Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2018 Nov;25(10):1766-1780.
doi: 10.1038/s41418-018-0076-9. Epub 2018 Mar 9.

The mTOR-S6 Kinase Pathway Promotes Stress Granule Assembly

Affiliations
Free PMC article

The mTOR-S6 Kinase Pathway Promotes Stress Granule Assembly

Aristeidis P Sfakianos et al. Cell Death Differ. .
Free PMC article

Erratum in

Abstract

Stress granules are cytoplasmic mRNA-protein complexes that form upon the inhibition of translation initiation and promote cell survival in response to environmental insults. However, they are often associated with pathologies, including neurodegeneration and cancer, and changes in their dynamics are implicated in ageing. Here we show that the mTOR effector kinases S6 kinase 1 (S6K1) and S6 kinase 2 (S6K2) localise to stress granules in human cells and are required for their assembly and maintenance after mild oxidative stress. The roles of S6K1 and S6K2 are distinct, with S6K1 having a more significant role in the formation of stress granules via the regulation of eIF2α phosphorylation, while S6K2 is important for their persistence. In C. elegans, the S6 kinase orthologue RSKS-1 promotes the assembly of stress granules and its loss of function sensitises the nematodes to stress-induced death. This study identifies S6 kinases as regulators of stress granule dynamics and provides a novel link between mTOR signalling, translation inhibition and survival.

Conflict of interest statement

The authors declare that they have no conflict of interest.

