Nonsteroidal Anti-Inflammatory Drug Flufenamic Acid Is a ... · Nonsteroidal Anti-Inflammatory Drug Flufenamic Acid Is a Potent Activator of AMP-Activated Protein Kinase S Yuan Chi,

Nonsteroidal Anti-Inflammatory Drug Flufenamic Acid Is aPotent Activator of AMP-Activated Protein Kinase□S

Yuan Chi, Kai Li, Qiaojing Yan, Schuichi Koizumi, Liye Shi, Shuhei Takahashi, Ying Zhu,Hiroyuki Matsue, Masayuki Takeda, Masanori Kitamura, and Jian YaoDepartments of Molecular Signaling (Y.C., K.L., Q.Y., S.T., Y.Z., M.K., J.Y.), Pharmacology (S.K.), and Urology (K.L., M.T.),University of Yamanashi, Yamanashi, Japan; First Affiliated Hospital of China Medical University, Shenyang, China (K.L., L.S.);and Department of Dermatology, Chiba University School of Medicine, Chiba, Japan (H.M.)

Received April 14, 2011; accepted July 13, 2011

ABSTRACTFlufenamic acid (FFA) is a nonsteroidal anti-inflammatory drug(NSAID). It has anti-inflammatory and antipyretic properties. Inaddition, it modulates multiple channel activities. The mecha-nisms underlying the pharmacological actions of FFA are pres-ently unclear. Given that AMP-activated protein kinase (AMPK)has both anti-inflammatory and channel-regulating functions,we examined whether FFA induces AMPK activation. 1) Expo-sure of several different types of cells to FFA resulted in anelevation of AMPK� phosphorylation at Thr172. This effect ofFFA was reproduced by functionally and structurally similarmefenamic acid, tolfenamic acid, niflumic acid, and meclofe-namic acid. 2) FFA-induced activation of AMPK was largelyabolished by the treatment of cells with 1,2-bis(2-aminophe-noxy)ethane-N,N,N�,N�-tetraacetic acid tetrakis(acetoxymethylester) (an intracellular Ca2� chelator) or depletion of extracel-lular Ca2�, whereas it was mimicked by stimulation of cells withthe Ca2� ionophore 5-(methylamino)-2-({(2R,3R,6S,8S,9R,11R)-

3,9,11-trimethyl-8-[(1S)-1-methyl-2-oxo-2-(1H-pyrrol-2-yl)ethyl]-1,7-dioxaspiro[5.5]undec-2-yl}methyl)-1,3-benzoxazole-4-car-boxylic acid (A23187) or ionomycin. 3) FFA triggered a rise inintracellular Ca2�, which was abolished by cyclosporine, a blockerof mitochondrial permeability transition pore. Cyclosporine alsoabolished FFA-induced activation of AMPK. 4) Inhibition of Ca2�/calmodulin-dependent kinase kinase � (CaMKK�) with 7-oxo-7H-benzimidazo[2,1-a]benz[de]isoquinoline-3-carboxylic acid acetate(STO-609) or down-regulation of CaMKK� with short interferingRNA largely abrogated FFA-induced activation of AMPK. 5) FFAsignificantly suppressed nuclear factor-�� activity and induciblenitric-oxide synthase expression triggered by interleukin-1� andtumor necrosis factor �. This suppression was also largely abro-gated by STO-609. Taken together, we conclude that FFA inducesAMPK activation through the Ca2�-CaMKK� pathway. Activationof AMPK is a presently unrecognized important mechanism un-derlying the pharmacological effects of FFA.

IntroductionAMP-activated protein kinase (AMPK) is a serine/threo-

nine protein kinase composed of a catalytic � subunit andregulatory � and � subunits. It serves as a sensor of theenergy state of the cells and is a key regulator of metabolichomeostasis (Towler and Hardie, 2007). AMPK is activatedin response to increased AMP/ATP ratio, as well as by phos-phorylation of Thr172 on its �-subunit by upstream kinasesLKB1 (liver kinase B1) and Ca2�/calmodulin-dependent ki-nase kinase (CaMKK) (Shaw et al., 2004; Hawley et al.,2005). Once activated, AMPK switches off ATP-consuming

This work was supported by Grants-in-Aid for Scientific Research from theMinistry of Education, Culture, Sports, Science, and Technology of Japan[Grants 17659255, 20590953 (to J.Y.); B21390324 (to H.M.)]; the TakedaScience Foundation; the Japan-China Medical Association; and a StrategicProject Grant from the University of Yamanashi.

Article, publication date, and citation information can be found athttp://jpet.aspetjournals.org.

doi:10.1124/jpet.111.183020.□S The online version of this article (available at http://jpet.aspetjournals.org)

contains supplemental material.

ABBREVIATIONS: AMPK, AMP-activated protein kinase; p-AMPK, phosphorylated AMPK; AICAR, 5-aminoimidazole-4-carboxamide-1-�-D-ribofuranoside; CaMKK, Ca2�/calmodulin-dependent kinase kinase; FFA, flufenamic acid; LKB1, liver kinase B1; MPTP, mitochondria permeabilitytransition pore; NSAID, nonsteroidal anti-inflammatory drug; TRP, transient receptor potential; IL, interleukin; TNF, tumor necrosis factor; siRNA, shortinterfering RNA; iNOS, inducible nitric-oxide synthase; ACC, acetyl-coA carboxylase; FBS, fetal bovine serum; NF�B, nuclear factor �B; SEAP, secretedalkaline phosphatase; NRK, normal rat kidney; MFA, mefenamic acid; TFA, tolfenamic acid; NFA, niflumic acid; MCFA, meclofenamic acid; PKC, proteinkinase C; ERK, extracellular signal-related kinase; BAPTA-AM, 1,2-bis(2-aminophenoxy)ethane-N,N,N�,N�-tetraacetic acid tetrakis(acetoxymethyl ester);A23187, 5-(methylamino)-2-({(2R,3R,6S,8S,9R,11R)-3,9,11-trimethyl-8-[(1S)-1-methyl-2-oxo-2-(1H-pyrrol-2-yl)ethyl]-1,7-dioxaspiro[5.5]undec-2-yl}methyl)-1,3-benzoxazole-4-carboxylic acid; STO-609, 7-oxo-7H-benzimidazo[2,1-a]benz[de]isoquinoline-3-carboxylic acid acetate; SC-514, 4-amino-[2�,3�-bithiophene]-5-carboxamide.

0022-3565/11/3391-257–266$25.00THE JOURNAL OF PHARMACOLOGY AND EXPERIMENTAL THERAPEUTICS Vol. 339, No. 1Copyright © 2011 by The American Society for Pharmacology and Experimental Therapeutics 183020/3717942JPET 339:257–266, 2011 Printed in U.S.A.

