First published online 15 July 2003
doi: 10.1242/jcs.00671
Formation of nuclear stress granules involves HSF2 and coincides with the nucleolar localization of Hsp70
Tero-Pekka Alastalo1,*,
Maria Hellesuo1,*,
Anton Sandqvist1,2,
Ville Hietakangas1,
Marko Kallio1,
and
Lea Sistonen1,3,
1 Turku Centre for Biotechnology, University of Turku, Åbo Akademi
University, BioCity, PO Box 123, FIN-20521 Turku, Finland
2 Department of Biochemistry and Pharmacy, Åbo Akademi University, Turku,
Finland
3 Department of Biology, Åbo Akademi University, Turku, Finland
View larger version (43K):
[in a new window]
|
Fig. 4. The oligomerization domain HR-A/B is essential for the physical interaction
between HSF1 and HSF2. (A) Schematic presentation of HSF2 deletion mutants.
The DNA-binding domain (DBD), N-terminal oligomerization domain (HR-A/B) and
C-terminal leucine zipper (HR-C) are indicated. (B) Anti-Myc antibodies were
used to immunoprecipitate HSF1, and antibodies against HSF1 and HSF2 were used
for western blotting. An empty Myc vector was used in the mock transfections.
The expression levels were analyzed by western blotting (WB), as indicated.
Arrows and asterisks indicate the inducibly phosphorylated HSF1 and the
endogenous HSF2, respectively. Molecular weights (kDa) are shown.
|
|
View larger version (71K):
[in a new window]
|
Fig. 1. Heat-induced localization of HSF1 and HSF2 in nuclear granules.
Fluorescence micrographs of HSF1 and HSF2 localization in untreated (C) and
1-hour heat-shocked (HS) HeLa cells, using polyclonal antibodies against HSF1
(left) and HSF2 (right). The phase contrast and DAPI-stained DNA of the same
fields are also shown. (B) Cells were untreated (C), heat-shocked for 30
minutes to 6 hours (30'-6h) or, after 1 hour of heat shock, allowed to
recover for 1 or 3 hours (HS+1R, HS+3R). (C) The graph represents the means
± SEM of three independent experiments in which the number of granule
positive cells was analyzed by counting nine groups of 50 cells. (D) The HSF
DNA-binding activity and HSF1 hyperphosphorylation were analyzed by gel
mobility shift assay with radiolabeled HSE (top) and by western blotting
(bottom), respectively. The arrow indicates the inducibly phosphorylated HSF1.
Scale bars, 5 µm.
|
|
View larger version (74K):
[in a new window]
|
Fig. 2. Co-localization of HSF1 and HSF2 in stress granules. (A) HSF1 and HSF2
co-localize to the stress granules upon heat shock (HS). Monoclonal antibodies
against HSF1 were used to detect endogenous HSF1 and polyclonal HSF2
antibodies to detect endogenous HSF2 by double staining of HeLa cells analyzed
by immunofluorescence microscopy. (B) The epitope-tagged HSF1 and HSF2
constructs were transiently transfected into HeLa cells. Anti-Myc antibody was
used to detect the ectopic HSFs (green) and polyclonal antibodies against HSF1
and HSF2 were used to detect endogenous HSFs (red). Yellow indicates
co-localization in the merged image (Merge). Scale bars, 5 µm.
|
|
View larger version (49K):
[in a new window]
|
Fig. 3. Coimmunoprecipitation of HSF1 and HSF2. (A) Coimmunoprecipitation of HSF2
with transiently (left) or stably (right) overexpressed Myc-tagged HSF1 (4H9)
using monoclonal anti-Myc antibodies (IP). The lower blot shows the expression
levels of HSF1 and HSF2 (WB). (B) Exogenous HSF1 was coimmunoprecipitated with
endogenous HSF2 using monoclonal HSF2 antibodies (IP). Anti-Flag antibodies
were used as an antibody control. Cells were either left untreated (C) or heat
shocked for 1 hour (HS). (C) The dynamics of the interacting complex was
analyzed by coimmunoprecipitation of endogenous HSF2 with endogenous HSF1
using monoclonal HSF1 antibodies in HeLa cells heat-shocked for indicated
times (IP). Thermotolerance was acquired at 1 or 3 hours of recovery after a
1-hour heat shock (HS+1R, HS+3R). The bottom blots show the expression levels
of HSF1 and HSF2 (WB). Hsc70 was an equal loading control. HSF2- and
-ß isoforms are indicated. Arrows and asterisks mark the inducibly
phosphorylated HSF1 and the endogenous HSF2, respectively.
