Oxidized Alkyl Phospholipids Are Specific, High Affinity Peroxisome Proliferator-activated Receptor γ Ligands and Agonists
2001; Elsevier BV; Volume: 276; Issue: 19 Linguagem: Inglês
10.1074/jbc.m100878200
ISSN1083-351X
AutoresSean S. Davies, Aaron V. Pontsler, Gopal K. Marathe, Kathleen A. Harrison, Robert C. Murphy, Jerald C. Hinshaw, Glenn D. Prestwich, Andy St. Hilaire, Stephen M. Prescott, Guy A. Zimmerman, Thomas M. McIntyre,
Tópico(s)Metabolism, Diabetes, and Cancer
ResumoSynthetic high affinity peroxisome proliferator-activated receptor (PPAR) agonists are known, but biologic ligands are of low affinity. Oxidized low density lipoprotein (oxLDL) is inflammatory and signals through PPARs. We showed, by phospholipase A1 digestion, that PPARγ agonists in oxLDL arise from the small pool of alkyl phosphatidylcholines in LDL. We identified an abundant oxidatively fragmented alkyl phospholipid in oxLDL, hexadecyl azelaoyl phosphatidylcholine (azPC), as a high affinity ligand and agonist for PPARγ. [3H]azPC bound recombinant PPARγ with an affinity (Kd(app) ≈40 nm) that was equivalent to rosiglitazone (BRL49653), and competition with rosiglitazone showed that binding occurred in the ligand-binding pocket. azPC induced PPRE reporter gene expression, as did rosiglitazone, with a half-maximal effect at 100 nm. Overexpression of PPARα or PPARγ revealed that azPC was a specific PPARγ agonist. The scavenger receptor CD36 is encoded by a PPRE-responsive gene, and azPC enhanced expression of CD36 in primary human monocytes. We found that anti-CD36 inhibited azPC uptake, and it inhibited PPRE reporter induction. Results with a small molecule phospholipid flippase mimetic suggest azPC acts intracellularly and that cellular azPC accumulation was efficient. Thus, certain alkyl phospholipid oxidation products in oxLDL are specific, high affinity extracellular ligands and agonists for PPARγ that induce PPAR-responsive genes. Synthetic high affinity peroxisome proliferator-activated receptor (PPAR) agonists are known, but biologic ligands are of low affinity. Oxidized low density lipoprotein (oxLDL) is inflammatory and signals through PPARs. We showed, by phospholipase A1 digestion, that PPARγ agonists in oxLDL arise from the small pool of alkyl phosphatidylcholines in LDL. We identified an abundant oxidatively fragmented alkyl phospholipid in oxLDL, hexadecyl azelaoyl phosphatidylcholine (azPC), as a high affinity ligand and agonist for PPARγ. [3H]azPC bound recombinant PPARγ with an affinity (Kd(app) ≈40 nm) that was equivalent to rosiglitazone (BRL49653), and competition with rosiglitazone showed that binding occurred in the ligand-binding pocket. azPC induced PPRE reporter gene expression, as did rosiglitazone, with a half-maximal effect at 100 nm. Overexpression of PPARα or PPARγ revealed that azPC was a specific PPARγ agonist. The scavenger receptor CD36 is encoded by a PPRE-responsive gene, and azPC enhanced expression of CD36 in primary human monocytes. We found that anti-CD36 inhibited azPC uptake, and it inhibited PPRE reporter induction. Results with a small molecule phospholipid flippase mimetic suggest azPC acts intracellularly and that cellular azPC accumulation was efficient. Thus, certain alkyl phospholipid oxidation products in oxLDL are specific, high affinity extracellular ligands and agonists for PPARγ that induce PPAR-responsive genes. peroxisome proliferator-activated receptor 1-hexadecyl-2-azelaoyl-sn-glycero-3-phosphocholine tris tosylamine low density lipoprotein oxidized LDL PPAR-response element retinoid X receptor hydroxyoctadecadienoic acid prostaglandin 15-deoxy-Δ12,14-prostaglandin J2 high pressure liquid chromatography reversed phase HPLC phosphate-buffered saline platelet-activating factor phosphatidylcholine fluorescein isothiocyanate