Sebastian Sasu, David LaVerda, Nilofer Qureshi, Douglas T. Golenbock and Debbie BeasleyMitogen-Activated Protein Kinase Activation

Human Vascular Smooth Muscle Cells via Toll-Like Receptor 4 and p44/p42 and Chlamydial Heat Shock Protein 60 Stimulate Proliferation ofChlamydia pneumoniae

Print ISSN: 0009-7330. Online ISSN: 1524-4571 Copyright © 2001 American Heart Association, Inc. All rights reserved.is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation Research

doi: 10.1161/hh1501.0941842001;89:244-250; originally published online July 19, 2001;Circ Res. 

http://circres.ahajournals.org/content/89/3/244World Wide Web at:

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

  http://circres.ahajournals.org//subscriptions/

is online at: Circulation Research Information about subscribing to Subscriptions: 

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

  document. Permissions and Rights Question and Answer about this process is available in the

located, click Request Permissions in the middle column of the Web page under Services. Further informationEditorial Office. Once the online version of the published article for which permission is being requested is

can be obtained via RightsLink, a service of the Copyright Clearance Center, not theCirculation Researchin Requests for permissions to reproduce figures, tables, or portions of articles originally publishedPermissions:

by guest on August 20, 2014http://circres.ahajournals.org/Downloaded from by guest on August 20, 2014http://circres.ahajournals.org/Downloaded from

Chlamydia pneumoniaeand Chlamydial Heat Shock Protein60 Stimulate Proliferation of Human Vascular Smooth

Muscle Cells via Toll-Like Receptor 4 and p44/p42Mitogen-Activated Protein Kinase Activation

Sebastian Sasu, David LaVerda, Nilofer Qureshi, Douglas T. Golenbock, Debbie Beasley

Abstract—An early component of atherogenesis is abnormal vascular smooth muscle cell (VSMC) proliferation. Thepresence ofChlamydia pneumoniaein many atherosclerotic lesions raises the possibility that this organism plays acausal role in atherogenesis. In this study,C pneumoniaeelementary bodies (EBs) rapidly activated p44/p42mitogen-activated protein kinases (MAPKs) and stimulated proliferation of VSMCs in vitro. Exposure of VSMCsderived from human saphenous vein toC pneumoniaeEBs (33107 inclusion forming units/mL) enhanced bromode-oxyuridine (BrdU) incorporation 1263-fold. UV- and heat-inactivatedC pneumoniaeEBs also stimulated VSMCproliferation, indicating a role of direct stimulation by chlamydial antigens. However, the mitogenic activity ofC pneumoniaewas heat-labile, thus excluding a role of lipopolysaccharide. Chlamydial hsp60 (25mg/mL) replicated theeffect of C pneumoniae, stimulating BrdU incorporation 763-fold. Exposure toC pneumoniaeor chlamydial hsp60rapidly activated p44/p42 MAPK, within 5 to 10 minutes of exposure. In addition, PD98059 and U0126, which are twodistinct inhibitors of upstream MAPK kinase 1/2 (MEK1/2), abolished the mitogenic effect ofC pneumoniaeandchlamydial hsp60. Toll-like receptors (TLRs) act as sensors for microbial antigens and can signal via the p44/p42MAPK pathway. Human VSMCs were shown to express TLR4 mRNA and protein, and a TLR4 antagonist abolishedchlamydial hsp60–induced VSMC proliferation and attenuatedC pneumoniae–induced MAPK activation and VSMCproliferation. Together these results indicate thatC pneumoniaeand chlamydial hsp60 are potent inducers of humanVSMC proliferation and that these effects are mediated, at least in part, by rapid TLR4-mediated activation of p44/p42MAPK. (Circ Res. 2001;89:244-250.)

Key Words: Chlamydia pneumoniaen heat shock proteinsn vascular smooth musclen cell divisionn mitogen-activated protein kinases

A therosclerosis is an inflammatory disease, with theearliest stages characterized by the invasion of the

intima by mononuclear phagocytes and by intimal hyperpla-sia.1 Accumulating evidence indicates that chronic infectionwith the ubiquitous respiratory pathogenChlamydia pneu-moniae, a Gram-negative obligate intracellular bacterium,may be an additional risk factor for atherosclerosis. Macro-phages are thought to become infected withC pneumoniaeinthe respiratory tract and then enter the circulation and crossthe endothelium at sites of preexisting vascular inflammation.The first report linking C pneumoniaeto atherosclerosisidentified the organism by electron microscopy in coronaryatherosclerotic plaques and localized it to intimal smoothmuscle cells (SMCs).2 C pneumoniaehas been found fre-quently in lesions of the aorta, iliac, carotid, and coronaryarteries,3–5 but is rarely found in normal arterial tissue.6 In

vitro evidence supports the notion thatC pneumoniaecaninfect human arterial SMCs.7–9 However, it is not clearwhetherC pneumoniaeorganisms that have been identifiedwithin SMCs of human atheromas are actively replicating orviable. Irrespective of whetherC pneumoniaereplicateswithin SMCs in vivo, its presence in atherosclerotic lesionsraises the issue of whether the organism plays a causal role oris an innocent bystander.3,10

Evidence supporting a causal role of chlamydia in athero-genesis comes from recent reports thatC pneumoniaeandchlamydial antigens can activate mononuclear phagocytesand vascular cells, including SMCs. A heat-stable componentof C pneumoniaeinduces macrophage foam cell formation,11

and this effect is replicated by chlamydial lipopolysaccharide(LPS). On the other hand, a heat-labile component ofC pneumoniaestimulates oxidation of LDL12 and synthesis of

Original received January 26, 2001; revision received May 11, 2001; accepted June 6, 2001.From the Department of Medicine (S.S., D.B.), New England Medical Center Hospitals and Tufts University School of Medicine, and Department of

Medicine (D.L., D.T.G.), Boston Medical Center, Boston, Mass, and Department of Animal Health and Biomedical Sciences (N.Q.), University ofWisconsin, Madison, Wis.

