Scavenger receptor class-A has a central role incerebral ischemia–reperfusion injury

Chen Lu1,2, Fang Hua3, Li Liu2, Tuanzhu Ha1, John Kalbfleisch4, John Schweitzer5,Jim Kelley6, Race Kao1, David Williams1 and Chuanfu Li1

1Department of Surgery, East Tennessee State University, Johnson City, Tennessee, USA; 2Department ofGeriatrics, The First Affiliated Hospital of Nanjing Medical University, Nanjing, China; 3Brain ResearchLaboratory, Department of Emergency Medicine, Emory University School of Medicine, Atlanta, Georgia,USA; 4Department of Biometry and Medical Computing, East Tennessee State University, Johnson City,Tennessee, USA; 5Department of Pathology, East Tennessee State University, Johnson City, Tennessee, USA;6Department of Internal Medicine, East Tennessee State University, Johnson City, Tennessee, USA

The innate immune response is involved in the pathophysiology of cerebral ischemia–reperfusion(I/R) injury. Recent evidence suggests that scavenger receptors have a role in the induction of innateimmunity. In this study, we examined the role of scavenger receptor A (SR-A) in focal cerebral I/Rinjury. Both SR-A�/� mice (n = 10) and age-matched wild-type (WT) mice (n = 9) were subjectedto focal cerebral ischemia (60 minutes), followed by reperfusion (for 24 hours). Infarct size wasdetermined by TTC (triphenyltetrazolium chloride) staining. The morphology of neurons in the brainsections was examined by Nissl’s staining. Activation of intracellular signaling was analyzed bywestern blot. Cerebral infarct size in SR-A�/� mice was significantly reduced by 63.9% comparedwith WT mice after cerebral I/R. In SR-A�/� mice, there was less neuronal damage in thehippocampus compared with WT mice. Levels of FasL, Fas, FADD, caspase-3 activity, and terminaldeoynucleotidyl transferase-mediated 20-deoxyuridine 50-triphosphate-biotin nick end labeling-positive apoptotic cells were significantly increased in WT mice after cerebral I/R, but not in SR-A�/� mice. Cerebral I/R increased nuclear factor-jB activation in WT mice, but not in SR-A�/� mice.These data suggest that SR-A has a central role in cerebral I/R injury and that suppression of SR-Amay be a useful approach for ameliorating brain injury in stroke patients.Journal of Cerebral Blood Flow & Metabolism (2010) 30, 1972–1981; doi:10.1038/jcbfm.2010.59; published online28 April 2010

Keywords: apoptosis; cerebral ischemia–reperfusion; inflammatory response; innate immunity; scavengerreceptor; SR-A

Introduction

A growing body of evidence suggests that innateimmune and inflammatory responses are involved inbrain ischemia–reperfusion (I/R) injury (Jordan et al,2008; Brea et al, 2009). Indeed, we along with othershave shown that modulation of innate immuneresponses significantly attenuated cerebral I/R injury(Caso et al, 2007; Tang et al, 2007; Hua et al, 2007b,

2008, 2009). However, the mechanisms by whichinnate immune and inflammatory responses areactivated in the ischemic brain remain unclear.

Pattern recognition receptors (PRRs), such asToll-like receptors (TLRs), have a critical role in theinduction of innate immunity and inflammatoryresponses (Medzhitov et al, 1997). Toll-like receptor-mediated signaling predominately activates nuclearfactor-kappaB (NF-kB), which is an important trans-cription factor controlling innate immune and inflam-matory cytokine gene expressions (Medzhitov et al,1997). We, along with others, have shown that TLR4deficiency protects the brain from cerebral ischemicinjury (Caso et al, 2007; Tang et al, 2007; Hua et al,2007b, 2008, 2009).

The macrophage scavenger receptor A (SR-A alsoknown as CD204) was initially discovered becauseof its ability to bind and internalize modified low-density lipoprotein (Goldstein et al, 1979). Subse-quently, SR-A has been shown to recognize and clear

Received 30 November 2009; revised 18 February 2010; accepted29 March 2010; published online 28 April 2010

Correspondence: Dr C Li, Department of Surgery, East TennesseeState University, Campus Box 70575, Johnson City, TN 37614-0575, USA.E-mail: Li@etsu.edu

This study was supported, in part, by NIH RO1HL071837 to CL,

NIH RO1GM53552 to DLW, NIH RO1GM083016 to CL and DLW, by

AHA SDG 0830481N to FH and by AHA grant 09GRNT2020111 to

RLK.

Journal of Cerebral Blood Flow & Metabolism (2010) 30, 1972–1981& 2010 ISCBFM All rights reserved 0271-678X/10 $32.00

www.jcbfm.com

modified host components, apoptotic cells, andpathogens (Platt and Gordon, 2001; Greaves andGordon, 2009). Numerous studies have shown thatSR-A has a critical role in the induction of innateimmune and inflammatory responses by recognitionof exogenous pathogen-associated molecular pat-terns and endogenous ligands (Zhu et al, 2001;Hollifield et al, 2007; Limmon et al, 2008). However,the intracellular signaling mediated by SR-A is stillunclear. Most studies have indicated that SR-A doesnot directly transduce a signal into the cell (Cotenaet al, 2004), because the intracellular domain of SR-Alacks signaling motifs (Bowdish and Gordon, 2009).Several other studies have suggested that SR-Aintracellular domains can be phosphorylated, whichmay facilitate the interaction of the SR-A transmem-brane domain with intracellular signaling compo-nents (Fong and Li, 1999; Hsu et al, 2001; Nikolicet al, 2007; Todt et al, 2008). Recent evidence alsosuggests that SR-A is a coreceptor for TLRs tofacilitate innate immune recognition and response,resulting in an overexuberant response (Seimon et al,2006). For example, TLR ligands synergize withSR-A to mediate bacterial phagocytosis (Amiel et al,2009), induce SR-A expression (Xu et al, 2007), andpromote SR-A binding to the TLR4 ligand, lipopoly-saccharide (Xu et al, 2007). Scavenger receptor Ainteracts with TLR4 to promote a proinflammatory,proapoptotic phenotype in lipopolysaccharide-exposed macrophages (Seimon et al, 2006). In addi-tion, SR-A suppresses prosurvival signaling path-ways, such as interferon regulatory factor-3-mediatedinterferon-b production (Seimon et al, 2006). Incontrast, SR-A ligands trigger apoptosis in theendoplasmic reticulum-stressed macrophages by co-operating with TLR4 (Seimon et al, 2006) and serveas a negative regulator of TLR4 in mediating immuneresponses (Yi et al, 2009). Collectively, these datasuggest that SR-A could contribute to cerebralischemic injury by promoting the activation of innateimmune and inflammatory responses (Zhu et al,2001; Hollifield et al, 2007; Limmon et al, 2008), byacting as a coreceptor to TLR4 (Seimon et al, 2006),and by suppressing the prosurvival signaling path-way (Seimon et al, 2006).