The transcription factor PPARγ,1 in association with its 9-cis-retinoate-binding RXR partner, controls metabolic and cellular differentiation genes that contain variations of a cognate PPAR-response element (1Desvergne B. Wahli W. Endocr. Rev. 1999; 20: 649-688Crossref PubMed Scopus (2744) Google Scholar). PPARγ, like other members of the broad nuclear hormone receptor family, undergoes a conformational change when it binds specific lipid ligands. This structural reorganization alters its associated proteins, releasing transcriptional inhibitors and recruiting transcriptional co-activators. The regulation of PPARγ function is therefore controlled by lipid ligand binding. A number of synthetic ligands for PPARγ are known. One of these, rosiglitazone (BRL49653), binds with high affinity and is widely prescribed as an insulin sensitizer in type II diabetes. However, defining relevant biologic ligands has been problematic. Several oxidatively modified fatty acids bind and activate PPARγ, including 15-deoxy-Δ12,14-prostaglandin J2(15-deoxy-PGJ2), other arachidonate metabolites (2Forman B.M. Tontonoz P. Chen J. Brun R.P. Spiegelman B.M. Evans R.M. Cell. 1995; 83: 803-812Abstract Full Text PDF PubMed Scopus (2739) Google Scholar, 3Kliewer S.A. Lenhard J.M. Wilson T.M. Patel I. Morris D.C. Lehmann J.M. Cell. 1995; 83: 813-819Abstract Full Text PDF PubMed Scopus (1871) Google Scholar), the linoleate derivatives 9-HODE and 13-HODE (4Lenz M.L. Hughes H. Mitchell J.R. Via D.P. Guyton J.R. Taylor A.A. Gotto Jr., A.M. Smith C.V. J. Lipid Res. 1990; 31: 1043-1050Abstract Full Text PDF PubMed Google Scholar), and several free fatty acids (5Kliewer S.A. Sundseth S.S. Jones S.A. Brown P.J. Wisely G.B. Koble C.S. Devchand P. Wahli W. Willson T.M. Lenhard J.M. Lehmann J.M. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 4318-4323Crossref PubMed Scopus (1898) Google Scholar, 6Forman B.M. Chen J. Evans R.M. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 4312-4317Crossref PubMed Scopus (1870) Google Scholar). However, none of these are particularly potent agonists, and for some their presence at concentrations sufficient to activate PPARγ can be questioned. For example, the oxygenated fatty acid products described above do not confer much advantage in potency over activation by free arachidonate (7Spiegelman B.M. Cell. 1998; 93: 153-155Abstract Full Text Full Text PDF PubMed Scopus (129) Google Scholar) where a concentration of several micromolar is required to elicit a response. Moreover, the 9- and 13-HODEs and their peroxides found in oxidized LDL (8Folcik V.A. Cathcart M.K. J. Lipid Res. 1994; 35: 1570-1582Abstract Full Text PDF PubMed Google Scholar) or in skin exposed to the tumor promoter phorbol myristate acetate (9Beckman J.K. Bagheri F. Ji C. Blair I.A. Marnett L.J. Carcinogenesis. 1994; 15: 2937-2944Crossref PubMed Scopus (16) Google Scholar) are esterified, yet only the free forms of these lipids, and not the intact phospholipids from these sources, are PPAR ligands (10Nagy L. Tontonoz P. Alvarez J.G. Chen H. Evans R.M. Cell. 1998; 93: 229-240Abstract Full Text Full Text PDF PubMed Scopus (1599) Google Scholar). Additionally, it is unlikely that the PPARγ and PPARα agonist 15-deoxy-PGJ2 (2Forman B.M. Tontonoz P. Chen J. Brun R.P. Spiegelman B.M. Evans R.M. Cell. 1995; 83: 803-812Abstract Full Text PDF PubMed Scopus (2739) Google Scholar, 3Kliewer S.A. Lenhard J.M. Wilson T.M. Patel I. Morris D.C. Lehmann J.M. Cell. 1995; 83: 813-819Abstract Full Text PDF PubMed Scopus (1871) Google Scholar) accumulates in vivo, and it now appears that little 15-deoxy-PGJ2 is actually present in commercial sources of this reactive and unstable lipid (11Maxey K.M. Hessler E. MacDonald J. Hitchingham L. Prostaglandins Other Lipid Mediat. 