257

http://jpet.aspetjournals.org/content/suppl/2011/07/15/jpet.111.183020.DC1Supplemental material to this article can be found at:

at ASPE

T Journals on A

ugust 30, 2017jpet.aspetjournals.org

Dow

nloaded from

at ASPE

T Journals on A


Dow

nloaded from

at ASPE

T Journals on A


Dow

nloaded from

at ASPE

T Journals on A


Dow

nloaded from

at ASPE

T Journals on A


Dow

nloaded from

at ASPE

T Journals on A


Dow

nloaded from

at ASPE

T Journals on A


Dow

nloaded from

http://jpet.aspetjournals.org/content/suppl/2011/07/15/jpet.111.183020.DC1

http://jpet.aspetjournals.org/







oryl

ated

AM

PK le

vel

400

600

800

*

*

*

* *

Con FFA MFA TFA NFA MCFA

phos

pho

0

200

K le

vel

1000

1200

**

phos

phor

ylat

ed A

MP

K

200

400

600

800

** * *

**

0 2 5 15 30 90 180 360 7200

el

1000

*

phos

phor

ylat

ed A

MPK

leve

200

400

600

800

**

0

Fig. 1. Effects of fenamates on AMPK activation. A, effects ofstructurally similar fenamates on the activation of AMPK.NRK-52E cells were exposed to 50 �M FFA, MFA, TFA, NFA,or MCFA for 5 min. Cellular protein was extracted and sub-jected to Western blot analysis for the phosphorylated levels ofAMPK�, AMPK�, and ACC. The equal loading of protein ineach lane was verified by probing the blots with an anti-�-actin antibody. Results are representative of three separateexperiments. B, densitometric analysis of the phosphorylatedAMPK� shown in A. Results are expressed as relative unit[mean � S.E.; n 3; �, P 0.01 versus untreated control(Con)]. C to F, time-course and concentration-dependent ef-fects of FFA on AMPK. NRK-52E cells were exposed to 50 �MFFA for the indicated time intervals (C and D) or differentconcentrations of FFA for 5 min (E and F). Cellular proteinwas extracted and subjected to Western blot analysis. D andF, densitometric analysis of the phosphorylated AMPK�shown in C and E, respectively (mean � S.E.; n 3; �, P 0.01versus untreated control).

LLC-PK1 Hepa 1c1c-7

HeLa3T3-L1

PC-3LNCaP

Kidney tissueLiver tissue

Fig. 2. Effects of FFA on activation of AMPK in several differ-ent cell lines and tissues. A to F, effects of FFA on activation ofAMPK in several different cell lines. Pig kidney proximal tu-bular epithelial cells (LLC-PK1; A), the mouse hepatoma cellline Hepa 1c1c-7 (B), mouse preadipocyte 3T3-L1 cells (C),human HeLa cells (D), and human prostate cancer epithelialLNCaP (E) and PC-3 cells (F) were exposed to the indicatedconcentrations of FFA for 5 min. The phosphorylation level ofAMPK� at Ser172 was determined by Western blot. �-Actinlevels shown at the bottom of the blots indicate the sameamount of loading of the protein. G and H, effects of FFA onAMPK activation in vivo. Mice were intraperitoneally injectedwith the indicated concentrations of FFA for 30 min. Proteinsfrom liver (G) and kidney (H) tissues were extracted and sub-jected to Western blot analysis for the phosphorylated AMPK�.�-Actin levels shown at the bottom of the blots indicate thesame amount of loading of the protein. Results are represen-tatives of two to three separate experiments.

258 Chi et al.

at ASPE

T Journals on A


Dow

nloaded from


pathways and switches on ATP-producing pathways. In ad-dition to its regulatory functions on cellular metabolic path-ways, AMPK has anti-inflammatory effects (Pilon et al.,2004; Cheng et al., 2007; Jeong et al., 2009; Aoki et al., 2010;Cai et al., 2010; Shin et al., 2010). In addition, it promotesangiogenesis, protects cells from apoptosis (Shaw et al.,2004), and modulates a variety of channel activities (Carat-tino et al., 2005; Mace et al., 2008; Klein et al., 2009; Kong-suphol et al., 2009; Kreneisz et al., 2009).

Flufenamic acid (FFA) is one of the nonsteroidal anti-inflammatory drugs (NSAIDs) used for the alleviation ofinflammation and pain in the clinic (Flower et al., 1972). Inaddition, FFA regulates multiple channel activities. FFA, onthe one hand, inhibits gap junction channels (Harks et al.,2001), Ca2�-activated chloride channels (White and Aylwin,1990), cystic fibrosis transmembrane conductance regulatorchloride channels (McCarty et al., 1993), voltage-gated so-dium channels (Yau et al., 2010), transient receptor potential(TRP) channels (Hill et al., 2004), and nonselective cationchannels (Poronnik et al., 1992). On the other hand, it acti-vates potassium channels (Ottolia and Toro, 1994), TRPC6channels (Foster et al., 2009), and TRPA1 channels (Hu etal., 2010). The channel-regulating property of FFA has beenextensively exploited for both experimental and therapeuticpurposes in a variety of pharmacologic and pathophysiologi-cal models.

At present, little is known about the molecular mecha-nisms underlying the actions of FFA. Several studies haveshown that FFA is able to elevate intracellular Ca2� (Mc-Dougall et al., 1988; Poronnik et al., 1992; Jordani et al.,2000; Gardam et al., 2008; Tu et al., 2009). Furthermore, apossible link between FFA-induced elevation of intracellularCa2� and the alterations of channel activities has been pro-posed (Poronnik et al., 1992; Gardam et al., 2008). However,the downstream molecular events implicated in the actions ofFFA are still poorly understood. Several considerationsprompted us to speculate a possible involvement of AMPK.First, AMPK can be activated by Ca2� through the CaMKK�pathway (Stahmann et al., 2006). Second, similar to FFA,AMPK has both anti-inflammatory and channel-regulatoryactivity (Pilon et al., 2004; Carattino et al., 2005; Cheng etal., 2007; Mace et al., 2008; Jeong et al., 2009; Klein et al.,2009; Kongsuphol et al., 2009; Kreneisz et al., 2009; Aoki etal., 2010; Cai et al., 2010; Shin et al., 2010). For example,both FFA and AMPK have been reported to suppress theinflammatory mediator-induced expression of inducible ni-tric-oxide synthase (iNOS) (Paik et al., 2000; Aoki et al.,2010) and inhibit sodium channel (Carattino et al., 2005;Yau et al., 2010). Third, FFA is reported to inhibit glucoseproduction and promote glucose glycolysis in a model ofisolated perfused liver (Lopez et al., 1998). These meta-bolic changes could also be achieved through the activationof AMPK (Towler and Hardie, 2007). Therefore, the pur-pose of this study was to determine whether FFA couldinduce AMPK activation.

Here, we present evidence showing that FFA potentlyactivates AMPK through the Ca2�-CaMKK� pathway. Ac-tivation of AMPK is a presently unrecognized importantmechanism underlying the pharmacological actions ofFFA.

Materials and MethodsMaterials. IL-1� and TNF� were purchased from R&D Systems

(Minneapolis, MN). 4-Amino-[2�,3�-bithiophene]-5-carboxamide (SC-514), aspirin, and anti-iNOS antibody were obtained from CaymanChemical (Ann Arbor, MI). Antiphospho-AMPK� (Thr172), antiphos-pho-AMPK�1 (Ser108), and antiphospho-acetyl-coA carboxylase(ACC; Ser79) antibodies were obtained from Cell Signaling Technol-ogy (Danvers, MA). Fetal bovine serum (FBS), trypsin/EDTA, anti-biotics, anti-�-actin antibodies, and all other chemicals were pur-chased from Sigma (Tokyo, Japan).