Correspondence to Debbie Beasley, PhD, Box 172, New England Medical Center, 750 Washington St, Boston, MA 02111. E-maildbeasley@lifespan.org

© 2001 American Heart Association, Inc.

Circulation Researchis available at http://www.circresaha.org

244 by guest on August 20, 2014http://circres.ahajournals.org/Downloaded from

proinflammatory cytokines, including interleukin (IL)–1,IL-6, and tumor necrosis factor (TNF)–a,13 by human bloodmononuclear cells. Heat shock protein 60 (hsp60) is a highlyexpressed chlamydial protein that can activate macrophagesand vascular cells by mechanisms that are not well charac-terized. Chlamydial hsp60 mimics the ability ofC pneu-moniaeto stimulate LDL oxidation by human blood mono-nuclear cells.12 Chlamydial hsp60 also induces synthesis ofTNF-a and matrix-degrading metalloproteinases by mousemacrophages14 and expression of adhesion molecules byhuman endothelial cells.15 We undertook the present study todetermine whetherC pneumoniaeor chlamydial hsp60 stim-ulates proliferation of human vascular SMCs (VSMCs), aprocess responsible for intimal hyperplasia in early athero-sclerotic lesions.

A crucial component of signaling via classical growthfactors involves sequential activation of Ras, Raf, and p44/p42 mitogen-activated protein kinase (MAPK). Exposure toC pneumoniaeinduces rapid activation of p44/p42 MAPK inhuman umbilical vein endothelial cells,16 although the signal-ing pathway that links chlamydia to MAPK activation is notknown. Recent studies have documented the role of trans-membrane Toll-like receptors (TLRs) in cellular activation bymicrobial pathogens.17 Microbial antigens may interact withthe extracellular domain of TLRs and subsequently activatemultiple intracellular signaling pathways. Bacterial LPS–induced activation of nuclear factor (NF)–kB and p44/p42MAPK pathways has been extensively studied, and is nowknown to involve TLR4.18–21An inhibitor of TLR4-mediatedLPS signaling also abolished induction of inflammatorycytokine synthesis after exposure toChlamydia trachomatis,a related species of chlamydia.22 The present study testedwhetherC pneumoniaestimulates human VSMC prolifera-tion via activation of TLR4 and/or p44/p42 MAPK.

Materials and MethodsHuman VSMC CultureVSMCs were obtained by explant technique from saphenous veinsharvested for coronary artery bypass surgery at New EnglandMedical Center and from segments of pulmonary artery harvestedfrom organ donors (National Disease Research Interchange, Phila-delphia, PA). VSMCs were cultured in DMEM supplemented with10% FCS, glutamine, penicillin, streptomycin, and fungizone, andused at passages 2 to 6.

Propagation and Purification of C pneumoniaeThe AR39 strain ofC pneumoniaewas inoculated into HEp-2 cellsat a multiplicity of infection of 10. After 48 to 72 hours, HEp-2 cellswere harvested on ice and sonicated. Elementary bodies (EBs) werepurified from host cell lysates on discontinuous gradients ofRenografin (E.R. Squibb and Sons) as described,23 and stored at280°C. In some cases, EBs were inactivated by exposure to UV light(30 W, 15 cm, 30 minutes) or to heat (56°C, 30 minutes), aspreviously described.12,24

Chlamydial hsp60Recombinant chlamydial hsp60 protein was a kind gift of Dr RichardP. Morrison (Montana State University, Bozeman, MT).C tracho-matisserovar A hsp60, fused with 8 additional amino acids (arginine,serine, and 6 histidine residues) at the carboxyl terminus, wasexpressed inEscherichia coli, and recombinant protein was purifiedby affinity chromatography with nickel-nitrilotriacetic acid resin, as

previously described.25 The endotoxin level of this preparation was,0.04 U/mg, as determined byLimulus amebocyte lysate assay(Associates of Cape Cod).

Bromodeoxyuridine (BrdU) IncorporationVSMCs were plated on glass coverslips (5000 cells/cm2), incubatedfor 24 hours, and then exposed toC pneumoniaeEBs or chlamydialhsp60 in DMEM/1% FCS. In some experiments, VSMCs werepreincubated for 30 minutes with MAPK kinase (MEK) inhibitorsPD98059 and U0126, or with vehicle (DMSO), before exposure toC pneumoniae. In others, the TLR4 antagonistRhodobacter spha-eroides diphosphoryl lipid A (RSLA; 1mg/mL) was added justbefore addition ofC pneumoniaeor hsp60. After 48 hours, BrdU wasadded and cells were incubated an additional 24 hours. To immu-nostain for BrdU, cells were washed in PBS, fixed in 3.7%formaldehyde (15 minutes), and permeabilized in ice-cold methanol(3 minutes). Nonspecific binding sites were blocked with 10%normal horse serum, and the cells were incubated for 90 minutes at37°C with BrdU monoclonal antibody (mAb; Amersham) andDNase I (10 U/mL), and then 45 minutes at room temperature withTexas Red–coupled donkey anti-mouse IgG. Cell nuclei werestained with Hoechst 33342 and evaluated by epifluorescencemicroscopy. For each treatment group, 200 cells were analyzed by anobserver blinded to the specimen treatment.

