MOLECULAR
AND
CELLULAR
BIOLOGY,
OCt.
1994,
p.
6552-6560
0270-7306/94/$04.00+0
Copyright
©
1994,
American
Society
for
Microbiology
Interaction
between
Heat
Shock
Factor
and
hsp7O
Is
Insufficient
To
Suppress
Induction
of
DNA-Binding
Activity
In
Vivo
SRIDHAR
K.
RABINDRAN,1t
JAN
WISNIEWSKI,1
LIGENG
LI,2
GLORIA
C.
LI,2
AND
CARL
WU`*
Laboratory
of
Biochemistry,
National
Cancer
Institute,
Bethesda,
Maryland
20892,1
and
Memorial
Sloan
Kettering
Cancer
Center,
New
York
New
York
100212
Received
27
April
1994/Returned
for
modification
3
June
1994/Accepted
29
June
1994
The
intracellular
level
of
free
heat
shock
proteins,
in
particular
the
70-kDa
stress
protein
family,
has
been
suggested
to
be
the
basis
of
an
autoregulatory
mechanism
by
which
the
cell
measures
the
level
of
thermal
stress
and
regulates
the
synthesis
of
heat
shock
proteins.
It
has
been
proposed
that
the
DNA-binding
and
oligomeric
state
of
the
heat
shock
transcription
factor
(HSF)
is
a
principal
step
in
the
induction
pathway
that
is
responsive
to
the
level
of
70-kDa
stress
protein.
To
test
this
hypothesis,
we
investigated
the
association
between
HSF
and
70-kDa
stress
protein
by
means
of
a
coimmunoprecipitation
assay.
We
found
that
70-kDa
stress
proteins
associate
to
similar
extents
with
both
latent
and
active
fonns
of
HSF,
although
unlike
other
70-kDa
stress
protein
substrates,
the
association
with
HSF
was
not
significantly
disrupted
in
the
presence
of
ATP.
Gel
mobility
shift
assays
indicated
that
active
HSF
trimers
purified
from
a
bacterial
expression
system
could
not
be
substantially
deactivated
in
vitro
with
purified
70-kDa
stress
protein
and
ATP.
In
addition,
elevated
concentrations
of
hsp7O
alone
could
not
significantly
inhibit
induction
of
the
DNA-binding
activity
of
endogenous
HSF
in
cultured
rat
cells,
and
the
induction
was
also
not
inhibited
in
cultured
rat
cells
or
Drosophila
cells
containing
elevated
levels
of
all
members
of
the
heat
shock
protein
family.
However,
the
deactivation
of
HSF
to
the
non-DNA-binding
state
after
prolonged
heat
stress
or
during
recovery
could
be
accelerated
by
increased
levels
of
heat
shock
proteins.
Hence,
the
level
of
heat
shock
proteins
may
affect
the
rate
of
disassembly
of
HSF
trimers,
but
another
mechanism,
as
yet
undefined,
appears
to
control
the
onset
of
the
oligomeric
transitions.
Organisms
respond
to
elevated
temperatures
and
to
a
variety
of
chemical
inducers
by
rapidly
inducing
the
synthesis
of
the
heat
shock
proteins
(26,
27, 33,
35).
The
regulation
of
this
response
in
eukaryotes
is
mediated
at
the
transcriptional
level
by
a
preexisting
transcriptional
activator
heat
shock
factor
(HSF),
which
binds
to
regulatory
heat
shock
elements
(HSEs)
present
upstream
of
all
heat
shock
genes
(28,
29,
48).
The
HSE
is
composed
of
contiguous,
alternating
repeats
of
the
5-bp
sequence
NGAAN;
three
NGAAN
repeats
are
required
for
high-affinity
interaction
with
HSF
(3,
13,
37,
38,
59).
HSF
responds
to
the
heat
shock
signal
by
the
induction
of
DNA-
binding
activity
and
transcriptional
competence
rather
than
by
increasing
its
own
synthesis
(22,
28,
48,
57,
58,
60).
For
multicellular
eukaryotes
such
as
Drosophila
melano-
gaster,
vertebrates,
and
plants,
the
first
stage
of
HSF
activation
involves
the
acquisition
of
high-affinity
binding
to
HSEs.
This
stress-inducible
binding
of
HSF
to
DNA
is
controlled
by
a
monomer-to-trimer
transition
of
HSF
protein
(4,
44,
55,
56).
An
isolated,
monomeric
HSF
DNA-binding
domain
is
capable
of
binding
specifically
to
a
single
NGAAN
box,
but
the
affinity
of
this
interaction
is
relatively
low
(Kd
=
10-7
to
10-8
M)
(21).
Trimerization
increases
the
affinity
for
the
HSE
by
several
orders
of
magnitude
through
the
assembly
of
three
DNA-
binding
domains
of
HSF
in
one
complex,
thereby
allowing
the
potential
for
concurrent
interactions
with
all
three
NGAAN
boxes
of
the
HSE.
The
binding
of
HSF
trimers
to
the
HSE
is
necessary
but
apparently
not
sufficient
for
transactivation
of
heat
shock
*
Corresponding
author.
Mailing
address:
Laboratory
of
Biochem-
istry,
National
Cancer
Institute,
National
Institutes
of
Health,
Building
37,
Room
4C-09,
Bethesda,
MD
20892.
t
Present
address:
American
Cyanamid
Co.,
Medical
Research
Division,
Pearl
River,
NY
10965.
promoters.
Under
certain
conditions,
binding
of
mammalian
HSF
to
DNA
can
be
uncoupled
from
the
acquisition
of
transcriptional
activity
(17,
20).
It
is
likely
that
metazoan
HSFs
undergo
a
secondary
activation
process
in
addition
to
the
monomer-trimer
transition
to
achieve
full
competence
in
transactivation.
Indeed,
regulation
at
the
level
of
DNA
binding
is
bypassed
in
the
yeasts
Saccharomyces
cerevisiae
and
Kluyveromyces
lactis,
which
possess
constitutively
trimeric
HSF
proteins
bound
to
their
cognate
sites
in
vivo
(16,
18,
49,
50).
For
S.
cerevisiae,
it
is
the
transactivation
function
of
bound
HSF
trimers
which
is
induced
upon
heat
stress,
and
this
activity
is
correlated
with
increased
phosphorylation
at
multiple
serine
and
threonine
residues
(36,
51).
The
transactivation
domain
of
S.
cerevisiae
HSF
and
K
lactis
HSF
is
masked
by
an
internal
region
of
the
protein
that
includes
a
conserved
heptapeptide
sequence;
mutations
in
this
sequence
lead
to
constitutive
activity
(6,
19,
36,
47).
In
addition,
the
transcriptional
activity
is
directly
or
indirectly
repressed
by
the
molecular
chaperone
hsp70,
as
mutations
in
two
yeast
hsp70
genes
cause
constitutive
expression
of
an
HSE-lacZ
construct
(7).
The
regulation
of
metazoan
HSF
binding
to
DNA
is
also
under
negative
control.
The
constitutive
assembly
of
active
HSF
trimers
in
Escherichia
coli
at
non-heat
shock
tempera-
tures,
contrasted
with
the
assembly
of
a
latent,
inducible
HSF
when
the
protein
is
expressed
in
frog
oocytes,
tissue
culture
cells,
or
reticulocyte
lysates,
suggests
that
factors
present
in
eukaryotic
cells
are
required
for
suppressing
the
assembly
of
HSF
trimers
under
normal
conditions
(4,
9,
40,
41,
45).
The
role
of
cellular
factors
in
dictating
the
temperature
response
of
HSF
is
underscored
by
the
behavior
of
human
HSF1
when
the
protein
is
expressed
in
Drosophila
cells,
tobacco
protoplasts,
and
frog
oocytes.
In
such
heterologous
cell
environments,
human
HSF1
is
activated
at
the
host
heat
shock
temperature,
about
10°C
below
its
induction
temperature
in
human
cells
(4,
6552
Vol.
14,
No.
10
EFFECT
OF
HSF-hsp70
INTERACrION
ON
DNA-BINDING
ACTIVITY
6553
8,
53).
These
cellular
factors
may
assist
in
constraining
HSF
in
a
latent,
monomeric
state
whose
stability
is
dependent
on
the
integrity
of
a
leucine
zipper
motif
located
in
the
C-terminal
end
of
the
protein
(41).
While
the
mechanism
of
suppression
of
HSF
trimerization
is
poorly
understood
in
vivo,
recent
studies
have
shown
an
association
between
HSF
and
the
hsp7O
protein
in
vitro
(1,
5,
34,
46).
This
association
has
led
to
a
model
for
the
regulation
of
the
DNA-binding
activity
of
HSF
that
is
principally
related
to
heat
shock-induced
changes
in
the
level
of
hsp7O
proteins
in
the
cell.
In
this
report,
we
have
investigated
the
association
of
HSF
with heat
shock
proteins
by
means
of
coimmunoprecipitation
experiments.
