JBC

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M408849200 on September 9, 2004

J. Biol. Chem., Vol. 279, Issue 47, 48855-48864, November 19, 2004
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
279/47/48855    most recent
M408849200v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Sauna, Z. E.
Right arrow Articles by Ambudkar, S. V.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sauna, Z. E.
Right arrow Articles by Ambudkar, S. V.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Multidrug Resistance Protein 4 (ABCC4)-mediated ATP Hydrolysis

EFFECT OF TRANSPORT SUBSTRATES AND CHARACTERIZATION OF THE POST-HYDROLYSIS TRANSITION STATE*

Zuben E. Sauna, Krishnamachary Nandigama, and Suresh V. Ambudkar{ddagger}

From the Laboratory of Cell Biology, Center for Cancer Research, NCI, National Institutes of Health, Department of Health and Human Services, Bethesda, Maryland 20892-4256

Received for publication, August 3, 2004 , and in revised form, August 30, 2004.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Multidrug resistance protein 4 (MRP4/ABCC4), transports cyclic nucleoside monophosphates, nucleoside analog drugs, chemotherapeutic agents, and prostaglandins. In this study we characterize ATP hydrolysis by human MRP4 expressed in insect cells. MRP4 hydrolyzes ATP (Km, 0.62 mM), which is inhibited by orthovanadate and beryllium fluoride. However, unlike ATPase activity of P-glycoprotein, which is equally sensitive to both inhibitors, MRP4-ATPase is more sensitive to beryllium fluoride than to orthovanadate. 8-Azido[{alpha}-32P]ATP binds to MRP4 (concentration for half-maximal binding ~3 µM) and is displaced by ATP or by its non-hydrolyzable analog AMPPNP (concentrations for half-maximal inhibition of 13.3 and 308 µM). MRP4 substrates, the prostaglandins E1 and E2, stimulate ATP hydrolysis 2- to 3-fold but do not affect the Km for ATP. Several other substrates, azidothymidine, 9-(2-phosphonylmethoxyethyl)adenine, and methotrexate do not stimulate ATP hydrolysis but inhibit prostaglandin E2-stimulated ATP hydrolysis. Although both post-hydrolysis transition states MRP4·8-azido[{alpha}-32P]ADP·Vi and MRP4·8-azido[{alpha}-32P]ADP·beryllium fluoride can be generated, nucleotide trapping is ~4-fold higher with beryllium fluoride. The divalent cations Mg2+ and Mn2+ support comparable levels of nucleotide binding, hydrolysis, and trapping. However, Co2+ increases 8-azido[{alpha}-32P]ATP binding and beryllium fluoride-induced 8-azido[{alpha}-32P]ADP trapping but does not support steady-state ATP hydrolysis. ADP inhibits basal and prostaglandin E2-stimulated ATP hydrolysis (concentrations for half-maximal inhibition 0.19 and 0.25 mM, respectively) and beryllium fluoride-induced 8-azido[{alpha}-32P]ADP trapping, whereas Pi has no effect up to 20 mM. In aggregate, our results demonstrate that MRP4 exhibits substrate-stimulated ATP hydrolysis, and we propose a kinetic scheme suggesting that ADP release from the post-hydrolysis transition state may be the rate-limiting step during the catalytic cycle.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
The multidrug resistance-associated proteins (MRPs)1 are ATP-binding cassette (ABC) transporters and belong to the ABCC subfamily (1, 2). The MRP family of ABC transporters consists of at least nine members (3). MRP1, the first member of this family to be characterized (4), transports a range of anticancer drugs out of cells and has been implicated in multidrug resistance (MDR). The preferred substrates for MRP1 are organic anions, e.g. drugs conjugated to glutathione, glucuronate, or sulfate (5–7). Similarly, MRP2 and to a lesser extent MRP3 have also been implicated in MDR (substrates and resistance profiles have been reviewed in Refs. 3 and 5). The MRPs fall into two groups on the basis of the degree of amino acid identity and predicted topology of the full-length proteins (3, 6). MRP1, MRP2, MRP3, MRP6, and MRP7 all have an extra N-terminal domain (often indicated as TMD0), whereas MRP4, MRP5, MRP8, and MRP9 do not possess the TMD0 (3, 8). Moreover, the former group has a higher amino acid identity with MRP1 (45–58% overall; 61–74% nucleotide binding folds). Nonetheless all MRPs are more homologous to each other than to P-glycoprotein (Pgp, ABCB1, and MDR1) or other subclasses of ABC transport proteins (2, 3, 6).

MRP4 (ABCC4) has garnered a great deal of attention in recent years, because it was demonstrated that this transporter can function as a cellular efflux pump for the anti-human immunodeficiency virus drugs 9-(2-phosphonylmethoxyethyl)adenine (PMEA) and azidothymidine monophosphate in PMEA-resistant cells (9). High levels of MRP4 severely impair the antiviral efficacy of several nucleoside analogs (10). Moreover, MRP4 has also been implicated in the MgATP-energized transport of cAMP and cGMP (11, 12), a feature that suggests their involvement in the regulation of intracellular cyclic nucleotide levels (10). MRP4 does not appear to confer resistance to natural product anticancer agents, although it does appear to confer resistance to the widely used antimetabolite methotrexate (13). Most recently, Reid and coworkers (14) have demonstrated that prostaglandins E1 and E2 (PGE1 and PGE2) are transported by MRP4. This suggests a unique and important physiological role for MRP4. Prostaglandins are involved in inflammatory responses (15), and recent studies have suggested that prostaglandins are overexpressed in several tumors (16, 17). The regulation of prostaglandins is central to the management of many disease conditions, and the inhibition of the enzymes cyclooxygenase-1 and -2 has been the principal pharmaceutical approach (18, 19). The prostaglandins exert their influence outside the cells where they are produced, and the existence of specific energy-dependent pumps involved in their efflux offers unique mechanistic insights and potential drug targets.

The reports described above suggest that the putative substrates of MRP4 have interesting physiological effects. There are, however, discrepancies in results from different laboratories vis à vis the primary role of MRP4 and its physiological substrate (see for example Ref. 11). Resolving these and other issues necessitates an understanding of the transport mechanism(s). The catalytic cycle and the ATP sites of Pgp and MRP1 have been studied in considerable detail (3, 20–24). A striking difference between Pgp- and MRP1-mediated ATP hydrolysis is that while the two ATP sites of Pgp are essentially symmetric (25), those of MRP1 do not appear to be symmetrical (24, 26, 27). These findings are consistent with the fact that the two ATP sites of MRP1 are more divergent than those of most ABC transport proteins (4). We have previously reported (28) that MRP4-mediated ATP hydrolysis, similar to Pgp and MRP1, is sensitive to disulfiram, a drug clinically used to treat alcoholism (29), which is also an MDR modulator (28, 30). However, the kinetics of MRP4 ATPase activity, modulation of the activity by transport substrates and the transition-state intermediates has not yet been characterized.

