Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2024 May 14;15(1):4083.
doi: 10.1038/s41467-024-48386-y.

mTORC1 regulates cell survival under glucose starvation through 4EBP1/2-mediated translational reprogramming of fatty acid metabolism

Tal Levy #  1   2 , Kai Voeltzke #  3 , Laura Hruby #  3 , Khawla Alasad #  1   2 , Zuelal Bas  3 , Marteinn Snaebjörnsson  4   5 , Ran Marciano  1   2 , Katerina Scharov  3   6 , Mélanie Planque  7   8 , Kim Vriens  7   8 , Stefan Christen  7   8 , Cornelius M Funk  9   10 , Christina Hassiepen  3 , Alisa Kahler  3 , Beate Heider  3 , Daniel Picard  3   6   11 , Jonathan K M Lim  3 , Anja Stefanski  12 , Katja Bendrin  13 , Andres Vargas-Toscano  14   15   16 , Ulf D Kahlert  17 , Kai Stühler  12 , Marc Remke  3   6   11 , Moshe Elkabets  18   19 , Thomas G P Grünewald  9   10   20 , Andreas S Reichert  13 , Sarah-Maria Fendt  7   8 , Almut Schulze  4   5 , Guido Reifenberger  3   11 , Barak Rotblat  21   22 , Gabriel Leprivier  23
Affiliations

mTORC1 regulates cell survival under glucose starvation through 4EBP1/2-mediated translational reprogramming of fatty acid metabolism

Tal Levy et al. Nat Commun. .

Abstract

Energetic stress compels cells to evolve adaptive mechanisms to adjust their metabolism. Inhibition of mTOR kinase complex 1 (mTORC1) is essential for cell survival during glucose starvation. How mTORC1 controls cell viability during glucose starvation is not well understood. Here we show that the mTORC1 effectors eukaryotic initiation factor 4E binding proteins 1/2 (4EBP1/2) confer protection to mammalian cells and budding yeast under glucose starvation. Mechanistically, 4EBP1/2 promote NADPH homeostasis by preventing NADPH-consuming fatty acid synthesis via translational repression of Acetyl-CoA Carboxylase 1 (ACC1), thereby mitigating oxidative stress. This has important relevance for cancer, as oncogene-transformed cells and glioma cells exploit the 4EBP1/2 regulation of ACC1 expression and redox balance to combat energetic stress, thereby supporting transformation and tumorigenicity in vitro and in vivo. Clinically, high EIF4EBP1 expression is associated with poor outcomes in several cancer types. Our data reveal that the mTORC1-4EBP1/2 axis provokes a metabolic switch essential for survival during glucose starvation which is exploited by transformed and tumor cells.

PubMed Disclaimer

Conflict of interest statement

B.R. and G.L. filed for a patent based on these findings. The remaining authors declare no competing interests.

