2845

Journal of Cell Science 113, 2845-2854 (2000)

Printed in Great Britain The Company of Biologists Limited 2000
JCS1406

Nuclear translocation and aggregate formation of heat shock cognate protein

70 (Hsc70) in oxidative stress and apoptosis
Zubin Dastoor and Jean-Luc Dreyer*
Institute of Biochemistry, University of Fribourg, CH-1700 Fribourg, Switzerland
*Author for correspondence (e-mail: jean-luc.dreyer@unifr.ch)

Accepted 3 June; published on WWW 20 July 2000

SUMMARY
Recent evidence has shown a role for the heat shock cognate
protein Hsc70 in the response to oxidative stress. We have
investigated the subcellular distribution of Hsc70 by means
of laser scanning confocal microscopy in neuroblastoma
NB41A3 cells, in fibroblasts R6 cells and in R6-Bcl-2, an
apoptosis-resistant cell line, and its function in oxidative
stress and in apoptosis has been evaluated. Endogenous
Hsc70 is localised predominantly in the cytoplasm in
unstressed cells, whereas oxidative stress but not apoptosis
induces its translocation into the nucleus. In transfected
cells overexpressing Hsc70 increased nuclear translocation
and aggregation of Hsc70 in intracellular speckles is
observed after oxidative stress and, to a lesser degree, after

exposure to apoptotic agents. Bcl-2 did not influence the

movement of Hsc70 nor the formation of Hsc70-containing
speckles. Nuclear translocation of Hsc70 can be modulated
by the expression of components from a previously
described plasma membrane oxidoreductase involved in
the cellular response against oxidative stress. Our data may
suggest a correlation between differential translocation of
Hsc70 with specific functions in apoptosis and a potential
role in the protection against reactive oxygen species.

INTRODUCTION

During embryonic neurulation Hsc70 levels depend on

endogenous (pro)insulin which also prevents apoptosis.
Apoptosis affects mainly cells with the lowest level of Hsc70,
suggesting a role for Hsc70 in the control of apoptosis in
embryogenesis (de la Rosa et al., 1998). Several studies have
further shown Hsp70 and Hsc70 binding to the anti-apoptotic
protein BAG-1 (Bimston et al., 1998; Hohfeld, 1998; Hohfeld
and Jentsch, 1997; Stuart et al., 1998; Takayama et al., 1997,
1998, 1999). Overexpression of BAG-1 protected from heat
shock-induced cell death whereas overexpression of Bcl-2
causes intracellular redistribution of BAG-1. Mechanistic basis
for BAG-1 as a negative regulator of Hsc70/Hsp70 chaperones
was given by Bimston et al. (1998). Three isoforms of human
BAG-1 (BAG-1, BAG-1M, and BAG-1L) exist which all retain
the ability to bind Hsc70 (Takayama et al., 1998). This reveals
an unexpected diversity in the regulation of Hsc70 and raises
the possibility that the anti-apoptotic function of BAG-1 may
be exerted through a modulation of the chaperone activity of
Hsc70 on specific protein folding and maturation pathways.
In a previous study we have isolated a neuronal transplasmamembrane oxidoreductase (PMO) complex involved in
the cellular response against oxidative stress. One major
components of the complex was identified as Hsc70 (Bulliard
et al., 1997). We therefore aimed in this study to define the role
of Hsc70 in oxidative stress and in apoptosis with respect to
the other components of the PMO complex, in both neuronal
and non-neuronal cells. To clarify these we investigated the
cellular distribution of Hsc70 by means of confocal laser

Heat shock protein Hsc70 and Hsp70 are molecular chaperones

that participate in important cellular processes (Ellis, 1987;
Gething and Sambrook, 1992; Hendrick and Hartl, 1993; Hartl,
1996; Hightower, 1980; Pelham, 1986; Rothman, 1989). Hsc70
plays an important role under non-stress conditions in
stabilising the folding of newly synthesised polypeptides in the
ER (Gething and Sambrook, 1992; Haas and Wabl, 1983) and
for uptake into the mitochondrion and the ER (Chirico et al.,
1988; Deshaies et al., 1988; Zimmermann et al., 1988). It
interacts with nascent polypeptide chains in the cytosol by
binding hydrophobic peptide segments in an ATP-dependent
manner (Beckmann et al., 1990; Flynn et al., 1991). Evidence
was given recently for a role of Hsc70 in the response to
oxidative stress, anti-oxidants preventing gene expression in
the focal regions (Turner et al., 1999). Also hypoxia induces
Hsp/Hsc70 production by increased superoxide generation.
Oxygen radicals and posthypoxic injury leads to the synthesis
of Hsps (Gebhardt et al., 1999) and inhibitors of superoxide
dismutase such as diethyldithiocarbamate induces whereas
superoxide dismutase represses Hsp/Hsc70 synthesis. Hsc70
also modulates heme-regulated eIF-2 kinase (HRI) in heminsupplemented rabbit reticulocyte lysate (RRL) in response to
heat and oxidative stress (Thulasiraman et al., 1998) and Hsc70
may modulate the activation of HRI not only in heme-deficient
RRL, but also in hemin-supplemented RRL in response to
oxidative stress.

Key words: Apoptosis, Confocal laser scanning microscopy, Heat

shock protein 70, Nuclear import, Oxidative stress

2846 Z. Dastoor and J.-L. Dreyer

scanning microscopy and immunocytochemistry. We studied
the effects of transient and stable overexpression of Hsc70GFP in different cell types, either central nervous system
(CNS)-derived cells (mouse neuroblastoma NB41A3), nonCNS derived cells (R6 fibroblast) or an apoptosis-resistant Bcl2 expressing R6 cell line (R6-Bcl-2).
MATERIALS AND METHODS
Cell culture
Neuroblastoma cells NB41A3 were obtained from ATCC (Bethesda,
MD) and maintained as monolayers on tissue culture plastic flasks at
37C in 5% CO2 as previously described (Augusti-Tocco and Sato,
1969). The growth medium was Dulbeccos modification of Eagles
medium (DMEM) at pH 7.4 containing 3.7 g/l NaHCO3, 4.5 g/l Dglucose, 0.11 g/l sodium pyruvate and 0.58 g/l glutamine, 10% foetal
calf serum, 50 units/l of penicillin and 50 g/l of streptomycin. The
rat 6 embryo fibroblast cell line (R6) and a derivative stably
overexpressing Bcl-2 (R6-Bcl-2) were kindly provided by Dr C.
Borner, Institute of Biochemistry, Fribourg, Switzerland (Borner,
1996). They were grown under the same conditions as the
neuroblastoma cells, except that the cell culture medium contained
5% instead of 10% foetal calf serum. For all cell lines the medium
was changed every 3 days. Starving of the cultures and growth to post
confluence were strictly avoided. The cells were trypsinised in the
presence of 0.25% trypsin and 0.1% EDTA for splitting or harvested
for RNA or protein extracts when they were subconfluent.
Construction of plasmids
The complete coding region of Hsc70 was amplified by PCR with the
Advantage Klentaq polymerase mixture (CLONTECH) from mouse
cDNA (mouse brain Marathon cDNA, CLONTECH) with 5-AAA
ACC CGG GGG TAC CAC CTC CAT CGA CCT CTT CAA TAG
TGG GGC CTG-3 as downstream and 5-TCG TCG TCA GCG CAG
CTG GGC CTACA-3 as upstream primer. The downstream primer
was designed such as to replace the stop codon by nucleotides coding
for three glycine residues. Additionally, two artificial restriction sites
KpnI/Asp781I and XmaI were added to facilitate ligation (Tomassini
et al., 1978). Frameshifted clones were produced either by using KpnI
which produces a 4-base 3 extension which is subsequently digested
in a T4 DNA polymerase reaction or by using Asp718I (=Acc65I) that
produces a 4-base 5 extension which serves as a template to add the
missing nucleotides by Klenow fragment. Blunt end re-ligation with
the KpnI cut fragment(s) produces a 1 nucleotide frameshift, whereas
the Asp718I cut fragment(s) produces a +1 nucleotide frameshift.
The amplified products were subcloned into the pKS+ Bluescript
(Stratagene, La Jolla, CA, USA) and the sequence confirmed by
sequencing. Subsequently, the amplified DNA sequence was
subcloned in-frame with the sequence coding for green fluorescent
protein (GFP) into expression vectors pEGFP-N1 or pEBFP-N1 (both
Clontech), respectively.
Control plasmids that do not produce GFP, but only the Hsc70
protein, were prepared by disrupting the GFP sequence by a frameshift mutation at the artificial KpnI/Asp718 restriction site between the
3-end of the Hsc70 gene and the start codon of GFP. Correct
expression of these constructs was confirmed by immunofluorescence
assays in transfected cells. All plasmids were purified on QIAGEN
Midi columns.
Transfection
Cells grown on 12 mm glass coverslips in 6-well plates were
transfected at 80-90% confluence using SuperFect (Qiagen). 1 g of
plasmid DNA suspended in 60 l fresh DMEM without serum and 5
l SuperFect were mixed, vortexed and incubated for 10 minutes. 600
l of complete growth medium was then added and immediately

