B R A I N R E S E A R C H 1 4 5 9 ( 2 0 1 2 ) 6 1 – 7 0

Ava i l ab l e on l i ne 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

Neuroprotection of early and short-time applying berberine inthe acute phase of cerebral ischemia: Up-regulated pAkt, pGSKand pCREB, down-regulated NF-κB expression, amelioratedBBB permeability

Xiaolin Zhanga, Xiangjian Zhanga, b, c,⁎, Chaohui Wanga, Yanhua Lia, Lipeng Donga,Lili Cuia, Lina Wanga, Zongjie Liua, Huimin Qiaoa, Chunhua Zhua, Yinxue Xinga,Xiaoyun Caoa, Ye Jia, Kang Zhaoa

aDepartment of Neurology, Second Hospital of Hebei Medical University, Shijiazhuang, Hebei, 050000, ChinabHebei Institute of Cardio-Cerebral Vascular Diseases, Shijiazhuang, Hebei, ChinacHebei Key Laboratory for Neurology, Shijiazhuang, Hebei, China

A R T I C L E I N F O

⁎ Corresponding author at:Department of Neu050000, China. Fax: +86 031166002915.

E-mail address: zhang6xj@yahoo.com.cn

0006-8993/$ – see front matter © 2012 Elseviedoi:10.1016/j.brainres.2012.03.065

A B S T R A C T

Article history:Accepted 28 March 2012Available online 4 April 2012

Background: Berberine (BBR) has gained attention for its vast beneficial biological effectsthrough immunomodulation, anti-inflammatory and anti-apoptosis properties. Inflamma-tory and apoptosis damage play an important role in cerebral ischemic pathogenesis andmay represent a target for treatment. The aim of this study was to explore BBR's effect inischemic injury and the role of the Akt/GSK (glycogen synthase kinase) signaling cascade inmediating the anti-apoptosis and anti-inflammatory effects in the rat brain of permanentmiddle cerebral artery occlusion (pMCAO). Male Sprague–Dawley rats were subjected topMCAO and randomly assigned into four groups: Sham (sham-operated) group, pMCAO(pMCAO+0.9% saline) group, BBR-L (pMCAO+BBR 10 mg/kg) and BBR-H (pMCAO+BBR40mg/kg) group. BBR was administered immediately after pMCAO and the neuroprotectionwas detected. Phospho-Akt (pAkt), phospho-glycogen synthase kinase 3-β (pGSK3β), phospho-cAMP response element binding protein (pCREB), nuclear factor-kappa B (NF-κB) and claudin-5in ischemic cerebral cortex were detected by immunohistochemistry, reverse transcription-polymerase chain reaction and western blotting. Compared with pMCAO group, BBRdramatically lessened neurological deficits scores, brain water contents and infarct sizes,upregulated the expression of pAkt, pGSK3β, pCREB and claudin-5, and decreased the nuclearaccumulation of NF-κB (P<0.05) in ischemic brain. The results showed that BBR reducedischemic brain injury after pMACO, and this effect may be via the increasing the activation ofAkt/GSK signaling and claudin-5, and decreasing NF-κB expression.

© 2012 Elsevier B.V. All rights reserved.

Keywords:Middle cerebral artery occlusionNeuroprotectionBerberineAkt/GSKCREBNF-κB

rology, Second Hospital of Hebei Medical University, 215 Hepingxi Road, Shijiazhuang, Hebei

(X. Zhang).

r B.V. All rights reserved.

H3CO

OCH3

N+

o

o

Fig. 1 – The chemical structure of berberine.

62 B R A I N R E S E A R C H 1 4 5 9 ( 2 0 1 2 ) 6 1 – 7 0

1. Introduction

Stroke is the most frequent cause of permanent disability inadults and the third leading cause of death worldwide (Donnanet al., 2008). Despite considerable advances in the understand-ing of the pathophysiology of acute cerebral ischemia (Lo et al.,2003), therapeutic options for acute ischemia are still limited.Berberine (BBR), an alkaloid derivative from Berberis vulgaris L,has been used extensively in traditional Chinese medicine totreat diarrhea and diabetes. Previous studies have reported thatBBR has many beneficial biological effects, including immuno-modulation (Kong et al., 2004; Zha et al., 2010), anti-diabeticmetabolic effect (Bhutada et al., 2011), anti-inflammatory effect(Kuo et al., 2004), and anti-tumor activity (Tan et al., 2011).Furthermore, studies have shown that BBR exerts a protectiveeffect on several models of cerebral ischemic injury and myo-cardium injury induced by ischemia/reperfusion (Wang et al.,2004; Zhou et al., 2008). However, the functional role of BBR inacute phase of cerebral ischemia is not clear so far.

After ischemia, depending on the severity of the insult andthe ability of the cell to maintain adenosine triphosphate (ATP)synthesis, there are mainly two forms of neuron death, necrosisand apoptosis (Christophe and Nicolas, 2006). The serine–threonine kinase, Akt, which is also known as protein kinase B,plays an important role in the cell death/survival pathway. Aktphosphorylation is neuroprotective against ischemic injury(Mangi et al., 2003). Activated Akt phosphorylates several down-stream targets of the PI(3)K pathway, including the constitutivelyactive serine–threonine kinase glycogen synthase kinase 3-β(GSK3β) (Ser9). GSK3β is involved in fundamental cellular pro-cesses that can determine cell death (Frame and Cohen, 2001).GSK3β is particularly abundant in the central nervous systemand is neuron-specific (Leroy and Brion, 1999). Akt phosphory-lates GSK3β on Ser9 to render it inactive, a proposedmechanismby which neurons become resistant to apoptotic stimuli (Bijurand Jope, 2000; Pap and Cooper, 1998). GSK3β has also beenreported to be involved in ischemic brain injury (Kelly et al.,2004). GSK3β inhibition augments the binding of cAMP responseelement binding protein (CREB) (Ser133) and suppresses thebinding of nuclear factor-kappa B (NF-κB) p65 (Ser276) to thenuclear co-activator CREB binding protein (CBP) (Martin et al.,2005). GSK3β regulated the inflammatory response by differen-tially affecting the nuclear amounts of transcription factors NF-κB subunit p65 (Martin et al., 2005). The nuclear transcriptionfactor CREB is known to regulate the expression of a largenumber of genes in a variety of physiological contexts, such asproliferation, differentiation, growth, survival and adaptation ofall types of cells. Since 1990s, phosphorylated CREB (pCREB),activated form of CREB, has been examined in several ischemiamodels. Cerebral ischemia caused a burst of CREB phosphoryla-tion and the corresponding expression of cAMP-responsiveelement (CRE)-targeted genes which encode neuroprotectivemolecules, such as the protein B cell lymphoma/lewkmia-2 andthe brain-derived neurotrophic factor in neuron. CREB is knownto play important roles in synaptic plasticity, neurogenesis andaxon growth after ischemia (Walton and Dragunow, 2000).Ample evidences have suggested that Akt/GSK signaling path-way plays a central role in physical and pathological conditionsto differently regulate inflammation factors as well as boostsurvival (Collino et al., 2009; Endo et al., 2006).

