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
. 2020 May 4;39(9):e102731.
doi: 10.15252/embj.2019102731. Epub 2020 Mar 9.

CLUH granules coordinate translation of mitochondrial proteins with mTORC1 signaling and mitophagy

David Pla-Martín #  1   2 , Désirée Schatton #  1   3 , Janica L Wiederstein  1   3 , Marie-Charlotte Marx  1   3 , Salim Khiati  4 , Marcus Krüger  1   3   5 , Elena I Rugarli  1   3   5
Affiliations

CLUH granules coordinate translation of mitochondrial proteins with mTORC1 signaling and mitophagy

David Pla-Martín et al. EMBO J. .

Abstract

Mitochondria house anabolic and catabolic processes that must be balanced and adjusted to meet cellular demands. The RNA-binding protein CLUH (clustered mitochondria homolog) binds mRNAs of nuclear-encoded mitochondrial proteins and is highly expressed in the liver, where it regulates metabolic plasticity. Here, we show that in primary hepatocytes, CLUH coalesces in specific ribonucleoprotein particles that define the translational fate of target mRNAs, such as Pcx, Hadha, and Hmgcs2, to match nutrient availability. Moreover, CLUH granules play signaling roles, by recruiting mTOR kinase and the RNA-binding proteins G3BP1 and G3BP2. Upon starvation, CLUH regulates translation of Hmgcs2, involved in ketogenesis, inhibits mTORC1 activation and mitochondrial anabolic pathways, and promotes mitochondrial turnover, thus allowing efficient reprograming of metabolic function. In the absence of CLUH, a mitophagy block causes mitochondrial clustering that is rescued by rapamycin treatment or depletion of G3BP1 and G3BP2. Our data demonstrate that metabolic adaptation of liver mitochondria to nutrient availability depends on a compartmentalized CLUH-dependent post-transcriptional mechanism that controls both mTORC1 and G3BP signaling and ensures survival.

Keywords: CLUH; G3BP; RNA metabolism; mTORC1; mitochondria.

PubMed Disclaimer

Conflict of interest statement

The authors declare that they have no conflict of interest.

Figures

Figure EV1
Figure EV1. CLUH forms granules upon starvation
  1. A

    Confocal images of liver cryosections and primary hepatocytes of Li‐Cluh WT and Li‐Cluh KO mice stained with anti‐CLUH antibody. Scale bar, 10 μm.

  2. B

    Confocal images of liver cryosections of fed and starved Li‐Cluh WT and Li‐Cluh KO mice stained with anti‐CLUH antibody. Right panels show 6.5× magnified boxed areas. Scale bar, 10 μm.

  3. C

    Confocal images of primary hepatocytes cultured in indicated media and stained with anti‐CLUH antibody. Right panels show 4.5× magnified boxed areas. Scale bar, 10 μm.

  4. D, E

    Confocal images of primary hepatocytes cultured in indicated media and stained with (D) anti‐G3BP1 or (E) anti‐DCP1A and anti‐CLUH antibodies. Right panel shows 5× enlargement of indicated area. Scale bar, 10 μm.

Figure 1
Figure 1. CLUH forms specific RNA granules with its targets
  1. A–C

    Confocal images of primary hepatocytes grown under indicated conditions and stained with anti‐CLUH antibody and Hadha (A), Pcx (B), and Actb (C) mRNA in situ hybridization. Right panels show 5× magnified boxed areas. Scale bar, 10 μm.

  2. D–F

    Manders’ colocalization coefficient between Hadha (D), Pcx (E), and Actb (F) mRNA molecules and CLUH signal (n ≥ 50 cells isolated from 3 to 6 mice).

  3. G, H

    Confocal images of primary hepatocytes grown in HBSS and stained with anti‐CLUH and anti‐TIA‐1 (G) or anti‐G3BP1 (H) antibodies. Right panels show 5× magnified boxed areas. Scale bar, 10 μm.

