Hydrophobic ligand binding properties of the human lipocalin apolipoprotein M
2007; Elsevier BV; Volume: 48; Issue: 8 Linguagem: Inglês
10.1194/jlr.m700103-jlr200
ISSN1539-7262
AutoresJosefin Ahnström, Kirsten Faber, Olof Axler, Björn Dahlbäck,
Tópico(s)Biomedical Research and Pathophysiology
ResumoApolipoprotein M (apoM) is a plasma protein associated mainly with HDL. ApoM is suggested to be important for the formation of preβ-HDL, but its mechanism of action is unknown. Homology modeling has suggested apoM to be a lipocalin. Lipocalins share a structurally conserved β-barrel, which in many lipocalins bind hydrophobic ligands. The aim of this study was to test the ability of apoM to bind different hydrophobic substances. ApoM was produced both in Escherichia coli and in HEK 293 cells. Characterization of both variants with electrophoretic and immunological methods suggested apoM from E. coli to be correctly folded. Intrinsic tryptophan fluorescence of both apoM variants revealed that retinol, all-trans-retinoic acid, and 9-cis-retinoic acid bound (dissociation constant = 2–3 μM), whereas other tested substances (e.g., cholesterol, vitamin K, and arachidonic acid) did not. The intrinsic fluorescence of two apoM mutants carrying single tryptophans was quenched by retinol and retinoic acid to the same extent as wild-type apoM, indicating that the environment of both tryptophans was affected by the binding. In conclusion, the binding of retinol and retinoic acid supports the hypothesis that apoM is a lipocalin. The physiological relevance of this binding has yet to be elucidated. Apolipoprotein M (apoM) is a plasma protein associated mainly with HDL. ApoM is suggested to be important for the formation of preβ-HDL, but its mechanism of action is unknown. Homology modeling has suggested apoM to be a lipocalin. Lipocalins share a structurally conserved β-barrel, which in many lipocalins bind hydrophobic ligands. The aim of this study was to test the ability of apoM to bind different hydrophobic substances. ApoM was produced both in Escherichia coli and in HEK 293 cells. Characterization of both variants with electrophoretic and immunological methods suggested apoM from E. coli to be correctly folded. Intrinsic tryptophan fluorescence of both apoM variants revealed that retinol, all-trans-retinoic acid, and 9-cis-retinoic acid bound (dissociation constant = 2–3 μM), whereas other tested substances (e.g., cholesterol, vitamin K, and arachidonic acid) did not. The intrinsic fluorescence of two apoM mutants carrying single tryptophans was quenched by retinol and retinoic acid to the same extent as wild-type apoM, indicating that the environment of both tryptophans was affected by the binding. In conclusion, the binding of retinol and retinoic acid supports the hypothesis that apoM is a lipocalin. The physiological relevance of this binding has yet to be elucidated. ERRATUMJournal of Lipid ResearchVol. 51Issue 7PreviewThe authors of "Hydrophobic ligand-binding properties of the human lipocalin apolipoprotein M" (J. Lipid Res. 2007. 48: 1754–1762) have advised the Journal that Equation 1 on page 1756 is incorrect. Full-Text PDF Open Access 1-anilinonaphthalene-8-sulfonic acid apolipoprotein M dissociation constant retinol binding protein Apolipoprotein M (apoM) is a 25 kDa (188 amino acid residues) protein expressed mainly in liver and kidney (1.Xu N. Dahlback B. A novel human apolipoprotein (apoM).J. Biol. Chem. 1999; 274: 31286-31290Abstract Full Text Full Text PDF PubMed Scopus (273) Google Scholar, 2.Faber K. Axler O. Dahlback B. Nielsen L.B. Characterization of apoM in normal and genetically modified mice.J. Lipid Res. 2004; 45: 1272-1278Abstract Full Text Full Text PDF PubMed Scopus (66) Google Scholar). In plasma, apoM