We demonstrate an MRP4-mediated, divalent cation-dependent, orthovanadate (Vi)- and beryllium fluoride (BeFx)-sensitive ATP hydrolysis that shows Henri-Michaelis-Menten kinetics. The ATP hydrolysis is affected by putative transport substrates of MRP4, and of all the potential substrates tested, the prostaglandins showed the largest stimulation of MRP4-mediated ATP hydrolysis (3- to 4-fold). Moreover, we found a close correlation between these results and previous transport studies to identify substrates of MRP4. The Km values reported for transport of PGE1 and PGE2 are 2.1 and 3.4 µM, whereas the concentrations required for half-maximal stimulation of ATP hydrolysis reported in the present study were 12.15 and 2.9 µM. Using either Vi or BeFx, it was possible to obtain the post-hydrolysis transition states, MRP4·Mg2+-8-azido[{alpha}-32P]-ADP·Vi/BeFx. Based on our data, we present a kinetic scheme for MRP4-mediated ATP hydrolysis.


   EXPERIMENTAL PROCEDURES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
RESULTS
 DISCUSSION
 REFERENCES
 
Chemicals—[{alpha}-32P]8-azidoATP (5–10 Ci/mmol) and 8-azidoATP were purchased from Affinity Labeling Technologies, Inc. (Lexington, KY). The polyclonal anti-MRP4 antibody was obtained from Kamiya Medical Co., (Seattle, WA). AZT and PMEA were from Moravek Biochemicals (Brea, CA). All other chemicals were obtained from Sigma.

Preparation of Crude Membranes from High Five Insect Cells Infected with Recombinant Baculovirus Carrying the Human MDR1, MRP1, or MRP4 cDNA—High Five insect cells (Invitrogen) were infected with the recombinant baculovirus carrying one of the following: the human MDR1 cDNA with a 6-histidine tag at the C-terminal end (BV-MDR1(H6)) (31), the human MRP1 cDNA with a 10-histidine tag at the C-terminal end (BV-MRP1(H10)), or human MRP4 cDNA (BV-MRP4) (the plasmid pVL1393-MRP4 was provided by Dr. Gary Kruh, Fox Chase Cancer Center) (32). The MRP4 cDNA was cloned into the pDest-8 baculovirus vector, and the cloned gene was sequenced in both directions. The sequence of MRP4 in pDest vector was found to be identical to that published (33). Crude membranes were prepared as described previously (34) and used in all experiments. Total protein was estimated using the Amido Black protein method of Schaffner and Weissmann (35) with bovine serum albumin as a standard. Crude membranes (10 µg per lane) were electrophoresed on a 7% NuPAGE precast gel (Invitrogen) and stained with Colloidal blue (Invitrogen) following the manufacturer's instructions. Following Western blotting to nitrocellulose paper, crude membranes (5 µg of protein/lane) were probed with the polyclonal anti-MRP4 antibody (Kamiya) at a dilution of 1:1000 (9).

Effect of Divalent Cations on 8-Azido[{alpha}-32P]ATP Binding, BeFx-induced 8-Azido[{alpha}-32P]ADP Trapping, and ATP Hydrolysis by MRP4—To remove all traces of bound divalent cations the crude membranes were diluted 10- to 20-fold with the ATPase assay buffer (see below) without MgCl2 but containing 5 mM EDTA. Following centrifugation at 300,000 x g at 4 °C for 10 min in an RC-M120EX micro-ultracentrifuge (Sorvall, Newtown, CT), the pellet was resuspended in the MgCl2-free ATPase assay buffer containing 10% glycerol and stored at –70 °C. These membranes were used to study [{alpha}-32P]8-azidoATP binding, BeFx-induced [{alpha}-32P]8-azidoADP trapping, and ATP hydrolysis as described below.

ATPase Assays—ATPase activity of MRP4 in crude membranes (10 µg of protein per assay) was measured by the end-point Pi assay as previously described (31, 36), with minor modifications. MRP4-specific activity was recorded as the Vi- or BeFx-sensitive ATPase activity. The assay measured the amount of Pi released over 20 min at 37 °C in the ATPase assay buffer (50 mM MES-Tris, pH 6.8, 50 mM KCl, 5 mM sodium azide, 2 mM EGTA, 2 mM dithiothreitol, and 10 mM MgCl2). The assay was carried out under basal conditions or in the presence of putative substrates or modulators. The reaction was initiated with 5 mM ATP and quenched with SDS (2.5% final concentration); the amount of Pi released was quantified using a colorimetric method (31, 37).

Binding of 8-Azido[{alpha}-32P]ATP to MRP4 —Crude membranes (50 µg of protein per assay) were incubated in the ATPase assay buffer containing 10 µM 8-azido[{alpha}-32P]ATP (8–10 µCi/nmol) in the dark on ice for 5 min. The samples were then illuminated with a UV lamp (365 nm) assembly (PGC Scientifics, Gaithersburg, MD) for 10 min on ice. Ice-cold ATP (12.5 mM) was added to displace excess non-covalently bound radionucleotides. Excess unbound nucleotides were removed by centrifugation at 300,000 x g at 4 °C for 10 min by using an S120-AT2 rotor in an RC-M120EX micro-ultracentrifuge (Sorvall), and the pellet was resuspended in SDS-PAGE sample buffer. Following electrophoresis on a 7% NuPAGE gel (Invitrogen) at constant voltage, gels were dried and exposures were made to Bio-Max MR film (Eastman Kodak, Rochester, NY) at –70 °C for 12–24 h. The radioactivity incorporated into the MRP4 band was quantified using a STORM 860 PhosphorImager system (Amersham Biosciences) and the software ImageQuaNT.

Vanadate or BeFx-induced 8-Azido[{alpha}-32P]ADP Trapping in MRP4 —The MRP4·8-azidoADP·Vi or MRP4·8-azidoADP·BeFx transition states were generated as described previously for Pgp (38, 39). Crude membranes (50 µg of protein per assay) were incubated in the ATPase assay buffer containing 50 µM 8-azido[{alpha}-32P]ATP (4–6 µCi/nmol) and either 250 µM Vi or 0.2 mM BeSO4 plus 2.5 mM NaF in the dark for 5 min at 37 °C. The reaction was stopped by adding 12.5 mM ice-cold ATP and placing the sample immediately on ice. The samples were then cross-linked by UV illumination, electrophoresed, and analyzed as described above.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
DISCUSSION
 REFERENCES
 
Expression and Identification of Human MRP4 (ABCC4) in High Five Insect Cells—Human MRP4 cDNA was cloned into the pDest-8 baculovirus vector (Invitrogen), and the recombinant virus was used to infect High Five insect cells at an multiplicity of infection of 5. The insect cells were harvested ~48–54 h post-infection with >70% cell viability, and crude membranes were prepared by hypotonic lysis as described previously (34). Crude membranes prepared from insect cells overexpressing MDR1, MRP1, or MRP4 all showed the presence of overexpressed protein on a Colloidal Blue-stained gel (Fig. 1A). The putative MRP4 band migrated at an apparent molecular mass of ~150 kDa, similar to the predicted molecular mass. Moreover, immunoblot analysis of crude membranes from MRP4-infected insect cells, with the anti-MRP4 polyclonal antibody, demonstrated that MRP4 was efficiently expressed in the baculovirus heterologous expression system (Fig. 1B). In addition, the anti-MRP4 antibody did not detect any protein in crude membranes prepared with insect cells infected with control baculovirus or those infected with human the MDR1or MRP1 cDNA (Fig. 1B).



