of July 2, 2018.This information is current as

by Targeting ERK1/2 SignalingSuppresses Host Innate Immune Responses

Mce3EMycobacterium tuberculosis

Xiao-Bo Qiu and Cui Hua LiuJie Li, Qi-Yao Chai, Yong Zhang, Bing-Xi Li, Jing Wang,

http://www.jimmunol.org/content/194/8/3756doi: 10.4049/jimmunol.1402679March 2015;

2015; 194:3756-3767; Prepublished online 16J Immunol 

MaterialSupplementary

9.DCSupplementalhttp://www.jimmunol.org/content/suppl/2015/03/14/jimmunol.140267

Referenceshttp://www.jimmunol.org/content/194/8/3756.full#ref-list-1

, 21 of which you can access for free at: cites 50 articlesThis article

        average*  

4 weeks from acceptance to publicationFast Publication! •    

Every submission reviewed by practicing scientistsNo Triage! •    

from submission to initial decisionRapid Reviews! 30 days* •    

Submit online. ?The JIWhy

Subscriptionhttp://jimmunol.org/subscription

is online at: The Journal of ImmunologyInformation about subscribing to

Permissionshttp://www.aai.org/About/Publications/JI/copyright.htmlSubmit copyright permission requests at:

Email Alertshttp://jimmunol.org/alertsReceive free email-alerts when new articles cite this article. Sign up at:

Print ISSN: 0022-1767 Online ISSN: 1550-6606. Immunologists, Inc. All rights reserved.Copyright © 2015 by The American Association of1451 Rockville Pike, Suite 650, Rockville, MD 20852The American Association of Immunologists, Inc.,

is published twice each month byThe Journal of Immunology

by guest on July 2, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

by guest on July 2, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

The Journal of Immunology

Mycobacterium tuberculosis Mce3E Suppresses Host InnateImmune Responses by Targeting ERK1/2 Signaling

Jie Li,* Qi-Yao Chai,* Yong Zhang,* Bing-Xi Li,* Jing Wang,* Xiao-Bo Qiu,† and

Cui Hua Liu*

Crucial to the pathogenesis of the tuberculosis (TB)-causing pathogen Mycobacterium tuberculosis is its ability to subvert host

immune defenses to promote its intracellular survival. The mammalian cell entry protein 3E (Mce3E), located in the region of

difference 15 of the M. tuberculosis genome and absent in Mycobacterium bovis bacillus Calmette-Guerin, has an essential role in

facilitating the internalization of mammalian cells by mycobacteria. However, relatively little is known about the role of Mce3E in

modulation of host innate immune responses. In this study, we demonstrate that Mce3E inhibits the activation of the ERK1/2

signaling pathway, leading to the suppression of Tnf and Il6 expression, and the promotion of mycobacterial survival within

macrophages. Mce3E interacts and colocalizes with ERK1/2 at the endoplasmic reticulum in a DEF motif (an ERK-docking

motif)–dependent manner, relocates ERK1/2 from cytoplasm to the endoplasmic reticulum, and finally reduces the association of

ERK1/2 with MEK1 and blocks the nuclear translocation of phospho-ERK1/2. A DEF motif mutant form of Mce3E (F294A) loses

its ability to suppress Tnf and Il6 expression and to promote intracellular survival of mycobacteria. Inhibition of the ERK1/2

pathway in macrophages using U0126, a specific inhibitor of the ERK pathway, also leads to the suppressed Tnf and Il6 expression

and the enhanced intracellular survival of mycobacteria. Taken together, these results suggest that M. tuberculosis Mce3E exploits

the ERK1/2 signaling pathway to suppress host innate immune responses, providing a potential Mce3E–ERK1/2 interface–based

drug target against M. tuberculosis. The Journal of Immunology, 2015, 194: 3756–3767.

Mycobacterium tuberculosis remains one of the majorglobal health threats, resulting in an annual casualty of∼2 million people worldwide (1). Although nearly one-

third of the human population is infected with M. tuberculosis,only ∼10% of the infected individuals develop active diseaseduring their lifetime. The pathogenesis of M. tuberculosis islargely because it can escape host immune defense system andthus survive within the hostile environment of human macro-phages (2, 3). The prolonged coevolution of M. tuberculosis withits hosts has led to multiple survival strategies to interfere witha wide range of host cellular processes, such as production ofcytokines, phagolysosome biogenesis, and modulation of macro-phage survival (3–6). The majority of the currently available drugsare only partially effective due to the impermeable nature of the

mycobacterial cell wall and the propensity of M. tuberculosis todevelop drug resistance (7). Thus, it is urgent to unravel the

molecular details underlying M. tuberculosis–macrophage inter-

actions, which could be useful to identify novel targets for anti-

tuberculosis (TB) drug development.Innate immunity constitutes the first line of defense against

pathogen infection (8). Two major immune signaling pathways,

including NF-kB and MAPK, mediate regulation of innate im-

mune responses through controlling synthesis of various cytokines

such as TNF and IL-6, etc. (9–11). Meanwhile, increasing evi-

dence suggests that bacterial pathogens target and manipulate

those signaling pathways for their benefits (12). For example,

Yersinia enterocolitica was indicated to suppress TNF production

by inhibiting ERK1/2, p38, and JNK activities (13). Macrophages

infected with Mycobacterium avium showed decreased MAPK

activation compared with the cells infected with nonpathogenic

mycobacteria (14). The NF-kB and MAPK signaling pathways

can also regulate production of a number of cytokines by the

macrophages infected with M. tuberculosis (15). But the molec-

ular mechanisms by which specific mycobacterial effectors mod-

ulate the host innate immune signaling pathways remain largely

unclear.The genome of M. tuberculosis H37Rv harbors four homolo-

gous copies of the gene cluster termed mce operons (mce1–4),

which includes two yrbE and six mce genes (mceA–F) (16, 17).

Mammalian cell entry (Mce) proteins were initially characterized

as invasion-like proteins with putative export signal sequences at

the N-terminal end (16, 17). Previous studies suggest that latex

beads coated with Mce1A, Mce3A, and Mce3E are internalized

by nonphagocytic HeLa cells (18, 19). Additional studies have

demonstrated that the M. tuberculosis strains with mce operon

disruption were attenuated in mice (20, 21). Among the four mce

operons, the mce3 operon is located in the region of difference 15

of theM. tuberculosis genome, which is of special interest because

*CAS Key Laboratory of Pathogenic Microbiology and Immunology, Instituteof Microbiology, Chinese Academy of Sciences, Beijing 100101, China; and†Department of Cell Biology, Ministry of Education Key Laboratory of Cell Prolif-eration and Regulation Biology, College of Life Sciences, Beijing Normal University,Beijing 100875, China

Received for publication October 21, 2014. Accepted for publication February 12,2015.

This work was supported by National Basic Research Programs of China Grants2012CB518700 and 2014CB744400 (to C.H.L.), National Natural Science Founda-tion of China Grants 81371769 (to C.H.L.) and 91319303 (to X.-B.Q.), Ministry ofHealth and the Ministry of Science and Technology, China Grant 2013ZX10003006(to C.H.L.), and Chinese Academy of Sciences Grant KJZD-EW-L02 (to J.W. andC.H.L.).

Address correspondence and reprint requests to Prof. Cui Hua Liu, CAS Key Lab-oratory of Pathogenic Microbiology and Immunology, Institute of Microbiology,Chinese Academy of Sciences, Beijing 100101, China. Email address: liucuihua@im.ac.cn

The online version of this article contains supplemental material.

Abbreviations used in this article: BCG, bacillus Calmette–Guerin; EGF, epidermalgrowth factor; ER, endoplasmic reticulum; MBP, maltose-binding protein; Mce,mammalian cell entry; MOI, multiplicity of infection; TB, tuberculosis.