is associated mainly with HDL and to a minor extent with chylomicrons, VLDL, and LDL (1.Xu N. Dahlback B. A novel human apolipoprotein (apoM).J. Biol. Chem. 1999; 274: 31286-31290Abstract Full Text Full Text PDF PubMed Scopus (273) Google Scholar, 3.Christoffersen C. Nielsen L.B. Axler O. Andersson A. Johnsen A.H. Dahlback B. Isolation and characterization of human apolipoprotein M-containing lipoproteins.J. Lipid Res. 2006; 47: 1833-1843Abstract Full Text Full Text PDF PubMed Scopus (154) Google Scholar). Studies of apoM in mice have suggested apoM to be important for the formation of preβ-HDL and for the efflux of cholesterol from cells to HDL, which are key events in reverse cholesterol transport. Furthermore, overexpression of apoM in LDL receptor-deficient mice challenged with a cholesterol-rich diet suggested apoM to have antiatherogenic properties (4.Wolfrum C. Poy M.N. Stoffel M. Apolipoprotein M is required for prebeta-HDL formation and cholesterol efflux to HDL and protects against atherosclerosis.Nat. Med. 2005; 11: 418-422Crossref PubMed Scopus (266) Google Scholar). A noteworthy feature of the apoM amino acid sequence is the lack of a signal peptidase cleavage site, explaining why circulating apoM contains its signal peptide (1.Xu N. Dahlback B. A novel human apolipoprotein (apoM).J. Biol. Chem. 1999; 274: 31286-31290Abstract Full Text Full Text PDF PubMed Scopus (273) Google Scholar). This unusual property is found in two other HDL-associated proteins: paraoxonase-1 and the haptoglobin-related protein (5.Raper J. Fung R. Ghiso J. Nussenzweig V. Tomlinson S. Characterization of a novel trypanosome lytic factor from human serum.Infect. Immun. 1999; 67: 1910-1916Crossref PubMed Google Scholar). The retained signal peptide of apoM serves as a hydrophobic anchor, binding apoM to the phospholipid layer of the lipoproteins (O. Axler et al., unpublished data). Structural analysis including homology modeling has predicted apoM to belong to the lipocalin protein family (6.Duan J. Dahlback B. Villoutreix B.O. Proposed lipocalin fold for apolipoprotein M based on bioinformatics and site-directed mutagenesis.FEBS Lett. 2001; 499: 127-132Crossref PubMed Scopus (84) Google Scholar). Members of the lipocalin protein family are typically small secreted proteins, which are functionally diverse and have weak sequence similarity but high similarity at the tertiary structural level (7.Flower D.R. The lipocalin protein family: structure and function.Biochem. J. 1996; 318: 1-14Crossref PubMed Scopus (1389) Google Scholar, 8.Schlehuber S. Skerra A. Lipocalins in drug discovery: from natural ligand-binding proteins to "anticalins.".Drug Discov. Today. 2005; 10: 23-33Crossref PubMed Scopus (145) Google Scholar). The majority of lipocalins are responsible for the storage and transport of compounds that have low solubility or are chemically sensitive, such as vitamins, steroids, and metabolic products (7.Flower D.R. The lipocalin protein family: structure and function.Biochem. J. 1996; 318: 1-14Crossref PubMed Scopus (1389) Google Scholar, 8.Schlehuber S. Skerra A. Lipocalins in drug discovery: from natural ligand-binding proteins to "anticalins.".Drug Discov. Today. 2005; 10: 23-33Crossref PubMed Scopus (145) Google Scholar). The lipocalins share a structurally conserved β-barrel as a central motif, which consists of an eight-stranded antiparallel β-sheet closed back on itself and a short α-helix. The β-barrel encloses a hydrophobic pocket, which in many lipocalins serves as a ligand binding site (7.Flower D.R. The lipocalin protein family: structure and function.Biochem. J. 1996; 318: 1-14Crossref PubMed Scopus (1389) Google Scholar, 8.Schlehuber S. Skerra A. Lipocalins in drug discovery: from natural ligand-binding proteins to "anticalins.".Drug Discov. Today. 