[Show abstract][Hide abstract] ABSTRACT: ATP Binding Cassette (ABC) transporter, ABCA1, plays a pivotal role in reverse cholesterol transport by mediating the cellular efflux of phospholipid and cholesterol. Studies using intact cells strongly suggest that ABCA1 acts as a phospholipid floppase, but there has been no direct demonstration that the protein is a primary active sterol transporter. Using membrane vesicles from insect Sf21 cells, we found that ABCA1 mediated ATP-dependent uptake of [(3)H]25-hydroxycholesterol with an apparent K(m) of 0.7 muM. Consistent with this high apparent affinity, expression of ABCA1 in human embryonic kidney cells both increased rapid efflux of 25-hydroxcholesterol and prevented oxysterol-mediated repression of low-density lipoprotein (LDL) receptor and 3-hydroxy-3-methylglutaryl-coenzyme A (HMG-CoA) reductase mRNAs. Comparison of wild-type and ABCA1(-/-) murine fibroblasts indicates that 25-hydroxycholesterol is effluxed approximately 5-fold more rapidly by wild-type cells. In addition, the rate of efflux from the wild-type but not the ABCA1(-/-) fibroblasts is increased a further twofold by inducers of ABCA1 expression. Thus under the experimental conditions employed, endogenous ABCA1 is a major contributor to 25-hydroxycholesterol efflux from wild-type fibroblasts. Evidence from in vitro studies indicates that oxysterols are potent inducers of genes involved in cellular cholesterol efflux and metabolism, including the ABCA1 gene, and repressors of genes involved in cholesterol synthesis or uptake. Our observations raise the possibility that efflux of oxysterols by ABCA1 could contribute to a homeostatic mechanism, which both attenuates oxysterol-induced expression of its cognate gene and alleviates repression of genes encoding proteins, such as HMG-CoA reductase and LDL receptor.
[Show abstract][Hide abstract] ABSTRACT: Human multidrug resistance protein 1 (MRP1) has a total of 17 transmembrane (TM) helices arranged in three membrane-spanning domains, MSD0, MSD1, and MSD2, with a 5 + 6 + 6 TM configuration. Photolabeling studies indicate that TMs 10 and 11 in MSD1 and 16 and 17 in MSD2 contribute to the substrate binding pocket of the protein. Previous mutational analyses of charged and polar amino acids in predicted TM helices 11, 16, and 17 support this suggestion. Mutation of Trp(553) in TM10 also affects substrate specificity. To extend this analysis, we mutated six additional polar residues within TM10 and the remaining uncharacterized polar residue in TM16, Asn(1208). Although mutation of Asn(1208) was without effect, two of six mutations in TM10, T550A and T556A, modulated the drug resistance profile of MRP1 without affecting transport of leukotriene C4, 17beta-estradiol 17-(beta-d-glucuronide) (E(2)17betaG), and glutathione. Mutation T550A increased vincristine resistance but decreased doxorubicin resistance, whereas mutation T556A decreased resistance to etoposide (VP-16) and doxorubicin. Although conservative mutation of Tyr(568) in TM10 to Phe or Trp had no apparent effect on substrate specificity, substitution with Ala decreased the affinity of MRP1 for E(2)17betaG without affecting drug resistance or the transport of other substrates tested. These analyses confirm that several amino acids in TM10 selectively alter the substrate specificity of MRP1, suggesting that they interact directly with certain substrates. The location of these and other functionally important residues in TM helices 11, 16, and 17 is discussed in the context of an energy-minimized model of the membrane-spanning domains of MRP1.
