It is generally accepted that P-glycoprotein binds its substrates in the lipid phase of the membrane. Quantification and characterization of the lipid-transporter binding step are, however, still a matter of debate. We therefore selected 15 structurally diverse drugs and measured the binding constants from water to the activating (inhibitory) binding region of P-glycoprotein, K(tw(1)) (K(tw(2))), as well as the lipid-water partition coefficients, K(lw). The former were obtained by measuring the concentrations of half-maximum activation (inhibition), K(1) (K(2)), in living NIH-MDR-G185 mouse embryo fibroblasts using a Cytosensor microphysiometer, and the latter were derived from surface activity measurements. This allowed determination of the membrane concentration of drugs at half-maximum P-glycoprotein activation (C(b(1)) = (0.02 to 67) mmol/L lipid), which is much higher than the corresponding aqueous concentration (K(1) = (0.02 to 376) microM). Moreover we determined the free energy of drug binding from water to the activating binding region of the transporter (DeltaG degrees (tw(1)) = (-30 to -54) kJ/mol), the free energy of drug partitioning into the lipid membrane (DeltaG degrees (lw) = (-23 to -34) kJ/mol), and, as the difference of the two, the free energy of drug binding from the lipid membrane to the activating binding region of the transporter (DeltaG degrees (tl(1)) = (-7 to -27) kJ/mol). For the compounds tested DeltaG degrees (tl(1)) was less negative than DeltaG degrees (lw) but varied more strongly. The free energies of substrate binding to the transporter within the lipid phase, DeltaG degrees (tl(1)), are consistent with a modular binding concept, where the energetically most efficient binding module comprises two hydrogen bond acceptor groups.

P-glycoprotein (MDR1, ABCB1) is an ATP-dependent efflux transporter of a large variety of compounds. To understand P-glycoprotein in more detail, it is important to elucidate its activity in the cellular ensemble as well as in plasma membrane vesicles (under conditions where other ATP dependent proteins are blocked). We measured P-glycoprotein activity in inside-out vesicles formed from plasma membranes of MDR1-transfected mouse embryo fibroblasts (NIH-MDR1-G185) for comparison with previous measurements of P-glycoprotein activity in living NIH-MDR1-G185 cells. In plasma membrane vesicles activity was measured by monitoring phosphate release upon ATP hydrolysis and in living cells by monitoring the extracellular acidification rate upon ATP synthesis via glycolysis. P-glycoprotein was stimulated as a function of the concentration with 19 structurally different drugs, including local anesthetics, cyclic peptides, and cytotoxic drugs. The concentrations of half-maximum P-glycoprotein activation, K1, were identical in inside-out plasma membrane vesicles and in living cells and covered a broad range of concentrations (K1 approximately (10(-8)-10(-3)) M). The influence of the pH, drug association, and vesicle aggregation on the concentration of half-maximum P-glycoprotein activation was investigated. The turnover numbers in plasma membrane vesicles and in living cells were also approximately identical if the latter were measured in the presence of pyruvate. However, in the absence of pyruvate they were higher in living cells. The rate of ATP hydrolysis/ATP synthesis decreased exponentially with decreasing free energy of drug binding from water to the transporter, DeltaG0(tw)(1) (or increasing binding affinity). This suggests that drug release from the transmembrane domains has to occur before ATP is hydrolyzed for resetting the transporter.

Macitentan is a dual endothelin receptor antagonist under phase 3 investigation in pulmonary arterial hypertension. We investigated the effect of cyclosporine (Cs) and rifampin on the pharmacokinetics of macitentan and its metabolites ACT-132577 and ACT-373898 in healthy male subjects. In addition, in vitro studies were performed to investigate interactions between macitentan and its active metabolite ACT-132577 with human organic anion-transporting polypeptides (OATPs). The clinical study (AC-055-111) was conducted as a two-part, one-sequence, crossover study. Ten subjects in each part received multiple-dose macitentan followed by multiple-dose co-administration of Cs (part A) or rifampin (part B). In the presence of Cs, steady-state area under the plasma concentration–time profiles during a dose interval (AUCτ) for macitentan and ACT-373898 increased 10% and 7%, respectively, and decreased 3% for ACT-132577. Steady-state AUCτ of macitentan and ACT-373898 in the presence of rifampin decreased 79% and 64%, respectively. For ACT-132577, no relevant difference in AUCτ between the two treatments was observed. Macitentan co-administered with Cs or rifampin was well tolerated. The complementary in vitro studies demonstrated no marked differences in uptake rates of macitentan and ACT-132577 between the wild-type and OATP over-expressing cells over the concentration range tested. Concomitant treatment with Cs did not have any clinically relevant effect on the exposure to macitentan or its metabolites, at steady-state. Concomitant treatment with rifampin reduced significantly the exposure to macitentan and its metabolite ACT-373898 at steady-state but did not affect the exposure to the active metabolite ACT-132577 to a clinically relevant extent.

