Expression of the multidrug resistance-associated protein (MRP) gene in human cancers

We determined the expression of a newly recognized drug resistance gene, the multidrug resistance-associated protein (MRP) gene, [Cole et al., Science (Washington DC), 258: 1650-1654, 1992], in normal human tissues and in >370 human tumor biopsies using a quantitative RNase protection assay and immunohistochemistry. MRP mRNA appeared to be ubiquitously expressed at low levels in all normal tissues, including peripheral blood, the endocrine glands (adrenal and thyroid), striated muscle, the lymphoreticular system (spleen and tonsil), the digestive tract (salivary gland, esophagus, liver, gall bladder, pancreas, and colon), the respiratory tract (lung), and the urogenital tract (kidney, bladder, testis, and ovary). The human cancers analyzed could be divided into three groups with regard to MRP expression. Group 1 consists of tumors that often exhibit high to very high MRP mRNA levels (e.g., chronic lymphocytic leukemia). Group 2 comprises the tumors that often exhibit low, but occasionally exhibit high MRP mRNA expression (e.g., esophagus squamous cell carcinoma, non-small cell lung cancer, and acute myelocytic leukemia). Group 3 comprises the tumors with predominantly low levels of MRP mRNA, comparable to the levels found in normal tissues (e.g., other hematological malignancies, soft tissue sarcomas, melanoma, and cancers of the prostate, breast, kidney, bladder, testis, ovary, and colon). Using the MRP-specific mAbs MRPr1 and MRPm6, we confirmed the elevated MRP mRNA levels in tumor tissues by immunohistochemistry. We conclude that hyperexpression of MRP is observed in several human cancers, and that additional studies are needed to assess the clinical relevance of MRP.     

Expression of the multidrug resistance-associated protein (MRP) gene in urothelial carcinomas

The intrinsic or acquired resistance of urothelial cancer to chemotherapy is one major obstacle to successful treatment. Generally, the expression level of P-glycoprotein in urothelial cancer is low, so we accordingly investigated the expression of multidrug resistance-associated protein (MRP). We examined the expression of MRP mRNA by means of slot-blotting samples of 11 renal pelvic and/or ureteral tumors, 33 bladder tumors, one lung metastasis from a ureter tumor, 7 non-cancerous urothelia from patients with transitional-cell carcinoma (TCC) and one urothelium from a patient with renal-cell carcinoma (RCC). We also estimated, by Southern blotting, whether or not the MRP gene was amplified in clinical specimens that overexpressed MRP mRNA. MRP was detected immunohistochemically using a polyclonal antibody against MRP. In all, 5 of 11 renal pelvic and/or ureter tumors (45.5%), 17 of 33 bladder tumors (51.5%) and 4 of 7 non-cancerous urothelia of TCC patients (57.1%) expressed more than 2-fold the MRP mRNA levels of drug-sensitive human KB cells. There was no significant difference in the MRP mRNA level between primary and recurrent tumors. Low-grade urothelial carcinomas (G1 and G2 TCCs) expressed significantly higher levels of MRP mRNA than the high-grade G3 TCC. The MRP gene was not amplified in urothelial carcinomas, irrespective of their expression levels of MRP mRNA. Immunohistochemically, MRP was located mainly on the plasma membrane, but also detected on the cytoplasm of cancer cells. MRP may be one mechanism responsible for intrinsic drug resistance in low-grade urothelial cancer.     

Expression of multidrug resistance-associated protein (MRP) in brain microvessel endothelial cells

Multidrug resistance-associated protein (MRP) is a recently identified drug efflux transport system that actively transports organic acids and selected glucuronide or glutathione conjugates out of the cell. The current study presents, for the first time, both functional and biochemical data demonstrating the presence of MRP in the brain microvessel endothelial cells that form the blood-brain barrier (BBB). Using known MRP inhibitors, such as indomethacin and probenecid, fluorescein accumulation in primary cultured bovine brain microvessel endothelial cell (BBMEC) monolayers was significantly enhanced compared to control. The specificity of the MRP inhibitors on cellular fluorescein accumulation was confirmed using both MRP positive (Panc-1) and MRP negative (KBv) cell lines. Furthermore, western blot analysis using a specific antibody for MRP (MRPm6) and RT-PCR studies using a complementary sequence probe for human MRP demonstrate the expression of MRP in BBMEC. Previous studies have demonstrated the significance of the P-glycoprotein drug efflux transporter in the BBB. Given its function as a drug efflux transport system, it is anticipated that MRP in the BBB will also have an important role in limiting the exposure of the brain to many endogenous and exogenous compounds, including both toxic and therapeutic agents.     