Copyright� 2015 by TheAmericanAssociation of Immunologists, Inc. 0022-1767/15/$25.00

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1402679

by guest on July 2, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

it is absent in the vaccine strain of Mycobacterium bovis bacillusCalmette–Guerin (BCG) (22). However, relatively little is knownabout the role of the mce3 operon in modulation of host innateimmune responses.Previous studies have shown that the six Mce proteins encoded

by mce3 operon, especially Mce3A, Mce3D, and Mce3E, areexpressed in M. tuberculosis during infection in humans and areimmunogenic (23). But among all mce3 operon proteins, onlyregion of difference 1504 (Mce3A) was demonstrated to be able toinduce a strong Th1 bias in healthy subjects and a weak Th1 biasin TB patients, whereas other proteins of mce3 operon inducedonly moderate to weak Th1 biases in both TB patients and healthysubjects (24), suggesting that other proteins among mce3 operonmight have potential immune-modulating effects. Despite thepotential function of Mce proteins on intracellular survival ofM. tuberculosis, little progress has been made to define the spe-cific functional mechanisms of each Mce protein. In this study, wefocus on the Mce3E, aiming at elucidating its potential role inmodulation of host innate immune responses. We demonstrate thatMce3E downregulates cytokine expression and promotes survivalof mycobacteria in macrophages through inhibiting activation ofthe ERK1/2 signaling pathway. We further show that Mce3E couldbe secreted from mycobacteria phagocytized by macrophages andthen translocate into cytosol to localize at the endoplasmic retic-ulum (ER). Furthermore, Mce3E interacts with ERK1/2 in a DEFmotif (an ERK-docking motif)–dependent manner, traps ERK1/2into ER, and blocks the interaction of ERK1/2 with MEK1 andthe nuclear translocation of p-ERK1/2, resulting in inhibition ofthe ERK1/2 signaling pathway. Our findings not only reveal thespecific mechanisms by which Mce3E modulates host innate im-mune responses during the course of mycobacterial infection, butalso provide a potential Mce3E–ERK1/2 interface–based drugtarget against M. tuberculosis.

Materials and MethodsBacterial strains

Escherichia coli DH5a, E. coli BL21 (DE3), Mycobacterium smegmatismc2155, and M. bovis BCG (Pasteur) were used for genetic manipulationor protein expression. E. coli DH5a and E. coli BL21 (DE3) were grown inflasks using LB medium. The mycobacterial strains were grown in Mid-dlebrook 7H9 broth (271310; BD Difco) supplemented with 10% oleicacid-albumin-dexrose-catalase and 0.05% Tween 80 (Sigma-Aldrich), oron Middlebrook 7H10 agar (262710; BD Difco) supplemented with 10%oleic acid-albumin-dexrose-catalase. The Mycobacterium shuttle vectorpMV261 (provided by W. Jacobs from Albert Einstein College) was usedto express M. tuberculosis Mce3E in M. smegmatis or BCG. The myco-bacterial expression vector with GFP tag pSC300 (Addgene) was used toexpress GFP-Mce3E in M. smegmatis or BCG. Expression of Mce3E inmycobacteria-infected RAW264.7 cells was examined by quantitative PCRanalysis.

Plasmids, Abs, and reagents

The gene for Mce3E was amplified from genome ofM. tuberculosisH37Rv.For expression in mammalian cells, the gene was cloned into pcDNA6A,p3XFlag-CMV-14, or pEGFP-C1, respectively. Bacterial expression plas-mids were constructed by inserting the genes into pGEX-6p-1, pMV261,or pSC300, respectively. Luciferase assay plasmids for Gal4-Elk, Gal4-Luc, pRL-TK, pFA-cJun, pBII-Luc, RacL61, RasV12, v-Raf, constitutiveactive MEK1 mutant (DN3/S218E/S222, MEK1-ED), as well as plasmidsfor maltose-binding protein (MBP), MBP-ERK2, and myc-MEK1 wereprovided by F. Shao (National Institute of Biological Sciences, Beijing,China). ERK2 was also subcloned into p3XFlag-CMV-14 and pET30avector. The gene of p38 was amplified from total cDNA obtained fromHEK293T and inserted into pET30a vector. Mce3A gene was synthesizedby Generay Biotechnology and cloned into pEGFP-C1 vector. Mutantforms of Mce3E were generated by overlap extension. SupplementalTable I lists the detailed information on strains, plasmids, and oligonu-cleotides used in this study. The following Abs were used in this study:anti-ERK1/2 (9102; CST), anti–p-ERK1/2 (9101; CST), anti–p-MEK1/2

(9121; CST), anti-Calnexin (H-70) (sc-11397; Santa Cruz), anti-GST(TA-03; ZSGB-BIO), anti-Myc (sc-40; Santa Cruz), anti-Flag (F3165;Sigma-Aldrich), anti-GFP (AG281; Beyotime), anti-Giantin (ab24586;Abcam), anti–M. tuberculosis Rv3134 (sc-52108; Santa Cruz), anti–a-tubulin (T6199; Sigma-Aldrich), and anti-poly(ADP-ribose) polymeraseAb (9542; CST). Amylose resin (E8021) was from New England BioLabs;glutathione Sepharose 4B was from GE Healthcare; Ni-NTA resin wasfrom Qiagen. Mouse anti-Flag M2 affinity gel (F3165), U0126 ethanolate(U120), epidermal growth factor (EGF) (E9644), and LPS (L4391) werefrom Sigma-Aldrich.

Cell culture, transfection, immunoblot analysis, andimmunoprecipitation

HEK293T and RAW264.7 cells were cultured in DMEM (Life Technol-ogies) supplemented with 10% (v/v) heat-inactivated FBS. Transienttransfection was performed with the polyethylenimine method or standardcalcium phosphate method for HEK293T cells and Lipofectamine 2000(Invitrogen, Carlsbad, CA) for RAW264.7 cells, according to themanufacturers’ instructions. For immunoblot analysis, transfectedHEK293T cells were lysed in the cell lysis buffer for Western and im-munoprecipitation (P0013; Beyotime), or in a buffer containing 10 mMHEPES (pH 7.5), 50 mM NaCl, 1% Triton X-100, 2 mM EDTA, anda protease inhibitor mixture (Roche Molecular Biochemicals). Proteinswere separated by SDS-PAGE and transferred to polyvinylidene difluoridemembrane (Millipore). The blots were blocked with 5% nonfat dry milk inTBS for 1 h at room temperature and subsequently incubated overnight at4˚C with primary Abs in TBS-Tween 20 (1%, v/v) (TBST) with 5% (w/v)BSA. Following three washes of 10 min each with TBST, the blots wereincubated with goat anti-mouse IgG or goat anti-rabbit IgG conjugated toHRP at a dilution of 1:5000 in blocking buffer for 1 h at room temperature.After three washes with TBST, the blots were developed by ImmobilonWestern Chemiluminescent HRP Substrate (WBKLS0500; Millipore) andexposed to x-ray film. For immunoprecipitation, HEK293T cells wereharvested 24 h after transfection and lysed in a buffer containing 20 mMTris (pH 7.5), 150 mM NaCl, 1% Triton X-100, and a protease inhibitormixture. Cell lysates were incubated with Flag M2 beads (Sigma-Aldrich),followed by extensive wash with the lysis buffer. The bound proteins weresubjected to Western blotting analysis.

Luciferase assay

Dual luciferase assay was performed as described by Li et al. (25) using thePromega luciferase reporter system. For the ERK pathway, RAW264.7cells grown on 12-well plate were cotransfected with 0.6 mg Gal4-Elk, 0.6mg Gal4-luc, 50 ng pRL-TK and 10 ng RasV12, or 100 ng v-Raf or 100 ngconstitutive active MEK1 (MEK1-ED) in the presence or absence of 1 mgMce3E. To measure JNK activation, RAW264.7 cells were cotransfectedwith 0.3 mg pFA-cJun, 0.9 mg Gal4-luc, 0.5 mg RacL61, and 50 ng pRL-TK with or without 1 mg Mce3E plasmid. For the NF-kB pathway,RAW264.7 cells were cotransfected with 1 mg pNF-kB-Luc and 50 ngpRL-TK in the presence or absence of 1 mg Mce3E plasmid. Twenty-fourhours later, cells were treated with 100 ng/ml LPS (Sigma-Aldrich) for 5 hto stimulate NF-kB activation.

