Báo cáo khoa học: Suppression of heat- and polyglutamine-induced cytotoxicity by nonsteroidal anti-inflammatory drugs potx

Suppression of heat- and polyglutamine-induced cytotoxicity
by nonsteroidal anti-inflammatory drugs
Keiichi Ishihara, Nobuyuki Yamagishi and Takumi Hatayama
Department of Biochemistry, Kyoto Pharmaceutical University, Misasagi, Yamashina-ku, Kyoto, Japan
We have shown that sodium salicylate activates the heat
shock promoter and induces the expression of heat shock
proteins (hsps), with a concomitant increase in the
thermotolerance of cells. To determine whether these
effects are generally displayed by nonsteroidal anti-
inflammatory d rugs (NSAIDs), w e examined the effects of
a cyclooxygenase inhibitor, indomethacin, and a lipoxyg-
enase inhibitor, nordihydroguaiaretic acid. Both inhibitors
up-regulated the hsp promoter at 37 °C through t he acti-
vation of heat shock factors, and i ncreased cellular levels
of hsps in mammalian cells, although the degree of the
expression of hsps and t hermotolerance of cells differed
depending on the d rugs. Furthermore, NSAIDs s uch as
sodium salicylate and indomethacin s uppressed the protein
aggregation a nd apoptosis caused by an expanded poly-
glutamine tract in a cellular model of polyglutamine dis-
ease. These findings suggest that NSAIDs generally induce
the e xpression of hsps in mammalian c ells and m ay be used
for the protection of cells against deleterious st ressors and
neurodegenerative diseases.
Keywords: heat s hock proteins; indomethacin; nonsteroidal
anti-inflammatory drugs; nordihydroguaiaretic acid; poly-
glutamine d isease.
Nonsteroidal anti-inflammatory drugs (NSAIDs) such as
sodium salicylate (SA) and indomethacin (IND) are widely
used as analgesic or antipyretic agents for the clinical
treatment of inflammatory diseases. Most NSAIDs exhibit

an inhibitory effect on cyclooxygenases (COXs), which
catalyze the b iosynthesis of prostaglandins and thromb-
oxanes from arachidonic acid. On the other hand, nor-
dihydroguaiaretic acid (NDGA) inhibits lipoxygenases
(LOXs), which produce leukotrienes from arachidonic acid.
Recently, the l ong-term u se of NSAIDs was s hown to
prevent the occurrence of Alzheimer’s disease without COX
inhibition [1]. In addition to having these a nti-inflammatory
effects, SA activates the heat shock promoter and induces
the expression of heat shock proteins (hsps) with a
concomitant increase in the thermotolerance of cells [2].
Cellular resistance against deleterious stress seems to be
regulated by the expression leve ls of hsps in cells [3]. Upon
exposure to a sublethal heat treatment, mammalian cells
acquire transient resistance to a subsequent heat shock that
would be normally lethal, and much evidence supports the
idea that hsps, e specially H sp70, play important roles in its
development [4]. Furthermore, in several polyglutamine
(polyQ) diseases s uch as Huntington’s disease and s pino-
cerebellar ataxia type 3/Machado–Joseph disease, the
polyQ-induced cytotoxicity was suppressed by the over-
expression of Hsp70, Hsp40 and Hsp27 [ 5–10]. Recently,
we have shown that H sp105a reduces the aggregation o f
proteins and cellular t oxicity caused by an expansion of the
polyQ tract using a cellular model of spinal and bulbar
muscular atrophy (SBMA) [11]. Overexpression of Hsp70
also ameliorates S BMA phenotypes in mice [12]. A ddition-
ally, geldanamycin ( a b enzoquinone ansamycin t hat i nduces
the stress response of cells) inhibits huntingtin aggregation
in a cellular model of Huntington’s disease [13]. Thus,

