Aea

SIB

a

ARRAA

KANEXPC

1

acdecc1im

dcnPX

0d

Toxicology 279 (2011) 189–195

Contents lists available at ScienceDirect

Toxicology

journa l homepage: www.e lsev ier .com/ locate / tox ico l

dipose-derived mesenchymal stem cells protect PC12 cells from glutamatexcitotoxicity-induced apoptosis by upregulation of XIAP through PI3-K/Aktctivation

han Lu, Chunhua Lu, Qin Han, Jing Li, Zhijian Du, Lianming Liao, Robert Chunhua Zhao ∗

nstitute of Basic Medical Sciences & School of Basic Medicine, Center of Excellence in Tissue Engineering, Chinese Academy of Medical Sciences & Peking Union Medical College,eijing, PR China

r t i c l e i n f o

rticle history:eceived 20 September 2010eceived in revised form 20 October 2010ccepted 26 October 2010vailable online 30 October 2010

eywords:dipose-derived mesenchymal stem celleuroprotectionxcitotoxicity

a b s t r a c t

Glutamate excitotoxicity has been implicated as one of the factors contributing to neuronal apoptosis andis involved in many neurodegenerative diseases. Previous studies suggest that mesenchymal stem cellshave the ability to protect cultured neurons from excitotoxicity-induced apoptosis, although the under-lying mechanisms are not clear. In this study, we evaluated whether adipose mesenchymal stem cells(AMSCs) could protect against glutamate-induced injury in PC12 cells by secreting neurotrophic factors.We found that AMSCs secreted neurotrophic factors including vascular endothelial growth factor (VEGF),hepatocyte growth factor (HGF), brain-derived neurotrophic factor (BDNF) and nerve growth factor (NGF)under both normoxic and hypoxic conditions. AMSC – conditioned medium (AMSC-CM) had a protectiveeffect on excitotoxicity-injured PC12 cells, as indicated by increased cell viability, decreased number of

IAPI3-K/Akt pathwayaspase-3 activity

TUNEL-staining positive nuclei and lowered caspase-3 activity. By using neutralizing monoclonal anti-bodies and specific inhibitors, VEGF, HGF and BDNF were identified as the mediators of AMSC effects andPI3-K/Akt and MAPK pathways were involved. Western blot analysis showed that AMSC-CM can increasethe level of p-Akt, up-regulate XIAP and reduce the level of cleaved-caspase-3 in PC12 cells. These resultssuggest that AMSCs can effectively protect PC12 cells from glutamate excitotoxicity-induced apopto-sis and support the hypothesis that AMSCs may be a useful treatment for stroke or neurodegenerative

lve e

diseases which often invo

. Introduction

Glutamate excitotoxicity is one of the major pathological mech-nisms that cause neuronal injury in nervous diseases such aserebral ischemia (stroke), brain trauma and neurodegenerativeisease (Lipton and Rosenberg, 1994). Excessive accumulation ofxtracellular glutamate may lead to hyperexcitability of neurons,alcium overload and increased intracellular oxygen free radi-

als, resulting in DNA injury and apoptosis (Kruman and Mattson,999; Reynolds and Hastings, 1995). Massive neuronal death will

nevitably lead to the dysfunction of nervous system. Therefore,inimizing neuronal apoptosis is the most direct and effective

Abbreviations: AMSC, adipose-derived mesenchymal stem cell; BDNF, brain-erived neurotrophic factor; ERK, extracellular signal-regulated kinase; FCS, fetalalf serum; HGF, hepatocyte growth factor; NGF, nerve growth factor; NTF-3,eurotrophin-3; MAPK, mitogen-activated protein kinase; MEK, MAPK/ERK kinase;I3-K, phosphatidylinositol 3-kinase; VEGF, vascular endothelial growth factor;IAP, X-chromosome-linked inhibitor of apoptosis protein.∗ Corresponding author. Tel.: +86 10 65125311; fax: +86 10 65125311.

E-mail address: chunhuaz@public.tpt.tj.cn (R.C. Zhao).

300-483X/$ – see front matter © 2010 Elsevier Ireland Ltd. All rights reserved.oi:10.1016/j.tox.2010.10.011

xcitotoxicity.© 2010 Elsevier Ireland Ltd. All rights reserved.

therapeutic approach. The PI3-K/Akt and MAPK pathways are reg-ulated and controlled by growth factors to guarantee neuronalviability. Active Akt inhibits apoptosis by regulating the expressionof Bcl-2 and Bax (Matsuzaki et al., 1999). Furthermore, active ERKsand Akt are translocated to the nucleus where they can phosphory-late certain transcription factors, thereby regulating the expressionof specific genes that contribute to cell survival (Hanada et al.,2004). In addition, studies showed that the Notch pathway whichwidely exists in nervous system is also involved in neural cell sur-vival and differentiation (Imayoshi et al., 2010; Shin et al., 2008).In recent years, the inhibitors of apoptosis protein (IAP) familyhas attracted attention, with one of its members, X-chromosome-linked IAP (XIAP), reported to be responsible for anti-apoptosis inneural cells (West et al., 2009).