Protein purification and in vitro binding assay

E. coli BL21 (DE3) strain was used as the host for expression of GST,GST-Mce3E, GST-Mce3E (F294A), MBP, MBP-ERK2, His-ERK2, andHis-p38. Protein expression was induced at 30˚C or 16˚C with 0.1–0.2 mMisopropyl-b-D-thiogalactopyranoside after OD600 reached 0.6–0.8. GST-tagged proteins, MBP fusion proteins, and His fusion proteins were puri-fied by affinity chromatography (amylose resin for MBP fusion proteins;glutathione Sepharose 4B for GST-tagged proteins; Ni-NTA resin for Hisfusion proteins), followed by size exclusion chromatography on a Super-dex column. For pulldown assay, bait proteins were immobilized ontoamylose beads (MBP and MBP-ERK2) or glutathione Sepharose 4B (GST,GST-Mce3E and GST-Mce3E [F294A]), followed by incubation withcorresponding prey proteins in a buffer containing 20 mM HEPES, 200mM NaCl, and 1% Nonidet P40 (pH 7.4) supplemented with completeprotease inhibitors at 4˚C for 2 h. After incubation, the beads were ex-tensively washed with the binding buffer, and the bound proteins wereanalyzed by immunoblot analysis.

Cell staining and confocal microscopy

RAW264.7 cells were seeded on glass coverslips and transfected withLipofectamine 2000. Twenty-four hours after transfection, the cells werefixed with 4% paraformaldehyde in PBS for 10 min, permeabilized with0.5% Triton X-100 in PBS for 10 min, blocked with 1% milk in PBS for 30

The Journal of Immunology 3757

by guest on July 2, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

min, and labeled with primary and secondary Abs. Confocal images weretaken with a Leica SP8 confocal system.

Quantification of infected RAW264.7 cells with nuclearlocalization of phospho-ERK

At 24 h after transfection of Flag-ERK2, RAW264.7 cells were infected for4 h with pSC300-BCG, pSC300-Mce3E-BCG, or pSC300-Mce3E (F294A)-BCG at multiplicity of infection (MOI) of 100 and then subjected to cellstaining, as described above. With regard to quantification of the numberof infected RAW264.7 cells with nuclear localization of p-ERK, onlyRAW264.7 cells that were infected with GFP-labeled BCG were counted.At least 100 infected cells were scored blind in each experiment, and allexperiments were repeated three times.

Infection of macrophages, CFU counting, quantitative PCRanalysis, and ELISA

RAW264.7 cells were seeded at a density of 5 3 105–1 3 106 cells/well in6-well plates and cultured for 12 h before infection. Frozen mycobacterialstrains were thawed and centrifuged, and the pellet was resuspended inDMEM supplemented with 0.05% Tween 80 by vortexing. Then the cellswere infected with mycobacterial strains for 2 h at a MOI of 10. After

incubation for 2 h, the cells were washed three times with PBS to removeany unphagocytosed bacteria, followed by culturing in fresh mediumcontaining 20 mg/ml gentamicin. At various time points postinfection,cell culture media were obtained for ELISA; cells were washed withPBS three times and harvested for CFU measurement or quantitativePCR assay. For CFU determination, macrophages were solubilized with0.025% (v/v) SDS. Then intracellular survival was determined by platingserially diluted cultures on 7H10 plates, and the colonies were enu-merated after 3 d for M. smegmatis and 3 wk for BCG. For quantitativePCR assay, total RNA was extracted from the collected cells and reversetranscribed into cDNA by using commercially available kits according tothe manufacturers’ instructions. The obtained cDNA was then subjectedto quantitative real-time PCR analysis (quantitative PCR) with KAPASYBR FAST qPCR Kit (KAPA Biosystems) on ABI 7300 system (Ap-plied Biosystems). GAPDH mRNA was used as a reference housekeep-ing gene for normalization. For ELISA, the cell-free culture supernatantswere harvested by centrifugation at 13,000 rpm for 1 min to removesuspension cells and mycobacteria. Levels of TNF protein and IL-6protein in the supernatants were quantified by RayBio ELISA kits(Mouse TNF ELISA Kit: ELM-TNFa-001; Mouse IL-6 ELISA Kit:ELM-IL6-001; Human TNF ELISA Kit: ELH-TNFa-001; Human IL-6ELISA Kit: ELH-IL6-001), according to the manufacturer’s instructions.Each sample was done in triplicates.

FIGURE 1. Inhibition of the ERK1/2 pathway by

M. tuberculosis Mce3E. (A–C) Luciferase assays of

RasV12-, V-Raf–, and MEK1-ED–induced ERK1/2

activation in the absence or presence of Mce3E.

Mce3E blocks ERK1/2 activation in RAW264.7

macrophage cells stimulated by RasV12 (A), V-Raf

(constitutive active Raf) (B), or MEK1-ED (consti-

tutive active MEK1) (C). (D) Western blotting

analysis showed the secretion of Mce3E into the

cytosol of RAW264.7 macrophage cells infected

with BCG-expressing M. tuberculosis Mce3E. The

membrane protein Rv3134 was used as a negative

control for secretion. a-Tubulin and poly(ADP-ri-

bose) polymerase were used as markers of the cy-

tosolic and nuclear fractions, respectively. (E and F)

Expression of M. tuberculosis Mce3E in M. smeg-

matis (M. smeg) (E) or BCG (F) reduces phosphor-

ylation of ERK1/2 in RAW264.7 macrophage cells

infected for 0–24 h with indicated mycobacterial

strains as analyzed by immunoblot analysis. (G)

Phosphospecific immunoblot analysis of inhibi-

tion of MEK1-ED–induced ERK1/2 activation by

Mce3E in HEK293T cells. (H) Phosphospecific

immunoblot analysis of RasV12-induced MEK1 or

ERK1/2 activation in the presence of Mce3E. Data

are representative of at least three independent

experiments (mean and SEM in A–C). **p , 0.01

(two-tailed unpaired t test).

3758 M. TUBERCULOSIS Mce3E TARGETS HOST ERK1/2 SIGNALING PATHWAY

by guest on July 2, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

Isolation of PBMCs and differentiation of monocytes intomacrophages

Human PBMCs were derived from whole blood by Ficoll-Hypaque densitygradient centrifugation and cultured in RPMI 1640 supplemented with 10%FBS, 4 mM L-glutamine, and 1% penicillin-streptomycin for 2 h at 37˚Cwith 5% CO2 to allow the adherence of monocytes. After 2 h of adherence,PBMCs were washed five times with warmed PBS to remove those non-adherent cells, including lymphocytes, and then cultured with mediumexchanged every 3 d until the cells differentiated into macrophages.PBMC-derived macrophages were infected with mycobacterial strains andsubjected to CFU counting, quantitative PCR analysis, and ELISA, asdescribed above.

Cellular fractionation of infected macrophages

RAW264.7 cells infected with pSC300-Mce3E-BCG at MOI of 100 weresubjected to hypotonic lysis buffer containing 10 mM HEPES (pH7.9), 1.5mM MgCl2, 10 mM KCl, 0.34 M sucrose, 10% glycerol, protease inhib-itors, and 0.1% Triton X-100. The lysis mixture was then centrifuged at1300 3 g for 10 min to pellet the macrophage nuclei and BCG. The su-pernatant was then collected as the cytosolic fraction of infected macro-phages for immunoblot analysis (26).

Statistical analysis

Two-tailed unpaired t test was used for statistical analysis. Each p value ,0.05 or , 0.01 was considered statistically significant.

ResultsInhibition of the ERK pathway by M. tuberculosis Mce3E

Given the important contribution of the NF-kB and MAPK signalingpathways in activating host innate immune responses, microbes haveevolved a variety of strategies to inhibit activation of those signalingpathways (27–30). To define the potential role of M. tuberculosis

Mce3E in modulating host innate immunity as well as to identify thespecific signaling pathways it targets, we examined the effects ofM. tuberculosis Mce3E on activation of the NF-kB and MAPK sig-naling pathways. Transient expression of M. tuberculosis Mce3E inRAW264.7 macrophage cells efficiently blocked RasV12-inducedERK1/2 activation (Fig. 1A), and had little, if any, inhibitory ef-fect on the RacL61-induced JNK/p38 activation, while promotingLPS-stimulated NF-kB activation (data not shown). We thus furthersought to define the specific step(s) in the ERK1/2 pathway sup-pressed by M. tuberculosis Mce3E. As shown in Fig. 1B and 1C,M. tuberculosisMce3E efficiently blocked v-Raf (constitutive activeRaf)– and MEK1-ED (constitutive active MEK1)–stimulated ERK1/2 activation in RAW264.7 macrophage cells. Immunoblot analysiswas used to confirm the secretion of Mce3E into the cytosolicfraction of macrophages infected with pSC300-Mce3E-BCG. TheRv3134 protein, which was identified in the cell membrane fractionof M. tuberculosis H37Rv (31), was used as a negative control forsecretion (Fig. 1D). To further examine whether Mce3E suppressedERK1/2 signaling in macrophages during the course of mycobac-terial infection, we expressed M. tuberculosis Mce3E in M. smeg-matis (Mce3E-M. smegmatis) or BCG (Mce3E-BCG) for infectionof RAW264.7 cells. We found that expression of M. tuberculosisMce3E in both M. smegmatis and BCG (as assessed by quantitativePCR analysis; data not shown) resulted in reduced ERK phosphor-ylation in RAW264.7 cells during the course of infection (Fig. 1E,1F). Furthermore, Mce3E abrogated the ERK1/2 phosphorylationinduced by MEK1-ED (Fig. 1G) or RasV12 (Fig. 1H), whereas itdid not affect RasV12-induced MEK1 phosphorylation in HEK293Tcells. Collectively, these data suggest that M. tuberculosis Mce3E

