SHPS-1 deficiency induces robust neuroprotection against experimental stroke by attenuating oxidative stress

Stroke is the second leading cause of mortality and the mostcommon cause of long-term disability worldwide (Matherset al. 2009). Over 15 million people suffer from stroke eachyear, and approximately 80–85% of these cases are ischemicstroke. Accumulating evidence from the past two decadeshas suggested that the oxidative stress associated with theexcessive production of reactive oxygen species (ROS) has aprofound effect on ischemic stroke pathogenesis (El Kossi

Received December 5, 2011; revised manuscript received May 31, 2012;accepted June 2, 2012.Address correspondence and reprint requests to Dr. Hong Jiang or Dr.

Hongliang Li, Department of Cardiology, Renmin Hospital of WuhanUniversity, Jiefang Road 238, Wuhan 430060, PR China.E-mails: [email protected] or [email protected] used: HO-1, heme oxygenase 1; Nrf2, nuclear factor-

E2-related factor 2; ROS, reactive oxygen speciesSHPS-1 MT, SHPS-1mutant; tMCAO, transient middle cerebral artery occlusion.

, , , ,

, ,

,

*Department of Cardiology, Renmin Hospital of Wuhan University, Wuhan, PR China

�Cardiovascular Research Institute of Wuhan University, Wuhan 430060, PR China

�Department of Cardiology, Union Hospital, Tongji Medical College, Huazhong University of Science

and Technology, Wuhan, PR China

§Department of Thoracic and Cardiovascular Surgery, Tongji Hospital, Tongji Medical College,

Huazhong University of Science and Technology, WuHan, PR China

¶Laboratory of Biosignal Sciences, Institute for Molecular and Cellular Regulation, Gunma University,

Gunma, Japan

**Division of Molecular and Cellular Signaling, Department of Biochemistry and Molecular Biology,

Kobe University Graduate School of Medicine, Kobe, Japan

Abstract

Src homology 2 domain–containing protein tyrosine phos-

phatase substrate–1 (SHPS-1), also known as Signal-regu-

latory protein alpha (SIRPa) or SIRPA is a transmembrane

protein that is predominantly expressed in neurons, dendritic

cells, and macrophages. This study was conducted to

investigate the role of SHPS-1 in the oxidative stress and

brain damage induced by acute focal cerebral ischemia. Wild-

type (WT) and SHPS-1 mutant (MT) mice were subjected to

middle cerebral artery occlusion (60 min) followed by reper-

fusion. SHPS-1 MT mice had significantly reduced infarct

volumes and improved neurological function after brain

ischemia. In addition, neural injury and oxidative stress were

inhibited in SHPS-1 MT mice. The mRNA and protein levels

of the antioxidant genes nuclear factor-E2-related factor 2

(Nrf2) and heme oxygenase 1 were up-regulated in SHPS-1

MT mice. The SHPS-1 mutation suppressed the phosphory-

lation of SHP-1 and SHP-2 and increased the phosphoryla-

tion of Akt and GSK3b. These results provide the first

demonstration that SHPS-1 plays an important role in the

oxidative stress and brain injury induced by acute cerebral

ischemia. The activation of Akt signaling and the up-regula-

tion of Nrf2 and heme oxygenase 1 likely account for the

protective effects that were observed in the SHPS-1 MT

mice.

Keywords: Akt, cerebral ischemia, heme oxygenase 1,

nuclear factor-E2-related factor 2, oxidative stress, SHPS-1.

J. Neurochem. (2012) 122, 834–843.

JOURNAL OF NEUROCHEMISTRY | 2012 | 122 | 834–843 doi: 10.1111/j.1471-4159.2012.07818.x

834 Journal of Neurochemistry � 2012 International Society for Neurochemistry, J. Neurochem. (2012) 122, 834–843� 2012 The Authors

and Zakhary 2000; Kelly et al. 2008; Lei et al. 2011). ROShave direct cellular effects, such as lipid peroxidation, proteindenaturation, and DNA and RNA damage, which result intissue destruction and cell death. ROS also act in severalsignal transduction pathways, including intrinsic and extrin-sic caspase activation and nuclear factor kappa B (NF-jB)activation, which may lead to excessive cell apoptosis andinflammatory gene expression (Chan 2001; Allen andBayraktutan 2009). In recent years, continuous efforts havebeen made to modulate oxidative stress after ischemic stroke.Although some free radical scavenging agents and radicaltrapping agents have shown therapeutic potential in animalmodels, they have failed in clinical trials. In the presentstudy, we found that the deletion of the signal regulatoryprotein (SIRP) family member SHPS-1 inhibits oxidativedamage and mitigates brain injury after ischemic stroke.

SHPS-1, which is also known as SIRP a or SIRPA, is atransmembrane protein that consists of three domains: animmunoglobulin (Ig)-like extracellular domain, a transmem-brane domain, and an intracellular domain (Yamao et al.1997; Oshima et al. 2002). The intracellular domain ofSHPS-1 contains two immunoreceptor tyrosine-based inhi-bition motifs (ITIMs) with four tyrosine residues, which canbe phosphorylated by growth factors and integrins thatmediate cell adhesion to extracellular matrix proteins(Galbaugh et al. 2010; Kapoor and O’Rourke 2010; Shenet al. 2010). The intracellular tyrosine-phosphorylated sitesof SHPS-1 then bind to and activate two Src homology-2(SH2) domain-containing protein tyrosine phosphatases,SHP-1 and SHP-2. The activated SHP-1 and SHP-2 act onmultiple signaling molecules and modulate downstreamsignal transduction via dephosphorylation.

