Yan Huang, Andrew J. Fleming, Shan Wu, Gabriel Virella and Maria ...

7
Yan Huang, Andrew J. Fleming, Shan Wu, Gabriel Virella and Maria F. Lopes-Virella in U937 Cells via Mitogen-Activated Protein Kinase Receptor Cross-Linking by Immune Complexes Induces Matrix Metalloproteinase-1 γ Fc- Print ISSN: 1079-5642. Online ISSN: 1524-4636 Copyright © 2000 American Heart Association, Inc. All rights reserved. Greenville Avenue, Dallas, TX 75231 is published by the American Heart Association, 7272 Arteriosclerosis, Thrombosis, and Vascular Biology doi: 10.1161/01.ATV.20.12.2533 2000;20:2533-2538 Arterioscler Thromb Vasc Biol. http://atvb.ahajournals.org/content/20/12/2533 World Wide Web at: The online version of this article, along with updated information and services, is located on the http://atvb.ahajournals.org/content/suppl/2000/11/29/20.12.2533.DC1.html Data Supplement (unedited) at: http://atvb.ahajournals.org//subscriptions/ at: is online Arteriosclerosis, Thrombosis, and Vascular Biology Information about subscribing to Subscriptions: http://www.lww.com/reprints Information about reprints can be found online at: Reprints: document. Question and Answer Permissions and Rights page under Services. Further information about this process is available in the which permission is being requested is located, click Request Permissions in the middle column of the Web Copyright Clearance Center, not the Editorial Office. Once the online version of the published article for can be obtained via RightsLink, a service of the Arteriosclerosis, Thrombosis, and Vascular Biology in Requests for permissions to reproduce figures, tables, or portions of articles originally published Permissions: by guest on February 20, 2013 http://atvb.ahajournals.org/ Downloaded from

Transcript of Yan Huang, Andrew J. Fleming, Shan Wu, Gabriel Virella and Maria ...

Page 1: Yan Huang, Andrew J. Fleming, Shan Wu, Gabriel Virella and Maria ...

Yan Huang, Andrew J. Fleming, Shan Wu, Gabriel Virella and Maria F. Lopes-Virellain U937 Cells via Mitogen-Activated Protein Kinase

Receptor Cross-Linking by Immune Complexes Induces Matrix Metalloproteinase-1γFc-

Print ISSN: 1079-5642. Online ISSN: 1524-4636 Copyright © 2000 American Heart Association, Inc. All rights reserved.

Greenville Avenue, Dallas, TX 75231is published by the American Heart Association, 7272Arteriosclerosis, Thrombosis, and Vascular Biology

doi: 10.1161/01.ATV.20.12.25332000;20:2533-2538Arterioscler Thromb Vasc Biol. 

http://atvb.ahajournals.org/content/20/12/2533World Wide Web at:

The online version of this article, along with updated information and services, is located on the

http://atvb.ahajournals.org/content/suppl/2000/11/29/20.12.2533.DC1.htmlData Supplement (unedited) at:

  http://atvb.ahajournals.org//subscriptions/

at: is onlineArteriosclerosis, Thrombosis, and Vascular Biology Information about subscribing to Subscriptions:

  http://www.lww.com/reprints

Information about reprints can be found online at: Reprints: 

document. Question and AnswerPermissions and Rightspage under Services. Further information about this process is available in the

which permission is being requested is located, click Request Permissions in the middle column of the WebCopyright Clearance Center, not the Editorial Office. Once the online version of the published article for

can be obtained via RightsLink, a service of theArteriosclerosis, Thrombosis, and Vascular Biologyin Requests for permissions to reproduce figures, tables, or portions of articles originally publishedPermissions:

by guest on February 20, 2013http://atvb.ahajournals.org/Downloaded from

Page 2: Yan Huang, Andrew J. Fleming, Shan Wu, Gabriel Virella and Maria ...

Fc-g Receptor Cross-Linking by Immune ComplexesInduces Matrix Metalloproteinase-1 in U937 Cells via

Mitogen-Activated Protein KinaseYan Huang, Andrew J. Fleming, Shan Wu, Gabriel Virella, Maria F. Lopes-Virella

Abstract—Matrix metalloproteinase-1 (MMP-1) secreted by macrophages potentially contributes to plaque rupture.Because large quantities of immunoglobulin G and ICs (ICs), including low density lipoprotein–containing ICs(LDL-ICs), are present in atherosclerotic lesions, we examined the effect of LDL-ICs on macrophage MMP-1expression. With the use of ICs prepared with human LDL and rabbit anti-LDL antiserum, our enzyme-linkedimmunosorbent assays and Northern blots showed that MMP-1 secretion and expression by U937 histiocytes wereinduced by LDL-ICs. Furthermore, our results showed that blocking of Fc-g receptor I and II (FcgRI and FcgRII)inhibited 70% and 55%, respectively, of the LDL-IC–induced secretion of MMP-1. Finally, our data showed that bothPD98059, an inhibitor of the mitogen-activated protein kinase pathway, and Ro-31-2880, an inhibitor of protein kinaseC, inhibited LDL-IC–stimulated MMP-1 secretion in a dose-dependent manner, with 96% and 95% inhibition,respectively, at the respective doses of 50mmol/L and 80 nmol/L. In conclusion, this study demonstrated for the firsttime that LDL-ICs induce macrophage MMP-1 secretion by cocross-linking FcgRI and FcgRII and triggering a proteinkinase C–dependent mitogen-activated protein kinase pathway. These results suggest that LDL-ICs and other ICslocalized in atherosclerotic plaques may be potent stimulators for macrophage MMP-1 expression and may contributeto plaque rupture and acute coronary events.(Arterioscler Thromb Vasc Biol. 2000;20:2533-2538.)

