ATP-binding Cassette Transporter A1 Contains a Novel C-terminal VFVNFA Motif That Is Required for Its Cholesterol Efflux and ApoA-I Binding Activities
2004; Elsevier BV; Volume: 279; Issue: 46 Linguagem: Inglês
10.1074/jbc.m409848200
ISSN1083-351X
AutoresMichael L. Fitzgerald, Keiichiro Okuhira, Glenn F. Short, Jennifer J. Manning, Susan A. Bell, Mason W. Freeman,
Tópico(s)Peroxisome Proliferator-Activated Receptors
ResumoThe stimulation of cellular cholesterol and phospholipid efflux by apolipoprotein A-I is mediated by the activity of the ATP-binding cassette transporter A1 (ABCA1). Individuals with Tangier disease harbor loss-of-function mutations in this transporter that have proven useful in illuminating its activity. Here, we analyze a mutation that deletes the last 46 residues of the 2261 amino acid transporter (Δ46) and eliminates its lipid efflux. As the final four amino acids of the C terminus represent a putative PDZ-binding motif, we initially characterized deletion mutants lacking only these residues. Although a moderate decline in lipid efflux was detected, this decline was not as profound as that seen in the Δ46 mutant. Subsequent systematic analysis of the ABCA1 C terminus revealed a novel, highly conserved motif (VFVNFA) that was required for lipid efflux. Alteration of this motif, which is present in some but not all members of the ABCA family, did not prevent trafficking of the transporter to the plasma membrane but did eliminate its binding of apoA-I. Chimeric transporters, generated by substituting the C termini of either ABCA4 or ABCA7 for the endogenous terminus, demonstrated that ABCA1 could stimulate cholesterol efflux without its PDZ-binding motif but not without the VFVNFA motif. When a peptide containing the VFVNFA sequence was introduced into ABCA1-expressing cells, ABCA1-mediated lipid efflux was also markedly inhibited. These results indicate that the C-terminal VFVNFA motif of ABCA1 is essential for its lipid efflux activity. The data also suggest that this motif participates in novel protein-protein interactions that may be shared among members of the ABCA family. The stimulation of cellular cholesterol and phospholipid efflux by apolipoprotein A-I is mediated by the activity of the ATP-binding cassette transporter A1 (ABCA1). Individuals with Tangier disease harbor loss-of-function mutations in this transporter that have proven useful in illuminating its activity. Here, we analyze a mutation that deletes the last 46 residues of the 2261 amino acid transporter (Δ46) and eliminates its lipid efflux. As the final four amino acids of the C terminus represent a putative PDZ-binding motif, we initially characterized deletion mutants lacking only these residues. Although a moderate decline in lipid efflux was detected, this decline was not as profound as that seen in the Δ46 mutant. Subsequent systematic analysis of the ABCA1 C terminus revealed a novel, highly conserved motif (VFVNFA) that was required for lipid efflux. Alteration of this motif, which is present in some but not all members of the ABCA family, did not prevent trafficking of the transporter to the plasma membrane but did eliminate its binding of apoA-I. Chimeric transporters, generated by substituting the C termini of either ABCA4 or ABCA7 for the endogenous terminus, demonstrated that ABCA1 could stimulate cholesterol efflux without its PDZ-binding motif but not without the VFVNFA motif. When a peptide containing the VFVNFA sequence was introduced into ABCA1-expressing cells, ABCA1-mediated lipid efflux was also markedly inhibited. These results indicate that the C-terminal VFVNFA motif of ABCA1 is essential for its lipid efflux activity. The data also suggest that this motif participates in novel protein-protein interactions that may be