The linker region of breast cancer resistance protein ABCG2 is critical for coupling of ATP-dependent drug transport

The ATP-binding cassette (ABC) transporters of class G display a different domain organisation than P-glycoprotein/ABCB1 and bacterial homologues with a nucleotide-binding domain preceding the transmembrane domain. The linker region connecting these domains is unique and its function and structure cannot be predicted. Sequence analysis revealed that the human ABCG2 linker contains a LSGGE sequence, homologous to the canonical C-motif/ABC signature present in all ABC nucleotide-binding domains. Predictions of disorder and of secondary structures indicated that this C2-sequence was highly mobile and located between an α-helix and a loop similarly to the C-motif. Point mutations of the two first residues of the C2-sequence fully abolished the transport-coupled ATPase activity, and led to the complete loss of cell resistance to mitoxantrone. The interaction with potent, selective and non-competitive, ABCG2 inhibitors was also significantly altered upon mutation. These results suggest an important mechanistic role for the C2-sequence of the ABCG2 linker region in ATP binding and/or hydrolysis coupled to drug efflux.

Improving cancer chemotherapy with modulators of ABC drug transporters

ATP-binding cassette (ABC) transporters, P-glycoprotein (P-gp, ABCB1) and ABCG2, are membrane proteins that couple the energy derived from ATP hydrolysis to efflux many chemically diverse compounds across the plasma membrane, thereby playing a critical and important physiological role in protecting cells from xenobiotics. These transporters are also implicated in the development of multidrug resistance (MDR) in cancer cells that have been treated with chemotherapeutics. One approach to blocking the efflux capability of an ABC transporter in a cell or tissue is inhibiting the activity of the transporters with a modulator. Since ABC transporter modulators can be used in combination with chemotherapeutics to increase the effective intracellular concentration of anticancer drugs, the possible impact of modulators of ABC drug transporters is of great clinical interest. Another possible clinical use of modulators that has recently attracted attention is their ability to increase oral bioavailability or increase tissue penetration of drugs transported by the transporters. Several preclinical and clinical studies have been performed to evaluate the feasibility and the safety of this approach. The primary focus of this review is to discuss progress made in recent years in the identification and applicability of compounds that may serve as ABC transporter modulators and the possible role of these compounds in altering the pharmacokinetics and pharmacodynamics of therapeutic drugs used in the clinic.

Single-step doxorubicin-selected cancer cells overexpress the ABCG2 drug transporter through epigenetic changes

Understanding the mechanisms of multidrug resistance (MDR) could improve clinical drug efficacy. Multidrug resistance is associated with ATP binding cassette (ABC) transporters, but the factors that regulate their expression at clinically relevant drug concentrations are poorly understood. We report that a single-step selection with low doses of anti-cancer agents, similar to concentrations reported in vivo, induces MDR that is mediated exclusively by ABCG2. We selected breast, ovarian and colon cancer cells (MCF-7, IGROV-1 and S-1) after exposure to 14 or 21 nM doxorubicin for only 10 days. We found that these cells overexpress ABCG2 at the mRNA and protein levels. RNA interference analysis confirmed that ABCG2 confers drug resistance. Furthermore, ABCG2 upregulation was facilitated by histone hyperacetylation due to weaker histone deacetylase 1-promoter association, indicating that these epigenetic changes elicit changes in ABCG2 gene expression. These studies indicate that the MDR phenotype arises following low-dose, single-step exposure to doxorubicin, and further suggest that ABCG2 may mediate early stages of MDR development. This is the first report to our knowledge of single-step, low-dose selection leading to overexpression of ABCG2 by epigenetic changes in multiple cancer cell lines.

Imatinib mesylate and nilotinib (AMN107) exhibit high-affinity interaction with ABCG2 on primitive hematopoietic stem cells

The majority of chronic phase chronic myeloid leukemia (CML) patients treated with the tyrosine kinase inhibitor (TKI) imatinib mesylate maintain durable responses to the drug. However, most patients relapse after withdrawal of imatinib and advanced stage patients often develop drug resistance. As CML is considered a hematopoietic stem cell cancer, it has been postulated that inherent protective mechanisms lead to relapse in patients. The ATP binding-cassette transporters ABCB1 (MDR-1; P-glycoprotein) and ABCG2 are highly expressed on primitive hematopoietic stem cells (HSCs) and have been shown to interact with TKIs. Herein we demonstrate a dose-dependent, reversible inhibition of ABCG2-mediated Hoechst 33342 dye efflux in primary human and murine HSC by both imatinib and nilotinib (AMN107), a novel aminopyrimidine inhibitor of BCR-ABL. ABCG2-transduced K562 cells were protected from imatinib and nilotinib-mediated cell death and from downregulation of P-CRKL. Moreover, photoaffinity labeling revealed interaction of both TKIs with ABCG2 at the substrate binding sites as they compete with the binding of [(125)I] IAAP and also stimulate the transporter's ATPase activity. Therefore, our evidence suggests for the role of ABC transporters in resistance to TKI on primitive HSCs and CML stem cells and provides a rationale how TKI resistance can be overcome in vivo.

Biochemical mechanism of modulation of human P-glycoprotein (ABCB1) by curcumin I, II, and III purified from Turmeric powder

P-glycoprotein (Pgp, ABCB1) is an ATP-dependent drug efflux pump linked to development of multidrug resistance (MDR) in cancer cells. Previously [Biochem Pharmacol 2002;64:573-82], we reported that a curcumin mixture could modulate both function and expression of Pgp. This study focuses on the effect of three major curcuminoids--curcumin I, II and III purified from a curcumin mixture--on modulation of Pgp function in a multidrug resistant human cervical carcinoma cell line (KB-V1). The similar IC(50) values for cytotoxicity of curcuminoids of KB-V1, and KB-3-1 (parental drug sensitive cell line) suggest that these curcuminoids may not be substrates for Pgp. Treating the cells with non-toxic doses of curcuminoids increased their sensitivity to vinblastine only in the Pgp expressing drug resistant cell line, KB-V1, and curcumin I retained the drug in KB-V1 cells more effectively than curcumin II and III, respectively. Effects of each curcuminoid on rhodamine123, calcein-AM, and bodipy-FL-vinblastine accumulation confirmed these findings. Curcumin I, II and III increased the accumulation of fluorescent substrates in a dose-dependent manner, and at 15 microM, curcumin I was the most effective. The inhibitory effect in a concentration-dependent manner of curcuminoids on verapamil-stimulated ATPase activity and photoaffinity labeling of Pgp with the [(125)I]-iodoarylazidoprazosin offered additional support; curcumin I was the most potent modulator. Taken together, these results indicate that curcumin I is the most effective MDR modulator among curcuminoids, and may be used in combination with conventional chemotherapeutic drugs to reverse MDR in cancer cells.

The mechanism of action of multidrug-resistance-linked P-glycoprotein

P-glycoprotein (Pgp), the ATP-binding cassette (ABC) transporter, confers multidrug resistance to cancer cells by extruding cytotoxic natural product amphipathic drugs using the energy of ATP hydrolysis. Our studies are directed toward understanding the mechanism of action of Pgp and recent work deals with the assessment of interaction between substrate and ATP sites and elucidation of the catalytic cycle of ATP hydrolysis. The kinetic analyses of ATP hydrolysis by reconstituted purified Pgp suggest that ADP release is the rate-limiting step in the catalytic cycle and the substrates exert their effect by modulating ADP release. In addition, we provide evidence for two distinct roles for ATP hydrolysis in a single turnover of Pgp, one in the transport of drug and the other in effecting conformational changes so as to reset the transporter for the next catalytic cycle. Detailed kinetic measurements determined that both nucleotide-binding domains behave symmetrically and during individual hydrolysis events the ATP sites are recruited in a random manner. Furthermore, only one nucleotide site hydrolyzes ATP at any given time, causing (in this site) a conformational change that drastically decreases (>30-fold) the affinity of the second site for ATP-binding. Thus, the blocking of ATP-binding to the second site while the first one is in catalytic conformation appears to be the basis for the alternate catalytic cycle of ATP hydrolysis by Pgp, and this may be applicable as well to other ABC transporters linked with the development of multidrug resistance.

