Da

JYa

b

c

d

e

f

a

ARRA

KDEREM

1

ama

ePsm

T

h0

Molecular Immunology 60 (2014) 135–142

Contents lists available at ScienceDirect

Molecular Immunology

j ourna l ho me pa ge: www.elsev ier .com/ locate /mol imm

aidzein enhances efferocytosis via transglutaminase 2 andugmentation of Rac1 activity

ia-Hau Yena, Deng-Jye Yangb, Meng-Chi Chena, Wu Yi-Yingc, Yu-Fan Hsieha,ueh-Mei Chenga, Wen-Nan Huangd, Zsuzsa Szondye, Gregory J. Tsaya,f,∗

Institute of Microbiology and Immunology, TaiwanSchool of Health Diet and Industry Management, Chung Shan Medical University, Taichung, TaiwanDepartment of Medical Laboratory Science and Biotechnology, China Medical University and Hospital, Taichung, TaiwanDepartment of Allergy, Immunology and Rheumatology, Taichung Veterans General Hospital, TaiwanDepartment of Biochemistry and Molecular Biology, Apoptosis and Genomics Research Group, Hungarian Academy of Sciences, Debrecen, HungaryDepartment of Internal Medicine, Chung Shan Medical University Hospital, Taichung, Taiwan

r t i c l e i n f o

rticle history:eceived 17 December 2013eceived in revised form 14 March 2014ccepted 20 April 2014

eywords:aidzeinfferocytosisac1rkitochondrial membrane potential

a b s t r a c t

Clearance of apoptotic cells, termed “efferocytosis”, is the mechanism required to prevent secondarynecrosis and release of proinflammatory cytokines. Defective efferocytosis is cumulatively regarded asone of mechanisms in the development of autoimmune and chronic inflammatory diseases. Our pre-vious finding showed that ethanolic extract from Glycine tomentella Hayata (GTH) can enhance mousemacrophage RAW264.7 efferocytosis (clearance of apoptotic cells). We have demonstrated that the majorcomponents of GTH are daidzein, catechin, epicatechin and naringin. Here, we explore the potential ofeach component in modulating efferocytic capability. For this, RAW264.7 cells were cultured with CFDA-stained apoptotic cells and assayed by flow cytometry. We found that daidzein is the main componentof GTH, and it can enhance RAW264.7 efferocytosis dose-dependently. Moreover, the enhancive effectof daidzein on macrophage efferocytic capability is accompanied by increased transglutaminase 2 (TG2)at both mRNA and protein levels. TG2 knockdown attenuated daidzein increased macrophage effero-cytic capability. After treatment with daidzein, increased phosphorylation was observed in Erk, but not

in p38 and JNK. Finally, we report that after daidzein treatment, Rac1 activity was markedly increasedand the mitochondrial membrane potential was decreased, which may contribute to efferocytosis. Takentogether, these data suggest that enhancement of macrophage efferocytic capability by daidzein treat-ment was mainly through up-regulation of TG2 expression and Rac1 activity. Daidzein may have thetherapeutical potential in the treatment of inflammatory diseases.

© 2014 Elsevier Ltd. All rights reserved.

. Introduction

Apoptosis is the process of programmed cell death that plays

n essential role in the development and maintenance of allammalian tissues. The apoptotic program ensures that the dam-

ged, aged, or excess cells are deleted in a regulated manner to

Abbreviations: TG2, transglutaminase 2; GTH, Glycine tomentella Hayata; Erk,xtracellular signal-regulated kinase; TGF-�, transforming growth factor �; RT-CR, reverse transcription-polymerase chain reaction; NF�B, nuclear factor-kappaB;iRNA, small interfering RNA; MAPK, mitogen-activated protein kinase; MDM,onocyte-derived macrophage.∗ Corresponding author at: 110 Section 1, Chien-Kuo N. Road, Taichung 402,aiwan. Tel.: +886 4934487707; fax: +886 4 2324 8172.

E-mail address: gjt@csmu.edu.tw (G.J. Tsay).

ttp://dx.doi.org/10.1016/j.molimm.2014.04.006161-5890/© 2014 Elsevier Ltd. All rights reserved.

maintain normal tissue homeostasis (Henson et al., 2001; Hensonand Hume, 2006). The apoptotic cells are removed by phagocyto-sis by professional phagocytes, such as macrophages or dendriticcells; or by nonprofessional phagocytes including fibroblasts, stro-mal cells, and epithelial cells. The ingestion of apoptotic cells(termed “efferocytosis”) by phagocytes has been shown to triggerrelease of molecules such as transforming growth factor � (TGF-�), interleukin-10 (IL-10), and prostaglandin E2 (PGE2), which arethought to be mediators of immunosuppression and immunolog-ical tolerance to self-antigens (Ravichandran, 2011; Savill et al.,2002; Michlewska et al., 2007). Apoptotic cells are recognized by

the receptors on the phagocytes through the so-called ‘eat-me’signals on the surface of the apoptotic cell (Ravichandran, 2011).The molecules downstream of these receptors are partly identi-fied, including small GTPase Rac-1, Cdc42, and RhoG which belong

1 Immun

tioticwsa2ae

idbc�hpadtemT

tmiaalrdttat(o(dami

caea

2

2

d1acbnaM

36 J.-H. Yen et al. / Molecular

o Rho family of GTPase. Activation of these molecules resultsn the reorganization of the actin cytoskeleton and engulfmentf apoptotic cells. In contrast, RhoA has been reported to inhibithis process (Tosello-Trampont et al., 2003). Thus, efferocytosiss tightly regulated by the small GTPase Rac-1 in the process ofytoskeletal reorganization, in which two parallel signaling path-ays are involved. The first pathway is initiated by CD91 or

tabilin-2 functioning as receptors, which acts in concert with thedaptor protein GULP, and consequently activates Rac1 (Su et al.,002; Park et al., 2008). The second pathway involves Dock180nd ELMO functioning together as a bipartite guanine nucleotidexchange factor (GEF) for Rac1 (Brugnera et al., 2002).

