B R A I N R E S E A R C H 1 2 9 5 ( 2 0 0 9 ) 1 8 6 – 1 9 1

ava i l ab l e a t www.sc i enced i r ec t . com

www.e l sev i e r . com/ loca te /b ra i n res

Research Report

Effects of non-steroidal anti-inflammatory drugs onAβ deposition in Aβ1–42 transgenic C. elegans

Masahiko Moritaa,⁎,1, Kazuhiko Osodab,1, Mayako Yamazakia, Fumiyuki Shiraib,Nobuya Matsuokaa, Hiroyuki Arakawac, Shintaro Nishimurad

aDepartment of Neuroscience, Pharmacology Research Labs, Astellas Pharma Inc., Miyukigaoka Research Center, 21 Miyukigaoka,Tsukuba, Ibaraki 305-8585, JapanbChemistry Research Labs, Astellas Pharma Inc., Miyukigaoka Research Center, 21 Miyukigaoka, Tsukuba, Ibaraki 305-8585, JapancLead Discovery Research Labs, Astellas Pharma Inc., Miyukigaoka Research Center, 21 Miyukigaoka, Tsukuba, Ibaraki 305-8585, JapandApplied Pharmacology Research Labs, Astellas Pharma Inc., Miyukigaoka Research Center, 21 Miyukigaoka, Tsukuba,Ibaraki 305-8585, Japan

A R T I C L E I N F O

⁎ Corresponding author. Fax: +81 29 856 2515.E-mail address: masahiko.morita@jp.aste

1 These authors contributed equally to this

0006-8993/$ – see front matter © 2009 Elsevidoi:10.1016/j.brainres.2009.08.002

A B S T R A C T

Article history:Accepted 2 August 2009Available online 8 August 2009

Although epidemiological studies have shown that long-term treatment with non-steroidalanti-inflammatory drugs (NSAIDs) may protect against Alzheimer's disease (AD), themechanism(s) by which NSAIDs reduce the risk of AD remain to be determined. As C. eleganspossess neither inflammatory cells nor the arachidonate cascade, we could evaluate theeffects of NSAIDs on amyloid β (Aβ) deposition in the absence of immune cells usingAβ-transgenic C. elegans. For this purpose, we established a strain of Aβ-transgenicC. elegans in which thioflavin S-reactive deposits are reproducibly detectable by confocalmicroscopy. Among the NSAIDs examined, ibuprofen and naproxen reduced the number ofthioflavin S-reactive deposits. Furthermore, ibuprofen and naproxen neither affect thethioflavin S binding to Aβ nor Aβ expression in transgenic C. elegans. These data suggest thatibuprofen and naproxen, the most frequently used NSAIDs for the treatment of AD, have aninhibitory effect on Aβ deposition that is independent of the arachidonate cascade andcellular immune systems.

© 2009 Elsevier B.V. All rights reserved.

Keywords:NaproxenIbuprofenNon-steroidal anti-inflammatorydrugs (NSAIDs)Amyloid β (Aβ)Thioflavin SAβ-transgenic C. elegans

1. Introduction

Alzheimer's disease (AD) pathology is characterized by exces-sive deposition of senile plaques in the extracellular space ofthe brain. Themajor component of senile plaques is amyloid β(Aβ) peptide, derived from the proteolysis of amyloid precursorprotein (APP) (Selkoe, 1998). In addition, Aβ oligomers areknown to have a toxic effect on synaptic function. The kineticsof and mechanism behind Aβ oligomerization and fibril

llas.com (M. Morita).work.

er B.V. All rights reserved

formation have been examined using a variety of syntheticAβ peptides. However, the conformation of synthetic Aβ issensitive to local conditions, and in vitro experiments thereforenecessarily require assumptions about the in vivo Aβ environ-ment (Simmons et al., 1994).

Chronic inflammatory responses have been observed inthe brains of AD patients, which are characterized by thepresence of activated astrocytes and microglia surroundingthe senile plaques (McGeer and McGeer, 1998). These injury-

.