2000; 62: 15-21Crossref PubMed Scopus (36) Google Scholar). Oxidation of LDL creates unknown PPARγ agonists (10Nagy L. Tontonoz P. Alvarez J.G. Chen H. Evans R.M. Cell. 1998; 93: 229-240Abstract Full Text Full Text PDF PubMed Scopus (1599) Google Scholar). This process also creates PPARα ligands (12Lee H. Shi W. Tontonoz P. Wang S. Subbanagounder G. Hedrick C.C. Hama S. Borromeo C. Evans R.M. Berliner J.A. Nagy L. Circ. Res. 2000; 87: 516-521Crossref PubMed Scopus (273) Google Scholar, 13Delerive P. Furman C. Teissier E. Fruchart J. Duriez P. Staels B. FEBS Lett. 2000; 471: 34-38Crossref PubMed Scopus (169) Google Scholar), the bulk of which depend on liberation by phospholipase A2 to free fatty acid oxidation products (13Delerive P. Furman C. Teissier E. Fruchart J. Duriez P. Staels B. FEBS Lett. 2000; 471: 34-38Crossref PubMed Scopus (169) Google Scholar). PPARα alters lipid metabolism, enhances lipid oxidation, and often dampens inflammatory events and signaling pathways (14Fruchart J.C. Duriez P. Staels B. Curr. Opin. Lipidol. 1999; 10: 245-257Crossref PubMed Scopus (405) Google Scholar, 15Poynter M.E. Daynes R.A. J. Biol. Chem. 1998; 273: 32833-32841Abstract Full Text Full Text PDF PubMed Scopus (493) Google Scholar). PPARγ has a distinct profile of activities as it promotes adipogenesis through differentiation of preadipocytes, and it may have a complex role in atherogenesis (1Desvergne B. Wahli W. Endocr. Rev. 1999; 20: 649-688Crossref PubMed Scopus (2744) Google Scholar, 16Auwerx J. Diabetologia. 1999; 42: 1033-1049Crossref PubMed Scopus (583) Google Scholar). In part, its pro-atherogenic effects may occur through the formation of foam cells by stimulating CD36 expression (17Tontonoz P. Nagy L. Alvarez J.G. Thomazy V.A. Evans R.M. Cell. 1998; 93: 241-252Abstract Full Text Full Text PDF PubMed Scopus (1617) Google Scholar, 18Feng J. Han J. Pearce S.F. Silverstein R.L. Gotto Jr., A.M. Hajjar D.P. Nicholson A.C. J. Lipid Res. 2000; 41: 688-696Abstract Full Text Full Text PDF PubMed Google Scholar). CD36 is a member of the scavenger receptor family that promotes the uptake of oxidized LDL, driving macrophages to a lipid-surfeit state characterized by foamy fatty inclusions. Cells in atherosclerotic lesions express PPARγ (19Marx N. Sukhova G. Murphy C. Libby P. Plutzky J. Am. J. Pathol. 1998; 153: 17-23Abstract Full Text Full Text PDF PubMed Scopus (510) Google Scholar, 20Ricote M. Huang J. Fajas L. Li A. Welch J. Najib J. Witztum J.L. Auwerx J. Palinski W. Glass C.K. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 7614-7619Crossref PubMed Scopus (683) Google Scholar), and CD36 is induced by PPARγ agonists present in oxidized LDL (10Nagy L. Tontonoz P. Alvarez J.G. Chen H. Evans R.M. Cell. 1998; 93: 229-240Abstract Full Text Full Text PDF PubMed Scopus (1599) Google Scholar, 18Feng J. Han J. Pearce S.F. Silverstein R.L. Gotto Jr., A.M. Hajjar D.P. Nicholson A.C. J. Lipid Res. 2000; 41: 688-696Abstract Full Text Full Text PDF PubMed Google Scholar). CD36 ligation and internalization of LDL particles oxidized by monocytes is blocked by an excess of oxidized phospholipids (21Podrez E.A. Febbraio M. Sheibani N. Schmitt D. Silverstein R.L. Hajjar D.P. Cohen P.A. Frazier W.A. Hoff H.F. Hazen S.L. J. Clin. Invest. 