QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) All Versions of this Article: jpet.103.056960v1 307/3/846    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Litman, T. Articles by Stein, W. D. Search for Related Content PubMed PubMed Citation Articles by Litman, T. Articles by Stein, W. D.

CELLULAR AND MOLECULAR

Pumping of Drugs by P-Glycoprotein: A Two-Step Process?

Bioinformatics Centre, University of Copenhagen, Copenhagen, Denmark (T.L.); Laboratory of Oncology, Herlev University Hospital, Herlev, Denmark (T.S., T.L.); and Silberman Institute of Life Sciences, Hebrew University of Jerusalem, Jerusalem, Israel (W.D.S.)

Received July 11, 2003; accepted August 21, 2003.

	Abstract

Top Abstract Materials and Methods Results Discussion Appendix References

 
The apparent inhibition constant, K_app, for the blockade of P-glycoprotein (P-gp) by four drugs, verapamil, cyclosporin A, XR9576 (tariquidar), and vinblastine, was measured by studying their ability to inhibit daunorubicin and calcein-AM efflux from four strains of Ehrlich cells with different levels of drug resistance and P-gp content. For daunorubicin as a transport substrate, K_app was independent of [P-gp] for verapamil but increased strictly linearly with [P-gp] for vinblastine, cyclosporin A, and XR9576. A theoretical analysis of the kinetics of drug pumping and its reversal shows that K_app for inhibition should increase linearly with the amount of pumps present in the membrane for a reverser that inhibits pumping from the cytoplasmic face. In contrast, if the reverser acts by blocking transport from the outer face, i.e., preemptively, K_app should be independent of the number of pumps present. The experimental data suggest that verapamil blocks pumping at the extracellular face of the membrane, whereas the other three blockers act on pumping from the cytoplasmic phase. The maximum degree of inhibition was the same for all four blockers; thus, they do not act in parallel but rather, in serial, i.e., a drug that is pumped from the cytoplasmic phase has to pass the preemptive route upon leaving the cell. Our results are consistent with the Sauna-Ambudkar two-step model for pumping by P-gp. We suggest that the vinblastine/cyclosporin A/XR9576-binding site accepts daunorubicin at the cytoplasmic face and transfers it to the verapamil-binding site, from where daunorubicin is emptied at the extracellular surface.

In the literature, variable values for IC₅₀ (the drug concentration that gives half-maximal inhibition) for P-glycoprotein reversal are reported for the same drug in different models. For example, for the blocker cyclosporin A, the IC₅₀ for inhibition of daunorubicin pumping by P-gp ranges from 0.04 µM (P388/ADR cells) (Ayesh et al., 1996) to ca. 4 µM (CEM/vinblastine1000 cells) (Hu et al., 1990), with a range of intermediate values (Cardarelli et al., 1995; Lan et al., 1996; Litman et al., 1997). For verapamil, the IC₅₀ for reversal of daunorubicin extrusion by P-gp ranges from 0.4 µM (SW2780AD cells) (Javaheri et al., 1983) to 10 µM (CEM/vinblastine1000 cells) (Hu et al., 1990), and the in vivo situation is further complicated because oxidative metabolites may act as substrates and inhibitors of P-gp (Pauli-Magnus et al., 2000).

Part of the variability can be ascribed to different measures of the reverser's effect on drug pumping, either by assessing the effect of the blocker on reversing resistance to cytotoxicity (Lee et al., 1994), or on reversing the drug accumulation deficit due to P-gp (Lan et al., 1996). The discrepancy remains, however, even when measurements are made using the same procedure. Thus, Hu et al. (1990) measured the effect of cyclosporin A and verapamil on daunorubicin accumulation in two resistant variants of CEM/CCRF cells and found, that for cyclosporin A, IC₅₀ increased from 1.7 µM in the moderately resistant subline CEM/vinblastine100 to 4 µM for the highly resistant CEM/vinblastine1000 line. For another P-gp blocker, GF120918, Chen et al. (2000) found that its IC₅₀ for reversal of Tc-tetrofosmin transport increased from 40 to 385 nM for a low P-gp-expressing and a high P-gp-expressing drug-resistant carcinoma cell line, respectively. In both these studies, the higher IC₅₀ was found for the more resistant cell line. We wondered whether there was some casual relationship between the rise in IC₅₀ and the increasing number of pump molecules. Therefore, we asked the question: Does IC₅₀ for a reverser vary with the pump/leak ratio, and what might be the basis for this?