We
find
that
hsp7O
can
be
associated
with
HSF,
although
the
extent
of
the
interaction
is
the
same
for
both
the
heat-shocked
and
the
unshocked
forms
of
HSF.
We
also
find
that
neither
transfected
cells
expressing
increased
levels
of
hsp70
nor
cells
previously
induced
to
express
the
entire
complement
of
heat
shock
proteins
show
significant
effects
on
the
induction
of
HSF
binding
to
DNA.
These
results
do
not
support
a
simple
model
for
the
DNA-binding
activity
of
HSF
solely
controlled
by
changes
in
the
levels
of
hsp70
or
other
heat
shock
proteins.
MATERIALS
AND
METHODS
Antibodies
and
proteins.
Murine
monoclonal
antibodies
C92
(specific
for
mammalian
hsp7o),
N27
(for
mammalian
hsc70
and
hsp70;
also
called
p73),
and
AC88
(for
mammalian
hsp9o)
were
purchased
from
Stressgen
Biotechnology
(Victo-
ria,
British
Columbia,
Canada);
antibody
3a3,
cross-reacting
with
Drosophila
hsc70
(see
Results),
was
obtained
from
Affinity
Bioreagents.
Rat
monoclonal
antibody
7FB
specific
for
Dro-
sophila
hsp7O
and
a
rabbit
polyclonal
antibody
against
mouse
HSF1
were
generous
gifts
of
Susan
Lindquist
and
Richard
Morimoto,
respectively.
Rabbit
polyclonal
antibodies
943
against
Drosophila
HSF
and
180
against
human
HSF1
were
described
previously
(41,
55).
Murine
monoclonal
antibody
3B3
against
human
HSF1
was
obtained
from
a
positive
hybrid-
oma
cell
line
(activity
was
assayed
by
electrophoretic
supershift
of
HSF1-HSE
complexes)
derived
from
a
mouse
injected
with
bacterially
expressed
human
HSF1
purified
as
described
pre-
viously
(40).
Purified
bovine
brain
uncoating
ATPase
(hsc70)
was
a
generous
gift
of
Lois
Greene
and
Evan
Eisenberg
(National
Institutes
of
Health).
Secondary
antibodies
conju-
gated
with
horseradish
peroxidase
(HRP)
(donkey
anti-rabbit
immunoglobulin
G
[IgG]-HRP,
sheep
anti-mouse
IgG-HRP,
and
goat
anti-rat
IgG-HRP)
were
obtained
from
Amersham
and
Chemicon,
respectively,
and
goat
anti-mouse
IgG
conju-
gated
with
alkaline
phosphatase
(AP)
was
purchased
from
Bethesda
Research
Laboratories.
The
enhanced
chemilumi-
nescence
(ECL)
kit
and
AP
detection
kit
II
were
from
Amer-
sham
and
Vector
Laboratories
(Burlingame,
Calif.),
respec-
tively.
Cell
culture
and
heat
shock
HeLa
cells
were
grown
in
Dulbecco's
modified
Eagle
medium
supplemented
with
10%
heat-inactivated
fetal
bovine
serum
(FBS;
HyClone)
in
humid-
ified
incubators
in
the
presence
of
5%
CO2.
Ratl
and
rat
M21
cells
were
cultured
as
described
previously
(24).
Drosophila
Schneider
line
2
(SL2)
cells
were
cultured
in
M3
medium
containing
10%
heat-inactivated
FBS
and
25
,ug
of
gentamicin
sulfate
per
ml
in
tightly
closed
flasks
at
room
temperature.
Cycloheximide
was
included
in
media
when
specified
at
18
,uM
for
rat
cells
and
118
,uM
for
Drosophila
cells.
For
heat
shock,
flasks
were
submerged
for
the
indicated
periods
in
a
water
bath.
After
heat
shock
(and
recovery
at
the
normal
culture
temperature,
where
indicated),
cells
were
washed
twice
with
ice-cold
phosphate-buffered
saline
(PBS)
and
harvested
by
scraping
and
centrifugation
at
12,000
rpm
at
4°C
for
30
s.
Cell
pellets
were
immediately
frozen
on
dry
ice.
Gel
mobility
shift
assay.
Whole
cell
and
nuclear
extracts
were
prepared
as
previously
described
(41).
Gel
mobility
shift
analysis
was
carried
out
as
described
previously
(41),
using
a
32P-labeled
HSE
oligonucleotide.
For
Drosophila
and
human
cell
extracts,
reaction
mixtures
were
incubated
for
10
min
on
ice
and
at
room
temperature,
respectively,
before
being
loaded
onto
a
0.8%
agarose-0.5
x
Tris-borate-EDTA
gel;
for
rat
cell
extracts,
the
mixture
was
incubated
for
10
min
at
37°C
and
10
min
at
room
temperature.
Where
indicated,
extracts
were
incubated
with
antibody
180
against
human
HSF1
(1:60
dilu-
tion
in
PBS)
for
10
min
at
room
temperature
prior
to
the
addition
of
other
components.
For
competition
experiments,
a
100-fold
excess
of
unlabeled
HSE
oligonucleotide
was
included
into
the
reaction
mixture.
For
some
experiments,
whole
cell
extracts
were
supplemented
with
purified
hsc70,
ATP,
15
mM
creatine
phosphate,
and
50
mg
of
creatine
kinase
per
ml,
as
indicated.
In
vivo
labeling
of
HeLa
cells.
Cells
were
seeded
into
25-cm2
tissue
culture
flasks
at
106
cells
per
flask.
After
24
h,
cells
were
washed
with
methionine-deficient
Dulbecco's
modified
Eagle
medium
(Quality
Biologicals,
Gaithersburg,
Md.)
and
labeled
for
8
h
in
1
ml
of
the
same
medium
supplemented
with
10%
FBS
(dialyzed
against
0.15
M
NaCl-0.015
M
sodium
citrate
[pH
7])
and
100
jiCi
of
[35S]methionine
(DuPont/New
En-
gland
Nuclear).
Cells
were
washed
and
harvested
as
described
below.
For
phosphate
labeling,
cells
were
washed
in
phos-
phate-free
medium
(Quality
Biologicals)
and
pulse-labeled
in
1
ml
of
medium
supplemented
with
dialyzed
FBS
and
1
mCi
of
32p;
(DuPont/New
England
Nuclear)
for
15
min
at
37
or
44°C.
Cells
were
washed
in
PBS
and
lysed
directly
in
solubilization
buffer
(see
below)
containing
0.1%
sodium
dodecyl
sulfate
(SDS).
Immunoprecipitation
of
HSF1-hsp7O
complexes.
Immuno-
precipitations
were
carried
out
essentially
as
described
else-
where
(14).
Briefly,
extracts
were
prepared
from
cells
lysed
directly
in
solubilization
buffer
(1%
Triton
X-100,
0.5%
so-
dium
deoxycholate,
5
mM
EDTA,
25
mM
Tris-HCl
[pH
7.5])
supplemented
with
0.1%
SDS.
Lysates
were
centrifuged
at
4°C
for
15
min
at
12,000
x
g
before
incubation
with
rabbit
antiserum
against
human
HSF1
or
preimmune
serum
(1
to
3
,ul)
for
20
min
at
room
temperature.
Fixed
Staphylococcus
aureus
cells
(Pansorbin;
Calbiochem)
were
then
added,
and
the
incubation
was
continued
for
another
5
min.
To
reduce
background,
extract
samples
were
cleared
with
Pansorbin
prior
to
addition
of
antiserum,
and
the
Pansorbin
used
for
immuno-
precipitation
was
preincubated
with
unlabeled
cell
extracts.
Immune
complexes
were
collected
by
centrifugation
through
a
sucrose
cushion,
washed
three
times
with
solubilization
buffer-
SDS,
and
washed
once
in
10
mM
Tris-HCl
(pH
7.5)-5
mM
EDTA.
Pellets
were
boiled
in
SDS
sample
buffer,
and
the
solubilized
material
was
analyzed
by
SDS-polyacrylamide
gel
electrophoresis
(PAGE)
and
fluorography
or
by
Western
blotting
(immunoblotting).
For
analysis
of
the
effects
of
ATP
(Fig.
2),
cells
were
disrupted
in
whole
cell
extraction
buffer
(see
above)
supple-
mented
with
0.1%
Nonidet
P-40,
5
mM
MgCl2,
and
2.5
mM
ATP
or
10
U
of
apyrase
(Sigma)
per
ml.
Extracts
were
incubated
at
room
temperature
for
1
h
and
then
prepared
for
immunoprecipitation
by
the
addition
of
Triton
X-100
to
1%,
sodium
deoxycholate
to
0.5%
and
SDS
to
0.1%.
Only
one
cycle
of
immunoprecipitation
was
carried
out
for
[35S]methionine-
labeled
samples.
For
32P-labeled
samples,
bound
proteins
were
VOL.
