B R A I N R E S E A R C H 1 3 0 4 ( 2 0 0 9 ) 1 5 5 – 1 6 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/ loca te /b ra i n res

Research Report

Heat shock protein 27 upregulation and phosphorylation in ratexperimental autoimmune encephalomyelitis

Heechul Kima,1, Changjong Moonb,1, Meejung Ahna, Jeesung Byuna, Yongduk Leea,Moon-Doo Kimc, Yoh Matsumotod, Chang-Sung Kohe, Taekyun Shina,⁎aDepartment of Veterinary Anatomy, College of Veterinary Medicine, Veterinary Medical Research Institute, Jeju National University,1 Ara-1-Dong, Jeju, Jeju 690-756, South KoreabDepartment of Veterinary Anatomy, College of Veterinary Medicine and Animal Medical Institute, Chonnam National University,Gwangju 500-757, South KoreacDepartment of Psychiatry, Jeju National University College of Medicine, Jeju 690-756, South KoreadDepartment of Molecular Neuropathology, Tokyo Metropolitan Institute for Neuroscience, Fuchu, Tokyo 183-8526, JapaneDepartment of Biomedical Laboratory Sciences, Shinshu University School of Health Sciences, 3-1-1 Asahi, Matsumoto 390-8621, Japan

A R T I C L E I N F O

⁎ Corresponding author. Fax: +82 64 756 3354.E-mail address: shint@jejunu.ac.kr (T. Sh

1 The first two authors equally contributed

0006-8993/$ – see front matter © 2009 Elsevidoi:10.1016/j.brainres.2009.09.060

A B S T R A C T

Article history:Accepted 15 September 2009Available online 23 September 2009

Following stress or inflammation, the 27-kDa heat shock protein (HSP27) is induced invarious cell types, where it promotes cell survival and inhibits inflammatory reactions. Weexamined the expression of HSP27 and phosphorylated HSP27 (p-HSP27) in the spinal cordsof Lewis rats with experimental autoimmune encephalomyelitis (EAE). Western blottinganalysis revealed low levels of HSP27 and p-HSP27 in the normal spinal cords andsignificantly higher levels in EAE-affected spinal cords. Immunohistochemistry revealedthat HSP27 was expressed constitutively in the neurons and some fibrous astrocytes of thespinal cords of normal rats. However, in EAE-affected spinal cords, HSP27 immunoreactivitywas higher and located primarily in the fibrous astrocytes of the white matter, whereas fewof the inflammatory cells were immunopositive for HSP27. Immunoreactivity for p-HSP27was detected predominantly in the fibrous astrocytes of the normal controls and wasmarkedly increased in EAE-affected spinal cords. Therefore, the levels of HSP27 expressionand phosphorylation of HSP27 were increased primarily during reactive astrogliosis ofspinal white matter affected by EAE. These observations suggest that in rat EAE, theincreased expression and elevated activation of HSP27 modulate host cell activity, survival,and inflammation to counter the autoimmune inflammatory injury. Our results also suggestthat HSP27 plays a role in spontaneous recovery from EAE-induced paralysis.

© 2009 Elsevier B.V. All rights reserved.

Keywords:AstrocytesNeuronExperimental autoimmuneencephalomyelitisHeat shock protein 27

1. Introduction

Heat shock proteins (HSPs) and stress proteins are produced incells in response to various types of insult, including heat

in).to this work.

er B.V. All rights reserved

shock, ischemia, and inflammatory mediators. The specificproperties of the 27-kDa heat shock protein (HSP27), which isinduced in response to cellular stress, include stabilization ofthe actin cytoskeleton, attenuation of the apoptotic pathways

.

156 B R A I N R E S E A R C H 1 3 0 4 ( 2 0 0 9 ) 1 5 5 – 1 6 3

through interactions with proapoptotic proteins, and thesuppression of aggregate formation (Bruey et al., 2000;Charette et al., 2000; Ehrnsperger et al., 1997; Guay et al.,1997). Although the protective effects of various HSPs on cellsurvival have been studied extensively, there is evidence thatHSP27 exerts a more potent protective effect in the centralnervous system (CNS). Under pathological conditions, e.g.,brain ischemia (Imura et al., 1999), Alexander's disease (Iwakiet al., 1993), multiple sclerosis (van Noort et al., 1995), andAlzheimer's disease (Shinohara et al., 1993), HSP27 ismarkedlyinduced in the reactive glial cells in the CNS.

Under certain stress stimuli, HSP27 is phosphorylated bymitogen-activated protein kinase-activated protein kinase 2 attwo or three serine residues (Rogalla et al., 1999). It is not clearwhether the protective effects of HSP27 on cell viability aredependent upon its phosphorylation. The overexpression of(unphosphorylated) HSP27 did not influence the ability of thecells to withstand heat shock (Geum et al., 2002; Lavoie et al.,1995). However, the chaperone activities of HSP27 after heatshock (for thermoresistance) were found to be independent ofphosphorylation (Knauf et al., 1994). In vivo, there was nosignificant difference between the overexpression of wild-type HSP27 and a non-phosphorylatable HSP27 mutant interms of protecting the heart from ischemia–reperfusioninjury (Hollander et al., 2004). These findings suggest thatthe activity of HSP27 in ensuring cell survival followingcellular stress is potentially regulated by phosphorylationand that the mechanism of the HSP27 response variesdepending on the system studied or the disease status.