A recent study has shown that tissue transglutaminase 2 (TG2)s needed for the formation of an efficient phagocyte portaluring efferocytosis (Toth et al., 2009). Tissue transglutaminaseelongs to the family of transglutaminases which catalyze theross-linking between an �-amino group of a lysine residue and a-carboxamide group of a glutamine residue. In addition, TG2 alsoas GTPase activity, suggesting that this molecule participates in Grotein-mediated signaling processes in the presence of GTP (Fesusnd Piacentini, 2002). Previous study reported that TG2−/− miceevelop age-dependent autoimmune disease on account of defec-ive efferocytosis in vivo. TG2 deficiency fundamentally impairsfferocytosis and alters release of TGF-� and other cytokines inacrophages (Szondy et al., 2003, 2011). Based on these findings,

G2 might play an important role in efferocytosis.We have previously reported that a total extract of Glycine

omentella Hayata (GTH) could enhance efferocytosis and TG2RNA up-regulation during this process (Yen et al., 2010). Daidzein

s a plant-derived diphenolic isoflavone found in a number of plantsnd herbs like Trifolium pratense (red clover) or Pueraria mirifica,s well as in food sources such as soybeans and soy productsike tofu (Wang et al., 2008; Hamalainen et al., 2007). It has beeneported that in murine macrophage J774 and RAW 264.7 cellsaidzein could inhibit expression of inducible nitric oxide syn-hase (iNOS) and nitric oxide production which may account forhe its anti-inflammatory effects (Franke et al., 1998). Daidzein haslso been shown to inhibit lipopolysaccharide (LPS)-induced NF-�Branscriptional activity in mouse macrophage cells and fibroblastsVanden Berghe et al., 2006). In another study, daidzein reducedsteoclastic differentiation in the murine RAW264.7 cell modelGarcia Palacios et al., 2005). Moreover, daidzein reduced myocar-ial injury in a rat ischemia/reperfusion model by inhibiting NF-�Bctivation (Kim et al., 2009). These studies suggest that daidzeinodulates cell signaling pathways and subsequently production of

nflammatory cytokines.In this study, we demonstrate that daidzein enhances efferocytic

apability in a dose-dependent manner by up-regulation of TG2nd Rac1. Daidzein may be a potent plant-derived substance fornhancing efferocytosis and may have therapeutic potentials forutoimmune and inflammatory diseases.

. Materials and methods

.1. Cell culture and isolation

Normal human PBMCs were isolated from five healthyonors by density Histopaque®-1077 sterile-filtered (density:.077 g/ml) gradient centrifugation. The cells were then washednd resuspended in Dulbeco’s Modified Eagle Medium (DMEM)ontaining 10% fetal bovine serum (Biological Industries, Kib-

utz Beit Haemek, Israel), 2 mM glutamine, 1 mM pyruvate, 1%on-essential amino acid, 100 U/ml penicillin, 0.0025 mg/mlmphotericin, and 1 mg/ml streptomycin (Biological Industries).onocyte-derived macrophages (MDM) were isolated from PBMCs

ology 60 (2014) 135–142

by plastic adherence. Phenotypical and functional characteriza-tions of MDM was performed after 6–7 days. RAW264.7 cellswere purchased from Bioresource Collection and Research Center(HsinChu, Taiwan) and cultured in DMEM.

2.2. Cell viability

Cell viability was assessed by using annexin V and propidiumiodide stain. Measurement of early and late apoptosis was per-formed by flow cytometry using annexin V: FITC assay kit (AbDSerotec, Oxford, UK) according to the manufacture’s instructions.The stained cells were analyzed by FACSCAN laser flow cytometry(Becton Dickenson, San Jose, CA, USA).

2.3. Reverse transcription-polymerase chain reaction (RT-PCR)

Total RNA was isolated from RAW264.7 cells by using theTrizol reagent protocol (Sigma, St. Louis, MO, USA). Two micro-grams of total RNA were denatured at 65 ◦C with 1 �l oligo-dT(Promega, Madison, WI, USA), and 4 �l dNTPs (10 mM) for 5 minin 12 �l final volume. The primer-RNA mixture was cooled onice, and 1 �l Moloney Murine Leukemia Virus (M-MLV) reversetranscriptase (Invitrogen, Carlsbad, CA, USA), 1 �l RNase inhibitor(Promega) and 4 � l 5× RT-buffer were added for a total volumeof 20 �l. PCRs were performed under the following conditions:94 ◦C for 5 min, annealing at 54 ◦C for 1 min, and DNA synthe-sis at 72 ◦C for 2 min, followed by 28 cycles. To assess the TG2mRNA expression of RAW264.7 cells, RAW264.7 cells were incu-bated with indicated component for 24 h and the mRNA wasanalyzed by RT-PCR. The amplified PCR products were subjectedto electrophoresis in a 2% agarose gel. Sequences for the PCRprimers were: 5′-TGATGACCGGGAGGACATCA-3′, 5′-GATTC TCCAGGTA G AGATCTC-3′ (forward and reverse mouse TG2 primers),5′-TCACTCAA GATT GT CAGCAA-3′, 5′-AGATCCACGACGGACACATT-3′ (forward and reverse mouse GADPH primers).

2.4. TG2 expression constructs and stable transfection

Full length human transglutaminase 2 was cloned intothe pGene/V5-HisB vectors (Invitrogen). RAW264.7 cells weretransfected with the pSwitch vectors containing the mifepristone-inducible system, and then co-transfected with pGene/V5-HisB TG2or pGene/V5-HisB vector alone using TurboFect (Fermentas) withseveral weeks of dual-select with zeocin and hygromycin B.

2.5. siRNA treatment

TG2 siRNA was purchased from Santa Cruz Biotechnology.Transfection was performed with TurboFect (Fermentas) accord-ing to the manufacturer’s instructions. Cells were then treated withdaidzein for an additional 24 h.