In our previous studies, we measured drug accumulation and P-gp content in four Ehrlich ascites tumor cell (EATC) lines selected for increasing drug resistance (Litman et al., 1997). We found that daunorubicin accumulation varied inversely with the P-gp content of the cell. Thus, these four EATC lines (termed P6, P12, P36, and P72, possessing increasing levels of P-gp) provide a unique opportunity for testing whether IC₅₀ values for blockers do in fact vary with the pump/leak ratio and, if they do, is this in all circumstances, or only with certain substrates? We studied this experimentally by measuring the blockade of P-gp using two substrates and a number of reversers.

We first present the theory for the relation between IC₅₀ and the pump-leak ratio in different pumping models and then follow our experimental studies, designed around this theory.

	Materials and Methods

Top Abstract Materials and Methods Results Discussion Appendix References

 
Tumor Cells. Five different Ehrlich cell lines were used for the drug accumulation experiments. The parental, drug-sensitive tumor EHR2 and cells from the daunorubicin-resistant subline EHR2/0.8, maintained with daunorubicin 0.8 mg/kg x 4 weekly (50% of the LD₁₀ dose). Cells from passages 6, 12, 36, and 72, hereafter termed P6, P12, P36, and P72, were used. All cells were maintained in vivo in NMRI/DBA mice after intraperitoneal inoculation. No drug was administered in the last passage before the experiments. These cell lines have previously been characterized in our laboratories with respect to drug resistance (Nielsen et al., 1994), kinetics of daunorubicin transport (Nielsen et al., 1995), lipid composition (Litman et al., 1995), ATPase activity (Litman et al., 1997), pH regulation (Litman et al., 1998), and taurin uptake (Poulsen et al., 2002).

P-gp Expression. The P-gp content was quantified by Western blot analysis as described in detail in Nielsen et al. (1994). In essence, membrane-enriched pellets were separated by SDS-polyacrylamide gel electrophoresis, and the protein was transferred to nitrocellulose membranes, which were incubated overnight with C219 anti-P-gp antibody (Signet Laboratories, Dedham, MA). The P-gp-specific bands were scanned on a densitometer (Shimadzu, Kyoto, Japan), and the relative concentration of P-gp was calculated using a standard curve composed of membrane preparations with defined amounts of P-gp.

Daunorubicin Accumulation. The uptake of 5 µM daunorubicin was followed with time (5, 15, 29, 45, and 60 min) at 37°C, essentially as described by Skovsgaard (1978), and the effect of drugs on the steady-state accumulation of daunorubicin (at time 60 min) was investigated. To the incubation medium [57 mM NaCl, 5 KCl, 1.3 mM MgSO₄, 9 mM NaH₂PO₄, 51 mM Na₂HPO₄, 5% (v/v) fetal calf serum, pH 7.45] was added glucose to 10 mM and the test drug: either verapamil, cyclosporin A, vinblastine, or XR9576, at various concentrations. Energy deprivation experiments without glucose but including 10 mM sodium azide were performed to obtain an estimate of the maximum accumulation level of daunorubicin. The uptake of daunorubicin was quantitated fluorometrically in a Hitachi F3010 spectrofluorometer (Hitachi, Tokyo, Japan) with excitation wavelength 470 nm, and emission wavelength set to 585 nm. The kinetic parameter for half-maximal inhibition of pumping by P-gp, K_app, was obtained by curve fitting using the Marquardt-Levenberg algorithm (SigmaPlot; SPSS Science, Chicago, IL).

Calcein Accumulation. Calcein accumulation was measured in a 96-well plate assay modified after Eneroth et al. (2001). Ehrlich cells were harvested, washed, and resuspended in 100 µl of transport medium in 96-well microtiter plates. Next, the test drug (cyclosporin A, verapamil, vinblastine, or XR9576) at various concentrations was added to each well. After a 15-min preincubation, calcein-AM was added to 250 nM, and the plate was incubated at 37°C for 60 min. Calcein-specific fluorescence was measured in a Bio-Tek FL600 (BioTek, Boule Nordic, Denmark) microplate fluorescence reader with 485/20-nm excitation and 530/25-nm emission filters. For each test drug, the calcein accumulation (in arbitrary fluorescence units) was plotted versus drug concentration. The kinetic parameter for half-maximal inhibition of pumping by P-gp, K_app, was obtained by curve fitting using the Marquardt-Levenberg algorithm (SigmaPlot; SPSS Science).

Chemicals. XR9576 was kindly provided by Dr. Susan Bates (National Cancer Institute, National Institutes of Health, Bethesda, MD). All other chemicals were of analytical grade, purchased either from Merck (Darmstadt, Germany) or from Sigma-Aldrich (St. Louis, MO).

