Elevated Levels of cAMP Inhibit Protein Kinase C– Independent Mechanisms of Endothelial Platelet-Derived Growth Factor–B Chain and Intercellular Adhesion Molecule-1 Gene Induction by Lysophosphatidylcholine
Abstract Lysophosphatidylcholine (lyso-PC), a polar phospholipid product increased in atherogenic lipoproteins and atherosclerotic lesions, has been shown to differentially induce functional intercellular adhesion molecule-1 (ICAM-1) and vascular cell adhesion molecule-1 and mRNA for platelet-derived growth factor (PDGF)–A and –B chains and heparin-binding epidermal growth factor–like growth factor in various cultured endothelial cells. In this study, we have demonstrated increased expression of cell- and matrix-associated forms of PDGF–B chain (PDGF-B) protein elicited by lyso-PC and further characterized potential signal transduction mechanisms responsible for lyso-PC–induced gene expression, focusing on PDGF-B and ICAM-1 genes in cultured human umbilical vein endothelial cell models. Cycloheximide almost completely inhibited PDGF-B but not ICAM-1 mRNA induction elicited by lyso-PC, suggesting that dependence on de novo protein synthesis for PDGF-B is different from that for ICAM-1. Prolonged exposure to phorbol myristate acetate (PMA), which depletes protein kinase C (PKC), or staurosporine, a PKC inhibitor, did not block lyso-PC–induced increases in PDGF-B or ICAM-1 mRNA. Forskolin and dibutyryl cAMP, which elevate intracellular cAMP levels, blocked both PDGF-B and ICAM-1 upregulation elicited by lyso-PC; however, these cAMP-elevating agents did not suppress ICAM-1 upregulation by PMA. Taken together, PDGF-B and ICAM-1 gene induction by lyso-PC may involve different signaling mechanisms; however, both appear to be independent of PMA-regulatable PKC activation but are suppressed by increased levels of intracellular cAMP.
- protein kinase C
- platelet-derived growth factor–B chain
- intercellular adhesion molecule-1
Vascular endothelium plays a pivotal role in the pathogenesis of vascular diseases, including atherosclerosis and inflammation, expressing endothelial-leukocyte adhesion molecules, growth factors, and cytokines in response to various pathophysiological stimuli.1 2 3 Lyso-PC, which is a prominent phospholipid component of atherogenic lipoproteins4 and is also generated in inflammatory lesions by extracellular phospholipase A2 activities,5 has been shown to differentially upregulate VCAM-1 and ICAM-1 expression in various cultured endothelial cells, showing its potential role in mononuclear leukocyte recruitment into these lesions.6 Lyso-PC has been demonstrated to induce gene expression of potent smooth muscle growth factors such as PDGF–A and –B chains and HB-EGF in cultured human endothelial cells, thus suggesting its potential role in smooth muscle proliferation.7 Lyso-PC has been shown to act on human monocytes as well to induce HB-EGF gene expression.8 These effects of lyso-PC on endothelial genes appear to be specific and distinct from those elicited by cytokines, such as IL-1 and TNF-α, since neither E-selectin nor IL-8 was upregulated in the same cells stimulated with lyso-PC. Nuclear runoff assays and the evaluation of mRNA half-lives have indicated that lyso-PC stimulates transcription of these genes7 ; however, signal transduction mechanisms responsible for the gene induction of adhesion molecules and growth factors by this lyso-PC stimulus have not been fully elucidated. Furthermore, the effects of lyso-PC on growth factor protein expression have not been determined in cultured endothelial cells.
In the present study, therefore, we have measured PDGF-B protein levels in lyso-PC–treated cultured HUVECs and explored the potential signal transduction mechanisms involved in both PDGF-B and ICAM-1 upregulation elicited by lyso-PC, focusing on roles of PKC and cAMP.
Materials and Methods
Cultured HUVECs were isolated by collagenase digestion and grown in medium 199 with 20% (vol/vol) heat-inactivated FBS (Irvine Scientific), supplemented with 100 U/mL penicillin, 100 mg/mL streptomycin, 25 U/mL heparin (porcine intestinal mucosa, Sigma Chemical Co), and 30 mg/mL ECGS (Sigma) as previously described.9 All experiments were carried out within 3 days after HUVECs reached confluence. Cells (passage numbers between 2 and 4) were incubated with or without test stimuli in medium 199 supplemented with 5% (vol/vol) FBS, without heparin or ECGS.