Overview: ABC transporters and human disease

ABC transporters are found in all known organisms, and approximately 1,100 different transporters belonging to this family have been described in the literature. The family is defined by homology within the ATP-binding cassette (ABC) region, which extends outside of the more typical Walker motifs found in all ATP-binding proteins. Most family members also contain transmembrane domains involved in recognition of substrates, which are transported across, into, and out of cell membranes, but some members utilize ABCs as engines to regulate ion channels. There are approximately 50 known ABC transporters in the human, and there are currently 13 genetic diseases associated with defects in 14 of these transporters. The most common genetic disease conditions include cystic fibrosis, Stargardt disease, age-related macular degeneration, adrenoleukodystrophy, Tangier disease, Dubin-Johnson syndrome and progressive familial intrahepatic cholestasis. At least 8 members of this family are involved in the transport of a variety of amphipathic compounds, including anticancer drugs, and some appear to contribute to the resistance of cancer cells to chemotherapy.

Evidence for the vectorial nature of drug (substrate)-stimulated ATP hydrolysis by human P-glycoprotein

P-glycoprotein (Pgp), the ATP-binding cassette multidrug transporter, exhibits a drug (substrate)-stimulatable ATPase activity, and vanadate (Vi) inhibits this activity by stably trapping the nucleoside diphosphate in the Pgp.ADP.Vi conformation. We recently demonstrated that Vi-induced 8-azido-[alpha-(32)P]ADP trapping into Pgp in the absence of substrate occurs both in the presence of 8-azido-[alpha-(32)P]ATP (following 8-azido-ATP hydrolysis) or 8-azido-[alpha-(32)P]ADP (without hydrolysis) and, the transition state intermediates generated under either condition are functionally indistinguishable. In this study, we compare the effect of substrates on Vi-induced 8-azido-[alpha-(32)P]ADP trapping into Pgp under both non-hydrolysis and hydrolysis conditions. We demonstrate that whereas substrates stimulate the Vi-induced trapping of 8-azido-[alpha-(32)P]ADP under hydrolysis conditions, they strongly inhibit Vi-induced trapping under non-hydrolysis conditions. This inhibition is concentration-dependent, follows first order kinetics, and is effected by drastically decreasing the affinity of nucleoside diphosphate for Pgp during trapping. However, substrates do not affect the binding of nucleoside diphosphate in the absence of Vi, indicating that the substrate-induced conformation exerts its effect at a step distinct from nucleoside diphosphate-binding. Our results demonstrate that during the catalytic cycle of Pgp, although the transition state, Pgp x ADP x P(i) (Vi), can be generated both via the hydrolysis of ATP or by directly providing ADP to the system, in the presence of substrate the reaction is driven in the forward direction, i.e. hydrolysis of ATP. These data suggest that substrate-stimulated ATP hydrolysis by Pgp is a vectorial process.

Functionally similar vanadate-induced 8-azidoadenosine 5'-[alpha-(32)P]Diphosphate-trapped transition state intermediates of human P-glycoprotin are generated in the absence and presence of ATP hydrolysis

P-glycoprotein (Pgp) is an ATP-dependent drug efflux pump whose overexpression confers multidrug resistance to cancer cells. Pgp exhibits a robust drug substrate-stimulable ATPase activity, and vanadate (Vi) blocks this activity effectively by trapping Pgp nucleotide in a non-covalent stable transition state conformation. In this study we compare Vi-induced [alpha-(32)P]8-azido-ADP trapping into Pgp in the presence of [alpha-(32)P]8-azido-ATP (with ATP hydrolysis) or [alpha-(32)P]8-azido-ADP (without ATP hydrolysis). Vi mimics P(i) to trap the nucleotide tenaciously in the Pgp.[alpha-(32)P]8-azido-ADP.Vi conformation in either condition. Thus, by using [alpha-(32)P]8-azido-ADP we show that the Vi-induced transition state of Pgp can be generated even in the absence of ATP hydrolysis. Furthermore, half-maximal trapping of nucleotide into Pgp in the presence of Vi occurs at similar concentrations of [alpha-(32)P]8-azido-ATP or [alpha-(32)P]8-azido-ADP. The trapped [alpha-(32)P]8-azido-ADP is almost equally distributed between the N- and the C-terminal ATP sites of Pgp in both conditions. Additionally, point mutations in the Walker B domain of either the N- (D555N) or C (D1200N)-terminal ATP sites that arrest ATP hydrolysis and Vi-induced trapping also show abrogation of [alpha-(32)P]8-azido-ADP trapping into Pgp in the absence of hydrolysis. These data suggest that both ATP sites are dependent on each other for function and that each site exhibits similar affinity for 8-azido-ATP (ATP) or 8-azido-ADP (ADP). Similarly, Pgp in the transition state conformation generated with either ADP or ATP exhibits drastically reduced affinity for the binding of analogues of drug substrate ([(125)I]iodoarylazidoprazosin) as well as nucleotide (2'(3')-O-(2,4,6-trinitrophenyl)adenosine 5'-triphosphate). Analyses of Arrhenius plots show that trapping of Pgp with [alpha-(32)P]8-azido-ADP (in the absence of hydrolysis) displays an approximately 2.5-fold higher energy of activation (152 kJ/mol) compared with that observed when the transition state intermediate is generated through hydrolysis of [alpha-(32)P]8-azido-ATP (62 kJ/mol). In aggregate, these results demonstrate that the Pgp.[alpha-(32)P]8-azido-ADP (or ADP).Vi transition state complexes generated either in the absence of or accompanying [alpha-(32)P]8-azido-ATP hydrolysis are functionally indistinguishable.

Characterization of the catalytic cycle of ATP hydrolysis by human P-glycoprotein. The two ATP hydrolysis events in a single catalytic cycle are kinetically similar but affect different functional outcomes

P-glycoprotein (Pgp) is a plasma membrane protein whose overexpression confers multidrug resistance to tumor cells by extruding amphipathic natural product cytotoxic drugs using the energy of ATP. An elucidation of the catalytic cycle of Pgp would help design rational strategies to combat multidrug resistance and to further our understanding of the mechanism of ATP-binding cassette transporters. We have recently reported (Sauna, Z. E., and Ambudkar, S. V. (2000) Proc. Natl. Acad. Sci. U. S. A. 97, 2515-2520) that there are two independent ATP hydrolysis events in a single catalytic cycle of Pgp. In this study we exploit the vanadate (Vi)-induced transition state conformation of Pgp (Pgp.ADP.Vi) to address the question of what are the effects of ATP hydrolysis on the nucleotide-binding site. We find that at the end of the first hydrolysis event there is a drastic decrease in the affinity of nucleotide for Pgp coincident with decreased substrate binding. Release of occluded dinucleotide is adequate for the next hydrolysis event to occur but is not sufficient for the recovery of substrate binding. Whereas the two hydrolysis events have different functional outcomes vis à vis the substrate, they show comparable t(12) for both incorporation and release of nucleotide, and the affinities for [alpha-(32)P]8-azido-ATP during Vi-induced trapping are identical. In addition, the incorporation of [alpha-(32)P]8-azido-ADP in two ATP sites during both hydrolysis events is also similar. These data demonstrate that during individual hydrolysis events, the ATP sites are recruited in a random manner, and only one site is utilized at any given time because of the conformational change in the catalytic site that drastically reduces the affinity of the second ATP site for nucleotide binding. In aggregate, these findings provide an explanation for the alternate catalysis of ATP hydrolysis and offer a mechanistic framework to elucidate events at both the substrate- and nucleotide-binding sites in the catalytic cycle of Pgp.