Drug Metabolism and Disposition 05/2006; 34(4):539-46. DOI:10.1124/dmd.105.007740 · 3.33 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: Human multidrug resistance protein 1 (MRP1) is an ATP binding cassette (ABC) transporter that confers resistance to many natural product chemotherapeutic agents and can transport structurally diverse conjugated organic anions. MRP1 has three polytopic transmembrane domains (TMDs) and a total of 17 TM helices. Photolabeling and mutagenesis studies of MRP1 indicate that TM11, the last helix in the second TMD, may form part of the protein's substrate binding pocket. We have demonstrated that certain polar residues within a number of TM helices, including Arg593 in TM11, are determinants of MRP1 substrate specificity or overall activity. We have now extended these analyses to assess the functional consequences of mutating the remaining seven polar residues within and near TM11. Mutations Q580A, T581A, and S585A in the predicted outer leaflet region of the helix had no detectable effect on function, while mutation of three residues close to the membrane/cytoplasm interface altered substrate specificity. Two of these mutations affected only drug resistance. N597A increased and decreased resistance to vincristine and VP-16, respectively, while S605A decreased resistance to vincristine, VP-16 and doxorubicin. The third, S604A, selectively increased 17β-estradiol 17-(β-D-glucuronide) (E217βG) transport. In contrast, elimination of the polar character of the residue at position 590 (Asn in the wild-type protein) uniformly impaired the ability of MRP1 to transport potential physiological substrates and to confer resistance to three different classes of natural product drugs. Kinetic and photolabeling studies revealed that mutation N590A not only decreased the affinity of MRP1 for cysteinyl leukotriene 4 (LTC4) but also substantially reduced the binding of ATP to nucleotide binding domain 1 (NBD1). Thus, polar interactions involving residues in TM11 influence not only the substrate specificity of MRP1 but also an early step in the proposed catalytic cycle of the protein.
[Show abstract][Hide abstract] ABSTRACT: Multidrug resistance protein (MRP) 1 is a member of the ABCC branch of the ATP binding cassette (ABC) transporter superfamily that can confer resistance to natural product chemotherapeutic drugs and transport a variety of conjugated organic anions, as well as some unconjugated compounds in a glutathione- (GSH-) dependent manner. In addition to the two tandemly repeated polytopic membrane-spanning domains (MSDs) typical of ABC transporters, MRP1 and its homologues MRP2, -3, -6, and -7 contain a third NH(2)-terminal MSD. The cytoplasmic loop (CL3) connecting this MSD, but apparently not the MSD itself, is required for MRP1 leukotriene C(4) (LTC(4)) transport activity, substrate binding and appropriate trafficking of the protein to the basolateral membrane. We have used a baculovirus dual-expression system to produce various functionally complementing fragments of MRP1 in insect Sf21 cells to precisely define the region in CL3 that is required for activity and substrate binding. Using a parallel approach in polarized MDCK-I cells, we have also defined the region of CL3 that is required for basolateral trafficking. The CL3 NH(2)- and COOH-proximal functional boundaries have been identified as Cys(208) and Asn(260), respectively. Cys(208) also corresponds to the NH(2)-proximal boundary of the region required for basolateral trafficking in MDCK-I cells. However, additional residues downstream of the CL3 COOH-proximal functional boundary extending to Lys(270) were found to be important for basolateral localization. Finally, we show that regions in CL3 necessary for LTC(4) binding and transport are also required for binding of the photoactivatable GSH derivative azidophenacyl-GSH.
[Show abstract][Hide abstract] ABSTRACT: Human multidrug resistance protein (MRP) 3 is the most closely related homologue of MRP1. Like MRP1, MRP3 confers resistance to etoposide (VP-16) and actively transports 17 beta-estradiol 17-(beta-D-glucuronide) (E(2)17 beta G), cysteinyl leukotriene 4 (LTC(4)), and methotrexate, although with generally lower affinity. Unlike MRP1, MRP3 also transports monovalent bile salts. We have previously demonstrated that hydrogen-bonding residues predicted to be in the inner-leaflet spanning segment of transmembrane (TM) 17 of MRP1 are important for drug resistance and E(2)17 beta G transport. We have now examined the importance of the hydrogen-bonding potential of residues in TM17 of MRP3 on both substrate specificity and overall activity. Mutation S1229A reduced only methotrexate transport. Mutations S1231A and N1241A decreased resistance to VP-16 and transport of E(2)17 beta G and methotrexate but not taurocholate. Mutation Q1235A also reduced resistance to VP-16 and transport of E(2)17beta G but increased taurocholate transport without affecting transport of methotrexate. Mutations Y1232F and S1233A reduced resistance to VP-16 and the transport of all three substrates tested. In contrast, mutation T1237A markedly increased VP-16 resistance and transport of all substrates. On the basis of the substrates analyzed, residues Ser(1229), Ser(1231), Gln(1235), and Asn(1241) play an important role in determining the specificity of MRP3, while mutation of Tyr(1232), Ser(1233), and Thr(1237) affects overall activity. Unlike MRP1, the involvement of polar residues in determining substrate specificity extends throughout the TM helix. Furthermore, elimination of the hydrogen-bonding potential of a single amino acid, Thr(1237), markedly enhanced the ability of the protein to confer drug resistance and to transport all substrates examined.
