ABCMdb

PMID: 10529234

Hrycyna CA, Ramachandra M, Germann UA, Cheng PW, Pastan I, Gottesman MM
Both ATP sites of human P-glycoprotein are essential but not symmetric.
Biochemistry. 1999 Oct 19;38(42):13887-99., 1999-10-19 [PubMed]

Sentences

No. Mutations Sentence Comment

60 ABCB1 p.Asp555Asn Two internal complementary primers were used, each containing the mutation of interest (G f A at position 1663 for D555N). Login to comment 61 ABCB1 p.Asp555Asn The coding sequence for the D555N mutant primer was 5'- CTCCTGCTGAATGAGGCCAC-3'. Login to comment 70 ABCB1 p.Asp555Asn The resulting plasmid was called pTM1-MDR1-D555N. Login to comment 71 ABCB1 p.Asp1200Asn ABCB1 p.Asp1200Asn The D1200N mutant DNA was constructed similarly with two internal complementary primers, each containing the mutation of interest (G f A at position 3598 for D1200N). Login to comment 72 ABCB1 p.Asp1200Asn The coding sequence for the D1200N mutant primer was 5'-ATATTTTGCTTTTGAATGAAG-3'. Login to comment 78 ABCB1 p.Asp1200Asn The resulting plasmid was named pTM1-MDR1-D1200N. Login to comment 79 ABCB1 p.Asp555Asn ABCB1 p.Asp555Asn ABCB1 p.Asp1200Asn ABCB1 p.Asp1200Asn The D555N/D1200N double mutant was constructed exactly like pTM1-MDR1-D1200N except that pTM1-MDR1-D555N was used as the template for the first PCR reaction. Login to comment 80 ABCB1 p.Asp555Asn ABCB1 p.Asp1200Asn The resulting expression plasmid was named pTM1-MDR1-D555N/ D1200N. Login to comment 84 ABCB1 p.Lys1061Arg To introduce the FspI site, nucleotides 3181-3183 were changed from AAG to CGC, which also resulted in a single amino acid substitution (K1061R). Login to comment 87 ABCB1 p.Gln1215Asp ABCB1 p.Val1214Ile These changes resulted in two amino acid changes (V1214I and Q1215D). Login to comment 106 ABCB1 p.Asp555Asn ABCB1 p.Asp555Asn ABCB1 p.Asp1200Asn ABCB1 p.Asp1200Asn Recombinant vaccinia viruses vvT7MDR1(WT) (wild-type P-gp), vvMDR1-CM (P-gp-D1200N), vvMDR1-Mg5 (P-gp-D555N), and vvMDR1-DM (P-gp-D555N/D1200N) were constructed as previously described (28) from the corresponding pTM1-MDR1 vectors described above. Login to comment 163 ABCB1 p.Asp555Asn ABCB1 p.Asp1200Asn The two mutants, designated P-gp-D555N and P-gp-D1200N, were constructed by site-directed mutagenesis, cloned into pTM1-MDR1, and transiently expressed in HeLa cells coinfected with the vaccinia virus vTF7-3 as described under Experimental Procedures. Login to comment 165 ABCB1 p.Asp555Asn ABCB1 p.Asp555Asn ABCB1 p.Asp1200Asn ABCB1 p.Asp1200Asn Total extracts from HeLa cells expressing wild-type P-gp, P-gp-D555N, P-gp-D1200N, and the double mutant P-gp-D555N/ D1200N were subjected to immunoblot analysis with the monoclonal antibody C219 (32). Login to comment 166 ABCB1 p.Asp555Asn ABCB1 p.Asp555Asn ABCB1 p.Asp1200Asn ABCB1 p.Asp1200Asn While P-gp-D1200N was found to be expressed at a level comparable to wild-type P-gp, in this preparation of membranes it appears that less P-gp-D555N and P-gp D555N/D1200N were expressed compared to wild-type. Login to comment 169 ABCB1 p.Asp555Asn ABCB1 p.Asp1200Asn Fluorescent substrate accumulation assays were performed on intact HeLa cells expressing wild-type P-gp, P-gp-D555N, and P-gp-D1200N with the P-gp substrate calcein-AM. Login to comment 170 ABCB1 p.Asp555Asn ABCB1 p.Asp1200Asn HeLa cells were transiently infected with vTF7-3 and transfected with pTM1-MDR1 (wild-type), pTM1-MDR1-D555N, and pTM1-MDR1-D1200N, with pTM1 as the negative control, and allowed to incubate for approximately 24 h. After incubation with 0.5 µM