Cell ProliferationVSMCs were plated in 12-well plates (5000 cells/cm2), incubatedovernight, and the numbers of attached cells determined by hemo-cytometer (day 0 cell counts). VSMCs were then exposed toC pneumoniaeor DMEM/1% FCS alone, and final cell counts weredetermined after 96 hours.

p44/p42 MAPK PhosphorylationCells were rapidly frozen in liquid nitrogen and then harvested inbuffer containing (in mmol/L) NaCl 150, HEPES (pH 7.4) 50,NaVO4 1, and NaF 1; 1% Triton X-100; 10% glycerol; andproteinase inhibitor cocktail (Boehringer Mannheim). Cell lysateswere gently rotated (15 minutes, 4°C), centrifuged (16 000g, 20minutes, 4°C), and the supernatants stored at280°C. Cell proteins(40 mg) were separated on a 12% SDS-polyacrylamide gel andtransferred to nitrocellulose membranes. The membranes were incu-bated sequentially with 5% nonfat dry milk, rabbit antibodiesspecific for threonine- and tyrosine-phosphorylated p44/p42 MAPK(New England Biolabs), biotinylated goat anti-rabbit IgG, andstreptavidin-biotinylated alkaline phosphatase complex. Blots weredeveloped with nitroblue tetrazolium and 5-bromo-4-chloro-3-indolyl phosphate. To verify equivalency of cell extracts, replicateblots were incubated with rabbit antibodies specific for total p44/p42MAPK. NIH Image 1.61 was used to quantify bands.

Reverse Transcriptase–Polymerase Chain Reaction(RT-PCR) AnalysisTotal RNA was isolated from VSMCs using an RNeasy kit(Qiagen). RNA (2mg) was reverse-transcribed for 1 hour at 37°Cwith 200 units of Moloney murine leukemia virus reversetranscriptase and oligo(dT) primer, in a volume of 20mL. Thereaction was terminated at 95°C, and 2mL of first-strand cDNAadded to each PCR reaction. The primer sets used to amplifyTLR4 were 59-TGCGGGTTCTACATCAAA-39 and 59-CCATCCGAAATTATAAGAAAAGTC-39 , yielding a 413-bpfragment, and 59-TGGTGTCCCAGCACTTCATCC-39 and 59-TTCCTGCCAATTGCATCCTGTA-39, yielding a 299-bp frag-ment. The primer sets used to amplify TLR2 were 59-ACTTTGTGGATGGTGTGGGT-39 and 59-GAATATGCAGCC-TCCGGATT-39, yielding a 949-bp fragment and 59-AGGCTGCATTCCCAAGACACT-39 and 59-AGCCAGGCCCA-CATCATTTT-39, yielding a 519-bp fragment. Primers wereannealed at 55°C, and samples were amplified for 25 cycles.

Sasu et al C pneumoniae–Induced VSMC Proliferation 245

by guest on August 20, 2014http://circres.ahajournals.org/Downloaded from

Immunostaining and Flow Cytometric AnalysisVSMCs were gently scraped from culture plates with a soft rubberscraper in PBS containing 1% BSA and 1 mmol/L EDTA. Cells wereincubated 1 hour with HTA125 (aTLR4; a gift from K. Miyake,Saga Medical School, Saga, Japan), TL2.1 (aTLR226), or controlmouse IgG (each 10mg/mL), followed by FITC-labeled donkeyanti-mouse IgG. Cells were then analyzed for intensity of FITCfluorescence in a FACSCalibur flow cytometer (Becton Dickinson).Immunoreactive TLR4 was localized in individual SMCs by epifluo-rescence microscopy after indirect immunofluorescent staining.VSMCs were fixed with formaldehyde and then incubated with goatpolyclonal antibody specific for human TLR4 (Santa Cruz Biotech-nology), followed by Texas Red–coupled donkey anti-goat IgG.

ResultsA Heat-Labile Component of C pneumoniaeInduces Proliferation of VSMCsBrdU incorporation was markedly increased 48 to 72 hoursafter exposure of saphenous vein SMC cultures to viableC pneumoniaeEBs. Exposure toC pneumoniae(107 inclusionforming units [IFU]/mL) stimulated DNA synthesis 2.2-foldrelative to control VSMCs (Figure 1A; n55 experiments).Exposure toC pneumoniae(33107 and 53107 IFU/mL)produced 5.6-fold and 5.1-fold increases in BrdU incorpora-tion, respectively.C pneumoniae(33107 IFU/mL) producedmaximal stimulation of VSMC proliferation and was chosenfor further study. In 10 experiments with saphenous veinSMCs from different patients,C pneumoniaeenhanced BrdUincorporation 1263-fold (range 2.4- to 31-fold), with the

largest increases elicited in cells with lower baseline prolif-eration rates. Exposure toC pneumoniae(33107 IFU/mL)stimulated BrdU incorporation 3.460.1-fold in three experi-ments with SMCs derived from human pulmonary artery(BrdU-positive cells increased from 9.260.9% to30.863.6%). Direct cell counting confirmed thatC pneu-moniae–induced DNA synthesis was followed by cell repli-cation. The increment in cell number (from 4941 cells/cm2 onday 0) was'3-fold greater in saphenous vein SMCs incu-bated 96 hours withC pneumoniae(10 11161526 cells/cm2),compared with VSMCs incubated in medium alone(DMEM/1% FCS; 35696158 cells/cm2).

Mitogenic Effect of C pneumoniaeDoes NotRequire Active InfectionThe proliferative response elicited by viableC pneumoniaeEBs could be triggered by a specific chlamydial antigen ormay be secondary to an active infection within VSMCs. Todetermine whether infection of SMCs contributed to thismitogenic effect, the effect of viable EBs was compared withthat of EBs that had been inactivated by exposure to UV lightor heat (56°C). Exposure toC pneumoniae(33107 IFU/mL)stimulated BrdU incorporation 5.6-fold, whereas exposure toUV-inactivated or 56°C-inactivatedC pneumoniaestimulatedBrdU incorporation 3.3-fold and 2.7-fold, respectively (Fig-ure 1B; n55 experiments). These results indicate that theeffect of C pneumoniaeis mediated, at least in part, viarecognition of a chlamydial antigen, although infection mayenhance this effect.

Molecular components ofC pneumoniaethat can activatecells include LPS, major outer membrane protein,27 andhsp60.15 The activity of LPS is not affected by heating to100°C,28 whereas the activity of most proteins is destroyed.The mitogenic effect ofC pneumoniaewas abolished byheating to 100°C in five independent experiments (Figure1B), consistent with a role of a chlamydial protein, and rulingout a primary role of LPS.