[Show abstract][Hide abstract] ABSTRACT: The ATP-binding cassette (ABC) proteins comprise a large superfamily of transmembrane transporters that utilize the energy of ATP hydrolysis to translocate their substrates across biological membranes. Multidrug resistance protein (MRP) 2 (ABCC2) belongs to subfamily C of the ABC superfamily and, when overexpressed in tumor cells, confers resistance to a wide variety of anticancer chemotherapeutic agents. MRP2 is also an active transporter of organic anions such as methotrexate (MTX), estradiol glucuronide (E217betaG), and leukotriene C4 and is located on the apical membrane of polarized cells including hepatocytes where it acts as a biliary transporter. We recently identified a highly conserved tryptophan residue in the related MRP1 that is critical for the substrate specificity of this protein. In the present study, we have examined the effect of replacing the analogous tryptophan residue at position 1254 of MRP2. We found that only nonconservative substitutions (Ala and Cys) of Trp1254 eliminated [3H]E217betaG transport by MRP2, whereas more conservative substitutions (Phe and Tyr) had no effect. In addition, only the most conservatively substituted mutant (W1254Y) transported [3H]leukotriene C4, whereas all other substitutions eliminated transport of this substrate. On the other hand, all substitutions of Trp1254 eliminated transport of [3H]MTX. Finally, we found that sulfinpyrazone stimulated [3H]E217betaG transport by wild-type MRP2 4-fold, whereas transport by the Trp1254 substituted mutants was enhanced 6-10-fold. In contrast, sulfinpyrazone failed to stimulate [3H]MTX transport by either wild-type MRP2 or the MRP2-Trp1254 mutants. Taken together, our results demonstrate that Trp1254 plays an important role in the ability of MRP2 to transport conjugated organic anions and identify this amino acid in the putative last transmembrane segment (TM17) of this ABC protein as being critical for transport of MTX.
[Show abstract][Hide abstract] ABSTRACT: The mRNA encoding the human low density lipoprotein (LDL) receptor is transiently stabilized after phorbol ester treatment of HepG2 cells and has been shown to associate with components of the cytoskeleton in this cell line (G. M. Wilson, E. A. Roberts, and R. G. Deeley, J. Lipid Res. 1997. 38: 437-446). Using an episomal expression system, fragments of the 3' untranslated region (3'UTR) of LDL receptor mRNA were transcribed in fusion with the coding region of beta-globin mRNA in HepG2 cells. Analyses of the decay kinetics of these beta-globin-LDL receptor fusion mRNA deletion mutants showed that sequences in the proximal 3'UTR of LDL receptor mRNA including several AU-rich elements (AREs) were sufficient to confer short constitutive mRNA half-life in the heterologous system. Stabilization of LDL receptor mRNA in the presence of PMA required sequences in the distal 3'UTR, at or near three Alu-like repetitive elements. Furthermore, the 3'UTR of LDL receptor mRNA conferred cytoskeletal association on the otherwise unassociated beta-globin mRNA, by a mechanism involving at least two distinct RNA elements. Comparisons of decay kinetics and subcellular localization of endogenous LDL receptor mRNA and beta-globin-LDL receptor mRNA fusions in HepG2 cells have demonstrated that several cis-acting elements in the receptor 3'UTR contribute to post-transcriptional regulation of receptor expression, and provide further support for involvement of the cytoskeleton in the regulation of LDL receptor mRNA turnover.