14,
1994
6554
RABINDRAN
ET
AL.
anti-
pre-
HSF
1
immune
temp
(CC)
v
co
et
3B3
_
(anti-HSF1)
-
-(
N27
(anti-hsc70/hsp7O)
C92
(anti-hsp7O)
AC88
(anti-hsp9O)
B
anti-HSF1
pre-immune
nhs
hs
nhs
hs
94
57
-
94
-
67
-94
-67
;
94
-67
anti-
HSF
1
temp
(°C)
37
44
pre-
immunE
37
44
hHSF1
I
70K
D
exogenous
e
hsc7O:
none
C
HS
anti-HSF1:
+
-
+
-
94K
hHSF1
67K
hsc7o/hsp7o
1
Oqppg/ml
C
HS
N+
-
-
43K
-
30K
FIG.
1.
Association
of
human
HSF1
with
heat
shock
proteins
in
normal
and
heat-shocked
HeLa
cells.
(A)
Extracts
from
control
cells
(37°C)
and
cells
heat
shocked
at
44°C
for
20
min
were
immunoprecipitated
with
rabbit
polyclonal
anti-human
HSF1
antibody
180
or
with
a
preimmune
rabbit
serum
under
stringent
conditions
(see
Materials
and
Methods).
Immunoglobulin
complexes
were
collected
on
fixed
S.
aureus
cells
and
analyzed
by
SDS-PAGE
(10%
gel)
and
Western
blotting.
The
blots
were
probed
with
mouse
monoclonal
antibodies
against
human
HSF1
(3B3),
hsc70
and
hsp70
(N27),
hsp70
alone
(C92),
and
hsp90
(AC88).
The
secondary
antibody
was
goat
anti-mouse
IgG-AP.
In
panels
A
to
C,
positions
of
size
markers
are
shown
in
kilodaltons
on
the
right.
(B)
Effect
of
heat
shock
on
the
phosphorylation
state
of
human
HSF1.
HeLa
cells
were
pulse-labeled
with
32P;
at
37
or
44°C
for
15
min.
Cell
extracts
were
immunoprecipitated
with
anti-HSF1
or
preimmune
serum
as
described
above
and
analyzed
by
SDS-PAGE
and
autoradiography.
nhs,
non-heat-shocked;
hs,
heat
shocked.
(C)
Radiolabeled
cellular
proteins
associating
with
human
HSF1
(hHSF1).
HeLa
cells
were
labeled
with
[35S]methionine
for
8
h.
Extracts
from
control
cells
and
cells
heat
shocked
for
20
min
were
immunoprecipitated
with
anti-human
HSF1
serum
as
described
above;
precipitated
proteins
were
analyzed
by
SDS-PAGE
(10%
gel)
and
fluorography.
(D)
Competition
with
unlabeled
hsc70
proteins.
Control
(C)
and
heat-shocked
(HS;
44°C
for
20
min)
HeLa
cells
were
labeled
with
[35S]methionine
for
10
h.
Cells
were
lysed
in
the
presence
or
absence
of
100
,ug
of
purified
bovine
uncoating
ATPase
(hsc70)
per
ml,
immunoprecipitated
with
anti-HSF1
(+)
or
with
preimmune
serum
(-),
and
analyzed
by
SDS-PAGE
(10%
polyacrylamide
gel)
and
autoradiography.
The
positions
of
human
HSF1
and
70-kDa
stress
proteins
(hsc7o/hsp7o)
are
indicated.
Assuming
uniform
labeling
and
13
and
9
methionine
residues
for
human
hsp70
and
hsc70,
respectively,
the
stoichiometry
of
the
interaction
would
be
approximately
0.2:1
(70-kDa
stress
protein/HSF1).
released
with
2%
SDS
at
37°C,
diluted,
and
immunoprecipi-
tated
again
as
described
above.
Western
blotting.
Proteins
fractionated
by
SDS-PAGE
were
electroblotted
onto
nitrocellulose;
the
filter
was
blocked
with
3%
bovine
serum
albumin
(BSA)
(for
AP
detection)
or
5%
powdered
milk
(for
ECL
detection)
in
PBS-0.1%
Tween
20.
Primary
antibodies
were
diluted
1:1,000
in
PBS-0.1%
Tween
20
containing
0.5%
BSA
(for
AP
detection)
or
1:10,000
to
1:20,000
(for
ECL),
and
filters
were
incubated
for
1
h
at
room
temperature
or
overnight
at
4°C.
After
being
washed
with
PBS-0.1%
Tween
20,
filters
were
incubated
with
secondary
antibody
diluted
1:5,000
to
1:10,000
(for
AP
detection)
or
1:20,000
to
1:40,000
(for
ECL)
in
PBS-0.1%
Tlween
20
con-
taining
0.5%
BSA.
Some
blots
were
processed
with
a
mixture
of
two
primary
antibodies
and
then
with
species-specific
anti-
IgG
antibody-AP
and
antibody-HRP
conjugates.
In
such
cases,
ECL
detection
was
carried
out
first,
and
filters
were
washed
with
PBS-0.1%
Tween
20
and
then
developed
for
AP.
RESULTS
Association
of
HSF1
with
70-kDa
stress
proteins
under
normal
and
heat-shocked
conditions.
We
analyzed
the
associ-
ation
of
the
70-kDa
stress
proteins
with
human
HSF1,
the
major
heat
shock-inducible
HSF
species
in
HeLa
cells
(40),
using
antiserum
specific
for
human
HSF1
in
a
coimmunopre-
cipitation
assay.
Proteins
associating
with
HSF1
were
analyzed
by
Western
blotting
(Fig.
1A).
Parallel
blots
were
probed
with
monoclonal
antibodies
specific
for
human
HSF1
(3B3),
for
both
the
constitutively
expressed
hsc70
and
the
heat-inducible
hsp70
(N27),
for
only
the
inducible
hsp70
(C92),
and
for
hsp90
(AC88).
As
previously
reported
(23),
human
HSF1
migrates
as
A
-
94k
-
67k
-
43k
-
30k
MOL.
CELL.
BIOL.
-7-d
*
_v
EFFECT
OF
HSF-hsp7O
INTERACTION
ON
DNA-BINDING
ACTIVITY
6555
anti-HSF
1
C')'
E
CF)
co)
o
c
a
C'
)t
t
hHSF
1
hsc7O/
'
hsp7O
hHSF1
_
hsC70/
a
t-
hsp70
o,
pre-immune
Ec
0
m
N
OJC'
N
cm
CM
co
t
le
-
94K
;
-
67K
-X
+
-
94K
-;
-
67K
la
FIG.
2.
Effect
of
ATP
and
prolonged
heat
shock
on
the
interaction
between
human
HSF1
and
70-kDa
stress
proteins.
Whole
cell
extracts
were
prepared
from
control
HeLa
cells
and
from
cells
heat
shocked
at
42°C
for
30
min
or
3
h.
Extracts
were
incubated
for
1
h
in
the
absence
or
presence
of
2.5
mM
ATP.
Apyrase
(10
U/ml)
was
used
to
deplete
the
endogenous
ATP
in
samples
not
supplemented
with
ATP.
The
immunoprecipitations
were
carried
out
as
described
in
the
legend
to
Fig.
1,
and
Western
blots
of
the
precipitated
proteins
were
probed
with
a
mixture
of
mouse
monoclonal
antibodies
against
human
HSF1
(3B3)
and
hsc7o/hsp70
(N27).
The
bands
representing
human
HSF1
(hHSF1)
and
70-kDa
stress
proteins
(hsc7O/hsp7O)
are
marked
on
the
left;
positions
of
molecular
weight
markers
are
indicated
in
kilodaltons
on
the
right.
a
tight
cluster
of
bands
whose
mobility
after
heat
shock
is
retarded
(Fig.
1A;
this
sample
shows
some
proteolysis
of
HSF1).
This
change
in
the
electrophoretic
mobility
of
HSF
after
heat
stress
is
associated
with
an
increase
in
phosphoryla-
tion
of
the
protein
(Fig.
1B).
We
observed
that
both
hsc70
and
hsp7O
from
HeLa
cells
coimmunoprecipitated
with
human
HSF1,
as
revealed
by
closely
spaced
doublet
bands
that
react
with
the
N27
antibody
(Fig.
1A).
hsp7O,
the
lower
band
of
the
doublet,
was
specifically
revealed
with
the
C92
antibody;
this
stress-inducible
70-kDa
isoform
is
expressed
at
a
significant
level
in
primate
cells
independent
of
heat
stress
(32,
54).
Precipitation
of
hsp7O
or
hsc70
was
not
observed
when
preim-
mune
serum
was
used,
and
another
heat
shock
protein,
hsp9O,
was
not
observed
to
coimmunoprecipitate
with
HSF1.
The
data
indicate
that
both
the
constitutive
and
inducible
forms
of
the
70-kDa
stress
protein
associate
specifically
with
HSF1
in
normal
and
heat-treated
HeLa
cells.
Consistent
with
these
findings,
a
32S-labeled
70-kDa
species
was
also
found
to
coimmunoprecipitate
with
HSF1
when
unshocked
and
heat-
shocked
HeLa
cells
labeled
with
[35S]methionine
were
ana-
lyzed
(Fig.