The MAPK, JAK/STAT, and PI3K-Akt pathways havebeen reported to be modulated by SHP-1 or SHP-2 underdiverse biological states. Akt, an important member of Arg-directed kinases, plays a central role in mediating criticalcellular responses including cell survival, metabolism,angiogenesis, and transcriptional regulation. GSK3b is adirect substrate of Akt, which can be directly phosphorylatedby activated Akt at Ser 9. Previous studies have demon-strated that Akt activation plays an important role in theexpression and activation of nuclear factor-E2-related factor2 (Nrf2) (Martin et al. 2004; Bak et al. 2012). Nrf2 is animportant transcription factor that induces the expression of anumber of genes including those that encode for severalantioxidant enzymes, and it plays a physiological role in theregulation of oxidative stress.

SHPS-1 is predominantly expressed in neurons, dendriticcells, and in macrophages, and previous studies have shownthat SHPS-1 is involved in a variety of physiologicalprocesses, including the regulation of immune cells, theself-recognition of red blood cells, macrophage multinucle-ation, vascular smooth muscle proliferation, neuronal devel-opment, and survival (Ikeda et al. 2006; Mitsuhashi et al.

2008; Sobota et al. 2008). To date, studies have notexamined the role of SHPS-1 in ischemic stroke or theimpact of SHPS-1 on oxidative stress after cerebral ischemia.Identifying the functional role and the mechanisms of SHPS-1 in the stroke pathological process may provide potentialtreatment targets for ischemic stroke.

Materials and methods

AnimalsAll the animal procedures were approved by the Wuhan UniversityAnimal Ethics Committee. The generation of SHPS-1 mutant (MT)mice was described previously (Inagaki et al. 2000; Ohnishi et al.2010). The SHPS-1 MT mice lack most of the cytoplasmic regioninstead of the wild-type protein. The SHPS-1 MT mice werebackcrossed to the C57BL/6J background for > 10 generations. Thepresent study used male WT and SHPS-1 MT mice on a C57BL/6Jbackground that were between 10 and 12 weeks of age.

Mouse transient focal cerebral ischemia modelThe procedure for transient middle cerebral artery occlusion(tMCAO) has been previously described (Connolly et al. 1997).Briefly, the mice were anesthetized with 2.5–3% isoflurane in O2.The rectal temperature was maintained at 37 ± 0.5�C with a heatingpad. A probe was fixed to the skull (2 mm posterior and 5 mmlateral to the bregma) and connected to a laser Doppler flow meter(Periflux System 5010; Perimed, Sweden) to continuously monitorcerebral blood flow (CBF). For tMCAO, a 6-0 silicon-coatedmonofilament surgical suture (Doccol, Redland, CA, USA) wasinserted into the left external carotid artery, advanced into theinternal carotid artery, and wedged into the cerebral arterial circle toobstruct the origin of the MCA (middle cerebral artery). Aninterruption of the cerebral blood flow in the MCA territory wasconfirmed by documenting a > 80% decline in the relative cerebralblood flow. The filament was left in place for 60 min and thenwithdrawn. A return to > 70% of basal cerebral blood flow within10 min of suture withdrawal confirmed the reperfusion of the MCAterritory.

Indian ink stainingThe Indian ink staining was utilized to show gross cerebralvasculatures of WT mice and SHPS-1 MT mice (Fujii et al.1997). Animals were perfused under sodium pentobarbital (Sigma,St. Louis, MO, USA) anesthesia (50 mg/kg IP) with 10 mL ofphysiological saline followed by 2 mL of preheated Indian inkstaining solution via the left cardiac ventricle until the tissues (e.g.,tongue, lips, and gums) turned black. The Indian ink stainingsolution contained 10% (W/V) gelatin (Amresco, Solon, OH,USAyy), 50% (V/V) Indian ink (Solarbio, Beijing, China). Afterdecapitation, the brains were carefully removed into 10% bufferedformalin for 24 h before examination. The vessels of the circle ofWillis and their branches were photographed using Nikon D700digital camera.

Neurological deficit scoresThree days after tMCAO, neurological deficits were assessed usinga 9-point scale (Xia et al. 2006). A lack of neurological deficit was

� 2012 The AuthorsJournal of Neurochemistry � 2012 International Society for Neurochemistry, J. Neurochem. (2012) 122, 834–843

SHPS-1 deficiency induces robust neuroprotection | 835

scored as 0, and left forelimb flexion when suspended by the tail orfailure to fully extend the right forepaw was scored as 1. Leftshoulder adduction when suspended by the tail was scored as 2, andreduced resistance to a lateral push toward the left side was scored as3. Spontaneous movement in all directions with circling to the leftthat was only exhibited when the animal was pulled by the tail wasscored as 4, spontaneously circling or walking solely to the left wasscored as 5 and only walking when stimulated was scored as 6. Alack of response to stimulation was scored as 7, and stroke-relateddeath was scored as 8.

Measurement of infarct volumeInfarct volume and swelling were measured at 24 h after tMCAO by2,3,5-triphenyl-2H-tetrazolium chloride (TTC) staining. After beinganesthetized by sodium pentobarbital (50 mg/kg IP), the mice werekilled by cervical dislocation. Brains were cut into 1-mm-thickcoronal sections using a mouse brain matrix and stained with 2%TTC (Sigma, St. Louis, MO) in phosphate buffer (pH 7.4) for15 min at 37�C. After staining, the sections were transferred to a10% formalin solution and fixed overnight. Fixed sections werephotographed, and the volume of the infarct area was quantifiedusing Image-Pro Plus 6.0 software (Media Cybernetics, SilverSpring, MD, USA). To correct for the effect of edema, the area ofinfarction was measured by subtracting the area of the non-lesionedhemisphere from the area of the lesioned hemisphere. The volume ofinfarction was calculated by integrating the lesioned areas from theseven measured levels of the brain.