View larger version (62K):
[in this window]
[in a new window]
  		FIG. 1.
Expression of MRP4 in High Five insect cells. Crude membranes prepared from High Five insect cells infected with control baculovirus and baculovirus containing MDR1(H6), MRP1(H10), or MRP4 were electrophoresed and stained with Colloidal Blue (A) or probed with the polyclonal anti-MRP4 antibody (B) as described under "Experimental Procedures." The lanes are labeled in the figure, and the arrows depict the position of Pgp, MRP1, and MRP4 bands. MWM, molecular weight markers (x103).

 
8-Azido[{alpha}-32P]ATP Binds to Human MRP4 —Many laboratories have implicated MRP4 in the MgATP-dependent transport of several physiological substrates such as cyclic nucleotides, steroid conjugates, folates, and the PGE1 and PGE2 (6, 10, 14). However, ATP hydrolysis activity attributable to MRP4 and the characteristics of nucleotide binding to the ATP sites have not been studied. Specific binding of nucleotides at the ATP site(s) is a prerequisite for catalysis, and the photoaffinity analog of ATP, 8-azido[{alpha}-32P]ATP, has been extensively used to probe the nucleotide binding sites of ABC transporters (24, 26, 36, 40, 41). We demonstrate that 8-azido[{alpha}-32P]ATP binds specifically to the ATP sites of MRP4 (Fig. 2A). The photocross-linked 8-azido[{alpha}-32P]ATP colocalizes with MRP4 as identified by immunoblot analysis (data not shown). Moreover, the binding of 8-azido[{alpha}-32P]ATP to MRP4 is completely inhibited by ATP. Control insect cells infected with baculovirus alone did not show the 8-azido[{alpha}-32P]ATP-labeled band. We also show that the binding of [{alpha}-32P]8azidoATP to MRP4 is saturable and that the concentration required for half-maximal binding is 2.9 µM (Fig. 2B). The non-hydrolyzable analog of ATP, AMPPNP, is a useful tool that can be exploited to distinguish the effects of nucleotide binding versus nucleotide hydrolysis (38, 42, 43). Both ATP and AMPPNP inhibit the binding of 8-azido[{alpha}-32P]ATP to MRP4 (Fig. 2C), although the concentrations required for 50% inhibition differ by an order of magnitude (13.3 µM for ATP and 308.2 µM for AMPPNP). The fact that AMPPNP has a lower affinity for MRP4 than ATP is consistent with previous findings with Pgp (36, 44).



View larger version (29K):
[in this window]
[in a new window]
  		FIG. 2.
Binding of 8-azido[{alpha}-32P]ATP to MRP4. A, binding of 8-azido[{alpha}-32P]ATP to MRP4 containing crude membranes and control membranes was monitored in the presence or absence of 10 mM ATP as described under "Experimental Procedures." The experimental conditions are depicted on the autoradiogram. B, crude membranes containing MRP4 were incubated with increasing concentrations of 8-azido[{alpha}-32P]ATP, and processed as described under "Experimental Procedures." The radioactivity in the MRP4 band was quantified using the software ImageQuaNT. The line, which represents a non-linear least squares fit to a hyperbolic function, was obtained using GraphPad Prizm (concentration required for half-maximal binding of 8-azidoATP = 2.9 ± 0.16 µM). The autoradiogram and its graphical representation are depicted. C, binding of 8-azido[{alpha}-32P]ATP to MRP4 was monitored in the presence of increasing concentrations of ATP or AMPPNP. The samples were then processed and quantified as described under "Experimental Procedures." The lines represent non-linear least squares fits to a hyperbolic function and concentrations required for half-maximal inhibition by ATP and AMPPNP are 13.3 ± 4.3 µM and 308 ± 15.7 µM, respectively (n = 3).

 
MRP4-mediated ATP Hydrolysis—A Vi- or BeFx-sensitive ATPase activity in the presence of inhibitors of Ca2+-ATPase, Na+, K+-ATPase, and mitochondrial ATPase has been demonstrated in plasma membranes from several cell lines overexpressing ABC transport proteins (37, 44–48). We have previously reported that crude membrane preparations of insect cells overexpressing MRP4 exhibit Vi-sensitive ATP hydrolysis (28). Because crude membranes were used in the ATPase assay, we measured MRP4-mediated ATP hydrolysis in the presence of ouabain, EGTA, and sodium azide to eliminate Na+, K+-ATPase, Ca2+-ATPase, and mitochondrial ATPase activities. We found a significant ATPase activity that was sensitive to both Vi and BeFx (Fig. 3, A and B). Although BeFx shows >95% inhibition, Vi inhibits the ATP hydrolysis by 40%. BeSO4 or NaF alone did not inhibit ATP hydrolysis (inset to Fig. 3B). Thus, in the presence of excess NaF, concentration-dependent inhibition of ATP hydrolysis by BeSO4 allowed an estimation of the concentration required for half-maximal inhibition for BeFx (12.7 ± 2.41 µM). The concentration required for half-maximal inhibition by Vi on the other hand was 92.7 ± 15.1 µM. Although membranes prepared from control insect cells infected with baculovirus alone exhibited some BeFx- and Vi-sensitive ATPase activity (~20 and 4 nmol/mg of protein/min, respectively), this activity was not modulated by MRP4 transport substrates (see Fig. 6, A and B). MRP4-mediated ATP hydrolysis follows simple Henri-Michaelis-Menten kinetics with a Km of 0.62 mM (Fig. 3C).



View larger version (8K):
[in this window]
[in a new window]
  		FIG. 3.
MRP4-mediated ATP hydrolysis. Steady-state ATP hydrolysis was measured in MRP4 containing crude membranes as described under "Experimental Procedures" in the presence of increasing concentrations of Vi (A) or BeFx (B). In B, the experiment was carried out with increasing concentrations of BeSO4 in the presence of a fixed (2.5 mM) concentration of NaF. The inset shows ATP hydrolysis in the presence of 2.5 mM NaF or 200 µM BeSO4 alone. The y-axis title for the inset is the same as that for the main figure. Concentrations required for half-maximal inhibition by Vi and BeFx are 92.7 ± 15.1 µM and 12.7 ± 2.41 µM. C, MRP4-mediated BeFx-sensitive, steady-state ATP hydrolysis was measured in the presence of increasing concentrations of ATP, the Km value was obtained by fitting the data to the Henri-Michaelis-Menten equation. Km(ATP) = 0.62 ± 0.07 mM. Values represent the average (±S.D.) of three independent experiments.

 



View larger version (9K):
[in this window]
[in a new window]
  		FIG. 6.
Effect of putative transport substrates on MRP4-mediated ATP hydrolysis. BeFx-sensitive, steady-state ATP hydrolysis was measured in control and MRP4 containing crude membranes in the presence of increasing concentrations of PGE1 (A), PGE2 (B), and GMP and cGMP (C). The values for half-maximal stimulation were obtained by fitting the data using non-linear regression analysis. PGE1- and PGE2-stimulated MRP4-mediated ATP hydrolysis, and the concentrations required for half-maximal stimulation were 12.15 ± 2.5 and 2.9 ± 0.12 µM (n = 3), respectively. The data plotted on the graphs are the average of three independent experiments, and the S.D. is depicted as error bars in A and B, and in these panels open circles represent activity in control (without MRP4) membranes.