Key Words: LDL n metalloproteinasen immune complexn collagen

Matrix metalloproteinases (MMPs) have been implicatedin the disruption of atherosclerotic plaques, which

leads to acute coronary artery events.1–3 In the MMP family,MMP-1 is responsible for the initial cleavage of fibrillar typeI collagen,4,5 which accounts for 50% to 75% of total collagenin the intima of atherosclerotic plaques.6,7 Immunocytochem-istry studies have shown that MMP-1 is expressed by all 3major cellular components in atherosclerotic lesions: endo-thelial cells, smooth muscle cells, and macrophages, but notby cells in the normal arterial wall.8 Focal overexpression ofMMP-1 in human atherosclerotic lesions was frequentlyfound in areas that would be anticipated to have increasedmechanical stress and be prone to rupture.4 These datastrongly suggested that MMP-1 might play an important rolein the vulnerability of atherosclerotic plaques.

Plaque rupture occurs most frequently at the shoulderregions of atherosclerotic plaques and results in hemorrhage,thrombosis, and rapid occlusion of the artery.9 On the basis ofthe observation that immunoreactive MMP-1 and macro-phages are colocalized in the ruptured shoulder regions, it isgenerally believed that macrophages are the major source of

MMP-1 contributing to plaque rupture.4,10–13 Inflammatorycytokines have been postulated to be responsible for inducingMMP-1 expression in the lesion-associated macrophages,because cytokines such as tumor necrosis factor-a TNFa andinterleukin-1b IL-1b have been detected in atheroscleroticlesions and have been shown to stimulate MMP-1 expressionin fibroblasts, smooth muscle cells, and other neoplastictissues.2 However, a recent in vitro study14 showing thatTNF-a, IL-1b, and interferon-g IFN-g had no effect onMMP-1 expression in human monocyte-derived macrophagesdid not support this hypothesis. Thus, evidence suggests thatlocal factors other than cytokines may stimulate MMP-1expression in lesion-associated macrophages.

It has been previously demonstrated that atheroscleroticlesions contain large quantities of immune complexes (ICs),including LDL-containing ICs (LDL-ICs).15–17In the presentin vitro study, we investigated the roles of LDL-ICs and otherICs in MMP-1 expression and secretion by human U937histiocytes (macrophages). We found that cocross-linking ofFc-g receptors I and II (FcgRI, FcgRII) by LDL-ICs inducedMMP-1 expression and secretion. We also found that induc-

Received August 3, 2000; revision accepted September 6, 2000.From the Division of Endocrinology, Diabetes, and Medical Genetics (Y.H., A.J.F., S.W., M.F.L.-V.), Department of Medicine, and the Department

of Immunology and Microbiology (G.V.), Medical University of South Carolina, and the Ralph H. Johnson Veterans Administration Medical Center(Y.H., M.F.L.-V.), Charleston, SC.

Part of this work was presented at the American Heart Association 72nd Scientific Sessions and was published in abstract form (Circulation1999;[supplI]100:I-252).

Correspondence to Yan Huang, MD, PhD, Division of Endocrinology, Diabetes, and Medical Genetics, Department of Medicine, Medical Universityof South Carolina, 114 Doughty St, Charleston, SC 29403. E-mail [email protected].

© 2000 American Heart Association, Inc.

Arterioscler Thromb Vasc Biol.is available at http://www.atvbaha.org

2533 by guest on February 20, 2013http://atvb.ahajournals.org/Downloaded from

Page 3: Yan Huang, Andrew J. Fleming, Shan Wu, Gabriel Virella and Maria ...

tion of MMP-1 expression was mediated by a protein kinaseC (PKC)– dependent mitogen-activated protein kinase(MAPK) signaling pathway. Thus, this study demonstratesfor the first time that interaction of ICs with macrophagesstimulates MMP-1 expression and secretion.

Methods

Cell Culture ExperimentsU937 histiocytes18 were cultured in a 5% CO2 atmosphere inIscove’s modified Dulbecco’s media supplemented with 10% fetalcalf serum. The medium was changed every 2 to 3 days. Thehistiocytic (resident macrophage) origin of U937 cells was shown bytheir capacity for lysozyme production and the strong esteraseactivity.18 Peripheral blood human monocytes were isolated fromplasma collected from healthy, normolipidemic human volunteers aspreviously described19 by using the method of Recalde,20 withmodifications described by Fogelman et al.21

Isolation of Lipoproteins and Preparation ofInsoluble ICsLDL (d51.019 to 1.063 g/mL) was isolated from the plasma ofnormal volunteers and oxidatively modified as described.22 InsolubleLDL-ICs were prepared with human native LDL and rabbit anti-LDLantiserum and quantified as described previously.23–25Our previousstudies had shown that LDL-ICs prepared with both human LDL andrabbit anti-LDL antiserum and those prepared with collagen I–im-mobilized human oxidized LDL (oxLDL) and anti–oxLDL autoan-tibodies activated MAPK in macrophages,22 suggesting that theformer can be used as a model for the latter in studies examining theinteraction of LDL-ICs with macrophages.

