Hsp105
b
upregulates hsp70 gene expression through
signal transducer and activator of transcription-3
Nobuyuki Yamagishi, Hajime Fujii, Youhei Saito and Takumi Hatayama
Department of Biochemistry & Molecular Biology, Division of Biological Sciences, Kyoto Pharmaceutical University, Japan
Introduction
Heat shock proteins are a set of highly conserved pro-
teins produced in response to physiological and envi-
ronmental stresses that serve to protect cells from
stress-induced damage by preventing protein denatur-
ation and ⁄ or repairing such damage [1]. Mammalian
heat shock proteins are classified into several families
on the basis of their apparent molecular weight and
function, such as Hsp105 ⁄ 110, Hsp90, Hsp70, Hsp60,
Hsp40, and Hsp27. The Hsp70 family is the major and
best-characterized group of heat shock proteins.
Several different species of Hsp70 family proteins are
present in different compartments of eukaryotic cells,
and play important roles as molecular chaperones that
prevent the irreversible aggregation of denatured
proteins. Hsp70 family proteins also assist in the
folding, assembly and translocation across the mem-
brane of cellular proteins [2,3]. Hsp70 family proteins
are commonly composed of three functional domains:
Keywords
heat shock protein; Hsp70; Hsp105b;
nuclear localization; STAT3
Correspondence
T. Hatayama, Department of Biochemistry &
Molecular Biology, Division of Biological

This may represent the mechanism that connects the heat shock protein
and STAT families for cell defense against deleterious stress.
Abbreviations
DOX, doxycycline; HSE, heat shock element; HSF, heat shock factor; INFa, interferon a; JAK, Janus kinase; NLS, nuclear localization signal;
siRNA, small interfering RNA; STAT, signal transducer and activator of transcription.
5870 FEBS Journal 276 (2009) 5870–5880 ª 2009 The Authors Journal compilation ª 2009 FEBS
the highly conserved N-terminal ATPase domain binds
ADP ⁄ ATP, and can hydrolyze ATP; the central
b-sheet domain directly binds peptide substrates; and
the C-terminal a-helix domain regulates substrate
binding [4–6].
Hsp105a and Hsp105b are mammalian members of
the Hsp105 ⁄ 110 family, a divergent subgroup of the
Hsp70 family. Hsp105a is expressed constitutively and
induced by various forms of stress, whereas Hsp105b
is an alternatively spliced form of Hsp105a that is
expressed specifically during mild heat shock [7–9].
These proteins suppress the aggregation of denatured
proteins caused by heat shock in vitro, as does
Hsp70 ⁄ Hsc70, but they have yet to be found to have
refolding activity [10]. In addition, Hsp105a and
Hsp105b were suggested to function as a substitute for
Hsp70 family proteins to suppress the aggregation of
denatured proteins in cells under severe stress, in which
the cellular ATP level decreases markedly [11]. In a
preceding report, we showed that Hsp105a localizes in
the cytoplasm of mammalian cells, whereas Hsp105b
localizes in the nucleus [12]. Furthermore, Hsp105b,
but not Hsp105a, induces the expression of Hsp70 in
mammalian cells [13].

activation of the hsp70 promoter. Furthermore,
Hsp105bmNLS, in which several amino acids in the
nuclear localization signal (NLS) sequence of Hsp105b
were replaced, also failed to activate the hsp70
promoter, as described previously [13].
We next examined the cellular localization of the
deletion mutants of Hsp105b in COS-7 cells by indirect
immunofluorescence, using antibody against myc-tag.
As shown in Fig. 1C, Hsp105b localized to the nucleus
of cells, but Hsp105bmNLS, which failed to activate
the hsp70 promoter, localized to the cytoplasm of cells.
In addition, Hsp105b(1–756), which activates the hsp70
promoter like Hsp105b, localized mainly to the nucleus
of cells, whereas Hsp105b(1–698), Hsp105b(1–677),
and Hsp105b(1–662), which activate the hsp70 pro-
moter to approximately 60% of the extent to which it
is activated by Hsp105b, localized not only to the
nucleus but also to the cytoplasm of cells. However,
Hsp105b(1–641), Hsp105
b(1–591), and Hsp105b(D642–
698), which failed to activate the hsp70 promoter, also
localized to the nucleus and cytoplasm. Interestingly,
Hsp105b(1–564) localized mainly to the nucleus of
cells, like Hsp105b, although it failed to activate the
hsp70 promoter. These results suggest that the nuclear
localization of Hsp105b was necessary but not suffi-
cient for the Hsp105b-induced expression of the hsp70
gene.
The promoter region between )206 and )187 bp
of the hsp70 gene is essential for activation of