In this study, we examined the role of SR-A infocal cerebral I/R injury. We observed that SR-Adeficiency significantly protected the brain fromfocal cerebral ischemic injury. This is the first reportto show that SR-A contributes to brain damagecaused by ischemic stroke. Our data suggest thatsuppression of SR-A could be a strategy for preven-tion and therapy in stroke patients.

Materials and methods

Animals

Breeding pairs of macrophage SR-A-deficient mice on theC57BL/6J background were provided by Dr Siamon Gordon(Sir William Dunn School of Pathology, Oxford University)

(Cotena et al, 2004). A breeding colony was establishedand maintained in the Division of Laboratory AnimalResources, Quillen College of Medicine, ETSU (EastTennessee State University). Male wild-type (WT) C57BLmice were obtained from Jackson Laboratory (Bar Harbor,ME, USA). The experiments outlined in this articleconform to the Guide for the Care and Use of LaboratoryAnimals published by the National Institutes of Health(NIH publication no. 85-23, revised 1996). The animal careand experimental protocols were approved by the ETSUCommittee on Animal Care.

Focal Cerebral Ischemia–Reperfusion

Focal cerebral I/R was induced by occlusion of the middlecerebral artery on the left side as described in our previousstudies (Hua et al, 2008, 2009). Briefly, mice were subjectedto anesthesia by 5.0% isoflurane and anesthesia wasmaintained by inhalation of 1.5% to 2% isoflurane drivenby 100% oxygen flow. Mice were ventilated (110 breaths perminute with volume 0.5 mL), and body temperature wasregulated at 37.01C by surface water heating. After skinincision, the left common carotid artery, the external carotidartery, and the internal carotid artery were carefullyexposed. Microvascular aneurysm clips were applied tothe left common carotid artery and the internal carotidartery. A coated 6-0 filament (6023PK, Doccol, Redlands,CA, USA) was introduced into an arteriotomy hole, feddistally into the internal carotid artery. After the internalcarotid artery clamp was removed, the filament wasadvanced 11 mm from the carotid bifurcation, and focalcerebral ischemia started. After ischemia for 60 minutes, thefilament and the common carotid artery clamp were gentlyremoved (reperfusion starts). The collar suture at the base ofthe external carotid artery stump was tightened. The skinwas closed, anesthesia discontinued, and the animal wasallowed to recover in prewarmed cages. Control miceunderwent a neck dissection and coagulation of the externalcarotid artery, but no occlusion of the middle cerebral artery.

Examination of Infarct Size

Infarct size was determined as described previously (Huaet al, 2008, 2009). After ischemia (60 minutes), followed byreperfusion (24 hours), mice were killed and perfused withice-cold phosphate-buffered saline through the ascendingaorta. Brains were removed and sectioned coronally into 2-mm-thick slices. The slices were stained with 2% TTC(triphenyltetrazolium chloride) solution at 371C for 15 min-utes, followed by fixation with 10% formalin neutralbuffer solution (pH 7.4). The infarct areas were tracedand quantified using an image-analysis system. Unstainedareas (pale color) were defined as ischemic lesions. Theareas of infarction and the areas of both hemispheres werecalculated for each brain slice. An edema index wascalculated by dividing the total volume of the left hemi-sphere by the total volume of the right hemisphere. Theactual infarct volume adjusted for edema was calculated bydividing the infarct volume by the edema index. Infarctvolumes are expressed as percentage of the total brainvolume±s.e.m.

SR-A mediates brain injuryC Lu et al

1973

Journal of Cerebral Blood Flow & Metabolism (2010) 30, 1972–1981

Evaluation of Neuronal Damage in HippocampalFormation

Neuronal damage in brain sections were determined byNissl’s method as described in our previous studies (Huaet al, 2008, 2009). Paraffin sections cut in the coronal planeat B1.5 mm behind the bregma with a thickness of 7 mmwere deparaffinized and then stained with 0.1% cresylviolet for 2 minutes. The sections were evaluated usinglight microscopy.

Western Blots

Cellular proteins were prepared from the brain tissuesand immunoblots were performed as described previously(Li et al, 2003; Hua et al, 2007a, b, 2008, 2009). Thecellular proteins were separated by SDS-PAGE andtransferred onto Hybond ECL membranes (AmershamPharmacia, Piscataway, NJ, USA). The ECL membraneswere incubated with the appropriate primary antibody(anti-phospho-Jun N-terminal kinase (JNK), anti-JNK,anti-phospho-MKK4/7, anti-MKK4/7, anti-phospho-IkBa(Cell Signaling Technology, Beverly, MA, USA), anti-Fas,anti-FasL, and anti-Fas-associated protein with deathdomain (FADD) (Santa Cruz Biotechnology, Santa Cruz,CA, USA)), respectively, followed by incubation withperoxidase-conjugated secondary antibodies (Cell Signal-ing Technology). The signals were detected using the ECLsystem (Amersham Pharmacia). To control for lane loading,the same membranes were probed with anti-GAPDH(glyceraldehyde-3-phosphate dehydrogenase, Biodesign,Saco, ME, USA) after being washed with the strippingbuffer. The signals were quantified by scanning densito-metry using a Bio-Image Analysis System (Bio-Rad,Hercules, CA, USA). The results from each experimentalgroup were expressed as relative integrated intensitycompared with that of control hearts measured at the sametime.