The Journal of Lipid Research 06/1998; 39(5):1025-32. · 4.73 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: In mammals, some of the effects of interferon (IFN) on gene transcription are known to be mediated by a family of IFN-inducible DNA-binding proteins, the IFN regulatory factor (IRF) family, which includes both activators and repressors of transcription. Although IFN activities have been described in many vertebrates, little is known about regulation of IFN- or IFN-stimulated genes in species other than human and mouse. Here, we report the cloning of a chicken cDNA, cIRF-3, encoding a protein with a DNA-binding domain similar to that found in the mammalian IRF family of proteins. Similarity between cIRF-3 and the mammalian IRFs is comparable with that between known members of the family. It is most similar to the IRF proteins ICSBP and ISGF3 gamma but is equally divergent from both. Gel mobility shift assays indicate that cIRF-3 is capable of binding a known IFN-stimulated response element that is conserved between the mammalian and chicken Mx genes. Expression of the cIRF-3 gene can be induced to high levels by poly(I).poly(C). Induction is rapid and transient with no requirement for protein synthesis. Co-treatment of cells with cycloheximide results in superinduction of cIRF-3 mRNA. The structural and regulatory characteristics of cIRF-3 indicate that it is the first example of a non-mammalian IRF protein.
Nucleic Acids Research 07/1995; 23(12):2137-46. DOI:10.1093/nar/23.12.2137 · 9.11 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: Expression of the avian very-low-density apolipoprotein II (apoVLDLII) gene is completely dependent on estrogen and restricted to the liver. We have identified binding sites for nonhistone nuclear proteins located between -1.96 and -2.61 kilobases. One of these sites, located at -2.6 kilobases (designated site 1), was found to span an MspI site that becomes demethylated between days 7 and 9 of embryogenesis, the stage of development at which competence to express the apoVLDLII gene begins to be acquired. Levels of the factor(s) involved were high at day 7 of embryogenesis, decreased two- to threefold by days 9 to 11, and continued to decline more slowly until hatching. Furthermore, the mobility of the complex formed underwent a well-defined shift between days 11 to 13 embryogenesis. Methylation interference studies showed that modification of the outer guanosines of the MspI site resulted in marked inhibition of the formation of the protein-DNA complex. Competition studies, fractionation of nuclear extracts, and tissue distribution indicated that the factor was not the avian homolog of hepatocyte nuclear factor 1, nuclear factor 1, or CCAAT/enhancer-binding protein (C/EBP). However, site 1 could complete for binding to an oligonucleotide, previously shown to be recognized by C/EBP, in a nonreciprocal fashion. These studies demonstrate that the sequence recognized by the protein includes a C/EBP consensus sequence but that elements in addition to the core enhancer motif are essential for binding.
[Show abstract][Hide abstract] ABSTRACT: We have isolated and sequenced the 5' proximal exons and flanking regions of the chicken very low density apolipoprotein II (apo-VLDLII) and serum albumin genes. These genes specify the most abundant mRNA species present in livers of hens or estrogen-treated roosters. Sequencing revealed that the promoter region of the estrogen-dependent apo-VLDLII gene contained at least two potential transcriptional initiation sites, preceded by appropriately positioned "ATA" sequences. One is homologous with the cap site of the ovalbumin gene, and the other exhibits a match of 10 out of 12 nucleotides with the cap site of the serum albumin gene. S1 nuclease mapping indicates that both sites are active in vivo, although the "ovalbumin"-like site is by far the most efficient at all stages of the estrogenic response. The relative efficiencies of these two sites are maintained during in vitro transcription of truncated templates. A third site, that was not anticipated from sequencing data, is active in vivo but inactive in vitro. A conserved 5' flanking element, initially identified during studies on egg white protein genes, is also present upstream from the apo-VLDLII and serum albumin genes. Its removal does not affect initiation site selection in vitro. Regions of the apo-VLDLII and ovalbumin genes extending 100 nucleotides downstream from the "TATA box" contain several striking homologies that suggest a common mode of evolution.
Journal of Biological Chemistry 05/1983; 258(7):4556-64. · 4.57 Impact Factor