shared among members of the ABCA family. The superfamily of ABC 1The abbreviations used are: ABC, ATP-binding cassette; ABCA1, ABC transporter A1; apoA-I, apolipoprotein A-I; BSA, bovine serum albumin; GFP, green fluorescent protein; nt, nucleotide; PBS, phosphate-buffered saline.1The abbreviations used are: ABC, ATP-binding cassette; ABCA1, ABC transporter A1; apoA-I, apolipoprotein A-I; BSA, bovine serum albumin; GFP, green fluorescent protein; nt, nucleotide; PBS, phosphate-buffered saline. transporters is one of the largest and most ancient gene families with representatives in all extant phyla (1Jones P.M. George A.M. Cell. Mol. Life Sci. 2004; 61: 682-699Crossref PubMed Scopus (435) Google Scholar). In humans, this family is thought to comprise at least 49 members, 13 of which belong to the ABCA class of transporters (2Dean M. Hamon Y. Chimini G. J. Lipid Res. 2001; 42: 1007-1017Abstract Full Text Full Text PDF PubMed Google Scholar). Of the A class members, ABCA1 and ABCA4 are best studied because of the fact that mutations in these two transporters are associated with Tangier disease and Stargardt's macular degeneration, respectively (3Allikmets R. Singh N. Sun H. Shroyer N.F. Hutchinson A. Chidambaram A. Gerrard B. Baird L. Stauffer D. Peiffer A. Rattner A. Smallwood P. Li Y. Anderson K.L. Lewis R.A. Nathans J. Leppert M. Dean M. Lupski J.R. Nat. Genet. 1997; 15: 236-246Crossref PubMed Scopus (1077) Google Scholar, 4Brooks-Wilson A. Marcil M. Clee S.M. Zhang L.H. Roomp K. van Dam M. Yu L. Brewer C. Collins J.A. Molhuizen H.O. Loubser O. Ouelette B.F. Fichter K. Ashbourne-Excoffon K.J. Sensen C.W. Scherer S. Mott S. Denis M. Martindale D. Frohlich J. Morgan K. Koop B. Pimstone S. Kastelein J.J. Hayden M.R. Nat. Genet. 1999; 22: 336-345Crossref PubMed Scopus (1481) Google Scholar, 5Bodzioch M. Orso E. Klucken J. Langmann T. Bottcher A. Diederich W. Drobnik W. Barlage S. Buchler C. Porsch-Ozcurumez M. Kaminski W.E. Hahmann H.W. Oette K. Rothe G. Aslanidis C. Lackner K.J. Schmitz G. Nat. Genet. 1999; 22: 347-351Crossref PubMed Scopus (1328) Google Scholar, 6Rust S. Rosier M. Funke H. Real J. Amoura Z. Piette J.C. Deleuze J.F. Brewer H.B. Duverger N. Denefle P. Assmann G. Nat. Genet. 1999; 22: 352-355Crossref PubMed Scopus (1249) Google Scholar, 7Lawn R.M. Wade D.P. Garvin M.R. Wang X. Schwartz K. Porter J.G. Seilhamer J.J. Vaughan A.M. Oram J.F. J. Clin. Investig. 1999; 104: R25-R31Crossref PubMed Scopus (648) Google Scholar, 8Brousseau M.E. Schaefer E.J. Dupuis J. Eustace B. Van Eerdewegh P. Goldkamp A.L. Thurston L.M. FitzGerald M.G. Yasek-McKenna D. O'Neill G. Eberhart G.P. Weiffenbach B. Ordovas J.M. Freeman M.W. Brown Jr., R.H. Gu J.Z. J. Lipid Res. 2000; 41: 433-441Abstract Full Text Full Text PDF PubMed Google Scholar). In Tangier disease, loss of ABCA1 activity results in the near absence of circulating high density lipoproteins and the deposition of massive amounts of cholesterol esters in peripheral tissues. At the cellular level, loss of ABCA1 function eliminates the efflux of cholesterol and phospholipids in response to stimulation with the major apolipoprotein of the high density lipoproteins, apoA-I. At the molecular level, a variety of evidence indicates that ABCA1 forms a high affinity complex with apoA-I by binding amphipathic helices within the apolipoprotein (9Wang N. Silver D.L. Costet P. Tall A.R. J. Biol. Chem. 2000; 275: 33053-33058Abstract Full Text Full Text PDF PubMed Scopus (496) Google Scholar, 10Oram J.F. Lawn R.M. Garvin M.R. Wade D.P. J. Biol. Chem. 2000; 275: 34508-34511Abstract Full Text Full Text PDF PubMed Scopus (466) Google Scholar, 11Fitzgerald M.L. Morris A.L. Rhee J.S. Andersson L.P. Mendez A.J. Freeman M.W. J. Biol. Chem. 2002; 277: 33178-33187Abstract Full Text Full Text PDF PubMed Scopus (183) Google Scholar, 12Fitzgerald M.L. Morris A.L. Chroni A. Mendez A.J. Zannis V.I. Freeman M.W. J. Lipid Res. 2004; 45: 287-294Abstract Full Text Full Text PDF PubMed Scopus (125) Google Scholar, 13Chroni A. Liu T. Fitzgerald M.L. Freeman M.W. Zannis V.I. Biochemistry. 