Title: Relation Between the Turnover Number for Vinblastine Transport and for Vinblastine-stimulated ATP Hydrolysis by Human P-glycoprotein
Abstract: Considerable uncertainty surrounds the stoichiometry of coupling of ATP hydrolysis to drug pumping by P-glycoprotein, the multidrug transporter. To estimate relative turnovers for pumping of the drug vinblastine and ATP hydrolysis, we began by measuring the number of P-glycoprotein molecules on the surface of murine NIH3T3 cells expressing the human MDR1 gene. Fluorescence of cells treated with monoclonal antibody UIC2 was determined as a function of (i) amount of antibody at a fixed number of cells and (ii) increasing cell number at constant antibody. The two together gives 1.95 × 106 P-glycoprotein molecules/cell. Initial uptake rates of vinblastine ± verapamil measure the ability of P-glycoprotein to extract vinblastine from the plasma membrane before it enters the cell. As a function of [vinblastine] at 37 °C, they give the maximum rate of this component of outward pumping as 2.1 × 106 molecules s−1 cell−1 or a turnover number of 1.1 s−1. Initial rates of one-way efflux as a function of [vinblastine] at 25 °C ± glucose give the maximum rate of this component of pumping as 0.59 × 106 molecules s−1 cell−1. The ratio of ATPase activity of P-glycoprotein at 37 and 25 °C is 4.6. Appropriating this ratio for pumping, maximum one-way efflux at 37 °C is 4.6 × 0.59 = 2.7 × 106 molecules s−1cell−1, a turnover number of 1.4 s−1. The vinblastine-stimulated ATPase activity of P-glycoprotein has a turnover number of 3.5 s−1 at 37 °C, giving 2.8 molecules of ATP hydrolyzed for every vinblastine molecule transported in a particular direction. These calculations involve several approximations, but turnover numbers for pumping of vinblastine and for vinblastine-stimulated ATP hydrolysis are comparable. Thus, ATP hydrolysis is probably directly linked to drug transport by P-glycoprotein. Considerable uncertainty surrounds the stoichiometry of coupling of ATP hydrolysis to drug pumping by P-glycoprotein, the multidrug transporter. To estimate relative turnovers for pumping of the drug vinblastine and ATP hydrolysis, we began by measuring the number of P-glycoprotein molecules on the surface of murine NIH3T3 cells expressing the human MDR1 gene. Fluorescence of cells treated with monoclonal antibody UIC2 was determined as a function of (i) amount of antibody at a fixed number of cells and (ii) increasing cell number at constant antibody. The two together gives 1.95 × 106 P-glycoprotein molecules/cell. Initial uptake rates of vinblastine ± verapamil measure the ability of P-glycoprotein to extract vinblastine from the plasma membrane before it enters the cell. As a function of [vinblastine] at 37 °C, they give the maximum rate of this component of outward pumping as 2.1 × 106 molecules s−1 cell−1 or a turnover number of 1.1 s−1. Initial rates of one-way efflux as a function of [vinblastine] at 25 °C ± glucose give the maximum rate of this component of pumping as 0.59 × 106 molecules s−1 cell−1. The ratio of ATPase activity of P-glycoprotein at 37 and 25 °C is 4.6. Appropriating this ratio for pumping, maximum one-way efflux at 37 °C is 4.6 × 0.59 = 2.7 × 106 molecules s−1cell−1, a turnover number of 1.4 s−1. The vinblastine-stimulated ATPase activity of P-glycoprotein has a turnover number of 3.5 s−1 at 37 °C, giving 2.8 molecules of ATP hydrolyzed for every vinblastine molecule transported in a particular direction. These calculations involve several approximations, but turnover numbers for pumping of vinblastine and for vinblastine-stimulated ATP hydrolysis are comparable. Thus, ATP