transferred to PBS-washed cells. After 3-4 hours of incubation at

37C/5% CO2, cells were washed once with PBS.
Induction of oxidative stress and apoptosis
Immediately after transient transfection, cells were trypsinised, split
1:3 and plated on glass coverslips placed in 6-well plates. After 24
hours the growth medium was replaced by new medium (control) or
new medium containing either 300 M FeCN, 100 M H2O2, 1.5 M
MG132 or 200 nM staurosporine. Cells were exposed for 3 hours (to
300 M FeCN or 100 M H2O2) and for 6 hours or 24 hours (to 200
nM staurosporine or 1.5 M MG132) and analysed under the
microscope with and without prior fixation. The percentage of dead
cells correlated logarithmically with increasing concentrations of
H2O2 and FeCN.
Immunocytochemistry
Mouse anti-Hsc70 monoclonal antibody (Mab) (IgG) (Affinity
BioReagents) and rabbit anti-GFP polyclonal antibody (IgG)
(Clontech Laboratories, Inc., Palo Alto, CA, USA) were used as
primary antibodies. As secondary antibodies Texas Red dyeconjugated goat anti-mouse IgG, Texas Red dye-conjugated goat antirabbit IgG, fluorescein isothiocyanate (FITC)-conjugated goat antimouse IgG or FITC-conjugated goat anti-rabbit IgG (all Jackson
ImmunoResearch Laboratories, Inc., Bar Harbor, ME, USA) were
used for immunocytochemistry. When native GFP fluorescence was
used to localise the GFP fusion proteins, the coverslip containing the
adherent cells was directly placed on top of a drop of growth medium
without Phenol Red on a microscope slide and immediately analysed.
Otherwise cells were rinsed twice in PBS and fixed for 15 minutes in
4% p-formaldehyde in 0.1 M PIPES (pH 6.8) containing 0.1%
Hoechst 33342. Cells were then rinsed twice in PBS, permeabilised
for 5 minutes with 0.05% saponin in PIPES, again rinsed twice in PBS
and incubated for 10 minutes at 20C in acetone. After 2 washes
with PBS, cells on the coverslip were incubated in 1% BSA in PBS
for 2 hours with the first antibody, rinsed twice in PBS and further
incubated for 2 hours with the fluorescent secondary anti-IgG before
two washes in PBS. Fluorescence was extended by using SlowFade
(Molecular Probes, Leiden, Netherlands). All incubations were
performed in a light protected environment and, except for acetone,
at room temperature.
Fluorescence microscopy
Cells were examined with a Nikon Eclipse E800 microscope under a
100 oil immersion objective (Nikon Inc.) using a MRC-1024
Bio-Rad laser confocal microscope system equipped with an
Krypton/Argon laser (Bio-Rad Labs, Hercules, CA, USA) and the
LaserSharp acquisition software (Bio-Rad). Serial sections (at 0.5-1.5
m intervals) in the z-axis of cells were collected in the slow scanning
mode (approx. 160 Hz) with 1.5-4.5 mm diam. iris aperture by
averaging pictures by Kalman (3 scan). Green fluorescence (GFP or
FITC) and Texas Red were detected in parallel at 512512 pixels. The
pictures were processed with the LaserSharp processing software
(Bio-Rad) or alternatively with ConfocalAssistant before picture files
were transferred to Adobe Photoshop 4.0 and printed on a Kodak
printer.
To visualise DNA staining with Hoechst 33342, a 100 oil
immersion Plan-Neofluar objective (Zeiss) with an inverted Zeiss
Axiovert 135 TV microscope equipped with a HBO 100-XBO 75
lamp was used.
Quantification
The serial sections through the z-axis of the cell were collected within
a linear range of fluorescence intensity, superimposed to a 3-D stack
and analysed as a 2-D projection. To compare GFP expression in
different subcellular structures, the mean pixel intensity in a selected
area was determined and normalised by subtracting the background.
In short, all sections collected in the z-axis of the appropriate picture

Nuclear import of HSC70 in stress and apoptosis 2847

by LSCM are stacked to a two-dimensional projection in which the
intensities of the various sections are accumulated. A histogram is
then prepared from the area under investigation in the 2-D-projection,
evaluating the mean pixel intensity and the distribution of pixel
intensities versus the standard deviation. A high mean pixel intensity
correlates to high amount of expressed protein, whereas a high
standard deviation indicates a patchy, irregular expression in the
specific area. The final average value for protein expression levels was
obtained by analysing each time at least 15 different cells of three
independent experiments.