FIGURE 2. Expression of M. tuberculosis

Mce3E in BCG decreases expression of TNF and

IL-6, and increases the survival of BCG in

RAW264.7 macrophages during the course of in-

fection. (A and B) Expression of M. tuberculosis

Mce3E in BCG decreases the mRNA expression of

Tnf (A) and Il6 (B) in RAW264.7 cells as examined

by quantitative PCR. Macrophage cells were

infected for 0–24 h with indicated mycobacterial

strains. (C and D) ELISA showed that expression

of M. tuberculosis Mce3E in BCG decreases the

TNF protein (C) and IL-6 protein (D) released

by RAW264.7 cells infected as in (A) and (B). (E)

Expression of M. tuberculosis Mce3E in BCG

increases the survival of BCG in RAW264.7 cells

infected as in (A) and (B). Quantitative PCR results

are presented relative to expression of the control

gene Gapdh (in A and B). Data are representative of

at least three independent experiments (mean and

SEM in A–E). *p , 0.05, **p , 0.01 (two-tailed

unpaired t test).

The Journal of Immunology 3759

by guest on July 2, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

blocks the ERK signaling downstream or at the level of MEK1along the Ras-Raf-MEK-ERK cascade.

M. tuberculosis Mce3E inhibits cytokine production andpromotes survival of mycobacteria in macrophages

Bacterial infection incites the activation of host immune signalingpathways and thereby induces the production of various cytokines suchas TNF and IL-6. Such mediators are essential for recruitment andactivation of immune cells to respond to bacterial infection (8). Incontrast, bacterial pathogens frequently target the MAPK pathway tosuppress the production of cytokines to counteract host defense (29,32). The altered ERK1/2 pathway activation by M. tuberculosisMce3E therefore raised the possibility that the expression of certaincytokines might be altered in macrophages infected with M. tuber-culosis Mce3E-expressing mycobacterial strain. To examineM. tuberculosis Mce3E’s ability in modulating cytokine expressionindependently of otherM. tuberculosis effector proteins, we analyzedthe expression of a number of cytokines in macrophages infected withMce3E-M. smegmatis or Mce3E-BCG strains by quantitative PCRanalysis and ELISA. Expression of M. tuberculosis Mce3E in eitherM. smegmatis (data not shown) or BCG (Fig. 2) greatly inhibitedthe expression of Tnf and Il6 and promoted bacterial survival inRAW264.7 macrophage cells during mycobacterial infection. Theseresults establish that M. tuberculosis Mce3E alone can effectivelysuppress the expression of certain inflammatory cytokines and pro-mote survival of mycobacteria in macrophages.

Mce3E interacts with ERK in a DEF-dependent manner

To elucidate the underlying mechanisms of Mce3E-mediated sup-pression of host innate immune responses, we attempted to search forits host targets along the ERK1/2 pathway. Based on the observa-tion that Mce3E blocks MEK1-ED–mediated activation of theERK1/2 pathway, we analyzed whether Mce3E interacted withMEK1 or ERK1/2. When Mce3E and MEK1 were coexpressed inHEK293T cells, Mce3E was able to coimmunoprecipitate with en-

dogenous ERK1/2, but not expressed MEK1 (Fig. 3A). Consistently,Flag-ERK2 could also coimmunoprecipate with Myc-Mce3E whencoexpressed in HEK293T cells (Fig. 3B). Thus, we proposed thatMce3E interacts with ERK1/2, but not MEK1. Several effectors frompathogenic bacteria such as Legionella effectors display distinctiveeukaryotic domains, among which are protein kinase or phosphatasedomains (33–35). In our efforts to search for the functional domains/motifs within Mce3E, we identified a potential DEF motif, an ERK-docking motif that possesses a consensus sequence of FXFP, inMce3E (FPFP) using in silico analysis (http://scansite.mit.edu/) (36).Because the former Phe residue is most critical among the threeconserved amino acids of the DEF motif, we thus mutated the Mce3EPhe294 into Ala (F294A), and found that this mutant could barelybind to ERK in vitro (Fig. 3C). Because existing reports suggest thatthe DEF domain also mediates interaction with p38 MAPK (37), wethus further performed experiments to examine whether this DEFmotif in Mce3E also interacts with p38 MAPK. As is shown in Fig.3D, Mce3E has little binding affinity to p38a, whereas F294A mu-tation completely abolishes such weak binding. Consistently, suchmutation almost completely abolished Mce3E-mediated inhibition ofERK1/2 phosphorylation in HEK293T cells (Fig. 3E) and ERKsignaling activation in RAW264.7 macrophage cells (Fig. 3F). Takentogether, these data demonstrate that Mce3E interacts with ERK ina DEF motif-dependent manner to inhibit ERK pathway activation.

Mce3E is localized at the ER apparatus, thus trapping ERKinto the ER apparatus

We then performed immunofluorescence assays to investigatewhether Mce3E colocalizes with ERK in vivo. Intriguingly, im-munofluorescent analysis of exogenously expressed Mce3E fusedto GFP in RAW264.7 cells revealed a staining pattern that spe-cifically overlapped with the ERmarker Calnexin, but not the Golgimarker Giantin. Subcellular distribution of Mce3E remained thesame pattern when using GFP, Flag, or Myc tags at its C terminus inseveral cell lines (data not shown). In addition, the Mce3E F294A

FIGURE 3. M. tuberculosis Mce3E binds to

ERK1/2 in a DEF motif-dependent manner. (A)

Flag-Mce3E coprecipitates with endogenous

ERK1/2, but not exogenous expressed Myc-

MEK1 from HEK293T cells. (B) Coimmunopre-

cipitation of M. tuberculosis Mce3E and ERK2 in

the lysates of the HEK293T cells cotransfected

with M. tuberculosis Mce3E and ERK2. (C) WT

Mce3E, but not its F294A mutant, interacts with

ERK2 as analyzed by MBP pulldown assay. (D)

Mce3E displays DEF motif-based strong binding

affinity to ERK2 and little binding affinity to

p38a. (E) F294A mutation abolishes the Mce3E-

mediated inhibition of ERK1/2 phosphorylation

as indicated by immunoblot assay. (F) Luciferase

assay showed that F294A mutation attenuates the

Mce3E-mediated inhibition of ERK1/2 activation

in RAW264.7 macrophage cells stimulated by

MEK1-ED. Data are representative of at least

three independent experiments (mean and SEM in

F). **p , 0.01 (two-tailed unpaired t test).

3760 M. TUBERCULOSIS Mce3E TARGETS HOST ERK1/2 SIGNALING PATHWAY

by guest on July 2, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

mutant shared the same localization pattern with the wild-typeprotein. Unlike the concentrated localization of Mce3E, Mce3Aencoded by the mce3 operon displayed diffused distribution mode(Fig. 4A). Thus, these results indicate that Mce3E specificallyresides in the ER apparatus in a DEF motif-independent manner.In resting cells, ERKs are diffusely distributed in the cytoplasmdue to their interactions with anchoring proteins, such as theirupstream activators MEKs. Upon stimulation, ERKs detach fromthe anchoring proteins and translocate to their sites of action,mainly at nucleus (38). Given the distinct localization of Mce3Eas well as its interaction with ERK, we thus sought to investigatewhether Mce3E could regulate the cellular location of ERKs.When cotransfected with Mce3E, but not its F294A mutant, inresting RAW264.7 cells, the majority of ERKs were translocatedto the ER apparatus and exhibited almost complete colocalizationwith Mce3E (Fig. 4B), consistent with our results from coimmu-noprecipitation and pulldown assays. In addition, the colocaliza-tion of Mce3E with ERK appeared to be specific for ERK, becauseMce3E had no impact on the localization of MEK1 (Fig. 4B).

