B R A I N R E S E A R C H 1 2 3 8 ( 2 0 0 8 ) 1 4 3 – 1 5 3

ava i l ab l e a t www.sc i enced i r ec t . com

www.e l sev i e r. com/ l oca te /b ra in res

Research Report

Motoneuronotrophic factor analog GM6 reduces infarctvolume and behavioral deficits following transient ischemiain the mouse

Jin Yua, Hong Zhua, Dorothy Kob, Mark S. Kindya,c,d,⁎aDepartment of Neurosciences, Medical University of South Carolina, 173 Ashley Avenue, BSB 403, Charleston, SC 29425, USAbGenervon Biopharmaceuticals, Montebello, CA, USAcRalph H. Johnson VA Medical Center, Charleston, SC 29401, USAdNeurological Testing Service, Inc., Charleston, SC 29425, USA

A R T I C L E I N F O

⁎ Corresponding author. Department of Neuro29425, USA. Fax: +1 843 876 5099.

E-mail address: kindyms@musc.edu (M.SAbbreviations: MNTF, motoneuronotrophic

blood pressure

0006-8993/$ – see front matter. Published bydoi:10.1016/j.brainres.2008.08.053

A B S T R A C T

Article history:Accepted 12 August 2008Available online 29 August 2008

Motoneuronotrophic factor (MNTF) is an endogenous neurotrophin that is highly specific forthe human nervous system, and some of the observed effects of MNTF include motoneurondifferentiation, maintenance, survival, and reinnervation of target muscles and organs.MNTF is a neuro-signaling molecule that binds to specific receptors. Using In Silico Analysis,one of the active sites of MNTF was identified as an analog of six amino acids (GM6). Theeffect of chemically synthesized GM6 on ischemic stroke was studied in the middle cerebralartery occlusion (MCAo) mouse model. Mice were subjected to 1 hur of ischemia followed by24 h of reperfusion. Micewere injected intravenouslywith a bolus of GM6, at various doses (1and 5 mg/kg) immediately after the start of reperfusion and examined for changes inphysiological parameters, neurological deficits and infarct volume. GM6 was able topenetrate the blood brain barrier, and at both 1 and 5 mg/kg showed a significant protectionfrom infarct damage, which translated to improvement of neurological deficits.Administration of GM6 demonstrated no changes in HR, BP, pO2, pCO2, or pH. Asignificant increase over the control group in CBF after reperfusion was observed withGM6 administration, which helped to mitigate the ischemic effect caused by the blockage ofblood flow. The time window of treatment was assessed at various times following cerebralischemia with GM6 demonstrating a significant protective effect up to 6–12 h post ischemia.In addition, GM6 increased neurogenesis, and decreased apoptosis and inflammation in themouse brain following cerebral ischemic injury. These data suggest that GM6 isneuroprotective to the brain following IV injection in the mouse model of MCAo.

Published by Elsevier B.V.

Keywords:StrokeTrophic factorMiddle cerebral artery occlusionMouseNeuroprotection

sciences, Medical University of South Carolina, 173 Ashley Avenue, BSB 403, Charleston, SC

. Kindy).factor; MCAo, middle cerebral artery occlusion; CBF, cerebral blood flow; HR, heart rate; BP,

Elsevier B.V.

144 B R A I N R E S E A R C H 1 2 3 8 ( 2 0 0 8 ) 1 4 3 – 1 5 3

1. Introduction

Neuronotrophic factors (NTFs) are a specialized group ofproteins which function to promote the survival, growth,maintenance, and functional capabilities of selected popula-tions of neurons (Barde, 1988, 1989; and Thoenen and Edgar,1985). Studies have demonstrated that neuronal death occursin the nervous systems of vertebrates during certain periods ofgrowth and development (Burek and Oppenheim, 1996;Mennerick and Zorumski, 2000; Sendtner et al., 2000; Bussand Oppenheim, 2004). However, the addition of solubleneuronal trophic factors from associated target tissues servesto mitigate this phenomenon of neuronal death (Chau et al.,1990; Kuno, 1990; Oppenheim, 1989; Krieglstein et al., 2002;Hennigan et al., 2007).

In the vertebrate neuromuscular system, the survival ofembryonic motoneurons has been found to be dependentupon specific trophic substances derived from the asso-ciated developing skeletal muscles. Skeletal muscles havebeen shown, by both in vivo and in vitro studies, to producesubstances which are capable of enhancing the survival anddevelopment of motoneurons by preventing the embryonicmotoneurons from degeneration and subsequent, naturalcellular death (O'Brien and Fischbach, 1986; Hollyday andHamburger, 1976). Similarly, several investigators havereported that chick and rat skeletal muscles possess certaintrophic factors which can prevent the natural cellular deathof embryonic motoneurons both in vivo and in vitro(McManaman et al., 1988; Oppenheim and Haverkamp,1988; Smith et al., 1986). These skeletal muscle derivedneuronotrophic factors have been demonstrated to befunctionally different from other trophic factors such asNerve Growth Factor (NGF), Ciliary Ganglion NeurotrophicFactor (CNTF), Brain-Derived Neurotrophic Factor (BDNF),and Retinal Ganglion Neurotrophic Factor (RGNTF) (Edgar,1985; Levi-Montalcini, 1982; Varon et al., 1988; Barde, 1989;McManaman et al., 1990; Chau et al., 1991a,b).

More recently, the isolation and characterization of twomotoneuronotrophic factors from rat muscle tissue withapparent molecular weights of 35 kDa and 22 kDa have beenreported (Chau et al., 1992). The 35 kDa protein was defined asmotoneuronotrophic factor 1 (MNTF1) and the apparent22 kDa protein as motoneuronotrophic factor 2 (MNTF2).These two trophic factors have been demonstrated in vitro tosupport the growth and/or regeneration of both isolatedanterior horn motoneurons and spinal explants of rat lumberspinal cord. A number of studies have demonstrated theefficacy of these MNTFs in various rat nerve systems,including the peripheral sciatic nerve, the peripheral muscu-locutaneous nerve, the cranial facial nerve, the cranialhypoglossal nerve, and the portion of the spinal cord thatcontrols muscles in the neck, chest and upper limbs (Zhou etal., 1992, 1993; Chau et al., 1992;Wang et al., 1995). In the hemi-sectioned rat spinal cord model, MNTFs reduced inflamma-tion, limited degeneration and enhanced regeneration of thegrafted nerves (Wang et al., 1995). Furthermore, the wobblermice with double recessive genes given one dose of 35 µg/kgMNTF1 at the age of six weeks slowed the neurodegenerativegenetic disease in this strain.