RESULTS
Subcellular translocation of endogenous Hsc70
from the cytoplasm into the nucleus after induction
of oxidative stress
We determined the subcellular localisation of endogenous
HSC70 in CNS-derived cells (neuroblastoma mNB41A3), in
non-CNS derived cells (R6 fibroblast) and in an apoptosis-

resistant Bcl-2 expressing cell line (R6-Bcl-2), using confocal

laser scanning microscopy. Endogenous and transfected
Hsc70 (or Hsc70-GFP) was monitored by means of
immunocytochemistry using several antibodies, mainly clones
5A5 and 2A4 of Affinity Bioreagents, Inc (Milan TM), to
exclude the possibility that the observed changes could be
underestimated because of protein-protein interactions. The
clone 5A5 recognises amino acids (AA) 122-264, which
correspond to a sequence within the ATP binding domain of
Hsc70, whereas clone 2A4 recognises AA 437-479, which are
within the substrate-binding domain. Always both antibodies
were used in separate experiments and yielded similar
quantitative data, so the use of any antibody does actually not
affect our data.
In unstressed NB41A3 cells about 70% of endogenous
Hsc70 is expressed in the cytoplasm as shown by means of
immunocytochemistry with anti-Hsc70 antibodies (Figs 1A,C,
2A). The ratio of the measured Texas Red fluorescence
between cytoplasm and nucleus is constant in all NB41A3

Fig. 1. Subcellular localisation of endogenous Hsc70 as detected by Texas Red (TR)-immunocytochemistry. (A-E) Unstressed control cells:
(A) Endogenous Hsc70, as well as (B) Hsc70-GFP is displayed predominantly in the cytoplasm of NB41A3 cells. (C) Overlay of A, B with
additional Hoechst staining of the nucleus. The most upper transfected cell (yellow=green+red) shows similar Hsc70 localisation like the other
untransfected cells (red). The nuclei (blue) are not fragmented. (D) R6 and (E) R6-Bcl-2 cells both showing cytoplasmic localisation of Hsc70.
(F,G) Induction of oxidative stress: Increased expression of endogenous Hsc70 in the nucleus of (F) NB41A3 and (G) R6 cells after exposure to
100 M H2O2 for 3 hours. (H-P) Induction of apoptosis: Apoptotic cells display neither overexpression nor nuclear accumulation of endogenous
Hsc70 (H,I) in NB41A3 cells after exposure to 200 nM staurosporine for 24 hours and in (J,K) in R6 cells or in (L) R6-Bcl-2 cells after exposure
to 1.5 M MG132 for 6 hours. The non-apoptotic cells express Hsc70 predominantly in the cytoplasm. (M, N) R6 cells form long processes after
6 hours induction of apoptosis by MG132. Most of them are not apoptotic. The arrow points to the only cell showing chromatin condensation.
(O, P) Non-apoptotic R6-Bcl-2 cell with intact nucleus (blue) showing some nuclear translocation after exposure to 1.5 M MG132 for 6 hours.
(C,H,J,L,M,O) Visualisation of DNA by Hoechst blue staining. Apoptotic cells are marked with an arrow. (A,C,D,E,F,G,I,K,L,N,O) Detection of
Hsc70 by TR staining of anti-Hsc70 Ab. (B,C) Visualisation of transfected Hsc70 by GFP tag. Bar, 50 M.

2848 Z. Dastoor and J.-L. Dreyer

Ratio of cytoplasmic : nuclear expression of Hsc70-GFP

Ratio of cytoplasmic : nuclear expression of Hsc70

Subcellular expression of endogenous Hsc70 in untransfected cells

3.0

2.5

2.0

1.5

1.01

0.5

0
control

H 2O 2
R6

STS

control

H 2 O 2 STS
R6-Bcl-2

control H 2 O 2
NB41A3

STS

Subcellular expression of Hsc70-GFP in transfected cells

3.0

2.5

2.0

1.5

1.0

0.5

0
control

H2O2
R6

STS

control

H 2O 2 STS
R6-Bcl-2

control H 2O 2
NB41A3

STS

Fig. 2. Subcellular expression of Hsc70 in untransfected and in

Hsc70-GFP transfected cells. (A) Untransfected and (B) transiently
transfected cells were analysed under unstressed conditions, after
exposure to 100 M H2O2 for 3 hours or after exposure to 200 nM
staurosporine (STS) for 24 hours. The ratio between cytoplasmic and
nuclear Hsc70 (or Hsc70-GFP) expression is indicated. Ratios over
1.0 display predominantly cytoplasmic expression. Nuclear
translocation of Hsc70 is more pronounced after oxidative stress than
after induction of apoptosis. Each experiment consisted of analysing
15 cells and was repeated three times. Thus, each data point indicates
the average of 45 individual measurements.

cells. In resting R6 fibroblasts endogenous Hsc70 is also

predominantly in the cytoplasm and bound to the cytoskeleton
(Fig. 1D). Its highest concentration is seen in the perinuclear
area and at the plasma membrane. Cytoplasmic Hsc70 levels
are higher in R6 cells and R6-Bcl-2 cells than in NB41A3 cells
(Fig. 2A). However, 20% R6 cells are very flat and show a
volcano-like Hsc70 distribution (Fig. 1E).
A series of tests were performed to obtain standard
conditions of oxidative stress at which no apoptotic cells with
nuclear fragmentation were observed and where the total
amount of death cells was less than 20%. These limits were
achieved by exposing cells to either 100 M H2O2 or to 300
M FeCN for 3 hours.
Oxidative stress induced by these conditions raises the
expression level of endogeneous Hsc70 in the nucleus in
NB41A3, whereas the cytoplasmic expression level does not

change (Figs 1F, 2A). Nuclear Hsc70 is often concentrated

around the nucleolus. Most R6 and R6-Bcl-2 cells still
expressed endogenous Hsc70 predominantly in the perinuclear
area of the cytoplasm 3 hours after induction of oxidative stress
with 100 M H2O2 (Fig. 1G), but Hsc70 bound less tightly to
cytoskeletal structures than in unstressed cells. In fibroblasts
20% endogenous Hsc70 was translocated into the nucleus
where it was often concentrated in nuclear patches or at the
nuclear membrane.
Nuclear translocation of endogenous Hsc70 is
observed after induction of apoptosis
For investigating apoptosis we tested three potent inducers of
apoptosis at different concentrations: FAS APO(100 ng/ml and
400 ng/ml), MG132 (1 M, 1.5 M, 2 M and 3 M) and
staurosporine (100 nM, 200 nM and 500 nM). Whereas FAS
APO was not able to induce cell death efficiently, 200 nM
staurosporine and 1.5 M MG132 yielded reasonable
percentages of apoptotic NB41A3 and R6 cells after 24 hours
of exposure. Therefore staurosporine and MG132 were used at
these concentrations to induce apoptosis.
After 6 hours exposure of NB41A3 to either staurosporine
or MG132, endogenous Hsc70 displays higher levels of
expression in the nucleus, but the effect is much less
pronounced than after oxidative stress. Even after 24 hours
only small changes in subcellular localisation of endogenous
Hsc70 were observed in non-apoptotic NB41A3 cells (Figs
1H,I, 2A). Nuclear fragmentation is observed in 20% of
NB41A3 cells (Figs 1H, 3A), but these apoptotic cells do not
accumulate or overexpress endogenous Hsc70 (Fig. 1I).
After exposure to staurosporine or to MG132 most R6 cells
have an unchanged morphology and display endogenous
Hsc70 predominantly in the cytoplasm (Fig. 2A). However a
subset of R6 cells typically start to form long processes
extending from their rounded-up cell body leading to a net-like
appearance (Fig. 1N). Endogenous Hsc70 is displayed in the
cytoplasm, including within the processes. 20% of these netlike cells concentrate endogenous Hsc70 in specific nuclear
regions and are apoptotic (Fig. 1M). Other apoptotic R6 cells
are mostly round and display endogenous Hsc70 in the
cytoplasm or all over the cell, at levels close to endogenous
Hsc70 expression in non-apoptotic control R6 cells (Fig. 1J,K).
The overall percentage of R6 cells with fragmented nuclei after
exposure to staurosporine or MG132 increases from 10% after
6 hours up to 60% after 24 hours (Fig. 3A).
Bcl-2 protects cells from the induction of apoptosis by
MG132 or staurosporine (Fig. 3A). After 6 hours exposure to
staurosporine, about 95% of the R6-Bcl-2 cells did not display
nuclear fragmentation. Different patterns of Hsc70 localisation
were observed: either (a) an unchanged cytoplasmic
localisation in most cells with unchanged morphology, or (b)
expression all over the cell (including the nucleus) within a
network of cells with long processes that display endogenous
Hsc70, or sometimes (c) localisation in a perinuclear ring and
in the nucleus within flat volcano-like cells (Fig. 1O,P). After
24 hours of staurosporine treatment most R6-Bcl-2 cells had
small rounded-up cell bodies and formed long processes (data
not shown). After incubation with MG132 for 24 hours, in
contrast, endogenous Hsc70 was predominantly in the
cytoplasm or in the blebs of the cells, of which the majority
did not have any processes at all (data not shown). After