Figures

Fig. 1
Fig. 1. The mTORC1 substrates 4EBP1/2 prevent cell death in response to glucose starvation in human, mouse and yeast cells.
A, B The indicated cell lines were grown in glucose starved medium (Glc strv) and treated with vehicle (V), Rapamycin (Rap), KU-0063794 (KU), cycloheximide (CHX) or 4EGI for 48 h. Cell death was measured by PI staining and flow cytometry. C Scheme of the mTORC1 downstream signaling pathways controlling mRNA translation initiation and the impact of the inhibitors used in (A, B). D The indicated cell lines were grown in complete media or starved for glucose (Glc) for 48 h. Cell death was analyzed as in (A, B). The level of the indicated proteins was analyzed by immunoblotting. E The indicated cell lines were grown in glucose starved medium (Glc strv) for 48 h. Cell death and protein levels were analyzed as in (A, B) and (D) respectively. F The indicated cell lines were grown in complete media or starved for glucose (Glc) for 48 h. Cell death and protein levels were analyzed as in (A, B) and (C) respectively. G The indicated cell lines were glucose (Glc) starved for 48 h. Med8a and HD-MB03 cells were treated with 1 μg/ml doxycycline for 72 h. Cell death was measured as in (A, B). H 4E KO MEF were transfected with control siRNA (scr) or siRNAs targeting Eif4e and grown in glucose starved medium (Glc strv) for 48 h. Cell death and protein levels were analyzed as in (A, B) and (D) respectively. I WT, caf20Δ, eap1Δ, or eap1Δ/caf20Δ S. cerevisiae strains were plated by serial dilution on solid complex medium with (YPD) or without (YP) 2% glucose at 37 °C. J WT or eap1Δ strains were grown in liquid medium containing no glucose (YP) for 2 weeks at 30 °C and were plated by serial dilutions onto complete YPD agar plates. Data are shown as the mean ± SD. Statistics: unpaired one-sided Student’s t test (A, B, D, E, F, G, H); n = 3 independent experiments for (A, B, D, E, F, G, H). Source data are provided as a Source Data file.
Fig. 2
Fig. 2. 4EBP1/2 maintain antioxidant power and preserve the redox balance under glucose starvation.
A Control (shScr) and sh4EBP1/2 HEK293 cells were grown in complete medium or glucose starved (Glc strv) for the indicated times and labeled with azidohomoalanine (AHA). Levels of AHA-labelled proteins were detected by immunoblotting. B WT and 4E KO MEF were grown in complete medium or glucose starved (Glc strv) for the indicated times, labeled with EdU and analyzed by flow cytometry. C WT and 4E KO MEF were grown in complete medium or glucose (Glc) starved for 24 h. Mitochondrial membrane potential (Δψm) with TMRE staining and mitochondrial mass were measured. D Control (shScr) and sh4EBP1/2 HEK293 cells were grown in complete medium or glucose (Glc) starved for 24 h and the corresponding metabolites were measured by LC-MS. E NAD+ and NADH levels measured in (C) were plotted as NADH/NAD+ ratio. F NADP+ and NADPH levels measured in (C) were plotted as NADPH/NADP+ ratio. G WT and 4E KO MEF were grown in complete medium or glucose (Glc) starved for 24 h, and NADP+ and NADPH levels were measured. H Scheme of the usage of NADPH in recycling oxidized glutathione for H2O2 detoxification. (I) WT and 4E KO MEF were grown as in (C), and reduced and total glutathione were measured and expressed as the ratio of reduced (GSH) to oxidized (GSSG) glutathione. J WT and 4E KO MEF were grown as in (C), and total glutathione was measured. K WT and 4E KO MEF grown as in (C) were labelled with CM-DCFDA and analyzed by flow cytometry. L 4E KO MEF were grown in glucose starved medium (Glc strv) and treated with vehicle (V), N-acetyl cysteine (NAC) or Catalase (CAT) for 48 h. Cell death was measured by PI staining and flow cytometry. Data are shown as the mean ± SD. Statistics: unpaired one-sided Student’s t test (A–C, E–G, I–L); n = 3 independent experiments for (A–C, E–G, I–L). Source data are provided as a Source Data file.
Fig. 3
Fig. 3. 4EBP1/2 control fatty acid synthesis activity in response to glucose starvation to preserve redox balance and protect cells.