 
ADP Inhibits MRP4-mediated ATP Hydrolysis, whereas Inorganic Pi Does Not—The products of ATP hydrolysis are ADP and Pi. It has been demonstrated that the affinities of ATP and ADP for Pgp are comparable (44, 49) and that the release of ADP from the transition state complex is the rate-limiting step (34). ADP inhibits both basal and substrate (PGE2)-stimulated ATP hydrolysis in a concentration-dependent manner. The inhibition by ADP is not influenced by the presence of substrate in the assay (concentration required for half-maximal inhibition (ADP) = 0.19 and 0.25 mM in the absence or presence of PGE2, respectively) (Fig. 4). However, we find that even very high concentrations of Pi (up to 10 mM) had no effect on ATP hydrolysis (data not shown). These data indicate that MRP4 has an extremely low affinity for Pi, similar to findings with Pgp (34, 40).



View larger version (13K):
[in this window]
[in a new window]
  		FIG. 4.
Effect of ADP on MRP4-mediated ATP hydrolysis. BeFx-sensitive, steady-state ATP hydrolysis was measured in MRP4 containing crude membranes as described under "Experimental Procedures" in the presence of increasing concentrations of ADP in the absence or presence of 50 µM PGE2. The concentrations required for half-maximal inhibition by ADP were 0.19 ± 0.04 mM and 0.25 ± 0.03 mM (n = 3) in the absence or presence of PGE2, respectively.

 
Binding of 8-Azido-[{alpha}-32P]ATP to MRP4 and ATP Hydrolysis by MRP4 Are Divalent Cation-dependent—Biochemical evidence suggests that the presence of a divalent cation such as Mg2+ is mandatory during hydrolysis by ABC proteins (44, 50, 51), and the crystal structures of the ATP sites of several ABC proteins show that Mg2+ coordinates the nucleoside triphosphate such that the {gamma}-phosphate can be aligned for a nucleophilic attack (49, 52, 53). Consistent with these findings, we found that EDTA, which chelates Mg2+, inhibits the binding of 8-azido[{alpha}-32P]ATP to MRP4 in a concentration-dependent manner, and the concentration required for half-maximal inhibition is 0.7 mM (Fig. 5A). To study the effect of different divalent cations on nucleotide binding and hydrolysis, we washed the crude membranes with buffer free of divalent cations as described under "Experimental Procedures." We demonstrate that the divalent cations Mg2+, Mn2+, Co2+, and Ca2+ all support the binding of 8-azido[{alpha}-32P]ATP to MRP4 (Fig. 5B) with maximal binding in the presence of Co2+ (Co2+ > Mn2+ > Mg2+ = Ca2+) as the divalent cation. Moreover, excess ATP completely abolishes binding of 8-azido[{alpha}-32P]ATP in the presence of all the divalent cations (data not given) suggesting that the signal is specific. However, although MRP4 shows both basal and PGE2-stimulated ATP hydrolysis in the presence of both Mg2+ and Mn2+, neither Co2+ nor Ca2+ support ATP hydrolysis (Fig. 5C). Furthermore, in EDTA-washed membranes, Mg2+ stimulates ATP hydrolysis in a concentration-dependent manner and the concentration for half-maximal stimulation is 0.59 mM (Fig. 5D). This value is similar to the Km(ATP) during hydrolysis (0.62 mM), suggesting that the Mg2+ ion is titrating for the ATP. Moreover, free Mg2+ even at a concentration of 10 mM did not inhibit MRP4-mediated ATP hydrolysis. This is comparable to findings for another ABC transport protein, Pgp (44), but significantly different from observations with the F1F0-ATPase (54) and the sarcoplasmic reticulum Ca2+-ATPase (55), which are sensitive to higher concentrations of Mg2+.



View larger version (32K):
[in this window]
[in a new window]
  		FIG. 5.
Effect of divalent cations on nucleotide binding to MRP4 and ATP hydrolysis by MRP4. A, binding of 8-azido[{alpha}-32P]ATP to MRP4 was monitored and quantified as described above in the presence of increasing concentrations of EDTA. Concentration required for half-maximal inhibition (EDTA) = 0.7 ± 0.09 mM; n = 3. Experiments depicted in B–D were carried out using crude membranes washed to remove all traces of bound divalent cations as described under "Experimental Procedures." B, binding of 8-azido[{alpha}-32P]ATP to MRP4 was monitored in the presence of MgCl2, MnCl2, CoCl2, or CaCl2 (10 mM). The different treatments are depicted on the autoradiogram, and the lower panel is a quantification of the amount of 32P incorporated in the MRP4 band. C, BeFx-sensitive, MRP4-mediated steady-state ATP hydrolysis was measured in the presence of the divalent cations MgCl2, MnCl2, CoCl2, or CaCl2 (10 mM). The assays were carried out in the absence or in the presence of 50 µM PGE2. D, ATP hydrolysis was measured in the presence of increasing concentrations of MgCl2 in the presence of 5 mM ATP. Concentration of Mg2+ required for half-maximal stimulation = 0.59 ± 0.07 mM; n = 3).

 
Putative Substrates of MRP4 Affect ATP Hydrolysis—The catalytic cycles of Pgp and MRP1 have been studied in detail, and there appears to be a conformational coupling between the drug- and nucleotide-binding sites (for reviews see Refs. 3 and 21). Recent reports suggest several substrates for MRP4, which may be broadly classified as (i) cyclic nucleoside monophosphates, and nucleoside analog drugs (10); (ii) anticancer agents (13); (iii) steroids (12); and (iv) prostaglandins (14). We have compared the effect of members of each of these classes on ATP hydrolysis in membranes prepared from insect cells infected with baculovirus containing MRP4 or baculovirus alone. We observed that PGE1 and PGE2 stimulate ATP hydrolysis (Fig. 6, A and B), whereas cyclic GMP has a biphasic effect on MRP4-mediated ATP hydrolysis, although GMP does not (Fig. 6C). Moreover, glutathione has no effect on MRP4-mediated ATP hydrolysis (data not shown). It must be noted that the maximum stimulation of MRP4-mediated ATPase activity by putative substrates is only about 2- to 3-fold. This is considerably less than the 3- to 10-fold stimulation of ATPase activity observed for Pgp (34, 56). Our observations, however, are robust compared with findings with MRP1. The difficulties associated with studying ATP hydrolysis in MRP1 have been addressed elsewhere (3). On the other hand, we found no stimulation of MRP4-mediated ATP hydrolysis by the antiviral agents AZT and PMEA or the anticancer agent methotrexate (Fig. 7, A and B). Although ATP hydrolysis mediated by ABC transporters is a useful surrogate assay to identify potential transport substrates based on the premise that the transport of substrates is powered by ATP hydrolysis, not all transport substrates stimulate ATP hydrolysis in crude membrane preparations (34). An alternative strategy has been to use an agent that strongly stimulates the ATP hydrolysis and monitor whether the putative substrates compete for the binding site of the transport substrate. In Fig. 7, we demonstrate that PGE2-stimulated ATP hydrolysis is inhibited by AZT, PMEA, and methotrexate. These results suggest that the drugs can displace PGE2 from the substrate-binding site, however the fact that they bind to the substrate-binding site does not necessarily mean that they are transported by MRP4.