Electrophoretic Mobility Shift Assay

Nuclear proteins were isolated from ischemic cerebralhemispheres as described previously (Li et al, 2003; Huaet al, 2007a, b, 2008, 2009). Nuclear factor-kB bindingactivity was determined by electrophoretic mobility shiftassay as described in our previous studies. Nuclear factor-kB binding activity was examined by electrophoreticmobility shift assay in a 15 mL binding reaction mixturecontaining 15 mg of nuclear proteins and 35 fmol of [g-32P]-labeled double-stranded NF-kB consensus oligonucleotide.A supershift assay using antibodies to p65 and p50 wasperformed to confirm NF-kB binding specificity andsubunits as described previously (Li et al, 2003).

Caspase-3 Activity Assay

Caspase-3 activity in the brain tissue was measured usinga Caspase-Glo assay kit (Promega, Madison, WI, USA)according to the manufacturer’s protocol.

Terminal Deoynucleotidyl Transferase-Mediated20-Deoxyuridine 50-Triphosphate-Biotin Nick EndLabeling Analysis

Terminal deoynucleotidyl transferase-mediated 20-deoxy-uridine 50-triphosphate-biotin nick end labeling (TUNEL)staining was performed using the In Situ Cell DeathDetection Kit (Promega) according to the manufacturer’sprotocol. The number of total cells and TUNEL-positivecells were counted throughout borders between infarctionand normal areas in the fields of the hippocampus (Huaet al, 2007b). The number of TUNEL-positive cells wasexpressed as the percentage of total counted cells.

Statistical Analysis

Data are expressed as mean±s.e. Comparisons of databetween groups were made using one-way ANOVA (analysisof variance), and Tukey’s procedure for multiple range testswas performed. A P-value < 0.05 was considered significant.

Results

SR-A Deficiency Reduced Cerebral Infarct Size AfterIschemia–Reperfusion

We have previously shown that TLR4 deficiencyattenuated cerebral infarction after I/R (Hua et al,2007b, 2009). Scavenger receptor A is consideredto be a PRR (Zhu et al, 2001; Hollifield et al, 2007;Limmon et al, 2008); therefore, we examinedwhether SR-A is involved in cerebral I/R injury. Asshown in Figure 1, focal cerebral I/R resulted in largeinfarcts in WT mice. Scavenger receptor A–deficientmice show significantly reduced cerebral infarctvolumes (64.2%), when compared with WT I/R mice(6.85%±1.91% versus 19.14%±2.85%, P < 0.05).

SR-A Deficiency Attenuated Neuronal Damagein Hippocampal Formation After CerebralIschemia–Reperfusion

We examined neuronal damage in hippocampalformation after cerebral I/R. Nissl’s staining showedneuronal damage in the CA1 field of the hippo-campal formation in WT-I/R mice characterized bynumerous, often confluent zones of shrunken cellbodies accompanied by shrunken and pyknoticnuclei (Figure 2). Similar changes were observed inthe dentate gyrus and CA4. In contrast, the neuronsin the CA1, CA4, and DG fields in SR-A-deficientmice brain showed less neuronal damage and themorphology was well preserved (Figure 2).

Ischemia–Reperfusion-Induced Nuclear Factor-jBBinding Activity is Prevented by SR-A Deficiency

Nuclear factor-kB is an important transcriptionfactor that regulates inflammatory cytokine geneexpression (Medzhitov et al, 1997). Inhibition of

SR-A mediates brain injuryC Lu et al

1974

Journal of Cerebral Blood Flow & Metabolism (2010) 30, 1972–1981

NF-kB binding activity has been shown to signifi-cantly attenuate cerebral ischemic injury (Schneideret al, 1999a, b). We examined I/R-induced NF-kBbinding activity in WT- and SR-A-deficient mice. Asshown in Figure 3A, cerebral I/R significantlyincreased NF-kB binding activity by 46.1%, com-pared with sham control (1.52±0.09 versus 1.04±0.141, P < 0.05). However, NF-kB binding activity inSR-A-deficient mice was not increased after cerebralI/R and was significantly lower than that observed inWT mice (1.04±0.05 versus 1.52±0.09, P < 0.05).The Supershift assay showed the specificity of NF-kB binding activity with p65 and p50 subunits(Figure 3B).

Ischemia–Reperfusion-Induced Phosphorylation ofIRAK and IjBa is Prevented in the Brain Tissue ofSR-A-Deficient Mice

In the NF-kB activation pathway, interleukin receptor-associated kinase (IRAK) and IkBa phosphorylation areimportant upstream regulators of NF-kB activation andnuclear translocation (Medzhitov et al, 1997). Therefore,we examined IRAK and IkBa phosphorylation in thebrain tissues after cerebral I/R. Figures 3C and 3D showthat the levels of p-IRAK and p-IkBa were significantlyincreased by 54.0% (1.09±0.04 versus 0.71±0.03) and

128.7% (1.76±0.17 versus 0.77±0.25), respectively, inWT mice after cerebral I/R compared with sham control.In contrast, cerebral I/R did not increase IRAK and IkBaphosphorylation in the brain tissues of SR-A-deficientmice. In fact, levels of p-IRAK (0.67±0.05 versus1.09±0.04) and p-IkBa (0.70±0.06 versus 1.76±0.17)in SR-A�/� brains were significantly lower than that inWT mice after I/R.

Ischemia–Reperfusion-Associated Increases in FasL,Fas, and FADD are not Observed in the Brain Tissuefrom SR-A-Deficient Mice

The Fas-mediated apoptotic signaling pathway hasan important role in cerebral ischemic injury

WT

SR-A-/-

#

Per

cent

age

of in

farc

tion(

%)

25

15

20

10

5

0WT SR-A-/-

Figure 1 Reduced cerebral infarct size in SR-A�/� mice after I/Rinjury. SR-A-deficient (n = 10) and C57BL (WT) mice (n = 9)were subjected to focal cerebral ischemia (60 minutes), followedby reperfusion (24 hours). Infarct size was determined by TTCstaining and expressed as the percentage of actual infarct sizevolume in the total cerebral volume. Representative brainsections stained with TTC are shown at the top. #P < 0.05compared with WT. I/R, ischemia–reperfusion; SR-A, scavengerreceptor A; TTC, triphenyltetrazolium chloride; WT, wild type.