Data information: In (D–F), data are presented as boxplots showing the median, the first quartile, and the third quartile. Error bars show minimum and maximum values. ***P ≤ 0.001 (Student's t‐test).
Figure 2
Figure 2. CLUH granules contain stalled and active translation sites

  1. Confocal images of primary hepatocytes grown under indicated conditions and treatments after ribopuromycylation assay stained with anti‐puromycin antibody. Scale bar, 10 μm.

  2. Quantification of fluorescence intensity per cell of experiment shown in (A). AU, arbitrary units (n = 90–110 cells per treatment isolated from 4 mice).

  3. Quantification of the number of puromycin granules under indicated conditions (n ≥ 50 cells isolated from 4 mice).

  4. Confocal images of primary hepatocytes stained with anti‐puromycin and anti‐CLUH antibodies. The cells from which the enlarged areas (400 μm2) have been magnified are shown in Appendix Fig S1B for each individual channel. Scale bar, 4 μm. Cells analyzed were isolated from 4 different mice with similar results. Arrows point to colocalizing particles.

  5. Manders’ colocalization coefficient between puromycin and CLUH from experiment shown in (D) (n ≥ 50 cells isolated from 4 mice).

  6. Quantification of puromycin granules containing CLUH signal (n ≥ 80 cells isolated from 4 mice).

Data information: In (B, C, E, F), data are presented as boxplots showing the median, the first quartile, and the third quartile. Error bars show minimum and maximum values. (C, F) ***P ≤ 0.001; **P ≤ 0.01 (Student's t‐test). (E) ***P ≤ 0.001 (one‐way ANOVA, Tukey's multiple comparison test).
Figure 3
Figure 3. CLUH granules are translationally active or dormant depending on the mRNA
  1. A–C

    Confocal images of primary hepatocytes after ribopuromycylation experiment combined with mRNA in situ hybridization for (A) Hadha, (B) Pcx, and (C) Hmgcs2. Scale bar, 4 μm.

  2. D–F

    Fluorescence profile of 100‐pixel line from the merged channel shown in (A–C). The cells from which the enlarged areas (400 μm2) have been magnified are shown in Appendix Fig S2 for each individual channel. Cells analyzed were isolated from 6 different mice with similar results.

Figure 4
Figure 4. CLUH granules form in the absence of G3BPs and are distinct from SGs

  1. Representative Western blot of HeLa cells downregulated for G3BPs and overexpressing untagged CLUH. Asterisks indicate signal of previous incubation with anti‐G3BP1 antibody. Pan‐actin was used as loading control.

  2. Confocal images of anti‐CLUH staining in HeLa cells downregulated for G3BPs and overexpressing untagged CLUH. Insets show 2.6× enlarged area. Scale bar, 10 μm.

  3. Quantification of cells with CLUH granules from experiments shown in (B) and Fig EV2B (n = 3 independent experiments, > 50 cells per experiment per condition).

  4. Live imaging of WT and CLUH KO HeLa cells transfected with G3BP1‐GFP plasmid and incubated in HBSS medium with or without CHX. Cells were recorded for a maximum of 2 h. Insets show 2.5× enlarged areas. Scale bar, 10 μm.

  5. Total number of cells analyzed by live imaging for the indicated experiments shown in (D). “Positive” relates to a cell which formed G3BP1 granules at the end of the recording.

  6. Confocal images of anti‐G3BP1 staining in primary hepatocytes derived from Li‐Cluh WT and Li‐Cluh KO mice grown in indicated media. Arrows point to G3BP1‐granules. Scale bar, 10 μm.

  7. Quantification of percentage of cells with G3BP1‐positive granules of experiment shown in (F). Cells were analyzed in a blind fashion (n = 3 mice per genotype; number of cells in total: WT basal, 449; KO basal, 303; WT HBSS, 146; KO HBSS, 161).

  8. Western blot of primary WT and KO hepatocytes stained with indicated antibodies. Asterisks indicate unspecific signal. Pan‐actin was used as loading control.