selective hsp inducers may be useful for t he treatment o f
some diseases and for medicinal applications.
In this study, we examined whether NSAIDs generally
induce the expression of hsps and the resistance of cells
using a COX inhibitor IND and a LOX inhibitor NDGA,
and revealed that NSAIDs d o induce the expression of hsps
in mammalian cells and may be used for the protection of
cells against deleterious stressors and diseases.
Materials and methods
Cell culture and drug treatment
Mouse C3H10T1/2 fibroblast cells (Riken cell bank,
Tsukuba, Japan), pGL105/C3H cells stably transfected
with a plasmid containing the mouse Hsp105 promoter
upstream of a luciferase reporter g ene [2] and African green
monkey kidney COS-7 cells (Riken cell b ank) were cultured
in Dulbecco’s modified Eagle’s medium (Nissui Pharma-
ceutical, T okyo, Japan) supplemented with 10% fetal
bovine serum (Equitech-Bio, Kerriville, TX, USA) in a
humidified atmosphere of 5% (v/v) CO
2
in air at 37 °C.
IND a nd NDGA (Nacalai Tesque, Kyoto, J apan) w ere
dissolved in dimethylsulfoxide at 250 m
M
, a nd SA (Nacalai
Correspondence to T. Hatayama, Department of Biochemistry,
Kyoto Pharmaceutical University, 5 Nakauchi-cho, Misasagi,
Yamashina-ku, Kyoto 607–8414, Japan. Fax: +81 75 595 4758,
Tel.: +81 75 595 4653, E-mail:
Abbreviations: AA, arachidonate; COX, cyclooxygenase; GFP, green

fluorescence protein; hsp(s), heat shock protein(s); HSF, heat shock
factor; HSE, heat shock element; luc, luciferase; NSAIDs, nonsteroi-
dal anti-inflammatory drugs; IND, indomethacin; LOX, lipoxygenase;
NDGA, nordihydroguaiaretic acid; polyQ, polyglutamine; SA,
sodium salicylate; SBMA, spinal and bulbar muscular atrophy.
(Received 5 August 2004, rev ised 27 S eptember 2004,
accepted 6 October 2004)
Eur. J. Biochem. 271, 4552–4558 (2004) Ó FEBS 2004 doi:10.1111/j.1432-1033.2004.04419.x
Tesque) was dissolved in phosphate-buffered saline
(NaCl/P
i
)at1
M
just before use. Cells were incubated in
the medium with or w ithout drugs at 3 7 °C for 1 h, washed
with medium three times and further incubated at 37 °C
without the drug.
Measurement of hsp promoter activity
To measure h sp promoter act ivity in mammalian pGL105/
C3H cells (2 · 10
5
cells per 3 5 mm dish), they w ere t reated
with a drug at 37 °C for 1 h and further incubated for 6 h
without the drug. Then, the cells were washed three times
with NaCl/P
i
,andlysedin50lL of Cell Lysis Reagent
(Promega,Madison,WI,USA).Thecelllysateswere
centrifuged at 20 000 g for 1 0 min, and the supernatants
recovered as cell e xtracts. Aliquots (5 lL) of the extracts

were added to 5 0 lL o f l uciferase assay reagent (Promega),
and luciferase ( luc) activity was measured using a T D-20/20
luminometer (Turner Designs, Sunnyvale, CA, USA).
Gel mobility shift assay
C3H10T1/2 cells (5 · 10
5
cells per 60 mm dish), treated
with or without a drug or heat-shocked, were washed with
NaCl/P
i
, and qu ickly frozen a t )80 °C. Frozen cells were
suspended in 100 lL of extraction buffer [20 m
M
Hepes /
KOH, pH 7.9, 1.5 m
M
MgCl
2
,0.2m
M
EDTA, 0.5 m
M
phenylmethanesulfonyl fluolide, 0.5 m
M
dithiothreitol,
0.42
M
NaCl and 25% (v/v) g lycerol], incubated at 4 °C
for 15 min, and vortexed for 15 min at 4 °C. After
centrifugation at 50 000 g for 5 min, aliquots of the

supernatant (15 lg protein) were incubated in 25 lLof
buffer containing 10 m
M
Tris/HCl, pH 7.8, 1 m
M
EDTA,
50 m
M
NaCl, 0.5 m
M
dithiothreitol, 5% (v/v) glycerol,
0.2 mgÆmL
)1
bovine serum albumin, 40 lgÆmL
)1
poly(dI-
dC)] and 0 .4 ngÆmL
)132
P-labeled heat s hock e lement (HSE)
corresponding to nucleotides )115 to )81 of the human
hsp70 gene [14] at 25 °C for 20 min. The mixtures were then
electrophoresed on a native 4% polyacrylamide gel, and the
gel was dried and subjected to autoradiography. T o define
the sp ecific HSF–HSE complex, unlabeled HSE was added
to the reaction mixture in a 100-fold molar excess o f the
labeled HSE. U nder the exp erimental conditions, t he
activated mouse HSF binds to the HSE of the human
hsp70 gene, as does the activated human HSF [15].
Western blot analysis
C3H10T1/2 or COS-7 cells (5 · 10