CA1 (40x)4x

WTSham

WTI/R

CA1(40x)4x

SR-A-/-

Sham

SR-A-/-

I/R

Figure 2 SR-A deficiency attenuated neuronal damage in HFafter cerebral I/R. SR-A-deficient (n = 6) and C57BL (WT) mice(n = 6) were subjected to focal cerebral ischemia (60 minutes),followed by reperfusion (24 hours). Sham surgically operatedmice (n = 4 per group) served as sham control. Brain sectionswere stained with 0.1% cresyl violet. The hippocampal neuronsin the CA1 field showed extreme changes in WT mice, withabnormal nuclei and neuronal shrinkage. In contrast, theneurons in this field showed less neuronal damage and themorphology of neurons was well preserved in the SR-A-deficientmice. HF, hippocampal formation; I/R, ischemia–reperfusion;SR-A, scavenger receptor A; WT, wild type.

SR-A mediates brain injuryC Lu et al

1975

Journal of Cerebral Blood Flow & Metabolism (2010) 30, 1972–1981

(Dzietko et al, 2008). Inhibition of apoptotic signal-ing pathways has been shown to protect the brainfrom ischemic injury (Dzietko et al, 2008). Weexamined the effect of SR-A deficiency on theFas-mediated apoptotic signaling pathway aftercerebral I/R. As shown in Figures 4A–4C, cerebralI/R significantly increased the levels of Fas, FasL,and FADD in the WT brain tissues by 33.3, 49.6, and52.2%, respectively, when compared with shamcontrol. In striking contrast, SR-A deficiency pre-vented I/R-induced Fas, FasL, and FADD in the braintissues (Figures 4A–4C).

SR-A Deficiency Effectively Prevented Ischemia–Reperfusion-Induced MKK4 and JNK Phosphorylationin the Brain Tissues

Activation of the JNK signaling pathway has beenshown to contribute to cerebral ischemic injury (Gaoet al, 2005). In addition, ligation of SR-A activates the

JNK signaling pathway (Ricci et al, 2004). MKK4 isan upstream regulator of JNK (Gao et al, 2005). Weexamined the effects of SR-A on MKK4 and JNKphosphorylation in the brain tissues after cerebralI/R. As expected, cerebral I/R increased the levels ofphosphorylated-JNK by 58.7% and phosphorylatedMkk4 by 57.7%, when compared with WT shamcontrols (Figures 5A and 5B). In striking contrast, thelevels of p-JNK and p-MKK4 were essentially normalin SR-A-deficient mice after cerebral I/R.

Attenuation of Ischemia–Reperfusion Induced BrainCaspase-3 Activity in SR-A-Deficient Mice

We examined caspase-3 activity in the brain tissuesafter cerebral I/R. Figure 6A shows that I/R increasedcaspase-3 activity by 57.9% in WT mice comparedwith sham control. In contrast, cerebral I/R did notincrease caspase-3 activity in SR-A-deficient micecompared with sham control mice. The TUNEL assay

SR-A-/-WTS

NF-κB

ShamI/R * *

NF

-κB

bin

ding

Act

ivity

Inte

grat

ed In

tens

ity

2.5e+5

2.0e+5

1.0e+5

1.5e+5

5.0e+4

0.0

WT SR-A-/-

NF-κB

Binding

Unlabe

led N

FκB

Unlabe

led A

P-ll

Anti-p

65

Anti-p

50

Anti-p

65 &

p-5

0

p65p50

I/R S I/R

Sham I/R I/RSham

ShamI/R

ShamI/R

WT

P-IRAK

IRAK

**

P-I

RA

K/IR

AK

1.2

1.4

0.8

1.0

0.4

0.6

0.0

0.2

Sham I/R Sham I/R

WT SR-A-/-

Sham I/R Sham I/R

WT SR-A-/-

**P

-IκB

-α/Iκ

B-α

2.0

2.5

1.5

1.0

0.0

0.5

SR-A-/-

S I/R S I/RWT

P-IκBα

IκBα

SR-A-/-

S I/R S I/R

Figure 3 I/R-increased IRAK and IkBa phosphorylation and NF-kB binding activity are effectively prevented in SR-A deficiency. SR-A-deficient (n = 6) and WT mice (n = 6) were subjected to focal cerebral ischemia (45 minutes)/reperfusion (6 hours). Shamsurgically operated mice served as sham control (n = 4 per group). Nuclear and cytoplasmic proteins were isolated from brain tissuesfor analysis of NF-kB binding activity by EMSA (A), supershift assay (B), and for examination of phospho-IRAK (C) and phospho-IkBa(D) by western blot. EMSA, electrophoretic mobility shift assay; I/R: ischemia–reperfusion; IRAK, interleukin receptor-associatedkinase; NF-kB, nuclear factor-kB; S, sham; SR-A, scavenger receptor A; WT, wild type;. *P < 0.05 compared with indicated groups.

SR-A mediates brain injuryC Lu et al

1976

Journal of Cerebral Blood Flow & Metabolism (2010) 30, 1972–1981

WT

GAPDH

FasL

ShamI/R

ShamI/R

ShamI/R

* *

FAS

-L/G

AP

DH

1.4

1.6

0.8

1.0

1.2

0.6

0.0

0.2

0.4

Sham I/R Sham I/R

WT SRA-/-

Sham I/R Sham I/R

WT SR-A-/-

Sham I/R Sham I/R

WT SR-A-/-

**

Fas/

GA

PD

H

0.5

0.3

0.4

0.2

0.1

0.0

* *

FAD

D/G

AP

DH

1.0

1.2

0.8

0.4

0.6

0.0

0.2

SR-A-/-

S I/R S I/R

WT

GAPDH

Fas

SR-A-/-

S I/R S I/RWT

GAPDH

FADD

SR-A-/-

S I/R S I/R

Figure 4 I/R-increased FasL, Fas, and FADD are attenuated by SR-A deficiency. SR-A-deficient (n = 6) and WT mice (n = 6) weresubjected to focal cerebral ischemia (45 minutes)/reperfusion (6 hours). Sham surgically operated mice served as sham control(n = 4 per group). Cytoplasmic proteins were isolated from brain tissues for western blot analysis of FasL (A), Fas (B), and FADD (C)expression. *P < 0.05 compared with indicated groups. I/R, ischemia–reperfusion; S, sham; SR-A, scavenger receptor A;WT, wild type.