Cells. Normal rat kidney proximal epithelial cells (NRK-52E), theporcine kidney epithelial cell line LLC-PK1, the mouse hepatoma cellline Hepa 1c1c-7, human epithelial carcinoma cell lines (HeLa, PC-3,and LNCaP), and the mouse preadipocyte cell line 3T3-L1 werepurchased from the American Type Culture Collection (Manassas,VA). For maintenance, these cells were cultured in Dulbecco’s mod-ified Eagle’s medium/F-12 containing 5 to 10% FBS. For experi-ments, they were cultured in Dulbecco’s modified Eagle’s medium/F-12 containing 1% FBS.

F380

ratio

2.0

2.5

3.0FFAFFA + Cyclosporin

FFA

Time (s)0 200 400 600 800

F340

/F

1.0

1.5

FFA

ratio

2.5

3.0

3.5- Cyclosporin+ Cyclosporin

FFA + FFA

F340

/F38

0 r

0.0

0.5

1.0

1.5

2.0*

- FFA + FFAFig. 3. Induction of intracellular Ca2� by FFA and its prevention bycyclosporine. NRK-52E cells were exposed to 30 �M FFA in the presenceof absence of 5 �M cyclosporine for the indicated seconds. The results arepresented as dynamic traces of Ca2� over time (A), representing averagelevels of intracellular Ca2� among 15 to 20 cells in a single study, orintracellular Ca2� level at basal and peak values after the addition ofFFA (B). Ca2� concentration is expressed as the ratio of emitted fluores-cence at 340 and 380 nm (F340/F380). �, P 0.01 versus FFA alone(mean � S.E.; n 15–20).

Fenamates Activate AMPK 259

at ASPE

T Journals on A


Dow

nloaded from


Animals. Adult female C57BL/6J mice, weighting 25 to 30 g, werepurchased from Japan SLC, Inc. (Hamamatsu, Japan). Animal ex-perimental plans and procedures were approved by the Animal Ex-perimental Committee of Yamanashi University.

Western Blot Analysis. Total cellular protein was extracted bysuspending the prewashed cells in SDS lysis buffer (62.5 mM Tris-HCl, 2% SDS, 10% glycerol) with freshly added proteinase inhibitorcocktail (Nacalai Tesque, Kyoto, Japan). Lysates were incubated onice for 30 min with intermittent mixing and then centrifuged at12,000 rpm for 10 min at 4°C. Supernatant was recovered andprotein concentration was determined using the Pierce Micro BCA

Protein Assay Kit (Thermo Fisher Scientific, Waltham, MA). West-ern blot was performed using the enhanced chemiluminescence sys-tem (Yao et al., 2005). In brief, extracted cellular proteins wereseparated by 10% or 4 to 20% gradient SDS-polyacrylamide gels andelectrotransferred onto polyvinylidine difluoride membranes. Afterblocking with 3% bovine serum albumin in phosphate-buffered sa-line, the membranes were incubated with antiphospho-AMPK� and-AMPK�, antiphospho-ACC, or anti-iNOS antibody. After washing,the membranes were probed with horseradish peroxidase-conjugatedanti-rabbit IgG, and the bands were visualized using the enhancedchemiluminescence system (GE Healthcare, Chalfont St. Giles,

β-ac�n

p-AMPKα

MPK

leve

lsm

ulat

ed)

80

100

120

Phos

phor

ylat

ed A

(% o

f FFA

-sti m

0

20

40

60

*

A MPK

leve

lsm

ulat

ed)

80

100

120

Phos

phor

ylat

ed A

(% o

f FFA

-sti

0

20

40

60

*A

Fig. 4. Modulation of intracellular Ca2� on FFA-induced activation of AMPK. A, induction of AMPK� phosphorylation by Ca2� inophores A23187 andionomycin. NRK-52E cells were exposed to 5 �M A23187 and 10 �M ionomycin for 5 min. B to D, FFA-induced activation of AMPK in the absence ofextracellular Ca2� or the presence of the Ca2� chelator BAPTA-AM. NRK-52E cells were either cultured in Ca2�-free medium (B and C) or normalCa2� medium with 100 �M BAPTA-AM (D) for 1 h before exposure to FFA for an additional 5 min. C, densitometric analysis of the blot shown in B.Data are expressed as the percentage of FFA-stimulated levels of p-AMPK (mean � S.E.; n 3). �, P 0.05 versus FFA alone. E and F, effects of theinhibition of mitochondria permeability transition pore with cyclosporine on FFA-induced activation of AMPK. E, NRK-52E cells were pretreated with5 �M cyclosporine for 30 min, and then exposed to 50 �M FFA for an additional 5 min. Cellular proteins were extracted and subjected to Westernanalysis for the phosphorylated AMPK�. �-Actin shown at the bottom of the blot indicates the same amount of loading of protein. F, densitometricanalysis of the results of E. Data are expressed as the percentage of FFA-stimulated levels of p-AMPK (mean � S.E.; n 3). �, P 0.05 versus FFAalone.

260 Chi et al.

at ASPE

T Journals on A


Dow

nloaded from


Buckinghamshire, UK). The chemiluminescent signal was capturedwith a Fujifilm luminescent image LAS-1000 analyzer (Fujifilm,Tokyo, Japan) and quantified with densitometric software. To con-firm equal loading of proteins, the membranes were probed for �-ac-tin protein.

Measurement of Ca2�. Cultured NRK-52E cells were loadedwith fura-2 by incubation with 5 �M fura-2 acetoxymethyl ester inHanks’ balanced salt solution containing 2.0 mM CaCl2 and 1 mMMgCl2 at room temperature. Ca2� was determined by the ratiomethod as reported previously (Yao et al., 2003).

Transient Transfection of Cells with siRNA. NRK-52E cellswere transiently transfected with siRNA specifically targetingCaMKK� or a negative control siRNA (AllStars Negative ControlsiRNA; QIAGEN, Tokyo, Japan) at a final concentration of 20 nMusing Hyperfect transfection reagent for 48 h. After that, cells wereeither left untreated or exposed to 50 �M FFA for 5 min. Cellularproteins were extracted and analyzed for phosphorylated AMPK�.

Establishment of Stable Transfectant. NRK/NF�B-SEAP re-porter cells were established by stably transfection of NRK-52E cellswith pNF�B-SEAP (BD Biosciences, San Jose, CA) as describedpreviously (Yao et al., 2005; Hayakawa et al., 2006a). pNF�B-SEAPencodes SEAP under the control of NF��.

SEAP Assay. Activity of SEAP in culture media was evaluated bya chemiluminescent method using the Great EscAPe SEAP detectionkit (BD Biosciences) as described previously (Yao et al., 2005; Hay-akawa et al., 2006a). In brief, 5 �l of culture media were mixed with

15 �l of 1� dilution buffer and incubated at 65°C for 30 min. Afterthe incubation, the samples were mixed with 20 �l of assay buffercontaining L-homoarginine, left at room temperature for 5 min,and added with 20 �l of chemiluminescent enhancer containing1.25 mM CSPD chemiluminescent substrate [disodium 3-(4-methox-yspiro{1,2-dioxetane-3,2�-(5�-chloro)tricyclo[3.3.1.13,7]decan}-4-yl) phenylphosphate]. After incubation in the dark for 30 min, the samples weresubjected to assays using a luminometer (Gene Light 55; Microtech Nition,Chiba, Japan). All assays were performed in quadruplicate.