Subsequently, the cloning of human MNTF1 and itsassociated receptor from a human retinoblastoma cDNAlibrary was reported (Chau et al., 1993). Human MNTF1cDNAs were subcloned into expression vectors and theMNTF1 polypeptides contained in the expressed fusionproteins exhibited biological activity similar to that of the“native” MNTF1 protein in that they supported the in vitrogrowth of rat anterior horn motoneurons. The amino acidsequences of the human MNTF1 polypeptides were eluci-dated by direct protein sequencing. One of the MNTF1polypeptides consisting of 33 amino acids was identified asMNTF33mer, and subsequently was successfully synthesizedby solid phase chemistry. The synthesized MNTF33mershowed biological activity in various in vitro and in vivofunctional assays. A number of studies have demonstratedthe trophic and tropic efficacy of the synthesizedMNTF33mer in well-established rat peripheral nerve modelsystems. In a rat sciatic nerve transection with an 8-mm gapstudy, MNTF33mer treated animals have significantimprovement of motoneuron regeneration in a doseresponse manner and promoted DRG neurons regeneration.In a transected femoral nerve rat model, the number ofmotoneurons projected correctly to muscle was enhanced inthe MNTF33mer treated animals in a dose dependentmanner. At the optimal dose, the number of motoneuronsprojected correctly to muscle was three times that of themotoneurons projected incorrectly to the skin.

In Silico Analysis was employed to search further for theactive sites within the MNTF33mer molecule, and six activedomain sites were identified. The smallest active site consistsof 6 amino acids and hence was named MNTF6mer or GM6(Chau, 2001). Recent studies have shown that GM6 had similaractivity as the parentmolecule (Chau, 2005, 2007). Studieswiththe synthesized GM6 also demonstrated similar trophic effectsin a transected femoral nerve rat model. In a zebrafishbioassay, GM6 protected the organism from L-2-hydroxyglu-taric acid (LGA) induced oxidative stress and apoptosis in theCNS, and reduced apoptosis by 85% in the midbrain. Thesestudies suggest that MNTF33mer and its smaller analog GM6provide protection from cell injury in a number of diseasemodels.

Human MNTF1 is an endogenous neurotrophin, is highlyspecific for the human nervous system and it is expressedrapidly during the first trimester of human fetal develop-ment of the complete nervous system, peaking at weeknine (Di and Huang, 1998). MNTF is a neuro-signalingmolecule that binds on very specific receptors. The specificfunctions of MNTF, as demonstrated in animal and in vitrostudies, include embryonic stem cell differentiation intomotoneurons, motoneuron maintenance and survival,motor axon regeneration with guidance, and reinnervationof target muscles and organs (Chau et al., 1992). When thecentral nervous system (CNS) and peripheral nervoussystem (PNS) are under attack caused by diseases, disordersor injuries, MNTF creates a protective and permissiveenvironment for nerve regeneration and repair that areneuroprotective, anti-apoptosis, anti-oxidation, anti-inflam-mation, and anti-scar.

Based on these data, we decided to test the ability of GM6 toprotect the brain from acute ischemia and reperfusion injury.

Fig. 1 – Levels of GM6 in brain of mice. C57BL/6 mice wereinjected with vehicle control (phosphate buffered saline),GM6 at 0.2 mg or GM6 at 2 mg/kg. The compounds wereadministered at time 0 and the brains were collected 4 h laterfor analysis by ELISA. *p<0.0001 compared to Saline;**p<0.0001 for 2000 µg/kg compared to 200 µg/kg.

Fig. 2 – Effects of GM6 on infarct volumes in themouse followingischemia followed by 24 h of reperfusion. (A) Animals were injecintravenously at the end of the ischemic period. Animals were svolume. p<0.0004 for GM6 (1 mg/kg) compared to control; p<0.00pictures of brains frommice subject to 1 h ischemia and 24 h repeor vehicle (right panel) at the end of ischemia.

145B R A I N R E S E A R C H 1 2 3 8 ( 2 0 0 8 ) 1 4 3 – 1 5 3

2. Results

2.1. Blood brain barrier penetration of GM6

To measure GM6 levels in brain samples following intrave-nous (i.v.) administration, we used a competitive enzymelinked immunosorbent assay (ELISA). As shown in Fig. 1,GM6 was detected in the cortex using the ELISA. The assaydetected endogenous GM6 in the brain (0.4 µM) of the controlanimals. In the animals injected with 0.2 mg/kg of GM6 a400% increase in GM6 was detected (1.760 µM) whereasinjection of 2 mg/kg of GM6 gave rise to a 3000% increase inGM6 in the brain after 4 h (12.92 µM). These data suggestthat intravenous injection of GM6 allowed for distribution ofthe peptide in the brain.

2.2. Dose effect of GM6 on ischemic injury

To determine the effect of GM6 on protection from ischemicinjury, mice were subjected to 1 h of ischemia followed by 24 h

transient ischemia. All mice were subjected to 1 h of cerebralted with vehicle (control) or GM6 at 1 mg/kg or 5 mg/kgacrificed on day 2 and processed to determine the infarct01 for GM6 (5 mg/kg) compared to control. (B) Representativerfusion. Animals were injectedwith GM6 (5mg/kg) (left panel)

Fig. 3 – Neurological deficits in mice subjected to GM6treatment. All micewere subjected to 1 h of cerebral ischemiafollowed by 24 h of reperfusion. Animals were injected withvehicle (control) or GM6 at 1 mg/kg or 5 mg/kg intravenouslyat the end of the ischemic period. Neurological deficits weremeasured at the end of reperfusion (22 h) as outlined inthe Experimental procedures. *p<0.04 for GM6 (1 mg/kg)compared to control; **p<0.0006 for GM6 (5 mg/kg) comparedto control.

Fig. 4 – Effects of GM6 on infarct volumes with varioustreatment times. Mice were subjected to 1 h of cerebralischemia followed by 14 days of reperfusion. Animals wereinjected with vehicle (control) or GM6 at 5 mg/kgintravenously at various times after the end of the ischemicperiod. Animals were sacrificed on day 14 and processed todetermine the infarct volume.*p<0.0001 for all groupscompared to control except for t=24 h, p<0.0821.

146 B R A I N R E S E A R C H 1 2 3 8 ( 2 0 0 8 ) 1 4 3 – 1 5 3

of reperfusion (Fig. 2). GM6 was administered i.v. at 1 or5 mg/kg at the start of reperfusion, to ensure maximumprotection. Compared with the vehicle-injected group, theinfarct volume in the brains was significantly decreased in theGM6 treated groups (at both 1 and 5 mg/kg). GM6 showed adose dependent reduction in lesion area. Infarct volumes vs.GM6 dosage was plotted in Fig. 2A. The lesion areas were73.37 mm3 for vehicle group, 45.93 mm3 and 20.29 mm3 forGM6 treated groups at 1 or 5 mg/kg, respectively. Thus, postischemia i.v. administration of GM6 at 1 or 5 mg/kg resulted in38% and 73% decrease in infarct volume respectively com-pared to vehicle. Fig. 2B shows the damaged area present inanimals subjected to either vehicle or 5 mg/kg of GM6.