Figures

Fig. 1
Fig. 1
S6 kinases accumulate in stress granules. a HeLa cells were treated with either 0.5 mM of NaAsO2 for 30 min or 0.5 μM FL3 for 24 h and stress granules labelled using an antibody against G3BP1 (red). The localisation of S6K1 and S6K2 were observed by immunofluorescence staining (green). Nuclei were stained with DAPI (blue). Merged images are shown. Yellow arrows indicate examples of stress granules. Scale bar = 25 μm. b HeLa cells were treated with either 0.5 mM of NaAsO2 for 30 min or 30 μM NaAsO2 for 1, 2, 4 or 6 h. Stress granule formation was assessed by immunofluorescence staining of G3BP1 (red) or TIA1 (green). c, d Quantification of the % of cells containing stress granules after treatment and the mean number of stress granules in the cells displaying stress granules. 100 cells were counted in each of the 3 biological repeats. Scale bar = 20 μm. e HeLa cells were treated with either 0.5 mM of NaAsO2 for 30 min or 30 μM NaAsO2 for 1, 2, 4 or 6 h and protein extracts immunoblotted for phosphorylated eIF2α (p-eIF2α), eIF2α and β-tubulin. Quantification was performed from three independent experiments. f HeLa cells were treated with 30 μM NaAsO2 for 1, 2, 4 or 6 h. In the last 5 min of treatment, cells were incubated with 5 μg/ml puromycin. Incorporation of puromycin into newly synthesised protein was assessed by immunoblotting. Band intensities for each lane were measured in the biological repeats and normalised against intensity of Coomassie Blue staining. For cf, error bars are s.e.m and the data were analysed using one-way Anova (*p < 0.04; **p < 0.0002). g HeLa cells were treated with 30 μM NaAsO2 for 2 h and stress granules were observed by immunofluorescence staining of G3BP1 (red). The distribution of S6K1 and S6K2 were analysed by immunostaining (green). Yellow arrows indicate examples of stress granules. Nuclei were stained with DAPI (blue). Scale bar = 25 μm
Fig. 2
Fig. 2
S6K1 and S6K2 are required for stress granule assembly. a HeLa cells were pre-treated or not with 0.5 mM LYS6K2 for 12 h and then treated with or without 30 μΜ of NaAsO2 for 2 h. Quantification of cells with G3BP1-positive granules under the indicated conditions is presented. 100 cells were analysed in each of the 3 biological repeats. Statistical analysis was carried out using one-way Anova (**p < 0.0002). b HeLa cells were transfected with siRNAs against S6K1 and S6K2 and the levels of knockdown were confirmed via immunoblotting of cell lysates. β-tubulin levels were analysed to ensure equal loading of lysates. A non-specific band in the S6K2 blot is indicated with a star (*). c–e Quantification of cells with stress granules after treatment with 30 μΜ of NaAsO2 for the indicated times and either siRNAs against S6K1 (S6K1-A and S6K1-B), S6K2 (S6K2-A and S6K2-B) or both S6K1 and S6K2 (S6K1-B and S6K2-B). 100 cells were analysed in each of the 4 biological repeats. Error bars are s.e.m. Data were analysed using two-way Anova (*p < 0.04; **p < 0.0002). f Mean number of granules per cell in those cells that displayed stress granules after 2 h treatment with 30 μΜ of NaAsO2. g, h Analysis of the size of stress granules following siRNA knockdown of S6K1 or S6K2 in cells treated for 2 h with 30 μΜ of NaAsO2. SGs were clustered into 3 groups according to their size: small, intermediate and large (see Materials and Methods) (g). Mean granule size was also measured (h). 100 cells were analysed in each of the 3 biological repeats. For f and h, error bars are s.e.m and data were analysed using one-way Anova (*p < 0.04; **p < 0.0002)
Fig. 3
Fig. 3
S6 kinases promote stress granule assembly and persistence dependent on their kinase activities. a–c Immunoblots of lysates from HeLa cells expressing HA-tagged S6K1p70 or S6K2p54 and corresponding kinase-inactive versions (KR). β-tubulin levels were analysed to ensure equal loading of lysates. Transfection with the parent vectors pRK7 and pCDNA3 was used as a control. Representative blots using antibodies against the HA-tag (a), S6K1 (b) and S6K2 (c) are shown. Comparison between the levels of ectopically expressed and endogenous S6K1 and S6K2 can be observed. The endogenous S6K1 (p85 and p70) can be observed in a longer exposure of the S6K1 blot (exp) and the endogenous S6K2 is labelled as p54. Levels of phosphorylated RPS6 (p-RPS6) and RPS6 are also shown. (d) HeLa cells expressing HA-S6K1p70 or HA-S6K2p54 were treated with 30 μM NaAsO2 for the indicated times and the % of cells with stress granules was quantified. e, f HeLa cells expressing HA-S6K1p70, HA-S6K2p54 or kinase-inactive mutants (KR) were treated with 30 μM NaAsO2 for 6 h. The % of cells with stress granules (e) and the mean number of granules in the cells that displayed stress granules were quantified (f). g HeLa cells expressing HA-tagged S6K1p70 or S6K2p54 were subjected to 30 μM NaAsO2 for 1 h and left to recover for the indicated times. The % of cells displaying stress granules was quantified. h, i HeLa cells expressing HA-S6K1p70, HA-S6K2p54 or kinase-inactive mutants (KR) were treated with 30 μM NaAsO2 for 1 h and left to recover for 1 h. The % of cells with stress granules (h) and the mean number of granules in the cells that displayed stress granules were quantified (i). All quantifications are from 100 cells per condition in each of the 3 biological repeats. Error bars are s.e.m. For d and g, statistical analysis was performed using two-way Anova and for e, f, h and i by one-way Anova (ns = not significant; *p < 0.04; **p < 0.0002)
Fig. 4
Fig. 4
The mTORC1-S6K signalling pathway is required for stress granule assembly. a HeLa cells were pre-treated with 50 nM Rapamycin for 2 h prior to exposure to 30 μM NaAsO2 for a further 2 h. Cells were immunostained for G3BP1 (red) and the phosphorylated form of RPS6 (p-RPS6) (green). Nuclei were stained with DAPI (blue). Yellow arrows indicate examples of stress granules. Scale bar = 25 μm. b Quantification of cells forming stress granules under the indicated conditions. c HeLa cells were subjected to siRNA directed against RAPTOR. Immunoblots of cell extracts with antibodies to RAPTOR, phosphorylated RPS6 (p-RPS6), RPS6 and β-tubulin are shown. d Images of cells treated with or without 30 μM NaAsO2 for 2 h and subjected to RAPTOR siRNA. Cells were immunostained for G3BP1 (red). Nuclei were stained with DAPI (blue). Scale bar = 50 μm. Quantification of cells forming stress granules under the indicated conditions is presented. e–g Cells expressing S6K1p70 or S6K2p54 were pre-treated with DMSO or rapamycin for 2 h and then subjected to either 30 μM NaAsO2 for a further 6 h (f) or 30 μM NaAsO2 for 1 h and left to recover for 1 h (g). For all quantifications, 100 cells per condition were counted in each of the 3 biological repeats. Error bars are s.e.m. Data were analysed using one-way Anova (**p < 0.0002)
Fig. 5
Fig. 5
S6 kinases promote the phosphorylation of eIF2α and translation inhibition. a Immunoblotting of HeLa cell lysates for Ser51 phosphorylation on eIF2α (p-eIF2α) after treating cells with or without 30 μM NaAsO2 for 1 h in the presence of either DMSO or 50 nM rapamycin. Levels of phosphorylated RPS6 (p-RPS6) and S6 kinases (p-S6K) are shown to indicate the effectiveness of rapamycin at inhibiting mTORC1 activity. Blots for total eIF2α, RPS6, S6K1 and S6K2 protein levels are also presented. Quantification of p-eIF2α band intensities from 4 independent experiments is shown. The p-eIF2α levels were normalised against β-tubulin levels and are presented as relative increase in phosphorylation between untreated samples and those treated with NaAsO2. b HeLa cells were treated with 30 μΜ NaAsO2 for 30 min and the incorporation of puromycin into newly synthesised protein was assessed by immunoblotting. Band intensities for each lane were measured in 4 independent experiments and normalised against the intensity of Coomassie blue staining and presented as % translation inhibition. c HeLa cells transfected with siRNAs against S6K1 or S6K2 were treated with either 30 μΜ NaAsO2 for 1 h or 0.5 mM NaAsO2 for 30 min. Protein extracts were immunoblotted for p-eIF2α and total eIF2α protein levels. Quantification of p-eIF2α band intensities from 5 independent experiments is shown. The p-eIF2α levels were normalised against β-tubulin levels. d, e HeLa cells transfected with siRNAs against S6K1 or S6K2 were treated with either 30 μΜ NaAsO2 for 1 h or 0.5 mM NaAsO2 for 30 min and the incorporation of puromycin into newly synthesised protein was assessed by immunoblotting. Band intensities for each lane were measured in 3 independent experiments and normalised against the intensity of Coomassie blue staining and presented as % translation inhibition (e). f, g HeLa cells transfected with empty plasmids or plasmids expressing HA-S6K1p70 or HA-S6K2p54 were assessed for the incorporation of puromycin into newly synthesised protein by immunoblotting. The expression of the S6 kinases increased the levels of phosphorylated RPS6 (p-RPS6). Quantification of puromycin incorporation normalised against the intensity of Coomassie blue staining in 3 independent experiments is presented (g). Error bars are s.e.m and the data analysed using one-way Anova (*p < 0.04; **p < 0.0002)
Fig. 6
Fig. 6
The C. elegans S6 kinase orthologue, RSKS-1, promotes stress granule formation in vivo. a Worms expressing a pharyngeal Venus::TIAR-2 reporter were fed rsks-1 RNAi and subjected to heat shock at 35 °C for 3 h. Images of the pharynx of worms are shown for the indicated conditions. Left-hand panels show reporter expression and the right-hand panels are corresponding bright field images. Scale bar = 25 μM. b Quantification of the number of TIAR-2 positive granules separated by size, either small (0.01–0.5 μm2) or large (0.5–2.5 μm2). 30 worms were analysed per condition in 3 biological repeats and data analysed by two-way Anova (ns = not significant; *p < 0.04). c Worm survival assay. N2 or rsks-1(ok1255) mutant worms were subjected to heat shock at 35 °C and the number of worms alive at the indicated times was scored. At least 30 worms were assessed in each of the 3 biological repeats and data were analysed using two-way Anova (*p < 0.04; **p < 0.0002). d Epistasis experiment to determine the relationship between rsks-1 and gtbp-1. N2 and rsks-1(ok1255) worms were fed with gtbp-1(RNAi) and subjected to heat stress (35 °C for 6 h). Worm survival was recorded. The smo-1 RNAi was used as a negative control as it does not suppress stress granule assembly. 30 worms were analysed per condition in 3 biological repeats and data were analysed using one-way Anova (ns = not significant; **p < 0.0002)
Fig. 7
Fig. 7
Schematic of the role of the mTORC1-S6 kinase pathway in stress granule assembly and maintenance. Distinct types of stress granule (SG) are formed depending on the level of oxidative stress. In response to acute oxidative stress, mTORC1 components are sequestered into solid SGs thus inhibiting mTORC1 activity [22, 23]. In response to mild oxidative stress, mTORC1 activates S6 kinases leading to inhibition of translation and the assembly of SGs that contain a solid core surrounded by a liquid shell. Both S6K1 and S6K2 can accumulate in SGs and S6K2 helps maintain SGs via an unknown mechanism. On recovery from the stress, the SGs dissolve and protein translation resumes