Mrp1 multidrug resistance-associated protein and P-glycoprotein expression in rat brain microvessel endothelial cells

Two membrane glycoproteins acting as energy-dependent efflux pumps, mdr-encoded P-glycoprotein (P-gp) and the more recently described multidrug resistance-associated protein (MRP), are known to confer cellular resistance to many cytotoxic hydrophobic drugs. In the brain, P-gp has been shown to be expressed specifically in the capillary endothelial cells forming the blood-brain barrier, but localization of MRP has not been well characterized yet. Using RT-PCR and immunoblot analysis, we have compared the expression of P-gp and Mrp1 in homogenates, isolated capillaries, primary cultured endothelial cells, and RBE4 immortalized endothelial cells from rat brain. Whereas the mdr1a P-gp-encoding mRNA was specifically detected in brain microvessels and mdr1b mRNA in brain parenchyma, mrp1 mRNA was present both in microvessels and in parenchyma. However, Mrp1 was weakly expressed in microvessels. Mrp1 expression was higher in brain parenchyma, as well as in primary cultured brain endothelial cells and in immortalized RBE4 cells. This Mrp1 overexpression in cultured brain endothelial cells was less pronounced when the cells were cocultured with astrocytes. A low Mrp activity could be demonstrated in the endothelial cell primary monocultures, because the intracellular [3H]vincristine accumulation was increased by several MRP modulators. No Mrp activity was found in the cocultures or in the RBE4 cells. We suggest that in rat brain, Mrp1, unlike P-gp, is not predominantly expressed in the blood-brain barrier endothelial cells and that Mrp1 and the mdr1b P-gp isoform may be present in other cerebral cells.     

Expression of multidrug resistance-associated protein (MRP) in head and neck squamous cell carcinoma

Vascular parkinsonism is thought to be a distinct parkinsonian syndrome associated with small deep infarcts and white matter lesions (WMLs). We studied the prevalence of parkinsonian features (bradykinesia, rigidity, tremor, and gait disorder) in relation to small deep or territorial infarcts and WMLs on computed tomography (CT) in 62 lacunar and 41 territorial stroke patients, at 3.0 (median) years of follow up. One or more parkinsonian signs were found in 36% of these patients; 11% clinically had parkinsonism. Parkinsonian signs were found more frequently in lacunar than in territorial stroke patients: bradykinesia in 45% and 7%, rigidity in 13% and 7%, tremor in 6% and 7%, and gait disorder in 16% and 7%, respectively. Patients with WMLs at study entry (n = 16) were compared with those without WMLs (n = 87): 56% and 25% had bradykinesia, 25% and 8% rigidity, 25% and 3% tremor, and 38% and 8% gait disorder, respectively. Regression analysis with adjusted odds ratios ([a]OR) showed that WMLs at study entry were associated with bradykinesia ([a]OR 8.0, 95% confidence interval [CI] 1.6-41.6), gait disorder ([a]OR 7.1, 95% CI 1.5-33.7), and tremor ([a]OR 7.0, 95% CI 1.2-40.3). Bradykinesia was associated with lacunar stroke at study entry ([a]OR 11.5, 95% CI 2.4-54.9). Thus, one third of our stroke patients had one or more parkinsonian signs, and 10% clinically had a parkinsonian syndrome that differed from Lewy body parkinsonism: infrequent resting tremor, but frequent gait disorder. Parkinsonian signs were associated with WMLs and lacunar stroke. Therefore, this study favors a distinct vascular parkinsonian syndrome.     