Nuclear import of HSC70 in stress and apoptosis 2849

Apoptosis in untransfected cells

70

% apoptotic cells

60
50
40
30
20
10
0
0
6 24
6 24
control MG132 STS

0
6 24
6 24
control MG132 STS

0
6 24
6 24
control MG132 STS

R6-Bcl-2

NB41A3

R6

% apoptotic cells

Apoptosis in Hsc70-GFP transfected cells

70

Table 1. Relative mean pixel intensity of total Hsc70

expression in NB41A3 cells

60

Relative mean pixel intensity (%)*

40

Endogenous
Hsc70

Non-apoptotic cells
Apoptotic cells

NB41A3
100
100

R6
100
100

30

Transfected
Hsc70-GFP

Non-apoptotic cells
Apoptotic cells

300
300

250
250

50

20
10
0
0
6 24
control MG132

6 24
STS

0
6 24
6 24
control MG132 STS

0
6 24
6 24
control MG132 STS

R6-Bcl-2

NB41A3

R6

C
% cells forming Hsc70-positive speckles

Fig. 3. Apoptosis and formation of intracellular Hsc70-positive

speckles in untransfected and in Hsc70-GFP transfected cells.
(A,B) Apoptosis in (A) untransfected and (B) transfected cells after
exposure to 200 nM staurosporine (STS) or to 1.5 M MG132. The
percentage of apoptotic cells showing nuclear fragmentation or
chromatin condensation as detected by Hoechst blue staining in cells
expressing endogenous Hsc70 or transfected Hsc70-GFP is indicated
after 0, 6 and 24 hours of exposure time. Bcl-2 overexpressing R6
cells are resistant to apoptosis induced by MG132 or staurosporine.
Each data point is an average of three individual analyses of different
populations, each containing at least 50 cells. (C) Percentage of
transfected cells exposed for different time (from 0-6 hours) to 100
M H2O2 or to 300 M FeCN forming Hsc70-GFP positive
speckles. Additionally, a set of cells (24+3) was first stressed for 6
hours, incubated further for 18 hours in normal growth medium and
then stressed again for 3 hours. The increase in formation of speckles
correlates with longer exposure times to stressing agents. Each data
point is an average of three individual analyses of different
populations, each containing at least 50 cells.

Time curve of Hsc70-GFP speckles formation

after oxidative stress in NB41A3 cells

40
35
30

300 M FeCN
100 M H2O2

25
20
15
10
5
0

0h

2h

3h

4h

5h

6h

24+3 h

Exposure to oxidative stress in hours

induction of apoptosis the morphology of R6-Bcl-2 and R6

cells was similar, in both apoptotic and non-apoptotic cells
(Fig. 1L).
GFP-tag does not influence subcellular localisation
of transfected HSC70-GFP
Overexpression of Hsc70 was conducted by means of

*The relative mean pixel intensity was obtained by measuring total Hsc70
expression and mean pixel intensity in several apoptotic and non-apoptotic
cells by means of histograms. The mean intensity of endogenous Hsc70 in
non-apoptotic cells was defined as 100%. Each experiment consisted of
analysing 15 cells and was repeated three times. Thus, each data point
indicates the average of 45 individual measurements. s.d. is smaller than 5%
in all tested conditions.

transfecting either Hsc70 or the Hsc70-GFP hybrid. Under our

experimental conditions transfection efficiency is about 15%
in all cell lines. Under resting conditions transfected R6 and
NB41A3 cells expressed in the average about 2.5-3 fold more
Hsc70 than untransfected cells (Table 1).
Transfection of GFP alone as a control showed that in
NB41A3 cells (Fig. 4A) as well as in R6 cells (Fig. 4B) GFP
was uniformly displayed all over the cell. No nuclear
fragmentation was observed and, independent of the level of
GFP expression, no differences in morphology or cell viability
were observed compared to untransfected cells (data not
shown). Induction of oxidative stress with 100 M H2O2 or
300 M FeCN for 3 hours did not change the cell morphology
nor the expression pattern of GFP, which was still distributed
all over the cell. After exposure to pro-apoptotic reagents, the
percentages of GFP-transfected apoptotic NB41A3 and R6
cells were comparable to untransfected cells (data not shown).
Cells were shrivelled up and showed slightly increased levels
of GFP, however, with no increased nuclear translocation of
GFP.
In order to test whether the subcellular localisation or
expression level of proteins is unchanged after cell fixation and
immunostaining we analysed the green fluorescence in living
cells transfected by Hsc70-GFP and compared the results with
immunofluorescent analysis where Hsc70-GFP was indirectly

2850 Z. Dastoor and J.-L. Dreyer

Fig. 4. Cellular localisation of Hsc70-GFP in transiently transfected NB41A3, R6 cells and R6-Bcl-2 cells. (A-B) Controls of transfection with
pEGFP: Cellular localisation of GFP in (A) NB41A3 neuroblastoma cells or (B) R6 fibroblast cells transfected with pEGFP (a control vector
devoid of Hsc70 gene). GFP is expressed in the entire cell. Viability of the cells was not influenced by transfection with GFP. (C-D) NB41A3
cells transfected with pHsc70 (a vector devoid of GFP-tag): The expression of Hsc70 (as detected by TR immunohistochemistry) was tested in
(C) unstressed NB41A3 cells and in (D) unstressed R6. The expression is similar like in Hsc70-GFP transfected NB41A3 and R6 cells under
the same conditions (E,F). (E- G) Unstressed cells transfected with HSC70-GFP: (E) Unstressed NB41A3 cells express Hsc70-GFP mainly in
the cytoplasm, whereas (F,G) unstressed R6 cells show some increase in nuclear localisation of Hsc70-GFP compared to (G) untransfected
cells. (H-L) Induction of oxidative stress: NB41A3 cells exposed to 300 M FeCN show (H) patchy accumulation of Hsc70-GFP in the nucleus
and (I) formation of Hsc70-GFP-positive speckles. (J) Nuclear translocation and (K) initial stage of intracellular speckles formation in NB41A3
cells exposed to 100 M H2O2 for 3 hours. (L) Increased nuclear expression of Hsc70-GFP in R6 cells exposed to 100 M H2O2 for 3 hours.
(H-P) Induction of apoptosis: (M) 30% NB41A3 cells show nuclear translocation 24 hours after induction of apoptosis by 200 nM
staurosporine. (N,O) Nuclear translocation in R6-Bcl-2 cells with intact nucleus (blue) 6 hours after induction of apoptosis by 200 nM
staurosporine leads to Hsc70-GFP expression in the entire cell. (N, G) Visualisation of DNA by Hoechst blue staining. (C,D,G) Detection of
Hsc70 by TR staining of anti-Hsc70 Ab. (E,F,H,I,J,K,L,M,O) Visualisation of transfected Hsc70 by GFP tag. Bar, 50 M.

detected by anti-GFP antibodies in fixed cells. Our experiments

demonstrated neither qualitative nor quantitative differences
between fixed and unfixed cells.
The effect of the GFP-tag on protein localisation was tested
by transfection with either the Hsc70-GFP or the Hsc70
(without tag) expression vector. In both cell lines the
subcellular distribution of transfected Hsc70 (red) (Fig. 4C,D)

was similar to that of transfected Hsc70-GFP (green) (Fig.