2005; 10: 23-33Crossref PubMed Scopus (145) Google Scholar). We previously constructed an apoM model based partly on the structure of retinol binding protein (RBP) and mouse major urinary protein (Fig. 1) (6.Duan J. Dahlback B. Villoutreix B.O. Proposed lipocalin fold for apolipoprotein M based on bioinformatics and site-directed mutagenesis.FEBS Lett. 2001; 499: 127-132Crossref PubMed Scopus (84) Google Scholar). RBP is a 21 kDa lipocalin present in plasma. Its main function is to transport vitamin A (all-trans-retinol). The three-dimensional structure of RBP has been experimentally determined, and the structural details of the retinol binding have been elucidated (9.Newcomer M.E. Ong D.E. Plasma retinol binding protein: structure and function of the prototypic lipocalin.Biochim. Biophys. Acta. 2000; 1482: 57-64Crossref PubMed Scopus (141) Google Scholar, 10.Zanotti G. Berni R. Plasma retinol-binding protein: structure and interactions with retinol, retinoids, and transthyretin.Vitam. Horm. 2004; 69: 271-295Crossref PubMed Scopus (127) Google Scholar). The lipocalin family contains a highly conserved tryptophan (position 24 in RBP and position 47 in apoM) that points toward the inside of the hydrophobic pocket (9.Newcomer M.E. Ong D.E. Plasma retinol binding protein: structure and function of the prototypic lipocalin.Biochim. Biophys. Acta. 2000; 1482: 57-64Crossref PubMed Scopus (141) Google Scholar). ApoM contains a second tryptophan at position 100 (Trp100) facing the hydrophobic pocket close to its opening (Fig. 1). These tryptophans can be used for assessing the binding of ligands in the hydrophobic pocket by intrinsic fluorescence quenching studies. The aim of this study was to investigate the binding of several small lipophilic ligands to apoM. The predicted lipocalin domain of human apoM was produced in a prokaryotic expression system as well as in stably transfected HEK 293 cells, and intrinsic fluorescence quenching was used to test the binding of a panel of hydrophobic substances to apoM. ApoM was found to bind retinol and retinoic acid, whereas other hydrophobic vitamins, steroid hormones, and fatty acid derivatives did not associate with apoM. These results suggest that apoM is indeed a lipocalin, but the physiological importance of retinol binding remains unknown. Arachidonic acid, β-estradiol, bilirubin, cholesterol, linoleic acid, oleic acid, palmitic acid, platelet-activating factor, progesterone, prostaglandin E1, vitamin A (all-trans-retinol), all-trans-retinoic acid, 9-cis-retinoic acid, and vitamin E (dl-α-tocopherol) were purchased from MP Biomedicals. 1-Anilinonaphthalene-8-sulfonic acid (1,8-ANS) was from Fluka. Acetyl-CoA, vitamin D3 (cholecalciferol), and 3-hydroxy butyric acid were from Sigma-Aldrich. Vitamin K was from Roche. Testosterone was kindly provided by Dr. Pirkko Härkönen (University Hospital Malmö). Human purified RBP was a kind gift from Dr. Bo Åkerström (Lund University). Polyclonal and monoclonal antibodies were generated in house using recombinant human apoM as antigen. Polyclonal antibodies were raised in rabbits and monoclonal antibodies (mab23, mab42, and mab58) were raised in mice using standard techniques (11.Axler O. Ahnstrom J. Dahlback B. An ELISA for apolipoprotein M reveals a strong correlation to total cholesterol in human plasma.J. Lipid Res. 2007; 48: 1772-1780Abstract Full Text Full Text PDF PubMed Scopus (101) Google Scholar). The cDNA encoding the lipocalin domain of apoM (amino acid residues 22–188) was introduced into a pET15b bacterial expression vector (Novagen) using the BamHI and NdeI cleavage sites. The pET15b vector contains the coding sequence for an N-terminal histidine tag followed by a thrombin