Lyso-PC (palmitoyl, C16:0) was purchased from Avanti Polar Lipids. Cycloheximide, PMA, forskolin, and Bt2cAMP were obtained from Sigma, and staurosporine was from Calbiochem. All other reagents were of reagent grade.
Northern Blot Analysis
Total cellular RNA, isolated from HUVECs by the acid guanidinium–phenol–chloroform method,10 was electrophoresed through 1% agarose gels containing formaldehyde, transferred onto 0.45-μm nylon membranes (Zeta-Probe, Bio-Rad), and fixed by UV cross-linking. Northern membranes were hybridized with human ICAM-1 and PDGF-B cDNA probes, which were labeled with [α-32P]dCTP (DuPont NEN) by using random hexanucleotide primers (Pharmacia) at 65°C for 18 hours in a mixture containing 1 mmol/L EDTA, 0.25 mol/L Na2HPO4 (pH 7.2), and 7% SDS. The filters were subsequently washed at 65°C twice with 1 mmol/L EDTA, 40 mmol/L Na2HPO4 (pH 7.2), and 5% SDS and once with 1 mmol/L EDTA, 40 mmol/L Na2HPO4 (pH 7.2), and 1% SDS and exposed to x-ray films. A 1.3-kb Xho I fragment of human ICAM-1, kindly provided by Dr Brian Seed (Massachusetts General Hospital, Boston), was used to detect ICAM-1 mRNA.6 7 11 A 2.1-kb BamHI fragment of human PDGF-B cDNA,12 obtained from American Type Culture Collection, and a Pst I–EcoRI fragment of human PDGF-B,13 kindly provided by Dr Tucker Collins (Brigham and Women’s Hospital, Boston, Mass) were used as hybridization probes. Northern analyses using these PDGF-B cDNAs gave similar results. Some blots were rehybridized with radiolabeled human β-actin cDNA to control the amounts of RNA loaded. Densitometric scanning was performed to quantify the amounts of mRNA; an Image Master laser densitometer (Pharmacia) was used. Relative amounts of mRNA for PDGF-B and ICAM-1 were normalized to β-actin mRNA levels.
Western Blot Analysis
Laemmli sample buffer (2% SDS, 10% glycerol, 60 mmol/L Tris [pH 6.8], and 0.001% bromophenol blue) was directly poured into HUVEC culture plates, and the cell lysates were passed through 25-gauge needles 10 times. After heating at 98°C for 10 minutes, samples were subjected to SDS–polyacrylamide (10% to ≈20% gradient) gel electrophoresis in nonreducing conditions and transferred onto nitrocellulose filters (Hybond ECL filters, Amersham Corp) by electroblotting. After preincubation with TBS (50 mmol/L Tris-Cl [pH 8.0], 2 mmol/L CaCl2, 100 mmol/L NaCl, and 5% [wt/vol] nonfat dry milk) for 3 hours at room temperature, filters were incubated with a rabbit polyclonal antibody directed to PDGF-BB homodimer (Genzyme Corp) diluted in TBS at room temperature for 2 hours, followed by washing twice with TBS without nonfat dry milk. Filters were then incubated with the horseradish peroxidase–conjugated anti-rabbit IgG antibody (Amersham) diluted in TBS for 2 hours at room temperature, washed twice in TBS without nonfat dry milk, and visualized by use of a chemiluminescence reagent (ECL kit, Amersham Corp).
Statistical analyses were carried out by paired Student’s t test. Percent changes in mRNA levels compared with respective controls are expressed as mean±SD.