P-glycoprotein ATPase activity has been studied almost exclusively by measuring inorganic phosphate release from inside-out cellular vesicles. We have recently proposed a new method based on measurements of the extracellular acidification rate (ECAR) of living cells with a Cytosensor microphysiometer. This method allows for systematic investigation of the various factors influencing P-glycoprotein activation in living cells. Basal metabolic rates or ECARs of different MDR1-transfected cell lines were compared with those of the Mdr1a(-/-)1b(-/-) knockout, MRP1-transfected, and corresponding wild-type cell lines. Basal ECARs of all cells were on the order of 10(7) protons/cell/s, whereby those of genetically modified cells were on average (over all cell lines) slightly lower than those of wild-type cells. The expression level of P-glycoprotein in MDR1-transfected cells had no influence on basal ECARs. Verapamil-induced ECARs were specific for MDR1-transfected cells and increased with the expression level of P-glycoprotein. Moreover, ECARs were dependent on the metabolic state of the cell and were (2.8 +/- 1.2) x 10(6) and (8.0 +/- 1.5) x 10(6) protons/cell/s in glucose-deficient and glucose-fed NIH-MDR-G185 cells, respectively, after verapamil (10 muM) stimulation. The ECARs were practically identical to the rates of lactate extrusion and thus reflect the rates of ATP synthesis via glycolysis. Taking into account the number of P-glycoprotein molecules per cell, the rate of ATP hydrolysis in inside-out vesicles of the same cells was determined as (9.2 +/- 1.5) x 10(6) phosphates/cell/s, in good agreement with the rate of ATP synthesized in glucose-fed cells. The energy required for P-glycoprotein activation relative to the basal metabolic energy was twice as large in glucose-deficient as in glucose-fed cells, suggesting cellular protection by P-glycoprotein even under conditions of starvation.

P-glycoprotein (ABCB1) prevents absorption (e.g., blood-brain barrier) or enhances excretion (e.g., kidney) by moving substrates from the cytosolic to the extracellular membrane leaflet at the expense of ATP hydrolysis. It translocates various drugs and functions in membranes exhibiting different lateral packing densities. To gain more functional insight, we measured the temperature dependence of the P-glycoprotein ATPase activity in NIH-MDR1-G185 cell membranes in the absence and presence of three drugs (promazine, verapamil, and PSC833), exhibiting significantly different transporter affinities. Activation enthalpies (Delta H(++)) and entropies ( TDelta S(++)) were derived from Eyring plots. In the absence of drugs, the activation enthalpy and the free energy of activation for P-glycoprotein ATPase activity was determined as Delta H(++) = 92.6 +/- 4.2 kJ/mol and Delta G(++) = 73.1 +/- 7.2 kJ/mol, respectively. Increasing the drug concentration reduced the activation enthalpy, whereby the drug with the highest transporter affinity had the strongest effect (DeltaDelta H(++) = -21%). The free energy of activation decreased for activating (DeltaDelta G(++) = approximately -3.8%) and increased for inhibitory compounds (DeltaDelta G(++) = approximately +0.7%). The drug-specific changes of the free energy of activation are thus barely above thermal energy. A comparison with literature data revealed that a decrease of the lateral membrane packing density reduces the enthalpic and the entropic contribution to the free energy of activation. Although the P-glycoprotein ATPase activity increases only slightly with decreasing lateral membrane packing density, the mode of action changes from strongly entropy-driven at high, to essentially enthalpy-driven at low packing densities. This suggests that the transporter and the membrane form a functional entity.