A Control (shScr) and sh4EBP1/2 HEK293 cells were grown in complete medium or glucose (Glc) starved for 30 h and proteomics was analyzed by MS. Differentially expressed proteins in shScr versus sh4EBP1/2 HEK293 cells under glucose starvation (p-value < 0.05) correspond to yellow and blue dots. Proteins involved in NADPH producing or NADPH consuming processes are highlighted. B Scheme of the fatty acid synthesis pathway highlighting the enzymatic steps and consumption of NAPDH. ACLY: ATP citrate lyase, ACC1/2: acetyl-CoA carboxylase 1/2, FASN: fatty acid synthase. C, D The indicated cell lines were grown in glucose starved medium (Glc strv) with or without TOFA for 48 h. Cell death was measured by PI staining and flow cytometry. E, F The indicated cell lines were transfected with control siRNA (scr) or siRNAs targeting FASN and glucose starved (Glc strv) for 48 h. Cell death was analyzed as in (C, D). FASN protein levels were analyzed by immunoblotting. G Scheme of the [14C] acetate labeling assay to measure fatty acid synthesis activity. ACS: acetyl-CoA synthetase, ACC: acetyl-CoA carboxylase, FASN: fatty acid synthase. H, I The indicated cell lines grown in complete medium or glucose (Glc) starved for 24 h were labeled with [14C] acetate in the last 18 h. [14C] was measured in the lipid fraction and normalized to total protein levels. In (H), n = 10 biological replicates, and in (I), n = 4 biological replicates. (J, K) The indicated cell lines were glucose starved (Glc strv) for 24 h with or without TOFA, and NADP+ and NADPH levels were measured. (L, M) The indicated cell lines were glucose starved (Glc strv) for 24 h with or without TOFA were labelled with CM-DCFDA and analyzed by flow cytometry. Data are shown as the mean ± SD. Statistics: unpaired one-sided Student’s t test (C–F, J–M), two way ANOVA (H, I); n = 3 independent experiments for (C–F, J–M). Source data are provided as a Source Data file.
Fig. 4
Fig. 4. 4EBP1/2 repress ACACA translation under glucose starvation.
A, B WT and 4E KO MEF (A), or shScr and sh4EBP1/2 HEK293 cells (B) were grown in complete medium or glucose starved (Glc strv) for the indicated times, and analyzed by immunoblotting using antibodies against the indicated proteins. Representative results of two independent experiments are shown. C ShScr and sh4EBP1/2 HEK293 cells were grown in complete medium or glucose (Glc) starved for 16 h, and ACACA mRNA expression was analyzed by qRT-PCR. D ShScr and sh4EBP1/2 HEK293 cells were grown in complete medium or glucose (Glc) starved for 6 h, and translation efficiency (TE) of ACACA mRNA was calculated by measuring the levels of polysomal and total ACACA mRNA by qRT-PCR. n = 4 independent experiments. E, F WT and 4E KO MEF (E), or shScr and sh4EBP1/2 HEK293 cells (F) were transfected with an ACACA 5’UTR-containing Firefly Luciferase construct and a control Renilla Luciferase vector. Cells were grown in complete medium or glucose (Glc) starved for 6 h, and luminescence was measured. G HEK293 cells were transfected with an HA-tagged ACC1 expressing vector containing or not the ACACA 5’UTR. Cells were grown in complete medium or glucose starved (Glc strv) for the indicated times, and analyzed by immunoblotting using antibodies against the indicated proteins. Representative results of three independent experiments are shown. H, I 4E KO MEF (H) or sh4EBP1/2 HEK293 cells (I) were transfected with control siRNA (scr) or siRNAs targeting ACACA and grown in glucose starved medium (Glc strv) for 48 h. Cell death was measured by PI staining and flow cytometry. ACC1 protein levels were analyzed by immunoblotting. Data are shown as the mean ± SD. Statistics: unpaired one-sided Student’s t test (C–F, H, I); n = 3 independent experiments for (C, E, F, H, I). Source data are provided as a Source Data file.
Fig. 5
Fig. 5. 4EBP1 supports oncogenic transformation in vitro and in vivo.