Immunofluorescence stainingMice were anesthetized with sodium pentobarbital and perfusedthrough the left ventricle with 0.1 mol/L sodium phosphate bufferunder 100 mmHg of pressure for 5 min. This was followed byperfusion with a fixative solution that contained 4% paraformalde-hyde in 0.1 mol/L phosphate buffer (pH 7.4) for 15 min. The brainswere carefully removed, post-fixed for 6–8 h in the same fixativesolution at 25�C, and immersed overnight in a 0.1 mol/L phosphatebuffer that contained 30% sucrose at 4�C. The brains were embeddedinOCT, and serial frontal sectionswere cut with a cryostat microtome.For immunofluorescence staining, the sections were washed in PBScontaining 10% goat serum. The sections were incubated with anti-4-Hydroxynonenal (4HNE, ab48506, Abcam, Cambridge, MA) or anti-8-Hydroxyguanosine (8OHdG, sc-66036, Santa cruz, CA), anti-Nrf2(ab31163, Abcam, Cambridge, MA) primary antibodies overnight at4�C. For NeuN (MAB377, Millipore, Billerica, MA) immunofluo-rescence staining, the sections were washed in PBS containing 10%goat serum and 0.1% Triton X-100. After washing, the sections wereincubated in the anti-NeuN antibody for 2 h prior to incubation insecondary antibody for 1 h at 37�C. After the sections were washed inPBS, they were incubated with secondary antibody for 1 h. Finally,the nuclei were labeled with DAPI. Visualization was performedunder a fluorescence microscope OLYMPUS DX51 (Olympus,Japan) with DP2-BSW Ver. 2.2 software, and the image analysis wasperformed with Image-Pro Plus 6.0 software.

TUNEL stainingSix hours after the onset of ischemia, the brains were collected andsliced as described above. After the NeuN immunofluorescencestaining was completed, TUNEL (terminal deoxynucleotidyl trans-ferase dUTP nick-end labeling) staining was performed using the

In Situ Cell Death Detection Kit (Roche Diagnostics, Indianapolis,IN, USA) according to the manufacturer’s protocol. We also labeledthe nuclei with DAPI, and DNA fragmentation was quantified underhigh-power magnification (200X). An investigator who was blindedto the study condition calculated the percentage of DAPI-positivecells that were also TUNEL-positive.

Tissue preparationFor quantitative real-time PCR (qRT-PCR) and western blottinganalysis, mice were anesthetized with sodium pentobarbital(50 mg/kg IP) and perfused through the left ventricle with coldsodium phosphate. After perfusion, the brains were quicklyremoved. To collect tissue in an unbiased manner that globallyreflected the infarct, the olfactory bulbs and the front and back 1 mmof brain tissue were excised from each animal. The remaining lefthemisphere was collected (including the infarct area and the peri-infarct area). The brain tissues were immediately frozen in liquidnitrogen and transferred to a )80�C freezer for storage.

Quantitative real-time PCRTotal RNA was prepared from snap-frozen tissue specimens usingTRIzol reagent (Invitrogen, Carlsbad, CA) and was reversetranscribed into cDNA using 2 lg of RNA from each sample andthe Transcriptor First Strand cDNA Synthesis Kit (Roche, India-napolis, IN). To examine the relative mRNA expression of Nrf2 andheme oxygenase 1 (HO-1), the specific mRNA expression levelswere normalized to GAPDH. Quantitative RT-PCR analysis wasperformed using the LightCycler 480 SYBR Green 1 Master Mix(Roche, Indianapolis, IN) and the LightCycler 480 QPCR System(Roche, Indianapolis, IN). The following sequence-specific primerswere used:

Nrf2 forward: 5¢-ATGATGGACTTGGAGTTGCC-3¢;Nrf2 reverse: 5¢-TCCTGTTCCTTCTGGAGTTG-3¢;HO-1 forward: 5¢-AGGAGATAGAGCGCAACAAGCAGA-3¢;HO-1 reverse: 5¢-CCAGTGAGGCCCATACCAGAAG-3¢.

Western blottingWestern blotting was conducted to determine the activation state ofprogrammed cell death using cleaved caspase 8 (Asp387), caspase8, cleaved caspase 3 (Asp175), and caspase 3 (all these antibodieswere from Cell Signaling Technology; Beverly, MA, USA). Wealso examined the expression of Nrf2 (Bioworld Technology,Minneapolis, MN, USA) and HO-1 (Cell Signaling Technology,Beverly, MA, USA). In addition, we examined the phosphorylationof SHP-1 (Bioworld Technology, Minneapolis, MN), SHP-2(Bioworld Technology, Minneapolis, MN), Akt (Cell SignalingTechnology, Beverly, MA) and GSK3b (Cell Signaling Technol-ogy, Beverly, MA). For the western blot analysis, 50 lg of proteinextract was separated on an 8–12% SDS-PAGE gel and subse-quently transferred to a PVDF membrane (Millipore, Bedford, MA).Membrane blocking (5% skimmed milk powder), washes (PBS) andsecondary antibody (goat anti-rabbit IRDye 800CW or goat anti-mouse IRDye 800CW, LI-COR Biosciences, Lincoln, NE) incuba-tions were all performed at 25�C for 1 h, whereas the primaryantibodies were allowed to incubate overnight at 4�C. The proteinsignals were detected using the Odyssey Infrared Imaging System(LI-COR Biosciences, Lincoln, NE). All the specific proteinexpression levels were normalized to GAPDH.

