Molecular Cell, Volume 48 Supplemental Information Cytoplasmic STAT3 Represses Autophagy by Inhibiting PKR Activity Shensi Shen, Mireia Niso-Santano, Sandy Adjemian, Tetsuo Takehara, Shoaib Ahmad Malik, Hervé Minoux, Sylvie Souquere, Guillermo Mariño, Sylvie Lachkar, Laura Senovilla, Lorenzo Galluzzi, Oliver Kepp, Gérard Pierron, Maria Chiara Maiuri, Hayato Hikita, Romano Kroemer, and Guido Kroemer Supplemental Information - Inventory Supplemental Figures Figure S1. Specific inhibitors of STAT3 stimulate ATG7- and Beclin 1-dependent autophagy (related to Figure 1c,d) Figure S2. STAT3 inhibitors stimulate the autophagic flux (related to Figure 1g,h) Figure S3. Cytoplasmic, not nuclear, STAT3 inhibits autophagy, irrespective of phosphorylation on Y705 (related to Figure 2i,j) Figure S4. STAT3 inhibitors induce autophagy independent of HSP90 (related to Figure 4a) Figure S5. STAT3 inhibitors induce the dissociation of polyribosomes and the appearance of stress granules (related to Figure 4b) Figure S6. Interaction between STAT family members and eIF2α kinases (related to Figure 4d) Figure S7. Characterization of the interaction between STAT3 and PKR (related to Figure 5b) Figure S8. The autophagic response to STAT3 inhibitors requires PKR and eIF2α but not PERK, GCN2 and HRI (related to Figure 6e-g) Supplemental Tables Table S1. Primers employed for site-directed mutagenesis (related to Experimental Procedures)
25
Embed
Cytoplasmic STAT3 represses autophagy by inhibiting PKR activity ...
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Molecular Cell, Volume 48
Supplemental Information
Cytoplasmic STAT3 Represses Autophagy by Inhibiting PKR Activity
Shensi Shen, Mireia Niso-Santano, Sandy Adjemian, Tetsuo Takehara, Shoaib Ahmad Malik, Hervé Minoux, Sylvie Souquere, Guillermo Mariño, Sylvie Lachkar, Laura Senovilla, Lorenzo Galluzzi, Oliver Kepp, Gérard Pierron, Maria Chiara Maiuri, Hayato Hikita, Romano Kroemer, and Guido Kroemer
Supplemental Information - Inventory
Supplemental Figures
Figure S1. Specific inhibitors of STAT3 stimulate ATG7- and Beclin 1-dependent autophagy
(related to Figure 1c,d)
Figure S2. STAT3 inhibitors stimulate the autophagic flux (related to Figure 1g,h)
Figure S3. Cytoplasmic, not nuclear, STAT3 inhibits autophagy, irrespective of phosphorylation
on Y705 (related to Figure 2i,j)
Figure S4. STAT3 inhibitors induce autophagy independent of HSP90 (related to Figure 4a)
Figure S5. STAT3 inhibitors induce the dissociation of polyribosomes and the appearance of stress
granules (related to Figure 4b)
Figure S6. Interaction between STAT family members and eIF2α kinases (related to Figure 4d)
Figure S7. Characterization of the interaction between STAT3 and PKR (related to Figure 5b)
Figure S8. The autophagic response to STAT3 inhibitors requires PKR and eIF2α but not PERK,
GCN2 and HRI (related to Figure 6e-g)
Supplemental Tables
Table S1. Primers employed for site-directed mutagenesis (related to Experimental Procedures)
Legends to Additional Supplementary Items
Movie S1. Molecular model of the STAT3-PKR interaction
Spreadsheet S1. Panel of autophagy inducers employed in this study (related to Figure 7c,d)
Supplemental Experimental Procedures (related to Experimental Procedures)
Supplemental References
Supplemental Information - Supplemental Figures
Figure S1. Specific inhibitors of STAT3 induce ATG7- and Beclin 1-dependent autophagy.
(a) Chemical structure of JSI-124, Stattic and WP1066.