Correlation between steady-state ATP hydrolysis and vanadate-induced ADP trapping in Human P-glycoprotein. Evidence for ADP release as the rate-limiting step in the catalytic cycle and its modulation by substrates

P-glycoprotein (Pgp) is a transmembrane protein conferring multidrug resistance to cells by extruding a variety of amphipathic cytotoxic agents using energy from ATP hydrolysis. The objective of this study was to understand how substrates affect the catalytic cycle of ATP hydrolysis by Pgp. The ATPase activity of purified and reconstituted recombinant human Pgp was measured using a continuous cycling assay. Pgp hydrolyzes ATP in the absence of drug at a basal rate of 0.5 micromol x min x mg(-1) with a K(m) for ATP of 0.33 mm. This basal rate can be either increased or decreased depending on the Pgp substrate used, without an effect on the K(m) for ATP or 8-azidoATP and K(i) for ADP, suggesting that substrates do not affect nucleotide binding to Pgp. Although inhibitors of Pgp activity, cyclosporin A, its analog PSC833, and rapamycin decrease the rate of ATP hydrolysis with respect to the basal rate, they do not completely inhibit the activity. Therefore, these drugs can be classified as substrates. Vanadate (Vi)-induced trapping of [alpha-(32)P]8-azidoADP was used to probe the effect of substrates on the transition state of the ATP hydrolysis reaction. The K(m) for [alpha-(32)P]8-azidoATP (20 microm) is decreased in the presence of Vi; however, it is not changed by drugs such as verapamil or cyclosporin A. Strikingly, the extent of Vi-induced [alpha-(32)P]8-azidoADP trapping correlates directly with the fold stimulation of ATPase activity at steady state. Furthermore, P(i) exhibits very low affinity for Pgp (K(i) approximately 30 mm for Vi-induced 8-azidoADP trapping). In aggregate, these data demonstrate that the release of Vi trapped [alpha-(32)P]8-azidoADP from Pgp is the rate-limiting step in the steady-state reaction. We suggest that substrates modulate the rate of ATPase activity of Pgp by controlling the rate of dissociation of ADP following ATP hydrolysis and that ADP release is the rate-limiting step in the normal catalytic cycle of Pgp.

Synthesis and preliminary analysis of a P-glycoprotein-specific [3H]-benzophenone photoaffinity label based on (-)-stipiamide

A benzophenone photoaffinity label 9 based on the polyene natural product (-)-stipiamide has been constructed using a diaminoethane spacer and the radioactive agent [3H]-BZDC (N-succinimidyl p-benzoyl-(2,3-3H)-dehydrocinnamate). Photoaffinity experiments show specific binding to human P-glycoprotein (Pgp) in the presence of cis-flupentixol but not with cyclosporin A.

The synthesis and evaluation of a solution phase indexed combinatorial library of non-natural polyenes for reversal of P-glycoprotein mediated multidrug resistance

A combinatorial library of polyenes, based on (-)-stipiamide, has been constructed and evaluated for the discovery of new multidrug resistance reversal agents. A palladium coupling was used to react each individual vinyl iodide with a mixture of the seven acetylenes at near 1:1 stoichiometry. The coupling was also used to react each individual acetylene with the mixture of six vinyl iodides to create 13 pools indexed in two dimensions for a total of 42 compounds. Individual compounds were detected at equimolar concentration. The vinyl iodides, made initially using a crotylborane addition to generate the anti1,2-hydroxylmethyl products, were now made using a more efficient norephedrine propionate boron enolate aldol reaction. The indexed approach, ideally suited for cellular assays that involve membrane-bound targets, allowed for the rapid identification of reversal agents using assays with drug-resistant human breast cancer MCF7-adrR cells. Intersections of potent pools identified new compounds with promising activity. Aryl dimension pools showed R = ph and naphthyl as the most potent. The acetylene dimension had R' = phenylalaninol and alaninol as the most potent. Isolated individual compounds, both active and nonpotent, were assayed to confirm the library results. The most potent new compound was 4ek (R = naphthyl, R' = phenylaninol) at 1.45 microM. Other nonnatural individual naphthyl-amide compounds showed potent MDR reversal including the morpholino-amide 4ej (1.69 microM). Synergistic activities attributed to the two ends of the molecule were also identified. Direct interaction with Pgp was established by ATPase and photoaffinity displacement assays. The results indicate that both ends of the polyene reversal agent are involved in Pgp interaction and can be further modified for increased potency.

Evidence for a requirement for ATP hydrolysis at two distinct steps during a single turnover of the catalytic cycle of human P-glycoprotein

P-glycoprotein (Pgp) is an ATP-dependent hydrophobic natural product anticancer drug efflux pump whose overexpression confers multidrug resistance to tumor cells. The work reported here deals with the elucidation of the energy requirement for substrate interaction with Pgp during the catalytic cycle. We show that the K(d) (412 nM) of the substrate analogue [(125)I]iodoarylazidoprazoin for Pgp is not altered by the presence of the nonhydrolyzable nucleotide 5'-adenylylimididiphosphate and vanadate (K(d) = 403 nM). Though binding of nucleotide per se does not affect interactions with the substrate, ATP hydrolysis results in a dramatic conformational change where the affinity of [(125)I]iodoarylazidoprazoin for Pgp trapped in transition-state conformation (Pgp x ADP x vanadate) is reduced >30-fold. To transform Pgp from this intermediate state of low affinity for substrate to the next catalytic cycle, i.e., a conformation that binds substrate with high affinity, requires conditions that permit ATP hydrolysis. Additionally, there is an inverse correlation (R(2) = 0.96) between 8AzidoADP (or ADP) release and the recovery of substrate binding. These results suggest that the release of nucleotide is necessary for reactivation but not sufficient. The hydrolysis of additional molecule(s) of ATP (or 8AzidoATP) is obligatory for the catalytic cycle to advance to completion. These data are consistent with the observed stoichiometry of two ATP molecules hydrolyzed for the transport of every substrate molecule. Our data demonstrate two distinct roles for ATP hydrolysis in a single turnover of the catalytic cycle of Pgp, one in the transport of substrate and the other in effecting conformational changes to reset the pump for the next catalytic cycle.

Functional characterization of glycosylation-deficient human P-glycoprotein using a vaccinia virus expression system

P-glycoprotein (P-gp), the product of human MDR1 gene, which functions as an ATP-dependent drug efflux pump, is N-linked glycosylated at asparagine residues 91, 94, and 99 located within the first extracellular loop. We report here the biochemical characterization of glycosylation-deficient (Gly(-)) P-gp using a vaccinia virus based transient expression system. The staining of HeLa cells expressing Gly(-) P-gp (91, 94, and 99N-->Q), with P-gp specific monoclonal antibodies, MRK-16, UIC2 and 4E3 revealed a 40 to 50% lower cell-surface expression of mutant P-gp compared to the wild-type protein. The transport function of Gly(-) P-gp, assessed using a variety of fluorescent compounds indicated that the substrate specificity of the pump was not affected by the lack of glycosylation. Additional mutants, Gly(-) D (91, 94, 99N-->D) and Gly(-) Delta (91, 94, 99 N deleted) were generated to verify that the reduced cell surface expression, as well as total expression, were not a result of the glutamine substitutions. Gly(-) D and Gly(-) Delta Pgps were also expressed to the same level as the Gly(-) mutant protein. (35)S-Methionine/cysteine pulse-chase studies revealed a reduced incorporation of (35)S-methionine/cysteine in full length Gly(-) P-gp compared to wild-type protein, but the half-life ( approximately 3 hr) of mutant P-gp was essentially unaltered. Since treatment with proteasome inhibitors (MG-132, lactacystin) increased only the intracellular level of nascent, mutant P-gp, the decreased incorporation of (35)S-methionine/cysteine in Gly(-) P-gp appears to be due to degradation of improperly folded mutant protein by the proteasome and endoplasmic reticulum-associated proteases. These results demonstrate that the unglycosylated protein, although expressed at lower levels at the cell surface, is functional and suitable for structural studies.

Biochemical, cellular, and pharmacological aspects of the multidrug transporter

Considerable evidence has accumulated indicating that the multidrug transporter or P-glycoprotein plays a role in the development of simultaneous resistance to multiple cytotoxic drugs in cancer cells. In recent years, various approaches such as mutational analyses and biochemical and pharmacological characterization have yielded significant information about the relationship of structure and function of P-glycoprotein. However, there is still considerable controversy about the mechanism of action of this efflux pump and its function in normal cells. This review summarizes current research on the structure-function analysis of P-glycoprotein, its mechanism of action, and facts and speculations about its normal physiological role.