View larger version (13K):
[in this window]
[in a new window]
  		FIG. 7.
Effect of AZT, PMEA, and methotrexate on the prostaglandin E2-stimulated ATP hydrolysis. BeFx-sensitive, steady-state ATP hydrolysis was measured in MRP4 containing crude membranes as described above in the presence of increasing concentrations of AZT (A), PMEA (B), or methotrexate (C) either in the absence or presence of 50 µM PGE2. The concentrations required for half-maximal inhibition were obtained by fitting the data using non-linear regression analysis to a hyperbolic function. AZT and PMEA do not affect the basal ATP hydrolysis but inhibit the PGE2-stimulated ATP hydrolysis; concentrations required for half-maximal inhibition were 2.7 ± 0.43 and 2.2 ± 0.7 µM (n = 3), respectively. Methotrexate inhibited both the basal and PGE2-stimulated ATP hydrolysis, and half-maximal inhibition occurred at 5.5 ± 0.97 and 11.5 ± 2.5 µM, (n = 3) respectively.

 
Presence of the Putative MRP4 Substrate PGE2 Does Not Affect the Affinity for ATP during Hydrolysis—The effect of putative substrates of MRP4 on the Km of ATP during hydrolysis can provide information on the interaction between the substrate binding and ATP sites. We find that PGE2 has no effect on the Km(ATP) for MRP4-mediated ATP hydrolysis (Km = 0.63 mM, in the presence of PGE2 versus Km = 0.66 mM in the absence). However, there is a large increase in the Vmax (Fig. 8). This is consistent with earlier work on Pgp (34, 57).



View larger version (16K):
[in this window]
[in a new window]
  		FIG. 8.
Effect of prostaglandin E2 on the Km (ATP) for MRP4-mediated ATP hydrolysis. BeFx-sensitive, steady-state ATP hydrolysis was measured in the presence of increasing concentrations of ATP as described in Fig. 3C either in the absence or presence of 50 µM PGE2. The data were fit to the Henri-Michaelis-Menten equation by non-linear regression analysis to obtain the Km values in the absence (Km = 0.63 ± 0.07 mM; n = 3) or presence (Km = 0.66 ± 0.1 mM; n = 3) of the transport substrate, PGE2.

 
The Vi- and BeFx-induced Transition States of MRP4 —The transition state ABC protein·8-azido[{alpha}-32P]ADP·Vi, generated using the nucleotide analog 8-azido[{alpha}-32P]ATP, has proved to be extremely useful in understanding the catalytic cycles of Pgp (25, 38, 40), MRP1 (24, 26), and Tap1/Tap2 (58). We have demonstrated above that the ATP hydrolysis associated with MRP4 is sensitive to both Vi and BeFx (Fig. 3, A and B). Thus, it should be possible to assess the Vi- and BeFx-induced trapping of the nucleoside diphosphate in MRP4. To determine whether the nucleoside is effectively trapped in MRP4, we incubated crude membranes with 50 µM 8-azido[{alpha}-32P]ATP in the presence of either Vi or BeFx for 5 min at 37 °C. We transferred the samples to ice and added 10 mM cold ATP followed by photocross-linking. Because 200-fold excess ATP was added prior to photocross-linking, it would be sufficient to displace the 8-azido[{alpha}-32P]ADP, unless the nucleoside were occluded. We show in Fig. 9A that, in the absence of Vi or BeFx, there is a negligible 32P signal associated with the MRP4 band. There is a significant increase in the signal in the presence of Vi (~2-fold), but it is only with BeFx that we observe a large (6- to 8-fold) increase in the accumulation of 8-azido[{alpha}-32P]ADP in MRP4. It should be noted that the nucleotide trapping in the presence of either Vi or BeFx was not observed under non-hydrolysis conditions when the incubation was carried out at 4 °C instead of 37 °C (data not shown) consistent with previous findings with Pgp and MRP1 showing that the trapped nucleotide is the nucleoside diphosphate. Both Vi- and BeFx have been shown to be equally effective in trapping 8-azido[{alpha}-32P]ADP in Pgp. However, the results with MRP4 are consistent with the more efficient inhibition of ATP hydrolysis by BeFx as compared with Vi (Fig. 3, A and B). Also as with the data in Fig. 5, we found that BeFx-induced trapping of 8-azido[{alpha}-32P]ADP is divalent cation-dependent. Of the divalent cations tested (Mg2+, Mn2+, Co2+, and Ca2+), Co2+ appears to support the strongest trapping of the nucleoside diphosphate (Fig. 9B). Finally, we examined the effect of ADP and Pi on BeFx-induced trapping of 8-azido[{alpha}-32P]ADP. ADP inhibits BeFx-induced trapping in a concentration-dependent manner, and the concentration required for half-maximal inhibition of 61 µM, whereas Pi at concentrations as high as 20 mM has only a marginal effect (Fig. 9C).



View larger version (30K):
[in this window]
[in a new window]
  		FIG. 9.
Vi- and BeFx-induced trapping of 8-azido[{alpha}-32P]ADP into MRP4. A, Vi-and BeFx-induced trapping of 8-azido[{alpha}-32P]ADP into MRP4 was monitored as described under "Experimental Procedures." B, BeFx-induced nucleotide trapping was monitored as described above in the presence of 10 mM MgCl2, MnCl2, CoCl2, or CaCl2. The experimental conditions are depicted on the autoradiograms, and the lower panels show the amount of 32P label incorporated in the MRP4 band. Note that the gel in panel A was exposed to the x-ray film for 14 h, whereas the gel in panel B was exposed to the film for 1 h. C, BeFx-induced nucleotide trapping was monitored as described above in the presence of indicated concentrations of ADP or KH2PO4 (pH 6.8). The concentration required for half-maximal inhibition by ADP was determined to be 61 µM by a non-linear least squares fit. The experimental conditions are depicted on the autoradiograms, and results from a typical experiment are depicted. Similar results were obtained with two additional experiments.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
REFERENCES
 
Since the discovery of MRP1 over a decade ago, 11 membrane proteins have been identified as members of the ABCC subfamily of transporters (2, 3). MRP1 (ABCC1) and cystic fibrosis transmembrane regulator (ABCC7) are the most extensively studied members of this family, however, only MRP1 has been implicated in MDR (3). MRP4, a recently discovered member of the ABCC subfamily of transport proteins (59), has been implicated in the transport of cyclic nucleotides and a few anti-cancer agents (2, 3, 10). In addition, an interesting new study finds that MRP4 may be involved in the export of prostanoids from cells in which they are generated (14). The substrate profile of MRP4 thus suggests that this ABC transporter may be clinically important.