1C).
This
labeled
species
was
further
identified
as
the
70-kDa
stress
protein
on
the
basis
of
its
electrophoretic
migration
on
a
two-dimensional
gel.
The
association
between
70-kDa
stress
protein
and
HSF1
could
not
be
competed
for
with
unlabeled,
exogenous
hsc70
protein
(Fig.
1D).
Effect
of
ATP
treatment
or
prolonged
heat
shock
on
the
association
of
HSF1
and
hsp7O.
To
explore
the
functional
significance
of
the
physical
interaction
between
HSF1
and
hsp70,
we
investigated
the
effects
of
ATP
on
the
coimmuno-
precipitation
of
the
two
proteins.
HeLa
cell
extracts
were
incubated
with
2.5
mM
ATP
prior
to
immunoprecipitation
and
analysis
by
Western
blotting.
As
shown
in
Fig.
2,
the
amount
of
hsp70
associated
with
HSF1
was
not
substantially
decreased
when
extracts
were
treated
with
ATP.
Similar
results
were
observed
with
the
nonhydrolyzable
ATP
analog
AMP-PCP
[adenylyl(3,y-methylene)-diphosphonate]
(data
not
shown).
FIG.
3.
Effect
of
70-kDa
stress
protein
on
HSF1
activity
in
vitro.
Purified,
bacterially
expressed
human
HSF1
was
incubated
for
1
h
at
room
temperature
in
the
absence
or
presence
of
1
mM
bovine
hsc70
(bovine
uncoating
ATPase)
and
5
mM
ATP,
along
with
an
ATP-
regenerating
system
(15
mM
creatine
phosphate,
50
mg
of
creatine
kinase
per
ml),
as
indicated.
Following
addition
of
the
labeled
HSE
for
10
min,
samples
were
analyzed
on
a
0.8%
agarose
gel
in
0.5
x
Tris-borate-EDTA
and
autoradiographed.
The
positions
of
the
HSF-
HSE
complex
(B)
and
free
HSE
(F)
are
indicated.
The
bulk
of
the
free
HSE
migrated
off
the
gel.
We
also
analyzed
the
possibility
of
an
increased
association
between
HSF1
and
hsp70
during
prolonged
heat
stress,
which
brings
about
an
attenuation
of
the
DNA-binding
activity
of
HSF
(2,
42).
When
the
interaction
between
HSF1
and
hsp7O
was
analyzed
by
immunoprecipitation
after
a
continuous
heat
shock
of
HeLa
cells
for
3
h
at
42°C,
no
significant
change
in
the
level
of
association
was
observed
(Fig.
2).
This
interaction
was
also
not
significantly
disrupted
upon
incubation
with
ATP.
Hence,
the
attenuation
of
DNA-binding
activity
of
HSF1
after
prolonged
heat
shock
is
apparently
not
correlated
with
a
measurable
increase
in
the
interaction
with
hsp70
under
the
conditions
of
the
immunoprecipitation
assay.
Interaction
between
HSF1
and
70-kDa
stress
proteins
in
vitro
does
not
affect
DNA-binding
activity.
To
assess
the
role
of
70-kDa
stress
proteins
on
the
DNA-binding
activity
of
HSF1
in
vitro,
purified
bovine
hsc70
active
as
uncoating
ATPase
(15)
was
incubated
with
purified,
bacterially
expressed
HSF1.
As
shown
by
a
gel
mobility
shift
assay,
the
addition
of
hsc7O
caused
a
retardation
in
the
mobility
of
the
HSF1-DNA
com-
plex,
suggesting
an
association
between
hsc70
and
the
HSF1-
DNA
complex
in
vitro.
However,
this
interaction
did
not
lead
to
a
change
in
the
amount
of
HSF1
bound
to
DNA
(Fig.
3).
Although
there
was
a
minor
shift
in
the
mobility
of
the
protein-DNA
complex
when
ATP
was
introduced
in
the
reac-
tion,
the
overall
level
of
DNA-binding
activity
was
not
affected.
Similar
results
were
obtained
with
HSF1
present
in
a
crude
nuclear
extract
isolated
from
heat-shocked
HeLa
cells
(42).
The
results
indicate that
there
is
an
interaction
between
HSF1
and
hsc70,
but
this
interaction
does
not
significantly
affect
the
DNA-binding
activity
of
HSF1.
No
effect
of
elevated
hsp7O
and
heat
shock
protein
levels
on
induction
of
HSF1
activity.
To
assess
the
role
of
hsp70
on
the
DNA-binding
activity
of
HSF1
in
vivo,
we
analyzed
the
induc-
tion
of
the
DNA-binding
activity
of
the
endogenous
HSF
in
a
rat
fibroblast
cell
line
(Ratl)
and
in
the
same
cell
line
stably
transfected
and
constitutively
expressing
the
human
hsp7O
recombinant
HSF1
CL
(L
+
C
+
00
m
C)
a
r_
VOL.
14,
1994
-,
B
6556
RABINDRAN
ET
AL.
1
1
42oC
HS:
°
°
°
°
CD
.....:
..
..
..
.:..
S.
,.,Iw
Fo
A
rHSF
1
iu
B
rhsc7lO..
rhsp7O0
hhsp7O
FIG.
4.
Induction
of
HSF
DNA-binding
activity
in
rat
fibroblast
cells
containing
elevated
levels
of
hsp70
and
heat
shock
proteins.
(A)
Gel
mobility
shift
assays
of
extracts
of
Ratl,
TTRatl,
and
M21
cells,
heat
shocked
(HS)
for
the
times
(minutes)
indicated.
The
positions
of
free
(F)
and
bound
(B)
HSE
are
marked.
(B
and
C)
Western
blot
analysis
of
the
same
samples,
showing
the
levels
of
rat
HSF1
(rHSF1),
rat
hsc70
and
hsp70
(rhsc7o/rhsp7o),
and
human
hsp7o
(hhsp7o)
proteins.
protein
(M21
cells
[24]).
We
also
analyzed
the
DNA-binding
activity
of
HSF
in
Ratl
cells
rendered
thermotolerant
by
heat
shock
at
45°C
for
15
min
followed
by
recovery
for
16
h
at
37°C.
These
thermotolerant
Ratl
(lTRatl)
cells
have
elevated
levels
of
the
entire
complement
of
heat
shock
proteins
and
allow
an
assessment
of
whether
the
increased
levels
affect
the
level
of
the
DNA-binding
activity
in
vivo
(30).
As
shown
by
the
gel
mobility
shift
assay
in
Fig.
4A,
we
did
not
observe
significant
differences
in
the
levels
of
DNA-
binding
activity
of
HSF1
in
extracts
prepared
from
Ratl,
TFRatl,
or
M21
cells
after
30
min
of
heat
shock
at
42°C.
The
specificity
of
the
HSF1-HSE
complex
was
confirmed
by
specific
inhibition
of
complex
formation
with
antibodies
to
HSF1
or
with
unlabeled
HSE
DNA
(data
not
shown).
Western
blot
analysis
showed
that
similar
levels
of
rat
HSF1
were
present
in
all
extract
samples
analyzed
and
revealed
the
characteristic
heat
shock-dependent
decrease
in
electrophoretic
mobility
of
HSF1
in
all
three
cell
types
(Fig.
4B).
The
levels
of
70-kDa
stress
proteins
in
Ratl,
T1Ratl,
and
M21
cells
were
measured
on
the
same
blot
with
antibodies
specific
for
both
hsc70
and
hsp70
(Fig.
4C).
As
shown
in
this
experiment,
unstressed
Ratl
cells
contain
endogenous
hsc70
protein,
while
TTRatl
cells
have
the
same
amount
of
hsc70
and
approximately
half
that
of
hsp70
(lower
band)
accumu-
lated
as
a
result
of
thermotolerance.
M21
cells
contain
a
level
of
hsc70
similar
to
that
of
Ratl
and
lTRat
1
cells,
along
with
a
comparable
amount
of
human
hsp7o
(lower
band).
Thus,
an
artificial
increase
in
the
level
of
hsp70
protein
approximating
that
found
physiologically
after
a
heat
stress
does
not
lead
to
an
inhibition
of
the
DNA-binding
activity
of
the
endogenous
HSF
when
the
M21
cells
were
subjected
to
a
heat
stress.
Indeed,
even
expression
of
the
full
set
of
heat
shock
proteins
in
TITRatl
cells
failed
to
significantly
affect
induction
of
the
DNA-binding
activity.
Similar
results
were
obtained
when
Ratl,
]TRatl,
or
M21
cells
were
heat
shocked
for
30
min
at
a
higher
temperature,
45°C
(data
not
shown).
The
only
detect-
able
effect
of
increased
heat
shock
protein
levels
on
HSF1
activity
was
a
modest
decrease
(about
30%)
in
the
DNA-
binding
activity
after
prolonged
heat
shock
of
lTRatl
cells
for
90
min,
while
Ratl
and
M21
cells
showed
a
10%
decrease
in
the
DNA-binding
activity
of
HSF
(Fig.