Chlamydial hsp60 Is a Heat-Labile Inducer ofVSMC ProliferationBrdU incorporation was markedly increased in humanVSMCs exposed to chlamydial hsp60 for 72 hours. Thepercentage of cells that incorporated BrdU increased 5.3-fold,6.7-fold, 20-fold, and 22.4-fold in VSMCs exposed to 5, 10,25, and 50mg/mL chlamydial hsp60, relative to controlVSMCs (Figure 1C). Chlamydial hsp60 (25mg/mL) pro-duced near-maximal stimulation of VSMC proliferation andwas chosen for further study. Notably, exposure to chlamyd-ial hsp60 (25mg/mL) produced a mitogenic effect similar tothat of C pneumoniae(33107 IFU/mL). In five experimentswith VSMCs from different patients, chlamydial hsp60 en-hanced BrdU incorporation 763-fold (range 2.2- to 20-fold),with the largest increases elicited in cells with low baselineproliferation rates. The effect of chlamydial hsp60 (25mg/mL) was also abolished by heating to 100°C (Figure 1D).

C pneumoniaeand Chlamydial hsp60 Activatep44/p42 MAPK in Human VSMCsTo determine whether exposure toC pneumoniaeactivatesp44/p42 MAPK in VSMCs, cells were left untreated or were

Figure 1. C pneumoniae and chlamydial hsp60 stimulate humanVSMC proliferation, and their activity is heat-labile. VSMCsderived from human saphenous vein were exposed to C pneu-moniae or chlamydial hsp60 for 72 hours, with BrdU incorpora-tion determined at 48 to 72 hours. A and C, VSMCs wereexposed to varying amounts of C pneumoniae (n55 experi-ments) or chlamydial hsp60. B, Inactivation of C pneumoniae(33107 IFU/mL) by exposure to UV light or heat (56°C) only par-tially attenuated the mitogenic effect (n55 experiments). B andD, Heating to 100°C abolished the mitogenic activity of C pneu-moniae (33107 IFU/mL; n55 experiments) and chlamydial hsp60(25 mg/mL; n53 experiments). *P,0.02, significantly differentfrom VSMCs exposed to medium alone; 1P,0.02, significantlydifferent from VSMCs exposed to C pneumoniae or chlamydialhsp60 without prior heat inactivation.

246 Circulation Research August 3, 2001

by guest on August 20, 2014http://circres.ahajournals.org/Downloaded from

stimulated withC pneumoniae(33107 IFU/mL) or chlamyd-ial hsp60 (25mg/mL) for 5 to 30 minutes. Cell lysates wereanalyzed by Western blot using a polyclonal antibody spe-cific for the active, dually phosphorylated forms of p44/p42MAPK. Levels of phosphorylated p44 and p42 MAPK wereincreased in VSMCs after only a 5-minute exposure toCpneumoniae, were maximally increased at 10 minutes (11-fold and 3-fold, respectively), and remained elevated after 30minutes (Figure 2, left). The levels of phosphorylated p44 andp42 MAPK were likewise increased (3-fold and 2-fold,respectively) in VSMCs incubated 10 minutes with chlamyd-ial hsp60 (25mg/mL) (Figure 2, right). In contrast, totalp44/p42 MAPK levels were similar in all groups. Phosphor-ylated p44 and p42 MAPKs were similarly increased inVSMCs incubated for 10 minutes with 10% FCS (6-fold and2-fold, respectively; data not shown). These results indicatethat MAPK is rapidly activated after exposure toC pneu-moniaeor chlamydial hsp60.

MEK Inhibition Abolishes C pneumoniae–InducedVSMC ProliferationTwo specific inhibitors of MEK, PD98059 and U0126, wereused to assess the role of MEK-induced p44/p42 MAPKactivation in the mitogenic effect ofC pneumoniae. U0126(10 mmol/L) largely attenuates p44/p42 MAPK activationand is equally effective at inhibiting MEK1 and MEK2.29

Concentrations of PD98059 5-fold to 10-fold higher arerequired to produce similar inhibition of MEK activity,30 andMEK1 is preferentially inhibited. VSMCs were pretreated for30 minutes with vehicle, PD98059 (50mmol/L), or U0126(10 mmol/L) before exposure toC pneumoniaeor mediumalone. BrdU incorporation increased 8.6-fold in VSMCsexposed toC pneumoniae(33107/mL) in the absence of aMEK inhibitor (Figure 3). The mitogenic effect ofC pneu-moniae was inhibited by 90% in VSMCs pretreated withPD98059 and was abolished in VSMCs pretreated withU0126. U0126 likewise inhibited the mitogenic effect ofchlamydial hsp60 (25mg/mL), by 73% and 100%, in twoindependent experiments (data not shown).

Human VSMCs Express TLR4 mRNA and ProteinRT-PCR analysis indicated that saphenous vein SMCs ex-press TLR4 but not TLR2 mRNA. cDNA was prepared from

VSMCs derived from three different patients and amplifiedby primers specific for TLR2 or TLR4. Each VSMC cDNAsample yielded TLR4 PCR product of similar size to thatobtained from a human bone marrow cDNA library, whenamplified with either TLR4 primer pair (Figure 4A). In

Figure 2. C pneumoniae and chlamydial hsp60 activate p44/p42MAPK in saphenous vein SMCs. VSMCs were serum-deprivedfor 24 hours (DMEM/1% FCS) and left untreated or stimulatedwith C pneumoniae (33107 IFU/mL; left panel) or chlamydialhsp60 (25 mg/mL; right panel). Cell lysates (40 mg protein) wereanalyzed by Western blot analysis using an antibody specific foractive phospho-p44/p42 MAPK (top) or total MAPK (bottom). Figure 3. C pneumoniae–induced proliferation is abolished by

MEK inhibitors. VSMCs derived from human saphenous veinwere preincubated 30 minutes with or without PD98059 (50mmol/L) or U0126 (10 mmol/L) and then exposed to C pneu-moniae (33107 IFU/mL) for 72 hours, with BrdU incorporationdetermined during the last 24 hours (n54 experiments).*P,0.03, significantly different from VSMCs exposed to mediumalone; 1P,0.02, significantly different from VSMCs exposed toC pneumoniae without MEK inhibitors.

