Portal de Periódicos da CAPES

Transcriptional Induction of Endothelial Nitric Oxide Synthase Type III by Lysophosphatidylcholine

1995; Elsevier BV; Volume: 270; Issue: 28 Linguagem: Inglês

10.1074/jbc.270.28.17006

1083-351X

Artur Zembowicz, Jih‐Luh Tang, Kenneth K. Wu,

Neutrophil, Myeloperoxidase and Oxidative Mechanisms

Endothelial synthesis of NO is catalyzed by constitutive NO synthase type III (NOS-III). NOS-III has been thought to be regulated mainly at the level of enzyme activity by intracellular calcium. We report that in human umbilical vein endothelial cells lysophosphatidylcholine (lyso-PC), a component of atherogenic lipoproteins and atherosclerotic lesions, increases NOS-III mRNA and protein levels. This leads to the augmentation of NOS-III activity and the enhancement of anti-platelet properties of endothelial cells. Importantly, nuclear run-off experiments demonstrate a transcriptional mechanism of regulation of NOS-III expression by lysophosphatidylcholine. As endothelium-derived NO appears to be an anti-atherogenic molecule, induction of NOS-III by lyso-PC may be a protective response that limits the progress of the atherosclerotic lesion and promotes its regression. Endothelial synthesis of NO is catalyzed by constitutive NO synthase type III (NOS-III). NOS-III has been thought to be regulated mainly at the level of enzyme activity by intracellular calcium. We report that in human umbilical vein endothelial cells lysophosphatidylcholine (lyso-PC), a component of atherogenic lipoproteins and atherosclerotic lesions, increases NOS-III mRNA and protein levels. This leads to the augmentation of NOS-III activity and the enhancement of anti-platelet properties of endothelial cells. Importantly, nuclear run-off experiments demonstrate a transcriptional mechanism of regulation of NOS-III expression by lysophosphatidylcholine. As endothelium-derived NO appears to be an anti-atherogenic molecule, induction of NOS-III by lyso-PC may be a protective response that limits the progress of the atherosclerotic lesion and promotes its regression. Smurf1 protein negatively regulates interferon-γ signaling through promoting STAT1 protein ubiquitination and degradationJournal of Biological ChemistryVol. 289Issue 43PreviewVOLUME 287 (2012) PAGES 17006–17015 Full-Text PDF Open Access Endothelium-derived NO is a key molecule regulating various physiological processes occurring at the interphase between the blood and vascular wall(1Moncada S. Higgs E.A. N. Engl. J. Med. 1993; 329: 2002-2011Crossref PubMed Scopus (5758) Google Scholar). Endothelial synthesis of NO is catalyzed by constitutively expressed calcium- and calmodulin-dependent NO synthase type III (NOS-III)1 1The abbreviations used are: NOSnitric oxide synthaselyso-PClysophosphatidylcholineHUVEChuman umbilical vein endothelial cellsNOnitric oxideL-NAMENω-nitro-L-arginineLDL(s)low density lipoprotein(s). (2Pollock J.S. Forstermann U. Mitchell J.A. Warner T.D. Schmidt H.H. Nakane M. Murad F. Proc. Natl. Acad. Sci. U. S. A. 1991; 88: 10480-10484Crossref PubMed Scopus (900) Google Scholar). In contrast to calcium-independent NOS type II that can be induced in many cell types by cytokines, NOS-III has been thought to be regulated mainly by changes in the intracellular concentrations of calcium (3-5). The importance of regulation of NOS-III expression as a means of regulation of endothelial synthesis of NO production has only recently emerged as a possibility. It has been shown that NOS-III mRNA levels in cultured human umbilical vein endothelial cells are reduced by tumor necrosis factor-β and that this is due to de-stabilization of NOS-III mRNA(6Yoshizumi M. Perrella M.A. Burnett Jr., J.C. Lee M.E. Circ. Res. 1993; 73: 205-209Crossref PubMed Scopus (705) Google Scholar). Several recent reports also indicated that NOS-III expression could be up-regulated. Increased levels of NOS-III mRNA levels have been reported in bovine endothelial cells subjected to shear stress(7Nishida K. Harrison D.G. Navas J.P. Fisher A.A. Dockery S.P. Uematsu M. Nerem R.M. Alexander R.W. Murphy T.J. J. Clin. Invest. 