Using purified P-glycoprotein to understand multidrug resistance

Synthesis and biological assessment of new pyrimidopyrimidines as inhibitors of breast cancer resistance protein (ABCG2)

Multidrug resistance constitutes a serious obstacle of the treatment success of cancer by chemotherapy. Mostly it is driven by expression of ABC transport proteins that actively efflux the anticancer agents out of the cell. This work describes the design and synthesis of 12 new pyrimidopyrimidines, as well as their inhibition of ABCG2 a transporter referred also to as breast cancer resistance protein, the selectivity versus ABCB1 (P-glycoprotein/P-gp) and ABCC1 as well as the investigation of their accumulation in single cells. From these results, N-(3,5-dimethoxyphenyl)-2-methyl-7-phenyl-5-(p-tolyl)pyrimido[4,5-d]pyrimidin-4-amine 7 h was identified as promising hit that deserves further investigation showing a selective and effective inhibition of ABCG2 with IC₅₀ equal to 0.493 µM only 2-fold less active than Ko143.

A Network Biology Approach to Decipher Stress Response in Bacteria Using Escherichia coli As a Model

The development of drug-resistant pathogenic bacteria poses challenges to global health for their treatment and control. In this context, stress response enables bacterial populations to survive extreme perturbations in the environment but remains poorly understood. Specific modules are activated for unique stressors with few recognized global regulators. The phenomenon of cross-stress protection strongly suggests the presence of central proteins that control the diverse stress responses. In this work, Escherichia coli was used to model the bacterial stress response. A Protein-Protein Interaction Network was generated by integrating differentially expressed genes in eight stress conditions of pH, temperature, and antibiotics with relevant gene ontology terms. Topological analysis identified 24 central proteins. The well-documented role of 16 central proteins in stress indicates central control of the response, while the remaining eight proteins may have a novel role in stress response. Cluster analysis of the generated network implicated RNA binding, flagellar assembly, ABC transporters, and DNA repair as important processes during response to stress. Pathway analysis showed crosstalk of Two Component Systems with metabolic processes, oxidative phosphorylation, and ABC transporters. The results were further validated by analysis of an independent cross-stress protection dataset. This study also reports on the ways in which bacterial stress response can progress to biofilm formation. In conclusion, we suggest that drug targets or pathways disrupting bacterial stress responses can potentially be exploited to combat antibiotic tolerance and multidrug resistance in the future.

Edoxaban drug-drug interactions with ketoconazole, erythromycin, and cyclosporine

AIMS

Edoxaban, a novel factor Xa inhibitor, is a substrate of cytochrome P450 3 A4 (CYP3A4) and the efflux transporter P-glycoprotein (P-gp). Three edoxaban drug-drug interaction studies examined the effects of P-gp inhibitors with varying degrees of CYP3A4 inhibition.

METHODS

In each study, healthy subjects received a single oral dose of 60 mg edoxaban with or without an oral dual P-gp/CYP3A4 inhibitor as follows: ketoconazole 400 mg once daily for 7 days, edoxaban on day 4; erythromycin 500 mg four times daily for 8 days, edoxaban on day 7; or single dose of cyclosporine 500 mg with edoxaban. Serial plasma samples were obtained for pharmacokinetics and pharmacodynamics. Safety was assessed throughout the study.

RESULTS

Coadministration of ketoconazole, erythromycin, or cyclosporine increased edoxaban total exposure by 87%, 85%, and 73%, respectively, and the peak concentration by 89%, 68%, and 74%, respectively, compared with edoxaban alone. The half-life did not change appreciably. Exposure of M4, the major active edoxaban metabolite, was consistent when edoxaban was administered alone or with ketoconazole and erythromycin. With cyclosporine, M4 total exposure increased by 6.9-fold and peak exposure by 8.7-fold, suggesting an additional interaction. Pharmacodynamic effects were reflective of increased edoxaban exposure. No clinically significant adverse events were observed.

CONCLUSIONS

Administration of dual inhibitors of P-gp and CYP3A4 increased edoxaban exposure by less than two-fold. This effect appears to be primarily due to inhibition of P-gp. The impact of CYP3A4 inhibition appears to be less pronounced, and its contribution to total clearance appears limited in healthy subjects.

Quercetin as a potential modulator of P-glycoprotein expression and function in cells of human pancreatic carcinoma line resistant to daunorubicin

P-glycoprotein (P-gp) is one of the ABC transporters responsible for the resistance of several tumours to successful chemotherapy. Numerous agents are capable of interfering with the P-gp-mediated export of drugs but unfortunately most of them produce serious side effects. Some plant polyphenols, including the flavonol quercetin (Q), manifest anti-neoplastic activity mainly due to their influence on cell cycle control and apoptosis. Reports are also available which show that Q may intensify action of cytostatic drugs and suppress the multidrug resistance (MDR) phenomenon. The study aimed at determination if Q sensitizes cells resistant to daunorubicin (DB) through its effect on P-gp expression and action. The experiments were conducted on two cell lines of human pancreatic carcinoma, resistant to DB EPP85-181RDB and sensitive EPP85-181P as a comparison. Cells of both lines were exposed to selected concentrations of Q and DB, and then membranous expression of P-gp and its transport function were examined. The influence on expression of gene for P-gp (ABCB1) was also investigated. Results of the studies confirmed that Q affects expression and function of P-gp in a concentration-dependent manner. Moreover it decreased expression of ABCB1. Thus, Q may be considered as a potential modulator of P-gp.

Apoptotic death mode of mitomycin C-treated HeLa cells and cellular localization of mitomycin C-induced P-glycoprotein

Mitomycin C (MMC) is an active antineoplastic agent and is suggested to induce apoptosis in a caspase- dependent manner in human gastric, bladder, and breast cancer cells. In this study, the death mode of human cervical cancer cells (HeLa) induced by MMC and the cellular localization of MMC-induced P-glycoprotein (P-gp) were investigated. The results of caspase-3 activity, Annexin V binding, and DNA fragmentation suggested that the degree of caspase-dependent apoptosis induced by MMC was in a dose-, but not time-dependent, manner. Further, in low-dose (0.0299 microM) and long-term (2 months) treatment with MMC, P-gp is itself extruded from the cells and colocalized with nuclear DNA and the overexpression was achieved.

New functional assay of P-glycoprotein activity using Hoechst 33342

In this study we describe a simplified, HTS-capable functional assay for the multidrug resistance (MDR) transporter P-glycoprotein (P-gp) based on its substrate Hoechst 33342. The physicochemical properties of Hoechst 33342 and the enormous milieu dependency of its fluorescence intensity allowed performing the assay in a homogeneous manner. This new assay served as an effective tool to estimate the potency of 10 well recognized P-gp substrates and modulators. Further, the potency of these compounds was also estimated in the calcein AM assay. The Hoechst 33342 and calcein AM assays yielded significantly comparable results for all compounds tested. Principal component analysis (PCA) applied to literature data on inhibition of P-gp activity and our results obtained in the Hoechst 33342 and calcein AM assay indicated similarity of compared functional transport assays. However, no correlation could be detected between these functional assays and the ATPase activity assay.