Mesenchymal stem cells (MSCs) are non-hematopoietic stemcells with the ability to differentiate into mesenchymal tissues such

as bone, cartilage and adipose tissue. MSCs have been shown tohave neuroprotective effects both in vitro and in vivo by secret-ing various nutrition factors and reducing apoptosis of nerve cellsunder pathological circumstances (Zhao et al., 2009). However, theunderlying mechanism of this protection remains unclear. PC12

1 ogy 27

cttwim

2

2

IipoC

2

tiAw

2h

eu(u

2

ccMppGATTAAAareU

2

tf

2

eiaAaiIvwUieFr

90 S. Lu et al. / Toxicol

ells, a clonal rat pheochromocytoma cell line, have been showno have many favourable properties as cellular models for neuro-oxicity testing (Almeida et al., 2005; Shan et al., 2004). In this study,e examined whether AMSCs could protect against glutamate-

nduced neurotoxicity in PC12 cells and explored the underlyingechanisms for this protection.

. Materials and methods

.1. Chemicals and reagents

Anti-human HGF, VEGF and BDNF antibodies were purchased from R&D Systemsnc. LY294002 and U0126 were bought from Cell Signaling Technology. �-secretasenhibitor I (GSI) was obtained from Calbiochem. Rabbit polyclonal antibodies againsthospho-ERK1/2, ERK, Akt and Actin were bought from Santa Cruz Biotechnol-gy. Rabbit polyclonal antibodies against phospho-Akt (Ser473), XIAP, Caspase3 andleaved-Caspase3 were purchased from Cell Signaling Technology.

.2. Preparation of human adipose mesenchymal stem cells (AMSCs)

Human subcutaneous adipose tissue samples were obtained from lipoaspira-ion/liposuction procedures after informed consent was obtained from the patients,n accordance with procedures approved by the Ethics Committee at Chinesecademy of Medical Sciences and Peking Union Medical College. All the experimentsere done with the 3rd–4th passage AMSCs.

.3. Preparation of conditioned medium from AMSCs cultured in normoxia andypoxia

Human AMSCs were cultured and expanded in culture flasks and used for thexperiments at 90% confluence. Cells were switched to serum-free DF12 mediumnder either normoxic (21% O2) or hypoxic (2% O2) condition. The conditioned mediaAMSC-CM) were collected after 12 h, 24 h and 48 h respectively and stored at −80 ◦Cntil use.

.4. RT-PCR

Total cellular RNA was extracted from AMSCs cultured in normoxic or hypoxiconditions for 24 h, using Trizol (Invitrogen) according to the manufacturer’s proto-ol. cDNA was made from 1 �g total mRNA using oligo dT and Reverse Transcriptase-MLV (Takara) at 42 ◦C for 1 h and at 95 ◦C for 5 min. Then PCR amplification was

erformed for human VEGF, HGF, BDNF, NGF, NTF3, NT4/5 and Actin with TaqDNAolymetase (Takara). The following primer pairs were used: VEGF: 5′-TTC ATG GATTC TAT CAG CG-3′/5′-CAT CTC TCC TAT GTG CTG GC-3′; HGF: 5′-ACG AAC ACA GCTTC GGG GTA-3′/5′-CAT CAA AGC CCT TGT CGG GAT-3′; BDNF: 5′-TCA TAC TTT GGTGC ATG AAG GC-3′/5′-GGC CGA ACT TTC TGG TCC T-3′; NGF: 5′-AGG GAG CAG CTTCT ATC CTG-3′/5′-GGC AGT GTC AAG GGA ATG C-3′; NTF3: 5′-CCG TGG CAT CCAGG TAA CAA-3′/5′-CAA TCA CCG GCT GGA ATG CT-3′; NT4/5: 5′-CCC ACC CTC AACTT GCC C-3′/5′ TGC AGC GGG TTT CAA AGA AGT-3′; Actin: 5′-GCT CCT CCT GAG CGCAG TA-3′/5′-GAT GGA GGG GCC GGA CT-3′ . PCR amplification was run by 35 cyclest 95 ◦C for 30 s, at 58 ◦C for 30 s and at 72 ◦C for 30 s, with 5 min of initial denatu-ation at 95 ◦C and 10 min of final synthesis at 72 ◦C. Amplification products werelectrophoresed by 2% agarose gel containing ethidium bromide and visualized byV-induced fluorescence.

.5. Determination of growth factors in AMSC-CM

The supernatants collected from cultured AMSCs were analyzed by ELISA, usinghe Quantikine® Human BDNF, VEGF, HGF and �-NGF Immunoassays (R&D Systems)ollowing the manufacturer’s instructions. All samples were analyzed in triplicate.