Experimental autoimmune encephalomyelitis (EAE) is anexperimentally induced autoimmune disease of the CNS thatis mediated by CD4+ T cells. EAE is used as an animal model ofmultiple sclerosis, which is a human demyelinating disease(Raine, 1994). EAE lesions are characterized by (1) the infiltra-tion of T cells and macrophages into the subarachnoid spaceduring the early stages of the disease and (2) the activation ofmicroglia and astrocytes during the peak symptomatic stagesof the disease (Kim et al., 2007; Shin et al., 1995).

In murine EAE, HSP60 (Gao et al., 1995) and HSP32 (Stahnkeet al., 2007) are expressed and upregulated in various cell types

Fig. 1 – Western blotting analysis of HSP27 and p-HSP27 levelsduring the peak stage (day 12 p.i.; EAE G.3) and recovery stageof Western blots for the detection of HSP27 (27 kDa), p-HSP27densitometric data analysis (mean±S.E., n=3 rats/group). The daof HSP27 and p-HSP27 relative to the normal control, after nor*P<0.05 vs. normal controls.

in the spinal cord. In addition, when injected prior to EAEinduction, HSP60 (Birnbaum et al., 1996) and HSP70 (Galazka etal., 2006) prevent the development of EAE. Another lowmolecular weight HSP, αB-crystallin, has also been investigat-ed in the EAE model (Ousman et al., 2007). Compared withtheir wild-type counterparts, αB-crystallin-knockout miceshowed more severe EAE pathology, comprised of intenseinflammation, high levels of T cell proliferation in response tomyelin oligodendrocyte glycoprotein (MOG), and high levels ofTh1 and Th17 cytokine production (Ousman et al., 2007).Furthermore, αB-crystallin-/- astrocytes and glial cells in micewith EAE showed higher levels of apoptosis in the presence orabsence of TNF stimulation, in comparison to wild-typeastrocytes and their wild-type counterparts, respectively(Ousman et al., 2007). These observations suggest that HSPsregulate the pathogenesis of EAE. However, the expressionand functions of HSP27 in EAE-affected spinal cords remainunclear.

In the present study, we examined the expression andlocalization of HSP27 and the phosphorylated form of HSP27(p-HSP27) in the spinal cords of Lewis rats with acute EAE toinvestigate the involvement of these proteins in the patho-genesis of EAE.

2. Results

2.1. Western blotting for HSP27 and p-HSP27 inEAE-affected spinal cords

The expression levels of HSP27 and p-HSP27 in the spinalcords during the course of EAE were evaluated semiquantita-tively by Western blotting (Fig. 1). Low levels of HSP27expression were detected in the spinal cords of the normalcontrol rats (Fig. 1A). However, the level of HSP27 expressionwas significantly higher in the spinal cords during the peakstages of EAE (G.3; relative OD values, 4.76±0.21-fold increase,n=3, P<0.05 vs. normal controls), and increased progressivelyduring the recovery stage of EAE (R.0, day 21 postimmuniza-tion [p.i.]; relative OD values, 7.64±0.59-fold increase, n=3,

in the spinal cords of normal control rats (Cont) and rats(day 21 p.i.; EAE R.0) of EAE. (A) Representative photographs(27 kDa), and β-actin (45 kDa). (B and C) Results ofta are presented as mean (±S.E.)-fold increases in the levels

malization for the β-actin band in each sample. **P<0.01,

157B R A I N R E S E A R C H 1 3 0 4 ( 2 0 0 9 ) 1 5 5 – 1 6 3

P<0.05 vs. normal controls) (Fig. 1B). Low levels of p-HSP27expression were detected in the spinal cords of the normalcontrol rats (Fig. 1A). However, the levels of p-HSP27 expres-sion were significantly higher in the spinal cords during thepeak stages of EAE (G.3; relative OD values, 4.4±0.19-foldincrease, n=3, P<0.05 vs. normal controls), and these levelsweremaintained through to the recovery stage of EAE (R.0, day21 p.i.; relative OD values, 4.98±0.11-fold increase, n=3,P<0.01 vs. normal controls) (Fig. 1C).

2.2. Immunohistochemical localization of HSP27 inEAE lesions

Constitutive HSP27 expression was detected immunohisto-chemically in some neuron (mean number±SEM, 9.8±0.8 in400× magnification fields) and fibrous astrocytes (3.4±0.75) inthe spinal cords of the normal rats (Figs. 2A and E). At the peakstage of EAE (day 12 p.i.), increased expression of HSP27 was