2004; 43: 2126-2139Crossref PubMed Scopus (95) Google Scholar). Although the formation of this complex appears central to efflux activity, the mechanism by which the formation and turnover of this complex results in the transfer of cholesterol and phospholipid to the apolipoprotein remains unclear (14Chambenoit O. Hamon Y. Marguet D. Rigneault H. Rosseneu M. Chimini G. J. Biol. Chem. 2001; 276: 9955-9960Abstract Full Text Full Text PDF PubMed Scopus (176) Google Scholar, 15Wang N. Silver D.L. Thiele C. Tall A.R. J. Biol. Chem. 2001; 276: 23742-23747Abstract Full Text Full Text PDF PubMed Scopus (388) Google Scholar, 16Fielding P.E. Nagao K. Hakamata H. Chimini G. Fielding C.J. Biochemistry. 2000; 39: 14113-14120Crossref PubMed Scopus (182) Google Scholar, 17Panagotopulos S.E. Witting S.R. Horace E.M. Hui D.Y. Maiorano J.N. Davidson W.S. J. Biol. Chem. 2002; 277: 39477-39484Abstract Full Text Full Text PDF PubMed Scopus (114) Google Scholar, 18Wang N. Lan D. Gerbod-Giannone M. Linsel-Nitschke P. Jehle A.W. Chen W. Martinez L.O. Tall A.R. J. Biol. Chem. 2003; 278: 42906-42912Abstract Full Text Full Text PDF PubMed Scopus (157) Google Scholar, 19Denis M. Haidar B. Marcil M. Bouvier M. Krimbou L. Genest Jr., J. J. Biol. Chem. 2004; 279: 7384-7394Abstract Full Text Full Text PDF PubMed Scopus (86) Google Scholar). Analysis of naturally occurring and engineered ABCA1 mutations has proven to be a fruitful approach in studying the efflux mechanism. We have previously elucidated the ABCA1 transporter's topology and characterized some of its interactions with apoA-I using this approach (11Fitzgerald M.L. Morris A.L. Rhee J.S. Andersson L.P. Mendez A.J. Freeman M.W. J. Biol. Chem. 2002; 277: 33178-33187Abstract Full Text Full Text PDF PubMed Scopus (183) Google Scholar, 12Fitzgerald M.L. Morris A.L. Chroni A. Mendez A.J. Zannis V.I. Freeman M.W. J. Lipid Res. 2004; 45: 287-294Abstract Full Text Full Text PDF PubMed Scopus (125) Google Scholar). In our ongoing effort to define the structure/function relationships that underlie the ABCA1 efflux mechanism, we focused on a truncation mutant that we had originally identified in a patient with Tangier disease (8Brousseau M.E. Schaefer E.J. Dupuis J. Eustace B. Van Eerdewegh P. Goldkamp A.L. Thurston L.M. FitzGerald M.G. Yasek-McKenna D. O'Neill G. Eberhart G.P. Weiffenbach B. Ordovas J.M. Freeman M.W. Brown Jr., R.H. Gu J.Z. J. Lipid Res. 2000; 41: 433-441Abstract Full Text Full Text PDF PubMed Google Scholar). This mutation would be predicted to delete the last 46 amino acids of the transporter. The deleted sequences were of interest, as they included a cytoplasmic motif (ESYV) that conforms to the consensus sequence for binding to PDZ domain-containing proteins. PDZ proteins are named for the founding members of the group (PSD-95, Dlg, and ZO-1) and have been shown to play a critical role in cellular physiology through their participation in the formation of multimeric protein complexes (20Kim E. Niethammer M. Rothschild A. Jan Y.N. Sheng M. Nature. 1995; 378: 85-88Crossref PubMed Scopus (892) Google Scholar). This diverse group of proteins contains one or more copies of a 90-amino acid domain that can bind the C termini of proteins ending in consensus motifs such as (S/T)XV-COOH, as is the case for class I PDZ binding domains (21Hung A.Y. Sheng M. J. Biol. Chem. 2002; 277: 5699-5702Abstract Full Text Full Text PDF PubMed Scopus (584) Google Scholar). Indeed, using yeast two-hybrid screens and either the last 165 or 120 amino acids of the ABCA1 C terminus, two groups have recently reported the interaction of ABCA1 with three proteins containing PDZ domains (the α1 and β2 syntrophins and Lin7) and one non-PDZ protein (Fas-associated death domain protein) (22Buechler C. Boettcher A. Bared S.M. Probst M.C. Schmitz G. Biochem. Biophys. Res. Commun. 