cleavage site. After the thrombin cleavage site, the vector construct encodes five amino acids, GSHMN, before the start of the apoM sequence. The thrombin cleavage site was thus placed a few amino acids before the start of the apoM sequence, the reason being that we had experienced poor cleavage in other constructs in which the apoM sequence followed directly upon the cleavage site. This is presumably attributable to steric hindrance, as the cysteine at position 23 is involved in a disulfide bridge with cysteine 167 (6.Duan J. Dahlback B. Villoutreix B.O. Proposed lipocalin fold for apolipoprotein M based on bioinformatics and site-directed mutagenesis.FEBS Lett. 2001; 499: 127-132Crossref PubMed Scopus (84) Google Scholar). The five extra amino acid residues did not seem to affect the folding, as demonstrated in Results. The BigDye Sequencing Kit (Perkin-Elmer) was used to confirm that the inserted apoM sequence was correct. ApoM (accession number NM 019101) contains two tryptophans, Trp47 and Trp100, which is why we denote the wild-type variant apoMWW (Fig. 1). Three mutants, Trp47Phe (apoMFW), Trp100Phe (apoMWF), and Trp47Phe, Trp100Phe (apoMFF), were constructed using the Quickchange Site-Directed Mutagenesis Kit (Stratagene). Two complementary oligonucleotides for each mutation were used for this purpose. The sequences for the sense oligonucleotides were 5′-GAGGTCCACTTGGGCCAGTTCTACTTTATCGCAGGG-3′ (FW and FF) and 5′-CTCTGTGTGCCCCGGAAATTCATCTACCACCTGACTG-3′ (WF and FF). Mutated nucleotides are highlighted by underlined letters. The oligonucleotides were ordered from MWG Biotech. The mutations were verified by DNA sequencing using a BigDye Sequencing Kit (Perkin-Elmer). The pET15b vector containing the apoM sequence was transfected into E. coli strain BL21 (DE3). Cultures were grown under agitation at 37°C in 2 liters of Luria-Bertani medium containing 100 μg/ml ampicillin. Gene expression was induced when the absorbance at 600 nm was 0.6–0.8 by the addition of isopropyl-1-thio-β-d-galactopyranoside to a final concentration of 1 mM. Expression was allowed for 4 h, and then the cells were harvested by centrifugation and resuspended in ice-cold 20 mM Tris-HCl and 0.15 M NaCl, pH 7.4. Lysozyme (Amersham Biosciences; 100 μg/ml final concentration) and benzamidine (0.5 M) were added to the bacteria, followed by incubation for 45 min at room temperature. The bacteria were sonicated in intermittent 10 s bursts for 30 min until the solution was clear and nonviscous and then centrifuged again. The pellet containing the inclusion bodies was suspended in 30 ml of 6 M guanidine-HCl and 20 mM Tris-HCl, pH 8.0, containing 10 mM reduced glutathione overnight at room temperature and then centrifuged again. The supernatant was collected and applied to a nickel column (20 ml) (nickel resin Superflow; Qiagen) equilibrated with the same buffer. After washing with the equilibration buffer, the apoM was eluted with a linear 0–0.5 M imidazole gradient (260 ml). Fractions containing recombinant apoM were identified by dot-blot analysis using a polyclonal anti-human apoM antiserum. The fractions with apoM were pooled and refolded by dialysis at 4°C. The pool (diluted to 100 μg/ml in the dialysis buffer) was first dialyzed against 3 M guanidine HCl, 20 mM Tris-HCl (pH 8.0), 4 mM reduced glutathione, 0.4 mM oxidized glutathione, and 10% glycerol for 2–3 days and then against 20 mM Tris-HCl (pH 8.0), 4 mM reduced glutathione, 0.4 mM oxidized glutathione, and 10% glycerol for 1 day. Iodacetamide (5 mM) was added to block free sulfhydryl groups before final dialysis against 20 mM Tris-HCl, pH 8.0, containing 10% glycerol for 1 day. The recombinant apoM was further purified through anion-exchange chromatography on a Q-Sepharose column (Amersham Biosciences). The column (100 ml) was equilibrated in 20 mM Tris-HCl (pH 8.0) and 10% glycerol, and after application of the refolded apoM and washing with the same buffer, the column was eluted with a 500 ml linear NaCl gradient (0–0.3 M). Fractions were analyzed by unreduced 15% PAGE in the presence of SDS, and those containing monomeric apoM were pooled and dialyzed against 20 mM Tris-HCl (pH 8.0) and 10% glycerol overnight at 4°C. The histidine tag was successfully cleaved off by thrombin. After thrombin cleavage, the apoM was repurified on another Q-Sepharose column (25 ml; 400 ml 0–0.3 M NaCl gradient). Finally, the pooled fractions from the second Q-Sepharose purification were applied on a cation-exchange chromatography column (SP-Sepharose; Amersham Biosciences) (2 ml) equilibrated with 20 mM Tris-HCl (pH 8.0) and 10% glycerol to remove the remaining thrombin; thrombin bound to the column, whereas apoM passed through. The apoM concentration was determined by absorption at 280 nm using an extinction coefficient (E1%, 1 cm) of 13.7, which was determined as described (12.Grimsley G.R. Pace C.N. Spectrophotometric determination of protein concentration.in: Coligan J.E. Dunn B.M. Speicher D.W. Wingfield P.T. Current Protocols in Protein Science. John Wiley & Sons, New York2003Google Scholar). The purified apoM was analyzed by 15% SDS-PAGE and by nondenaturing gel electrophoresis (13.Johansson B.G. Agarose gel electrophoresis.Scand. J. Clin. Lab. Invest. Suppl. 1972; 124: 7-19Crossref PubMed Scopus (511) Google Scholar). To express soluble apoM, a signal peptidase cleavage site was generated by mutagenesis of glutamine 22 to alanine (apoMQ22A) in full-length apoM cDNA in pcDNA3 (Invitrogen) (6.Duan J. Dahlback B. Villoutreix B.O. Proposed lipocalin fold for apolipoprotein M based on bioinformatics and site-directed mutagenesis.FEBS Lett. 2001; 499: 127-132Crossref PubMed Scopus (84) Google Scholar) using the Quikchange Site-Directed Mutagenesis Kit (Stratagene). The sequence for the sense oligonucleotide (MWG Biotech) was 5′-CCTTAACTCCATCTACGCGTGCCCTGAGCACAGTC-3′, with the mutated oligonucleotides being underlined. The presence of the mutation was verified by DNA sequencing using the BigDye Sequencing Kit (Perkin-Elmer). The apoMQ22A cDNA was used to transfect HEK 293 cells using Lipofectin (Invitrogen), and stable transfectants were selected using G-418 and tested with the apoM ELISA. Clones secreting high levels of apoM were expanded, and serum-free Optimem medium was collected and stored at −20°C. To purify apoM, ammonium sulfate (70%) was added to the medium (1 liter) and the precipitate was collected by centrifugation at 5,000 rpm for 25 min. The pellet was dissolved in 20 mM Tris, pH 8.0, dialyzed against the same buffer overnight at 4°C, and loaded onto a 5 ml mab23 (9 mg/ml) HiTrap column (Amersham Biosciences) (3.Christoffersen C. Nielsen L.B. Axler O. Andersson A. Johnsen A.H. Dahlback B. Isolation and characterization of human apolipoprotein M-containing lipoproteins.J. Lipid Res. 2006; 47: 1833-1843Abstract Full Text Full Text PDF PubMed Scopus (154) Google Scholar). The column was washed with TBS, pH 7.4, and apoM was eluted with 0.1 M glycine, pH 2.2, the pH being immediately neutralized with unbuffered Tris. The purity of the apoM was assessed by 15% SDS-PAGE. A sandwich ELISA for apoM based on two monoclonal antibodies, mab58 and mab42, was used to quantify apoM, as described previously (3.Christoffersen C. Nielsen L.B. Axler O. Andersson A. Johnsen A.H. Dahlback B. Isolation and characterization of human apolipoprotein M-containing lipoproteins.J. Lipid Res. 2006; 47: 1833-1843Abstract Full Text Full Text PDF PubMed Scopus (154) Google Scholar, 11.Axler O. Ahnstrom J. Dahlback B. An