A, B WT and 4EBP1/4EBP2 DKO (4E KO) NT2197 (A), or empty vector (EV) and 4EBP1AA expressing HeLa cells (B) were grown in soft agar for 21 days. Colonies and single cells were counted, and colony formation efficiency was calculated and normalized to respective control. Protein expression of 4EBP1 and 4EBP2 was analyzed by immunoblotting. C 4E KO NT2197 cells were grown in soft agar for 21 days and treated with DMSO, NAC, CAT, TROLOX or TOFA. Colonies and single cells were counted, and colony formation efficiency was calculated and normalized to DMSO. D Control (sgCtrl) and ACC1 targeting CRISPRi (sgAcaca) 4E KO NT2197 cells were grown in soft agar for 21 days. Colony formation efficiency and proteins level were analyzed as in (A, B). n = 3 independent experiments for (A–D). E, F WT or 4E KO NT2197 cells were injected in the mammary fat pad of NOD SCID gamma mice. Tumors were harvested, photographed (E) and weighed (F). n = 12 mice per cell line. G, H EV, 4EBP1AA or 4EBP1AA, YL expressing HeLa cells were injected in the flank of NOD SCID gamma mice. Tumors were harvested, photographed (G) and weighed (H). n = 6–8 mice per cell line. I, J Control (shGFP) or stable ACC1 knock down (shAcaca) 4E KO NT2197 cells were injected in the flank of NOD SCID gamma mice. Tumors were harvested, photographed (I) and weighed (J). n = 8–10 mice per cell line. Data are shown as the mean ± SD. Statistics: unpaired one-sided Student’s t test (A–D), two way ANOVA (F, H, J). Source data are provided as a Source Data file.
Fig. 6
Fig. 6. 4EBP1 has clinical relevance in glioma and promotes glioma tumorigenesis.
A Kaplan-Meier survival estimates of glioblastoma patients stratified by their EIF4EBP1 mRNA levels (cut off first quartile) in the CGGA cohort. p-values were calculated using a log rank test. B Expression levels of EIF4EBP1 per glioma grade in the CGGA cohort. Data were shown as boxplots with medians, interquartile ranges and lower/upper whiskers in. p-values were calculated using an unpaired and two-tailed parametric t test. C 4EBP1 protein levels plotted against the expression levels of ACC1 protein using CPTAC and TCGA GBM proteomic data. Co-expression level was quantified by calculating the Spearman’s correlation coefficient. D, E The indicated cell lines were grown in soft agar for 21 days and were treated (E) or not (D) with the indicated compounds. n = 3 independent experiments. 4EBP1 protein levels were analyzed by immunoblotting (D). p-values were calculated using an unpaired and one-tailed Student’s t test. F The indicated cell lines were injected in the flank of NOD SCID gamma mice. Tumors were harvested, photographed and weighed. n = 11–12 mice per cell line. p-values were calculated using two way ANOVA. G The indicated cell lines were injected in the flank of NOD SCID gamma mice. When tumors reached 100 mm3, mice were given doxycycline (DOX) or vehicle. Tumor volumes were measured at the indicated times. n = 12 mice per cell line. p-values were calculated using two way ANOVA. H ShScr (n = 7 mice) or sh4EBP1#1 (n = 8 mice) U-87 MG cells were injected intracranially in NOD SCID gamma mice. Survival of mice was monitored post injection. I ShScr (n = 10 mice) or sh4EBP1#1 (n = 10 mice) GL-261 cells were injected intracranially in C57WT mice. Survival of mice was monitored post injection. J sh4EBP1#1 containing shGFP (n = 9 mice) or shAcaca (n = 8 mice) GL-261 cells were injected intracranially in C57WT mice. Survival of mice was monitored post injection. p value was calculated using a log rank test for (H–J). Data are shown as the mean ± SD. Source data are provided as a Source Data file.
See this image and copyright information in PMC

Similar articles

See all similar articles

References

    1. Gullino PM, Clark SH, Grantham FH. The Interstitial Fluid of Solid Tumors. Cancer Res. 1964;24:780–794. - PubMed
    1. Nagy JA, Chang SH, Dvorak AM, Dvorak HF. Why are tumour blood vessels abnormal and why is it important to know? Br. J. Cancer. 2009;100:865–869. doi: 10.1038/sj.bjc.6604929. - DOI - PMC - PubMed
    1. Sullivan MR, et al. Quantification of microenvironmental metabolites in murine cancers reveals determinants of tumor nutrient availability. Elife. 2019;8:e44235. doi: 10.7554/eLife.44235. - DOI - PMC - PubMed
    1. Caro-Maldonado A. Dying for Something to Eat: How Cells Respond to Starvation. Open Cell Signaling J. 2011;3:42–51. doi: 10.2174/1876390101103010042. - DOI
    1. Lin SC, Hardie DG. AMPK: Sensing Glucose as well as Cellular Energy Status. Cell Metab. 2018;27:299–313. doi: 10.1016/j.cmet.2017.10.009. - DOI - PubMed

MeSH terms

LinkOut - more resources

-