lable at ScienceDirect

Microbial Pathogenesis 69-70 (2014) 1e8

Contents lists avai

Microbial Pathogenesis

journal homepage: www.elsevier .com/locate/micpath

MAPKePPARa/g signal transduction pathways are involved inChlamydia pneumoniae-induced macrophage-derived foamcell formation

Bei Cheng a,**,1, Xiaohua Wu a,1, Shan Sun a, Qinqin Wu a, Chunli Mei b, Qiumei Xu a,Jianping Wu c, Ping He a,*

aDepartment of Gerontology, Union Hospital, Tongji Medical College of Huazhong University of Science and Technology, Wuhan 430022, ChinabDepartment of Rheumatology, Union Hospital, Tongji Medical College of Huazhong University of Science and Technology, Wuhan 430022, ChinacDepartment of Cardiology, Union Hospital, Tongji Medical College of Huazhong University of Science and Technology, Wuhan 430022, China

a r t i c l e i n f o

Article history:Received 22 January 2014Received in revised form9 March 2014Accepted 10 March 2014Available online 19 March 2014

Keywords:Chlamydia pneumoniaeFoam cellsPeroxisome proliferator-activated receptorsMitogen-activated protein kinaseAtherosclerosis

* Corresponding author. Department of GerontolMedical College of Huazhong University of Science aAvenue, Wuhan 430022, China. Tel./fax: þ86 (0) 27 8** Corresponding author.

E-mail address: doctorpeace74@gmail.com (P. He)1 These authors contributed equally to this study.

http://dx.doi.org/10.1016/j.micpath.2014.03.0010882-4010/� 2014 Elsevier Ltd. All rights reserved.

a b s t r a c t

Chlamydia pneumoniae (C. pneumoniae) is now widely accepted as an independent risk of atherosclerosisdevelopment. In this paper, our results showed that C. pneumoniae infection significantly increased thenumber of foam cells in LDL-treated THP-1 macrophages. C-Jun NH2 terminal kinase (JNK1/2) inhibitorSP600125 and extracellular signal-regulated kinase (ERK1/2) inhibitor PD98059 strongly inhibitedC. pneumoniae-induced accumulation of lipid droplet, whereas p38 inhibitor SB203580 had no obviouseffect on lipid accumulation. Furthermore, we found that C. pneumoniae not only stimulated the phos-phorylation of Mitogen-activated protein kinase (MAPK) including JNK1/2, ERK1/2 and p38 but alsodown-regulated the expression of peroxisome proliferator-activated receptors (PPARg and PPARa) atmRNA and protein levels. However, the phosphorylation of JNK1/2, ERK1/2 and p38 MAPK byC. pneumoniae was substantially reversed after PPARg agonist (rosiglitazone) or PPARa agonist (fenofi-brate) treatment while PPARg inhibitor (GW9662) and PPARa antagonist (MK886) enhancedC. pneumoniae-induced phosphorylation of JNK1/2, ERK1/2 and p38. In addition, we demonstrated thatC. pneumoniae-induced PPARg and PPARa down-regulation were significantly suppressed by JNK1/2inhibitor (SP600125) and ERK1/2 inhibitor (PD98059), but not p38 inhibitor (SB203580). These resultsfirst declare that MAPKePPARa/g reciprocal signal pathways are involved in C. pneumoniae, whichinduces foam cell formation, thus facilitating atherogenesis.

� 2014 Elsevier Ltd. All rights reserved.

1. Introduction

In recent decades, there is extensive evidence indicating thatatherosclerosis is a chronic inflammatory disorder and Chlamydiapneumoniae (C. pneumoniae) plays a key role in the development ofatherosclerosis and coronary artery disease [1]. C. pneumoniae is anobligate intracellular bacterial pathogen that has a biphasicelementary body and reticulate body life cycle and causes pneu-monia, bronchitis and other respiratory tract disease as a

ogy, Union Hospital, Tongjind Technology, 1277 Jie-Fang5351551.

.

widespread respiratory pathogen [2]. Saikku et al. [3] firstmentioned the role for C. pneumoniae in the pathogenesis ofatherosclerosis in 1988. Subsequently large amounts of epidemio-logical and clinical studies suggested that C. pneumoniae infectionaggravated the progression of atherosclerotic lesions [4].C. pneumoniae is able to disseminate via the circulation throughoutthe body within monocytes/macrophages and promote inflamma-tory atherogenous process [5]. Furthermore, in vitro study sug-gested that C. pneumoniae induced macrophage foam cellformationwhen cultured in the presence of low density lipoprotein(LDL), implicating that this organism is intimately linked withatherosclerosis progression [6]. However, the underlying mecha-nisms and pathways involved in C. pneumoniae induced foam cellformation have not yet been fully understood.