  9. Quantification of Western blot shown in (H) (n = 3 per genotype per condition).

  10. mRNA levels of G3bp1 in primary hepatocytes grown under indicated conditions (n = 4 per genotype per condition).

Data information: In (C, G, I, J), data are presented as histograms showing the mean ± SEM. *P ≤ 0.05; **P ≤ 0.01; ***P ≤ 0.001 (one‐way ANOVA, Tukey's multiple comparisons test).Source data are available online for this figure.
Figure EV2
Figure EV2. Overexpressed CLUH forms CHX resistant granules in HeLa cells
  1. A

    Confocal images of WT and CLUH KO HeLa cells stained with anti‐CLUH antibody. Scale bar, 10 μm.

  2. B

    Confocal images of HeLa cells downregulated for G3BPs and overexpressing untagged CLUH (marked with asterisks). These images were overexposed to detect cells with CLUH expression at endogenous level CLUH. Asterisks indicate overexpressing cells. Scale bar, 10 μm.

  3. C, D

    Confocal images of HeLa cells overexpressing untagged CLUH (C) or FLAG‐tagged CLUH (D) stained with indicated antibodies. Scale bar, 10 μm.

  4. E

    Confocal images of HeLa cells overexpressing untagged CLUH treated with or without CHX and stained with the indicated antibodies. Scale bar, 10 μm.

  5. F

    Quantification of percentage of cells with CLUH granules of experiment shown in (E) (n = 3 independent experiments, > 50 cells per condition per replicate).

  6. G

    Confocal images of HeLa cells treated with arsenite with or without CHX and stained with anti‐G3BP1 antibody. Scale bar, 10 μm.

  7. H

    Quantification of percentage of cells with G3BP1 granules of experiment shown in (G) (n = 3 independent experiments, > 50 cells per condition per replicate).

  8. I

    Live imaging of WT and CLUH KO HeLa cells transfected with G3BP1‐GFP plasmid and treated with arsenite with and without CHX. Cells were recorded for a maximum of 30 min. Scale bar, 10 μm.

  9. J

    Total number of cells analyzed by live imaging for the indicated experiments. “Positive” indicates a cell which forms G3BP1 granules at the end of the recording.

Data information: In (F, H), data are presented as histograms showing the mean ± SEM. (H) ***P ≤ 0.001 (Student's t‐test).
Figure 5
Figure 5. Transcriptomic and proteomic profile of WT and KO hepatocytes
  1. A, B

    Volcano plots showing significantly changed proteins in KO hepatocytes relative to WT in basal (A) and HBSS (B) conditions.

  2. C

    2D score plots of enriched pathways in transcriptomics analysis of KO and WT hepatocytes under basal condition and upon HBSS (2 h).

  3. D

    2D score plots of enriched pathways in proteomics analysis of KO and WT hepatocytes under basal condition and upon HBSS (2 h).

  4. E

    2D score plots of enriched pathways in proteomics versus transcriptomics analysis of KO and WT hepatocytes under basal condition.

  5. F

    Correlation of protein fold changes in KO with respect to WT hepatocytes under basal condition and upon HBSS (2 h). Only genes previously found in CLUH RIP experiments in HeLa cells (Gao et al, 2014) are plotted. Inset shows magnification of upper part of the graph.

Figure 6
Figure 6. CLUH regulates mTORC1 activation and inhibits apoptosis upon starvation
  1. A

    Representative Western blot of primary hepatocytes starved in HBSS for the indicated time points and probed with indicated antibodies. Pan‐actin was used as loading control.

  2. B, C

    Quantification of Western blots as shown in (A). Antibody signal was normalized to pan‐actin signal, and signal of phospho‐protein was normalized to signal of total protein (n = 4 independent experiments).

  3. D

    Representative Western blot of total protein extracts from livers of starved Li‐Cluh WT and Li‐Cluh KO mice probed with indicated antibodies. GAPDH was used as loading control.

  4. E, F

    Quantification of Western blots as shown in (D). Antibody signal was normalized to GAPDH signal, and signal of phospho‐protein was normalized to signal of the total protein (n = 8 mice per genotype).