In this study, we have characterized ATP hydrolysis mediated by human MRP4 expressed in High Five insect cells and demonstrated that ATP hydrolysis is modulated by putative substrates of MRP4. To our knowledge, this is the first report describing detailed properties of ATP hydrolysis by a member of the subfamily of MRPs lacking the TMD0. We show that MRP4 exhibits a very high affinity for 8-azido[{alpha}-32P]ATP (concentration required for half-maximal binding = 2.9 µM), and the binding of 8-azido[{alpha}-32P]ATP can be displaced by ATP or the non-hydrolyzable ATP analog AMPPNP (concentrations required for half-maximal inhibition, 13.3 and 308 µM, respectively) (Fig. 2, B and C). The BeFx-sensitive ATP hydrolysis by MRP4 has a Km(ATP) of 0.62 mM (Fig. 3C). These kinetic parameters are for the most part comparable to those obtained for Pgp (3, 21). The Km(ATP) for Pgp has been reported to be in the range of 0.5–1.5 mM depending on the source of the Pgp and the laboratory (20, 34, 60). The Vmax for MRP4-mediated ATP hydrolysis is 20 nmol of Pi/mg of protein/min, which is significantly higher than the 4–5 nmol of Pi/mg of protein/min reported for MRP1 and MRP2 in a comparable system (61). Although the affinity of 8-azido[{alpha}-32P]ATP for MRP4 is greater than that for Pgp (25), it is comparable to that for MRP1.2 MRP4-mediated ATP hydrolysis is sensitive to both BeFx (>95% inhibition) and Vi (40% inhibition). Similarly, the concentration required for half-maximal inhibition by BeFx was 12.7 µM (Fig. 3B), which is comparable to that observed for Pgp (62). The kinetic parameters of BeFx-mediated inhibition of ATP hydrolysis by MRP1 or other MRPs have not been reported. The concentration required for half-maximal inhibition by Vi (Fig. 3A), on the other hand, is ~15-fold higher than the comparable value for Pgp (40, 57). Based on x-ray crystallography, it has been demonstrated that in myosin the complex generated with BeFx shows tetrahedral geometry around the beryllium atom, which is in the position thought to be occupied by the {gamma}-phosphorus of MgATP (63). In contrast, the myosin·MgADP·Vi complex shows a trigonal bipyramidal geometry around the pentacovalent vanadium (64). Moreover, the Vi to bridge-oxygen distance is longer than the equivalent Be to bridge-oxygen distance (63, 64). Thus, it is plausible that the MRP4·MgADP·BeFx and the MRP4·MgADP·Vi complexes differ in detail accounting for the variation in extent of inhibition.

Based on numerous studies with Pgp (for reviews see Refs. 22, 65–67) and a more limited number with MRP1 (for reviews see Refs. 1–3, 5), there is a general consensus that substrates influence the ATPase activity of ABC transporters. We therefore studied the effect of putative substrates of MRP4 on ATP hydrolysis. We found that the both PGE1 and PGE2 showed strong stimulation of MRP4-mediated ATP hydrolysis (Fig. 6, A and B) and that, although PGE2 increases the Vmax of MRP4-mediated ATP hydrolysis ~2- to 3-fold, it has no effect on the Km(ATP) of MRP4 (Figs. 6 and 8). Moreover, the values for apparent Km (ATP hydrolysis), 12.15 and 2.9 µM, respectively, for PGE1 and PGE2 are comparable to the Km values obtained previously in transport studies, 2.1 and 3.4 µM (14). This strong correlation between ATP hydrolysis and transport along with the fact that the prostaglandins have a relatively high affinity for MRP4 provide another line of evidence that MRP4 is involved in the transport of prostanoids as postulated by Reid et al. (14).

Drugs used in anti-viral therapy such as AZT and PMEA have also been reported to be substrates of MRP4. We, however, found no stimulation of MRP4-mediated ATP hydrolysis by either of these drugs (Fig. 7, A and B). The antimetabolite agent methotrexate on the other hand strongly inhibits MRP4-mediated ATP hydrolysis (Fig. 7C), consistent with evidence that MRP4 may confer resistance to this anti-cancer agent (13). In the case of Pgp, it has been suggested that modulators that are either bulky (e.g. cyclosporin A) or bind too tightly to Pgp are not easily transported from the "on" to the "off" site and thus inhibit ATP hydrolysis (34), and it is possible that methotrexate interacts with MRP4 in such a manner. Interestingly, methotrexate inhibits both basal and substrate (PGE2)-stimulated ATP hydrolysis (Fig. 7C). Similarly, although AZT and PMEA do not show any effect on the basal ATPase activity of MRP4, they inhibit PGE2-stimulated ATP hydrolysis in a concentration-dependent manner (Fig. 7, A and B). Our results with cyclic nucleotides, on the other hand, show that cGMP has a very modest effect on MRP4-mediated ATP hydrolysis, whereas GMP has no effect. The transport of cyclic nucleotides by MRP4 has been a controversial issue. Several reports have shown that MRP4 transports cGMP and cAMP in vesicular uptake experiments (12, 13) and in intact cells overexpressing MRP4 (11). The study with the intact cells, contrary to earlier reports, found that neither cAMP nor cGMP transport was affected by glutathione depletion (11). This is consistent with our finding that glutathione does not affect ATP hydrolysis, nor does it increase ATP hydrolysis in the presence of cyclic nucleotides (data not shown). Furthermore, our results show that cGMP has a very low affinity for MRP4 (Fig. 6C), which is consistent with the suggestion that in intact cells MRP4 mediates extrusion of cyclic nucleotides with low affinity (11) and cyclic nucleotide efflux may not be its primary role under physiological conditions.

In recent years, the catalytic cycles of ATP hydrolysis by Pgp and MRP1 have been elucidated in considerable detail (3, 21, 22), and the emerging models suggest that there are clear differences between the two. Although MRP4, similar to MRP1, belongs to the ABCC family, it belongs to a distinct subgroup (see the introduction for details). Thus, it would be important to compare the catalytic mechanism of MRP4 with those of MRP1 and Pgp. A significant technical impediment to studying the kinetics of ATP hydrolysis by ABC transport proteins is the low affinity that these proteins have for ATP (Kms in the range of 0.3–1.5 mM) compared with, for example, myosin or the mitochondrial F1F0-ATP synthase (20, 60). The use of photoaffinity analogs helps circumvent some of these difficulties and in particular the ability to generate and study the transition-state conformation of Pgp and MRP1 has been central to understanding the catalytic cycles of Pgp (21, 22) and MRP1 (23, 24, 68). In Fig. 9 we demonstrate that MRP4 can trap 8-azido[{alpha}-32P]ADP in the transition state using the inhibitors Vi and BeFx. We have shown above that, similar to other ABC transport proteins and other ATPases, MRP4-mediated ATP hydrolysis is divalent cation-dependent. Besides Mg2+ (the physiological divalent cation) Mn2+, Co2+, and Ca2+ support 8-azido[{alpha}-32P]ATP binding. However, in the presence of Co2+ there is very strong BeFx-induced trapping of 8-azido[{alpha}-32P]ADP even though Co2+ cannot support steady-state ATP hydrolysis effectively. Binding of 8-azido[{alpha}-32P]ATP in the presence of Co2+ is completely abolished by ATP, and trapping is BeFx- and Vi-dependent suggesting that Co2+ does not increase affinity of nucleotides for MRP4. A plausible explanation is that Co2+ stabilizes the MRP4·8-azido[{alpha}-32P]ADP·BeFx transition-state complex by slowing the release of the nucleoside diphosphate. This would have the effect of reducing the rate of steady-state ATP hydrolysis as is observed in Fig. 5C.