1992; 90: 2092-2096Crossref PubMed Scopus (616) Google Scholar), in aortas isolated form dogs undergoing exercise training(8Sessa W.C. Pritchard K. Seyedi N. Wang J. Hintze T.H. Circ. Res. 1994; 74: 349-353Crossref PubMed Scopus (827) Google Scholar), and in guinea pigs following treatment with estrogens (9). However, in no case mechanisms leading to the increase of NOS-III mRNA levels have been elucidated. Recent cloning of the human NOS-III gene revealed the presence in a putative promoter region of the NOS-III gene of several potentially cis-acting regulatory elements which in other genes have been demonstrated to regulate gene expression in response to cAMP, cholesterol, protein kinase C activation, transforming growth factor β, and shear stress(10Marsden P.A. Heng H.H. Scherer S.W. Stewart R.J. Hall A.V. Shi X.M. Tsui L.C. Schappert K.T. J. Biol. Chem. 1993; 268: 17478-17488Abstract Full Text PDF PubMed Google Scholar). Whether any of these regulatory elements is involved in regulation of the NOS-III gene in response to extracellular stimulation has not been established yet. In fact, no molecule has been conclusively demonstrated to induce NOS-III expression at the level of gene transcription so far. nitric oxide synthase lysophosphatidylcholine human umbilical vein endothelial cells nitric oxide Nω-nitro-L-arginine low density lipoprotein(s). Although under certain conditions high levels of NO may induce tissue injury(11Beckman J.S. Beckman T.W. Chen J. Marshall P.A. Freeman B.A. Proc. Natl. Acad. Sci. U. S. A. 1990; 87: 1620-1624Crossref PubMed Scopus (6731) Google Scholar), the inhibitory actions of endothelium-derived NO on vascular tone(12Furchgott R.F. Zawadzki J.V. Nature. 1980; 288: 373-376Crossref PubMed Scopus (10026) Google Scholar), platelet activation(13Radomski M.W. Palmer R.M. Moncada S. Lancet. 1987; 2: 1057-1058Abstract PubMed Scopus (1234) Google Scholar), leukocyte adhesion(14Bath P.M. Hassall D.G. Gladwin A.M. Palmer R.M. Martin J.F. Arterioscler. Thromb. 1991; 11: 254-260Crossref PubMed Scopus (301) Google Scholar), vascular smooth muscle cell proliferation, and de-differentiation (15Cornwell T.L. Arnold E. Boerth N.J. Lincoln T.M. Am. J. Physiol. 1994; 267: C1405-C1413Crossref PubMed Google Scholar, 16Lincoln T.M. Komalavilas P. Cornwell T.L. Hypertension. 1994; 23: 1141-1147Crossref PubMed Scopus (173) Google Scholar) suggest that NO is a protective and anti-atherogenic molecule(17Cooke J.P. Tsao P.S. Arterioscler. Thromb. 1994; 14: 653-655Crossref PubMed Google Scholar). Likewise, inhibitors of NOS enhance atherosclerosis in hypercholesterolemic rabbits(18Cayatte A.J. Palacino J.J. Horten K. Cohen R.A. Arterioscler. Thromb. 1994; 14: 753-759Crossref PubMed Scopus (465) Google Scholar, 19Naruse K. Shimizu K. Muramatsu M. Toki Y. Miyazaki Y. Okumura K. Hashimoto H. Ito T. Arterioscler. Thromb. 1994; 14: 746-752Crossref PubMed Scopus (244) Google Scholar), whereas L-arginine, a substrate for NOS, has a protective effect(20Cooke J.P. Singer A.H. Tsao P. Zera P. Rowan R.A. Billingham M.E. J. Clin. Invest. 1992; 90: 1168-1172Crossref PubMed Scopus (638) Google Scholar). Several lines of evidence suggest a role for lysophosphatidylcholine (lyso-PC) in atherogenesis(21Vidaver G.A. Ting A. Lee J.W. J. Theor. Biol. 1985; 115: 27-41Crossref PubMed Scopus (39) Google Scholar). Lyso-PC content of atherosclerotic arteries is severalfold higher than that of normal vessels(22Portman O.W. Alexander M. J. Lipid Res. 1969; 10: 158-165Abstract Full Text PDF PubMed Google Scholar). Following pro-atherogenic modification of low density lipoproteins lyso-PC may constitute up to 40% of their total lipid content(23Parthasarathy S. Steinbrecher U.P. Barnett J. Witztum J.L. Steinberg D. Proc. Natl. Acad. Sci. U. S. A. 1985; 82: 3000-3004Crossref PubMed Scopus (306) Google Scholar). Lyso-PC is a chemoattractant for human monocytes(24Quinn M.T. Parthasarathy S. Steinberg D. Proc. Natl. Acad. Sci. U. S. A. 1988; 85: 2805-2809Crossref PubMed Scopus (585) Google Scholar, 25Parthasarathy S. Quinn M.T. Schwenke D.C. Carew T.E. Steinberg D. Arteriosclerosis. 1989; 9: 398-404Crossref PubMed Google Scholar). Lyso-PC induces monocytic cell expression of heparin binding epidermal growth factor(26Nakano T. Raines E.W. Abraham J.A. Klagsbrun M. Ross R. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 1069-1073Crossref PubMed Scopus (123) Google Scholar). Importantly, lyso-PC causes induction of several endothelial genes expressed in early atherosclerosis, such as vascular adhesion molecule-1, intercellular adhesion molecule-1, platelet-derived growth factor chains A and B, and heparin-binding epidermal growth factor(27Kume N. Cybulsky M.I. Gimbrone Jr., M.A. J. Clin. Invest. 