Measurement of Nitrite Levels. NO production was assayed bydetecting nitrite accumulation in the culture medium using Griessreagent (Green et al., 1982). In brief, 100 �l of a solution containing1% sulfanilamide and 0.1% naphthylethylenediamine in 2 M HClwas added to 100 �l of conditioned medium. Samples were incubatedat room temperature for 10 min, and then the absorbance wasmeasured with a microtiter plate reader at 550 nm. Nitrite levelswere expressed in picomoles of NO2 per microgram of total cellularprotein.

Statistical Analysis. Values are expressed as mean � S.E.Comparison of two populations was made by Student’s t test. Formultiple comparisons with a single control, one-way analysis ofvariance followed by Dunnett’s test was used. Both analyses werecarried out using SigmaStat statistical software (Systat Software,Inc., San Jose, CA). P 0.05 was considered to be a statisticallysignificant difference.

120

Phos

phor

ylat

ed A

MPK

leve

ls (%

of F

FA-s

timul

ated

)

20

40

60

80

100

*

0

Fig. 5. Involvement of CaMKK� in FFA-induced activation of AMPK. A and B, abrogation of FFA-induced activation of AMPK� by kinase inhibitors.NRK-52E cells were pretreated with 50 nM calphostin (A) or 5 �M STO-609 (B) for 15 min, and then exposed to 50 �M FFA for an additional 5 min.C, densitometric analysis of blot shown in B. Data are expressed as the percentage of FFA-stimulated levels of p-AMPK (mean � S.E.; n 3). �, P 0.05 versus FFA alone. D, inhibition of FFA-induced activation of AMPK� by specific siRNA against CaMKK�. NRK-52E cells were transfected withCaMKK� siRNA or control siRNA for 48 h. After that, cells were exposed to 50 �M FFA for 5 min. Cellular proteins were extracted and subjected toWestern analysis for phosphorylated AMPK� and CaMKK�. Equal loading of protein per lane was verified by probing the blot with an anti-�-actinantibody. Results are representatives of two to three separate experiments.


at ASPE

T Journals on A


Dow

nloaded from


mul

ated

)

100

120

140

iNO

S le

vels

(% o

f cyt

okin

es-s

ti m

0

20

40

60

80

* *

evel

ss-

stim

ulat

ed)

80

100

120

**

NS

*

iNO

S le

(% o

f cyt

okin

e s

0

20

40

60

cellu

lar p

rote

in

300

400

500

- Cytokines+ Cytokines

* *NS

Control FFA FFA + STO

pM N

O2/

mg

c

0

100

200

- Cytokines+ Cytokines

activ

ity (R

LU x

100

0)

20

30

40

50

+ Cytokines

*

*

*

Control FFA FFA + STO

SE

AP

0

10

Fig. 6. Suppressive effects of FFA on cytokine-induced iNOS expression and NF�B activation. A, suppression of cytokine-induced iNOS expression byFFA and AICAR. NRK-52E cells were stimulated with 2 ng/ml IL-1� and 20 ng/ml TNF� in the presence or absence of 50 �M FFA or 500 �M AICARfor 24 h. Cellular proteins were extracted and subjected to Western analysis for iNOS. Equal loading of protein per lane was verified by probing theblot with an anti-�-actin antibody. B, densitometric analysis of iNOS expression shown in A. Results are expressed as the percentage of thecytokine-stimulated level of iNOS (mean � S.E.; n 3). �, P 0.05 versus the cytokine-stimulated cells. C, attenuation of FFA-induced suppression

262 Chi et al.

at ASPE

T Journals on A


Dow

nloaded from


ResultsFFA Induces AMPK Activation. FFA is one of the N-

ary-anthranilic acid derivatives, belonging to the fenamategroup of NSAIDs. Other members of the fenamate group ofNASIDs are mefenamic acid (MFA), tolfenamic acid (TFA),niflumic acid (NFA), and meclofenamic acid (MCFA). All ofthese chemicals have similar structures and functions(Winder et al., 1963; Poronnik et al., 1992). To test whetherfenamates activate AMPK, we examined the influence ofthese chemicals on phosphorylation levels of AMPK�at Thr172 in NRK-52E cells. Previous studies had estab-lished that phosphorylation of this site correlates withAMPK activity (Towler and Hardie, 2007). As shown in Fig.1,A and B, incubation of NRK cells with fenamates resultedin increased levels of phosphorylated AMPK� and AMPK�,which was associated with paralleled elevation of phosphor-ylated ACC, one of the AMPK substrates (Towler and Hardie,2007). Densitometric analysis of the blots in Fig. 1A revealedthat all the chemicals significantly activated AMPK. Amongthem, the effects of FFA and TFA were more pronounced(Fig. 1B). Considering the widespread use of FFA in a varietyof in vivo and in vitro experimental systems, we chose FFAfor further analysis.

Time-course analysis of the effect using FFA revealed thatthe activation of AMPK was rapid, which was detectable asearly as 2 min after FFA addition, peaked at approximately15 min, and was retained at a relatively high level for at least12 h (Fig. 1, C and D). Concentration-effect analysis revealedthat the activation of AMPK was concentration-dependent.The clear activation could be observed at the concentrationsof FFA as low as 10 �M (Fig. 1, E and F). These results thusindicate that N-ary-anthranilic acid derivative is a novelclass of AMPK activator.

Activation of AMPK by FFA Is Not Cell Type- andSpecies-Specific. To determine whether the effect of FFA iscell type- and species-specific, we evaluated AMPK activationin several different types of cells. As shown in Fig. 2, FFAcaused concentration-dependent activation of AMPK in pigkidney proximal tubular epithelial cells (LLC-PK1; Fig. 2A),the mouse hepatoma cell line Hepa 1c1c-7 (Fig. 2B), themouse preadipocyte cell line 3T3-L1 (Fig. 2C), human HeLacells (Fig. 2D), and human prostate cancer epithelial cells(LNCaP and PC-3; Fig. 2, E and F, respectively).

FFA also induced AMPK activation in vivo. Intraperitonealinjection of FFA into mice caused an elevation of AMPKphosphorylation in both liver (Fig. 2G) and kidney (Fig. 2H)tissues. These results indicate that the effect of FFA is notcell type- and species-specific and can be detected both invitro and in vivo.

FFA-Induced Activation of AMPK Depends on Intra-cellular Ca2�. Several studies have demonstrated that FFAis able to elevate intracellular Ca2� through the induction of

Ca2� releases from mitochondria (McDougall et al., 1988;Poronnik et al., 1992; Jordani et al., 2000; Gardam et al.,2008; Tu et al., 2009). We, therefore, evaluated the role ofCa2� in FFA-induced activation of AMPK. First, we con-firmed the calcium-elevating effect of FFA in NRK-52Ecells. Consistent with previous reports (McDougall et al.,1988; Poronnik et al., 1992; Jordani et al., 2000; Gardam etal., 2008; Tu et al., 2009), FFA elevated intracellular Ca2�

in NRK-52E cells (Fig. 3). This effect was completelyblocked by cyclosporine, an inhibitor of the mitochondriapermeability transition pore (MPTP) (Broekemeier andPfeiffer, 1995).