Fig. 2 – Immunohistochemical stainingofHSP27 in the spinal cord12 p.i.) (B, D and F) and recovery stage (day 21 p.i.) (C and G) of EAdetected in some of the fibrous astrocytes in the white matter (A,some vessels in the graymatter (E, arrowhead). At the peak stageastrocytes in the white matter (B, arrowheads), whereas the lev(F, arrows) and in the infiltrating inflammatory cells in the perivain immunoreactivity in the endothelial cells or vessels (F, arrowh(E, arrowhead). At the recovery stage of EAE, the levels of HSP27 i(C, arrowheads) and the numbers of HSP27-positive neurons werspecimens were counterstained with hematoxylin. Scale bars, 60

observed in the fibrous astrocytes (22.6±2.25; P<0.05 vs.controls) in the white matter (Fig. 2B), whereas a very lowlevel of HSP27 immunoreactivity was detected in the proto-plasmic astrocytes in the gray matter. The number of HSP27-positive neurons (5±0.84; P<0.05 vs. controls) was decreasedcompared with the normal controls (Fig. 2F). In contrast, therewas little HSP27 immunoreactivity in the infiltrating inflam-matory cells in the perivascular cuffing (Fig. 2D). During therecovery stage (day 21 p.i.), the number of HSP27-positiveastrocytes was significantly decreased compared with thatduring the peak stage of EAE, although it was still higher thanthat of the normal controls (14.75±1.11; P<0.05 vs. controls,and peak stage of EAE) (Fig. 2C). However, the number ofHSP27-positive neurons was significantly higher in therecovery stage (9±0.71; P<0.05) than in the peak stage of EAE(Fig. 2G). The immunostaining patterns for HSP27 in the EAElesions corresponded well with the results of Western blottinganalysis.

s of normal control rats (A andE) and rats at thepeak stage (dayE. In the spinal cords of the normal control rats, HSP27 wasarrowhead) as well as in most of the neurons (E, arrow) andof EAE, increased levels of HSP27were detected in the fibrousel of HSP27 immunoreactivity was very low in the neuronsscular cuffing (D, arrow). There were no significant changesead) between the peak stage and the normal controlsmmunoreactivity in the fibrous astrocytes were decreasede increased (G, arrow) in comparison to the peak stage. Theμm (A–C) and 30 μm (D–G).

Fig. 3 – Double-immunofluorescence colocalization of HSP27 (A andD)with either GFAP (B) or ED1 (E) in the spinal cords of Lewisrats with EAE (day 12 p.i.). Colocalization of HSP27 (A; red, arrows) with GFAP (B; green, arrows) was observed in some cells inthe EAE-affected spinal cords (C; merged, arrows). Very low levels of HSP27 immunoreactivity (D; red, arrow) were observedin the ED1-positive macrophages (E; green, arrow) in the EAE-affected spinal cord (F; merged, arrow). Scale bars, 20 μm.

158 B R A I N R E S E A R C H 1 3 0 4 ( 2 0 0 9 ) 1 5 5 – 1 6 3

Double-immunofluorescence experiments were performedto confirm theHSP27phenotype. In theEAE lesions (day 12p.i.),HSP27 immunoreactivity (Fig. 3A; red, arrows) in the whitematter was colocalized primarily with the GFAP-positiveastrocytes (Fig. 3B; green, arrows), indicating that most of thefibrous astrocytes were positive for HSP27 (Fig. 3C; merged,arrows). Few HSP27-positive cells (Fig. 3D; red, arrow) coloca-lizedwith the ED1-positivemacrophages (Fig. 3E; green, arrow)in the EAE-affected spinal cords (Fig. 3F, merged).

2.3. Immunohistochemical localization of p-HSP27 inEAE lesions

p-HSP27 was expressed constitutively in some neurons (meannumber±SEM: 10.2±1.16 in 400× magnification fields) andfibrous astrocytes (7.4±1.17) in the spinal cords of normal rats(Figs. 4A andD). At the peak stage of EAE (day 12 p.i.), increasedexpression of p-HSP27 was observed in the fibrous astrocytesin the white matter (27±3.58; P<0.05 vs. controls) (Fig. 4B),whereas p-HSP27 immunoreactivity was rarely detected in theprotoplasmic astrocytes of the gray matter. The number of p-HSP27-positive neurons was decreased compared with that innormal control rats (3±1.47; P<0.05 vs. controls) (Fig. 4E). p-HSP27 immunoreactivity was weak in the infiltrating inflam-matory cells in the perivascular cuffing (Fig. 4E). At therecovery stage (day 21 p.i.), the number of p-HSP27-positiveastrocytes was significantly lower than that at the peak stageof EAE, although it was still higher than that in the normalcontrols (19.6±1.81; P<0.05 vs. controls, and rats at peak stageof EAE) (Fig. 4C). The number of p-HSP27-positive neurons atthe recovery stage was significantly higher than that at the

peak stage of EAE (10±0.71; P<0.05 vs. rats at peak stage ofEAE) (Fig. 4F).

We performed double-immunofluorescence experiments,in which the majority of the p-HSP27-postive cells were GFAP-positive astrocytes of the EAE-affected spinal cords (Figs. 5A–C). Moreover, most of the p-HSP27-positive astrocytes weredetected in the white matter, with few detected in the graymatter (Figs. 5A–C). These observations suggest that most ofthe fibrous astrocytes in the white matter are phosphorylatedwhen inflammatory cells attack the parenchyma of the spinalcord. However, p-HSP27 was detected only weakly in the ED1-positive macrophages (Figs. 5D–F) and NeuN-positive neuronsin the EAE lesions (Figs. 5G–I).

Furthermore, p-HSP27 immunoreactivity was detected inthe HSP27-positive neurons (Figs. 6A and B) and fibrousastrocytes (Figs. 6C and D) in the normal controls and EAE-affected rats, respectively.