The present project is to characterize the profound impactof BBR on outcome after acute ischemic stroke, and identify itstargets and protective signaling pathways in permanentmiddle cerebral artery occlusion (pMCAO) brain (Fig. 1).

2. Results

2.1. BBR reduced the brain infarction volume

No infarction was observed in the sham-operated group. InpMCAO group, an extensive lesion was developed in lateralcortex (Fig. 2A). The infarct volume was significantly reducedfrom 44.17±7.44% in pMCAO group to 34.73±2.63% in the BBR-Lgroup (P<0.05) and 30.93±6.66% in the BBR-H group (P<0.01)(Figs. 2A and C) (n=6 in each group).

2.2. BBR attenuated the neurological deficits

Neurological deficitwas examined and scored ona 5-point scaleand for statistical analysis the Mann–Whitney U-tests wereconducted. The neurological deficit scores were significantlyhigher in pMCAO group (3.50±0.89) compared with Shamcontrol. Berberine decreased these scores in BBR-L (3.13±0.68)and BBR-H (2.88±0.85) groups (10 mg/kg vs. pMCAO group,P<0.05; 40mg/kg vs. pMCAOgroup, P<0.01; Fig. 2B). And there isno significant difference between BBR-L group and BBR-H group(10 mg/kg vs. 40mg/kg, P>0.05).

2.3. BBR decreased the brain water content

Ipsilateral brain water content of Sham group was 78.59±1.31%.ComparedwithpMCAOcontrol, BBR-LandBBR-Hgroups showedan intense decline in the percentage of brain water content(pMCAO group vs. BBR-L group: 84.35±1.69% vs. 82.44±1.83%,P<0.05; pMCAO group vs. BBR-H group: 84.35±1.69% vs. 81.84±1.91%, P<0.05) (Fig. 2D) (n=6 in each group).

2.4. BBRupregulated the expression of pAkt, pGSK3β, pCREBand claudin-5 and downregulated the expression of NF-κB (p65)

The expression of pAkt, pGSK3β, pCREB and claudin-5 wasupregulated at protein level after systemic administration ofBBR in ischemia brain. The localization of pAkt, pGSK3β, pCREB

pMCAO BBR-L BBR-H

pMCAO BBR-L BBR-H

Infa

rct

volu

me

0

0.1

0.2

0.3

0.4

0.5

0.6

Cor

rect

ed %

infa

rct a

rea

C

A

0.74

0.76

0.78

0.8

0.82

0.84

0.86

0.88contralateral

ipsilateral

Bra

in w

ater

con

tent

D

0

1

2

3

4

5

Sham pMCAO BBR-L BBR-H

Sham pMCAO BBR-L BBR-H

Neu

rolo

gica

l def

icit

scor

es

B

Fig. 2 – Effects of BBR administration on ischemic rat brains. (A) BBR reduced the infarct volume in the brain. Brain sampleswere obtained from neocortical tissues of pMCAO group, BBR-L group and BBR-H group. TTC stain: comparison of infarct sizebetween the pMCAO group and BBR treatment group at 24 h. The pale region was the infarct brain tissue and the red regionwas normal. (C) BBR reduced %HLV. The %HLV were significantly lower in BBR-L group (※P<0.05) and BBR-H group (★P<0.01)than in pMCAO group. Results are means±SD (n=6) of each group. One-way ANOVA followed by the LSD test. (B) BBRattenuated neurological deficit. Brain samples obtained from neocortical infarction tissues of Sham group, pMCAO group,BBR-L group and BBR-H group. The behavioral scores were significantly reduced in BBR-L group (※P<0.05) and BBR-H group(★P<0.01) vs. pMCAO group. Mann–Whitney test were used (n=24). (D) BBR decreased brain edema in bilateral brain. Rats wererespectively from four groups described above. The water content of ipsilateral hemispheres was reduced in BBR-L group andBBR-H group (both ★P<0.01) vs. pMCAO group; whereas no difference was found in contralateral hemispheres. Data areexpressed as mean±SD (n=6). One-way ANOVA followed by the LSD test.

63B R A I N R E S E A R C H 1 4 5 9 ( 2 0 1 2 ) 6 1 – 7 0

protein was identified by immunohistochemistry at 24 h afterpMCAO. Representative immunohistochemistry photographsfor pAkt and NF-κB were exhibited in Fig. 3A. As shown inFigs. 3B and C the expressions of pAkt, pGSK3β and pCREB inBBR-L and BBR-H groups (n=6 in each group) were significantlyincreased at 24 h after surgery vs. pMCAO (P<0.01). In agree-ment with results of immunohistochemistry, western blottinganalyses (Figs. 4A, B) also showed a significant increase of pAkt,pGSK3β and pCREB in BBR-L and BBR-H groups (n=6 in eachgroup) vs. pMCAO (P<0.01).

The expression of NF-κB (p65) was upregulated after ische-mia, and was downregulated significantly at mRNA and proteinlevels after systemic administration of BBR in ischemia brain. Insham operated animals, few cells stained by NF-κB (p65) wereseen in the cortex (Fig. 3 A5). 24 h after the injury, an intensestaining ofNF-κBwasobservedat both cytoplasmandnucleus inthe ischemic cortex (Fig. 3 A6). In both BBR-L and BBR-H groups,the number of cells labeled with NF-κB (p65) was declinedsignificantly vs. pMCAO (P<0.01) (Fig. 3 A7, A8). Compared withpMCAO group, the nuclear NF-κB was significantly decreased in

BBR-L and BBR-H groups (both P<0.01) by immunohistochemis-try (Figs. 3A, C). In agreementwith the results ofwestern blotting(Figs. 4C, D) and RT-PCR (Figs. 5A, B), the expression of NF-κBwassignificantly decreased in BBR-L and BBR-H groups vs. pMCAOgroup (both P<0.01).