ELISA of Secreted MMPsSecreted MMPs from U937 cells were quantified by using sandwichELISA kits according to the protocol provided by the manufacturer(Oncogene).

Northern Blot AnalysisTotal cellular RNA was isolated from U937 cells by using theUltraspec RNA isolation reagent according to the instructions fromthe manufacturer (Biotecx Laboratories). Northern blotting ofMMP-1 mRNA was performed as described previously.26

FcgR BlockingFcgRI was blocked with human monomeric IgG1 isolated fromserum by NH4(SO4)2 precipitation followed by affinity chromatog-raphy on a protein A/G column (ImmunoPure, Pierce) as describedpreviously.19 Human monomeric IgG2 was used as negative controlfor IgG1.19,27 Analysis of IgG subclasses by radial immunodiffusion(Binding Site) on the purified IgG1 showed mild contamination withIgG2 (2.4% of the total IgG). To eliminate aggregates, purified IgG1

and IgG2 were centrifuged at 100 000gfor 30 minutes before beingadded to the culture medium. FcgRII was blocked with a monoclonalanti-CD32 antibody (composition, IgG2b,k, clone FLI8.26; PharM-ingen). A monoclonal antibody (clone 27-35) with the same isotype(IgG2b,k) was used as a negative control for anti-CD32.

MAPK PhosphorylationPhosphorylation of MAPK was detected by Western blot analysis byusing monoclonal anti-phosphorylated and anti-p42/p44 MAPKantibodies (Santa Cruz Biotechnology) as described previously.22

DNA AssayCellular DNA was quantified with a CyQUANT cell proliferationassay kit according to the procedures provided by the manufacturer(Molecular Probes).

Collagenase Activity AssayCollagenase activity in conditioned medium was measured with theEnzChek assay kit according to the protocol provided by themanufacturer (Molecular Probes).

Statistical AnalysisData are presented as mean6SEM. Comparison between treatmentswas performed by using a 1-way ANOVA. A value ofP,0.05 wasconsidered significant.

ResultsConcentration- and Time-Dependent Stimulationof MMP-1 Secretion by LDL-ICsU937 cells were treated with increasing concentrations (0 to200 mg/mL) of LDL-ICs for 24 hours, and the amount ofsecreted MMP-1 in conditioned medium was assayed byELISA. Results showed that MMP-1 secretion was LDL-ICconcentration–dependent and reached a plateau at 150mg/mL of LDL-ICs (please see Figure I, published online athttp://atvb.ahajournals.org). The kinetic study on MMP-1secretion in response to LDL-ICs showed that MMP-1secretion was time-dependent and started to plateau after24-hour stimulation (please see Figure II, published online athttp://atvb.ahajournals.org). On the basis of these results, wechose 150mg/mL and 24 hours as the optimal concentrationof LDL-IC and the stimulation time, respectively, for theremaining experiments.

Effects of LDL-ICs on Secretion of MMPsand TIMP-1The amount of MMP-1, -2, -3, and -9 secreted by U937 cellsin response to LDL-ICs was measured by ELISA. Our resultsshowed that LDL-ICs stimulated the secretion of MMP-1 by20-fold (Figure 1A). In contrast, LDL-ICs inhibited MMP-2secretion by 50% and had no effect on the secretion ofMMP-3 and -9 (data not shown). Our data also showed thatboth native and oxLDL had no effect on the secretion ofMMP-1 (Figure 1A) or MMP-2 (data not shown). Phorbol-12-myristate-13-acetate (PMA) as a positive control inducedmarked secretion of MMP-1 (Figure 1A). These resultsdemonstrated that LDL-ICs selectively stimulated MMP-1secretion by U937 cells. We also investigated whether or notthe stimulation of MMP-1 was accompanied by an increase inthe secretion of tissue inhibitor of metalloproteinase-1(TIMP-1). Data showed that TIMP-1 secretion was notinduced by LDL-ICs (Figure 1B) but was inhibited byoxLDL. The inhibition of TIMP-1 by oxLDL in macrophageswas also recently reported by Shah and coworkers.28 Incontrast, a 4-fold increase in TIMP-1 secretion was observedin cells stimulated with PMA (Figure 1B).

Induction of MMP-1 Expression in U937 Cellsby LDL-ICsThe effect of LDL-ICs on the steady-state level of MMP-1mRNA in U937 cells was determined by Northern blotting.As shown in Figure 2, MMP-1 mRNA was undetected incontrol cells but induced by LDL-ICs. PMA, which has beenshown to stimulate MMP-1 transcription in human monocyte-derived macrophages,29 markedly increased the MMP-1mRNA level. These results suggest that stimulation ofMMP-1 secretion by LDL-ICs is most likely secondary to anincrease in the MMP-1 mRNA level.

2534 Arterioscler Thromb Vasc Biol. December 2000

by guest on February 20, 2013http://atvb.ahajournals.org/Downloaded from

Page 4: Yan Huang, Andrew J. Fleming, Shan Wu, Gabriel Virella and Maria ...