.6. Cell viability experiments

Cell viability experiments were performed as described previously (Almeidat al., 2005) with modifications. After the PC12 cells (105 cells/well) were culturedn 96-well plates for 24 h, 0.5 mM glutamate was added and cultured for 15 mint room temperature. The medium was then replaced with either DF12 medium,MSC-CM alone or AMSC-CM with neutralization antibodies against VEGF, HGFnd BDNF-(either alone or combined) for 24 h. In some experiments, either Aktnhibitor LY294002 (10 �M), MEK inhibitor U0126 (10 �M) or �-secretase inhibitorGSI (1 �M) was added to AMSC-CM instead of neutralization antibodies. For celliability determination, 20 �l MTS solution was added to each well and the cells

ere incubated at 37 ◦C for 2 h. Absorbance was subsequently measured at 490 nm.ntreated cells were considered as control and the culture medium without cells

n the presence of MTS solution was used as solution background. Cell viability inach group was expressed as a percentage relative to the control group absorbance.or the MTS assay, there were three samples in each group, and the experiment wasepeated 5 times.

9 (2011) 189–195

For Western blot analysis, the PC12 cells were cultured in six-well plates(1 × 106 cells/well). The concentrations of LY294002 (10 �M), U0126 (10 �M) andGSI (1 �M) used in the present study were not toxic to PC12 cells (data not shown).

2.7. TUNEL staining

An apoptotic cell death detection kit (TMR red, Roche Applied Science) basedon the TUNEL assay was used to detect apoptotic cells following the manufacturer’sinstructions. Each experiment included negative controls by omitting the TUNELenzyme TdT reaction mixture and incubating the cells with the label solution pro-vided in the kit. Cells were stained with fluorescent dye Hoechst 33342 (0.5 mg/ml)to stain the total nuclei and were examined with a fluorescence microscope (Olym-pus BX51).

2.8. Caspase-3 activity assay

PC12 cells were cultured in six-well plates at a density of 1 × 106 cells/well for24 h. The cells were then treated with 0.5 mM glutamate at room temperature for15 min, and then replaced with DF12 medium or AMSC-CM for additional 24 h.Caspase-3 activity was measured using a caspase-3 colorimetric activity assay kitfrom BioVision. Colorimetric reaction was developed and measured at 405 nm in aBioRad plate reader.

2.9. Western blotting analysis

The PC12 cells (1 × 106 cells/well) were washed twice with cold PBS and lysedon ice with RIPA buffer. Equal amounts of proteins (40 �g per lane) was separated bySDS-PAGE gels and then transferred to PVDF membranes (Millipore). After blocking,the membranes were incubated with the appropriate primary antibodies overnightat 4 ◦C (anti-phospho-Akt, 1:2000; anti-Akt, 1:2000; anti-phospho-ERK1/2, 1:1000;anti-ERK1/2, 1:2000; anti-XIAP, 1:1000; anti-cleaved-caspase-3, 1:1000; anti-caspase-3, 1:1000; anti-Actin, 1:2000). Following three washes with TBST, the blotswere incubated with the secondary horseradish peroxidase-conjugated antibody(1:3000) at room temperature for 1 h. Immunocomplexes were visualized by usingenhanced chemiluminescence (Millipore) following the manufacturer’s instruc-tions. The bands were subsequently analyzed densitometrically with Quantity One4.6.2. Software (BioRad).

2.10. Statistical analysis

Results were presented as means ± SEM of the indicated experiments, per-formed in independent preparations. All of the results were analyzed using one-wayANOVA, followed by multiple comparison tests. All analyses were performed withSPSS 11.5 software. P-values of less than 0.05 were considered to indicate statisticalsignificance.

3. Results

3.1. Secretion of growth factors by human AMSCs

To determine which growth factors were secreted from AMSCs,we performed RT-PCR and ELISA assays. By RT-PCR, we found thatunder normoxic culture condition, AMSCs expressed VEGF, HGF,BDNF and NGF mRNA, but not NTF3 or NT4/5 mRNA. Addition-ally, the mRNA expression pattern of these factors under hypoxiawas very similar to those under normoxia (Fig. 1A). Further exper-iment with ELISA showed AMSCs secreted 862 ± 31 pg/ml of VEGF,1504 ± 41 pg/ml of HGF, 359 ± 32 pg/ml of BDNF and 29 ± 2 pg/ml ofNGF into the cell culture medium (Fig. 1B). Furthermore, the secre-tion of VEGF and HGF increased under hypoxic condition (Fig. 1C).

3.2. AMSC-CM protects PC12 cells from glutamate-inducedapoptosis

In order to investigate whether AMSC-CM could protect PC12cells from glutamate excitotoxicity, we performed a cell viabil-ity test by MTS assay. The results showed that AMSC-CM could

protect PC12 cells from glutamate-induced apoptosis. Cell viabil-ity decreased when either 1 �g/ml VEGF neutralization antibody,10 �g/ml BDNF neutralization antibody, or 10 �g/ml HGF neutral-ization antibody was added. When three neutralization antibodieswere added simultaneously, the viability further decreased (Fig. 2).