Peroxisome proliferator-activated receptors (PPARs) are aligand-activated nuclear hormone receptor superfamily existing in

B. Cheng et al. / Microbial Pathogenesis 69-70 (2014) 1e82

three isoforms: PPARa, PPARb/d and PPARg. As members of tran-scription factors, PPARg and PPARa can regulate key genes involvedin all stages of atherosclerosis, such as monocyte recruitments,macrophage activation, foam cell formation, vascular inflammationand thrombosis [7]. Recent studies revealed that activated PPARgand PPARa significantly suppressed the development of athero-sclerosis in LDL receptor-deficientmice [8,9]. In addition, our earlierstudy has demonstrated that ligand-activated PPARg inhibited theexpression of Acylcoenzyme A: cholesterol acyltransferase 1(ACAT1) in macrophages, which is a key and exclusive microsomalenzyme that esterifies free cholesterol with fatty acids to formcholesteryl ester [10]. Subsequently, we further confirmed theinvolvement of PPARg dependent pathway in the foam cell for-mation induced by C. pneumoniae, while C. pneumoniae up-regulated ACAT1 mRNA and protein levels via PPARg dependentpathway [11,12]. These studies indicate that PPARg expression bymacrophages has antiatherogenic effects via modulation of cellcholesterol trafficking.

Mitogen-activated protein kinase (MAPK) is a key component ofthe signaling pathways that transduct extracellular stimuli intocells and regulate expression of numerous genes [13]. It was re-ported that C. pneumoniae infection induced activation of MAPK[14,15]. Kitazawa et al. have demonstrated that chlamydophilalantigens-induced macrophage foam cell formation was mediatedby JNK [16]. Moreover, our previous works have demonstrated thatC. pneumoniae induced disequilibrium of intracellular cholesterolhomeostasis by up-regulating ACAT1 expression and down-regulating ATP binding cassette transporters (ABCA1/G1) expres-sion via JNK-PPARg signal transduction pathway [17].

Although a number of studies indicated the vital roles ofMAPK and PPARs in the modulation of foam cell formation, theeffect of MAPKePPARa/g crosstalk on C. pneumoniae-inducedfoam cell formation has not been fully investigated. Therefore, weexamined the role of MAPKePPARa/g crosstalk in THP-1 macro-phage-derived foam cell formation induced by C. pneumoniaeinfection.

2. Materials and methods

2.1. Ethics statement

This experiment involving fresh plasma was approved by nor-molipidemic volunteers and Wuhan Blood Centre (authorizations:2010-8) and conformed to the Declaration of Helsinki.

PPARa 50-GCCACCGATTTCATACAACA-30 (forward)50-TCTGGGATCTGCCTCATTCT-30 (reverse)

PPARg 50-CGTGGCCGCAGATTTGAA-30 (forward)50-CTTCCATTACGGAGAGATCCAC-30 (reverse)

b-actin 50-GTCCACCTTCCAGCAGATGT-30 (forward)50-CACCTTCACCGTTCCAGTTT-30 (reverse)

2.2. Cell lines and other reagents

Human monocytic THP-1 cells were purchased from AmericanType Culture Collection (ATCC, Rockville, MA, USA). Epithelial HEp-2 cells were obtained fromWuhan University, China. C. pneumoniaestrain AR-39 was purchased from ATCC. RPMI1640 powder andfetal bovine serum (FBS) were purchased from Hyclone (ThermoScientific, Waltham, MA, USA). Fluorescein isothiocyanate (FITC)-conjugated specific anti-chlamydial monoclonal antibody was ob-tained from Dako (Copenhagen, Denmark). SB203580, PD98059,SP60125, fenofibrate, rosiglitazone, MK886, GW9662, Phorbol 12-myristate 13-acetate (PMA) and cycloheximide were from Sigma(St. Louis, MO, USA). PrimeScript RT Reagent Kit and qRT-PCR usingSYBR� Premix Ex Taq�were from TAKARA Bio INC. PPARa antibodywas fromAbcam (Cambridge, UK). The following protein antibodieswere obtained from Cell Signaling Technology (Beverly, MA, USA):phospho-p44/42 MAPK (ERK1/2) (Thr202/Tyr204), phospho-SAPK/JNK (Thr183/Tyr185), Phospho-p38 MAPK (Thr180/Tyr182), p44/42MAPK (Erk1/2), SAPK/JNK, p38 MAPK and PPARg.

2.3. Low density lipoprotein (LDL) isolation and assessment

Amounts of plasma LDL (d ¼ 1.006e1.063 g/ml) was isolated bydensity-gradient ultracentrifugation as described [18]. The isolatedLDL was dialyzed in phosphate-buffered saline (PBS) for 24 h (h),and then was concentrated with polyoxyl 20,000. LDL was storedfor no longer than 2 weeks in the dark at 4 �C before use.