Figure 4. VSMCs express TLR4 mRNA and protein but do notexpress TLR2. A, RT-PCR analysis of saphenous vein SMCcDNA; 25 cycles of amplification with two distinct TLR4-specificprimer sets produced the expected 413-bp (left) and 299-bp(right) products. TLR4 product was absent when cDNA wasexcluded from the reaction (2). cDNA prepared from humanbone marrow (HBM) yielded product of similar size. B, VSMCswere incubated with mAb HTA125 (aTLR4), TL2.1 (aTLR2), orcontrol mouse IgG (10 mg/mL), followed by anti-mouse IgG-FITC, and analyzed by flow cytometry. Relative fluorescentintensity and relative cell number are shown on the x- andy-axes, respectively. The mean relative fluorescent intensities ofeach group are shown in parentheses. C, VSMCs were stainedwith goat anti-human TLR4 Ab, followed by anti-goat IgG TexasRed, and Hoechst 33342. Membrane-associated TLR4 wasvisualized by epifluorescence microscopy.

Sasu et al C pneumoniae–Induced VSMC Proliferation 247

by guest on August 20, 2014http://circres.ahajournals.org/Downloaded from

contrast, two sets of primers specific for TLR2 produced noPCR product with cDNAs from VSMCs, but yielded theexpected TLR2 PCR product when human bone marrowcDNA was amplified (data not shown).

Human VSMCs also expressed immunoreactive TLR4, butnot TLR2, as determined by immunostaining followed byflow cytometric analysis or fluorescence microscopy. Themean relative fluorescent intensities of VSMCs were 81, 97,and 253, when incubated in the presence of control IgG,TLR2 mAb, or TLR4 mAb, respectively (Figure 4B). Flowcytometric analysis of VSMCs derived from a differentpatient yielded similar results (not shown). Microscopicanalysis of VSMCs that were immunostained with polyclonalTLR4 antibodies revealed specific staining that appearedlocalized to the cell membrane (Figure 4C). In contrast, TLR2antigen was not detected in VSMCs by immunostaining withTLR2 mAb (data not shown).

RSLA Inhibits C pneumoniae– and Chlamydialhsp60–Induced VSMC ProliferationDiphosphoryl lipid A prepared fromR. sphaeroides(RSLA)is a competitive antagonist of LPS signaling31 and thus mayact as a competitive inhibitor of TLR4. Therefore, RSLA wasused to investigate the role of TLR4 in recognition ofC pneumoniaeand hsp60 by VSMCs. Addition of RSLA (1mg/mL) to VSMC cultures immediately before addition ofchlamydial hsp60 abolished the subsequent proliferative re-sponse to the chlamydial protein (Figure 5;P,0.01; n55experiments). The mitogenic effect of FCS was not altered byaddition of RSLA, indicating that the lipid A analogue does nothave nonspecific effects on VSMC proliferation. RSLA attenu-ated proliferation induced byC pneumoniaeby 46616% (Fig-ure 5; P,0.05, n57 experiments). These results support thehypothesis thatC pneumoniaeand chlamydial hsp60 stimulatehuman VSMC proliferation via activation of TLR4.

DiscussionIdentification of subendothelial macrophages in early athero-sclerotic lesions1 led to the hypothesis that atherosclerosis is

an inflammatory disease. More recently, correlative studieshave supported a possible association between atherosclerosisand chronic infection with the Gram-negative obligate intra-cellular bacteriumC pneumoniae.C pneumoniaeis a com-mon cause of respiratory tract infections, with most adultsexperiencing several infections over the course of a lifetime.32

C pneumoniaebacteria occur frequently in atheroscleroticlesions, but only rarely in normal arterial tissue.3–6 Evidencethat C pneumoniaecan induce proatherogenic effects onmononuclear phagocytes and vascular cells supports a rolefor the organism in the pathogenesis of vascular disease. Thepresent studies have four principal findings. First,C pneu-moniaeand chlamydial hsp60 are potent stimuli to humanVSMC proliferation. Second, the mitogenic effect ofC pneu-moniaeEBs is independent of chlamydial LPS and is likelymediated by a heat-labile chlamydial protein. Third, p44/p42MAPK activation is crucial toC pneumoniaeand chlamydialhsp60–induced VSMC proliferation. Finally, the mitogeniceffects of bothC pneumoniaeand hsp60 involve TLR4.

Several observations support the hypothesis that the mito-genic effect ofC pneumoniaeis mediated, at least in part, byrapid recognition of a heat-labile component of the organism.First, p44/p42 MAPK is activated 5 minutes after exposure tothe organism, and its activation is crucial to the subsequentincrease in proliferation. Second, the mitogenic effect ismimicked by chlamydial hsp60. Third, the mitogenic effectcan occur in the absence of active infection. The develop-mental cycle ofC pneumoniaeis biphasic.33 EBs released intothe extracellular space are metabolically inactive, yet infec-tious, and can be taken up by cells. Once inside the cell,chlamydia reside within inclusions in the cytoplasm, utilizingthe host cell’s metabolic machinery to develop into maturereticulate bodies, divide, and create new infectious EBs,which are released on host cell lysis. In the present study,C pneumoniaeEBs that were rendered noninfectious byexposure to UV light or heat (56°C) also stimulated prolifer-ation of VSMCs, although the response was partially attenu-ated. These results suggest that the mitogenic effect can occurin the absence of secondary events linked to active infection.However, they do not rule out the possibility that activeinfection may contribute to SMC activation in other settings,either in SMCs cultured in vitro7–9 or in atheroscleroticlesions in vivo.2–4