4E,F) irrespective of whether cells were unstressed, or exposed
to either H2O2, FeCN, staurosporine or MG132 (data not
shown). Also the number of cells undergoing apoptosis in
response to staurosporine or MG132 were comparable. These
observations indicate that the GFP tag does not influence the
expression and localisation of Hsc70.

Nuclear import of HSC70 in stress and apoptosis 2851

Increased nuclear translocation and accumulation of
Hsc70-GFP in intracellular speckles after induction
of oxidative stress is not prevented by Bcl-2
After transient transfection with Hsc70-GFP NB41A3 cells
contained in the average about 65% GFP fluorescence in the
cytoplasm, 1.7 times more than in the nucleus (Figs 4E, 2B).
This ratio was constant in all cells and similar to ratios in
untransfected cells (Fig. 1B,C). Hsc70 was displayed at
different intensities in the cytoplasm in a patchy pattern. The
nucleolus was devoid of Hsc70-GFP and signs of apoptosis
were not detected (Figs 1C, 3B). The distribution pattern of
Hsc70-GFP was very similar among all unstressed NB41A3
cells and did not depend on the shape of the cells. In R6 and
R6-Bcl-2 cells most transiently transfected Hsc70-GFP is
localised in the perinuclear area of the cytoplasm, but their
Hsc70-GFP levels in the nucleus are higher than those of
untransfected fibroblasts (Fig. 4G). No change in shape or cell
viability was observed compared to untransfected R6 and R6Bcl-2 cells.
After exposure to 100 M H2O2 slightly over 50% of Hsc70GFP was localised in the nucleus of stressed NB41A3 cells
(Figs 4J, 2B). This stress resulted in formation of speckles
accumulating Hsc70-GFP in 25% of NB41A3 cells. The
remaining NB41A3 cells were mostly of round shape, had
short processes and their nuclear Hsc70-GFP level was
increased. A correlation between the exposure time to 100 M
H2O2 and the amount of speckles formation was observed (Fig.
3C). Between 15 and 35% of NB41A3 cells accumulated
Hsc70-GFP fluorescence in speckles depending upon the
incubation time. Besides NB41A3 cells expressing Hsc70-GFP
in speckles only, but neither in the nucleus nor in the
cytoplasm, we also observed cells at intermediate stages, where
speckle formation is observed in particular regions of the cell
(Fig. 4K). With both stress agents, H2O2 and FeCN, the ratio
of nuclear translocation increases in parallel with exposure
times.
Oxidative stress induced by 300 M FeCN for 4 hours raised
the level of Hsc70-GFP in the nucleus in most cells (80%; data
not shown). As a consequence the Hsc70-GFP level was about
equal in the nucleus and in the cytoplasm (Fig. 4H). Nuclear
Hsc70-GFP fluorescence was strongest in patches around the
nucleolus (which is devoid of Hsc70-GFP) and weaker in other
parts of the nucleus. Formation of Hsc70-GFP positive
speckled spots was sometimes observed (5% of NB41A3 cells;
Fig. 4I). The percentage of such cells expressing Hsc70-GFP
in small speckles in response to 300 M FeCN was time and
dose dependent and increased up to 15% after 6 hours (Fig.
3C).
The response of R6 and R6-Bcl-2 cells to induction of
oxidative stress by either H2O2 or FeCN was very similar. In
both cases increased nuclear localisation of Hsc70-GFP was
observed (Figs 4L, 2B), but it was less pronounced than in
transfected NB41A3 cells. R6-Bcl-2 did not influence the
cellular response to oxidative stress, in particular it did not
inhibit translocation of Hsc70-GFP into the nucleus.
Hsc70-GFP overexpression does not inhibit
apoptosis but protects from cell body shrinkage
After 6 hours exposure to staurosporine or MG132 the majority
of NB41A3 cells had not changed their morphology and
displayed Hsc70-GFP at 1.7 times higher levels in the

Table 2. Hsc70-GFP overexpression partially prevents

formation of shrunk cell bodies with long processes after
exposure to MG132
Shrunk cell bodies with
long processes (%)
Untransfected R6
Hsc70-GFP transfected R6

30
15

Untransfected R6-Bcl-2
Hsc70-GFP transfected R6-Bcl-2

20
10

Percentage of cells with shrunk cell body and long processes after exposure
to 1.5 M MG132 for 24 hours decreased in both R6 and R6-Bcl-2 cells
when Hsc70-GFP was overexpressed. Each data point is an average of three
individual analyses of different populations, each containing at least 50 cells.
s.d. is smaller than 5% in all tested conditions.

cytoplasm than in the nucleus. After 24 hours, however, in

contrast to untransfected cells, this ratio decreases up to 1.3
(Fig. 2B), because of nuclear translocation of Hsc70-GFP in
30% of non-apoptotic NB41A3 cells (Fig. 4M). The percentage
of apoptotic NB41A3 cells was comparable to untransfected
cells, i.e. 2-5% after 6 hours, 20% after 24 hours exposure to
staurosporine or MG132 (Fig. 3B). Apoptotic NB41A3 cells
did not significantly change their morphology and the
expression of Hsc70-GFP was identical to the one in nonapoptotic cells.
In contrast to untransfected R6 cells, Hsc70-GFP transfected
R6 cells exposed to MG132 or staurosporine displayed only
rarely long processes branching off from rounded-up cell
bodies were observed (Table 2), indicating a protective role of
Hsc70-GFP overexpression against these effects in response to
induction of apoptosis. The number of apoptotic R6 cells after
6 hours and after 24 hours exposure was similar to
untransfected R6 cells (Fig. 3B). Localisation and expression
level of Hsc70-GFP were not increased and similar to nonapoptotic R6 cells. In contrast most R6-Bcl-2 cells transfected
with Hsc70-GFP and exposed to staurosporine for 6 hours had
no sign of apoptosis (Fig. 3B) and displayed Hsc70-GFP along
intracellular fibres within the entire cell (Fig. 4N,O). Also 24
hours after induction of apoptosis with staurosporine the cell
viability of Hsc70-GFP transfected R6-Bcl-2 cells was about
80%, not significantly altered compared to untransfected R6Bcl-2 cells, but significantly higher than in Hsc70-GFP
transfected R6 cells (Fig. 3B). 24 hours incubation with either
MG132 or staurosporine yielded a comparable percentage of
nuclear fragmentation, but the number of R6-Bcl-2 cells with
altered shape was much higher with MG132. Most cells were
devoid of processes and the cell body had shrunk and was
rounded up.
Overexpression of other proteins of the neuronal
PMO complex can modulate the subcellular
localisation of Hsc70-GFP
Because Hsp70 is a component of the neuronal PMO complex,
we investigated whether its nuclear translocation is affected by
other components of the complex, e.g. glyceraldehyde-3phosphate dehydrogenase (GAPDH), aldolase C, enolase- and
Ulip-2. These were prepared as blue fluorescent protein (BFP)
fusion proteins into expression vectors and co-transfected with
Hsc70-GFP vectors.
When Hsc70-GFP was transfected together with enolase-

