inhibition of glycogen synthase kinase-3β prevents nsaid-induced acute kidney injury

Inhibition of glycogen synthase kinase-3b preventsNSAID-induced acute kidney injuryHao Bao1,2, Yan Ge2, Shougang Zhuang2, Lance D. Dworkin2, Zhihong Liu1 and Rujun Gong2

1Research Institute of Nephrology, Jinling Hospital, Nanjing University School of Medicine, Nanjing, China and 2Division of KidneyDisease and Hypertension, Department of Medicine, Brown University School of Medicine, Providence, Rhode Island, USA

Clinical use of nonsteroidal anti-inflammatory drugs (NSAIDs)

like diclofenac (DCLF) is limited by multiple adverse effects,

including renal toxicity leading to acute kidney injury.

In mice with DCLF-induced nephrotoxicity, TDZD-8,

a selective glycogen synthase kinase (GSK)3b inhibitor,

improved acute kidney dysfunction and ameliorated tubular

necrosis and apoptosis associated with induced cortical

expression of cyclooxygenase-2 (COX-2) and prostaglandin E2.

This renoprotective effect was blunted but still largely

preserved in COX-2-null mice, suggesting that other GSK3btargets beyond COX-2 functioned in renal protection.

Indeed, TDZD-8 diminished the mitochondrial permeability

transition in DCLF-injured kidneys. In vitro, GSK3b inhibition

reinstated viability and suppressed necrosis and apoptosis

in DCLF-stimulated tubular epithelial cells. DCLF elicited

oxidative stress, enhanced the activity of the redox-sensitive

GSK3b, and promoted a mitochondrial permeability

transition by interacting with cyclophilin D, a key component

of the mitochondrial permeability transition pore. TDZD-8

blocked GSK3b activity and prevented GSK3b-mediated

cyclophilin D phosphorylation and the ensuing mitochondrial

permeability transition, concomitant with normalization

of intracellular ATP. Conversely, ectopic expression of a

constitutively active GSK3b abolished the effects of TDZD-8.

Hence, inhibition of GSK3b ameliorates NSAID-induced acute

kidney injury by induction of renal cortical COX-2 and direct

inhibition of the mitochondrial permeability transition.

Kidney International (2012) 81, 662–673; doi:10.1038/ki.2011.443;

published online 18 January 2012

KEYWORDS: acute kidney injury; cyclooxygenase-2; diclofenac; glycogen

synthase kinase-3b; mitochondrial permeability transition; nonsteroidal

anti-inflammatory drugs

Diclofenac (2-[(2,6-dichlorophenyl)amino]phenyl-acetate,DCLF) is one of the most frequently prescribed nonsteroidalanti-inflammatory drugs (NSAIDs). Its clinical use is limitedby the occurrence of DCLF-induced renal toxicity, whichcan manifest as acute kidney injury (AKI).1,2 Acute DCLF-induced nephrotoxicity involves both cyclooxygenase (COX)-dependent and COX-independent mechanisms. The blockingof COX and subsequent inhibition of prostaglandin (PG)synthesis by DCLF leads to renovascular constriction, renalischemia, a decline in glomerular hydraulic pressure, andAKI. Moreover, recent evidence suggests that mitochondriaare a primary target of DCLF.3–5 DCLF can directly inducethe mitochondrial membrane permeability transition (MPT),which subsequently triggers cell death in multiple organs,including the kidneys.6–8

Glycogen synthase kinase-3b (GSK3b) is a ubiquitousserine–threonine protein kinase that participates in a multi-tude of cellular processes, including proliferation, apoptosis,and necrosis, and has an important role in the pathophysi-ology of a number of diseases, including kidney diseases.Previous studies have demonstrated that GSK3b inhibitioninduces COX-2 expression in the renal cortex.9,10 In addition,a growing body of evidence suggests that GSK3b inhibitionhas a central role in both the ischemic and pharmacologicalpreconditioning–induced modulation of MPT induction.11,12

Inhibitory phosphorylation of GSK3b at serine 9 resulted inan elevated threshold for MPT activation and, consequently,protected from cell death.

In this study, we examined the effects of 4-benzyl-2-methyl-1,2,4-thiadiazolidine-3,5-dione (TDZD-8), a novel, non-ATPcompetitive, highly selective small-molecule inhibitor of GSK3b,on the development of DCLF-induced acute nephrotoxicityin vivo in a murine model of DCLF-induced nephrotoxicityand in vitro in cultured renal proximal tubular cells.

RESULTSGSK3b inhibition improves general behavior and kidneyfunction in DCLF-injured mice

Mice were pretreated with TDZD-8 or vehicle 1 h beforeinjury by a nephrotoxic dose of DCLF (200 mg/kg) via oralgavages. After 6 h, illness and weakness became evident in theanimals that received DCLF and vehicle alone. They werelethargic and reluctant to eat. Kidneys from DCLF-treated

o r i g i n a l a r t i c l e http://www.kidney-international.org

& 2012 International Society of Nephrology

Received 24 February 2011; revised 24 September 2011; accepted 18

October 2011; published online 18 January 2012

Correspondence: Rujun Gong, Division of Kidney Disease and Hypertension,

Department of Medicine, Brown University School of Medicine, 593 Eddy

Street, Providence, Rhode Island 02903, USA.

E-mail: [email protected]

662 Kidney International (2012) 81, 662–673

http://dx.doi.org/10.1038/ki.2011.443

http://www.kidney-international.org

mailto:[email protected]

mice appeared enlarged and swollen, and markedly higherserum creatinine levels were detected in these mice, denotingacute kidney dysfunction (Table 1). In contrast, treatmentwith TDZD-8 significantly improved animal activity andfood intake, attenuated weight loss, and preserved renalfunction in DCLF-injured mice.

GSK3b inhibition prevents tubular necrosis and apoptosis inthe kidneys of DCLF-injured mice

As shown in Figure 1, exposure to DCLF induced a typicalpattern of acute tubular necrosis in the proximal and distal

tubules, which was characterized by isometric vacuolizationof the tubular epithelium, luminal ectasia, sloughing oftubular cells into the lumen, loss of brush border, nuclearenlargement, and pleomorphism. To visualize apoptosis,TdT-mediated dUTP nick-end labeling (TUNEL) stainingwas performed (Figure 1b, d, f, h, and j). Scattered, brightTUNEL-positive nuclei could be evidently observed over theentire cortex of the DCLF-injured kidneys, but were rarelyobserved in the specimens from control mice. Both tubularnecrosis and apoptosis were significantly attenuated inTDZD-8-treated mice.

Table 1 | Changes in body weight, kidney weight, and serum creatinine (SCr) levels after different treatments

Group Ctrl T D T�L+D T+D

Base weight (g) 19.1±1.0 19.8±0.8 19.8±0.9 19.6±0.7 19.3±0.6Weight change (g) 0.01±0.16 �0.07±0.09 �1.44±0.20* �0.85±0.10** �0.55±0.12**Kidney weight (mg) 127.0±3.6 124.6±3.3 176.4±9.9* 158.7±6.8** 148.0±3.5**SCr (mg/dl) 0.19±0.03 0.21±0.02 1.36±0.16* 0.98±0.20** 0.67±0.10**

Abbreviations: Group Ctrl, control mice treated with vehicle; Group D, 2-[(2,6-dichlorophenyl)amino]phenyl-acetate (DCLF; 200 mg/kg)-injured mice; Group T, 4-benzyl-2-methyl-1,2,4-thiadiazolidine-3,5-dione (TDZD-8; 5 mg/kg)-treated mice; Group T�L+D, low-dose TDZD-8 (1 mg/kg) + DCLF (200 mg/kg)-treated mice; Group T+D, TDZD-8(5 mg/kg) + DCLF (200 mg/kg)-treated mice.*Po0.05 vs. group Ctrl.**Po0.05 vs. group D.