3. Discussion

This study confirmed that the levels of HSP27 expression andphosphorylation increase primarily in the reactive fibrousastrocytes of the white matter in the spinal cords of rats withEAE, which corresponds to an increase in astrogliosis affectedby EAE.

HSP27, which is a member of the small heat shockprotein family, is induced by various stress conditions.HSP27 interacts with actin filaments, regulating actin-dependent cell functions (Lavoie et al., 1993) and binds tothe intermediate filaments, including GFAP, to modulate the

Fig. 4 – Immunohistochemical staining of p-HSP27 in the spinal cords of normal control rats (A and D) and rats at the peak stage(day 12 p.i.) (B and E) and recovery stage (day 21 p.i.) (C, F) of EAE. In the spinal cords of the normal control rats, p-HSP27 wasdetected in someof the fibrous astrocytes in thewhitematter (A, arrow) and inmost of theneurons in the graymatter (D, arrows).At the peak stage of EAE, increased levels of p-HSP27 immunoreactivity were observed in the fibrous astrocytes in the whitematter (B, arrows), whereas very low levels of p-HSP27 immunoreactivity were detected in a few neurons (E, arrows) andinflammatory cells (E, arrowhead). At the recovery stage of EAE, the levels of p-HSP27 immunoreactivity in the fibrous astrocyteswere decreased (C, arrows) and the numbers of p-HSP27-positive neurons were increased (F, arrows) in comparison with thecorresponding values during the peak stage of EAE. Sections were counterstained with hematoxylin. Scale bars, 30 μm (A–F).

159B R A I N R E S E A R C H 1 3 0 4 ( 2 0 0 9 ) 1 5 5 – 1 6 3

intermediate filament network in astrocytes (Perng et al.,1999). In vitro and in vivo studies indicated that HSP27 isexpressed mainly in glial cells (Acarin et al., 2002; Faucon-neau et al., 2002). Bajramović et al. (2000) reported thedifferential expression of HSP27 in response to proinflam-matory and immunoregulatory cytokines in human adultastrocytes. In EAE, the increased expression of GFAP isassociated with reactive gliosis (Hardin-Pouzet et al., 1997;Raine, 1985). In the present study, in the EAE-affected spinalcords, HSP27 activity increased primarily in fibrous astro-cytes of the white matter, which contains abundant myelinbasic protein (target auto-antigen). Therefore, the up-regu-lation of HSP27 activity in the white matter may be due tothe local environmental conditions involved in the activa-tion of astrocytes.

The mechanism underlying the protective effect of HSP27against apoptosis may involve (1) decreases in the levels ofreactive oxygen species (Garrido et al., 1997); (2) restoration ofprotein homeostasis and the promotion of cell survival, eitherthrough the repair of damaged proteins (protein refolding) ortheir degradation (Jolly and Morimoto, 2000); (3) stabilizationof the actin cytoskeleton (Richards et al., 1996); (4) a delay in

the release of cytochrome c from the mitochondria and Bidintracellular redistribution (Paul et al., 2002); and (5) inhibitionof caspase-3 activation (Xanthoudakis and Nicholson, 2000).There is also accumulating evidence for a neuroprotectiveeffect of HSP27 on motor neurons, in that the exogenousdelivery of HSP27 promotes motor neuron survival for 1 weekfollowing neonatal nerve axotomy (Benn et al., 2002), althoughthe precise roles of HSP27 in neurons remain unclear.

In the multiple sclerosis brain, high levels of HSP27immunoreactivity were detected in reactive astrocytes, peri-vascular macrophages, endothelial cells, and some oligoden-drocytes within the lesions, whereas HSP27 immunoreactivitywas weak in endothelial cells and astrocytes in the adjacentwhite matter (Aquino et al., 1997). In addition, strongexpression of HSP27 was detected in the astrocytes and inthe walls of the blood vessels of the cortices of epilepsypatients with multiple sclerosis, as well in the astrocytes inthe white matter, whereas such immunoreactivity in astro-cytes was rarely seen in the gray matter, with the exception ofthe blood vessels (Bidmon et al., 2004). These patterns ofHSP27 immunoreactivity are in accordance with the findingsof the present study. We showed that the level of HSP27

Fig. 5 – Double-immunofluorescence colocalization of p-HSP27 (A, D, and G) with GFAP (B), ED1 (E), and NeuN (H) in the spinalcords of Lewis rats with EAE (day 12 p.i.). p-HSP27 immunoreactivity (A; red) was colocalized with GFAP (B; green) in some cellsin the white matter of the EAE-affected spinal cords (C; merged). p-HSP27 immunoreactivity (D; red) was detected rarely in theED1-positive macrophages (E; green) and NeuN-positive neurons (H, green) in the EAE-affected spinal cords (F and I; merged).Scale bars, 40 μm.

160 B R A I N R E S E A R C H 1 3 0 4 ( 2 0 0 9 ) 1 5 5 – 1 6 3

expression in astrocytes was different between the white andgray matter of the spinal cords of rates with EAE; strongexpression of HSP27 was observed in the white matter. Theseobservations indicated that HSP27 is selectively expressed

Fig. 6 – Immunohistochemical staining for HSP27 (A and C) andrats (A and B) and rats at the peak stage of EAE (day 12 p.i.) (Cwith p-HSP27-positive neurons (B, arrows) in the control spina(C, arrows) was colocalized with p-HSP27 in the astrocytes (D,adjacent sections. Sections A and B and sections C and D are s

during the astrocyte response to cellular stress, such asinflammation.

