Experimental Neurology 247 (2013) 720–729

Contents lists available at SciVerse ScienceDirect

Experimental Neurology

j ourna l homepage: www.e lsev ie r .com/ locate /yexnr

Effects of hypothermia on oligodendrocyte precursor cell proliferation, differentiationand maturation following hypoxia ischemia in vivo and in vitro

Man Xiong a,b, Jin Li c, Si-Min Ma a, Yi Yang a,b,⁎, Wen-Hao Zhou a,⁎⁎a Key Laboratory of Neonatal Diseases, Ministry of Health, Fudan University, 138 Yi Xue Yuan Road, Shanghai 201102, Chinab Institute of Pediatrics, Children's Hospital, Fudan University, 399 Wanyuan Road, Shanghai 201102, Chinac Institutes of Biomedical Sciences, Fudan University, 138 Yi Xue Yuan Road, Shanghai 201102, China

⁎ Correspondence to: Y. Yang, Institute of PediatricsUniversity, 399 Wanyuan Road, Shanghai 201102, Ch⁎⁎ Correspondence to:W. Zhou, Key Laboratory of NeonaChildren's Hospital of Fudan University, 399Wanyuan RFax: +86 21 64931914.

E-mail addresses: yyang@shmu.edu.cn (Y. Yang), zh(W.-H. Zhou).

0014-4886/$ – see front matter © 2013 Elsevier Inc. Allhttp://dx.doi.org/10.1016/j.expneurol.2013.03.015

a b s t r a c t

a r t i c l e i n f o

Article history:Received 29 January 2013Revised 2 March 2013Accepted 14 March 2013Available online 22 March 2013

Keywords:HypothermiaHypoxic-ischemiaOligodendrocyte precursorMaturationImmature brain

Hypoxic-ischemia (HI) not only causes gray matter injury but also white matter injury, leading to severe neu-rological deficits and mortality, and only limited therapies exist. The white matter of animal models andhuman patients with HI-induced brain injury contains increased oligodendrocyte precursor cells (OPCs).However, little OPC can survive and mature to repair the injured white matter. Here, we test the effects ofmild hypothermia on OPC proliferation, differentiation and maturation. Animals suffered to left carotid arteryligation followed by 8% oxygen for 2 h in 7-day-old rats. They were divided into a hypothermic group (rectaltemperature 32–33 °C for 48 h) and a normothermic group (36–37 °C for 48 h), then animals were sacrificedat 3, 7, 14 and 42 days after HI surgery. Our results showed that hypothermia successfully enhanced early OLprogenitors (NG2+) and its proliferation in the corpus callosum (CC) after HI. Late OL progenitor (O4+) ac-cumulation decreased accompanied with increased OL maturation which is detected by myelin basic protein(MBP) and proteolipid protein. (PLP) immunostaining and immunoblotting in hypothermia compared tonormothermia. Additionally, using an in vitro hypoxic-ischemia model-oxygen glucose deprivation (OGD),we demonstrated that hypothermia decreased preOL accumulation and promoted OPC differentiation andmaturation. Further data indicated that OPC death was significantly suppressed by hypothermia in vitro.The myelinated axons and animal behavior both markedly increased in hypothermic- compared tonormothermic-animals after HI. In summary, these data suggest that hypothermia has the effects toprotect OPC and to promote OL maturation and myelin repair in hypoxic–ischemic events in the neonatalrat brain. This study proposed new aspects that may contribute to elucidate the mechanism of hypothermicneuroprotection for white matter injury in neonatal rat brain injury.

© 2013 Elsevier Inc. All rights reserved.

Introduction

Hypoxic–ischemic encephalopathy (HIE) remains one of the mostimportant neurological disorders with high risks of severe morbidityand death in full- and near-full-term newborns (Phelan et al.,2005). HIE causes neuronal apoptosis as well as diffuse primary dam-age to subcortical white matter in the term infant brain (Khwaja andVolpe, 2008; Martinez-Biarge et al., 2012; Okereafor et al., 2008).Such diffuse white matter injuries (WMI) in the newborn brain can

, Children's Hospital of Fudanina. Fax: +86 21 64931883.tal Diseases, Ministry of Health,oad, Shanghai 201102, China.

ou_wenhao@yahoo.com.cn

rights reserved.

result in cerebral palsy (CP), cognitive disability and neurologicaldysfunction.

WMI mainly correlated with damage to cells of the oligodendroglialineage via necrotic and apoptotic death, as could be demonstrated inrodents (Carloni et al., 2007; Rothstein and Levison, 2005; Segovia etal., 2008) and in brain tissue of preterm infants (Volpe, 2009). Oligoden-drocyte development is characterized by four major stages: NG2+ orA2B5+ early OL progenitors, O4+ late OLprogenitors (preOLs), O1+ im-mature OLs and mature OLs expressing MBP to myelinate axons (Back,2006). In rat brain, OLmaturation is almost completed at P7 in thewhitematter in contrast to P14 in the cortex and hippocampus (Dean et al.,2011). After neonatal injury, significant loss of oligodendrocyte anddemyelination was reported in the corpus callosum (Levison et al.,2001). Although oligodendroglia progenitors within the SVZ (Dizon etal., 2010) and CC (Zaidi et al., 2004) regenerated locally after hypoxia–ischemia injury, high vulnerability of premature OL to HI injury (Backet al., 2002) caused significant apoptosis of oligodendrocyte progenitorsand arrested preOL maturation in corpus callosum, which resulted incompromised OL production and failed regeneration in white matter

721M. Xiong et al. / Experimental Neurology 247 (2013) 720–729

(Buser et al., 2012; Levison et al., 2001; Ness et al., 2001; Segovia et al.,2008). In neonatal brain, factors that affect oligodendrocyte progenitorcell (OPC) survival and oligodendrocyte regeneration and maturationare poorly understood. Previous studies suggested pharmacologicalmeans of treatment such as stimulation of neurotrophic properties inthe neonatal brain can promote oligodendrocyte regeneration, but fur-ther studies should be directed to enhance endogenous repair while atthe same time avoiding adverse effects of the drugs used (Fan et al.,2010).