187B R A I N R E S E A R C H 1 2 9 5 ( 2 0 0 9 ) 1 8 6 – 1 9 1

responsive cells likely produce a large number of inflamma-tory mediators, including cytokines such as interleukin (IL)-1and tumor necrosis factor (TNF)-α. These observations indi-cate that inflammatory processes play a pivotal role in thepathogenesis of AD, and anti-inflammatory drugs, such asnon-steroidal anti-inflammatory drugs (NSAIDs), may beeffective in the prevention of AD (In't Veld et al., 2001). NSAIDsmay influence the inflammatory response by inhibitingcyclooxygenase (COX) activity and activating peroxisomeproliferator γ (Lehmann et al., 1997).

In an epidemiological study, long-term use of NSAIDsreduced the risk of developing AD (In't Veld et al., 2001). Itwas also shown that ibuprofen, which is one of the mostfrequently used NSAIDs in these studies, significantly sup-pressed amyloid plaque pathology in a transgenic mousemodel for AD (Lim et al., 2000, 2001). On the other hand,Weggen et al. (2001) reported that a subset of NSAIDs reducedthe rate of amyloidogenic Aβ1–42 peptide release from culturedcells, independent of COX activity. This result was confirmedby another study that found NSAID R-enantiomers, theinactive formofCOX (as opposed to S-enantiomers), selectivelyinhibited Aβ1–42 production in a human cell culture (Moriharaet al., 2002). Some reports suggestedNSAIDs lowerAβ1–42 targetγ-secretase derived fromHela cells (Takahashi et al., 2003) andfrom mutated APP cDNA-transfected CHO cells (Eriksen et al.,2003).Moreover, someNSAIDs inhibit the aggregation of Aβ1–42

(Hirohata et al., 2005). Therefore, these are a number ofproposedmechanisms bywhich NSAIDs reduce the risk of AD.

To address some of these areas of uncertainty, we haveused a previously described strain of Aβ1–42 transgenicC. elegans (CL2006) (Link, 1995; Gutierrez-Zepeda and Luo,2004). In this in vivo model, the animals overexpress the 42-amino acids of Aβ and show muscle-specific deposits that areimmunoreactive to anti-Aβ antibodies. A subset of thesedeposits reproducibly bind to the Aβ dye, thioflavin S. Thismodel offers a distinct advantage over in vitro assays, as thetrue characteristics of Aβ may not be reflected in vitro (Iversenet al., 1995). An additional advantage is that neither cellularimmune systems nor the arachidonate cascade, which involvethe cyclooxygenase (COX) gene pathway, are present in C.elegans (Hodgkin, 1999). We therefore pharmacologicallyanalyzed the formation of deposits in Aβ-transgenic C. elegansto evaluate the effect of NSAIDs in the absence of a cellularimmune system. Among the NSAIDs studied, ibuprofen andnaproxen were found to inhibit the formation of thioflavinS-reactive deposits. This indicates that ibuprofen andnaproxen have an inhibitory effect on Aβ deposits indepen-dent of COX activity and the cellular immune system.

Fig. 1 – Thioflavin S and anti-Aβ antibody staining oftransgenic C. elegans. Animalswere visualized using confocalmicroscopy. Representative images of Aβ-transgenicC. elegans (A) and thewild-type C. elegans (N2) (B) stainedwithanti-Aβ antibody (4G8) and Alexa Fluor 488-conjugatedanti-mouse IgG. Yellow arrows indicate positions of Aβdeposit in the muscle of the body wall. Representativeimages of non-treated (C) and 10 μM ibuprofen-treated(D) Aβ-transgenic animals, which were stained withthioflavin S as described in Experimental procedures. Whitearrow heads indicate the positions of Aβ deposits in themuscles of the body wall. Bars: 20 μm.

2. Results

We established three transgenic C. elegans strains (A17, A25,and A35) that express an unc-54/Aβ1–42 gene engineered toproduce a potentially secretary form of human Aβ1–42 in themuscle of the body wall. Unc-54 promoter/enhancersequences in the pPD30.38 vector produce high muscle-geneexpression in C. elegans. The pPD30.38 expression vectorsignal peptide [(SP)-Aβ1–42] contains a synthetic signal peptidesequence and a 189-bp Aβ1–42 sequence derived from human

APP cDNA. These transgenic strains produce extensive anti-Aβimmunoreactive deposits in the muscle of the body wall(Fig. 1A); however, no deposits were observed in non-trans-genic C. elegans (Fig. 1B).

