Skip to main page content
U.S. flag
See More

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
. 2015 Sep 18:6:8261.
doi: 10.1038/ncomms9261.

eIF6 coordinates insulin sensitivity and lipid metabolism by coupling translation to transcription

Daniela Brina  1 , Annarita Miluzio  1 , Sara Ricciardi  1 , Kim Clarke  2 , Peter K Davidsen  2 , Gabriella Viero  3   4 , Toma Tebaldi  4 , Nina Offenhäuser  5 , Jan Rozman  6 , Birgit Rathkolb  6   7 , Susanne Neschen  6 , Martin Klingenspor  8 , Eckhard Wolf  7 , Valerie Gailus-Durner  6 , Helmut Fuchs  6 , Martin Hrabe de Angelis  6   9   10 , Alessandro Quattrone  4 , Francesco Falciani  2 , Stefano Biffo  1   11
Affiliations

eIF6 coordinates insulin sensitivity and lipid metabolism by coupling translation to transcription

Daniela Brina et al. Nat Commun. .

Abstract

Insulin regulates glycaemia, lipogenesis and increases mRNA translation. Cells with reduced eukaryotic initiation factor 6 (eIF6) do not increase translation in response to insulin. The role of insulin-regulated translation is unknown. Here we show that reduction of insulin-regulated translation in mice heterozygous for eIF6 results in normal glycaemia, but less blood cholesterol and triglycerides. eIF6 controls fatty acid synthesis and glycolysis in a cell autonomous fashion. eIF6 acts by exerting translational control of adipogenic transcription factors like C/EBPβ, C/EBPδ and ATF4 that have G/C rich or uORF sequences in their 5' UTR. The outcome of the translational activation by eIF6 is a reshaping of gene expression with increased levels of lipogenic and glycolytic enzymes. Finally, eIF6 levels modulate histone acetylation and amounts of rate-limiting fatty acid synthase (Fasn) mRNA. Since obesity, type 2 diabetes, and cancer require a Fasn-driven lipogenic state, we propose that eIF6 could be a therapeutic target for these diseases.