2.4. Propagation of C. pneumoniae and purification

C. pneumoniae strain AR39 was propagated in HEp-2 cells bycentrifugation-driven infection at 750 � g and 25 �C for 1 h fol-lowed by incubation at 37 �C and 5% CO2 for 1 h as described [19].Fresh RPMI1640 medium containing 2 mg/ml cycloheximide and10% heat-inactivated FBS was added. Infected HEp-2 cells wereharvested after 72 h incubation, and disrupted by freezing,thawing and ultrasonication. After centrifugation at 500 � g and4 �C for 30 min (min) in order to remove cell debris, the super-natants containing elementary bodies were harvested and storedat 70 �C. Chlamydial inclusion forming units (IFU) were deter-mined by counting chlamydial inclusions formed in HEp-2 cellswith FITC-conjugated specific anti-chlamydial monoclonalantibody.

2.5. Cell culture and infection of THP-1-derived macrophages

THP-1 cells were cultured in RPMI1640 medium containing 10%FBS at 37 �C in 5% CO2 and differentiated into macrophage by160 nmol/l PMA for 48 h. THP-1-derived macrophages were pre-treated with 50 mM fenofibrate, 20 mM rosiglitazone, 20 mMMK886,20 mMGW9662 for 2 h or 20 mM SP600125, 50 mM PD98059, 10 mMSB203580 for 1 h. Then macrophages were infected or uninfectedwith C. pneumoniae (1 � 106 IFU) in the presence of LDL (50 mg/l).

2.6. Oil red O staining

Cells were counterstained with hematoxylin and oil red O. Theintracellular lipid droplets were stained red, and cell nuclei werestained blue. Cells with a lipid droplet area no less than thewidth ofthe nucleus were designated oil red O positive. Foam cells ofmorphological criterion were defined as cells that have ten Oil redO-positive lipid droplets or more (�10) under a microscope [20].

2.7. Real-time quantitative RT-PCR

Total cellular RNA was isolated from treated cells using Trizol(TAKARA Bio INC). Real-time quantitative RT-PCR assay was per-formed using the two-step method following the manufacturer’sprocedure. Transcript levels of selected genes were measured bythe ABI Prism 7900 sequence detection system (Applied Bio-systems, Foster City, CA). The gene specific primers and probeswere purchased from Invitrogen life technologies (Shanghai,China). b-actin was used as an endogenous control. The sequencesof the PCR primers were as follows:

B. Cheng et al. / Microbial Pathogenesis 69-70 (2014) 1e8 3

2.8. Western blotting analysis

Protein samples (60 mg) were separated on 10% SDS-PAGE gels,transferred to polyvinylidene fluoride membrane (Invitrogen,Grand Island, NY, USA) and detected by enhanced chem-iluminescence method. For quantification, PPARa and PPARg pro-tein levels were normalized to the mouse polyclonal anti-b-actin(AntGene Biotech, Wuhan, China) protein level, respectively. Toinvestigate the effects of MAPK phosphorylation status in equalloading of samples, the previous phosphorylation antibody wasstripped from the membrane in a stripping buffer (Beyotime, Hai-men, P. R. China) for 5 min at 37 �C under permanent agitation.After being blocked, the membrane was incubated with a rabbitpolyclonal anti-p38, ERK1/2 or JNK1/2 antibody respectively. For

Fig. 1. MAPK pathway was involved in C. pneumoniae (C.pn) induced THP-1-derived foam cdoses of C.pn in the presence of LDL (50 mg/l) for 48 h. (a) Control, (b) 1 �105 IFU C.pn, (c) 4 �were infected with or without 1 � 106 IFU C.pn in the presence of LDL (50 mg/l) for 0e72 hwithout MAPK inhibitors for 1 h and then infected with or without C. pneumoniae (1 �106 IFUplus C.pn, (d) 20 mM SP600125, (e) 50 mM PD98059 plus C.pn, (f) 50 mM PD98059, (g) 10 mM SO staining at a magnification of 400�. (B), (D) and (F) Percent positive foam cells were quanrepresent the mean � SD of three independent experiments (n ¼ 3). *p < 0.05 compared wcompared with C. pneumoniae infection group.

quantification, the ratio of the intensity of the protein concerned(p-p38, p-ERK1/2 or p-JNK1/2) was presented in relation to themeasured levels of p38, ERK1/2 or JNK1/2 in THP-1-derivedmacrophages.

