Mitochondria Take Up and Dispose of Cytosolic Protein Aggregates

Protein quality control (PQC) plays an essential role in maintaining cellular homeostasis by preventing the accumulation of misfolded proteins. The quality of cytosolic proteins is maintained through the coordinated actions of molecular chaperones, the ubiquitin-proteasome system, and the autophagy-lysosome pathway. Failure of these mechanisms results in protein aggregation and contributes to the development of heart failure ^1,2. PQC is particularly important in cardiomyocytes, which are long-lived, terminally differentiated cells that need to continuously maintain highly organized sarcomeric structures. There is emerging evidence that mitochondria also participate in cytosolic PQC. In yeast, aggregation-prone cytosolic proteins are imported into mitochondria and degraded by mitochondrial proteases, a process termed mitochondria as guardians in cytosol ³. Similarly, mitochondria in HeLa cells have also been shown to take up and discard unfolded cytosolic proteins ⁴. However, the molecular machinery allowing mitochondria to sort, import and destroy cytosolic protein aggregates remains elusive particularly in the heart. Given the critical importance of mitochondria in energy/metabolism and their fragility in stressed cardiomyocytes, if mitochondria in cardiomyocytes do import and destroy cytosolic aggregation-prone proteins, such missions would require sophistication and security.

In a recent article in Circ Res, Rawnsley et al. showed that aggregation-prone proteins, including CRYAB-R120G and BAG3-P209L, are found in the mitochondrial matrix, as evidenced by their retention within mitoplasts ⁵. Proteotoxic stress was accompanied by increased mitochondrial localization of Tumor necrosis factor receptor associated factor 2 (TRAF2) and accumulation of aggregation-prone proteins in mitochondria. Loss of TRAF2 increased protein aggregates in both cytosol and mitochondria, whereas TRAF2 augmentation promoted aggregate clearance and preserved cardiac function. TRAF2 is an E3-ligase and the authors have shown previously that TRAF2 is involved in mitophagy at baseline conditions ⁶. Thus, the authors proposed that aggregation-prone proteins in the mitochondrial matrix can be disposed of through mitophagy (Figure 1).

Figure 1.

Aggregation-prone cytosolic proteins can be taken up by mitochondria through an HSC70-dependent mechanism and subsequently translocated into the mitochondrial matrix, possibly through mitochondrial translocases. FUNDC1 may also participate in this process. Mitochondria containing protein aggregates appear to undergo fission-mediated segregation followed by selective elimination through TRAF2-mediated mitophagy. TRAF2, an E3 ubiquitin ligase, mediates basal mitophagy in cardiomyocytes by promoting K63-linked polyubiquitination of mitochondrial proteins, which are recognized by p62, an LC3 adaptor protein. Although another E3 ubiquitin ligase, PARKIN, is not essential for this process, PARKIN overexpression can facilitate the elimination of mitochondria containing protein aggregates, including those that have become dysfunctional as a result of aggregate accumulation. When TRAF2-mediated mitophagy is impaired, protein aggregates accumulate in both the cytosol and mitochondria, leading to proteotoxic and mitochondrial stress and ultimately contributing to the development of heart failure. FUNDC1, FUN14 domain-containing protein 1; HSC70, heat shock cognate 71kDa protein; TRAF2, tumor necrosis factor receptor-associated factor 2; PARKIN, parkin RBR E3 ubiquitin-protein ligase; TOM, translocase of the outer mitochondrial membrane; TIM, translocase of the inner mitochondrial membrane; FIS1, fission, mitochondrial 1; DRP1, dynamin-related protein 1. Created in BioRender. Sung, E. (2026) https://BioRender.com/n1hq6sd

The most surprising but important finding in the current study is that aggregation-prone cytosolic proteins can be taken up by mitochondria and disposed of through mitophagy in the mammalian heart. Increases in aggregation-prone proteins in the mitochondrial matrix would induce mitochondrial dysfunction if not removed promptly. Due to the absence of a proteasome degradation system in the mitochondria matrix, however, protein aggregates should, in theory, be removed by proteases, autophagy or re-exported to the cytosol. Rawnsley et al. demonstrated for the first time that TRAF2 mitigates both cytosolic and mitochondrial proteotoxicity during cardiac proteinopathy by stimulating mitophagy ⁵. Mitophagy is generally viewed as a mechanism for selectively eliminating damaged or excessive mitochondria. However, Rawnsley et al. showed that mitochondria function as a carrier of cytosolic proteotoxic cargo for lysosomal disposal through autophagy. Thus, the study expands the territory of PQC by mitophagy from mitochondria alone to a wider area, including the cytosol. The study also suggested that either insufficient activation or downregulation of TRAF2-mediated mitophagy promotes mitochondrial accumulation of misfolded proteins, thereby representing a novel mechanism driving mitochondrial dysfunction in proteinopathy.