S. Lu et al. / Toxicology 279 (2011) 189–195 191

Fig. 1. Expression of growth factors in human AMSCs. (A) Results of RT-PCR showedhuman AMSCs expressed VEGF, HGF, BDNF and NGF mRNA but not NT3 or NT4/5mRNA. Actin gene was used as positive control. (B) ELISA analysis showed AMSCscultured in normoxia secreted VEGF, HGF, BDNF and NGF. (C) Secretion of VEGFae

3n

nceidtwLAioaTseU

Fig. 2. AMSC-CM protects PC12 cells from glutamate-induced apoptosis. (A)After exposure to 0.5 mM glutamate for 15 min, PC12 cells were culturedin basic medium (Glu), AMSC-CM (Glu + CM), AMSC-CM with anti-VEGF neu-tralization antibody (Glu + CM + antiV), AMSC-CM with anti-HGF neutralizationantibody (Glu + CM + antiH), AMSC-CM with anti-BDNF neutralization antibody(Glu + CM + antiB) or AMSC-CM with neutralization antibodies combination(Glu + CM + antiVHB), respectively. Cells in control group were not exposed to glu-

apoptosis by binding to initiator caspase-9 and effector caspases-3and -7. We evaluated the expression of caspase-3, cleaved-caspase-

nd HGF peaked at 48 h while BDNF paeked at 12 h. Data from three independentxperiments are presented as means ± SEM.

.3. PI3-K/Akt and MAPK/ERK pathways are involved ineuroprotection of AMSC-CM against glutamate toxicity

The PI3-K/Akt and MAPK/ERK pathways, which can promoteeuronal growth and survival, represent two of the key signalingascades regulated by growth factor-receptor activation (Almeidat al., 2005; Hanada et al., 2004), while the Notch pathway is mainlynvolved in cell survival and differentiation (Shin et al., 2008). Toetermine whether these signaling pathways were involved inhe protective effects of AMSC-CM, corresponding signal inhibitorsere used in the following experiments. We found that both

Y294002 and U0126 can counteract the neuroprotective effect ofMSC-CM (Fig. 3A). Unexpectedly, addition of the Notch pathway

nhibitor GSI significantly increased cell viability in the presencef AMSC-CM and further experiments showed that GSI alone couldlso decrease glutamate-induced cell injury (Fig. 3B). In addition,UNEL staining showed that AMSC-CM and AMSC-CM with GSI

ignificantly inhibited the glutamate-induced apoptosis, while theffect of AMSC-CM was partly counteracted by LY294002 and0126 (Fig. 3C–H).

tamate. Cell viability was assessed 24 h later by MTS assay and represented aspercentages of the control group. Bars represent the means ± S.E.M. from five inde-pendent experiments. *P < 0.05 vs. Glu group, #P < 0.05 vs. AMSC-CM group. ANOVAfollowed by Dunnett’s test.

3.4. PI3-K/Akt pathway is critical for neuroprotection ofAMSC-CM against glutamate toxicity

To examine which signaling pathway was more important tothe neuroprotection of AMSC-CM, we used Western blot analysisto determine p-Akt and p-ERK levels in the PC12 cells. In the pres-ence of 0.5 mM glutamate acid, p-Akt level in PC12 cells decreasedsignificantly. This was almost reversed after cells were cultured inAMSC-CM for 24 h. p-Akt level in PC12 cells changed slightly wheneither BDNF, VEGF or HGF neutralization antibodies were added.However, when these antibodies were added simultaneously, theysignificantly counteracted the effect of AMSC-CM (Fig. 4). LY294002inhibited Akt phosphorylation, while U0126 appeared to slightlyincrease p-Akt level (Fig. 5A).

As shown by Western blot, phosphorylation of ERK, which is theactive form in the MAPK signaling pathway, increased significantlywhen PC12 cells were injured by glutamate. This may have beencaused by the activation of cellular self-repair pathways by exci-totoxicity. However, p-ERK level in cells from the AMSC-CM groupshowed no significant changes, indicating that PC12 cells were notseverely injured in the presence of AMSC-CM. The p-ERK levelslightly decreased when neutralization antibody(ies) was/wereadded to the culture medium (Fig. 4). Addition of U0126 into AMSC-CM further inhibited the phosphorylation of ERK (Fig. 5B), whileaddition of LY294002 may increase the p-ERK level. Taken together,these findings indicate possible cross-talk between the PI3-K/Aktand MAPK pathways.

3.5. AMSC-CM reduces caspase-3 activity by up-regulating XIAP

Since caspase-3 is the major executioner caspase in neurons,we measured caspase-3 activity in different treatment groups afterglutamate exposure and found that caspase-3 activity was signifi-cantly lower in the presence of AMSC-CM (Fig. 6A).

