|
|
||||||||
J. Biol. Chem., Vol. 278, Issue 23, 20449-20452, June 6, 2003
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
ACCELERATED PUBLICATIONS

From the Canadian Institutes of Health Research Group in Membrane Biology, Department of Medicine and Department of Biochemistry, University of Toronto, Toronto, Ontario M5S 1A8, Canada
Received for publication, April 10, 2003
, and in revised form, April 21, 2003.
| ABSTRACT |
|---|
|
|
|---|
| INTRODUCTION |
|---|
|
|
|---|
P-gp is a member of the ATP-binding cassette family of transporters. The 1280 amino acids of P-gp are arranged as two repeating units of 610 amino acids that are joined by a linker region of about 60 amino acids (5). Each repeat has six transmembrane (TM) segments and a hydrophilic domain containing an ATP-binding site (6, 7, 8). The minimum functional unit is a monomer (9), but the two halves of the molecule do not have to be covalently linked for function (10, 11). Both ATP-binding sites are required for activity (7, 12, 13, 14) and likely function in an alternating mechanism (15). Studies on deletion mutants have shown that the TM domains alone are sufficient to mediate drug binding (11). Studies on the activity of cysteine mutants and their inhibition by different thiol-reactive substrate analogs indicate that residues from multiple TM segments contribute to the drug-binding site (16, 17, 18, 19, 20, 21).
To determine the mechanism of P-gp, it is important to understand how drug binding and subsequent efflux are coupled to ATP hydrolysis. A key step in the reaction cycle is activation of ATP hydrolysis when the drug-binding site is occupied. It is not known whether drug-stimulated ATPase activity requires that the drug-binding site alternate between occupied and unoccupied states or whether ATPase activity is permanently activated when the drug-binding site is permanently occupied.
In this study we used cysteine-scanning mutagenesis and reaction with the thiol-reactive analog of verapamil, MTS-verapamil, to test for the presence of a permanently activated P-gp intermediate.
| EXPERIMENTAL PROCEDURES |
|---|
|
|
|---|
Expression, Reaction with MTS-verapamil, Purification, and Measurement
of Drug-stimulated ATPase Activities of P-gp MutantsThe
histidine-tagged P-gp mutants were expressed in HEK 293 cells. After 24 h, the
medium was replaced with fresh medium containing 10 µM
cyclosporin A. Cyclosporin A acts as a specific chemical chaperone to promote
maturation of P-gp and increases the yield of P-gp
(24,
25). After another 24 h, the
transfected cells from 50 10-cm-diameter culture plates were washed three
times with phosphate buffered saline (PBS, pH 7.4) and suspended in 1.5 ml of
PBS. The cells were solubilized by addition of 1 volume of PBS containing 2%
(w/v) n-dodecyl
-D-maltoside. Insoluble material was
removed by centrifugation at 16,000 x g for 15 min at 4 °C.
Half of the supernatant (1.3 ml) was mixed with 8 µl of Me2SO
(control), and the other half was mixed with 8 µl of 50 mM
MTS-verapamil (in Me2SO; final concentration, 0.3 mM).
The mixtures were incubated for 10 min at 22 °C, and cooled to 4 °C,
and then 150 µl of 3 M NaCl and 50 µl of 1 M
imidazole (pH 7.0) were added. P-gp(His)10 was then isolated by
nickel-chelate chromatography as described previously
(23). The recovery of P-gp was
monitored by immunoblot analysis with rabbit anti-P-gp polyclonal antibody
(9). To protect P-gp from
labeling with MTS-verapamil, 0.5 mM cyclosporin A, 5 mM
demecolcine, 5 mM verapamil, 0.5 mM vinblastine, 10
mM colchicine, or 2 mM calcein-AM was added to the
solubilized material before addition of MTS-verapamil. The highest
concentrations possible (solubility in aqueous solution) were chosen to
maximize protection from modification by MTS-verapamil.