Interestingly, the neurons and most inflammatory cells,including ED1-positive macrophages, in the EAE lesions

p-HSP27 (B and D) in the spinal cords of the normal controland D). HSP27 immunoreactivity (A; arrows) was colocalizedl cords in adjacent sections. HSP27 immunoreactivityarrows) in the white matter of EAE-affected spinal cords inerial mirror sections. Scale bars, 30 μm.

161B R A I N R E S E A R C H 1 3 0 4 ( 2 0 0 9 ) 1 5 5 – 1 6 3

transiently showed no or only weak expression of HSP27 and p-HSP27 during the peak stage of EAE. With regard to thepathology of EAE, antigen-responsive inflammatory cells infil-trate the CNS, and are eliminated by apoptosis (Moon et al.,2000). In addition, neuronal cell death occurs after induction ofEAE through multiple apoptosis pathways (Das et al., 2008). Arepresentative signalingpathway for apoptosis in inflammatorycells and neuronal cells in EAE involves caspase-3 (Ahmed et al.,2002; Das et al., 2008; Irony-Tur-Sinai et al., 2009). Therefore, itseems likely that the lack of expression and/or phosphorylationof HSP27 in inflammatory cells and/or neuronal cells in thespinal cords of rats with EAE would lead to loss of inhibition ofcaspase-3 activation, thusmay lead toapoptosis. Additionally, itis not clear whether the cytoprotective effects of HSP27 aremediatedby itsphosphorylation.Theoverexpressionof (unpho-sphorylated) HSP27 does not influence the ability of cells towithstand heat shock (Geum et al., 2002; Lavoie et al., 1995).However, the chaperone activities of HSP27 after heat shock (forthermoresistance) have been shown to be independent of phos-phorylation (Knauf et al., 1994). In the present study, Westernblotting analysis indicated that the temporal patterns of HSP27expression and phosphorylation show different changes duringEAE progression. Further studies are needed to clarify theprecise roles of HSP27 and p-HSP27 in the EAE spinal cord.

In conclusion, the expression and phosphorylation ofHSP27 increase mainly during reactive astrogliosis of thewhite matter after the induction of EAE, which suggests thatthe upregulation and phosphorylation of HSP27modulate hostcell activity and survival, as well as inflammation, to counterthe autoimmune inflammatory injury, thereby possibly play-ing a role in recovery from EAE-induced paralysis.

4. Experimental procedures

4.1. Animals

Lewis rats were obtained from Harlan (Indianapolis, IN) andwere bred at our animal facility. Female rats 7–12 weeks oldand weighing 160–200 g were used. All animal experimentswere carried out in accordance with the Jeju NationalUniversity Guide for the Care and Use of Laboratory Animals.

4.2. EAE induction

The footpads of the hind feet of each rat in the EAE groupwereinjected with 100 μL of an emulsion containing equal parts ofmyelin basic protein (1 mg/mL) and complete Freund'sadjuvant (CFA) supplemented with Mycobacterium tuberculosisH37Ra (5 mg/mL; Difco, Detroit, MI). After injection, the ratswere observed daily for any clinical signs of EAE. Theprogression of EAE was classified as follows: Grade 0 (G.0),no signs; G.1, floppy tail; G.2, mild paraparesis; G.3, severeparaparesis; G.4, tetraparesis; G.5, moribund condition ordeath; and R.0, in the recovery stage.

4.3. Antibodies

The rabbit polyclonal anti-HSP27 (Santa Cruz Biotechnology,Santa Cruz, CA), rabbit polyclonal anti-p-HSP27 (Ser 82;

Upstate Biotechnology, Lake Placid, NY) and mouse monoclo-nal anti-β-actin (Sigma-Aldrich, St. Louis, MO) antibodies wereused in this study. The cell phenotypes were identified usingthe followingmonoclonal antibodies:mousemonoclonal anti-glial fibrillary acidic protein (GFAP; Sigma-Aldrich); ED1(mouse monoclonal anti-rat macrophage; Serotec, London,UK); and anti-NeuN (Chemicon, Billerica, MA).

4.4. Tissue sampling

Spinal cords were sampled on days 12–15 p.i. and day 21 p.i.,corresponding to the peak (G.3) and recovery stages (R.0) ofEAE, respectively. The spinal cords of the normal rats wereused as controls. At each sampling, rats from each group weresacrificed under deep anesthesia, and samples of the lumbarspinal cords were collected. The spinal cord samples wereprocessed for paraffin embedding after fixation in 4% para-formaldehyde in phosphate-buffered saline (PBS) (n=5 pergroup) or stored at -70 °C for biochemical analysis (n=6 pergroup).