2852 Z. Dastoor and J.-L. Dreyer

Table 3. Influence of co-transfection of components from
the neuronal PMO complex on the subcellular localization
of Hsc70-GFP
Cells accumulating
Hsc70-GFP in speckles*
Protein co-expressed
with Hsc70-GFP
Hsc70-GFP only
Enolase-BFP
GAPDH-BFP
Aldolase-BFP
Ulip-BFP

Nuclear Hsc70-GFP*

Control
%

H2O2
%

Control
%

H2O2
%

0
0
0
0
0

20
50
25
20
20

35
35
35
35
35

50
60
85
50
50

*The percentage of cells accumulating Hsc70-GFP in intracellular speckles

and the percentage of Hsc70-GFP expressed in the nucleus is indicated.
Transient transfection was performed in NB41A3 cells, which were analysed
under unstressed conditions or after 3 hours of oxidative stress with either 100
M H2O2. Each data point is an average of two individual analyses of
different populations, each containing at least 50 cells. s.d. is smaller than 5%
in all tested conditions.

BFP in neuroblastoma cells, no difference in Hsc70-GFP

distribution was observed compared to cells transfected with
Hsc70-GFP alone. However, after oxidative stress with 100 M
H2O2 the percentage of cells forming intracellular Hsc70-GFPspeckles augmented from 35% to 50%. This effect was not
observed with 300 M FeCN. Nuclear translocation of Hsc70GFP was 10% higher under both conditions (Table 3).
Hsc70-GFP localisation is not influenced by GAPDH-BFP
co-overexpression in resting cells. But after exposure to either
100 M H2O2 or 300 M FeCN elevated nuclear Hsc70
localisation was observed, i.e. 40% more in double-transfected
cells (85%) than in cells overexpressing Hsc70-GFP alone
(Table 3). On the other hand, concomitant overexpression of
Hsc70 with GAPDH increased the cytoplasmic level of
GAPDH in both stressed and unstressed cells (unpublished
data). These observations indicate that overexpression of
Hsc70 partially prevents nuclear translocation of GAPDH,
which has been reported to induce apoptosis.
Aldolase-BFP or Ulip-BFP co-transfected with Hsc70-GFP
had no influence on Hsc70-GFP localisation neither before nor
after oxidative stress (Table 3).
These observations show that co-expression of two
components of the neuronal PMO complex (GAPDH and
enolase-) can modulate the expression of Hsc70-GFP. Vice
versa, overexpression of Hsc70 also has an impact on the
subcellular localisation of components of the PMO complex.
DISCUSSION
Translocation of endogenous Hsc70 into the
nucleus is pronounced during oxidative stress but
not during apoptosis
Nuclear translocation of endogenous Hsc70 is only observed
during oxidative stress and much less during apoptosis. In all
conditions Hsc70-GFP behaved like the untagged Hsc70
protein, as detected by immunofluorescence, enabling us to
follow the translocation of Hsc70 by monitoring the
fluorescence of GFP. Apoptotic cells did not display
increased levels of either endogenous Hsc70 or overexpressed
Hsc70-GFP. Under similar experimental conditions, in

contrast, GAPDH shows nuclear translocation under

oxidative stress and during apoptosis and is increased in
apoptotic cells (unpublished data). In control experiments
with enolase--GFP and GFP alone no changes in the
subcellular localisation of these proteins was observed
under the same conditions, clearly indicating that nuclear
translocation after oxidative stress and apoptosis is a specific
phenomenon for Hsc70. It is possible that the nuclear
translocation mechanisms of Hsc70 for reactive oxygen
species (ROS)-induced oxidative stress and for MG132 or
staurosporine-induced apoptosis are different. Hsc70
functions as a molecular chaperone in the correct refolding
of proteins. When oxidative stress is induced, nuclear
proteins become damaged by oxygen radicals and Hsc70 may
probably be imported into the nucleus in order to catalyze
refolding them into a suitable conformation.
The shuttling of Hsc70 between the nucleus and the
cytoplasm observed in this report confirms previous
observations (Lamian et al., 1996). These studies established
that the nuclear import of purified Hsc70 is not inhibited by a
nuclear localisation signal peptide conjugated with BSA. A
basic domain (246KRKHKKDISENKRAVRR262) of Hsc70
acts as a nuclear localisation signal and promotes nuclear
import. However, inactivation of this signal by deletion of the
first six amino acids has no effect on Hsc70 import, so Hsc70
could use a different import signal and enter the nucleus by a
mechanism different from classical nuclear localisation signals
(Lamian et al., 1996). Nuclear translocation of Hsc70, but not
of Hsp70, was observed during early S phase of the cell cycle,
indicating a role for Hsc70 in the process of S phase entry
(Zeise et al., 1998). From our data nuclear translocation of
Hsc70 plays also a role in oxidative stress.
It has been shown that stress factors result in increased hsp70
expression in human monocytes (Mariethoz et al., 1997), but
the subcellular distribution pattern is different depending on
the type of stress. Heat-shock induces a rapid translocation of
hsp70 into the nucleus, but not phorbol ester nor
erythrophagocytosis, consistent with our findings showing a
different type of subcellular localisation after oxidative stress
or apoptosis. This differential expression probably relates to
distinct regulation and functions of Hsc70 in apoptosis or
oxidative stress.
Programmed cell death involves a shut down of protein
expression and protein breakdown. To show that the observed
changes in the ratio between cytosolic and nuclear Hsc70 is not
influenced by this fact we performed controls with enolaseGFP and GFP alone and did not observe changes in such a
ratio, indicating that nuclear translocation is specific for Hsc70.
Overexpression of Hsc70 renders cells more
resistant to oxidative stress
Untransfected cells exposed to oxidative stress displayed often
a shrivelled cell body and long processes, an effect that was
rarely observed in Hsc70-GFP-transfected cells. Probably
overexpression of Hsc70-GFP protects cells from damages
induced by oxidative stress, and the formation of long
processes might result from decreasing attachment of stressed
cells to the growth surface. Despite the fact that Bcl-2
efficiently protects cells from apoptosis, Bcl-2 does not
influence nuclear translocation and speckle formation that arise
in response to oxidative stress.