Epitope insertion favors a six transmembrane domain model for the carboxy-terminal portion of the multidrug resistance-associated protein

The overexpression of the multidrug resistance protein, MRP, in mammalian cells is associated with pleiotropic resistance to cytotoxic drugs. MRP is an integral membrane protein which belongs to the family of ATP-binding cassette transporters. Secondary structure predictions combined with biochemical analyses suggest that MRP encodes 11 transmembrane (TM) domains in the amino-terminal half of the protein and four or six transmembrane domains in the carboxy-terminal half of the protein. To gain insight into the membrane topology of the carboxy-terminal half of MRP, small, antigenic hemagglutinin (HA) epitopes (YPYDVPDYAS) were inserted within six predicted hydrophilic subfragments of this region (938, 1001, 1084, 1175, 1222, 1295). These epitope-tagged MRP variants were expressed in HeLa cells to evaluate their ability to confer resistance to the drug etoposide (VP-16). Insertion of the HA epitopes at positions 938, 1001, and 1222 resulted in functional proteins, while epitope insertion at positions 1084, 1175, and 1295 abrogated MRP function. The intracellular versus extracellular location of the HA epitopes present in biologically active MRP variants was then established in intact and permeabilized cells by immunofluorescence using an anti-HA antibody. Epitopes inserted at positions 1001 and 1222 were located on the extracellular side of the plasma membrane, while the epitope inserted at position 938 was located intracellularly. These results are consistent with a six TM rather than a four TM domain model for the membrane portion of the carboxy-terminal half of MRP.     

Retrovirus-mediated gene transfer of the multidrug resistance-associated protein (MRP) cDNA protects cells from chemotherapeutic agents

Transduction of hematopoietic progenitors with a multidrug resistance gene like mdr-1 or mrp aims to protect bone marrow from toxicity of chemotherapeutic agents. The interest in the use of mrp as an alternative to mdr-1 gene transfer for bone marrow protection lies in its different modulation. Indeed, classical P-gp reversal agents, tested in the clinic to decrease mdr-1 tumor resistance, have little or no effect on MRP function. This would allow, in the same patient, the use of reversal agents to decrease P-gp tumor resistance without reversing bone marrow protection of the transduced hematopoietic cells provided by multidrug resistance-associated protein (MRP). As a first step, we have constructed and tested two different mrp-containing vectors with either the Harvey retroviral long terminal repeat (LTR) or PGK as promoters and generated ecotropic producer cells. We have shown by Southern blot analysis that retroviral supernatant from these producer cells can efficiently transmit the mrp gene to target cells. Mrp expression could be detected by fluorescence-activated cell sorting (FACS) analysis in the producer cells. The transduced cells have increased resistance to doxorubicin, vincristine, and etoposide. Furthermore, chemoprotection of the transduced cells was increased after selection with chemotherapeutic agents in the presence of glutathione, a co-factor for MRP function. These data indicate that mrp retroviral vectors may be useful for chemoprotection and selection.     

Adenosine triphosphate-dependent transport of anionic conjugates by the rabbit multidrug resistance-associated protein Mrp2 expressed in insect cells

The multidrug resistance-associated protein Mrp2 is expressed in liver, kidney, and small intestine and mediates ATP-dependent transport of conjugated organic anions across the apical membrane of epithelial cells. We recently cloned a rabbit cDNA encoding a protein that on basis of highest amino acid homology and tissue distribution was considered to be the rabbit homolog of rat Mrp2. To investigate whether rabbit Mrp2 mediates ATP-dependent transport similar to rat Mrp2, we expressed rabbit Mrp2 in Spodoptera frugiperda (Sf9) cells using recombinant baculovirus. Mrp2 was expressed as an underglycosylated protein in Sf9 cells and to a higher level compared with rabbit liver and renal proximal tubules. Both 17beta-estradiol-17-beta-D-glucuronide ([3H]E217betaG, 50 nM) and [3H]leukotriene C4 (3 nM) were taken up by Sf9-Mrp2 membrane vesicles in an ATP-dependent fashion. Uptake of [3H]E217betaG was dependent on the osmolarity of the medium and saturable for ATP (Km = 623 microM). Leukotriene C4, MK571, phenolphthalein glucuronide, and fluorescein-methotrexate were good inhibitors of [3H]E217betaG transport. The inhibitory potency of cyclosporin A and methotrexate was moderate, whereas fluorescein, alpha-naphthyl-beta-D-glucuronide, and p-nitrophenyl-beta-D-glucuronide did not inhibit transport. In conclusion, we show direct ATP-dependent transport by recombinant rabbit Mrp2 and provide new data on Mrp2 inhibitor specificity.     