WT

P-MKK4

MKK4

ShamI/R

ShamI/R**

P-M

KK

4/M

KK

4

1.4

1.6

1.0

1.2

0.4

0.6

0.8

0.2

0.0

WT SR-A-/-

P-JNK

JNK

* *

P-J

NK

/JN

K

2.0

2.5

1.5

1.0

0.0

0.5

SR-A-/-

S I/R S I/R

WT SR-A-/-

S I/R S I/R

Sham I/R I/RSham

WT SR-A-/-

Sham I/R I/RSham

Figure 5 SR-A deficiency prevented I/R-increased MKK4 and JNK phosphorylation in the brain. SR-A-deficient (n = 6) and WT mice(n = 6) were subjected to focal cerebral ischemia (45 minutes)/reperfusion (6 hours). Sham surgically operated mice served as shamcontrol (n = 4 per group). Cytoplasmic proteins were isolated from brain tissues for western blot analysis of MKK4/7 (A) and JNK (B)phosphorylation. *P < 0.05 compared with indicated groups. I/R, ischemia–reperfusion; JNK, Jun N-terminal kinase; S, sham;SR-A, scavenger receptor A; WT, wild type.

ShamI/R

ShamI/R* *

Cas

pase

-3 A

ctiv

ityLu

min

esce

nce

(RU

L)

1000

600

800

400

200

0Sham Sham I/RI/R

WT SR-A-/-

Sham Sham I/RI/R

WT SR-A-/-

**

*

TU

NE

L P

ositi

ve C

ells

(%

) 60

40

50

30

10

20

0

Figure 6 Attenuation of I/R induced brain caspase-3 activity in SR-A-deficient mice. SR-A-deficient (n = 6) and WT mice (n = 6)were subjected to focal cerebral ischemia (45 minutes)/reperfusion (6 hours). Sham surgically operated mice served as sham control(n = 4 per group). (A) Cytoplasmic proteins were isolated from brain tissues for analysis of caspase-3 activity. (B) TUNEL assaywas performed on the brain sections. *P < 0.05 compared with indicated groups. I/R, ischemia–reperfusion; S, sham; SR-A,scavenger receptor A; TUNEL, terminal deoynucleotidyl transferase-mediated 20-deoxyuridine 50-triphosphate-biotin nick endlabeling; WT, wild type.

SR-A mediates brain injuryC Lu et al

1977

Journal of Cerebral Blood Flow & Metabolism (2010) 30, 1972–1981

indicated that I/R significantly increased the numberof apoptotic brain cells, when compared with thesham control (45.6%±4.34% versus 21.4%±2.1%),which is consistent with caspase-3 activity. However,apoptotic cells were significantly reduced in thebrain tissues of SR-A�/� mice after cerebral I/Rcompared with WT mice (Figure 6B).

Discussion

This study provides compelling evidence that themacrophage SR-A contributes to the pathophysio-logy of cerebral ischemic injury. In particular, weobserved that mice deficient in SR-A showeddramatically smaller brain infarcts in response toearly cerebral I/R injury, implying that SR-A med-iates, in part, brain I/R pathology. Our data indicatethat SR-A deficiency attenuated cerebral I/R-acti-vated apoptotic signaling pathways and NF-kBbinding activity, both of which are known tocontribute to cerebral ischemic injury. When con-sidered as a whole, these data suggest that SR-A has acritical role in early cerebral ischemic injury. Thedata also suggest that SR-A may be an importanttarget for preventing and treating cerebral ischemicinjury.

The SR-A is a trimeric, type II membrane glyco-protein that was initially described on the basis ofits ability to bind and internalize modified low-density lipoprotein (Goldstein et al, 1979). The SR-Aexpression is principally on macrophages and den-dritic cells and can be induced in the endotheliumand smooth muscle cells in atheroscleroticplaques (Krieger, 1997; Yi et al, 2009). Interestingly,Matsumoto et al (1990) reported the presence of SR-A in the human brain in 1990. Since then, SR-A hasbeen detected in the microglia (Grewal et al, 1997;El Koury et al, 1998), in senile plaques of braintissue from patients with Alzheimer’s disease(Christie et al, 1996), and on Mato’s fluorescentgranular perithelial perivascular macrophages in thenormal brain tissue (Mato et al, 1996). We observedin this study that SR-A deficiency significantlyreduced infarction after cerebral I/R, suggesting thatSR-A, that is expressed on the microglia and Matocells, could mediate pathophysiological responseto cerebral I/R injury. Cho et al (2005) (Kim et al,2008; Kunz et al, 2008) recently reported that theclass B scavenger receptor (CD36) has a role in brainischemic injury. CD36 is a membrane glycoproteinthat is found on many cell types, including platelets,endothelial cells, macrophages, adipocytes, andmicroglia, and has been implicated in a host ofnormal and disease processes (Silverstein andFebbraio, 2009). Mice deficient in CD36 werepartially protected from experimental brain ischemicinjury (Cho et al, 2005; Kim et al, 2008; Kunz et al,2008). The mechanism of protection was shownto be due to a decreased inflammatory response(Silverstein and Febbraio, 2009).