hydrolysis is probably directly linked to drug transport by P-glycoprotein. Most people who die from cancer do so because their tumors have metastasized and become resistant to chemotherapy. Until we find ways to prevent cancer entirely, overcoming drug resistance is the main hope to save lives. P-glycoprotein (P-gp), 1The abbreviations used are: P-gp, P-glycoprotein; Ab, antibody; FACS, fluorescence-activated cell sorter; VBL, vinblastine; PBS, phosphate-buffered saline; BSA, bovine serum albumin.1The abbreviations used are: P-gp, P-glycoprotein; Ab, antibody; FACS, fluorescence-activated cell sorter; VBL, vinblastine; PBS, phosphate-buffered saline; BSA, bovine serum albumin. the product of theMDR1 gene, contributes to multidrug resistance in many cell types (1Schinkel A.H. Borst P. Semin. Cancer Biol. 1991; 2: 213-226PubMed Google Scholar, 2Gottesman M.M. Hrycyna C.A. Schoenlein P.V. Germann U.A. Pastan I. Ann. Rev. Genet. 1995; 29: 607-649Crossref PubMed Scopus (458) Google Scholar) and is expressed in many tumors (3He L. Hao C. Lin B. Wang Y. Gao F. Chin. Med. Sci. J. 1995; 10: 12-15PubMed Google Scholar, 4Decker D.A. Morris L.W. Levine A.J. Pettinga J.E. Grudzien J.L. Farkas D.H. Ann. Clin. Lab. Sci. 1995; 25: 52-59PubMed Google Scholar, 5Redmond S.M.S. Joncourt F. Buser K. Ziemiecki A. Altermatt H.J. Fey M. Margison G. Cerny T. Cancer Res. 1991; 51: 2092-2097PubMed Google Scholar, 6Marie J.-P. Hematol. Oncol. Clin. N. Am. 1995; 9: 239-249Abstract Full Text PDF PubMed Google Scholar, 7Holmes J.A. West R.R. Br. J. Cancer. 1994; 69: 382-384Crossref PubMed Scopus (59) Google Scholar, 8Baldini N. Scotlandi K. Barbanti Brodano G. Manara M.C. Maurici D. Bacci G. Bertoni F. Picci P. Sottili S. Campanacci M. N. Engl. J. Med. 1995; 333: 1380-1385Crossref PubMed Scopus (375) Google Scholar). P-gp pumps out its drug substrate from the tumor cell, reducing the effectiveness of administered chemotherapeutic agents (9Gottesman M.M. Pastan I. Ambudkar S.V. Curr. Opin. Genet. & Dev. 1996; 6: 610-617Crossref PubMed Scopus (505) Google Scholar). It has ATPase (10Ambudkar S.V. Lelong I.H. Zhang J. Cardarelli C.O. Gottesman M.M. Pastan I. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 8472-8476Crossref PubMed Scopus (378) Google Scholar, 11Sarkadi B. Price E.M. Boucher R.C. Germann U.A. Scarborough G.A. J. Biol. Chem. 1992; 267: 4854-4858Abstract Full Text PDF PubMed Google Scholar, 12Shapiro A.B. Ling V. J. Biol. Chem. 1994; 269: 3745-3754Abstract Full Text PDF PubMed Google Scholar, 13Urbatsch I.L. Al-Shawi M.K. Senior A.E. Biochemistry. 1994; 33: 7069-7076Crossref PubMed Scopus (224) Google Scholar, 14Ambudkar S.V. J. Bioenerg. Biomembr. 1995; 27: 23-29Crossref PubMed Scopus (46) Google Scholar) activity enhanced by numerous substrates and substrate analogs. A very wide range of substrates are pumped out of cells by P-gp (15Gottesman M.M. Pastan I. Annu. Rev. Biochem. 1993; 62: 385-427Crossref PubMed Scopus (3549) Google Scholar). Major efforts have been made toward finding clinically useful reversers of P-gp that can block its action, leading to renewed accumulation of drugs within erstwhile resistant cells (16Raderer M. Scheithauer W. Cancer ( Phila. ). 1993; 72: 3553-3563Crossref PubMed Scopus (265) Google Scholar, 17Lan L.-B. Ayesh S. Lyubimov E. Pashinsky I. Stein W.D. Cancer Chemother. Pharmacol. 