moter was dependent on the length of the 5¢-flanking
sequence of the hsp70 gene (Fig. 2C). Thus, Hsp105b
seemed to activate the hsp70 promoter by a mechanism
different from the heat-induced activation of hsp70
promoter, and the region between )206 and )187 bp
of the hsp70 promoter is necessary for Hsp105b-
induced expression of the hsp70 gene.
STAT3, but not HSF1, binds to the region
between )206 and )187 bp of the hsp70
promoter
Next, we investigated protein binding to the region
between )206 and )187 bp of the hsp70 promoter,
using a biotin-mediated oligonucleotide pull-down
assay (Fig. 3A). When the extracts from the cells over-
expressing Hsp105b were incubated with a biotinylated
oligonucleotide probe, GL70, containing the sequence
from )212 to )185 bp of the hsp70 promoter, STAT3
was found to bind to the GL70 oligonucleotide but not
the GL70mt oligonucleotide, which was mutated in the
HSF1 ⁄ STAT3 consensus sequences. On the other hand,
HSF1 binding to GL70 or GL70mt was not observed.
To further examine whether STAT3 bound to the
hsp70 promoter in vivo, chromatin immunoprecipita-
tion analyses were performed using HeLa-tet ⁄ Hsp105b
cells (Fig. 3B). In the cells incubated with doxycycline
(DOX), the chromatin containing the hsp70 promoter
fragment was not coimmunoprecipitated with STAT3
antibody. However, when Hsp105b was overexpressed,
the binding of STAT3 to the region between )272 and
)13 bp of the hsp70 promoter was observed using

tion, and DNA binding [14–16]. We next examined
whether Hsp105b stimulated the phosphorylation of
STAT3 at Tyr705 using antibody against phospho-
STAT3 (Tyr705). When HeLa-tet ⁄ Hsp105b cells were
treated with interferon a (INF a), an activator of the
Janus kinase (JAK)–STAT pathway, STAT3 was phos-
phorylated at Tyr705 and translocated into the nucleus
in cells incubated with or without DOX (Fig. 5A,B).
Furthermore, when Hsp105b was overexpressed in
HeLa-tet ⁄ Hsp105b cells, STAT3 was phosphorylated
and translocated into the nuclei without INFa treat-
ment (Fig. 5A,B), suggesting that Hsp105b stimulates
the phosphorylation of STAT3 at Tyr705 and trans-
location into the nuclei of cells.
Fig. 1. The region between amino acids 642 and 662 of Hsp105b is necessary for activation of the hsp70 promoter by Hsp105b. (A) Sche-
matic representation of a series of C-terminal deletion mutants of Hsp105b. The asterisk in the Hsp105bmNLS mutant indicates the location
of the mutated NLS sequence. (B) The expression plasmids for Hsp105b and its deletion mutants were cotransfected with the
pGL70()2616) reporter construct into COS-7 cells, and luciferase activities were measured. To estimate the transfection efficiency of these
plasmids, the levels of a series of C-terminal deletion mutants of Hsp105b and a-tubulin were determined by western blotting using antibod-
ies against myc-tag and a-tubulin, respectively. Relative activity is expressed as ratio to that of cells cotransfected with pGL70()2616) plas-
mid and pcDNA3.1(+)myc ⁄ His vector. Values represent the means ± standard deviations of four independent experiments, and asterisks
indicate significant differences (*P < 0.01, **P < 0.05). (C) The expression plasmids for Hsp105b and its deletion mutants were transfected
into COS-7 cells. At 48 h after transfection, cells were stained with Hoechst 33342 (blue), and the intracellular distribution of Hsp105b and
its mutants was determined by indirect immunofluorescence microscopy using antibody against myc-tag (red).
N. Yamagishi et al. Mechanism of Hsp105b-induced Hsp70 expression
FEBS Journal 276 (2009) 5870–5880 ª 2009 The Authors Journal compilation ª 2009 FEBS 5873
Downregulation of Hsp105 reduces the induction
of Hsp70 expression during heat shock
To elucidate the physiological role of Hsp105b-medi-
ated regulation of Hsp70 expression, we examined