Scavenger receptor A has recently been shownto act as a PRR and has an important role in theinduction of innate immune and inflammatoryresponses (Zhu et al, 2001; Hollifield et al, 2007;Limmon et al, 2008). For example, SR-A canrecognize several pathogen-associated molecularpatterns, such as lipopolysaccharide, lipoteichoicacid, bacterial CpG DNA, double-stranded RNA,and yeast zymosan/b-glucans (Zhu et al, 2001;Mukhopadhyay and Gordon, 2004; Hollifield et al,2007; Limmon et al, 2008; Areschoug and Gordon,2008). The contribution of the innate immune andinflammatory responses to cerebral I/R has been welldemonstrated (Jordan et al, 2008; Brea et al, 2009).Cerebral ischemia activates innate immunity andtriggers inflammatory responses through the activa-tion of TLR-mediated signaling pathways (Caso et al,2007; Tang et al, 2007; Hua et al, 2007b, 2008, 2009).We, along with other investigators, have shown thatTLR4 contributes to cerebral ischemic injury (Casoet al, 2007; Tang et al, 2007; Hua et al, 2007b, 2008,2009). Toll-like receptor 4 deficiency protects thebrain against cerebral ischemic injury by inhibitionof NF-kB binding activity, resulting in the down-regulation of inflammatory responses (Caso et al,2007; Tang et al, 2007; Hua et al, 2007b, 2008, 2009).Interestingly, recent studies have suggested thatSR-A can act as a coreceptor with TLR4 in modulat-ing the inflammatory response to TLR agonists(Seimon et al, 2006; Amiel et al, 2009). For example,TLR-specific stimuli synergized with SR-A to med-iate bacterial phagocytosis (Amiel et al, 2009).Toll-like receptor ligands dramatically induced SR-A expression and promoted macrophages to bindand internalize TLR4 ligand through SR-A (Xu et al,2007). On the other hand, SR-A ligands triggeredapoptosis in the endoplasmic reticulum-stressedmacrophages by cooperating with TLR4 (Seimonet al, 2006) and may serve as a physiologic negativeregulator of TLR4-mediated immune responses(Yi et al, 2009). Collectively, SR-A may contributeto cerebral I/R injury by deleterious augmentation ofinnate immune and inflammatory challenges.

An increasing body of evidence indicates that focaland global cerebral ischemia induce apoptosis,which is an active process (Gao et al, 2005; Dzietkoet al, 2008). Death receptor signaling moleculessuch as FasL and TRAIL have been implicated inischemia-induced neuronal apoptosis. Fas-mediatedneuronal apoptosis during cerebral I/R is deleterious(Dzietko et al, 2008). Mice that lack functional Fasand FasL are protected from ischemic neuronalinjury (Graham et al, 2004). We observed in thisstudy that cerebral I/R significantly increased thelevels of FasL and Fas in the brain tissues of WTmice. Caspase-3 activity and apoptotic cells in thebrains of WT mice were significantly increased aftercerebral I/R. However, SR-A deficiency significantlyprevented cerebral I/R-increased Fas and FasLlevels in the brain tissues. The caspase-3 activityand apoptotic cells in the brains were significantly

SR-A mediates brain injuryC Lu et al

1978

Journal of Cerebral Blood Flow & Metabolism (2010) 30, 1972–1981

attenuated in SR-A-deficient mice after cerebral I/R.Our observations suggest that SR-A may promotethe Fas/FasL-mediated apoptotic signaling pathwayafter cerebral I/R. However, it is unclear how SR-Apromotes Fas/FasL-mediated apoptotic signalingduring cerebral I/R. Scavenger receptor A has beenconsidered as a coreceptor in the induction of innateimmune and inflammatory responses (Seimon et al,2006); therefore, it could be possible that SR-A maycooperate with Fas in activating the apoptoticsignaling pathway during cerebral I/R.

Activation of the JNK pathway has a role incerebral ischemic injury through the apoptotic path-way (Gao et al, 2005). Specific inhibition of JNKactivation has been shown to attenuate cerebralischemic injury (Gao et al, 2005). We observed thatSR-A deficiency attenuated cerebral I/R-increasedphosphorylation of MKK4/7 and JNK. Attenuation ofJNK activation after cerebral I/R could be an addi-tional mechanism by which SR-A deficiency pre-vents activation of the apoptotic signaling pathwayafter cerebral I/R. It has been shown that stimulationof SR-A by its ligands will activate the JNK signalingpathway (Ricci et al, 2004). Collectively, SR-Acould be a target for preventing apoptosis duringcerebral I/R.

Nuclear factor-kB is an important transcriptionfactor, which regulates the expression of genesassociated with immune and inflammatory re-sponses. Activation of NF-kB has been shown tocontribute to cerebral ischemic injury (Schneideret al, 1999a; Stephenson et al, 2000). For example,inhibition of NF-kB binding activity showed atherapeutic effect on experimental cerebral I/R injury(Schneider et al, 1999a, b). We have reported thatTLR4 deficiency protects against cerebral ischemicinjury (Hua et al, 2007b, 2008, 2009). Nuclear factor-kB binding activity in the particularly ischemia-sensitive hippocampal formation was significantlyattenuated in TLR4-deficient mice after cerebral I/Rcompared with WT mice (Hua et al, 2007b, 2008,2009). In this study, we observed that SR-A defi-ciency significantly attenuated cerebral I/R-in-creased NF-kB activation. Although we do notknow the mechanism by which deficiency of SR-Aattenuated NF-kB activation after cerebral I/R, recentstudies have shown a cooperation between SR-A andTLR4 (Seimon et al, 2006; Amiel et al, 2009; Yi et al,2009). Therefore, activation of SR-A by its ligandsduring I/R may trigger TLR4 to activate the NF-kBpathway. In addition, SR-A is now considered to be aPRR which can recognize several ligands, such asdouble-stranded RNA (Limmon et al, 2008), CpGDNA (Zhu et al, 2001), apoptotic cells, and unknownendogenous ligands, which may be released fromstressed or damaged cells after I/R, resulting in theactivation of NF-kB through ligation of TLRs. Weobserved that NF-kB binding activity was increasedby 46.1% after cerebral I/R in this study, which isconsistent with our previous report (Hua et al, 2009).However, we observed in a global brain ischemia

(12 minutes)/reperfusion (24 hours) that NF-kB bind-ing activity was significantly increased by 91% (Huaet al, 2007b). These data suggest that the time ofischemia and the proportion of dead tissue couldaffect the degree of NF-kB binding activity in acerebral I/R model.