Mutational analysis of conserved aromatic residues in the A-loop of the ABC transporter ABCB1A (mouse Mdr3)

The A-loop is a recently described conserved region in the NBDs of ABC transporters [Ambudkar, S.V., Kim, I.-W., Xia, D. and Sauna, Z.E. (2006) The A-loop, a novel conserved aromatic acid subdomain upstream of the Walker A motif in ABC transporters, is critical for ATP binding. FEBS Lett. 580, 1049-1055; Kim, I.W., Peng, X.H., Sauna, Z.E., FitzGerald, P.C., Xia, D., Muller, M., Nandigama, K. and Ambudkar, S.V. (2006) The conserved tyrosine residues 401 and 1044 in ATP sites of human P-glycoprotein are critical for ATP binding and hydrolysis: evidence for a conserved subdomain, the A-loop in the ATP-binding cassette. Biochemistry 45, 7605-7616]. In mouse P-glycoprotein (Abcb1a), the aromatic residue of the A-loop in both NBDs is a tyrosine: Y397 in NBD1 and Y1040 in NBD2. Another tyrosine residue (618 in NBD1 and 1263 in NBD2) also appears to lie in proximity to the ATP molecule. We have mutated residues Y397, Y618, Y1040, and Y1263 to tryptophan and analyzed the effect of these substitutions on transport properties, ATP binding, and ATP hydrolysis by Abcb1a (mouse Mdr3). Y618W and Y1263W enzymes had catalytic characteristics similar to WT Abcb1a. On the other hand, Y397W and Y1040W showed impaired transport and greatly reduced ATPase activity, including a approximately 10-fold increase in Km for MgATP. Thus, Y397 and Y1040 play an important role in Abcb1a catalysis.

Is P-glycoprotein (ABCB1) a phase 0 or a phase 3 colchicine transporter depending on colchicine exposure conditions?

This study investigates the P-glycoprotein (Pgp)-mediated transport of its substrates in accumulation or efflux modes under steady-state conditions. The kinetics of colchicine uptake and efflux, a substrate of both Pgp and intracellular tubulin, were studied in HL60 and HL60/DNR cells; HL60/DNR cells contain 25 times more Pgp than do HL60 cells. HL60/DNR cells in a medium containing 6.25 nM colchicine, which mimics therapeutic conditions, reached steady-state twice as rapidly as did HL60 cells, and accumulated 24-times less colchicine than did HL60 cells. The Pgp inhibitor GF120918, increased colchicine uptake by HL60 cells 1.2-fold and that of HL60/DNR cells 17-fold, while it had no effect on colchicine efflux from either cell line that had been incubated with colchicine for 24 h. Colchicine kinetics fitted well a two closed-compartment model, showing that the low intracellular accumulation of colchicine in HL60/DNR cells resulted from a 11-fold decrease in colchicine uptake and a 2.3-fold increase in colchicine efflux, that could be attributed to Pgp-mediated efflux activity in HL60/DNR cells. Intracellular colchicine was mainly and similarly distributed in the cytosol in both cell lines. These data demonstrate that the kinetics of the intracellular colchicine accumulation depend on the density of Pgp and that Pgp is more a phase 0 (preventing cellular uptake) than a phase 3 (effluxing intracellular substrate) transporter under steady-state conditions, although the situation is reversed after a short incubation time (30 min), when intracellular free colchicine concentration is probably high enough for it to be removed from the cell by Pgp.

Preface: the concept and consequences of multidrug resistance

The problem of multidrug resistance (MDR) in human cancers led to the discovery 30 years ago of a single protein P-glycoprotein (P-gp), capable of mediating resistance to multiple structurally diverse drugs. P-gp became the archetypal eukaryotic ABC transporter gene, and studies of P-gp and related ABC transporters in both eukaryotes and bacteria have led to a basic mechanistic understanding of the molecular basis of MDR. Particular milestones along the way have been the identification of the homology between P-gp and bacterial transport proteins, the purification and functional reconstitution of P-gp into synthetic lipid systems, and the development of targeted therapies that attempt to overcome MDR by inhibiting P-gp. This preface places into this context some of the less well-explored themes developed in the MDR field, particularly various alternative models of P-gp action, evidence for parallel physiological roles for P-gp, and the unusual relationship between the substrate recognition capabilities of ABC transporters and their evolutionary history.

The molecular basis of multidrug resistance in cancer: the early years of P-glycoprotein research

The discovery and characterization of P-glycoprotein, an energy-dependent multidrug efflux pump, as a mechanism of multidrug resistance in cancer is generally accepted as a significant contribution to the ongoing effort to end death and suffering from this disease. The historical reflections of Victor Ling and Michael Gottesman concerning the early years of this research highlight the important contributions of the multidisciplinary teams involved in these studies, and illustrate how technological developments in biochemistry and molecular and cell biology enabled this discovery.

Imaging of P-glycoprotein of H69/VP small-cell lung cancer lines by scanning near-field optical microscopy and confocal laser microspectrofluorometer

Chemoresistance remains the major obstacle to successful therapy of the lung cancer. The multi-drug resistance (MDR) is generally associated with altered expression of drug transporter proteins, such as P-glycoprotein (P-gp). So the distribution of P-gp on the membrane is of great importance to further study the interaction between drug and P-gp. In the present work, the P-gp of the H69/VP small-lung cancer cells was detected using monoclonal antibody UIC2. A secondary goat-anti mouse antibody coupled with biotin was used. The fluorescence emission was detected from a streptavidin-Texas Red. Results were investigated by a homemade scanning near-field optical microscope (SNOM) coupled to a confocal laser microspectrofluorometer (CLMF). Topographical images and localized spectra were obtained at the level of one cell membrane. It was found that the distribution of P-gp is not homogeneous and this observation is basically in accord with the fluorescent images obtained by classical microscopy. The distribution of P-gp would be localized in a higher region on a cell surface. This methodology would also enhance our understanding of MDR under physiological conditions.

Cancer: a single disease with a multitude of manifestions?

The relationships of critical nutrients such as plant phenolics, vitamins, minerals and lipids are considered with respect to the incidence of a variety of cancers, and analyzed in terms of how these nutrient deficiencies alter immune function, DNA integrity and cell proliferation. With a significant correlation found between cancer and these nutrient deficiencies, the hypothesis is presented here that nutrition could provide a unifying perception of cancer and recast it as a single disease. This further suggests that a coordinated administration of specific, critical nutrients to cancer patients could lead to the reversal of the disease. It is also proposed that the concurrent presence of a variety of nutritional deficiencies in cancer patients requires a multilevel, systemic approach to this disease as opposed to the single active therapeutic agent approach that is the cornerstone of contemporary research and pharmacology.

Functional study of intracellular P-gp- and MRP1-mediated pumping of free cytosolic pirarubicin into acidic organelles in intrinsic resistant SiHa cells

We sought to determine the efficiency of the intracellular functional P-gp- and MRP1-mediated pumping of THP into acidic organelles in SiHa cells and etoposide-resistant SiHa/VP16 cells. The expression of both MDR1 and MRP1 genes of SiHa and SiHa/VP16 cells was clearly shown by using RT-PCR. The functional studies of both intracellular functional P-gp- and MRP1-mediated pumping were performed by using THP in a conventional spectrofluorometer, and they demonstrated that SiHa and SiHa/VP16 cells are good models to illustrate the functional role of intracellular P-gp and MRP1 in the transport of free cytosolic drug into acidic organelles. The functional P-gp and MRP1 proteins were identified both on plasma membranes and on intracellular vesicle membranes. Within the limit of experimental error, similar efficiencies in THP transport were observed in the two proteins at both locations in SiHa and SiHa/VP16 cells. The P-gp- and MRP1-mediated pump coefficient (k v a), Michealis-Menten's constant (K V m), and maximal pumping rate (V V max) values of those located on vesicular membranes were 1.87 +/- 0.30 pL x cell-1 x s-1, 1.63 +/- 0.21 microM, and 4.95 +/- 0.45 nM x s-1</sup>, respectively. Drug retention inside acidic organelles (C mon V) of SiHa cells was significantly higher than that of SiHa/VP16 cells, perhaps a consequence of slower movement of recycling endosomes and (or) lysosomes to the cell membrane of SiHa cells, leading to distended organelles and cell death. Our results suggest that intracellular P-gp and MRP1 proteins play an important role in the transport of free drug from cytosol to cytoplasmic acidic organelles.