4;
also
see
below).
Elevated
expression
of
heat
shock
proteins
accelerates
decrease
of
HSF1
DNA-binding
activity
during
recovery.
Al-
though
there
was
no
detectable
effect
of
increased
hsp70
and
heat
shock
protein
levels
on
induction
of
the
DNA-binding
activity
of
HSF1,
we
did
observe
a
reproducible
effect
of
heat
shock
protein
levels
on
the
kinetics
of
recovery
from
heat
shock.
Ratl,
TTRatl,
and
M21
cells
were
given
a
severe
heat
shock
at
45°C
for
30
min
and
allowed
to
recover
at
37°C
for
up
to
4
h.
To
exclude
de
novo
synthesis
of
heat
shock
proteins
that
would
augment
the
level
of
previously
expressed
stress
proteins
in
TrRatl
and
M21
cells,
the
protein
synthesis
inhibitor
cycloheximide
was
introduced
prior
to
the
beginning
of
the
heat
shock.
As
controls,
each
of
the
three
cell
types
was
also
maintained
on
cycloheximide
for
4.5
h
without
heat
stress.
As
shown
in
Fig.
5A,
the
DNA-binding
activity
of
HSF1
was
induced
to
the
same
extent
in
all
three
cell
types.
However,
the
decay
of
the
DNA-binding
activity
during
recovery
at
37°C
was
accelerated
in
3TRatl
cells.
The
DNA-binding
activities
of
HSF1
relative
to
the
amount
of
HSF1
protein
in
the
Ratl
and
M21
samples
(Fig.
5B)
were
observed
to
decrease
to
50
and
40%,
respectively,
of
maximal
levels
at
4
h
of
recovery.
At
the
same
time
point,
only
10%
of
HSF
activity
was
detectable
in
TTRatl
cells.
These
results
show
that
increased
levels
of
hsp70
alone
have
a
modest
effect
on
the
kinetics
of
HSF
deactivation,
while
a
full
complement
of
stress
proteins
can
accelerate
recovery
to
the
inactive
state
(Fig.
5D).
This
increased
rate
of
recovery
in
PTRatl
cells
is
also
correlated
with
a
reversion
to
the
hypophosphorylated
form
of
HSF1,
as
shown
by
the
change
in
SDS-gel
mobility
(Fig.
SB).
Elevated
heat
shock
protein
levels
accelerate
attenuation
but
do
not
block
induction
of
HSF
DNA-binding
activity
in
Drosophila
cells.
Since
the
results
presented
above
could
be
peculiar
to
Ratl
cells,
we
analyzed
the
effects
of
elevated
heat
shock
protein
levels
on
the
DNA-binding
activity
of
the
endogenous
HSF
in
Drosophila
SL2
cells.
We
first
investigated
the
effects
of
prolonged
heat
shock
on
the
DNA-binding
activity
of
Drosophila
HSF.
As
shown
by
the
gel
shift
assay
and
Western
blotting,
the
DNA-binding
activity
of
Drosophila
HSF
normalized
to
the
amount
of
HSF
protein
was
progressively
attenuated
at
1,
2,
and
4
h
of
continuous
heat
stress
at
a
moderate
temperature,
34.5°C
(Fig.
6).
This
attenuation
was
roughly
correlated
with
the
de
novo
synthesis
of
hsp70,
as
shown
by
Western
blotting.
Attenuation
of
the
DNA-binding
activity
was
not
observed
up
to
4
h
when
heat
shock
was
conducted
at
a
higher
temperature,
37.5°C,
which
resulted
in
a
greater
level
of
hsp70
production.
Thus,
the
potential
effects
of
elevated
heat
shock
protein
levels
are
operative
on
HSF
only
under
moderate
heat
stress.
To
further
investigate
the
kinetics
of
Drosophila
HSF
deac-
tivation
under
moderate
stress
conditions,
we
analyzed
the
DNA-binding
activity
of
HSF
in
thermotolerant
SL2
(TTSL2)
cells
(which
were
previously
induced
to
express
heat
shock
proteins
with
a
20-min
incubation
at
37.5°C
and
a
5-h
recovery
at
22°C).
Cells
were
treated
with
cycloheximide
30
min
before
the
start
of
the
34.5°C
heat
shock
to
prevent
de
novo
synthesis
of
heat
shock
proteins
in
the
course
of
the
continuous
heat
stress.
As
shown
by
the
gel
shift
assay
and
Western
blotting,
the
DNA
binding
activity
of
Drosophila
HSF
relative
to
the
MOL.
CELL.
BIOL.
EFFECT
OF
HSF-hsp7O
INTERACrION
ON
DNA-BINDING
ACTIVITY
6557
M21
+
CHX
recovery
C-)
I-o=
cJ
F
Am
1i1hl~
C
HS
30'
lh
2h
recovery
at
370C
,lrhsc7O
hhsp7O
FIG.
5.
(A
to
C)
Accelerated
decrease
of
HSF
DNA-binding
activity
during
recovery
from
heat
shock.
(A)
Gel
mobility
shift
assay
of
extracts
of
Ratl,
TJRatl,
and
M21
cells.
Cells
were
heat
shocked
for
30
min
at
45°C
(HS)
and
allowed
to
recover
at
37°C
for
the
times
indicated
(recovery)
in
the
presence
of
cyclobeximide
(CHX)
(see
text
for
details).
C,
control;
B,
bound
HSE;
F,
free
HSE.
(B
and
C)
Western
blot
analysis
of
the
same
samples,
showing
the
levels
of
rat
HSF1
(rHSF1),
rat
hsc70
and
hsp70
(rhsc70/rhsp7o),
and
human
hsp70
(hhsp7O)
(D)
Graph
showing
the
relative
DNA-binding
activity
of
HSF
during
recovery
from
heat
shock.
Densitometer
scans
of
the
films
in
panels
A
and
B
were
used
to
normalize
the
DNA-binding
activities
of
HSF;
the
highest
value
for
each
cell
type
was
set
at
1.0.
amount
of
HSF
protein
was
not
significantly
affected
when
TISL2
cells
heat
shocked
for
15
min
were
compared
with
nonthermotolerant
SL2
cells
stressed
similarly
(Fig.
7A
and
B).
However,
the
decay
of
DNA-binding
activity
occurred
faster
in
TTSL2
cells:
quantitatively
the
activity
was
decreased
to
10%
of
the
maximal
level
at
1
h
and
to
1%
after
2.5
h
of
heat
shock,
while
it
was
decreased
to
50
and
25%,
respectively,
for
the
nonthermotolerant
control
cells
(Fig.
7E).
A
faster
decay
of
the
DNA-binding
activity
of
HSF
was
also
observed
when
1TSL2
cells
were
analyzed
in
the
absence
of
cycloheximide
(data
not
shown).
Hence,
the
presence
of
previously
synthe-
sized
heat
shock
proteins
is
correlated
with
an
increased
rate
of
attenuation
but
not
with
an
inhibition
of
the
DNA-binding
activity
of
Drosophila
HSF.
SL2
/
34.5°C
HS
o
_I
cI
cI
t
.B..
DISCUSSION
The
intracellular
level
of
free
heat
shock
proteins
in
general,
and
the
hsp70
protein
family
in
particular,
has
been
suggested
to
be
part
of
an
autoregulatory
loop
by
which
the
cell
measures
the
level
of
thermal
stress
(10,
11).
This
autoregulatory
loop
has
been
proposed
to
affect
the
DNA-binding
and
trimeriza-
tion
of
HSF.
Accordingly,
the
DNA-binding
activity
of
HSF
would
be
suppressed
under
normal
conditions
by
the
constitu-
tive
level
of
free
70-kDa
stress
protein,
which
stabilizes
the
latent
HSF
monomer.
As
a
consequence
of
the
decreased
level
of
free
70-kDa
stress
protein
at
the
onset
of
heat
shock,
the
monomeric
form
of
HSF
is
able
to
trimerize,
leading
to
high-affinity
binding
to
HSEs
and
the
initiation
of
the
pathway
for
heat
shock
protein
synthesis.
The
ensuing
increase
in
heat
shock
protein
levels
drives
the
equilibrium
between
HSF
monomers
and
trimers
toward
the
monomeric
species,
thus
reestablishing
the
inert
form
of
the
transcription
factor
(5,
9,
31,
34).
A
prediction
of
this
model
is
that
an
artificial
elevation
of
heat
shock
proteins
to
the
level
observed
after
heat
shock
dHSFD
__
___W
on
we
mso
B
dhsp7O0
-
_
C
dhsc7O.
moon
_ _ _ _
D
FIG.
6.
DNA-binding
activity
of
Drosophila
HSF
is
attenuated
during
prolonged
heat
shock.
(A)
Gel
mobility
shift
assays
of
extracts
from
SL2
cells
heat
shocked
at
34.5
and
37.5°C
for
the
indicated
times
('
indicates
minutes).