Lyso-PC Induces Cell- and Matrix-Associated Forms of PDGF-B Protein
Lyso-PC has been shown to increase the amounts of mRNA for PDGF-B.7 To examine the effects of lyso-PC on PDGF-B protein levels, we performed Western blot analyses using cell lysate from lyso-PC–treated HUVECs. As shown in Fig 1⇓, multiple bands with approximate molecular masses between 38 and 43 kD, which appear to be compatible with previous reports,14 15 16 were detected by treatments with lyso-PC for 8 hours and continuously increased for at least 20 hours. Similar bands of 38 to 45 kD were detectable in HUVECs treated with PMA for 8 hours. These bands disappeared when excess amounts of recombinant human PDGF-BB homodimer were included in the incubation buffer containing anti-human PDGF-B antibody (data not shown). Two additional bands with approximate molecular masses of 85 and 55 kD, respectively, appeared to be nonspecific, since these bands were similarly detectable when anti-human PDGF-B was replaced at a concentration equivalent to that of nonimmune rabbit serum (data not shown).
PDGF-B mRNA Upregulation by Lyso-PC Depends on De Novo Protein Synthesis, but ICAM-1 Does Not
In previous studies, lyso-PC has been shown to upregulate mRNA for both ICAM-1 and PDGF-B as early as 4 hours of treatment. To determine whether de novo protein synthesis is required in this process, we treated HUVECs with or without lyso-PC (or PMA as a control) in the presence or absence of cycloheximide for 4 hours, and Northern blot analyses were performed. Cycloheximide completely blocked lyso-PC–induced, as well as PMA-induced, increases in PDGF mRNA levels, although cycloheximide alone modestly increased the amount of PDGF-B mRNA (Fig 2A⇓). In contrast, cycloheximide dramatically increased ICAM-1 mRNA levels in both untreated and lyso-PC–treated or PMA-treated cells (Fig 2B⇓). These results clearly indicate that dependence of de novo protein synthesis in lyso-PC–induced, as well as PMA-induced, gene upregulation is different between PDGF-B and ICAM-1.
Induction of mRNA for Both PDGF-B and ICAM-1 Is Independent of PMA-Regulatable PKC Activation
To investigate the potential role of PKC in PDGF-B and ICAM-1 mRNA induction by lyso-PC, we followed two strategies: depletion of PKC activities by prolonged exposure to PMA and inhibition of PKC by a pharmacological reagent. As shown in Fig 3⇓, pretreatment with PMA for 24 hours significantly blocked PDGF-B mRNA induction elicited by the subsequent stimulation with PMA (63±18% reduction); however, this PMA pretreatment did not inhibit lyso-PC–induced increases in PDGF-B mRNA. Effects of prolonged exposure to PMA on ICAM-1 mRNA upregulation is similar to those observed in PDGF-B; pretreatment with PMA did not prohibit lyso-PC–induced increases in ICAM-1 mRNA levels but completely blocked the effect of the subsequent stimulation with PMA (>99% reduction). We also examined the effects of staurosporine, an inhibitor of PKC. Staurosporine (10 nmol/L) significantly reduced PMA-induced increases in PDGF-B mRNA levels (44±12% reduction); however, this agent did not inhibit but rather increased those elicited by lyso-PC (Fig 4⇓). These results indicate that signal transduction mechanisms responsible for PDGF-B and ICAM-1 mRNA upregulation elicited by lyso-PC in HUVECs appear to be dissociated from PMA-regulatable PKC activation. We further tested the effects of higher concentrations of staurosporine; however, >50 nmol/L concentrations of this reagent with 60 μmol/L lyso-PC caused morphologically apparent cytotoxicity.
Elevated Levels of cAMP Inhibit PDGF-B and ICAM-1 mRNA Induction Elicited by Lyso-PC
To explore the potential role of cAMP in lyso-PC–induced PDGF-B and ICAM-1 gene expression, we examined the effects of forskolin and Bt2cAMP, both of which can increase intracellular cAMP levels. We treated HUVECs with or without lyso-PC or PMA in the presence or absence of forskolin and measured mRNA levels for PDGF-B and ICAM-1 by Northern analyses. As shown in Fig 5⇓, forskolin significantly reduced lyso-PC–induced, as well as PMA-induced, increases in PDGF-B mRNA (59±19% and 68±19% reduction, respectively). In contrast, this agent inhibited only lyso-PC–induced increases in ICAM-1 mRNA (67±22% reduction) and did not significantly block those induced by PMA. Bt2cAMP exhibited effects similar to those observed with forskolin. Upregulated expression of PDGF-B mRNA elicited by lyso-PC, as well as PMA, was significantly prohibited by Bt2cAMP (54% and 30% reduction, respectively); however, this agent blocked only lyso-PC–induced ICAM-1 mRNA upregulation (46% reduction) but did not show any significant inhibitory effect on PMA-induced ICAM-1 mRNA expression (Fig 6⇓). To examine whether elevated levels of cAMP decrease PDGF-B protein levels, we performed Western blot analyses. Treatment with optimal doses of lyso-PC or PMA for 20 hours significantly increased the levels of cell- and matrix-associated forms of PDGF-BB, which were effectively suppressed by forskolin (Fig 7⇓).