Role of the lung resistance-related protein (LRP) in the drug sensitivity of cultured tumor cells

Drug resistance, one of the major obstacle in the successful anticancer therapy, can be observed at the outset of therapy (intrinsic resistance) or after exposure to the antitumor agent (acquired resistance). To gain a better insight into the mechanisms of intrinsic resistance we have analyzed two human cell types derived from untreated tumors: MCF-7 breast cancer and A549 non small cell lung cancer (NSCLC). We have examined: the cytotoxic effect induced by doxorubicin (DOX); the time course of drug accumulation by flow cytometry and intracellular drug distribution by confocal microscopy; the expression and distribution of proteins related to anthracycline resistance, such as P-gp (P-glycoprotein), MRP1 (multidrug resistance-associated protein) and LRP (lung resistance-related protein). The cytotoxicity assays showed that A549 cells were less sensitive than MCF-7 cells to the DOX treatment in agreement with the different DOX uptake. Moreover, while in A549 cells DOX was mostly located in well defined intracytoplasmic vesicles, in MCF-7 cells it was mainly revealed inside the nuclei. The analysis of P-gp and MRP expression did not show significant differences between the two cell lines while a high expression of LRP was detected at the nuclear envelope and cytoplasmic levels in A549 cells. These findings suggest that the lower sensitivity to DOX treatment showed by lung carcinoma cells could be ascribed to drug sequestration by LRP inside the cytoplasmic compartments.

Immobilized receptor- and transporter-based liquid chromatographic phases for on-line pharmacological and biochemical studies: a mini-review

This review addresses the synthesis and characterization of two different types of receptor-based liquid chromatographic supports, one based upon a trans-membrane ligand gated ion channel receptor (the nicotinic acetylcholine receptor) and the other a soluble nuclear receptor (the estrogen receptor). In addition, studies with the P-glycoprotein transporter are also reported. The nicotinic receptor was immobilized via hydrophobic insertion into the interstitial spaces of an immobilized artificial membrane (IAM) stationary phase. the estrogen receptor was tethered to a hydrophilic stationary phase and the membranes containing the Pgp transporter were coated on the surface of the IAM stationary phase. The stationary phases were characterized using known ligands and substrates for the respective non-immobilized proteins. The results from zonal and frontal chromatographic experiments demonstrated that the stationary phases could be used to determine binding affinities (expressed as dissociation constants, Kd,'s) and to resolve mixtures of ligands according to their relative affinities. In addition. competitive ligand binding studies on the P-glycoprotein-based stationary phase have established that this phase can be used to identify and characterize competitive displacement and allosteric interactions. These studies demonstrate that immobilized-receptor phases can be used for on-line pharmacological studies and as rapid screens for the isolation and identification of lead drug candidates from complex biological or chemical mixtures.

The ABC transporter genes of Plasmodium falciparum and drug resistance

The seminal observations that (a) chloroquine-resistant Plasmodium falciparum strains accumulate less drug than more sensitive parasites, and (b) chloroquine resistance could be modulated in vitro by the classic multidrug-resistance (MDR) modulator verapamil, suggested not only that parasite resistance to multiple drugs may be similar to the MDR phenotype described in mammalian cancer cells, but that homologous proteins may be involved. These findings prompted search for MDR-like genes in the parasite. To date, three full-length ABC transporter genes have been isolated from P. falciparum: two P-glycoprotein-like homologues, pfmdr1 and pfmdr2, and a homologue of the yeast GCN20 gene, pfgcn20.

Mutational analysis of conserved carboxylate residues in the nucleotide binding sites of P-glycoprotein

Mutagenesis was used to investigate the functional role of six pairs of aspartate and glutamate residues (D450/D1093, E482/E1125, E552/E1197, D558/D1203, D592/D1237, and E604/E1249) that are highly conserved in the nucleotide binding sites of P-glycoprotein (Mdr3) and of other ABC transporters. Removal of the charge in E552Q/E1197Q and D558N/D1203N produced proteins with severely impaired biological activity when the proteins were analyzed in yeast cells for cellular resistance to FK506 and restoration of mating in a ste6Delta mutant. Mutations at other acidic residues had no apparent effect in the same assays. These four mutants were expressed in Pichia pastoris, purified to homogeneity, and biochemically characterized with respect to ATPase activity. Studies with purified proteins showed that mutants D558N and D1203N retained 14 and 30% of the drug-stimulated ATPase activity of wild-type (WT) Mdr3, respectively, and vanadate trapping of 8-azido[alpha-(32)P]nucleotide confirmed slower basal and drug-stimulated 8-azido-ATP hydrolysis compared to that for WT Mdr3. The E552Q and E1197Q mutants showed no drug-stimulated ATPase activity. Surprisingly, drugs did stimulate vanadate trapping of 8-azido[alpha-(32)P]nucleotide in E552Q and E1197Q at a level similar to that of WT Mdr3. This suggests that formation of the catalytic transition state can occur in these mutants, and that the bond between the beta- and gamma-phosphates is hydrolyzed. In addition, photolabeling by 8-azido[alpha-(32)P]nucleotide in the presence or absence of drug was also detected in the absence of vanadate in these mutants. These results suggest that steps after the transition state, possibly involved in release of MgADP, are severely impaired in these mutant enzymes.

Nucleotide-induced conformational changes in P-glycoprotein and in nucleotide binding site mutants monitored by trypsin sensitivity

Limited trypsin digestion was used to monitor nucleotide-induced conformational changes in wild-type P-glycoprotein (Pgp) as well as in nucleotide binding domain (NBD) Pgp mutants. Purified and reconstituted wild-type or mutant mouse Mdr3 Pgps were preincubated with different hydrolyzable or nonhydrolyzable nucleotides, followed by limited proteolytic cleavage at different trypsin:protein ratios. The Pgp tryptic digestion products were separated by SDS-PAGE followed by immunodetection with the mouse monoclonal anti-Pgp antibody C219, which recognizes a conserved epitope (VVQE/AALD) in each half of the protein. Different trypsin digestion patterns were observed for wild-type Pgp incubated with MgCl(2) alone, MgADP, MgAMP.PNP, MgATP, and MgATP + vanadate. A unique trypsin digestion profile suggestive of enhanced resistance to trypsin was observed under conditions of vanadate-induced trapping of nucleotides (MgATP + vanadate). The trypsin sensitivity profiles of Pgp mutants bearing either single or double mutations in Walker A (K429R, K1072R) and Walker B (D551N, D1196N) sequence signatures of NBD1 and NBD2 were analyzed under conditions of vanadate-induced trapping of nucleotides. The proteolytic cleavage pattern observed for the double mutants K429R/K1072R and D551N/D1196N, and for the single mutants K429R, K1072R, and D1196N were similar and clearly distinct from wild-type Pgp under the same conditions. This is consistent with the absence of ATP hydrolysis and of vanadate-induced trapping of 8-azido-ADP previously reported for these mutants [Urbatsch et al. (1998) Biochemistry 37, 4592-4602]. Interestingly, the trypsin digestion profiles observed under vanadate-induced trapping for the D551N and D1196N mutants were quite different, with the D551N mutant showing a profile resembling that seen for wild-type Pgp. The different sensitivity profiles of Pgp mutants bearing mutations at the homologous residue in NBD1 (D551N) and NBD2 (D1196N) suggest possible structural and functional differences between the two sites.