  5. G

    Confocal images of primary hepatocytes grown in HBSS for 16 h and stained with anti‐cleaved caspase 3 antibody. Scale bar, 25 μm.

  6. H

    Quantification of cleaved caspase 3‐positive hepatocytes treated with or without rapamycin. More than 100 cells per replicate were analyzed in a blind fashion (n = 3 independent experiments).

  7. I

    IHC staining of cleaved caspase 3 in liver sections of starved Li‐Cluh WT and Li‐Cluh KO mice. Arrows point to positive apoptotic foci. Scale bar, 200 μm.

  8. J

    Quantification of positive apoptotic foci of pictures shown in (I). The number of positive spots was relativized to the area analyzed (n = 5 mice per genotype).

  9. K

    Confocal images of primary hepatocytes grown in indicated media and stained with anti‐G3BP1 and anti‐mTOR. Scale bar, 10 μm.

  10. L

    Manders’ colocalization coefficient between G3BP1 and mTOR staining (n > 50 cells isolated from 3 mice).

Data information: In (B, C), data show the mean ± SEM. **P ≤ 0.01 for genotype and ***P ≤ 0.001 for time (two‐way ANOVA). In (E, F, H, J), data are presented as histograms showing the mean ± SEM. In (L), data are presented as boxplots showing the median, the first quartile, and the third quartile. Error bars show minimum and maximum values. (E, F, L) ***P ≤ 0.001 (Student's t‐test). (H, J) *P ≤ 0.05; ***P ≤ 0.001 (one‐way ANOVA, Tukey's multiple comparison test).Source data are available online for this figure.
Figure 7
Figure 7. Lack of CLUH promotes bulk autophagy
  1. A

    Confocal images of liver cryosections from Li‐Cluh WT and Li‐Cluh KO‐LC3‐GFP mice. Scale bar, 25 μm.

  2. B

    Quantification of LC3 signal morphology of images shown in (A). Graphs show relative number of LC3 autophagosomes and the relative area occupied per nuclei. Four images were quantified per animal and averaged. Nuclear staining was excluded for the analysis (n = 4 mice per genotype).

  3. C

    Confocal images of primary hepatocytes derived from Li‐Cluh WT and Li‐Cluh KO‐LC3‐GFP mice. Hepatocytes were starved for 2 h in HBSS. Scale bar, 10 μm.

  4. D, E

    Quantification of LC3 puncta number (D) and area (E) per cell (n ≥ 100 cells isolated from 3 mice).

  5. F

    Western blot of total protein extracts from livers of Li‐Cluh WT and Li‐Cluh KO mice probed with LAMP1. GAPDH was used as loading control.

  6. G

    Quantification of Western blot shown in (F). WT: n = 4; KO: n = 3.

  7. H

    Representative Western blot of primary hepatocytes treated for 2 h with bafilomycin A in the indicated media and probed with indicated antibodies. Two different exposure times are shown for LC3. I refers to total LC3, and II indicates lipidated form of LC3. Asterisk indicates signal of previous incubation.

  8. I

    Quantification of LC3 levels in Western blots as shown in (H) (n = 4 independent experiments).

  9. J

    Quantification of autophagic flux of Western blot as shown in (H). The flux was calculated as the increase in LC3II upon bafilomycin A treatment in each condition. The flux in KO cells was normalized to flux in WT cells (n = 4 independent experiments).

  10. K, L

    Quantification of p62 and CLUH levels in Western blots as shown in (H) (n = 4 independent experiments).