Theory. Figure 1A is a cartoon depicting the steady-state accumulation of a P-gp substrate that is pumped out from the cytoplasmic phase (where it is at concentration S_i) by a pump, which is operating at a velocity k_i against a bidirectional leak with rate p. (In the following treatment, the subscript i will be used to denote processes occurring at the inner phase of the membrane, while subscript o denotes processes at the outer surface). The pump is inhibited by a blocker having the intrinsic affinity K_i for the pump and being at concentration I. The following analyses are in terms of the concentration of free substrate. If a substrate is bound to some intracellular component (e.g., DNA, lipids, or trapped inside lysosomes) the concentration measured is the total concentration, which will be proportional to the free concentration where this is below saturation of the binding component.

View larger version (19K): [in this window] [in a new window]   Fig. 1. Three models for drug pumping by P-glycoprotein and their theoretical predictions. A and B, pumping is from the cytoplasmic face of the membrane. A, cartoon. The bidirectional leak rate constant is p and the pump rate ki. I is the blocker. The curves in B give the predictions of this model (see text, eq. 2) with pump/leak ratios set at 0.1, 1, 5, 10, and 100. The y-axis is the drug distribution ratio, the x-axis is the concentration of blocker where the value of Ki (see text) is set at 10. C and D, pumping is preemptive, i.e., drug is pumped out on its way into the membrane. C, cartoon. Symbols as in A. The curves in D give the predictions of this model (see text, eq. 3) with pump/leak ratios set at 0.1, 1, 5, and 10, and Ko set at 10. E, pumping is both preemptively and from the cytoplasmic face of the membrane. The curves give the predictions of the membrane-symmetric case of this model (see text, eq. 6) with pump/leak ratios set at 0.1, 1, 5, and 10, and Ki = Ko set at 10.

The ratio of free substrate concentrations (inside to outside) at steady state is given by

as elaborated on by Stein (1997). When substrate permeability is very high in comparison with the pump rate, i.e., p » k_i, the ratio of S_i/S_o approaches unity. Thus, it seems that even if drug is rapidly pumped out of the cell, an even faster back-diffusion will overwhelm the pump. The concept of "fast" versus "slow" diffusing agents has been comprehensively reviewed by Gera Eytan and coworkers (Eytan et al., 1996; Eytan and Kuchel, 1999), and the influence of passive permeability on the kinetics of P-gp has subsequently been demonstrated experimentally (Lentz et al., 2000; Westerhoff et al., 2000; Wielinga et al., 2000).

The fraction of pumping activity in the presence of I is

(1)

It is simple to show that at the steady state:

(2)

where k_i,0 is the uninhibited pump rate.

Figure 1B is computed from eq. 2 where S_i/S_o is plotted at different inhibitor concentrations I for five different ratios of the pumping rate constant k_i,0 to the leak rate p.

Figure 1C shows the steady-state level of a P-gp substrate that is pumped out preemptively, that is, en route across the cell membrane, as the compound enters the extracellular leaflet of the membrane bilayer. Here, the ratio of free substrate concentrations (inside to outside) at steady state is given by

Thus, if substrate permeability is very high in comparison with the pump rate, i.e., p » k_o, the ratio of S_i/S_o approaches unity. Thus, in this case too, even if drug is rapidly pumped out of the cell, back-diffusion will dominate and the effect of pumping will not be seen.

The pump is again inhibited by a blocker having intrinsic affinity K_o for the pump and being at concentration I (eq. 1 still applies).

For this case, at the steady state

(3)

Figure 1D is now computed from eq. 3 where S_i/S_o is plotted as a function of the inhibitor concentration I for four different ratios of the pumping rate constant k_i,o to the leak rate p. For this case, a constraint exists: k_o p, i.e., the pump cannot extract from the membrane more than enters it.

Visual inspection of Fig. 1, B and D, suggests that in the former case, the apparent half-saturation concentration K_app (i.e., the observed IC₅₀) increases with the rate of pumping, k_i,0, whereas in the latter case the K_app does not depend on the pumping rate, k_o.

Mathematically (see Appendix), eq. 2 can be solved to yield

(4)

Thus, in the case of a pump acting from the cytoplasmic phase, the half-saturation concentration for the effect of a blocker on the steady-state accumulation level (K_app) depends on the rate of pumping, that is, on the number of pumps present in the cell membrane.

If eq. 3 is similarly treated (see Appendix), one obtains

(5)

That is, for preemptive pumping, the apparent coefficient for inhibition by a blocker equals the intrinsic K_o.

Finally, one can analyze the situation where the pump acts both on substrate present in the inner layer and outer layer of the membrane.

For this case, at the steady state

(6)

Again, the constraint k_o p applies.