We also have examined the effects of combinations of reagents used in the present study on cell viability. We did not detect any evidence of cytotoxicity by either microscopic observation or measuring lactate dehydrogenase released into culture media (data not shown).
Endothelial expression of smooth muscle growth factors and endothelial-leukocyte adhesion molecules appears to play an important role in atherogenesis and inflammatory disease processes.1 2 3 Lyso-PC, a phospholipid stimulus relevant to atherosclerosis and inflammation, has been shown to selectively upregulate gene expression of adhesion molecules and growth factors, including PDGF-B and ICAM-1. In the present study, we have shown, for the first time, that cell- and matrix-associated forms of PDGF-B protein are increased by lyso-PC treatment in HUVECs. We have also explored the potential signal transduction mechanisms responsible for the PDGF-B and ICAM-1 upregulation by lyso-PC in HUVEC models.
Inhibition of nascent protein synthesis by cycloheximide provided contrasting results between PDGF-B and ICAM-1 gene regulation. PDGF-B mRNA induction by lyso-PC, as well as that elicited by PMA, requires de novo protein synthesis, but ICAM-1 does not. In fact, cycloheximide enhanced the accumulation of ICAM-1 mRNA in both lyso-PC–treated and untreated HUVECs; this finding appears to be similar to the previous finding with PMA-treated and TNF-α–treated HUVECs.17 These results clearly indicate that signal transduction mechanisms responsible for lyso-PC–induced gene expression might be different between PDGF-B and ICAM-1.
PKC activation has been reported to be involved in lyso-PC–induced inhibition of endothelium-dependent vasorelaxation ex vivo, and transient activation of PKC has been detected in lyso-PC–stimulated endothelial cells.18 Furthermore, a recent report has shown that lyso-PC treatment resulted in increased expression of ICAM-1 protein in isolated porcine coronary arteries, which was efficiently blocked by PKC inhibitors.19 Therefore, we have explored the potential role of PKC in lyso-PC–induced endothelial gene upregulation of PDGF-B and ICAM-1 in our HUVEC models, in which selective patterns of gene regulation by lyso-PC have been well characterized.7 Data presented here have clearly demonstrated that upregulated expression of PDGF-B and ICAM-1 by lyso-PC in our HUVEC models appears to depend on mechanisms other than PMA-regulatable PKC activation. Previous studies by others, using PMA and staurosporine, have indicated that ICAM-1 upregulation by TNF-α also is dissociated from PKC activation,20 21 thus supporting the existence of PKC-independent mechanisms of ICAM-1 gene induction.
Transient increases in cytosolic calcium levels and turnover of phosphoinositides have been detected in lyso-PC–treated cultured endothelial cells and are reported to be involved in the inhibitory actions of lyso-PC on endothelium-dependent vasorelaxation.22 Our preliminary experiments, however, have revealed that the calcium ionophore ionomycin did not induce PDGF-B or ICAM-1 mRNA (data not shown), suggesting that calcium mobilization alone is not sufficient to induce endothelial genes such as PDGF-B and ICAM-1.
