Loss of proteostasis underlies ageing and neurodegeneration characterized by the accumulation of protein aggregates and mitochondrial dysfunction1,2,3,4,5. Although many neurodegenerative-disease-associated proteins can be found in mitochondria4,6, it remains unclear how mitochondrial dysfunction and protein aggregation could be related. In dividing yeast cells, protein aggregates that form under stress or during ageing are preferentially retained by the mother cell, in part through tethering to mitochondria, while the disaggregase Hsp104 helps to dissociate aggregates and thereby enables refolding or degradation of misfolded proteins7,8,9,10. Here we show that, in yeast, cytosolic proteins prone to aggregation are imported into mitochondria for degradation. Protein aggregates that form under heat shock contain both cytosolic and mitochondrial proteins and interact with the mitochondrial import complex. Many aggregation-prone proteins enter the mitochondrial intermembrane space and matrix after heat shock, and some do so even without stress. Timely dissolution of cytosolic aggregates requires the mitochondrial import machinery and proteases. Blocking mitochondrial import but not proteasome activity causes a marked delay in the degradation of aggregated proteins. Defects in cytosolic Hsp70s leads to enhanced entry of misfolded proteins into mitochondria and elevated mitochondrial stress. We term this mitochondria-mediated proteostasis mechanism MAGIC (mitochondria as guardian in cytosol) and provide evidence that it may exist in human cells.
Subscribe to Journal
Get full journal access for 1 year
only $3.90 per issue
All prices are NET prices.
VAT will be added later in the checkout.
Rent or Buy article
Get time limited or full article access on ReadCube.
All prices are NET prices.
Lin, M. T. & Beal, M. F. Mitochondrial dysfunction and oxidative stress in neurodegenerative diseases. Nature 443, 787–795 (2006)
Balch, W. E., Morimoto, R. I., Dillin, A. & Kelly, J. W. Adapting proteostasis for disease intervention. Science 319, 916–919 (2008)
Hartl, F. U., Bracher, A. & Hayer-Hartl, M. Molecular chaperones in protein folding and proteostasis. Nature 475, 324–332 (2011)
Schapira, A. H. Mitochondrial diseases. Lancet 379, 1825–1834 (2012)
López-Otín, C., Blasco, M. A., Partridge, L., Serrano, M. & Kroemer, G. The hallmarks of aging. Cell 153, 1194–1217 (2013)
Hansson Petersen, C. A. et al. The amyloid β-peptide is imported into mitochondria via the TOM import machinery and localized to mitochondrial cristae. Proc. Natl Acad. Sci. USA 105, 13145–13150 (2008)
Parsell, D. A., Kowal, A. S., Singer, M. A. & Lindquist, S. Protein disaggregation mediated by heat-shock protein Hsp104. Nature 372, 475–478 (1994)
Doyle, S. M., Genest, O. & Wickner, S. Protein rescue from aggregates by powerful molecular chaperone machines. Nat. Rev. Mol. Cell Biol. 14, 617–629 (2013)
Zhou, C. et al. Organelle-based aggregation and retention of damaged proteins in asymmetrically dividing cells. Cell 159, 530–542 (2014)
Miller, S. B., Mogk, A. & Bukau, B. Spatially organized aggregation of misfolded proteins as cellular stress defense strategy. J. Mol. Biol. 427, 1564–1574 (2015)