To determine the effect of GM6 on behavioral outcomes inmice following ischemic injury, animals were examined at22 h for neurological deficits (Fig. 3). Animalswere assessed forneurological deficits based on a scale of 0 to 4. Animals treatedwith GM6 showed a dose dependent decrease in neurologicaldeficits (Vehicle, 2.7±0.300; GM6 at 1 mg/kg, 1.8±0.249; GM6 at5 mg/kg, 1.3±0.153). There were no significant differences inphysiological parameters (mean arterial pressure, blood pO2,pCO2, and pH) between the vehicle and treated mice atbaseline, during ischemia, or after reperfusion. A significantincrease over vehicle group in cerebral blood flow after

Table 1 – Cerebral blood flow after ischemic injury in GM6treated mice

Compound(group)

CBF afterreperfusion

p value compareto vehicle

Vehicle 84.9±1.567GM6 (1) 89.6±1.893 p<0.07GM6 (5) 91.2±0.987 p<0.003

reperfusion was observed in the group treated with 5 mg/kgof GM6 (p<0.003) (Table 1).

2.3. Time course of GM6 on ischemic injury

To determine the time course of protection afforded by GM6 inthe mouse model of ischemia and reperfusion injury, 5 mg/kgof GM6 was selected because of the effect seen in the dosingstudies. For these studies, mice were subjected to 1 h ofischemia and 14 days of reperfusion for assessment. Animals

Fig. 5 – Neurological deficits in mice subjected to time courseof GM6 treatment. All mice were subjected to 1 h of cerebralischemia followed by 14 days of reperfusion. Animalswere injected with vehicle (control) or GM6 at 5 mg/kgintravenously at various times after the end of the ischemicperiod. Neurological deficits were measured at the end ofreperfusion as outlined in the Experimental procedures.*p<0.0001 for all groups compared to control except fort=12 h (**p<0.0002) and t=24 h (p<0.1387).

Table 2 – Cerebral blood flow after ischemic injury in micetreatedwith GM6 at various times after the ischemic period

Compound(group)

CBF afterreperfusion

p value compareto vehicle

Vehicle 82.5GM6 (0 h) 92.5 p<0.0001GM6 (3 h) 87.3 p<0.044

147B R A I N R E S E A R C H 1 2 3 8 ( 2 0 0 8 ) 1 4 3 – 1 5 3

were treatedwith GM6 at 0, 3, 6, 12 and 24 h after the end of theischemic period and examined for infarct volumes andbehavioral deficits (Figs. 4 and 5). Post ischemia i.v. adminis-tration of GM6 reduced the infarct volume in the animal brains(Fig. 4). GM6 administered initially at 0, 3, 6 and 12 h afterischemia showed 70, 68, 58 and 36% reduction in lesion area,respectively. The infarct volume of the animals treated at 24 hafter ischemia, although decreased, did not show a significantdifference from the vehicle treated animals.

Animals were assessed for neurological deficits based ona scale of 0 to 4. The neurological deficits for vehicle, ortreated with GM6 at 0, 3, 6, 12 and 24 h after ischemia were3.6, 1.2, 1.67, 1.8, 2.5 and 3.2 respectively. Animals treatedwith GM6 showed a treatment time dependent decrease inneurological deficits (Fig. 5), with p<0.0001 for all treatedgroups compared to control except for t=12 h (p<0.0002) andt=24 h (p<0.1387). There were no significant differences in

Fig. 6 – Effect of GM6 on neurogenesis. Micewere subjected to 1 hinjected with BrdU on day 7 for 3 days and examined for neurogimmunostained with antibodies directed against BrdU. (B) Tissuagainst NeuN. (C) Quantitative analysis of neurogenesis in ische

physiological parameters (mean arterial pressure, blood pO2,pCO2, and pH) between the vehicle and treated mice atbaseline, during ischemia, or after reperfusion. However,GM6-mediated increase in cerebral blood flow after reperfu-sion was observed in the groups initially treated with 5 mg/kg of GM6 at 0 h and 3 h after ischemia compared to control(Table 2).

2.4. Effect of GM6 on neurogenesis, apoptosisand inflammation

In an attempt to determine the mechanism of action ofGM6 in the mouse brain following ischemic injury, animalswere examined for neurogenesis, apoptosis and inflamma-tory markers. To study the effect of tGM6 on cell prolifera-tion and neurogenesis, we identified proliferating cells bylabeling with bromodeoxyuridine (BrdU). BrdU cells weredetected in the hippocampus and dentate gyrus as well asin the cortex (Fig. 6A). Double labeling experiments withBrdU and doublecortin (DCX) indicated that the cells wereneuronal in nature and mature neurons by staining withthe neuron specific marker NeuN (Fig. 6B and data notshown). Fig. 6C shows the quantitative assessment of theneurogenesis induced by GM6. The effect of GM 6 on celldeath and inflammation were validated by measurement ofapoptosis (TUNEL assay) and glial fibrillary acidic protein

of ischemia followed by 14 days of reperfusion. Animalswereenesis on day 14 of reperfusion. (A) Tissue sections weree sections were immunostained with antibodies directedmic brain. *p<0.0001 for all groups compared to controls.

Fig. 7 – Effect of GM6 on apoptosis and inflammation. Mice were subjected to 1 h of ischemia followed by 24 h of reperfusion.(A) Animals were examined at 24 h of reperfusion for apoptosis using a TUNEL assay. (B) Animals were examined at 24 hfor the astroglial inflammatory marker GFAP. (C) Quantitative analysis of apoptosis in the ischemic brain. (D) Quantitativeanalysis of GFAP staining in the ischemic brain. *p<0.0001 for all groups compared to controls.

148 B R A I N R E S E A R C H 1 2 3 8 ( 2 0 0 8 ) 1 4 3 – 1 5 3

(GFAP) levels (Fig. 7). Both apoptosis and GFAP levels wereattenuated in the GM6 treated animals in a dose dependentfashion (Figs. 7A and B). Quantitative analysis demonstrateda significant decrease in both following ischemia/reperfu-sion injury (Figs. 7C and D).