Based on the results presented in this report we propose a kinetic scheme for the ATP hydrolysis by MRP4 (Fig. 10). Initially MgATP binds to the ATP sites of MRP4 (Step I), and the presence of a divalent cation appears to be mandatory (Fig. 5, A and B). Subsequent ATP hydrolysis (Step II) results in the products MgADP and Pi. We find that ADP has a greater affinity for MRP4 than does Pi. The concentration of ADP required for half-maximal inhibition of ATP hydrolysis is 0.19 mM, whereas even 10 mM Pi has no inhibitory effect on ATP hydrolysis. Similarly, BeFx-induced trapping of 8-azido[{alpha}-32P]ADP is inhibited in a concentration-dependent manner by ADP (Fig. 9C). Thus Pi would be released first from the post-hydrolysis transition-state MRP4·MgADP·Pi (Step III) followed by ADP (Step IV). This is consistent with previous work with Pgp that demonstrated that ADP release is the rate-limiting step during the catalytic cycle of ATP hydrolysis (34).



View larger version (10K):
[in this window]
[in a new window]
  		FIG. 10.
Proposed scheme for ATP hydrolysis by MRP4. Step I: MgATP binds to the ATP-sites of MRP4, and the presence of a divalent cation is obligatory. Step II: binding of ATP is followed by ATP hydrolysis, and ADP and Pi are the products. Step III: because ADP has a higher affinity for MRP4 than Pi; the Pi would be released prior to ADP. Step IIIA/IIIB: if either BeFx or Vi is present in the reaction during ATP hydrolysis, these transition-state analogs trap the nucleoside diphosphate in a post-hydrolysis transition state, MRP4·MgADP·BeFx or MRP4·MgADP·Vi. Step IV: following Pi release, the ADP is released and MRP4 is ready for another cycle of ATP hydrolysis. Similar to Pgp (34), ADP release appears to be the rate-limiting step during MRP4-mediated ATP hydrolysis.

 
In this study, we have characterized MRP4-mediated ATP hydrolysis and presented a working model for the catalytic cycle, the basic features of which are comparable to that proposed for Pgp (21, 34, 60). However, it is likely that in many of the details MRP4 may be significantly different from Pgp. As a case in point, the two ATP sites of Pgp exhibit a great deal of functional symmetry (21, 25, 40, 69), and this is reflected in an almost 70% sequence identity between the two ATP sites. On the other hand, there is <30% identity between the two ATP sites of MRP4. Do the two ATP sites of MRP4 also show functional asymmetry? Do they have different roles during the catalytic cycle? The answers to these and other questions will allow us to elucidate in detail the catalytic cycle of MRP4 and compare it to those of Pgp and MRP1.


   FOOTNOTES
 
* The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Back

{ddagger} To whom correspondence should be addressed: Laboratory of Cell Biology, Center for Cancer Research, NCI, National Institutes of Health, Dept. of Health and Human Services, Bethesda, MD 20892-4256. Tel.: 301-402-4178; Fax: 301-435-8188; E-mail: ambudkar{at}helix.nih.gov.

1 The abbreviations used are: MRP, multidrug resistance-associated protein; ABC, ATP-binding cassette; AZT, azidothymidine; BeFx, beryllium fluoride; MDR, multidrug resistance; PGE1, prostaglandin E1; PGE2, prostaglandin E2; Pgp, P-glycoprotein; PMEA, 9-(2-phosphonylmethoxyethyl)adenine; Vi, sodium orthovanadate; MES, 4-morpholineethanesulfonic acid; AMPPNP, adenosine 5'-({beta},{gamma}-imino)triphosphate; TMD0, extra N-terminal domain of MRPs. Back

2 M. Smith, Z. Sauna, and S. V. Ambudkar, unpublished observations. Back


   ACKNOWLEDGMENTS
 
We thank Dr. Gary Kruh (Fox Chase Cancer Center) for the gift of plasmid pV21393-MRP4, Dr. Michael M. Gottesman for discussions and encouragement, and Drs. Jill Paterson and Suneet Shukla for comments on the manuscript. The editorial assistance of George Leiman is gratefully acknowledged.