2002; 293: 759-765Crossref PubMed Scopus (66) Google Scholar, 23Buechler C. Bared S.M. Aslanidis C. Ritter M. Drobnik W. Schmitz G. J. Biol. Chem. 2002; 277: 41307-41310Abstract Full Text Full Text PDF PubMed Scopus (26) Google Scholar, 24Munehira Y. Ohnishi T. Kawamoto S. Furuya A. Shitara K. Imamura M. Yokota T. Takeda S. Amachi T. Matsuo M. Kioka N. Ueda K. J. Biol. Chem. 2004; 279: 15091-15095Abstract Full Text Full Text PDF PubMed Scopus (107) Google Scholar). To functionally characterize the importance of the putative PDZ-binding motif and to explore the impact of the loss of the C-terminal 46 amino acids in detail, we generated a series of ABCA1 C-terminal mutants. Initially, it was observed that the loss of the four amino acids composing the PDZ-binding motif, although detrimental to function, did not recapitulate the loss of all 46 amino acids at the C terminus. Further analysis demonstrated that the sequence between -41 and -46 (VFVNFA) was critical for efflux activity and that its loss abrogated apoA-I binding to ABCA1. This sequence is shared by some but not all members of the ABCA class of transporters. This finding was exploited to demonstrate that ABCA1 can efflux lipid normally when the VFVNFA-containing ABCA4 C terminus is substituted for that of ABCA1. That this activity is reconstituted despite the absence of the PDZ-binding motif in this chimeric protein indicates that the motif is not required for ABCA1 activity in all cell types. Substitution of the ABCA7 C terminus, which lacks the VNVNFA sequence but retains the PDZ motif, did not produce an active transporter, further corroborating this conclusion. When these results are coupled with our finding that an intracellular peptide containing the VFVNFA motif was able to strongly inhibit ABCA1-dependent efflux, the data suggest that this motif in the C terminus of ABCA1 plays a critical role in facilitating its interactions with other proteins required for the transporter's lipid efflux activity. Reagents—The following reagents were purchased from the indicated suppliers: LipofectAMINE 2000, the Anti-Xpress mouse monoclonal antibody, and Dulbecco's modified Eagle's cell culture medium were from Invitrogen; dithiobis(succinimidyl propionate) was from Pierce; Gamma Bind Plus Sepharose came from Amersham Biosciences; the anti-FLAG M2 mouse monoclonal antibody was from Sigma; poly-d-lysine-coated tissue culture plates were from BD Biosciences; apoA-I was purchased from Biodesign, Saco, ME; radionucleotides were from Perkin Elmer Life Sciences; nickel-nitrilotriacetic acid-agarose and pQE-30 vector were from Qiagen (Valencia, CA); and the rabbit anti-Rab6 antibody came from Santa Cruz Biotechnology (Santa Cruz, CA). DNA Constructs—Deletions in the C terminus of ABCA1 (Δ117, Δ46, Δ40, Δ30, Δ20, Δ10, and Δ4) were generated by PCR using a common forward primer that anneals to nt position 6088 of the open reading frame and unique reverse primers annealing at nt 6719, nt 6935, nt 6953, nt 6983, nt 7013, nt 7043, and nt 7061, respectively. The C-terminal mutant products were cloned into the ABCA1 cDNA using the BamHI site at nt position 6275 of the ABCA1 open reading frame. To generate tagged mutants, GFP was fused at the N terminus or a FLAG tag (DYKDDDDK) was inserted in the N-terminal extracellular loop as described previously (11Fitzgerald M.L. Morris A.L. Rhee J.S. Andersson L.P. Mendez A.J. Freeman M.W. J. Biol. Chem. 2002; 277: 33178-33187Abstract Full Text