PubMed Disclaimer

Figures

Figure 1
Figure 1. eIF6 depletion impairs polysome formation in the liver after feeding, reduces lipidemia, protects from HFD and liver steatosis.
All graphs, black, wt; red, eIF6 het mice (a) Representative liver polysomal profiles of wt (black) and eIF6 het (red, +/−) show reduced polysomes in het after overnight fasting and 4 h refeeding. Experiment repeated at least five times. (b) GTT is normal in eIF6 het (red) compared with wt. N=7. (c,d) Cholesterol and triglycerides levels are reduced in blood of eIF6 het compared with wt ones (N=10). (e) Liver triglyceride content is reduced in eIF6 het mice (N=4). (f) ATP liver content is reduced in eIF6 het. N=3. (g) eIF6 het mice are protected from HFD: body weight gain is reduced in eIF6 het mice compared with wt. N=10. Top bars show absolute body weight gain. Lower graph, body weight. (h) GTT eIF6 het mice have lower insulinaemia at 2 h after glucose injection. N=10. (i) ITT (AUC ITT, area under curve) shows partial amelioration of diet-induced insulin resistance. N=5 (j,k). Increased liver-X-ray attenuation (N=8) indicating reduced fat deposition in the liver (j) confirmed by autoptic liver triglyceride content on separate animals (N=5) (k). (i) In humans, eIF6 levels inversely correlate with insulin sensitivity, as measured by HOMA-IR. In all panels, data are represented as mean±s.d. Statistical P values were calculated by two-tailed t-test (*P value ≤0.05).
Figure 2
Figure 2. eIF6 activity cell autonomously controls fatty acid synthesis, glycolysis and ATP levels.
(a) Outline of the analysis. Primary hepatocytes from mice were isolated and assayed as specified. Experiments performed on biological replicates. All data are expressed as percentage of controls (wt), analysing primary cells from littermate couples of mice. (b) De novo lipogenesis, measured by labelling with D-[6-14C]-glucose and subsequent fatty acid analysis, is reduced in cells from eIF6 het mice. Cells were kept in 100 nM insulin and 20 mM glucose. N=6 (c) Lactate secretion, an index of glycolysis flux, is reduced in hepatocytes from eIF6 het mice. Basal values are around 30 pm μg−1 proteins per hour. N=6. (d) Fatty acid oxidation is not significantly affected (basal values of control around 1,000 c.p.m. μg−1 proteins). N=4 (e–g) ATP content depends from eIF6 levels: (e) ATP decreased in eIF6 het mice compared with wt ones. (f) Acute depletion of eIF6 leads to a reduction of ATP levels in eIF6+/+ cells, whereas (g) restoration of eIF6 levels leads to an increase in ATP in eIF6+/− cells. (h) Lactate secretion as in (f) after eIF6 shRNA in AML12 cells. (i) ATP as in (d), after eIF6 shRNA in AML12 cells. In all panels, data are represented as mean±s.d. Statistical P values were calculated by two-tailed t-test (*P value ≤0.05, ***P≤0.001). (e–i), N=3.
Figure 3
Figure 3. eIF6 translational activity controls the expression of lipogenic transcription factors.
(a) Global initiation is reduced on eIF6 inhibition and insulin administration, as shown by polysome/80S ratio. Data are represented as mean±s.d. Statistical P values were calculated by two-tailed t-test (n=4. ***P value ≤0.001). (b) Outline of the analysis and quantitative results: a subset of mRNA is differentially depleted (green) from eIF6 shRNA polysomes. The normalized ratio was obtained by the quantile method. Correction taking in account the reduction of polysomal area shows that more genes are affected by eIF6 deficiency. (c) Gene Ontology analysis shows that metabolism predominates in eIF6-affected genes, from the normalized ratio pool (P value ≤10−12). (d) Structural 5′-UTR analysis of genes affected by eIF6 deficiency. Polysomes from eIF6-depleted cells show reduction of G/C rich 5′ mRNAs (green box) (e) Transcription factors involved in lipogenesis are depleted from polysomes upon eIF6 shRNA. Structural features of their 5′ UTR are included. (f) C/EBPδ protein expression is modulated by eIF6. eIF6 increase, by lentiviral administration, induces C/EBPδ. eIF6 depletion, by lentiviral-mediated shRNA, reduces C/EBPδ. Representative experiments done in triplicates on AML12 hepatocytes and reconstituted mesenchymal stem cells (g) Lipogenic transcription factor C/EBPβ (LIP isoform) is reduced in eIF6-depleted cells. Left, mRNA structure of C/EBPβ generating three isoforms by differential translation (arrows: translation start sites). Blue small box with blue arrows is the C/EBP uORF. Right panel show a blot with LAP, LIP and C/EBPβ isoforms: LIP isoform is specifically downregulated. (Representative experiment done in triplicates). (h) ATF4 reporter translation depends from eIF6. Reporter assay of reinitiation with luciferase (blue box) cloned downstream of natural ATF4 uORF. eIF6 upregulation increases ATF4 translational reporter activity, eIF6 downregulation decreases it. Data are normalized on translation of a cap-dependent firefly control. The large black arrow represents the stop codon of the second uORF. Small arrows are start codons. Blue box is the reporter. Statistical P values were calculated by two-tailed t-test, as above (*P value ≤0.05; **P≤0.01; ***P≤0.001).
Figure 4
Figure 4. eIF6 depletion induces a specific transcriptional signature in insulin-responsive tissues.