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Borst, P., and Elferink, R. O. (2002) Annu. Rev. Biochem. 71, 537–592[CrossRef][Medline] [Order article via Infotrieve]
  2. Kruh, G. D., and Belinsky, M. G. (2003) Oncogene 22, 7537–7552[CrossRef][Medline] [Order article via Infotrieve]
  3. Haimeur, A., Conseil, G., Deeley, R. G., and Cole, S. P. C. (2004) Curr. Drug Metab. 5, 21–53[CrossRef][Medline] [Order article via Infotrieve]
  4. Cole, S. P., Bhardwaj, G., Gerlach, J. H., Mackie, J. E., Grant, C. E., Almquist, K. C., Stewart, A. J., Kurz, E. U., Duncan, A. M., and Deeley, R. G. (1992) Science 258, 1650–1654[Abstract/Free Full Text]
  5. Borst, P., Evers, R., Kool, M., and Wijnholds, J. (1999) Biochim. Biophys. Acta 1461, 347–357[Medline] [Order article via Infotrieve]
  6. Borst, P., Evers, R., Kool, M., and Wijnholds, J. (2000) J. Natl. Cancer Inst. 92, 1295–1302[Abstract/Free Full Text]
  7. Kruh, G. D., Zeng, H., Rea, P. A., Liu, G. S., Chen, Z. S., Lee, K., and Belinsky, M. G. (2001) J. Bioenerg. Biomembr. 33, 493–501[CrossRef][Medline] [Order article via Infotrieve]
  8. Bera, T. K., Iavarone, C., Kumar, V., Lee, S., Lee, B., and Pastan, I. (2002) Proc. Natl. Acad. Sci. U. S. A. 99, 6997–7002[Abstract/Free Full Text]
  9. Schuetz, J. D., Connelly, M. C., Sun, D. X., Paibir, S. G., Flynn, P. M., Srinivas, R. V., Kumar, A., and Fridland, A. (1999) Nat. Med. 5, 1048–1051[CrossRef][Medline] [Order article via Infotrieve]
  10. Adachi, M., Reid, G., and Schuetz, J. D. (2002) Adv. Drug Deliver. Rev. 54, 1333–1342[CrossRef][Medline] [Order article via Infotrieve]
  11. Wielinga, P. R., van der Heijden, I., Reid, G., Beijnen, J. H., Wijnholds, J., and Borst, P. (2003) J. Biol. Chem. 278, 17664–17671[Abstract/Free Full Text]
  12. Chen, Z. S., Lee, K., and Kruh, G. D. (2001) J. Biol. Chem. 276, 33747–33754[Abstract/Free Full Text]
  13. Chen, Z. S., Lee, K., Walther, S., Raftogianis, R. B., Kuwano, M., Zeng, H., and Kruh, G. D. (2002) Cancer Res. 62, 3144–3150[Abstract/Free Full Text]
  14. Reid, G., Wielinga, P., Zelcer, N., van der Heijden, I., Kuil, A., de Haas, M., Wijnholds, J., and Borst, P. (2003) Proc. Natl. Acad. Sci. U. S. A. 100, 9244–9249[Abstract/Free Full Text]
  15. Vane, J. R., and Botting, R. M. (1998) Inflamm. Res. 47, S78–S87[CrossRef][Medline] [Order article via Infotrieve]
  16. Sheng, H. M., Shao, J. Y., Washington, M. K., and DuBois, R. N. (2001) J. Biol. Chem. 276, 18075–18081[Abstract/Free Full Text]
  17. Adam, L., Mazumdar, A., Sharma, T., Jones, T. R., and Kumar, R. (2001) Cancer Res. 61, 81–87[Abstract/Free Full Text]
  18. Vane, J. R., Bakhle, Y. S., and Botting, R. M. (1998) Annu. Rev. Pharmacol. Toxicol. 38, 97–120[CrossRef][Medline] [Order article via Infotrieve]
  19. Warner, T. D., and Mitchell, J. A. (2003) Proc. Natl. Acad. Sci. U. S. A. 100, 9108–9110[Free Full Text]
  20. Senior, A. E. (1998) Acta Physiol. Scand. Suppl. 643, 213–218[Medline] [Order article via Infotrieve]
  21. Sauna, Z. E., Smith, M. M., Muller, M., Kerr, K. M., and Ambudkar, S. V. (2001) J. Bioenerg. Biomembr. 33, 481–491[CrossRef][Medline] [Order article via Infotrieve]
  22. Ambudkar, S. V., Kimchi-Sarfaty, C., Sauna, Z. E., and Gottesman, M. M. (2003) Oncogene 22, 7468–7485[CrossRef][Medline] [Order article via Infotrieve]
  23. Hou, Y.-x., Cui, L., Riordan, J. R., and Chang, X.-b. (2000) J. Biol. Chem. 275, 20280–20287[Abstract/Free Full Text]
  24. Hou, Y. X., Riordan, J. R., and Chang, X. B. (2003) J. Biol. Chem. 278, 3599–3605[Abstract/Free Full Text]
  25. Sauna, Z. E., and Ambudkar, S. V. (2001) J. Biol. Chem. 276, 11653–11661[Abstract/Free Full Text]
  26. Yang, R. Y., Cui, L. Y., Hou, Y. X., Riordan, J. R., and Chang, X. B. (2003) J. Biol. Chem. 278, 30764–30771[Abstract/Free Full Text]
  27. Gao, M., Cui, H. R., Loe, D. W., Grant, C. E., Almquist, K. C., Cole, S. P. C., and Deeley, R. G. (2000) J. Biol. Chem. 275, 13098–13108[Abstract/Free Full Text]
  28. Sauna, Z. E., Peng, X. H., Nandigama, K., Tekle, S., and Ambudkar, S. V. (2004) Mol. Pharmacol. 65, 675–684[Abstract/Free Full Text]
  29. Chick, J. (1999) Drug Safety 20, 427–435[CrossRef][Medline] [Order article via Infotrieve]
  30. Loo, T. W., and Clarke, D. M. (2000) J. Natl. Cancer Inst. 92, 898–902[Abstract/Free Full Text]
  31. Ramachandra, M., Ambudkar, S. V., Chen, D., Hrycyna, C. A., Dey, S., Gottesman, M. M., and Pastan, I. (1998) Biochemistry 37, 5010–5019[CrossRef][Medline] [Order article via Infotrieve]
  32. Lee, K., Klein-Szanto, A. J. P., and Kruh, G. D. (2000) J. Natl. Cancer Inst. 93, 1934–1940
  33. Lee, K., Belinsky, M. G., Bell, D. W., Testa, J. R., and Kruh, G. D. (1998) Cancer Res. 58, 2741–2747[Abstract/Free Full Text]
  34. Kerr, K. M., Sauna, Z. E., and Ambudkar, S. V. (2001) J. Biol. Chem. 276, 8657–8664[Abstract/Free Full Text]
  35. Schaffner, W., and Weissmann, C. (1973) Anal. Biochem. 56, 502–514[CrossRef][Medline] [Order article via Infotrieve]
  36. Sauna, Z. E., Smith, M. M., Muller, M., and Ambudkar, S. V. (2001) J. Biol. Chem. 276, 21199–21208[Abstract/Free Full Text]
  37. Ambudkar, S. V. (1998) Methods Enzymol. 292, 504–514[Medline] [Order article via Infotrieve]
  38. Sauna, Z. E., and Ambudkar, S. V. (2000) Proc. Natl. Acad. Sci. U. S. A. 97, 2515–2520[Abstract/Free Full Text]
  39. Sauna, Z. E., Mller, M., Peng, X. H., and Ambudkar, S. V. (2002) Biochemistry 41, 13989–14000[CrossRef][Medline] [Order article via Infotrieve]
  40. Urbatsch, I. L., Sankaran, B., Weber, J., and Senior, A. E. (1995) J. Biol. Chem. 270, 19383–19390[Abstract/Free Full Text]
  41. Hrycyna, C. A., Ramachandra, M., Ambudkar, S. V., Ko, Y. H., Pedersen, P. L., Pastan, I., and Gottesman, M. M. (1998) J. Biol. Chem. 273, 16631–16634[Abstract/Free Full Text]
  42. Rosenberg, M. F., Velarde, G., Ford, R. C., Martin, C., Berridge, G., Kerr, I. D., Callaghan, R., Schmidlin, A., Wooding, C., Linton, K. J., and Higgins, C. F. (2001) EMBO J. 20, 5615–5625[CrossRef][Medline] [Order article via Infotrieve]
  43. Loo, T. W., and Clarke, D. M. (2002) Proc. Natl. Acad. Sci. U. S. A. 99, 3511–3516[Abstract/Free Full Text]
  44. Al-Shawi, M. K., and Senior, A. E. (1993) J. Biol. Chem. 268, 4197–4206[Abstract/Free Full Text]
  45. Sarkadi, B., Price, E. M., Boucher, R. C., Germann, U. A., and Scarborough, G. A. (1992) J. Biol. Chem. 267, 4854–4858[Abstract/Free Full Text]
  46. Sharom, F. J., Yu, X., Chu, J. W., and Doige, C. A. (1995) Biochem. J. 308, 381–390[Medline] [Order article via Infotrieve]
  47. Hooijberg, J. H., Pinedo, H. M., Vrasdonk, C., Priebe, W., Lankelma, J., and Broxterman, H. J. (2000) FEBS Lett. 469, 47–51[CrossRef][Medline] [Order article via Infotrieve]
  48. Chang, X. B., Hou, Y. X., and Riordan, J. R. (1997) J. Biol. Chem. 272, 30962–30968[Abstract/Free Full Text]
  49. Kerr, I. D. (2002) Biochim. Biophys. Acta 1561, 47–64[Medline] [Order article via Infotrieve]
  50. Al-Shawi, M. K., Urbatsch, I. L., and Senior, A. E. (1994) J. Biol. Chem. 269, 8986–8992[Abstract/Free Full Text]