Full Text PDF PubMed Scopus (183) Google Scholar, 25Fitzgerald M.L. J. Biol. Chem. 2001; 276: 15137-15145Abstract Full Text Full Text PDF PubMed Scopus (106) Google Scholar). Wild type tagged ABCA1 transporters interact with apoA-I and retain normal efflux activity (11Fitzgerald M.L. Morris A.L. Rhee J.S. Andersson L.P. Mendez A.J. Freeman M.W. J. Biol. Chem. 2002; 277: 33178-33187Abstract Full Text Full Text PDF PubMed Scopus (183) Google Scholar, 25Fitzgerald M.L. J. Biol. Chem. 2001; 276: 15137-15145Abstract Full Text Full Text PDF PubMed Scopus (106) Google Scholar). Mutations of the VFVNFA motif of ABCA1 (2215VFVNFA2220 → 2215AAAAAA2220, 2215VFVNFA2220 → 2215VFLYFS2220, V2215A, F2216A, V2217A, V2217L, N2218A, and F2219A) were generated by overlap PCR. The ABCA1/A4 and A7 chimeras were generated by cloning the last 171 and 189 amino acids from the ABCA4 and A7 open reading frames, respectively, into the BamHI site at nt position 6275 of the ABCA1 open reading frame, replacing the terminal 184 amino acids of the protein. The human ABCA4 cDNA template was a generous gift of Dr. Jeremy Nathans (Johns Hopkins University School of Medicine). ABCA1/A4 and A7 chimeras in which the C terminus of ABCA1 was replaced from Val-2215 of the VFVNFA motif were generated by overlap PCR. All mutations were verified by sequencing of the resulting cDNA. Efflux Assays and Parallel Measurement of Total and Cell Surface ABCA1—Cholesterol efflux of both HEK293 (henceforth referred to as 293) and 293-EBNA-T cells transfected with mutant and wild type ABCA1 was measured in 24-well plates (two plates with FLAG constructs and two plates with GFP constructs). 24 h post-transfection, the media were replaced with Dulbecco's modified Eagle's/10% fetal bovine serum media containing 0.5 μCi/ml [3H]cholesterol or with media containing an equivalent amount of unlabeled cholesterol. To measure efflux activity, after 24 h the cells receiving [3H]cholesterol were washed twice with warm PBS, incubated with Dulbecco's modified Eagle's medium (1 mg/ml BSA) for 2 h, washed twice with PBS, and incubated in Dulbecco's modified Eagle's medium (1 mg/ml BSA) with or without 10 μg/ml apoA-I for an additional 20 h. In parallel, the plates of unlabeled cells were chilled on ice for 10 min and cell surface expression of ABCA1 was detected by an M2 anti-FLAG antibody, while total ABCA1 expression was measured in GFP-ABCA1-transfected cells by flow cytometry (FACSCalibur System; BD Biosciences). The percentage of cholesterol efflux (cpm media/(cpm media + cell-associated cpm) × 100) was calculated by scintillation counting. Cell surface ABCA1 expression was calculated as cpm 125I-labeled secondary antibody bound per milligram of total cell protein (Bio-Rad protein assay), and total transporter expression was expressed as the average GFP-ABCA1 fluorescent intensity. Cholesterol efflux was normalized by subtracting the efflux activity of mock transfected cells and then dividing by the cell surface or total expression of ABCA1. All measurements were performed in triplicate. The S.D. of these measures was calculated and propagated after normalization using formulas for the subtraction and division of errors. PDZ Domain Binding Assays—The interaction of PDZ domains with the ABCA1 C terminus was measured by overlay and oligopeptide binding assays (26Niethammer M. Kim E. Sheng M. J. Neurosci. 1996; 16: 2157-2163Crossref PubMed Google Scholar). For overlay assays, His-tagged ABCA1 polypeptides containing amino acids 2076–2261 or 2076–2257 (Δ4) were generated. To express the PDZ domains 1 and 2 of PSD-95, a pRSETB plasmid (generous gift of Dr. Morgan Sheng, Massachussetts Institute of Technology, Cambridge, MA) containing a sequence corresponding to amino acids 84–398 of human PSD-95 was used. As a positive control for binding to PSD-95, glutathione S-transferase fused with amino acids 1453 to 1482 of the N-methyl-d-aspartate receptor subunit NR2BC was used (26Niethammer M. Kim E. Sheng M. J. Neurosci. 