Stimulation of Collagenase Activity by LDL-ICsCollagenase activity in cell-conditioned medium was deter-mined bu using fluorescein-conjugated type I collagen as asubstrate. Our results showed that collagenase activity inmedium conditioned by cells exposed to LDL-ICs wassignificantly higher than that observed with medium collected

from control cells (data not shown). Our data also showedthat the collagenase activity stimulated by LDL-ICs wascompletely inhibited by 1 mmol/L 1,10-phenanthroline (datanot shown), indicating that the collagenase activity wassecondary to the presence of metalloproteinases.

Engagement of FcgRI and FcgRII by LDL-ICs IsRequired for Induction of MMP-1 SecretionOur previous studies have shown that LDL-ICs engage FcgRIpredominantly and engage FcgRII to a lesser extent in humanmacrophages and THP-1 macrophage-like cells.19 To deter-mine whether MMP-1 secretion and expression induced byLDL-ICs in U937 cells was due to the engagement of FcgRIor FcgRII by LDL-ICs, we conducted experiments in whichincubation of the cells with LDL-ICs was performed in thepresence of FcgRI or FcgRII blockers. Human monomericIgG1, which binds to FcgRI with high affinity, and themonoclonal anti-CD32 (FcgRII) antibodies (clone FLI8.26)were used to block the interactions of LDL-ICs with FcgRIand FcgRII, respectively. Our results showed that humanmonomeric IgG1 and anti-CD32 inhibited LDL-IC–stimu-lated MMP-1 secretion in a dose-dependent manner, with70% and 55% inhibition at 20 and 5mg/mL, respectively(Figure 3). Human monomeric IgG2 was used as a negativecontrol, because it does not block either FcgRI or FcgRII. Asexpected, human monomeric IgG2 showed no effect onblocking the LDL-IC–stimulated MMP-1 secretion. Mousemonoclonal antibody clone 27-35, an isotype control foranti-CD32 (IgG2b,k), also had no effect on MMP-1 secretioninduced by LDL-ICs (Figure 3). Because U937 cells lackFcgRIII,30 anti-CD16 (FcgRIII) antibody was also used as anirrelevant antibody to exclude the nonspecific interactionbetween antibodies and U937 cells. Our results showed thatanti-CD16 did not block MMP-1 secretion (Figure 3). Thesedata strongly suggest that cocross-linking of FcgRI andFcgRII by LDL-ICs might be responsible for the induction ofMMP-1 secretion in U937 cells.

Figure 1. Effects of LDL-ICs on secretion of MMP-1 and TIMP-1by U937 cells. U937 cells were incubated at 37°C for 24 hourswith culture medium alone (control) or with medium containing150 mg/mL LDL-IC, 100 mg/mL native LDL, 100 mg/mL oxLDL,50 mL/mL anti-LDL antiserum, or 100 nmol/L PMA. After incuba-tion, secreted MMP-1 (A) and TIMP-1 (B) in the conditionedmedium were quantified by ELISA, and the cells were lysed forDNA assay as described in Methods. Data represent themean6SEM of 3 experiments run in duplicate.

Figure 2. Northern blot of MMP-1 mRNA expression in U937cells stimulated with LDL-ICs. U937 cells were incubated for 24hours with medium alone (C) or with medium containing 100nmol/L PMA (P) or 150 mg/mL LDL-ICs (IC). After incubation,total RNA was isolated and 20 mg of RNA for each sample wasused for Northern blot analysis of MMP-1 and GAPDH mRNA asdescribed in Methods. The experiment was performed withduplicate dishes for control and LDL-IC–stimulated cells.

Figure 3. Blocking of LDL-IC–stimulated MMP-1 secretion bymonomeric human IgG1 and anti-CD32 monoclonal antibodies.U937 cells were incubated for 24 hours with 150 mg/mL LDL-ICs in the absence or presence of increasing concentrations ofhuman IgG1, IgG2, anti-CD32 monoclonal antibodies, an anti-CD32 isotype control antibody (mouse IgG2b,k), or anti-CD16monoclonal antibodies as indicated. After incubation, the condi-tioned medium was subjected to ELISA to determine theamount of secreted MMP-1. The amount of LDL-IC–stimulatedMMP-1 in the absence of blocking antibodies was designatedas 100%. Data presented are the means of 3 experiments run intriplicate.

Huang et al Immune Complexes Stimulate Macrophage MMP-1 2535

by guest on February 20, 2013http://atvb.ahajournals.org/Downloaded from

Page 5: Yan Huang, Andrew J. Fleming, Shan Wu, Gabriel Virella and Maria ...

LDL-ICs Stimulate MMP-1 Secretion viaActivation of a PKC-Dependent MAPK PathwayCross-linking of FcgRI triggers activation of MAPK path-ways.31 Our recent study has demonstrated that LDL-ICsactivate MAPK (extracellular signal–related protein kinase[ERK]) in THP-1 macrophage-like cells through engagingFcgRI.22 On the basis on these observations, we determinedwhether the altered MMP-1 expression/secretion had resultedfrom activation of an MAPK pathway induced by LDL-ICs.First, we examined the effect of LDL-ICs on phosphorylationof p42/p44 MAPK (ERK1/2) in U937 cells by Western blotanalysis with the use of both anti-phosphorylated and anti–total ERK antibodies. Our data showed that LDL-ICs inducedphosphorylation of ERK, mainly ERK2, in a time-dependentmanner, with peak phosphorylation occurring at 30 minutes(Figure 4). We then performed experiments in which U937cells were treated with LDL-ICs in the absence or presence ofPD98059,32 and the amount of MMP-1 released into theculture medium was determined after treatment. Our resultsshowed that PD98059 inhibited MMP-1 secretion in a dose-dependent manner, with complete inhibition at 50mmol/L(Figure 5). Dimethyl sulfoxide (DMSO), a vehicle forPD98059, had no blocking effect on MMP-1 secretion. Toensure that PD98059 inhibited ERK phosphorylation in U937cells, Western blotting was performed and the results showed

that 50mmol/L PD98059 significantly inhibited ERK phos-phorylation (data not shown).