Journal of Neurochemistry � 2012 International Society for Neurochemistry, J. Neurochem. (2012) 122, 834–843� 2012 The Authors

836 | L. Wang et al.

Statistical analysisData are expressed as the means ± SEM. Differences among thegroups were determined by two-way ANOVA followed by a post hocTukey test. Comparisons between the two groups were performedby the unpaired Student’s t-test. A p-value of < 0.05 was accepted asthe level of statistical significance.

Results

SHPS-1 deficiency decreased the infarct volume at 24 hafter tMCAOGross cerebral vasculatures showed by Indian ink staining.There was no macroscopic difference in the vessels of thecircle of Willis and their branches between WT mice andSHPS-1 MT mice (Fig. 1a). According to the laser Dopplerflowmetry monitoring, the CBF was similar between the WTand the SHPS-1 MT mice during both ischemia andreperfusion (Fig. 1b). The infarct volume was determinedusing TTC staining. The infarct volume was 30.77± 4.12 mm3 in the SHPS-1 MT mice, which was 30%smaller than the volume in the WT mice (47.18 ± 3.10 mm3)(Fig. 1c). The edema volume percentage was 7.03 ± 1.43%in the SHPS-1 MT mice, and was 7.93 ± 2.09% in the WTmice, there was no statistical significance between the twogroups.

Neurological function was improved in SHPS-1 mutantmiceOn the basis of previous study (Xia et al. 2006), theneurological deficits were assessed 24 h after tMCAO using

a 9-point scale. The neurological deficit score was4.00 ± 0.31 in the SHPS-1 MT mice compared with5.29 ± 0.18 in the WT mice (Fig. 2), which suggested thatneurological function improved in the SHPS-1 MT micefollowing transient cerebral ischemia.

Neuronal injury was inhibited in the SHPS-1 mutant miceNeuronal injury was detected with TUNEL and NeuNstaining 6 h after tMCAO. There was an abundance ofTUNEL-positive cells in the infarct and peri-infarct area. Inaddition, NeuN immunofluorescence staining revealed thatmost of the TUNEL-positive cells also expressed NeuN,which indicated that the injured cells were mainly neurons.We calculated the number of TUNEL-positive nuclei andtotal nuclei (DAPI labeled), and the TUNEL-positive ratewas calculated with the following formula: TUNEL-positiverate (%)=TUNEL-positive nuclei/total nuclei. The TUNEL-positive rate in the peri-infarct area in the SHPS-1 MT micewas 36.73 ± 0.04%, which was much lower than that of theWT mice (63.58 ± 0.41%) (Fig. 3a). Western blot analysisrevealed that the protein levels of cleaved caspase 3 andcleaved caspase 8 were much lower in the SHPS-1 MT micecompared with the WT mice (Fig. 3b).

Oxidative injury was attenuated in the SHPS-1 mutantmice8OHdG is a biomarker of oxidative DNA damage, and 4HNE is a stable product of lipid peroxidation and a keymediator of oxidative stress-induced cell death (Imai et al.2001; Kawai et al. 2011). Immunofluorescence staining for

(a) (b)

(c)Fig. 1 The SHPS-1 deficiency reduced

the infarct volume in a mouse transient

middle cerebral artery occlusion model.

(a) Cerebral vascular anatomy. Repre-

sentative images showing gross cerebral

vasculatures stained by India ink (n = 3

for WT and MT mice). (b) There was no

significant difference in the CBF of the

SHPS-1 mutant mice and the WT mice

during the ischemia and reperfusion pro-

cess. (c) The SHPS-1 deficiency signifi-

cantly reduced the infarct volume

(*p < 0.05 vs. WT, n = 7 for each group)

and slightly reduced edema formation.



8OHdG and 4HNE was used to estimate the oxidative injury.The 8OHdG- or 4HNE-positive cells were counted under a200X field. The numbers of both the 8OHdG-positive cellsand the 4HNE-positive cells were reduced by approximately50% in the SHPS-1 MT mice compared with the WT mice(Fig. 4a).

The expression levels of the antioxidant genes Nrf2 andHO-1 were elevated in SHPS-1 mutant miceNrf2 is an important transcription factor that regulates theexpression of a large number of antioxidant genes (Wanget al. 2011). HO-1 is a target gene of Nrf2 and plays aprotective role in cerebral ischemic injury via an anti-oxidative mechanism (Kweon et al. 2006). Quantitative real-time PCR and western blot analysis revealed that the mRNAand protein levels of Nrf2 and HO-1 were elevated in thebrains of the ischemic SHPS-1 MT mice compared with theWT mice (Fig. 4b). Immunofluorescence staining showedthat the Nrf2 positive cell numbers and the fluorescenceintensity of Nrf2 were increased in SHPS-1 MT mice,indicating that the expression of Nrf2 was up-regulated in theischemic brain of SHPS-1 MT mice. Immunofluorescencestaining also showed that, the expression of Nrf2 was moreconcentrated in the nucleus in SHPS-1 MT mice, implyingthat there was more Nrf2 transferred into the nucleus inSHPS-1 MT mice (Fig. 4c).