Multixenobiotic resistance as a cellular defense mechanism in aquatic organisms

Multixenobiotic resistance in aquatic organisms exposed to natural toxins or anthropogenic contaminants is a phenomenon analogous to multidrug resistance in mammalian tumor cell lines tolerant of anti-cancer drugs. Multidrug resistance is commonly due to the elevated expression of transmembrane P-glycoproteins (P-gp) which actively transport a wide variety of structurally and functionally diverse compounds. The purpose of this review is to place aquatic ecotoxicological data in context of the larger multidrug resistance field of study. Information on P-glycoproteins structure, mechanism of transport, and substrate specificity gained through traditional mammalian and cell culture models is examined in conjunction with recent work on aquatic species exposed to xenobiotics both in the field and in the laboratory. The physiological function of P-glycoproteins is explored through studies of gene knockout models and expression patterns in normal tissues and tumors. The effect of xenobiotic exposures on P-gp activity and protein titer is examined in wild and captive populations of aquatic invertebrates and vertebrates. Substrate overlap and evidence of co-expression of phase I detoxification enzymes (e.g. cytochromes P450) and P-gp are presented. The role of P-gp chemosensitizers as environmental pollutants and the ecotoxicological consequences of P-gp inhibition are highlighted. The overwhelming evidence suggests that P-glycoproteins provide aquatic organisms with resistance to a wide range of natural and anthropogenic toxins.

Activation of the human P-glycoprotein ATPase by trypsin

The human MDR1 gene product, P-glycoprotein (Pgp), a tandemly duplicated molecule containing two putative ATP- and perhaps two drug-binding sites, is responsible for multidrug resistance in tumors. In this report, we characterized the effects of trypsinization of Pgp on its ATPase function. Incubation of Pgp-containing membranes with trypsin at a ratio of 1000:1 (w/w) resulted in a gradual increase in the basal- and the drug-stimulated ATPase activities of Pgp in a time-dependent manner. The maximal basal-, verapamil-, and vinblastine-stimulated ATPase activities of the trypsinized Pgp were approximately 1.8-, 1.5-, and 1.75-fold higher than the activities of the native Pgp, respectively. Increased basal- and drug-stimulated ATPase activities of the Pgp were also observed when the ratio of membrane protein to trypsin in the incubation mixtures was raised to 10:1 (w/w). Immunoblotting analysis of Pgp tryptic digests using Pgp-specific NH(2)11, C219, and C494 antibodies together revealed the degradation of full-length Pgp and formation of at least eight peptides migrating in the 36-60 kDa range. Immunoprecipitation reactions using NH(2)11 and C494 antibodies have suggested that the peptides originating from the NH(2) half of Pgp are in strong association with the COOH half of the peptide. These findings suggest that while Pgp fragments together exhibit the ATPase functional characteristics, Pgp possesses a cleavage activation site or region, and its cleavage leads to the activation of basal ATPase function of Pgp.

A bacterial model system for understanding multi-drug resistance

Mankind stands at the crossroads, recognizing the need for a radical change in bacterial disease management. The development of several antimicrobial agents in the 1940s and 1950s allowed man to gain the upper hand in controlling these diseases. However, the horizon is now clouded by the activation in bacteria of cryptic multi-drug resistance (MDR) genes and the spread of plasmid- and integron-born MDR genes through bacterial populations. Unless remedial measures are taken, nearly all currently available antimicrobial agents are likely to soon lose their efficacies. We briefly review the bacterial MDR phenomenon and focus on a recently emerging family of small multi-drug resistance (SMR) pumps which may provide an ideal model system for understanding the MDR phenomenon in general.

Simple purification of highly active biotinylated P-glycoprotein: enantiomer-specific modulation of drug-stimulated ATPase activity

A simplified method for the expression and purification of P-glycoprotein (Pgp) is presented. This method is based on the in-frame fusion of both a polyhistidine tail and a 100-amino acid residue biotin acceptor domain of oxaloacetate decarboxylase from Klebsiella pneumoniae at the carboxyl terminus end of Pgp (Pgp-H6BD). The expression/purification protocol for Pgp-H6BD involves high-level expression of the fusion protein in the yeast Pichia pastoris, biotinylation in vitro with biotin ligase, solubilization of crude membrane fractions in detergent, and affinity purification by a combination of nickel and avidin chromatography. Biotinylated Pgp binds to immobilized monomeric avidin and can be eluted with free biotin in a high state of purity. This protocol is rapid and efficient and yields purified Pgp which shows robust ATPase activity, as determined by vanadate-induced trapping of photoactive nucleotides and by direct measurement of ATP hydrolysis by Pgp-H6BD. This method should be useful for structural studies of the protein by spectroscopic or crystallographic approaches. This purified Pgp-H6BD preparation has been used to study the enantiomer-specific effects of inhibitors of Pgp-mediated drug transport on the drug-stimulated ATPase activity of the protein. A series of 1, 4-disubstituted piperazine derivatives with a central chiral carbon and modified at the head and tail groups are shown to stimulate Pgp ATPase activity in a dose-dependent fashion. Some of these compounds are also capable of inhibiting either vinblastine or verapamil stimulation of ATPase activity of Pgp in an enantiomer-specific fashion. The enantiomeric specific inhibitory activity of these compounds suggests complex interactions at a single substrate binding site(s) on Pgp.

Large scale purification of detergent-soluble P-glycoprotein from Pichia pastoris cells and characterization of nucleotide binding properties of wild-type, Walker A, and Walker B mutant proteins

P-glycoprotein (Pgp; mouse MDR3) was expressed in Pichia pastoris, grown in fermentor culture, and purified. The final pure product is of high specific ATPase activity and is soluble at low detergent concentration. 120 g of cells yielded 6 mg of pure Pgp; >4 kg of cells were obtained from a single fermentor run. Properties of the pure protein were similar to those of previous preparations, except there was significant ATPase activity in absence of added lipid. Mutant mouse MDR3 P-glycoproteins were purified by the same procedure after growth of cells in flask culture, with similar yields and purity. This procedure should open up new avenues of structural, biophysical, and biochemical studies of Pgp. Equilibrium nucleotide-binding parameters of wild-type mouse MDR3 Pgp were studied using 2'-(3')-O-(2,4,6-trinitrophenyl)adenosine tri- and diphosphate. Both analogs were found to bind with K(d) in the low micromolar range, to a single class of site, with no evidence of cooperativity. ATP displacement of the analogs was seen. Similar binding was seen with K429R/K1072R and D551N/D1196N mutant mouse MDR3 Pgp, showing that these Walker A and B mutations had no significant effect on affinity or stoichiometry of nucleotide binding. These residues, known to be critical for catalysis, are concluded to be involved primarily in stabilization of the catalytic transition state in Pgp.

Localization of a substrate specificity domain in the multidrug resistance protein

Multidrug resistance protein (MRP) confers resistance to a number of natural product chemotherapeutic agents. It is also a high affinity transporter of some physiological conjugated organic anions such as cysteinyl leukotriene C(4) and the cholestatic estrogen, 17beta-estradiol 17(beta-D-glucuronide) (E(2)17betaG). We have shown that the murine orthologue of MRP (mrp), unlike the human protein, does not confer resistance to common anthracyclines and is a relatively poor transporter of E(2)17betaG. We have taken advantage of these functional differences to identify region(s) of MRP involved in mediating anthracycline resistance and E(2)17betaG transport by generating mrp/MRP hybrid proteins. All hybrid proteins conferred resistance to the Vinca alkaloid, vincristine, when transfected into human embryonic kidney cells. However, only those in which the COOH-terminal third of mrp had been replaced with the corresponding region of MRP-conferred resistance to the anthracyclines, doxorubicin, and epirubicin. Exchange of smaller segments of the COOH-terminal third of the mouse protein by replacement of either amino acids 959-1187 or 1188-1531 with those of MRP produced proteins capable of conferring some level of resistance to the anthracyclines tested. All hybrid proteins transported cysteinyl leukotriene C(4) with similar efficiencies. In contrast, only those containing the COOH-terminal third of MRP transported E(2)17betaG with an efficiency comparable with that of the intact human protein. The results demonstrate that differences in primary structure of the highly conserved COOH-terminal third of mrp and MRP are important determinants of the inability of the murine protein to confer anthracycline resistance and its relatively poor ability to transport E(2)17betaG.