2.9. Statistical analysis

All experiments were repeated at least three times and theexperimental data were presented as the means � standard deri-vation (SD). SPSS 17.0 software was used for statistical analysis.Concentration and time response curves were performed using aone-way ANOVA, followed by analysis of correlation. Direct com-parisons between two groups were analyzed using the Student’s t-test. Statistical significance was accepted at p-value < 0.05.

ell formation. (A) THP-1-derived macrophages were infected with or without different105 IFU C.pn, (d) 5 � 105 IFU C.pn, (e) 1 �106 IFU C.pn. (C) THP-1-derived macrophages

. (a) 0 h, (b) 24 h, (c) 48 h, (d) 72 h. (E) THP-1-derived macrophages pretreated with or) in the presence of LDL (50 mg/l). (a) Control, (b) 1 �105 IFU C.pn, (c) 20 mM SP600125B203580 plus C.pn, (h) 10 mM SB203580. Black arrow indicates lipid droplets by Oil Redtified. On-screen counts were performed by a person blinded to sample identity. Dataith control, :p < 0.05 compared with 0 h C. pneumoniae infection group, 6p < 0.05

B. Cheng et al. / Microbial Pathogenesis 69-70 (2014) 1e84

3. Results

3.1. MAPK pathway was involved in C. pneumoniae induced THP-1-derived foam cell formation

As shown in Fig. 1A and B, LDL-treated THP-1-derived macro-phages were infected with or without different doses ofC. pneumoniae (1 �105 IFU to 1 �106 IFU) for 48 h. Compared withuninfected macrophages (control group, Fig. 1Aa), C. pneumoniaeinfection increased the accumulation of intracellular Oil Red O-stained lipid droplets especially induced foam cell formation athigher doses of C. pneumoniae infection (5 � 105 IFU to 1 �106 IFU)(Fig. 1A and B). In addition, Oil Red O-stained lipid droplets weresteadily increased by 1 � 106 IFU C. pneumoniae infection in 72 h(Fig. 1C and D). Overall, these results demonstrate thatC. pneumoniae introduces the accumulation of lipid droplets in adose-dependent manner (r ¼ 0.87, p < 0.05, Fig. 1B), and also in atime-dependent manner (r ¼ 0.89, p < 0.05, Fig. 1D).

Compared with C. pneumoniae-infected macrophages (Fig. 1Eb),in the presence of LDL, the inhibitors of JNK (SP600125) and ERK(PD98059) significantly decreased the accumulation intracellularlipid (p < 0.05, respectively. Fig. 1Ec, e and F). Furthermore, 20 mMSP600125 and 50 mM PD98059 alone had little effect on the accu-mulation of cellular lipid droplets, comparedwith control (p> 0.05,respectively. Fig. 1Ed, f and F). However, p38 inhibitor SB203580

Fig. 2. C. pneumoniae stimulated the phosphorylation of JNK1/2, ERK1/2 and p38 MAPK in L5, 15, 30, 45 or 60 min. Representative images for both the total and phosphorylated forms othe time course of phosphorylated-JNK1/2 (p-JNK1/2) and total JNK1/2 (JNK1/2) (A), phosphp38) and total p38 (p38) (E). Ratios of phosphorylated over total JNK1/2 (B), ERK1/2 (D) anactivity, control ¼ 1 arbitrary unit) and used as a measure of pathway activation. Data represe0 min C. pneumoniae infection group.

had little effect on the reduction of C. pneumoniae-induced intra-cellular lipid accumulation (Fig. 1Eg). These results suggest thatC. pneumoniae induces foam cell formation via JNK and ERK, but notp38.

3.2. C. pneumoniae mediated phosphorylation of MAPK in THP-1-derived macrophages

C. pneumoniae was added to the THP-1-derived macrophagesmedium for different periods of time in the presence of LDL.Western blot analysis demonstrated that C. pneumoniae inducedincrease of JNK1/2 (Fig. 2A and B), ERK1/2 (Fig. 2C and D) and p38MAPK (Fig. 2E and F) activity in THP-1-derived macrophages 15e45min. Phosphorylation level peaked at 30 min (ERK1/2) to 45 min(JNK1/2 and p38) and decreased to almost baseline at 60min. Theseresults indicate that C. pneumoniae induced the phosphorylation ofMAPK in LDL-treated THP-1-derived macrophages.

3.3. PPAR pathway took part in C. pneumoniae regulating MAPKphosphorylation

We examined the effects of PPARg agonist (rosiglitazone, 20 mM)and PPARg inhibitor (GW9662, 20 mM) on C. pneumoniae (1 � 106

IFU)-induced MAPK phosphorylation in THP-1 macrophages. Asshown in Fig. 3A and B, C. pneumoniae-induced JNK1/2, ERK1/2 and

DL-treated THP-1-derived macrophages. Cells were incubated with C. pneumoniae for 0,f JNK1/2, ERK1/2 and p38 proteins were collected. Western blot analysis demonstratedorylated-ERK1/2 (p-ERK1/2) and total ERK1/2 (ERK1/2) (C) and phosphorylated-p38 (p-d p38 (F) were assessed from semiquantitative analysis (expressed as relative kinasent the mean � SD of three independent experiments (n ¼ 3). :p < 0.05 compared with

B. Cheng et al. / Microbial Pathogenesis 69-70 (2014) 1e8 5

p38 MAPK phosphorylation were completely eliminated whenpretreated with rosiglitazone (p < 0.05, respectively). However,pretreatment with GW9662 significantly promoted the phosphor-ylation level of JNK1/2, ERK1/2 and p38MAPK to 1.88-fold,1.61-foldand 2.17-fold by C. pneumoniae infection compared with controls(p < 0.05, respectively, Fig. 3C and D). These results reveal thatC. pneumoniae-induced MAPK phosphorylation is mediated byPPARg in THP-1 macrophages.