(b,c) Human osteosarcoma U2OS cells were left untreated or incubated with 10 μM Stattic or 10 μM
WP1066 for 1 h, followed or not by the administration of 100 ng/mL recombinant interleukin-6 (IL-6)
for additional 30 min. Finally, cells were processed for the immunoblotting-assisted detection of
STAT3 (b) or STAT1 (c) phosphorylation. β actin was monitored as a loading control. Numbers
illustrate band intensity upon normalization to the STAT3 (b) or STAT1 (c) total protein levels.
(d,e) U2OS cells stably expressing a GFP-LC3 chimera were transfected with a control siRNA
(siUNR), or with siRNAs targeting ATG7 (siATG7) and Beclin 1 (siBeclin1) for 48h, followed by the
administration of 10 μM Stattic for 6 h. Representative images are shown in (d), quantitative data in
(e). Scale bar = 5 μm. Columns report the percentage of cells exhibiting GFP-LC3 dots in the
cytoplasm (GFP-LC3vac cells, Mean ± SEM, n = 3; **p<0.01, Student’s t test, compared to equally
treated siUNR-transfected cells). The inset in (e) depicts the efficacy of siRNA-mediated ATG7 and
Beclin 1 downregulation, as assessed by immunoblotting. β actin levels were monitored to ensure
equal loading of lanes.
Figure S2. STAT3 inhibitors stimulate the autophagic flux.
(a) Human osteosarcoma U2OS cells transiently expressing a RFP-GFP-LC3 chimera were
maintained in control (Co) or nutrient-free (NF) conditions, or administered with 10 μM Stattic, in the
absence or in the presence of 50 nM bafilomycin A1 (BafA1) for the indicated time, followed by the
quantification of the number of RFP+GFP+ or RFP+GFP- dots per cell by fluorescence microscopy
(Mean ± SEM, n = 3; *p<0.05, **p<0.01 Student’s t test, RFP+GFP- dots compared to untreated cells,
#p<0.05, ##p<0.01, RFP+GFP+ dots compared to cells treated with the same inducer of autophagy in
the absence of BafA1).
(b) U2OS cells were treated as in (a), followed by immunoblotting for the assessment of LC3
lipidation and p62 levels. β actin was monitored as a loading control. Numbers illustrate the
abundance of LC3-II and p62 upon normalization to β actin levels.
(c,d) U2OS cells were treated with 1 μM JSI-124 (c) or 10 μM WP1066 (d), alone or in combination
with 50 nM BafA1 for 6 h, followed by the immunoblotting-assisted determination of LC3 lipidation
and p62 levels. β actin was monitored as a loading control. Numbers illustrate the abundance of LC3-
II and p62 upon normalization to β actin levels.
(e,f) U2OS cells stably expressing a GFP-LC3 fusion protein were treated as in (d), followed by
immunostaining with a LAMP2a-specific antibody and confocal fluorescence microscopy. Columns
in (e) indicate the percentage of cells in which GFP-LC3 and LAMP2a colocalized (Mean ± SEM, n =
3; **p<0.01, Student’s t test, compared to untreated cells, #p<0.05, compared to Stattic-treated cells).
In (f) representative images are reported together with co-localization profiles as determined by the
ImageJ software along the indicated α--ω axis.
Figure S3. Cytoplasmic, not nuclear, STAT3 inhibits autophagy, irrespective of phosphorylation
on Y705.
(a,c) Full version of Figure 2i. Human osteosarcoma U2OS cells were co-transfected with a plasmid
coding for a RFP-LC3 chimera and either a GFP-coding control vector (pcDNA3-GFP, a) or
constructs coding for merely cytoplasmic (NES, b) or merely nuclear (NLS, c) GFP-tagged STAT3
variants for 24 h. Then, cells were maintained in control (Co) or nutrient-free (NF) conditions, or
administered with 1 μM thapsigargin (Thaps) or 10 μM Stattic, for 6 h, followed by nuclear
counterstaining with Hoechst 33342 (blue) and confocal fluorescence microscopy. Representative
images are reported. Scale bar = 5 μm.