Recent research shows that XIAP may one of the targetmolecules in PI3-K/Akt pathway (Dan et al., 2004). XIAP suppresses

3 (the active form of caspase-3) and XIAP in different treatmentgroups to investigate the relationship between them (Fig. 6B).The results indicate that AMSC-CM can significantly increase XIAP

192 S. Lu et al. / Toxicology 279 (2011) 189–195

Fig. 3. Signaling pathway inhibitors counteract the neuroprotective effect of AMSC-CM. (A) Cell viability of injured PC12 cells was determined after culturing in basicmedium (Glu), AMSC-CM (Glu + CM), AMSC-CM + LY294002 (Glu + CM + LY) or AMSC-CM + U0126 (Glu + CM + U), respectively. (B) Cell viability was also evaluated after PC12cells exposure to 0.5 mM glutamate, the cells were cultured in basic medium (Glu), AMSC-CM (Glu + CM), DF12 + GSI (Glu + GSI) or AMSC-CM + GSI (Glu + CM + GSI), respectively.Cell viability was represented as percentages of the control group. Bars represent the means ± SEM of five independent experiments. *P < 0.05 vs. Glu group, #P < 0.05 vs.A icrogrn withA retativ

eoClLrPa

4

tiat2tfaschv

mT

MSC-CM group. ANOVA followed by Dunnett’s test. (C–H) Representative photomuclei (red) are shown. Merged photo of untreated PC12 cells (C). PC12 cells treatedMSC-CM + LY294002 (F), AMSC-CM + U0126 (G), or AMSC-CM + GSI (H). (For interpersion of the article.)

xpression, reduce cleaved-caspase-3 level, and have no impactn caspase-3 expression. When LY294002 was added to AMSC-M, the expression of XIAP was inhibited and cleaved-caspase-3

evel elevated. U0126 had similar effect, though was weaker thanY294002 (Fig. 6C). These results indicate that AMSC-CM caneduce glutamate-induced apoptosis mainly through activatingI3-K/Akt pathway and up-regulating XIAP to inhibit the caspase-3ctivity.

. Discussion

Neurotrophic factors such as BDNF and NGF are important forhe survival of neural cells (Almeida et al., 2005). In recent years,t has been shown that other growth factors such as VEGF, HGFnd bFGF can also have direct or indirect neurotrophic protec-ive effects (Deng et al., 2009; Ishihara et al., 2005; Tolosa et al.,009). In this study, we sought to determine which growth fac-ors were involved in the neuroprotective effects of AMSCs. Weound that under normoxic conditions, AMSCs secreted BDNF, VEGFnd HGF into supernatant at relatively high levels. Furthermore,ecretion of VEGF and HGF increased significantly under hypoxiconditions, indicating that AMSCs may play an important role in

ypoxia-ischemic lesions. The results are in accordance with pre-ious reports (Rehman et al., 2004).

When cerebral ischemia happens, a mass of glutamic acid accu-ulates between cells and causes hyperexcitability of neural cells.

he massive calcium influx activates NOS, causes accumulation

aphs (×100) of total nuclei stained with Hoechst 33342 (blue) and TUNEL-stainedglutamate alone (D). Glutamate-treated PC12 cells in the presence of AMSC-CM (E),on of the references to color in this figure legend, the reader is referred to the web

of oxygen free radicals, and results in DNA injury and apopto-sis (White et al., 2000). During this process, caspase-3 drivesapoptosis (Eldadah and Faden, 2000). In this study, we treatedglutamate-injured PC12 cells with AMSC-CM, finding that it effec-tively reduced apoptosis in PC12 cells (as determined by MTS andTUNEL assays). Previous studies have demonstrated that MSCscan secrete various growth factors including neurotrophic factors(Crigler et al., 2006); however, the factor types differ from onelaboratory to another (a result possibly due to differences in cul-tures systems and/or protocols for MSCs). BDNF was reported to bethe major neuroprotective factor secreted by MSC (Wilkins et al.,2009). In the present study we found that AMSCs also secretedother neuroprotective factors. When HGF neutralization antibodywas added to the AMSC-CM, neuroprotection of PC12 cells byAMSC-CM was counteracted, indicating that HGF is a critical factorsecreted by AMSC. In addition, simultaneous addition of neutral-ization antibodies against three nutrition factors further decreasedcell viability, indicating that all three nutrition factors are involvedin the neuroprotective effect of AMSC-CM. As the protective effectof AMSC-CM cannot be eliminated completely by neutralizationantibodies, it is likely that there are other factors in AMSC-CM thatcan protect PC12 cells, such as osteopontin, clusterin, cystatin-c,

tissue inhibitor of metalloproteinase 1, insulin-like growth factor-1 (IGF-1) and anti-oxidization molecules such as SOD3 (Kemp et al.,2009; Singla et al., 2008; Wei et al., 2009).

The PI3-K/Akt pathway is a central signal transduction path-ways involved in cell growth, survival and metabolism (Brazil et al.,

S. Lu et al. / Toxicology 279 (2011) 189–195 193

Fig. 4. All three nutrition factors are involved in the neuroprotective effect of AMSC-CM. After exposure to glutamate for 15 min, PC12 cells were treated with AMSC-CMalone or AMSC-CM plus neutralization antibodies. (A) Cells were lysed and total cellextracts (40 �g protein) were analyzed by Western blot using the anti-phospho-Aktand anti-phospho-ERK1/2 antibodies. Protein loading was checked by stripping andreprobing the membrane with an anti-Actin antibody. The bands were analyzedddvf

2icptapow

totHclg

Fig. 5. The neuroprotective effect of AMCS-CM is mediated by the PI3-K/Akt path-way. The injured PC12 cells were treated with AMSC-CM, AMSC-CM + LY294002and AMSC-CM + U0126 for 24 h respectively. Cells were lysed and total cell extractswere analyzed by Western blot, using the antibodies against Akt, phospho-Akt,ERK1/2 and phospho-ERK1/2. Representative Western blot analysis of p-Akt(A) and

ensitometrically with Quantity One 4.6.2. Data are means ± SEM of 3 indepen-ent experiments of (B) p-Akt and (C) p-ERK1/2 levels in different groups. +P < 0.05s. untreated group, *P < 0.05 vs. Glu group, #P < 0.05 vs. AMSC-CM group. ANOVAollowed by Tukey test.