The cAMP–protein kinase A pathway is another possible signal transduction cascade that can transmit biological signals. Forskolin or Bt2cAMP, which elevates intracellular cAMP levels, alone did not upregulate PDGF-B or ICAM-1 expression but rather suppressed lyso-PC–induced increases in both PDGF-B and ICAM-1 mRNA levels. These effects of cAMP-elevating agents do not appear to result from general inhibitory action, since forskolin or Bt2cAMP inhibited lyso-PC–induced ICAM-1 mRNA upregulation but did not affect that induced by PMA. Although previous studies by others have demonstrated that elevated levels of intracellular cAMP can also antagonize the effects of thrombin or TGF-β on PDGF-B mRNA levels23 but do not suppress ICAM-1 expression induced by TNF-α,24 our results appear to be the first to demonstrate that elevated levels of cAMP can counteract the effect of lyso-PC on both PDGF-B and ICAM-1 expression.
Previous studies with Northern analyses using actinomycin D and nuclear runoff assays have revealed that lyso-PC does not appear to act on PDGF-B and ICAM-1 genes by stabilizing mRNA but rather by stimulating gene transcription.7 Consensus sequences for binding of known transcription factors, including activating protein 1 (AP-1), have been detected in 5′ promoter regions of PDGF-B25 26 27 and the ICAM-128 29 gene. Studies are in progress in our laboratory to identify which promoter elements and transcription factors are involved in lyso-PC–induced endothelial gene upregulation.
In summary, we have demonstrated that PDGF-B protein expression is upregulated by lyso-PC and have partially characterized potential signal transduction mechanisms responsible for endothelial gene induction elicited by lyso-PC; PDGF-B and ICAM-1 gene induction by lyso-PC appears to depend on signal transduction mechanisms other than PMA-regulatable PKC, which can be suppressed by elevated levels of cAMP. Dependence of de novo protein synthesis for PDGF-B is different from that for ICAM-1, suggesting that lyso-PC might stimulate multiple and diverse signaling pathways. Further studies related to signal transduction pathways and transcriptional regulatory mechanisms involved in this lipid stimulus relevant to atherogenesis and inflammation may provide new insights into endothelial activation in these pathophysiological settings and might provide potential therapeutic targets in preventing vascular diseases.
Selected Abbreviations and Acronyms
|ECGS||=||endothelial cell growth supplement|
|FBS||=||fetal bovine serum|
|HB-EGF||=||heparin-binding epidermal growth factor–like growth factor|
|HUVEC||=||human umbilical vein endothelial cell|
|ICAM-1||=||intercellular adhesion molecule-1|
|PDGF||=||platelet-derived growth factor|
|PKC||=||protein kinase C|
|PMA||=||phorbol 12-myristate 13-acetate|
|TNF-α||=||tumor necrosis factor-α|
|VCAM-1||=||vascular cell adhesion molecule-1|
This study was supported by a research grant from the Ministry of Education, Science, and Culture of Japan (No. 06671022) and a grant-in-aid from the Japan Research Foundation for Clinical Pharmacology, Tokyo, Japan. We thank the doctors and nursing staff in the obstetrics ward of Kyoto University Hospital for their help in obtaining human umbilical cords. We also gratefully acknowledge Dr Tucker Collins (Brigham and Women’s Hospital, Boston, Mass) for his generous gift of PDGF-B cDNA and helpful discussions.
- Received October 24, 1994.
- Accepted April 19, 1995.
- © 1995 American Heart Association, Inc.
Gimbrone MA Jr, Kume N, Cybulsky MI. Vascular endothelial dysfunction and the pathogenesis of atherosclerosis. In: Weber PC, Leaf A, eds. Atherosclerosis Reviews. New York, NY: Raven Press Publishers; 1993;25:1-9.
Witztum JL, Steinberg D. Role of oxidized low density lipoprotein in atherogenesis. J Clin Invest. 1991;88:1785-1792.
Kume N, Cybulsky MI, Gimbrone MA Jr. Lysophosphatidylcholine, a component of atherogenic lipoproteins, induces mononuclear adhesion molecules in cultured human and rabbit arterial endothelial cells. J Clin Invest. 1992;90:1138-1144.
Kume N, Gimbrone MA Jr. Lysophosphatidylcholine transcriptionally induces growth factor gene expression in cultured human endothelial cells. J Clin Invest. 1994;93:907-911.
Nakano T, Raines EW, Abraham JA, Klagsbrun M, Ross R. Lysophosphatidylcholine upregulates the level of heparin-binding epidermal growth factor-like growth factor mRNA in human monocytes. Proc Natl Acad Sci U S A. 1994;91:1069-1073.