Taken together, Mce3E is located at and traps ERK to the ERapparatus in host cells.

Mce3E blocks interactions between ERK and MEK andinterrupts nuclear translocation of p-ERK

Because Mce3E could interact with ERK1/2 and trap ERK1/2 tothe ER, we thus hypothesized that the Mce3E-ERK1/2 interactionsmight compete for the interactions between ERK1/2 and MEK1.We addressed this issue by coimmunoprecipitation assay withvarious expression levels of Mce3E. The interactions betweenERK1/2 and Mce3E were increased upon increasing expression ofMce3E, concurrent with decreased interactions between ERK1/2and MEK1 (Fig. 5A). Similarly, Mce3E (F294A) could not af-fect the association between ERK1/2 and MEK1 (Fig. 5B). Thesedata suggest that Mce3E, through sequestering ERK1/2 at the ER,reduces the accessibility of ERK1/2 to MEK1 and consequentlyblocks the signaling from MEK to ERK. Because the nucleartranslocation of p-ERK is stimulated by many stimuli such as EGFand could be modulated by pathogen effectors (39–41), we then

FIGURE 4. M. tuberculosis Mce3E colocalizes

with ERK1/2 at the ER apparatus in RAW264.7

cells. (A) M. tuberculosis Mce3E is specifically

localized at the ER in a DEF motif-independent

manner. RAW264.7 cells were transfected with

GFP-Mce3E or its mutant F294A or GFP-Mce3A

(green). The ER, Golgi, and cell nuclei were la-

beled by Calnexin (red), Giantin (red), and DAPI

(blue), respectively. (B) M. tuberculosis Mce3E

colocalizes with ERK1/2 in RAW264.7 cells.

RAW264.7 cells were transfected with the indi-

cated plasmids and followed by serum starvation.

Coexpression of Mce3E, but not its mutant form

F294A, causes the translocation of ERK1/2 from

cytosol to the ER. Cells were stained with anti-

ERK1/2 or anti-Myc Abs, followed by labeling

with anti-rabbit/mouse Alexa594 or anti-mouse

FITC Abs (Molecular Probes). After Ab labeling,

cells were counterstained with DAPI for nuclei

(blue). Scale bars, 10 mm. Data are representative

of at least three independent experiments.

The Journal of Immunology 3761

by guest on July 2, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

further examined whether Mce3E could affect the nuclear trans-location of p-ERK1/2.We found that Mce3E, but not its F294Amutant, also colocalizes

with p-ERK1/2 and thus blocks the nuclear translocation of

p-ERK1/2 in RAW264.7 cells stimulated with EGF (Fig. 5C).Consistently, RAW264.7 cells infected with pSC300-Mce3E-BCGshowed less p-ERK1/2 localization in nucleus than their coun-terparts infected with pSC300-BCG or pSC300-Mce3E (F294A)-

FIGURE 5. M. tuberculosis Mce3E blocks ERK1/2-MEK1 interaction and nuclear translocation of phospho-ERK1/2. (A and B) Mce3E, but not its mutant

form F294A, interferes with the interaction between ERK1/2 and MEK1. HEK293T cells were transiently transfected with the indicated plasmids. The

Flag-tagged ERK2 was immunoprecipitated, and the ERK2-associated Mce3E and MEK1 were analyzed by immunoblot analysis. (C) Mce3E blocks EGF-

stimulated nuclear translocation of p-ERK1/2 in RAW264.7 cells, as shown in immunostaining assays. RAW264.7 cells transfected with GFP-Mce3E (or its

mutant form F294A, green) were treated with EGF (100 ng/ml) for 5 min and then stained with anti–p-ERK1/2 Ab and anti-rabbit Alexa 594 Ab (red).

After Ab labeling, cells were counterstained with DAPI for nuclei (blue). Scale bars, 10 mm. (D) Expression of M. tuberculosis Mce3E in BCG, but not its

mutant form F294A, blocks the nuclear translocation of p-ERK1/2 in RAW264.7 macrophage cells during infection. RAW264.7 macrophage cells were

infected with indicated mycobacterial strains for 4 h and then stained with anti–p-ERK1/2 Ab and anti-rabbit Alexa 594 Ab (red). After Ab labeling, cells

were counterstained with DAPI for nuclei (blue). Scale bars, 10 mm. Data are representative of at least three independent experiments. (E) Fraction (%) of

infected RAW264.7 cells with nuclear localization of p-ERK1/2. Data are representative of at least three independent experiments (mean and SEM in E).

**p , 0.01 (two-tailed unpaired t test).

3762 M. TUBERCULOSIS Mce3E TARGETS HOST ERK1/2 SIGNALING PATHWAY

by guest on July 2, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

BCG at the identical time points postinfection (Fig. 5D, 5E).Together, these observations reveal that Mce3E is exclusivelylocalized at the ER apparatus and might inhibit ERK activation byspecifically sequestering cytoplasmic ERK1/2 at the ER apparatusand thereby antagonizing the interactions between ERK1/2 andMEK1. In addition, through colocalization with p-ERK1/2, Mce3Ealso blocks their nuclear translocation, thus further attenuatingERK signaling.

ERK interaction is required for Mce3E to downregulate Tnfand Il6 expression and to enhance intracellular survival ofmycobacteria

We further examined whether ERK interaction through Mce3EDEF domain is essential for the Mce3E-mediated suppressionof cytokine expression in RAW264.7 cells and primary humanmonocyte-derived macrophage cells during the course of myco-bacterial infection. We found that expression of Mce3E inM. smegmatis or BCG efficiently downregulated the expression levelof Tnf and Il6 in both RAW264.7 macrophage cells and primaryhuman monocyte-derived macrophage cells during mycobacterialinfection (Supplemental Figs. 1A–D, 2A–D, Figs. 6A–D, 7A–D).Consistently, expression of Mce3E also promoted the intracellularsurvival of mycobacteria in macrophage cells. In contrast, F294Amutation abolished Mce3E-mediated suppression of proin-flammatory cytokine production as well as promotion of intra-cellular survival of mycobacteria (Supplemental Figs. 1E, 2E,Figs. 6E, 7E). We next further examined whether chemical inhi-bition of the ERK1/2 signaling pathway could affect the expres-sion of proinflammatory cytokines and bacterial intracellularsurvival during mycobacterial infection. We pretreated RAW264.7cells for 2 h with U0126, a specific inhibitor of the ERK pathway,and then subjected them to M. smegmatis or BCG infection.

Infection of U0126-pretreated RAW264.7 cells elicited lowerexpression of Tnf and Il6 than did infection of untreated RAW264.7cells (Supplemental Fig. 3A–D, Fig. 8A–D). In addition, U0126pretreatment promoted the survival of M. smegmatis or BCG inRAW264.7 cells (Supplemental Fig. 3E, Fig. 8E). Collectively,these data suggest that Mce3E might suppress cytokine productionand enhance the intracellular survival of mycobacteria via inhi-bition of ERK1/2 activation in a DEF motif-dependent manner(Fig. 9).

DiscussionThe human intracellular pathogen M. tuberculosis can persist inhost macrophage cells for long periods, even in a fully function-ing immune system. Among the repertoire of strategies thatM. tuberculosis has evolved to evade, divert, or subvert immuneresponses, the best documented are those that regulate host innateimmune signaling pathways to avoid the elimination of the bacilliwithin macrophages. Many cytokines regulated by NF-kB andMAPK signaling pathways have been implicated in host–myco-bacteria interactions. There are increasing numbers of examples ofbacterial pathogens that produce and secrete effector moleculesthat dampen inflammatory cytokine production. For example, theYersinia effector YopJ, which is a ubiquitin-like cysteine protease,targets and downregulates both the NF-kB and MAPK pathways(42). Shigella flexneri effector OspG, which is a protein kinase,targets ubiquitin-conjugated enzymes, thereby affecting phospho-IkBa degradation and subsequent NF-kB activation (43). OspF,another Shigella type III effector, was also shown to inactivateMAPKs, including ERK1/2, JNK, and p38, by irreversibly re-moving phosphate groups from the phosphothreonine, but notfrom the phosphotyrosine residue in the activation loop of MAPKs(25). But up to now, few mycobacterial effectors as well as their

FIGURE 6. Expression of M. tuberculosis Mce3E,

but not Mce3E (F294A), in M. smegmatis decreases

production of TNF and IL-6 and increases the sur-

vival of M. smegmatis in primary human monocyte-

derived macrophages during the course of infection.