1992; 90: 1138-1144Crossref PubMed Scopus (724) Google Scholar, 28Kume N. Gimbrone Jr., M.A. J. Clin. Invest. 1994; 93: 907-911Crossref PubMed Scopus (299) Google Scholar). In this communication we present experimental data that suggest that lyso-PC may also induce vasoprotective endothelial genes. Our findings demonstrate that lyso-PC enhances NOS-III gene transcription in cultured human umbilical vein endothelial cells. Lyso-PC (1-palmitoyl-2-hydroxy-sn-glycero-3-phosphocholine) was from Avanti Polar Lipids (Birmingham, AL). Unless otherwise indicated all the other reagents were from Sigma. HUVEC were cultured as described previously (29) in Medium 199 containing 20% bovine calf serum (Hyclone, Logon, UT), 12.5 μg/ml endothelium cell mitogen (Biomedical Technologies, Stoughton, MA), and 100 μg/ml heparin. Only second and third passage cells were used. Twelve hours before lyso-PC treatment cells were transferred to Medium 199 containing 5% fetal bovine serum and no other additives. Total RNA was isolated using Ultraspec (Biotecx Laboratories, Houston, TX). 15-25 μg of RNA was fractionated on 1% agarose and transferred to a positively charged nylon membrane. As NOS-III probe we used a 1.6-kilobase BglII/EcoRI restriction fragment of full-length human NOS-III cDNA cloned in our laboratory (30Chen P.F. Tsai A.L. Wu K.K. J. Biol. Chem. 1994; 269: 25062-25066Abstract Full Text PDF PubMed Google Scholar) and subcloned into EcoRI sites of pcDNA-3 (Invitrogen). Probe labeling, hybridization, and chemiluminescent detection were performed using components of Genius system (Boehringer Mannheim). Hybridization conditions (5 × SSC, 0.02% SDS, 0.1% sarcosyl, 2% blocking reagent, 68°C, overnight) and high stringency washes (0.1 × SSC, 0.1% SDS, 60°C, 45 min) assured no cross-hybridization with any other HUVEC mRNA or NOS-II mRNA induced in cultured human aortic vascular smooth muscle cells by 12-h stimulation with human interleukin-1β (100 pg/ml) and γ-interferon (250 units/ml) (n = 2). Membranes were stripped by boiling in 0.1 × SSC containing 1% SDS for 15 min and re-hybridized to 5′-digoxigenin-labeled nucleotide, 5′-ACGGTATCTGATCGTCTTCGAACC-3′, complementary to human 18 S ribosomal RNA or/and to a digoxigenin-labeled glyceraldehyde-3-phosphate dehydrogenase cDNA probe. Blots were quantified using the Bio Image system (MilliGene). Nuclear Run-Off-Experiments were performed according to an established laboratory procedure(31Greenberg M.E. Bender T.P. Ausubel F.M. Short Protocols in Molecular Biology. Green Publishing Associates and John Wiley and Sons, New York1992: 4-25-4-29Google Scholar). Nuclei (107), isolated from three to five T75 flasks of confluent HUVEC, were incubated in the presence of 0.25 mCi of [32P]GTP and other unlabeled nucleotides (1 mM) in a volume of 200 μl at 26°C for 25 min. Transcribed RNAs were isolated using Ultraspec and equal amounts (6-8 × 106 cpm) were hybridized to denatured plasmids (15 μg) containing cDNAs of NOS-III and chloramphenicol acetyltransferase (CAT) (both in pcDNA-3) or hamster βtubulin cDNA in pAcUW51 (Pharmingen) immobilized on nitrocellulose membranes. Conditions for hybridization and washes were identical to those during Northern blotting. Densities of bands were quantified using the Bio Image system (MilliGene). To normalize the results, the intensities of β-tubulin bands were assigned an arbitrary densitometric unit of one. Intensities of other bands were expressed as a fraction of the density of β-tubulin bands. Following incubation with lyso-PC for the desired times, HUVEC were washed with ice-cold phosphate-buffered saline containing 0.5 mM EDTA and lysed in 300 μl of buffer containing 0.05 M Tris-HCl (pH 7.4), 1% Nonidet P-40, 60 mML-arginine HCl, 1 mM EDTA, 1 mM EGTA, 1 mM dithiothreitol, 100 μM tetrahydrobiopterin, and all components of Boehringer Mannheim Protease Inhibitors Set. Following centrifugation at 13,000 × g for 2 min, the protein concentrations in lysates were determined using Bio-Rad DC protein assay. Lysates (15 μg of protein) were analyzed by electrophoresis on 7% SDS-polyacrylamide gels and electroblotted onto nitrocellulose membranes. Membranes were probed using monoclonal antibodies from Transduction Laboratories (Lexington, Kentucky). Antibody N30020 is selective for NOS-III, and antibody N32020 binds to all isoforms of NOS. In some experiments, to confirm the specificities of antibodies, NOS-III standard (1 μg of lysate of stably transfected human kidney cell line 293 overexpressing bovine NOS-III) and NOS-II standard (4 μg of lysate of human aortic vascular smooth muscle cells stimulated with 100 pg/ml human interleukin-1β and 250 units/ml γ-interferon for 24 h) were also analyzed. Immunoreactive bands were visualized using ECL system (Amersham Corp.). NOS activity was assayed in the same lysates that were used for Western blotting. NOS activity was determined as L-NAME-inhibitable conversion of L-arginine to L-citrulline, as described previously(32Zembowicz A. Hatchett R.J. Radziszewski W. Gryglewski R.J. J. Pharmacol. Exp. Ther. 1993; 267: 1112-1118PubMed Google Scholar). Incubations were carried out with or without L-NAME (300 μM) and in the presence of 150-200 μg of protein, 106 cpm [2,3-3H]L-arginine HCl, 60 μM unlabeled L-arginine HCl, 10 mM NADPH, 1 mM dithiothreitol, 100 nM calmodulin, 1 mM Ca2+, and 100 μM tetrahydrobiopterin for 30 min. Human washed platelets were prepared as described previously(33Radomski M.W. Moncada S. Thromb. Res. 1983; 30: 383-389Abstract Full Text PDF PubMed Scopus (197) Google Scholar). Anti-platelet properties of HUVEC were assayed using an adaptation of an established method of bioassay of NO (34). HUVEC were harvested using trypsin/EDTA into ice-cold Tyrode buffer composed of 8 g/liter NaCl, 0.2 g/liter KCl, 0.225 g/liter MgCl2× 6H2O, 0.05 g/liter NaH2PO4, 1 g/liter NaHCO3, 1.5 g/liter glucose. Typically, equal numbers of control and lyso-PC-treated cells were isolated and 70-80% of those excluded trypan blue. Platelet aggregation was studied using a light transmission aggregometer (Chronolog) in the volume of 500 μl. Incubations contained 2.5 × 108 platelets/ml suspended in Tyrode buffer containing 1 mM Ca2+. HUVEC suspensions or the corresponding volumes of Tyrode buffer (3-20 μl) were added to stirred platelets 2 min before thrombin. L-NAME was added 1 min before the addition of HUVEC. Lyso-PC (30-100 μM) caused a time- and concentrationdependent increase of NOS-III mRNA levels in cultured human umbilical vein endothelial cells (HUVEC) (Fig. 1, A and B). NOS-III mRNA levels reached a maximum (11 ± 2-fold increase by densitometry, n = 7) 6 h after the stimulation with lyso-PC (100 μM). Phosphatidylcholine (100 μM) was without effect (n = 2, not shown). The long half-life of NOS-III mRNA in HUVEC (6Yoshizumi M. Perrella M.A. Burnett Jr., J.C. Lee M.E. Circ. Res. 1993; 73: 205-209Crossref PubMed Scopus (705) Google Scholar) suggested that lyso-PC may enhance transcription of the NOS-III gene rather then stabilize its mRNA. Indeed, actinomycin D (5 μM) abolished lyso-PC-induced increases of NOS-III mRNA levels (n = 2, not shown). Consistently, in three separate nuclear run-off experiments, treatment of HUVEC with lyso-PC (100 μM) for 3 h enhanced the rate of NOS-III mRNA synthesis by isolated nuclei (Fig. 2, A and B). This provides direct evidence for transcriptional activation of the NOS-III gene by lyso-PC. The induction by lyso-PC of NOS-III mRNA in HUVEC was not inhibited by cycloheximide, an inhibitor of protein synthesis (Fig. 2C). We investigated the possibility that the elevation of intracellular concentration of Ca2+ or activation of protein kinase C, known actions of lyso-PC in endothelial cells(35Inoue N. Hirata K. Yamada M. Hamamori Y. Matsuda Y. Akita H. Yokoyama M. Circ. Res. 1992; 71: 1410-1421Crossref PubMed Scopus (97) Google Scholar, 36Ohgushi M. Kugiyama K. Fukunaga K. Murohara T. Sugiyama S. Miyamoto E. Yasue H. Arterioscler. Thromb. 