a b g h

jid

f

c

e k l

HE STAIN HE STAINTUNEL STAIN TUNEL STAIN

5 40

30

20

10

0

4

3

2

1

0Ctrl

*#

# ##

*

Ctrl– – –

– ––5

200 200 200 200 200 2001 15 55

– –

T TD DT-L+D T-L+DT+D T+D

Tubu

lar

inju

ry s

core

Apo

ptot

ic c

ell/m

m2

Figure 1 | Glycogen synthase kinase-3b (GSK3b) inhibition protects mice from 2-[(2,6-dichlorophenyl)amino]phenyl-acetate (DCLF)-induced renal tubular injury and cell apoptosis in vivo. Representative micrographs of hematoxylin and eosin (HE) staining (a, c, e, g, i)and TdT-mediated dUTP nick-end labeling (TUNEL) staining (b, d, f, h, j) of kidney specimens. (a, b) Group Ctrl, control (vehicle treated); (c,d) Group T, 4-benzyl-2-methyl-1,2,4-thiadiazolidine-3,5-dione (TDZD-8; 5 mg/kg); (e, f) Group D, DCLF (200 mg/kg); (g, h) Group T�LþD,low-dose TDZD-8 (1 mg/kg)þDCLF (200 mg/kg); (i, j) Group TþD, TDZD-8 (5 mg/kg)þDCLF (200 mg/kg). (k, l) Quantitative analysisof tubular injury and TUNEL-labeled cells among the groups. The results are expressed as the mean±s.d. (n¼ 6). *Po0.05 vs. Group Ctrl;#Po0.05 vs. Group D. DCLF-elicited injury to both proximal and distal tubular cells. TUNEL-labeled nuclei were revealed as bright spotsin the renal cortex in kidneys from DCLF-treated mice. TDZD-8 significantly attenuated both tubular necrosis and apoptosis in mousekidneys.

Kidney International (2012) 81, 662–673 663

H Bao et al.: Role of GSK3b in NSAID-induced AKI o r i g i n a l a r t i c l e

TDZD-8 inhibits GSK3b activity and induces cortical COX-2in DCLF-injured kidneys

In DCLF-injured kidneys, enhanced GSK3b activity wasdemonstrated by reduced inhibitory phosphorylation ofGSK3b at serine 9. TDZD-8, a highly selective small-moleculeinhibitor of GSK3b, induced prominent phosphorylation ofGSK3b in both control and DCLF-injured kidneys, denotingthe inhibition of GSK3b kinase activity (Figure 2a). Oneimportant mechanism of NSAID-induced kidney injury isthe inhibition of COX activity in the kidney. Indeed, PGE2

levels in the renal cortices were significantly decreased inDCLF-injured kidneys (Figure 2c). The compensatoryinduction of renal cortical COX-2 was observed in DCLF-injured mice, and this was augmented following TDZD-8treatment (Figure 2b). This is consistent with previousreports in other animal models that GSK3b inhibitioninduces COX-2 expression.9,10 The decreased PGE2 levelsin the renal cortex were partially abrogated after TDZD-8treatment (Figure 2c).

Renal protective effects of GSK3b inhibition on DCLF-inducedAKI is largely preserved in COX-2-null mice

To determine whether the beneficial effects of GSK3b inhi-bition were entirely mediated by the induction of COX-2expression, the COX-2-knockout mice were administered anephrotoxic dose of DCLF following pretreatment with eitherTDZD-8 or vehicle. As shown in Figure 3, the renoprotectiveefficacy of TDZD-8 was slightly but significantly blunted inthe COX-2-null mice, as evidenced by the difference in thetubular injury scores, as well as in the serum creatinine levels(Figure 3b and c) between the wild-type and the knockout

mice injured with DCLF and treated with TDZD-8,suggesting that COX-2 is involved in the mechanism ofkidney protection. However, TDZD-8 could still effectivelyameliorate the DCLF-induced AKI and acute tubular necrosisin COX-2-null mice, and the renoprotective effect ofTDZD-8 was actually largely preserved in COX-2-null mice,indicating that a COX-2/PGE2-independent pathway alsocontributes, at least in part, to kidney protection.

GSK3b inhibition suppresses MPT in DCLF-injured kidneys

Both persistent ischemia and direct DCLF toxicity can induceMPT, a critical mechanism causing cell death. WhetherTDZD-8 protects kidneys from DCLF-induced MPT wasstudied next. In isolated mitochondria, the capacity totolerate a calcium challenge is inversely correlated with thesusceptibility to the MPT. To assess the MPT in vivo, renalmitochondria were purified, and calcium-induced mitochon-drial swelling was measured as a decrease in spectrophoto-metric absorbance at 540 nm. As can be seen in Figure 4,mitochondria isolated from the kidneys of DCLF-treatedmice displayed a greater decrease in absorbance at 540 nmupon calcium overload, denoting an accelerated rate ofmitochondrial swelling and marked mitochondrial dysfunc-tion. TDZD-8 treatment significantly ameliorated this effectin a dose-dependent manner (Figure 4). More importantly,the extent of calcium-induced mitochondrial swelling wasnegatively correlated with both the number of apoptotic cellsin the kidneys and the severity of renal histological injury.These findings suggest that TDZD-8 protected renal cellsfrom apoptosis and necrosis primarily by inhibiting MPTin vivo.

a b cCtrl

*

**

*

*

*

#

# ##

#

#

Ctrl

GSK3β

p-GSK3β

p-GSK3β/ActinGSK3β/Actin

Actin

1.2

Den

sito

met

ric r

atio

Den

sito

met

ric r

atio

0.8

0.4

0.0

T D T-L+D T+D

TTDZD-8(mg/kg)

TDZD-8(mg/kg)

DCLF(mg/kg)

DCLF(mg/kg)

TDZD-8(mg/kg)

DCLF(mg/kg)

D T-L+D T+D Ctrl T D T-L+D T+D Ctrl0

1

2

3

4

5

PG

E2

leve

ls in

cor

tex

(ng/

g)

–

–

5

–

–

200

1 5

200200

T D T-L+D T+D

Ctrl

COX-1

COX-1/Actin

COX-2

COX-2/Actin

Actin

1.2

0.8

0.4

0.0

T D T-L+D T+D

–

– –

– –

– –

–1

200 200200

5 5 5 51

200 200 200

Figure 2 | Effects of 4-benzyl-2-methyl-1,2,4-thiadiazolidine-3,5-dione (TDZD-8) on glycogen synthase kinase-3b (GSK3b)phosphorylation and the levels of cyclooxygenase-2 (COX-2) and prostaglandin E2 (PGE2) in kidney cortices of 2-[(2,6-dichloro-phenyl)amino]phenyl-acetate (DCLF)-treated mice. (a) Western blot analysis of GSK3b and p-GSK3b in cortical homogenate. (b) Westernblot analysis of COX-1 and COX-2 in cortical homogenate. (c) Analysis of cortical PGE2 in cortical homogenate. Group Ctrl, control (vehicletreated); Group T, TDZD-8 (5 mg/kg); Group D, DCLF (200 mg/kg); Group T�LþD, TDZD-8 (1 mg/kg)þDCLF (200 mg/kg); Group TþD,TDZD-8 (5 mg/kg)þDCLF (200 mg/kg). *Po0.05 vs. Group Ctrl; #Po0.05 vs. Group D. COX-1 expression was maintained and COX-2expression was increased following treatment with DCLF. GSK3b inhibition led to a further increase in COX-2 expression in the renal corticesof DCLF-treated mice. Accordingly, PGE2 synthesis was increased in TDZD-8-treated mice.


o r i g i n a l a r t i c l e H Bao et al.: Role of GSK3b in NSAID-induced AKI

GSK3b inhibition improves the cell viability of renal tubularepithelial cells exposed to DCLF

To determine whether treatment with TDZD-8 amelioratedDCLF-induced AKI via a direct protective effect on renaltubular cells, cultured murine tubular epithelial (TKPT) cellswere exposed to DCLF at different doses for differentintervals in the presence or absence of TDZD-8. DCLFstimulation reduced cellular viability in a dose-dependent(Figure 5a) and time-dependent (Figure 5b) manner, asassessed by tetrazolium (MTT) assays, and this effect couldbe significantly counteracted by treatment with TDZD-8(Figure 5c). In contrast, in cells expressing S9A-GSK3b—anuninhibitable mutant of GSK3b in which serine 9 wasreplaced by alanine—the protective activity of TDZD-8 wasblunted (Figure 5d), suggesting that the inhibitory phos-phorylation of GSK3b is required for the protective effect ofTDZD-8.