Fig. 3. Effects of PPARg/a agonists and inhibitors on C. pneumoniae-induced MAPK phosph(20 mM), GW9662 (20 mM), fenofibrate (50 mM) or MK886 (20 mM) and then infected with oblotting estimated JNK1/2, ERK1/2 and p38 MAPK phosphorylation levels relative to the totaor MK886 (G), respectively. The bar graphs represented the ratios of phosphorylated overfenofibrate (F) or MK886 (H) and estimated by semiquantitative analysis (expressed as relativof three independent experiments (n ¼ 3). *p < 0.05 compared with control, 6p < 0.05 co

In similar experiments, compared with C. pneumoniae infectiongroup, PPARa agonist fenofibrate (50 mM) reversed C. pneumoniae-induced activation of JNK1/2, ERK1/2 and p38 MAPK (p < 0.05,respectively, Fig. 3E and F). In contrast, PPARa inhibitor MK886(20 mM) promoted C. pneumoniae-induced JNK1/2, ERK1/2 and p38MAPK phosphorylation level up to 2.57-fold, 1.79-fold and 2.12-foldcompared with controls (p < 0.05, respectively, Fig. 3G and H).These results suggest that PPARa modulates the activation of the

orylation. THP-1-derived macrophages were pretreated with or without rosiglitazoner without C. pneumoniae (C.pn, 1 � 106 IFU) in the presence of LDL (50 mg/l). Westernl protein in the macrophages treated with rosiglitazone (A), GW9662 (C), fenofibrate (E)total JNK1/2, ERK1/2 and p38 in cells pretreated with rosiglitazone (B), GW9662 (D),e kinase activity, control¼ 1 arbitrary unit), respectively. Data represent the mean � SDmpared with C. pneumoniae infection group.

B. Cheng et al. / Microbial Pathogenesis 69-70 (2014) 1e86

MAPK signaling pathway in C. pneumoniae-stimulated THP-1macrophages.

3.4. MAPK pathway participated in C. pneumoniae-induced down-regulation of PPARa/g mRNA and protein expression

As shown in Fig. 4, compared with control, the expression ofPPARg and PPARa mRNA was reduced 60% and 70% (compared tocontrol, p < 0.05, respectively) by C. pneumoniae infection. So didthe expression of PPARg and PPARa protein (p < 0.05). JNK1/2 in-hibitor (SP600125, 20 mM) and ERK1/2 inhibitor (PD98059, 50 mM)reversed the down-regulation of PPARg and PPARa gene expressioncompared with controls (p < 0.05, respectively, Fig. 4AeF). How-ever, pretreatment of THP-1 macrophages with p38 inhibitor

Fig. 4. Effects of MAPK inhibitors on C. pneumoniae-induced PPARg/a mRNA and protein expSP600125, 50 mM PD98059 or 10 mM SB203580) and then infected with or without C. pneumreaction estimating PPARg and PPARa mRNA expression relative to b-actin in cells pretreaestimating gene protein levels relative to b-actin in cells pretreated with SP600125 (PPARg-BResults are expressed as fold induction of control (control ¼ 1 arbitrary unit). Data representcontrol, 6p < 0.05 compared with C. pneumoniae infection group.

(SB302580,10 mM) had no obvious effect on inhibition of PPARg andPPARa gene expression by C. pneumoniae infection (p > 0.05,Fig. 4GeI). Collectively, these results indicate that activation ofJNK1/2 and ERK1/2 is involved in C. pneumoniae-induced down-regulation of PPARa/g, while p38 pathway may not be involved.