Because our recent study22 showed that activation of theMAPK signaling pathway in macrophage by LDL-ICs wasPKC dependent, the role of PKC in MAPK-mediated MMP-1secretion was also investigated. Our results showed that thePKC inhibitor Ro-31-822033 blocked MMP-1 secretion in aconcentration-dependent manner, and complete inhibitionwas achieved at 80 nmol/L (please see Figure III, publishedonline at http://atvb.ahajournals.org). The effect of Ro-31-8220on ERK phosphorylation was determined, and the resultsshowed that 80 nmol/LRo-31-8220 significantly inhibited ERKphosphorylation (please see Figure IV, published online at http://atvb.ahajournals.org). These data suggest that LDL-ICs inducedMMP-1 secretion via a PKC-dependent MAPK signaling pathway.

Stimulation of MMP-1 Secretion byIgG–Anti-IgG ICsBecause atherosclerotic lesions also contain ICs other thanLDL-ICs,17 we determined whether these non-LDL ICs alsostimulated MMP-1 secretion by U937 cells. Insoluble IgG–anti-IgG ICs IgG-ICs were prepared by incubating humanIgG with rabbit anti-human IgG antiserum. Our resultsshowed that insoluble IgG–anti-IgG ICs also stimulatedMMP-1 secretion by U937 cells (please see Figure V,published online at http://atvb.ahajournals.org). Thus, thesedata indicate that insoluble IgG-containing ICs are capable ofinducing MMP-1 secretion by U937 cells, regardless of theirantigen content.

Stimulation of MMP-1 Secretion From HumanMonocyte-Derived Macrophages by LDL-ICsWe further determined whether LDL-ICs also stimulatedMMP-1 secretion by human monocyte-derived macrophages.Our results showed that treatment of macrophages withLDL-ICs resulted in a 3-fold increase in MMP-1 secretionover control levels (please see Figure VI, published online athttp://atvb.ahajournals.org).

DiscussionThe titer of anti-oxLDL autoantibodies has been measured inhuman serum by our and other laboratories.34–38In addition toserum, the unbound autoantibodies as well as those com-plexed with oxLDL have been also localized in atheroscle-rotic lesions.17 The potential role of LDL-ICs in atherogenesishas been extensively investigated by our and other groups(for a review, see Reference 38). In our previous studies ofinsoluble LDL-ICs prepared with human LDL and rabbitanti-LDL antiserum, we demonstrated that interaction be-

Figure 4. Time-dependent stimulation of MAPK inU937 cells by LDL-ICs. U937 cells were stimulatedwith 150 mg/mL LDL-ICs for the times indicated andthen lysed. Twenty-five micrograms of cell proteinwas electrophoresed on a 10% SDS polyacrylamidegel and then transferred to a polyvinylidene difluoridemembrane. The membrane was immunoblotted withanti-phosphorylated or anti-p42/p44 MAPK antibod-ies as described in Methods. MAPK was visualizedby incubating the membrane with chemilumines-cence reagent for 1 minute and exposure to x-rayfilm for 15 seconds. Data are representative of 3experiments with similar results.

Figure 5. Inhibition of LDL-IC–stimulated MMP-1 secretion byPD98059. U937 cells were incubated for 24 hours with 150mg/mL LDL-ICs in the presence of increasing concentrations ofPD98059 as indicated. After incubation, the conditionedmedium was subjected to ELISA to quantify secreted MMP-1.The amount of secreted MMP-1 by LDL-IC–stimulated cells inthe absence of PD98059 was designated as 100%. Dimethylsulfoxide (DMSO), a vehicle for PD98059, was 0.1% of themedium volume. Data presented are the mean6SEM of 3 differ-ent experiments run in duplicate.

2536 Arterioscler Thromb Vasc Biol. December 2000

by guest on February 20, 2013http://atvb.ahajournals.org/Downloaded from

Page 6: Yan Huang, Andrew J. Fleming, Shan Wu, Gabriel Virella and Maria ...

tween LDL-ICs and macrophages led to foam cell formationand macrophage activation.24,39 Recently, we have demon-strated that this interaction was mediated predominantly byFcgRI.19 Our present study shows that MAPK activationelicited by interaction between LDL-ICs and the FcgRs onU937 histiocytes led to induction of MMP-1 expression andsecretion. Because MMP-1 has been shown to be potentiallyinvolved in the disruption of atherosclerotic plaques,2 ourpresent study suggests that LDL-ICs may play an importantrole not only in the initiation and progression of atheroscle-rosis but also in triggering acute coronary events.