2000; 105: 1095-1108Crossref PubMed Scopus (366) Google Scholar), suggesting that one or more oxidized phospholipids is a CD36 ligand responsible for the internalization of oxLDL. Here we show that a complex lipid (i.e. a phospholipid) formed by oxidative attack on a subclass of LDL phospholipids is effectively internalized by CD36 and is a high affinity, selective PPARγ ligand and agonist. Alkyl phosphatidylcholines, which consist of a small portion of the LDL phosphatidylcholine pool (22Diagne A. Fauvel J. Record M. Chap H. Douste-Blazy L. Biochim. Biophys. Acta. 1984; 793: 221-231Crossref PubMed Scopus (154) Google Scholar), are the sole precursors for these agonists because there is selectivity for thesn-1 bond in both binding and PPRE reporter activation. We conclude that certain oxidized alkyl phospholipids define a new class of high affinity agonists for PPARγ, and because these are found in oxidized LDL, they may contribute to its biologic effects. PAF was from Avanti Polar Lipids; lyso-PAF (1-O-hexadecyl-sn-glycero-3-phosphocholine), 9-cis-retinoate, pirinixic acid (WY14643), and 15-deoxy-PGJ2 were from Biomol; Rhizopus arrhizus lipase was from Roche Molecular Biochemicals and then Sigma; [3H]rosiglitazone and rosiglitazone were from the American Radiolabeled Chemicals. ECL kits were from Amersham Pharmacia Biotech; the SV40-β-galactosidase reporter was from Promega (Madison, WI); the CD36 reporter constructs with (−273) and without (−261) its PPRE (17Tontonoz P. Nagy L. Alvarez J.G. Thomazy V.A. Evans R.M. Cell. 1998; 93: 241-252Abstract Full Text Full Text PDF PubMed Scopus (1617) Google Scholar) were constructed from its reported sequence (23Armesilla A.L. Vega M.A. J. Biol. Chem. 1994; 269: 18985-18991Abstract Full Text PDF PubMed Google Scholar). The forward primers (for CD36−273) were 5′-GCGACGCGTCTGGCCTCTGACTTACTTGG-3′ or (CD36−261) 5′-GCGACGCGTTTACTTGGATGGGAACTAGCC-3′, and the reverse primer was 5′-GGAAGATCTAGTCCTACACTGCAGTCCTC-3′. The amplicon was inserted into pGL3b at the MluI and BcgII sites. pGL3b was fromPromega, and Probond Ni+ beads were from Invitrogen. The blocking anti-CD36 antibody 185–1G2, without azide, was from NeoMarkers (Fremont, CA); the FITC-conjugated anti-CD36 antibody CLB-IVC7 used for flow analysis was from Accurate Chemicals (Westbury, NY); and the anti-ICAM-3 antibody CAL3.10 (BBA29) was from R & D Systems (Minneapolis MN). The flippase mimetic I (24Boon J.M. Smith B.D. J. Am. Chem. Soc. 1999; 121: 11924-11925Crossref Scopus (42) Google Scholar) tris (tosylaminoethyl)amine (TTA) was synthesized as described (25Valiyaveeettil S. Engbersen J.F.J. Verboom W. Reinhoudt D.N. Angew. Chem. Int. Ed. Engl. 1993; 32: 900-901Crossref Scopus (229) Google Scholar). LDL was oxidized with 20 μmCuSO4 at 37 oC overnight, and oxidized phospholipids were purified by RP-HPLC (26Heery J.M. Kozak M. Stafforini D.M. Jones D.A. Zimmerman G.A. McIntyre T.M. Prescott S.M. J. Clin. Invest. 1995; 96: 2322-2330Crossref PubMed Scopus (279) Google Scholar). A portion of the recovered fractions was treated with phospholipase A1 as before (27Marathe G.K. Davies S.S. Harrison K.A. Silva A.R. Murphy R.C. Castro-Faria Neto H. Prescott S.M. Zimmerman G.A. McIntyre T.M. J. Biol. Chem. 1999; 274: 28395-28405Abstract Full Text Full Text PDF PubMed Scopus (157) Google Scholar), except that the enzyme was from Sigma. The lipid azPC was synthesized from 1-O-hexadecyl-sn-glycero-3-phosphocholine (hexadecyl lyso-PC; after mild alkaline hydrolysis; 0.5 nNaOH in methanol; 4 h; 24 oC) and 1-palmitoyl-2-azelaoyl-sn-glycero-3-phosphocholine (palmitoyl azelaoyl-PC) was synthesized in a similar fashion from palmitoyl lyso-PC. After neutralization, purified lipid (2 mg) was reacted with 10 mg of azelaic anhydride (University of Utah Chemical Synthesis Facility) in the presence of 1 mg of 4-(N,N-dimethylamino)pyridine in CHCl3:pyridine (4:1) for 36 h before purification by RP-HPLC. The mass of each synthetic phospholipid was determined by phosphorus analysis (28Ames B.N. Dubin D.T. J. Biol. Chem. 1960; 235: 769-775Abstract Full Text PDF PubMed Google Scholar). Mass spectroscopy of lipid oxidation products was performed as before (27Marathe G.K. Davies S.S. Harrison K.A. Silva A.R. Murphy R.C. Castro-Faria Neto H. Prescott S.M. Zimmerman G.A. McIntyre T.M. J. Biol. Chem. 1999; 274: 28395-28405Abstract Full Text Full Text PDF PubMed Scopus (157) Google Scholar). [3H]Hexadecyl azelaic phosphatidylcholine was synthesized from [3H]hexadecyl-sn-glycero-3-phosphocholine (PerkinElmer Life Sciences) and HPLC-purified in a similar fashion. Human monocytes were isolated by counter-current elutriation (29Elstad M.R. Prescott S.M. McIntyre T.M. Zimmerman G.A. J. Immunol. 