5
and 6 · 10
5
cells per
60 mm dish, respectively) were lysed in 200 lLof0.1%
(w/v) SDS. Cellular proteins (15 lg) were separated by
SDS/PAGE, and blotted onto a n itrocellulose membrane.
The membrane w as washed with Tris -buffered saline [0.1
M
Tris/HCl, pH 7 .5 and 0.9% (w/v) NaCl] containing 0.1%
(v/v) Tween 20 (TTBS), and reacted with rabbit anti-
Hsp105 [16] or mouse anti-Hsp70 (Sigma Chemical, St.
Louis, MO, USA) a ntibody at room temperature for 1 h .
After w ashes with TTBS, the membrane was further
incubated with horseradish peroxidase-conjugated anti-
(rabbit or anti-(mouse I gG) Igs (Santa Cruz Biotechnology,
Santa Cruz, CA, USA) at room temperature for 1 h.
Hsp105a and Hsp70/Hsc70 were detected using electro-
chemiluminescence reagent (Santa Cruz Biotechnology).
Films were digitized by scanning into
ADOBE P HOTO SHOP
5,
and the intensity of the bands was quantified on a
Macintosh computer using the public domain NIH
IMAGE
program (developed a t US National Institutes of H ealth
and available on the Internet at />nih-image/).
Thermotolerance of cells
C3H10T1/2 cells (7 · 10
4

cells per well) grown in 24-well
plates containing collagen-coated coverslips were incubated
with or without drugs at 37 °C for 1 h. These cells were
heat-shocked a t 4 5 °C f or 45 min a fter incubation at 37 °C
for 6 h i n the absence of drugs, and further incubated at
37 °C. At appropriate times, cells were then washed three
times with NaCl/P
i
, fixed with 4% (v/v) paraformaldehyde
at room temperature for 20 min, and observed under a
phase-contrast microscope (Nikon, Tokyo, Japan).
The viability o f c ells was also assessed b ased on the ability
of living cells to incorporate neutral red into lysozomes. For
the neutral red uptake assay, C3H10T1/2 c ells (7 · 10
4
cells
per well) in 24-well plates were treated with or without
a drug or heat-shocked, then incubated at 3 7 °Cfor3h
with 50 lgÆmL
)1
of neutral red, and fixed with 1% (v/v)
formaldehyde co ntaining 1% (w/v) CaCl
2
for 1 min. The
dye incorporated into cells was extracted with 50% (v/v)
ethanol containing 1% (v/v) acetic acid, and absorbance at
540 nm was measured.
Cellular model of polyQ disease
COS-7 cells grown on c overslips t o 7 0–80% confluence were
treated with or without dr ugs at 3 7 °Cfor1hand

incubated a t 3 7 °C f or 3 h without the d rug. The cells were
then transfected with the e xpression plasmid of a t runcated
androgen receptor containing 97 glutamine repeats fused
to green fluorescence protein (GFP) (tAR97) [9] with
DMRIE-C reagent (Invitrogen, Carlsbad, CA, USA). A fter
incubation at 37 °C f or 72 h, the cells were washed with
NaCl/P
i
, fi xed w ith 4% (v/v) paraformaldehyde f or 3 0 min
at room temperature, and stained with 10 l
M
Hoechst
33342 for 15 min at room temperature. The cells were
observed using a confocal laser s canning microscope (Z eiss,
Jena, Germany). The numbers of transfected cells with and
without visible aggregates were counted independently in
randomly chosen microscopic fields in different areas of a
coverslip. Approximately 300–600 transfected cells were
analyzed for data in each experiment. Apoptotic cells were
identified by their nuclear morphology stained with H oechst
33342 and the TdT-mediated dUTP-biotin end labeling
(TUNEL) method [11]. The TUNEL method was per-
formed using a DeadEnd
TM
apoptosis detection kit (Pro-
mega) according to the manufacturer’s instructions.
Results
Induction of hsp promoter activity, activation of HSF
and accumulation of hsps by IND and NDGA
We first examined the effect of IND and NDGA on the