3. Discussion

Stroke is the third most common cause of death and the maincause of disability in the United States (American HeartAssociation, 2006). The outcome and infarction size afterfocal cerebral ischemia is determined by both “necrotic” celldeath and by delayed neuronal cell loss in the borderzone ofischemia (programmed cell death or apoptosis) (Ekshyyan andAw, 2004). Recent therapies have emerged to treat ischemicstroke. However, these treatments mostly dealt with dissol-ving the blood clot but did not address neuroprotection,reduction of behavioral deficit or brain infarct volume oncethe neuronal cell death cycle has been triggered (Grupper etal., 2007). Past and current neuroprotective strategies havebeen successful in animal models but have failed significantlyin clinical trials (Young et al., 2007; Arumugam et al., 2008).Understanding the basic mechanisms that influence cell losswill help in the design of drugs and applications to reduce celldeath associated with ischemic injury.

MNTF is a trophic factor that may provide protection fromneurological diseases and allow for regeneration of neuronaltissue following injury (Zhou et al., 1993, 1997). Recent studieshave suggested that MNTFmay be essential in the differentia-tion of embryonic stem (ES) cells into motor neurons (Zhouet al., 1994; Di et al., 1997; Di and Huang, 1998). Because of theexpression pattern of MNTF during injury, its importance inneuronal protection and in differentiation of ES cells, wedecided to test the efficacy of GM6, a six amino acid analog ofMNTF, in an animalmodel of stroke.We showed that systemicadministration of GM6 was capable of penetrating the bloodbrain barrier with reasonable efficiency. Subsequent studiesperformed here demonstrated the ability of GM6 to protect thebrain from the detrimental effects of cerebral ischemia andreperfusion injury in an effective and efficient way. Intrave-nous administration of GM6 at 1 and 5mg/kg single bolus dosedemonstrated a dose dependent protective effect in the brainagainst ischemia/reperfusion injury by a decrease in infarctvolume, improved behavioral attributes, and an increase incerebral blood flow. In addition, we demonstrated that GM6had a broad window of opportunity in the protection fromischemic injury. This may be due to several possibilities. First,GM6 showed an increase in cerebral blood flow followingreperfusion suggesting itmight be able to enhance reperfusionof the brain. This has been suggested in the endothelial nitricoxide synthase (eNOS) deficientmice that vascular nitric oxide

149B R A I N R E S E A R C H 1 2 3 8 ( 2 0 0 8 ) 1 4 3 – 1 5 3

(NO) productionmay be important for preservation of vascularfunction (Huang et al., 1996). Second, the effect of variousMNTF analogs on ES cells suggests that MNTF may enhanceneurogenesis in the brain following ischemic injury (Wang etal., 2007). Further studies are required to determine themechanisms associated with GM6 and/or MNTF action inischemic/reperfusion injury.

The rate of increase in cell proliferation that we find in theGM6 treated animals was greater than that seen in othermodels of neurogenesis (Sivilia et al., 2008). This suggests thatGM6 helps not only to initiate but promote neural stem cells toproliferate and differentiate into neuronal cells. The increaseis significantly greater than the vehicle treated animals. Assuggested previously, the role of astrocytes in the inhibition ofneurogenesis may be important and the reduction in glialresponse by GM6 may help to provide the niche necessary fornew neuronal growth (Wachs et al., 2006). The underlyingmolecular and cellular mechanisms are still unknown how-ever suggest that the neurotrophic factor basis to GM6 andMNTF may be involved in the potent proliferative anddifferentiation activity (Chau, 2007). One concern is thatBrdU stains for a multitude of cell proliferation and may notnecessarily indicate neurogenesis (Liu et al., 1998). However, itdoes indicate an increase in proliferation in the brainfollowing injury. In addition, the increase in NeuN stainingin the presence of GM6 suggests that at least some of the cellsdifferentiate into neuronal like cells. The unique capabilitiesof GM6 to differentiate neural stem cells may provide a novelmechanism for treating ischemia and reperfusion injury. Notonly can GM6 protect against neuronal injury but theinduction of differentiation helps to stimulate recovery andrepair to a greater extent.

Here we show that animals infused with GM6 demon-strated a reduction in apoptosis and inflammation. We foundthat in ischemia and reperfusion injury, apoptotic cells andintercellular apoptotic fragments were detectable at theborder of the lesion in both vehicle and GM6 treated mice,and that GM6 reduced the amount of apoptotic cells. Cerebralischemia results in necrotic cell death in the core of the infarct,and this is surrounded by apoptotic cells in the penumbraregion (Rami et al., 2008). GM6 appeared to limit the extent ofapoptotic cell death indicating that it provided neuroprotec-tion. The preclusion of cell death paralleled the behavioralchanges. Cerebral ischemia promotes apoptosis of neurons,oligodendrocytes, and astrocytes (Petito et al., 1998). Cellapoptosis begins as early as 4 h after the trauma and peaksat 1–2 days post-injury but can continue to develop over time(Borsello and Forloni, 2007). Cytokine and inflammatorymolecules have been shown to increase dramatically afterischemia, suggesting that they are involved in the apoptoticprocess (Rodrίguez-Yáñez and Castillo, 2008). Studies haveshown that cytokines (TNF-α, IL-1, IL-2, etc) and inflammatorymediators (reactive oxygen species, ROS) can contribute to theinfarct volume (Rothwell and Luheshi, 1996; Wong and Crack,2008). In addition, attenuation of inflammation can restrict thedevelopment of the lesion (Khan et al., 2004). We have shownthat GFAP is dramatically elevated following ischemia andreperfusion injury (Kindy et al., 1992). In the acute period afterischemia, cytokines can also induce the expression ofendothelial leukocyte adhesionmolecules, which can activate

neutrophils adhering to endothelial cells and damage theendothelium by releasing inflammatory mediators such asneutrophil elastase and reactive oxygen species (ROS)(Ducruet et al., 2008). The damage of endothelial cells leadsto further ischemia by increasing the vascular permeabilityand decreasing the tissue blood flow (Kahles et al., 2007). Ourdata suggest that GM6 reduced the levels of apoptosis andinflammation in a dose dependent fashion and resulted in areduction in lesion volume and improved physical outcome.

Our data establishes MNTF as a major factor in thepathogenesis of stroke. Alterations in MNTF expression oractivitymay affect the development and progression of stroke,and therapies to enhance MNTF expression or to provideMNTF following ischemic injurymay attenuate the expressionof the disease. These studies lay the groundwork for futurestudies to determine the beneficial effects of GM6 in strokeand other neurodegenerative disorders.

4. Experimental procedures

4.1. Animals

C57BL/6 mice (Jackson Laboratory, Bar Harbor, ME), weighing22–25 g each were given free access to food and water beforethe experiment.