4. Discussion

Various studies have demonstrated that C. pneumoniae infectionwas involved in the development and progression of atheroscle-rosis, by triggering inflammatory reaction in monocytes or mac-rophages and especially by promoting macrophage-derived foamcell formation [5,11,12]. However, the underlying biochemicalmechanisms of C. pneumoniaemodulated foam cell formation have

ression. THP-1 macrophages were pretreated with or without MAPK inhibitors (20 mMoniae (C.pn, 1 � 106 IFU) in the presence of LDL (50 mg/l). Quantitative real-time PCR

ted with SP600125 (A), PD98059 (D) or SB203580 (G), respectively. Western blotting, PPARa-C), PD98059 (PPARg-E, PPARa-F) or SB203580 (PPARg-H, PPARa-I), respectively.the mean � SD from three independent experiments (n ¼ 3). *p < 0.05 compared with

B. Cheng et al. / Microbial Pathogenesis 69-70 (2014) 1e8 7

not been fully elucidated. In the current study, we demonstratedthat C. pneumoniae infection significantly increased the number offoam cells. C. pneumoniae could not only up-regulate the phos-phorylation of MAPK but also down-regulate the expression ofPPARa/g. Moreover, we found that there was a close correlationbetween MAPK and PPARs signal pathways.

MAPKs and PPARa/g signal pathways are key elements inintracellular signal transduction. MAPKs are considered key regu-lators of pro-inflammatory cytokine production and play importantroles in the signaling cascades downstream of immune cells [13,21].On the contrary, PPARa and PPARg are known to inhibit the acti-vation of inflammatory factor and negatively interfere withpro-inflammatory [22]. Noteworthy is both PPARa/g and MAPKpathways can regulate intracellular lipid metabolism [7,23]. Ourstudy revealed that C. pneumoniae significantly stimulated thephosphorylation of MAPK including JNK1/2, ERK1/2 and p38,whereas the expression of PPARg and PPARa was strongly sup-pressed after infection. It indicates that MAPK and PPARa/g signalpathways may be involved in the C. pneumoniae-induced pertur-bation of cholesterol homeostasis.

PPARs are capable of inhibiting the phosphorylation and sub-sequent activation of certain members of the MAPK signalingcascade. Recent studies confirmed that rosiglitazone, a PPARgspecial antagonist, attenuated phosphorylation of ERK1/2 and JNKin human aortic endothelial cells (HAECs) [24]. Appel et al. [25] alsofound that activation of PPARg by its natural ligand 15d-PGJ2inhibited the phosphorylation of ERK1 and p38 kinase signalpathways. In addition, PPARa agonist fenofibrate could decreasephosphorylation of JNK and p38 MAPK in vivo and ERK in vitro[26,27]. In our study, pretreatment with rosiglitazone suppressedthe stimulation of JNK1/2, ERK1/2 and p38 phosphorylation byC. pneumoniae infection, while fenofibrate showed a similar effectof rosiglitazone. Additionally, both a specific PPARg inhibitorGW9662 and a PPARa inhibitor MK886 significantly increasedC. pneumoniae-induced phosphorylation of JNK1/2, ERK1/2 andp38. Thus, our results demonstrate that PPARa/g signal pathwaysplay an important role in C. pneumoniae-induced MAPKphosphorylation.

In vitro studies have shown that PPARg and PPARa were thepotential downstream target of the MAPK signaling [28,29]. JNKphosphorylation negatively regulated PPARg transcriptional activ-ity [30]. ERK1/2 pathway inhibited PPARa transcriptional activitythrough a mechanism involving direct binding of the ERK1/2 up-stream kinase MEK1 to PPARa [29]. Our present study showed thatJNK1/2 inhibitor SP600125 and ERK1/2 inhibitor PD98059 reversedthe down-regulation of PPARg and PPARa induced byC. pneumoniae. These data reveal that the phosphorylation of JNK1/2 and/or ERK1/2 take part in C. pneumoniae down-regulating PPARgand PPARa gene expression.

Importantly, we found that SP600125 and PD98059 stronglyinhibited the accumulation of intracellular lipid droplets byC. pneumoniae infection. More notably, the previous studies fromour laboratory demonstrated that rosiglitazone and fenofibratedrastically decreased C. pneumoniae-induced accumulation ofintracellular lipid droplets [11]. Therefore, it suggests that MAPKePPARa/g crosstalk signal pathways are involved in the underlyingmechanism regulating C. pneumoniae-induced foam cell formation.

Results in this study showed that the three pathways of MAPKwere activated in C. pneumoniae-induced intracellular lipid accu-mulation. The peak of phosphorylation appeared at 30 min in ERKand at 45 min in JNK and p38. The different kinetics of phosphor-ylation of the pathways may lead to different roles they played inC. pneumoniae-induced foam cell formation. We further demon-strated that JNK inhibitor SP600125 and ERK inhibitor PD98059strongly inhibited the accumulation of intracellular lipid droplets