4.5. Western blotting analysis

Spinal cord tissue was homogenized in lysis buffer (40 mMTris, 120 mM NaCl, 0.1% Nonidet-40, 2 mM Na3VO4, 1 mMphenylmethylsulfonyl fluoride, 10 μg/mL aprotinin, 10 μg/mLleupeptin) with 20 strokes in a homogenizer. The homogenatewas transferred to microtubes and centrifuged at 14,000 rpmfor 20 min. The supernatant was then harvested. Forimmunoblotting, aliquots of the supernatant containing40 μg of protein were subjected to 15% sodium dodecylsulfate-polyacrylamide gel electrophoresis (SDS-PAGE) andimmunoblotted onto nitrocellulose membranes (Schleicherand Schuell BioScience, Keene, NH). The residual binding siteson the membrane were blocked by incubation with 5% nonfatmilk in Tris-buffered saline (TBS; 10 mM Tris–HCl, pH 7.4,150 mM NaCl) for 1 h, followed by incubation with the rabbitpolyclonal anti-HSP27 and anti-p-HSP27 antibodies for 2 h.The blots were washed three times in TBS containing 0.1%Tween-20, and then incubated with horseradish peroxidase-conjugated anti-rabbit IgG (Vector Laboratories, Burlingame,CA) for 1 h. The bound antibody was detected using enhancedchemiluminescence (ECL) reagents (Amersham, ArlingtonHeights, IL) according to the manufacturer's instructions.After imaging, the membranes were stripped and reprobedusing the mouse monoclonal anti-β-actin antibody. Thedensity of each band was measured using a GS-700 scanninglaser densitometer (Bio-Rad Laboratories, Hercules, CA). Theratio of the density of the HSP27 band to that of the β-actinband was compared using the Molecular Analyst software(Bio-Rad Laboratories). The results were analyzed using one-way analysis of variance (ANOVA) followed by post hocStudent–Newman–Keuls t test for multiple comparisons. Inall cases, P<0.05 was considered statistically significant.

4.6. Immunohistochemistry

Paraffin-embedded spinal cord sections (5 μm thick) weredeparaffinized, treated with 0.01 M citrate buffer (pH 6.0) in amicrowave for 3 min for antigen retrieval, and then treated

162 B R A I N R E S E A R C H 1 3 0 4 ( 2 0 0 9 ) 1 5 5 – 1 6 3

with 0.3% hydrogen peroxide in methyl alcohol for 20 min, toblock endogenous peroxidase activity. After threewasheswithPBS, the sectionswere incubated with 10% normal goat serum,and then with the primary antibodies, i.e., rabbit polyclonalanti-HSP27 and anti-p-HSP27 antibodies, for 1 h at roomtemperature. Immunoreactivity was visualized using theVector Elite avidin–biotin peroxidase complex (Vector Labora-tories). The peroxidase reaction was developed using adiaminobenzidine substrate kit (Vector Laboratories).

Paraffin sections were used for double-immunofluores-cence analyses of HSP27 and p-HSP27 in rat macrophages orastrocytes. The paraffin sections were deparaffinized andhydrated. The paraffin sections were washed three timeswith PBS and incubated with 10% normal goat serum for 1h at room temperature, followed by incubation with rabbitpolyclonal anti-HSP27 or anti-p-HSP27 overnight at 4 °C.The sections were then reacted with biotinylated anti-rabbitIgG (Vector Laboratories), followed by tetramethyl rhoda-mine isothiocyanate (TRITC)-labeled streptavidin (Zymed,South San Francisco, CA). The slides were incubated withthe second primary reagent including ED1, anti-GFAP, andanti-NeuN, and sequentially reacted with fluorescein iso-thiocyanate (FITC)-labeled goat anti-mouse IgG (Sigma-Aldrich). The double-labeled specimens were examinedunder an FV 500 laser confocal microscope (Olympus,Tokyo, Japan).

For semiquantitative analysis of immunostaining forHSP27 and p-HSP27 in the neurons of the ventral horn andfibrous astrocytes in the white matter, the numbers of HSP27-and p-HSP27-positive neurons in the ventral horn, and GFAP-positive astrocytes in the white matter were assessed inrandomly chosen fields (400× magnification) from each group(n=5).

R E F E R E N C E S

Acarin, L., Paris, J., Gonzalez, B., Castellano, B., 2002. Glialexpression of small heat shock proteins following anexcitotoxic lesion in the immature rat brain. Glia 38, 1–14.

Ahmed, Z., Doward, A.I., Pryce, G., Taylor, D.L., Pocock, J.M.,Leonard, J.P., Baker, D., Cuzner, M.L., 2002. A role forcaspase-1 and -3 in the pathology of experimental allergicencephalomyelitis : inflammation versus degeneration. Am.J. Pathol. 161, 1577–1586.

Aquino, D.A., Capello, E., Weisstein, J., Sanders, V., Lopez, C.,Tourtellotte, W.W., Brosnan, C.F., Raine, C.S., Norton, W.T.,1997. Multiple sclerosis: altered expression of 70- and 27-kDaheat shock proteins in lesions and myelin. J. Neuropathol. Exp.Neurol. 56, 664–672.

Bajramović, J.J., Bsibsi, M., Geutskens, S.B., Hassankhan, R.,Verhulst, K.C., Stege, G.J., de Groot, C.J., van Noort, J.M., 2000.Differential expression of stress proteins in human adultastrocytes in response to cytokines. J. Neuroimmunol. 106,14–22.