hsp105 promoter (Fig. 1). When pGL105/C3H cells were
Ó FEBS 2004 Suppression of cytotoxicity by NSAIDs (Eur. J. Biochem. 271) 4553
treated with various concentrations of IND a t 37 °Cfor1h
and further incubated for 6 h with out the drug, luc a ctivity
increased depending on the concentration, and an approxi-
mately 25-fold i ncrease w as observed i n cells pretreated with
0.75 m
M
IND compared to control cells (Fig. 1A ). Under
these conditions, cell viability was not reduced by IND at
concentrations up to 1 m
M
(Fig. 1B). On t he other hand,
luc activity was increased approximately fivefold in cells
pretreated with 0.2 m
M
NDGA compared to the control,
whereas cell death was observed at concentrations of
NDGA of more than 0.2 m
M
.
We next examined whether these drugs e nhance hsp
promoter activity through activation of HSF, by perform-
ing a gel mobility shift assay ( Fig. 2A). When C3H10T1/2
cells were treated with IND or NDGA at 37 °Cfor1h,
HSF was activated in cells treated with 0.5 and 0.75 m
M
IND or 0.2 m
M
NDGA, as in the heat-shocked cells,

although the activation of HSF was at high er levels in cells
treated with IND than with NDGA. Furthermore, w hen
C3H10T1/2 cells were treated with 0.75 m
M
IND at 37 °C
for 1 h and furt her incubated for 6 h , the cellular levels of
Hsp105a and Hsc70 (a constitutive isoform of Hsp70)
increased, and t he expression of Hsp70 (an inducible
isoform of Hsp70) was markedly induced (Fig. 2B). On
the other hand, the increase i n the levels of Hsp105a and
Hsc70 a nd the induction of Hsp70 e xpression were observed
at low levels in t he cells treated with 0.2 m
M
NDGA. T hus,
although NSAIDs seemed to commonly induce hsps at
physiological temperatures, the levels of the expression of
hsps differed depending on the drug.
Induction of thermotolerance of cells by IND
As IND and NDGA caused hsps to accumulate in
mammalian cells, we next examined whether these drugs
Fig. 1. Effect o f IND or NDGA on hsp promoter in pGL105/C3H
cells. (A) pGL105/C3H cells were incubated with or w ithout IND and
NDGA a t 3 7 °C for 1 h, and further incubated at 37 °Cfor6h
without the d rug. Then luc activi ty was assayed, with relative activity
shown as a ratio to that of u ntreated control c ells. Each value repre-
sents the mean ± SE of three independent experiments. Statistical
significance was determined with Student’s t-test; *P <0.01 vs.
respective control c ells. (B) Viability of cells was a ssessed by the neutral
red upt ake assay. Values re present the mean ± SE of three inde-
pendent experiments.

Fig. 2. Effect of IND or NDGA on activation of HSF a nd accumulation
of hsps in mammalian cells. (A) C 3H10T1/2 cells were incubated w ith
or withou t I ND or NDGA a t 3 7 °C for 1 h, or heat-shock ed at 42 °C
for 1 h (HS) a s a positive control. Cell extracts from these cells were
subjected to a gel mobility shift assay using
32
P-labeled HSE. S pecific
HSF–HSE comp lexes we re d etermine d by a dding a 1 0 0-fold ex cess of
unlabeled HSE. A rrows indicate specific HSF–HSE complexes. (B)
C3H10T1/2 cells were incubated with or without IND o r N DGA a t
37 °C for 1 h , a nd then fu rther incubated at 37 °C f or 6 h withou t t he
drug. As a positive control, cells were heat-shocked at 41 °Cfor6h
(HS). Cellular proteins (15 lg) we re sep arate d by 10% S DS/ PAG E,
blotted onto nitrocellulose membranes, and immunostained using
anti-Hsp105 or anti-Hsp70.
4554 K. Ishihara et al.(Eur. J. Biochem. 271) Ó FEBS 2004
induced resistance against s ubsequent heat shock ( Fig. 3A).
The treatment of cells with 0.5 and 0.75 m
M
IND or 0.1 and
0.2 m
M
NDGA at 37 °C for 1 h did not cause marked
changes of cell morphology, and i mmediately a fter exposure
to a lethal h eat shock, the number of cells attached to culture
dishes was not significantly altered. However, when these
cells were further incubated a t 37 °C for 48 h, the number of
cells attach ed to the dishes was markedly dec reased in
untreated controls. H owever, many c ells remained attached
to dishes when pretreated with 0.75 m