Mild hypothermia has been proven as an effective rescue therapyto improve outcomes in specific patient populations and animalmodels after hypoxic-ischemia (Akiko Ohmura et al., 2005; Marionand Bullock, 2009; Marks et al., 2010; Wilkinson et al., 2007; Xionget al., 2011; Xiong et al., 2009; Zhou et al., 2010). In clinics, therapeu-tic hypothermia is now the standard of care for brain injury control interm infants with perinatal HIE (Gancia and Pomero, 2012). However,the underlying mechanisms and the critical process are complex andmultifactorial. Previous investigations have demonstrated that hypo-thermia reduces immature oligodendrocyte loss in sheep brain(Bennet et al., 2007; George et al., 2011). Thus, we thought that hypo-thermia may regulate immature oligodendrocyte differentiation andmaturation in the corpus callosum in injured rat brain. In this study,we examined long-term effects of hypoxic–ischemic hypothermiaon OL lineage cell proliferation and maturation in vivo and in vitro.This study may provide a possible novel role for hypothermia onwhite matter injury in the neonatal brain after HI.

Material and methods

Animal models and tissue preparation

This study was conducted in accordance with the National Insti-tute of Health Guide for the Care and Use of Laboratory Animals.Sprague–Dawley rats were purchased from Shanghai ExperimentalAnimal Center of the Chinese Academy of Sciences. Neonatal HI wasinduced in rats on P7, and the mild hypothermia was induced 2 h re-covery after HI surgery. The detailed protocol was described previ-ously (Xiong et al., 2009, 2011). BrdU was injected (50 mg/kg)intraperitoneally three times at 4 h intervals before the day animalswere sacrificed, based on a previously described protocol (Tatsumiet al., 2005). Animals were sacrificed at 3, 7, 14 and 42 days after HIand perfused with 4% paraformaldehyde in 0.1 M PBS. The brainswere removed and immersion-fixed in the same solution at 4 °C for24 h, and dehydrated with a graded series of sucrose solutions (20%,30%) until sank. Coronal sections were cut on a freezing microtome(Jung Histocut, Model 820-II; Leica, Germany) at a thickness of30 μm at bregma level from 1.60 mm to −4.80 mm, and stored at−20 °C in cryoprotectant solution.

Cell culture and oxygen glucose deprivation

OPCswere cultured as previously described (Zhang et al., 1998). Pri-mary cultures were isolated from the cerebral cortex of SD rat embryos(E14). Briefly, cerebral cortex of rat brain was dissected under a micro-scope in ice cold PBS (Gibco, Carlsbad, CA, USA). The tissuewas triturat-ed and filtrated before centrifugation, cell suspensions were plated at adensity of 1 × 106 cells/ml with DMEM/F12medium containing 1% N2,2% B27 (all from Gibco, Carlsbad, CA, USA), EGF (R&D systems, USA;20 ng/ml) and bFGF (R&D systems, USA; 20 ng/ml). Five days later,cells were seeded into flasks coated with poly-ornithine (100 μg/ml,Sigma, Louis, MO, USA) using DMEM/F12 medium containing 1%N2,2%B27, bFGF (20 ng/ml) and PDGF-AA (R&D Systems, USA; 10 ng/ml).Cells were maintained at 37 °C in a CO2 incubator for 3 days followedby 2 h OGD as described previously (Yan et al., 2010). Then cells wererandomly divided into two groups: the normo- or hypothermic groupwhich was placed in a cell incubator at a stable temperature of 37 °C

or 33 °C for 48 h, respectively. Oligodendrocyte differentiation was in-duced by T3 (triiodothyronine, Sigma, USA; 40 ng/ml) and T4 (thyrox-ine, Shengong, China; 30 ng/ml) for 12 days (Baas et al., 1997).

Immunohistochemical staining

Sections at 1.20 to 0.20 mm from bregma were used for immu-nohistochemical staining. For rabbit anti-NG2 (Millipore, USA;1:200 dilutions), mouse anti-O4 (Millipore, USA; 1:200 dilutions)and mouse anti-MBP (Abcam, USA; 1:200 dilutions) immunohisto-chemical staining, sections were incubated with primary antibodiesovernight at 4 °C. Sections were then incubated with correspondingbiotinylated secondary antibodies and avidin–biotin-peroxidase(Vectastain Elite ABC kit; Vector Laboratories, Burlingame; 1:200dilution) for 45 min respectively at 37 °C. Immunoreactivity was vi-sualized with diaminobenzidine (DAB). For cell immunostaining,coverslip cultures were fixed by 4% paraformaldehyde in phosphate-buffered saline (PBS) for 15–20 min at RT. Cells were blocked with 5%normal sheep serum in 0.06% TritonX-100 for 30 min. Cells were incu-bated overnight at 4 °C with primary antibodies including mouseanti-A2B5 (R&D system, USA; 1:200), mouse anti-O4(1:200 dilutions),mouse anti-O1 (Millipore, USA; 1:200 dilutions) and mouse anti-MBP.Then, cells were incubated with fluorescent IgG-Rhodamine, IgG-FITC,IgG-Rhodamine and anti-mouse IgM-FITC (all from Invitrogen, USA;1:500 dilutions) at 37 °C for 45 min. After washing, sections weremounted on glass slides and cover slipped using fluorescencemountingmedium (Vector Laboratories). The fluorescent signals were detected atexcitation of 535 nmand 565 nm(Rhodamine), or 490 nmand525 nm(FITC) by microscopy (Leica DMRA2).

Electron microscopy

Rats were perfused with 2.5% glutaraldehyde in 4% paraformalde-hyde. The brain was removed, coronally sectioned to 1 μm includingthe corpus callosum formation at 1.70 mm to 3.40 mm posterior tobregma, and placed in fresh fixative (4.0% paraformaldehyde/2.5%glutaraldehyde in 0.1 mol/L PBS). The sections were further trimmedto include the corpus callosum. Samples were rinsed in the phosphatebuffer following a post-fixation for 2 h with 1% osmium tetroxide inphosphate buffer at 4 °C. After that they were dehydrated and em-bedded in resin. Ultrathin sections were prepared using a Reichertultra microtome, contrasted with uranyl acetate and lead citrate.Samples were examined under a Philips CM120 electron microscopeat 60 kV.

Immunoblotting

For immunoblotting analysis, rat corpus callosum which dissectedfrom brain tissues or cells was homogenized in RIPA buffer containing aprotease inhibitor cocktail. Equal amounts of total protein (50 μg forbrain tissue; 30 μg for cells) were separated on 12% SDS-polyacrylamidegels, and electrophoretically transferred onto nitrocellulose membranes.The membranes were incubated with rat anti-MBP (Chemicon,USA; 1:1000 dilutions) or mouse anti-PLP (Millipore, USA; 1:1000dilutions) at 4 °C overnight. After washing, membranes were incu-bated with horseradish-peroxidase-conjugated anti-rat or mouseIgG (1:5000 dilutions) and HRP-conjugated actin, the immunoreac-tivity was visualized by enhanced chemiluminescence (LAS-4000mini; FUJIFILM).

Open field test and rotating rod test

Animal behavior performed as previous paper described (Miki et al.,2009). HI + Normo, HI + Hypo, and sham animals were subjected tobehavioral tests at 42 days after surgery. In all experiments, the re-searcher conducting the behavioral testing and scoring was blind to

722 M. Xiong et al. / Experimental Neurology 247 (2013) 720–729

the experimental conditions. An open field test was performed to evalu-ate spontaneous locomotor activity. Animals were placed individually inan open field apparatus (43.2 cm × 43.2 cm × 30.5 cm; Med associatesInc., Georgia, VT, USA) and allowed to start from one of the four cornersselected randomly by the experimenter. Data were collected for 8 minafter a 2-min habituation period. All the movements of each animal inthe open field apparatus were traced and total distance traveled was an-alyzed with software activity monitor based on the NIH Image program.Rotating rod testingwas performed to evaluate sensorimotor skill acqui-sition. An accelerated rotating rod (Med Associates Inc., Georgia, VT,USA) test was conducted for 2 consecutive days: animals were placedon the rod (shaft diameter: 7 cm, lane width: 8.9 cm, fall height:26.7 cm) for three trials each day. Each trial lasted for a maximum of10 min. The rotating rod underwent linear acceleration from 4 to40 rpm over the first 5 min of the trial and then remained at maximumspeed for the remaining 5 min. Latency to step out of the rod (latency tostep out) was recorded.

Fig. 1. Immunohistochemical analysis of early OL progenitor proliferation and preOL actographs reveal immunopositive labeling of early OL progenitors by NG2 in each grouhypothermia compared to normothermia and sham controls after HI. (I–K) Showed g(L–O) The immunopositive staining of preOLs was detected by O4 in the CC region. (genitors in the hypothermia compared to normothermic animals and sham controls aenhanced in normothermia in contrast to hypothermia at all time points (3, 7, 14 aSham + Hypo: normothermia or hypothermia sham control; HI + Normo/HI + Hypo: normgroup).

Quantification analysis and statistics

For tissue immunostaining, three continuous sectionswere taken forquantification. Cells positive for NG2 or O4 were analyzed by a comput-erized Stereo Investigator 6.5 (MicroBrightField, Inc.Williston, VT, USA)(Xiong et al., 2008). Single immunostained cells were counted using theoptical-fractionator probe under 20× microscope lens (n = 8–10 ani-mals per group). For cultured cells, six images/section were acquiredrandomly under a 40× object lens, immunolabeled cells were countedmanually. Three independent experiments were conducted in culturedcells. For immunoblotting, analysis of the sumdensity of a given proteinband was performed using Image Pro Plus 5.1. software. All experi-mental results were initially analyzed with observers blinded tothe experimental conditions. Data were expressed as mean ± SEM.Differences between groups were analyzed by ANOVA followed byStudent-Newman-Keuls test. Statistical significance was set atP b 0.05.

cumulation in the CC at different time points after HI. (A–D) Representative pho-p. (E–H and Q) Indicated that BrdU incorporation was significantly enhanced byreat co-localization of BrdU with NG2 in the CC of hypothermic-treated animals.P) Time-course changes of NG2+ suggested significant increases of early OL pro-t 3 and 7 days after HI. (R) Statistical data indicated that the preOLs were greatlynd 42 days) after HI. Data are shown by bars as mean ± SEM. (Sham + Normo/othermia or hypothermia treatment group. *P b 0.05 vs HI + Hypo, n = 5–10 in each

723M. Xiong et al. / Experimental Neurology 247 (2013) 720–729

Results

Hypothermia enhanced early OL progenitor proliferation and suppressedpreOL accumulation after HI

To examine hypothermic effects on early OL progenitors, weperformed NG2 immunostaining in the corpus callosum. Fig. 1showed that there are more obvious NG2 positive cells in hypother-mic (C) than in normothermic animals (A). Statistical data (Fig. 1P)displayed that hypothermia significantly increased early OL progen-itors at the early time points (3 and 7 days) compared to normother-mia. Early OL progenitors showed a slight increase at 14 days butdeclined to similar level in hypothermia compared with normother-mia at 42 days after HI. In the hypothermic group, NG2+ cells in-creased greatly at 3 days, and gradually decreased from 7 to42 days after HI. However, in normothermia, NG2 positive cells didnot change obviously from 7 to 42 days after HI. We tested cell

Fig. 2. Effects of hypothermia on OL maturation in CC after HI. MBP immunostaining in thehypothermia and sham controls than in normothermia. (E–F)Western blot analysis of MBP (Densitometric quantification revealed greatly increased density of MBP and PLP in hypoth(*P b 0.05 vs HI + Hypo, n = 5–10 in each group).

proliferation by BrdU incorporation at different time points afterHI. Figs. 1E–H showed more BrdU immunoreactivity in hypothermiathan in normothermia and sham controls in the corpus callosum ofrat brain. Statistical data (Fig. 1Q) revealed that cell proliferation in-creased from 3 till 14 days, and then reduced to very low level at42 days in hypothermia and normothermia compared to sham con-trols. Significant differences were observed between hypothermiaand normothermia at 3, 7 and 14 days after HI. Early OL progenitorproliferation was examined by double immunofluorescent stainingof BrdU with NG2. Figs. 1I–K showed obvious co-existence of BrdUwith NG2 in the CC of hypothermic rat brain. There are about 65%NG2+ cells that were co-stained with BrdU in hypothermic animalscompared with 31% in normothermia in the CC at 3 days after HI.Late OL progenitors (PreOLs) were explored by O4 immunostainingin the CC region (Figs. 1L–O). PreOLs in hypothermia (Fig. 1N) weremorphologically distinct from normothermia (Fig. 1L) which accu-mulated as tightly-packed cluster of cells with few processes. Data