2.6. Phagocytosis assay

The human keratinocyte cell line (HaCat) was offered by Profes-sor Jen-Hung Yang of the Department of Dermatology at ChungShan Medical University in Taichung, Taiwan. For induction ofUV-induced apoptosis, HaCat cells were exposed to UV radia-tion at 400 J/m2 using the Spectroline UV Crosslinker (Spectroline,New York, NY, USA). For phagocytosis of apoptotic cells, theUV-irradiated HaCat cells were labeled with carboxy-fluoresceindiacetate (CFDA) followed by incubation with RAW264.7 cells at

10:1 target/macrophage ratio at 37 ◦C for 4 h. After washing, cellson the dish were resuspended in phosphate buffered saline (PBS)solution and 10,000–20,000 cells were analyzed for fluorescenceintensity by flow cytometry (Becton Dickenson, San Jose, CA, USA).

Immun

Ta

2

SdMf(EmbC(sad

2

lCAm

2r

dc1wmouaJ

2

otge<

3

3m

weiActsdd

J.-H. Yen et al. / Molecular

he phagocytosis index was calculated as the number of ingestedpoptotic cells divided by the total number of macrophages.

.7. Western blot analysis

The whole-cell protein lysates were size fractionated by 10%DS-polyacrylamide gels, and then transferred into polyvinyli-ene fluoride (PVDF) membranes (Millipore, Billerica, MA, USA).embranes were blotted overnight at 4 ◦C with one of the

ollowing primary antibodies: rabbit anti-transglutaminase 2Thermo Fisher Scientific), mouse anti- Erk, mouse anti-phospho-rk, mouse anti-JNK, mouse anti-phospho-JNK, mouse anti-p38,ouse anti-phospho-p38 (all from BD Biosciences), mouse anti-

eta-actin (Novus Biologicals), rabbit anti-beta-tubulin (Santaruz Biotechnology), and mouse anti-heat shock protein 60Sigma–Aldrich). Blots were then probed with appropriateecondary horseradish peroxidase (HRP)-conjugated secondaryntibodies (Jackson ImmunoResearch) and detected by an ECLetection system (Millipore, Billerica, MA, USA).

.8. Detection of active Rac1 and CDC42

Cells plated overnight were exposed to daidzein for 24 h oreft untreated as controls to detect GTP-bound forms of Rac1 andDC42. A pull-down assay was performed with the Rac1/CDC42ctivation Assay Kit (Millipore, Billerica, MA, USA) according to theanufacturer’s instruction.

.9. Mitochondrial membrane potential determination withhodamine 123

Mitochondrial membrane potential was measured using rho-amine 123 (Qiu et al., 2013) (Molecular Probes). Cells wereultured in DMEM without phenol red containing 3 �M rhodamine23 for 15 min. The cells were then washed two times with DMEMithout phenol red and resuspended in PBS. Depolarization ofitochondrial membrane potential caused rhodamine 123 to leak

ut of mitochondria into the cytosol where rhodamine 123 wasnquenched, which increased fluorescence. The stained cells werenalyzed by FACSCAN laser flow cytometry (Becton Dickenson, Sanose, CA, USA).

.10. Statistics

The data were analyzed with GraphPad Prism 4 software byne-way analysis of variance (one-way ANOVA) or Student’s t-testo determine the significance of differences between sets of cate-orical data. Data reported were verified in at least three differentxperiments and are represented as the mean ± SEM. A p-value of0.05 was considered to be significant.

. Results

.1. Daidzein enhanced efferocytosis and up-regulated TG2RNA expression

Our previous data showed that the major components of GTHere daidzein, epicatechin, naringin and catechin, which could

nhance efferocytosis (Yen et al., 2010). In the present study wenvestigated the effect of each component of GTH on efferocytosis.fter pretreatment of each component of GTH for 24 h, RAW264.7ells were incubated with CFDA-stained apoptotic cells for 4 h and

hen assayed by flow cytometry. As shown in Fig. 1A, only daidzeinignificantly enhanced the uptake of apoptotic cells in a dose-ependent manner, whereas epicatechin, naringin and catechinid not. The percentage of daidzein (100 �M) treated RAW264.7

ology 60 (2014) 135–142 137

cells to phagocytose the apoptotic cells was found to increase by67.3 ± 17.52% compared with the untreated group. Since mono-cytes and monocyte-derived macrophages play a key role in theremoval of apoptotic cell, we tested the effect of daidzein on effe-rocytosis by human monocyte-derived macrophages (HMDMs).HMDMs were isolated from peripheral blood mononuclear cells(PBMCs) of healthy blood donors by density gradient centrifuga-tion. Fig. 1B shows that daidzein enhanced efferocytosis in theseHMDM compared with the control without daidzein treatment.

We next investigated the effect of daidzein on TG2 expressionbecause TG2 was reported to promote efferocytosis, and defi-ciency of this molecule leads to impaired efferocytic capabilityof macrophages. As shown in Fig. 1C, only daidzein increasedTG2 mRNA expression. Daidzein dose-dependently increased TG2mRNA (Fig. 1D) and protein levels in a time-dependent manner(Fig. 1E). The effect of daidzein on cell viability was also mea-sured by V-FITC and propidium iodide staining and analyzed byflow cytometry. No significant differences were observed amongvarious concentrations of daidzein-treated cells (5–100 �M) andthe untreated control (Fig. 1F). Taken together, these data indicatethat daidzein can enhance efferocytic capability and up-regulateTG2 expression in macrophages.

3.2. The effect of daidzein on efferocytosis was dependent on TG2

To investigate whether the effect of daidzein on efferocytosis isdependent on TG2, we used small-interfering RNA (siRNA) to knockdown the expression of TG2. After transfection with TG2 siRNAfor 48 h, RAW264.7 macrophages were cultured with daidzein for24 h, and decrease of TG2 was observed in RT-PCR and Westernblotting experiments (Fig. 2A and B, respectively). We found thatdaidzein did not enhance efferocytosis when TG2 expression wasinhibited (Fig. 2C). Consistent with the finding in macrophages,daidzein did not enhance efferocytosis in TG2 knockout mouse-derived embryonic fibroblast cell lines (data not shown). Therefore,daidzein-induced TG2 expression appears to be required for effe-rocytosis in these cells.