|
|
View larger version (60K):
[in a new window]
|
Fig. 8. Nucleolar localization of Hsp70 correlates with the activation and
deactivation of HSF1. (A) An autoradiograph image of proteins interacting with
Myc-HSF1 in labeled 4H9 cells. Anti-Myc antibodies were used for
immunoprecipitation. K562 cells were used as a negative control. The two
obtained bands were detected also by silver staining and analyzed by
matrix-assisted laser desorption ionization mass spectrometry. (B)
Coimmunoprecipitation of endogenous HSF2 and Hsp70 with endogenous HSF1 in
heat-shocked HeLa cells. Monoclonal HSF1 antibodies were used for
immunoprecipitations (IP). Cells were untreated (C) or heat shocked for 30
minutes to 6 hours (30'-6h). Thermotolerance was obtained by a 1-hour
heat shock, followed by 1 or 3 hours of recovery (HS+1R or HS+3R).
Thermotolerant cells were heat shocked for 30 minutes or 1 hour
(HS+3R+HS30' or HS+3R+HS1h). The bottom blots (WB) show the expression
levels of HSF1, HSF2 and Hsp70. Hsc70 was a control for equal loading. Arrows
indicate the inducibly phosphorylated HSF1. (C) Fluorescence micrographs
showing the localization of HSF1 and Hsp70 in untreated (C), heat-shocked (HS)
for 1 hour or 6 hours, thermotolerant (HS+3R), and heat-shocked thermotolerant
HeLa cells (HS+3R+HS1h). Merge lane indicates possible co-localization
(yellow) and DNA is stained by DAPI (blue). Scale bar, 5 µm.
|
|
View larger version (54K):
[in a new window]
|
Fig. 5. HSF1 is the major component of the heat-induced HSE-binding complex. (A)
Cells were subjected to different periods of heat shock (HS; 1-6 h) or left
untreated (C). The HSF DNA-binding activity in the whole cell extracts was
analyzed by gel mobility shift assay with radiolabeled HSE (left panel).
Antibody perturbations with Flag and Myc antibodies (1:10) were used to detect
the presence of HSF1 or HSF2 in the DNA-binding complex, respectively (right
panel). A hemin-treated sample was used as a control. An empty Myc vector was
used in the mock transfections. (B) Flag antibodies were used to
immunoprecipitate HSF1 from the same lysates as in panel A. Western blotting
was performed as in Fig. 3. The
expression levels are presented in the right hand panel (WB). Arrows and
asterisks indicate the inducibly phosphorylated HSF1 and the endogenous HSF2,
respectively.
|
|
View larger version (46K):
[in a new window]
|
Fig. 6A,B,C. HSF2 affects the localization of HSF1 in stress granules. (A-C) HeLa cells
were transiently transfected with Flag-tagged HSF2 and constructs shown in
Fig. 4A. Cells were subjected
to a 1-hour heat shock (HS) or left untreated (C), and immunodetected with
anti-Flag antibodies for HSF2 deletion mutants (green) and with polyclonal
HSF1 antibodies for the endogenous HSF1 (red). Colocalization is shown in the
merge lane (yellow) and DNA is stained with DAPI (blue).
|
|
View larger version (23K):
[in a new window]
|
Fig. 6D. HSF2 affects the localization of HSF1 in stress granules. (D) Granule
formation was analyzed by counting 50 Flag-positive cells from three
independent experiments (total 150 cells), and the graph represents the means
± s.e.m. Scale bars, 5 µm.
|
|
View larger version (52K):
[in a new window]
|
Fig. 7. The localization of HSF1 and HSF2 in stress granules is suppressed in
thermotolerant HeLa cells. (A) Fluorescence micrographs of the localization of
HSF1 and HSF2 in untreated (C), heat-shocked (30' or 1h), thermotolerant
(HS+3R) and heat-shocked thermotolerant (HS+3R+HS30' or HS+3R+HS1h)
cells. Proteins were visualized using monoclonal antibodies against HSF1 (red)
and polyclonal antibodies against HSF2 (green). Co-localization is shown in
the merge lane (yellow) and DNA is stained with DAPI (blue). Scale bars, 5
µm. (B) The graph represents the means ± s.e.m. of the three
independent experiments in which the number of granule-positive cells was
analyzed by counting nine groups of 50 cells.
|
|
© The Company of Biologists Ltd 2003