GSK3b inhibition protects from DCLF-induced apoptosis andnecrosis in renal tubular epithelial cells

As shown in Figure 6, DCLF induced significant cell apop-tosis within 24 h. Treatment with TDZD-8 significantlydecreased the number of apoptotic cells, as detected byTUNEL staining, and the anti-apoptotic effect of TDZD-8was attenuated in cells expressing S9A-GSK3b, indicating theindispensable role of phosphorylation in the inhibition ofGSK3b. To assess necrosis, we undertook both a morpho-logical examination of propidium iodide staining of culturedcells (Figure 7a and b) and a biochemical assay for theenzymatic activity of cytoplasmic lactate dehydrogenase(Figure 7c), which is released into conditioned media uponcellular necrosis and rupture of the plasma membrane. DCLFstimulation augmented both lactate dehydrogenase activityand propidium iodide positivity, and this was substantiallyreduced by treatment with TDZD-8. In cells expressing theS9A-GSK3b mutant, the antinecrotic effect of TDZD-8 wassimilarly blunted, suggesting that inhibitory phosphory-lation of GSK3b is essential for the protection of the cells byTDZD-8.

Ladder WT WT KO KOa d

COX-2256 bp

6

5

4

3

2

1

0

Tub

ular

inju

rysc

ore

bP< 0.05

##

* *

TDZD-8 (mg/kg)DCLF (mg/kg)

– ––

D T+D5

200200

Ctrl

P< 0.05

* *

##

TDZD-8 (mg/kg)DCLF (mg/kg)

Ctrl––

D–

200

T+D5

200

2.0

1.5

1.0

0.5

0.0

2.5

SC

r (m

g/dl

)

cTDZD-8+DCLF

Ctrl

DCLF

WTCOX-2 –/–

COX-2 –/–WT

Figure 3 | Glycogen synthase kinase-3b (GSK3b) inhibitionprotects cyclooxygenase-2 (COX-2�/�) mice from 2-[(2,6-dichlorophenyl)amino]phenyl-acetate (DCLF)-induced renaltubular injury. (a) mRNA expression of COX-2 detected byreverse transcription-PCR (RT-PCR) in the kidneys of wild-type(WT) and COX-2-knockout (KO) mice. (b) Tubular injury score ofWT and COX-2 KO mice treated with DCLF (200 mg/kg) and/or4-benzyl-2-methyl-1,2,4-thiadiazolidine-3,5-dione (TDZD-8;5 mg/kg). (c) Serum creatinine (SCr) of WT and COX-2 KO micetreated with DCLF (200 mg/kg) and/or TDZD-8 (5 mg/kg).(d) Representative micrographs of hematoxylin and eosin stainingof kidney specimens from COX-2 KO mice treated with DCLF(200 mg/kg) and/or TDZD-8 (5 mg/kg). Ctrl, control (vehicletreated). *Po0.05 vs. control; #Po0.05 vs. DCLF-treated mice.DCLF produced injury to both proximal and distal tubular cells.This kidney-protective effect of TDZD-8 was blunted but stilllargely preserved in COX-2-null mice.

d

OD 540 change

–0.20 –0.15 –0.10 –0.05

P< 0.05

Apo

ptot

ic c

ell/m

m2

30

20

10

0

0.00

c

–0.20 –0.15 –0.10

OD 540 change

–0.05 0.00

4P< 0.05

3

1

Tub

ular

inju

ry s

core

0

2

b0.0

Ctrl T D

*

##

T-L+D T+D

0.1

Cha

nge

in O

D 5

40nm

–0.2

–0.3TDZD-8 (mg/kg)

DCLF (mg/kg) 2002002001–

–––

5 5

0.8a0.70.60.5

Group CtrlGroup TGroup DGroup T-L+DGroup T+D

OD

540

nm

0.40.30.2

0 4 8 12

Time (min)

16 20

Figure 4 | Effects of glycogen synthase kinase-3b(GSK3b) inhibition on the induction of the mitochondrialmembrane permeability transition (MPT) by 2-[(2,6-dichloro-phenyl)amino]phenyl-acetate (DCLF) in mouse kidneys.(a) Mitochondrial swelling was spectrophotometrically monitoredat 540 nm following induction with 20 mM CaCl2 for 20 min.(b) Change in absorbance at optical density (OD) 540 nmafter 20 min. (c) Correlation between decreased absorbanceat 540 nm and the renal tubular injury score. (d) Correlationbetween decreased absorbance at 540 nm and the numberof apoptotic cells in renal tissues. Group Ctrl, control (vehicletreated); Group T, 4-benzyl-2-methyl-1,2,4-thiadiazolidine-3,5-dione (TDZD-8; 5 mg/kg); Group D, DCLF (200 mg/kg);Group T�LþD, TDZD-8 (1 mg/kg)þDCLF (200 mg/kg); GroupTþD, TDZD-8 (5 mg/kg)þDCLF (200 mg/kg). *Po0.05 vs.Group Ctrl; #Po0.05 vs. Group D. The decrease in absorbanceat 540 nm was negatively correlated with the number of apoptoticcells and the renal tubular injury scores in mouse kidneys.Mitochondria that were isolated from kidneys treated withDCLF displayed an increased rate of mitochondrial swelling anda greater decrease in absorbance at 540 nm vs. the controls,whereas TDZD-8 treatment ameliorated this process in adose-dependent manner.



GSK3b inhibition attenuates DCLF-induced MPT in renaltubular epithelial cells

The mitochondrial dye tetramethylrhodamine methyl ester(TMRM) was used to monitor the onset of the MPT incultured TKPT cells. Under basal conditions, the lipophilicTMRM, which bears a delocalized positive charge, entersthe negatively charged mitochondria and accumulates in apotential-dependent manner along the inner membrane(Figure 8a). Upon MPT, the MPT pore opens, and theTMRM potentiometric dye no longer accumulates insidethe mitochondria, but is instead distributed throughout thecytosol. In our study, following DCLF stimulation, theoverall cellular fluorescence intensity dropped significantly(Figure 8a), denoting the onset of the MPT. TDZD-8successfully counteracted the DCLF-elicited quenching ofTMRM, and this effect was similarly blunted in cellsexpressing the mutant S9A-GSK3b (Figure 8a).

Calcium overload–induced mitochondrial swelling, asmeasured by a decrease in spectrophotometric absor-bance at 540 nm, was similarly used to detect the MPT(Figure 8b and c). There was a negative correlation between

the reduction in absorbance at 540 nm and the numberof apoptotic or necrotic cells (Figure 8c). Mitochondriathat were isolated from cells treated with DCLF displayedincreased swelling, whereas treatment with TDZD-8 attenu-ated this effect, which was similarly diminished in the

100 100

80

60

40

20

0

100

80

60

40

20

0Ctrl T D T+D T+D

1010– ––– –

10D

0 h 6 h

EV S9A

12 h 24 h

*

** *

**

*

*

#

#

**

80

60

40

% C

ell v

iabi

lity

% C

ell v

iabi

lity

% C

ell v

iabi

lity

% C

ell v

iabi

lity

20

0

100

80

60

40

20

00

200 200 200 200200 200 200 200

20105

DCLF (μM)

DCLF (μM)

TDZD-8 (μM)

DCLF (μM)TDZD-8 (μM)

DCLF (200 μM)

0 μM 100 μM 200 μM 400 μM

a b

c d

Figure 5 | Effects of glycogen synthase kinase-3b (GSK3b)inhibition on the viability of renal tubular epithelial cellsinjured with 2-[(2,6-dichlorophenyl)amino]phenyl-acetate(DCLF). (a) Viability of murine tubular epithelial (TKPT) cellsfollowing treatment with different doses of DCLF (0, 100, 200, and400mM) for 24 h. *Po0.05 vs. normal controls; n¼ 6. (b) Survivalrate of TKPT cells after treatment with 200 mM DCLF for differenttimes. *Po0.05 vs. the beginning; n¼ 6. (c) Survival rate of TKPTcells pretreated with different doses of 4-benzyl-2-methyl-1,2,4-thiadiazolidine-3,5-dione (TDZD-8; 0, 5, 10, and 20 mM) and thentreated with 200mM DCLF for 24 h. *Po0.05 vs. DCLF-treated cells;n¼ 6. (d) Survival rate of TKPT cells expressing either emptyvector or S9A-GSK3b after treatment with 200 mM DCLF, 10 mM

TDZD-8, or combined treatment. Ctrl, control; D, DCLF 200 mM; EV,empty vector–transfected TKPT cells; S9A, S9A-GSK3b-transfectedTKPT cells; T, TDZD-8 10mM. *Po0.05 vs. control cells; #Po0.05 vs.the value of empty vector–transfected cells treated with DCLF.Treatment with DCLF markedly decreased cell viability in a time-and dose-dependent manner, whereas treatment with TDZD-8improved survival rates (Po0.05). In contrast, the supportiveeffect of TDZD-8 was blunted in cells transfected with vectorscarrying S9A-GSK3b, a mutant of GSK3b that is unable to bephosphorylated at serine 9.