The deficiency in SHPS-1 led to increased Akt activationPhosphorylated SHP-1, SHP-2, Akt, and GSK3b weredetected using western blot analysis. The phosphorylationof SHP-1 and SHP-2 was inhibited in the SHPS-1 MT miceafter cerebral ischemia, whereas the phosphorylation of Aktand GSK3b was markedly elevated. The total protein levelsof SHP-1, SHP-2, Akt, and GSK3b were not changed in theSHPS-1 MT mice compared with the WT mice (Fig. 5).

Discussion

This study suggests that SHPS-1 plays an important role inthe pathological process of focal cerebral ischemia. In theSHPS-1 MT mice, the infarct volume was decreased,neurological function was improved, neuronal apoptosiswas reduced, the activities of caspase 3 and caspase 8 wereinhibited, oxidative injury was attenuated and the expressionlevels of the antioxidant genes Nrf2 and HO-1 wereup-regulated. The phosphorylation of SHP-1 and SHP-2was also inhibited, and the activity of Akt was enhanced inthe SHPS-1 MT mice.

SHPS-1 is a ubiquitously expressed receptor-type trans-membrane glycoprotein that is abundantly expressed in thebrain (Yamao et al. 1997; Oshima et al. 2002). Previousstudies (Stofega et al. 2000; Ikeda et al. 2006) havedemonstrated that SHPS-1 plays a key role in the negativeregulation of tyrosine kinase-coupled cellular responses thatare induced by cell adhesion, growth factors and/or insulin.SHPS-1 has been shown to be involved in various biologicalfunctions, such as cell migration, phagocytosis, and mast anddendritic cell activation. In the brain, SHPS-1 has beenidentified as a neural adhesion molecule that participates inbrain-derived neurotrophic factor (BDNF)-mediated neuro-nal survival via Akt activation (Araki et al. 2000). Inaddition, transfection with wild-type and mutant SHPS-1both have been shown to enhance Akt activation in neurons(Araki et al. 2000). CD47 is a ligand of SHPS-1, Koshimizuet al. and Xing et al. suggested that the activation of CD47by its activating peptide induces oxidative injury andcytotoxicity in cultured neurons (Koshimizu et al. 2002;Xing et al. 2009); however, previous studies have not shownthat SHPS-1 is involved in oxidative stress.

Oxidative stress has been identified as an important factorin the pathological process of stroke because it has directeffects on cellular injury and activates downstream signalingpathways, which may aggravate the post-ischemic injury (ElKossi and Zakhary 2000; Kelly et al. 2008; Lei et al. 2011).In the present study, the biomarkers of oxidative damage(i.e., 8OHdG and 4HNE) were markedly decreased in thebrains of the ischemic SHPS-1 MT mice, which suggestedthat oxidative damage is mitigated by the SHPS-1 mutation.The severity of oxidative stress depends on the balancebetween antioxidants and pro-oxidants. We estimated theexpression of the antioxidant genes Nrf2 and HO-1. Nrf2 isan important transcription factor that regulates the expressionof a large number of antioxidant genes (Venugopal andJaiswal 1996; Solis et al. 2002; Kweon et al. 2006), and HO-1 is a target gene of Nrf2 that plays a protective role incerebral ischemic injury via an anti-oxidative mechanism(Kweon et al. 2006; Aztatzi-Santillan et al. 2010; Kim et al.2010). The mRNA and protein levels of Nrf2 and HO-1 weremarkedly up-regulated in the brains of the ischemic SHPS-1MT mice, which indicated that the SHPS-1 mutation

Fig. 2 Neurological function was improved in the SHPS-1 mutant

mice. Neurological deficits were assessed 24 hours after transient

middle cerebral artery occlusion using a 9-point scale. The SHPS-1

deficiency significantly improved neurological function following tran-

sient middle cerebral artery occlusion (*p < 0.05 vs. WT, n = 7 for

each group).



mitigated oxidative stress, likely through the up-regulation ofan anti-oxidative mechanism.

The number of TUNEL-positive neurons decreased, andthe protein levels of cleaved caspase 3 and caspase 8 were

down-regulated in the SHPS-1 knockout mice in this study.TUNEL detects DNA fragmentation which appeared inprogrammed cell death. Caspases, or cysteine-aspartic pro-teases or cysteine-dependent aspartate-directed proteases are

(a)

(b)

Fig. 3 Neuronal injury was inhibited in the

SHPS-1 mutant mice. (a) TUNEL and NeuN

double staining shows that, most TUNEL-

positive nuclei can overlap with NeuN

positive nuclei, indicates that the majority of

TUNEL-positive cells were neurons. The

TUNEL-positive nuclei numbers and total

nuclei numbers were counted, TUNEL-

positive rate (%)=TUNEL-positive nuclei

number/total nuclei number. SHPS-1

mutation significantly reduced the TUNEL-

positive rate in the peri-infarct area

(*p < 0.05 vs. WT, n = 4 for each group).

(b) western blot analysis shows the protein

levels of cleaved caspase 3 (c-caspase 3),

caspase 3, cleaved caspase 8 (c-caspase

8) and caspase 8. All the caspase proteins

were normalized with GAPDH. The protein

levels of c-caspase 3 and c-caspase 8 were

reduced in the SHPS-1 mutant mice

(*p < 0.05 vs. WT, #p < 0.05 vs. sham,

n = 6 for each group).