Fig. 3 – Effects of COX-2-selective inhibitors rofecoxiband celecoxib on thioflavin S-reactive deposition inAβ-transgenic C. elegans (A35). Square, rofecoxib (n=30, 27,30, 30, at 0, 0.1, 1.0, 10 μM, respectively); Circle, celecoxib(n=28, 30, 30, at 0.1, 1.0, 10 μM, respectively). Ibuprofen (Bar:10 μM) significantly reduced the number of Aβ-deposits,*P<0.05 vs. control (F(7, 227)=2.30, N=30). Dunnett's multiplecomparison analysis, following ANOVA.

188 B R A I N R E S E A R C H 1 2 9 5 ( 2 0 0 9 ) 1 8 6 – 1 9 1

Thioflavin S-stainedanimals produced thioflavin S-reactivedeposits that were viewed by confocal microscopy (Fig. 1C).Thioflavin S-reactive deposits approximately 2–7 μm in sizewere observed in the muscle of the body wall in transgenicC. elegans. These data indicate that thioflavin S-reactivedeposition is a result of Aβ aggregation. For pharmacologicalanalysis, the thioflavin S-reactive deposits in the anteriorbody wall muscles of C. elegans were measured. Very fewthioflavin S-reactive deposits were observed in wild-typeorganisms (2.8±0.3) (Fig. 2) due to non-specific binding ofthioflavin S; however, the amounts in transgenic C. elegans(23.6±2.8) were significantly greater than that in the wild-typeC. elegans [F(16,472)=8.86, P<0.001, ANOVA, followed byDunnett's test].

Ibuprofen significantly inhibited thioflavin S-reactive de-position in a dose-dependent manner (Fig. 2, 12.4±1.2, 14.4±1.3, P<0.01, P<0.05 at 1, 10 μM, respectively) (Image; Fig. 1D),and also inhibited Aβ deposition in all Aβ-transgenic strains(A17 and A25) (data not shown). Again, the inhibition byibuprofen was observed in Fig. 3 (control; 18.4±1.8, ibuprofen;10.9±0.8) [F(7,227)=2.3, P<0.001, ANOVA, followed by Dun-nett's test]. The anti-Aβ antibody immunohistochemistryresults for ibuprofen-treated Aβ-transgenic C. elegans (datanot shown) agreed with those for thioflavin S staining.Ibuprofen (10 μM) did not inhibit the binding of thioflavinS-bound to Aβ peptide (Table 1). No growth defects wereobserved in ibuprofen-treated transgenic animals (data notshown). In this study, we demonstrated that ibuprofeninhibits Aβ deposition in Aβ42 transgenic C. elegans, and

Fig. 2 – Ibuprofen and naproxen reduced the number ofthioflavin S-stained deposits in Aβ-transgenic C. elegans.NSAIDs- or vehicle-treated Aβ-transgenic and wild-typeC. elegans were fixed, permeabilized, and stained withthioflavin S. A Z-series of images was collected by confocalmicroscopy. Thioflavin S-reactive deposits were counted onthe anterior side of the organisms in the merged Z-seriesimage. The bar indicates the number of thioflavin S-reactivedeposits in wild-type C. elegans. Filled square, naproxen(n=25, 30, 29, 30, at 0, 0.1, 1.0, 10 μM, respectively); Filledcircle, ibuprofen (n=30 in each group); Triangle, flurbiprofen(n=27, 22, 30, at 0.1, 1.0, 10 μM, respectively); Square,indomethacin (n=30, 28, 30, at 0.1, 1.0, 10 μM, respectively);Diamond, diclofenac (n=30, 28, 30, at 0.1, 1.0, 10 μM,respectively). Dunnett's multiple comparison analysis,following ANOVA. *<0.05, **<0.01 vs. control.