M
IND but not 0.1
and 0.2 m
M
NDGA. Furthermore, the viability of c ells was
assessed w ith the neutral red uptake assay (Fig. 3B). The
uptake o f d ye was not affected by IND or NDGA treatment
alone. However, the uptake was markedly suppressed in
untreated control cells 72 h after heat shock at 45 °Cfor
45 min, while pretreatment of cells with 0.75 m
M
IND but
not 0.2 m
M
NDGA significantly suppressed the inhibition.
Thus, cellular r esistance to a subsequent heat stress seemed
to be enhanced by the drug that markedly induced the
expression and accumulation o f hsps.
Suppression of the aggregation of protein containing
an expanded polyQ tract by NSAIDs
Hsps such as Hsp105a, Hsp70 and Hsp40 have been
identified as potent modulators of aggregation and/or cell
death c aused b y the expression of proteins with an expanded
polyQ tract in cellular models of neurodegenerative diseases
[9,11]. We then examined whether NSAIDs suppress the
protein aggregation and apoptosis in a cellular model of
SBMA (Figs 4 and 5 ). When COS-7 cells were transfected
with an expression plasmid of t AR97, approximately
50% of c ells expressing GFP fluorescence of tAR97 were
found to contain protein aggregates and also c ondensed

Fig. 3. Effect o f I ND or NDG A o n t hermotolerance of cells. C3H 10T1/2 cells we re i ncubated w ith or without IND or NDGA a t 3 7 °C for 1 h (a),
and incubated at 37 °C for 6 h with out the d rug. Then , these ce lls were hea t-shock ed at 45 °C for 45 min (b), and further incubated at 37 °Cfor
48 h (c) or 72 h (d). ( A) Cells in (a), (b ) and (c) w ere fixed with 4% ( v/v) paraformaldhyde and o bserved using a phase contrast m icroscope. ( B)
Viability of cells at (b) and (d) was assessed b y neutral red uptake assay. Values repre sent the m ean ± SE of three independ ent exp eriments.
Statistical significance was determined with Student’s t-test; *P < 0.001 vs. respective control cells .
Ó FEBS 2004 Suppression of cytotoxicity by NSAIDs (Eur. J. Biochem. 271) 4555
chromatin, a characteristic of apoptosis. The cells contain-
ing condensed chromatin were TUNEL-positive, another
characteristic of apoptosis (Fig. 4A). Pretreatment of cells
with IND at concentrations of 0.75 m
M
and above
significantly suppressed the protein a ggregation and apop-
tosis in a dose-dependent manner. (Fig. 4B). Under these
conditions, Hsp105a, Hsc70 and Hsp70 we re marke dly
accumulated in cells treated with 0.75 and 1 m
M
IND.
(Fig. 4 C). Furthermore, SA (a potent hsp inducer [2]) at
concentrations of 40 and 80 m
M
significantly suppressed t he
aggregation of tAR97 and apopto sis, with a concomitant
marked induction of Hsp105a and Hsp70 (Fig. 5). How-
ever, NDGA did not suppress the aggregation and apop-
tosis caused b y a n e xpanded polyQ tract (data not sho wn).
Fig. 4. Effects of IND on protein aggregation and cytotoxicity caused by an expanded poly Q tract. (A) COS-7 cells were transfected with tAR97
expression p lasmid and incubated further at 37 °C for 72 h. Cells were t hen fi xed, stained with H oechst 33 342, and o bserved u sing a confocal laser
scanning m icroscope. T ypical images of cells expressing tAR97 (GFP) d iffusely (upper pa nels) or a s aggregates (lower p anels) are sh own. Apoptotic
cells were identified based on t he nuclear morphology (Hoechst) and by t h e TUNEL method. (B) C OS-7 cells were treated w ith 0, 0.5, 0 .75 and