1993; 13: 1525-1532Crossref PubMed Scopus (73) Google Scholar), mediates the induction of NOS-III. However, 2-12-h stimulation of HUVEC with calcium ionophore A23187 (3 μM), phorbol myristate acetate (0.3 μM), or their combination did not increase NOS-III mRNA levels (n = 3, not shown).Figure 2:Induction of NOS-III gene transcription in HUVEC by lyso-PC. A, nuclear run-off analysis of transcription rates of NOS-III and β-tubulin (β-Tub) genes in control HUVEC and HUVEC stimulated with lyso-PC (100 μM) for 3 h. Background hybridization to the CAT-pcDNA-3 plasmid is also shown. B, Densitometric analysis of the autoradiograph shown in A. Intensities of β-tubulin (β-Tub) bands were assigned an arbitrary densitometric unit of 1. This graph shows a 2.0-fold increase of NOS-III band intensity in the blot hybridized to RNA synthesized by nuclei isolated from lyso-PC-treated HUVEC. In two other experiments performed similar analysis revealed 2.0- and 2.15-fold increases. Intensities of NOS-III and pcDNA-3 bands in control were identical. This indicates that the basal transcription of the NOS-III gene was below the level of detection of the assay. This is not unexpected, considering low levels of NOS-III mRNA and its stability in HUVEC. Therefore, densitometric analysis may underestimate the actual level of induction of NOS-III gene transcription by lyso-PC. C, Northern blot analysis of 20 μg of total RNA isolated from control HUVEC (lane 1) and HUVEC incubated for 6 h with cycloheximide (CHEX, 10 μM; lane 2), lyso-PC (100 μM; lane 3), or their combination (LysoPC&CHEX; lane 4).View Large Image Figure ViewerDownload Hi-res image Download (PPT) In agreement with changes of NOS-III mRNA levels, lyso-PC (100 μM) caused a time-dependent increase of NOS-III protein in HUVEC lysates (Fig. 3A). NOS-III levels reached a peak (5.5 ± 1.5 -fold increase by densitometry, n = 3) 12 h following stimulation with lyso-PC (100 μM). In the same lysates, lyso-PC induced a biphasic change in NOS activity (Fig. 3B). NOS activity decreased to 42 ± 6% of control (n = 4, p < 0.05) 3 h after stimulation with lyso-PC. At later time points NOS activity increased and reached a plateau after 12 h at the level 2-fold above that in control lysates (Fig. 3B). We found no evidence for NOS-II expression in HUVEC by immunoblotting (Fig. 3A). Our NOS-III antibody recognized only the NOS-III protein expressed in HUVEC and the NOS-III protein standard. The slight difference in electrophoretic mobility of NOS-III of HUVEC and that of NOS-III standard is due to species differences. The NOS-III standard contained bovine NOS-III. Consistently, NOS activity was abolished in the absence of Ca2+ (100% inhibition, n = 2), a property of NOS-III but not NOS-II. To determine whether induction of NOS-III by lyso-PC has the expected physiological consequences, we evaluated the effect of treatment with lyso-PC on the anti-platelet properties of HUVEC. We used a modification of the established method developed to bioassay the release of NO from endothelial cells. As predicted, treatment of HUVEC with lyso-PC (100 μM) for 24 h decreased the number of HUVEC required to inhibit thrombin-induced aggregation of human washed platelets (Fig. 4). The effect of lyso-PC was reversed by NOS inhibitor, L-NAME. The major finding of this paper is the identification of lyso-PC as a transcriptional inducer of constitutive endothelial NOS-III. Our results demonstrate that in cultured HUVEC, lyso-PC causes significant (11-fold) time- and concentration-dependent increase of NOS-III mRNA levels. Increases of NOS-III mRNA levels are accompanied by corresponding elevations (5.5-fold) of NOS-III protein levels. This results in a 2-fold increase of NOS-III activity in HUVEC lysates. At present, we do not know the basis of discrepancies between the levels and time courses of lyso-PC-induced changes of NOS-III mRNA and protein levels and those of NOS activity. These discrepancies suggest, however, that the regulation of NOS-III expression in HUVEC is complex and that it may also involve a lyso-PC-induced post-translational modification (e.g. phosphorylation) of NOS-III. Tetrahydrobiopterin has recently been demonstrated to have a structural role in NO synthases (37). Therefore, it is also possible that low levels of tetrahydrobiopterin present in cultured HUVEC (38Werner-Felmayer G. Werner E.R. Fuchs D. Hausen A. Reibnegger G. Schmidt K. Weiss G. Wachter H. J. Biol. Chem. 1993; 268: 1842-1846Abstract Full Text PDF PubMed Google Scholar, 39Rosenkranz-Weiss P. Sessa W.C. Milstien S. Kaufman S. Watson C.A. Pober J.S. J. Clin. Invest. 