1988; 140: 1618-1624PubMed Google Scholar) and resuspended (1 × 106/ml) in Hanks' balanced salt solution with 0.5% human serum albumin and 10 μg/ml polymyxin B. Monocytes were added to plates coated with 10 μg/ml CAL3.10 anti-ICAM-3 monoclonal antibody (30Kessel J.M. Hayflick J. Weyrich A.S. Hoffman P.A. Gallatin M. McIntyre T.M. Prescott S.M. Zimmerman G.A. J. Immunol. 1998; 160: 5579-5587PubMed Google Scholar). CV-1 cells were obtained from ATCC and grown as suggested. Surface expression of CD36 on primary human monocytes was determined by allowing elutriated monocytes to adhere to anti-ICAM3-coated wells for 1 h before the cells were exposed to the lipid agonists, or not, as stated in the figures. Some cells were maintained in a suspended state by gently rocking on a platform rocker in polypropylene tubes as a control. Adherent cells were released from the plate by gentle agitation and scraping and washed three times in PBS containing 1% goat serum. Recovered cells were stained with FITC-labeled anti-CD36 antibody CLB-IVC7 for flow analysis by the University of Utah flow analysis core facility. The acyl-CoA oxidase-luciferase plasmid was described previously (31Meade E.A. McIntyre T.M. Zimmerman G.A. Prescott S.M. J. Biol. Chem. 1999; 274: 8328-8334Abstract Full Text Full Text PDF PubMed Scopus (268) Google Scholar). Plasmids were transformed into TOP10F′Escherichia coli strain using the TA cloning kit. Plasmids from log phase cells were isolated using a Bigger Prep kit (5 Prime → 3 Prime, Inc., Boulder, CO), and purified by CsCl gradients. The His6-tagged PPARγ was constructed similarly using the M13 primer in pCR2.1 that contained full-length PPARγ1 and a primer (5′-CTA ATG ATG ATG ATG ATG ATG GTA CAA GTC CTT GTA G-3′) containing the His6 tag sequence. PPARγ and PPARα expression plasmids were a gift from Beth Meade (University of Utah). Inserts in all plasmids were sequence-verified by the University of Utah sequencing core facility. When PPAR expression plasmids were co-transfected with a reporter construct, 0.5 μg of the relevant plasmid was combined with 1 μg of pGL3-PPRE and 0.1 μg of the SV40-β-galactosidase reporter to normalize transfection efficiencies. All transfections included 1–2 μg of total plasmid, and 5–10 μl of LipofectAMINE per ml of Opti-MEM. Transfection solution was added to CV-1 cells overnight and then removed, and agonist was added in fresh media for 18–20 h. [3H]Rosiglitazone displacement from PPARγ was determined with a carboxy His6-tagged molecule. HeLa cells were transfected with pCR3.1-PPARγ-His6 or pCR3.1 for 21 h with LipofectAMINE and then grown (48 h). The cells were washed twice with PBS and lysed in PBS containing 0.1% Triton X-100 and frozen at −70 °C until used. Transfection with PPARγ-His6 was assessed by immunoblotting with anti-His6 antibody (Santa Cruz Biotechnology). Debris was removed from thawed samples, and 200 μl of lysate was incubated with 50 μl of Probond beads at 4 °C for 1 h in PBS. The beads were washed once by centrifugation before [3H]rosiglitazone, and then unlabeled competitor was added in a final volume of 300 μl of PBS. Samples were incubated with shaking for 3 h at 4 °C before washing three times with PBS and quantitating retained3H. Binding of [3H]azelaic phosphatidylcholine to PPARγ was performed in a similar fashion, although the amount of protein lysate was increased to account for its lower specific radioactivity. Accumulation of [3H]azPC was estimated by incubating monolayers with carrier-free [3H]azPC (22.9 Ci/mmol) for 30 min in the presence or absence of the lipids specified in the figure at a concentration of 10 μm (except HPLC fraction 6 from oxLDL that was used at a concentration that maximally induced ACox reporter expression since there was insufficient material for phosphorus quantitation). Some cells were preincubated for 30 min with 10 μg of the blocking anti-CD36 antibody 185–1G2 or an irrelevant IgG2a isotype-matched control monoclonal antibody. We oxidized human LDL, extracted the lipids, and separated nonpolar lipids (which we found to have no activity in this assay; not shown) from the polar phospholipid oxidation products. We separately examined these polar phospholipid-containing fractions as agonists using CV-1 cells that had been transiently transfected with an acyl-CoA oxidase PPRE-firefly luciferase reporter construct and SV40-β-galactosidase to normalize for transfection efficiency. These purified polar phospholipids stimulated luciferase transcription under the control of this PPRE (Fig. 1), and this activity was concentrated in RP-HPLC fraction 6. We found this material to be as effective an agonist for PPAR-induced transcription as rosiglitazone (BRL49653), and a synthetic oxidized phospholipid (azPC) is discussed in detail below. Fig. 1 shows that an equivalent fraction from the same batch of LDL that had not been oxidized contained little stimulatory activity. Oxidation of LDL creates oxidized phosphatidylcholines that are potent inflammatory agents because they structurally resemble PAF and activate the cloned receptor for PAF. Oxidized phospholipids of this class are all derived from the small pool of alkyl phosphatidylcholines in LDL (22Diagne A. Fauvel J. Record M. Chap H. Douste-Blazy L. Biochim. Biophys. Acta. 1984; 793: 221-231Crossref PubMed Scopus (154) Google Scholar) that are resistant to phospholipase A1digestion (27Marathe G.K. Davies S.S. Harrison K.A. Silva A.R. Murphy R.C. Castro-Faria Neto H. Prescott S.M. Zimmerman G.A. McIntyre T.M. J. Biol. Chem. 1999; 274: 28395-28405Abstract Full Text Full Text PDF PubMed Scopus (157) Google Scholar). To determine whether the oxidatively generated agonists that induce transcription from a PPRE reporter construct also fall into this class, we digested the fractions isolated by HPLC with phospholipase A1. This removes the oxidized diacyl phospholipids, derived from the 99.5% of LDL phospholipids that have an sn-1 ester bond, as assessed by phosphorus analysis (not shown). Phospholipase A1 digestion had no effect on the ability of the fractions isolated from oxidized LDL to stimulate luciferase expression from the PPRE-reporter construct (Fig. 1), indicating that only alkyl phosphatidylcholine oxidation products were effective agonists in this assay. We determined which alkyl phosphatidylcholine oxidation products were present in oxidized LDL by resolving the phospholipase A1-treated phospholipids by reversed phase HPLC and examining these by electrospray tandem mass spectroscopy as precursors of the phosphocholine ion m/z 184. An abundant ion in the HPLC effluent was observed at m/z 652 (Fig.2 a), potentially corresponding to 1-O-hexadecyl-2-azelaoyl-sn-glycero-3-phosphocholine (azPC). This component was maximal in RP-HPLC fraction 6 (Fig. 2,inset), whereas the C4-PAF analogs were most abundant in fractions 7 and 8, as reported previously (27Marathe G.K. Davies S.S. Harrison K.A. Silva A.R. Murphy R.C. Castro-Faria Neto H. Prescott S.M. Zimmerman G.A. McIntyre T.M. J. Biol. Chem. 1999; 274: 28395-28405Abstract Full Text Full Text PDF PubMed Scopus (157) Google Scholar). A structurally unique diagnostic product of sn-2 ω-carboxyl glycerophosphocholine lipids is the collision-induced rearrangement of a methyl group and decomposition to a monomethyl acid and dimethyl lyso-PC (32Kayganich-Harrison K.A. Murphy R.C. Anal. Biochem. 