004). Many research groups have demonstrated that PI3-K activ-ty is responsible for as much as 80% of neurotrophin-regulatedell survival, indicating that PI3-K is the major survival-promotingrotein for neurons (Hanada et al., 2004). Our findings supporthese earlier results. As shown in Western blot analysis, AMSC-CMctivated the PI3-K/Akt pathway, significantly promoted Akt phos-horylation and reduced apoptosis of PC12 cells. With the additionf Akt inhibitor LY294002, protection of PC12 cells by AMSC-CMas significantly counteracted.

Neurotrophic factors and growth factors have been reportedo protect neuronal cells gainst injury of excitotoxicity, calciumverload, oxidative injury, hypoxia, or neurotoxic viruses by activa-

ion of the ERK signaling (Fukunaga and Miyamoto, 1998; Han andoltzman, 2000). The MAPK/ERK pathway is the major pathway forell survival when cells are injured. Apoptosis happens when cellu-ar self-repair processes are insufficient to repair cell injury. Whenlutamate-injured PC12 cells were treated with AMSC-CM, ERK1/2

p-ERK1/2(B) from different groups is shown. Data are means ± SEM of three inde-pendent experiments. +P < 0.05 vs. untreated group, *P < 0.05 vs. Glu group, #P < 0.05vs. AMSC-CM group. ANOVA followed by Tukey test.

phosphorylation was slightly suppressed. This was aggravated bythe addition of ERK1/2 inhibitor U0126. This indicates that the ERKpathway also plays a role, albeit a smaller one, in the protectiveeffect of AMSC-CM.

Cross-talk between the PI3K/Akt and Raf/MEK/ERK pathwayswas observed in the present study. It has been reported that

PI3K activity is essential for the Raf/MEK/ERK cascade activation(Wennstrom and Downward, 1999). While other studies sug-gest that Akt can negatively regulate the Raf/MEK/ERK pathway(Rommel et al., 1999), in this study, it may be a protective mecha-

194 S. Lu et al. / Toxicology 27

Fig. 6. AMSC-CM decreases caspase-3 activation and induces up-regulation of XIAPlevel in cultured PC12 cells. After exposure to glutamate, cells were incubatedwith basic medium (glutamate) or AMSC-CM (glu + CM) for an additional 24 h.Untreated cells served as a control. (A) Caspase-3 activity assay. Bars representtWte

ne

t2

Brazil, D.P., Yang, Z.Z., Hemmings, B.A., 2004. New advances in Protein Kinase B

he means ± SEM from three independent experiments. *P < 0.05 vs. glutamate. (B)estern blot analysis of XIAP, cleaved-caspase-3 and total caspase-3 in different

reatment groups. (C) LY294002 and U0126 counter the effect of AMSC-CM on thexpression of XIAP and cleaved-caspase-3 in injured PC12 cells respectively.

ism utilized by PC12 cells to adapt to excitatory toxicity. But the

xact mechanisms need to be further investigated.

Studies have shown that PI3-K/Akt signaling inhibits apop-osis by regulating the expression of Bcl2 and Bax (Lim et al.,008). Besides Bcl2, the family of inhibitors of apoptosis (IAP)

9 (2011) 189–195

has drawn great attention in recent years, and one of its mem-bers X-chromosome-linked IAP (XIAP) is a potent anti-apoptosisfactor. The BIR2 domain of XIAP can bind to and directly inhibitcaspase-3 and 7 (Deveraux et al., 1997; Liston et al., 2003; Vauxand Silke, 2005). Some studies have shown that XIAP is one of thetarget molecules in the PI3-K/Akt pathway (Dan et al., 2004). Weobserved that the expression of XIAP in PC12 cells decreased sig-nificantly when exposed to glutamate and that this process couldbe reversed in the presence of AMSC-CM (which was accompa-nied by lower cleaved-caspase-3 activity). When adding the Aktinhibitor LY294002 to AMSC-CM, this function was eliminated.Thus AMSC-CM might increase XIAP expression by activating PI3-K/Akt signaling pathway and ultimately result in lower caspase-3activity and prevention of glutamate-induced apoptosis of PC 12cells.

We also found that GSI can prevent apoptosis induced byglutamate toxicity. The Notch family of transmembrane proteinreceptors is activated through proteolytic cleavage by �-secretase.Notch pathway activity exists widely in nervous system and isinvolved in the differentiation and survival of neural cells. Wefound that GSI and AMSC-CM can synergistically protect glutamate-injured PC12 cells. Western blot analysis showed that GSI canincrease the levels of p-ERK, suggesting that GSI may exert itsneuroprotective effect via MAPK pathway activation. GSI has beenreported to be able to inhibit proteasome activity, whether thiscapacity is involved in its effect against neuronal apoptosis is asubject of ongoing investigation.