Contribution to substrate specificity and transport of nonconserved residues in transmembrane domain 12 of human P-glycoprotein

P-glycoprotein (Pgp), the product of the MDR1 gene, confers multidrug resistance on cancer cells by ATP-dependent extrusion of anticancer drugs. Biochemical and genetic studies with Pgp have identified the putative transmembrane (TM) region 12 (residues 974-994) as a major region involved in drug interactions with amino acid residues conserved among Pgp family members shown to be essential for transport. To determine whether nonconserved residues might be involved in substrate specificity, seven amino acid residues were identified within TM 12 that were not strictly conserved among the MDR1 and MDR2 family of proteins from different mammalian species. We replaced all seven of these amino acid residues with alanine, one at a time and in combinations, and used a vaccinia virus based transient expression system to analyze function. None of the single replacements caused any alteration in transport function. However, when residues L975, V981, and F983 were replaced collectively, drug transport, drug-stimulated ATP hydrolysis, and photoaffinity labeling with the drug analogue, [125I]iodoarylazidoprazosin (IAAP), were abrogated, with little effect on [alpha-32P]-8-azido-ATP labeling and basal ATPase activity. Pairwise alanine substitutuions showed variable effects on function. Substitutions including L975A in combination with any one of the other two replacements had the least effect on Pgp function. The V981A and F983A double mutant showed the most effect on transport of fluorescent substrates. In contrast, alanine substitutions of all four nonconserved residues M986, V988, Q990, and V991 at the putative carboxy-terminal half of TM 12 showed no effect on drug transport except for a partial reduction in bodipy-verapamil extrusion. These results suggest that nonconserved residues in the putative amino-proximal half of TM 12 of Pgp play a more direct role in determining specificity of drug transport function than those in the putative carboxy-terminal half of TM 12.

Contribution to substrate specificity and transport of nonconserved residues in transmembrane domain 12 of human P-glycoprotein

P-glycoprotein (Pgp), the product of the MDR1 gene, confers multidrug resistance on cancer cells by ATP-dependent extrusion of anticancer drugs. Biochemical and genetic studies with Pgp have identified the putative transmembrane (TM) region 12 (residues 974-994) as a major region involved in drug interactions with amino acid residues conserved among Pgp family members shown to be essential for transport. To determine whether nonconserved residues might be involved in substrate specificity, seven amino acid residues were identified within TM 12 that were not strictly conserved among the MDR1 and MDR2 family of proteins from different mammalian species. We replaced all seven of these amino acid residues with alanine, one at a time and in combinations, and used a vaccinia virus based transient expression system to analyze function. None of the single replacements caused any alteration in transport function. However, when residues L975, V981, and F983 were replaced collectively, drug transport, drug-stimulated ATP hydrolysis, and photoaffinity labeling with the drug analogue, [125I]iodoarylazidoprazosin (IAAP), were abrogated, with little effect on [alpha-32P]-8-azido-ATP labeling and basal ATPase activity. Pairwise alanine substitutuions showed variable effects on function. Substitutions including L975A in combination with any one of the other two replacements had the least effect on Pgp function. The V981A and F983A double mutant showed the most effect on transport of fluorescent substrates. In contrast, alanine substitutions of all four nonconserved residues M986, V988, Q990, and V991 at the putative carboxy-terminal half of TM 12 showed no effect on drug transport except for a partial reduction in bodipy-verapamil extrusion. These results suggest that nonconserved residues in the putative amino-proximal half of TM 12 of Pgp play a more direct role in determining specificity of drug transport function than those in the putative carboxy-terminal half of TM 12.

Structural flexibility of the linker region of human P-glycoprotein permits ATP hydrolysis and drug transport

P-Glycoprotein (Pgp), an energy-dependent drug efflux pump responsible for multidrug resistance of many cancer cells, is comprised of two homologous halves connected by a peptide segment approximately 75 amino acids (aa) in length. The effects of length and composition of this connecting region on Pgp cell surface expression and the ability of the two halves to interact were explored using both stable transfections of Pgp mutants in mammalian cell lines and a vaccinia virus transient expression system. A 17 aa insertion of predicted flexible structure between amino acids 681 and 682 resulted in a functional Pgp molecule that was capable of conferring drug resistance. In contrast, an 18 aa peptide insertion with a predicted alpha-helical structure was unstable when expressed transiently. A 34 aa deletion from the central core of the linker region (Delta653-686) resulted in a protein expressed at the cell surface in amounts comparable to that of wild-type Pgp but unable to confer drug resistance. No apparent differences in drug or [alpha-32P]-8-azido-ATP photoaffinity labeling were observed. However, both ATP hydrolysis and drug transport activities of the deletion mutant were completely abrogated, indicating that the linker deletion disconnected substrate binding from ATP hydrolysis and transport. This mutant also failed to exhibit an ATP hydrolysis-dependent enhancement of binding of a conformation-sensitive monoclonal antibody, UIC2. Upon replacement with a 17 aa linker peptide having a predicted flexible secondary structure, but bearing no homology to the deleted 34 aa segment, normal Pgp transport and basal and drug-stimulated ATPase activities were restored along with increased UIC2 binding in the presence of substrate, suggesting a dramatic conformational change between the nonfunctional and functional molecules. Taken together, these data suggest a flexible secondary structure of the connector region is sufficient for the coordinate functioning of the two halves of Pgp, likely specifically required for the proper interaction of the two ATP binding sites.

Purification and reconstitution of human P-glycoprotein

Human Pgp from the vinblastine-resistant cell line, KB-V1, can be purified by sequential conventional chromatography on DEAE-sepharose CL-6B resin followed by a wheat germ agglutinin column. By including glycerol (osmolyte protectant) and lipid during the solubilization and chromatography procedures most of the biological activity of Pgp can be retained. The activity of Pgp in the detergent extract or in the concentrated column fractions is stable for at least 8-10 months when stored at -80 degrees. However, repeated cycles of freezing and thawing of fractions result in considerable loss of activity. We have purified Pgp from KB-C1 (a subclone of KB 3-1 that is resistant to 1 microgram/ml colchicine) by following the same protocol. When this method was used for purification of Pgp from MDR1-transfected NIH 3T3 transfectants (N3-V2400, grown in the presence of 2.4 micrograms/ml vinblastine), the protein was eluted with 0.1 M NaCl from the DEAE-Sepharose CL-6B column as usual. However, during WGA lectin chromatography, the protein was eluted with a lower concentration of sugar (0.1 M instead of 0.25 M NAG). This altered elution pattern appears to be due to a difference in the glycosylation of human Pgp in mouse NIH 3T3 cells. This is consistent with the observation that human Pgp expressed in NIH 3T3 cells migrates faster compared to the protein from KB-V1 cells on 8-10% acrylamide gel. Similarly, other workers have purified Chinese hamster Pgp either by a single-step chromatography on Reactive Red 120 agarose or by a combination of anion exchange and immunoaffinity chromatography (see the article by Senior et al. for the purification and properties of ATPase activity of Chinese hamster Pgp). The high level of drug-stimulated ATP hydrolysis by Pgp (Table I), like other ion-transporting ATPases, indicates that this is a high-capacity pump that can function as an effective multidrug transporter. This is further supported by the qualitative demonstration of ATP-dependent vinblastine transport in proteoliposomes reconstituted with pure Pgp (see Fig. 2). Thus, these experiments provide strong evidence that purified Pgp retains its activity and that it functions as an ATP-dependent drug transporter.

Mechanism of action of human P-glycoprotein ATPase activity. Photochemical cleavage during a catalytic transition state using orthovanadate reveals cross-talk between the two ATP sites

Human P-glycoprotein (P-gp), an ATP-dependent efflux pump responsible for cross-resistance of human cancers to a variety of lipophilic compounds, is composed of two homologous halves, each containing six transmembrane domains and an ATP-binding/utilization domain. To determine whether each site can hydrolyze ATP simultaneously, we used an orthovanadate (Vi)-induced ADP-trapping technique (P-gp.MgADP.Vi). In analogy with other ATPases, a photochemical peptide bond cleavage reaction occurs within the Walker A nucleotide binding domain consensus sequence (GX4GK(T/S)) when the molecule is trapped with Vi in an inhibited catalytic transition state (P-gp.MgADP.Vi) and incubated in the presence of ultraviolet light. Upon reconstitution into proteoliposomes, histidine-tagged purified P-gp from baculovirus-infected insect cells had drug-stimulated ATPase activity. Reconstituted P-gp was incubated with either ATP or 8-azido-ATP in the presence or absence of Vi under ultraviolet (365 nm) light on ice for 60 min. The resultant products were separated by SDS-polyacrylamide gel electrophoresis and subjected to immunoblotting with seven different human P-gp-specific antibodies covering the entire length of the molecule. Little to no degradation of P-gp was observed in the absence of Vi. In the presence of Vi, products of approximately 28, 47, 94, and 110 kDa were obtained, consistent with predicted molecular weights from cleavage at either of the ATP sites but not both sites. An additional Vi-dependent cleavage site was detected at or near the trypsin site in the linker region of P-gp. These results suggest that both the amino- and carboxyl-terminal ATP sites can hydrolyze ATP. However, there is no evidence that ATP can be hydrolyzed simultaneously by both sites.