Treatment of pulmonary arterial hypertension with the endothelin receptor antagonist bosentan has been associated with transient increases in liver transaminases. Mechanistically, bosentan inhibits the bile salt export pump (BSEP) leading to an intrahepatic accumulation of cytotoxic bile salts, which eventually results in hepatocellular damage. BSEP inhibition by bosentan is amplified by its accumulation in the liver as bosentan is a substrate of organic anion-transporting polypeptide (OATP) transport proteins. The novel endothelin receptor antagonist macitentan shows a superior liver safety profile. Introduction of the less acidic sulfamide moiety and increased lipophilicity yield a hepatic disposition profile different from other endothelin receptor antagonists. Passive diffusion rather than OATP-mediated uptake is the driving force for macitentan uptake into the liver. Interaction with the sodium taurocholate cotransporting polypeptide and BSEP transport proteins involved in hepatic bile salt homeostasis is therefore limited due to the low intrahepatic drug concentrations. Evidence for this conclusion is provided by in vitro experiments in drug transporter-expressing cell lines, acute and long-term studies in rats and dogs, absence of plasma bile salt changes in healthy human volunteers after multiple dosing, and finally the liver safety profile of macitentan in the completed phase III morbidity/mortality SERAPHIN (Study with an Endothelin Receptor Antagonist in Pulmonary Arterial Hypertension to Improve Clinical Outcome) trial.

We compared the P-glycoprotein ATPase activity in inside-out plasma membrane vesicles and living NIH-MDR1-G185 cells with the aim to detect substrate transport. To this purpose we used six substrates which differ significantly in their passive influx through the plasma membrane. In cells, the cytosolic membrane leaflet harboring the substrate binding site of P-glycoprotein has to be approached by passive diffusion through the lipid membrane, whereas in inside-out plasma membrane vesicles, it is accessible directly from the aqueous phase. Compounds exhibiting fast passive influx compared to active efflux by P-glycoprotein induced similar ATPase activity profiles in cells and inside-out plasma membrane vesicles, because their concentrations in the cytosolic leaflets were similar. Compounds exhibiting similar influx as efflux induced in contrast different ATPase activity profiles in cells and inside-out vesicles. Their concentration was significantly lower in the cytosolic leaflet of cells than in the cytosolic leaflet of inside-out membrane vesicles, indicating that P-glycoprotein could cope with passive influx. P-glycoprotein thus transported all compounds at a rate proportional to ATP hydrolysis (i.e. all compounds were substrates). However, it prevented substrate entry into the cytosol only if passive influx of substrates across the lipid bilayer was in a similar range as active efflux.

AIMSThis study investigated the effect of a fixed dose combination of lopinavir/ritonavir on the pharmacokinetics (PK) of selexipag and its active metabolite ACT-333679. METHODSThis was an open label, randomized, single centre, two way, crossover study. Twenty healthy male subjects were treated with a single dose of 400 μg selexipag alone and in combination with multiple doses of lopinavir/ritonavir (400/100 mg) twice daily. RESULTSThe results showed that lopinavir/ritonavir approximately doubled the exposure to selexipag. The area under the plasma concentration-time curve from time zero to infinity (AUC(0,1) and the maximum plasma concentration (C max ) of selexipag were 2.2-and 2.1-fold higher, respectively, than under selexipag alone, with a 90% confidence interval (CI) of the geometric mean ratio (GMR) of 1.9, 2.7 and 1.7, 2.6, respectively. For ACT-333679, the clinically more relevant component of selexipag, systemic exposure was increased by 8% (GMR of AUC(0,1) 1.1, 90% CI 0.9, 1.3), when lopinavir/ritonavir was co-administered with selexipag. The most frequently reported adverse event (AE) was headache. A single dose of selexipag, administered either alone or together with multiple doses of lopinavir/ritonavir, was safe and well tolerated. CONCLUSIONSLopinavir/ritonavir does not affect the PK parameters of selexipag and ACT-333679 to a clinically relevant extent. Therefore, adaptation of the selexipag dose is not required when co-administered with inhibitors of the organic anion-transporting polypeptide (OATP) 1B1/ 1B3, P-glycoprotein (P-gp) and/or CYP3A4.