(d) Stat3Δ/Δ mouse embryonic fibroblasts (MEFs) were transiently transfected with an empty plasmid
(pBabe) or with constructs encoding wild-type (WT) STAT3 or a non-phosphorylatable STAT3
variant (STAT3Y705F). Twenty-four h later, MEFs were either maintained in Co conditions or
stimulated with 100 ng/mL recombinant interleukin-6 (IL-6) for 30 min and processed for the
immunoblotting-assisted determination of STAT3 phosphorylation. GAPDH abundance was
monitored as a loading control. Numbers illustrate STAT3 phosphorylation normalized to total
STAT3 levels.
(e) U2OS cells stably expressing a GFP-LC3 fusion protein were transiently transfected with an
empty plasmid (pBabe) or with constructs encoding WT STAT3 orSTAT3Y705F. Twenty-four h later,
cells were maintained in Co or NF conditions, or administered with 1 μM rapamycin (Rapa) or 1 μM
Thaps for 15 h, and then subjected to fluorescence microscopy for the quantification of cells
exhibiting GFP-LC3 dots in the cytoplasm (GFP-LC3vac cells, Mean ± SEM, n = 3; **p<0.01,
Student’s t test, compared to equally treated pBabe-transfected cells).
Figure S4. STAT3 inhibitors induce autophagy independent of HSP90.
(a,b) Human osteosarcoma U2OS cells stably expressing a GFP-LC3 fusion protein were maintained
in control (Co) or nutrient-free (NF) conditions, or treated with 1 μM rapamycin (Rapa), 10 μM Stattic
or 10 μM WP1066 for 6 h, alone or combined with 1 μM 17-allylamino-17-demethoxygeldanamycin
(17-AAG). Representative images are shown in (a), quantitative data in (b). Scale bar = 5 μm.
Columns report the percentage of cells exhibiting GFP-LC3 dots in the cytoplasm (GFP-LC3vac cells,
Mean ± SEM, n = 3; *p<0.05, Student’s t test, compared to cells stimulated to undergo autophagy by
the same inducer in the absence of 17-AAG).
Figure S5. STAT3 inhibitors induce the dissociation of polyribosomes and the appearance of
stress granules.
(a) Human osteosarcoma U2OS cells were left untreated (Co) or incubated with 10 μM Stattic for 12
h, and then processed for transmission electron microscopy. Representative images of ribosomes
(dissociated by Stattic) and stress granules (induced by Stattic, indicated by asterisks) are reported in
upper and lower panels respectively. Scale bars are provided.
(b,c) U2OS cells stably expressing a G3BP-GFP chimera were maintained in Co or nutrient-free (NF)
conditions, or treated with 1 μM thapsigargin (Thaps), 10 μM Stattic or 10 μM WP1066 for 6 h,
followed by immunostaining with an anti-PKR antibody and confocal fluorescence microscopy.
Representative images are shown in (d), quantitative data in (e). Scale bar = 5 μm. Columns report the
number of G3BP-GFP+ dots/cell (Mean ± SEM, n = 3; *p<0.05, **p<0.01, Student’s t test, compared
to untreated cells).
Figure S6. Interaction between STAT family members and eIF2α kinases.
(a) Human osteosarcoma U2OS cells were maintained in control (Co) of nutrient-free (NF)
conditions, or treated with 10 μM Stattic, 10 μM WP1066, 1 μM rapamycin (Rapa) or 0.5 mM
palmitate (PAL) for 15h. Thereafter, cells were lysed and lysates were immunoprecipitated with a
STAT3-specific antibodies, followed by the immunoblotting-assisted detection of the indicated eIF2α
kinases. Numbers illustrate PKR abundance within immunoprecipitates (normalized to PKR levels in
input lysates).
(b) U2OS cells were treated as in (a), followed by lysis, immunoprecipitation with a PKR-specific
antibody and immunoblotting-assisted determination of STAT1 and STAT3 levels. Numbers illustrate
STAT3 abundance within immunoprecipitates (normalized to STAT3 levels in input lysates).
(c) Superimposition of the crystal structures of eIF2α (in pink the relevant domain) and STAT3 (in
light blue the relevant domain). The panel on the right focuses on STAT3 SH2 domain, folding as a α-
helix that mimics the C-terminal of eIF2α.
(d) Diagram of the STAT3 mutants employed in this study. CC, coiled-coil domain; DBD, DNA-