3.3. Daidzein enhanced Rac1 activity

It is known that activation of the Rho-GTPase Rac1 is requiredfor efferocytosis and TG2 increases active GTP-bound Rac throughblocking Bcr activity (Nakaya et al., 2008; Yi et al., 2009). Becausedaidzein enhanced efferocytosis via up-regulation of TG2, wehypothesized that daidzein might augment Rac1 activity. Toaddress this hypothesis, RAW264.7 macrophages were treated withor without daidzein (20 �M and 100 �M) for 24 h, and PAK1-PBD agarose beads pull-down assays were used to measure Rac-1and CDC42 activity. The results showed that daidzein increasedGTP-bound Rac-1 in a dose-dependent manner and increased GTP-bound CDC42 (Fig. 3A). However, we found that TG2 knockdownabolished effect of daidzein on increasing GTP-bound Rac-1 ratherthan GTP-bound CDC42 (Fig. 3B). The data suggest a mechanism forthe positive effect of daidzein on efferocytosis.

3.4. The signaling event, Erk phosphorylation acting as anupstream of Rac-1 activation, was enhanced after daidzeintreatment

Engulfment of extracellular targets involves a number ofprocesses, including activation of Elmo-1/DOCK 80/CrkII and Erk,thereby leading to Rac-1 activation, cytoskeleton rearrangement,

membrane ruffling, and phagocytic cup formation (Gumiennyet al., 2001; Kress et al., 2007; Flinder et al., 2011). To investigatethe mechanism involved in daidzein-enhanced efferocytosis, westudied the signaling pathways of MAP kinase pathway activity

138 J.-H. Yen et al. / Molecular Immunology 60 (2014) 135–142

Fig. 1. Daidzein increased efferocytosis and TG2 expression specifically in a dose- and time-dependent manner. (A) RAW264.7 macrophages were treated with catechin,epicatechin, naringin or daidzein for various concentrations and then incubated with CFDA-labeled apoptotic cells. The data are expressed as the efferocytosis index andcompared with untreated controls. Only daidzein increased efferocytosis in a dose dependent manner. (B) MDMs were incubated with or without daidzein (100 �M) for 24 hand then incubated with CFDA-labeled apoptotic cells. Efferocytosis activity was analyzed by flow cytometry. Human MDM showed increased efferocytosis after daidzeintreatment. (C) RAW264.7 macrophages were cultured for 24 h in the absence or presence of catechin, epicatechin, naringin or daidzein at 100 �M. TG2 mRNA expressionafter each component treatment was measured by RT-PCR. Only daidzein increased TG2 mRNA expression. (D) RAW264.7 macrophages were cultured with daidzein atconcentrations of 20 and 100 �M for 24 h. Then TG2 mRNA expression was measured by RT-PCR. Daidzein increased TG2 mRNA expression in a dose-dependent manner.(E) RAW264.7 macrophages were cultured with daidzein for different time periods and the TG2 protein level was analyzed by Western blot. Daidzein increased TG2 level ina time-dependent manner. (F) RAW264.7 macrophages were cultured for 24 h in the absence or presence of various concentrations of daidzein, after which the cells werestained with Annexin V-FITC and propidiumiodide and analyzed by flow cytometry. There was no significant difference in cell death among each concentration of daidzeintreatment. The data are expressed as the means ± SEM from three independent experiments each done in triplicate (n ≥ 3, *p < 0.05 compared with control).

J.-H. Yen et al. / Molecular Immunology 60 (2014) 135–142 139

Fig. 2. TG2 knockdown abolished daidzein-induced efferocytosis. RAW264.7 macrophages were transfected with TG2 siRNA for 48 h. Cells were then treated with daidzein(100 �M) for 24 h. (A) TG2 mRNA expression after daidzein teatment was measured by RT-PCR. (B) The protein level of TG2 after daidzein treatment was measured byimmunoblotting. TG2 siRNA suppressed daidzein-induced TG2 expression efficiently. (C)daidzein did not increase efferocytosis. The data are expressed as the efferocytosis index anfrom three independent experiments each done in triplicate (n ≥ 3, *p < 0.05, **p < 0.01 co

Fig. 3. Daidzein enhanced Rac1 and CDC42 activation. RAW264.7 macrophageswere incubated with daidzein (100 �M) for 24 h and GTP-Rac1 and GTP-CDC42 weremeasured by affinity precipitation assay. Glutathione agarose beads conjugated tothe Rac1 effector PAK-1 PBD (PAK-1 PBD, agarose), which specifically binds GTP-Rac1/CDC42 (active), are used to “pull-down” active Rac1/CDC42 from cell lysates.The amounts of GTP-Rac1/CDC42 isolated by GST-PBD immobilized on glutathionebeads were measured by a Rac1/CDC42 immunoblotting. After daidzein treatment,both GTP-Rac1 and GTP-CDC42 were increased (A). Cells transfected withTG2-siRNAwere treated with daidzein. TG2 knockdown abolished effect of daidzein on increas-ing GTP-bound Rac-1 rather than GTP-bound CDC42 (B).The data are expressed fromthree independent experiments each done in triplicate.

Efferocytic capability was analyzed by flow cytometry. After TG2 siRNA treatmentd compared with untreated controls. The data also are expressed as the means ± SEMmpared with control).

which is important for macrophage differentiation. Phosphory-lation of Erk1/2, p38, and JNK was analyzed by immunoblotting.Fig. 4A shows that daidzein enhanced the level of phospho-Erk1/2, in a time-dependent manner, but not phospho-JNK andphospho-p38.

To further study the relationship between TG2 and Erk acti-vation in daidzein treatment, TG2 mRNA was measured in thepresence or absence of Erk inhibitor (PD98059) in macrophages.We found that inhibition of Erk increased TG2 mRNA in theabsence or presence of daidzein (Fig. 4B). PD98059 treatmentalso increased efferocytosis. Furthermore, this inhibitor enhanceddaidzein-increased TG2 expression and efferocytosis (Fig. 4C). Tounderstand the interrelation of TG2 and Erk phosphorylation, weknockeddown TG2 and analyzed phosphorylation level of Erk. Weobserved that TG2 silencing in the RAW264.7 cells also decreasedErk phosphorylation (Fig. 4D). In addition, TG2 overexpressionresulted in increased level of phosphor-Erk (Fig. 4E), suggesting thatErk phosphorylation was affected by TG2. In summary, the resultsshow that daidzein increases both TG2 level and Erk phosphory-lation, and activation of this kinase is dependent on TG2. Erk mayplay a role in the negative regulation of TG2.