ELISA for apolipoprotein M reveals a strong correlation to total cholesterol in human plasma.J. Lipid Res. 2007; 48: 1772-1780Abstract Full Text Full Text PDF PubMed Scopus (101) Google Scholar). In brief, a 96-well Costar plate (Corning, Inc.) was coated with mab58 and quenched according to standard techniques. Serial dilutions of wild-type apoM (apoMWW and apoMQ22A) and apoM mutants and Li-heparin plasma were made in TBS (50 mM Tris-HCl and 0.15 M NaCl, pH 7.4) containing 1% BSA (Sigma-Aldrich) and 1% Triton X-100 (MP Biomedicals) and added to the wells. ApoM bound to mab58 was detected by biotinylated mab42, streptavidin-avidin-horseradish peroxidase (DAKOCytomation A/S), and 1,2-phenylenediamine dihydrocloride tablets (DAKOCytomation A/S), according to the manufacturer's instructions. Recombinant apoM (0.5 ml of 20 μg/ml) in TBS containing 1% BSA (Sigma-Aldrich) was applied to a gel filtration column, Superose 12 10/300 GL (Amersham Biosciences), and equilibrated in TBS. Fractions of 0.27 ml were collected and analyzed using the apoM ELISA. The total protein concentration in the fractions was measured with the Bio-Rad Protein Assay according to the manufacturer's instructions. The ligand binding experiments were performed at 25 ± 0.1°C with a Fluoromax-3 spectrofluorometer (Jabin Yvon Horiba Group) with slits set at 5 nm bandwidth. The binding studies were performed using excitation at 295 nm. In initial experiments, increasing concentrations of the proteins (in TBS) were tested until the signals for intrinsic fluorescence were high enough. The emission maximum for each protein was determined from the emission spectra received during these titrations. Ligands were added stepwise to a 1 ml apoM protein solution in 0.5–5 μl aliquots from a 1 mM stock solution of the ligand, the maximum volume of ligand added not exceeding 20 μl. The slight volume increase in the course of the titration was neglected. After each added aliquot, the protein sample was mixed and incubated in the dark for 1 min before the fluorescence measurement. In each of the measurements, a blank consisting of the ligand in TBS was subtracted from the actual fluorescence. For each ligand, three to five independent titrations were performed. Binding was evaluated by following the quenching of protein fluorescence attributable to energy transfer. The ligands tested were the following: bilirubin, cholesterol, β-estradiol, 3-hydroxy butyric acid, oleic acid, palmitic acid, platelet-activating factor, progesterone, prostaglandin E1, all-trans-retinol, all-trans-retinoic acid, 9-cis-retinoic acid, vitamin D3 that had been dissolved in DMSO, arachidonic acid, testosterone, vitamin K that had been dissolved in 99.5% ethanol, vitamin E that had been dissolved in acetone, and acetyl-CoA that had been dissolved in distilled water. The binding affinity between apoM and the different ligands was calculated by fitting to two different equations, 1 and 2, according to the theory of bimolecular complex formation. I=IP1-[L][L]Kd+IPL[L][L]KdEq. 1 [L]=[L]t-Kd-[P]t2+[L]t-Kd-[P]t22+Kd[L]tEq. 2 [P]t and [L]t represent the total concentration of protein and ligand, respectively, and [L] represents the concentration of free ligand. I indicates the corrected fluorescence. IP and IPL represent the fluorescence of free protein without added ligand and the protein-ligand complex (at saturating concentrations of ligand), respectively. Kd is the apparent dissociation constant for the binding between apoM and the ligand. Kd and IPL were fitted as free parameters. See the supplementary material for the derivation of the equation. 1,8-ANS is a hydrophobic probe that is known to interact with hydrophobic regions in proteins (14.Fu X. Zhang X. Chang Z. 4,4′-Dianilino-1,1′-binaphthyl-5,5′-sulfonate, a novel molecule having chaperone-like activity.Biochem. Biophys. Res. Commun. 