1.0 m
M
IND f or 1 h, i ncubated at 37 °C for 3 h without the drug, and t hen transfected with tAR97 expression plasmid. After 72 h , cells we re fixed,
stained w ith Hoechst 33342 and obser ved using a confocal l aser scanning mic roscope. The a rrowheads represent typical cells c ontaining aggregated
tAR97 or c onde nsed c hromatin . Prop ortions o f cells contain ing tAR97 aggregates or c on densed ch roma tin i n G FP-positiv e c ells rep resen t t he
mean ± SE of four independent expe rimen ts. Statistical signifi cance was dete rmined with S tudent’s t-test; *P < 0.01 vs. respective controls.
(C) Western blot s of Hsp105a and Hsp70/Hsc70 in cells treated with or w it hout IND.
4556 K. Ishihara et al.(Eur. J. Biochem. 271) Ó FEBS 2004
Discussion
SA activates the hsp promoter and induces the expression o f
hsps with a concomitant i ncrease i n t he thermotoleran ce of
cells [2]. Here, we further examined whether other NSAID s
such as IND (a COX inhibitor) and NDGA (a LOX
inhibitor) induced the expression of hsps in mammalian
cells. IND has been shown to activate HSF, but not to
induce a stress response in mammalian c ells [17]. A
pharmacological dose of I ND lowers the temperature
threshold of HSF activation, and a complete heat shock
response can be attained at a temperature that is otherwise
insufficient. In contrast, IND induces the expression of
Hsp70 in human umbilical vein endothelial cells [18] or
induces the nuclear localiz ation of Hsc70 in mammalian
cells [19]. I n this study, we revealed that IND and NDGA
activate HSF at 37 °C and induce hsp expression in
mammalian cells, suggesting that NSAIDs generally induce
the expression of hsps at physiological temperatures.
As to the mechanism by which the stress response is
induced by NSAIDs, these drugs activated HSF at a higher
dose than those needed for the inhibition of COX a nd LOX.
These drugs may activate HSF and induce hsp exp ression

independently of the i nhibition of these e nzymes. H owever,
arachidonate, an essential f atty acid, is released f rom
phospholipids o f the cell membrane b y phospholipase A
2
in response t o e xtracellular s ignals. A rachidonate is oxidized
by COX and LOX to produce various eicosanoids such as
prostaglandins, thromboxanes and leukotrienes [20]. As
arachidonate induces the activation and phosphorylation of
HSF in mammalian cells and in synergy with an elevated
temperature [21], the activation of HSF by NSAIDs may be
in part caused by the inhibition of these e nzymes, b y w hich
the cellular concentration of arachidonate increases.
IND induced thermotolerance of cells, whereas NDGA
did no t s ignificantly induce tolerance. Because hsps play an
important role in the acquisition of cellular resistance to
various forms o f stress [ 4,22,23], the d ifference b etween IN D
and NDGA seems to be due to their ability t o in duce the
expression of hsps. Furthermore, in a cellular model of
SBMA, NSAIDs such as SA and IND suppressed forma-
tion of the protein aggregates and c ytotoxicity caused by a n
Fig. 5. Effects of SA on pro tein aggregation a nd cytotoxicity c aused by an expanded polyQ tract. (A) COS-7 cells were treated with 0, 20, 4 0 and
80 m
M
SA at 37 °C for 1 h, incubated at 37 °C for 3 h without the drug, and then transfe cted with tAR97 expression plasmid. After 7 2 h, cells were
fixed, stained with Hoechst 33342 and observed using a confocal laser scann ing mic roscope. The arrowh eads re present the c ells containing
aggregated tAR97 or c onde nsed chromatin. Propo rtions of cells containing tAR97 aggregates or condensed chromatin in GFP-positive ce lls
represent the mean ± SE of four ind epende nt experiments. S tatistical sign ifican ce was determine d with Stu dent’s t-test; *P < 0.01 vs. r esp ecti ve
controls. (B) Western blots of Hsp105a and Hsp70/Hsc70 in cells treated with or without SA.
Ó FEBS 2004 Suppression of cytotoxicity by NSAIDs (Eur. J. Biochem. 271) 4557
expansion of t he polyQ tract. As the cellular toxicity caused