3.5. Daidzein decreased mitochondrial membrane potential (�m)

Because it was recently found that increased efferocytosiswas correlated with decreased mitochondrial membrane poten-tial (Park et al., 2011), we tested whether daidzein has similareffect on mitochondria in RAW264.7 macrophages. The resultsshowed that daidzein-treated cells had decreased mitochondrialmembrane potential compared to that in the untreated controls(Fig. 5A).

We next investigated whether daidzein-suppressed mitochon-

drial membrane potential was dependent on TG2. RAW264.7macrophages were treated with or without daidzein for 24 h,and mitochondrial fractions were prepared for TG2 analysis byimmunoblotting. The TG2 protein levels were increased both in

140 J.-H. Yen et al. / Molecular Immunology 60 (2014) 135–142

Fig. 4. Daidzein induced TG2 which acted as an upstream regulator of Erk. (A) RAW264.7 macrophages were cultured with daidzein for different durations, and phosphor-ylation of Erk, p38, and JNK was determined by Western blot. �-actin was blotted as an internal control. Daidzein induced Erk phosphorylation from 8 h until at least 24 h,but not for p38 and JNK. (B) RAW264.7 macrophages were cultured with daidzein in the absence or presence of PD98059 (20 �M) for 24 h. TG2 mRNA expression afterdaidzein treatment was measured by RT-PCR. TG2 mRNA was increased by PD98059 alone or in combination with daidzein. (C) Efferocytic capability was analyzed by flowcytometry. (D) RAW264.7 macrophages were transfected with TG2 siRNA for 48 h. Then cells were treated with daidzein (100 �M) for 24 h. Erk phosphorylation was analyzedby Western blot. TG2 siRNA abolished daidzein-induced Erk phosphorylation. (E) Cell lysates derived from RAW264.7 macrophages with stable transfection of TG2 or emptyv overem 5, **p

tNitkpdd

4

totoe2afpttt2seabca

ector (mock) were subjected to Western blot analysis of Erk phosphorylation. TG2eans ± SEM from three independent experiments done in triplicate (n ≥ 3, *p < 0.0

he cytosol and mitochondria in daidzein-treated cells (Fig. 5B).ext we sought to elucidate whether TG2 induced by daidzein

s involved in the decrease of mitochondrial membrane poten-ial. As shown in Fig. 5C, in daidzein-treated RAW264.7 cells TG2nockdown did not lead to decrease in mitochondrial membraneotential. Taken together, the data showed that in phagocytesaidzein induced TG2 up-regulation and localization in mitochon-ria, resulting in decreased mitochondrial membrane potential.

. Discussion

It is cumulatively accepted that defective clearance of apop-otic cells (efferocytosis) is one of mechanisms in the developmentf in autoimmune and chronic inflammatory diseases, indicatinghat effective efferocytosis may be essential for the maintenancef homeostasis (Park et al., 2011; Hodge et al., 2003; Firesteint al., 1995; Munoz et al., 2005; Baumann et al., 2002; Bosch,011). Because this process is not only important in the innate anddaptive immune response but also in removing autoantigens, theailure of this highly conserved process may lead to disease by ham-ering both the resolution of inflammation and the maintenance ofissue integrity (Odaka et al., 2003; Filardy et al., 2010). Therefore,herapeutic approaches to enhance efferocytosis may be useful toreat the diseases in which this process is impaired (Schutters et al.,013). In this study, we found that daidzein enhanced efferocyto-is in human primary monocyte-derived macrophages. Its effect onfferocytosis depends on the ability to induce TG2 expression, Erk

nd Rac1 activity, and to moderately decrease mitochondrial mem-rane potential. Thus daidzein may have therapeutic potential forhronic inflammatory diseases with impaired efferocytosis and isn important nutraceutical component in our diet.

xpression resulted in increased Erk phosphorylation. The data are expressed as the< 0.01 compared with control).

Daidzein is an isoflavone and is present in a number of plantsand herbs such as soybeans. A previous study showed that daidzeinbinds to cell membrane ER-� to activate the phospholipase C-�2(PLC-�2)/PKC and PI3K/c-Src pathways in osteoblasts (Anne DeWilde et al., 2006). This led to the activation of transcriptionfactors and subsequent expression of several groups of genesinvolved in cell differentiation, proliferation, and modulation forthe actin cytoskeleton that controlled cell adhesion and motility. Inmacrophages, daidzein inhibited the activation of the NF-�B path-way, thereby decreasing the expression of iNOS and the productionof NO (Hamalainen et al., 2007). Indeed, the present experimentsproved that daidzein could enhance efferocytosis via upregulationof TG2 in macrophages. Interestingly, Toth et al. reported thatduring efferocytosis formation of the complex between TG2 andMFG-E8 induces molecular events, including proper apoptotic cellrecognition, receptor clustering, and signaling by integrin-�3 (Tothet al., 2009). Daidzein may exert its effect on efferocytosis directlythrough affecting the association between macrophages andapoptotic cells, or through modification of efferocytosis-associateddownstream signaling events, including the activation of Erk andRac-1 when TG2 and MFG-E8 binds to integrin-�3.