1994; 93: 2236-2243Crossref PubMed Scopus (344) Google Scholar) result in the inability of HUVEC to form active NOS-III in spite of increased synthesis of NOS-III peptide. Our nuclear run-off experiments and the inhibitory effect of actinomycin D demonstrate that the induction of NOS-III expression by lyso-PC is at least in part mediated by the transcriptional activation of the NOS-III gene. The major implication of these results is that the commonly held view of NOS-III as a "housekeeping gene" regulated mainly at the level of enzyme activity should be revised. Increased de novo synthesis of NOS-III via induction of gene expression may be an important mechanism that augments vasoprotective properties of endothelium in response to extracellular insults. To the best of our knowledge, lyso-PC is the first molecule to be shown to enhance the transcription of the NOS-III gene. Whether signals other then lyso-PC are also capable of transcriptional induction of NOS-III remains to be elucidated. However, recent studies showing the enhancement of NOS-III mRNA levels in cultured cells by shear stress (7) and in vivo by chronic exercise or estrogens(8Sessa W.C. Pritchard K. Seyedi N. Wang J. Hintze T.H. Circ. Res. 1994; 74: 349-353Crossref PubMed Scopus (827) Google Scholar, 9Weiner C.P. Lizasoain I. Baylis S.A. Knowles R.G. Charles I.G. Moncada S. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 5212-5216Crossref PubMed Scopus (1085) Google Scholar) suggest that the regulation of NOS-III at the transcriptional level may be a more common phenomenon. Like the induction by lyso-PC of heparin-binding epidermal growth factor mRNA in human monocytes(26Nakano T. Raines E.W. Abraham J.A. Klagsbrun M. Ross R. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 1069-1073Crossref PubMed Scopus (123) Google Scholar), induction of NOS-III mRNA levels was independent of new protein synthesis. This indicates that the activation of the NOS-III gene by lyso-PC involves the activation of a pre-existing transcriptional factor or factors. We investigated the possibility that the elevation of intracellular concentration of Ca2+ or activation of protein kinase C, known actions of lyso-PC in endothelial cells(35Inoue N. Hirata K. Yamada M. Hamamori Y. Matsuda Y. Akita H. Yokoyama M. Circ. Res. 1992; 71: 1410-1421Crossref PubMed Scopus (97) Google Scholar, 36Ohgushi M. Kugiyama K. Fukunaga K. Murohara T. Sugiyama S. Miyamoto E. Yasue H. Arterioscler. Thromb. 1993; 13: 1525-1532Crossref PubMed Scopus (73) Google Scholar), mediates the induction of NOS-III. However, calcium ionophore A23187, phorbol myristate acetate, or their combination did not increase NOS-III mRNA levels. Lyso-PC has been shown to modify G-proteins (40Flavahan N.A. Am. J. Physiol. 1993; 264: H722-H727PubMed Google Scholar) and activate adenylate cyclase(41Resnick R.J. Tomaska L. J. Biol. Chem. 1994; 269: 32336-32341Abstract Full Text PDF PubMed Google Scholar). However, we did not explore a possibility that any of these effects is involved in transduction pathway of the NOS-III gene activation by lyso-PC. Thus, as in the case of other lyso-PC-activated genes, the signal transduction pathway and transcriptional factors involved in activation of the NOS-III gene by lyso-PC remain to be established. It is also important to note that lyso-PC is the major product of action of cellular phospholipases A2 on phosphatidylcholine, the most abundant phospholipid(42Dennis E.A. J. Biol. Chem. 1994; 269: 13057-13060Abstract Full Text PDF PubMed Google Scholar). In contrast to arachidonic acid, the other product of action of phospholipases A2, which is a substrate for enzymes involved in the synthesis of prostanoids and leukotrienes, a role of lyso-PC in signal transduction has only recently been suggested(43Nishizuka Y. Science. 1992; 258: 607-614Crossref PubMed Scopus (4232) Google Scholar). Therefore, a putative role of intracellularly generated lyso-PC in the regulation of NOS-III or other lyso-PC-inducible genes in response to extracellular signals might be an interesting area of investigation. We have also conducted experiments designed to determine whether the treatment of HUVEC with lyso-PC potentiates the ability of HUVEC to inhibit platelet aggregation, a key physiological function of endothelial cells. As predicted, the treatment of HUVEC with lyso-PC reduced the number of HUVEC required to inhibit aggregation of human washed platelets. The effect of lyso-PC was reversed by an inhibitor of NO synthase. These results demonstrate that lyso-PC augments the anti-platelet properties of HUVEC and, considered together with the evidence for the simultaneous induction of NOS-III expression, suggest the induction of NOS-III