induced by C. pneumoniae infection, while p38 inhibitor SB203580showed no obvious effect on it. Similar results have been reportedelsewhere. Kitazawa et al. found that treatment with SP600125inhibited chlamydophilal antigens-induced foam cell formation,although PD98059 or SB203580 had little effect on it in humanmonocytes. And treatment with SP600125 and SB203580 inhibitedchlamydophilal antigens-induced foam cell formation, whilePD98059 enhanced foam cell formation in RAW cells [16]. Yet,different pathway acted differently in regulating PPARa/g expres-sion. Our results showed that SB203580 performed little effect onregulating PPARa/g expression while the other two inhibitorssignificantly prevented the down-regulating of PPARa/g expressionby C. pneumoniae. Reports also found that pretreatment of THP-1cells with p38 inhibitor SB203580 did not suppress PPARg inacti-vation induced by ox-LDL [31] and glucose-regulated PPARaexpression did not involve p38 MAPK, except JNK played a vital rolein it [32]. Taken together, MAPK phosphorylation participated inC. pneumoniae-induced foam cell formation and down-regulationof PPARa/g expression and ERK and JNK pathways play majorroles in it. But what part of p38 phosphorylation might take in it isstill need to be elucidated.

In conclusion, the current investigation demonstrates thatC. pneumoniae-induced foam cell formation via MAPKePPARa/gcrosstalk signal pathways, which may provide a new insight for themechanism of atherosclerosis initiated by C. pneumoniae infection.Understanding the signaling pathway involved in the process willhopefully shed light on possible target choices of developing po-tential interventions in atherosclerosis.

Acknowledgments

This study was supported by grants of the National NaturalScience Foundation of China (Grant No. 30900599) and theFundamental Research Funds for the Central Universities of China(Grant No. 2012QN228).

References

[1] Xu Q. Infections, heat shock proteins, and atherosclerosis. Curr Opin Cardiol2003;18:245e52.

[2] Wolf K, Fischer E, Hackstadt T. Ultrastructural analysis of developmental eventsin Chlamydia pneumoniae-infected cells. Infect Immun 2000;68:2379e85.

[3] Saikku P, Leinonen M, Mattila K, Ekman MR, Nieminen MS, Makela PH, et al.Serological evidence of an association of a novel Chlamydia, TWAR, withchronic coronary heart disease and acute myocardial infarction. Lancet1988;2:983e6.

[4] Liu C, Waters DD. Chlamydia pneumoniae and atherosclerosis: from Kochpostulates to clinical trials. Prog Cardiovasc Dis 2005;47:230e9.

[5] Sessa R, Nicoletti M, Di Pietro M, Schiavoni G, Santino I, Zagaglia C, et al.Chlamydia pneumoniae and atherosclerosis: current state and future pro-spectives. Int J Immunopathol Pharmacol 2009;22:9e14.

[6] Kalayoglu MV, Byrne GI. Induction of macrophage foam cell formation byChlamydia pneumoniae. J Infect Dis 1998;177:725e9.

[7] Barbier O, Torra IP, Duguay Y, Blanquart C, Fruchart JC, Glineur C, et al. Pleio-tropic actions of peroxisome proliferator-activated receptors in lipid meta-bolism and atherosclerosis. Arterioscler Thromb Vasc Biol 2002;22:717e26.

[8] Babaev VR, Ishiguro H, Ding L, Yancey PG, Dove DE, Kovacs WJ, et al.Macrophage expression of peroxisome proliferator-activated receptor-alphareduces atherosclerosis in low-density lipoprotein receptor-deficient mice.Circulation 2007;116:1404e12.

[9] Li AC, Brown KK, Silvestre MJ, Willson TM, Palinski W, Glass CK. Peroxisomeproliferator-activated receptor gamma ligands inhibit development ofatherosclerosis in LDL receptor-deficient mice. J Clin Invest 2000;106:523e31.

[10] Bai Z, Cheng B, Yu Q, Li C, He P, Mao X. Effects of leptin on expression of acyl-coenzymea: cholesterol acyltransferases-1 in cultured human monocyte-macrophages. J Huazhong Univ Sci Technol Med Sci 2004;24:563e5. 90.

[11] He P, Mei C, Cheng B, Liu W, Wang Y, Wan J. Chlamydia pneumoniae inducesmacrophage-derived foam cell formation by up-regulating acyl-coenzyme A:cholesterol acyltransferase 1. Microbes Infect 2009;11:157e63.

[12] Mei CL, He P, Cheng B, Liu W, Wang YF, Wan JJ. Chlamydia pneumoniae inducesmacrophage-derived foam cell formation via PPAR alpha and PPAR gamma-dependent pathways. Cell Biol Int 2009;33:301e8.

B. Cheng et al. / Microbial Pathogenesis 69-70 (2014) 1e88

[13] Johnson GL, Lapadat R. Mitogen-activated protein kinase pathways mediatedby ERK, JNK, and p38 protein kinases. Science 2002;298:1911e2.

[14] Krull M, Kramp J, Petrov T, Klucken AC, Hocke AC, Walter C, et al. Differencesin cell activation by Chlamydophila pneumoniae and Chlamydia trachomatisinfection in human endothelial cells. Infect Immun 2004;72:6615e21.