1996; 38: 181-190Crossref PubMed Scopus (33) Google Scholar). Secure knowledge of the mechanism of action of P-gp is the basis for designing new and more effective reversers.Four lines of evidence indicate that P-gp can expel its substrates directly out of the cell membrane (reviewed in Ref. 18Stein W.D. Physiol. Rev. 1997; 77: 545-590Crossref PubMed Scopus (241) Google Scholar). First, the substrates of P-gp are lipophilic and reside, most of the time, within cell membranes. Thus, it is within the membrane that the pump will find it most easy to locate its substrate. Second, kinetic analyses show that drug accumulation is reduced by the action of P-gp from the earliest times that measurements can be made, i.e. before significant amounts of the drug can enter the cell yet when it is already present within the membrane (19Stein W.D. Cardarelli C. Pastan I. Gottesman M.M. Mol. Pharmacol. 1994; 45: 763-772PubMed Google Scholar). Hence it seems to be pumped out from the membrane itself before it crosses it. Third, Ravivet al. (20Raviv Y. Pollard H.B. Bruggemann E.P. Pastan I. Gottesman M.M. J. Biol. Chem. 1990; 265: 3975-3980Abstract Full Text PDF PubMed Google Scholar), making use of the ability of the photosensitive probe 5-iodonaphthalene-1-azide to label membrane proteins, showed that doxorubicin was expelled from cell membranes of P-gp-containing but not drug-sensitive cells. In addition, fluorescent dyes such as Hoechst 33342 (21Shapiro A.B. Ling V. J. Biol. Chem. 1995; 270: 16167-16175Abstract Full Text Full Text PDF PubMed Scopus (187) Google Scholar) and calcein-AM (22Hollo Z. Homolya L. Hegedus T. Sarkadi B. FEBS Lett. 1996; 383: 99-104Crossref PubMed Scopus (188) Google Scholar) have been used to demonstrate removal of substrate from the lipid bilayer by P-gp. These considerations favor a “vacuum cleaner” model (20Raviv Y. Pollard H.B. Bruggemann E.P. Pastan I. Gottesman M.M. J. Biol. Chem. 1990; 265: 3975-3980Abstract Full Text PDF PubMed Google Scholar) for P-gp in which this protein is an ATP-driven pump that pumps its substrates directly out of the plasma membrane. A major criticism of this model is, however, the lack of reliable information on the stoichiometry of ATP hydrolysis to drug pumping. Attempts to measure this stoichiometry in phospholipid vesicles containing pure P-gp suggested a minimum of 50 ATP molecules hydrolyzed per drug molecule transported, however, such a figure is difficult to reconcile with the metabolic potential of multidrug resistant cells (12Shapiro A.B. Ling V. J. Biol. Chem. 1994; 269: 3745-3754Abstract Full Text PDF PubMed Google Scholar). Recently, Eytan et al. (23Eytan G.D. Regev R. Assaraf Y.G. J. Biol. Chem. 1996; 271: 3172-3178Abstract Full Text Full Text PDF PubMed Scopus (95) Google Scholar) attempted to derive the stoichiometry by measuring the effect of P-gp on valinomycin-facilitated transport of 86Rb+ into proteoliposomes containing P-gp, valinomycin being a P-gp substrate. They reported 0.5–0.8 molecules of the complex valinomycin-Rb+ transported for every ATP molecule hydrolyzed. They assumed, however, that only Rb+-complexed valinomycin molecules were transported by P-gp, whereas we observed 2S. V. Ambudkar, unpublished data.2S. V. Ambudkar, unpublished data. that valinomycin's stimulation of ATP hydrolysis by P-gp is unaffected by its charged state. Transport of charged valinomycin molecules in the experiments by Eytan et al. (23Eytan G.D. Regev R. Assaraf Y.G. J. Biol. Chem. 1996; 271: 3172-3178Abstract Full Text Full Text PDF PubMed Scopus (95) Google Scholar) was carried out in the presence of a substantial excess of the uncharged form, which would not contribute to the measured flux but would contribute to ATP hydrolysis. In addition, their assay was carried out in solutions that were not free of potassium, which would compete with rubidium, complicating their calculations of the stoichiometry. We deemed it important, therefore, to determine this stoichiometry using an independent