1994; 221: 16-24Crossref PubMed Scopus (29) Google Scholar). Proof that azPC was present was obtained by collision-induced decomposition of the corresponding [M − H]− of m/z 650 (Fig. 2 b) that yielded the expected product ions at m/z 201 and 466 corresponding to monomethyl azelaic acid and the dimethyl lyso-PAF adduct resulting from the loss of the sn-2 methylazelaoyl ketene. azPC is therefore an abundant oxidation product of LDL alkyl phosphatidylcholines. We confirmed this deduction by synthesizing azPC and finding that synthetic azPC produced the same fragmentation pattern as the material isolated from oxidized LDL (not shown). We experimentally determined whether a complex lipid like azPC could function as a ligand for PPARγ. We synthesized [3H]hexadecyl azelaoyl phosphatidylcholine ([3H]azPC) and incubated it with full-length recombinant human PPARγ1. The PPARγ in transfected HeLa cell lysates was immobilized on Ni+ beads through an introduced His6 tag, so tight binding could be assessed after collecting and washing the beads. We found (Fig.3 a) that [3H]azPC bound to PPARγ, and that this binding was dependent on the concentration of immobilized hPPARγ1. We next varied the concentration of [3H]azPC to establish its apparent affinity for PPARγ under these conditions. We found that the binding of [3H]azPC was concentration-dependent and that its apparent affinity was ≈40 nm (Fig. 3 b). However, we also found that different lysates provided different apparent affinities, perhaps reflecting a similar wide disparity in reported apparent affinities for rosigitazone (2Forman B.M. Tontonoz P. Chen J. Brun R.P. Spiegelman B.M. Evans R.M. Cell. 1995; 83: 803-812Abstract Full Text PDF PubMed Scopus (2739) Google Scholar, 33Lehmann J.M. Moore L.B. Smith-Oliver T.A. Wilkison W.O. Willson T.M. Kliewer S.A. J. Biol. Chem. 1995; 270: 12953-12956Abstract Full Text Full Text PDF PubMed Scopus (3469) Google Scholar, 34Shao D. Rangwala S.M. Bailey S.T. Krakow S.L. Reginato M.J. Lazar M.A. Nature. 1998; 396: 377-380Crossref PubMed Scopus (311) Google Scholar). To determine better whether azPC bound PPARγ as effectively as rosigitazone, we directly compared [3H]azPC binding with [3H]rosiglitazone binding at low concentrations and found (Fig. 3 c) that [3H]azPC binding precisely mirrored the binding of [3H]rosiglitazone to immobilized PPARγ1. Rosiglitazone co-crystallizes with the ligand-binding domain of PPARγ (35Nolte R.T. Wisely G.B. Westin S. Cobb J.E. Lambert M.H. Kurokawa R. Rosenfeld M.G. Willson T.M. Glass C.K. Milburn M.V. Nature. 1998; 395: 137-143Crossref PubMed Scopus (1699) Google Scholar, 36Gampe R.T. Montana V.G. Lambert M.H. Miller A.B. Bledsoe R.K. Milburn M.V. Kliewer S.A. Wilson T.M. Xu H.E. Mol. Cell. 2000; 5: 545-555Abstract Full Text Full Text PDF PubMed Scopus (521) Google Scholar) in the ligand-binding pocket, so displacement of [3H]rosiglitazone tests binding in this pocket. We found (Fig. 4 a) that azPC displaced the standard ligand [3H]rosiglitazone in a concentration-dependent fashion and that the concentration relationship of this competition was identical to that of unlabeled rosiglitazone. azPC bound PPARγ only through this ligand-binding pocket because this competition with rosiglitazone was complete. We examined the effect of other lipids as competitors at 6.7 μm where azPC and rosiglitazone competition was complete and found (Fig. 4 b) that PAF and 15-deoxy-PGJ2could only displaced about a third of the [3H]rosiglitazone at this concentration. Free azelaic acid, like arachidonate, was an ineffective competitor. We performed the converse experiment where unlabeled rosiglitazone or azPC was used to