No TDZD-8a

b

No

DC

LFD

CLF

DC

LF

S9A

EV

Ctrl T

T+D

T+D

D

D

PROPIDIUM IODIDE + TUNEL STAIN

TDZD-8

20

*

#

#T

UN

EL-

posi

tive

cells

/mm

218

12

8

4

0Ctrl

TDZD-8 (μM)

DCLF (μM)

T D D

200

EV S9A

200 200 200

T+DT+D10 10– –10–

– –

Figure 6 | Effect of glycogen synthase kinase-3b (GSK3b)inhibition on the apoptosis of murine tubular epithelial(TKPT) cells treated with 2-[(2,6-dichlorophenyl)amino]phenyl-acetate (DCLF). (a) TKPT cells were treated as indicated,fixed, and exposed to TdT-mediated dUTP nick-end labeling(TUNEL) to assess apoptosis. (b) Quantitative assessment of cellapoptosis. Cells in a � 200 field were manually counted in 10fields per slide, with three replicates. Ctrl, control; D, DCLF 200 mM;EV, empty vector–transfected TKPT cells; S9A, S9A-GSK3b-transfected TKPT cells; T, 4-benzyl-2-methyl-1,2,4-thiadiazolidine-3,5-dione (TDZD-8) 10mM. *Po0.05 vs. control cells; #Po0.05 vs.empty vector–transfected cells treated with DCLF; n¼ 6. Therewas a significant increase in the number of apoptotic cellsfollowing exposure to DCLF vs. controls after 24 h (Po0.05).The number of apoptotic cells was decreased to approximatelyfive per field in cultures pretreated with TDZD-8, whereas theanti-apoptotic effect of TDZD-8 was attenuated in cells expressingS9A-GSK3b.



mitochondria of cells expressing S9A-GSK3b (Figure 8b).All of these observations imply that the inhibition of GSK3bprotects tubular cells from apoptosis and necrosis byinhibiting MPT pore opening in vitro.

GSK3b inhibition prevents DCLF-induced ATP reduction inrenal tubular epithelial cells

MPT pore opening allows the reentry of protons, thus by-passing adenosine triphosphate (ATP) synthase and prevent-ing the generation of ATP by the mitochondria. Consistentwith induction of the MPT, intracellular ATP levels weredrastically diminished in DCLF-treated TKPT cells (Figure 9).Treatment with TDZD-8 retained intracellular ATP levels,whereas this effect was not apparent in cells expressing S9A-GSK3b, implying that the inhibition of GSK3b protectstubular cells from death by preserving intracellular ATPin vitro.

DCLF enhances the activity of redox-sensitive GSK3b in TKPTcells

Upon exposure to DCLF, the production of reactive oxygenspecies (ROS) was significantly increased, concomitant withincreased GSK3b kinase activity and diminished inhi-bitory phosphorylation of GSK3b (Figure 10a, c, and d),which is consistent with previous findings that GSK3b is a

reduction-oxidation (redox)-sensitive kinase.13 Mitochondriaare a major source of ROS production, and two main sites ofO2� and H2O2 generation in the inner mitochondrial mem-

brane have been identified: complex I and complex III.To identify the source of ROS production in response toDCLF stimulation, rotenone and stigmatellin, inhibitors forrespiratory complexes I and III, respectively, were applied.Treatment with rotenone did not prevent the increase of ROSproduction or enhance GSK3b activity, denoting thatcomplex I is not the main site of DCLF-increased ROSproduction. Consistently, DCLF failed to produce a directeffect on the activity of reduced nicotinamide adeninedinucleotide (NADH) dehydrogenase (Figure 10b). In contrast,treatment with stigmatellin significantly suppressed the pro-duction of ROS and the activation of GSK3b (Figure 10a, c,and d), suggesting that mitochondrial ROS formation,particularly at respiratory chain complex III, is involved inDCLF-enhanced activity of redox-sensitive GSK3b.

GSK3b inhibition prevents cyclophilin D phosphorylationin the mitochondria of TKPT cells exposed to DCLF

To explore whether and how GSK3b directly regulates MPT,the expression of GSK3b in the mitochondria of TKPT cellswas examined. Fluorescent immunocytochemistry indicatedthat a significant amount of GSK3b staining colocalized with

No TDZD-8a b

c

No

DC

LFD

CLF

DC

LF

Ctrl

D

D

T+D

T+D

PROPIDIUM IODIDE STAIN

T

TDZD-850

40

30

Nec

rosi

s %

LDH

(fo

ld in

duct

ion)

20

10

0

2.5

2.0

1.5

0.5

0.0

TDZD-8 (μM)DCLF (μM)

1.0

CtrlTDZD-8 (μM)

DCLF (μM)

T D T+D

#

#

#

#*

*

T+D D–1010–10–

– – 200 200 200 200

Ctrl T D

200

EV

EV

200 200 200

S9A

S9A

T+D T+D D–1010–10–

– –

Figure 7 | Effects of glycogen synthase kinase-3b (GSK3b) inhibition on the necrosis of murine tubular epithelial (TKPT) cellsexposed to 2-[(2,6-dichlorophenyl)amino]phenyl-acetate (DCLF). (a) Necrotic cell death was assessed using the propidium iodideexclusion assay in live cells. (b) Quantitative assessment of cell necrosis by propidium iodide exclusion assay. Cells in a � 200 field weremanually counted in 10 fields per slide, with three replicates. (c) Necrotic cell death was assessed by lactate dehydrogenase (LDH) assayof the conditioned medium. Ctrl, control; D, DCLF 200 mM; EV, empty vector–transfected TKPT cells; S9A, S9A-GSK3b-transfected TKPT cells;T, 4-benzyl-2-methyl-1,2,4-thiadiazolidine-3,5-dione (TDZD-8) 10 mM. *Po0.05 vs. control cells; #Po0.05 vs. empty vector–transfected cellstreated with DCLF; n¼ 6. There was a significant increase in the number of necrotic cells and in LDH release in cultures exposed to DCLF vs.controls after 24 h (Po0.05). TDZD-8 treatment decreased the number of necrotic cells and the LDH level, whereas the effects of TDZD-8were attenuated in cells transfected with S9A-GSK3b.



mitochondrial staining, in agreement with the specific,subcellular distribution of GSK3b in the mitochondria(Figure 11a). To corroborate the morphological findings,immunoprecipitation was performed and revealed thatGSK3b coprecipitated with cyclophilin D (Figure 11b),indicating that GSK3b physically interacts with cyclophilinD—a component of the MPT pore complex.

To further explore the mechanism by which GSK3bregulates cyclophilin D activation, the cyclophilin D amino-acid sequence was analyzed. Residues Ser118 and Ser190 ofmitochondrial cyclophilin D reside in consensus motifs (SXXXS)for phosphorylation by GSK3b. As shown in Figure 11c,DCLF treatment significantly increased the phosphorylationof cyclophilin D. Conversely, inhibition of GSK3b by TDZD-8 suppressed the phosphorylation of cyclophilin D, whereasthis effect was diminished in cells transfected with S9A-GSK3b. Together, these results suggest that GSK3b inhibitionstabilizes the MPT by suppressing cyclophilin D phosphor-ylation in the mitochondria of TKPT cells exposed to DCLF.

