The Transmission Interfaces Contribute Asymmetrically to the Assembly and Activity of Human P-glycoprotein

Abstract

P-glycoprotein (P-gp; ABCB1) is an ABC drug pump that protects us from toxic compounds. It is clinically important because it confers multidrug resistance. The homologous halves of P-gp each contain a transmembrane (TM) domain (TMD) with 6 TM segments followed by a nucleotide-binding domain (NBD). The drug- and ATP-binding sites reside at the interface between the TMDs and NBDs, respectively. Each NBD is connected to the TMDs by a transmission interface involving a pair of intracellular loops (ICLs) that form ball-and-socket joints. P-gp is different from CFTR (ABCC7) in that deleting NBD2 causes misprocessing of only P-gp. Therefore, NBD2 might be critical for stabilizing ICLs 2 and 3 that form a tetrahelix bundle at the NBD2 interface. Here we report that the NBD1 and NBD2 transmission interfaces in P-gp are asymmetric. Point mutations to 25 of 60 ICL2/ICL3 residues at the NBD2 transmission interface severely reduced P-gp assembly while changes to the equivalent residues in ICL1/ICL4 at the NBD1 interface had little effect. The hydrophobic nature at the transmission interfaces was also different. Mutation of Phe-1086 or Tyr-1087 to arginine at the NBD2 socket blocked activity or assembly while the equivalent mutations at the NBD1 socket had only modest effects. The results suggest that the NBD transmission interfaces are asymmetric. In contrast to the ICL2/3-NBD2 interface, the ICL1/4-NBD1 transmission interface is more hydrophilic and insensitive to mutations. Therefore the ICL2/3-NBD2 transmission interface forms a precise hydrophobic connection that acts as a linchpin for assembly and trafficking of P-gp.

Keywords: ABC transporter; membrane enzyme; membrane protein; protein cross-linking; protein folding.

FIGURE 1.

Models of human P-gp. A, secondary structure of human P-gp showing the four ball-and-socket joints of the NBD-TMD transmission interfaces. The intracellular loops (ICLs) containing intracellular helices (IHs) that interact with the NBDs are shown in color. The cylinders represent TM segments, and the branched lines in the loops connecting TM segments 1 and 2 represent glycosylated sites. B, predicted structure of human P-gp in an open conformation was based on the crystal structure of mouse P-gp (56). The intracellular loops interacting with the NBDs are colored. The model was viewed using the PyMol system (57).

FIGURE 2.

Maturation of P-gp is highly sensitive to point mutations at the NBD2 transmission interface (ICL2/ICL3) but not point mutations at the NBD1 transmission interface (ICL1/ICL4). HEK 293 cells were transiently transfected with A-52-tagged P-gps containing point mutations in the ICLs. A, representative immunoblot of SDS extracts of whole cells transfected with wild-type P-gp (WT), control vector (Cont), or P-gp mutants R789A, V907S, Y790A, or V908S showing no maturation (R789A), partial maturation (V907S, Y790A) or maturation similar to wild-type P-gp (V908S). The positions of mature (170 kDa) and immature (150 kDa) forms of P-gp are indicated. The amount of mature 170 kDa P-gp relative to total (mature 170 kDa plus immature 150 kDa protein) was determined for wild-type P-gp (WT) or mutants with point mutations in ICL2 (B), ICL4 (C), ICL3 (D), or ICL1 (E) was determined (Percent Mature). Each value is the mean ± S.D. Residues in the predicted IH segments are indicated.

FIGURE 3.

Rescue of P-gp Processing Mutants with Tariquidar. HEK 293 cells were transiently transfected with A52-tagged wild-type P-gp or ICL processing mutants that yielded immature P-gp as the major product (Fig. 2) and expressed with (+) or without (−) 0.5 μm tariquidar (Tar). A, representative immunoblot of SDS extracts of whole cells transfected with wild-type P-gp (WT) or mutant A250L. The positions of mature (170 kDa) and immature (150 kDa) forms of P-gp are indicated. The amount of mature P-gp relative to total for wild-type P-gp or mutants with processing mutations in ICL2 (B) or ICL3 (C) was determined (Percent Mature). Each value is the mean ± S.D. (n = 3 different transfections).

FIGURE 4.

Point mutations at the NBD2-TMD contact point have a greater impact on P-gp maturation and activity than those at the NBD1-TMD site. A, HEK 293 cells were transiently transfected with A52-tagged wild-type P-gp (WT) or mutants with changes to residues in NBD1 that is adjacent to IH4 (Leu-443, Tyr-444) or to homologous residues in NBD2 (Phe-1086, Tyr-1087). Whole cell SDS extracts were subjected to immunoblot analysis and the amount of mature P-gp relative to total was determined (Percent Mature). Each value is the mean ± S.D. (n = 3). B, histidine-tagged mutants were expressed in the presence of cyclosporine A to promote maturation. P-gps were isolated, mixed with lipid and ATPase activity determined in the presence of verapamil. The results are derived from three different transfections + S.D. (n = 3 different transfections).

FIGURE 5.

IH1/IH4 cross-linking inhibits P-gp ATPase activity. A, model of human P-gp showing parts of the various ICLs. The boxed inset is expanded in B to show the positions of residues I160C(IH1), F904C(IH4), F163C(IH1), and R905C(IH4). The distances (Å) between the α carbons of I160(IH1)/F904(IH4) and F163(IH1)/R905(IH4) are indicated. C, histidine-tagged Cys-less P-gp or mutants containing pairs of cysteines introduced into IH1 and IH4 (I160C(IH1)/F904C(IH4), F163C(IH1)/R905C(IH4)) (in Cys-less background) were transiently expressed in HEK 293 cells and isolated by nickel-chelate chromatography. Samples of the isolated P-gps were treated without (−) or with (+) copper phenanthroline (CuP). The reactions were stopped by addition of EDTA. Samples from the mutant P-gps were subjected to immunoblot analysis before (−) or after (+) treatment with dithiothreitol (DTT). The positions of cross-linked (X-link), mature (170 kDa), and immature (150 kDa) P-gps are indicated. D, samples were also assayed for ATPase activity in the presence of verapamil. Each value is the mean ± S.D. (n = 3).

FIGURE 6.

Double mutations introduced into equivalent positions in IH2 and IH4 caused modest reductions in ATPase activity. Histidine-tagged wild-type P-gp (WT) or mutants containing point mutations to equivalent positions in IH2 and IH4 (R262A(IH2)/R905A(IH4), T263A(IH2)/T906A(IH4)), or NBD1 and NBD2 (Q475A(NBD1)/Q1118A(NBD2)) were expressed in HEK 293 cells, isolated by nickel-chelate chromatography and assayed for ATPase activity in the presence of verapamil. The results are derived from the mean of three different transfections + S.D.

FIGURE 7.

P-gp maturation is highly sensitive to point mutations at the second transmission interface but relatively insensitive to changes at the first transmission interface. Predicted structure of human P-gp in an open conformation (based on the crystal structure of mouse P-gp) (56). Mutation of residues in ICL2 and ICL3 inhibited maturation of P-gp. The side chains of these residues are shown in blue and red, respectively. None of the mutations in ICLs 1 and 4 inhibited maturation of P-gp and their side chains are shown in green.