We then examined the role of the elevated intracellular Ca2� inAMPK activation. As shown in Fig. 4A, induction of intracellularCa2� with the Ca2� inophores 5-(methylamino)-2-({(2R,3R,6S,8S,9R,11R)-3,9,11-trimethyl-8-[(1S)-1-methyl-2-oxo-2-(1H-pyrrol-2-yl)ethyl]-1,7-dioxaspiro[5.5]undec-2-yl}methyl)-1,3-benzoxazole-4-carboxylic acid (A23187) and ionomycin increasedphosphorylation levels of AMPK�. In contrast, inhibition of intra-cellular Ca2� by culture of cells in a calcium-free medium oraddition of the Ca2� chelator BAPTA-AM largely abrogatedAMPK activation (Fig. 4, B-D). Consistent with the causative roleof MPTP opening in FFA-induced elevation in intracellular Ca2�,cyclosporine also significantly blocked AMPK activation (Fig. 4, Eand F). Thus the elevated Ca2� is required for FFA-induced acti-vation of AMPK.

CaMKK� Underlies FFA-Induced Activation of AMPK.Increased intracellular Ca2� activates various kinases, includ-ing a well documented AMPK, CaMKK� (Hawley et al., 2005).To assess the role of kinases, especially CaMKK�, we examinedthe influence of the PKC inhibitor calphostin and a specificCaMKK� inhibitor 7-oxo-7H-benzimidazo[2,1-a]benz[de]iso-quinoline-3-carboxylic acid acetate (STO-609) on the activationof AMPK. As shown in Fig. 5, A to C, both agents effectivelysuppressed AMPK phosphorylation. Furthermore, down-regu-lation of CaMKK� with the specific siRNA also abolished theeffect of FFA (Fig. 5D).

AMPK Contributes to FFA-Induced Suppression ofNF�B Activity and iNOS Expression. In addition to itscrucial role in the control of metabolic processes, AMPKsuppresses inflammatory responses (Cheng et al., 2007;Peairs et al., 2009). Therefore, we tested the possible impli-cation of AMPK on the anti-inflammatory effect of FFA. Forthis purpose, we examined proinflammatory cytokines IL-1�-and TNF�-induced expression of iNOS. As shown in Fig. 6, Aand B, FFA markedly inhibited the cytokine-induced expres-sion of iNOS in NRK-52E cells. This effect was similarlyproduced by a well known AMPK activator, 5-aminoimida-zole-4-carboxamide-1-�-D-ribofuranoside (AICAR), suggest-ing a possible involvement of AMPK. Because activation ofAMPK by FFA in NRK-52E cells was mediated by CaMKK�,we, therefore, determined the role of AMPK by inhibition of

of iNOS by inhibition of the CaMKK�. NRK-52E cells were pretreated with 5 �M STO-609 for 15 min and then exposed to 2 ng/ml IL-1� plus 20 ng/mlTNF� in the presence of absence of FFA for 24 h. D, densitometric analysis of iNOS expression shown in C. Results are expressed as the percentageof the cytokine-stimulated level of iNOS (mean � S.E.; n 3). �, P 0.05. NS, not significantly different (P 0.05). E, abrogation of FFA-inducedsuppression of NO formation by inhibition of the AMPK CaMKK�. NRK-52E cells were treated as in D. The conditioned media were harvested formeasurement of nitrite levels. Data are expressed as mean � S.E. (n 4). �, P 0.05. NS, not significantly different (P 0.05). F, role of NF�B incytokine-induced iNOS expression. NRK cells were treated with the NF�B inhibitor SC-514 (100 �M) for 30 min and then exposed to 2 ng/ml IL-1�and 20 ng/ml TNF� for 24 h. Cellular proteins were extracted and subjected to Western blot analysis using an anti-iNOS antibody. Expression of�-actin was used as loading control. G, abrogation of FFA-induced suppression of NF�B activity by inhibition of the CaMKK�. NRK/NF�B-SEAPreporter cells were exposed to 2 ng/ml IL-1� and 20 ng/ml TNF� in the presence or absence of 50 �M FFA and/or 5 �M STO-609 for 24 h. Theconditioned media were harvested for measurement of SEAP activity. Data are expressed as mean � S.E. (n 4). �, P 0.05.


at ASPE

T Journals on A


Dow

nloaded from


CaMKK�. As shown in Fig. 6, C to E, the CaMKK� inhibitorSTO-609 significantly abrogated the suppressive effect ofFFA on iNOS expression and NO formation.

One recent study indicated that AMPK suppresses iNOSexpression through the inhibition of NF�B (Aoki et al.,2010). We, therefore, examined the possible effect of FFAon NF�B. First, we confirmed that the cytokine-inducedexpression of iNOS was controlled by NF�B. Inhibition ofNF�B with SC-514 completely abrogated cytokine-elicitediNOS expression (Fig. 6F). To determine the influence ofFFA on NF�B, we transfected NRK-52E cells with pNF�B-SEAP and monitored SEAP activity in the conditionedmedia. As shown in Fig. 6G, FFA significantly inhibitedcytokine-induced NF�B activation, which was also signif-icantly blocked by STO-609. These observations indicatethat FFA suppresses iNOS expression through CaMKK�-dependent inhibition of NF�B.

DiscussionIn this study, we provide the first evidence showing that

FFA and other members of the fenamate class of NSAIDs arepotent activators of AMPK. The mechanisms involved areschematically illustrated in Fig. 7. Given that FFA has beenwidely used as an anti-inflammatory and channel-regulatingagent in clinical and basic research, our findings may havesignificant implications for understanding the pharmacolog-ical actions of FFA.

FFA induced AMPK activation through the Ca2�-CaMKK�pathway. This is shown by the fact that activation of AMPK byFFA was Ca2�-dependent and abolished by inhibition or down-regulation of CaMKK�. Apart from CaMKK�, AMPK is alsophosphorylated by LKB1 (Shaw et al., 2004). However, it is lesslikely that LKB1 played a major role in this study, because FFAsimilarly triggered AMPK activation in LKB1-deficient Helacells (Fig. 2D) (Shaw et al., 2004).

CaMKK is a Ca2�-dependent kinase. It is activated subse-quent to the elevation of intracellular Ca2� (Hawley et al.,2005; Stahmann et al., 2006). Indeed, FFA elicited a rise inintracellular Ca2� in NRK-52E cells. Consistent with previ-ous reports (McDougall et al., 1988; Poronnik et al., 1992;Jordani et al., 2000; Tu et al., 2009), this action of FFA wascaused by the opening of MPTP. Blockade of MPTP withcyclosporine attenuated the increase of Ca2� and abolishedthe activation of AMPK.

The mechanisms by which FFA alters mitochondria Ca2�

fluxes are still unclear and needed to be clarified in thefuture. FFA has also been reported to uncouple mitochondria(McDougall et al., 1988; Poronnik et al., 1992; Jordani et al.,2000; Gardam et al., 2008; Tu et al., 2009). Treatment of cellswith FFA caused inhibition in ATP production (Lopez et al.,1998; Tu et al., 2009). Given that the increased AMP/ATPratio can allosterically activate AMPK (Towler and Hardie,2007), an involvement of this mechanism in this study is alsolikely. It is noteworthy that the loss of mitochondria Ca2�

has been recognized to be causative of the reduced biogenesisin mitochondria (Cardenas et al., 2010). Therefore, it can besaid that FFA-induced activation of AMPK resides in itsability to alter Ca2� concentrations inside and outside ofmitochondria.