2005; 329: 1087-1093Crossref PubMed Scopus (29) Google Scholar – 17.Matulis D. Baumann C.G. Bloomfield V.A. Lovrien R.E. 1-Anilino-8-naphthalene sulfonate as a protein conformational tightening agent.Biopolymers. 1999; 49: 451-458Crossref PubMed Scopus (238) Google Scholar). The affinity of other ligands for the 1,8-ANS binding site can be assessed by observing the decrease of 1,8-ANS-apoM fluorescence in the presence of increasing concentrations of competitor ligand (16.Kane C.D. Bernlohr D.A. A simple assay for intracellular lipid-binding proteins using displacement of 1-anilinonaphthalene 8-sulfonic acid.Anal. Biochem. 1996; 233: 197-204Crossref PubMed Scopus (86) Google Scholar). 1,8-ANS also shows a reversible shift of its own fluorescence emission maximum from 512 to 470 nm when excited at 370 nm after binding to hydrophobic surfaces. 1,8-ANS-bound fluorescence measurements were performed with the ligands binding to apoM in the intrinsic fluorescence measurements. A precomplex was formed by adding 2 μl of a 5 mM stock solution of 1,8-ANS in DMSO to 1 ml of 1.6 μM apoMWW. Fluorescence intensity was measured at 500 nm after excitation at 370 nm. Just as in the intrinsic fluorescence experiments, ligands were added to the protein in 0.5–5 μl aliquots of 1 mM stock solutions to a total of 20 μl. After each added aliquot of ligand, the protein sample was mixed and incubated in the dark for 1 min before the fluorescence measurement. Slit widths of 5 nm were used for both excitation and emission. For each ligand, five independent titrations were performed. The lipocalin domain of human apoM was expressed both in prokaryotic (E. coli) and eukaryotic (HEK 293) cells. The hydrophobic signal peptide (residues 1–21) was excluded from the apoM constructs, as we aimed to obtain a soluble apoM for ligand binding studies. In the E. coli system, in addition to apoMWW, three tryptophan mutants, Trp47Phe (apoMFW), Trp100Phe (apoMWF), and Trp47Phe, Trp100Phe (apoMFF), were produced. The recombinant proteins expressed in E. coli were recovered from inclusion bodies using reducing and denaturing conditions. The N-terminal histidine tag allowed purification of the proteins using chromatography on a nickel column. The apoM protein was refolded through dialysis, a protocol including oxidized and reduced glutathione. Pure apoM was obtained after ion-exchange chromatography on Q-Sepharose. The N-terminal histidine tag was cleaved off with thrombin, the apoM was again purified on a Q-Sepharose column, and the remaining thrombin was removed on a small SP-Sepharose column. The purified apoM variants migrated as single bands on SDS-PAGE under both reducing and nonreducing conditions, the apparent molecular masses being just below 20 kDa (Fig. 2). Even on heavily overloaded gels (data not shown), no additional bands were seen and the apoM variants were judged to be >95% pure. The yield of refolded apoMWW per liter of bacterial culture varied between 50 and 75 mg. The yield of apoMWF was similar to that of apoMWW, whereas the yields of apoMFW and apoMFF were much lower (i.e., 10–30% of the yield of apoMWW). The low yield of these two mutants was attributable mainly to poor recovery during the refolding process. On nondenaturing agarose gel electrophoresis, apoM migrated as a homogenous band to the α2 region (data not shown). On gel filtration chromatography, the elution volume of apoMWW was similar to that of RBP, suggesting apoM to be monomeric. ApoMWW was stable upon storage in a refrigerator for weeks. It was highly soluble and could have been concentrated to 30 mg/ml without any visible precipitation. ApoMQ22A was expressed in HEK 