Similar articles

See all similar articles

Cited by 5 articles

References

    1. Kultz D. Molecular and evolutionary basis of the cellular stress response. Annu Rev Physiol. 2005;67:225–57. doi: 10.1146/annurev.physiol.67.040403.103635. - DOI - PubMed
    1. Spriggs KA, Bushell M, Willis AE. Translational regulation of gene expression during conditions of cell stress. Mol Cell. 2010;40:228–37. doi: 10.1016/j.molcel.2010.09.028. - DOI - PubMed
    1. Buchan JR, Parker R. Eukaryotic stress granules: the ins and outs of translation. Mol Cell. 2009;36:932–41. doi: 10.1016/j.molcel.2009.11.020. - DOI - PMC - PubMed
    1. Kedersha N, Ivanov P, Anderson P. Stress granules and cell signaling: more than just a passing phase. Trends Biochem Sci. 2013;38:494–506. doi: 10.1016/j.tibs.2013.07.004. - DOI - PMC - PubMed
    1. Aulas A, Fay MM, Lyons SM, Achorn CA, Kedersha N, Anderson P, et al. Stress-specific differences in assembly and composition of stress granules and related foci. J Cell Sci. 2017;130:927–37. doi: 10.1242/jcs.199240. - DOI - PMC - PubMed

Publication types

MeSH terms

Feedback