Benn, S.C., Perrelet, D., Kato, A.C., Scholz, J., Decosterd, I., Mannion,R.J., Bakowska, J.C., Woolf, C.J., 2002. Hsp27 upregulation andphosphorylation is required for injured sensory and motorneuron survival. Neuron 36, 45–56.

Bidmon, H.J., Görg, B., Palomero-Gallagher, N., Behne, F., Lahl, R.,Pannek, H.W., Speckmann, E.J., Zilles, K., 2004. Heat shockprotein-27 is upregulated in the temporal cortex of patientswith epilepsy. Epilepsia 45, 1549–1559.

Birnbaum, G., Kotilinek, L., Schlievert, P., Clark, H.B., Trotter, J.,Horvath, E., Gao, E., Cox, M., Braun, P.E., 1996. Heat shockproteins and experimental autoimmune encephalomyelitis(EAE): I. Immunization with a peptide of the myelinprotein 2′,3′ cyclic nucleotide 3′ phosphodiesterase that iscross-reactive with a heat shock protein alters the courseof EAE. J. Neurosci. Res. 44, 381–396.

Bruey, J.M., Ducasse, C., Bonniaud, P., Ravagnan, L., Susin, S.A.,Diaz-Latoud, C., Gurbuxani, S., Arrigo, A.P., Kroemer, G., Solary,E., Garrido, C., 2000. Hsp27 negatively regulates cell death byinteracting with cytochrome C. Nat. Cell Biol. 2, 645–652.

Charette, S.J., Lavoie, J.N., Lambert, H., Landry, J., 2000. Inhibitionof Daxx-mediated apoptosis by heat shock protein 27. Mol. Cell.Biol. 20, 7602–7612.

Das, A., Guyton, M.K., Matzelle, D.D., Ray, S.K., Banik, N.L., 2008.Time-dependent increases in protease activities for neuronalapoptosis in spinal cords of Lewis rats during developmentof acute experimental autoimmune encephalomyelitis.J. Neurosci. Res. 86, 2992–3001.

Ehrnsperger, M., Graber, S., Gaestel, M., Buchner, J., 1997. Bindingof nonnative protein to Hsp25 during heat shock creates areservoir of folding intermediates for reactivation. EMBO J. 16,221–229.

Fauconneau, B., Petegnief, V., Sanfeliu, C., Piriou, A., Planas, A.M.,2002. Induction of heat shock proteins (HSPs) by sodiumarsenite in cultured astrocytes and reduction of hydrogenperoxide-induced cell death. J. Neurochem. 83, 1338–1348.

Galazka, G., Stasiolek, M., Walczak, A., Jurewicz, A., Zylicz, A.,Brosnan, C.F., Raine, C.S., Selmaj, K.W., 2006. Brain-derivedheat shock protein 70-peptide complexes induce NKcell-dependent tolerance to experimental autoimmuneencephalomyelitis. J. Immunol. 176, 1588–1599.

Gao, Y.L., Brosnan, C.F., Raine, C.S., 1995. Experimentalautoimmune encephalomyelitis. Qualitative andsemiquantitative differences in heat shock protein 60expression in the central nervous system. J. Immunol. 154,3548–3556.

Garrido, C., Ottavi, P., Fromentin, A., Hammann, A., Arrigo, A.P.,Chauffert, B., Mehlen, P., 1997. HSP27 as a mediator ofconfluence dependent resistance to cell death induced byanticancer drugs. Cancer Res. 57, 2661–2667.

Geum, D., Son, G.H., Kim, K., 2002. Phosphorylation-dependentcellular localization and thermoprotective role of heat shockprotein 25 in hippocampal progenitor cells. J. Biol. Chem. 277,19913–19921.

Guay, J., Lambert, H., Gingras-Breton, G., Lavoie, J.N., Huot, J.,Landry, J., 1997. Regulation of actin filament dynamics by p38map kinase-mediated phosphorylation of heat shock protein27. J. Cell Sci. 110, 357–368.

Hardin-Pouzet, H., Krakowski, M., Bourbonnière, L., Didier-Bazes,M., Tran, E., Owens, T., 1997. Glutamate metabolism isdown-regulated in astrocytes during experimental allergicencephalomyelitis. Glia 20, 79–85.

Hollander, J.M., Martin, J.L., Belke, D.D., Scott, B.T., Swanson, E.,Krishnamoorthy, V., Dillmann, W.H., 2004. Overexpression ofwild-type heat shock protein 27 and a nonphosphorylatableheat shock protein 27 mutant protects against ischemia/reperfusion injury in a transgenic mouse model. Circulation110, 3544–3552.

Imura, T., Shimohama, S., Sato, M., Nishikawa, H., Madono, K.,Akaike, A., Kimura, J., 1999. Differential expression of smallheat shock proteins in reactive astrocytes after focal ischemia:possible role of beta-adrenergic receptor. J. Neurosci. 19,9768–9779.

Irony-Tur-Sinai, M., Grigoriadis, N., Tsiantoulas, D., Touloumi, O.,Abramsky, O., Brenner, T., 2009. Immunomodulation of EAE byalpha-fetoprotein involves elevation of immune cell apoptosismarkers and the transcription factor FoxP3. J. Neurol. Sci. 279,80–87.