Gupta, R. et al. Firefly luciferase mutants as sensors of proteome stress. Nat. Methods 8, 879–884 (2011)
Wallace, E. W. et al. Reversible, specific, active aggregates of endogenous proteins assemble upon heat stress. Cell 162, 1286–1298 (2015)
Neupert, W. & Herrmann, J. M. Translocation of proteins into mitochondria. Annu. Rev. Biochem. 76, 723–749 (2007)
Martin, J., Mahlke, K. & Pfanner, N. Role of an energized inner membrane in mitochondrial protein import. ΔΨ drives the movement of presequences. J. Biol. Chem. 266, 18051–18057 (1991)
Serrano, R. Energy requirements for maltose transport in yeast. Eur. J. Biochem. 80, 97–102 (1977)
Stevens, H. C. & Nichols, J. W. The proton electrochemical gradient across the plasma membrane of yeast is necessary for phospholipid flip. J. Biol. Chem. 282, 17563–17567 (2007)
Pareek, G., Krishnamoorthy, V. & D’Silva, P. Molecular insights revealing interaction of Tim23 and channel subunits of presequence translocase. Mol. Cell. Biol. 33, 4641–4659 (2013)
Cabantous, S., Terwilliger, T. C. & Waldo, G. S. Protein tagging and detection with engineered self-assembling fragments of green fluorescent protein. Nat. Biotechnol. 23, 102–107 (2005)
Fehrenbacher, K. L., Yang, H. C., Gay, A. C., Huckaba, T. M. & Pon, L. A. Live cell imaging of mitochondrial movement along actin cables in budding yeast. Curr. Biol. 14, 1996–2004 (2004)
Lagier-Tourenne, C., Polymenidou, M. & Cleveland, D. W. TDP-43 and FUS/TLS: emerging roles in RNA processing and neurodegeneration. Hum. Mol. Genet. 19 (R1), R46–R64 (2010)
Becker, J., Walter, W., Yan, W. & Craig, E. A. Functional interaction of cytosolic hsp70 and a DnaJ-related protein, Ydj1p, in protein translocation in vivo . Mol. Cell. Biol. 16, 4378–4386 (1996)
Kummer, E., Oguchi, Y., Seyffer, F., Bukau, B. & Mogk, A. Mechanism of Hsp104/ClpB inhibition by prion curing guanidinium hydrochloride. FEBS Lett. 587, 810–817 (2013)
Ryan, M. T., Voos, W. & Pfanner, N. Assaying protein import into mitochondria. Methods Cell Biol. 65, 189–215 (2001)
Devi, L., Raghavendran, V., Prabhu, B. M., Avadhani, N. G. & Anandatheerthavarada, H. K. Mitochondrial import and accumulation of α-synuclein impair complex I in human dopaminergic neuronal cultures and Parkinson disease brain. J. Biol. Chem. 283, 9089–9100 (2008)
Baker, M. J., Tatsuta, T. & Langer, T. Quality control of mitochondrial proteostasis. Cold Spring Harb. Perspect. Biol. 3, a007559 (2011)
Rep, M. et al. Promotion of mitochondrial membrane complex assembly by a proteolytically inactive yeast Lon. Science 274, 103–106 (1996)
Lee, D. H. & Goldberg, A. L. Selective inhibitors of the proteasome-dependent and vacuolar pathways of protein degradation in Saccharomyces cerevisiae . J. Biol. Chem. 271, 27280–27284 (1996)
Collins, G. A., Gomez, T. A., Deshaies, R. J. & Tansey, W. P. Combined chemical and genetic approach to inhibit proteolysis by the proteasome. Yeast 27, 965–974 (2010)
Knott, A. B., Perkins, G., Schwarzenbacher, R. & Bossy-Wetzel, E. Mitochondrial fragmentation in neurodegeneration. Nat. Rev. Neurosci. 9, 505–518 (2008)
Tyedmers, J., Mogk, A. & Bukau, B. Cellular strategies for controlling protein aggregation. Nat. Rev. Mol. Cell Biol. 11, 777–788 (2010)