Glutamate excitotoxicity is the common pathological effectorof nervous system diseases such as brain ischemia, traumatic braininjury, Alzheimer’s disease and other neurodegenerative diseases,as well as the cause of neuronal necrosis and apoptosis. We havedemonstrated that AMSCs exert neuroprotective effects in a well-established glutamate injury model. Our investigation suggests thatthe mechanisms underpinning this phenomenon rely at least partlyon secretion of neuroprotective factors (such as BDNF, VEGF andHGF), activation of the PI3-K/Akt pathway, increased XIAP expres-sion and reduced caspase-3 activity. Furthermore, we have shownthat GSI can cooperate with AMSC-CM to protect the PC12 cellsagainst glutamate toxicity. This work lends strong support to thehypothesis that AMSCs may ultimately be a useful treatment forstroke and other neurodegenerative diseases that involve excito-toxicity.

Conflict of interest

None.

Acknowledgments

Supported by grants from the ‘863 Projects’ of Ministryof Science and Technology of PR China (no. 2006AA02A109);National Natural Science Foundation of China (nos. 30830052,30911130363, U0970181); Beijing Ministry of Science and Tech-nology (no. D07050701350701) and major National Science andTechnology Project (nos. 2008zx09101-044, 2009ZX09503-025).

References

Almeida, R.D., Manadas, B.J., Melo, C.V., Gomes, J.R., Mendes, C.S., Graos, M.M., Car-valho, R.F., Carvalho, A.P., Duarte, C.B., 2005. Neuroprotection by BDNF againstglutamate-induced apoptotic cell death is mediated by ERK and PI3-kinase path-ways. Cell Death Differ. 12, 1329–1343.

signaling: AKTion on multiple fronts. Trends Biol. Sci. 29, 233–242.Crigler, L., Robey, R.C., Asawachaicharn, A., Gaupp, D., Phinney, D.G., 2006. Human

mesenchymal stem cell subpopulations express a variety of neuro-regulatorymolecules and promote neuronal cell survival and neuritogenesis. Exp. Neurol.198, 54–64.

ogy 27

D

D

D

E

F

H

H

I

I

K

K

L

L

L

M

R

bone marrow-derived mesenchymal stem cells secrete brain-derived neu-rotrophic factor which promotes neuronal survival in vitro. Stem Cell Res. 3,

S. Lu et al. / Toxicol

an, H.C., Sun, M., Kaneko, S., Feldman, R.I., Nicosia, S.V., Wang, H.G., Tsang, B.K.,Cheng, J.Q., 2004. Akt phosphorylation and stabilization of X-linked inhibitor ofapoptosis protein (XIAP). J. Biol. Chem. 279, 5405–5412.

eng, Y., Ye, W., Hu, Z., Yan, Y., Wang, Y., Takon, B., Zhou, G., Zhou, Y., 2009. Intra-venously administered BMSCs reduce neuronal apoptosis and promote neuronalproliferation through the release of VEGF after stroke in rats. Neurol. Res. 32,148–156.

everaux, Q.L., Takahashi, R., Salvesen, G.S., Reed, J.C., 1997. X-linked IAP is a directinhibitor of cell-death proteases. Nature 388, 300–304.

ldadah, B.A., Faden, A.I., 2000. Caspase pathways, neuronal apoptosis, and CNSinjury. J. Neurotrauma 17, 811–829.

ukunaga, K, Miyamoto, E., 1998. Role of MAP kinase in neurons. Mol. Neurobiol. 16,79–95.

an, B.H., Holtzman, D.M., 2000. BDNF protects the neonatal brain from hypoxic-ischemic injury in vivo via the ERK pathway. J. Neurosci. 20, 5775–5781.

anada, M., Feng, J., Hemmings, B.A., 2004. Structure, regulation and function ofPKB/AKT—a major therapeutic target. Biochim. Biophys. Acta 1697, 3–16.

mayoshi, I., Sakamoto, M., Yamaguchi, M., Mori, K., Kageyama, R., 2010. Essentialroles of Notch signaling in maintenance of neural stem cells in developing andadult brains. J. Neurosci. 30, 3489–3498.

shihara, N., Takagi, N., Niimura, M., Takagi, K., Nakano, M., Tanonaka, K., Funakoshi,H., Matsumoto, K., Nakamura, T., Takeo, S., 2005. Inhibition of apoptosis-inducing factor translocation is involved in protective effects of hepatocytegrowth factor against excitotoxic cell death in cultured hippocampal neurons. J.Neurochem. 95, 1277–1286.

emp, K., Hares, K., Mallam, E., Heesom, K., Scolding, N., Wilkins, A., 2009. Mesenchy-mal stem cell-secreted superoxide dismutase promotes cerebellar neuronalsurvival. J. Neurochem. 114, 1569–1580.