Nuclear import of HSC70 in stress and apoptosis 2853

After induction of oxidative stress and apoptosis
Hsc70 accumulates in aggregates similar to those
formed by polyglutamine-expanded proteins
We observed that formation of Hsc70-positive intracellular
speckles does not occur in resting cells, but only in stressed
cells and correlates with increasing concentrations of stressing
agents. These speckles are very similar to aggregates formed
by polyglutamine-expanded proteins in pathological cells.
Polyglutamine expansion encoded by CAG repeats is now
recognised to be a major cause of inherited neurodegenerative
disease including Huntingtons disease, spinobulbar muscular
atrophy and spinocerebellar ataxia (Koshy and Zoghbi, 1997;
Perutz, 1999). The toxicity of the disease protein (e.g. ataxin
or androgen receptor) with expanded polyglutamine seems
linked to its tendency to assume an abnormal conformation,
which promotes aggregation. Recent studies have now shown
that pathological cells that express expanded polyglutamine
protein in aggregates elicit a stress response manifested by
marked induction of Hsp70 and Hsc70 and that these two
co-localise with the aggregates (Chai et al., 1999; Stenoien
et al., 1999). A common hypothesis is that polyglutamine
aggregation may be necessary, but not sufficient, to elicit a
stress response. However, our observations indicate that
aggregation may be a secondary phenomenon and not
responsible for the stress response induction. Induction and
accumulation of Hsc70 in aggregates may represent an effort
by the cell to increase refolding, disaggregation or elimination
of misfolded (polyglutamine) proteins. This view is supported
by the fact that overexpression of other heat shock proteins,
Hsp40 chaperones, can suppress aggregation of polyglutamine
protein (Cummings et al., 1998; Chai et al., 1999; Stenoien et
al., 1999).
Hsc70 may also have a role in hypoxia and
apoptosis
Besides Hsc70 expression, the expression of two other major
components of the neuronal PMO complex, GAPDH and
enolase-, is also significantly changed after hypoxia (Graven
and Farber, 1995, 1998; Graven et al., 1994, 1998). Enolase-
is up-regulated in the cytoplasm of hypoxic cells, GAPDH is
up-regulated both in the cytoplasm and in the nucleus and its
glycolytic activity is decreased. Therefore GAPDH, and
perhaps enolase-, exert functions during hypoxia aside from
their catalytic function in glycolysis. This is an interesting
observation since we isolated a multi-enzyme complex
containing GAPDH and enolase- tightly bound to Hsc70
(Bulliard et al., 1997).
It is well established that overexpression of both inducible
(hsp70i) and constitutive (hsp70c) isoforms confers resistance
to oxidative challenges generated by ROS, again showing an
antioxidant role for these proteins (Chong et al., 1998). Lee
also showed that glucose deprivation induces metabolic
oxidative stress via a stress-activated protein kinase,
accompanied by an increase in Hsp70 mRNA which is
suppressed by overexpression of Bcl-2 (Lee and Corry, 1998).
ROS production is common to apoptosis induced by agents
such as e.g. camptothecin and actinomycin D, against which
Hsps conferred protection, but not by Fas-mediated apoptosis,
against which Hsps showed no protective effect (Creagh and
Cotter, 1999). Selective protection against these agents is
mimicked by pre-treatment with antioxidant compounds, and

occurs downstream of ROS production. Consistent with our

results, Creagh found no correlation between Bcl-2- and
Hsp70-mediated protection. Similarly Hsc70 protection
against hypoxia-induced apoptosis has been described in
hypoxia-induced apoptosis-resistant macrophages (HARMs;
Yun et al., 1997; Kim et al., 1997). A significant increase in
TNF production in HARM but a decrease in other
macrophages was observed after hypoxic treatment, indicating
that a selective population of macrophages can adapt to
hypoxic conditions by overcoming the apoptotic signal, in
agreement with our observations.
Together, our findings demonstrate that the differential
translocation of Hsc70 observed after induction of oxidative
stress or apoptosis is in direct correlation with specific actions
of Hsc70.
The authors are indebted to Mrs C. Deforel-Poncet and Mrs A.
Hayoz for skilled technical assistance, to Dr M. Celio and Dr M.
Ibrahim for their invaluable help in the immunocytochemical studies,
to Dr S. Rusconi and Dr S. Brenz-Verca for their help in molecular
biology and to Dr C. Borner for supplying R6-Bcl2 cell lines and for
his many advices. The present work was supported by grant no. 3153206.97 from the Swiss National Foundation.

REFERENCES
Augusti-Tocco, G. and Sato, G. (1969). Establishment of functional clonal
lines of neurons from mouse neuroblastoma. Proc. Nat. Acad. Sci. USA 64,
311-315.
Beckmann, R. P., Mizzen, L. E. and Welch, W. J. (1990). Interaction of Hsp
70 with newly synthesized proteins: implications for protein folding and
assembly. Science 248, 850-854.
Bimston, D., Song, J., Winchester, D., Takayama, S., Reed, J. C. and
Morimoto, R. I. (1998). BAG-1, a negative regulator of Hsp70 chaperone
activity, uncouples nucleotide hydrolysis from substrate release. EMBO J.
17, 6871-6878.
Borner, C. (1996). Diminished cell proliferation associated with the deathprotective activity of Bcl-2. J. Biol. Chem. 271, 12695-12698.
Bulliard, C., Zurbriggen, R., Tornare, J., Faty, M., Dastoor, Z. and Dreyer,
J. L. (1997). Purification of a dichlorophenol-indophenol oxidoreductase
from rat and bovine synaptic membranes: tight complex association of a
glyceraldehyde-3-phosphate dehydrogenase isoform, TOAD64, enolasegamma and aldolase C. Biochem. J. 324, 555-563.
Chai, Y., Koppenhafer, S. L., Bonini, N. M. and Paulson, H. L. (1999).
Analysis of the role of heat shock protein (Hsp) molecular chaperones in
polyglutamine disease. J. Neurosci. 19, 10338-10347.
Chirico, W. J., Waters, M. G. and Blobel, G. (1988). 70K heat shock related
proteins stimulate protein translocation into microsomes. Nature 332, 805810.
Chong, K. Y., Lai, C. C., Lille, S., Chang, C. and Su, C. Y. (1998). Stable
overexpression of the constitutive form of heat shock protein 70 confers
oxidative protection. J. Mol. Cell Cardiol. 30, 599-608.
Creagh, E. M. and Cotter, T. G. (1999). Selective protection by hsp 70
against cytotoxic drug-, but not Fas- induced T-cell apoptosis. Immunology
97, 36-44.
de la Rosa, E. J., Vega-Nunez, E., Morales, A. V., Serna, J., Rubio, E. and
de Pablo, F. (1998). Modulation of the chaperone heat shock cognate 70 by
embryonic (pro)insulin correlates with prevention of apoptosis. Proc. Nat.
Acad. Sci. USA 95, 9950-9955.
Cummings, C. J., Mancini, M. A., Antalffy, B., DeFranco, D. B., Orr, H.
T. and Zoghbi, H. Y. (1998). Chaperone suppression of aggregation and
altered subcellular proteasome localization imply protein misfolding in
SCA1. Nature Genet. 19, 148-154.
Deshaies, R. J., Koch, B. D., Werner-Washburne, M., Craig, E. A. and
Schekman, R. (1988). A subfamily of stress proteins facilitates
translocation of secretory and mitochondrial precursor polypeptides. Nature
332, 800-805.
Ellis, J. (1987). Proteins as molecular chaperones. Nature 328, 378-379.