Longtine, M. S. et al. Additional modules for versatile and economical PCR-based gene deletion and modification in Saccharomyces cerevisiae . Yeast 14, 953–961 (1998)
Mumberg, D., Müller, R. & Funk, M. Yeast vectors for the controlled expression of heterologous proteins in different genetic backgrounds. Gene 156, 119–122 (1995)
Huh, W. K. et al. Global analysis of protein localization in budding yeast. Nature 425, 686–691 (2003)
Tran, P. T., Paoletti, A. & Chang, F. Imaging green fluorescent protein fusions in living fission yeast cells. Methods 33, 220–225 (2004)
Zhou, C. et al. Motility and segregation of Hsp104-associated protein aggregates in budding yeast. Cell 147, 1186–1196 (2011)
Imamura, H. et al. Visualization of ATP levels inside single living cells with fluorescence resonance energy transfer-based genetically encoded indicators. Proc. Natl Acad. Sci. USA 106, 15651–15656 (2009)
Florens, L. & Washburn, M. P. Proteomic analysis by multidimensional protein identification technology. Methods Mol. Biol. 328, 159–175 (2006)
Washburn, M. P., Wolters, D. & Yates, J. R., III . Large-scale analysis of the yeast proteome by multidimensional protein identification technology. Nat. Biotechnol. 19, 242–247 (2001)
Eng, J. K., McCormack, A. L. & Yates, J. R. An approach to correlate tandem mass spectral data of peptides with amino acid sequences in a protein database. J. Am. Soc. Mass Spectrom. 5, 976–989 (1994)
Tabb, D. L., McDonald, W. H. & Yates, J. R., III . DTASelect and Contrast: tools for assembling and comparing protein identifications from shotgun proteomics. J. Proteome Res. 1, 21–26 (2002)
Zhang, Y., Wen, Z., Washburn, M. P. & Florens, L. Refinements to label free proteome quantitation: how to deal with peptides shared by multiple proteins. Anal. Chem. 82, 2272–2281 (2010)
Reinders, J., Zahedi, R. P., Pfanner, N., Meisinger, C. & Sickmann, A. Toward the complete yeast mitochondrial proteome: multidimensional separation techniques for mitochondrial proteomics. J. Proteome Res. 5, 1543–1554 (2006)
Lam, S. S. et al. Directed evolution of APEX2 for electron microscopy and proximity labeling. Nat. Methods 12, 51–54 (2015)
Boldogh, I. R. & Pon, L. A. Purification and subfractionation of mitochondria from the yeast Saccharomyces cerevisiae . Methods Cell Biol. 80, 45–64 (2007)
Stojanovski, D., Pfanner, N. & Wiedemann, N. Import of proteins into mitochondria. Methods Cell Biol. 80, 783–806 (2007)
DeSantis, M. E. et al. Operational plasticity enables hsp104 to disaggregate diverse amyloid and nonamyloid clients. Cell 151, 778–793 (2012)
Fernández-Higuero, J. A. et al. Allosteric communication between the nucleotide binding domains of caseinolytic peptidase B. J. Biol. Chem. 286, 25547–25555 (2011)
The authors thank K. Si for technical advice and reagents, and F.-U. Hartl, E. Craig, P. Silva, B. Bukau, S. Claypool and J. Wang for reagents. This work was supported by the grant R35 GM118172 from the National Institute of Health to R.L., and a fellowship from the American Heart Association to C.Z.
The authors declare no competing financial interests.
Extended data figures and tables
Extended Data Figure 1 Aggregate purification and role of mitochondrial import in aggregate dissolution.