Data information: In (B, G, I–L), data are presented as histograms showing the mean ± SEM. In (D, E), data are presented as boxplots showing the median, the first quartile, and the third quartile. Error bars show minimum and maximum values. (B, G, J) *P ≤ 0.05; **P ≤ 0.01; ***P ≤ 0.001 (Student's t‐test). (D, E, I, K, L) *P ≤ 0.05; **P ≤ 0.01; ***P ≤ 0.001 (one‐way ANOVA, Tukey's multiple comparison test).Source data are available online for this figure.
Figure EV3
Figure EV3. G3BP1 granules colocalize with LAMP1
  1. A, B

    Confocal images of primary hepatocytes isolated from Li‐Cluh WT (A) and Li‐Cluh KO (B) mice and stained with anti‐G3BP1 and anti‐LAMP1 antibodies. 7× magnified areas are shown on the right side. Scale bar, 10 μm.

  2. C

    Manders’ colocalization coefficient between G3BP1 and LAMP1 signals of experiments shown in (A, B) (n = 35 cells with granules from 3 mice per genotype).

Data information: In (C), data show the mean ± SEM. (C) *P ≤ 0.05 (Student's t‐test); ***P ≤ 0.001 for time; and **P ≤ 0.01 for interaction time–genotype (two‐way ANOVA).
Figure 8
Figure 8. CLUH promotes mitophagy

  1. Western blot analysis of total protein extracts from livers of 8‐week‐old Li‐Cluh WT and Li‐Cluh KO mice probed with indicated antibodies. GAPDH was used as loading control.

  2. Quantification of Western blot shown in (A). WT: n = 4; KO: n = 3.

  3. Western blot analysis of isolated mitochondria from Li‐Cluh WT and Li‐Cluh KO mice probed with indicated antibodies. Ponceau S staining was used as a loading control.

  4. Quantification of Western blot shown in (C) (n = 3 mice per genotype).

  5. Confocal images of liver cryosections derived from Li‐Cluh WT and Li‐Cluh KO‐LC3‐GFP mice and stained with anti‐TOM20 antibody. On the right side, indicated boxes were 3.5× magnified. Arrows point to colocalizing spots. Scale bar, 25 μm.

  6. Manders’ colocalization coefficient between TOM20 and LC3 from experiments shown in (E). Five images were analyzed per animal to get an averaged value (n = 5 mice per genotype).

  7. Confocal images of primary hepatocytes derived from Li‐Cluh WT and Li‐Cluh KO‐LC3‐GFP mice and stained with anti‐TOM20 antibody. When indicated, cells were treated with rapamycin (200 nM for 4 h). Bottom panels show 2.3× magnified areas for each channel of indicated boxes. Arrows point to colocalizing spots. Scale bar, 10 μm.

  8. Manders’ colocalization coefficient from experiment shown in (G) (n ≥ 50 cells isolated from 3 mice per genotype).

  9. Confocal images of WT and Cluh KO MEFs transfected with mCherry‐GFP‐FIS101–152 and grown for 16 h in galactose medium or in medium containing 10 μM antimycin A/oligomycin. Arrows indicate mitophagosomes. Scale bar, 10 μm.

  10. Quantification of percentage of cells with mitophagosomes. Cells were considered positive when red signal was clearly recognizable (n = 3–5 independent experiments, > 100 cells per experiment).

  11. Confocal images of WT and Cluh KO MEFs transfected with mCherry‐GFP‐FIS101–152, incubated with 200 nM rapamycin, and grown for 16 h in galactose medium or in medium containing 10 μM antimycin A/oligomycin. Arrows indicate mitophagosomes. Scale bar, 10 μm.

  12. Quantification of percentage of cells with mitophagosomes. Cells were considered positive when red signal was clearly recognizable (n = 3–5 independent experiments, > 100 cells per experiment).

Data information: In (B, D, F, J, L), data are presented as histograms showing the mean ± SEM. In (H), data are presented as boxplots showing the median, the first quartile, and the third quartile. Error bars show minimum and maximum values. (B, D, F) ***P ≤ 0.001; *P ≤ 0.05 (Student's t‐test). (H, J, L) *P ≤ 0.05; **P ≤ 0.01; ***P ≤ 0.001 (one‐way ANOVA, Tukey's multiple comparison test).Source data are available online for this figure.
Figure 9
Figure 9. Inhibition of mTORC1 and downregulation of G3BPs recover mitochondrial clustering in the absence of CLUH

  1. Confocal images of liver cryosections from Li‐Cluh WT and Li‐Cluh KO mice injected with rapamycin or mock solution and stained with anti‐TOM20 antibody. Scale bar, 25 μm.