DISCUSSION

NSAIDs are prescribed to millions worldwide for the treat-ment of osteoarthritis, rheumatoid arthritis, and muscle

No TDZD-8a

c

b

Ctrl

Ctrl0.0

0.0

–0.3

–0.3

30

20

10

Apo

ptot

ic c

ells

/mm

2

Apoptotic cells/mm2

Necrosis%

Nec

rosi

s%

0

– – –101010

200

EV S9A

50

40

30

20

10

0

200 200 200––

TDZD-8 (µM)

DCLF (µM)

–0.2

–0.2

Change in OD 540 nm

–0.1

–0.1C

hang

e in

OD

540

nm

T D

#

#*

T+D T+D D

D

D T+D

T+D

T

No

DC

LFD

CLF

DC

LF

S9A

EV

TDZD-8

Figure 8 | Effects of glycogen synthase kinase-3b (GSK3b) inhibition on mitochondrial membrane permeability transition (MPT)in murine tubular epithelial (TKPT) cells exposed to 2-[(2,6-dichlorophenyl)amino]phenyl-acetate (DCLF). (a) Tetramethyl-rhodamine methyl ester (TMRM) was used to detect the onset of the MPT by fluorescence microscopy. (b) Mitochondrial swelling wasspectrophotometrically monitored at 540 nm in mitochondria that had been induced with 20 mM CaCl2 for 20 min. Ctrl, control; D, DCLF200mM; EV, empty vector–transfected TKPT cells; S9A, S9A-GSK3b-transfected TKPT cells; T, 4-benzyl-2-methyl-1,2,4-thiadiazolidine-3,5-dione(TDZD-8) 10mM. *Po0.05 vs. control cells; #Po0.05 vs. empty vector–transfected cells treated with DCLF; n¼ 6. (c) Correlation between thedecrease in absorbance at 540 nm and the number of apoptotic or necrotic cells. Mitochondria that were isolated from cells treated withDCLF displayed quenching of TMRM and decreased absorbance at 540 nm compared with mitochondria from control cells. The extent ofthe decrease in absorbance at 540 nm was correlated with both the number of apoptotic and necrotic cells. Pretreatment with TDZD-8inhibited the quenching of TMRM and restored the capacity of mitochondria that had been subjected to DCLF to accumulate calcium. All ofthese effects were blunted in cells expressing S9A-GSK3b.

100

*

#

#

80

60

40

20

Intr

acel

lula

r AT

P (

% c

ontr

ol)

0Ctrl

TDZD-8 (μM)

DCLF (μM)

–

–

T10

–

D–

200

EV S9A

D–

200

T+D10

200

T+D10

200

Figure 9 | Effect of glycogen synthase kinase-3b (GSK3b)inhibition on intracellular adenosine triphosphate (ATP)content in murine tubular epithelial (TKPT) cells treated with2-[(2,6-dichlorophenyl)amino]phenyl-acetate (DCLF). ATPconcentration was measured using the luciferase method. Ctrl,control; D, DCLF 200 mM; EV, empty vector–transfected cells; S9A,S9A-GSK3b-transfected TKPT cells; T, 4-benzyl-2-methyl-1,2,4-thiadiazolidine-3,5-dione (TDZD-8) 10 mM. *Po0.05 vs. controlcells; #Po0.05 vs. empty vector–transfected cells treated withDCLF; n¼ 6. Treatment with DCLF significantly decreased ATPproduction. Treatment with TDZD-8 increased intracellular ATP,and the effect was blunted in cells transfected with S9A-GSK3b.



pain.14 In 2008 alone, over 107,000 NSAID-related toxicitycases were reported, more than 14,000 cases of whichrequired treatment in health-care facilities. DCLF is one ofthe most frequently prescribed NSAID. A literature search onDCLF-induced renal toxicity in humans identified a numberof cases, most of them associated with acute renal failure andsome of which were fatal.1,2

Basal PGs maintain normal glomerular perfusion andglomerular filtration rate. The inhibition of PG synthesis byDCLF leads to renal ischemia, a decline in glomerularhydraulic pressure, and acute renal failure. GSK3b, a consti-tutively active protein kinase with an important regulatoryrole in cell signaling, differentiation, gene expression, andapoptosis, has been implicated in the pathogenesis ofmultiple kidney diseases, both chronic and acute.15–19 In thisstudy, we found that the inhibition of GSK3b activity byTDZD-8, a small-molecule inhibitor of GSK3b, amelioratedDCLF-induced AKI. This was associated with suppressedGSK3b activity, induced inhibitory phosphorylation ofGSK3b, and increased COX-2 expression in the renal cortex.Consistently, PGE2 concentrations in the renal cortex were

significantly elevated. Our finding is in agreement withprevious reports indicating that GSK3b inhibition mayinduce COX-2 expression via the negative regulation ofextracellular signal-regulated kinases.20 Of note, the renalprotective effect of GSK3b inhibition by TDZD-8 wasblunted but still largely preserved in COX-2-null mice,suggesting that GSK3b inhibition may attenuate renal injurythrough mechanisms that are beyond COX-2 induction.

In addition to the induction of acute ischemic kidneyinjury through the disturbance of PG homeostasis in thekidney, treatment with DCLF can impinge on renal tubularepithelial cells directly. Significantly, tubular cell mito-chondria are a prime target of DCLF.4 We found thatmitochondria treated with DCLF were more likely toundergo mitochondrial swelling induced by calcium andwere more susceptible to the shift of TMRM through theopened pores. MPT, an increase in the permeability of themitochondrial membrane, is a common pathway leading tocell death. DCLF-induced apoptosis and necrosis weredetected in vivo and in vitro, consistent with previousreports.8,21 40OH-DCLF and 5OH-DCLF are the major urinemetabolites of DCLF in rodents and humans.22 DCLF, 40OH-DCLF, and 5OH-DCLF induce concentration-dependentapoptosis, and at equimolar concentrations the greatestpro-apoptotic activity was produced by 5OH-DCLF.7

The MPT is under orchestrated regulation by a myriad ofcell signaling transducers, including GSK3b. Recent studieshave demonstrated that GSK3b inhibition has a central rolein the transduction of protective signals by raising the MPTinduction threshold.11,12,23 The phosphorylation of GSK3b atserine 9 raised the MPT threshold and protected cells byinhibiting the signaling cascades associated with the induc-tion of the MPT.24,25 In this study, the inhibition of GSK3bactivity by TDZD-8 successfully prevented the swelling ofmitochondria that were isolated from DCLF-injured cells andtissues, and this effect was attenuated in mitochondria fromcells expressing mutant GSK3b that is not able to be inhibitedvia phosphorylation at serine 9, demonstrating that GSK3binhibition diminishes DCLF-triggered MPT pore opening.Our findings are in agreement with a previous study byPlotnikov et al.26 in which treatment with lithium ion, aninhibitor of GSK3b, restored the mitochondrial membranepotential in tubular cells from rat kidneys that were subjectedto ischemic reperfusion. In the present study, in murineTKPT cells, DCLF induced a strong ROS burst and oxidativestress. Although ROS can cause cell destruction by massivelipid peroxidation, in most cases, ROS modulates signaltransduction pathways by affecting redox-sensitive enzymes,organelles, and signaling transducers, such as GSK3b. Indeed,the DCLF-elicited ROS burst was associated with theenhanced activation of GSK3b, and both were abrogated bytreatment with stigmatellin but not rotenone. This suggeststhat DCLF can interfere with the mitochondrial electrontransport chain, leading to augmented activation of theredox-sensitive GSK3b kinase. Moreover, a significant sub-cellular distribution of GSK3b was found to be colocalized

3a b

dc

*

*

*

#

#

#

#

2

1

1

2

1.00.80.60.40.20.0

RO

S (

fold

indu

ctio

n)

GS

K3β

act

ivity

(fol

d in

duct

ion)

NA

DH

fluo

resc

ence

inte

nsity

(fo

ld in

duct

ion)

Den

sito

met

ric r

atio

0

0

3

2

1

0Ctrl Ctrl

DCLF (μM) DCLF (μM)

p-GSK3βGSK3β

Rot (μM) Rot (μM)