Residues in P-glycoprotein catalytic sites that react with the inhibitor 7-chloro-4-nitrobenzo-2-oxa-1,3-diazole

7-Chloro-4-nitrobenzo-2-oxa-1,3-diazole (NBD-Cl) is a specific covalent inhibitor of P-glycoprotein ATPase activity (M. K. Al-Shawi, and A. E. Senior, 1993, J. Biol. Chem. 268, 4197-4206). Complete inhibition occurs at a reaction stoichiometry of 1 mol NBD/mol P-glycoprotein, and the reagent has proved valuable in understanding catalytic mechanisms, particularly in relation to catalytic site cooperativity (A. E. Senior, and S. Bhagat, 1998, Biochemistry 37, 831-836). The actual location of reaction in the amino acid sequence has not yet been determined. Using a combined mutagenesis and biochemical approach we establish here that the initial reaction of NBD-Cl is with Cys within the Walker A consensus sequence of the N- or C-terminal nucleotide site (Cys-431 or Cys-1074 of human P-glycoprotein). Reaction with either Cys yields full inhibition. It was further found that inhibition consists of dithiothreitol (DTT)-reversible and DTT-irreversible components. The former predominates at low pH and the latter at higher pH. This demonstrates that, at higher pH, intramolecular transfer of NBD from Cys to Lys occurs, probably to the proximate Walker A Lys (Lys-433 or Lys-1076 of human P-glycoprotein). After transfer of NBD to Lys, P-glycoprotein ATPase remains fully inhibited.

ATPase activity of Chinese hamster P-glycoprotein

We have developed two defined experimental systems for biochemical investigation of P-glycoprotein, namely, plasma membranes highly enriched in Pgp, obtained from the CR1R12 Chinese hamster ovary cell line, and pure, reconstituted Pgp, obtained by solubilization of Pgp from CR1R12 plasma membranes, Reactive Red 120 chromatography, and reconstitution in liposomes. Studies of the ATPase catalytic mechanism by kinetic methods and covalent inactivation have been greatly facilitated by the availability of these experimental systems. The technique of vanadate trapping of nucleotide has been particularly useful. As a result of these studies, we now have explicit, testable, proposals for (1) the normal catalytic pathway of ATP hydrolysis, (2) a postulated alternating catalytic site cycle, and (3) coupling of ATP hydrolysis to drug transport. The experimental methods described here should prove valuable for future studies of Pgp and of ABC transporters in general.

Mutations in the nucleotide-binding sites of P-glycoprotein that affect substrate specificity modulate substrate-induced adenosine triphosphatase activity

The amino- and carboxy-terminal nucleotide-binding domains (NBD1 and NBD2) of P-glycoprotein (P-gp) share over 80% sequence identity. Almost all of NBD1 can be exchanged by corresponding NBD2 segments with no significant loss of function, except for a small segment around the Walker B motif. Within this segment, we identified two sets of residues [ERGA --> DKGT (522-525) and T578C] that, when replaced by their NBD2 counterparts, cause dramatic alterations of the substrate specificity of the protein [Beaudet, L., and Gros, P. (1995) J. Biol. Chem. 270, 17159-17170]. We wished to gain insight into the molecular basis of this defect. For this, we overexpressed the wild-type mouse Mdr3 and variants bearing single or double mutations at these positions in the yeast Pichia pastoris. P-gp-specific ATPase activity was measured in yeast plasma membrane preparations after detergent solubilization and reconstitution in Escherichia coli proteoliposomes. P-gp proteoliposomes from P. pastoris showed a strong verapamil- and valinomycin-stimulated ATPase activity, with characteristics (KM, Vmax) similar to those measured in mammalian cells. Mutations did not appear to affect the KM for Mg2+ATP ( approximately 0.4 mM), but maximum velocity (Vmax) of the drug-stimulated ATPase activity was severely affected in a substrate/modulator-specific fashion. Indeed, all mutants showed complete loss of verapamil-induced ATPase, while all retained at least some degree of valinomycin-induced ATPase activity. Photolabeling studies with [125I]iodoarylazidoprazosin, including competition with MDR drugs and modulators, suggested that drug binding was not affected in the mutants. The altered drug resistance profiles of the ERGA --> DKGT(522-525) and T578C mutants in vivo, together with the observed alterations in substrate-induced ATPase activity of these proteins, suggest that the residues involved may form part of a signal pathway between the membrane regions (substrate binding) and the ATP binding sites.

Functional intracellular P-glycoprotein

Efflux of chemotherapy drugs by P-glycoprotein (P-gp) at the plasma membrane is thought to be a major cause of cancer multidrug resistance. In this report, we show by flow cytometry that P-gp also concentrates large amounts of 2 different drugs, Hoechst 33342 and daunorubicin, within a cytoplasmic compartment of multidrug resistant CHRC5 cells. A quantitative assay of Hoechst 33342 revealed that cytoplasmic sequestration by P-gp in CHRC5 cells accounted for about half of the amount of Hoechst 33342 accumulated by the drug-sensitive parental Aux BI cells. Daunorubicin sequestered in the cytoplasm of CHRC5 cells could be released by inhibiting P-gp function with cyclosporin A, resulting in cell death. A likely site of drug sequestration is P-gp-containing cytoplasmic vesicles, in which the P-gp is oriented so that drugs are transported and concentrated in the interior of the vesicles. P-gp was detected in the membranes of cytoplasmic vesicles of CHRC5 cells by confocal immunofluorescence microscopy and immunoelectron microscopy with anti-P-gp monoclonal antibodies (MAbs). Vesicular localization of daunorubicin was observed by epifluorescence microscopy. The origin and nature of the P-gp-containing vesicles are unknown, but they do not correspond to endocytic vesicles. Our results directly demonstrate that chemosensitizer-induced release of drugs sequestered in cytoplasmic vesicles by P-gp can be used to overcome multidrug resistance.

Mutations in either nucleotide-binding site of P-glycoprotein (Mdr3) prevent vanadate trapping of nucleotide at both sites

Vanadate trapping of nucleotide and site-directed mutagenesis were used to investigate the role of the two nucleotide-binding (NB) sites in the regulation of ATP hydrolysis by P-glycoprotein (mouse Mdr3). Mdr3, tagged with a hexahistidine tail, was overexpressed in the yeast Pichia pastoris and purified to about 90% homogeneity by Ni-affinity chromatography. This protocol yielded purified, reconstituted Mdr3 which exhibited high verapamil stimulation of ATPase activity with a Vmax of 4.2 micromol min-1 mg-1 and a KM of 0.7 mM, suggesting that Mdr3 purified from P. pastoris is highly functional. Point mutations were introduced into the core consensus sequence of the Walker A or B motifs in each of the two NB sites. The mutants K429R, K1072R (Walker A) and D551N, D1196N (Walker B) were functionally impaired and unable to confer cellular resistance to the fungicide FK506 in the yeast Saccharomyces cerevisiae. Single and double mutants (K429R/K1072R, D551N/D1196N) were expressed in P. pastoris, and the effect of these mutations on the ATPase activity of Mdr3 was characterized. Purified reconstituted Mdr3 mutants showed no detectable ATPase activity compared to proteoliposomes purified from negative controls (<5% of wild-type Mdr3). Vanadate readily induced trapping of 8-azido-nucleotide in the wild-type enzyme after a short 10 s incubation, and specific photolabeling of Mdr3 after UV irradiation. No such vanadate-induced trapping/photolabeling was observed in any of the mutants, even after a 60 min trapping period at 37 degrees C. Since vanadate trapping with 8-azido-ATP requires hydrolysis of the nucleotide, the data suggest that 8-azido-ATP hydrolysis is dramatically impaired in all of the mutant proteins (<0.3% activity). These results show that mutations in either NB site prevent single turnover and vanadate trapping of nucleotide in the nonmutant site. These results further suggest that the two NB sites cannot function independently as catalytic sites in the intact molecule. In addition, the N- or C-terminal NB sites appear functionally indistinguishable, and cooperative interactions absolutely required for ATP hydrolysis may originate from both sites.