163B R A I N R E S E A R C H 1 3 0 4 ( 2 0 0 9 ) 1 5 5 – 1 6 3

Iwaki, T., Iwaki, A., Tateishi, J., Sakaki, Y., Goldman, J.E., 1993.Alpha B-crystallin and 27-kd heat shock protein are regulatedby stress conditions in the central nervous system andaccumulate in Rosenthal fibers. Am. J. Pathol. 143,487–495.

Jolly, C., Morimoto, R.I., 2000. Role of the heat shock response andmolecular chaperones in oncogenesis and cell death. J. Natl.Cancer Inst. 92, 1564–1572.

Kim, H., Moon, C., Ahn,M., Lee, Y., Kim, S., Matsumoto, Y., Koh, C.S.,Kim, M.D., Shin, T., 2007. Increased phosphorylation of cyclicAMP response element-binding protein in the spinal cord ofLewis rats with experimental autoimmune encephalomyelitis.Brain Res. 1162, 113–120.

Knauf, U., Jakob, U., Engel, K., Buchner, J., Gaestel, M., 1994.Stress- andmitogen-induced phosphorylation of the small heatshock protein Hsp25 by MAPKAP kinase 2 is not essential forchaperone properties and cellular thermoresistance. EMBO J.13, 54–60.

Lavoie, J.N., Hickey, E., Weber, L.A., Landry, J., 1993. Modulation ofactin microfilament dynamics and fluid phase pinocytosis byphosphorylation of heat shock protein 27. J. Biol. Chem. 268,24210–24214.

Lavoie, J.N., Lambert, H., Hickey, E., Weber, L.A., Landry, J., 1995.Modulation of cellular thermoresistance and actin filamentstability accompanies phosphorylation-induced changes inthe oligomeric structure of heat shock protein 27. Mol. Cell.Biol. 15, 505–516.

Moon, C., Kim, S., Wie, M., Kim, H., Cheong, J., Park, J., Jee, Y.,Tanuma, N., Matsumoto, Y., Shin, T., 2000. Increasedexpression of p53 and Bax in the spinal cords of rats withexperimental autoimmune encephalomyelitis. Neurosci. Lett.289, 41–44.

Ousman, S.S., Tomooka, B.H., van Noort, J.M., Wawrousek, E.F.,O'Connor, K.C., Hafler, D.A., Sobel, R.A., Robinson, W.H.,Steinman, L., 2007. Protective and therapeutic role foralphaB-crystallin in autoimmune demyelination. Nature 448,474–479.

Paul, C., Manero, F., Gonin, S., Kretz-Remy, C., Virot, S., Arrigo, A.P.,2002. Hsp27 as a negative regulator of cytochrome C release.Mol. Cell. Biol. 22, 816–834.

Perng, M.D., Cairns, L., van den Ijssel, P., Prescott, A., Hutcheson,A.M., Quinlan, R.A., 1999. Intermediate filament interactionscan be altered by HSP27 and alphaB-crystallin. J. Cell Sci. 112,2099–2112.

Raine, C.S., 1985. Experimental allergic encephalomyelitis. In:Koetsier, J.C. (Ed.), Handbook of Clinical Neurology, 3. Elsevier,Amsterdam, pp. 429–466.

Raine, C.S., 1994. The Dale E. McFarlin Memorial Lecture: theimmunology of the multiple sclerosis lesion. Ann. Neurol. 36(Suppl.), S61–S72.

Richards, E.H., Hickey, E., Weber, L., Master, J.R., 1996. Effect ofoverexpression of the small heat shock protein HSP27 on theheat and drug sensitivities of human testis tumor cells. CancerRes. 56, 2446–2451.

Rogalla, T., Ehrnsperger, M., Preville, X., Kotlyarov, A., Lutsch, G.,Ducasse, C., Paul, C., Wieske, M., Arrigo, A.P., Buchner, J.,Gaestel, M., 1999. Regulation of Hsp27 oligomerization,chaperone function, and protective activity against oxidativestress/tumor necrosis factor alpha by phosphorylation. J. Biol.Chem. 274, 18947–18956.

Shin, T., Kojima, T., Tanuma, N., Ishihara, Y., Matsumoto, Y., 1995.The subarachnoid space as a site for precursor T cellproliferation and effector T cell selection in experimentalautoimmune encephalomyelitis. J. Neuroimmunol. 56,171–178.

Shinohara, H., Inaguma, Y., Goto, S., Inagaki, T., Kato, K., 1993.Alpha B crystallin and HSP28 are enhanced in the cerebralcortex of patients with Alzheimer's disease. J. Neurol. Sci. 119,203–208.

Stahnke, T., Stadelmann, C., Netzler, A., Bruck, W.,Richter-Landsberg, C., 2007. Differential upregulation ofheme oxygenase-1 (HSP32) in glial cells after oxidativestress and in demyelinating disorders. J. Mol. Neurosci. 32,25–37.

van Noort, J.M., van Sechel, A.C., Bajramovic, J.J., el Ouagmiri,M., Polman, C.H., Lassmann, H., Ravid, R., 1995. The smallheat-shock protein alpha B-crystallin as candidate autoantigenin multiple sclerosis. Nature 375, 798–801.

Xanthoudakis, S., Nicholson, D.W., 2000. Heat-shock proteins asdeath determinants. Nat. Cell Biol. 2, E163–E165.