2.5. BBR protected the integrity of BBB

The integrity of brain blood barrier (BBB) was assessed byclaudin-5 which was primarily present in brain tight junctionsof endothelia. Compared with pMCAO group, claudin-5'sexpression was increased in both BBR-L and BBR-H groups(P<0.01) by western blotting and RT-PCR (Figs. 6A–D) at 24 h(n=6 in each group).

3. Discussion

The past decades have seen unprecedented advances in ourunderstanding of pathophysiological processes in cerebral

0

10

20

30

40

50

60

70

80

90 pGSK3β(Ser9)

pAkt(Ser473)

B

0

10

20

30

40

50

60

70

80

Sham pMCAO BBR-L BBR-HSham pMCAO BBR-L BBR-H

pCREB

NF-κB

pCR

EB

pos

itive

cel

ls

C

pAkt

(S

er 4

73)

A1

A5A7

A8A6

A

A2 A3A4

200um 200um

200um200um

200um200um

200um200um

Fig. 3 – Representative immunohistochemistry photographs of NF-κB and pAkt (Ser473) (A) and positive cell number analysis(B and C) of pGSK3β (Ser9), pAkt (Ser473), pCREB and NF-κB (p65). Brain samples obtained from Sham group, pMCAO group,BBR-L group and BBR-H group. (A1) The pAkt (Ser473) expression could be observed dominantly in the Sham group at 24 h inthe Sham group. (A2) The pAkt (Ser473) expression evidently decreased in the pMCAO group at 24 h than Sham group.(A3, A4) The pAkt (Ser473) expression increased in the BBR-L and BBR-H treatment groups at 24 h than that of pMCAO group.The number of NF-κB positive cells increased in pMCAO group (A6) than Sham group (A5) at 24 h. (A7, A8) NF-κB expressiondecreased relatively in the BBR-L and BBR-H treatment groups at 24 h. The expression of pAkt (Ser473), pGSK3β (Ser9), andpCREB was increased in BBR-L group (★P<0.01) and BBR-H group (★P<0.01) vs. pMCAO group, and the expression of NF-κB(p65) was decreased in BBR-L group (★P<0.01) and BBR-H group (★P<0.01) vs. pMCAO group. Data are expressed as mean±SD(n=6). One-way ANOVA followed by the LSD test.

64 B R A I N R E S E A R C H 1 4 5 9 ( 2 0 1 2 ) 6 1 – 7 0

ischemia, due in large part to the focus on MCAO as a well-characterized and classical experimental model (Longa et al.,1989; Sun et al., 2003; Yang et al., 2009). Evidences of increasedinflammatory and apoptosis damage are observed in thismodel (Dirnagl et al., 1999; Liu et al., 2009b; Lo et al., 2003). Thismodel allows us to explore BBR's effect in ischemic injury andthe role of the Akt/GSK pathway in mediating the anti-apoptosis and anti-inflammatory effects.

The main finding of the present study is that systemicadministration of BBR in a rat pMCAO model 24 h after strokepromoted neurological functional recovery via up-regulatingpAkt, pGSK3β, pCREB and claudin-5, and down-regulating thenuclear accumulation of NF-κB (p65). To our knowledge, ourstudy is the first time to use pMCAO model to explore theinterrelation between BBR's neuroprotective effect and activityof Akt/GSK pathway in vivo. The preservation of Akt activity byBBR treatment is likely the primary basis of its cytoprotective

action in brain infarction. pAkt is believed to suppress apoptosisthrough phosphorylation of several substrates including GSK3β.In this study, we defined the relationship between BBR and Akt/GSK signaling, as well as CREB, NF-κB and claudin-5 in ischemicbrain tissue. BBR induced rescue of Akt activity and consequent-ly blocked GSK3β dephosphorylation after cerebral ischemia.Furthermore, BBR administration also promoted CREB phos-phorylation, decreased NF-κB nuclear transposition and in-creased claudin-5 expression.

The causes of cell death in stroke are multifactorial and areinfluenced by the ischemic environment lacing nutrients andoxygen, coupled with the loss of survival signals (Zhang et al.,2001). The Akt/PKB signaling pathway is known as one of themost relevant pathways in regulating neuronal survival (Songet al., 2005). As a primary mediator of survival signals, it is bothnecessary and sufficient for cell survival by targeting apoptoticfamily members such as Ced-9/Bcl-2, Ced-3/caspases, forkhead

0

1

2

3

4

5

6

7

Sham pMCAO BBR-L BBR-H

Sham pMCAO BBR-L BBR-H

pGSK3 β (Ser9)pAkt(Ser473)pCREB

Nor

mal

ized

pro

tein

leve

l

B

NF-

0

0.5

1

1.5

2

2.5

3NF- κ B nuclear

NF- κ B cytoplasmic

D

β

β

C

A

Fig. 4 – Immunoblots (A and C) and quantitative analysis (B and D) of pGSK3β (Ser9), pAkt (Ser473), pCREB and, nuclear andcytoplasm NF-κB. Brain samples obtained from neocortical tissues of four groups described in Fig. 3. The expression of pAkt(Ser473), pGSK3β (Ser9) and pCREB was increased in BBR-L group (★P<0.01) and BBR-H group (★P<0.01) vs. pMCAO group, butthe expression of Akt, GSK3β, CREB was constant in control and experimental groups. The expression of nuclear NF-κB (p65)was decreased in BBR-L group (★P<0.01) and BBR-H group (★P<0.01) vs. pMCAO group. On the contrary, the expression ofcytosolic NF-κB (p65) was increased in BBR-L group (★P<0.01) and BBR-H group (★P<0.01) vs. pMCAO group. β-actin was usedas an endogenous protein loading control. Data represent mean±SD (n=6). One-way ANOVA followed by LSD test.