One previous report demonstrated that PKC mediated theischemic precondition-induced activation of AMPK (Nishinoet al., 2004). It is noteworthy that activation of AMPK byFFA was also abolished by calphostin C, a widely used in-hibitor of protein kinase C. In this context, a possible involve-ment of PKC in this study cannot be excluded. However,calphostin C has also been described to be able to blockL-type Ca2� channels (Nishino et al., 2004). It is unclearwhether the effect of calphostin C was caused by its inhibi-tion on PKC or intracellular Ca2�. More detailed analysis onthis aspect may be needed in the future.

It is noteworthy that in this study we observed that thedifferent types of cells varied in their response to FFA-in-duced activation of AMPK. For example, FFA induced adetectable activation of AMPK in NRK-52E cells at a concen-tration as low as 5 to 10 �M. In contrast, the similar extentof activation in LLC-PK1 cells required more than 50 �MFFA. The reasons for the discrepancy are unclear. It could berelated to the different capacity of the cells in uptaking,processing, and metabolizing FFA. It could also be caused bythe difference in the abundance of CaMKK� among the dif-ferent types of cells. It should be mentioned that FFA at theconcentrations used in this study did not exhibit significantcytotoxic effect, as revealed by lactate dehydrogenase releas-ing and 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazoliumassay (data not shown).

FFA has multifaceted functions. It is conceivable that theeffect of FFA on AMPK activation could be a secondary eventresulting from its effects on other target molecules. However,

Fig. 7. Schematic diagram illustrating potential mecha-nisms involved in FFA-induced activation of AMPK andthe suppression of iNOS. FFA elevates intracellular Ca2�

through the induction of the release of Ca2� from mitochon-dria, causing the activation of the CaMKK�-AMPK path-way. The activated AMPK contributes to the inhibition ofNF�B and NF�B-regulated iNOS expression.

264 Chi et al.

at ASPE

T Journals on A


Dow

nloaded from


by using structurally different cyclooxygenase-2 inhibitorsand various channel blockers, we excluded this possibility(Supplemental Fig. 1). On the contrary, we believe that theactivation of AMPK might be behind the regulatory effects ofFFA on these molecules. In support of this notion, implica-tion of AMPK in the inhibition of cyclooxygenase-2 expres-sion has been reported (Lee et al., 2009). In addition, thechannels regulated by FFA are, in fact, modulated by AMPKin a similar way. For example, both FFA and AMPK are ableto inhibit cystic fibrosis transmembrane conductance regula-tor and sodium channels (McCarty et al., 1993; Carattino etal., 2005; Kongsuphol et al., 2009; Yau et al., 2010). More-over, the effective concentrations of FFA used in those stud-ies (1–300 �M) are consistent with the concentrations re-quired for AMPK activation shown in this study.

Activation of AMPK also underlies the anti-inflammatoryactions of FFA (Pilon et al., 2004; Cheng et al., 2007; Jeonget al., 2009; Aoki et al., 2010; Cai et al., 2010; Shin et al.,2010). A series of studies demonstrated that AMPK mediatesthe anti-inflammatory effects of a variety of agents, includingnicotine, berberine, cilostazol, and adiponectin (Pilon et al.,2004; Cheng et al., 2007; Jeong et al., 2009; Aoki et al., 2010;Cai et al., 2010). Consistent with these findings, we alsofound that the suppressive effects of FFA on IL-1�/TNF�-induced NF�B activation and iNOS expression critically de-pended on the CaMKK�-AMPK pathway. In addition toiNOS, our preliminary result demonstrated that the cyto-kine-elicited expression of MCP-1, another NF�B-regulatedgene product (Hayakawa et al., 2006b), was suppressed byFFA (Supplemental Fig. 2). Thus the well documented anti-inflammatory effect of FFA could be attributable to AMPK.

FFA has also been documented to be able to inhibit vascu-lar smooth muscle cell proliferation and suppress p44/42-mitogen-activated protein kinase [also known as extracellu-lar signal-related kinase (ERK)] expression (Schober et al.,2002). It is noteworthy that an antagonistic relationshipbetween AMPK and ERK in the regulation of cell growth andother cell behaviors has been well described previously(Hwang et al., 2006; Du et al., 2008). It is likely that AMPKalso contributes to the growth inhibitory action of FFAthrough suppression of ERK activation.

In addition to unraveling an important molecular mecha-nism mediating the pharmacological actions of FFA, our find-ings suggest that the fenamate group of NSAIDs may be usedfor the treatment of metabolic disorders. As a key regulator ofcell metabolism, activation of AMPK underlies the therapeu-tic benefits of some important antidiabetic drugs such asmetformin and thiazolidinediones (Towler and Hardie, 2007).In comparison with the well known AMPK activator AICAR,FFA induced an even rapid and potent activation of AMPK(Supplemental Fig. 3). In addition, cells that are criticallyinvolved in lipid and glucose metabolism, such as hepato-cytes and preadepocytes, displayed a similar response toFFA. Moreover, FFA also induced AMPK activation in vivo.As a commonly used anti-inflammatory NSAID, FFA mightbe a promising therapeutic option for treating metabolic dis-eases, especially for those associated with inflammatorylesions.

In summary, our study indicates that FFA is a potentactivator of AMPK. Activation of AMPK could be an impor-tant mechanism by which FFA exerts its anti-inflammatoryand channel-regulating actions. As a widely used NSAID,

FFA may be exploited for the therapeutic intervention ofmetabolic disorders.

Authorship Contributions

Participated in research design: Chi, Takeda, Kitamura, and Yao.Conducted experiments: Chi, Li, Yan, Koizumi, Shi, Takahashi,

and Zhu.Contributed new reagents or analytic tools: Koizumi, Matsue, and

Takeda.Performed data analysis: Chi, Zhu, and Yao.Wrote or contributed to the writing of the manuscript: Chi and Yao.

ReferencesAoki C, Hattori Y, Tomizawa A, Jojima T, and Kasai K (2010) Anti-inflammatory role

of cilostazol in vascular smooth muscle cells in vitro and in vivo. J AtherosclerThromb 17:503–509.

Broekemeier KM and Pfeiffer DR (1995) Inhibition of the mitochondrial permeabilitytransition by cyclosporin A during long time frame experiments: relationshipbetween pore opening and the activity of mitochondrial phospholipases. Biochem-istry 34:16440–16449.

Cai XJ, Chen L, Li L, Feng M, Li X, Zhang K, Rong YY, Hu XB, Zhang MX, ZhangY, et al. (2010) Adiponectin inhibits lipopolysaccharide-induced adventitial fibro-blast migration and transition to myofibroblasts via AdipoR1-AMPK-iNOS path-way. Mol Endocrinol 24:218–228.

Carattino MD, Edinger RS, Grieser HJ, Wise R, Neumann D, Schlattner U, JohnsonJP, Kleyman TR, and Hallows KR (2005) Epithelial sodium channel inhibition byAMP-activated protein kinase in oocytes and polarized renal epithelial cells. J BiolChem 280:17608–17616.