calcein-AM for 10 min, cells expressing wild-type P-gp accumulated less substrate than the pTM1 control as measured by a decrease in fluorescence intensity (Figure 2). Login to comment 172 ABCB1 p.Asp555Asn ABCB1 p.Asp1200Asn Strikingly, the two Walker B mutants P-gp-D555N and P-gp-D1200N were completely devoid of transport function (Figure 2). Login to comment 178 ABCB1 p.Asp555Asn ABCB1 p.Asp555Asn ABCB1 p.Asp555Asn ABCB1 p.Asp555Asn ABCB1 p.Asp1200Asn ABCB1 p.Asp1200Asn ABCB1 p.Asp1200Asn ABCB1 p.Asp1200Asn The ability of Walker B mutants P-gp-D555N, P-gp-D1200N, and P-gp-D555N/D1200N to hydrolyze ATP either in the absence (basal) or presence (drug-stimulated) of 25 µM verapamil was measured as the vanadate-sensitive release of inorganic phosphate from Mg2+ ‚ATP as described under Experimental Procedures, with membrane preparations from vTF7-3-infected HeLa cells alone or vTF7-3-infected cells that were coinfected with recombinant vaccinia viruses encoding wild-type P-gp, P-gp-D555N, P-gp-D1200N, and P-gp-D555N/D1200N. Login to comment 184 ABCB1 p.Asp555Asn ABCB1 p.Asp1200Asn (A) vTF7-3-infected-transfected HeLa cells (500 000) expressing wild-type P-gp (bold line), P-gp-D555N (---), and P-gp-D1200N (-‚-) were subjected to FACS analysis after incubation with 5 µg of MRK-16 for 30 min at 37 °C, washing (200g, 5 min), and staining with FITC-conjugated anti-mouse IgG secondary antibody (1 µg) for 30 min at 37 °C. Login to comment 188 ABCB1 p.Asp555Asn ABCB1 p.Asp555Asn ABCB1 p.Asp1200Asn HeLa cells were infected with vTF7-3 recombinant vaccinia viruses expressing wild-type P-gp, P-gp-D555N, P-gp-D1200N, and P-gp-D555N/D12000N and harvested after 48 h. Membranes (1 µg/lane) prepared from these cells were subjected to SDS-PAGE and immunoblotting with monoclonal antibody C219 (1:1500). Login to comment 191 ABCB1 p.Asp555Asn ABCB1 p.Asp1200Asn HeLa cells were infected with vTF7-3 and transfected with pTM1-MDR1 (wild type), pTM1-MDR1-D555N, and pTM1-MDR1-D1200N for 24 h. Calcein (0.5 µM) accumulation was determined in these cells by FACS after a 10 min incubation at 37 °C in the presence (thin line) and absence (bold line) of 2 µM cyclosporin A. vTF7-3-infected cells transfected with pTM1 vector DNA were included as a negative control. Login to comment 195 ABCB1 p.Asp555Asn ABCB1 p.Asp555Asn ABCB1 p.Asp1200Asn ABCB1 p.Asp1200Asn Crude membrane preparations derived from HeLa cells infected with vTF7-3 alone and from cells coinfected with recombinant vaccinia viruses encoding wild-type P-gp, P-gp-D555N, P-gp-D1200N, and P-gp-D555N/D1200N were labeled with [125I]IAAP at a concentration of 7.5 nM, immunoprecipitated with anti-P-gp polyclonal antibody PEPG-2 (33), and subjected to SDS-PAGE and autoradiography (Figure 4A). Login to comment 202 ABCB1 p.Asp555Asn ABCB1 p.Asp555Asn ABCB1 p.Asp1200Asn ABCB1 p.Asp1200Asn To further demonstrate that ATP hydrolysis is essential and to evaluate the roles of the two NBDs for inhibition of [125I]IAAP labeling as a result of vanadate trapping, we analyzed the effects of the Walker B consensus motif mutations D555N, D1200N, and D555N/D1200N. Login to comment 207 ABCB1 p.Asp555Asn ABCB1 p.Asp1200Asn Human P-gp-D555N and P-gp-D1200N Differ Dramatically in Their Ability To Be Labeled with [R-32P]-8-Azido-ATP. Login to comment 210 ABCB1 p.Asp555Asn ABCB1 p.Asp555Asn ABCB1 p.Asp1200Asn ABCB1 p.Asp1200Asn Membrane preparations from HeLa cells infected with vTF7-3 and coinfected with vvT7MDR1 (WT) (wild-type P-gp), vvMDR1-CM (P-gp-D1200N), vvMDR1-Mg5 (P-gp-D555N), or vvMDR1-DM (P-gp-D555N/D1200N). Login to comment 216 ABCB1 p.Asp555Asn ABCB1 p.Asp555Asn ABCB1 p.Asp1200Asn (A) Wild type and the mutant P-gps D555N, D1200N, and D555N/1200N were expressed in HeLa cells by use of recombinant vaccinia viruses, and crude membranes (45 µg) were used for [125I]IAAP photoaffinity labeling. Login to comment 220 ABCB1 p.Asp555Asn ABCB1 p.Asp555Asn ABCB1 p.Asp1200Asn ABCB1 p.Asp1200Asn Wild-type (b) and mutant Pgps, D555N ([), D1200N (4), and D555N/D1200N (O) were expressed in HeLa cells by use of recombinant vaccinia viruses and crude membranes (15 µg) were used for [125I]IAAP labeling in the presence of varying concentrations of vanadate, 5 mM MgCl2, and 2.5 mM ATP. Login to comment 225 ABCB1 p.Asp1200Asn Membranes expressing wild-type P-gp and P-gp-D1200N were labeled with [R-32 P]-8-azido-ATP. Login to comment 227 ABCB1 p.Asp555Asn ABCB1 p.Asp555Asn ABCB1 p.Asp1200Asn Strikingly, the [R-32P]-8-azido-ATP labeling observed in membranes expressing P-gp-D555N and P-gp-D555N/D1200N was severely impaired. Login to comment 229 ABCB1 p.Asp555Asn These results suggest that the ability to bind ATP to the N-half ATP site, as determined by photoaffinity labeling, is eliminated or reduced significantly in the protein containing a mutation in the magnesium binding site (D555N). Login to comment 232 ABCB1 p.Asp555Asn The same experiment was subsequently performed at a saturating concentration of 77.5 µM [R-32P]-8-azido-ATP, and under these conditions, the same differential labeling phenomenon is observed, although slightly more label is incorporated into P-gp-D555N (Figure 5B). Login to comment 238 ABCB1 p.Asp1200Asn As can be seen, the N-half of P-gp is labeled predominantly in the wild-type and D1200N proteins (Figure 6A). Login to comment 239 ABCB1 p.Asp555Asn As was demonstrated in Figure 5, again, no substantial labeling was observed for P-gp-D555N. Login to comment 251 ABCB1 p.Asp555Asn ABCB1 p.Asp555Asn ABCB1 p.Asp1200Asn ABCB1 p.Asp1200Asn Crude membrane preparations (50 µg) from HeLa cells infected with vTF7-3 and recombinant vaccinia viruses expressing wild-type P-gp, P-gp-D555N, P-gp-D1200N, or P-gp-D555N/ D1200N were photoaffinity-labeled on ice (0 °C) with either (A) 2.5 µM or (B) 77.5 µM [R-32P]-8-azido-ATP. Login to comment 255 ABCB1 p.Asp555Asn ABCB1 p.Asp555Asn ABCB1 p.Asp1200Asn ABCB1 p.Asp1200Asn Crude membrane preparations (50-60 µg) from HeLa cells infected with vTF7-3 and recombinant vaccinia viruses expressing wild-type P-gp, P-gp-D555N, P-gp-D1200N, or P-gp-D555N/D1200N were photoaffinity-labeled with either (A) [R-32P]- 8-azido-ATP (2.5 µM) or (B) [125I]IAAP (15 nM) as described in the legend to Figures 4 and 5. Login to comment 270 ABCB1 p.Asp1200Asn There was no apparent increase in the amount of label associated with the C-half of P-gp-D1200N under hydrolysis conditions, suggesting that the mutant is incapable of catalyzing even one ATP hydrolysis reaction (Figure 7B, lane 8). Login to comment 271 ABCB1 p.Asp1200Asn However, the N-half of P-gp-D1200N retained its ability to be labeled with [R-32P]-8-azido-ATP (Figure 7B, lanes 6 and 8) under both binding and hydrolysis conditions. Login to comment 272 ABCB1 p.Asp555Asn ABCB1 p.Asp555Asn ABCB1 p.Asp1200Asn ABCB1 p.Asp1200Asn Similar to P-gp-D1200N, P-gp-D555N and P-gp-D555N/D1200N were not capable of being vanadate-trapped, confirming their lack of ATPase activity (data not shown). Login to comment 277 ABCB1 p.Asp1200Asn Wild-Type and D1200N Human P-gp Proteins Do Not Differ Significantly in Their Apparent Affinity for ATP. Login to comment 279 ABCB1 p.Asp1200Asn The specific P-gp bands were quantified by phosphorimaging and the data were analyzed as described under Experimental FIGURE 7: [R-32P]-8-Azido-ATP photoaffinity labeling and vanadate trapping of wild-type and P-gp-D1200N from HeLa cell membranes. Login to comment 280 ABCB1 p.Asp1200Asn Crude membrane preparations (100 µg) from HeLa cells infected with vTF7-3 and the recombinant vaccinia virus expressing wild-type or P-gp-D1200N were photoaffinity-labeled with 2.5 µM [R-32P]-8-azido-ATP for 10 min. Login to comment 286 ABCB1 p.Asp1200Asn (B) Examination of wild-type P-gp (lanes 1-4) and D1200N P-gp (lanes 5-8) under [R-32P]-8-azido-ATP hydrolysis/binding conditions: lanes 1 and 5, 0 °C, (-) trypsin, (-) 300 µM vanadate; lanes 2 and 6, 0 °C, (+) trypsin, (-) 300 µM vanadate; lanes 3 and 7, 37 °C, (+) 300 µM vanadate, (+) 25 µM verapamil; lanes 4 and 8, 37 °C, (+) 300 µM vanadate, (+) 25 µM verapamil, (+) trypsin. Login to comment 292 ABCB1 p.Asp555Asn Under the conditions of this assay, P-gp-D555N was labeled poorly, precluding the reliable estimation of the concentration required for 50% inhibition of labeling by ATP for this construct. Login to comment 293 ABCB1 p.Asp1200Asn These data suggest that the wild-type and P-gp D1200N mutant proteins have similar apparent affinities for ATP. Login to comment 294 ABCB1 p.Asp555Asn Under the conditions of the assay, the binding capacity of the C-site, as demonstrated by labeling of P-gp-D555N, is limited and could not be reliably quantitated. Login to comment 297 ABCB1 p.Asp555Asn ABCB1 p.Asp1200Asn To determine if conformational differences exist between active wild-type Pgp and inactive P-gp-D555N and P-gp-D1200N, the reactivity of these constructs expressed transiently in HeLa cells was assessed by use of the conformation-sensitive monoclonal anti-human P-gp antibody UIC2 (5, 45). Login to comment 300 ABCB1 p.Asp555Asn ABCB1 p.Asp1200Asn Furthermore, the UIC2 reactivity of cells expressing either P-gp-D555N and P-gp-D1200N is indistinguishable from each other, in either the presence or absence of cyclosporin A, and is equivalent to that of the cyclosporin A-treated wild-type-expressing cells. Login to comment 301 ABCB1 p.Asp555Asn ABCB1 p.Asp1200Asn These UIC2 reactivity data, taken together with the ATPase data that demonstrated complete abrogation of ATPase activity for the D555N and D1200N P-gp proteins (Figure 3) and the ATP labeling data that demonstrated marked differences in the ability of these molecules to bind ATP (Figure 5), strongly suggest that the intrinsic ability of the P-gp molecule to hydrolyze ATP or associate with hydrolysis products accounts for the difference in conformation detected by UIC2. Login to comment 304 ABCB1 p.Asp1200Asn Crude membrane preparations from HeLa cells expressing either wild-type P-gp or P-gp-D1200N were incubated and photoaffinity-labeled in the presence of increasing MgCl2 (0-10 mM) followed by immunoprecipitation with PEPG-2, SDS-PAGE, and autoradiography. Login to comment 306 ABCB1 p.Asp1200Asn ABCB1 p.Asp1200Asn Figure 9 demonstrates that for both the wild-type P-gp (Figure 9A) and P-gp-D1200N (Figure 9B) the labeling with [R-32P]-8-azido-ATP increased in a magnesium concentration-dependent manner with half-maximum incorporation at 152.9 ( 29.9 µM for wild-type P-gp and 161.5 ( 39.7 µM for P-gp-D1200N. Login to comment 307 ABCB1 p.Asp555Asn The extremely low labeling efficiency of P-gp-D555N