Stig (μM)

DCLF (μM)Rot (μM)Stig (μM)

DCLF DCLF200 200– –

– –

–

1–

–

–––

–

–

200

––

1 –

– 1

200

1–

200

200

–1

200

Rot RotStig

Ctrl DCLF Rot StigDCLF (μM)

Rot (μM)Stig (μM)

–––

200

–– 1

–

200–1

200Ctrl DCLF Rot Stig

Figure 10 | Effect of 2-[(2,6-dichlorophenyl)amino]phenyl-acetate (DCLF) on reduction-oxidation (redox)-sensitiveglycogen synthase kinase-3b (GSK3b) in murine tubularepithelial (TKPT) cells. TKPT cells were treated with 200 mM DCLFor pretreated with 1 mM rotenone (Rot) or 1 mM stigmatellin (Stig).(a) Reactive oxygen species (ROS) formation was evaluated by theoxidation of 20,70-dichlorodihydrofluorescein diacetate (DCFH-DA)to the fluorescent DCF in TKPT cells. (b) NADH dehydrogenaseactivity in mitochondrial extracts was compared by thefluorescence of reduced nicotinamide adenine dinucleotide(NADH) at 5 min after the start of assay. (c) GSK3b activity wasassayed by measuring the incorporation of 32P into the substrate.(d) Western blot analysis of GSK3b and p-Ser9-GSK3b expressionin cells. *Po0.05 vs. control cells; #Po0.05 vs. cells treated withDCLF; n¼ 6. DCLF treatment induced ROS production and GSK3bactivation in TKPT cells but did not affect NADH dehydrogenaseactivity. Stigmatellin significantly suppressed ROS production andGSK3b activation in DCLF-treated TKPT cells, whereas rotenonehad no effect.



with the mitochondria, suggesting a potential regulatory rolefor GSK3b in mitochondrial function. Thus far, the only MPTcomponents to be identified are the peripheral benzodiazepinereceptor, located in the mitochondrial outer membrane, andcyclophilin D, localized to the matrix.27,28 Residues Ser118 andSer190 of mitochondrial cyclophilin D reside in GSK3bphosphorylation consensus motifs. DCLF treatment signifi-cantly increased the phosphorylation of cyclophilin D, whichmay trigger the opening of the end-effector MPT pore.Permeability of the mitochondrial inner membrane allowssolutes with a molecular weight of o1500 Da to leak from themitochondria, and the subsequent breaking of mitochondrialouter membrane leads to the release of proteins and celldeath.28,29 Conversely, inhibition of GSK3b successfully pre-vented the phosphorylation of cyclophilin D in the mitochon-dria, which may lead to the stabilization of the MPT.

Mitochondria have a crucial role in maintaining ATP-dependent cellular processes.30 MPT pore opening allows themassive reentry of protons into the mitochondrial matrix,thus bypassing ATP synthase and preventing ATP synthesis.31

In this study, we observed a significant reduction in ATPlevels in DCLF-injured cells, consistent with the findings ofboth Mingatto et al.5 and Ng et al.32 that DCLF depressed therate of ATP synthesis. The progression to necrotic or

apoptotic cell death depends, in part, on the effect of theopening of the MPT pore on cellular ATP levels.33 If manypores open, the major drop in cellular ATP levels triggerscellular swelling and rupture of the plasma membrane, andthus necrotic cell death ensues. If ATP levels are at leastpartially maintained, the MPT is followed by apoptosis.GSK3b inhibition has been shown to be a successful thera-peutic strategy to prevent hepatic mitochondrial dysfunctionand improve the preservation of cellular ATP contentfollowing ischemic reperfusion injury.34 In this study, weobserved that, following the inhibition of GSK3b by TDZD-8, DCLF-elicited MPT was suppressed and intracellular ATPlevels were restored, leading to reduced tubular cell death.

In conclusion, TDZD-8, a highly selective inhibitor ofGSK3b, attenuates DCLF-induced AKI, suppresses tubularcell death, and improves kidney function by inducing renalcortical COX-2 expression and by inhibiting MPT in aDCLF-induced nephrotoxicity model (Figure 12). Ourfindings suggest that the inhibition of GSK3b by either novelselective small-molecule inhibitors or existing FDA (Foodand Drug Administration)-approved drugs with GSK3b inhi-bitory activity, such as lithium and valproate, represents anovel therapeutic strategy for AKI, particularly for NSAID-induced AKI.

EV

IP:CyP-D

IgG

WB:

GSK3β

α-CyP-DIP:

Ctrl T D DT+D T+D

Ctrl––

T10TDZD-8 (μM)

DCLF (μM) –

D–

200

EV S9A

D–

200

T+D10200

T+D10200

p-CyP-D

a c

b

CyP-D

1.2

GS

K3β

Mito

Gre

enM

erge

0.8

0.4 *

*#

#

Den

sito

met

ric r

atio

0.0

S9A

Figure 11 | Glycogen synthase kinase-3b (GSK3b) colocalizes with cyclophilin D, a component of the mitochondrial membranepermeability transition (MPT) pore complex, and regulates its activation. (a) Representative images of the colocalization of GSK3b withmitochondria in murine tubular epithelial (TKPT) cells. (b) TKPT cell mitochondria were purified, immunoprecipitated with anti-cyclophilin Dantibody, and immunoblotted with anti-GSK3b antibody. (c) TKPT cell mitochondria were separated, immunoprecipitated with anti-cyclophilin D antibody, and immunoblotted with anti-phosphorylated serine antibody. Ctrl, control; D, 2-[(2,6-dichlorophenyl)amino]phenyl-acetate (DCLF) 200mM; CyP-D, cyclophilin D; EV, empty vector–transfected cells; IP, immunoprecipitation; S9A, S9A-GSK3b-transfected TKPTcells; T, 4-benzyl-2-methyl-1,2,4-thiadiazolidine-3,5-dione (TDZD-8) 10 mM; WB, western blot. *Po0.05 vs. control cells; #Po0.05 vs. emptyvector–transfected cells treated with DCLF; n¼ 6. GSK3b colocalized with cyclophilin D in the mitochondria. DCLF treatment significantlyincreased the phosphorylation of cyclophilin D. Inhibition of GSK3b by TDZD-8 suppressed the phosphorylation of cyclophilin D, whereasthis effect was diminished in cells transfected with S9A-GSK3b.



MATERIALS AND METHODSAnimal experimental designAnimal experimental studies were approved by the institution’sAnimal Care and Use Committee and they conform to theUSDA (United States Department of Agriculture) regulations andthe NIH (National Institutes of Health) guidelines for humanecare and use of laboratory animals. Drugs were given as previouslydescribed.8,35 BALB/c mice aged 6–8 weeks were randomlyassigned to one of the five groups, and killed 24 h after DCLFadministration:

Group C (n¼ 6), control mice received vehicle alone;Group T (n¼ 6), mice were only treated with selective GSK3b

inhibitor TDZD-8 (5 mg/kg, intravenous, dissolved in 10% dimethylsulfoxide);

Group D (n¼ 6), mice were subjected to DCLF (200 mg/kg)injury by oral gavage and vehicle was given 1 h before DCLF injury;

Group T�LþD (n¼ 6), mice were subjected to DCLF(200 mg/kg) injury by oral gavage. Low-dose TDZD-8 (1 mg/kg,

intravenous, dissolved in 10% dimethyl sulfoxide) was given 1 hbefore DCLF injury; and

Group TþD (n¼ 6), mice were subjected to DCLF (200 mg/kg)injury by oral gavage. TDZD-8 (5 mg/kg, intravenous, dissolved in10% dimethyl sulfoxide) was given 1 h before DCLF injury.

Congenic COX-2-knockout mice on a BALB/c backgroundwere bred by backcrossing the COX-2-knockout mice on a mixed129/C57 background, which were originally obtained from theJackson Laboratory (Bar Harbor, ME), with the inbred BALB/c micefor more than 10 generations. Congenic homozygous COX-2-nullmice were bred by brother/sister mating of heterozygous animals ofthe tenth generation. Genotyping was performed as previouslydescribed.36

Renal pathologyKidney sections were prepared and stained as previously described.37

Acute tubular injury was accessed using semiquantitative measure-ments according to the proportion relative to the total sectionarea and classified as follows: 0 (nil), 1 (o25%), 2 (25–50%),3 (50–75%), and 4 (475% of tubules).