CC was shown in pictures A–D. The immunoreactivity of MBP was more prominent in17–22 kDa) and PLP (23–25 kDa) in CC lysates from rat brain at 42 days after HI. (G–H)ermia compared with normothermia which is shown graphically with mean ± SEM

724 M. Xiong et al. / Experimental Neurology 247 (2013) 720–729

analysis revealed that hypothermic treatment significantly reducedO4+ cells in contrast to normothermia at 3, 7, 14 and 42 days afterHI. In normothermic animals, O4+ cells increased at 3 days andreached to the highest level at 7 days, and gradually decreased at14 days after HI.

Hypothermia significantly promoted OL maturation in rat corpus callosumafter HI

Mature OLs were detected by MBP immunostaining combinedwith MBP and PLP immunoblotting (Fig. 2). Figs. 2A–D showedmore MBP immunopositive staining in the hypothermia and shamcontrols than in the normothermic group. To further confirm hypo-thermic effects on OL maturation after HI, we tested protein changesof MBP and PLP in the corpus callosum at 42 days after HI. Figs. 2E andG suggested that MBP (17–22 kDa) intensity markedly enhanced byhypothermia compared to normothermia. Consistent with MBP, PLP(23–25 kDa) which was also expressed in mature OL dramatically in-creased in hypothermia compared to normothermia at 42 days afterHI (Figs. 2F and H).

Hypothermia decreased preOL accumulation and promoted OL matura-tion in cultured OPC after OGD

We also evaluated hypothermic effects on OPC differentiationand maturation in vitro. As previous paper described (Xu et al.,2012; Zhang et al., 1998), the experimental protocol for OPC culturein this study was shown in Fig. 3D. Representative images ofneurospheres and OPC are shown in Figs. 3A and B respectively. Im-munocytochemical characterizations of OPCs are tested by A2B5 inFig. 3C. The cytotoxicity induced by 2 h-OGD in OPC was exploredby LDH release. Fig. 3E showed that normothermia significantly in-creased cytotoxicity after 2-h OGD compared to sham controls, andhypothermic-treatment reduced LDH release in cultured OPC at 48 hafter OGD. To obtain various maturational OL lineage cells, culturedOPC was allowed to differentiate for 0, 4, 8 or 12 days. Figs. 4A–P

Fig. 3. Cultured OL lineage cells and OGD model. Photograph A showed cultured neurospheferentiated from neurosphere. (C) Almost all of the cultured OPC were stained by A2B5. DLDH release. The statistical data suggested that hypothermia significantly reduced cell cytopendent experiments. *P b 0.05 compared with normothermia.

displayed the representative immunocytochemical staining of stage-specific markers: A2B5, O4, O1 or MBP at 8 days after OPC differentia-tion in each group respectively. To analyze the impact of hypothermiaon OL differentiation and maturation, cell composition of each stagewas analyzed by the ratio of stage-specific markers to total cells. Thestatistical data about cell composition of each stage were shown inFigs. 4Q–T. At 0 day after OPC differentiation, there are about 99%A2B5+ and 97% O4+ cells. The ratio of O1 andMBPwas very low in nor-mothermic- or hypothermic-animals and sham controls after 2 h-OGD(Fig. 4Q). At 4 days after differentiation, the ratio of A2B5+ and O4+

cells has a slight decline, and the ratio of O1+ andMBP+ has a small en-hancement compared to 0 day after OPC differentiation in each group.There are a little more A2B5+ (93%) and O4+ (95%) in normothermiathan in hypothermia (A2B5+: 90%; O4+: 89%), however, no significantdifferences were detected between normothermia and hypothermia(Fig. 4R). After 8 days differentiation (Fig. 4S), the ratio of A2B5+ andO4+ cells significantly decreased in hypothermic (A2B5+: 64%; O4+:67%) compared to normothermia (A2B5+: 72%; O4+: 86%) after OGD.Correspondingly, the ratio of O1 andMBP positive cells increased signif-icantly in hypothermia (O1: 18%; MBP: 16%) compared to normother-mia (O1: 8%; MBP: 9%). The ratio of A2B5+ and O4+ cells furtherdecreased in hypothermic animals (A2B5+: 37%; O4+: 42%), but it didnot change greatly in the normothermia (A2B5+: 67%; O4+: 79%) at12 days compared to 8 days after OPC differentiation (Fig. 4T). Howev-er, the ratio of O1+ and MBP+ cells was significantly elevated by hypo-thermia (O1: 28%; MBP: 33%) compared to normothermia (O1: 15%;MBP: 12%). This result indicated that hypothermia significantly reducedthe ratio of preOLs but dramatically elevated the ratio of mature OL at 8and 12 days after OPC differentiation in contrast to normothermia. Toconfirm the enhanced OPCmaturation in hypothermic-treated culturesafter OGD, we further examined protein changes of MBP bywestern-blotting (Fig. 5). In Fig. 5A we can see that there is almost noMBP expression in OL lineage cells at 0 day. Only very few MBP wereexpressed at 4 days after OPC differentiation in sham control animals(17–22 kD). MBP protein was detected in OGD-treated animals till8 days. Consistent with the results in Fig. 4, the statistical data

res from E14 rat brain. Picture B displayed oligodendrocyte precursor cells which dif-iagram D introduced protocols for OPC culture. (E) Cell cytotoxicity was detected bytoxicity at 48 h after 2 h-OGD. Data are shown by bars as mean ± SEM for three inde-

Fig. 4. Effects of hypothermia on OPC differentiation and maturation at different days after 2 h-OGD. (A–P) Immunocytochemical staining of stage-specific OL markers by A2B5, O4,O1, or MBP at 8 days after OPC differentiation in each group. (Q–T) Quantitative analysis of stage-specific OL composition at different days after OPC differentiation. The ratio ofA2B5+ and O4+ cells significantly reduced, but the ratio of O1+ and MBP+ cells greatly enhanced in hypothermia at 8 and 12 days after 2 h-OGD. Data are shown by bars asmean ± SEM for three independent experiments.*P b 0.05 compared with normothermia.