The isolated P-gp(His)10 was mixed with lipid and sonicated, and ATPase activity was determined (26). A sample of the P-gp-lipid mixture was mixed with an equal volume of buffer containing 100 mM Tris-HCl, pH 7.4, 100 mM NaCl, 20 mM MgCl2, 10 mM ATP, and either no drug, 1.2 mM calcein-AM, 0.3 mM cyclosporin A, 6 mM demecolcine, 1.2 mM trans-(E)-flupentixol, 2 mM verapamil, or 0.2 mM vinblastine. These concentrations caused maximal stimulation or inhibition of the ATPase activity of Cys-less P-gp. The samples were then incubated for 30 min at 37 °C, and the amount of inorganic phosphate liberated was determined (27).
Inhibition of ATPase activity by vanadate was done by addition of an equal volume of buffer containing 100 mM Tris-HCl, pH 7.4, 100 mM NaCl, 20 mM MgCl2, 10 mM ATP, and 0.2 mM orthovanadate to the P-gp-lipid mixture (26). Orthovanadate was prepared from Na3VO4, pH 10 and boiled for 2 min to break down polymeric species before use (28).
| RESULTS |
|---|
|
|
|---|
A useful thiol-reactive drug substrate is MTS-verapamil
(Fig. 1). It is an analog of
verapamil and contains an alkylthiosulfonate group that reacts selectively
with cysteine residues (29,
30). We showed previously that
MTS-verapamil is a good substrate of Cys-less P-gp as it stimulated its ATPase
activity 10-fold and with a Km of 25
µM (Km for verapamil, 25
µM) (19).
Accordingly a cysteine was introduced at each position in the predicted TM
segments of Cys-less P-gp(His)10. A total of 252 mutants were made,
expressed in HEK 293 cells, and then solubilized with 1% (w/v)
n-dodecyl
-D-maltoside. Insoluble material was
removed by centrifugation, and the samples were treated with or without 0.3
mM MTS-verapamil. This concentration of MTS-verapamil caused
maximum activation of Cysless P-gp
(19). P-gp was then
immobilized on a nickel-chelate column, washed with buffer to remove unbound
MTS-verapamil, and eluted from the column. The ATPase activities of the P-gp
treated with and without MTS-verapamil were determined. One mutant, I306C,
showed an 8-fold increase in ATPase activity after treatment with
MTS-verapamil. Activation by MTS-verapamil was concentration-dependent.
Fig. 1 shows that maximal
stimulation occurred after treatment of mutant I306C with 0.11
mM MTS-verapamil, and half-maximal stimulation was 39
µM. Nickel-chelate chromatography was effective in removing
unreacted MTS-verapamil because the activity of Cys-less P-gp remained at
basal levels even after pretreatment with 1 mM MTS-verapamil
(Fig. 1).
|
To test whether all of mutant I306C had been modified by MTS-verapamil, we assayed for stimulation or inhibition of the ATPase activity by other drug substrates. The rationale is that if a significant amount of unmodified mutant I306C is present, then the presence of other substrates or inhibitors should affect the activity of the mutant. We tested compounds that are potent stimulators (calcein-AM, demecolcine, vinblastine, and verapamil) (17, 31, 32, 33) or inhibitors (cyclosporin A and trans-(E)-flupentixol) of P-gp ATPase activity (33, 34, 35). Fig. 2 shows the effect of various stimulators and inhibitors on the ATPase activity of mutant I306C before and after treatment with MTS-verapamil. Before treatment with MTS-verapamil, the ATPase activity of mutant I306C was stimulated by demecolcine, calcein-AM, verapamil, and vinblastine by 11.9-, 9.8-, 7-, and 3-fold, respectively. By contrast, cyclosporin A and trans-(E)-flupentixol inhibited the ATPase activity (0.9- and 0.5-fold, respectively). When mutant I306C was pretreated with MTS-verapamil, the presence of other stimulators or inhibitors of P-gp had little effect on its ATPase activity (7.78.1-fold versus 8-fold increase) (Fig. 2). This inability to further stimulate or inhibit the activity of the MTS-verapamil-treated mutant I306C suggests that most (more than 90%) of the mutant I306C P-gp was modified and that covalent attachment of verapamil in the drug-binding site blocks access of other drug substrates to the drug-binding site.