4.2. Experimental groups

4.2.1. Blood brain barrier penetrationAnimals were injected with a bolus i.v. via tail vein withvehicle or GM6 at 0.2 or 2 mg/kg. After 4 h, animals wereperfused with cold PBS and then the cortex was examined forGM6 in the brain via ELISA. Four hours was chosen asmaximalaccumulated dose of GM6 in preliminary studies (data notshown).

4.2.2. Dose effect of GM6Animals were subjected to 1.0 h ischemia followed by 24 hreperfusion. Animals were randomly assigned to a vehiclegroup (n=10) or groups (n=10) treated with an intravenousinjection of a 6mer active component of MNTF (GM6) at a doseof 1 or 5 mg/kg. Formulation of GM6 (CS Bio Co., Menlo Park,CA, Catalog# CS1507, GMP013, lot C811) was performed byreconstituting GM6 with normal saline solution that wasstored at 4 °C. Vehicle control received phosphate bufferedsaline (PBS) solution. The bolus IV injections via tail vein weregiven immediately after the onset of reperfusion or at varioustimes after reperfusion.

4.2.3. Time course of GM6 effectAnimals were subjected to 1.0 h ischemia followed by 14 daysreperfusion. Animals were randomly assigned to a vehiclegroup (n=10) or one of the five treatment groups (n=10 foreach group) that would receive an initial intravenous injectionof GM6 at a dose of 5 mg/kg at 0, 3, 6, 12 or 24 h after ischemia.Vehicle control received saline solution. The bolus i.v. injec-tion of GM6 via tail vein was initiated in the differenttreatment groups at 0, 3, 6, 12, or 24 h after the start ofreperfusion and subsequently one additional dose of GM6was

150 B R A I N R E S E A R C H 1 2 3 8 ( 2 0 0 8 ) 1 4 3 – 1 5 3

i.v. injected every day for the next 3 days in all treatmentgroups. The investigators were blinded to the treatmentgroups.

4.3. Induction of ischemia

The animals were anesthetized with halothane (1% in 70%/30% NO2/O2 by mask). Monitoring of mean arterial bloodpressure (MABP) via tail cuff apparatus, and blood sampleswere collected to determine arterial pH levels and PaCO2 andPaO2. The MABP and heart rate were recorded using a VisitechSystem blood pressure monitor. Brain temperature wasmonitored using a rectal thermometer and thermistor probeinserted into the temporalis muscle. The animals' bodytemperature was maintained at 37 °C by using a water-jacketed heating pad. Brain temperaturewasmonitored for 1 hprior to ischemia to 6 h following ischemia and was recordedat 30-minute intervals. Each mouse was anesthetized and theexternal carotid artery (ECA) and common carotid artery (CCA)were isolated (Gary et al., 1998; Mattson et al., 2000; Dubal etal., 2001; Ellsworth et al., 2003, 2004). The left common carotidartery (CCA) was exposed through a midline incision in theneck. The superior thyroid and occipital arteries were electro-coagulated and divided. A microsurgical clip was placedaround the origin of the external carotid artery (ECA). Thedistal end of the ECA was ligated with 6-0 silk and transected.A 6-0 silk was tied loosely around the ECA stump. The clip wasremoved and the fire-polished tip of a 5-0 nylon suture(silicone coated) was gently inserted into the ECA stump.The loop of the 6-0 silk was tightened around the stump andthe nylon suture was advanced approximately 13 mm(adjusted for body weight) into and through the internalcarotid artery (ICA) until it rested in the anterior cerebralartery (ACA), thereby occluding the anterior communicatingand middle cerebral arteries. After the nylon suture was inplace for 1 h, it was pulled back into the ECA and the incisionclosed.

4.4. Histological examination

For histological examination, the animals were anesthetizedwith an intraperitoneal injection of sodium pentobarbital(50 mg/kg) 24 h after ischemia was induced. The brains weretranscardially perfused with 4 °C, 10% phosphate bufferedsaline (PBS). The brainswere removed and chilled for 15min at−20 °C before being placed in a Rodent Brain Matrix. Coronalsections (1-mm thickness) were prepared and subjected to 2%triphenyltetrazolium chloride (TTC) staining at 37 °C (Mattsonet al., 2000). TTC stains live tissue (red) versus dead or dyingtissue (white). Seven serial one-mm thick coronal sectionsthrough the rostral to caudal extent of the infarction wereobtained from each brain, beginning two-mm from the frontalpole. The TTC stained sections were placed in 10% neutralbuffered formalin and kept in darkness at 4 °C for at least 24 h.The infarct area in each section was determined with acomputer-assisted image analysis system, consisting of aPower Macintosh computer equipped with a Quick Captureframe grabber card, Hitachi CCD camera mounted on anOlympus microscope and camera stand. NIH Image AnalysisSoftware, v. 1.55 was used. The images were captured and the

total area of damage determined over the seven sections. Asingle operator blinded to treatment status performed allmeasurements. The infarct volume was calculated by sum-ming the infarct volumes of the sections. Infarct size (%) wascalculated by using the following formula: (contralateralvolume − ipsilateral undamaged volume)×100/contralateralvolume to eliminate effects of oedema.

4.5. ELISA analysis

GM6 levels in samples were measured using the competitiveELISA kit. Coating buffer containing affinity purified rabbitanti-6 Mer at 10 µg/ml was coated on ELISA plate. 6 Mer-biotinat 1 µM (final dilution) was used in the assay. The knownconcentrations of GM6 (competitor) were used as referencestandards in establishing the standard curve starting from80 µM and serially diluted 2-fold down to 0.625 µM. Concen-tration of tested samples was estimated from their OD450observation based on the standard curve. The brain tissue wasprepared in cell lysis buffer containing 100 mM Tris/HCl, pH7.0, containing 2% BSA, 1 MNaCl, 4mM EDTA, 2% Triton X-100,0.1% sodiumazide and protease inhibitors (CompleteTM,Mini,Boehringer Mannheim). Homogenates were prepared in 10volumes of buffer to tissuewetweight. The homogenateswerecentrifuged for 30 min at 14,000 ×g. The resulting supernatantwas used for ELISA analysis with the appropriate volumeadjustment. For the assay, the anti-6 Mer pAb was diluted to10 µg/ml in ELISA coating buffer. The 96-well micro-titer plateswere coated with 100 µl/well of the diluted pAb. The plateswere covered and kept refrigerated overnight. Next day, theplates were blocked with 200 µl/well TBS with 3% BSA, andkept at RT for 60 min. The plates were washed 3× with TBST(TBS+0.05% Tween 20). For each plate, lanes 1 and 2 were usedfor the standard curve. One hundredmicroliters of 1 µM 6Mer-biotin was added to all wells except A1 and A2. 200 µl/well ofthe standard mix of 1 µM 6 Mer-biotin with 80 µM GM6 wasadded to A1 and A2. A serial 2-fold dilution from wells A1 andA2 down to wells H1 and H2 was performed by taking out100 µl sample from each well and mix it with next well—pipetting up and down at least 8 times. Test samples (brainhomogenate) were diluted in Ab dilution buffer and mixed at1:50 with the 1 µM 6 Mer-biotin in the wells. The plate wasmixed on a shaker at 400 rpm for 2 h at RT. The samples werewashed 6× with TBST. Streptavidin-HRP (10 ml per plate)solution was prepared by diluting Streptavidin-HRP to 1:2000in Ab dilution buffer. The plate was mixed on the shaker foranother 1 h at RT. The samples were washed 8× with TBST.50 µl/well substrate (TMBS, Genetel) was added and developedfor 5 min. The reaction was stopped with 50 µl 1 M HCl andOD450 was measured immediately.