precluded analysis of that protein in this assay. Login to comment 310 ABCB1 p.Asp1200Asn Taken together, these data demonstrate that the P-gp-D1200N and wild-type P-gp proteins have similar requirements for Mg2+ and that the requirement is 50-60 times greater than the [R-32 P]-8-azido-ATP concentration (2.5 µM) used in the assay, suggesting another role for magnesium in addition to complexing with ATP in the active site. Login to comment 315 ABCB1 p.Asp555Asn ABCB1 p.Asp1200Asn FIGURE 8: UIC2 reactivity shift induced by incubation with cyclosporin A. vTF7-3-infected-transfected HeLa cells expressing wild-type P-gp, P-gp-D555N, or P-gp-D1200N were subjected to FACS analysis after staining with UIC2 in the presence (thin line) and absence (bold line) of 5 µM cyclosporin A. Login to comment 330 ABCB1 p.Lys1061Arg ABCB1 p.Gln1215Asp ABCB1 p.Val1214Ile In constructing this protein, three amino acid changes were introduced: K1061R, V1214I, and Q1215D. Login to comment 335 ABCB1 p.Asp1200Asn As shown in Figure 11, both the wild-type (WT) P-gp and the P-gp-1N/1N molecules were labeled FIGURE 9: Effect of increasing concentrations of MgCl2 on [R-32P]- 8-azido-ATP photoaffinity labeling of wild-type P-gp and P-gp- D1200N in membrane preparations from HeLa cells. Login to comment 336 ABCB1 p.Asp1200Asn Crude membrane preparations (75 µg) from HeLa cells expressing (A) wild-type P-gp and (B) P-gp-D1200N were photoaffinity-labeled with 2.5 µM [R-32P]-8-azido-ATP on ice (0 °C) in the presence of increasing concentrations of MgCl2 (0-10 mM). Login to comment 354 ABCB1 p.Asp555Asn ABCB1 p.Asp1200Asn These mutations at positions D555N and D1200N, made separately or in concert, did not affect the ability of the transporter to bind substrate but completely abrogated transport function and both basal and drug-stimulated ATPase activity. Login to comment 365 ABCB1 p.Asp555Asn ABCB1 p.Asp1200Asn Whereas P-gp-D1200N, the mutant transporter with a C-terminal amino acid substitution, labels comparably to the wild-type protein, P-gp-D555N, containing a homologous N-terminal mutation, is severely impaired in its ability to be labeled with [R-32P]-8-azido-ATP. Login to comment 366 ABCB1 p.Asp555Asn The inability of P-gp-D555N to be labeled efficiently even though the C-half NBD was intact may be because the mutation severely affected the ability of Mg‚ATP to bind to the N-half NBD, and the protein exists in such a conformation that either the C-half NBD site is obscured from nucleotide or the amino acid to which the azido group is to be cross-linked is improperly positioned. Login to comment 367 ABCB1 p.Asp555Asn This hypothesis that the N-half NBD was impaired for labeling in the D555N protein was supported through analysis of the labeling pattern of the two halves of the molecule. Login to comment 368 ABCB1 p.Asp555Asn ABCB1 p.Asp1200Asn After mild trypsinization, it was apparent that the N-half is predominantly labeled in the D1200N and wild-type proteins, whereas little to no labeling of the N-half of the P-gp-D555N was observed. Login to comment 393 ABCB1 p.Asp555Asn ABCB1 p.Asp1200Asn This model is also consistent with the data from the nonfunctional Walker B motif mutants, P-gp-D555N and P-gp-D1200N. Login to comment 395 ABCB1 p.Asp555Asn In the case of P-gp-D555N, only the C-half is intact. Login to comment 397 ABCB1 p.Asp1200Asn In the case of P-gp-D1200N, although the N-half can bind ATP, the C-half is disabled by the mutation and the molecule remains hydrolysis-incompetent. Login to comment