(A and B) Quantitative PCR analysis showed that

expression of M. tuberculosis Mce3E, but not Mce3E

(F294A), in M. smegmatis decreases the mRNA ex-

pression of Tnf (A) and Il6 (B) in primary human

monocyte-derived macrophages. Macrophage cells

were infected for 0–24 h with indicated mycobacte-

rial strains. (C and D) ELISA showed that expression

of M. tuberculosis Mce3E, but not Mce3E (F294A),

in M. smegmatis decreases the TNF protein (C) and

IL-6 protein (D) released by primary human mono-

cyte-derived macrophages infected as in (A) and (B).

(E) Expression of M. tuberculosis Mce3E, but not

Mce3E (F294A), increases the survival of M. smeg-

matis in primary human monocyte-derived macro-

phages infected as in (A) and (B). Quantitative PCR

results are presented relative to expression of the

control gene Gapdh (in A and B). Data are repre-

sentative of at least three independent experiments

(mean and SEM in A–E). *p , 0.05, **p , 0.01

(two-tailed unpaired t test).

The Journal of Immunology 3763

by guest on July 2, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

host-modulatory mechanisms have been revealed. In this study, wedemonstrate that Mce3E, a secreted effector protein from M. tu-berculosis, can inhibit ERK1/2 pathway activation, concurrentwith decreased expression of a specific pool of inflammatorycytokines, including Tnf and Il6, which in turn leads to inhibitionof the development of efficient immune responses against themycobacteria.Previous studies involving Mce3E and other Mce proteins were

mainly focused on their potential role in facilitating the uptake ofmycobacteria by macrophages, although relatively little is knownabout howMce3E and other Mce proteins function after entering ofthe pathogen into host macrophage cells. The fact that inactivationof mce operons could attenuate M. tuberculosis virulence in themurine model provides a convincing evidence that Mce proteinsare essential for intracellular survival of mycobacteria; we thus inthis study sought to investigate the potential role of Mce3E in themodulation of host innate immune responses during mycobacterialinfection of macrophage cells. Mce3E was presumably defined asa lipoprotein and anchored in the outer membrane of M. tuber-culosis (16). Previous studies suggest that the Mce3 are localizedextracellularly, and, based on their topology, they are most likelyinvolved either in transport activity or in functions involvinginteractions with the environment or the host (17, 19). But up tonow, it is still unclear whether Mce3E could be secreted into thecytosol of host cells during mycobacterial infection. In this work,based on our study, we propose that Mce3E is likely to be secretedoutside of the mycobacteria and gain access to the cytosol ofinfected macrophages, leading to manipulation of host cellularactivities, including immune signaling pathways, which is con-sistent with our data demonstrating that M. tuberculosis Mce3Einhibits ERK pathway activation and downregulates the expres-sion of certain proinflammatory cytokines.

Microbial pathogens have evolved mechanisms to subvert ERfunctions, thereby influencing the host immune response. The ER isone of the organelles of the host cell that is targeted by intracellularpathogens. Certain bacterial toxins also hijack the ER for entry andintracellular survival (35). In addition, in the Ras/Raf/MEK/ERKsignaling cascade, several mediators within this pathway are ex-quisitely regulated by subcellular compartmentalization (44, 45).For example, b-arrestin could function as a scaffold protein totarget the MAPK protein complex to early endosome (46). It wasreported that Raf kinase can be recruited to the Golgi by Raf ki-nase trapping to Golgi, and that such spatial regulation inhibitsRaf-1 activation and reduces the association of Raf-1 with Ras andMEK, thus blocking the ERK pathway (47). In this study, weunveil another aspect of the ER function that mycobacteria hijackto maintain infection. Our study reveals that M. tuberculosisMce3E modulates the ERK signaling pathway through spatialregulation of subcellular localization of ERK/p-ERK. Becausehydrophobicity analysis indicated Mce3E contains no trans-membrane domains, we thus hypothesized that Mce3E localizes atER apparatus by interacting with certain ER resident protein.Further identification of such protein(s) is currently underway inthe laboratory.Phosphorylation plays important roles in regulating numerous

eukaryotic cellular processes, including cell cycle progression,vesicular trafficking, and innate immune responses (48). Severaleffector proteins from pathogenic bacteria were shown to mimickhost proteins such as phosphatases or kinases to evade host im-mune defenses (33, 35, 49–51). Because Mce3E has a DEF motifand could interact with ERK1/2 directly, we initially examinedwhether Mce3E could be able to dephosphorylate p-ERK directly,but we were not able to demonstrate the phosphatase activity ofMce3E toward p-ERK in vitro. We also conducted yeast two-

FIGURE 7. Expression of M. tuberculosis

Mce3E, but not Mce3E (F294A), in BCG decreases

production of TNF and IL-6 and increases the

survival of BCG in primary human monocyte-de-

rived macrophages during the course of infection.

(A and B) Quantitative PCR analysis showed that

expression of M. tuberculosis Mce3E, but not

Mce3E (F294A), in BCG decreases the mRNA

expression of Tnf (A) and Il6 (B) in primary human

monocyte-derived macrophages. Macrophage cells

were infected for 0–24 h with indicated mycobac-

terial strains. (C and D) ELISA showed that ex-

pression of M. tuberculosis Mce3E, but not Mce3E

(F294A), in BCG decreases the TNF protein (C)

and IL-6 protein (D) released by primary human

monocyte-derived macrophages infected as in (A)

and (B). (E) Expression of M. tuberculosis Mce3E,

but not Mce3E (F294A), increases the survival of

BCG in primary human monocyte-derived macro-

phages infected as in (A) and (B). Quantitative PCR

results are presented relative to expression of the

control gene Gapdh (in A and B). Data are repre-

sentative of at least three independent experiments

(mean and SEM in A–E). *p , 0.05, **p , 0.01

(two-tailed unpaired t test).

3764 M. TUBERCULOSIS Mce3E TARGETS HOST ERK1/2 SIGNALING PATHWAY

by guest on July 2, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

hybrid assay to identify potential Mce3E-interacting phosphatase(s)that might be brought to p-ERK1/2 to function more efficiently,but the only Mce3E-interacting phosphatase we identified was

Ptpn23, and Mce3E-mediated inhibition of ERK1/2 activation wasnot affected in Ptpn23 knocked-down cells (data not shown). Wethus turned to other explanations for the mechanisms underlying

FIGURE 8. Inhibition of ERK1/2 pathway acti-

vation causes downregulation of the expression of

Tnf and Il6 and promotion of the survival of BCG

in RAW264.7 macrophage cells during the course

of infection. (A and B) Inhibition of ERK1/2 acti-

vation by U0126 causes decreased expression of

Tnf (A) and Il6 (B) in BCG-infected RAW264.7

macrophage cells as examined by quantitative PCR.

Macrophage cells were infected for 0–24 h with

indicated mycobacterial strains. (C and D) ELISA

showed that inhibition of ERK1/2 activation by

U0126 decreases the TNF protein (C) and IL-6

protein (D) released by RAW264.7 cells infected as

in (A and B). (E) Inhibition of ERK1/2 activation by

U0126 promotes survival of BCG in RAW264.7

cells infected as in (A) and (B). Quantitative PCR

results are presented relative to expression of the

control gene Gapdh (in A and B). Data are repre-

sentative of at least three independent experiments

(mean and SEM in A–E). *p , 0.05, **p , 0.01

(two-tailed unpaired t test).

FIGURE 9. Proposed model for

the mechanism of M. tuberculosis

Mce3E-mediated host innate im-

mune suppression. Mce3E interacts

with and colocalizes with ERK1/2 at

the ER apparatus in a DEF motif-

dependent manner. Through associ-

ation with ERK1/2, Mce3E changes

the subcellular localization of

ERK1/2 from cytoplasm to the ER

apparatus, reduces the association of

ERK1/2 with MEK1, and blocks the

nuclear translocation of phospho-

ERK1/2, leading to inhibition of

ERK1/2 signaling pathway activa-

tion, suppression of Tnf and Il6

expression, and promotion of my-

cobacterial survival within macro-

phages.