65B R A I N R E S E A R C H 1 4 5 9 ( 2 0 1 2 ) 6 1 – 7 0

transcription factors IKK-α and IKK-β. And it can phosphorylateand inactivate GSK3β at its N-terminus (at Ser9), and activatingCREB phosphorylation, which regulates the expression of genescritical for survival, such as brain-derived neurotrophic factor.In addition, Akt also has a role in modulating intracellularglucose metabolism, and consequently enhances energy pro-duction after ischemia. Thus, Akt is an excellent therapeutictarget for preserving neuron viability in the acute ischemicperiod. There are twomammalian isoforms for serine/threoninekinase GSK3: GSK3α and GSK3β. Unlike most kinases, GSK3 isconstitutively active in cells and can be inactivated by phos-phorylation (Cohen and Frame, 2001). Under stimulation, GSK3βis phosphorylated at serine 9 for GSK3α or serine 9 for GSK3β,resulting in the inhibition of GSK3 kinase activity. Akt phosphor-ylates GSK3 at both of these sites. Under conditions when Aktactivity is increased, GSK3β can be inactivated (Cohen andFrame, 2001). GSK3β is highly expressed in the central nervoussystem (Leroy and Brion, 1999). Inhibition of GSK3β has beendemonstrated to reduce apoptosis and enhance cell survival(Pap and Cooper, 1998). GSK3 has emerged as a key regulatoryswitch in themodulation of inflammation (Martinez et al., 2002).The present study showed that BBR's protective effects may bethrough activation of Akt/GSK signaling pathway.

Our previous studies have proved that the expression ofpCREB is upregulated in the brain cortex at early stage of brainischemia (Liu et al., 2010). Ischemic spinal cord injury inducedthe CREB phosphorylation at the anterior horn of the spinal cord.CREB phosphorylation was marked in the peri-infarct area(Irving et al., 2000). CREB is abundant in the brain andparticularlyin neurons. Various extracellular stimuli cause phosphorylationof CREB by protein kinase A, protein kinase B (PKB), extra-cellular signal-related protein kinase, and calcium–calmodulin-dependent protein kinase. Recently CREB phosphorylation hasbeen found to be crucial in neurotrophin-mediated neuronsurvival (Finkbeiner, 2000). In our study, it was interesting thatthe expression of pCREB was increased after ischemic stroke,and BBR can further intensify its phosphorylation. So it waspossible that BBR treatment promotes cell survival and decreaseinflammation through Akt/GSK activation pathway and pCREBactivation cascade.

Inflammation is proved playing a role in human stroke andits animal models several hours after ischemia (Iadecola andAlexander, 2001). Apoptosis is another main mechanism ofneuronal death in stroke models evidenced by the ischemiastudy using transgenic and knockoutmice (Dirnagl et al., 1999;Lo et al., 2003). Transcription factor NF-κB plays a pivotal role

A

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

Sham pMCAO BBR-L BBR-H

Nor

mal

ized

NF

-kB

mR

NA

leve

l

B

Fig. 5 – RT-PCR of NF-κB genes in the rat brain. Agarose gelphotographs (A) and the bar graph (B) showed the RT-PCR ofBBR induction (10 mg/kg or 40 mg/kg, ip) NF-κB mRNA.Samples were the same with Fig. 4. The expression of NF-κBmRNA was reduced in BBR-L group (★P<0.01) and BBR-Hgroup (★P<0.01) vs. pMCAO group. An endogenous GAPDHwas used as RT-PCR control. Data are mean±SD (n=6).One-way ANOVA followed by LSD test.

A

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2

Sham pMCAO BBR-L BBR-H

Claudin-5

Nor

mal

ized

mR

NA

leve

l

C

Fig. 6 – Claudin-5 expression in mRNA and protein level. Agarosinduction (10 mg/kg or 40 mg/kg, ip) of claudin-5mRNA. Westernbar chart (D) showed the BBR's effect on claudin-5 protein level. ThmRNA and protein levels in BBR-L group (★P<0.01) and BBR-H grendogenous protein loading control. The results were expressedData are mean±SD (n=6). One-way ANOVA followed by LSD test

66 B R A I N R E S E A R C H 1 4 5 9 ( 2 0 1 2 ) 6 1 – 7 0

in the regulation of immune and inflammatory responses (Liuet al., 2009a; Tak and Firestein, 2001). Particularly in the brainNF-κB regulates the expression of both proinflammatorygenes, such as IL-1β, COX-2, tumor necrosis factor-α, matrixmetalloproteinase-9 (MMP-9), and inducible nitric oxide synthase,and genes related to apoptosis such as Bcl-2, manganesesuperoxide dismutase (Mattson and Camandola, 2001). ForpMCAO it was demonstrated that the role of NF-κB wasdetrimental (Nurmi et al., 2004). In our study, ischemiainduced NF-κB nuclear accumulation was ameliorated byBBR administration.

After the onset of cerebral ischemia the cytotoxic responseoccurs within minutes and encompasses oxidative stress,inflammatory responses, cell death, and neurologic injury (Ji etal., 2012; Liu et al., 2009b; Yang et al., 2009). And the proteinlevels of pAkt, pGSK, pCREB and NF-κB nuclear accumulationare implicated in the early stage of ischemia (Liu et al., 2010; Sunet al., 2011;Wang et al., 2010). So that original administration ofBBR will be effective in the acute phase of cerebral ischemia.Although the clinical application of berberine is not limited bythe short therapeutic window, it is better to be applied in theacute phase. Numerous animal trials have demonstrated thatberberine administration both from 7 days prior to ischemiaand 1 day later showed a neuroprotective effect (Benaissa et al.,2009; Hong et al., 2012; Zhou et al., 2008). And in our studyberberine was administrated immediately after the onset ofischemia. Prior research indicated that the potent stroke-protection effect of BBR occurred via reducing MMP-9 activity(Hong et al., 2012), blockade on K+ currents (Wang et al., 2004),decreasing reactive oxygen species level (Zhou et al., 2008),COX-2 expression and PGE2 production (Yoo et al., 2008). In ourstudy, we investigated BBR's effect on stroke and the close

B

Sham pMCAO BBR-L BBR-H

D

0

0.2

0.4

0.6

0.8

1

1.2 Claudin-5

Nor

mal

ized

pro

tein

leve

l

e gel photographs (A) and the bar graph (C) showed BBRblotting photographs (B) and the respective quantity analysise expression of claudin-5was significantly promoted both inoup (★P<0.01) vs. pMCAO group. β-actin was used as annormalized to the GAPDH endogenous control in RT-PCR..