additionally, naproxen significantly inhibited this depositionat a dose of 0.1 μM (14.2±1.5, P<0.01) (Fig. 2). On other hand,neither indomethacin, diclofenac, nor flurbiprofen signifi-cantly suppressed Aβ deposition in these animals. Similarly,neither of the COX-2 selective inhibitors examined (rofecoxiband celecoxib) reduced the number of thioflavin S-reactivedeposits (Fig. 3) (F(7,227)=2.30). To determine whether ibupro-fen and naproxen affect the expression of Aβ in vivo, wemeasured the amount of Aβ in transgenic C. elegans. Guani-dine-soluble Aβ1–42 in the transgenic animals was quantifiedby sandwich ELISA. Ibuprofen and naproxen did not changethe total amount of Aβ1–42 in the C. elegans compared with theuntreated transgenic C. elegans (Fig. 4). These data suggestedthat among the NSAIDs examined, ibuprofen and naproxeninhibited the formation of deposits in Aβ-transgenic C. elegans.

3. Discussion

Here, we demonstrated that ibuprofen and naproxen reducethe number of thioflavin S-reactive deposits in the body wallmuscles of Aβ-transgenic C. elegans. This is consistent withresults from epidemiological studies showing the risk of ADis reduced with NSAIDs therapy (In't Veld et al., 2001; Stewart

Table 1 – No effect of ibuprofen and naproxen onthioflavin S-binding to Aβ1–42 deposits.

1 2 Average %

Water 5.45 5.26 5.36 0Aβ1–42 399.1 397.6 398.5 100Aβ1–42 + ibuprofen (10 μM) 411.3 410.5 410.9 103.2Aβ1–42 + naproxen (0.1 μM) 394.2 386.4 390.3 97.9

Data presented as fluorescence intensity.

Fig. 4 – Aβ1–42 amounts in the transgenic C. elegans treatedwith ibuprofen and naproxen. Sandwich ELISA wasperformed to assay Guanidine-soluble Aβ1–42 from vehicle,ibuprofen (10 μM), or naproxen (0.1 μM)-treatedAβ-transgenic C. elegans (A35) (see Experimentalprocedures). Data are presented as percentage of control.

189B R A I N R E S E A R C H 1 2 9 5 ( 2 0 0 9 ) 1 8 6 – 1 9 1

et al., 1997; McGeer, 2000). NSAIDs probably protect against ADvia an anti-inflammatory effect on the microglial cells andastrocytes activated in the brains of AD patients (McGeer andMcGeer, 1998), as well as by anti-amyloidogenic effects on Aβformation (Hirohata et al., 2005). As C. elegans has no suchinflammatory cells, it was considered that NSAIDs unlikelyhave an effect in this organism. We therefore studied theeffect of NSAIDs on thioflavin S-reactive deposition in Aβ-transgenic C. elegans to determine whether these drugs haveany affect on C. elegans in the absence of cellular inflammatorysystems. Our results showed that ibuprofen and naproxenreduced the number of deposits in these animals. Further, nodifference was observed in the fluorescence intensity ofthioflavin S-bound synthetic Aβ peptide in the presence orabsence of ibuprofen or naproxen (Table 1). These resultsindicate that neither ibuprofen nor naproxen inhibit thebinding of thioflavin S to Aβ deposits, and that fluorescenceof bound thioflavin S-Aβ was not inhibited by ibuprofen.Neither ibuprofen nor naproxen changed Aβ expression intransgenic C. elegans (Fig. 4). Aβ-deposit formation was clearlyreduced by ibuprofen and naproxen, leading to the subse-quent evaluation of the effects of COX-2-selective inhibitorsand non-selective NSAIDs on deposits in Aβ-transgenic C.elegans (Fig. 3).