displace [3H]azPC bound to immobilized PPARγ. We found (Fig. 4 c) that unlabeled rosiglitazone displaced nearly all of the [3H]azPC from PPARγ as its concentration was increased, just as unlabeled azPC displaced this bound [3H]azPC. We tested other lipids for their ability to displace [3H]azPC and found (Fig. 4 d) that PAF, and to a lesser extent lyso-PAF, was a modest competitor, but that 9-HODE or 13-HODE were unable to displace bound [3H]azPC. We tested palmitoyl azelaoyl-PC, the diacyl homolog of hexadecyl azelaoyl PC, as a PPARγ ligand, and we found it to be 10–100-fold less potent as a competitor (not shown). azPC was a ligand for PPARγ, so we next determined whether it was an agonist. We transfected CV-1 cells with a luciferase reporter under the control of the PPRE from acyl-CoA oxidase, along with SV40-β−galactosidase as a transfection control, and then treated the cells with rosiglitazone, synthetic azPC, or 9-cis-retinoate to activate RXR. Rosiglitazone induced a 3.4-fold increase in reporter expression, and azPC induced a 3.9-fold increase (Fig.5 a) in this assay. Fig. 1presented a similar result where azPC induced a 3.9-fold increase in ACox reporter expression. Activation of just the RXR subunit with 9-cis-retinoate, which can act as a phantom ligand (37Schulman I.G. Li C. Schwabe J.W. Evans R.M. Genes Dev. 1997; 11: 299-308Crossref PubMed Scopus (115) Google Scholar), was not as effective as either PPARγ ligand in stimulating reporter expression. We compared the concentration-response relationship of azPC and its acyl analog with rosiglitazone as PPARγ agonists. CV-1 cells were transfected with the ACox reporter and SV40-β-galactosidase for normalization, and we then treated with increasing concentrations of azPC, diacyl azPC, or rosiglitazone. We found that azPC induced a concentration-dependent increase in reporter expression starting by about 10−8m (Fig.5 b). Half-maximal activation occurred by 10−7m, and maximal expression was achieved by 1 μm. Rosiglitazone also induced a concentration-dependent increase, and this relationship was identical to that of azPC. We also examined the diacyl homolog of azPC, 1-palmitoyl-2-azelaoyl-sn-glycero-3-phosphocholine, as an agonist; we found that it was about 100-fold less effective than azPC and only began to elicit an effect at 10 μm. We determined the specificity of azPC as a PPARγ agonist by transfecting CV-1 cells with PPARα or PPARγ expression plasmids in addition to the ACox reporter and an SV40 β-galactosidase transfection control. Rosiglitazone induced an 8-fold increase in reporter expression in this experiment (Fig. 6) in untransfected cells that rely on their endogenous PPARs. This was a more robust response than the 5.3-fold increase induced by the PPARα-selective agonist WY14643 or the 5-fold increase induced by azPC in these control cells. When PPARγ was overexpressed in these cells, the response to rosiglitazone was markedly enhanced (a 25-fold induction), as was the response to azPC (an 18-fold induction). In contrast, overexpression of PPARα failed to enhance the existing response to either azPC or rosiglitazone. The ectopic PPARα was functional because the PPARα-selective agonist WY14643 enhanced expression in cells overexpressing PPARα and not in cells transfected with PPARγ. We determined whether endogenous PPAR-regulated genes were induced by azPC, and for this we chose CD36, a scavenger receptor that binds and internalizes oxidized LDL particles. Intriguingly, transcription of CD36 in monocytes is stimulated by unknown ligands associated with oxidized LDL (10Nagy L. Tontonoz P. Alvarez J.G. Chen H. Evans R.M. Cell. 1998; 93: 229-240Abstract Full Text Full Text PDF PubMed Scopus (1599) Google Scholar, 18Feng J. Han J. Pearce S.F. Silverstein R.L. Gotto Jr.,
Referência(s)