approach.Here we determine the number of P-gp molecules present on the surface of P-gp-expressing cells using an antibody (Ab) titration procedure and FACS analysis. In the same cell type, we measure the maximum capacity of P-gp to reduce the entry of its substrate vinblastine (VBL) by determining the difference between initial rates of VBL uptake in the presence and absence of the P-gp blocker verapamil. These two numbers, taken together, can determine the catalytic constant (molecules pumped per second) for the effect of P-gp on reducing VBL accumulation. Also, we measure the maximum capacity of such cells to accelerate efflux of loaded VBL in the presence of an energy source, giving another measure of the transport capacity and of the catalytic constant for the effect of P-gp on VBL transport. Comparing these catalytic constants with that obtained for the effect of VBL on accelerating ATP hydrolysis by P-gp, we compute the ratio of the maximum rate of VBL transport to the maximum rate of VBL-stimulated ATP hydrolysis. This ratio is not far from unity. Thus, only a small number of ATP molecules seem hydrolyzed for each VBL molecule transported by P-gp.RESULTSUsing FACS, we first studied binding of P-gp-specific monoclonal Ab UIC2 to cells of various cell lines engineered to express human P-gp. Fig. 1 depicts the results of experiments where we used control mouse IgG2a Ab (A) and UIC2 Ab (B) binding to NIH3T3 cells transfected with the human MDR1 gene (Gly-185 cells) at four concentrations of Ab, 0, 0.2, 1.6, and 4.8 μg/tube containing 200 μl of medium. Inpanel C, at these same concentrations with binding being to Val-185 cells, NIH3T3 cells transfected with a mutant strain ofMDR1 where valine replaces glycine in position 185 of the polypeptide chain of P-gp; in panel D, binding was done with a single amount (1 μg) of antibody to five different cell strains, using the non-transfected NIH3T3 cells, cells transfected with wild type MDR1 gene and selected at low dose (30 ng/ml vincristine; N3–30), with a higher dose (600 ng/ml; N3–600), and with a still higher dose (2400 ng/ml; N3–2400), as well as theMDR1 Gly-185 cell strain grown in the presence of 60 ng/ml colchicine (see panel B). In each case, the data are plotted as the number of cells on the ordinate that are labeled with the fluorescence intensity denoted on the abscissa. Cells tested with the control IgG Ab showed very little fluorescence with no increase as the concentration of Ab is raised, whereas the Gly-185 and Val-185 strains, tested with monoclonal Ab UIC2, demonstrate a clear and similar fluorescence signal that increases in intensity as the concentration of Ab increases. The various cell strains depicted in panel Dshow a marked difference in signal intensity, consistent with differences in the amount of P-gp present on the surface of these different cells (24Germann U.A. Chambers T.C. Ambudkar S.V. Licht T. Cardarelli C.O. Pastan I. Gottesman M.M. J. Biol. Chem. 1996; 271: 1708-1716Abstract Full Text Full Text PDF PubMed Scopus (157) Google Scholar).To determine the number of available P-gp molecules on the cell surface, we used the same monoclonal Ab UIC2 and a depletion assay. We first quantitate how the degree of reaction depends on the amount of Ab. Fig. 2 (panels A andC) depicts the median fluorescence of a sample of Gly-185 and Val-185 3T3 cells, respectively, when treated with increasing Ab concentration (0–8 μg of Ab/sample mixture of 200 μl) using a fixed number of cells, 5 × 105/reaction at 4 °C. We fitted the data by the simple binding equation.BOUND=BOUNDMAX·[Ab]Km+[Ab]Equation 1 where [Ab] is the concentration of the Ab present in the