The initial steps following interactions between integrins on themacrophage surface with apoptotic cells in efferocytosis involveSrc-induced phosphorylation and activation of focal adhesionkinase (FAK), ultimately leading to activation of Erk, actin poly-merization, and cytoskeletal rearrangement (David-Pfeuty et al.,1993; Mitra et al., 2005). We speculated that daidzein could

affect the transmission of extracellular signals and lead to thephosphorylation of mitogen-activated protein kinase (MAPKs),which include three major classes, Erk, JNK/SAPK and p38 MAPK.Herein, we found that daidzein triggered only the phosphorylation

J.-H. Yen et al. / Molecular Immun

Fig. 5. Daidzein decreased mitochondrial membrane potential. (A) RAW264.7macrophages were incubated with daidzein (100 �M) for the indicated dura-tions and were then stained with rhodamine 123 (�m loss). Staining intensitywas detected by flow cytometry. (B) RAW264.7 macrophages were cultured withdaidzein for different durations. Mitochondrial and cytosolic fractions were pre-pared, and the expression of TG2, �-tubulin and HSP60 in each sample were analyzedby Western blot. (C) RAW264.7 macrophages were transfected with TG2 siRNA for48 h. Cells were then treated with daidzein (100 �M) for the indicated times. �mafter daidzein treatment was measured by flow cytometry. TG2 siRNA abolishedthe effect of daidzein on �m. The data are expressed as �m levels (%) comparedwith that in control cells. The data also are expressed as the means ± SEM fromtc

obosr2aaoa(esesv

Taiwanese bilateral agreement. Flow cytometry was performed in

hree independent experiments each done in triplicate (n ≥ 3, *p < 0.05, **p < 0.01ompared with control).

f Erk which started at 8 h and persisted over 24 h. Moreover, incu-ating RAW264.7 macrophages with PD98059, a specific inhibitorf MEK1/2, prevented the Erk phosphorylation by daidzein. A recenttudy showed that Erk phosphorylation seems to be a negativeegulator of TG2 expression (Kang et al., 2004; Antonyak et al.,003). Overexpression of TG2 decreased Erk phosphorylation andffected erythroid differentiation (Kang et al., 2004). In addition,ctivation of Erk was sufficient for the Ras-mediated suppressionf retinoic acid-induced TG2 expression, indicating that retinoiccid-induced TG2 expression was antagonized by the Erk pathwayAntonyak et al., 2003). To identify biological actions of daidzein, wevaluated the relationship between Erk activation and TG2 expres-ion in daidzein-treated cells. Indeed, inhibition of Erk by PD98059

nhanced daidzein-induced TG2 mRNA expression. In contrast,uppression of TG2 with small interfering RNA decreased Erk acti-ation, and increased TG2 expression in these RAW264.7 cells led to

ology 60 (2014) 135–142 141

Erk activation. Thus it is likely that Erk activation is TG2 dependent,and that TG2 expression also regulates Erk activation.

The molecular mechanism of efferocytosis involves reorgani-zation of the actin cytoskeleton to produce membrane ruffles,allowing for development of “engulfment synaps” during theapoptotic cell clearance process that differs from phagocytosis ofopsonized particles (Gumienny et al., 2001; Kress et al., 2007;Flinder et al., 2011). This process ultimately requires activationof the Rho-family GTPase Rac. Furthermore, the Erk-uPAR path-way can regulate cytoskeletal dynamics through Rac1 activity incolon carcinoma cells (Vial et al., 2003; Ray et al., 2007). In linewith this, Toth et al. and Yi et al. proposed that TG2 plays a role inRac activation (Toth et al., 2009; Yi et al., 2009). The data presentedhere demonstrate that daidzein not only significantly enhances TG2expression and Erk phosphorylation but also increases Rac1 andCDC42 activation. We propose that signals elicited by daidzein toinduce TG2 expression and Erk phosphorylation may contribute tothe Rac1 and CDC42 activation in vitro.

Macrophages often ingest multiple apoptotic cells sequentially.In addition to how they recognize and ingest these apoptotic cells,how they maintain cellular homeostasis after the ‘meal’ and the fac-tors regulating the ‘engulfment capacity’ still need to be elucidated.Park et al. revealed that crosstalk exists between the mitochon-dria and the engulfment machinery within phagocytes, with the‘sensing’ of the total mitochondrial membrane potential criticallymodulating efferocytic capability. For example, lower mitochon-drial membrane potential enhanced engulfment and vice versa(Park et al., 2011). We found that daidzein decreased mitochon-drial membrane potential in RAW264.7 cells, which may accountfor enhanced efferocytosis in daidzein-treated phagocytes. We alsofound that daidzein increased the levels of TG2 expression inboth cytosol and mitochondria. A recent study showed that TG2localizes in mitochondria under physiological conditions (Malorniet al., 2009). Here we demonstrate that siRNA specifically silencingTG2 expression eliminates daidzein-induced lower mitochondrialmembrane potential. On this basis, we hypothesize that daidzeinmight regulate mitochondrial membrane potential by increasingthe level of TG2 expression.

In conclusion, our data reveal that daidzein is capable ofenhancing efferocytosis, suggesting a new mechanism by whichisoflavones may modulate inflammatory diseases. Accumulation ofapoptotic cells and defect in efferocytosis have also been implicatedin systemic inflammatory diseases, such as chronic obstructive pul-monary disease and systemic lupus erythematosus.

Conflict of interest statement

The authors have declared no conflict of interest.

Acknowledgements

This study was supported by NSC (NSC 99-2811-B-040-005) andChung Shan Medical University Hospital grand (CSH-2010-D-2002and CSH-2012-D-001), Hungarian National Research Fund (OTKAK77587, 83865, 104228, NK105046), TÁMOP 4.2.2.A -11/1/KONV-2012-0023 “VÉD-ELEM” project (implemented through the NewHungary Development Plan co-financed by the European SocialFund and the European Regional Development Fund), a Hungarian

the Instrument Center of Chung Shan Medical University, which issupported by National Science Council, Ministry of Education andChung Shan Medical University.