1996; 16: 2157-2163Crossref PubMed Google Scholar). His6-tagged Rab6 and Rab11 polypeptides were used as negative controls for PDS-95 binding (27Fitzgerald M.L. Reed G.L. Biochem. J. 1999; 342: 353-360Crossref PubMed Scopus (66) Google Scholar). Polypeptides were subjected to 12% SDS-PAGE, transferred to nitrocellulose, and allowed to renature in 1× PBS (with 3% BSA) overnight. The transferred polypeptides were incubated with PSD-95 (1× 250 nm PBS and 3% BSA) for 2 h at room temperature, washed with 1× PBS (with 0.1% Tween 20), and incubated with anti-express mouse monoclonal antibody (1:5000 dilution). Membranes were washed, and bound antibody was detected with an anti-mouse IgG-horseradish peroxidase antibody (1:30,000 dilution) and enhanced chemiluminescence (Super Signal; Pierce). Blots were stripped and sequentially re-probed with rabbit anti-ABCA1 antibody (25Fitzgerald M.L. J. Biol. Chem. 2001; 276: 15137-15145Abstract Full Text Full Text PDF PubMed Scopus (106) Google Scholar) and rabbit anti-Rab6 antibody. To assess whether ABCA1-PDZ interactions have class I binding specificity, we used 20- and 16-mer biotinylated peptides representing the C termini of ABCA1 with and without the PDZ binding motif, respectively, immobilized on a streptavidin column. In vitro transcribed RNAs representing E-tag PDZ fusions of the first PDZ domain of PSD-95 (residues 41–161), the second PDZ domain of PSD-95 (residues 123–267), and Dlg-1 (residues 209–332) or the first and second PDZ domains of syntenin (residues 101–285) were translated in 500 μl of rabbit reticulocyte lysate translation reactions (Novagen) in the presence of 200 μCi of [35S]methionine (1000 Ci/mmol; PerkinElmer Life Sciences). The E-tag PDZ fusions were purified from the lysate using an anti-E-tag column (Amersham Biosciences) and buffer was exchanged into 1× TnT (40 mm triethylamine, pH 7.5, 150 mm NaCl, 5 mm EDTA, 0.05% Tween 20, and 100 μg/ml BSA) using NAP-10 columns (Amersham Biosciences). The purified PDZ domains were applied to 200 μl of the streptavidin-charged agarose and incubated at 4 °C for 2 h. The columns were washed with 10-column volumes of 1× TnT, and the bound fraction was eluted with free wild type ABCA1 peptide. The amount of input and eluted protein was quantitated by scintillation counting and expressed as percent bound. Peptide Synthesis and Inhibition Assays—To test whether the VFVNFA motif alone could act in trans to modulate ABCA1 efflux activity, we generated peptides containing VFVNFA or six alanines fused to the C terminus of the third α-helix of the Antennapedia homeo-domain (N-RQIKIWFQNRRMKWKKVFVNFA-C and N-RQIKIWFQNRRMKWKKAAAAAA-C) (28May M.J. D'Acquisto F. Madge L.A. Glockner J. Pober J.S. Ghosh S. Science. 2000; 289: 1550-1554Crossref PubMed Scopus (607) Google Scholar, 29Gallouzi I.E. Steitz J.A. Science. 2001; 294: 1895-1901Crossref PubMed Scopus (233) Google Scholar). The peptides were purified to >95% homogeneity by high pressure liquid chromatography and analyzed by mass spectrometry. To measure the efficiency of cellular uptake, peptides were biotinylated and incubated with 293-EBNA-T cells for 2 h. The cells were washed extensively, trypsinized to remove any cell surface-associated peptide, and lysed to determine the amount of internalized peptide. Equivalent amounts of cellular protein were spotted onto nitrocellulose and probed with streptavidin-conjugated horseradish peroxidase (Pierce). For efflux assays, 293-EBNA-T cells transfected with wild type ABCA1 or empty vector were loaded with [3H]cholesterol for 24 h, washed