(B
to
D)
Western
blots
of
the
same
samples,
showing
the
levels
of
Drosophila
HSF
(dHSF),
Drosophila
hsp7o
(dhsp7O),
and
Drosophila
hsc70
(dhsc70).
RatI
+
CHX
recovery
OmCOb
1-
C
a
c
C=J
A
TTRat1
+
CHX
recovery
0
I-
'i
_
1,
D
B
rHSF1.o
_
Is
-..
(0
L.
c)
z
0)
._
Ca
C
rhsc7O
,
rhsp7O'
Ratl
M21
TTRat1
4h
SL2
/
37.5°C
HS
ii.o
-
m
e
6
-COV
CV
F
1
-
A
VOL.
14,
1994
i..
-
li.
..
'.'-
'...
__40S
.,O!
ioio
WOON
:,*.
-,.
..
6558
RABINDRAN
ET
AL.
SL2
+
CHX
34.50C
HS:
b
o
°
TTSL2
+
CHX
-
C:>
c
i
A
dHSF
>
+*
*
t
-
t
...
........
dhsp7O
-
c
dhsclO
ip.
4~
48
D
E
1.00
0)
-Q
C:
C)
01)
O-5
0.75
0.50
0.25
0.00
,
SL-2
g
TTSL-2
0
15'30'
lh
2.5
h
time
at
34.50C
4h
FIG.
7.
(A
to
D)
Accelerated
decrease
of
the
DNA-binding
activity
of
Drosophila
HSF
in
cells
containing
accumulated
heat
shock
pro-
teins.
(A)
Gel
mobility
shift
assay
of
extracts
from
control
SL2
cells
and
TTSL2
cells.
Both
SL2
and
TTSL2
cells
were
heat
shocked
(HS)
continuously
at
34.5°C
for
the
times
('
indicates
minutes)
indicated
in
the
presence
of
118
,M
cycloheximide
(CHX)
introduced
30
min
before
the
start
of
heat
shock;
see
text
for
details.
B,
bound
HSE;
F,
free
HSE.
(B
to
D)
Western
blots
of
the
same
samples,
showing
the
levels
of
Drosophila
HSF
(dHSF),
Drosophila
hsp70
(dhsp7O),
and
Drosophila
hsc70
(dhsc7O).
(E)
Graph
showing
the
relative
DNA-
binding
activity
of
HSF
during
prolonged
heat
shock.
Densitometer
scans
of
the
films
in
panels
A
and
B
were
used
to
normalize
the
DNA-binding
activities
of
HSF;
the
highest
value
for
each
cell
type
was
set
at
1.0.
should
inhibit
induction
of
the
DNA-binding
activity
of
HSF.
Our
test
of
this
prediction
with
rat
M21
cells
constitutively
expressing
hsp70
at
a
level
close
to
that
found
physiologically
after
heat
shock
and
with
ITRatl
cells
expressing
a
full
complement
of
heat
shock
proteins
indicates
that
elevated
concentrations
of
hsp7O,
alone
or
in
combination
with
the
other
heat
shock
proteins,
do
not
significantly
block
induction
of
the
DNA-binding
activity
of
the
endogenous
HSF.
Similar
findings
with
Drosophila
TTSL2
cells
suggest
that
in
general,
the
equilibrium
between
HSF
monomers
and
trimers
is
not
solely
sensitive
to
changes
in
the
concentration
of
heat
shock
proteins
and
that
some
other
mechanism
for
controlling
tri-
merization
and
the
DNA-binding
activity
of
HSF
must
exist.
However,
the
inactivation
of
HSF
to
the
non-DNA-binding
state
is
indeed
accelerated
by
the
increased
levels
of
heat
shock
proteins.
This
finding
confirms
previous
reports
(34,
39)
and
suggests
that
heat
shock
proteins
do
assist,
directly
or
indi-
rectly,
in
the
disassembly
of
the
HSF
trimer.
It
should
be
noted
that
in
a
previous
study,
constitutive
overexpression
of
hsp7O
in
Drosophila
cells
led
to
the
sequestration
of
the
protein
in
granules,
where
it
appears
to
be
irreversibly
inactivated
and
unable
to
contribute
to
thermoresistance
(12).
While
it
is
plausible
that
some
of
the
constitutively
expressed
hsp7O
protein
in
our
rat
M21
cells
is
similarly
inactivated,
most
of
the
protein
appears
not
to
be
sequestered
in
granules
(25,
42)
and
is
functional
in
providing
thermoresistance
(24,
25,
30).
Previous
studies
detecting
in
vitro
interactions
between
HSF
and
hsp7O
were
based
on
the
electrophoretic
retardation
of
HSF-HSE
complexes
with
hsp70-specific
antibodies
in
a
gel
shift
assay
and
on
the
analysis
of
HSF-hsp7O
comnplexes
on
nondenaturing
pore
gradient
gels
(1,
5,
34,
46).
The
gel
shift
assay
also
revealed
a
greater
association
of
HSF
with
hsp7O
upon
prolonged
heat
stress
(1).
The
coimmunoprecipitation
of
the
70-kDa
stress
proteins
and
HSF
that
we
observed
in
extracts
prepared
from
heat-shocked
cells
generally
confirms
these
findings
and
indicates
further
that
the
complexes
be-
tween
HSF
and
hsp70
or
hsc7O
are
most
likely
formed
in
the
cell
prior
to
the
preparation
of
cell
extracts,
since
the
interac-
tion
could
not
be
displaced
with
an
excess
of
(unlabeled)
hsc70
protein.
However,
an
increased
HSF-hsp70
interaction
after
prolonged
heat
stress
was
not
observed
in
our
coimmunopre-
cipitation
studies.
This
difference
may
be
related
to
the
differ-
ent
assays
used
and
may
imply
a
separate
mode
of
interaction
between
HSF
and
hsp7O
that
is
detectable
only
by
the
gel
shift
assay
after
prolonged
heat
stress.
The
interaction
observed
between
the
70-kDa
stress
proteins
and
the
latent
(monomeric)
form
of
HSF
is
a
new
finding.
The
extent
of
this
interaction,
similar
to
the
interaction
detected
on
active
HSF,
augments
previous
studies
using
the
gel
shift
assay,
which
could
not
detect
the
inactive
form
of
HSF.
The
coim-
munoprecipitation
of
both
latent
and
active
forms
of
HSF
with
the
70-kDa
stress
proteins
is
not
necessarily inconsistent
with
sedimentation
and
gel
filtration
studies
from
our
laboratory
suggesting
that
the
two
native
forms
of
HSF
are
composed
of
solely
one
and
three
HSF
subunits,
respectively.
It
is
probable
that
substoichiometric
associations
of
a
transient
nature
cap-
tured
by
immunoprecipitation
would
not
be
detected
or
survive
biophysical
fractionation.
As
with
the
interaction
observed
for
the
activated
HSF,
the
functional
implications
of
the
interac-
tions
of
hsp7O
and
hsc70
with
the
latent
HSF
are
unclear,
but
they
might
reflect
the
participation
of
molecular
chaperones
in
both
the
assembly
and
disassembly
of
HSF
trimers.
We
detected
only
a
limited
effect
of
ATP
on
the
interaction
between
HSF
and
hsp7O.
Under
no
circumstances
was
substan-
tial
disruption
of
the
complex
observed.
In
contrast,
the
known
chaperone
action
of
hsp70
utilizes
ATP
hydrolysis
to
alter
the
folded
state
of
proteins
and
to
dissociate
oligomeric
complexes
(43).
Thus,
the
observed
interaction
between
HSF
and
hsp70
may
only
be
a
relic
of
a
larger
chaperone
complex
that
failed
to
survive
cell
extraction.
This
possibility
is
reinforced
by
the
MOL.
CELL.
BIOL.
EFFECT
OF
HSF-hsp7O
INTERACTION
ON
DNA-BINDING
ACTIVITY
6559
inability
to
deactivate
HSF
in
vitro
by
using
purified
70-kDa
stress
protein,
ATP,
and
HSF
trimers
purified
from
a
bacterial
expression
system.
A
similar
resistance
of
HSF-hsp7O
com-
plexes
to
disassociation
in
the
presence
of
ATP
has
been
reported
(1,
5).
Our
finding
that
the
induction
of
the
DNA-binding
activity
of
endogenous
HSF1
is
not
inhibited
in
rat
M21
cells
consti-
tutively
expressing
hsp7O
is
in
conflict
with
a
recent
report
that
showed
an
inhibition
of
HSF1
in
human
PEER
cells
constitu-
tively
expressing
hsp7O
at
a
comparable
level
(34).
Aside
from
methodological
differences
in
the
mode
of
hsp7O
expression
and
the
choice
of
a
different
mammalian
cell
line,
we
are
unable
to
reconcile
the
discrepant
findings.
However,
in
the
absence
of
a
direct
measurement
of
HSF1
levels
in
the
extracts
of
the
human
PEER
cells
that
were
induced
by
heat
stress,
it
remains
difficult
to
interpret
the
observed
changes
in
DNA-
binding
activity
(34).