expression of Hsp70 through the transactivation of
STAT3 in mammalian cells. Indeed, it was found that
STAT3 bound to the sequence between )206 and
)187 bp of the hsp70 promoter, and mutation of the
conserved motif for STAT3 binding in this region
abrogated activation of the hsp70 promoter by
Hsp105b. Furthermore, downregulation of STAT3
expression resulted in reduced Hsp105b-induced
expression of the hsp70 gene.
B
A
Fig. 3. STAT3, but not HSF1, binds to the region between )206
and )187 bp of the hsp70 promoter. (A) HeLa-tet ⁄ Hsp105b cells
were grown in medium without DOX for a period of 48 h. Extracts
from these cells were incubated with biotinylated GL70 or GL70mt
oligonucleotides. After unbound proteins were removed by wash-
ing, proteins bound to the oligonucleotides were eluted with
excess biotin and detected by western blotting, using antibody
against STAT3 or HSF1. The input represented 5% of the protein
used in the pull-down assay. (B) Chromatin immunoprecipitation
analysis was performed with chromatin extracts from HeLa-
tet ⁄ Hsp105b cells grown in medium with or without DOX for a
period of 48 h. DNAs from chromatins immunoprecipitated with
an antibody against STAT3, acetylated histone H4 or normal rabbit
IgG were amplified by PCR with two sets of specific primers for
the hsp70 promoter. The input represents 5% of the material used
in the chromatin immunoprecipitation assays. Similar results were
obtained with two independent experiments.
A
B

48 h, and were treated without or with 100 ngÆmL
)1
INFa for
10 min. (A) Phosphorylated STAT3 (Tyr705) (p-STAT3), STAT3,
Hsp105, Hsp70 and a-tubulin were analyzed by western blotting,
using the respective antibodies. (B) These cells were stained with
Hoechst 33342 (blue), and the intracellular distribution of phosphor-
ylated STAT3 was determined by indirect immunofluorescence
microscopy using antibody against phospho-STAT3 (red).
A
B
C
Fig. 6. Hsp105b upregulates the expression of Hsp70 during mild
heat shock. (A) Hsp105 or control siRNA was transfected into
HeLa-tet ⁄ Hsp105b cells grown in medium with DOX. At 48 h after
transfection, the cells were treated at 42 °C for 5 h, and the
expression of Hsp105, Hsp70 and a-tubulin was analyzed by wes-
tern blotting, using the respective antibodies. (B, C) The density of
bands was quantified by densitometry, and was corrected with the
density of a-tubulin as loading control. Relative levels of Hsp70 (B)
and Hsp105b (C) are represented as ratios of respective levels in
the cells heat-shocked at 42 °C for 5 h with transfection of control
siRNA. Values represent the means ± standard deviations of three
independent experiments, and the asterisks indicate significant
differences at P < 0.01.
Mechanism of Hsp105b-induced Hsp70 expression N. Yamagishi et al.
5876 FEBS Journal 276 (2009) 5870–5880 ª 2009 The Authors Journal compilation ª 2009 FEBS
sites that are located close to the HSEs in the hsp70
and hsp90b promoters. The activation of hsp promot-
ers is mediated through the STAT3 signaling pathway

nucleotide exchange factors for Hsp70 and its ortho-
logs in Saccharomyces cerevisiae, Ssa1p and Ssb1p,
respectively, and enhance Hsp70-mediated chaperone
activity [24–26]. However, although HSP105 family
proteins are important components of the Hsp70 chap-
erone machinery, excess Hsp110 seems to have a nega-
tive effect on Hsp70-mediated chaperone activity,
owing to it accelerating substrate cycling to such an
extent that the reaction becomes unproductive for fold-
ing [25]. Hsp105b-induced expression of Hsp70 may be
important for Hsp70-mediated protein refolding by the
correction of the ratio between Hsp105 and Hsp70
chaperones in the cells. Furthermore, as Hsp105b is
necessary for the marked expression of Hsp70 under
mild heat shock conditions, Hsp105b seems to play an
important role in protection against deleterious stres-
sors by Hsp70. Although further study will be required
to clarify the mechanisms by which Hsp105b induces
the activation of STAT3, the present findings may pro-
vide clues to the cellular function of HSP105 family
proteins in the chaperone network of mammalian cells.
Experimental procedures
Antibodies
The following antibodies were used for western blotting,
immunofluorescence, and gel shift experiments: Hsp105,
rabbit anti-(human Hsp105 IgG) [27]; Hsp70, mouse anti-
(human Hsp70 IgG), which only reacted with inducible
Hsp70 (clone C92F3A-5; Stressgen, Ann Arbor, MI, USA);
myc tag, mouse anti-myc IgG (Invitrogen, Carlsbad, CA,
USA); HSF1, rabbit anti-(human HSF1 IgG) (Stressgen);