P-glycoprotein shows strong catalytic cooperativity between the two nucleotide sites

P-Glycoprotein (Pgp) (also known as multidrug-resistance protein) contains two nucleotide binding sites, both of which are catalytic ATPase sites. The covalent reagent 7-chloro-4-nitrobenzo-2-oxa-1,3-diazole (NBD-Cl) reacts in catalytic sites, and full inactivation of ATPase activity occurs at a reaction stoichiometry of 1 mol of NBD-Cl/mol of Pgp. We show that, at reaction stoichiometry of < or = 1 mol/mol, both nucleotide sites become labeled in relatively nonselective fashion. There is therefore strong interaction between the two nucleotide sites because (a) reaction of one site with NBD-Cl severely impedes reaction of reagent with the other site, and (b) reaction of one site inhibits steady-state ATPase, i.e. both sites are inhibited. Vanadate-trapping experiments revealed that when one nucleotide site was reacted with NBD-Cl, not even a single ATPase turnover event could occur in the other, intact, nucleotide site. The data demonstrate therefore that catalytic cooperativity between the two nucleotide sites in Pgp is extremely strong and mandatory for catalysis.

A comparison of the phenomenology and genetics of multidrug resistance in cancer cells and quinoline resistance in Plasmodium falciparum

Plasmodium falciparum is the causative agent of the most deadly form of human malaria. Chemotherapy traditionally has been the main line of defense against this parasite, and chloroquine, the drug of choice, has been one of the most successful drugs ever developed. Unfortunately, the evolution and spread of resistance to chloroquine and other quinoline-containing drugs means that these compounds are now virtually useless in many endemic areas. Future prospects for the use of quinoline compounds improved considerably when it was demonstrated that chloroquine resistance could be circumvented in vitro by a number of structurally and functionally unrelated compounds such as verapamil and desipramine. The phenomenon of resistance reversal by compounds such as verapamil is also a key feature of drug resistance in mammalian cells, and this has raised the possibility that the underlying mechanisms of drug resistance of the two cell types could be similar. This hypothesis has prompted a large number of studies into the genetics and biochemistry of resistance to quinoline-containing drugs in P. falciparum. Both the genetic and the biochemical studies have raised issues of controversy and stimulated much debate. These issues are discussed in this review, in the context of a comparison with the genetics and biochemistry of multidrug resistance in mammalian cells.

P-glycoprotein-mediated Hoechst 33342 transport out of the lipid bilayer

High-level expression of P-glycoprotein, a 170-kDa mammalian plasma membrane ATPase, is the cause of an important and widespread form of cancer multidrug resistance. P-glycoprotein reduces cellular accumulation of an enormous variety of lipophilic compounds. The basis for this broad substrate specificity is not well understood. We explored this issue by measuring the kinetics of transport of the lipophilic P-glycoprotein substrate Hoechst 33342 by P-glycoprotein-rich plasma membrane vesicles from CH(R)B30 cells. Hoechst 33342 is fluorescent when bound to the membrane, but not when in the aqueous medium, allowing movement of the dye out of the membrane to be quantitated by fluorescence intensity. The initial specific rate of transport was directly proportional to the amount of Hoechst 33342 in the lipid phase and inversely proportional to the concentration in the aqueous phase. This demonstrates that P-glycoprotein removes Hoechst 33342 from the lipid membrane, where it concentrates due to its hydrophobicity. Because the membrane concentration of hydrophobic P-glycoprotein substrates is high, it may be that P-glycoprotein need not recognize them with high affinity. Transport of hydrophobic substrates out of the lipid bilayer instead of the cytoplasm thus helps to explain the broad substrate specificity of P-glycoprotein.

Extraction of Hoechst 33342 from the cytoplasmic leaflet of the plasma membrane by P-glycoprotein

P-glycoprotein is an ATP-dependent plasma membrane multidrug transporter of broad specificity. A common chemical property of its substrates is that all are lipophilic. Using Hoechst 33342 as the substrate, we have previously shown that P-glycoprotein extracts the substrate directly from the lipid bilayer [Shapiro, A. B., Corder, A. B. & Ling, V. (1997) Eur. J. Biochem. 250, 115-121]. In this paper, we determined the leaflet of the plasma membrane from which P-glycoprotein extracts Hoechst 33342. The initial rate of Hoechst 33342 transport upon ATP addition to P-glycoprotein-rich inside-out plasma membrane vesicles decreased slightly with the amount of time previously elapsed for slow diffusion of Hoechst 33342 to the extracellular leaflet. This result is consistent with transport from the cytoplasmic leaflet. Fluorescence resonance energy transfer from donor Hoechst 33342 to acceptor 2-[6-(7-nitrobenz-2-oxa-1,3-diazol-4-yl)amino]hexanoyl-sn-glycero- 3-phosphocholine (Nbd-C6-HPC) in the cytoplasmic leaflet was used to monitor the amount of Hoechst 33342 in the cytoplasmic leaflet versus time. The initial rate of decrease of the energy-transfer-related Nbd-C6-HPC fluorescence after ATP addition exceeded that of the Hoechst 33342 fluorescence and continued to decrease after decrease of the Hoechst 33342 fluorescence had ceased. These effects were consistent with transport of Hoechst 33342 from the cytoplasmic leaflet to the aqueous interior of the vesicles, followed by rebinding to the extracellular leaflet. This demonstrates that P-glycoprotein transports drugs from the cytoplasmic leaflet of the plasma membrane directly to the aqueous extracellular medium. This finding has implications for efforts to localize the drug-binding site(s) within P-glycoprotein.

Photoaffinity labelling of P-glycoprotein catalytic sites

Photoaffinity labelling of hamster P-glycoprotein was carried out after trapping of radioactive Mg-8-azido-ADP in the catalytic sites by vanadate or beryllium fluoride. With either trapping agent the same labelled peptide was obtained in homogeneous form, with the sequence -FNEVVFNxPTRPDI-, corresponding to residues 1034-1037 in the C-terminal nucleotide binding site. The missing residue 'x' corresponds to Tyr-1041, which is therefore a primary reaction target of 8-azido-ADP. This tyrosine is conserved in all hamster, mouse and human P-glycoproteins. A second major labelled peptide fraction was also identified. The major sequence in this fraction was -NIHFSxPSR-, corresponding to residues 393-401 of hamster P-glycoprotein, where 'x' corresponds to Tyr-398 in the N-terminal nucleotide binding site. Therefore Tyr-398, which is also conserved in other P-glycoproteins, is also a reaction target for 8-azido-ADP. In sequence alignment of the two nucleotide binding sites, Tyr-398 exactly corresponds to Tyr-1041. The data indicate that these two tyrosines lie close to the adenine ring of bound substrate MgATP in the respective catalytic sites of P-glycoprotein.