725M. Xiong et al. / Experimental Neurology 247 (2013) 720–729

Fig. 5. Western blot analysis of the effects of hypothermia on MBP expression in OL lineage cells at different days after OPC differentiation. (A) Representative photographs of MBPand actin bands at 0, 4, 8 and 12 days after OPC differentiation. Histograms B and C displayed that hypothermia enhanced MBP expression compared to normothermia at 8 and12 days after OPC differentiation respectively. Densitometric quantification is shown graphically with mean values ± SEM. Data are shown for three independent experiments.*P b 0.05 compared with normothermia.

726 M. Xiong et al. / Experimental Neurology 247 (2013) 720–729

(Figs. 5B and C) indicated that MBP protein level significantly increasedin hypothermia compared to normothermia at 8 and 12 days after OPCdifferentiation.

Hypothermic-treatment reduced OL death after HI in vitro

To explore the effects of hypothermia on cell death, cultured OL lin-eage cells were assessed by tunnel staining and CCK-8 Kit. Figs. 6A–Dshowed that there are obviously more tunnel positive staining in nor-mothermia compared to hypothermia and sham controls. The statisti-cal data suggested that hypothermia significantly increased OLlineage survival (Fig. 6E) and reduced cell death (Fig. 6F) comparedto normothermia at 0, 4, 8 and 12 days after OPC differentiation. Celldeath in normothermia maintained at higher level at 0 and 4 days,and gradually reduced at 8 and 12 days. Hypothermia successfullysuppressed cell death after 2 h-OGD at each time point after OPCdifferentiation.

Myelinated axons and animal behavior were elevated by hypothermia inrat after HI

To elucidate if the increased mature OLs would enhance myelin re-pair, we performed scanning electron microscopy. Figs. 7A–D showedmoremyelinated axons in hypothermic and shamcontrols thannormo-thermia, and significant differences were detected between hypother-mia and normothermia in the corpus callosum of rat brain (Fig. 7E).Animal motor function was examined by Open field and rotarod test.Open field results showed a significant difference between normother-mia and hypothermia in the traveled distance: normothermiapresented a greatly lower traveled distance than all other groups(Fig. 7F). A similar result was found in the latency to fall off the cylinderin the rotarod test (Fig. 7G). Paired sampled T-test showed that ratsfrom sham control groups had significant greater latencies to fall off inthis test compared to HI injured animals. And hypothermic treatmentsignificantly enhanced their performance compared to normothermicanimals.

Discussion

Therapy to prevent OL maturation and myelination failure inwhite matter injury is limited because of little information aboutthe cellular mechanisms related to this form of brain injury. In new-born brain, the oligodendrocyte precursor cells are the importantcomponent of white matter. Considerable clinical and experimentaldata have demonstrated that OL series are maturation-dependentvulnerability to HI injury (Back et al., 2001, 2002). Excitotoxicity, ox-idative stress, inflammation and apoptosis are some of the mecha-nisms involved in the vulnerability of OLs to HI in immature brain(Volpe et al., 2011). Treatments aiming at the above mechanismsmight provide protection to OLs.

In this research, hypothermia significantly promoted early OL pro-genitor proliferation and reduced preOL accumulation in the CC(Fig. 1) at different time points after HI. Hypothermia may promoteOL progenitor proliferation by initiating cell cycle proteins (Imada etal., 2010). These new generated OL progenitors are mainly derivedfrom local OPCs where generating within and surrounding the infarctafter injury (Zaidi et al., 2004; Zawadzka et al., 2010), that would pro-vide more precursor cells for expansion and differentiation into ma-ture OLs in hypothermia-treated animals. Nevertheless, the preOLswere greatly reduced in hypothermia compared with normothermiain CC after HI (Fig. 1). Previous paper displayed a combination of pro-liferative, degenerative and maturational processes that resulted in anet expansion in the pool of preOLs (Buser et al., 2012; Riddle et al.,2011; Segovia et al., 2008). Our results in vitro (Figs. 1 and 2) andin vivo (Figs. 4 and 5) both confirmed obstructed maturation ofpreOLs in normothermia after hypoxic–ischemic injury. The arrestedPreOL maturation (Segovia et al., 2008) led to a further elevatedpreOLs in normothermia. Hypothermic treatment successfully re-duced preOL accumulation and promoted OL maturation in CC afterHI. And this result was also found in cultured OL lineage cells thatmore preOLs (O4+) differentiated into immature (O1+) and matureOL (MBP+) in hypothermia compared to normothermia after2 h-OGD (Figs. 4 and 5). So, the enhanced differentiation of preOLs inhypothermia may be the main reason why there are less preOLs in

Fig. 6.Hypothermia inhibits OL lineage cell death after 2 h-OGD. (A–D) Tunnel staining in primary cultures of OPCs. There are more tunnel+ cells in the normothermia than in hypother-mia and sham controls. (E) Statistical data showed the ratio of cell survival significantly increased in hypothermic-treatment at all time points after OPC differentiation. (F) Quantitativeanalysis of the ratio of tunnel+ to total cells indicated dramatically reduced cell death in hypothermia compared with normothermia at 0, 4, 8 and 12 days after OPC differentiation. Dataare shown by bars for three independent experiments. *P b 0.05 compared with normothermia.

727M. Xiong et al. / Experimental Neurology 247 (2013) 720–729

hypothermia than in normothermia after brain injury. More important-ly, the myelin repair and animal motor behavior were greatly enhancedby hypothermia after HI (Fig. 7). Themechanismof hypothermic axonalprotection was proposed to be related to the reduction in brain meta-bolic rate, blockade of excitotoxicity, calcium antagonism, and modula-tion of the inflammatory response in previous paper (Sahuquillo andVilalta, 2007). Our research firstly revealed that hypothermia enhancedmature oligodendrocytes which could myelinate axons and rescue ani-mal behavior. Nevertheless, the underlying mechanism of hypothermicprotection in OL maturation and myelination after HI still needs to bestudied in future.