TIMP-1 and MMP-1 have been shown to be coordinatelyregulated by some factors, such as PMA and interleukin-1b,and reciprocally regulated by others, such as transforminggrowth factor-b1, which downregulates TIMP-1 while up-regulating MMPs.40 Our present study has shown that LDL-ICs stimulate MMP-1 secretion but have no effect onTIMP-1, resulting in a net increase in collagenase activity.The complete inhibition of stimulated collagenase activity byphenanthroline indicates that the enhanced collagenase activ-ity is secondary to the increased secretion of MMPs. Becausephenanthroline is not a specific inhibitor of MMP-1, our datado not ascertain that MMP-1 is the responsible enzyme.Although MMP-1 might be at least partially responsible forincreased collagenase activity, because our results show thatLDL-ICs selectively stimulate MMP-1 and the fluorescence-labeled type I collagen is the substrate of MMP-1, theinvolvement of MMP-13 needs to be excluded. Libby andcoworkers41 recently demonstrated increased levels ofMMP-13 and MMP-1 and the loss of interstitial collagen typeI in atheromatous versus fibrous plaques.

An intriguing result of this study is that cocross-linking ofFcgRI and FcgRII by LDL-ICs seems to be essential forinduction of MMP-1 secretion, because human monomericIgG1 and anti-CD32 monoclonal antibody blocked theMMP-1 secretion induced by LDL-ICs by 70% and 55%,respectively. To the best of our knowledge, this is the firstreport to show that cocross-linking of FcgRI and FcgRIIcoordinates gene expression. Because ICs are capable ofbinding either FcgRI or FcgRII on the surface of U937 cells,they can either cocross-link FcgRI and FcgRII separately andsimultaneously or cocross-link FcgRI with FcgRII. Ourblocking study did not distinguish the cocross-linking thatplays an essential role in MMP-1 stimulation, and furtherstudies are required to address this issue. Previously, anumber of studies on lymphocytes have reported that cocross-linking of 2 different cell surface receptors by ligands isneeded for regulating specific cell functions.42–44 For exam-ple, it has been shown that in B cells, cocross-linking ofFc-«RII and the B-cell receptor modulates B-cell activation42

and that cocross-linking of CD27 and the B-cell receptoraugments CD27-mediated B-cell apoptosis.43 In T cells, it hasbeen shown that cocross-linking of CD3 and CD4 results inenhanced mobilization of free intracellular calcium.44 Ourprevious studies on human monocyte-derived macrophagesshowed that FcgRI was engaged by LDL-ICs predominantlyand that FcgRII was engaged to a lesser extent.19 Despite thefact that FcgRII is less engaged by LDL-ICs, our presentstudy showed that 55% of MMP-1 secretion by U937 cellswas inhibited by blocking FcgRII, suggesting that FcgRIImay play an important role in MMP-1 expression.

Unlike FcgRI as a single form, FcgRII has several iso-forms: FcgRIIA, IIB1, IIB2, and IIC.45 It has been shown thatFcgRIIB molecules are preferentially expressed by lympho-cytes, whereas FcgRIIA and IIC are preferentially expressedby neutrophils, and that human monocytes and macrophagesexpress all classes.45 However, the surface expression ofthese isoforms on human monocytes/macrophages remainsunknown, because the extracellular and transmembrane do-mains of these isoforms are nearly identical and the mono-clonal antibodies that distinguish among the surface epitopesof FcgRIIA, IIB, and IIC have not been successfullyproduced.45

In contrast to the extracellular and transmembrane do-mains, the cytoplasmic portions of FcgRII isoforms aredivergent.31 FcgRIIA/C contains an immunoreceptortyrosine-based activation motif, whereas FcgRIIB1/B2 con-tains an immunoreceptor tyrosine-based inhibition motif(ITIM). 31 Owing to the complexity of the surface expressionof FcgRII isoforms on monocytes/macrophages, the role ofthe ITIM-containing FcgRIIB in monocyte/macrophage acti-vation remains unknown. However, many studies have shownthat cross-linking of FcgRII on monocytes/macrophages ledto activation of signaling pathways,30,46,47suggesting that theinhibitory signal transmitted through ITIM-containingFcgRIIB may not mediate a complete shutdown of FcgRIIA/C-mediated tyrosine phosphorylation in human monocytes.Further extensive investigation is necessary to define the roleof FcgRIIB in signal transduction and gene expression inmonocytes and macrophages.

In summary, our present study demonstrates for the firsttime that cocross-linking of FcgRI and FcgRII by LDL-ICsinduces MMP-1 expression and secretion by U937 histiocytesvia activation of a PKC-dependent MAPK signaling pathway.These results suggest that the interaction between ICs andmacrophages in atherosclerotic plaques may lead to inductionof MMP-1 secretion, thus contributing to the disruption ofatherosclerotic plaques and acute coronary events.

AcknowledgmentsThis work was supported by an institutional grant from the

Medical University of South Carolina, the Atorvastatin ResearchAward, a Merit Review Grant from the Research Service of theDepartment of Veterans Affairs (to Y.H.), and in part by grantHL-55782 from the National Institutes of Health (to M.F.L.-V.). Wethank Charlyne Chassereau for the preparation of lipoproteins.