Western immunoblot analysis and immunoprecipitationWestern immunoblot analysis was performed as previously de-scribed.15 The antibodies against GSK3b, p-GSK3b, COX-1, andCOX-2 were purchased from Santa Cruz Biotechnology (Santa Cruz,CA). For detection of phosphorylated cyclophilin D, cyclophilin Dantibody (Santa Cruz Biotechnology) was used as the immuno-precipitation antibody and the antibody against phosphorylatedserine (Santa Cruz Biotechnology) was used to probe the immuno-precipitation products by immunoblot analysis.

Measurement of PGE2 in kidney cortexThe PGE2 was measured using an enzymatic immunoassay kit fromCayman Chemical (Ann Arbor, MI) according to the manufacturer’sinstruction.

Reverse transcription-PCRReverse transcription-PCR was performed as previously describedusing specific primers (COX-2 sense, 50-GTGGAAAAACCTCGTCCAGA-30, COX-2 antisense, 50-TGATGGTGGCTGTTTTGGTA-30).38

Cell culture and plasmid transfectionTKPT cells were grown in Dulbecco’s modified Eagle’s medium/F12that contained 5% fetal bovine serum. The eukaryotic expressionvectors encoding the uninhibitable mutant (S9A-GSK3b-HA/pcDNA3) were provided by Dr Johnson (Birmingham, AL),39 andwere transfected as previously described.15 Immunofluorescentstaining revealed that 475% of the cells expressed the hemagglutinin-tagged constructs 24 h after transfection. Cells were then subjected todifferent treatments and assessed by MTT viability assay.40

Measurement of cell apoptosisTUNEL staining was performed on fixed tissue sections or cellcultures with a cell apoptosis detection kit (Roche Applied Science,Indianapolis, IN) according to the manufacturer’s instructions.

Measurement of cell necrosisNecrotic cell death was assessed by the propidium iodide exclusionassay and lactate dehydrogenase release in the medium as previouslydescribed.41

NSAIDs(DCLF)

GSK3βinhibition

GSK3βinhibition

Vaso-constriction

Renalischemia

MPT

Tubular apoptosisand necrosis

AKI/NSAIDsnephrotoxicity

↑ COX-2expression

↓ COXactivity

↓ PGE2

↑ ROS

↑ p-CyP-D

↑ Redox-sensitiveGSK3β

Figure 12 | Glycogen synthase kinase-3b (GSK3b) inhibitionprevents nonsteroidal anti-inflammatory drug (NSAID)-induced acute kidney injury by dual mechanisms.Overinhibition of prostaglandin (PG) synthesis by NSAIDs leads torenal ischemia, a decline in glomerular hydraulic pressure, andacute renal failure. The 2-[(2,6-dichlorophenyl)amino]phenyl-acetate (DCLF) is also able to injure renal tubular epithelial cellsdirectly by triggering reactive oxygen species (ROS) production,enhancing the activity of reduction-oxidation (redox)-sensitiveGSK3b, and thereby promoting cyclophilin D phosphorylation andthe ensuing mitochondrial membrane permeability transition(MPT) pore opening. GSK3b inhibition prevents NSAID-inducedacute kidney injury via both the induction of corticalcyclooxygenase-2 (COX-2) and the inhibition of the MPT.AKI, acute kidney injury; CyP-D, cyclophilin D.



MPT assayMitochondria were isolated from kidney cells as previouslydescribed.4,42 The protein concentration was determined withbovine serum albumin as the standard. Mitochondrial swellingwas estimated on the basis of the decrease in the absorbance ofmitochondria (1.0 mg protein) at 540 nm in 1 ml of a mediumcontaining 125 mmol/l sucrose, 65 mmol/l KCl, 5 mmol/l succinate,5 mM rotenone, 20 mM CaCl2, and 10 mmol/l Hepes-KOH, pH 7.2,at 30 1C.

Fluorescent analysis of MPTMPT pore opening in cultured cells was assessed using TMRM(Sigma, St Louis, MO, USA). In brief, after different treatments,TMRM (100 nM) was added to the culture and incubated at 37 1Cfor 15 min.43 Results were interpreted using fluorescence microscopy.

ATP assayATP content in cells was measured by the luciferase method infreshly prepared cellular lysates using the ATP bioluminescence assaykit provided by Roche Applied Science according to the manufac-turer’s instructions.

Measurement of NADH dehydrogenase activityThe rotenone-sensitive reduction of decylubiquinone was measuredfollowing the procedure reported for respiratory complex I activityusing a freeze–thawed mitochondrial extract (containing 0.3 mgprotein).32 NADH fluorescence intensity was monitored at theexcitation/emission of 352/464 nm in the absence or presence of200mM DCLF or 1 mM rotenone, which served as a positive control.

ROS production assayROS production in isolated cells was measured using 20,70-dichloro-dihydrofluorescein diacetate as previously described.44

GSK3b kinase activity assayGSK3b kinase activity was determined using an assay kitcommercially available from Sigma according to the manufacturer’sinstructions.

Fluorescent immunocytochemistryTKPT cells were stained with MitoTracker Green FM (150 nM,Molecular Probes, Carlsbad, CA) for 20 min, and then fixed with 4%paraformaldehyde and counterstained with anti-GSK3b antibodiesas previously described.45

Statistical analysesFor immunoblot analysis, bands were scanned and the integratedpixel density was determined using a densitometer and an NIHimage analysis program (ImageJ, NIH, Bethesda, MD). All data areexpressed as mean±s.d. Statistical analysis of the data from multiplegroups was performed by one-way analysis of variance followed bythe Student–Newman–Kuels test. Data from two groups werecompared using t-test. The Po0.05 was considered significant.

DISCLOSUREAll the authors declared no competing interests.

ACKNOWLEDGMENTSHB is an International Society of Nephrology (ISN) Fellow and arecipient of the ISN fellowship. This work was made possible bythe funding from the National Basic Research Program of China 973

Program No. 2012CB517600, the ISN Sister Renal Center Program, theDevelopmental Grant from Brown Medical School Department ofMedicine, and the National Institutes of Health grant R01DK092485.We are indebted to Drs Rifai and Yang for stimulating discussions andhelpful suggestions.

REFERENCES1. Rubio GJA, Tellez MMJ. [Acute renal failure and nephrotic syndrome

associated with treatment with diclofenac]. Rev Clin Esp 1992; 191:289–290.

2. Cicuttini L, Colatutto A, Pellegrini MA et al. [Acute reversible renalinsufficiency during treatment with diclofenac]. Clin Ter 1989; 128:81–86.

3. Masubuchi Y, Nakayama S, Horie T. Role of mitochondrial permeabilitytransition in diclofenac-induced hepatocyte injury in rats. Hepatology2002; 35: 544–551.

4. Uyemura SA, Santos AC, Mingatto FE et al. Diclofenac sodium andmefenamic acid: potent inducers of the membrane permeabilitytransition in renal cortex mitochondria. Arch Biochem Biophys 1997; 342:231–235.

5. Mingatto FE, Santos AC, Uyemura SA et al. In vitro interaction ofnonsteroidal anti-inflammatory drugs on oxidative phosphorylation of ratkidney mitochondria: respiration and ATP synthesis. Arch BiochemBiophys 1996; 334: 303–308.

6. Gomez-Lechon MJ, Ponsoda X, O’Connor E et al. Diclofenac inducesapoptosis in hepatocytes. Toxicol In Vitro 2003; 17: 675–680.

7. Gomez-Lechon MJ, Ponsoda X, O’Connor E et al. Diclofenacinduces apoptosis in hepatocytes by alteration of mitochondrialfunction and generation of ROS. Biochem Pharmacol 2003; 66:2155–2167.

8. Hickey EJ, Raje RR, Reid VE et al. Diclofenac induced in vivo nephrotoxicitymay involve oxidative stress-mediated massive genomic DNAfragmentation and apoptotic cell death. Free Radic Biol Med 2001; 31:139–152.