In previous study, preOLs were found time-dependently correlatedand highly co-localized with active caspase-3 in normothermia but notin hypothermia suggested decreased preOL degeneration after hypother-mic treatment in rat brain. Actually, in culturedOPC, 2 h-OGD significant-ly increased the ratio of tunnel+ cells at 0 day and peaked up at 4 days innormothermia compared to hypothermia (Fig. 7), these trends are coin-cident with the highest ratio of preOLs in OL lineage (0 and 4 days)after OPC differentiation (Fig. 6). These data further indicated that hypo-thermia reduced delayed preOL degeneration after hypoxic injury. Takentogether, the suppressed degeneration of preOLs in hypothermia maycontribute to enhanced OL maturation in hypothermia. The reasons fordelayed preOL degeneration in normothermia are unknown, but it

may be related to the loss of growth factors such as neurotrophinsand brain-derived neurotrophic factor (BDNF) (Francis et al., 2006),which are critical for preOL survival after HI. Hypothermia increasesneurotrophins and BDNF following cardiac arrest and ischemia (Boris-Moller et al., 1998; D'Cruz et al., 2002), which helps to explain thereduced preOL degeneration and accumulation in hypothermic animalsin our study. Secondly, recent studies have indicated that preOLsfailed to differentiate surrounding astroglial scars in chronicwhitematterlesions (Buser et al., 2012). It was reported that hyaluronan or interferon-gammawhich released by reactive astrocyte arrested OPCmaturation byactivating TLR2 receptors or interferon-gamma receptors on preOLs(Back et al., 2005; Folkerth et al., 2004; Sloane et al., 2010). White matterastrocytes are much more sensitive to ischemic-reperfusion injury thanare gray matter astrocytes, a feature that may form more glia scars inthe CC after HI (Shannon et al., 2007). Glial scar formation in the CCmay block OL maturation in normothermia after HI injury. Our (Xionget al., 2009) and other (Gresle et al., 2006) studies have shown that hypo-thermia decreases astrocyte activation and glial scar formation after braininjury. So, the increasedmaturation of OLs in the hypothermic groupmayassociate with inhibited activation of astrocytes in the CC after hypoxic–ischemic injury. Thirdly, OPC differentiation and maturation weresuppressed by inflammation (Fancy et al., 2011, Volpe et al., 2011).Local and systemic inflammation including microglia activation is long

Fig. 7. Long-term effects of hypothermia on myelin repair and animal behavior in rat after HI. (A–D) Electron microscope examination of myelinated axons in the CC. (E) Statisticaldata revealed that the myelinated axons dramatically enhanced by hypothermia compared to normothermia. (F) Open field test showed that the total distances traveled by ratswere greatly elevated by hypothermia. (G) Rotarod test suggested the latency to fall was significantly restored by hypothermia after brain injury. Data are analyzed by means ofthree trials. *P b 0.05 compared with normothermia. (n = 8–10 in each group).

728 M. Xiong et al. / Experimental Neurology 247 (2013) 720–729

lasting in the neonatal brain after HI (Winerdal et al., 2012). TNF-alphafrom microglia likely played a critical potentiating role in mediatingpreOL death in white matter injury. Hypothermia shifts the balance ofcytokine release towards the anti-inflammatory cytokines in stimulatedmicroglial cells (Diestel et al., 2010) and reduced the release of TNF-alpha and IL-6 after brain injury (Xiong et al., 2009). All these findingsindicated that hypothermia treatment may stimulate OL maturation viasuppressing inflammation after HI.

Concluding, we consider that hypothermia promotes OPC regenera-tion and maturation and rescues animal behavior associated with de-creased preOL accumulation and degeneration in vivo and in vitroafter hypoxic–ischemic injury. Nevertheless, future studies are neededto determine the pivotal factors regulating OL maturation and myelinrepair after HI in hypothermic condition.

Acknowledgment

This study was supported by a grant from National Natural ScienceFoundation of China (no. 81000519).

References

Akiko Ohmura, W.N., Ishida*, Akira, Yasuoka, Noriko, Kawamura, Masanari, ShinobuMiura, G.T., 2005. Prolonged hypothermia protects neonatal rat brain againsthypoxic-ischemia by reducing both apoptosis and necrosis. Brain Dev. 27, 517–526.

Baas, D., Bourbeau, D., Sarlieve, L.L., Ittel, M.E., Dussault, J.H., Puymirat, J., 1997. Oligo-dendrocyte maturation and progenitor cell proliferation are independently regu-lated by thyroid hormone. Glia 19, 324–332.

Back, S.A., 2006. Perinatal white matter injury: the changing spectrum of pathologyand emerging insights into pathogenetic mechanisms. Ment. Retard. Dev. Disabil.Res. Rev. 12, 129–140.

Back, S.A., Luo, N.L., Borenstein, N.S., Levine, J.M., Volpe, J.J., Kinney, H.C., 2001.Late oligodendrocyte progenitors coincide with the developmental window

of vulnerability for human perinatal white matter injury. J. Neurosci. 21,1302–1312.

Back, S.A., Han, B.H., Luo, N.L., Chricton, C.A., Xanthoudakis, S., Tam, J., Arvin, K.L.,Holtzman, D.M., 2002. Selective vulnerability of late oligodendrocyte progenitorsto hypoxia–ischemia. J. Neurosci. 22, 455–463.

Back, S.A., Tuohy, T.M., Chen, H., Wallingford, N., Craig, A., Struve, J., Luo, N.L., Banine, F.,Liu, Y., Chang, A., Trapp, B.D., Bebo Jr., B.F., Rao, M.S., Sherman, L.S., 2005. Hyaluronanaccumulates in demyelinated lesions and inhibits oligodendrocyte progenitor matu-ration. Nat. Med. 11, 966–972.

Bennet, L., Roelfsema, V., George, S., Dean, J.M., Emerald, B.S., Gunn, A.J., 2007. The ef-fect of cerebral hypothermia on white and grey matter injury induced by severehypoxia in preterm fetal sheep. J. Physiol. 578, 491–506.