by an expanded polyQ tract is suppressed by o verexpression
of Hsp105a or Hsp70 in the cellular model [9,11], the
suppression by NSAIDs seemed to be due to the a ccumu-
lation of hsps by these drugs. These findings strongly
support an additional role for NSAIDs as a cytoprotective
agent through the induction of hsps, and suggest that
NSAIDs may be useful for the treatment of stress and
neurodegenerative diseases.
Acknowledgements
This work was supported in part by Gran t-in-Aid for Scientific
Research (T. H.) from the Ministry of E ducation, S cience, C ulture an d
Sports of J apan.
References
1. Sascha,W.,Jason,L.E.,Pritam,D.,Sarah,A.S.,Rong,W.,Claus,
U.P., K irk, A.F., Tawnya, E.S., Michael, P.M., Thomas, B.,
David, E.K., Numa, M., Todd, E.G. & Edward, H.K. (2 001) A
subset of NSAIDs lower amyloidogenic Ab 42 independently o f
cyclooxygenase a ctivity. Nature 41 4 , 212–216.
2. Ishihara, K., Horiguchi, K., Y amagishi, N. & Hatayama, T.
(2003) Identification of sodium salicylate as an hsp i nducer using a
simple screening system for stress response modulators in mam-
malian cells. Eur. J. Biochem. 270, 3 461–3468.
3. Lindquist, S. (1986) Th e heat shock response. Annu. Re v. Biochem.
55, 1151–1191.
4. Landry, J., Bernier, D., Chretien, P., Nicole, L.M., Tanguay,
R.M. & Marceau, N . (1982) Synthesis and degradation of heat
shock proteins d uring development and decay of the rmotoleranc e.
Cancer Res. 42 , 2457–2461.
5. Chai, Y., Koppenhafer, S.L., Bonini, N.M. & Paulson, H.L.
(1999) Analysis of the role of heat shock protein (Hsp) m olecular

chaperones in polyglutamine d isease. J. Neurosci. 19, 10338–
10347.
6. Stenoien, D.L., Cummings, C.J., Adams, H.P., Mancini, M.G.,
Patel, K., DeMartino, G.N., Marcelli, M., Weigel, N.L. &
Mancini, M.A. (1999) Polyglutamine-expanded androgen recep-
tors form aggregates that sequester heat shock proteins, protea-
some components and SRC-1, and are suppressed by t he HDJ-2
chaperone. Hum. Mol. Genet. 8, 7 31–741.
7. Cummings, C.J., Mancini, M.A.,Antalffy,B.,DeFranco,D.B.,
Orr, H.T. & Zoghbi, H.Y. (1998) Chaperone suppression of
aggregation and a ltered s ubcellular proteasome localization imply
protein misfolding in S CA1. Nat. Ge net. 19, 148–154.
8. Jana, N .R., Tanaka, M ., Wang, G . & N ukina, N. ( 2000) Poly-
glutamine length–dependent inte raction of Hsp40 and H sp70
family chaperones with truncated N-terminal huntingtin: their role
in suppression of aggregation and cellular toxicity. Hum. Mol.
Genet. 9, 2009–2018.
9. Kobayashi, Y., Kume, A., Li, M., Doyu, M., Hata, M.,
Ohtsuka, K. & Sobue, G . (2000) Chaperones Hsp70 a nd Hsp40
suppress aggregate formation and apoptosis in cultured n euronal
cells expressing trunc ated androge n receptor prote in with
expanded polyglutamine tract. J. B iol. Chem. 275, 8772–8778.
10. Wyttenbach, A., Sauvageot, O., Carmichael, J., Diaz-Latoud, C.,
Arrigo, A.P. & Rubinsztein, D.C. (2002) Heat shock protein 27
prevents cellular polyglutamine toxicity and suppresses the
increase of reactive oxygen species caused b y huntingtin. Hum.
Mol. Genet. 11, 1 137–1151.
11. Ishihara, K., Yamagishi, N ., Saito, Y., Adachi, H., Kobayashi, Y.,
Sobue, G., Ohtsuka, K. & Hatayama, T. (2003) Hsp105a sup-
presses the aggregation of truncated androgen receptor with