Figure 1E shows how S_i/S_o varies with the inhibitor concentration I at four different ratios of the pumping rate constant k to the leak rate p. Here, for simplicity, we have assumed that the properties of the pumping system are the same at both faces of the membrane, that is, that the affinity and the rate of pumping do not differ at the two faces.

Visual inspection and computer fitting of these predicted curves to a Michaelis-Menten kinetics model, shows that in this case, the apparent K_app increases with the pump rate, but to a maximum that is 2-fold the value at limiting low concentrations of the inhibitor.

	Results

Top Abstract Materials and Methods Results Discussion Appendix References

 
We have attempted to apply the theoretical analysis derived above to an experimental system, where we have a series of cell lines with different abilities to accumulate drugs and with measured (and differing) amounts of P-gp. This is equivalent to varying the pump rate k_i,o in the different cells, assuming that the pump rate is proportional to the amount of P-gp.

Figure 2A depicts the steady-accumulation of daunorubicin as a function of verapamil concentration for the four different EAT cell lines. The curves were fitted to the Michaelis-Menten equation

(7)

where V is the amount of daunorubicin accumulated at a concentration I of blocker, V₀ is the intracellular daunorubicin level in the absence of blocker, V_max is the daunorubicin accumulation at maximum concentration of blocker, and K_app is the blocker concentration which brings the daunorubicin accumulation to halfway between the value at I = 0 and that at I = .

View larger version (16K): [in this window] [in a new window]   Fig. 2. Experimental tests of the theoretical models of Fig. 1, using daunorubicin (DNR) as substrate. Accumulation of daunorubicin was measured during 60 min as described under Materials and Methods. Increasing levels of three blockers were used on four EATC lines, as indicated. In A, the blocker was verapamil, in B vinblastine, and in C XR9576. The lines are the best fits to the modified Michaelis-Menten equation as given in the text, eq. 7.

Figure 2B is a similar plot showing the effect of vinblastine used as a blocker of daunorubicin transport, and Fig. 2C shows a parallel experiment with the high-affinity blocker XR9576.

Table 1 records the values of K_app obtained by curve fitting for these three blockers as well as for cyclosporin A where similar data were obtained (not shown).

It would seem that for the blockers XR9576, cyclosporin A, and vinblastine the K_app increases with the amount of P-gp.

Figure 3 is a plot of K_app versus the P-gp level for all these blockers, where the K_app values have been normalized to the value found at the lowest P-gp concentration. The regression line through the data has r² = 0.90. The values of K_app at the highest values of P-gp content were close to 4 times the value at the lowest resistance level. In contrast (Table 1), the K_app for verapamil as a blocker does not seem to increase with the P-gp level.

View larger version (12K): [in this window] [in a new window]   Fig. 3. Kapp values derived from the curve fits of Fig. 2 for the blockers cyclosporin A, XR9576, and vinblastine plotted against the P-gp level of the four cell lines studied. Symbols are cyclosporin A, ; vinblastine, ; and XR9576, . In each case, the Kapp is normalized to the value found for the P6 line set at 1.

By transforming eq. 4, one can obtain the result that

Now, p/(p + k_i,0) is simply the ratio of unblocked to blocked accumulation of daunorubicin, which can be found from the extrapolates to zero and infinite blocker of the curves of Fig. 2 and the similar data for cyclosporin A.

Table 1 records also these estimates of p/(p + k_i,0) for each blocker and the corresponding values of K_i, calculated from K_app and the corresponding value of p/(p + k_i,0). It seems that the derived values of K_i do not vary with increasing amounts of P-gp. The mean of the recorded values of K_i for each blocker can therefore be taken as an estimate of the K_i for that blocker in its action of inhibiting P-gps function from the cytoplasmic surface.

Similar sets of data were obtained for calcein accumulation in the presence of these four blockers. Figure 4 depicts the data gathered for the blockers verapamil, vinblastine and cyclosporin A. The values for K_app obtained by curve fitting to eq.7 are collected in Table 1 together with the parameters found for XR9576. For none of these blockers does K_app increase significantly with the amount of P-gp, in contrast to the situation found for daunorubicin with the blockers cyclosporin A, XR9576, and vinblastine.

View larger version (16K): [in this window] [in a new window]   Fig. 4. Experimental tests of the theoretical models of Fig. 1, using calcein-AM as substrate. Accumulation of calcein was measured during 60 min as described under Materials and Methods. Increasing levels of three blockers were used on four EATC lines, as indicated. In A, the blocker was verapamil, in B vinblastine, and in C cyclosporin A. The lines are the best fits to the modified Michaelis-Menten equation as given in the text, eq. 7.