The mitogenic activity ofC pneumoniaewas heat-labile,ruling out a primary role of chlamydial LPS. Evidence for apossible role of chlamydial hsp60 was provided by the factthat chlamydial hsp60 replicated the mitogenic effect ofC pneumoniae. Recent studies suggest that chlamydial hsp60can activate cells in a manner similar to IL-1 or LPS.Chlamydial hsp60 induces synthesis of TNF-a and matrix-degrading metalloproteinases by mouse macrophages,14 ex-pression of adhesion molecules by human endothelial cells,15

and oxidation of LDL by human blood mononuclear cells.12

Chlamydial hsp60 is found in macrophage-rich areas ofhuman atherosclerotic plaque.14 Whether the presence ofchlamydial hsp60 is due to an active or persistentC pneu-moniaeinfection or to the persistence of the hsp60 antigen isnot known. It has been proposed thatC pneumoniaecanestablish persistent infections that could serve as a source of

Figure 5. RSLA partially attenuates C pneumoniae–induced pro-liferation and abolishes hsp60-induced proliferation. Saphenousvein SMCs were incubated for 72 hours in DMEM/1% FCS withor without RSLA (1 mg/mL) and with or without C pneumoniae(33107 IFU/mL), chlamydial hsp60 (25 mg/mL), or 10% FCS.BrdU incorporation was determined during the last 24 hours. A,n55 experiments. B, n57 experiments. *P,0.02, significantlydifferent from VSMCs incubated in medium alone; 1P,0.03,significantly different from VSMCs incubated withC pneumoniae or hsp60 and without RSLA.

248 Circulation Research August 3, 2001

by guest on August 20, 2014http://circres.ahajournals.org/Downloaded from

antigen to chronic inflammatory responses. In this regard, it isinteresting that the related organism,C trachomatis, contin-ues to synthesize hsp60 during a persistent infection, whereassynthesis of major outer membrane protein and LPS is greatlyreduced.34 The fact that chlamydial hsp60 is present inatherosclerotic lesions and is also a potent human VSMCmitogen supports a possible role of chlamydial hsp60 inintimal hyperplasia.

Recent evidence supports the hypothesis that hsp60 acti-vates cells via TLRs, the sensors of the innate immunesystem. TLRs are transmembrane proteins with an extracel-lular domain consisting of leucine-rich repeats involved inrecognition of microbial components. To date, nine TLRshave been identified in humans, but only a few of theirligands have been established.17 In one study, human hsp60stimulated TNF-a and nitric oxide production in mousemacrophages, whereas macrophages derived from C3H/HeJmice that express a nonfunctional form of TLR4 werenonresponsive to hsp60, supporting a role of TLR4 in humanhsp60–induced cell activation.35 In another study,36 expres-sion of CD14 conferred responsiveness to human hsp60 inU373 cells, an astrocytoma cell line that expresses TLR4 butnot TLR2,37 thereby implicating a CD14-dependent TLR inhsp60 signaling. Human VSMCs used in the present studyexpressed TLR4 mRNA and protein but did not expressTLR2 mRNA or protein, likewise suggesting a role of TLR4.Studies with RSLA, a lipid A analogue that antagonizesLPS-induced signaling,31 further supported a role of TLR4rather than TLR2 in the mitogenic effect of bothC pneu-moniaeand chlamydial hsp60. RSLA appears to be a specificinhibitor of TLR4, given that it markedly attenuates cellularactivation by LPS,31 and by lipoteichoic acid,38 an activecomponent of Gram-positive bacteria that can activateTLR4.39 In contrast, RSLA does not inhibit macrophageactivation by either heat-killedStaphylococcus aureusorextracts of Mycobacterium tuberculosis,31 organisms thatstimulate TLR2.40,41 However, it is possible that a distinctTLR is also expressed in VSMCs and is inhibited by RSLA.Further studies are required to establish whether RSLAattenuates signaling by other TLRs.

The cytosolic domains of TLRs have high homology withthe intracellular domain of the type I IL-1 receptor. Aspredicted by this homology, TLRs and the IL-1 receptorrecruit and activate common intermediate proteins, leading toactivation of the p44/p42 MAPK pathway and NF-kB. Afterbinding of their respective ligands, TLR4 and the IL-1receptor associate with an adapter protein, MyD88,42 leadingto recruitment of IL-1 receptor–associated kinases (IRAKs)to the receptor complex. IRAKs then interact with TRAF6, amember of the TNF receptor–associated factor family.TRAF6 activation, in turn, can lead to phosphorylation ofp44/p42 MAPK by undefined mechanisms.43 The p44/p42group of MAPKs is a central component of signaling viagrowth factors. Sequential activation of Ras and Raf activatesMEK. MEK then activates MAPK by dual phosphorylation ofkey threonine and tyrosine residues, and MAPK in turnphosphorylates serine and threonine residues on severaltranscription factors, including c-Myc, activator protein-1,NF–IL-6, activating transcription factor-2, and Elk-1, leading

ultimately to cell growth and differentiation.44 C pneumoniaeinduces rapid p44/p42 MAPK activation in human VSMCs,as previously reported in human umbilical vein endothelialcells.16 The mechanism of chlamydia-induced MAPK activa-tion in human VSMCs and endothelial cells is not known, butgiven the rapidity of the effect, it is likely to be mediated bydirect recognition of a chlamydial component, and in SMCsmay involve chlamydial hsp60-induced activation of TLR4.

The present studies have shown thatC pneumoniaeis apotent stimulus for human VSMC proliferation. Furthermore,the mitogenic effect is mediated by rapid TLR4-mediatedrecognition ofC pneumoniae, which may be due in part torecognition of chlamydial hsp60 and subsequent activation ofp44/p42 MAPK. The mitogenic effect ofC pneumoniaemaycontribute to intimal hyperplasia during the early stages ofatherogenesis.