These
differences
indicate
the
necessity
for
additional
investigations
to
clarify
the
role
of
heat
shock
proteins
in
the
cycling
of
HSF
between
latent
and
active
DNA-binding
forms.
In
this
respect,
it
should
be
noted
that
there
is
clear
genetic
evidence
for
a
functional
interaction
between
the
70-kDa
stress
proteins
and
HSF
in
S.
cerevisiae
(7,
52).
These
studies
indicate
that
the
70-kDa
stress
proteins
are
involved
in
repressing
the
activity
of
heat
shock
promoters
by
two
different
pathways.
One
pathway
is
independent
of
the
HSE
(and
HSF)
and
involves
an
adjacent
SRS1
(self-regulat-
ing
sequence)
element
(52).
The
other
pathway
implicates
HSF,
but
it
involves
repression
of
the
transcriptional,
not
the
DNA-binding
activity
of
the
constitutively
trimeric
yeast
HSF
(7).
Perhaps
it
is
this
repression
of
HSF
activity
by
hsp7O
that
operates
in
yeast
and
multicellular
eukaryotes.
A
different
mechanism
may
have
evolved
to
control
the
oligomeric
tran-
sition
and
DNA-binding
activity
of
HSF.
ACKNOWLEDGMENTS
We
thank
Susan
Lindquist
and
Rick
Morimoto
for
gifts
of
antibod-
ies,
Lois
Greene
and
Evan
Eisenberg
for
purified
70-kDa
heat
shock
protein,
and
members
of
our
laboratories
for
helpful
suggestions.
The
first
two
authors
contributed
equally
to
this
work.
REFERENCES
1.
Abravaya,
K.,
M.
P.
Myers,
S.
P.
Murphy,
and
R.
I.
Morimoto.
1992.
The
human
heat
shock
protein
hsp7O
interacts
with
HSF,
the
transcription
factor that
regulates
heat
shock
gene
expression.
Genes
Dev.
6:1153-1164.
2.
Abravaya,
K.,
B.
Phillips,
and
R.
I.
Morimoto.
1991.
Attenuation
of
the
heat
shock
response
in
HeLa
cells
is
mediated
by
the
release
of
bound
heat
shock
transcription
factor
and
is
modulated
by
changes
in
growth
and
in
heat
shock
temperatures.
Genes
Dev.
5:2117-2127.
3.
Amin,
J.,
J.
Ananthan,
and
R.
Voellmy.
1988.
Key
features
of heat
shock
regulatory
elements.
Mol.
Cell.
Biol.
8:3761-3769.
4.
Baler,
R.,
G.
Dahl,
and
R.
Voellmy.
1993.
Activation
of
heat
shock
genes
is
accompanied
by
oligomerization,
modification,
and
rapid
translocation
of
heat
shock
transcription
factor
HSF1.
Mol.
Cell.
Biol.
13:2486-2496.
5.
Baler,
R.,
W.
J.
Welch,
and
R.
Voellmy.
1992.
Heat
shock
gene
regulation
by
nascent
polypeptides
and
denatured
proteins:
hsp7o
as
a
potential
autoregulatory
factor.
J.
Cell
Biol.
117:1151-1159.
6.
Bonner,
J.,
S.
Heyward,
and
D.
Fackenthal.
1992.
Temperature-
dependent
regulation
of
a
heterologous
transcriptional
activation
domain
fused
to
yeast
heat
shock
transcription
factor.
Mol.
Cell.
Biol.
12:1021-1030.
7.
Boorstein,
W.
R.,
and
E.
A.
Craig.
1990.
Transcriptional
regulation
of
SSA3,
and
HSP70
gene
from
Saccharomyces
cerevisiae.
Mol.
Cell.
Biol.
10:3262-3267.
8.
Clos,
J.,
S.
K.
Rabindran,
J.
Wisniewski,
and
C.
Wu.
1993.
Induction
temperature
of
human
heat
shock
factor
is
repro-
grammed
in
a
Drosophila
cell
environment.
Nature
(London)
364:252-255.
9.
Clos,
J.,
J.
T.
Westwood,
P.
B.
Becker,
S.
Wilson,
K.
Lambert,
and
C.
Wu.
1990.
Molecular
cloning
and
expression
of
a
hexameric
Drosophila
heat
shock
factor
subject
to
negative
regulation.
Cell
63:1085-1097.
10.
Craig,
E.
A.,
and
C. A.
Gross.
1991.
Is
hsp7o
the
cellular
thermometer?
Trends
Biochem.
Sci.
16:135-140.
11.
DiDomenico,
B.
J.,
G.
E.
Bugaisky,
and
S.
Lindquist.
1982.
The
heat
shock
response
is
self-regulated
at
both
the
transcriptional
and
posttranscriptional
levels.
Cell
31:593-603.
12.
Feder,
J.
H.,
J.
M.
Rossi,
J.
Solomon,
N.
Solomon,
and
S.
Lindquist.
1992.
The
consequences
of
expressing
hsp7o
in
Dro-
sophila
cells
at
normal
temperatures.
Genes
Dev.
6:1402-1413.
13.
Fernandes,
M.,
H.
Xiao,
and
J.
T.
Lis.
1994.
Fine
structure
analysis
of
the
Drosophila
and
Saccharomyces
heat
shock
factor-heat
shock
element
interactions.
Nucleic
Acids
Res.
22:167-173.
14.
Firestone,
G.
L.,
and
S.
D.
Winguth.
1990.
Immunoprecipitation
of
proteins.
Methods
Enzymol.
182:688-700.
15.
Gao,
B.,
Y.
Emoto,
L.
Greene,
and
E.
Eisenberg.
1993.
Nucleotide
binding
properties
of
bovine
brain
uncoating
ATPase.
J.
Biol.
Chem.
268:8507-8513.
16.
Gross,
D.
S.,
K.
E.
English,
K.
W.
Collins,
and
S.
Lee.
1990.
Genomic
footprinting
of
the
yeast
HSP82
promoter
reveals
marked
distortion
of
the
DNA
helix
and
constitutive
occupancy
of
heat
shock
and
TATA
elements.
J.
Mol.
Biol.
216:611-631.
17.
Hensold,
J.
O.,
C.
R.
Hunt,
S.
K.
Calderwood,
D.
E.
Housman,
and
R.
E.
Kingston.
1990.
DNA
binding
of
heat
shock
factor
to
the
heat
shock
element
is
insufficient
for
transcriptional
activation
in
murine
erythroleukemia
cells.
Mol.
Cell.
Biol.
10:1600-1608.
18.
Jakobsen,
B.
K.,
and
H.
R.
B.
Pelham.
1988.
Constitutive
binding
of
yeast
heat
shock
factor
to
DNA
in
vivo.
Mol.
Cell.
Biol.
8:5040-5042.
f
19.
Jakobsen,
B.
K.,
and
H.
R.
B.
Pelham.
1991.
A
conserved
heptapeptide
restrains
the
activity
of
the
yeast
heat
shock
tran-
scription
factor.
EMBO
J.
10:369-375.
20.
Jurivich,
D.,
L.
Sistonen,
R.
Kroes,
and
R.
Morimoto.
1992.
Effect
of
sodium
salicylate
on
the
human
heat
shock
response.
Science
255:1243-1245.
21.
Kim,
S.-J.,
T.
Tsukiyama,
M.
S.
Lewis,
and
C.
Wu.
1994.
The
interaction
of
the
DNA
binding
domain
of
Drosophila
heat
shock
factor
with
its
cognate
DNA
site:
a
thermodynamic
analysis
using
analytical
ultracentrifugation.
Protein
Sci.
3:1040-1051.
22.
Kingston,
R.
E.,
T.
J.
Schuetz,
and
Z.
Larin.
1987.
Heat-inducible
human
factor
that
binds
to
a
human
hsp7O
promoter.
Mol.
Cell.
Biol.
7:372-375.
23.
Larson,
J.
S.,
T.
J.
Schuetz,
and
R.
E.
Kingston.
1988.
Activation
in
vitro
of
sequence-specific
DNA
binding
by
a
human
regulatory
factor.
Nature
(London)
335:372-375.
24.
Li,
G.
C.,
L.
Li,
R.
Y.
Liu,
M.
Rehman,
and
W.
M.
F.
Lee.
1992.
Heat
shock
protein
hsp7o
protects
cells
from
thermal
stress
even
after
deletion
of
its
ATP-binding
domain.
Proc.
Natl.
Acad.
Sci.
USA
89:2036-2040.
25.
Li,
G.
C.,
L.
Li,
Y.-K.
Liu,
J.
Y.
Mak,
L.
Chen,
and
W.
M.
F.
Lee.
1991.
Thermal
response
of
rat
fibroblast
stably
transfected
with
the
human
70-kDa
heat
shock
protein-encoding
gene.
Proc.
Natl.
Acad.
Sci.
USA
88:1681-1685.
26.
Lindquist,
S.
1986.
The
heat-shock
response.
Annu.
Rev.
Bio-
chem.
55:1151-1191.