67B R A I N R E S E A R C H 1 4 5 9 ( 2 0 1 2 ) 6 1 – 7 0

relationship between BBR and Akt/GSK signaling. BBR admin-istration, both dosage of 10mg/kg and 40mg/kg, could relievenerve defection, reduce brain edema, and decrease infarct size.BBR also could markedly promote the activation of Akt/GSKsignaling, increase CREB phosphorylation and NF-κB transloca-tion from cytoplasm to nucleus. The expression of pCREB inbrain ischemicwas knownas a protective role for brain. And theNF-κB translocation was crucial for inflammation progression.Thus, activating Akt/GSK signal pathway may be an attractivecandidate to explain protective effects of BBR in the acute stageof ischemic stroke. As a salt form, BBR is thought to be water-soluble. Thus it could be difficult for BBR to pass the BBB underphysiological conditions. However, it is suggested that BBRcould have a direct action on neuron and accumulate in thehippocampus and permeable to the brain in many braindiseases (Wang et al., 2005). In ratmodel of Alzheimer's disease,BBR can ameliorate the spatial memory impairment andincrease the expression of interleukin-1β and inducible nitricoxide synthase (Zhu and Qian, 2006).

Edema is the major cause of cerebrovascular death withinthe first week after stroke. And disruption of the BBB integrity isan early and prominent event in brain ischemia (Petito, 1979).Tight junctions are well-developed between adjacent endothe-lial cells in the blood vessels of central nervous system, andplaya primary role in establishing the BBB (Mark and Davis, 2002).Claudin-5, a major cell adhesion molecule of tight junctions incerebral endothelial cells, was important for BBB integrity study(Nitta et al., 2003). In our research, the expression of claudin-5was promoted, and thismay be themajor reason for BBR's brainedema reduction effect.

BBRdecreasedneurologic impairment and tissue injury aftercerebral ischemia and this effect may be through activation ofAkt/GSK pathway and up-regulation of claudin-5 expression.These results in particular indicate that Akt/GSK signalingmight be an attractive therapeutic target for BBR during cerebralischemia and ameliorating brain injury in stroke.

4. Conclusions

In conclusion, our study provides beneficial evidences for earlyadministration of BBR and its underling mechanisms afteracute brain ischemia. Specifically, the protective effect of BBRagainst ischemic injury may be through Akt/GSK signalingactivation, upregulation of pCREB, downregulation of NF-κBnuclear transposition and ameliorated BBB permeability.

5. Experimental procedures

5.1. Animals and ischemia protocol

All procedures were performed in compliance with the NationalInstitutes ofHealthGuide for Care andUse of LaboratoryAnimals. Theexperimental protocolwas approved by the institutional animalcare and use committee and the local experimental ethicscommittee and conformed to internationally accept ethicalstandards.All effortsweremade to alleviate animal suffering, tominimize the number of animals used, and to utilize alterna-tives to in vivo techniques, if permitted. Male Sprague–Dawley

rats (250–280 g) were kept on a 12 h light/12 h dark regime,with free access to food and water, which were supplied bythe Laboratory Animal Centre, Hebei Medical University,Shijiazhuang, Hebei, China. Ninety-six rats were randomlydivided into four groups (24 rats in each group): Sham operatedgroup that received equal volume 0.9% NaCl (Sham); pMCAOgroup that received equal volume 0.9% NaCl after pMCAO(pMCAO); low dose group that received BBR at 10mg/kg afterpMCAO (BBR-L); and high dose group that received BBR at40 mg/kg after pMCAO (BBR-H).

A standard model of pMCAO was used to make permanentfocal ischemia as previously described (Longa et al., 1989; Yanget al., 2009). Briefly, rats were anesthetized by 10% chloralhydrate (350mg/kg, intraperitoneal). In anesthetized rats, theright side of common carotid artery was exposed and isolated.Middle cerebral artery (MCA) was occluded by inserting amonofilament nylon suture with a heat-rounded tip into theinternal carotid artery, which was advanced further until itclosed the origin of the MCA. Body temperature was monitoredand maintained at 36.5 °C to 37.5 °C throughout the surgery.Sham-operated control rats received the same surgical proce-dure without inserting a filament.

5.2. Drug administration

BBR (Nanjing Zelang Medical Technology Co. Ltd, Nanjing,Jiangsu, China) with purity of more than 98%, was dissolved insaline to prepare concentration of 10 mg/ml. Rats were treatedwith BBR by intraperitoneal injection at different doses of 10and 40 mg/kg (added with 0.9% saline to a total volume of1 ml) per day immediately after cerebral ischemia. In thecases of the pMCAO and Sham group, equal volume 0.9%saline was administered in the same manner.

5.3. Evaluation of neurological deficit

Neurological deficit scores were evaluated by an examinerblinded to the experimental groups at 24 h after pMCAOfollowing a modified scoring system based on that developedfrom Longa et al. (1989) and Ding et al. (2002), as follows: 0, nodeficits; 1, difficulty in fully extending the contralateral forelimb;2, unable to extend the contralateral forelimb; 3, mild circling tothe contralateral side; 4, severe circling; and 5, falling to thecontralateral side. The higher the neurological deficit score, themore severe impairment of motor motion injury. The rats werefrom groups of brain water content, infarct volume, immunohis-tochemistry, western blot and reverse transcription-polymerasechain reaction (n=24).

5.4. Measurement of brain water content

Brain water content was measured by the standard wet/dryweight method. Six rats of each group were deeply anesthetizedwith 10% chloral hydrate and killed by decapitation at 24 h afterpMCAO. The brains were quickly removed and placed on acooled surface. A coronal brain slice (about 3mm thick) was cutand the slice was divided into the ipsilateral and contralateralhemispheres after dissecting free 4mm frontal pole. The twohemisphere slices packaged with pre-weighed tin foils wereimmediately weighed on an electronic balance to obtain the wet

68 B R A I N R E S E A R C H 1 4 5 9 ( 2 0 1 2 ) 6 1 – 7 0

weight, and then weighed again to obtain the dry weight afterdried for 24 h in an oven at 100 °C. Brain water content wascalculatedwith theequationas follows: Brainwater content (%)=(wet weight-dry weight)/(wet weight-tin foil weight)×100%.