Cardenas C, Miller RA, Smith I, Bui T, Molgo J, Muller M, Vais H, Cheung KH, YangJ, Parker I, et al. (2010) Essential regulation of cell bioenergetics by constitutiveInsP3 receptor Ca2� transfer to mitochondria. Cell 142:270–283.

Cheng PY, Lee YM, Law KK, Lin CW, and Yen MH (2007) The involvement ofAMP-activated protein kinases in the anti-inflammatory effect of nicotine in vivoand in vitro. Biochem Pharmacol 74:1758–1765.

Du J, Guan T, Zhang H, Xia Y, Liu F, and Zhang Y (2008) Inhibitory crosstalkbetween ERK and AMPK in the growth and proliferation of cardiac fibroblasts.Biochem Biophys Res Commun 368:402–407.

Flower R, Gryglewski R, Herbaczynska-Cedro K, and Vane JR (1972) Effects ofanti-inflammatory drugs on prostaglandin biosynthesis. Nat New Biol 238:104–106.

Foster RR, Zadeh MA, Welsh GI, Satchell SC, Ye Y, Mathieson PW, Bates DO, andSaleem MA (2009) Flufenamic acid is a tool for investigating TRPC6-mediatedcalcium signalling in human conditionally immortalised podocytes and HEK293cells. Cell Calcium 45:384–390.

Gardam KE, Geiger JE, Hickey CM, Hung AY, and Magoski NS (2008) Flufenamicacid affects multiple currents and causes intracellular Ca2� release in Aplysia bagcell neurons. J Neurophysiol 100:38–49.

Green LC, Wagner DA, Glogowski J, Skipper PL, Wishnok JS, and Tannenbaum SR(1982) Analysis of nitrate, nitrite, and [15N]nitrate in biological fluids. AnalBiochem 126:131–138.

Harks EG, de Roos AD, Peters PH, de Haan LH, Brouwer A, Ypey DL, van Zoelen EJ,and Theuvenet AP (2001) Fenamates: a novel class of reversible gap junctionblockers. J Pharmacol Exp Ther 298:1033–1041.

Hawley SA, Pan DA, Mustard KJ, Ross L, Bain J, Edelman AM, Frenguelli BG, andHardie DG (2005) Calmodulin-dependent protein kinase kinase-� is an alternativeupstream kinase for AMP-activated protein kinase. Cell Metab 2:9–19.

Hayakawa K, Meng Y, Hiramatsu N, Kasai A, Yamauchi K, Yao J, and Kitamura M(2006a) Priming of glomerular mesangial cells by activated macrophages causesblunted responses to proinflammatory stimuli. J Immunol 176:2529–2537.

Hayakawa K, Meng Y, Hiramatsu N, Kasai A, Yao J, and Kitamura M (2006b)Spontaneous activation of the NF-�B signaling pathway in isolated normal glom-eruli. Am J Physiol Renal Physiol 291:F1169–F1176.

Hill K, Benham CD, McNulty S, and Randall AD (2004) Flufenamic acid is apH-dependent antagonist of TRPM2 channels. Neuropharmacology 47:450–460.

Hu H, Tian J, Zhu Y, Wang C, Xiao R, Herz JM, Wood JD, and Zhu MX (2010)Activation of TRPA1 channels by fenamate nonsteroidal anti-inflammatory drugs.Pflugers Arch 459:579–592.

Hwang JT, Kim YM, Surh YJ, Baik HW, Lee SK, Ha J, and Park OJ (2006) Seleniumregulates cyclooxygenase-2 and extracellular signal-regulated kinase signalingpathways by activating AMP-activated protein kinase in colon cancer cells. CancerRes 66:10057–10063.

Jeong HW, Hsu KC, Lee JW, Ham M, Huh JY, Shin HJ, Kim WS, and Kim JB (2009)Berberine suppresses proinflammatory responses through AMPK activation inmacrophages. Am J Physiol Endocrinol Metab 296:E955–E964.

Jordani MC, Santos AC, Prado IM, Uyemura SA, and Curti C (2000) Flufenamic acidas an inducer of mitochondrial permeability transition. Mol Cell Biochem 210:153–158.

Klein H, Garneau L, Trinh NT, Prive A, Dionne F, Goupil E, Thuringer D, Parent L,Brochiero E, and Sauve R (2009) Inhibition of the KCa3.1 channels by AMP-activated protein kinase in human airway epithelial cells. Am J Physiol CellPhysiol 296:C285–C295.

Kongsuphol P, Cassidy D, Hieke B, Treharne KJ, Schreiber R, Mehta A, andKunzelmann K (2009) Mechanistic insight into control of CFTR by AMPK. J BiolChem 284:5645–5653.

Kreneisz O, Benoit JP, Bayliss DA, and Mulkey DK (2009) AMP-activated proteinkinase inhibits TREK channels. J Physiol 587:5819–5830.


at ASPE

T Journals on A


Dow

nloaded from


Lee YK, Park SY, Kim YM, and Park OJ (2009) Regulatory effect of the AMPK-COX-2 signaling pathway in curcumin-induced apoptosis in HT-29 colon cancercells. Ann NY Acad Sci 1171:489–494.

Lopez CH, Bracht A, Yamamoto NS, and dos Santos MD (1998) Metabolic effects anddistribution space of flufenamic acid in the isolated perfused rat liver. Chem BiolInteract 116:105–122.

Mace OJ, Woollhead AM, and Baines DL (2008) AICAR activates AMPK and altersPIP2 association with the epithelial sodium channel ENaC to inhibit Na� trans-port in H441 lung epithelial cells. J Physiol 586:4541–4557.

McCarty NA, McDonough S, Cohen BN, Riordan JR, Davidson N, and Lester HA(1993) Voltage-dependent block of the cystic fibrosis transmembrane conductanceregulator Cl� channel by two closely related arylaminobenzoates. J Gen Physiol102:1–23.

McDougall P, Markham A, Cameron I, and Sweetman AJ (1988) Action of thenonsteroidal anti-inflammatory agent, flufenamic acid, on calcium movements inisolated mitochondria. Biochem Pharmacol 37:1327–1330.

Nishino Y, Miura T, Miki T, Sakamoto J, Nakamura Y, Ikeda Y, Kobayashi H, andShimamoto K (2004) Ischemic preconditioning activates AMPK in a PKC-dependent manner and induces GLUT4 up-regulation in the late phase of cardio-protection. Cardiovasc Res 61:610–619.

Ottolia M and Toro L (1994) Potentiation of large conductance KCa channels byniflumic, flufenamic, and mefenamic acids. Biophys J 67:2272–2279.

Paik JH, Ju JH, Lee JY, Boudreau MD, and Hwang DH (2000) Two opposing effectsof non-steroidal anti-inflammatory drugs on the expression of the inducible cyclo-oxygenase. Mediation through different signaling pathways. J Biol Chem 275:28173–28179.

Peairs A, Radjavi A, Davis S, Li L, Ahmed A, Giri S, and Reilly CM (2009) Activationof AMPK inhibits inflammation in MRL/lpr mouse mesangial cells. Clin ExpImmunol 156:542–551.