2854 Z. Dastoor and J.-L. Dreyer

Flynn, G. C., Pohl, J., Flocco, M. T. and Rothman, J. E. (1991). Peptidebinding specificity of the molecular chaperone BiP. Nature 353, 726-730.
Gebhardt, B. R., Ries, J., Caspary, W. F., Boehles, H. and Stein, J. (1999).
Superoxide: a major factor for stress protein induction in reoxygenation
injury in the intestinal cell line Caco-2. Digestion 60, 238-245.
Gething, M. J. and Sambrook, J. (1992). Protein folding in the cell. Nature
355, 33-45.
Graven, K. K., Troxler, R. F., Kornfeld, H., Panchenko, M. V. and
Farber, H. W. (1994). Regulation of endothelial cell glyceraldehyde-3phosphate dehydrogenase expression by hypoxia. J. Biol. Chem. 269,
24446-24453.
Graven, K. K. and Farber, H. W. (1995). Hypoxia-associated proteins. New
Horiz. 3, 208-218.
Graven, K. K. and Farber, H. W. (1998). Endothelial cell hypoxic stress
proteins. J. Lab. Clin. Med. 132, 456-463.
Graven, K. K., McDonald, R. J. and Farber, H. W. (1998). Hypoxic
regulation of endothelial glyceraldehyde-3-phosphate dehydrogenase. Am.
J. Physiol. 274, C347-355.
Haas, I. G. and Wabl, M. (1983). Immunoglobulin heavy chain binding
protein. Nature 306, 387-389.
Hartl, F. U. (1996). Molecular chaperones in cellular protein folding. Nature
381, 571-579.
Hendrick, J. P. and Hartl, F. U. (1993). Molecular chaperone functions of
heat-shock proteins. Annu. Rev. Biochem. 62, 349-384.
Hightower, L. E. (1980). Cultured animal cells exposed to amino acid
analogues or puromycin rapidly synthesize several polypeptides. J. Cell
Physiol. 102, 407-427.
Hohfeld, J. (1998). Regulation of the heat shock conjugate Hsc70 in the
mammalian cell: the characterization of the anti-apoptotic protein BAG-1
provides novel insights. Biol. Chem. 379, 269-274.
Hohfeld, J. and Jentsch, S. (1997). GrpE-like regulation of the hsc70
chaperone by the anti-apoptotic protein BAG-1. EMBO J. 16, 6209-6216.
Kim, Y. M., de Vera, M. E., Watkins, S. C. and Billiar, T. R. (1997). Nitric
oxide protects cultured rat hepatocytes from tumor necrosis factor-alphainduced apoptosis by inducing heat shock protein 70 expression. J. Biol.
Chem. 272, 1402-1411.
Koshy, B. T. and Zoghbi, H. Y. (1997). The CAG/polyglutamine tract
diseases: gene products and molecular pathogenesis. Brain Pathol. 7, 927942.
Lamian, V., Small, G. M. and Feldherr, C. M. (1996). Evidence for the
existence of a novel mechanism for the nuclear import of Hsc70. Exp. Cell
Res. 228, 84-91.
Lee, Y. J. and Corry, P. M. (1998). Metabolic oxidative stress-induced HSP70
gene expression is mediated through SAPK pathway. Role of Bcl-2 and cJun NH2-terminal kinase. J. Biol. Chem. 273, 29857-29863.
Mariethoz, E., Jacquier-Sarlin, M. R., Multhoff, G., Healy, A. M.,
Tacchini-Cottier, F. and Polla, B. S. (1997). Heat shock and
proinflammatory stressors induce differential localization of heat shock
proteins in human monocytes. Inflammation 21, 629-642.

Pelham, H. R. (1986). Speculations on the functions of the major heat shock

and glucose-regulated proteins. Cell 46, 959-961.
Perutz, M. F. (1999). Glutamine repeats and neurodegenerative diseases:
molecular aspects. Trends Biochem. Sci. 24, 58-63.
Rothman, J. E. (1989). Polypeptide chain binding proteins: catalysts of
protein folding and related processes in cells. Cell 59, 591-601.
Stenoien, D. L., Cummings, C. J., Adams, H. P., Mancini, M. G., Patel, K.,
DeMartino, G. N., Marcelli, M., Weigel, N. L. and Mancini, M. A.
(1999). Polyglutamine-expanded androgen receptors form aggregates that
sequester heat shock proteins, proteasome components and SRC-1, and are
suppressed by the HDJ-2 chaperone. Hum. Mol. Genet. 8, 731-741.
Stuart, J. K., Myszka, D. G., Joss, L., Mitchell, R. S., McDonald, S. M.,
Xie, Z., Takayama, S., Reed, J. C. and Ely, K. R. (1998). Characterization
of interactions between the anti-apoptotic protein BAG- 1 and Hsc70
molecular chaperones. J. Biol. Chem. 273, 22506-22514.
Takayama, S., Bimston, D. N., Matsuzawa, S., Freeman, B. C., AimeSempe, C., Xie, Z., Morimoto, R. I. and Reed, J. C. (1997). BAG-1
modulates the chaperone activity of Hsp70/Hsc70. EMBO J. 16, 48874896.
Takayama, S., Krajewski, S., Krajewska, M., Kitada, S., Zapata, J. M.,
Kochel, K., Knee, D., Scudiero, D., Tudor, G., Miller, G. J. et al. (1998).
Expression and location of Hsp70/Hsc-binding anti-apoptotic protein BAG1 and its variants in normal tissues and tumor cell lines. Cancer Res. 58,
3116-3131.
Takayama, S., Xie, Z. and Reed, J. C. (1999). An evolutionarily conserved
family of Hsp70/Hsc70 molecular chaperone regulators. J. Biol. Chem. 274,
781-786.
Thulasiraman, V., Xu, Z., Uma, S., Gu, Y., Chen, J. J. and Matts, R. L.
(1998). Evidence that Hsc70 negatively modulates the activation of the
heme-regulated eIF-2alpha kinase in rabbit reticulocyte lysate. Eur. J.
Biochem. 255, 552-562.
Tomassini, J., Roychoudhury, R., Wu, R. and Roberts, R. J. (1978).
Recognition sequence of restriction endonuclease KpnI from Klebsiella
pneumoniae. Nucl. Acids Res. 5, 4055-4064.
Turner, C. P., Panter, S. S. and Sharp, F. R. (1999). Anti-oxidants prevent
focal rat brain injury as assessed by induction of heat shock proteins
(HSP70, HO-1/HSP32, HSP47) following subarachnoid injections of lysed
blood. Brain Res. Mol. Brain Res. 65, 87-102.
Yun, J. K., McCormick, T. S., Villabona, C., Judware, R. R., Espinosa, M.
B. and Lapetina, E. G. (1997). Inflammatory mediators are perpetuated in
macrophages resistant to apoptosis induced by hypoxia. Proc. Nat. Acad.
Sci. USA 94, 13903-13908.
Zeise, E., Kuhl, N., Kunz, J. and Rensing, L. (1998). Nuclear translocation
of stress protein Hsc70 during S phase in rat C6 glioma cells. Cell Stress
Chaperones 3, 94-99.
Zimmermann, R., Sagstetter, M., Lewis, M. J. and Pelham, H. R. (1988).
Seventy-kilodalton heat shock proteins and an additional component from
reticulocyte lysate stimulate import of M13 procoat protein into
microsomes. EMBO J. 7, 2875-2880.