Although themechanism(s) underlying the inhibition of Aβdeposition in C. elegans by ibuprofen and naproxen remains tobe determined, their effects in these animals, which secrete Aβindependent of secretases and endogenous APP (see inExperimental procedures), provide some implications. Ibupro-fen may have an effect on transcriptional factors and nucleus-to-membrane transporters. It should be noted, however, thatalthough ibuprofen and naproxen inhibits the activation andtranslocation of the key translocation factor NF-κB into thenucleus, the 50% inhibitory concentrations of these drugs(≈1.0 mM) are known to be higher than those of indomethacin,diclofenac, andcelecoxib (0.19, 0.36, and0.013mM, respectively)(Takada et al., 2004). Neither ibuprofen nor naproxen changedthe Aβ expression in transgenic C. elegans (Fig. 4). At the dosesused in this study (0.1–10 μM), it is therefore unlikely thatibuprofen or naproxen would inhibit the activation of NF-κB.

In previous studies, 2-(1-[6-[(2-[(18)F]fluoroethyl) (methyl)amino]-2-naphthyl]ethylidene ([F-18]FDDNP) has been used todemonstrate the in vitro anti-aggregation of Aβ peptide, andits labeled form, [F-18]FDDNP, has been used for in vivo plaqueimaging in positron emission tomography (Agdeppa et al.,2003). It has been shown that binding of [F-18]FDDNP to in vitroAβ fibrils is inhibited by ibuprofen and naproxen, but notflurbiprofen or diclofenac (Pignatello et al., 2008). Moreover, ithas been shown by electron microscopy and thioflavin S-staining fluorescence spectroscopy that ibuprofen andnaproxen have anti-amyloidogenic effects against Aβ fibrilsin vitro (Hirohata et al., 2005). These data suggest that theanti-aggregation of Aβ peptide by naproxen and ibuprofenmay result in inhibition of Aβ deposition in C. elegans.

Recently, EGb 761 Ginkgo biloba extract, which is used as adietary supplement for patients suffering from AD, was foundto suppress Aβ oligomerization and Aβ-induced pathologicalbehavior, including paralysis, chemotaxis dysfunction, and 5-HT hypersensitivity, in Aβ-expressing transgenic C. elegans(Gutierrez-Zepeda and Luo, 2004). Suppression of Aβ-inducedparalysis and 5-HT sensitivity is associated with inhibition ofAβ oligomerization (Wu et al., 2006). Naproxen only signifi-cantly inhibited the deposition at a dose of 0.1 μM (P<0.01)(Fig. 2). Although it is known that naproxen is well absorbedorally in rodents and human, there is little information aboutthis in C. elegans. Further experiments investigating the effectsof ibuprofen and naproxen on these Aβ-related physiologicalphenomena in C. elegans are warranted.

In conclusion, we demonstrated that ibuprofen affects theformation of Aβ deposits in C. elegans independent of thecellular immune system and COX activity. This transgenic C.elegans system may be useful for in vivo screening of anti-amyloidosis drugs without the involvement of inflammatorycells such as astrocytes and glial cells.

4. Experimental procedures

Ibuprofen, indomethacin, and diclofenacwere purchased fromSigma (St. Louis, USA). Naproxen, flurbiprofen, celecoxib, androfecoxib were synthesized. The plasmid pPD30.38-signalpeptide (SP)-Aβ42 contained human Aβ1–42 cDNA and a signalpeptide. Aβ cDNA was amplified via PCR (forward primer;5′-GGGGGTACCGATGCAGAATTCCGACATGA-3′, backwardprimer; 5′-CCC GAG CTC ACG CTA TGA CAA CAC CGC CAA-3′),digestedwith Sac-1 and BamH1 (Link, 1995), and subcloned intopcDNA3.1. Signal peptide sequences were attached to Aβ1–42

cDNA via PCR (forward primer; 5′-GCTAGCAAAAATGCATAAGGT TTT GCT GGC ACT GTT CTT TAT CTT TCT GGC ACC AGCAGG TAC CGA TGC AGA ATT CCG A-3′, backward primer;5′-CCC GAG CTC ACG CTA TGA CAA CAC CGC CAA-3′) andcloned into pPD30.38 vector.