	Discussion

Top Abstract Materials and Methods Results Discussion Appendix References

 
In the theory section, it was shown that K_app, the half-inhibition constant for the action of a blocker on a drug transport pump, increases with the amount of pumps present if pumping is occurring from the cytoplasmic phase. In contrast, if pumping occurs preemptively, K_app is independent of the number of transporters. Chen et al. (2000) studied the action of the blocker GF120918 on the accumulation of the P-gp substrate Tc-tetrofosmin in drug-resistant epidermoid carcinoma cell lines. It was found that the K_app values for GF120918 were 40 and 385 nM for a moderately (low P-gp expression) and a highly (high P-gp expression) resistant cell line, respectively. This would be understandable if the effect of GF120918 was to block the pump at the cytoplasmic surface. Our data (Fig. 2; Table 1) show that for the accumulation of the P-gp substrate daunorubicin, the K_app for vinblastine, cyclosporin A, and XR9576 increase with the amount of P-gp expression (Fig. 3). This suggests that the action of these blockers could also be on the cytoplasmatic side of the cell membrane (Fig. 1A). If this interpretation is correct, one would be entitled to compute the intrinsic affinity of blocker for the pump, K_i, from the several values of K_app and the corresponding value for the steady-state accumulation ratio for each cell line. The estimates of K_i are collected in Table 1 for cyclosporin A, XR9576, and vinblastine.

In contrast, for verapamil K_app does not seem to increase with P-gp expression, suggesting that verapamil interferes with preemptive pumping (Fig. 1C). Thus, the K_i for verapamil is given directly by the measured value of K_app (Table 1).

The situation with calcein-AM seems more complex. Here (Fig. 4; Table 1), for all the blockers, K_app does not seem to increase with P-gp expression. It could be that calcein-AM itself is preemptively pumped out of the membrane, differing in this way from daunorubicin. This would be in accordance with the suggestion of Homolya et al. (1994), the pioneers in the use of calcein-AM as a P-gp substrate, namely, that calcein-AM and other fluorescent methyl esters are extruded directly from the cell membrane, before reaching the cytoplasmic phase. Preemptive pumping of calcein-AM is also in accordance with the theoretical analysis of Stein (1997), who showed that the initial rate of substrate accumulation is reduced by pumping only if such pumping is preemptive. Calcein-AM uptake is always measured as an initial rate, because free calcein-AM does not accumulate within the cell, calcein-AM being cleaved by intracellular esterases to free calcein as soon as it enters the cell (Holló et al., 1994). Because the blockers do affect the initial rate of calcein accumulation, and hence the rate of calcein-AM extrusion, they must be affecting the initial rate of the process.

If this interpretation is correct, that the effect of blockers on daunorubicin accumulation can occur at the cytoplasmic surface, whereas the effect on calcein accumulation is on preemptive pumping, then one might be able to understand why the intrinsic K_i values for cyclosporin A, vinblastine, and XR9576 found for daunorubicin accumulation differ markedly from the values found for calcein accumulation (Table 1). This finding would imply that the intrinsic affinity of the blockers at the cytoplasmic surface is some 4 times higher than the affinity at the outer half of the bilayer. For verapamil as a blocker, on the other hand, the K_app value for its action on daunorubicin accumulation is not different from its value for blocking calcein-AM extrusion. This would be expected if both measures were of the same process, namely, preemptive pumping of substrate.

Sauna and Ambudkar (2001) have proposed a two-step model for the catalytic cycle of P-gp. Here, drug is bound at a high-affinity site at the cytoplasmic surface and then moves to a low-affinity site, from which it is extruded to the extracellular medium. It is tempting to identify the high-affinity blocker site that our analysis suggests at the cytoplasmic surface with the high-affinity site for drugs postulated in the Sauna-Ambudkar model. The low-affinity site for blockers, which our analysis suggests is concerned in preemptive pumping, might correspond to the low-affinity drug-binding site in the above-mentioned model. This interpretation is also supported by transport data obtained with a series of human lymphoblastoid cell lines (CEM wt, CEM ADR5000, and CEM Col1000); here, the affinity of verapamil to P-gp differed at the two sides of the plasma membrane (Köhler and Stein, 2003).

What still has to be accounted for is that verapamil, which we postulate is blocking at the preemptive site, can block all of the effect of the pump on daunorubicin accumulation, that is, both from the cytoplasmic surface and the preemptive pumping. This would be expected if the pathway for drug extrusion occurs by a serial process, i.e., drug that is being pumped from the cytoplasmic phase has to go through the preemptive passage on its way out of the cell (Fig. 5). However, an alternative explanation of the experimental results could be that verapamil being a fast-diffusing drug is likely to quickly partition across the membrane and accumulate also in the cytosolic leaflet of the membrane (which is rich in negatively charged lipids that would tend to interact electrostatically with the cationic compound), thus, exerting its inhibitory effect at both sites of the membrane. That the biophysical properties of membrane lipids are affected by chemosensitizers such as verapamil is suggested in a recent review by Hendrich and Michalak (2003).