(a) Microarray analysis on adult livers show extensive remodelling of gene expression in eIF6 het mice. Co-ordinated transcriptional changes in eIF6 het livers identify a profound metabolic reprogramming, at the steady-state level, consistent with reduction of lipid biosynthesis, and additional effects on chromatin remodelling and cell cycle. (b) Microarray analysis on multiple organs. Heat map analysis of liver, fat and brain shows a common signature of inhibition of lipid biosynthetic processes in fat and liver but not in the brain. The scale represents statistical significance of the enrichment of Gene Ontology terms obtained by using a modified Fisher's exact test (EASE score). This produces a P value, which is then corrected for multiple testing using the FDR (Benjamini). 1.3=P value ≤0.05, 2=P value ≤0.01, 3=P value ≤0.001, 5=P value ≤0.00001 (c) Outlines of genes affected by eIF6 levels in the fatty acid synthesis, glycolysis and cholesterol synthesis pathway. All blue-labelled genes are downregulated at the transcriptional level. (d) Real-time analysis of selected targets validates all target genes (liver tissues). Here we show representative biological triplicates of experiments performed on separate cohorts of animals, (five for eIF6 and Fasn and two for the others) (data shown for n=3; mean±s.d.). (e) Consistent changes are also found at the protein level. Representative blots of Hmgcr, Fasn (downregulated at the mRNA level) and of Pgc-1 α (upregulated at the mRNA level), done in triplicates. See also Supplementary Fig. 4. (n=3).
Figure 5
Figure 5. Fasn levels are controlled by eIF6 activity.
(a) Rescue strategy in hepatocytes and primary stem cells. Cells were transduced with either eIF6 shRNA or wt eIF6, eIF6 protein levels were checked and Fasn mRNA and protein were analysed. (b) Downregulation of eIF6 by an inducible shRNA in AML12 hepatocytes leads to reduction of Fasn mRNA. (c–d) eIF6 has a direct effect on Fasn mRNA levels in mesenchymal stem cells differentiated into adipocytes. wt or het stem cells were either transduced with constitutive eIF6 shRNA (c) or wt eIF6 (d) and assayed for Fasn expression. A direct correlation between eIF6 levels and Fasn levels is observed. (e) Fasn protein is controlled by eIF6 levels in AML12 and EMSC adipocytes. Experiments on EMSC show both downregulation of eIF6 in wt cells or upregulation in het cells. Note eIF6 dependent levels of both Fasn and LIP isoform. In all panels, data are represented as mean±s.d. Panel e shows the action of two independent eIF6 shRNA, constitutive (right) or inducible. Statistical P values were calculated by two-tailed t-test (*P value ≤0.05, **P≤0.01, ***P≤0.001). Real-time experiments and western blotting were performed on biological triplicates of at least two independent experiments. One triplicate is shown for real time.
Figure 6
Figure 6. eIF6 targeting leads to Fasn reduction and histone acetylation.
(a) Strategy through the connectivity map: drugs that generate a signature similar to eIF6 are identified and tested. (b) The connectivity map identifies two translational inhibitors, puromycin and ciclopirox and two HDAC inhibitors, TSA and MS-275. (c–d) HDAC inhibitors downregulate Fasn; HDAC inhibitor TSA, but not MS-275, reduces eIF6 mRNA (c), but both HDAC inhibitors downregulate Fasn mRNA (d). (e) Consistent with the Gene Ontology analysis (Fig. 4a) and the connectivity map, eIF6 depletion increases H3K9 acetylation. Western blot analysis of livers at indicated ages or AML12 cells. Representative blots with densitometry are shown. (f) Strategy through inhibition of eIF6 activation: mutation of eIF6 S235A is tested. S235A mutant is unable to restore translation following insulin stimulation. (g–h) eIF6 reintroduction in EMSC cells induces Fasn. (g) Real-time analysis of restored wt and mutant eIF6 levels. (h) Analysis in the same cells of Fasn mRNA. wt eIF6 but not eIF6 S235A mutant, recovers Fasn mRNA levels. In all panels, data are represented as mean±s.d. Statistical P values were calculated by two-tailed t-test (n=3). Repeated at least twice. **P value ≤0.01, ***P≤0.001 (i) Model based on data of Figs 1, 2, 3, 4, 5, 6. Insulin induces translation. eIF6 favours translation of uORF-containing and G/C rich mRNAs encoding for lipogenic transcription factors. The net result is activation of the lipid biosynthetic pathway. eIF6 inhibition blocks lipogenesis and reverts insulin resistance.
See this image and copyright information in PMC

References

    1. Wilcox G. Insulin and insulin resistance. Clin. Biochem. Rev. 26, 19–39 (2005). - PMC - PubMed
    1. Clark J. M. The epidemiology of nonalcoholic fatty liver disease in adults. J. Clin. Gastroenterol. 40, S5–10 (2006). - PubMed
    1. Marchesini G. et al.. Association of nonalcoholic fatty liver disease with insulin resistance. Am. J. Med. 107, 450–455 (1999). - PubMed
    1. Saltiel A. R. & Kahn C. R. Insulin signalling and the regulation of glucose and lipid metabolism. Nature 414, 799–806 (2001). - PubMed
    1. Proud C. G. Regulation of protein synthesis by insulin. Biochem. Soc. Trans. 34, 213–216 (2006). - PubMed

Publication types

MeSH terms

Associated data