reaction sample, BOUND is the measured median fluorescence due to binding of Ab to the cells, BOUNDMAX is the maximum fluorescence determined by the maximum Ab bound to the amount of cells present, while Km is the concentration of Ab that gives one-half maximal binding (a measure of the affinity of Ab for ligand, here P-gp). Next, we titrated the Ab, present at an amount less than the values of Km found in the previous titrations, against an increasing number of added cells, using up to 8 million cells/reaction sample. As the number of cells per sample increases, less and less Ab is available for each cell. The median fluorescence per cell decreases as Ab is depleted from the reaction mixture by being bound to the cells themselves. We fitted the data by the depletion equation.BOUND=BOUNDinit·KnKn+[Cells]Equation 2 where [Cells] is the number of cells present per reaction sample, BOUND is the measured median fluorescence due to binding of Ab to cells, BOUNDinit is the initial fluorescence determined by the maximum Ab bound per cell at the weight of Ab present, whileKn is that number of cells that depletes by one-half the maximal amount of binding (a measure of the number of sites that bind the Ab, i.e. of P-gp molecules per cell).Figure 2Titration of cell surface P-gp against increasing amounts of UIC2 Ab (panels A and C) or against increasing numbers of cells added (panels B andD). In panels A and B, cells were NIH3T3 transfected with wild-type (Gly-185) MDR1, whereas in panels C and D, these were the mutant Val-185 transfectants. In each case the ordinate is the median fluorescence from an experiment, as depicted in Fig. 1. The fitted curves are binding isotherms of Equations 1 and 2 in the text forpanels A and C, and panels B andD, respectively.View Large Image Figure ViewerDownload Hi-res image Download (PPT)For the Gly-185 cells, we found Km to be 1.87 ± 0.24 μg of Ab, while Kn was 2.97 ± 0.77 million cells/reaction sample. We calculate the number of P-gp molecules exposed to Ab as follows. In Fig. 2 B, 1 μg of Ab is present per reaction sample, and Kn is 2.97. Therefore 2.97 million cells are sufficient to deplete one-half or 0.5 μg of Ab from the reaction mixture. Now 0.5 μg of Ab is 2.02 × 1012 molecules (assuming molecular mass of the Ab to be 150 kDa). Hence each cell binds 2.02/2.97 or 0.68 million molecules of Ab. But from Fig. 2 A we find the Km of Ab for P-gp as 1.87 μg. Thus at the 1 μg concentration used in Fig. 2 B, Ab-binding sites are only saturated to a fraction of 1/(1 + 1.87) or 0.345. Therefore, the true number of Ab-binding sites is 0.68/0.345 = 1.95 ± 0.53 million P-gp sites/Gly-185 3T3 cell on computing for the combination of errors. Performing the Ab binding procedures at 25 and 37 °C gave essentially the same values (data not shown).For the Val-185 3T3 cells, similarly, the number of P-gp sites per cell is 2.63 ± 0.87 million, not significantly different from the number present per Gly-185 3T3 cell. We performed the same experiments also for the N3–2400 strain of human MDR1-transfected 3T3 cells (see Ref. 24Germann U.A. Chambers T.C. Ambudkar S.V. Licht T. Cardarelli C.O. Pastan I. Gottesman M.M. J. Biol. Chem. 1996; 271: 1708-1716Abstract Full Text Full Text PDF PubMed Scopus (157) Google Scholar and Fig. 1 D). For these cells (data not shown), the value of Km was found to be 2.9 ± 0.4 μg of Ab while Kn was 0.56 ± 0.14 × 106 cells. Calculating, as described above, gives 14.2 ± 3.8 × 106 molecules of P-gp/cell for this highly resistant strain.In addition, with the MDR1 Gly-185-transfected 3T3 cells, we used another human P-gp-specific monoclonal Ab with an external epitope, MRK-16 (27Hamada