Raines EW, Ross R. Compartmentalization of PDGF on extracellular binding sites dependent on exon-6-encoded sequences. J Cell Biol. 1992;116:533-543.
Kelly JL, Sanchez A, Brown GS, Chesterman CN, Sleigh MJ. Accumulation of PDGF B and cell-binding forms of PDGF A in the extracellular matrix. J Cell Biol. 1993;121:1153-1163.
Soyombo AA, DiCorleto PE. Stable expression of human platelet-derived growth factor B chain by bovine aortic endothelial cells. J Biol Chem. 1994;269:17734-17740.
Wertheimer SJ, Myers CL, Wallace RW, Parks TP. Intercellular adhesion molecule-1 gene expression in human endothelial cells: differential regulation by tumor necrosis factor-α and phorbol myristate acetate. J Biol Chem. 1992;267:12030-12035.
Kugiyama K, Ohgushi M, Sugiyama S, Murohara T, Fukunaga K, Miyamoto E, Yasue H. Lysophosphatidylcholine inhibits surface receptor–mediated intracellular signals in endothelial cells by a pathway involving protein kinase C activation. Circ Res. 1992;71:1422-1428.
Sugiyama S, Kugiyama K, Ohgushi M, Fujimoto K, Yasue H. Lysophosphatidylcholine in oxidized low-density lipoprotein increases endothelial susceptibility to polymorphonuclear leukocyte–induced endothelial dysfunction in porcine coronary arteries. Circ Res. 1994;74:565-575.
Ritchie AJ, Johnson DR, Ewenstein BM, Pober JS. Tumor necrosis factor induction of surface antigen is independent of protein kinase C activation or inactivation: studies with phorbol myristate acetate and staurosporine. J Immunol. 1991;146:3056-3062.
Myers CL, Desai SN, Schembri-King J, Letts GL, Wallace RW. Discriminatory effects of protein kinase inhibitors and calcium ionophore on endothelial ICAM-1 induction. Am J Physiol. 1992;262:C365-C373.
Inoue N, Hirata K, Yamada M, Hamamori Y, Matsuda Y, Akita H, Yokoyama M. Lysophosphatidylcholine inhibits bradykinin-induced phosphoinositide hydrolysis and calcium transients in cultured bovine aortic endothelial cells. Circ Res. 1992;71:1410-1421.
Daniel TO, Gibbs VC, Milfay DF, Williams LT. Agents that increase cAMP accumulation block endothelial c-sis induction by thrombin and transforming growth factor-β. J Biol Chem. 1987;262:11893-11896.
Pober JS, Slowik MR, De Luca LG, Ritchie AJ. Elevated levels of cyclic AMP inhibits endothelial cell synthesis and expression of TNF-induced endothelial-leukocyte adhesion molecule-1, and vascular cell adhesion molecule-1, but not intercellular adhesion molecule-1. J Immunol. 1993;150:5114-5123.
Ratner L, Thielan B, Collins T. Sequences of 5′ portion of the human c-sis gene: characterization of the transcriptional promoter and regulation of expression of the protein product by 5′ untranslated mRNA sequence. Nucleic Acids Res. 1987;15:6017-6036.
Pech M, Rao CD, Robbins KC, Aaronson SA. Functional identification of regulatory elements within the promoter region of platelet-derived growth factor 2. Mol Cell Biol. 1989;9:396-405.
Khachigian LM, Fries JWU, Benz MW, Bonthron DT, Collins T. Novel cis-acting elements in the human platelet-derived growth factor B-chain core promoter that mediate gene expression in cultured vascular endothelial cells. J Biol Chem. 1994;269:22647-22656.
Degitz K, Lian-Jie L, Caughman SW. Cloning and characterization of the 5′-transcriptional regulatory region of the human intercellular adhesion molecule 1 gene. J Biol Chem. 1991;266:14024-14030.
Voraberger G, Schafer R, Stratowa C. Cloning of the human gene for intercellular adhesion molecule 1 and analysis of its 5′-regulatory region: induction by cytokines and phorbol ester. J Immunol. 1991;147:2777-2786.