Membrane topology of the Escherichia coli TolR protein required for cell envelope integrity

TolR is a 142-amino-acid protein required for the import of colicins and bacteriophage and for maintenance of cell envelope integrity. The topology of TolR in the inner membrane was analyzed by two methods. First, bacteria expressing a series of TolR-beta-galactosidase, TolR-alkaline phosphatase, and TolR-beta-lactamase fusions were assayed for the appropriate enzymatic activity. Second, the accessibility of TolR to proteinase K was determined in permeabilized cells and everted vesicles with an antibody elicited against the carboxyl-terminal 70% of TolR. The results are consistent with TolR spanning the inner membrane once via residues 23 to 43 and with the carboxyl-terminal moiety being exposed to the periplasm. Quantitative studies with the anti-TolR antibody indicated the presence of 2 x 10(3) to 3 x 10(3) TolR molecules per cell.     

Dual topology of the large envelope protein of duck hepatitis B virus: determinants preventing pre-S translocation and glycosylation

The biosynthesis and topology of the large envelope protein (L protein) of hepadnaviruses was investigated using the duck hepatitis B virus (DHBV) model, which also allows the study of hepadnavirus morphogenesis in experimentally infected hepatocytes. Results from proteolysis of virus particles and from the analysis of topology and posttranslational modification of L chains synthesized in vivo or in a cell-free system both support the presence of a mixed population of L-protein molecules with their N-terminal pre-S domain located either inside or outside the virus particle. During L biosynthesis and DHBV morphogenesis, pre-S, together with the neighboring transmembrane domain (TM-I), initially remained cytoplasmically disposed and was translocated only posttranslationally. Delayed pre-S translocation into a post-endoplasmic reticulum compartment is also indicated by the absence of glycosylation at a modification-competent pre-S glycosylation site. Major features of L-protein biosynthesis and of the resulting dual topology appear to be conserved between avian and mammalian hepadnaviruses, supporting the model that pre-S domains function in part either as an internal matrix for capsid envelopment or externally as a ligand for cellular receptor binding. However, differences in the mechanisms controlling pre-S translocation were revealed by the results of mutational analyses identifying and characterizing cis-acting determinants in pre-S that delay its cotranslational translocation. Our data from DHBV demonstrate the negative influence of a cluster of positively charged amino acid residues next to TM-I, a motif that is conserved among the avian but absent from mammalian hepadnaviruses. Additional control elements, which are apparently shared between both virus groups and which may serve in chaperone binding, were mapped by deletion analysis in the central part of pre-S.     

Membrane topology of the human dipeptide transporter, hPEPT1, determined by epitope insertions