(a)

(b)

(c)

Fig. 4 The oxidative injury was attenuated

in the SHPS-1 MT (mutant)mice. (a)

Immunofluorescence staining indicates that

the numbers of both 8OHdG-positive cells

(red) and 4HNE-positive cells (red) in the

peri-infarct area were significantly reduced

in the SHPS-1 MT mice, blue is nucleus

stained by DAPI. (*p < 0.05 vs. WT, n = 4

for each group). (b) Quantitative real-time

PCR and western blot analysis indicate that

the mRNA and protein levels of the anti-

oxidative genes Nrf2 and heme oxygenase

1 were up-regulated in the SHPS-1 MT

mice (*p < 0.05 vs. WT, # p < 0.05 vs.

sham, n = 6 for each group). (c) Immuno-

fluorescence staining reveals that the Nrf2

positive cell number and light intensity of

Nrf2 are increased in the peri-infarct area of

SHPS-1 MT mice, in addition, the expres-

sion of Nrf2 protein is more concentrated in

the nucleus in SHPS-1 MT mice.



a family of cysteine proteases that play essential roles inprogrammed cell death, necrosis, and inflammation. Caspase8 is involved in the programmed cell death induced by Fasand various apoptotic stimuli. Activated caspase 8 cleavesand activates downstream effector caspases, such as caspase1, caspase 3, caspase 6, and caspase 7. Caspase 3 ultimatelyelicits the morphological hallmarks of apoptosis, includingDNA fragmentation and cell shrinkage. Several studies havedemonstrated that oxidative stress-induced neural injury ismediated by caspase 8 and caspase 3 (Russell et al. 2002;Wang et al. 2002). In the present study, the inhibition ofneural injury appeared to be attributed to the mitigation ofoxidative stress in the SHPS-1 MT mice.

We observed that the phosphorylation of SHP-1 and SHP-2 was inhibited after cerebral ischemia in the SHPS-1 MT

mice. SHP-1 and SHP-2 are SH2 domain-containing non-transmembrane protein tyrosine phosphatases. When SHPS-1is activated by growth factors, cell adhesion or other stimuli,the phosphorylated cytoplasmic region binds to and activatesSHP-1 and SHP-2. SHP-1 and SHP-2 have been implicatedin several signaling pathways, such as the MAPK, JAK/STAT, and PI3K-Akt pathways (Dubois et al. 2006; Pandeyet al. 2009; Won et al. 2011). In addition, SHP-1 and SHP-2have been shown to interact with a variety of signalingintermediates, and they dephosphorylate associated signalingmolecules, which negatively regulate the local signal path-ways. In the present study, higher levels of Akt phosphor-ylation were detected in the SHPS-1 MT mice, which waslikely because of decreased activity of SHP-1 and SHP-2(Zhang et al. 2002; Lodeiro et al. 2011).

In previous studies, it has been demonstrated that thePI3K/Akt signaling pathway is involved in the expressionand activation of Nrf2 and that the induction of Nrf2 proteinexpression contributes to the transcriptional activity of Nrf2(Martin et al. 2004; Bak et al. 2012). Not only does it affectthe accumulation of Nrf2, but the PI3K/Akt signalingpathway also contributes to the activation of Nrf2 bypromoting transfer of cytoplasmic Nrf2 to the nucleus(Wang et al. 2008). GSK-3b is a downstream signalingmolecule of the PI3K/Akt pathway whose activity can beinhibited by Akt-mediated phosphorylation at Ser9. Aprevious study demonstrated that GSK-3b down-regulatedNrf2 activity; thus, phosphorylation of GSK-3b can increasethe activity of Nrf2 (Rojo et al. 2008). In the present study,the mRNA level and protein level of Nrf2 were up-regulatedin SHPS-1 MT mice, the expression of Nrf2 protein wasmore concentrated in the nucleus in SHPS-1 MT mice, themRNA, and protein level of Nrf2 targeted gene HO-1 werealso up-regulated in SHPS-1 MT mice. The increasingexpression and activity of Nrf2 is probably attributed to thephosphorylation of Akt/GSK-3b.

In conclusion, this study suggested that the SHPS-1deficiency protects the brain from acute ischemic injury.Oxidative stress was mitigated and neural injury wasinhibited in SHPS-1 MT mice. In addition, we demonstratedthat Akt was activated in the brains of ischemic SHPS-1 MTmice. The activated Akt signaling may account for theup-regulation of Nrf2 and HO-1 and may eventually lead tothe decline of oxidative stress. These findings may provide anew therapeutic target for the treatment of acute ischemicstroke.

Acknowledgements

The SHPS-1 mutant mice were provided by the RIKEN BRCthrough the National Bio-Resource Project of The MEXT Japan.This work was supported in part by National Natural ScienceFoundation of China (NO. 81100230), National Science andTechnology Support Project (NO. 2011BAI15B02).

Fig. 5 The SHPS-1 deficiency led to increased Akt activation. Wes-

tern blot analysis indicates that the phosphorylation of SHP-1 and

SHP-2 (p-SHP-1, p-SHP-2) was inhibited in the ischemic brain of

SHPS-1 MT (mutant) mice, whereas the phosphorylation of Akt and

GSK3b (p-Akt, p-GSK3b) was elevated in the SHPS-1 MT mice

compared with the WT mice (*p < 0.05 vs. WT, # p < 0.05 vs. sham,

n = 6 for each group). The total protein levels of SHP-1, SHP-2, Akt

and GSK3b were not changed in the SHPS-1 MT mice.



References

Allen C. L. and Bayraktutan U. (2009) Oxidative stress and its role in thepathogenesis of ischaemic stroke. Int. J. Stroke 4, 461–470.

Araki T., Yamada M., Ohnishi H., Sano S. I. and Hatanaka H. (2000)BIT/SHPS-1 enhances brain-derived neurotrophic factor-promotedneuronal survival in cultured cerebral cortical neurons. J. Neuro-chem. 75, 1502–1510.