ously [28]. These cells were maintained in DMEM supple-
mented with 10% fetal bovine serum at 37 ° C with 95% air
and 5% CO
2
.
Plasmids
The expression plasmids for mouse Hsp105a, Hsp105b and
a mutant with substitutions in the NLS or a series of
C-terminal deletion mutants of Hsp105b in mammalian
cells have been described previously [12,13,29]. The construc-
tion of the reporter plasmid pGL70()2616), which contains
the firefly luciferase cDNA driven by a human hsp70 pro-
moter sequence from )2616 to +150, has been described
previously [13]. A series of 5¢-deletions of the human hsp70
promoter sequence were generated by self-ligation of the
DNA sequence made by PCR, with pGL70()2616) as the
template DNA and a specific set of primers (Table 1). The
substitution construct of pGL70()218) was made using a
QuickChange site-directed mutagenesis kit (Stratagene, La
Jolla, CA, USA), according to the manufacturer’s instruc-
tions. The following oligonucleotides were used as primers
for mutagenesis: 5¢-CAG AAC TCT CCA GAG TCT
GAT
G
AG ATC TAC TGG AGG GGA CAG GGT T-3¢ and
5¢-AAC CCT GTC CCC TCC AGT
AGA TCTCAT
CAG ACT CTG GAG AGT TCT G-3¢ (substituted nucleo-
tides underlined).
Western blotting

of the cell lysates was also determined, to normalize for
protein content. For the knockdown of STAT3 expression,
10 pmol of the STAT3 siRNA (Santa Cruz Biotechnology)
or RISC-free control siRNA (Dharmacon, Chicago, IL,
USA) was cotransfected into HEK293 cells with 1.5 lgof
reporter plasmid and 0.75 lg of the expression vector
for Hsp105b, using Lipofectamine 2000, according to the
manufacturer’s instructions (Invitrogen).
Indirect immunofluorescence
Cells grown on collagenized coverslips (2 · 10
4
cellsÆcm
)2
in
24-well plates) were fixed with 4% paraformaldehyde for
30 min at room temperature and permeabilized with ice-
cold methanol for 10 min. After being washed twice with
NaCl ⁄ P
i
, the cells were incubated with blocking solution
(NaCl ⁄ P
i
containing 5% BSA) at 37 °C for 1 h. Then, the
cells were incubated with mouse anti-myc Ig (1 : 100) or
rabbit anti-phospho-STAT3 (Tyr705) IgG for 1–2 h at
37 °C. After multiple washes with NaCl ⁄ P
i
, the cells were
Table 1. Primers used for construction of 5¢-serial deletions of
hsp70 promoter reporter constructs.

G-3¢; GL70mt, 5¢-biotin-CTC CAG TAG ATC TCA AGA
GAC TCT GGA G-3¢) in binding buffer [50 mm Tris ⁄ HCl,
pH 7.8, 100 mm NaCl, 1.5 mm MgCl
2
,1mm EDTA, 0.1%
NP-40, 10% (v ⁄ v) glycerol, 0.5 mgÆmL
)1
salmon sperm
DNA] at 4 °C for 30 min. The samples were precleared
with CL-4B Sepharose at 4 °C for 30 min, and the remain-
ing DNA was precipitated with 30 lL of a 50% slurry of
Ultralink streptavidin gel (Pierce, Rockford, IL, USA) at
4 °C for 1 h. Bound fractions were washed five times with
binding buffer, and DNA-bound proteins were eluted with
binding buffer containing 1 mgÆmL
)1
biotin; this was fol-
lowed by SDS ⁄ PAGE and western blotting.
Chromatin immunoprecipitation
HeLa-tet ⁄ Hsp105b cells were grown in medium with or
without DOX for a period of 48 h and subjected to chro-
matin immunoprecipitation analysis. Cells (5 · 10
7
cells)
were cross-linked with a final concentration of 1% for-
maldehyde for 8 min at room temperature, and this was
followed by quenching with a final concentration of 0.25 m
glycine. The cell pellets were then collected by centrifuga-
tion (450 g,4°C, 5 min) and washed three times with lysis
buffer (10 mm Tris ⁄ HCl, pH 7.5, 10 mm NaCl, 3 mm