View larger version (47K): [in this window] [in a new window]   Fig. 5. Drug pumping as a serial process. Cartoon depicting daunorubicin being pumped across the membrane by P-gp in two-site manner. The blockers cyclosporin A (CsA), XR9576 (XR), and vinblastine (VL) block at the cytoplasmic face, verapamil (VER) at the extracellular face (preemptively), but daunorubicin (DNR) must egress through the extracellular site if it is to leave the membrane.

	Appendix

Top Abstract Materials and Methods Results Discussion Appendix References

 
Derivation of

For pumping from the cytoplasmic face S_i/S_o = p/(p + k_i). At zero concentration of blocker (I = 0), this becomes S_i/S_o = p/(p + k_i,0). At infinitely high concentration of I, S_i/S_o = 1, which equals (p + k_i,0)/(p + k_i,0).

Define {S_i/S_o}_1/2 as the value of S_i/S_o halfway between none and complete inhibition, and K_app as the value of I at this S_i/S_o.

Then, at I = K_app:

(A1)

which is halfway between S_i/S_o at I = 0, and S_i/S_o at I = .

Substituting K_app for I in eq. 2,

(A2)

Equating the right-hand sides of equations A1 and A2, and solving for K_app in terms of p and k_i,0, we obtain:

Derivation of K_app = K_o (eq. 5). For preemptive pumping, S_i/S_o = (p – k_o)/p. At zero concentration of blocker (I = 0), this becomes S_i/S_o = (p – k_o,0)/p. At infinitely high concentration of I, S_i/S_o = 1, which equals p/p.

Again, define {S_i/S_o}_1/2 as the value of S_i/S_o halfway between none and complete inhibition, and K_app as the value of I this S_i/S_o.

Then, at I = K_app:

(A3)

which is halfway between S_i/S_o at I = 0, and S_i/S_o at I = .

Substituting K_app for I in eq. 3,

(A4)

Again, equating the right-hand sides of eqs. A3 and A4, and solving for K_app, we obtain K_app = K_o.

	Acknowledgements

 
We thank Dr. Susan E. Bates for providing the XR9576. For excellent technical assistance, we are grateful to Marianne Knudsen.

	Footnotes

 
During the course of this work, T.L. was a Weimann Associate Research Professor and recipient of research grants from the Novo Nordisk Foundation, and "Eva and Henry Frænkels Mindefond".

DOI: 10.1124/jpet.103.056960.

ABBREVIATIONS: P-gp, P-glycoprotein; EATC, Ehrlich ascites tumor cell; AM, acetoxymethyl ester; K_app, apparent half-saturation concentration (e.g., IC₅₀); I, inhibitor concentration; k_i, the pump rate for pumping from the inside; k_i,0, the uninhibited pump rate for pumping from the inside; K_i, intrinsic affinity for P-gp at the inner leaflet of the cell membrane; k_o, the pump rate for (preemptive) pumping from the outer face of the membrane; k_o,0, the uninhibited pump rate for pumping from the outer face of the membrane; K_o, intrinsic affinity for P-gp at the outer face of the membrane (preemptive pumping); p, the leak rate of a compound that crosses the membrane by passive diffusion; S_i, intracellular substrate concentration; S_o, extracellular substrate concentration; GF120918, elacridar; XR9576, tariquidar.

Address correspondence to: Dr. Thomas Litman, Bioinformatics Centre, University of Copenhagen, Universitetsparken 15, Bldg. 10, DK-2100 Copenhagen, Denmark. E-mail: tlitman{at}binf.ku.dk

	References

Top Abstract Materials and Methods Results Discussion Appendix References

 

Ayesh S, Shao Y-M, and Stein WD (1996) Co-operative, competitive and noncompetitive interactions between modulators of P-glycoprotein. Biochim Biophys Acta 1316: 8–18.[Medline]

Bates SE, Wilson WH, Fojo AT, Alvarez M, Zhan Z, Regis J, Robey R, Hose C, Monks A, Kang YK, and Chabner B (1996) Clinical reversal of multidrug resistance. Stem Cells 14: 56–63.[Abstract]

Borst P, Evers R, Kool M, and Wijnholds J (2000) A family of drug transporters: the multidrug resistance-associated proteins. J Natl Cancer Inst 92: 1295–1302.[Abstract/Free Full Text]

Cardarelli CO, Aksentijevich I, Pastan I, and Gottesman MM (1995) Differential effects of P-glycoprotein inhibitors on NIH3T3 cells transfected with wild-type (G185) or mutant (V185) multidrug transporters. Cancer Res 55: 1086–1091.[Abstract/Free Full Text]

Chabner BA and Fojo AT (1998) Multidrug resistance: P-glycoprotein and its allies - the elusive foes. J Natl Cancer Inst 81: 910–913.