as a mechanism involved. However, whether this is the only mechanism remains to be established. Identification of lyso-PC as an inducer of the NOS-III gene raises several intriguing questions regarding the biological significance of lyso-PC-triggered induction of NOS-III. It is well established that both in experimental models of atherosclerosis (44Verbeuren T.J. Jordaens F.H. Zonnekeyn L.L. Van Hove C.E. Coene M.C. Herman A.G. Circ. Res. 1986; 58: 552-564Crossref PubMed Scopus (431) Google Scholar, 45Jayakody L. Senaratne M. Thomson A. Kappagoda T. Circ. Res. 1987; 60: 251-264Crossref PubMed Scopus (160) Google Scholar, 46Harrison D.G. Freiman P.C. Armstrong M.L. Marcus M.L. Heistad D.D. Circ. Res. 1987; 61: 74-80Google Scholar, 47Bossaller C. Habib G.B. Yamamoto H. Williams C. Wells S. Henry P.D. J. Clin. Invest. 1987; 79: 170-174Crossref PubMed Scopus (432) Google Scholar) and in human disease (48Ludmer P.L. Selwyn A.P. Shook T.L. Wayne R.R. Mudge G.H. Alexander R.W. Ganz P. N. Engl. J. Med. 1986; 315: 1046-1051Crossref PubMed Scopus (2032) Google Scholar, 49Forstermann U. Mugge A. Alheid U. Haverich A. Frolich J.C. Circ. Res. 1988; 62: 185-190Crossref PubMed Scopus (373) Google Scholar, 50Werns S.W. Walton J.A. Hsia H.H. Nabel E.G. Sanz M.L. Pitt B. Circulation. 1989; 79: 287-291Crossref PubMed Scopus (292) Google Scholar, 51Creager M.A. Cooke J.P. Mendelsohn M.E. Gallagher S.J. Coleman S.M. Loscalzo J. Dzau V.J. J. Clin. Invest. 1990; 86: 228-234Crossref PubMed Scopus (950) Google Scholar) atherosclerosis leads to the impairment of endothelium-dependent relaxation, which in most vascular beds is mediated by endothelium-derived NO. It has also been convincingly demonstrated by several laboratories that in vitro incubation of isolated blood vessels with oxidized LDL leads to the impairment of endothelium-dependent relaxation similar to that observed in atherosclerotic vessels(52Jacobs M. Plane F. Bruckdorfer K.R. Br. J. Pharmacol. 1990; 100: 21-26Crossref PubMed Scopus (163) Google Scholar, 53Tanner F.C. Noll G. Boulanger C.M. Luscher T.F. Circulation. 1991; 83: 2012-2020Crossref PubMed Scopus (304) Google Scholar, 54Kugiyama K. Kerns S.A. Morrisett J.D. Roberts R. Henry P.D. Nature. 1990; 344: 160-162Crossref PubMed Scopus (780) Google Scholar, 55Yokoyama M. Hirata K. Miyake R. Akita H. Ishikawa Y. Fukuzaki H. Biochem. Biophys. Res. Commun. 1990; 168: 301-308Crossref PubMed Scopus (170) Google Scholar). Lyso-PC has been identified as the critical diffusible component of oxidized LDL that is responsible for a rapid inhibition of endothelium-dependent relaxation by oxidized LDL(54Kugiyama K. Kerns S.A. Morrisett J.D. Roberts R. Henry P.D. Nature. 1990; 344: 160-162Crossref PubMed Scopus (780) Google Scholar, 55Yokoyama M. Hirata K. Miyake R. Akita H. Ishikawa Y. Fukuzaki H. Biochem. Biophys. Res. Commun. 1990; 168: 301-308Crossref PubMed Scopus (170) Google Scholar). It has also been recently reported that chemically oxidized LDLs reduce NOS-III mRNA levels in human endothelial cell via destabilization of NOS-III mRNA(56Liao J.K. Shin W.S. Lee W.Y. Clark S.L. J. Biol. Chem. 1995; 270: 319-324Abstract Full Text Full Text PDF PubMed Scopus (534) Google Scholar). Thus, in the context of the studies summarized above, augmentation of endothelial expression and activity of NOS-III by lyso-PC comes as a surprise and may at first appear to be contradictory. However, a careful analysis demonstrates that these results are not necessary conflicting. It should be emphasized that the reduction of endothelium-dependent relaxation in isolated vessel can be due to factors other then the reduction of endothelial capacity to synthesize NO. For instance, the impairment of endothelium-dependent relaxation could be caused by the reduction of the half-life of NO in atherosclerotic vessels. Indeed, oxidized LDLs have been demonstrated to reduce the half-life of NO in vitro(57Galle J. Mulsch A. Busse R. Bassenge E. Arterioscler. Thromb. 1991; 11: 198-203Crossref PubMed Scopus (189) Google Scholar, 58Chin J.H. Azhar S. Hoffman B.B. J. Clin. Invest. 1992; 89: 10-18Crossref PubMed Scopus (406) Google Scholar), and both hypercholesterolemia in experimental animals (59Ohara Y. Peterson T.E. Harrison D.G. J. Clin. Invest. 1993; 91: 2546-2551Crossref PubMed Scopus (1656) Google Scholar) and lyso-PC in isolated blood vessels (60Ohara Y. Peterson T.E. Zheng B. Kuo J.F. Harrison D.G. Arterioscler. Thromb. 