We have used epitope insertion to analyze the transmembrane topology of the human H+-dipeptide symporter hPEPT1. An epitope tag, EYMPME (EE), was inserted into different locations at amino acids 39, 78, 106, 412, and 708 of hPEPT1 by site-directed mutagenesis. The functional integrity of the tagged protein was tested by measuring its dipeptide transport activity in transfected Cos7 cells. Further, cells expressing hPEPT1 or EE-tagged hPEPT1 derivatives were labeled with an anti-EE-monoclonal antibody (anti-EE-mAb) or an antiserum raised against the carboxyl terminus of hPEPT1 (anti-hPEPT1) and examined by immunofluorescence confocal microscopy. EE106-, 412-, and 708-hPEPT1 transported the dipeptide tracer as well as wild-type hPEPT1. Tags at position 106 and 412 were shown to be extracellular because they were accessible to anti-epitope antibody in nonpermeabilized cells. In contrast, the carboxyl-terminal domain and EE708 were shown to be intracellular since they were only accessible to the antibodies in permeabilized cells. These results are consistent with a 12-transmembrane domain (TMD) topological model of PEPT1. Epitope insertions at regions linking the putative TMD1 and TMD2, and TMD2 and TMD3 (EE39- and EE78-hPEPT1), abolished the dipeptide transport into the cells. In transfected Cos7 cells, these tagged proteins remained largely intracellular rather than at the plasma membrane. These results suggest that the integrity of these regions is essential for transporter trafficking and/or function. Thus, the topology of the amino-terminal portion, including putative TMD1 and -2, remains to be clarified.     

Altered drug-stimulated ATPase activity in mutants of the human multidrug resistance protein

The characteristics of P-glycoprotein (MDR1), an ATP-dependent drug extrusion pump responsible for the multidrug resistance of human cancer, were investigated in an in vitro expression system. The wild-type and several mutants of the human MDR1 cDNA were engineered into recombinant baculoviruses and the mutant proteins were expressed in Sf9 insect cells. In isolated cell membrane preparations of the virus-infected cells the MDR1-dependent drug-stimulated ATPase activity, and 8-azido-ATP binding to the MDR1 protein were studied. We found that when lysines 433 and/or 1076 were replaced by methionines in the ATP-binding domains, all these mutations abolished drug-stimulated ATPase activity independent of the MgATP concentrations applied. Photoaffinity labeling with 8-azido-ATP showed that the double lysine mutant had a decreased ATP-binding affinity. In the MDR1 mutant containing a Gly185 to Val replacement we found no significant alteration in the maximum activity of the MDR1-ATPase or in its activation by verapamil and vinblastine, and this mutation did not modify the MgATP affinity or the 8-azido-ATP binding of the transporter either. However, the Gly185 to Val mutation significantly increased the stimulation of the MDR1-ATPase by colchicine and etoposide, while slightly decreasing its stimulation by vincristine. These shifts closely correspond to the effects of this mutation on the drug-resistance profile, as observed in tumor cells. These data indicate that the Sf9-baculovirus expression system for MDR1 provides an efficient tool for examining structure-function relationships and molecular characteristics of this clinically important enzyme.     

Membrane topology of Borrelia burgdorferi and Treponema pallidum lipoproteins

A critical issue regarding the molecular architectures of Treponema pallidum and Borrelia burgdorferi, the agents of venereal syphilis and Lyme disease, respectively, concerns the membrane topologies of their major lipoprotein immunogens. A related question is whether these lipid-modified membrane proteins form intramembranous particles during freeze fracture electron microscopy. To address these issues, native borrelial and treponemal lipoproteins were reconstituted into liposomes of diverse composition. The importance of the covalently associated lipids for membrane association of lipoproteins was revealed by the observation that nonlipidated recombinant forms of both B. burgdorferi OspA and the T. pallidum 47-kDa immunogen (Tpp47) showed very weak or no binding to model bilayer vesicles. In contrast to control liposomes reconstituted with bacteriorhodopsin or bovine rhodopsin, two well-characterized transmembrane proteins, none of the lipoprotein-liposomes contained particles when examined by freeze fracture electron microscopy. To extend these findings to prokaryotic lipoproteins with relatively amphiphilic polypeptides, similar experiments were conducted with a recombinant nonlipidated form of Escherichia coli TraT, a lipoprotein which has putative transmembrane domains. The nonlipidated TraT oligomers bound vesicles derived from E. coli lipids but, surprisingly, did not form particles in the freeze-fractured liposomes. These findings support (i) a proposed topology of spirochetal lipoproteins in which the polypeptide is extrinsic to the membrane surface and (ii) the contention that particles visualized in freeze-fractured spirochetal membranes represent poorly characterized transmembrane proteins.     