a, Schematics of aggregate purification. Yeast strains expressing FlucSM–GFP–3×Flag or FlucSM–GFP as control were treated with heat shock at 42 °C to induce aggregate formation (green dots). Sucrose gradient centrifugation was then used to separate FlucSM–GFP monomers from aggregates in the cell lysates. The fraction enriched for protein aggregates was applied to an anti-Flag column to separate FlucSM–GFP–3×Flag aggregates from other cellular debris (grey shapes). b, Representative images (n = 8) of anti-flag resin and associated aggregates isolated from FlucSM–GFP (left) or FlucSM–GFP–3×Flag (right) strains. c, Total non-redundant peptides identified by MudPIT of independent repeats using FlucSM–GFP–3×Flag (blue) and FlucSM–GFP (red) strains. d, Representative images of examples of proteins identified by MudPIT that form aggregates after 30 min heat shock. Not3 is a negative control. Three images for each. e, Tom70–GFP did not colocalize with aggregates after 30 min heat shock. FlucSM–RFP was used as a marker for cytosolic aggregates. Image is representative of five captured. f, Anti-HA immunoblot showing co-purification of HA-tagged Tom70 and Tom40 but not Om45 with aggregates. g, Additional negative controls for experiment in f: anti-HA immunoblotting of two mitochondrial outer membrane proteins Mdm10 and Mdm34 that face the cytosol and were not identified by proteomics to be enriched in aggregates. h, Dissolution kinetics of heat shock-induced FlucSM–GFP aggregates in cells treated with CHX (green) or CHX+CCCP (red). Shown are fluorescent traces of three biological repeats for each condition. i, Measurement of ATP level under the same experimental conditions as for h using a FRET-based sensor showing that CCCP did not deplete cellular ATP. The FRET efficiencies of the CCCP+CHX-treated group and the CHX-treated group were normalized to the mean of the CHX-treated group at each indicated time point. 36 cells were measured at time 0 (before drug addition), 43 and 74 cells for the CHX condition (at 30 and 60 min, respectively), and 39 and 74 cells for CHX/CCCP (at 30 and 60 min, respectively). Shown are means and s.e.m. j, Images, representative of about 300 cells imaged, from three biological repeats of TIM23 and tim23ts cells stained with TMRM (red) after heat shock to demonstrate that the membrane potentials of these cells were similar. Scale bar, 2.5 μm for (e), 5 μm for other panels. For gel source data, see Supplementary Fig. 1. Source data
Extended Data Figure 2 Additional data demonstrating import of aggregate proteins into mitochondria after heat shock.
a, Schematic diagram explaining the split-GFP assay to detect translocation of a cytosolic protein into mitochondria. b, Positive (Grx5) and negative (Hsp104, Not3) controls used for the split-GFP assay. The split-GFP signal for the stable mitochondrial matrix protein Grx5 did not diminish in cells treated at 30 °C with CCCP for 15 min. This result is to be compared with that in Extended Data Fig. 6a. Top, split GFP images; middle, MTS–mCherry-labelled mitochondria; bottom, merged images. Images are representative of 9, 9, 4, 7, 8, 6 and 7 images captured from left to right. c, Additional examples of heat shock-induced translocation of cytosolic aggregate proteins into mitochondria, as shown by split-GFP signal after heat shock. Tsl1 and Tma19 are aggregate proteins confirmed by both imaging and proteomics (n = 9 from 3 biological repeats). d–f, Representative images (d) and quantifications (e, f) showing that CCCP treatment blocked the heat shock-induced mitochondrial translocation of Lsg1–GFP11. e and f show mean and s.e.m. of, from left to right, 874, 503 and 385 cells counted (e) and 351, 164 and 261 cells quantified (f); three biological repeats. In f the intensity ratio from each cell is normalized to the mean of 30 °C control samples before treatment. g, h, Images (g) and quantification (h) of TDP43–GFP11 split-GFP (top) and mitochondria (bottom). h shows mean and s.e.m. of 96 (30 °C) and 133 (heat shock) cells imaged and quantified; three biological repeats. i, Representative images showing the effect of inhibition of Hsp104 with GdnHCl on FlucSM import. j, Quantification of images in i and Fig. 2e with the corresponding controls. Shown are mean and s.e.m., from left to right, of 174, 195, 181, 183, 180 and 168 cells. k–n, Protease protection assay. k shows aggregates with FlucSM–GFP attached to purified mitochondria (red), representative of eight images acquired. l and n show immunoblots of purified post-heat shock mitochondria treated with or without detergents and proteases as indicated. TR, trypsin; PK, protease K. m, Anti-HA immunoblot of Tom70–HA as a mitochondria outer membrane protein in the protease protection assay in various treated samples. o, Schematics of the APEX assay to detect mitochondrial import. p, Image showing localization of GFP–APEX in mitochondria, representative of three images acquired. Tukey’s multiple comparisons test for e, f; unpaired two-tailed t-test for h, j. **P < 0.01, ***P < 0.001. Scale bars, 5 μm. For gel source data, see Supplementary Fig. 1. Source data
Extended Data Figure 3 Mitochondrial proteases and peptidases are important for efficient dissolution of aggregates after heat shock.