5.5. Measurement of infarct volume

Infarct volume after pMCAO was determined by 2,3,5-triphenyltetrazolium chloride (TTC) at 24 h after pMCAO.Also 6 rats of each group were euthanized with chloral hy-drate and the brains were collected quickly. Then the brainswere sliced into 5 coronal sections 3-mm thick each andstained with 2% solution of TTC at 37 °C for 20 min, followedby fixation with 4% paraformaldehyde. The normal tissuewas stained deep red while the infarct area pale gray. Stainedsections were photographed and the digital images wereanalyzed using Image-Pro Plus 5.1 system to calculate theipsilateral and contralateral hemispheric volumes and in-farct volumes. To compensate for the effect of brain edema,the percentage hemisphere lesion volumes (%HLV) werecalculated by the following formula:

%HLV ¼ f½total infarct volume−ðthe volume of intact ipsilateral hemisphere−the volume of intact contralateral hemisphereÞ�=contralateral hemisphere volumeg � 100%:

5.6. Immunohistochemistry

Brains (n=6 in each group) at 24 h after pMCAO and deepanesthetizationwere removed and immersed in 4% paraformal-dehyde over 24 h at 4 °C, and then dehydrated in gradientalcohol, embedded in paraffin. Coronal brain slices containingthird ventricle and basal ganglia about 4 mm thickwere cut afterdissecting free 7mm frontal pole (3 mm of which were used forbrain water content detection described in the Section 2.4).Coronal brain slices, containing third ventricle and basal ganglia,were used. Standard histological processing was performedfor paraffin-embedded sections. The slices were incubated with3% H2O2, 3% normal goat serum and incubated with interestprimary antibodies respectively in 0.01 mol/L phosphate-buffered saline overnight. Rabbit monoclonal antibody againstpGSK3β (1:100, Santa Cruz Biotechnology), mouse monoclonalantibody against pAkt (1:50, Cell Signaling Technology), rabbitpolyclonal antibody claudin-5 antibody (1:200, Santa CruzBiotechnology) and rabbit polyclonal antibody NF-κB p65 (1:100,Santa Cruz Biotechnology) were used to detect the expres-sions. Immunohistochemistry was performed via the avidin–biotin technique, and then hematoxylin stainingwas selectedas counterstaining. The secondary antibodies, secondarybiotinylated conjugates and diaminobezidine were from theSP kit (Zhongshan Biology Technology Company, China). Slideswere viewed and photographed with a ×400 light microscope(Nikon, Japan).

5.7. Western blot

The cytosolic and nuclear protein and total protein wereextracted respectively from rat ischemic and control cortexfollowing the manufacturer's protocols (Applygen TechnologiesInc., Beijing, China) at 24 h after pMCAO. Protein concentration

of the supernatant was determined using a BCA Protein AssayReagent Kit (Novagen, Madison, WI, USA) with bovine serumalbumin as the standard. An equivalent amount of 50 μg totalprotein samples, as well as 40 μg cytosolic or nuclearsamples, was separated respectively by sodium dodecylsulfate-polyacrylamide gels prior and transferred 2 h on toPVDF membranes (Millipore Corporation, USA). After block-ing 1 h with 5% non-fat dry milk in phosphate buffered saline(PBS), membranes were incubated overnight at 4 °C with rabbitpolyclonal antibody anti-Akt (1:800, Cell Signaling Technology),anti-pAkt (1:500), rabbitmonoclonal antibody anti-GSK3β(1:1000,Cell Signaling Technology), anti-pGSK3β (1:500), NF-κB (1:500),claudin-5 (1:200) and anti-β-actin (1:500, Santa Cruz Biotechnol-ogy). The second day, membranes were washed with PBScontaining 0.1% Tween-20 (TPBS) (10 min×3) each time andsubsequently incubated with fluorescent labeling second anti-bodies (IRDye® 800-conjugated goat anti-rabbit or anti-mouseIgG, 1:5000 dilution, Rockland, Gilbertsville, PA) for 1 h at roomtemperature. Membranes were then again washed with TPBS(10min×3) and the relative density of bandswas analyzed on anOdyssey infrared scanner (LI-COR Bioscience, USA). The densi-tometric values were normalized with respect to the values ofβ-actin immunoreactivity to correct for any loading and transferdifferences between samples. Six rats each group were used.

5.8. Reverse transcription-polymerase chain reaction (RT-PCR)

RT-PCR was used to analyze the levels of claudin-5, NF-κB andGAPDH mRNA. At 24 h after pMCAO, rats were reanesthetizedand brains were removed and frozen in liquid nitrogen. TotalRNA from ischemia cortex was extracted from the brain (n=6 ineach group) after pMCAO using Trizol reagent (Invitrogen,Carlsbad, CA, USA) following the manufacturer's instructions,cDNA was synthesized using ThermoScript RT-PCR System(Invitrogen). Forward and reverse primers are as follows: NF-κB:5′-AGAGAAGCACAGATACCACTAAG-3′ and 5′-CAGCCTCATA-GAAGCCATC-C-3′; claudin-5: 5′-CACAGAGAGGGGTCGTTGAT-3′and 5′-ACTGTTAGCGGCA-GTTTGGT-3′, GAPDH: 5′-ACAGCAA-CAGGGTGGTGGAC-3′, and 5′-TTTGAGG-GTGCAGCGAACTT-3′.Reverse transcription was carried out using RevertAid firstStrand cDNA Synthesis Kit (Fermentas International Inc,Canada) following the manufacturer's instructions. Thepolymerase chain reaction was performed in a total volumeof 20 μL using GoTaq®Green Master Mix (Promega, Madison,WI, USA). GAPDH was used as an internal standard gene.The RT-PCR products were separated on 2% agarose gel andthe intensity of each band was quantified using gel Smart-View analysis software (Cognex, Natick, MA, USA). Resultswere expressed relative to the corresponding intensity ofthe GAPDH bands from the same RNA sample.

5.9. Data analysis

Group data in this study was analyzed using SPSS 13.0software. Quantitative data was represented as mean±SD.Statistical analysis was performed by One-way ANOVAfollowed by LSD test for intergroup comparisons. For neuro-logical deficits, Mann–Whitney U-test was used for compari-son between two groups. Differences with P<0.05 wereconsidered statistically significant.

69B R A I N R E S E A R C H 1 4 5 9 ( 2 0 1 2 ) 6 1 – 7 0

Acknowledgments

Thisworkwas fundedbyHebei Province (Grantnos. C2010000564and 10276104D); we thank technicians Ruichun Liu and HongranWu for their technical assistance, and Prof. Yansu Guo M.D.PhD. and Prof. Weisong Duan M.D. PhD. for providing valuablesuggestions.