References1. Fernandez-Ortiz A, Badimon J, Falk E, Fuster V, Meyer B, Mailhac A,

Weng D, Shah PK, Badimon L. Characterization of the relative throm-bogenicity of atherosclerotic plaque components: implications for conse-quences of plaque rupture.J Am Coll Cardiol. 1994;23:11562–11569.

2. Libby P. Molecular bases of the acute coronary syndromes.Circulation.1995;91:2844–2850.

3. Falk E, Shah PK, Fuster V. Coronary plaque disruption.Circulation.1995;92:657–671.

4. Lee RT, Schoen FJ, Loree HM, Lark MW, Libby P. Circumferentialstress and matrix metalloproteinase 1 in human coronary atherosclerosis:implication for plaque rupture.Arterioscler Thromb Vasc Biol. 1996;16:1070–1073.

5. Matrisian LM. The matrix-degrading metalloproteinases.Bioessays.1992;14:455–463.

6. Morton LF, Barnes MJ. Collagen polymorphism in the normal anddiseased blood vessel wall: investigation of collagens types I, III and V.Atherosclerosis. 1982;42:41–51.

Huang et al Immune Complexes Stimulate Macrophage MMP-1 2537

by guest on February 20, 2013http://atvb.ahajournals.org/Downloaded from

Page 7: Yan Huang, Andrew J. Fleming, Shan Wu, Gabriel Virella and Maria ...

7. Hanson AN, Bentley JP. Quantitation of type I to type III collagen ratiosin small samples of human tendon, blood vessels, and atheroscleroticplaques.Anal Biochem. 1983;130:32–40.

8. Galis ZS, Sukhova GK, Lark MW, Libby P. Increased expression ofmatrix metalloproteinases and matrix degrading activity in vulnerableregions of human atherosclerotic plaques.J Clin Invest. 1995;94:2493–2503.

9. Davies MJ, Thomas AC. Plaque fissuring: the cause of acute myocardialinfarction, sudden ischaemic death, and crescendo angina.Br Heart J.1985;53:363–373.

10. Cheng GC, Loree HM, Kamm RD, Fishbein MC, Lee RT. Distribution ofcircumferential stress in ruptured and stable atherosclerotic lesions: astructural analysis with histopathological correlation.Circulation. 1993;87:1179–1187.

11. Lendon CL, Davies MJ, Born GVR, Richard PD. Atherosclerotic plaquecaps are locally weakened when macrophage density is increased.Ath-erosclerosis. 1991;87:87–90.

12. Libby P. Lesion versus lumen.Nat Med. 1995;1:17–18.13. Richard PD, Davies MJ, Born GVR. Influence of plaque configuration

and stress distribution on fissuring of coronary atherosclerotic plaques.Lancet. 1989;2:941–944.

14. Saren P, Welgus HG, Kovanen PT. TNF-a, and IL-1b belectively induceexpression of 92-kDa gelatinase by human macrophages.J Immunol.1996;157:4159–4165.

15. Parums O, Mitchinson MJ. Demonstration of immunoglobulin in theneighborhood of advanced atherosclerotic plaques.Atherosclerosis. 1981;38:211–216.

16. Hansson GK, Bondjers G, Bylock A, Hjalmarsson L. Ultrastructuralstudies on the localization of IgG in the aortic endothelium and suben-dothelial intima of atherosclerotic and non-atherosclerotic rabbits.ExpMol Pathol. 1980;33:302–315.

17. Yla-Herttuala S, Palinski W, Butler SW, Picard S, Steinberg D, WitztumJL. Rabbit and human atherosclerotic lesions contain IgG that recognizesepitopes of oxidized LDL.Arterioscler Thromb. 1994;14:32–40.

18. Sundstrom C, Nilsson K. Establishment and characterization of a humanhistiocytic lymphoma cell line (U937).Int J Cancer. 1976;17:565–577.

19. Lopes-Virella MF, Binzafar N, Rackley S, Takei A, LaVia M, Virella G.The uptake of LDL-containing immune complexes by human macro-phages: predominant involvement of the FcgRI receptor.Atherosclerosis.1997;135:161–170.

20. Recalde HR. A simple method of obtaining monocytes in suspension.J Immunol Methods. 1984;69:71–77.

21. Fogelman AM, Elahi F, Sykes K, Van Lenten BJ, Territo M, Berliner J.Modification of the Recalde method for the isolation of humanmonocytes.J Lipid Res. 1988;29:1243–1247.

22. Huang Y, Jaffa A, Koskinen S, Takei A, Lopes-Virella MF. OxidizedLDL–containing immune complexes induce Fc-g geceptor I–mediatedmitogen-activated protein kinase activation in THP-1 macrophages.Arte-rioscler Thromb Vasc Biol. 1999;19:1600–1607.

23. Huang Y, Ghosh MJ, Lopes-Virella MF. Transcriptional and post-transcriptional regulation of LDL receptor gene expression in PMA-treated THP-1 cells by LDL-containing immune complexes.J Lipid Res.1997;38:110–120.

24. Griffith RL, Virella GT, Stevenson HC, Lopes-Virella MF. Low densitylipoprotein metabolism by human macrophages activated with lowdensity lipoprotein immune complexes.J Exp Med. 1988;168:1041–1059.

25. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurementwith the Folin phenol reagent.J Biol Chem. 1957;193:265–275.