a, b, Representative images (a) and quantification (b) over time showing that the mitochondrial split-GFP signal of Lsg1–GFP11 increased after 30 min heat shock and gradually diminished during the 90-min recovery after returning to 30 °C. Top, split-GFP images; middle, MTS–mCherry-labelled mitochondria; bottom, merged images. b shows the mean and s.e.m. of the fraction of cells from three experiments that had a split-GFP signal at each time point. A total of 2,153 cells were counted; three biological repeats. c, Representative images (n = 8) of purified aggregates labelled by GFP-tagged FlucSM, bound to purified mitochondria, labelled with MTS–mCherry. d, Quantification of the immunoblot shown in Fig. 3c. e, f, Immunoblots of FlucSM–HA after aggregates purified from Δhsp104 cells were incubated with wild-type mitochondria for various amounts of time as indicated (e) and quantification of the immunoblot (f). g, h, Dissolution curves of aggregates (g) and their half-decay times (h) showing that the deletion of different mitochondrial proteases delayed the dissolution of cytosolic protein aggregates. g shows mean curves; h shows data points and mean half decay times extracted from fitted curves of three biological repeats. Original data for each repeat are available in Supplementary Information. Scale bars, 5 μm. For gel source data, see Supplementary Fig. 1. Source data
Extended Data Figure 4 The protease activity of Pim1 is important for efficient degradation of misfolded cytosolic proteins.
a, b, pim1S1015A:PIM1 grows normally under fermentable (a) and non-fermentable conditions (b). c, Representative images showing that delayed split-GFP of FlucSM disappearance in pim1S1015A:PIM1 cells was not affected by CHX during the recovery phase. Top, split GFP images; bottom, MTS–mCherry-labelled mitochondria. d, Quantification of mean split-GFP/mCherry ratio for pim1S1015A:PIM1. CHX was added after heat shock. Plotted are mean and s.e.m. from 747 cells that were imaged and measured; three biological repeats. e, f, Representative immunoblots (e) and quantification (f) from three biological repeats showing that the mitochondrial import (inhibited by CCCP) is the major source for degrading stress-damaged endogenous Lsg1–HA, but vacuole-mediated degradation (inhibited by PMSF) also plays a role, while the proteasome pathway (inhibited by MG132) has the least effect. f shows data points and mean. g, h, Representative immunoblots (g) and quantification (h) of wild-type or tim23ts cells treated with CHX after heat shock showing that mitochondrial import (inhibited by tim23ts) is important for the degradation of stress-damaged FlucSM–HA. h shows data points and mean plots from three biological replicates. i, j, Representative immunoblots (i) and quantification (j) showing that without heat shock, proteasome-mediated degradation of FlucSM–HA was inhibited by MG132. j shows data points and mean plots from three biological replicates. Scale bar, 5 μm. For gel source data, see Supplementary Fig. 1. Source data
Extended Data Figure 5 Impairment of cytosolic Hsp70 leads to enhanced import of unfolded proteins into mitochondria and mitochondrial damage.