Human P-glycoprotein exhibits reduced affinity for substrates during a catalytic transition state

Human P-glycoprotein (Pgp), a plasma membrane protein that confers multidrug resistance, functions as an ATP-dependent drug efflux pump. Pgp contains two ATP binding/utilization sites and exhibits ATPase activity that is stimulated in the presence of substrates and modulating agents. The mechanism of coupling of ATP hydrolysis to drug transport is not known. To understand the role of ATP hydrolysis in drug binding, it is necessary to develop methods for purifying and reconstituting Pgp that retains properties including stimulation of ATPase activity by known substrates to an extent similar to that in the native membrane. In this study, (His)6-tagged Pgp was expressed in Trichoplusia ni (High Five) cells using the recombinant baculovirus system and purified by metal affinity chromatography. Upon reconstitution into phospholipid vesicles, purified Pgp exhibited specific binding to analogues of substrates and ATP in affinity labeling experiments and displayed a high level of drug-stimulated ATPase activity (specific activity ranging from 4.5 to 6.5 micromol min-1 mg-1). The ATPase activity was inhibited by ADP in a competitive manner, and by vanadate and N-ethylmaleimide at low concentrations. Vanadate which is known to inhibit ATPase activity by trapping MgADP at the catalytic site inhibited photoaffinity labeling of Pgp with substrate analogues, [125I]iodoarylazidoprazosin and [3H]azidopine, only under ATP hydrolysis conditions. Because vanadate-trapped Pgp is known to resemble the ADP and phosphate-bound catalytic transition state, our findings indicate that ATP hydrolysis results in a conformation with reduced affinity for substrates. A catalytic transition conformation with reduced affinity would essentially result in substrate dissociation and supports a model for drug transport in which an ATP hydrolysis-induced conformational change leads to drug release toward the extracellular medium.

HIV-1 protease inhibitors are substrates for the MDR1 multidrug transporter

The FDA approved HIV-1 protease inhibitors, ritonavir, saquinavir, and indinavir, are very effective in inhibiting HIV-1 replication, but their long-term efficacy is unknown. Since in vivo efficacy depends on access of these drugs to intracellular sites where HIV-1 replicates, we determined whether these protease inhibitors are recognized by the MDR1 multidrug transporter (P-glycoprotein, or P-gp), thereby reducing their intracellular accumulation. In vitro studies in isolated membrane preparations from insect cells infected with MDR1-expressing recombinant baculovirus showed that these inhibitors significantly stimulated P-gp-specific ATPase activity and that this stimulation was inhibited by SDZ PSC 833, a potent inhibitor of P-gp. Furthermore, photoaffinity labeling of P-gp with the substrate analogue [125I]iodoarylazidoprazosin (IAAP) was inhibited by all three inhibitors. Cell-based approaches to evaluate the ability of these protease inhibitors to compete for transport of known P-gp substrates showed that all three HIV-1 protease inhibitors were capable of inhibiting the transport of some of the known P-gp substrates but their effects were generally weaker than other documented P-gp modulators such as verapamil or cyclosporin A. Inhibition of HIV-1 replication by all three protease inhibitors was reduced but could be restored by MDR1 inhibitors in cells expressing MDR1. These results indicate that the HIV-1 protease inhibitors are substrates of the human multidrug transporter, suggesting that cells in patients that express the MDR1 transporter will be relatively resistant to the anti-viral effects of the HIV-1 protease inhibitors, and that absorption, excretion, and distribution of these inhibitors in the body may be affected by the multidrug transporter.

Evidence for two nonidentical drug-interaction sites in the human P-glycoprotein

Human P-glycoprotein (Pgp) confers multidrug resistance to cancer cells by ATP-dependent extrusion of a great many structurally dissimilar hydrophobic compounds. The manner in which Pgp recognizes these different substrates is unknown. The protein shows internal homology between its N- and C-terminal halves, each comprised of six putative transmembrane helices and a consensus ATP binding/utilization site. Photoactive derivatives of certain Pgp substrates specifically label two regions, one on each half of the protein. In this study, using [125I]iodoarylazidoprazosin ([125I]IAAP), a photoactive analog of prazosin, we have demonstrated the presence of two nonidentical drug-interaction sites within Pgp. Taking advantage of a highly susceptible trypsin cleavage site in the linker region of Pgp, we characterized the [125I]IAAP binding to the N- and C-terminal halves. cis(Z)-Flupentixol, a modulator of Pgp function, preferentially increased the affinity of [125I]IAAP for the C-terminal half of the protein (C-site) by reducing the Kd from 20 to 6 nM without changing the labeling or affinity (Kd = 42-46 nM) of the N-terminal half (N-site). Also, the concentration of vinblastine (Pgp substrate) and cyclosporin A (Pgp modulator) required for 50% inhibition of [125I]IAAP binding to the C-site was increased 5- to 6-fold by cis(Z)-flupentixol without any effect on the N-site. In addition, [125I]IAAP binding to the N-site was less susceptible than to C-site to inhibition by vanadate which blocks ATP hydrolysis and drug transport. These data demonstrate the presence of at least two nonidentical substrate interaction sites in Pgp.

Relation between the turnover number for vinblastine transport and for vinblastine-stimulated ATP hydrolysis by human P-glycoprotein

Considerable uncertainty surrounds the stoichiometry of coupling of ATP hydrolysis to drug pumping by P-glycoprotein, the multidrug transporter. To estimate relative turnovers for pumping of the drug vinblastine and ATP hydrolysis, we began by measuring the number of P-glycoprotein molecules on the surface of murine NIH3T3 cells expressing the human MDR1 gene. Fluorescence of cells treated with monoclonal antibody UIC2 was determined as a function of (i) amount of antibody at a fixed number of cells and (ii) increasing cell number at constant antibody. The two together gives 1.95 x 10(6) P-glycoprotein molecules/cell. Initial uptake rates of vinblastine +/- verapamil measure the ability of P-glycoprotein to extract vinblastine from the plasma membrane before it enters the cell. As a function of [vinblastine] at 37 degrees C, they give the maximum rate of this component of outward pumping as 2.1 x 10(6) molecules s-1 cell-1 or a turnover number of 1.1 s-1. Initial rates of one-way efflux as a function of [vinblastine] at 25 degrees C +/- glucose give the maximum rate of this component of pumping as 0.59 x 10(6) molecules s-1 cell-1. The ratio of ATPase activity of P-glycoprotein at 37 and 25 degrees C is 4.6. Appropriating this ratio for pumping, maximum one-way efflux at 37 degrees C is 4.6 x 0.59 = 2.7 x 10(6) molecules s-1 cell-1, a turnover number of 1.4 s-1. The vinblastine-stimulated ATPase activity of P-glycoprotein has a turnover number of 3.5 s-1 at 37 degrees C, giving 2.8 molecules of ATP hydrolyzed for every vinblastine molecule transported in a particular direction. These calculations involve several approximations, but turnover numbers for pumping of vinblastine and for vinblastine-stimulated ATP hydrolysis are comparable. Thus, ATP hydrolysis is probably directly linked to drug transport by P-glycoprotein.

Functional characterization of a glycine 185-to-valine substitution in human P-glycoprotein by using a vaccinia-based transient expression system

Human P-glycoprotein (Pgp) is a 170-kDa plasma membrane protein that confers multidrug resistance to otherwise sensitive cells. A mutation in Pgp, G185-->V, originally identified as a spontaneous mutation, was shown previously to alter the drug resistance profiles in cell lines that are stably transfected with the mutant MDR1 cDNA and selected with cytotoxic agents. To understand the mechanism by which the V185 mutation leads to an altered drug resistance profile, we used a transient expression system that eliminates the need for drug selection to attain high expression levels and allows for the rapid characterization of many aspects of Pgp function and biosynthesis. The mutant and wild-type proteins were expressed at similar levels after 24-48 h in human osteosarcoma (HOS) cells by infection with a recombinant vaccinia virus encoding T7 RNA polymerase and simultaneous transfection with a plasmid containing MDR1 cDNA controlled by the T7 promoter. For both mutant and wild-type proteins, photolabeling with [3H]azidopine and [125I]iodoarylazidoprazosin, drug-stimulated ATPase activity, efflux of rhodamine 123, and accumulation of radiolabeled vinblastine and colchicine were evaluated. In crude membrane preparations from HOS cells, a higher level of basal Pgp-ATPase activity was observed for the V185 variant than for the wild-type, suggesting partial uncoupling of drug-dependent ATP hydrolysis by the mutant. Several compounds, including verapamil, nicardipine, tetraphenylphosphonium, and prazosin, stimulated ATPase activities of both the wild-type and mutant similarly, whereas cyclosporin A inhibited the ATPase activity of the mutant more efficiently than that of the wild-type. This latter observation explains the enhanced potency of cyclosporin A as an inhibitor of the mutant Pgp. No differences were seen in verapamil-inhibited rhodamine 123 efflux, but the rate of accumulation was slower for colchicine and faster for vinblastine in cells expressing the mutant protein, as compared with those expressing wild-type Pgp. We conclude that the G185-->V mutation confers pleiotropic alterations on Pgp, including an altered basal ATPase activity and altered interaction with substrates and the inhibitor cyclosporin A.