9. Kotnik P, Nielsen J, Kwon TH et al. Altered expression of COX-1, COX-2,and mPGES in rats with nephrogenic and central diabetes insipidus.Am J Physiol Renal Physiol 2005; 288: F1053–F1068.

10. Kwon TH. Dysregulation of renal cyclooxygenase-2 in rats with lithium-induced nephrogenic diabetes insipidus. Electrolyte Blood Press 2007; 5:68–74.

11. Juhaszova M, Zorov DB, Kim SH et al. Glycogen synthase kinase-3betamediates convergence of protection signaling to inhibit themitochondrial permeability transition pore. J Clin Invest 2004; 113:1535–1549.

12. Xi J, Wang H, Mueller RA et al. Mechanism for resveratrol-inducedcardioprotection against reperfusion injury involves glycogen synthasekinase 3beta and mitochondrial permeability transition pore. Eur JPharmacol 2009; 604: 111–116.

13. Wang SH, Shih YL, Kuo TC et al. Cadmium toxicity toward autophagythrough ROS-activated GSK-3beta in mesangial cells. Toxicol Sci 2009;108: 124–131.

14. Murray MD, Brater DC. Renal toxicity of the nonsteroidal anti-inflammatory drugs. Annu Rev Pharmacol Toxicol 1993; 33: 435–465.

15. Gong R, Rifai A, Ge Y et al. Hepatocyte growth factor suppressesproinflammatory NFkappaB activation through GSK3beta inactivation inrenal tubular epithelial cells. J Biol Chem 2008; 283: 7401–7410.

16. Gong R, Rifai A, Dworkin LD. Activation of PI3K-Akt-GSK3beta pathwaymediates hepatocyte growth factor inhibition of RANTES expression inrenal tubular epithelial cells. Biochem Biophys Res Commun 2005; 330:27–33.

17. Nelson PJ, Cantley L. GSK3beta plays dirty in acute kidney injury. J Am SocNephrol 2010; 21: 199–200.

18. Ge Y, Si J, Tian L et al. Conditional ablation of glycogen synthase kinase3beta in postnatal mouse kidney. Lab Invest 2011; 91: 85–96.

19. Wang Z, Havasi A, Gall J et al. GSK3beta promotes apoptosis after renalischemic injury. J Am Soc Nephrol 2010; 21: 284–294.

20. Wang Q, Zhou Y, Wang X et al. Glycogen synthase kinase-3 is a negativeregulator of extracellular signal-regulated kinase. Oncogene 2006; 25:43–50.

21. Ng LE, Halliwell B, Wong KP. Nephrotoxic cell death by diclofenac andmeloxicam. Biochem Biophys Res Commun 2008; 369: 873–877.

22. Stierlin H, Faigle JW. Biotransformation of diclofenac sodium (Voltaren) inanimals and in man. II. Quantitative determination of the unchangeddrug and principal phenolic metabolites, in urine and bile. Xenobiotica1979; 9: 611–621.



23. Zhai P, Sadoshima J. Overcoming an energy crisis?: an adaptive role ofglycogen synthase kinase-3 inhibition in ischemia/reperfusion. Circ Res2008; 103: 910–913.

24. Yao LL, Huang XW, Wang YG et al. Hydrogen sulfide protectscardiomyocytes from hypoxia/reoxygenation-induced apoptosis bypreventing GSK-3beta-dependent opening of mPTP. Am J Physiol HeartCirc Physiol 2010; 298: H1310–H1319.

25. Zorov DB, Juhaszova M, Yaniv Y et al. Regulation and pharmacology ofthe mitochondrial permeability transition pore. Cardiovasc Res 2009;83: 213–225.

26. Plotnikov EY, Kazachenko AV, Vyssokikh MY et al. The role ofmitochondria in oxidative and nitrosative stress during ischemia/reperfusion in the rat kidney. Kidney Int 2007; 72: 1493–1502.

27. Sileikyte J, Petronilli V, Zulian A et al. Regulation of the inner membranemitochondrial permeability transition by the outer membranetranslocator protein (peripheral benzodiazepine receptor). J Biol Chem2011; 286: 1046–1053.

28. Baines CP, Kaiser RA, Purcell NH et al. Loss of cyclophilin D revealsa critical role for mitochondrial permeability transition in cell death.Nature 2005; 434: 658–662.

29. Green DR, Reed JC. Mitochondria and apoptosis. Science 1998; 281: 1309–1312.30. Bereiter-Hahn J, Jendrach M. Mitochondrial dynamics. Int Rev Cell Mol Biol

2010; 284: 1–65.31. Pessayre D, Mansouri A, Berson A et al. Mitochondrial involvement in

drug-induced liver injury. Handb Exp Pharmacol 2010; 196: 311–365.32. Ng LE, Vincent AS, Halliwell B et al. Action of diclofenac on kidney

mitochondria and cells. Biochem Biophys Res Commun 2006; 348: 494–500.33. Lemasters JJ, Qian T, Bradham CA et al. Mitochondrial dysfunction in

the pathogenesis of necrotic and apoptotic cell death. J BioenergBiomembr 1999; 31: 305–319.

34. Varela AT, Simoes AM, Teodoro JS et al. Indirubin-30-oxime preventshepatic I/R damage by inhibiting GSK-3beta and mitochondrialpermeability transition. Mitochondrion 2010; 10: 456–463.

35. Dugo L, Abdelrahman M, Murch O et al. Glycogen synthase kinase-3betainhibitors protect against the organ injury and dysfunction causedby hemorrhage and resuscitation. Shock 2006; 25: 485–491.

36. Alam I, Warden SJ, Robling AG et al. Mechanotransduction in bonedoes not require a functional cyclooxygenase-2 (COX-2) gene. J BoneMiner Res 2005; 20: 438–446.

37. Gong R, Rifai A, Tolbert EM et al. Hepatocyte growth factor amelioratesrenal interstitial inflammation in rat remnant kidney by modulatingtubular expression of macrophage chemoattractant protein-1 andRANTES. J Am Soc Nephrol 2004; 15: 2868–2881.

38. Ma X, Yang Q, Wilson KT et al. Promoter methylation regulatescyclooxygenase expression in breast cancer. Breast Cancer Res 2004; 6:R316–R321.

39. Cho JH, Johnson GV. Primed phosphorylation of tau at Thr231 byglycogen synthase kinase 3beta (GSK3beta) plays a critical role inregulating tau’s ability to bind and stabilize microtubules. J Neurochem2004; 88: 349–358.

40. Gong R, Rifai A, Dworkin LD. Anti-inflammatory effect of hepatocytegrowth factor in chronic kidney disease: targeting the inflamed vascularendothelium. J Am Soc Nephrol 2006; 17: 2464–2473.

41. Zhuang S, Kinsey GR, Yan Y et al. Extracellular signal-regulated kinaseactivation mediates mitochondrial dysfunction and necrosis induced byhydrogen peroxide in renal proximal tubular cells. J Pharmacol Exp Ther2008; 325: 732–740.

42. Stoyanovsky DA, Vlasova II, Belikova NA et al. Activation of NO donorsin mitochondria: peroxidase metabolism of (2-hydroxyamino-vinyl)-triphenyl-phosphonium by cytochrome c releases NO and protectscells against apoptosis. FEBS Lett 2008; 582: 725–728.

43. Nieminen AL, Saylor AK, Tesfai SA et al. Contribution of themitochondrial permeability transition to lethal injury after exposureof hepatocytes to t-butylhydroperoxide. Biochem J 1995; 307(Pt 1):99–106.

44. Chen S, Ge Y, Si J et al. Candesartan suppresses chronic renalinflammation by a novel antioxidant action independent of AT1Rblockade. Kidney Int 2008; 74: 1128–1138.

45. Ohori K, Miura T, Tanno M et al. Ser9 phosphorylation of mitochondrialGSK-3beta is a primary mechanism of cardiomyocyte protection byerythropoietin against oxidant-induced apoptosis. Am J Physiol Heart CircPhysiol 2008; 295: H2079–H2086.



inhibition of glycogen synthase kinase-3β prevents nsaid-induced acute kidney injury

Documents