Chen WS, Luker KE, Dahlheimer JL, Pica CM, Luker GD, and Piwnica-Worms D (2000) Effects of MDR1 and MDR3 P-glycoproteins, MRP1 and BCRP/MXR/ABCP on the transport of ^99mTc-tetrofosmin. Biochem Pharmacol 60: 413–426.[CrossRef][Medline]

Eneroth A, Åström E, Hoogstraate J, Schrenk D, Conrad S, Kauffmann H-M, and Gjellan K (2001) Evaluation of a vincristine resistant Caco-2 cell line for use in a calcein AM extrusion screening assay for P-glycoprotein interaction. Eur J Pharm Sci 12: 205–214.[CrossRef][Medline]

Eytan GD and Kuchel PW (1999) Mechanism of action of P-glycoprotein in relation to passive membrane permeation. Int Rev Cytol 190: 175–250.[Medline]

Eytan GD, Regev R, Oren G, and Assaraf YG (1996) The role of passive transbilayer drug movement in multidrug resistance and its modulation. J Biol Chem 271: 12897–12902.[Abstract/Free Full Text]

Hendrich AB and Michalak K (2003) Lipids as a target for drugs modulating multidrug resistance of cancer cells. Curr Drug Targets 4: 23–30.[CrossRef][Medline]

Holló Z, Homolya L, Davis CW, and Sarkadi B (1994) Calcein accumulation as a fluorometric functional assay of the multidrug transporter. Biochim Biophys Acta 1191: 384–388.[Medline]

Homolya L, Holló Z, Germann UA, Pastan I, Gottesman MM, and Sarkadi B (1994) Fluorescent cellular indicators are extruded by the multidrug resistance protein. J Biol Chem 268: 21493–21496.

Hu XF, de Luise M, Martin TJ, and Zalcberg JR (1990) Effect of cyclosporin and verapamil on the cellular kinetics of daunorubicin. Eur J Cancer 26: 814–817.

Javaheri S, De Hemptinne A, Vanheel B, and Leusen I (1983) Changes in brain ECF pH during metabolic acidosis and alkalosis: a microelectrode study. J Appl Physiol 55: 1849–1853.[Abstract/Free Full Text]

Krishna R and Mayer LD (2001) Modulation of P-glycoprotein (PGP) mediated multidrug resistance (MDR) using chemosensitizers: recent advances in the design of selective MDR modulators. Curr Med Chem Anticancer Agents 1: 163–174.

Köhler S and Stein WD (2003) Optimizing chemotherapy by measuring reversal of P-glycoprotein activity in plasma membrane vesicles. Biotechnol Bioeng 81: 507–517.[Medline]

Lan LB, Ayesh S, Lyubimov E, Pashinsky I, and Stein WD (1996) Kinetic parameters for reversal of the multidrug pump as measured for drug accumulation and cell killing. Cancer Chemother Pharmacol 38: 181–190.[CrossRef][Medline]

Lee J-S, Paull K, Alvarez M, Hose C, Monks A, Grever M, Fojo AT, and Bates SE (1994) Rhodamine efflux patterns predict P-glycoprotein substrates in the National Cancer Institute drug screen. Mol Pharmacol 46: 627–638.[Abstract]

Lentz KA, Polli JW, Wring SA, Humphreys JE, and Polli JE (2000) Influence of passive permeability on apparent P-glycoprotein kinetics. Pharmaceut Res 17: 1456–1460.[CrossRef][Medline]

Litman T, Druley T, Stein WD, and Bates SE (2001) From MDR to MXR: new understanding of multidrug resistance systems, their properties and clinical significance. Cell Mol Life Sci 58: 931–959.[CrossRef][Medline]

Litman T, Nielsen D, Skovsgaard T, and Bukhave K (1995) Lipid composition of sensitive and multidrug resistant Ehrlich ascites tumour cells. Cell Pharm 2: 9–14.