R E F E R E N C E S

Benaissa, F., Mohseni-Rad, H., Rahimi-Moghaddam, P.,Mahmoudian, M., 2009. Berberine reduces the hypoxic–ischemicinsult in rat pup brain. Acta Physiol. Hung. 96, 213–220.

Bhutada, P., Mundhada, Y., Bansod, K., Tawari, S., Patil, S., Dixit, P.,Umathe, S., Mundhada, D., 2011. Protection of cholinergic andantioxidant system contributes to the effect of berberineameliorating memory dysfunction in rat model ofstreptozotocin-induced diabetes. Behav. Brain Res. 220, 30–41.

Bijur, G.N., Jope, R.S., 2000. Opposing actions of phosphatidylinositol3-kinase and glycogen synthase kinase-3beta in the regulationof HSF-1 activity. J. Neurochem. 75, 2401–2408.

Christophe, M., Nicolas, S., 2006. Mitochondria: a target forneuroprotective interventions in cerebral ischemia–reperfusion.Curr. Pharm. Des. 12, 739–757.

Cohen, P., Frame, S., 2001. The renaissance of GSK3. Nat. Rev. Mol.Cell Biol. 2, 769–776.

Collino, M., Aragno, M., Castiglia, S., Tomasinelli, C.,Thiemermann, C., Boccuzzi, G., Fantozzi, R., 2009. Insulinreduces cerebral ischemia/reperfusion injury in thehippocampus of diabetic rats: a role for glycogen synthasekinase-3beta. Diabetes 58, 235–242.

Ding, Y., Li, J., Rafols, J.A., Phillis, J.W., Diaz, F.G., 2002. Prereperfusionsaline infusion into ischemic territory reduces inflammatoryinjury after transient middle cerebral artery occlusion in rats.Stroke 33, 2492–2498.

Dirnagl, U., Iadecola, C., Moskowitz, M.A., 1999. Pathobiology ofischaemic stroke: an integrated view. Trends Neurosci. 22,391–397.

Donnan, G.A., Fisher, M., Macleod, M., Davis, S.M., 2008. Stroke.Lancet 371, 1612–1623.

Endo, H., Nito, C., Kamada, H., Nishi, T., Chan, P.H., 2006.Activation of the Akt/GSK3beta signaling pathway mediatessurvival of vulnerable hippocampal neurons after transientglobal cerebral ischemia in rats. J. Cereb. Blood Flow Metab. 26,1479–1489.

Finkbeiner, S., 2000. CREB couples neurotrophin signals to survivalmessages. Neuron 25, 11–14.

Frame, S., Cohen, P., 2001. GSK3 takes centre stage more than 20years after its discovery. Biochem. J. 359, 1–16.

Hong, J.S., Chu, Y.K., Lee, H., Ahn, B.H., Park, J.H., Kim, M.J., Lee, S.,Ryoo, H.S., Jang, J.H., Lee, S.R., Park, J.W., 2012. Effects ofberberine on hippocampal neuronal damage and matrixmetalloproteinase-9 activity following transient globalcerebral ischemia. J. Neurosci. Res. 90, 489–497.

Iadecola, C., Alexander, M., 2001. Cerebral ischemia andinflammation. Curr. Opin. Neurol. 14, 89–94.

Irving, E.A., Barone, F.C., Reith, A.D., Hadingham, S.J., Parsons,A.A., 2000. Differential activation of MAPK/ERK and p38/SAPKin neurones and glia following focal cerebral ischaemia in therat. Brain Res. Mol. Brain Res. 77, 65–75.

Ji, H., Miao, J., Zhang, X., Du, Y., Liu, H., Li, S., Li, L., 2012. Inhibitionof sonic hedgehog signaling aggravates brain damage associatedwith the down-regulation of Gli1, Ptch1 and SOD1 expression inacute ischemic stroke. Neurosci. Lett. 506, 1–6.

Kelly, S., Zhao, H., Hua Sun, G., Cheng, D., Qiao, Y., Luo, J., Martin,K., Steinberg, G.K., Harrison, S.D., Yenari, M.A., 2004. Glycogensynthase kinase 3beta inhibitor Chir025 reduces neuronaldeath resulting from oxygen-glucose deprivation, glutamateexcitotoxicity, and cerebral ischemia. Exp. Neurol. 188,378–386.

Kong, W., Wei, J., Abidi, P., Lin, M., Inaba, S., Li, C., Wang, Y., Wang,Z., Si, S., Pan, H., Wang, S., Wu, J., Wang, Y., Li, Z., Liu, J., Jiang,J.D., 2004. Berberine is a novel cholesterol-lowering drugworking through a unique mechanism distinct from statins.Nat. Med. 10, 1344–1351.

Kuo, C.L., Chi, C.W., Liu, T.Y., 2004. The anti-inflammatorypotential of berberine in vitro and in vivo. Cancer Lett. 203,127–137.

Leroy, K., Brion, J.P., 1999. Developmental expression and localizationof glycogen synthase kinase-3beta in rat brain. J. Chem.Neuroanat. 16, 279–293.

Liu, P., Lu, M., Tian, B., Li, K., Garofalo, R.P., Prusak, D., Wood, T.G.,Brasier, A.R., 2009a. Expression of an IKKgamma splice variantdetermines IRF3 and canonical NF-kappaB pathway utilizationin ssRNA virus infection. PLoS One 4, e8079.

Liu, Y., Zhang, X.J., Yang, C.H., Fan, H.G., 2009b. Oxymatrineprotects rat brains against permanent focal ischemia anddownregulates NF-kappaB expression. Brain Res. 1268,174–180.

Liu, L., Zhang, X., Wang, L., Yang, R., Cui, L., Li, M., Du, W., Wang,S., 2010. The neuroprotective effects of Tanshinone IIA areassociated with induced nuclear translocation of TORC1 andupregulated expression of TORC1, pCREB and BDNF in theacute stage of ischemic stroke. Brain Res. Bull. 82, 228–233.

Lo, E.H., Dalkara, T., Moskowitz, M.A., 2003. Mechanisms,challenges and opportunities in stroke. Nat. Rev. Neurosci. 4,399–415.