ruman, I.I., Mattson, M.P., 1999. Pivotal role of mitochondrial calcium uptake inneural cell apoptosis and necrosis. J. Neurochem. 72, 529–540.

ipton, S.A., Rosenberg, P.A., 1994. Excitatory amino acids as a final common pathwayfor neurologic disorders. N. Engl. J. Med. 330, 613–622.

im, J.Y., Park, S.I., Oh, J.H., Kim, S.M., Jeong, C.H., Jun, J.A., Lee, K.S., Oh, W., Lee,J.K., Jeun, S.S., 2008. Brain-derived neurotrophic factor stimulates the neuraldifferentiation of human umbilical cord blood-derived mesenchymal stem cellsand survival of differentiated cells through MAPK/ERK and PI3K/Akt-dependentsignaling pathways. J. Neurosci. Res. 86, 2168–2178.

iston, P., Fong, W.G., Korneluk, R.G., 2003. The inhibitors of apoptosis: there is moreto life than Bcl2. Oncogene 22, 8568–8580.

atsuzaki, H., Tamatani, M., Mitsuda, N., Namikawa, K., Kiyama, H., Miyake, S.,Tohyama, M., 1999. Activation of Akt kinase inhibits apoptosis and changesin Bcl-2 and Bax expression induced by nitric oxide in primary hippocampalneurons. J. Neurochem. 73, 2037–2046.

ehman, J., Traktuev, D., Li, J., Merfeld-Clauss, S., Temm-Grove, C.J., Bovenkerk, J.E.,Pell, C.L., Johnstone, B.H., Considine, R.V., March, K.L., 2004. Secretion of angio-

9 (2011) 189–195 195

genic and antiapoptotic factors by human adipose stromal cells. Circulation 109,1292–1298.

Reynolds, I.J., Hastings, T.G., 1995. Glutamate induces the production of reactive oxy-gen species in cultured forebrain neurons following NMDA receptor activation.J. Neurosci. 15, 3318–3327.

Rommel, C., Clarke, B.A., Zimmermann, S., Nunez, L., Rossman, R., Reid, K., Moelling,K., Yancopoulos, G.D., Glass, D.J., 1999. Differentiation stage-specific inhibitionof the Raf-MEK-ERK pathway by Akt. Science 286, 1738–1741.

Shan, K.R., Qi, X.L., Long, Y.G., Nordberg, A., Guan, Z.Z., 2004. Decreased nicotinicreceptors in PC12 cells and rat brains influenced by fluoride toxicity—a mecha-nism relating to a damage at the level in post-transcription of the receptor genes.Toxicology 200, 169–177.

Shin, D.M., Shaffer, D.J., Wang, H., Roopenian, D.C., Morse 3rd., H.C., 2008. NOTCH ispart of the transcriptional network regulating cell growth and survival in mouseplasmacytomas. Cancer Res. 68, 9202–9211.

Singla, D.K., Singla, R.D., McDonald, D.E., 2008. Factors released from embryonic stemcells inhibit apoptosis in H9c2 cells through PI3K/Akt but not ERK pathway. Am.J. Physiol. 295, H907–913.

Tolosa, L., Mir, M., Olmos, G., Llado, J., 2009. Vascular endothelial growth fac-tor protects motoneurons from serum deprivation-induced cell death throughphosphatidylinositol 3-kinase-mediated p38 mitogen-activated protein kinaseinhibition. Neuroscience 158, 1348–1355.

Vaux, D.L., Silke, J., 2005. IAPs, RINGs and ubiquitylation. Nat. Rev. Mol. Cell Biol. 6,287–297.

Wei, X., Zhao, L., Zhong, J., Gu, H., Feng, D., Johnstone, B., March, K., Farlow, M., Du,Y., 2009. Adipose stromal cells-secreted neuroprotective media against neuronalapoptosis. Neurosci. Lett. 462, 76–79.

Wennstrom, S., Downward, J., 1999. Role of phosphoinositide 3-kinase in activationof ras and mitogen-activated protein kinase by epidermal growth factor. Mol.Cell. Biol. 19, 4279–4288.

West, T., Stump, M., Lodygensky, G., Neil, J.J., Deshmukh, M., Holtzman, D.M., 2009.Lack of X-linked inhibitor of apoptosis protein leads to increased apoptosis andtissue loss following neonatal brain injury. ASN Neuro, 1.

White, B.C., Sullivan, J.M., DeGracia, D.J., O’Neil, B.J., Neumar, R.W., Grossman, L.I.,Rafols, J.A., Krause, G.S., 2000. Brain ischemia and reperfusion: molecular mech-anisms of neuronal injury. J. Neurol. Sci. 179, 1–33.

Wilkins, A., Kemp, K., Ginty, M., Hares, K., Mallam, E., Scolding, N., 2009. Human

63–70.Zhao, L., Wei, X., Ma, Z., Feng, D., Tu, P., Johnstone, B.H., March, K.L., Du, Y., 2009.

Adipose stromal cells-conditional medium protected glutamate-induced CGNsneuronal death by BDNF. Neurosci. Lett. 452, 238–240.