Aztatzi-Santillan E., Nares-Lopez F. E., Marquez-Valadez B., AguileraP. and Chanez-Cardenas M. E. (2010) The protective role of hemeoxygenase-1 in cerebral ischemia. Cent. Nerv. Syst. Agents Med.Chem. 10, 310–316.

Bak M. J., Jun M. and Jeong W. S. (2012) Procyanidins from wild grape(vitis amurensis) seeds regulate ARE-mediated enzyme expressionvia Nrf2 coupled with p38 and PI3K/Akt pathway in HepG2 cells.Int. J. Mol. Sci. 13, 801–818.

Chan P. H. (2001) Reactive oxygen radicals in signaling and damage inthe ischemic brain. J. Cereb. Blood Flow Metab. 21, 2–14.

Connolly Jr E. S., Winfree C. J., Springer T. A., Naka Y., Liao H., Yan S.D., SternD.M., SolomonR.A.,Gutierrez-Ramos J. C. andPinskyD.J. (1997)Cerebral protection in homozygous null ICAM-1mice aftermiddle cerebral artery occlusion. Role of neutrophil adhesion in thepathogenesis of stroke. J. Clin. Invest. 97, 209–216.

Dubois M. J., Bergeron S., Kim H. J. et al. (2006) The SHP-1 proteintyrosine phosphatase negatively modulates glucose homeostasis.Nat. Med. 12, 549–556.

El Kossi M. M. and Zakhary M. M. (2000) Oxidative stress in thecontext of acute cerebrovascular stroke. Stroke 31, 1889–1892.

Fujii M., Hara H., Meng W., Vonsattel J. P., Huang Z. and MoskowitzM. A. (1997) Strain-related differences in susceptibility to transientforebrain ischemia in SV-129 and C57black/6 mice. Stroke 28,1805–1810; discussion 1811.

Galbaugh T., Feeney Y. B. and Clevenger C. V. (2010) Prolactinreceptor-integrin cross-talk mediated by SIRPa in breast cancercells. Mol. Cancer Res. 8, 1413–1424.

Ikeda H., Okazawa H., Ohnishi H., Murata Y., Oldenborg P. A. andMatozaki T. (2006) Mutational analysis of the mechanism ofnegative regulation by SRC homology 2 domain-containing pro-tein tyrosine phosphatase substrate-1 of phagocytosis in macro-phages. J. Immunol. 177, 3123–3132.

Imai H., Masayasu H., Dewar D., Graham D. I. and Macrae I. M. (2001)Ebselen protects both gray and white matter in a rodent model offocal cerebral ischemia. Stroke 32, 2149–2154.

Inagaki K., Yamao T., Noguchi T., Matozaki T., Fukunaga K., TakadaT., Hosooka T., Akira S. and Kasuga M. (2000) SHPS-1 regulatesintegrin-mediated cytoskeletal reorganization and cell motility.EMBO J. 19, 6721–6731.

Kapoor G. S. and O’Rourke D. M. (2010) SIRPalpha1 receptors interferewith the EGFRvIII signalosome to inhibit glioblastoma celltransformation and migration. Oncogene 29, 4130–4144.

Kawai H., Deguchi S., Deguchi K. et al. (2011) Synergistic benefit ofcombined amlodipine plus atorvastatin on neuronal damage afterstroke in Zucker metabolic rat. Brain Res. 1368, 317–323.

Kelly P. J., Morrow J. D., Ning M. et al. (2008) Oxidative stress andmatrix metalloproteinase-9 in acute ischemic stroke: the BiomarkerEvaluation for Antioxidant Therapies in Stroke (BEAT-Stroke)study. Stroke 39, 100–104.

Kim K. C., Kang K. A., Zhang R., Piao M. J., Kim G. Y., Kang M. Y.,Lee S. J., Lee N. H., Surh Y. J. and Hyun J. W. (2010)Up-regulation of Nrf2-mediated heme oxygenase-1 expression byeckol, a phlorotannin compound, through activation of Erk andPI3K/Akt. Int. J. Biochem. Cell Biol. 42, 297–305.

Koshimizu H., Araki T., Takai S., Yokomaku D., Ishikawa Y., KubotaM., Sano S., Hatanaka H. and Yamada M. (2002) Expression of

CD47/integrin-associated protein induces death of cultured cere-bral cortical neurons. J. Neurochem. 82, 249–257.

Kweon M. H., Adhami V. M., Lee J. S. and Mukhtar H. (2006) Con-stitutive overexpression of Nrf2-dependent heme oxygenase-1 inA549 cells contributes to resistance to apoptosis induced byepigallocatechin 3-gallate. J. Biol. Chem. 281, 33761–33772.

Lei C., Deng J., Wang B., Cheng D., Yang Q., Dong H. and Xiong L.(2011) Reactive oxygen species scavenger inhibits STAT3 acti-vation after transient focal cerebral ischemia-reperfusion injury inrats. Anesth. Analg. 113, 153–159.

Lodeiro M., Alen B. O., Mosteiro C. S. et al. (2011) The SHP-1 proteintyrosine phosphatase negatively modulates Akt signaling in theghrelin/GHSR1a system. Mol. Biol. Cell 22, 4182–4191.

Martin D., Rojo A. I., Salinas M., Diaz R., Gallardo G., Alam J., DeGalarreta C. M. and Cuadrado A. (2004) Regulation of hemeoxygenase-1 expression through the phosphatidylinositol 3-kinase/Akt pathway and the Nrf2 transcription factor in response to theantioxidant phytochemical carnosol. J. Biol. Chem. 279,8919–8929.