|
It was possible that reaction of the mutant with MTS-verapamil partially denatured the protein so that the ATP-binding sites are uncoupled from the drug-binding sites. If modified I306C retained a "native" structure, then it should be possible to restore basal levels of ATPase activity if covalent attachment of MTS-verapamil was removed. Accordingly modified I306C was treated with 20 mM dithiothreitol to reduce the disulfide bond between P-gp and MTS-verapamil, and then P-gp was reisolated by nickel-chelate chromatography. Fig. 3 shows that reduction of the disulfide bond by dithiothreitol reduced the activity of the mutant as its ATPase activity was only slightly higher (1.5-fold) than the untreated I306C mutant. Mutant I306C remained active after removal of the covalently bound MTS-verapamil since it retained the ability to be stimulated by verapamil (6.9-fold).
|
The characteristics of the ATPase activity of modified I306C mutant were also examined. We tested for vanadate trapping of nucleotide. P-gp traps nucleotides in the presence of vanadate plus Mg-ATP and results in the formation of an inhibitory transition state that inhibits ATP hydrolysis at the second ATP site (36). Fig. 3 shows that mutant I306C that had been labeled with MTS-verapamil was still inhibited by vanadate. In addition, the modified mutant I306C had a Km for ATP (1.1 mM) that was very similar to that of Cys-less P-gp (1 mM) (data not shown).
If MTS-verapamil occupied the drug-binding site in mutant I306C, then pretreatment of the mutant with other drug substrates should protect it from labeling by MTS-verapamil if they shared a common drug-binding site. Accordingly mutant I306C was treated with or without the drug substrates calcein-AM, demecolcine, verapamil, vinblastine, cyclosporin A, or colchicine before treatment with MTS-verapamil for 10 min at 22 °C. P-gp was isolated by nickel-chelate chromatography, and ATPase activity was determined. Fig. 4 shows that all of the drug substrates protected mutant I306C from labeling with MTS-verapamil since they prevented modification and activation of I306C by MTS-verapamil by 7085%. These results suggest that the compounds tested likely share a common drug-binding site.
|
| DISCUSSION |
|---|
|
|
|---|
A model to explain the activated state of P-gp is shown in Fig. 5. In the resting state (state I), P-gp can bind ATP with normal affinity, but ATPase activity is low because the opposing signature "LSGGQ" motifs are farther away from the Walker A sites in the nucleotide-binding domains (NBDs) (33, 37). Binding of verapamil in the TM segments induces conformation changes in the NBDs to bring the LSGGQ and Walker A sequences closer (state II) and promote ATP hydrolysis (38, 39). ATP hydrolysis then changes the conformation of the TM segments (26, 40) leading to drug release (state III). Expulsion of verapamil from the drug-binding site allows the LSGGQ and Walker A sequences to move apart, and P-gp returns to the resting state (state IV) (41, 42). Covalent attachment of verapamil, however, results in the permanent occupation of the drug-binding site (state IIA). The inability of MTS-verapamil to exit the activated state keeps the LSGGQ and Walker A sequences close together and in a conformation that ensures continued hydrolysis of ATP and recycling between states IIA and IIIA (Fig. 5, dotted arrow). It is possible that Ile-306 may be part of the mechanism that senses that the drug-binding site is occupied and then initiates conformational changes in the molecule.
|
Our model predicts that covalent modification of the drug-binding site can result in permanent activation of ATP hydrolysis in the absence of transport because hydrolysis of ATP and release of inorganic phosphate (Pi) occurs before vectorial displacement of the drug ligand. In this respect, ATP-binding cassette transporters appear to be different from the P-type ATPases such as the sarcoplasmic reticulum (SERCA) Ca-ATPase. In the P-type ATPases, ATP hydrolysis is linked to conformational changes in the calcium-binding site, but Pi is not released until after vectorial transport of calcium (43, 44).