Boris-Moller, F., Kamme, F., Wieloch, T., 1998. The effect of hypothermia on the expres-sion of neurotrophin mRNA in the hippocampus following transient cerebral ische-mia in the rat. Brain Res. Mol. Brain Res. 63, 163–173.

Buser, J.R., Maire, J., Riddle, A., Gong, X., Nguyen, T., Nelson, K., Luo, N.L., Ren, J., Struve,J., Sherman, L.S., Miller, S.P., Chau, V., Hendson, G., Ballabh, P., Grafe, M.R., Back, S.A.,2012. Arrested preoligodendrocyte maturation contributes to myelination failurein premature infants. Ann. Neurol. 71, 93–109.

Carloni, S., Carnevali, A., Cimino, M., Balduini, W., 2007. Extended role of necrotic celldeath after hypoxia–ischemia-induced neurodegeneration in the neonatal rat.Neurobiol. Dis. 27, 354–361.

D'Cruz, B.J., Fertig, K.C., Filiano, A.J., Hicks, S.D., DeFranco, D.B., Callaway, C.W., 2002. Hy-pothermic reperfusion after cardiac arrest augments brain-derived neurotrophic fac-tor activation. J. Cereb. Blood Flow Metab. 22, 843–851.

Dean, J.M., Moravec, M.D., Grafe, M., Abend, N., Ren, J., Gong, X., Volpe, J.J., Jensen, F.E.,Hohimer, A.R., Back, S.A., 2011. Strain-specific differences in perinatal rodent oligo-dendrocyte lineage progression and its correlation with human. Dev. Neurosci. 33,251–260.

Diestel, A., Troeller, S., Billecke, N., Sauer, I.M., Berger, F., Schmitt, K.R., 2010. Mecha-nisms of hypothermia-induced cell protection mediated by microglial cells invitro. Eur. J. Neurosci. 31, 779–787.

Dizon, M., Szele, F., Kessler, J.A., 2010. Hypoxia-ischemia induces an endogenous repar-ative response by local neural progenitors in the postnatal mouse telencephalon.Dev. Neurosci. 32, 173–183.

Fan, X., Kavelaars, A., Heijnen, C.J., Groenendaal, F., van Bel, F., 2010. Pharmacologicalneuroprotection after perinatal hypoxic-ischemic brain injury. Curr. Neuropharmacol.8, 324–334.

Fancy, S.P., Chan, J.R., Baranzini, S.E., Franklin, R.J., Rowitch, D.H., 2011.Myelin regeneration:a recapitulation of development? Annu. Rev. Neurosci. 34, 21–43.

729M. Xiong et al. / Experimental Neurology 247 (2013) 720–729

Folkerth, R.D., Keefe, R.J., Haynes, R.L., Trachtenberg, F.L., Volpe, J.J., Kinney, H.C., 2004.Interferon-gamma expression in periventricular leukomalacia in the human brain.Brain Pathol. 14, 265–274.

Francis, J.S., Olariu, A., McPhee, S.W., Leone, P., 2006. Novel role for aspartoacylase inregulation of BDNF and timing of postnatal oligodendrogenesis. J. Neurosci. Res.84, 151–169.

Gancia, P., Pomero, G., 2012. Therapeutic hypothermia in the prevention of hypoxic-ischaemic encephalopathy: new categories to be enrolled. J. Matern-Fetal NeonatalMed. 25 (Suppl. 4), 94–96.

George, S., Bennet, L., Weaver-Mikaere, L., Fraser, M., Bouwmans, J., Mathai, S., Skinner,S.J., Gunn, A.J., 2011. White matter protection with insulin-like growth factor 1 andhypothermia is not additive after severe reversible cerebral ischemia in term fetalsheep. Dev. Neurosci. 33, 280–287.

Gresle, M.M., Jarrott, B., Jones, N.M., Callaway, J.K., 2006. Injury to axons and oligodendro-cytes following endothelin-1-induced middle cerebral artery occlusion in consciousrats. Brain Res. 1110, 13–22.

Imada, S., Yamamoto, M., Tanaka, K., Seiwa, C., Watanabe, K., Kamei, Y., Kozuma, S.,Taketani, Y., Asou, H., 2010. Hypothermia-induced increase of oligodendrocyteprecursor cells: Possible involvement of plasmalemmal voltage-dependent anionchannel 1. J. Neurosci. Res. 88, 3457–3466.

Khwaja, O., Volpe, J.J., 2008. Pathogenesis of cerebral white matter injury of prematurity.Arch. Dis. Child. Fetal Neonatal Ed. 93, F153–F161.

Levison, S.W., Rothstein, R.P., Romanko, M.J., Snyder, M.J., Meyers, R.L., Vannucci, S.J.,2001. Hypoxia/ischemia depletes the rat perinatal subventricular zone of oligoden-drocyte progenitors and neural stem cells. Dev. Neurosci. 23, 234–247.

Marion, D., Bullock, M.R., 2009. Current and future role of therapeutic hypothermia.J. Neurotrauma 26, 455–467.

Marks, K., Shany, E., Shelef, I., Golan, A., Zmora, E., 2010. Hypothermia: a neuroprotectivetherapy for neonatal hypoxic ischemic encephalopathy. Isr. Med. Assoc. J. 12, 494–500.

Martinez-Biarge, M., Bregant, T., Wusthoff, C.J., Chew, A.T., Diez-Sebastian, J., Rutherford,M.A., Cowan, F.M., 2012. White matter and cortical injury in hypoxic–ischemicencephalopathy: antecedent factors and 2-year outcome. J. Pediatr. 161, 799–807.

Miki, K., Ishibashi, S., Sun, L., Xu, H., Ohashi, W., Kuroiwa, T., Mizusawa, H., 2009. Inten-sity of chronic cerebral hypoperfusion determines white/gray matter injury andcognitive/motor dysfunction in mice. J. Neurosci. Res. 87, 1270–1281.

Ness, J.K., Romanko, M.J., Rothstein, R.P., Wood, T.L., Levison, S.W., 2001. Perinatal hyp-oxia–ischemia induces apoptotic and excitotoxic death of periventricular whitematter oligodendrocyte progenitors. Dev. Neurosci. 23, 203–208.