Longa, E.Z., Weinstein, P.R., Carlson, S., Cummins, R., 1989.Reversible middle cerebral artery occlusion without craniotomyin rats. Stroke 20, 84–91.

Mangi, A.A., Noiseux, N., Kong, D., He, H., Rezvani, M., Ingwall, J.S.,Dzau, V.J., 2003. Mesenchymal stem cells modified with Aktprevent remodeling and restore performance of infarctedhearts. Nat. Med. 9, 1195–1201.

Mark, K.S., Davis, T.P., 2002. Cerebral microvascular changes inpermeability and tight junctions induced byhypoxia-reoxygenation. Am. J. Physiol. Heart Circ. Physiol. 282,H1485–H1494.

Martin, M., Rehani, K., Jope, R.S., Michalek, S.M., 2005. Toll-likereceptor-mediated cytokine production is differentially regulatedby glycogen synthase kinase 3. Nat. Immunol. 6, 777–784.

Martinez, A., Castro, A., Dorronsoro, I., Alonso, M., 2002. Glycogensynthase kinase 3 (GSK-3) inhibitors as new promising drugsfor diabetes, neurodegeneration, cancer, and inflammation.Med. Res. Rev. 22, 373–384.

Mattson, M.P., Camandola, S., 2001. NF-kappaB in neuronalplasticity and neurodegenerative disorders. J. Clin. Invest. 107,247–254.

Nitta, T., Hata, M., Gotoh, S., Seo, Y., Sasaki, H., Hashimoto, N.,Furuse, M., Tsukita, S., 2003. Size-selective loosening of theblood–brain barrier in claudin-5-deficient mice. J. Cell Biol. 161,653–660.

Nurmi, A., Lindsberg, P.J., Koistinaho, M., Zhang, W., Juettler, E.,Karjalainen-Lindsberg, M.L., Weih, F., Frank, N., Schwaninger,M., Koistinaho, J., 2004. Nuclear factor-kappaB contributes toinfarction after permanent focal ischemia. Stroke 35, 987–991.

Pap, M., Cooper, G.M., 1998. Role of glycogen synthase kinase-3 inthe phosphatidylinositol 3-Kinase/Akt cell survival pathway.J. Biol. Chem. 273, 19929–19932.

Petito, C.K., 1979. Early and late mechanisms of increased vascularpermeability following experimental cerebral infarction.J. Neuropathol. Exp. Neurol. 38, 222–234.

70 B R A I N R E S E A R C H 1 4 5 9 ( 2 0 1 2 ) 6 1 – 7 0

Song, G., Ouyang, G., Bao, S., 2005. The activation of Akt/PKBsignaling pathway and cell survival. J. Cell. Mol. Med. 9,59–71.

Sun, Y., Jin, K., Xie, L., Childs, J., Mao, X.O., Logvinova, A.,Greenberg, D.A., 2003. VEGF-induced neuroprotection,neurogenesis, and angiogenesis after focal cerebral ischemia.J. Clin. Invest. 111, 1843–1851.

Sun, B., Chen, L., Wei, X., Xiang, Y., Liu, X., Zhang, X., 2011. TheAkt/GSK-3β pathway mediates flurbiprofen-inducedneuroprotection against focal cerebral ischemia/reperfusioninjury in rats. Biochem. Biophys. Res. Commun. 409, 808–813.

Tak, P.P., Firestein, G.S., 2001. NF-kappaB: a key role in inflammatorydiseases. J. Clin. Invest. 107, 7–11.

Tan, W., Li, Y., Chen, M., Wang, Y., 2011. Berberine hydrochloride:anticancer activity and nanoparticulate delivery system. Int. J.Nanomedicine 6, 1773–1777.

Walton, M.R., Dragunow, I., 2000. Is CREB a key to neuronalsurvival? Trends Neurosci. 23, 48–53.

Wang, F., Zhao, G., Cheng, L., Zhou, H.Y., Fu, L.Y., Yao, W.X., 2004.Effects of berberine on potassium currents in acutely isolatedCA1 pyramidal neurons of rat hippocampus. Brain Res. 999,91–97.

Wang, X., Wang, R., Xing, D., Su, H., Ma, C., Ding, Y., Du, L., 2005.Kinetic difference of berberine between hippocampus andplasma in rat after intravenous administration of Coptidisrhizoma extract. Life Sci. 77, 3058–3067.

Wang, L., Zhang, X., Liu, L., Cui, L., Yang, R., Li, M., Du, W., 2010.Tanshinone II A down-regulates HMGB1, RAGE, TLR4,NF-kappaB expression, ameliorates BBB permeability andendothelial cell function, and protects rat brains against focalischemia. Brain Res. 1321, 143–151.

Yang, C., Zhang, X., Fan, H., Liu, Y., 2009. Curcumin upregulatestranscription factor Nrf2, HO-1 expression and protects ratbrains against focal ischemia. Brain Res. 1282, 133–141.

Yoo, K.Y., Hwang, I.K., Kim, J.D., Kang, I.J., Park, J., Yi, J.S., Kim, J.K.,Bae, Y.S., Won, M.H., 2008. Antiinflammatory effect of theethanol extract of Berberis koreana in a gerbil model of cerebralischemia/reperfusion. Phytother. Res. 22, 1527–1532.

Zha,W., Liang, G., Xiao, J., Studer, E.J., Hylemon, P.B., Pandak Jr., W.M.,Wang, G., Li, X., Zhou, H., 2010. Berberine inhibits HIV proteaseinhibitor-induced inflammatory response bymodulating ER stresssignaling pathways in murine macrophages. PLoS One 5, e9069.

Zhang, M., Methot, D., Poppa, V., Fujio, Y., Walsh, K., Murry, C.E.,2001. Cardiomyocyte grafting for cardiac repair: graft cell deathand anti-death strategies. J. Mol. Cell. Cardiol. 33, 907–921.

Zhou, X.Q., Zeng, X.N., Kong, H., Sun, X.L., 2008. Neuroprotectiveeffects of berberine on stroke models in vitro and in vivo.Neurosci. Lett. 447, 31–36.

Zhu, F., Qian, C., 2006. Berberine chloride can ameliorate thespatial memory impairment and increase the expression ofinterleukin-1beta and inducible nitric oxide synthase in the ratmodel of Alzheimer's disease. BMC Neurosci. 7, 78.