This study also shows that TM5 must contribute residues to the drug-binding site since other substrates and inhibitors could protect I306C from labeling by MTS-verapamil. We had predicted that I306C likely lined the drug-binding site because it could be cross-linked to other cysteines in TMs 10, 11, and 12 with thiol-reactive cross-linker substrates (20).
| FOOTNOTES |
|---|
Recipient of the Canada Research Chair in Membrane Biology. To whom
correspondence should be addressed: Dept. of Medicine, University of Toronto,
Rm. 7342, Medical Sciences Bldg., 1 King's College Circle, Toronto, Ontario
M5S 1A8, Canada. Tel./Fax: 416-978-1105; E-mail:
david.clarke{at}utoronto.ca.
1 The abbreviations used are: P-gp, P-glycoprotein; TM, transmembrane; HEK,
human embryonic kidney; MTS, methanethiosulfonate; PBS, phosphate-buffered
saline; NBD, nucleotide-binding domain; AM, acetoxymethyl ester. ![]()
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
T. W. Loo, M. C. Bartlett, and D. M. Clarke Arginines in the First Transmembrane Segment Promote Maturation of a P-glycoprotein Processing Mutant by Hydrogen Bond Interactions with Tyrosines in Transmembrane Segment 11 J. Biol. Chem., September 5, 2008; 283(36): 24860 - 24870. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. W. Loo, M. C. Bartlett, and D. M. Clarke Suppressor Mutations in the Transmembrane Segments of P-glycoprotein Promote Maturation of Processing Mutants and Disrupt a Subset of Drug-binding Sites J. Biol. Chem., November 2, 2007; 282(44): 32043 - 32052. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. W. Loo, M. C. Bartlett, and D. M. Clarke Insertion of an Arginine Residue into the Transmembrane Segments Corrects Protein Misfolding J. Biol. Chem., October 6, 2006; 281(40): 29436 - 29440. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. Y. Chen, M. C. Bartlett, T. W. Loo, and D. M. Clarke The {Delta}F508 Mutation Disrupts Packing of the Transmembrane Segments of the Cystic Fibrosis Transmembrane Conductance Regulator J. Biol. Chem., September 17, 2004; 279(38): 39620 - 39627. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. W. Loo, M. C. Bartlett, and D. M. Clarke Val133 and Cys137 in Transmembrane Segment 2 Are Close to Arg935 and Gly939 in Transmembrane Segment 11 of Human P-glycoprotein J. Biol. Chem., April 30, 2004; 279(18): 18232 - 18238. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. W. Loo, M. C. Bartlett, and D. M. Clarke Disulfide Cross-linking Analysis Shows That Transmembrane Segments 5 and 8 of Human P-glycoprotein Are Close Together on the Cytoplasmic Side of the Membrane J. Biol. Chem., February 27, 2004; 279(9): 7692 - 7697. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. W. Loo, M. C. Bartlett, and D. M. Clarke Methanethiosulfonate Derivatives of Rhodamine and Verapamil Activate Human P-glycoprotein at Different Sites J. Biol. Chem., December 12, 2003; 278(50): 50136 - 50141. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. W. Loo, M. C. Bartlett, and D. M. Clarke Simultaneous Binding of Two Different Drugs in the Binding Pocket of the Human Multidrug Resistance P-glycoprotein J. Biol. Chem., October 10, 2003; 278(41): 39706 - 39710. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| All ASBMB Journals | Molecular and Cellular Proteomics |
| Journal of Lipid Research | ASBMB Today |