Okereafor, A., Allsop, J., Counsell, S.J., Fitzpatrick, J., Azzopardi, D., Rutherford, M.A.,Cowan, F.M., 2008. Patterns of brain injury in neonates exposed to perinatal senti-nel events. Pediatrics 121, 906–914.

Phelan, J.P., Martin, G.I., Korst, L.M., 2005. Birth asphyxia and cerebral palsy. Clin.Perinatol. 32, 61–76 (vi.).

Riddle, A., Dean, J., Buser, J.R., Gong, X., Maire, J., Chen, K., Ahmad, T., Cai, V., Nguyen, T.,Kroenke, C.D., Hohimer, A.R., Back, S.A., 2011. Histopathological correlates of magneticresonance imaging-defined chronic perinatal white matter injury. Ann. Neurol. 70,493–507.

Rothstein, R.P., Levison, S.W., 2005. Gray matter oligodendrocyte progenitors and neu-rons die caspase-3 mediated deaths subsequent to mild perinatal hypoxic/ischemicinsults. Dev. Neurosci. 27, 149–159.

Sahuquillo, J., Vilalta, A., 2007. Cooling the injured brain: how does moderate hypo-thermia influence the pathophysiology of traumatic brain injury. Curr. Pharm.Des. 13, 2310–2322.

Segovia, K.N., McClure, M., Moravec, M., Luo, N.L., Wan, Y., Gong, X., Riddle, A.,Craig, A., Struve, J., Sherman, L.S., Back, S.A., 2008. Arrested oligodendrocyte lin-eage maturation in chronic perinatal white matter injury. Ann. Neurol. 63,520–530.

Shannon, C., Salter, M., Fern, R., 2007. GFP imaging of live astrocytes: regional differ-ences in the effects of ischaemia upon astrocytes. J. Anat. 210, 684–692.

Sloane, J.A., Batt, C., Ma, Y., Harris, Z.M., Trapp, B., Vartanian, T., 2010. Hyaluronanblocks oligodendrocyte progenitor maturation and remyelination through TLR2.Proc. Natl. Acad. Sci. U. S. A. 107, 11555–11560.

Tatsumi, K., Haga, S., Matsuyoshi, H., Inoue, M., Manabe, T., Makinodan, M., Wanaka, A.,2005. Characterization of cells with proliferative activity after a brain injury.Neurochem. Int. 46, 381–389.

Volpe, J.J., 2009. Brain injury in premature infants: a complex amalgam of destructiveand developmental disturbances. Lancet Neurol. 8, 110–124.

Volpe, J.J., Kinney, H.C., Jensen, F.E., Rosenberg, P.A., 2011. Reprint of “The developingoligodendrocyte: key cellular target in brain injury in the premature infant”. Int.J. Dev. Neurosci. 29, 565–582.

Wilkinson, D.J., Casalaz, D., Watkins, A., Andersen, C.C., Duke, T., 2007. Hypothermia: aneuroprotective therapy for neonatal hypoxic–ischemic encephalopathy. Pediatrics119, 422–423.

Winerdal, M., Winerdal, M.E., Kinn, J., Urmaliya, V., Winqvist, O., Aden, U., 2012. Long last-ing local and systemic inflammation after cerebral hypoxic ischemia in newbornmice. PLoS One 7, e36422.

Xiong, M., Zhang, T., Zhang, L.M., Lu, S.D., Huang, Y.L., Sun, F.Y., 2008. Caspase inhibitionattenuates accumulation of beta-amyloid by reducing beta-secretase productionand activity in rat brains after stroke. Neurobiol. Dis. 32, 433–441.

Xiong, M., Yang, Y., Chen, G.Q., Zhou, W.H., 2009. Post-ischemic hypothermia for 24 hin P7 rats rescues hippocampal neuron: association with decreased astrocyte acti-vation and inflammatory cytokine expression. Brain Res. Bull. 79, 351–357.

Xiong, M., Cheng, G.Q., Ma, S.M., Yang, Y., Shao, X.M., Zhou, W.H., 2011. Post-ischemichypothermia promotes generation of neural cells and reduces apoptosis by Bcl-2in the striatum of neonatal rat brain. Neurochem. Int. 58, 625–633.

Xu, C., Lv, L., Zheng, G., Li, B., Gao, L., Sun, Y., 2012. Neuregulin1beta1 protects oli-godendrocyte progenitor cells from oxygen glucose deprivation injury inducedapoptosis via ErbB4-dependent activation of PI3-kinase/Akt. Brain Res. 1467,104–112.

Yan, H., Zhou, W., Wei, L., Zhong, F., Yang, Y., 2010. Proteomic analysis of astrocytic se-cretion that regulates neurogenesis using quantitative amine-specific isobaric tag-ging. Biochem. Biophys. Res. Commun. 391, 1187–1191.

Zaidi, A.U., Bessert, D.A., Ong, J.E., Xu, H., Barks, J.D., Silverstein, F.S., Skoff, R.P., 2004.New oligodendrocytes are generated after neonatal hypoxic-ischemic brain injuryin rodents. Glia 46, 380–390.

Zawadzka, M., Rivers, L.E., Fancy, S.P., Zhao, C., Tripathi, R., Jamen, F., Young, K.,Goncharevich, A., Pohl, H., Rizzi, M., Rowitch, D.H., Kessaris, N., Suter, U.,Richardson, W.D., Franklin, R.J., 2010. CNS-resident glial progenitor/stem cells pro-duce Schwann cells as well as oligodendrocytes during repair of CNS demyelin-ation. Cell Stem Cell 6, 578–590.

Zhang, S.C., Lundberg, C., Lipsitz, D., O'Connor, L.T., Duncan, I.D., 1998. Generation of ol-igodendroglial progenitors from neural stem cells. J. Neurocytol. 27, 475–489.

Zhou, W.H., Cheng, G.Q., Shao, X.M., Liu, X.Z., Shan, R.B., Zhuang, D.Y., Zhou, C.L., Du, L.Z.,Cao, Y., Yang, Q., Wang, L.S., 2010. Selective head cooling with mild systemic hypo-thermia after neonatal hypoxic–ischemic encephalopathy: a multicenter random-ized controlled trial in China. J. Pediatr. 157, 367–372 (372 e361-363.).