The animals were propagated at 30 °C on NGM agar (NaCl3 g/L, Bacto Peptone 2.5 g/L, Bacto Agar 17 g/L)-containing 6 cmdishes seeded with E. coli as a food source. Transgenes wereintroduced into C. elegans by gonad microinjection (Melloet al., 1991). Aβ-expressing transgenes were co-injected withconstructs displaying the dominant morphological marker,rol-6. Transgenic organisms in which the rol-6 marker gene isexpressedmove in a distinctive, non-sinusoidal manner; thus,

190 B R A I N R E S E A R C H 1 2 9 5 ( 2 0 0 9 ) 1 8 6 – 1 9 1

expression can be easily detected by phenotype. Strainscontaining chromosomally integrated transgenes were gen-erated via UV-irradiation. The progeny of irradiated animalswere screened for stable lines using the rol-6 marker pheno-type. Complete transmittance of the chimeric constructs wasconfirmed by immunohistochemically examining large popu-lations of the putative integrated line (Hodgkin, 1999). Asa result, three strains (A17, A25, and A35) of Aβ-transgenicC. elegans were established.

For histological analysis, the Aβ-expressing transgenicC. elegans and the wild-type C. elegans were fixed in 4%paraformaldehyde-PBS for 24 h at 4 °C and permeabilized byincubation in permeabilizing solution (5% mercaptoethanol,1% triton X, 125mMTris–HCl pH 7.4) for 24 h at 4 °C (Link, 1995;Fay et al., 1998). After being washed with PBS, the animalswere stained with 0.125% thioflavin S (Sigma) in 50% ethanolfor 2min andwashedwith 50% ethanol for 2min. The sampleswere kept in PBS until examination. They were then soaked inobservation buffer (50 mM Tris–HCl pH 6.8, 10 % glycerol and10 mM dithiothreitol), and fluorescence images were takenusing a confocal microscope system (model TCS SP2; Leica,Heidelberg, Germany) equipped with an Ar/ArKr laser. Thespecimens were excited at 458 nm and fluorescence emissionwas collected between 470 and 490 nm. A stack of confocal Z-series sections were collected at 0.8-μm intervals.

Immunohistochemical analysis was performed as previ-ously described (Link, 1995). Animals were prepared by fixingin 4% paraformaldehyde-PBS for 16 h and permeabilizing in asolution of 5%mercaptoethanol, 1% triton X, and 125mMTris–HCl (pH 7.4) for 16 h. The permeabilized animals were washedwith staining buffer (5% skim milk and 0.3 % Tween 20 PBS),and then stainedwith anti-Aβmonoclonal antibody 4G8 (EMD,San Diego, USA) and Alexa 488-conjugated anti-mouse IgG(Invitrogen, California, USA). The images were taken byconfocal microscopy, as described above.

Each Aβ-transgenic C. elegans was placed on a singleNSAID-containing agar plate. These transgenic animals con-sumed the drugs from this point. After incubation for 7 days,the adult animals were collected and analyzed by counting thethioflavin S-reactive deposits as described above. For phar-macological analysis, 50 animals were collected from fiveplates per group. The number of thioflavin S-reactive depositsin the anterior muscle wall was determined using a confocalmicroscopy Z-series compilation image, in accordance withthe method established by Morley et al. (2002). Data arepresented as themean±S.E.M. For multiple comparisons, datawere analyzed using analysis of variance (ANOVA), followingDunnett's test.

Aβ polymerization was assayed as described by Hirohataet al. (2005). The reaction mixture contained 10 μM Aβ1–42

(WAKO, Tokyo Japan), 25 mM Tris–HCl, pH 9.0 and 100 mMNaCl. After overnight incubation at 37 °C, aliquots (30 μL) of themixture were put into a 96-well black plate (Corning, NewYork). One hundred and seventy microliters of thioflavin Treaction mixture containing 5 μM thioflavin T, 50 mM phos-phate buffer, pH 7.5, 100 μM ibuprofen or 1 μM naproxen wasadded to polymerized Aβ and incubated for 15 min at roomtemperature. Thioflavin T fluorescence measurements wereconducted using an ARVOsx plate reader (PerkinElmer, MA)with excitation at 430 nm and emission at 486 nm.