Pharmacological characterization of the murine and human orthologs of multidrug-resistance protein in transfected human embryonic kidney cells

Overexpression of the human multidrug-resistance protein (MRP) causes a form of multidrug resistance similar to that conferred by P-glycoprotein, although the two proteins are only distantly related. In contrast to P-glycoprotein, human MRP has also been shown to be a primary active transporter of a structurally diverse range of organic anionic conjugates, some of which may be physiological substrates. At present, the mechanism by which MRP transports these compounds and mediates multidrug resistance is not understood. With the objective of developing an animal model for studies on the normal functions of MRP and its ability to confer multidrug resistance in vivo, we recently cloned the murine ortholog of MRP (mrp). To assess the degree of functional conservation between mrp and MRP, we directly compared the drug cross-resistance profiles they confer when transfected into human embryonic kidney cells, as well as their ability to actively transport leukotriene C4, 17beta-Estradiol 17beta-(D-glucuronide), and vincristine; mrp and MRP conferred similar drug resistance profiles, with the exception that only MRP conferred resistance to the anthracyclines tested. Consistent with these findings, accumulation of [3H]vincristine and [3H]VP-16 was decreased, and efflux of [3H]vincristine was increased in both murine and human MRP-transfected cell populations, whereas only human MRP-transfected cells displayed decreased accumulation and increased efflux of [3H]daunorubicin. Membrane vesicles derived from both transfected cell populations transported leukotriene C4 in an ATP-dependent manner with comparable efficiency, although the efficiency of 17beta-estradiol 17beta-(D-glucuronide) transport was somewhat higher with MRP transfectants. ATP-dependent transport of vincristine was also observed with vesicles from mrp and MRP transfectants but only in the presence of glutathione. These studies reveal intrinsic differences between the murine and human MRP orthologs with respect to their ability to confer resistance to a major class of chemotherapeutic drugs.

ATPase activity of P-glycoprotein related to emergence of drug resistance in Ehrlich ascites tumor cell lines

We have characterized the ATPase activity of a sensitive and five progressively daunorubicin resistant Ehrlich ascites tumor cell lines passaged in mice. For the nine different modulators of drug resistance that we have studied, the ATPase activity first rose with the modulator concentration and then declined. We analyzed the ATPase activity profiles in terms of an activation constant and an inhibition constant for each of the nine drugs and six cell lines. In this series of cell lines, the drug-stimulatable ATPase activity was directly proportional to the amount of P-glycoprotein. Pumping of daunorubicin was also correlated with the amount of P-glycoprotein, except that, for a highly passaged line more daunorubicin was pumped than could be accounted for by the content of P-glycoprotein. Between the 12th and the 36th passage an additional source of resistance emerged, which was not correlated with P-glycoprotein. Pumping of daunorubicin was negatively correlated with the cell volume for the different lines.

Mechanisms of multidrug transporters

Drug resistance, mediated by various mechanisms, plays a crucial role in the failure of the drug-based treatment of various infectious diseases. As a result, these infectious diseases re-emerge rapidly and cause many victims every year. Another serious threat is imposed by the development of multidrug resistance (MDR) in eukaryotic (tumor) cells, where many different drugs fail to perform their therapeutic function. One of the causes of the occurrence of MDR in these cells is the action of transmembrane transport proteins that catalyze the active extrusion of a large number of structurally and functionally unrelated compounds out of the cell. The mode of action of these MDR transporters and their apparent lack of substrate specificity is poorly understood and has been subject to many speculations. In this review we will summarize our current knowledge about the occurrence, mechanism and molecular basis of (multi-)drug resistance especially as found in bacteria.

Inhibition of P-glycoprotein ATPase activity by beryllium fluoride

ATPase activity of P-glycoprotein (multidrug-resistance protein) was found to be potently inhibited by beryllium fluoride (BeFx) in combination with MgATP, MgADP, or corresponding Mg-8-azido-nucleotides. Inhibition was due to trapping of nucleoside diphosphate at catalytic sites. Full inhibition was achieved on trapping of 1 mol of nucleotide per mol of Pgp. Reactivation was slow (t(1/2) = 32 min at 37 degrees C), and release of trapped nucleotide correlated with recovery of ATPase. Trapping of 8-azido-ADP followed by UV irradiation yielded permanent inactivation and specific labeling of Pgp in plasma membranes. Both N- and C-terminal nucleotide binding sites were labeled. These findings give strong confirmation of the concepts that in intact Pgp both nucleotide sites are active in MgATP hydrolysis, and that they interact strongly. The characteristics of inhibition by BeFx were similar in general to those seen with vanadate. However, PPi gave strong protection against BeFx inhibition, and in this respect, inhibition by BeFx was clearly different from vanadate inhibition.

P-glycoprotein mediated multidrug resistance and its implications for pathology

The discovery of P-glycoprotein has revealed a fundamental mechanism by which cancer cells evade chemotherapy and this principle has proven relevant to general cellular defence mechanisms in normal physiology. To date this knowledge has promised to improve current cancer chemotherapy through the manipulation of drug combinations according to the P-glycoprotein status of the tumor. Furthermore, the discovery of inhibitors of the protein may provide new therapeutic tools in the treatment of multidrug resistant neoplasia, provided the benefits are deemed greater than the potential detrimental side effects. When looking towards future therapies, however, we must also consider additional mechanisms which undoubtedly contribute to clinical drug resistance. Complete elucidation of this complex cellular defence network will hopefully translate into therapeutic opportunities to circumvent all mechanisms of multidrug resistance, thus positively impacting on patient survival.

Inhibition of P-glycoprotein ATPase activity by procedures involving trapping of nucleotide in catalytic sites

Fluoroaluminate in combination with nucleotide inhibited ATPase activity of P-glycoprotein (Pgp) in plasma membranes and in pure reconstituted form. Low nucleotide concentrations were effective, e.g., half-maximal inhibition was obtained with 10 microM MgATP. With MgATP or MgADP, reactivation occurred with t1/2 = 7 min at 37 degrees C. With 8-azido-ATP, UV irradiation of inhibited Pgp gave specific photolabeling of both nucleotide sites. Fluoroaluminate therefore provides a valuable tool for functional and structural characterization of P-glycoprotein and probably of other ABC transporters. 2-Azido-ATP, in combination with vanadate, fluoroaluminate, or beryllium fluoride, inhibited Pgp ATPase activity. Low concentrations of 2-azido-ATP were effective. However, after UV irradiation of the inhibited Pgp species, in no case was there evidence of covalent labeling of nucleotide sites. Therefore in the Pgp catalytic sites, under conditions of nucleotide trapping, there is no suitable amino acid side chain adjacent to the photoactivated 2-position of bound 2-azido-nucleotide, and 8-azido-ATP is the preferred photolabeling analog.

Biophysical aspects of P-glycoprotein-mediated multidrug resistance

In the 45 years since Burchenal's observation of chemotherapeutic drug resistance in tumor cells, many investigators have studied the molecular basis of tumor drug resistance and the phenomenon of tumor multidrug resistance (tumor MDR). Examples of MDR in microorganisms have also become topics of intensive study (e.g., Plasmodium falciparum MDR and various types of bacterial MDR) and these emerging fields have, in some cases, borrowed language, techniques, and theories from the tumor MDR field. Serendipitously, the cloning of MDR genes overexpressed in MDR tumor cells has led to elucidation of a large family of membrane proteins [the ATP-binding cassette (ABC) proteins], an important subset of which confer drug resistance in many different cells and microorganisms. In trying to decipher how ABC proteins confer various forms of drug resistance, studies on the structure and function of both murine and human MDR1 protein (also called P-glycoprotein or P-gp) have often led the way. Although various theories of P-gp function have become popular, there is still no precise molecular-level description for how P-gp overexpression lowers intracellular accumulation of chemotherapeutic drugs. In recent years, controversy has developed over whether the protein protects cells by translocating drugs directly (as some type of drug pump) or indirectly (through modulating biophysical parameters of the cell). In this ongoing debate over P-gp function, detailed consideration of biophysical issues is critical but has often been neglected in considering cell biological and pharmacological issues. In particular, P-gp overexpression also changes plasma membrane electrical potential (delta psi zero) and intracellular pH (pHi), and these changes will greatly affect the cellular flux of a large number of compounds to which P-gp overexpression confers resistance. In this chapter, we highlight these biophysical issues and describe how delta psi zero and pHi may in fact be responsible for many MDR-related phenomena that have often been hypothesized to be due to direct drug translocation (e.g., drug pumping) by P-gp.