P-glycoprotein and multidrug resistance

Although the phenomenon of simultaneous resistance to multiple cytotoxic drugs (multidrug resistance) in cancer cells has been discussed for more than two decades, and the human and mouse genes encoding an energy-dependent transporter (the multidrug transporter or P-glycoprotein) responsible for multidrug resistance were cloned 10 years ago, there is still considerable controversy about the mechanism of action of this efflux pump and its true biological function. This review summarizes the current research on the mechanism of action of the multidrug transporter, including the hydrophobic cleaner and altered partitioning models, the possible function of P-glycoprotein as a chloride and/or ATP channel, the role of phosphorylation in its function and fact and speculation about its physiological role.

Characterization of phosphorylation-defective mutants of human P-glycoprotein expressed in mammalian cells

To assess the role of phosphorylation of the human multidrug resistance MDR1 gene product P-glycoprotein for its drug transport activity, phosphorylation sites within its linker region were subjected to mutational analysis. We constructed a 5A mutant, in which serines at positions 661, 667, 671, 675, and 683 were replaced by nonphosphorylatable alanine residues, and a 5D mutant carrying aspartic acid residues at the respective positions to mimic permanently phosphorylated serine residues. Transfection studies revealed that both mutants were targeted properly to the cell surface and conferred multidrug resistance by diminishing drug accumulation. In contrast to wild-type P-glycoprotein, the overexpressed 5A and the 5D mutants exhibited no detectable levels of phosphorylation, either in vivo following metabolic labeling of cells with [32P]orthophosphate or in vitro in phosphorylation assays with protein kinase C, cAMP-dependent protein kinase, or a P-glyco-protein-specific protein kinase purified from multidrug-resistant KB-V1 cells. These results reconfirm that the major P-glycoprotein phosphorylation sites are located within the linker region. Furthermore, the first direct evidence is provided that phosphorylation/dephosphorylation mechanisms do not play an essential role in the establishment of the multidrug resistance phenotype mediated by human P-glycoprotein.

Reconstitution of a passive Ca(2+)-transport pathway from the basolateral plasma membrane of rat parotid gland acinar cells

We have previously reported that rat parotid gland basolateral plasma membrane vesicles (BLMV) have a relatively high affinity Ca2+ transport pathway and an unsaturable Ca2+ flux component (Lockwich et al., 1994. J. Membrane Biol. 141:289-296). In this study, we have solubilized BLMV with octylglucoside (1.5%) and have reconstituted the solubilized proteins into proteoliposomes (PrL) composed of E. coli bulk phospholipids, by using a detergent dilution method. PrL exhibited 3-5-fold higher 45Ca2+ influx than control liposomes (without protein). Ca2+ uptake into PrL was dependent on the [protein] in PrL and steady state [Ca2+] in PrL was in equilibrium with external [Ca2+]. These data demonstrate that a passive, protein-mediated Ca2+ transport has been reconstituted from BLMV into PrL. 45Ca2+ influx into liposomes did not saturate with increasing [Ca2+] in the assay medium. In contrast, PrL displayed saturable 45Ca2+ influx and exhibited a single Ca2+ flux component with an apparent KCa = 242 +/- 50.9 microM and Vmax = 13.5 +/- 1.14 nmoles Ca2+/mg protein/ minute. The KCa of Ca(2+)-transport in PrL was similar to that of the high affinity Ca2+ influx component in BLMV while the Vmax was about 4-fold higher. The unsaturable Ca2+ flux component was not detected in PrL. 45Ca2+ influx in PrL was inhibited by divalent cations in the order of efficacy, Zn2+ > Mn2+ > Co2+ = Ni2+, and appeared to be more sensitive to lower concentrations of Zn2+ than in BLMV. Consistent with our observations with BLMV, the carboxyl group reagent N,N'-dicyclohexylcarbodiimide (DCCD) inhibited the reconstituted Ca2+ transport in PrL. Importantly, in both BLMV and PrL, DCCD induced a 40-50% decrease in Vmax of Ca2+ transport without an alteration in KCa. These data strongly suggest that the high affinity, passive Ca2+ transport pathway present in BLMV has been functionally reconstituted into PrL. We suggest that this approach provides a useful experimental system towards isolation of the protein(s) involved in mediating Ca2+ influx in the rat parotid gland basolateral plasma membrane.

Bacterial expression of the linker region of human MDR1 P-glycoprotein and mutational analysis of phosphorylation sites

Phosphorylation may play a role in modulating multidrug resistance by P-glycoprotein (P-gp). The linker region between the two homologous halves of human P-gp harbors several serine residues which are phosphorylated by protein kinase C (PKC) in vitro. We used the glutathione S-transferase gene fusion system to express and purify a series of fusion proteins containing the relevant portion (residues 644-689) of the linker region of the human MDR1 gene product. The fusion proteins were subjected to in vitro phosphorylation and phosphopeptide mapping analysis to identify specific phosphorylation sites. On the basis of a mutational strategy in which individual serine residues were systematically replaced with nonphosphorylatable alanine residues, Ser-661 and Ser-667 were identified as major PKC sites and Ser-683 was identified as a minor PKC site. Ser-661 and Ser-667 were also found to be the primary sites of phosphorylation for a novel membrane-associated P-gp specific kinase isolated from the multidrug-resistant KB-V1 cell line. Individual phosphorylation sites were recognized independently of each other. These data show that the linker region of P-gp represents a target for multisite phosphorylation not only for PKC but also for the P-gp specific V1 kinase. Specific serine phosphorylation sites are identified, and evidence is presented that the V1 kinase has a specificity which overlaps, but is more restricted than, that of PKC. In addition, these studies also suggest that the use of GST fusion peptides may be applicable for the analysis of multisite and ordered protein phosphorylation in other systems.

Purification and reconstitution of functional human P-glycoprotein

The overexpression of the P-glycoprotein, the MDR1 gene product, has been linked to the development of resistance to multiple cytotoxic natural product anticancer drugs in certain cancers and cell lines derived from tumors. P-glycoprotein, a member of the ATP-binding cassette (ABC) superfamily of transporters, is believed to function as an ATP-dependent drug efflux pump with broad specificity for chemically unrelated hydrophobic compounds. We review here recent studies on the purification and reconstitution of P-glycoprotein to elucidate the mechanism of drug transport. P-glycoprotein from the human carcinoma multidrug resistant cell line, KB-V1, was purified by sequential chromatography on anion exchange followed by a lectin (wheat germ agglutinin) column. Proteoliposomes reconstituted with pure protein exhibited high levels of drug-stimulated ATPase activity as well as ATP-dependent [3H]vinblastine accumulation. Both the ATPase and vinblastine transport activities of the reconstituted P-glycoprotein were inhibited by vanadate. In addition, the vinblastine transport was inhibited by verapamil and daunorubicin. These studies provide strong evidence that the human P-glycoprotein functions as an ATP-dependent drug transporter. The development of the reconstitution system and the availability of recombinant protein in large amounts due to recent advances in overexpression of P-glycoprotein in a heterologous expression system should facilitate a better understanding of the function of this novel protein.

Effects of phosphorylation of P-glycoprotein on multidrug resistance

Cells expressing elevated levels of the membrane phosphoprotein P-glycoprotein exhibit a multidrug resistance phenotype. Studies involving protein kinase activators and inhibitors have implied that covalent modification of P-glycoprotein by phosphorylation may modulate its biological activity as a multidrug transporter. Most of these reagents, however, have additional mechanisms of action and may alter drug accumulation within multidrug resistant cells independent of, or in addition to, their effects on the state of phosphorylation of P-glycoprotein. The protein kinase(s) responsible for P-glycoprotein phosphorylation has(ve) not been unambiguously identified, although several possible candidates have been suggested. Recent biochemical analyses demonstrate that the major sites of phosphorylation are clustered within the linker region that connects the two homologous halves of P-glycoprotein. Mutational analyses have been initiated to confirm this finding. Preliminary data obtained from phosphorylation- and dephosphorylation-defective mutants suggest that phosphorylation of P-glycoprotein is not essential to confer multidrug resistance.