4.6. Measurement of cerebral blood flow

Cerebral blood flow (CBF) was monitored by using a laserDoppler flowmeter (). The CBF values were determined as apercentage, because the values displayed by the laser Dopplerflowmeter were not absolute. As described above, the animalswere anesthetized with halothane (1% in 70%/30% NO2/O2 bymask) and were mounted in a stereotaxic frame and theprobes were fixed to the head. In the hemisphere ipsilateral to

151B R A I N R E S E A R C H 1 2 3 8 ( 2 0 0 8 ) 1 4 3 – 1 5 3

the MCA occlusion, coordinates were as follows: point A,0.5 mm posterior to the bregma and 2 mm lateral to themidline; point B, 1 mm posterior to the bregma and 1.2 mmlateral to the midline; point D, 1 mm anterior to the bregmaand 1.7 mm lateral to the midline; and point C in thecontralateral hemisphere, 1 mm posterior to the bregma and2 mm from the midline. CBF was compared at 15 min prior tothe onset of ischemia, during ischemia (15 min after the startof ischemia) before injection of test articles and at 30 min postinjection (continuous measurements were taken from 15 minprior to ischemia to 30 min after the end of injection of thecompound). Animals were re-anesthetized and the CBF wasmeasured at 3 h following reperfusion. The mean valuesbefore MCA occlusion were taken as baseline and the datathereafter were expressed as percentages of the baselinevalue.

4.7. Behavioral assessment

Behavioral analysis (neurological deficit) was determined inthe mice before and after ischemic injury (Dubal et al., 2001).Neurological scores were as follows: 0, normal motorfunction; 1, flexion of torso and contralateral forelimbwhen animal was lifted by the tail; 2, circling to thecontralateral side when held by tail on flat surface, butnormal posture at rest; 3, leaning to the contralateral side atrest; 4, no spontaneous motor activity.

4.8. Analysis of neurogenesis, apoptosisand inflammation

BrdU administration was chosen to identify the lineage ofnewly generated cells as described previously (Kempermannet al., 2003). Briefly, BrdU was administered intraperitoneallytwice daily (75mg/kg, Sigma, St Louis, MO, USA) for 3 days, andthe rats were perfused 7 days after the last BrdU injection.Animals were then sacrificed and examined by immunohis-tochemistry. Sections were first incubated in 0.1 M PBS atroom temperature for 30 min, followed by incubation at 37 °Cfor 30 min with HCl 2 N. After rinsing in PBS for 20 min, thesections were incubated at 37 °C for 4 min with pepsin 0.025%.Sections were then processed for BrdU visualization by usingrat anti-BrdU diluted in PBS/Triton 0.3%. Cryosections werewashed 3 times (5 min/wash) with tris-buffered saline (TBS,pH:7.4) buffer, followed by washing one time with 0.1% TritonX-100–TBS buffer for 5 min. Sections were then incubated in3% H2O2–TBS buffer for 30 min at room temperature toeliminate endogenous peroxidase activity. Following 1 hblocking with 5.0% serum (goat), the sections were incubatedovernight with primary antibodies: GFAP positive astrocytes(2E1; BD Biosciences, San Jose, CA, 1:200 dilution); DCX, (DCX,Santa Cruz Biotechnology, Santa Cruz, CA, 1:200 dilution);NeuN, (NeuN, Chemicon, Temecula, CA, 1:250 dilution). Thenext day, sections were washed three times (5min/wash) with0.1% Triton X-100/TBS buffer to remove excess primaryantibody. Thereafter, primary antibody was detected usingHRP-conjugated rabbit VECTASTAIN ABC kit and DAB/sub-strate reagents (Burlingame, CA) according to the manufac-turer's instructions. At 24 h ischemia, the terminaldeoxynucleotidyl transferase-mediated dUTP-digoxygenin

nick-end labeling (TUNEL) technique was used to detectapoptotic cells (ApopTag® Peroxidase In Situ Apoptosis Detec-tion Kit, Chemicon International, Inc., Temecula, CA, USA).The preparation of tissue cryosections was the same asdescribed above. TUNEL assay was performed according tothe manufacturer's instruction. In brief, tissue sections werefixed in 1% paraformaldehyde for 10min at room temperature,and then post-fixed in pre-cooled ethanol: acetic acid (2:1) for5 min at −20 °C. Endogenous peroxidase was quenched in 3%H2O2 for 5 min at room temperature. Subsequently, sectionswere incubated with equilibration buffer for at least 10 s atroom temperature, followed by incubation with terminaldeoxynucleotidyl transferase in a humidified chamber at37 °C for 1 h. The reaction was stopped by adding in stop/wash buffer for 10 min at room temperature. Sections werethen incubated with anti-digoxigenin peroxidase conjugate ina humidified chamber for 30 min at room temperature and,developed with 3, 3′-diaminobenzidine (DAB) for 5 min atroom temperature. Sections were counterstained with 0.5%methyl green and cover-slipped. Sections from 5 to 8 separateanimals were used to assess the number of TUNEL-positivecells.

4.9. Statistical analysis

The results were expressed as the mean±standard deviation(SD). The statistical significance of the results in the GM6level in brain by ELISA, infarct volume, neurological deficit,physiological and histological data were analyzed using aone-way analysis of variance (ANOVA) followed by Fisher'spost hoc test. Repeated-measures ANOVA were computedon the monitoring data and the significance of thedifference among groups were evaluated by Fisher's posthoc test.

Acknowledgments

The authors wish to acknowledge Genervon Biopharmaceu-ticals for providing the MNTF 6 mer for the studies. Wethank Dr. Sebastiano Gattoni-Celli and Dr. Pui-Chu Yu forreviewing the manuscript prior to submission and EileenMcFadden for assistance with preparation of the manu-script. Dorothy Ko hold equity in Genervon Biopharmaceu-ticals and Dr. Kindy holds equity in Neurological TestingService, Inc. These studies were funded by GenervonBiopharmaceuticals, Inc.