AcknowledgmentsThis work was supported by NIH Grants HL47569 (to D.B.);GM50870 (to N.Q.); and DK50305, GM54060, and AI32725 (toD.T.G.). We thank Dr Michael Mendelsohn for critical review of themanuscript.

References1. Aqel NM, Ball RY, Waldmann H, Mitchinson MJ. Identification of

macrophages and smooth muscle cells in human atherosclerosis usingmonoclonal antibodies.J Pathol. 1985;146:197–204.

2. Shor A, Kuo CC, Patton D. Detection ofChlamydia pneumoniaeincoronary arterial fatty streaks and atheromatous plaques.S Afr Med J.1992;82:158–161.

3. Shor A, Phillips JI, Ong G, Thomas BJ, Taylor-Robinson D.Chlamydiapneumoniaein atheroma: consideration of criteria for causality.J ClinPathol. 1998;51:812–817.

4. Shor A, Phillips J.Chlamydia pneumoniaeand atherosclerosis.JAMA.1999;282:2071–2073.

5. Yamashita K, Ouchi K, Shirai M, Gondo T, Nakazawa T, Ito H. Distri-bution of Chlamydia pneumoniaeinfection in the atherosclerotic carotidartery.Stroke. 1998;29:773–778.

6. Kuo CC, Gown AM, Benditt EP, Grayston JT. Detection ofChlamydiapneumoniaein aortic lesions of atherosclerosis by immunocytochemicalstain.Arterioscler Thromb. 1993;13:1501–1504.

7. Godzik KL, O’Brien ER, Wang S, Kuo CC. In vitro susceptibility ofhuman vascular wall cells to infection withChlamydia pneumoniae.J Clin Microbiol. 1995;33:2411–2414.

8. Gaydos CA, Summersgill JT, Sahney NN, Ramirez JA, Quinn TC. Rep-lication of Chlamydia pneumoniaein vitro in human macrophages, en-dothelial cells, and aortic artery smooth muscle cells.Infect Immun.1996;64:1614–1620.

9. Maass M, Gieffers J, Solbach W. Atherogenetically relevant cells supportcontinuous growth ofChlamydia pneumoniae.Herz. 2000;25:68–72.

10. Jackson LA, Campbell LA, Schmidt RA, Kuo CC, Cappuccio AL, LeeMJ, Grayston JT. Specificity of detection ofChlamydia pneumoniaeincardiovascular atheroma: evaluation of the innocent bystander hypothesis.Am J Pathol. 1997;150:1785–1790.

11. Kalayoglu MV, Byrne GI. AChlamydia pneumoniaecomponent thatinduces macrophage foam cell formation is chlamydial lipopolysaccha-ride. Infect Immun. 1998;66:5067–5072.

12. Kalayoglu MV, Hoerneman BDL, Morrison SG, Morrison RP, Byrne GI.Cellular oxidation of low-density lipoprotein byChlamydia pneumoniae.J Infect Dis. 1999;180:780–790.

13. Netea MG, Selzman CH, Kullberg BJ, Galama JMD, Weinberg A,Stalenhoef AFH, Van der Meer JWM, Dinarello CA. Acellular com-ponents ofChlamydia pneumoniaestimulate cytokine production inhuman blood mononuclear cells.Eur J Immunol. 2000;30:541–549.

14. Kol A, Sukhova G, Lichtman AH, Libby P. Chlamydial heat shockprotein 60 localizes in human atheroma and regulates macrophage tumornecrosis factor-a and matrix metalloproteinase expression.Circulation.1998;98:300–307.

Sasu et al C pneumoniae–Induced VSMC Proliferation 249

by guest on August 20, 2014http://circres.ahajournals.org/Downloaded from

15. Kol A, Bourcier T, Lichtman AH, Libby P. Chlamydial and human heatshock protein 60s activate human vascular endothelium, smooth musclecells, and macrophages.J Clin Invest. 1999;103:571–577.

16. Krull M, Klucken AC, Wuppemann FN, Fuhrmann O, Magerl C,Seybold J, Hippenstiel S, Hegemann JH, Jantos CA, Suttorp N. Signaltransduction pathways activated in endothelial cells following infectionwith Chlamydia pneumoniae.J Immunol. 1999;162:4834–4841.

17. Means TK, Golenbock DT, Fenton MJ. The biology of Toll-likereceptors.Cytokine Growth Factor Rev. 2000;11:219–232.

18. Chow JC, Young DW, Golenbock DT, Christ WJ, Gusovsky F. Toll-likereceptor-4 mediates lipopolysaccharide-induced signal transduction.J Biol Chem. 1999;274:10689–10692.

19. Lien E, Means TK, Heine H, Yoshimura A, Kusumoto S, Fukase K,Fenton MJ, Oikawa M, Qureshi N, Monks B, Finberg RW, Ingalls RR,Golenbock DT. Toll-like receptor 4 imparts ligand-specific recognitionof bacterial lipopolysaccharide.J Clin Invest. 2000;105:497–504.

20. Faure E, Equils O, Sieling PA, Thomas L, Zhang FX, Kirschning CJ,Polentarutti N, Muzio M, Arditi M. Bacterial lipopolysaccharide acti-vates NF-kB through Toll-like receptor 4 (TLR-4) in cultured humandermal endothelial cells.J Biol Chem. 2000;275:11058–11063.

21. Yang H, Young DW, Gusovsky F, Chow JC. Cellular events mediatedby lipopolysaccharide-stimulated toll-like receptor 4.J Biol Chem.2000;275:20861–20866.

22. Ingalls RR, Rice PA, Qureshi N, Takayama K, Lin JS, Golenbock DT.The inflammatory cytokine response toChlamydia trachomatisinfection is endotoxin mediated.Infect Immun. 1995;63:3125–3130.