1 Immun

A

i0

R

A

A

B

B

B

D

F

F

F

F

F

G

G

H

H

H

H

K

K

K

42 J.-H. Yen et al. / Molecular

ppendix A. Supplementary data

Supplementary data associated with this article can be found,n the online version, at http://dx.doi.org/10.1016/j.molimm.2014.4.006.

eferences

nne De Wilde, A., Heberden, C., Chaumaz, G., Bordat, C., Lieberherr, M., 2006.Signaling networks from Gbeta1 subunit to transcription factors and actinremodeling via a membrane-located ER beta-related protein in the rapid actionof daidzein in osteoblasts. J. Cell. Physiol. 209, 786–801.

ntonyak, M.A., McNeill, C.J., Wakshlag, J.J., Boehm, J.E., Cerione, R.A., 2003. Activa-tion of the Ras-ERK pathway inhibits retinoic acid-induced stimulation of tissuetransglutaminase expression in NIH3T3 cells. J. Biol. Chem. 278, 15859–15866.

aumann, I., Kolowos, W., Voll, R.E., Manger, B., Gaipl, U., Neuhuber, W.L., Kirch-ner, T., Kalden, J.R., Herrmann, M., 2002. Impaired uptake of apoptotic cells intotingible body macrophages in germinal centers of patients with systemic lupuserythematosus. Arthritis Rheum. 46, 191–201.

osch, X., 2011. Systemic lupus erythematosus and the neutrophil. N. Eng. J. Med.365, 758–760.

rugnera, E., Haney, L., Grimsley, C., Lu, M., Walk, S.F., Tosello-Tampont, A.C., Macara,I.G., Madhani, H., Fink, G.R., Ravichandran, K.S., 2002. Unconventional Rac-GEFactivity is mediated through the Dock180-ELMO complex. Nat. Cell Biol. 4,74–582.

avid-Pfeuty, T., Bagrodia, S., Shalloway, D., 1993. Differential localization patternsof myristoylated and nonmyristoylated c-Src proteins in interphase and mitoticc-Src overexpresser cells. J. Cell Sci. 105 (Pt 3), 613–628.

esus, L., Piacentini, M., 2002. Transglutaminase 2: an enigmatic enzyme withdiverse functions. Trends Biochem. Sci. 27, 534–539.

ilardy, A.A., Pires, D.R., Nunes, M.P., Takiya, C.M., Freire-de-Lima, C.G.,Ribeiro-Gomes, F.L., DosReis, G.A., 2010. Proinflammatory clearance of apop-totic neutrophils induces an IL-12(low)IL-10(high) regulatory phenotype inmacrophages. J. Immunol. 185, 2044–2050.

irestein, G.S., Yeo, M., Zvaifler, N.J., 1995. Apoptosis in rheumatoid arthritis syn-ovium. J. Clin. Invest. 96, 1631–1638.

linder, L.I., Timofeeva, O.A., Rosseland, C.M., Wierod, L., Huitfeldt, H.S., Skarpen,E., 2011. EGF-induced ERK-activation downstream of FAK requires rac1-NADPHoxidase. J. Cell. Physiol. 226, 2267–2278.

ranke, A.A., Custer, L.J., Wang, W., Shi, C.Y., 1998. HPLC analysis of isoflavonoids andother phenolic agents from foods and from human fluids. Proc. Soc. Exp. Biol.Med. (New York, N.Y) 217, 263–273.

arcia Palacios, V., Robinson, L.J., Borysenko, C.W., Lehmann, T., Kalla, S.E., Blair,H.C., 2005. Negative regulation of RANKL-induced osteoclastic differentia-tion in RAW264.7 Cells by estrogen and phytoestrogens. J. Biol. Chem. 280,13720–13727.

umienny, T.L., Brugnera, E., Tosello-Trampont, A.C., Kinchen, J.M., Haney, L.B., Nishi-waki, K., Walk, S.F., Nemergut, M.E., Macara, I.G., Francis, R., Schedl, T., Qin, Y.,Van Aelst, L., Hengartner, M.O., Ravichandran, K.S., 2001. CED-12/ELMO, a novelmember of the CrkII/Dock180/Rac pathway, is required for phagocytosis and cellmigration. Cell 107, 27–41.

amalainen, M., Nieminen, R., Vuorela, P., Heinonen, M., Moilanen, E., 2007.Anti-inflammatory effects of flavonoids: genistein, kaempferol, quercetin, anddaidzein inhibit STAT-1 and NF-kappaB activations, whereas flavone, isorham-netin, naringenin, and pelargonidin inhibit only NF-kappaB activation alongwith their inhibitory effect on iNOS expression and NO production in activatedmacrophages. Mediat. Inflamm. 2007, 45673.

enson, P.M., Hume, D.A., 2006. Apoptotic cell removal in development and tissuehomeostasis. Trends Immunol. 27, 244–250.

enson, P.M., Bratton, D.L., Fadok, V.A., 2001. Apoptotic cell removal. Curr. Biol. 11,R795–R805.

odge, S., Hodge, G., Scicchitano, R., Reynolds, P.N., Holmes, M., 2003. Alveo-lar macrophages from subjects with chronic obstructive pulmonary diseaseare deficient in their ability to phagocytose apoptotic airway epithelial cells.Immunol. Cell Biol. 81, 289–296.

ang, S.K., Lee, J.Y., Chung, T.W., Kim, C.H., 2004. Overexpression of transglutamin-ase 2 accelerates the erythroid differentiation of human chronic myelogenousleukemia K562 cell line through PI3 K/Akt signaling pathway. FEBS Lett. 577,361–366.

im, J.W., Jin, Y.C., Kim, Y.M., Rhie, S., Kim, H.J., Seo, H.G., Lee, J.H., Ha, Y.L., Chang,K.C., 2009. Daidzein administration in vivo reduces myocardial injury in a rat

ischemia/reperfusion model by inhibiting NF-kappaB activation. Life Sci. 84,227–234.

ress, H., Stelzer, E.H., Holzer, D., Buss, F., Griffiths, G., Rohrbach, A., 2007. Filopodiaact as phagocytic tentacles and pull with discrete steps and a load-dependentvelocity. Proc. Natl. Acad. Sci. U. S. A. 104, 11633–11638.

ology 60 (2014) 135–142

Malorni, W., Farrace, M.G., Matarrese, P., Tinari, A., Ciarlo, L., Mousavi-Shafaei, P.,D’Eletto, M., Di Giacomo, G., Melino, G., Palmieri, L., Rodolfo, C., Piacentini, M.,2009. The adenine nucleotide translocator 1 acts as a type 2 transglutamin-ase substrate: implications for mitochondrial-dependent apoptosis. Cell DeathDiffer. 16, 1480–1492.

Michlewska, S., McColl, A., Rossi, A.G., Megson, I.L., Dransfield, I., 2007. Clearance ofdying cells and autoimmunity. Autoimmunity 40, 267–273.

Mitra, S.K., Hanson, D.A., Schlaepfer, D.D., 2005. Focal adhesion kinase: in commandand control of cell motility. Nat. Rev. 6, 56–68.