R E F E R E N C E S

American Heart Association, Heart Disease and StrokeStatistics — 2006 Update, http://www.americanheart.org/downloadable/heart/1136308648540Statupdate2006.pdf.

Arumugam, T.V., Selvaraj, P.K., Woodruff, T.M., Mattson, M.P.,2008. Targeting ischemic brain injury with intravenousimmunoglobulin. Expert. Opin. Ther. Targets 12, 19–29.

Barde, Y.A., 1988. What, if anything, is a neurotrophic factor? TINS11, 343–346.

Barde, Y.A., 1989. Trophic factors and neuronal survival. Neuron 2,1525–1534.

152 B R A I N R E S E A R C H 1 2 3 8 ( 2 0 0 8 ) 1 4 3 – 1 5 3

Borsello, T., Forloni, G., 2007. JNK signalling: a possible target toprevent neurodegeneration. Curr. Pharm. Des. 13, 1875–1886.

Burek, M.J., Oppenheim, R.W., 1996. Programmed cell death in thedeveloping nervous system. Brain Pathol. 6, 427–446.

Buss, R.R., Oppenheim, R.W., 2004. Role of programmed cell deathin normal neuronal development and function. Anat. Sci. Int.79, 191–197.

Chau, R.M.W., 2001. Polynucleotides encodingmotoneuronotrophic factors. US Patent #6,309,877.

Chau, R.M.W., 2005. Methods and use of motoneuronotrophicfactors. US Patent #6,841,531.

Chau, R.M.W., 2007. MNTF peptides and compositions andmethods of use. US Patent #7,183,373.

Chau, R.M.W., Wu, X.Y., Zhao, L.P., Ren, F., Jen, J.S., 1990.Neuronotrophic factor. Chin. J. Neuroanat. 6, 129–138.

Chau, R.M.W., Zhao, L.P., Jen, L.S., 1991a. The effect of a 30 kDaprotein from tectal extract on cultured retinal ganglion cell.Scient. Sin. B1, 50–53.

Chau, R.M.W., Zhao, L.P., Jen, L.S., 1991b. The effect of a 30 kDaprotein from tectal extract on cultured retinal neuron. Sci.China 34, 33–39.

Chau, R.M.W., Ren, F., Huang, W., Jen, L.S., 1992. Muscleneurotrophic factors specific for anterior horn motoneurons ofrat spinal cord. Recent Adv. Cell Mol. Biol. 5, 89–94.

Chau, R.M.W., Ren, F., Liang, Z.H., 1993. Cloning of genes formuscle-derived motoneuronotrophic factor 1 (MNTF1) and itsreceptor bymonoclonal antibody probes. Soc. Neurosci. 19, 252,#109.23.

Di, X., Huang, W., 1998. Localization and morphometric studyon motoneuronotrophic factor 1 and its receptor indeveloping chorionic villi of human placenta. Acta Anat. Sin.29, 86–89.

Di, X., Huang, W.Q., Sun, L., 1997. Immunohistochemicallocalization of c-fos p53 protein and MNTF1 receptor in earlyhuman placental villi. Acta Anat. Sin. 28, 404–406.

Dubal, D.B., Zhu, H., Yu, J., Rau, S.W., Shughrue, P.J., Merchenthaler,I., Kindy, M.S., Wise, P.M., 2001. Estrogen receptor alpha, notbeta, is a critical link in estradiol-mediated protectionagainst brain injury. Proc. Natl. Acad. Sci. U. S. A. 98,1952–1957.

Ducruet, A.F., Hassid, B.G., Mack, W.J., Sosunov, S.A., Otten, M.L.,Fusco, D.J., Hickman, Z.L., Kim, G.H., Komotar, R.J., Mocco, J.,Connolly, E.S., 2008. C3a receptor modulation of granulocyteinfiltration after murine focal cerebral ischemia isreperfusion dependent. J. Cereb. Blood Flow Metab. 28,1048–1058.

Edgar, D., 1985. Nerve growth factors and molecules of theextracellular matrix in neuronal development. J. Cell Sci.(Suppl. 3), 107–113.

Ekshyyan, O., Aw, T.Y., 2004. Apoptosis in acute and chronicneurological disorders. Front Biosci. 9, 1567–1576.

Ellsworth, J.L., Garcia, R., Yu, J., Kindy, M.S., 2003. Fibroblast growthfactor-18 reduced infarct volumes and behavioral deficits aftertransient occlusion of the middle cerebral artery in rats. Stroke34, 1507–1512.

Ellsworth, J.L., Garcia, R., Yu, J., Kindy, M.S., 2004. Time window offibroblast growth factor-18-mediated neuroprotection afterocclusion of the middle cerebral artery in rats. J. Cereb. BloodFlow Metab. 24, 114–123.

Gary, D.S., Bruce-Keller, A.J., Kindy, M.S., Mattson, M.P., 1998.Ischemic and excitotoxic brain injury is enhanced in micelacking the p55 tumor necrosis factor receptor. J. Cereb. BloodFlow Metab. 18, 1283–1287.

Grupper, M., Eran, A., Shifrin, A., 2007. Ischemic stroke, aorticdissection, and thrombolytic therapy—the importance of basicclinical skills. J. Gen. Intern. Med. 22, 1370–1372.

Hennigan, A., O'Callaghan, R.M., Kelly, A.M., 2007. Neurotrophinsand their receptors: roles in plasticity, neurodegeneration andneuroprotection. Biochem. Soc. Trans. 35, 424–427.

Hollyday, M., Hamburger, V., 1976. Reduction in naturallyoccurring motor neuron loss by enlargement of the periphery.J. Comp. Neurol. 170, 311–320.

Huang, Z., Huang, P.L., Ma, J., Meng, W., Ayata, C., Fishman,M.C., Moskowitz, M.A., 1996. Enlarged infarcts in endothelialnitric oxide synthase knockout mice are attenuated bynitro-L-arginine. J. Cereb. Blood Flow Metab. 16, 981–987.

Kahles, T., Luedike, P., Endres, M., Galla, H.J., Steinmetz, H., Busse,R., Neumann-Haefelin, T., Brandes, R.P., 2007. NADPH oxidaseplays a central role in blood-brain barrier damage inexperimental stroke. Stroke 38, 3000–3006.

Kempermann, G., Gast, D., Kronenberg, G., Yamaguchi, M., Gage,F.H., 2003. Early determination and long-term persistence ofadult-generated new neurons in the hippocampus of mice.Development 130, 391–399.