The Journal of Immunology 3765

by guest on July 2, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

Mce3E-mediated downregulation of ERK1/2 phosphorylation inhost cells. Because Mce3E not only locates at the ER, but alsodirectly and specifically interacts with host ERK1/2 and p-ERK1/2through a DEF motif, we thus hypothesized that Mce3E mightantagonize the interactions between ERK1/2 and MEK1. In ad-dition, through colocalization with p-ERK1/2, Mce3E could alsoblock the nuclear translocation p-ERK1/2, thus further attenuatingERK signaling.We found that DEF motif-based Mce3E-ERK interaction is

required for Mce3E to inhibit ERK1/2 signaling activation,downregulate cytokine expression, and enhance intracellular sur-vival of mycobacteria. Notably, the novel DEF motif identified inM. tuberculosis Mce3E shared conserved residues with typicalDEF motifs of molecules from eukaryotic cells. Moreover, con-sistent to the previous finding that the DEF motif promotes tar-geting by ERK2 and, to a lesser extent, p38a, but not p38b2 (37),we found that Mce3E displayed DEF motif-based strong bindingaffinity to ERK2 and little binding affinity to p38a. Such dis-covery might help to explain the slight, if any, inhibitory effect onthe RacL61-induced JNK/p38 activation by Mce3E. Our discoveryof this novel DEF motif in M. tuberculosis Mce3E might lead tofurther identification of similar DEF motifs in mycobacteria andperhaps other pathogenic bacteria. Our results suggest that, al-though strategies to reduce damaging inflammation and facilitatetissue repair could therefore be helpful in augmenting standard TBdrug regimens, great care could be necessary in targeting theERK1/2 signaling pathway in host as it may lead to unwantedexacerbation of, or increased susceptibility to, certain bacterialinfections such as M. tuberculosis infection. Instead, the DEFmotif of M. tuberculosis Mce3E might provide more efficient andselective target for the development of novel anti-TB therapies.In summary, this study reveals a previously unknown, yet im-

portant, function of M. tuberculosis Mce3E in exploiting the hostERK1/2 signaling pathway to evade host immune defense againstmycobacterial infection. We demonstrate that Mce3E localizes atER, interacts with ERK in a DEF motif-dependent manner, andsuch interaction traps ERK into the ER and compromises theassociation between ERK and MEK as well as the nuclear trans-location of p-ERK, thus abolishing the activation of ERK. Theseresults suggest that M. tuberculosis Mce3E might exploit theERK1/2 signaling pathway to suppress host innate immuneresponses and promote intracellular survival of mycobacteria,providing a novel potential Mce3E–ERK1/2 interface–based drugtarget against TB diseases.

AcknowledgmentsWe thank Dr. Feng Shao from National Institute of Biological Sciences for

providing multiple plasmids, for helpful discussions, and for critical reading

of the manuscript.

DisclosuresThe authors have no financial conflicts of interest.

References1. World Health Organization. 2013. Global tuberculosis report 2013. Geneva,

Switzerland: WHO. Available at: http://apps.who.int/iris/bitstream/10665/91355/1/9789241564656_eng.pdf. Accessed: February 25, 2015.

2. Guirado, E., L. S. Schlesinger, and G. Kaplan. 2013. Macrophages in tubercu-losis: friend or foe. Semin. Immunopathol. 35: 563–583.

3. Gagneux, S. 2012. Host-pathogen coevolution in human tuberculosis. Philos.Trans. R. Soc. Lond. B Biol. Sci. 367: 850–859.

4. Keane, J., H. G. Remold, and H. Kornfeld. 2000. Virulent Mycobacterium tu-berculosis strains evade apoptosis of infected alveolar macrophages. J. Immunol.164: 2016–2020.

5. Seto, S., K. Tsujimura, and Y. Koide. 2012. Coronin-1a inhibits autophagosomeformation around Mycobacterium tuberculosis-containing phagosomes andassists mycobacterial survival in macrophages. Cell. Microbiol. 14: 710–727.

6. Welin, A., J. Raffetseder, D. Eklund, O. Stendahl, and M. Lerm. 2011. Impor-tance of phagosomal functionality for growth restriction of Mycobacterium tu-berculosis in primary human macrophages. J. Innate Immun. 3: 508–518.

7. Warner, D. F., and V. Mizrahi. 2006. Tuberculosis chemotherapy: the influenceof bacillary stress and damage response pathways on drug efficacy. Clin.Microbiol. Rev. 19: 558–570.

8. van Crevel, R., T. H. M. Ottenhoff, and J. W. M. van der Meer. 2002. Innateimmunity to Mycobacterium tuberculosis. Clin. Microbiol. Rev. 15: 294–309.

9. Zhang, Y. L., and C. Dong. 2005. MAP kinases in immune responses. Cell. Mol.Immunol. 2: 20–27.

10. Azad, A. K., W. Sadee, and L. S. Schlesinger. 2012. Innate immune genepolymorphisms in tuberculosis. Infect. Immun. 80: 3343–3359.

11. Lawrence, T. 2009. The nuclear factor NF-kappaB pathway in inflammation.Cold Spring Harb. Perspect. Biol. 1: a001651.

12. Roy, C. R., and E. S. Mocarski. 2007. Pathogen subversion of cell-intrinsic in-nate immunity. Nat. Immunol. 8: 1179–1187.

13. Ruckdeschel, K., J. Machold, A. Roggenkamp, S. Schubert, J. Pierre,R. Zumbihl, J. P. Liautard, J. Heesemann, and B. Rouot. 1997. Yersinia entero-colitica promotes deactivation of macrophage mitogen-activated protein kinasesextracellular signal-regulated kinase-1/2, p38, and c-Jun NH2-terminal kinase:correlation with its inhibitory effect on tumor necrosis factor-alpha production.J. Biol. Chem. 272: 15920–15927.

14. Roach, S. K., and J. S. Schorey. 2002. Differential regulation of the mitogen-activated protein kinases by pathogenic and nonpathogenic mycobacteria. Infect.Immun. 70: 3040–3052.

15. Song, C. H., J. S. Lee, S. H. Lee, K. Lim, H. J. Kim, J. K. Park, T. H. Paik, andE. K. Jo. 2003. Role of mitogen-activated protein kinase pathways in the pro-duction of tumor necrosis factor-alpha, interleukin-10, and monocyte chemo-tactic protein-1 by Mycobacterium tuberculosis H37Rv-infected humanmonocytes. J. Clin. Immunol. 23: 194–201.

16. Ahmad, S., S. El-Shazly, A. S. Mustafa, and R. Al-Attiyah. 2005. The sixmammalian cell entry proteins (Mce3A-F) encoded by the mce3 operon areexpressed during in vitro growth of Mycobacterium tuberculosis. Scand. J.Immunol. 62: 16–24.

17. Casali, N., and L. W. Riley. 2007. A phylogenomic analysis of the Actino-mycetales mce operons. BMC Genomics 8: 60.

18. Chitale, S., S. Ehrt, I. Kawamura, T. Fujimura, N. Shimono, N. Anand, S. Lu,L. Cohen-Gould, and L. W. Riley. 2001. RecombinantMycobacterium tuberculosisprotein associated with mammalian cell entry. Cell. Microbiol. 3: 247–254.

19. El-Shazly, S., S. Ahmad, A. S. Mustafa, R. Al-Attiyah, and D. Krajci. 2007.Internalization by HeLa cells of latex beads coated with mammalian cell entry(Mce) proteins encoded by the mce3 operon of Mycobacterium tuberculosis. J.Med. Microbiol. 56: 1145–1151.

20. Gioffre, A., E. Infante, D. Aguilar, M. P. Santangelo, L. Klepp, A. Amadio,V. Meikle, I. Etchechoury, M. I. Romano, A. Cataldi, et al. 2005. Mutation inmce operons attenuates Mycobacterium tuberculosis virulence. Microbes Infect.7: 325–334.

21. Senaratne, R. H., B. Sidders, P. Sequeira, G. Saunders, K. Dunphy,O. Marjanovic, J. R. Reader, P. Lima, S. Chan, S. Kendall, et al. 2008. Myco-bacterium tuberculosis strains disrupted in mce3 and mce4 operons are attenu-ated in mice. J. Med. Microbiol. 57: 164–170.

22. Behr, M. A., M. A. Wilson, W. P. Gill, H. Salamon, G. K. Schoolnik, S. Rane,and P. M. Small. 1999. Comparative genomics of BCG vaccines by whole-genome DNA microarray. Science 284: 1520–1523.