Heterologous expression systems for P-glycoprotein: E. coli, yeast, and baculovirus

Chemotherapy, though it remains one of the front-line weapons used to treat human cancer, is often ineffective due to drug resistance mechanisms manifest in tumor cells. One common pattern of drug resistance, characterized by simultaneous resistance to multiple amphipathic, but otherwise structurally dissimilar anticancer drugs, is termed multidrug resistance. Multidrug resistance in various model systems, covering the phylogenetic range from bacteria to man, can be conferred by mammalian P-glycoproteins (PGPs), often termed multidrug transporters. PGPs are 170-kD polytopic membrane proteins, predicted to consist of two homologous halves, each with six membrane spanning regions and one ATP binding site. They are members of the ATP-binding cassette (ABC) superfamily of transporters, and are known to function biochemically as energy-dependent drug efflux pumps. However, much remains to be learned about PGP structure-function relationships, membrane topology, posttranslational regulation, and bioenergetics of drug transport. Much of the recent progress in the study of the human and mouse PGPs has come from heterologous expression systems which offer the benefits of ease of genetic selection and manipulation, and/or short generation times of the organism in which PGPs are expressed, and/or high-level expression of recombinant PGP. Here we review recent studies of PGP in E. coli, baculovirus, and yeast systems and evaluate their utility for the study of PGPs, as well as other higher eukaryotic membrane proteins.

Ultrastructure, pharmacologic inhibition, and transport selectivity of aquaporin channel-forming integral protein in proteoliposomes

Reconstitution of highly purified aquaporin CHIP (channel-forming integral protein) into proteoliposomes was previously shown to confer high osmotic water permeability (Pf) to the membranes [Zeidel et al. (1992) Biochemistry 31, 7436-7440]. Here we report detailed ultrastructural, pharmacologic, and transport studies of human red cell CHIP in proteoliposomes. Freeze-fracture and transmission electron microscopy revealed a uniform distribution of CHIP which was incorporated into the membranes in both native and inverse orientations. Morphometric analysis of membranes reconstituted at three different concentrations of CHIP revealed that the intramembrane particles correspond to tetramers or possible higher order oligomers, and the Pf increased in direct proportion to the CHIP density. Proteolytic removal of the 4-kDa C-terminal cytoplasmic domain of CHIP did not alter the Pf or oligomerization in red cell membranes. CHIP exhibited a similar conductance for water when reconstituted into membranes of varied lipid compositions. The sensitivities of CHIP-mediated Pf to specific sulfhydryl reagents were identical to known sensitivities of red cell Pf, including a delayed response to p-(chloromercuri)benzenesulfonate. CHIP did not increase the permeability of the proteoliposome membranes to H+/OH- or NH3. These studies demonstrate that CHIP proteoliposomes exhibit all known characteristics of water channels in native red cells and therefore provide a defined system for biophysical analysis of transmembrane water movements.

Partial purification and reconstitution of the human multidrug-resistance pump: characterization of the drug-stimulatable ATP hydrolysis

Multidrug-resistant human tumor cells overexpress the MDR1 gene product P-glycoprotein, which is believed to function as an ATP-dependent efflux pump. In this study we demonstrate that the partially purified P-glycoprotein, when reconstituted in an artificial membrane, catalyzes drug-stimulated ATP hydrolysis. Plasma membrane proteins of a human multidrug-resistant cell line, KB-V1, were solubilized with 1.4% (wt/vol) octyl beta-D-glucopyranoside in the presence of 0.4% phospholipid and 20% (vol/vol) glycerol, and the crude detergent extract was chromatographed on DEAE-Sepharose CL-6B. The 0.1 M NaCl fraction, enriched in P-glycoprotein but devoid of Na,K-ATPase, was reconstituted by the detergent-dilution method. P-glycoprotein constituted 25-30% of the reconstituted protein in proteoliposomes. ATP hydrolysis by proteoliposomes was stimulated 3.5-fold by the addition of vinblastine but was unaffected by the hydrophobic antitumor agent camptothecin, which is not transported by P-glycoprotein. The stimulatory effect of vinblastine was observed only if the protein was reconstituted in proteoliposomes, suggesting that either the substrate binding site(s) was masked by detergent or that the conformation of the soluble P-glycoprotein might not be suitable for substrate-induced activation. Several other drugs that are known to be transported by P-glycoprotein enhanced the ATPase activity in a dose-dependent manner with relative potencies as follows: doxorubicin = vinblastine greater than daunomycin greater than actinomycin D greater than verapamil greater than colchicine. The basal and vinblastine-stimulated ATPase activities were inhibited by vanadate (50% inhibition observed at 7-10 microM) but were not affected by agents that inhibit other ATPases and phosphatases. These data indicate that the P-glycoprotein, similar to other ion-transporting ATPases, exhibits a high level of ATP hydrolysis (5-12 mumol per min per mg of protein).

Reconstitution of functional water channels in liposomes containing purified red cell CHIP28 protein

Water rapidly crosses the plasma membranes of red blood cells (RBCs) and renal tubules through highly specialized channels. CHIP28 is an abundant integral membrane protein in RBCs and renal tubules, and Xenopus laevis oocytes injected with CHIP28 RNA exhibit high osmotic water permeability, Pf [Preston et al. (1992) Science 256, 385-387]. Purified CHIP28 from human RBCs was reconstituted into proteoliposomes in order to establish if CHIP28 is itself the functional unit of water channels and to characterize its physiological behavior. CHIP28 proteoliposomes exhibit Pf which is up to 50-fold above that of control liposomes, but permeability to urea and protons is not increased. Like intact RBC, the Pf of CHIP28 proteoliposomes is reversibly inhibited by mercurial sulfhydryl reagents and exhibits a low Arrhenius activation energy. The magnitude of CHIP28-mediated water flux (11.7 x 10(-14) cm3/s per CHIP28) corresponds to the known Pf of intact RBCs. These results demonstrate that CHIP28 protein functions as a molecular water channel and also indicate that CHIP28 is responsible for most transmembrane water movement in RBCs.

Identification, purification, and reconstitution of OxlT, the oxalate: formate antiport protein of Oxalobacter formigenes

We had proposed earlier that the anaerobe Oxalobacter formigenes sustains a proton-motive force by exploiting a secondary carrier rather than a primary proton pump. In this view, a carrier protein would catalyze the exchange of extracellular oxalate, a divalent anion, and intracellular formate, the monovalent product of oxalate decarboxylation. Such an electrogenic exchange develops an internally negative membrane potential, and since the decarboxylation reaction consumes an internal proton, the combined activity of the carrier and the soluble decarboxylase would constitute an "indirect" proton pump with a stoichiometry of 1H+ per turnover. This model is now verified by identification and purification of OxlT, the protein responsible for the anion exchange reaction. Membranes of O. formigenes were solubilized at pH 7 with 1.25% octyl glucoside in 20 mM 3-(N-morpholino)propanesulfonic acid/K, in the presence of 0.4% Escherichia coli phospholipids and with 20% glucerol present as the osmolyte stabilant. Rapid methods for reconstitution were developed to monitor the distribution of OxlT during biochemical fractionation, allowing its purification by sequential anion and cation exchange chromatography. OxlT proved to be a single hydrophobic polypeptide, of 38 kDa mobility during sodium dodecyl sulfate-polyacrylamide gel electrophoresis, with a turnover number estimated as at least 1000/s. The properties of OxlT point to an indirect proton pump as the mechanism by which a proton-motive force arises in O. formigenes, and one may reasonably argue that indirect proton pumps take part in bacterial events such as acetogenesis, malolactate fermentation, and perhaps methanogenesis.