23. Caldwell HD, Kromhout J, Schachter J. Purification and partial charac-terization of the major outer membrane protein ofChlamydia tra-chomatis.Infect Immun. 1981;31:1161–1176.

24. Coombes BK, Mahony JB.Chlamydia pneumoniaeinfection of humanendothelial cells induces proliferation of smooth muscle cells via an endo-thelial cell-derived soluble factor(s).Infect Immun. 1999;67:2909–2915.

25. LaVerda D, Albanese LN, Ruther P, Morrison SG, Morrison RP, AultKA, Byrne GI. Seroreactivity toChlamydia trachomatishsp10 cor-relates with disease severity in women.Infect Immun. 2000;68:303–309.

26. Flo TH, Halaas O, Lien E, Ryan L, Teti G, Golenbock DT, Sundan A,Espevik T. Human toll-like receptor 2 mediates monocyte activation byListeria monocytogenes, but not by group B streptococci or lipopolysac-charide.J Immunol. 2000;164:2064–2069.

27. Vehmaan-Kreula P, Puolakkainen M, Sarvas M, Welgus HG, KovanenPT. Chlamydia pneumoniae proteins induce secretion of the 92-kDagelatinase by human monocyte-derived macrophages.ArteriosclerThromb Vasc Biol. 2001;21:E1–E8.

28. Rietschel ET. Chemistry of endotoxins. In: Rietschel ET, ed.Chemistryof Endotoxins. Vol 1. New York, NY: Elsevier; 1984:1–417.

29. Favata MF, Horiuchi KY, Manos EJ, Daulerio AJ, Stradley DA, FeeserWS, Van Dyk DE, Pitts WF, Earl RA, Hobbs F, Copeland RA, MagoldaRL, Scherle PA, Trzaskos JM. Identification of a novel inhibitor ofmitogen-activated protein kinase kinase.J Biol Chem. 1998;273:18623–18632.

30. Alessi DR, Cuenda A, Cohen P, Dudley DT, Saltiel AR. PD 098059 isa specific inhibitor of the activation of mitogen-activated protein kinasekinase in vitro and in vivo.J Biol Chem. 1995;270:27489–27494.

31. Golenbock DT, Hampton RY, Qureshi N, Takayama K, Raetz CRH.Lipid A-like molecules that antagonize the effects of endotoxins onhuman monocytes.J Biol Chem. 1991;266:19490–19498.

32. Kauppinen M, Saikku P. Pneumonia due toChlamydia pneumoniae:prevalence, clinical features, diagnosis, and treatment.Clin Infect Dis.1995;21:S244–S252.

33. Leinonen M. Pathogenetic mechanisms and epidemiology ofChlamydiapneumoniae.Eur Heart J. 1993;14:57–61.

34. Beatty WL, Byrne GI, Morrison RP. Morphologic and antigenic char-acterization of interferong-mediated persistentChlamydia trachomatisinfection in vitro.Proc Natl Acad Sci U S A. 1993;90:3998–4002.

35. Ohashi K, Burkart V, Flohe S, Kolb H. Cutting edge: heat shock protein60 is a putative endogenous ligand of the Toll-like receptor-4 complex.J Immunol. 2000;164:558–561.

36. Kol A, Lichtman AH, Finberg RW, Libby P, Kurt-Jones EA. Cuttingedge: heat shock protein (HSP) 60 activates the innate immune response:CD14 is an essential receptor for HSP60 activation of mononuclearcells.J Immunol. 2000;164:13–17.

37. Hirschfeld M, Kirschning CJ, Schwandner R, Wesche H, Weis JH,Wooten RM, Weis JJ. Cutting edge: inflammatory signaling byBorreliaburgdorferi lipoproteins is mediated by Toll-like receptor 2.J Immunol.1999;163:2382–2386.

38. Cleveland MG, Gorham JD, Murphy TL, Tuomanen E, Murphy KM. Lipo-teichoic acid preparations of Gram-positive bacteria induce interleukin-12through a CD14-dependent pathway.Infect Immun. 1996;64:1906–1912.

39. Takeuchi O, Hoshino K, Kawai T, Sanjo H, Takada H, Ogawa T,Takeda K, Akira S. Differential roles of TLR2 and TLR4 in recognitionof Gram-negative and Gram-positive bacterial cell wall components.Immunity. 1999;11:443–451.

40. Lien E, Sellati TJ, Yoshimura A, Flo TH, Rawadi G, Finberg RW,Carroll JD, Espevik T, Ingalls RR, Radolf JD, Golenbock DT. Toll-likereceptor 2 functions as a pattern recognition receptor for diverse bac-terial products.J Biol Chem. 1999;274:33419–33425.

41. Underhill DM, Ozinsky A, Smith KD, Aderem A. Toll-like receptor-2mediates mycobacteria-induced proinflammatory signaling in macro-phages.Proc Natl Acad Sci U S A. 1999;96:14459–14463.

42. Zhang FX, Kirschning CJ, Mancinelli R, Xu X-P, Jin Y, Faure E,Mantovani A, Rothe M, Muzio M, Ardit M. Bacterial lipopolysaccha-ride activates nuclear factor-kB through interleukin-1 signalingmediators in cultured human dermal endothelial cells and mononuclearphagocytes.J Biol Chem. 1999;274:7611–7614.

43. Kashiwada M, Shirakata Y, Inoue J-I, Nakano H, Okazaki K, OkumuraK, Yamamoto T, Nagaoka H, Takemori T. Tumor necrosis factorreceptor-associated factor 6 (TRAF6) stimulates extracellular signal-regulated kinase (ERK) activity in CD40 signaling along a Ras-independent pathway.J Exp Med. 1998;187:237–244.

44. Thomson S, Mahadevan LC, Clayton AL. MAP kinase-mediated sig-nalling to nucleosomes and immediate-early gene induction.Semin CellDev Biol. 1999;10:205–214.

250 Circulation Research August 3, 2001

by guest on August 20, 2014http://circres.ahajournals.org/Downloaded from