H. Tsuruo T. Proc. Natl. Acad. Sci. U. S. A. 1986; 83: 7785-7789Crossref PubMed Scopus (576) Google Scholar), and also the Fab and F(ab′)2fragments prepared from this Ab. The depletion titration gave aKn value of 1.90 ± 0.37 × 106 cells for the MRK-16 Ab and 2.76 ± 0.62 × 106 for the F(ab′)2 fragment, not significantly different. Similar results were obtained with Fab (monovalent) fragment (data not shown). Considering these data it is reasonable to assume that 1 molecule of Ab binds to 1 P-gp molecule.P-gp can act on its substrates in two ways; it can pump drug from the cytoplasm of the cell and also pump it out from the membrane before it reaches the cytoplasm (19Stein W.D. Cardarelli C. Pastan I. Gottesman M.M. Mol. Pharmacol. 1994; 45: 763-772PubMed Google Scholar). We determined the maximum rate of VBL pumping by P-gp by these two pathways. First, we studied the extraction of VBL from the membrane before it accumulated inside the cell. Fig. 3 shows the time course of uptake of VBL into Gly-185 3T3 cells at 37 °C in PBS. (In Figs. 3 and4, the ordinate is the amount of VBL taken up and expressed as the volume of external medium cleared of VBL (in μl) per million cells at the time stated (19Stein W.D. Cardarelli C. Pastan I. Gottesman M.M. Mol. Pharmacol. 1994; 45: 763-772PubMed Google Scholar). Multiplying this measure by the concentration of VBL in the external medium would give the amount of VBL that enters the cell in the given time period). In Fig. 3 A, the time course was measured over an extended range. The final level of uptake is dramatically increased by adding 50 μm of the reverser verapamil to the cells (filled circles). In Fig. 3 B, however, uptake is measured for 12 s at 37 °C. Uptake is linear from zero time during this interval. In succeeding experiments, uptakes were performed at 10 s in the initial rate range. Fig. 4 A depicts the uptake of VBL during 10 s in the presence (filled circles) and absence (open squares) of 50 μm verapamil. In the presence of verapamil there is little change in the VBL uptake as the concentration of the drug is increased. In its absence, uptake increases with the concentration of VBL. Two curves, with and without verapamil, begin to approach one another, which is consistent with increasing concentrations of VBL saturating the pumping ability of P-gp so that it cannot cope with the inflow of the drug. Thus, the difference between uptake of VBL in presence and absence of verapamil gives the component of pumping that takes place as the drug crosses the membrane but before it enters the cytoplasm (18Stein W.D. Physiol. Rev. 1997; 77: 545-590Crossref PubMed Scopus (241) Google Scholar, 19Stein W.D. Cardarelli C. Pastan I. Gottesman M.M. Mol. Pharmacol. 1994; 45: 763-772PubMed Google Scholar). In Fig. 4 B we plot this difference, i.e. pumping of VBL by P-gp, as a function of [VBL] by combining data from different experiments. The filled circles are from Fig. 4 A,open circles are from a similar experiment at other concentrations of VBL, whereas the open square is from the 10-s data points of Fig. 3 B. The ordinate here, the volume of external medium cleared at the time chosen, is proportional to the velocity of uptake (v) divided by the concentration of VBL in the external medium (S). Using Michaelis-Menten kinetics to describe the pumping action of P-gp leads us tov=Vmax·SKp+S,givingvS=VmaxKp+SEquation 3 where Vmax is the maximum velocity of pumping (in moles/cell/second), whereas Kp is that [VBL] giving one-half maximum pumping velocity. We fitted the data of Fig. 4 B to Equation 3 where v/S is the uptake of VBL (volume of external medium cleared per 10 s) and obtained an estimate