CCA ACA CCC T-3¢; reverse, 5¢-CCC TGG GCT TTT
ATA AGT CGT-3¢), or the human hsp70 gene between
)1860 and ) 656 (forward, 5¢-TCT ATC TCT CGA TGG
ATA CAG A-3¢; reverse, 5¢-AGG ACA GTA GAA TTA
GGT CAC T-3¢).
Knockdown of Hsp105a and Hsp105b
The double-stranded RNA targeting Hsp105 (Dharmacon;
5¢-GCA AAU CAC UCA UGC AAA CUU-3¢) was resus-
pended to make a 20 lm solution, following the manufac-
turer’s instructions. siRNA transfections were carried out
using siLentFect reagent, according to the manufacturer’s
instructions (Bio-Rad, Hercules, CA, USA). At 72 h after
transfection, cells were harvested for western blotting.
Statistical analysis
The significance of differences was assessed using an
unpaired Student’s t-test. A probability level (P) of less
than 0.05 was considered to be statistically significant.
Acknowledgements
This study was supported in part by a Grant-in-Aid
for Scientific Research (C) (No. 1750903) and Young
Scientists (B) (No. 17790077), and a grant for the
Frontier Research Program of Kyoto Pharmaceutical
University from the Ministry of Education, Science,
Sports and Culture of Japan.
References
1 Samali A & Orrenius S (1998) Heat shock proteins:
regulators of stress response and apoptosis. Cell Stress
Chaperones 3, 228–236.
2 Hendrick JP & Hartl F-U (1993) Molecular chaperone
functions of heat-shock proteins. Annu Rev Biochem 62,

Hatayama T (2000) Modulation of the chaperone activi-
ties of Hsc70 ⁄ Hsp40 by Hsp105alpha and Hsp105beta.
Biochem Biophys Res Commun 272, 850–855.
11 Yamagishi N, Ishihara K, Saito Y & Hatayama T
(2003) Hsp105 but not Hsp70 family proteins suppress
the aggregation of heat-denatured protein in the pres-
ence of ADP. FEBS Lett 555, 390–396.
12 Saito Y, Yamagishi N & Hatayama T (2007) Different
localization of Hsp105 family proteins in mammalian
cells. Exp Cell Res 313, 3707–3717.
13 Saito Y, Yamagishi N & Hatayama T (2009) Nuclear
localization mechanism of Hsp105b and its possible func-
tion in mammalian cells. J Biochemistry 145, 185–191.
14 Kishimoto T, Taga T & Akira S (1994) Cytokine signal
transduction. Cell 76, 253–262.
15 Darnell JE, Kerr IM & Stark GR (1994) Jak-STAT
pathways and transcriptional activation in response to
IFNs and other extracellular signaling proteins. Science
264, 1415–1421.
16 Ihle JN, Witthuhn BA, Quelle FW, Yamamoto K,
Thierfelder W, Kreider B & Silvennoinen O (1994)
Signaling by the cytokine receptor superfamily: JAKs
and STATs. Trends Biochem Sci 19, 222–227.
17 Pirkkala L, Nyka
¨
nen P & Sistonen L (2001) Roles of the
heat shock transcription factors in regulation of the heat
shock response and beyond. FASEB J 15, 1181–1131.
18 Baler R, Dahl G & Voellmy R (1993) Activation of
human heat shock genes is accompanied by oligomeri-

Hsp110 is revealed as an Hsp70 nucleotide exchange
factor. EMBO J 25, 2510–2018.
25 Dragovic Z, Broadley SA, Shomura Y, Bracher A &
Hartl FU (2006) Molecular chaperones of the Hsp110
family act as nucleotide exchange factors of Hsp70s.
EMBO J 25, 2519–2028.
26 Shaner L, Sousa R & Morano KA (2006) Characteriza-
tion of Hsp70 binding and nucleotide exchange by the
yeast Hsp110 chaperone Sse1. Biochemistry 45, 15075–
15084.
27 Honda K, Hatayama T & Yukioka M (1988) Com-
mon antigenicity of mouse 42 degrees C-specific
heat-shock protein with mouse HSP 105. J Biochem
103, 81–85.
28 Yamagishi N, Ishihara K, Saito Y & Hatayama T
(2006) Hsp105 family proteins suppress staurosporine-
induced apoptosis by inhibiting the translocation of Bax
to mitochondria in HeLa cells. Exp Cell Res 312, 3215–
3223.
29 Yamagishi N, Saito Y, Ishihara K & Hatayama T
(2002) Enhancement of oxidative stress-induced apopto-
sis by Hsp105alpha in mouse embryonal F9 cells. Eur J
Biochem 269, 4143–4151.
30 Ishihara K, Horiguchi K, Yamagishi N, Saito Y &
Hatayama T (2003) Identification of sodium salicylate
as an hsp inducer using a simple screening system for
stress response modulators in mammalian cells. Eur J
Biochem 270, 3461–3468.
Mechanism of Hsp105b-induced Hsp70 expression N. Yamagishi et al.
5880 FEBS Journal 276 (2009) 5870–5880 ª 2009 The Authors Journal compilation ª 2009 FEBS