expanded CAG repeats and cell toxicity. J. B iol. Chem. 278,
25143–25150.
12. Adachi, H., Ka tsuno, M., Minamiyama, M., Sang, C. , P agoula-
tos, G., A ngelidis, C., Kusakabe, M ., Yo shiki, A., K obayashi, Y.,
Doyu, M . & Sobue, G. (2003) Heat shock protein 70 chaperone
overexpression ameliorates phenotypes of the spin al and bulbar
muscular atrophy transgenic mouse mod el by reducing nuclear-
localized m utant an drogen receptor pro tein. J. Neurosci. 23, 2203 –
2211.
13. Sittler, A., Lurz, R., Lueder, G ., Priller, J., L ehrach, H ., Hayer-
Hartl, M.K., Hartl, F.U. & Wanker, E.E. (200 1) Geldanamycin
activates a heat shock resp onse and inh ibits hu ntingtin aggrega-
tion in a cell culture model of Huntington’s disease. Hu m. Mol.
Genet. 10, 1307–1315.
14. Williams, G.T., McClanahan, T.K.&Morimoto,R.I.(1989)E1a
transactivation of the human HSP70 promoter is mediated
through t he basal t ranscriptional complex. Mol. Cell Biol. 9, 2574–
2587.
15. Hatayama, T. & H ayakawa, M. (1999) Differential t emperature
dependency of chemical stressors in HSF1-mediated stress re-
sponse in mammali an cells. B ioche m. Biophys . Res. Com m un. 265,
763–769.
16. Honda, K., Hatayama, T. & Yukioka, M. (1989) Common anti-
genicity of mouse 42°C-specific heat-shock protein with mouse
HSP105. Biochem. Biop hys. Res. C ommun. 160, 60–6 6.
17. Lee,B.S.,Chen,J.,Angelidis,C.,Jurivich,D.A.&Morimoto,R.I.
(1995) Pharmacological modulation of heat shock factor 1 by
antiinflammatory drugs results in protection against stress-
induced cellular damage. Proc. Natl Acad. Sci. USA 92, 7207–
7211.

18. Amberger, A., Hala, M., Saurwein-Teissl, M., Metzler, B.,
Grubeck-Loebenstein, B., Xu, Q . & W ick, G. (1999 ) Suppressive
effects of anti-inflammatory agents o n human endothelial cell
activation and induction of heat shock proteins. Mo l. Med. 5, 117–
128.
19. Lagunas, L., Bradbury, C.M., Laszlo, A., H unt, C.R. & G ius, D.
(2004) Indomethacin and ibprofen induce Hsc70 nuclear locali-
zation and a ctivation of the he at shock response i n HeLa cells.
Biochem. Bioph ys. Res. C ommun. 313, 863–870.
20. Goodman, D.S. (1987) The role of arachidonic acid metabolites
in cardiovascular homeostasis. Biochemical, histological and
clinical cardiovascular effects of non-steroidal anti-inflam matory
drugs and their interactions with cardiovascular drugs. Drugs 1,
47–55.
21. Jurivich, D.A., Sistonen, L., Sarge, K.D. & Morimoto, R.I. (1994)
Arachidonate is a potent modulator of h uman heat shock gene
transcription. Proc. Natl Acad. Sc i. USA 91 , 2280–2284.
22. Mosser, D.D., Caron, A.W., Bourget, L., Denis-Larose, C. &
Massie, B. (1997) Role of the h uman heat shock protein hsp70 in
protection against stress-induced a poptosis. Mol. Cell Biol. 17,
5317–5327.
23. Hatayama, T., Yamagishi, N., M inobe, E. & Sakai, K. (2001)
Role of hsp105 in protection against stress-induced apoptosis
in neuronal PC12 cells. Biochem. Biophys. Res. Commun. 288,
528–534.
4558 K. Ishihara et al.(Eur. J. Biochem. 271) Ó FEBS 2004