a, b, Representative images (a) and quantification (b) of split-GFP signal for FlucSM–GFP11 in the Δssa2 Δssa3 Δssa4 ssa1ts strain. Cells growing at 23 °C were treated with CHX for 30 min. Shown are mean and s.e.m. of 126 and 133 cells imaged and measured; three biological repeats. Unpaired two-tailed t-test: **P < 0.01. c, The fraction of cells with intact mitochondria was decreased after heat shock and the Δssa2 Δssa3 Δssa4 ssa1ts mutant showed more severe fragmentation and delayed recovery compared to the wild type (see also representative mitochondrial images in Fig. 3a). Shown are mean and s.e.m. quantified from three biological repeats with 1,335 cells for the wild type and 967 cells for the mutant. d, Representative images (from the 160-min time point of the plot in Fig. 3h) of ROS indicated by DHE signal. 1,621 mutant and 2,151 wild-type cells were imaged and quantified. Scale bars, 5 μm. Source data
Extended Data Figure 6 Unstably folded cytosolic proteins are imported into mitochondria in both yeast and human RPE1 cells.
a–d, Representative images (a and c) and quantification (b and d, as in Fig. 2c) of the split-GFP signal for the super-aggregator Ded1 in cells grown at 30 °C without or with CCCP for 15 min or without or with treatment with CHX for 30 min at 30 °C, with 205 and 182 cells imaged and quantified in a, b, respectively; 283 and 287 cells imaged and quantified in c, d, respectively; three biological repeats. Top, split GFP images; middle, MTS–mCherry-labelled mitochondria; bottom, merged images. b and d show mean and s.e.m. of measurements from the indicated number of cells. Unpaired two-tailed t-test: **P < 0.01, ***P < 0.001. e, Images of the split-GFP signal of cells expressing the super-aggregators Fas1–GFP11, Ola1–GFP11 or FlucDM–GFP11 under non-stressful growth conditions (30 °C) showing that these proteins are imported into mitochondria even without imposed proteotoxic stress. Top, split GFP images; middle, MTS–mCherry-labelled mitochondria; bottom, merged images. n = 7, 9 and 9 from left to right. f, Immunoblot showing expression of different luciferase mutants relative to wild-type luciferase in human RPE1 cells (corresponding to the quantification in Fig. 4c), representative of five biological repeats for the Fluc proteins. Loading control, GAPDH. g, Working model of MAGIC. Cytosolic aggregates are attached to mitochondria through interaction with import receptors such as Tom70. Individual aggregate proteins, which may be dissociated from aggregates by Hsp104, are imported through the outer membrane import complex to the intermembrane space, where they are either degraded by intermembrane proteases and peptidases, or imported through an inner membrane channel to be degraded by matrix proteases such as Pim1. Scale bars, 5 μm. For gel source data, see Supplementary Fig. 1. Source data
This file contains Supplementary Table 1. (XLSX 18 kb)
This file contains Supplementary Table 2. (XLSX 72 kb)
This file contains Supplementary Table 3. (XLSX 11 kb)
This file contains Supplementary Figure 1, the uncropped blots with size marker indications. (PDF 1980 kb)
3D reconstructed SIM images showing Lsg1 split-GFP to be present inside mitochondria after heat shock
Mitochondria outer membrane was labeled with mCherry-Fis1TM; GFP1-10 was fused with the mitochondrial matrix protein Grx5 (Grx5-GFP1-10) and the native aggregate protein Lsg1 was tagged with GFP11 (details in Methods). After 30 min heat shock at 42°C, the split GFP signal was imaged with mitochondria by using structured illumination microscope (Nikon, N‐SIM). 3D reconstruction was done using the Elements N-SIM software and the video shown represents 6 reconstructions. (WMV 5005 kb)
About this article
Cite this article
Ruan, L., Zhou, C., Jin, E. et al. Cytosolic proteostasis through importing of misfolded proteins into mitochondria. Nature 543, 443–446 (2017). https://doi.org/10.1038/nature21695
Vesicular transport mediates the uptake of cytoplasmic proteins into mitochondria in Drosophila melanogaster
Nature Communications (2020)
Mitochondrial unfolded protein-related responses across kingdoms: similar problems, different regulators
The FEBS Journal (2020)
Inter-Organelle Membrane Contact Sites and Mitochondrial Quality Control during Aging: A Geroscience View
Aging Cell (2020)