23. Ahmad, S., S. El-Shazly, A. S. Mustafa, and R. Al-Attiyah. 2004. Mammalian cell-entry proteins encoded by the mce3 operon of Mycobacterium tuberculosis areexpressed during natural infection in humans. Scand. J. Immunol. 60: 382–391.

24. Al-Attiyah, R., and A. S. Mustafa. 2010. Characterization of human cellularimmune responses to Mycobacterium tuberculosis proteins encoded by genespredicted in RD15 genomic region that is absent in Mycobacterium bovis BCG.FEMS Immunol. Med. Microbiol. 59: 177–187.

25. Li, H., H. Xu, Y. Zhou, J. Zhang, C. Long, S. Li, S. Chen, J. M. Zhou, andF. Shao. 2007. The phosphothreonine lyase activity of a bacterial type III effectorfamily. Science 315: 1000–1003.

26. Cowley, S. C., R. Babakaiff, and Y. Av-Gay. 2002. Expression and localizationof the Mycobacterium tuberculosis protein tyrosine phosphatase PtpA. Res.Microbiol. 153: 233–241.

27. Orth, K., L. E. Palmer, Z. Q. Bao, S. Stewart, A. E. Rudolph, J. B. Bliska, andJ. E. Dixon. 1999. Inhibition of the mitogen-activated protein kinase kinasesuperfamily by a Yersinia effector. Science 285: 1920–1923.

28. Barry, A. O., N. Boucherit, G. Mottola, P. Vadovic, V. Trouplin, P. Soubeyran,C. Capo, S. Bonatti, A. Nebreda, R. Toman, et al. 2012. Impaired stimulation ofp38a-MAPK/Vps41-HOPS by LPS from pathogenic Coxiella burnetii preventstrafficking to microbicidal phagolysosomes. Cell Host Microbe 12: 751–763.

29. Cho, J. E., S. Park, S. N. Cho, H. Lee, and Y. S. Kim. 2012. c-Jun N-terminalkinase (JNK) and p38 mitogen-activated protein kinase (p38 MAPK) are in-volved in Mycobacterium tuberculosis-induced expression of Leukotactin-1.BMB Rep. 45: 583–588.

30. Rahman, M. M., and G. McFadden. 2011. Modulation of NF-kB signalling bymicrobial pathogens. Nat. Rev. Microbiol. 9: 291–306.

31. Gu, S., J. Chen, K. M. Dobos, E. M. Bradbury, J. T. Belisle, and X. Chen. 2003.Comprehensive proteomic profiling of the membrane constituents of a Myco-bacterium tuberculosis strain. Mol. Cell. Proteomics 2: 1284–1296.

32. Zheng, Y., S. Lilo, I. E. Brodsky, Y. Zhang, R. Medzhitov, K. B. Marcu, andJ. B. Bliska. 2011. A Yersinia effector with enhanced inhibitory activity on theNF-kB pathway activates the NLRP3/ASC/caspase-1 inflammasome in macro-phages. PLoS Pathog. 7: e1002026.

3766 M. TUBERCULOSIS Mce3E TARGETS HOST ERK1/2 SIGNALING PATHWAY

by guest on July 2, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

33. Whitmore, S. E., and R. J. Lamont. 2012. Tyrosine phosphorylation and bacterialvirulence. Int. J. Oral Sci. 4: 1–6.

34. Zhu, H., J. Zhou, Y. Ni, Z. Yu, A. Mao, Y. Hu, W. Wang, X. Zhang, L. Wen,B. Li, et al. 2014. Contribution of eukaryotic-type serine/threonine kinase tostress response and virulence of Streptococcus suis. PLoS One 9: e91971.

35. Hervet, E., X. Charpentier, A. Vianney, J. C. Lazzaroni, C. Gilbert, D. Atlan, andP. Doublet. 2011. Protein kinase LegK2 is a type IV secretion system effectorinvolved in endoplasmic reticulum recruitment and intracellular replication ofLegionella pneumophila. Infect. Immun. 79: 1936–1950.

36. Sheridan, D. L., Y. Kong, S. A. Parker, K. N. Dalby, and B. E. Turk. 2008.Substrate discrimination among mitogen-activated protein kinases through dis-tinct docking sequence motifs. J. Biol. Chem. 283: 19511–19520.

37. Galanis, A., S. H. Yang, and A. D. Sharrocks. 2001. Selective targeting ofMAPKs to the ETS domain transcription factor SAP-1. J. Biol. Chem. 276: 965–973.

38. Zehorai, E., Z. Yao, A. Plotnikov, and R. Seger. 2010. The subcellular locali-zation of MEK and ERK—a novel nuclear translocation signal (NTS) pavesa way to the nucleus. Mol. Cell. Endocrinol. 314: 213–220.

39. Mochizuki, M., Y. W. Kwon, J. Yodoi, and H. Masutani. 2009. Thioredoxinregulates cell cycle via the ERK1/2-cyclin D1 pathway. Antioxid. Redox Signal.11: 2957–2971.

40. Yoo, M. H., B. A. Carlson, V. N. Gladyshev, and D. L. Hatfield. 2013. Abrogatedthioredoxin system causes increased sensitivity to TNF-a-induced apoptosis viaenrichment of p-ERK 1/2 in the nucleus. PLoS One 8: e71427.

41. Samuelson, D. R., and M. E. Konkel. 2013. Serine phosphorylation of cortactinis required for maximal host cell invasion by Campylobacter jejuni. Cell Com-mun. Signal. 11: 82.

42. Orth, K., Z. Xu, M. B. Mudgett, Z. Q. Bao, L. E. Palmer, J. B. Bliska,W. F. Mangel, B. Staskawicz, and J. E. Dixon. 2000. Disruption of signaling by

Yersinia effector YopJ, a ubiquitin-like protein protease. Science 290: 1594–1597.

43. Kim, D. W., G. Lenzen, A. L. Page, P. Legrain, P. J. Sansonetti, and C. Parsot.2005. The Shigella flexneri effector OspG interferes with innate immuneresponses by targeting ubiquitin-conjugating enzymes. Proc. Natl. Acad. Sci.USA 102: 14046–14051.

44. Harding, A., T. Tian, E. Westbury, E. Frische, and J. F. Hancock. 2005. Sub-cellular localization determines MAP kinase signal output. Curr. Biol. 15: 869–873.

45. Mor, A., and M. R. Philips. 2006. Compartmentalized Ras/MAPK signaling.Annu. Rev. Immunol. 24: 771–800.

46. Luttrell, L. M., F. L. Roudabush, E. W. Choy, W. E. Miller, M. E. Field,K. L. Pierce, and R. J. Lefkowitz. 2001. Activation and targeting of extracellularsignal-regulated kinases by beta-arrestin scaffolds. Proc. Natl. Acad. Sci. USA98: 2449–2454.

47. Feng, L., X. Xie, Q. Ding, X. Luo, J. He, F. Fan, W. Liu, Z. Wang, and Y. Chen.2007. Spatial regulation of Raf kinase signaling by RKTG. Proc. Natl. Acad. Sci.USA 104: 14348–14353.

48. Fischer, E. H. 2013. Cellular regulation by protein phosphorylation. Biochem.Biophys. Res. Commun. 430: 865–867.

49. Bach, H., K. G. Papavinasasundaram, D. Wong, Z. Hmama, and Y. Av-Gay.2008. Mycobacterium tuberculosis virulence is mediated by PtpA dephosphor-ylation of human vacuolar protein sorting 33B. Cell Host Microbe 3: 316–322.

50. Dukuzumuremyi, J. M., R. Rosqvist, B. Hallberg, B. Akerstrom, H. Wolf-Watz,and K. Schesser. 2000. The Yersinia protein kinase A is a host factor inducibleRhoA/Rac-binding virulence factor. J. Biol. Chem. 275: 35281–35290.

51. Walburger, A., A. Koul, G. Ferrari, L. Nguyen, C. Prescianotto-Baschong,K. Huygen, B. Klebl, C. Thompson, G. Bacher, and J. Pieters. 2004. Proteinkinase G from pathogenic mycobacteria promotes survival within macrophages.Science 304: 1800–1804.

The Journal of Immunology 3767

by guest on July 2, 2018http://w

ww

.jimm

unol.org/D

ownloaded from