Proton-dependent multidrug efflux systems

Multidrug efflux systems display the ability to transport a variety of structurally unrelated drugs from a cell and consequently are capable of conferring resistance to a diverse range of chemotherapeutic agents. This review examines multidrug efflux systems which use the proton motive force to drive drug transport. These proteins are likely to operate as multidrug/proton antiporters and have been identified in both prokaryotes and eukaryotes. Such proton-dependent multidrug efflux proteins belong to three distinct families or superfamilies of transport proteins: the major facilitator superfamily (MFS), the small multidrug resistance (SMR) family, and the resistance/ nodulation/cell division (RND) family. The MFS consists of symporters, antiporters, and uniporters with either 12 or 14 transmembrane-spanning segments (TMS), and we show that within the MFS, three separate families include various multidrug/proton antiport proteins. The SMR family consists of proteins with four TMS, and the multidrug efflux proteins within this family are the smallest known secondary transporters. The RND family consists of 12-TMS transport proteins and includes a number of multidrug efflux proteins with particularly broad substrate specificity. In gram-negative bacteria, some multidrug efflux systems require two auxiliary constituents, which might enable drug transport to occur across both membranes of the cell envelope. These auxiliary constituents belong to the membrane fusion protein and the outer membrane factor families, respectively. This review examines in detail each of the characterized proton-linked multidrug efflux systems. The molecular basis of the broad substrate specificity of these transporters is discussed. The surprisingly wide distribution of multidrug efflux systems and their multiplicity in single organisms, with Escherichia coli, for instance, possessing at least nine proton-dependent multidrug efflux systems with overlapping specificities, is examined. We also discuss whether the normal physiological role of the multidrug efflux systems is to protect the cell from toxic compounds or whether they fulfil primary functions unrelated to drug resistance and only efflux multiple drugs fortuitously or opportunistically.

Altered drug translocation mediated by the MDR protein: direct, indirect, or both?

Overexpression of the MDR protein, or p-glycoprotein (p-GP), in cells leads to decreased initial rates of accumulation and altered intracellular retention of chemotherapeutic drugs and a variety of other compounds. Thus, increased expression of the protein is related to increased drug resistance. Since several homologues of the MDR protein (CRP, ItpGPA, PDR5, sapABCDF) are also involved in conferring drug resistance phenomena in microorganisms, elucidating the function of the MDR protein at a molecular level will have important general applications. Although MDR protein function has been studied for nearly 20 years, interpretation of most data is complicated by the drug-selection conditions used to create model MDR cell lines. Precisely what level of resistance to particular drugs is conferred by a given amount of MDR protein, as well as a variety of other critical issues, are not yet resolved. Data from a number of laboratories has been gathered in support of at least four different models for the MDR protein. One model is that the protein uses the energy released from ATP hydrolysis to directly translocate drugs out of cells in some fashion. Another is that MDR protein overexpression perturbs electrical membrane potential (delta psi) and/or intracellular pH (pHi) and thereby indirectly alters translocation and intracellular retention of hydrophobic drugs that are cationic, weakly basic, and/or that react with intracellular targets in a pHi or delta psi-dependent manner. A third model proposes that the protein alternates between drug pump and Cl- channel (or channel regulator) conformations, implying that both direct and indirect mechanisms of altered drug translocation may be catalyzed by MDR protein. A fourth is that the protein acts as an ATP channel. Our recent work has tested predictions of these models via kinetic analysis of drug transport and single-cell photometry analysis of pHi, delta psi, and volume regulation in novel MDR and CFTR transfectants that have not been exposed to chemotherapeutic drugs prior to analysis. This paper reviews these data and previous work from other laboratories, as well as relevant transport physiology concepts, and summarizes how they either support or contradict the different models for MDR protein function.

Characterization and epitope mapping of several new anti-P-glycoprotein monoclonal antibodies

Monoclonal antibodies (MAbs) were raised against partially purified Class I P-glycoprotein from multidrug-resistant Chinese hamster ovary CHRB30 cells. Fifteen stable monoclonal hybridoma cell lines were established, and the secreted antibodies were classified into 8 groups on the basis of banding pattern on immunoblots of P-glycoprotein digested with cyanogen bromide or partially digested with proteases. One representative of each group was tested further for several activities. Six of the 8 recognized human P-glycoprotein in the multidrug-resistant SKVLBI cell line. None of the antibodies recognized P-glycoprotein in unfixed cells, suggesting that all recognize cytoplasmic epitopes or extracellular epitopes not accessible in native P-glycoprotein. All 8 antibodies were able to immunoprecipitate P-glycoprotein from non-denaturing detergent solution. The linear epitopes of the antibodies were mapped to 11-27 amino acids. Two of the antibodies had epitopes in the linker region, 3 in the N-terminal nucleotide binding domain, 2 in the C-terminal nucleotide binding domain and 1 in the predicted cytoplasmic loop between predicted transmembrane helices 8 and 9. These antibodies, with known epitopes, could have uses for P-glycoprotein detection, structure/function studies, purification and quantitation.

The application of pH-sensitive fluorescent dyes in lactic acid bacteria reveals distinct extrusion systems for unmodified and conjugated dyes

Intracellular pH in bacteria can be measured efficiently between internal pH values of 6.5 and 8.5 with the fluorescent pH indicator 2',7'-bis-(2-carboxyethyl)-5[and-6]-carboxyfluorescein (BCECF). A new fluorescent pH probe with a lower pKa(app) than BCECF was synthesized from fluorescein isothiocyanate and glutamate. The new probe, N-(fluorescein thio-ureanyl)-glutamate (FTUG), was much less sensitive to changes in concentrations of KCl than was BCECF. Similar to BCECF, an efflux of FTUG independent of the proton motive force, but dependent on ATP, was observed both in Lactobacillus plantarum and Lactococcus lactis. Corrections for probe efflux allowed accurate measurements of the pHin. Similar intracellular pH values were determined with FTUG and BCECF, in the range where both probes can be applied, and the pH values correlated well with those estimated from the distribution of radio-labelled benzoic acid. Since FITC can easily be coupled to substrates containing an amino group, it is possible to develop other FITC derivatives as well. The mechanisms of probe excretion and the nature of the excreted product(s) were studied in further detail for BCECF and FTUG. BCECF was excreted from wild-type L. lactis in an unmodified form as was determined by chromatographic and mass spectrometry analysis. In the case of FTUG, the excreted product was a conjugated derivative. Unmodified FTUG was not excreted, although it was present in cellular extracts from L. lactis. Exit of BCECF was completely inhibited in a BCECF efflux mutant (Bef-) of L. lactis, whereas FTUG-conjugate efflux in this mutant was similar to the wild-type. Addition of indomethacin, a known inhibitor of BCECF efflux in human epithelial cells, resulted in complete inhibition of BCECF efflux in wild-type L. lactis, whereas FTUG-conjugate exit was only slightly affected. The results of the mutant and inhibitor studies suggest that FTUG-conjugate and BCECF efflux in L. lactis are mediated by different ATP-driven extrusion systems for organic anions.