27.
Lindquist,
S.,
and
E.
A.
Craig.
1988.
The
heat-shock
proteins.
Annu.
Rev.
Genet.
22:631-677.
28.
Lis,
J.
T.,
and
C.
Wu.
1992.
Heat
shock
factor,
p.
907-930.
In
S.
L.
McKnight
and
K.
R.
Yamamoto
(ed.),
Transcriptional
regulation.
Cold
Spring
Harbor
Laboratory
Press,
Cold
Spring
Harbor,
N.Y.
29.
Lis,
J.
T.,
and
C.
Wu.
1993.
Protein
traffic
on
the
heat
shock
promoter:
parking,
stalling
and
trucking
along.
Cell
74:1-20.
30.
Liu,
R.
Y.,
X.
Li,
L.
Li,
and
G.
C.
Li.
1992.
Expression
of
human
hsp7o
in
rat
fibroblasts
enhances
cell
survival
and
facilitates
recovery
from
translational
and
transcriptional
inhibition
follow-
ing
heat
shock.
Cancer
Res.
52:3667-3673.
31.
Morimoto,
R.
1993.
Cells
in
stress:
transcriptional
activation
of
heat
shock
genes.
Science
269:1409-1410.
VOL.
14,
1994
6560
RABINDRAN
ET
AL.
32.
Morimoto,
R.
I.,
and
K.
Milarski.
1990.
Expression
and
function
of
vertebrate
hsp70
genes,
p.
323-359.
In
R.
I.
Morimoto,
A.
Tissieres,
and
C.
Georgopoulos
(ed.),
Stress
proteins
in
biology
and
medicine.
Cold
Spring
Harbor
Laboratory
Press,
Cold
Spring
Harbor,
N.Y.
33.
Morimoto,
R.
I.,
A.
Tissieres,
and
C.
Georgopoulos.
1990.
The
stress
response,
function
of
the
proteins,
and
perspectives,
p.
1-36.
In
R.
I.
Morimoto,
A.
Tissieres,
and
C.
Georgopoulos
(ed.),
Stress
proteins
in
biology
and
medicine.
Cold
Spring
Harbor
Laboratory
Press,
Cold
Spring
Harbor,
N.Y.
34.
Mosser,
D.
D.,
J.
Duchaine,
and
B.
Massie.
1993.
The
DNA-
binding
activity
of
human
heat
shock
transcription
factor
is
regulated
in
vivo
by
hsp70.
Mol.
Cell.
Biol.
13:5427-5438.
35.
Nover,
L.,
D.
Helimund,
D.
Neumann,
K.-D.
Scharf,
and
E.
Serfling.
1984.
The
heat
shock
response
of
eukaryotic
cells.
Biol.
Zentralbl.
103:357-435.
36.
Nieto-Sotelo,
J.,
G.
Wiederrecht,
A.
Okuda,
and
C.
S.
Parker.
1990.
The
yeast
heat
shock
transcription
factor
contains
a
tran-
scriptional
activation
domain
whose
activity
is
repressed
under
nonshock
conditions.
Cell
62:807-817.
37.
Pelham,
H.
R.
B.
1982.
A
regulatory
upstream
promoter
element
in
the
Drosophila
hsp7o
heat-shock
gene.
Cell
30:517-528.
38.
Perisic,
O.,
H.
Xiao,
and
J.
T.
Lis.
1989.
Stable
binding
of
Drosophila
heat
shock
factor
to
head-to-head
and
tail-to-tail
repeats
of
a
conserved
5
bp
recognition
unit.
Cell
59:797-806.
39.
Price,
B.
D.,
and
S.
K.
Calderwood.
1992.
Heat-induced
transcrip-
tion
from
RNA
polymerases
II
and
III
and
HSF
binding
activity
are
co-ordinately
regulated
by
the
products
of
the
heat
shock
genes.
J.
Cell.
Physiol.
153:392-401.
40.
Rabindran,
S.
K.,
G.
Giorgi,
J.
Clos,
and
C.
Wu.
1991.
Molecular
cloning
and
expression
of
a
human
heat
shock
factor,
HSF1.
Proc.
Natl.
Acad.
Sci.
USA
88:6906-6910.
41.
Rabindran,
S.
K.,
R.
I.
Haroun,
J.
Clos,
J.
Wisniewski,
and
C.
Wu.
1993.
Regulation
of
heat
shock
factor
trimerization:
role
of
a
conserved
leucine
zipper.
Science
259:230-234.
42.
Rabindran,
S.
K.,
G.
C.
Li,
and
C.
Wu.
Unpublished
results.
43.
Rothman,
J.
E.
1989.
Polypeptide
chain
binding
proteins:
catalysts
of
protein
folding
and
related
processes
in
cells.
Cell
59:591-601.
44.
Sarge,
K.,
S.
P.
Murphy,
and
R.
I.
Morimoto.
1993.
Activation
of
heat
shock
transcription
by
HSF1
involves
oligomerization,
acqui-
sition
of
DNA
binding
activity,
and
nuclear
localization
and
can
occur
in
the
absence
of
stress.
Mol.
Cell.
Biol.
13:1392-1407.
45.
Sarge,
K.,
V.
Zimarino,
K.
Holm,
C.
Wu,
and
R.
I.
Morimoto.
1991.
Cloning
and
characterization
of
two
mouse
heat
shock
factors
with
distinct
inducible
and
constitutive
DNA
binding
activity.
Genes
Dev.
5:1902-1911.
46.
Schlesinger,
M.,
and
C.
Ryan.
1993.
An
ATP-
and
hsc7o-depen-
dent
oligomerization
of
nascent
heat-shock
factor
(HSF)
polypep-
tide
suggests
that
HSF
itself
could
be
a
"sensor"
for
the
cellular
stress
response.
Protein
Sci.
2:1356-1360.
47.
Sorger,
P.
K.
1990.
Yeast
heat
shock
factor
contains
separable
transient
and
sustained
response
transcriptional
activators.
Cell
62:793-805.
48.
Sorger,
P.
K
1991.
Heat
shock
factor
and
the
heat
shock
response.
Cell
65:363-366.
49.
Sorger,
P.
K.,
M.
J.
Lewis,
and
H.
R
B.
Pelham.
1987.
Heat
shock
factor
is
regulated
differently
in
yeast
and
HeLa
cells.
Nature
(London)
329:81-84.
50.
Sorger,
P.
K.,
and
H.
C.
M.
Nelson.
1989.
Trimerization
of
a
yeast
transcriptional
activator
via
a
coiled-coil
motif.
Cell
59:807-813.
51.
Sorger,
P.
K.,
and
H.
R.
B.
Pelham.
1988.
Yeast
heat
shock
factor
is
an
essential
DNA-binding
protein
that
exhibits
temperature-
dependent
phosphorylation.
Cell
54:855-864.
52.
Stone,
D.
E.,
and
E.
A.
Craig.
1990.
Self-regulation
of
70-
kilodalton
heat
shock
proteins
in
Saccharomyces
cerevisiae.
Mol.
Cell.
Biol.
10:1622-1632.
53.
Treuter,
E.,
L.
Nover,
K.
Ohme,
and
K.-D.
Scharf.
1993.
Promoter
specificity
and
deletion
analysis
of
three
tomato
heat
stress
transcription
factors.
Mol.
Gen.
Genet.
240:113-125.
54.
Welch,
W.
J.
1990.
The
mammalian
stress
response:
cell
physiology
and
biochemistry
of
stress
proteins,
p.
223-278.
In
R.
I.
Morimoto,
A.
Tissieres,
and
C.
Georgopoulos
(ed.),
Stress
proteins
in
biology
and
medicine.
Cold
Spring
Harbor
Laboratory
Press,
Cold
Spring
Harbor,
N.Y.
55.
Westwood,
J. T., J.
Clos,
and
C.
Wu.
1991.
Stress-induced
oli-
gomerization
and
chromosomal
relocalization
of
heat-shock
fac-
tor.
Nature
(London)
353:822-827.
56.
Westwood,
J.
T.,
and
C.
Wu.
1993.
Activation
of
Drosophila
heat
shock
factor:
conformational
change
associated
with
monomer
to
trimer
transition.
Mol.
Cell.
Biol.
13:3481-3486.
57.
Wu,
C.
1984.
Two
protein-binding
sites
in
chromatin
implicated
in
the
activation
of heat
shock
genes.
Nature
(London)
309:229-234.
58.
Wu,
C.
1984.
Activating
protein
factor
binds
in
vitro
to
upstream
control
sequences
in
heat
shock
gene
chromatin.
Nature
(London)
311:81-84.
59.
Xiao,
H.,
and
J.
T.
Lis.
1988.
Germline
transformation
used
to
define
key
features
of
heat-shock
response
elements.
Science
239:1139-1142.
60.
Zimarino,
V.,
and
C.
Wu.
1987.
Induction
of
sequence-specific
binding
of
Drosophila
heat
shock
activator.
Nature
(London)
327:727-730.
MOL.
CELL.
BIOL.