The amount of Aβ in Aβ-transgenic C. elegans wasmeasured by sandwich ELISA as the for the procedure forhuman β-Amyloid (1–42) ELISA kit (WAKO). Aβ-transgenicC. elegans treatedwith NSAIDs or vehicle were collected on day7 of incubation and homogenized in 300 μL RIPA buffer (Pierce,Rockford). After protein determination, the homogenatesweresolubilized with 1500 μL of 6 M guanidine buffer (Tris–Cl, pH8.0), and diluted 1:50 with blocking buffer (WAKO). Equalvolumes of the sample were analyzed in sandwich ELISA. Theamount of Aβ1–42 in each sample was adjusted with proteinconcentration. All ELISA was carried out in duplicate. Datawere presented as a percentage of control.

R E F E R E N C E S

Agdeppa, E.D., Kepe, V., Petric, A., Satyamurthy, N., Liu, J., Huang,S.C., Small, G.W., Cole, G.M., Barrio, J.R., 2003. In vitro detectionof (S)-naproxen and ibuprofen binding to plaques in theAlzheimer's brain using the positron emission tomographymolecular imaging probe 2-(1-[6-[(2-[(18)F]fluoroethyl) (methyl)amino]-2-naphthyl]ethylidene) malononitrile. Neuroscience117, 723–730.

Eriksin, J.L., Sagi, S.A., Smith, T.E.,Weggen, S., Das, P., McLendon, D.C.,Ozols, V.V., Jessing, K.W., Zavitz, K.H., Koo, E.H., Golde, T.E., 2003.NSAIDs and enantiomers of flurbiprofen target γ-secretase andlower Aβ42 in vivo. J. Clin. Invest. 112, 440–449.

Fay, D.S., Fluet, A., Johnson, C.J., Link, C.D., 1998. In vivoaggregation of β-amyloid peptide variants. J. Neurochem. 71,1616–1625.

Gutierrez-Zepeda, A., Luo, Y., 2004. Testing the amyloid toxicityhypothesis of Alzheimer's disease in transgenic Caenorhabditiselegans model. Front. Biosci. 9, 3333–3338.

Hodgkin, J., 1999. Conventional genetics. In: Hope, I.A (Ed.),C. elegans: a practical approach. InOxford University Press,New York, pp. 245–270.

Hirohata, M., Ono, K., Naiki, H., Yamada, M., 2005. Non-steroidalanti-inflammatory drugs have anti-amyloidogenic effects forAlzheimer's β-amyloid fibrils in vitro. Neuropharmacol. 49,1088–1099.

In't Veld, B.A., Ruitenberg, A., Hofman, A., Launer, L.J., van Duijn,C.M., Stijnen, T., Breteler, M.M.B., Stricker, B.H.C., 2001.Nonsteroidal anti-inflammatory drugs and the risk ofAlzheimer's disease. New Engl. Med. 345, 1515–1521.

Iversen, L.L., Mortishire-Smith, R.J., Pollack, S.J., Shearman, M.S.,1995. The toxicity in vitro of β-amyloid protein. Biochem. J. 311,1–16.

Lehmann, J.M., Lenhard, J.M, Oliver, B.B., Ringold, G.M., Kliewer, S.A.,1997. Peroxisome proliferator-activated receptors α and γ areactivated by indomethacin and other non-steroidalanti-inflammatory drugs. J. Biol. Chem. 272, 3406–3410.

Lim, G.P., Yang, F., Chu, T., Chen, T., Beech, W., Teter, B., Tran, T.,Ubeda, O., Hsiao-Ashe, K., Frautschy, S.A., Cole, G.M., 2000.Ibuprofen suppresses plaque pathology and inflammation in amouse model for Alzheimer's disease. J. Neurosci. 20,5709–5714.

Lim, G.P., Yang, F., Chu, T., Gahtan, E., Ubeda, O., Beech, W.,Overmier, J.B., Hsiao-Ashe, K., Frautschy, S.A., Cole, G.M., 2001.Ibuprofen effects on Alzheimer pathology and open fieldactivity in APPsw transgenic mice. Neurobiol. Aging 22,982–991.