Litman T, Nielsen D, Skovsgaard T, Zeuthen T, and Stein WD (1997) ATPase activity of P-glycoprotein related to emergence of drug resistance in Ehrlich ascites tumor cell lines. Biochim Biophys Acta 1361: 147–158.[Medline]

Litman T, Pedersen SF, Kramhøft B, Skovsgaard T, and Hoffmann EK (1998) pH regulation in sensitive and multidrug resistant Ehrlich ascites tumor cells. Cell Physiol Biochem 8: 138–150.[CrossRef][Medline]

Nielsen D, Maare C, and Skovsgaard T (1994) Kinetics of daunorubicin transport in Ehrlich ascites tumor cells with different expression of P-glycoprotein. Biochem Pharmacol 47: 2125–2135.[CrossRef][Medline]

Nielsen D, Maare C, and Skovsgaard T (1995) Influx of daunorubicin in multidrug resistant Ehrlich ascites tumour cells: correlation to expression of P-glycoprotein and efflux. Influence of verapamil. Biochem Pharmacol 50: 443–450.[CrossRef][Medline]

Pauli-Magnus C, von Richter O, Burk O, Ziegler A, Mettang T, Eichelbaum M, and Fromm MF (2000) Characterization of the major metabolites of verapamil as substrates and inhibitors of P-glycoprotein. J Pharmacol Exp Ther 293: 376–382.[Abstract/Free Full Text]

Poulsen KA, Litman T, Eriksen J, Mollerup J, and Lambert IH (2002) Downregulation of taurine uptake in multidrug resistant Ehrlich ascites tumor cells. Amino Acids 22: 333–350.[Medline]

Sauna ZE and Ambudkar SV (2001) Characterization of the catalytic cycle of ATP hydrolysis by human P-glycoprotein. The two ATP hydrolysis events in a single catalytic cycle are kinetically similar but affect different functional outcomes. J Biol Chem 276: 11653–11661.[Abstract/Free Full Text]

Seelig A and Landwojtowicz E (2000) Structure-activity relationship of P-glycoprotein substrates and modifiers. Eur J Pharm Sci 12: 31–40.[CrossRef][Medline]

Skovsgaard T (1978) Carrier-mediated transport of daunorubicin, adriamycin and rubidazone in Ehrlich ascites tumour cells. Biochem Pharmacol 27: 1221–1227.[CrossRef][Medline]

Skovsgaard T, Danø K, and Nissen NI (1984) Chemosensitizers counteracting acquired resistance to anthracyclines and vinca alkaloides in vivo. A new treatment principle. Cancer Treat Rev 11: 63–72.

Sparreboom A, Danesi R, Ando Y, Chan J, and Figg WD (2003) Pharmacogenomics of ABC transporters and its role in cancer chemotherapy. Drug Resist Updat 6: 71–84.[CrossRef][Medline]

Stein WD (1997) Kinetics of the multidrug transporter (P-glycoprotein) and its reversal. Physiol Rev 77: 545–590.[Abstract/Free Full Text]

Westerhoff HV, Riethorst A, and Jongsma AP (2000) Relating multidrug resistance phenotypes to the kinetic properties of their drug-efflux pumps. Eur J Biochem 267: 5355–5368.[Medline]

Wielinga PR, Westerhoff HV, and Lankelma J (2000) The relative importance of passive and P-glycoprotein mediated anthracycline efflux from multidrug-resistant cells. Eur J Biochem 267: 649–657.[Medline]

This article has been cited by other articles:

				B. Sarkadi, L. Homolya, G. Szakacs, and A. Varadi Human Multidrug Resistance ABCB and ABCG Transporters: Participation in a Chemoimmunity Defense System. Physiol Rev, October 1, 2006; 86(4): 1179 - 1236. [Abstract] [Full Text] [PDF]

				E. F. Choo, D. Kurnik, M. Muszkat, T. Ohkubo, S. D. Shay, J. N. Higginbotham, H. Glaeser, R. B. Kim, A. J. J. Wood, and G. R. Wilkinson Differential in Vivo Sensitivity to Inhibition of P-glycoprotein Located in Lymphocytes, Testes, and the Blood-Brain Barrier J. Pharmacol. Exp. Ther., June 1, 2006; 317(3): 1012 - 1018. [Abstract] [Full Text] [PDF]

				R. B. Walter, J. L. Pirga, M. R. Cronk, S. Mayer, F. R. Appelbaum, and D. E. Banker PK11195, a peripheral benzodiazepine receptor (pBR) ligand, broadly blocks drug efflux to chemosensitize leukemia and myeloma cells by a pBR-independent, direct transporter-modulating mechanism Blood, November 15, 2005; 106(10): 3584 - 3593. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: jpet.103.056960v1 307/3/846    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Litman, T. Articles by Stein, W. D. Search for Related Content PubMed PubMed Citation Articles by Litman, T. Articles by Stein, W. D.