extensively, and treated with the Antennapedia peptides (50 μm) or vehicle (1% EtOH) for 2 h. Cells were then exposed or not exposed to apoA-I (10 μg/ml) for an additional 20 h, and the amount of cholesterol efflux was measured as described above. For acute inhibition assays, after incubation with the Antennapedia peptides for 2 h a wash step was added to remove free peptide before the addition of apoA-I. Cell viability was assessed by trypan blue uptake and 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) cell proliferation assays (American Type Culture Collection, Manassas, VA) to ensure that the peptides were not cytotoxic. All assays were performed on triplicate samples. Pattern Searches for the VFVNFA Motif—A BLAST-P (2.2.9 release) search of the non-redundant translated data bases was done using the VFVNFA motif as a query sequence (30Altschul S.F. Madden T.L. Schaffer A.A. Zhang J. Zhang Z. Miller W. Lipman D.J. Nucleic Acids Res. 1997; 25: 3389-3402Crossref PubMed Scopus (58771) Google Scholar). Search parameters optimized for finding short, nearly exact matches were used (word size, 2; SEG filter, off; expect value, 20,000; composition based statistics, off, score matrix, PAM30). An additional search using a pattern-hit initiated BLAST using the entire ABCA1 protein sequence confirmed that the VFVNFA sequence was uniquely conserved in the ABCA class members A1, A2, and A4 from vertebrate organisms. Additional searches for ABCA transporters encoded by the genomes of Fugu rubripes and Ciona intestinalis were done through the web portal maintained by the Department of Energy Joint Genome Institute (www.jgi.doe.gov). The ABCA1 C Terminus Is Highly Conserved and Is Bound by Class I PDZ Domains—To characterize the Δ46 Tangier mutation, we first determined whether the deleted sequences were conserved and, if so, whether they could be bound by PDZ domains. The alignment of six ABCA1 orthologues showed that the sequences are strongly conserved (Fig. 1A). There is 91.3% identity between the C termini of human and chicken ABCA1 transporters, which are separated by 300 million years of evolution. This conservation is most striking relative to a comparably sized region in the variable first large extracellular loop of ABCA1 (40.4% identity between the sequences comprising codons 190 to 236). As the final four residues of the C terminus constituted a putative class I PDZ-binding motif, we examined the binding of several PDZ proteins to the ABCA1 C terminus using an overlay binding assay. We first tested the founding class I PDZ protein, PSD-95, in this assay (21Hung A.Y. Sheng M. J. Biol. Chem. 2002; 277: 5699-5702Abstract Full Text Full Text PDF PubMed Scopus (584) Google Scholar, 31Songyang Z. Fanning A.S. Fu C. Xu J. Marfatia S.M. Chishti A.H. Crompton A. Chan A.C. Anderson J.M. Cantley L.C. Science. 1997; 275: 73-77Crossref PubMed Scopus (1209) Google Scholar). The last 186 amino acids of ABCA1 were expressed as a His-tagged polypeptide ending either with a wild type C terminus or with the C-terminal PDZ binding motif truncated (Δ4). The ABCA1 polypeptides were separated by SDS-PAGE and transferred to nitrocellulose along with a known binding partner of PSD-95, the N-methyl-d-aspartate receptor subunit NR2B (26Niethammer M. Kim E. Sheng M. J. Neurosci. 