Munoz, L.E., Herrmann, M., Gaipl, U.S., 2005. An impaired detection and clearance ofdying cells can lead to the development of chronic autoimmunity. Z. Rheumatol.64, 370–376.

Nakaya, M., Kitano, M., Matsuda, M., Nagata, S., 2008. Spatiotemporal activationof Rac1 for engulfment of apoptotic cells. Proc. Natl. Acad. Sci. U. S. A. 105,9198–9203.

Odaka, C., Mizuochi, T., Yang, J., Ding, A., 2003. Murine macrophages producesecretory leukocyte protease inhibitor during clearance of apoptotic cells:implications for resolution of the inflammatory response. J. Immunol. 171,1507–1514.

Park, S.Y., Kang, K.B., Thapa, N., Kim, S.Y., Lee, S.J., Kim, I.S., 2008. Requirement ofadapter protein GULP-6 for stabilin-2 mediated cell corpse engulfment. J. Biol.Chem. 283, 10593–10600.

Park, D., Han, C.Z., Elliott, M.R., Kinchen, J.M., Trampont, P.C., Das, S., Collins,S., Lysiak, J.J., Hoehn, K.L., Ravichandran, K.S., 2011. Continued clearance ofapoptotic cells critically depends on the phagocyte Ucp2 protein. Nature 477,220–224.

Qiu, J., Tan, Y.W., Hagenston, A.M., Martel, M.A., Kneisel, N., Skehel, P.A., Wyllie, D.J.,Bading, H., Hardingham, G.E., 2013. Mitochondrial calcium uniporter Mcu con-trols excitotoxicity and is transcriptionally repressed by neuroprotective nuclearcalcium signals. Nat. Commun. 4, 2034.

Ravichandran, K.S., 2011. Beginnings of a good apoptotic meal: the find-me andeat-me signaling pathways. Immunity 35, 445–455.

Ray, R.M., Vaidya, R.J., Johnson, L.R., 2007. MEK/ERK regulates adherens junctionsand migration through Rac1. Cell Motil. Cytoskeleton 64, 143–156.

Savill, J., Dransfield, I., Gregory, C., Haslett, C., 2002. A blast from the past: clear-ance of apoptotic cells regulates immune responses. Nat. Rev. Immunol. 2, 965–975.

Schutters, K., Kusters, D.H., Chatrou, M.L., Montero-Melendez, T., Donners, M.,Deckers, N.M., Krysko, D.V., Vandenabeele, P., Perretti, M., Schurgers, L.J.,Reutelingsperger, C.P., 2013. Cell surface-expressed phosphatidylserine as ther-apeutic target to enhance phagocytosis of apoptotic cells. Cell Death Differ. 20,49–56.

Su, H.P., Nakada-Tsukui, K., Tosello-Trampont, A.C., Li, Y., Bu, G., Henson, P.M.,Ravichandran, K.S., 2002. Interaction of CED-6/GULP, an adapter proteininvolved in engulfment of apoptotic cells with CED-1 and CD91/low densitylipoprotein receptor-related protein (LRP). J. Biol. Chem. 277, 1772–11779.

Szondy, Z., Sarang, Z., Molnar, P., Nemeth, T., Piacentini, M., Mastroberardino, P.G.,Falasca, L., Aeschlimann, D., Kovacs, J., Kiss, I., Szegezdi, E., Lakos, G., Rajnavolgyi,E., Birckbichler, P.J., Melino, G., Fesus, L., 2003. Transglutaminase 2−/− mice reveala phagocytosis-associated crosstalk between macrophages and apoptotic cells.Proc. Natl. Acad. Sci. U. S. A. 100, 7812–7817.

Szondy, Z., Korponay-Szabo, I., Kiraly, R., Fesus, L., 2011. Transglutaminase 2 dysfunc-tions in the development of autoimmune disorders: celiac disease and TG2−/−

mouse. Adv Enzymol. Relat. Areas Mol. Biol. 78, 295–345.Tosello-Trampont, A.C., Nakada-Tsukui, K., Ravichandran, K.S., 2003. Engulfment of

apoptotic cells is negatively regulated by Rho-mediated signaling. J. Biol. Chem.278, 49911–49919.

Toth, B., Garabuczi, E., Sarang, Z., Vereb, G., Vamosi, G., Aeschlimann, D., Blasko, B.,Becsi, B., Erdodi, F., Lacy-Hulbert, A., Zhang, A., Falasca, L., Birge, R.B., Balajthy,Z., Melino, G., Fesus, L., Szondy, Z., 2009. Transglutaminase 2 is needed for theformation of an efficient phagocyte portal in macrophages engulfing apoptoticcells. J. Immunol. 182, 2084–2092.

Vanden Berghe, W., Dijsselbloem, N., Vermeulen, L., Ndlovu, M.N., Boone, E.,Haegeman, G., 2006. Attenuation of mitogen- and stress-activated proteinkinase-1-driven nuclear factor-kappaB gene expression by soy isoflavones doesnot require estrogenic activity. Cancer Res. 66, 4852–4862.

Vial, E., Sahai, E., Marshall, C.J., 2003. ERK-MAPK signaling coordinately regulatesactivity of Rac1 and RhoA for tumor cell motility. Cancer Cell 4, 67–79.

Wang, S.W., Chen, Y., Joseph, T., Hu, M., 2008. Variable isoflavone content of redclover products affects intestinal disposition of biochanin A, formononetin,genistein, and daidzein. J. Altern. Complement. Med. (New York, N.Y.) 14,287–297.

Yen, J.H., Yang, D.J., Chen, M.C., Hsieh, Y.F., Sun, Y.S., Tsay, G.J., 2010. Glycine

tomentella Hayata inhibits IL-1beta and IL-6 production, inhibits-9 activity, andenhances RAW264.7 macrophage clearance of apoptotic cells. J. Biomed. Sci. 17,83.

Yi, S.J., Groffen, J., Heisterkamp, N., 2009. Transglutaminase 2 regulates the GTPase-activating activity of Bcr. J. Biol. Chem. 284, 35645–35651.