Khan, M., Sekhon, B., Jatana, M., Giri, S., Gilg, A.G., Sekhon, C.,Singh, I., Singh, A.K., 2004. Administration of N-acetylcysteineafter focal cerebral ischemia protects brain and reducesinflammation in a rat model of experimental stroke. J.Neurosci. Res. 76, 519–527.

Kindy, M.S., Bhat, A.N., Bhat, N.R., 1992. Transient ischemiastimulates glial fibrillary acid protein and vimentin geneexpression in the gerbil neocortex, striatum and hippocampus.Brain Res. Mol. Brain Res. 13, 199–206.

Krieglstein, K., Strelau, J., Schober, A., Sullivan, A., Unsicker, K.,2002. TGF-β and the regulation of neuron survival and death.J. Physiol. 96, 25–30.

Kuno, M., 1990. Target dependence of motoneuronal survival: thecurrent status. Neurosci. Res. 9, 155–172.

Levi-Montalcini, R., 1982. Developmental neurobiology and thenatural history of nerve growth factor. Annu. Rev. Neurosci. 5,341–362.

Liu, J., Solway, K., Messing, R.O., Sharp, F.R., 1998. Increasedneurogenesis in the dentate gyrus after transient globalischemia in gerbils. J. Neurosci. 18, 7768–7778.

Mattson, M.P., Zhu, H., Yu, J., Kindy, M.S., 2000. Presenilin-1mutation increases neuronal vulnerability to focal ischemia invivo and to hypoxia and glucose deprivation in cell culture:involvement of perturbed calcium homeostasis. J. Neurosci. 20,1358–1364.

McManaman, J.L., Crawford, F.G., Stewart, S.S., Appel, S.H., 1988.Purification of a skeletal muscle polypeptide which stimulatescholine acetyltransferase activity in cultured spinal cordneurons. J. Biol. Chem. 263, 5890–5897.

McManaman, J.L., Oppenheim, R.W., Prevette, D., Marchetti, D.,1990. Rescue of motoneuron from cell death by a purifiedskeletal muscle polypeptide: effects of the ChAT developmentfactor, CDF. Neuron 9, 155–172.

Mennerick, S., Zorumski, C.F., 2000. Neural activity and survival inthe developing nervous system. Mol. Neurobiol. 22, 41–54.

O'Brien, R.J., Fischbach, G.D., 1986. Modulation of embryonic chickmotoneuron glutamate sensitivity by interneurons andagonists. J. Neurosci. 6, 3290–3296.

Oppenheim, R.W., 1989. The neurotrophic theory and naturallyoccurring motoneuron death. Trends Neurosci. 12, 252–255.

Oppenheim, R.W., Haverkamp, L.J., 1988. Neurotrophicinteractions in the development of spinal cord motoneurons.Ciba Found Symp. 138, 152–171.

Petito, C.K., Olarte, J.P., Roberts, B., Nowak Jr., T.S., Pulsinelli,W.A., 1998. Selective glial vulnerability following transientglobal ischemia in rat brain. J. Neuropathol. Exp. Neurol. 57,231–238.

Rami, A., Bechmann, I., Stehle, J.H., 2008. Exploiting endogenousanti-apoptotic proteins for novel therapeutic strategies incerebral ischemia. Prog. Neurobiol. 85, 273–296.

Rodríguez-Yáñez, M., Castillo, J., 2008. Role of inflammatorymarkers in brain ischemia. Curr. Opin. Neurol. 21, 353–357.

Rothwell, N.J., Luheshi, G.N., 1996. Brain TNF: damage limitation ordamaged reputation? Nat. Med. 2, 746–747.

153B R A I N R E S E A R C H 1 2 3 8 ( 2 0 0 8 ) 1 4 3 – 1 5 3

Sendtner, M., Pei, G., Beck, M., Schweizer, U., Wiese, S., 2000.Developmental motoneuron cell death and neurotrophicfactors. Cell Tissue Res. 301, 71–84.

Sivilia, S., Giuliani, A., Del Vecchio, G., Giardino, L., Calza, L., 2008.Age-dependent impairment of hippocampal neurogenesis inchronic cerebral hypoperfusion. Neuropath. Appl. Neurobiol.34, 52–61.

Smith, R.G., Vaca, K., McManaman, J., Appel, S.H., 1986. Selectiveeffects of skeletal muscle extract fractions on motoneurondevelopment in vitro. J. Neurosci. 6, 439–447.

Thoenen, H., Edgar, D., 1985. Neurotrophic factors. Science 229,238–242.

Varon, S., Manthorpe, M., Davis, G.E., Williams, L.R., Skaper, S.D.,1988. Growth factors. Adv. Neurol. 47, 493–521.

Wachs, F.P., Winner, B., Couillard-Despres, S., Schiller, T., Aigner,R., Winkler, J., Bogdahn, U., Aigner, L., 2006. Transforminggrowth factor-beta1 is a negative modulator of adultneurogenesis. J. Neuropathol. Exp. Neurol. 65, 358–370.

Wang, A.M., Chau, R.M.W., Chow, S.P., Zhang, Z.Y., Li, Z.M., 1995.Effects ofmyogenic 22 and35kDneurotrophic factors onaxonalregeneration in free peripheral autografts into rat spinal cord.Chin. J. Spine Spinal Cord 5, 248–252.

Wang, Y., Jin, K., Greenberg, D.A., 2007. Neurogenesis associatedwith endothelin-induced cortical infarction in the mouse.Brain Res. 1167, 118–122.

Wong, C.H., Crack, P.J., 2008. Modulation of neuro-inflammationand vascular response by oxidative stress following cerebralischemia–reperfusion injury. Curr. Med. Chem. 15, 1–14.

Young, A.R., Ali, C., Duretête, A., Vivien, D., 2007. Neuroprotectionand stroke: time for a compromise. J. Neurochem. 103,1302–1309.

Zhou, M.H., Wu, X.Y., Ren, F., Zhao, L.P., Huang, W.Q., Yang, Z.Y.,Ren, L.S., 1992. Effect of 22kD, 35 kD molecules from skeletalmuscle extract on ant. Horn motoneuron of lumbar spine inrat. Chin. Sci. Bull. 37, 1742–1745.

Zhou, M.H., Yu, W.H., Reb, F., 1993. Changes in MNTF and itsreceptor in tonguemuscle post-denervation of the hypoglossalnerve. Acta Anat. Sin. 24, 391–395.

Zhou, M.H., et al., 1994. Immunohistochemical localization ofMotoneuronotrophic factor in fetal and neonatal rats. ActaAnat. Sin. 25, 189–192.

Zhou, M.H., Wu, X., Chen, S., 1997. Distribution of MNTF1 in spinalcord and limb muscles of mice with motoneuron disease. ActaAcad. Med. Sin. 19, 171–178.