  2. Quantification of morphological parameters of experiments shown in (A). Four images were quantified per animal and averaged. Nuclear staining was excluded for the analysis (n = 3 mice per genotype).

  3. Confocal images of liver cryosections from Li‐Cluh WT and Li‐Cluh KO mice injected with rapamycin or mock solution and stained with anti‐LAMP1 antibody. Scale bar, 25 μm.

  4. Quantification of morphological parameters of experiments shown in (C). Four images were quantified per animal and averaged. Nuclear staining was excluded for the analysis (n = 3 mice per genotype).

  5. Confocal images of COS‐7 cells downregulated for the indicated genes and stained with anti‐TOM20 antibody. Scale bar, 10 μm.

  6. Representative Western blot of COS‐7 cells downregulated for the indicated genes of experiments shown in (E) and probed with indicated antibodies. Asterisks indicate unspecific signal or signal of previous incubation.

  7. Quantification of morphological parameters of experiments shown in (E) (n > 100 cells from 3 independent experiments).

Data information: In (B, D), data are presented as histograms showing the mean ± SEM. In (G), data are presented as boxplots showing the median, the first quartile, and the third quartile. Error bars show minimum and maximum values. (B, D) **P ≤ 0.01; *P ≤ 0.05 (Student's t‐test). (G) ***P ≤ 0.001; **P ≤ 0.01 *P ≤ 0.05 (one‐way ANOVA, Tukey's multiple comparison test).Source data are available online for this figure.
Figure EV4
Figure EV4. Mitochondrial clustering in the absence of CLUH is recovered by mTORC1 inhibition
  1. A

    Confocal images of COS‐7 cells downregulated for CLUH and treated with rapamycin or torin and stained with anti‐TOM20 antibody. Scale bar, 10 μm.

  2. B, C

    Quantification of morphological parameters of experiments shown in (A) (n > 100 cells analyzed from 3 independent experiments).

Data information: In (B, C), data are presented as boxplots showing the median, the first quartile, and the third quartile. Error bars show minimum and maximum values. (B, C) **P ≤ 0.01; ***P ≤ 0.001 (one‐way ANOVA, Tukey's multiple comparison test).
Figure EV5
Figure EV5. Model of CLUH function in the liver
CLUH forms granules containing target mRNAs and regulates their translation. In addition, CLUH granules recruit G3BP1, G3BP2, and mTOR, thereby enhancing mitophagy and inhibiting mitochondrial anabolic pathways. Together, these roles of CLUH are crucial in the liver to survive starvation. In the absence of CLUH, CLUH granules do not form resulting in reduced translation of Hmgcs2, Hadha, and Pcx mRNAs, in failure to suppress mTORC1‐dependent anabolic pathways, and in a block in mitophagy.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Alam U, Kennedy D (2019) Rasputin a decade on and more promiscuous than ever? A review of G3BPs. Biochim Biophys Acta Mol Cell Res 1866: 360–370 - PMC - PubMed
    1. Allen GFG, Toth R, James J, Ganley IG (2013) Loss of iron triggers PINK1/Parkin‐independent mitophagy. EMBO Rep 14: 1127–1135 - PMC - PubMed
    1. Anders S, Huber W (2010) Differential expression analysis for sequence count data. Genome Biol 11: R106 - PMC - PubMed
    1. Anderson P, Kedersha N (2009) RNA granules: post‐transcriptional and epigenetic modulators of gene expression. Nat Rev Mol Cell Biol 10: 430–436 - PubMed
    1. Bolte S, Cordelieres FP (2006) A guided tour into subcellular colocalization analysis in light microscopy. J Microsc 224: 213–232 - PubMed

Publication types

Substances