1996; 16: 2157-2163Crossref PubMed Google Scholar). An epitope-tagged polypeptide encompassing the PDZ domains 1 and 2 of PSD-95 was incubated with the transferred proteins, and the binding of the polypeptide was detected using an anti-epitope antibody. PSD-95 bound both the wild type ABCA1 polypeptide and the NR2B C terminus (Fig. 1B). Truncation of the ABCA1 ESYV motif largely ablated the PSD-95 interaction. In contrast, PSD-95 did not bind two other His-tagged proteins that did not contain consensus PDZ-binding motifs. Because PSD-95 expression is largely confined to neuronal tissues whereas ABCA1 is ubiquitously expressed, we screened for additional class I PDZ proteins in a library constructed from cells known to express high levels of ABCA1. RNA for reverse transcription PCR reactions was prepared from human THP-1 macrophages treated with the phorbol ester phorbol 12-myristate 13-acetate, which up-regulates ABCA1 expression and cholesterol efflux (32Arakawa R. Yokoyama S. J. Biol. Chem. 2002; 277: 22426-22429Abstract Full Text Full Text PDF PubMed Scopus (180) Google Scholar). Reactions were run using primers designed to amplify class I PDZ proteins (PSD-95, Dlg-1, NE-dlg, and chapsyn-110) as well as the class II PDZ protein syntenin. In data not shown, we were able to amplify a small amount of a PSD-95 product. In contrast, Dlg-1 expression was easily detected whereas the expression of NE-dlg or chapsyn-110 was not. A product representing syntenin was also reliably amplified. To determine whether ABCA1-PDZ protein interactions have class I specificity, the PDZ domains from PSD-95, Dlg-1, and syntenin, were assessed for their ability to bind the ABCA1 C terminus. As with PSD-95, the second PDZ domain of Dlg-1 bound a C-terminal ABCA1 peptide, whereas the PDZ domains of syntenin did not (Fig. 1C). These results indicated that the highly conserved C terminus of ABCA1 can interact with proteins containing class I but not class II PDZ domains. The ABCA1 C Terminus Is Essential for Efflux but the PDZ Protein-binding Motif Is Not—To test the functional importance of the C terminus, we engineered the 46-amino acid Tangier deletion (Q2215X, Δ46) as well as a larger deletion described by Clee et al. (R2144X, Δ117) (33Clee S.M. Zwinderman A.H. Engert J.C. Zwarts K.Y. Molhuizen H.O. Roomp K. Jukema J.W. van Wijland M. van Dam M. Hudson T.J. Brooks-Wilson A. Genest Jr., J. Kastelein J.J. Hayden M.R. Circulation. 2001; 103: 1198-1205Crossref PubMed Scopus (271) Google Scholar) into our cDNA constructs. Both mutants showed no efflux activity (data not shown). The activity of a mutant with a deletion of the PDZ binding motif alone (elimination of only the last four amino acids of ABCA1, Δ4) was then compared with the activity of the Δ46 mutant. Both mutants were similarly expressed, as determined by immunoblotting with an antibody directed against the N terminus of ABCA1 (Fig. 2A). Δ4, when expressed at moderate levels in 293 cells, consistently showed efflux deficits of up to 60% as compared with wild type ABCA1 activity. This defect, however, was much less apparent in 293-EBNA-T cells, which express our constructs at a much higher level than traditional 293 cells because of replication of the transfected DNA in the EBNA-T-expressing cells (Fig. 2, B and C). In contrast to Δ4, Δ46 showed a complete loss in efflux activity regardless of expression level. The failure of the deletion of the PDZ-binding motif sequence to recapitulate the loss of the terminal 46 amino acids of ABCA1 indicated that other sequences in the C terminus were playing a functionally important role. We next turned our attention to identifying those sequences. C-terminal Deletion Mutants Reveal a Negative Regulatory Region between -30 and -40 —To refine the sequences responsible for the loss of activity of Δ46, a series of 10 amino acid deletions in ABCA1 were constructed. Like Δ4, the Δ10, Δ20, and Δ30 transporters showed similar losses in efflux activity when expressed in 293 cells but were again largely indistinguishable from wild type ABCA1 when expressed in 293-EBNA-T cells (Fig. 3, A and B). Surprisingly, in both cell types Δ40 showed less functional impairment than the shorter truncations, suggesting that a negative regulatory e
Referência(s)