Mathers C. D., Boerma T. and Ma Fat D. (2009) Global and regionalcauses of death. Br. Med. Bull. 92, 7–32.

Mitsuhashi H., Futai E., Sasagawa N., Hayashi Y., Nishino I. and IshiuraS. (2008) Csk-homologous kinase interacts with SHPS-1 and en-hances neurite outgrowth of PC12 cells. J. Neurochem. 105,101–112.

Ohnishi H., Murata T., Kusakari S. et al. (2010) Stress-evoked tyrosinephosphorylation of signal regulatory protein a regulates behavioralimmobility in the forced swim test. J. Neurosci. 30, 10472–10483.

Oshima K., Ruhul Amin A. R., Suzuki A., Hamaguchi M. and MatsudaS. (2002) SHPS-1, a multifunctional transmembrane glycoprotein.FEBS Lett. 519, 1–7.

Pandey M. K., Sung B., Ahn K. S. and Aggarwal B. B. (2009) Buteinsuppresses constitutive and inducible signal transducer and acti-vator of transcription (STAT) 3 activation and STAT3-regulatedgene products through the induction of a protein tyrosine phos-phatase SHP-1. Mol. Pharmacol. 75, 525–533.

Rojo A. I., Sagarra M. R. and Cuadrado A. (2008) GSK-3beta down-regulates the transcription factor Nrf2 after oxidant damage:relevance to exposure of neuronal cells to oxidative stress. J.Neurochem. 105, 192–202.

Russell J. W., Golovoy D., Vincent A. M., Mahendru P., Olzmann J. A.,Mentzer A. and Feldman E. L. (2002) High glucose-induced oxi-dative stress and mitochondrial dysfunction in neurons. FASEB J.16, 1738–1748.

Shen X., Xi G., Radhakrishnan Y. and Clemmons D. R. (2010) PDK1recruitment to the SHPS-1 signaling complex enhances insulin-likegrowth factor-i-stimulated AKT activation and vascular smoothmuscle cell survival. J. Biol. Chem. 285, 29416–29424.

Sobota R. M., Muller P. J., Khouri C., Ullrich A., Poli V., Noguchi T.,Heinrich P. C and Schaper F. (2008) SHPS-1/SIRP1alpha con-tributes to interleukin-6 signalling. Cell. Signal. 20, 1385–1391.

Solis W. A., Dalton T. P., Dieter M. Z., Freshwater S., Harrer J. M., HeL., Shertzer H. G. and Nebert D.W. (2002) Glutamate-cysteineligase modifier subunit, mouse Gclm gene structure and regulationby agents that cause oxidative stress. Biochem. Pharmacol. 63,1739–1754.

Stofega M. R., Argetsinger L. S., Wang H., Ullrich A. and Carter-Su C.(2000) Negative regulation of growth hormone receptor/JAK2signaling by signal regulatory protein alpha. J. Biol. Chem. 275,28222–28229.

Venugopal R. and Jaiswal A. K. (1996) Nrf1 and Nrf2 positively andc-Fos and Fra1 negatively regulate the human antioxidant responseelement-mediated expression of NAD(P)H: quinone oxidoreduc-tase1 gene. Proc. Natl Acad. Sci. USA 93, 14960–14965.



Wang X., Mori T., Sumii T. and Lo E. H. (2002) Hemoglobin-inducedcytotoxicity in rat cerebral cortical neurons: caspase activation andoxidative stress. Stroke 33, 1882–1888.

Wang L., Chen Y., Sternberg P. and Cai J. (2008) Essential roles of thePI3 kinase/Akt pathway in regulating Nrf2-dependent antioxidantfunctions in the RPE. Invest Ophthalmol. Vis. Sci. 49, 1671–1678.

Wang B., Cao W., Biswal S. and Dore S. (2011) Carbon monoxide-activated Nrf2 pathway leads to protection against permanent focalcerebral ischemia. Stroke 42, 2605–2610.

Won K. J., Lee H. M., Lee C. K. et al. (2011) Protein tyrosinephosphatase SHP-2 is positively involved in platelet-derivedgrowth factor-signaling in vascular neointima formation via thereactive oxygen species-related pathway. J. Pharmacol. Sci. 115,164–175.

Xia C. F., Smith Jr R. S., Shen B., Yang Z. R., Borlongan C. V., Chao L.and Chao J. (2006) Postischemic brain injury is exacerbated inmice lacking the kinin B2 receptor. Hypertension 47, 752–761.

Xing C., Lee S., Kim W. J., Jin G., Yang Y. G., Ji X., Wang X. and LoE. H. (2009) Role of oxidative stress and caspase 3 in CD47-mediated neuronal cell death. J. Neurochem. 108, 430–436.

Yamao T., Matozaki T., Amano K., Matsuda Y., Takahashi N., Ochi F.,Fujioka Y. and Kasuga M. (1997) Mouse and human SHPS-1:molecular cloning of cDNAs and chromosomal localization ofgenes. Biochem. Biophys. Res. Commun. 231, 61–67.

Zhang S. Q., Tsiaras W. G., Araki T., Wen G., Minichiello L., Klein R.and Neel B. G. (2002) Receptor-specific regulation of phosphati-dylinositol 3¢-kinase activation by the protein tyrosine phosphataseShp2. Mol. Cell. Biol. 22, 4062–4072.



Documents

SHPS-1 deficiency induces robust neuroprotection against experimental stroke by attenuating oxidative stress