26. Huang Y, Mironova M, Lopes-Virella MF. Oxidized LDL stimulatesmatrix metalloproteinase-1 expression in human vascular endothelialcells.Arterioscler Thromb Vasc Biol. 1999;19:2640–2647.

27. Gergely J, Sarmay G. The two binding-site models of human IgG bindingFcg receptors.FASEB J. 1990;4:3275–3283.

28. Xu X, Meisel SR, Ong JM, Kaul S, Cercek B, Rajavashisth TB, SharifiB, Shah RK. Oxidized low density lipoprotein regulates matrixmetalloproteinase-9 and its tissue inhibitor in human monocyte-derivedmacrophages.Circulation. 1999;99:993–998.

29. Borden P, Heller RA. Transcriptional control of matrix metallopro-teinases and the tissue inhibitors of matrix metalloproteinases.Crit RevEukayot Gene Exp. 1997;7:159–178.

30. Liao F, Shin H, Rhee SG. Tyrosine phosphorylation of phospholipaseC-g1 induced by crosslinking of the high-affinity or low-affinity Fcreceptor for IgG in U937 cells.Proc Natl Acad Sci U S A. 1992;89:3659–3663.

31. Daeron M. Fc receptor biology.Annu Rev Immunol. 1997;15:203–234.32. Rose DM, Winston BW, Chan ED, Riches DWH, Gerwins P, Johnson

GL, Henson PM. Fcg receptor crosslinking activates p42, p38, andJNK/SAPK mitogen-activated protein kinases in murine macrophages:role for p42MAPK in Fcg receptor-stimulated TNF-a synthesis.J Immunol.1997;158:3433–3438.

33. Paolucci L, Rozengurt E. Protein kinase D in small cell lung cancer cells:rapid activation through protein kinase C.Cancer Res. 1999;59:572–577.

34. Mironova M, Lopes-Virella MF, Virella G. Isolation and characterizationof human anti-oxidized LDL autoantibodies.Arterioscler Thromb VascBiol. 1996;16:222–229.

35. Boullier A, Hamon M, Walters-Laporte E, Martin-Nizart F, Mackereel R,Fruchart J-C, Bertrand M, Furiez P. Detection of autoantibodies againstoxidized LDL and of IgG-bound LDL in patients with coronary heartdisease.Clin Chim Acta. 1995;238:1–10.

36. Salonen JT, Yla-Herttuala S, Yamamoto R, Butler S, Korpela H, SalonenR, Nyyssonen K, Palinski W, Witztum JL. Autoantibody against oxidizedLDL and progression of carotid atherosclerosis.Lancet. 1992;339:883–887.

37. Maggi E, Chiesa R, Melissano G, Castellano R, Astore D, Grossi A,Finardi G, Bellomo G. LDL oxidation in patients with severe carotidatherosclerosis: a study of in vitro and in vivo oxidation markers.Arte-rioscler Thromb. 1994;14:1892–1899.

38. Mironova M, Virella G, Virella-Lowell I, Lopes-Virella MF. Anti-modified LDL antibodies, and LDL-containing immune complexes inIDDM patients and healthy controls.Clin Immunol Immunopathol. 1997;85:73–82.

39. Virella G, Munoz JF, Galbraith GM, Gissinger C, Chassereau CH, Lopes-Virella MF. Activation of human monocyte-derived macrophage byimmune complexes containing low density lipoprotein.Clin ImmunolImmunopathol. 1995;75:179–189.

40. Overall CM. Regulation of tissue inhibitor of matrix metalloproteinaseexpression.Ann N Y Acad Sci. 1994;732:51–64.

41. Sukhova GK, Schonbeck U, Rabkin E, Schoen FJ, Poole AR, BillinghurstRC, Libby P. Evidence for increased collagenolysis by interstitial colla-genase-1 and–3 in vulnerable human atheromatous plaques.Circulation.1999;99:2503–2509.

42. Campbell KA, Lees A, Finkelman FD, Conrad DH. Co-crosslinkingFc-«RII/CD23 and B cell surface immunoglobulin modulates B cellactivation.Eur J Immunol. 1992;22:2107–2112.

43. Prasad KV, Ao Z, Yoon Y, Wu MX, Rizk M, Jacquot S, Schlossman SF.CD27, a member of the tumor necrosis factor receptor family, inducesapoptosis and binds to Siva, a proapoptotic protein.Proc Natl Acad SciU S A. 1997;94:6346–6351.

44. Baier-Bitterlich G, Baier G, Gulbins E, Coggeshall KM, Altman A. Therole of p56ck in CD4-mediated suppression of CD3-induced early sig-naling events in T lymphocytes.Life Sci. 1995;56:1287–1297.

45. Ravetch JV, Kinet J. Fc receptors.Annu Rev Immunol. 1991;9:457–492.46. Ghazizadeh S, Bolen JB, Fleit HB. Physical and functional association of

Src-related protein tyrosine kinases with FcgRII in monocytic THP-1cells.J Biol Chem. 1994;269:8878–8884.

47. Pan X, Darby C, Indik ZK, Schreiber AD. Activation of three classes ofnonreceptor tyrosine kinases following Fcg receptor crosslinking inhuman monocytes.Clin Immunol. 1999;90:55–64.

2538 Arterioscler Thromb Vasc Biol. December 2000

by guest on February 20, 2013http://atvb.ahajournals.org/Downloaded from