UhpT, the sugar phosphate antiporter of Escherichia coli, functions as a monomer

We have characterized the minimal functioning unit of UhpT, the secondary carrier that mediates exchange of phosphate and glucose 6-phosphate in Escherichia coli. Membranes of a UhpT overproducing strain were solubilized with 1.25% octyl beta-D-glucopyranoside, in the presence of 0.1% E. coli phospholipid and with 20% glycerol as the osmolyte stabilant. That soluble UhpT could bind its natural substrates was indicated by the protections afforded by sugar phosphates against thermal inactivation or chemical modification with pyridoxal 5'-phosphate. Moreover, the degree of protection correlated with the strength of interaction between UhpT and the test substrate (2-deoxyglucose 6-phosphate = glucose 6-phosphate greater than galactose 6-phosphate = glucose 1-phosphate much greater than glucose 6-sulfate). Other experiments demonstrated that soluble UhpT existed as a monomer. For example, during both high performance liquid chromatography and conventional gel permeation chromatography, the elution pattern of UhpT activity was measured directly by a rapid reconstitution technique. In both cases, and in the presence and absence of substrate, UhpT activity traveled as a single component of Mr 53,000, corresponding closely to the sequence prediction of 50,600. Finally, reconstitution was studied at protein to lipid ratios low enough to achieve between 0.075 and 1.5 UhpT monomers/proteoliposome. Specific activity was constant throughout this range, a finding consistent with the idea of a functional monomer. Mitochondria and chloroplasts provide the only other anion exchange carriers described at this level of biochemical resolution, and these organelle antiporters function as dimers. By contrast, work summarized here places their bacterial counterpart, UhpT, in the same class as the lactose carrier of E. coli and the glucose carrier of the human erythrocyte, both of which function as monomers. Consideration of this pattern in conjunction with the known hydropathy profiles of these proteins suggests a novel scheme for the classification of all secondary carriers, with implications for both the structure and origin of these transport proteins.

Anion-exchange mechanisms in bacteria

This article discusses the physiological, biochemical, and molecular properties of bacterial anion-exchange reactions, with a particular focus on a family of phosphate (Pi)-linked antiporters that accept as their primary substrates sugar phosphates such as glucose 6-phosphate (G6P), mannose 6-phosphate, or glycerol 3-phosphate. Pi-linked antiporters may be found in both gram-positive and gram-negative cells. As their name suggests, these exchange proteins accept both inorganic and organic phosphates, but the two classes of substrate interact very differently with the protein. Thus, Pi is always accepted with a relatively low affinity, and when it participates in exchange, it is always taken as the monovalent anion. By contrast, when the high-affinity organic phosphates are used, these same systems fail to discriminate between monovalent and divalent forms. Tests of heterologous exchange (e.g., Pi: G6P) indicate that these proteins have a bifunctional active site that accepts a pair of negative charges, whether as two monovalent anions or as a single divalent anion. For this reason, exchange stoichiometry moves between limits of 2:1 and 2:2, according to the ratio of mono- and divalent substrates at either membrane surface. Since G6P has a pK2 within the physiological range (pK of 6.1), this predicts a novel reaction sequence in vivo because internal pH is more alkaline than external pH. Accordingly, one expects an asymmetric exchange as two monovalent G6P anions from the relatively acidic exterior move against a single divalent G6P from the alkaline interior. In this way an otherwise futile self-exchange of G6P can be biased towards a net inward flux driven (indirectly) by the pH gradient. Despite the biochemical complexity exhibited by Pi-linked antiporters, they resemble all other secondary carriers at a molecular level and show a likely topology in which two sets of six transmembrane alpha-helices are connected by a central hydrophilic loop. Speculations on the derivation of this common form suggest a limited number of structural models to accommodate such proteins. Three such models are presented.

Reconstitution of a bacterial periplasmic permease in proteoliposomes and demonstration of ATP hydrolysis concomitant with transport

The histidine periplasmic permease of Salmonella typhimurium has been partially purified and reconstituted into proteoliposomes. In this in vitro preparation, transport activity is completely dependent on the presence of all four permease proteins (HisJ, HisQ, HisM, and HisP) and on internal ATP. The reconstituted system shows initial rates of transport that are comparable to those obtained with right-side-out membrane vesicles and it establishes a 100-fold concentration gradient for histidine. Proteoliposomes also transport histidine when GTP replaces ATP. Proteoliposomes do not catalyze significant ATP hydrolysis until histidine transport is initiated by addition of substrate along with HisJ, the water-soluble histidine-binding protein. Both initially and throughout the course of substrate transport there is a concomitant hydrolysis of ATP, with an apparent stoichiometry (ATP/histidine) of 5:1. These experiments demonstrate directly that ATP is the source of energy for periplasmic permeases, thus resolving previous controversies on this topic.

Functional reconstitution of prokaryote and eukaryote membrane proteins

A new strategy for the functional reconstitution of membrane proteins is described. This approach introduces a new class of protein stabilizing agents--osmolytes--whose presence at high concentration (10-20%) during detergent solubilization prevents the inactivations that normally occur when proteins are extracted from natural membranes. Osmolytes that act in this way include compounds such as glycerol and higher polyols (erythritol, xylitol, sorbitol), sugars (glucose, trehalose), and certain amino acids (glycine, proline, betaine). The beneficial effects of osmolytes are documented by reconstitution of a variety of prokaryote and eukaryote membrane proteins, including several proton- and calcium-motive ATPases, cation- and anion-linked solute carriers (symport and antiport), and a membrane-bound hydrolase from endoplasmic reticulum. In all cases, the presence of 20% glycerol or other osmolyte during detergent solubilization led to 10-fold or more increased specific activity in proteoliposomes. These positive effects did not depend on use of any specific detergent for protein solubilization, nor on any particular method of reconstitution, but for convenience most of the work reported here has used octylglucoside as the solubilizing agent, followed by detergent-dilution to form proteoliposomes. The overall approach outlined by these experiments is simple and flexible. It is now feasible to use reconstitution as an analytical tool to study the biochemical and physiological properties of membrane proteins.

The mechanism of glucose 6-phosphate transport by Escherichia coli

To evaluate anion exchange as the mechanistic basis of sugar phosphate transport, natural and artificial membranes were used in studies of glucose 6-phosphate (Glc-6-P) and inorganic phosphate (Pi) accumulation by the uhpT-encoded protein (UhpT) of Escherichia coli. Experiments with intact cells demonstrated that UhpT catalyzed the neutral exchange of internal and external Pi, and work with everted as well as right-side-out membrane vesicles showed further that UhpT mediated the heterologous exchange of Pi and Glc-6-P. When loaded with Pi, but not when loaded with morpholinopropanesulfonate (MOPS), everted vesicles took up Glc-6-P to levels 100-fold above medium concentration in a reaction unaffected by the ionophores valinomycin, valinomycin plus nigericin, and carbonyl cyanide p-trifluoromethoxyphenylhydrazone. Similarly, right-side-out vesicles were capable of Glc-6-P transport, but only if a suitable internal countersubstrate was available. Thus, in MOPS-loaded vesicles, oxidative metabolism established a proton-motive force that supported proline or Pi accumulation, but transport of Glc-6-P was found only if vesicles could accumulate Pi during a preincubation. After reconstitution of UhpT into proteoliposomes it was possible to show as well that the level of accumulation of Glc-6-P (17 to 560 nmol/mg of protein) was related directly to the internal concentration of Pi. These results are most easily understood if the transport of glucose 6-phosphate in E. coli occurs by anion exchange rather than by nH+/anion support.

Reconstitution of the lysosomal proton pump

Lysosomal membrane proteins solubilized with octyl beta-D-glucopyranoside were reconstituted into proteoliposomes using acetone/ether-washed phospholipids from Escherichia coli. Assays of the quenching of acridine orange fluorescence showed that addition of both ATP and valinomycin to K+-loaded proteoliposomes led to the formation of a pH gradient that was acidic inside. ATP-driven acidification took place in the absence of permeant anions and was inhibited by the "protonophore", carbonylcyanide p-trifluoromethoxyphenylhydrazone, indicating that only H+ was transported actively. Proton translocation was readily blocked by N-ethylmaleimide (10 microM gave 50% inhibition of fluorescence quenching) but was unaffected by oligomycin (50 nM), orthovanadate (50 microM), or ouabain (0.5 mM); similarly, only N-ethylmaleimide affected ATP hydrolysis by proteoliposomes (88% inhibition). Other work showed that reconstitution of ATP-driven proton translocation required the presence of glycerol during protein solubilization and that optimal recovery depended on the use of both glycerol and phospholipid at this stage. We conclude that acidification of the lysosome is mediated by an ATPase capable of electrogenic H+ translocation without molecular coupling to other ionic species.