1994; 14: 1007-1013Crossref PubMed Scopus (127) Google Scholar) reduce the half-life of NO via induction of superoxide anion generation. Importantly, intimal thickening, which is a typical characteristic of atherosclerotic vessels, can be expected to decrease the amount of endothelium-derived NO that reaches vascular smooth muscle. Thus, it is possible that the impairment of endothelium-dependent relaxation and the augmentation of endothelial synthesis of NO co-exist. In this context, it is of interest that the release of NO measured by a chemiluminescent technique is increased in aortas from hypercholesterolemic rabbits (61). Recently, enhancement of the release of NO from cultured rabbit endothelial cells by oxidized LDL has been reported(62Fries D.M. Penha R.G. D'Amico E.A. Abdalla D.S. Monteiro H.P. Biochem. Biophys. Res. Commun. 1995; 207: 231-237Crossref PubMed Scopus (21) Google Scholar). Moreover, intensive staining for NOS-III has been observed in atherosclerotic vessels(5Forstermann U. Closs E.I. Pollock J.S. Nakane M. Schwarz P. Gath I. Kleinert H. Hypertension. 1994; 23: 1121-1131Crossref PubMed Scopus (1003) Google Scholar). Interestingly, the same paper that demonstrated that chemically oxidized LDLs decrease NOS-III mRNA levels via reduction of NOS mRNA half-life (56Liao J.K. Shin W.S. Lee W.Y. Clark S.L. J. Biol. Chem. 1995; 270: 319-324Abstract Full Text Full Text PDF PubMed Scopus (534) Google Scholar) also showed data demonstrating that oxidized LDLs double the NOS-III transcription rate. In the same paper, although not discussed in detail by the authors, results were presented that showed that the incubation of native LDL with endothelial cells, which results in oxidative modification of LDL, evidenced by increased content of thiobarbituric acid reactive substances, results in the enhancement of NOS-III mRNA levels. Thus, it is tempting to speculate that the increase of NOS-III mRNA levels by endothelial cell-oxidized LDL and the augmentation of the NOS-III transcription rate by chemically oxidized LDL may be due to lyso-PC-induced activation of the NOS-III gene. It is possible that chemically modified LDL has additional effects on endothelial cells that cause destabilization of NOS-III mRNA. As outlined in detail in the Introduction, lyso-PC causes induction of several potentially pro-atherogenic endothelial genes involved in leukocyte recruitment, mitogenesis, and inflammation(27Kume N. Cybulsky M.I. Gimbrone Jr., M.A. J. Clin. Invest. 1992; 90: 1138-1144Crossref PubMed Scopus (724) Google Scholar, 28Kume N. Gimbrone Jr., M.A. J. Clin. Invest. 1994; 93: 907-911Crossref PubMed Scopus (299) Google Scholar). Our findings add another inducible protein to this growing list. However, this may be an important addition, as in contrast to other genes showed to be induced by lyso-PC so far, the biological actions of endothelium derived NO suggest that NOS-III is a key vasoprotective gene. These findings are intriguing because they introduce a novel idea that a single molecule can induce both pro-atherogenic and vasoprotective mechanisms. The attractiveness of this hypothesis stems from the fact that it may help to rationalize several poorly understood issues concerning atherogenesis. Induction of both pro-atherogenic and vasoprotective mechanisms may explain a long time course of atherosclerotic lesions development. This may also explain the potential of early atherosclerotic lesions to regress after removing of initiating environmental factors. We acknowledge, however, that further work, beyond the limitations of the present study, is required to verify the significance of lyso-PC-induced activation of the NOS-III gene in the proper pathophysiological setting. In conclusion, our results demonstrate that lyso-PC enhances expression of NOS-III in HUVEC by a transcriptional mechanism. Induction of NOS-III by lyso-PC may be a protective response that limits the progress of the atherosclerotic lesion and promotes its regression. We thank D. Loose-Mitchell for review of the manuscript, P.-F. Chen for cloning NOS-III, Sang Lee and J. Juneja for supplying HUVEC, W. C. Sessa for NOS-III transfected cell line, T. Scott-Burden for human aortic vascular smooth muscle cells, and M. Kruzel for β-tubulin plasmid.