In summary, SR-A has a central role in thepathophysiology of cerebral ischemic injury. Themechanisms involve prevention of I/R-increasedactivation of apoptotic signaling and NF-kB bindingactivity. The data show that SR-A mediates activa-tion of inflammatory signaling and apoptosis inischemic stroke, both of which contribute to cerebralinjury. These data also suggest that modulation ofSR-A activity may be a useful approach for amelio-rating brain injury in stroke patients.

Conflict of interest

The authors declare no conflict of interest.

References

Amiel E, Alonso A, Uematsu S, Akira S, Poynter ME,Berwin B (2009) Toll-like receptor regulation ofscavenger receptor-A-mediated phagocytosis. J Leuko-cyte Biol 85:595–605

Areschoug T, Gordon S (2008) Pattern recognitionreceptors and their role in innate immunity: focuson microbial protein ligands. Contrib Microbiol 15:45–60

Bowdish DME, Gordon S (2009) Conserved domains of theclass A scavenger receptors: evolution and function.Immuno Rev 227:19–31

Brea D, Sobrino T, Ramos-Cabrer P, Castillo J (2009)Inflammatory and neuroimmunomodulatory changesin acute cerebral ischemia. Cerebrovasc Dis 27:48–64

Caso J, Pradillo J, Hurtado O, Lorenzo P, Moro M, LizasoainI (2007) Toll-like receptor 4 is involved in brain damageand inflammation after experimental stroke. Circulation115:1599–608

Cho S, Park E-M, Febbraio M, Anrather J, Park L, RacchumiG, Silverstein RL, Iadecola C (2005) The class Bscavenger receptor CD36 mediates free radical produc-tion and tissue injury in cerebral ischemia. J Neurosci25:2504–12

Christie RH, Freeman M, Hyman BT (1996) Expression ofthe macrophage scavenger receptor, a multifunctionallipoprotein receptor, in microglia associated withsenile plaques in Alzheimer’s disease. Am J Pathol148:399–403

Cotena A, Gordon S, Platt N (2004) The class A macro-phage scavenger receptor attenuates CXC chemokinproduction and the early infiltration of neutrophils insterile peritonitis. J Immunol 173:6427–32

Dzietko M, Boos V, Sifringer M, Polley O, Gerstner B,Genz K, Endesfelder S, Borner C, Jacotot E, Clauvier D,Obladen M, Buhrer C, Felderhoff-Mueser U (2008)A critical role for Fas/CD-95 dependent signalingpathways in the pathogenesis of hyperoxia-inducedbrain injury. Ann Neurol 64:664–73

SR-A mediates brain injuryC Lu et al

1979

Journal of Cerebral Blood Flow & Metabolism (2010) 30, 1972–1981

El Koury J, Hickman SE, Thomas CA, Loike JD, SilversteinSC (1998) Microglia, scavenger receptors, and thepathogenesis of Alzheimer’s disease. Neurobiol Aging19:S81–4

Fong LG, Li D (1999) The processing of Ligands by the classA scavenger receptor is dependent on signal informa-tion located in the cytoplasmic domain. J Biol Chem274:36808–16

Gao Y, Signore AP, Yin W, Cao G, Yin X-M, Sun F, Luo Y,Graham SH, Chen J (2005) Neuroprotection against focalischemic brain injury by inhibition of c-Jun N-terminalkinase and attenuation of the mitochondrial apoptosis-signaling pathway. J Cereb Blood Flow Metab 25:694–712

Goldstein JL, Ho YK, Basu SK, Brown MS (1979) Bindingsite on macrophages that mediates uptake and degrada-tion of acetylated low density lipoprotein, producingmassive cholesterol deposition. Proc Natl Acad Sci76:333–7

Graham EM, Sheldon RA, Flock DL, Ferriero DM, MartinLJ, O’Riordan DP, Northington FJ (2004) Neonatal micelacking functional Fas death receptors are resistant tohypoxic-ischemic brain injury. Neurobiol Dis 17:89–98

Greaves DR, Gordon S (2009) The macrophage scavengerreceptor at 30 years of age: current knowledge andfuture challenges. J Lipid Res 50(Suppl):S282–6

Grewal RP, Yoshida T, Finch CE, Morgan TE (1997)Scavenger receptor mRNAs in rat brain microglia areinduced by kainic acid lesioning and by cytokines.Neuroreport 8:1077–81

Hollifield M, Bou Ghanem E, de Villiers WJ, Garvy BA(2007) Scavenger receptor A dampens induction ofinflammation in response to the fungal pathogenPneumocystis carinii. Infect Immun 75:3999–4005

Hsu H-Y, Chiu S-L, Wen M-H, Chen K-Y, Hua K-F (2001)Ligands of macrophage scavenger receptor inducecytokine expression via differential modulation ofprotein kinase signaling pathways. J Biol Chem276:28719–30

Hua F, Ha T, Ma J, Li Y, Kelley J, Gao X, Browder IW,Kao RL, Williams DL, Li C (2007a) Protection againstmyocardial ischemia/reperfusion injury in TLR4deficient mice is mediated through a phosphoinositide3-kinase dependent mechanism. J Immunol 178:7317–24

Hua F, Ma J, Ha T, Kelley J, Williams DL, Kao RL,Kalbfleisch JH, Browder IW, Li C (2008) Preconditioningwith a TLR2 specific ligand increases resistance tocerebral ischemia/reperfusion injury. J Neuroimmunol199:75–82

Hua F, Ma J, Ha T, Kelley JL, Kao RL, Schweitzer JB,Kalbfleisch JH, Williams DL, Li C (2009) Differential rolesof TLR2 and TLR4 in acute focal cerebral ischemia/reperfusion injury in mice. Brain Res 1262:100–8

Hua F, Ma J, Ha T, Xia Y, Kelley J, Williams DL, Kao RL,Browder IW, Schweitzer JB, Kalbfleisch JH, Li C (2007b)Activation of Toll-like receptor 4 signaling contributesto hippocampal neuronal death following global cere-bral ischemia/reperfusion. J Neuroimmunol 190:101–11