Pilon G, Dallaire P, and Marette A (2004) Inhibition of inducible nitric-oxide syn-thase by activators of AMP-activated protein kinase: a new mechanism of action ofinsulin-sensitizing drugs. J Biol Chem 279:20767–20774.

Poronnik P, Ward MC, and Cook DI (1992) Intracellular Ca2� release by flufenamicacid and other blockers of the non-selective cation channel. FEBS Lett 296:245–248.

Schober W, Wiskirchen J, Kehlbach R, Gebert R, Rodegerdts E, Betsch A, Johst U,

Claussen CD, and Duda SH (2002) Flufenamic acid: growth modulating effects onhuman aortic smooth muscle cells in vitro. J Vasc Interv Radiol 13:89–96.

Shaw RJ, Kosmatka M, Bardeesy N, Hurley RL, Witters LA, DePinho RA, andCantley LC (2004) The tumor suppressor LKB1 kinase directly activates AMP-activated kinase and regulates apoptosis in response to energy stress. Proc NatlAcad Sci USA 101:3329–3335.

Shin MJ, Lee YP, Kim DW, An JJ, Jang SH, Cho SM, Sheen SH, Lee HR, Kweon HY,Kang SW, et al. (2010) Transduced PEP-1-AMPK inhibits the LPS-induced ex-pression of COX-2 and iNOS in Raw264.7 cells. BMB Rep 43:40–45.

Stahmann N, Woods A, Carling D, and Heller R (2006) Thrombin activates AMP-activated protein kinase in endothelial cells via a pathway involving Ca2�/calmodulin-dependent protein kinase kinase �. Mol Cell Biol 26:5933–5945.

Towler MC and Hardie DG (2007) AMP-activated protein kinase in metabolic controland insulin signaling. Circ Res 100:328–341.

Tu P, Brandolin G, and Bouron A (2009) The anti-inflammatory agent flufenamicacid depresses store-operated channels by altering mitochondrial calcium homeo-stasis. Neuropharmacology 56:1010–1016.

White MM and Aylwin M (1990) Niflumic and flufenamic acids are potent reversibleblockers of Ca2�-activated Cl� channels in Xenopus oocytes. Mol Pharmacol37:720–724.

Winder CV, Wax J, Serrano B, Jones EM, and McPhee ML (1963) Anti-inflammatoryand antipyretic properties of N-(�,�,�-trifluoro-m-tolyl) anthranilic acid (CI-440;flufenamic acid). Arthritis Rheum 6:36–47.

Yao J, Hiramatsu N, Zhu Y, Morioka T, Takeda M, Oite T, and Kitamura M (2005)Nitric oxide-mediated regulation of connexin43 expression and gap junctionalintercellular communication in mesangial cells. J Am Soc Nephrol 16:58–67.

Yao J, Suwa M, Li B, Kawamura K, Morioka T, and Oite T (2003) ATP-dependentmechanism for coordination of intercellular Ca2� signaling and renin secretion inrat juxtaglomerular cells. Circ Res 93:338–345.

Yau HJ, Baranauskas G, and Martina M (2010) Flufenamic acid decreases neuronalexcitability through modulation of voltage-gated sodium channel gating. J Physiol588:3869–3882.

Address correspondence to: Dr. Jian Yao, Department of Molecular Signaling,Interdisciplinary Graduate School of Medicine and Engineering, Universityof Yamanashi, Chuo, Yamanashi 409-3898, Japan. E-mail: [email protected]

266 Chi et al.

at ASPE

T Journals on A


Dow

nloaded from


Nonsteroidal Anti-inflammatory Drug Flufenamic Acid Is a Potent Activator of AMPK

Yuan Chi, Kai Li, Qiaojing Yan, Schuichi Koizumi, Liye Shi, Shuhei Takahashi, Ying Zhu, Hi ki M M ki T k d M i Ki d Ji Y

Supplemental Fig. 1

Hiroyuki Matsue, Masayuki Takeda, Masanori Kitamura and Jian Yao

The Journal of Pharmacology and Experimental Therapeutics

A

p-AMPKα p-AMPKα

B

β ti

pp g

β-actin β-actin

C

β ti

p-AMPKα

β-actin

Supplemental Figure 1. Effects of COX-2 inhibitors and channel blockers on activationof AMPK NRK 52E cells were exposed to COX 2 inhibitors FFA (50 M) indomethacinof AMPK. NRK-52E cells were exposed to COX-2 inhibitors FFA (50 M), indomethacin(100 M), aspirin (4 mM), and ibuprofen (100 M; A), or gap junction inhibitor lindane(100 M), -glycyrrhetinic acid (10 M), carbenoxolone (10 M) and heptanol (3 M; B),or nonspecific channel inhibitors lanthanum (La3+; 1 mM) and gadolinium (Gd3+; 500 M),or TRP channel inhibitor 2-aminoethoxydiphenyl borate (2-APB; 10 M; C) for 5 min.Cellular proteins were extracted and subjected to Western blot analysis for thephosphorylated AMPK. -actin shown at the bottom of the blots indicates the samephosphorylated AMPK. actin shown at the bottom of the blots indicates the sameamount of loading of protein. Results are representatives of two to three separateexperiments.


Yuan Chi, Kai Li, Qiaojing Yan, Schuichi Koizumi, Liye Shi, Shuhei Takahashi, Ying Zhu, Hi ki M M ki T k d M i Ki d Ji YHiroyuki Matsue, Masayuki Takeda, Masanori Kitamura and Jian Yao


Supplemental Fig. 2

MCP 1MCP-1

β-actin

- ＋- ＋＋-＋- TNFα/IL-1β

FFA

Supplemental Figure 2. Suppression of the cytokines-induced MCP-1 expression byFFA. NRK-52E cells were stimulated with 2 ng/ml IL-1 and 20 ng/ml TNF in thepresence or absence of 50 M FFA. Cellular proteins were extracted and subjected toWestern analysis for MCP-1 Equal loading of protein per lane was verified by probing theWestern analysis for MCP 1. Equal loading of protein per lane was verified by probing theblot with an anti--actin antibody. Result shown are representative of two separateexperiments.


Yuan Chi, Kai Li, Qiaojing Yan, Schuichi Koizumi, Liye Shi, Shuhei Takahashi, Ying Zhu, Hi ki M M ki T k d M i Ki d Ji Y

Supplemental Fig 3

Hiroyuki Matsue, Masayuki Takeda, Masanori Kitamura and Jian Yao


p-AMPKα

A

Supplemental Fig. 3

β-actin

0 5 30 60 180 360

FFA (min)

p-AMPKα

β-actin

B

0 5 30 60 180 360

AICAR (min)

Supplemental Figure 3. Effects of FFA and AICAR on AMPK activation. NRK-E52 cellswere exposed to 50 M FFA (A) or 500 M AICAR (B) for the indicated time intervals. Cellularprotein was extracted and subjected to Western blot analysis using an anti-p-AMPK antibody.The same loading of protein in each lane was verified by probing the blots with an anti--actinantibody.

Documents

Nonsteroidal Anti-Inflammatory Drug Flufenamic Acid Is a ... · Nonsteroidal Anti-Inflammatory Drug Flufenamic Acid Is a Potent Activator of AMP-Activated Protein Kinase S Yuan Chi,