[15] Kim TB, Moon KA, Lee KY, Park CS, Bae YJ, Moon HB, et al. Chlamydophilapneumoniae triggers release of CCL20 and vascular endothelial growth factorfrom human bronchial epithelial cells through enhanced intracellular oxida-tive stress and MAPK activation. J Clin Immunol 2009;29:629e36.

[16] Kitazawa T, Fukushima A, Okugawa S, Yanagimoto S, Tsukada K, Tatsuno K,et al. Chlamydophilal antigens induce foam cell formation via c-Jun NH2-terminal kinase. Microbes Infect 2007;9:1410e4.

[17] Liu W, He P, Cheng B, Mei CL, Wang YF, Wan JJ. Chlamydia pneumoniae dis-turbs cholesterol homeostasis in human THP-1 macrophages via JNK-PPARgamma dependent signal transduction pathways. Microbes Infect2010;12:1226e35.

[18] Wang CB, Feng YM, Zong YQ, Deng YZ, Feng ZC. Rapid isolation of largeamount of plasma VLDL and LDL by a two step ultracentrifugation. J TongjiMed Univ 1995;15:198e200.

[19] Roblin PM, Dumornay W, Hammerschlag MR. Use of HEp-2 cells for improvedisolation and passage of Chlamydia pneumoniae. J Clin Microbiol 1992;30:1968e71.

[20] Schaffner T, Taylor K, Bartucci EJ, Fischer-Dzoga K, Beeson JH, Glagov S, et al.Arterial foam cells with distinctive immunomorphologic and histochemicalfeatures of macrophages. Am J Pathol 1980;100:57e80.

[21] Harnett MM, Katz E, Ford CA. Differential signalling during B-cell maturation.Immunol Lett 2005;98:33e44.

[22] Moraes LA, Piqueras L, Bishop-Bailey D. Peroxisome proliferator-activatedreceptors and inflammation. Pharmacol Ther 2006;110:371e85.

[23] Eder K, Guan H, Sung HY, Francis SE, Crossman DC, Kiss-Toth E. LDL uptake bymonocytes in response to inflammation is MAPK dependent but independentof tribbles protein expression. Immunol Lett 2008;116:178e83.

[24] Han CJ, Liu JT, Li M, Cui M, Pang XM, Mao JJ, et al. Rosiglitazone inhibitsangiotensin II-induced C-reactive protein production in human aortic endo-thelial cells through regulating AT(1)-ROS-MAPK signal pathway. Inflamm Res2012;61:1031e7.

[25] Appel S, Mirakaj V, Bringmann A, Weck MM, Grunebach F, Brossart P.PPAR-gamma agonists inhibit toll-like receptor-mediated activation ofdendritic cells via the MAP kinase and NF-kappaB pathways. Blood2005;106:3888e94.

[26] Hou X, Shen YH, Li C, Wang F, Zhang C, Bu P, et al. PPARalpha agonist feno-fibrate protects the kidney from hypertensive injury in spontaneously hy-pertensive rats via inhibition of oxidative stress and MAPK activity. BiochemBiophys Res Commun 2010;394:653e9.

[27] Liang H, Kowalczyk P, Junco JJ, Klug-De Santiago HL, Malik G, Wei SJ, et al.Differential effects on lung cancer cell proliferation by agonists of glucocor-ticoid and PPARalpha receptors. Mol Carcinog; 2013 Apr 26. http://dx.doi.org/10.1002/mc.22029. [Epub ahead of print].

[28] Adams M, Reginato MJ, Shao D, Lazar MA, Chatterjee VK. Transcriptionalactivation by peroxisome proliferator-activated receptor gamma is inhibitedby phosphorylation at a consensus mitogen-activated protein kinase site.J Biol Chem 1997;272:5128e32.

[29] el Azzouzi H, Leptidis S, Bourajjaj M, van Bilsen M, da Costa Martins PA, DeWindt LJ. MEK1 inhibits cardiac PPARalpha activity by direct interaction andprevents its nuclear localization. PLoS ONE 2012;7:e36799.

[30] Camp HS, Tafuri SR, Leff T. c-Jun N-terminal kinase phosphorylates peroxi-some proliferator-activated receptor-gamma1 and negatively regulates itstranscriptional activity. Endocrinology 1999;140:392e7.

[31] Yin R, Dong YG, Li HL. PPARgamma phosphorylation mediated by JNKMAPK: apotential role in macrophage-derived foam cell formation. Acta Pharmacol Sin2006;27:1146e52.

[32] Joly E, Roduit R, Peyot ML, Habinowski SA, Ruderman NB, Witters LA, et al.Glucose represses PPARalpha gene expression via AMP-activated protein ki-nase but not via p38 mitogen-activated protein kinase in the pancreatic beta-cell. J Diabetes 2009;1:263e72.