Jordan J, Segura T, Brea D, Galindo MF, Casvillo J (2008)Inflammation as therapeutic objective in stroke. CurrPharm Des 14:3549–64

Kim E, Tolhurst AT, Qin LY, Chen X-Y, Febbraio M, Cho S(2008) CD36/fatty acid translocase, an inflammatorymediator, is involved in hyperlipidemia-inducedexacerbation in ischemic brain injury. J Neurosci28:4661–70

Krieger M (1997) The other side of scavenger receptors:pattern recognition for host defense. Curr Opin Lipidol8:275–80

Kunz A, Abe T, Hochrainer K, Shimamura M, Anrather J,Racchumi G, Zhou P, Iadecola C (2008) Nuclear factor-kB activation and postischemic inflammation are sup-pressed in CD36-null mice after middle cerebral arteryocclusion. J Neurosci 28:1649–58

Li C, Ha T, Kelley J, Gao X, Qiu Y, Kao RL, Browder W,Williams DL (2003) Modulating Toll-like receptormediated signaling by (1-3)-b-D-glucan rapidly in-duces cardioprotection. Cardiovasc Res 61:538–47

Limmon GV, Arredouani M, McCann KL, Minor RAC,Kobzik L, Imani F (2008) Scavenger receptor class-A is anovel cell surface receptor for double-stranded RNA.FASEB J 22:159–67

Mato M, Ookawara S, Sakamoto A, Aikawa E, Ogawa T,Mitsuhashi U, Masuzawa T, Suzuki H, Honda M,Yazaki Y, Watanabe E, Luoma J, Yla-Herttuala S, Fraser I,Gordon S, Kodama T (1996) Involvement of specificmacrophage-lineage cells surrounding arterioles inbarrier and scavenger function in brain cortex. ProcNatl Acad Sci USA 93:3269–74

Matsumoto A, Naito M, Itakura H, Ikemoto S, Asaoka H,Hayakawa I, Kanamori H, Aburatani H, Tanaku F,Suzuki H, Kobari Y, Miyai T, Takahashi K, Cohen EH,Wydro R, Housman DE, Kodama T (1990) Humanmacrophage scavenger receptors: primary structure,expression, and localization in atherosclerotic lesions.Proc Natl Acad Sci USA 87:9133–7

Medzhitov R, Preston-Hurlburt P, Janeway CA, Jr (1997)A human homologue of the Drosophila Toll proteinsignals activation of adaptive immunity. Nature388:394–7

Mukhopadhyay S, Gordon S (2004) The role of scavengerreceptors in pathogen recognition and innate immunity.Immunobiol 209:39–49

Nikolic DM, Cholewa J, Gass C, Gong MC, Post SR (2007)Class A scavenger receptor-mediated cell adhesionrequires the sequential activation of Lyn and PI3-kinase.Am J Physiol Cell Physiol 292:1450–8

Platt N, Gordon S (2001) Is the class A macrophagescavenger receptor (SR-A) multifunctional? The mouse’stale. J Clin Invest 108:649–54

Ricci R, Sumara G, Sumara I, Rozenberg I, Kurrer M,Akhmedov A, Hersberger M, Eriksson U, Ererli FR, BecherB, Boren J, Chen M, Cybulsky MI, Moore KJ, FreemanMW, Wagner EF, Matter CM, Luscher TF (2004) Require-ment of JNK2 for scavenger receptor A-mediated foam cellformation in atherogenesis. Science 306:1558–61

Schneider A, Martin-Villalba A, Weih F, Vogel J, Wirth T,Schwaninger M (1999a) NF-kB is activated and pro-motes cell death in focal cerebral ischemia. Nat Med5:554–9

Schneider A, Martin-Villalba A, Weih F, Vogel J, Wirth T,Schwaninger M (1999b) NF-kB is activated and pro-motes cell death in focal cerebral ischemia. Nat Med5:554–9

Seimon TA, Obstfeld A, Moore KJ, Golenbock DT, Tabas I(2006) Combinatorial pattern recognition receptor sig-naling alters the balance of life and death in macro-phages. Proc Natl Acad Sci 103:19794–9

Silverstein R, Febbraio M (2009) CD36: a scavengerreceptor involved in immunity, metabaolism, angiogen-esis and behavior. Sci Signal 2:1–8

Stephenson D, Yin T, Smalstig EB, Hsu MA, Panetta J,Little S, Clemens J (2000) Transcription factor nuclear

SR-A mediates brain injuryC Lu et al

1980

Journal of Cerebral Blood Flow & Metabolism (2010) 30, 1972–1981

factor-kappa B is activated in neurons after focalcerebral ischemia. J Cereb Blood Flow Metab 20:592–603

Tang SC, Arumugam TV, Xu X, Cheng A, Hughal MR,Jo DG, Lathia JD, Siler DA, Chigurapati S, Ouyang X,Magnus T, Camadola S, Mattson MP (2007) Pivotal rolefor neuronal Toll-like receptors in ischemic brain injuryand functional deficits. Proc Natl Aad Sci USA104:13798–803

Todt JC, Hu B, Curtis JL (2008) The scavenger receptor SR-A I/II (CD204) signals via the receptor tyrosine kinaseMertk during apoptotic cell uptake by murine macro-phages. J Leukocyte Biol 84:510–8

Xu WY, Wang L, Wang HM, Wang YQ, Liang YF, Zhao TT,Wu YZ (2007) TLR2 and TLR4 agonists synergisticallyup-regulate SR-A in RAW264.7 through p38. MolImmunol 44:2315–23

Yi H, Yu X, Gao P, Wang Y, Baek S-H, Chen X, Kim HL,Subjeck JR, Wang X-Y (2009) Pattern recognitionscavenger receptor SRA/CD204 down-regulates Toll-likereceptor 4 signaling-dependent CD8 T-cell activation.Blood 113:5819–28

Zhu FG, Reich CF, Pisetsky DS (2001) The role of themacrophage scavenger receptor in immune stimulationby bacterial DNA and synthetic oligonucleotides.Immunobiol 103:226–34

SR-A mediates brain injuryC Lu et al

1981

Journal of Cerebral Blood Flow & Metabolism (2010) 30, 1972–1981