of Kp andVmax. We find that Kp is 37.0 ± 9.9 μm, while Vmax is 3.5 ± 0.83 pmol/106 cells/sec or 2.1 ± 0.5 × 106 molecules/cell/s. Since the number of P-gp molecules present on the surface of each cell is 1.95 × 106, this gives a turnover number of 1.08 ± 0.39/s.Figure 3Uptake of vinblastine from an external concentration of 0.1 μm into Gly-185-transfected NIH3T3 cells as a function of time at 37 °C in the presence (circles) or absence (squares) of 50 μm verapamil. A, long term uptakes.B, initial rates. Vinblastine uptake assays were carried out as described under “Experimental Procedures.”View Large Image Figure ViewerDownload Hi-res image Download (PPT)Figure 4The initial rate of vinblastine uptake. A, vinblastine uptake during 10 s at 37 °C into Gly-185-transfected NIH3T3 cells in the presence (circles) or absence (squares) of 50 μm verapamil as a function of [VBL]. B, verapamil-sensitive uptake during 10 s at 37 °C from panel A (filled circles) and from a similar experiment (open circles) and from panel B of Fig. 3 (open square) as a function of [VBL]. The fitted line is for Equation 3 in the text.View Large Image Figure ViewerDownload Hi-res image Download (PPT)We next studied zero trans-efflux (28Stein W.D. Transport and Diffusion Across Cell Membranes. Academic Press, Orlando, FL1986Google Scholar) from cells loaded with VBL. In Fig. 5 A, Gly-185 3T3 cells were loaded with 1.4 nm VBL at 25 °C for 40 min in the presence of 10 mm sodium azide and 10 mm 2-deoxyglucose to deplete the energy (19Stein W.D. Cardarelli C. Pastan I. Gottesman M.M. Mol. Pharmacol. 1994; 45: 763-772PubMed Google Scholar). (These experiments were performed at 25 rather than 37 °C since we found that during loading of the cells at 37 °C, in the absence of an energy source, they became less adherent to the dishes, preventing an accurate determination of efflux rates.) Cells were washed rapidly with ice-cold PBS and then exposed to a wash-out medium at 25 °C, free of VBL but containing either PBS + 10 mm sodium azide + 10 mm 2-deoxyglucose (filled circles) or PBS + 5 mm glucose to restore energy (filled squares). The data were fitted by an exponential efflux equation, giving values for t1/2 of 37 and 4.4 min into azide/2-deoxyglucose or glucose medium, respectively. Efflux into glucose medium is thus some 8× that into azide/2-deoxyglucose. A parallel experiment performed with [VBL] at 100 μm is depicted in Fig. 5 B. Rates into azide/2-deoxyglucose and glucose are now similar to t1/2 values of 13 and 8.4 min, respectively. In addition, results of similar experiments with different concentrations of VBL are depicted in Fig. 5 C. Here we plot efflux during 2 min, reported as the difference between efflux into glucose and into azide/2-deoxyglucose and expressed as a fraction of the zero time value, as a function of [VBL]. Data from Fig. 5, A and B are shown as open squares with circles being from another experiment andtriangles from a third. As for Fig. 4 B, we fitted these data by Equation 3, obtaining a value for the maximum velocity of pumping (fractional loss) as 0.179 ± 0.011/min. The volume of external medium cleared by 106 cells at zero time of efflux was 21.5 ± 1.5 μl (n = 5). TheKp from Fig. 5 C was 15.1 ± 3.0 μm. Thus, at a concentration of VBL in the external medium equal to Kp, there would be 21.5 × 15.1 or 325 pmol of vinblastine present within the cell. This is pumped out at one-half the maximum rate but represents one-half of the pumping capacity of the cell's P-gp. Thus the maximum rate of VBL efflux is 0.179 × 325 pmol/106 cells/min or 0.59 ± 0.12 × 106 molecules/cell/s at 25 °C.Figure 5Time course of efflux of vinblastine at 25 °C from Gly-185-transfected NIH3T3 cells loaded with VBL during 40 min at 25 °C in PBS containing 10 mm sodium azide and 10 mm 2-deoxyglucose at VBL concentrations of 1.4 nm (A) or 100 μm (B). Efflux of 3H-VBL was followed into this same medium (filled circles) or into PBS containing 5 mmglucose (filled squares). The solid lines are the best-fitted exponential decay curves. Panel C plots the energy-dependent efflux, calculated as the difference between the VBL efflux during 2 min at 25 °C into glucose and azide/deoxyglucose at various VBL concentrations as indicated on the abscissa. The filled circles and triangles are from different experiments as in panels A and B, whereas the open squares are the data from panels A and B themselves. The fitted line is for Equation 3in the text.View Large Image Figure ViewerDownload Hi-res image Download (PPT)To compare these results with the turnover number for the ATPase, we measured the ATPase activity both at 25 and 37 °C. We determined this activity for the Val-185 3T3 cell line and the N3–2400 cell line that is enriched for P-gp (24Germann U.A. Chambers T.C. Ambudkar S.V. Licht T. Cardarelli C.O. Pastan I. Gottesman M.M. J. Biol. Chem. 1996; 271: 1708-1716Abstract Full Text Full Text PDF PubMed Scopus (157) Google Scholar). We assessed the effect of the drugs such as vinblastine, verapamil, and colchicine on stimulation of the ATPase activity (Table I). The VBL-stimulated ATPase activity is reduced 4.63-fold as the temperature is decreased from 37 to 25 °C. Were this ratio to be appropriate for the pumping function of P-gp on VBL, one would conclude that a maximum efflux of 0.59 at 25 °C might be equivalent to a value of 2.73 ± 0.55 × 106 molecules/cell/s at 37 °C or to a turnover number of 1.40 ± 0.43 s−1, which is indistinguishable from the value determined for the inward component of pumping (Fig. 4 B).Table IDrug-stimulated ATPase activity in NIH3T3 transfectants expressing varying levels of human P-glycoproteinDrugGly-1851-aEach column is for the strain ofMDR1-transfected 3T3 cells as indicated at either 37 or 25 °C. The data are reported as nmol of Pi released per min/mg of protein. (37 °C)Gly-1851-aEach column is for the strain ofMDR1-transfected 3T3 cells as indicated at either 37 or 25 °C. The data are reported as nmol of Pi released per min/mg of protein.(25 °C)Val-1851-aEach column is for the strain ofMDR1-transfected 3T3 cells as indicated at either 37 or 25 °C. The data are reported as nmol of Pi released per min/mg of protein. (37 °C)Val-1851-aEach column is for the strain ofMDR1-transfected 3T3 cells as indicated at either 37 or 25 °C. The data are reported as nmol of Pi released per min/mg of protein.(25 °C)N3–24001-aEach column is for the strain ofMDR1-transfected 3T3 cells as indicated at either 37 or 25 °C. The data are reported as nmol of Pi released per min/mg of protein. (37 °C)N3–24001-aEach column is for the strain ofMDR1-transfected 3T3 cells as indicated at either 37 or 25 °C. The data are reported as nmol of Pi released per min/mg of protein.(25 °C)ControlMe2SO285.0477315.8Verapamil8019962012422Δ-Verapamil521449139316Vinblastine417.8275.0789.8Δ-Vinblastine132.8−211-bAt this concentration, vinblastine inhibits activity in this mutant.−2.01-bAt this concentration, vinblastine inhibits activity in this mutant.464.3Colchicine457.56811238.5Δ-Colchicine172.5213.5202.7ATP hydrolysis by crude membranes was measured as described under “Experimental Procedures” at 37 and 25 °C in the presence or absence of verapamil (30 μm), vinblastine (20 μm), and colchicine (200 μm), respectively. Only the vanadate-sensitive activity is given, and the values are the average