Link, C.D., 1995. Expression of human β-amyloid peptide intransgenic Caenorhabditis elegans. Proc. Natl. Acad. Sci. U.S.A.92, 9368–9372.

McGeer, E.G., McGeer, P.L., 1998. The importance of inflammatorymechanisms in Alzheimer disease. Exp. Gerontol. 33, 371–378.

191B R A I N R E S E A R C H 1 2 9 5 ( 2 0 0 9 ) 1 8 6 – 1 9 1

McGeer, P.L., 2000. Cyclo-oxygenase-2 inhibitors: rationale andtherapeutic potential for Alzheimer's disease. Drugs Aging 17,1–11.

Mello, C.C., Kramer, J.M., Stinchcomb, D., Ambros, V., 1991.Efficient gene transfer in C. elegans: extrachromosomalmaintenance and integration of transforming sequences.EMBO J. 10, 3959–3970.

Morihara, T., Chu, T., Ubeda, O., Beech, W., Cole, G.M., 2002.Selective inhibition of Aβ42 production by NSAIDR-enantiomers. J. Neurochem. 83, 1009–1012.

Morley, J.F., Brignull, H.R., Weyers, J.J., Morimoto, R.I., 2002. Thethreshold for polyglutamine-expansion protein aggregationand cellular toxicity is dynamic and influenced by aging inCaenorhabditis elegans. Proc. Natl. Acad. Sci. U.S.A. 99,10417–10422.

Pignatello, R., Pantò, V., Salmaso, S., Bersani, S., Pistarà, V., Kepe,V., Barrio, J.R., Puglisi, G., 2008. Flurbiprofen derivatives inAlzheimer's disease: synthesis, pharmacokinetic andbiological assessment of lipoamino acid prodrugs.Bioconjugate Chem. 19, 349–357.

Selkoe,D.J., 1998. Thecell biologyofβ-amyloidprecursorproteinandpresenilin in Alzheimer's disease. Trends Cell Biol. 8, 447–453.

Simmons, L.K., Patrick, C.M., Tomaselli, K.J., Rydel, R.E., Fuson,K.S., Brigham, E.F., Wright, S., Lieberburg, I., Becker, G.W.,Brems, D.N., Li, W.Y., 1994. Secondary structure of amyloid β

peptide correlates with neurotoxic activity in vitro. Mol. Pharm.45, 373–379.

Stewart, W.F., Kawas, C., Corrada, M., Metter, E.J., 1997. Risk ofAlzheimer's disease and duration of NSAID use. Neurology 48,626–632.

Takada, Y., Bhardwaj, A., Potdar, P., Aggarwal, B.B., 2004.Nonsteroidal anti-inflammatory agents differ in their abilityto suppress NF-κ activation, inhibition of expression ofcyclooxygenase-2 and cyclin D1, and abrogation of tumor cellproliferation. Oncogene 23, 9247–9258.

Takahashi, Y., Hayashi, I., Tominari, Y., Rikumaru, K., Morohashi,Y., Kan, T., Natsugari, H., Fukuyama, T., Tomita, T., Iwatsubo,T., 2003. Sulindacsulfide is a noncompetitive γ–secretaseinhibitor that preferentially reduces Aβ42 generation. J. Biol.Chem. 278, 18664–18670.

Weggen, S., Eriksen, J.L., Das, P., Sagi, S.A., Wang, R., Pietrzik, C.U.,Findlay, K.A., Smith, T.E., Murphy, M.P., Bulter, T., Kang, D.E.,Marquez-Sterling, N., Golde, T.E., Koo, E.H., 2001. A subset ofNSAIDs lower amyloidogenic Aβ42 independently ofcyclooxygenase activity. Nature 414, 212–216.

Wu, Y., Wu, Z., Butko, P., Christen, Y., Lambert, M.P., Klein, W.L.,Link, C.D., Luo, Y., 2006. Amyloid-β-induced pathologicalbehaviors are suppressed by Ginkgo biloba extract EGb 761 andGinkgolides in transgenic Caenorhabditis elegans. J. Neurosci. 26,13102–13113.