293 cells and used to compare with the prokaryotically expressed apoM. The Q22A mutation created a signal peptidase cleavage site, and stably transfected HEK 293 cells expressed soluble apoM (residues 22–188) lacking the signal peptide at a concentration of ∼3 mg/l. The yield of purified apoMQ22A was 200 μg/l Optimem medium. The purified apoMQ22A migrated as two bands on SDS-PAGE under both reducing and nonreducing conditions, which corresponded to glycosylated (20 kDa) and nonglycosylated (15 kDa) apoM (Fig. 2A, B). On gel filtration chromatography, the elution volume of apoMQ22A was similar to that of apoMWW. A sandwich ELISA for apoM based on two monoclonal antibodies (mab42 and mab58) was used to compare the immunoreactivities of the recombinant apoM variants with that of apoM in plasma. On Western blotting, the two monoclonal antibodies only reacted with unreduced apoM (Fig. 2C, D), showing that their epitopes were conformation-dependent. Similar dose-response curves were obtained with the five recombinant apoM variants as with apoM from plasma (Fig. 3). The ELISA was used to determine the concentration of the mutants, using the apoMWW values to create the standard curve. The reason for this approach was that the 280 nm absorbance is dependent on the number of tryptophans in the molecule. A semiquantitative Western blot using polyclonal antibodies against apoM confirmed that the concentration determination made with the ELISA was correct (data not shown). Tryptophan is the dominant amino acid contributor to the fluorescence of a protein. The quantum yield at the maximum wavelength associated with intrinsic fluorescence is very sensitive to the polarity of the environment (18.Hutnik C.M. MacManus J.P. Banville D. Szabo A.G. Metal-induced changes in the fluorescence properties of tyrosine and tryptophan site-specific mutants of oncomodulin.Biochemistry. 1991; 30: 7652-7660Crossref PubMed Scopus (13) Google Scholar, 19.Xia X. Lin J.T. Kinne R.K. Binding of phlorizin to the isolated C-terminal extramembranous loop of the Na+/glucose cotransporter assessed by intrinsic tryptophan fluorescence.Biochemistry. 2003; 42: 6115-6120Crossref PubMed Scopus (31) Google Scholar) and therefore is used to monitor interactions. When investigating the emission spectra of apoM after excitation at 295 nm, emission maxima for the different proteins were found at 335–350 nm (Fig. 4). ApoMWW expressed in E. coli had a similar spectrum as apoMQ22A expressed in HEK 293 cells (Fig. 4). The apoM mutant without any tryptophans (apoMFF) did not give any intrinsic fluorescence and served as a negative control, showing that the intrinsic fluorescence was highly specific for tryptophans (Fig. 4). The emission spectra for apoMWW, apoMFW, and apoMWF had different emission maxima. The sensitivity to the polarity of the immediate environment and the fact that the two mutants apoMFW and apoMWF have only one tryptophan were sufficient to alter the emission maximum of the protein (20.Grimsley G.R. Pace C.N. Determining the fluorescence spectrum of a protein.in: Coligan J.E. Dunn B.M. Speicher D.W. Wingfield P.T. Current Protocols in Protein Science. John Wiley & Sons, New York2004Google Scholar). A panel of small hydrophobic substances was tested in the ligand binding assay. All-trans-retinol and its derivatives all-trans-retinoic acid and 9-cis-retinoic acid quenched the intrinsic fluorescence of apoM in a dose-dependent manner, suggesting binding of these molecules (Figs. 5, 6) . The emission maxima of all apoM variants shifted slightly toward red with increasing concentrations of all-trans-retinol, possibly as a result of a conformational change in apoM upon the binding of retinol. The quenching data fitted well to equations 1 a
Referência(s)