Evolving views of protein glycosylation

BACKGROUND: To study if the presence of the G/A polymorphism at the apo A-I gene promoter region could determine the lipid profile in patients with hyperlipidemia after heart transplantation, or if it is related with the type of heart disease that determined the transplantation. PATIENTS AND METHODS: This study included 31 patients with hyperlipidemia after heart transplantation. Anthropometric parameters, basic analytic and lipid study were measured in these subjects. Identification of the G/A mutation in the promoter region of the apo A-I gene was performed. RESULTS: 22 patients had the G/G genotype and 9 the G/A. 14 were transplanted by coronary heart disease and 17 by non ischemic heart disease. Patients with the A allele had higher cHDL (63 [SD 15] vs 53 [10]; p = 0.034) and apo A-I plasma levels (156 [34] vs 132 [24]; p = 0.040) than G/G subjects. The A allele was present in the 18% of the patients transplanted by ischemic heart disease and in the 43% of the transplanted by another etiology (p = 0.073). CONCLUSIONS: The presence of the G/A genotype in the promoter region of the apo A-I gene determines higher plasma levels of cHDL in patients with hyperlipidemia after heart transplantation.     

A possible role for multidrug resistance-associated protein in the secretion of basic fibroblast growth factor by osteogenic sarcoma cell line (MG-63)

Basic fibroblast growth factor (bFGF) lacks signal sequence and therefore the mechanism of its secretion is unknown. Multidrug resistance-associated protein (MRP) has been shown to transport a variety of molecules. Therefore, in this study we examined the role of MRP in the secretion of bFGF by osteogenic sarcoma MG-63 cells which spontaneously secrete bFGF. We show that MG-63 cells express MRP both at the protein and at the mRNA level. Furthermore, probenecid (a putative inhibitor of transport activity of MRP), in a concentration-dependent manner, inhibited secretion of bFGF from MG-63 cells with concomitant increase in intracellular contents of bFGF. These results suggest that MRP may have a possible role in the secretion of bFGF.     

Role of glycosylation in expression and function of the human parathyroid hormone/parathyroid hormone-related protein receptor

Parathyroid hormone (PTH) regulates calcium metabolism through a specific G protein-coupled, seven-transmembrane helix-containing receptor. This receptor also binds and is activated by PTH-related protein (PTHrP). The human (h) PTH/PTHrP receptor is a membrane glycoprotein with an apparent molecular weight of approximately 85000 which contains four putative N-glycosylation sites. To elucidate the functional role of receptor glycosylation, if any, we studied hormone binding and signal transduction in human embryonic kidney cells transfected with hPTH/PTHrP receptor (HEK-293/C-21). These cells stably express 300000-400000 receptors per cell. Inhibition of N-glycosylation with an optimized concentration of tunicamycin yielded completely nonglycosylated hPTH/PTHrP receptor (approximately 60 kDa). This receptor form is fully functional; it maintains nanomolar binding affinity for PTH- and PTHrP-derived agonists and antagonists. PTH and PTHrP agonists stimulate cyclic AMP accumulation and increases in cytosolic calcium levels. In addition, the highly potent benzophenone (pBz2)-containing PTH-derived radioligand [Nle8,18,Lys13(epsilon-pBz2),L-2-Nal23,Tyr34 3-125I)]bPTH(1-34)NH2 can photoaffinity cross-link specifically to the nonglycosylated receptor. The molecular weight (approximately 60000) of the band representing the photo-cross-linked, nonglycosylated receptor (obtained from the tunicamycin-treated HEK-293/C-21 cells) was similar to that of the deglycosylated photo-cross-linked receptor (obtained by enzymatic treatment with Endoglycosidase-F/N-glycosidase-F). Our findings indicate that glycosylation of the hPTH/PTHrP receptor is not essential for its effective expression on the plasma membrane or for the binding of ligands known to interact with the native receptor. The nonglycosylated hPTH/PTHrP receptor remains fully functional with regard to both of its known signal transduction pathways: cAMP-protein kinase A and phospholipase C-cytosolic calcium.