This study is highly provocative but leaves many issues for clarification in a future investigation. First and foremost, why do cardiomyocytes endanger themselves by potentially harming mitochondria and subjecting them to extra burdens? Although providing clear answers is difficult, mitochondria already contain efficient protein-import machineries, chaperones and matrix proteases, allowing them to provide degradative capacity without requiring construction of a new disposal compartment. Perinuclear and subsarcolemmal mitochondria are located in cells where protein aggregation takes place. This strategy may be especially useful in long-lived, largely non-dividing cardiomyocytes, which cannot easily dilute protein aggregates through cell division. However, since mitochondrial accumulation of unfolded proteins impairs mitochondrial integrity, how cytosolic aggregation-prone proteins are handled by mitochondria needs to be scrutinized.

For example, greater insight is needed into the molecular mechanisms by which aggregation-prone proteins accumulate in the mitochondrial matrix. Aggregation-prone cytosolic proteins generally lack conventional mitochondrial targeting sequences, raising the important question of how cytosolic misfolded proteins are sorted and delivered to the mitochondrial matrix. A previous study suggested that Heat shock cognate 71kDa protein (HSC70) interacts with FUN14 domain-containing protein 1 (FUNDC1), a mitochondrial outer membrane protein, to direct unfolded cytosolic client proteins toward mitochondria and that TOM components are required for their import in HeLa cells ⁴. If protein aggregates are taken up through the same mechanism, however, they would first need to be disaggregated and unfolded before passing through the mitochondrial translocase machinery as narrow, linear polypeptide chains. This raises several important questions. Why would aggregation-prone proteins undergo this elaborate process rather than being degraded directly through the ubiquitin–proteasome system, chaperone-mediated autophagy, or aggrephagy in the cytosol? Furthermore, how would cytosolic misfolded proteins or disaggregated polypeptides compete with the large number of nuclear-encoded mitochondrial proteins for access to the mitochondrial translocase machinery?

Another important question is how cardiomyocytes avoid mitochondrial dysfunction caused by cytosolic protein aggregates. Rawnsley et al proposed that mitochondria can segregate the aggregates and dispose of them by mitophagy, thereby maintain the PQC in mitochondria. Unlike PARKIN-mediated mitophagy, TRAF2-mediated mitophagy can eliminate polarized mitochondria in the unstressed heart ⁶. Thus, clearance of cytosolic protein aggregates through TRAF2-mediated mitophagy may result in the loss of otherwise functional mitochondria, in contrast to more selective degradation of misfolded proteins by mitochondrial proteases. This raises an important question: how do cardiomyocytes cope with the loss of mitochondria associated with aggregate clearance?

In HeLa cells, when cytosolic proteins imported through FUNDC1/HSC70 escape degradation mediatd by Long peptidase 1 (LONP1), a mitochondrial protease, they can accumulate within discrete mitochondrial regions that subsequently undergo Fission, mitochondrial 1 (FIS1)-dependent segregation ⁴. These structures contain both mitochondrial components and ubiquitinated proteins and p62 and are selectively removed by autophagy. Thus, autophagic degradation may be spatially restricted to mitochondrial regions enriched in proteotoxic cargo, such that the benefit of eliminating protein aggregates outweighs the cost of losing mitochondrial material. Cardiomyocytes may similarly segregate aggregate-enriched mitochondrial subdomains before their autophagic removal. Consistent with this possibility, Ser616 phosphorylated Dynamin-related protein 1 (DRP1), a mitochondrial fission protein, is upregulated in the cardiomyocytes of proteinopathic hearts ⁵. It remains unknown whether fission precedes TRAF2 recruitment to isolate proteotoxic cargo, whether TRAF2 recruitment promotes subsequent segregation or whether the two processes occur in parallel. Further investigation is required to determine how cytosolic protein aggregates are spatially segregated within mitochondria and how TRAF2 selectively marks these aggregate-containing mitochondrial compartments. Defining the molecular signal that recruits, upregulates or activates TRAF2 on these mitochondria would substantially strengthen the proposed mechanism. For example, it would be interesting to investigate whether the mitochondrial unfolded protein response is involved in the upregulation of TRAF2.

Finally, because cardiomyocytes are unlikely to maintain a large pool of excess mitochondria that can simply be sacrificed for PQC, the relationship between TRAF2-mediated mitophagy and mitochondrial biogenesis should also be clarified. In particular, it will be important to determine whether enhanced elimination of aggregate-containing mitochondria is coupled with compensatory mitochondrial biogenesis, thereby preserving mitochondrial mass and energetic capacity in cardiomyocytes.

As a cautionary note, TRAF2 may have diverse functions in proteinopathy, including inhibition of necroptosis and apoptosis, regulation of systemic inflammation, and K63-linked ubiquitination of multiple targets ⁷. Thus, further investigation is required to determine the extent to which TRAF2 protects against cardiac proteinopathy specifically through mitophagy, as opposed to its broader roles in cell death and inflammatory signaling. Electron microscopic evidence of TRAF2-positive autophagic structures containing mitochondria that harbor cytosolic protein aggregates would further strengthen the proposed mechanism. More importantly, it will be necessary to determine whether selective restoration of the mitophagic disposal of aggregate-containing mitochondria can rescue cardiac proteinopathy in a mouse model with a loss of TRAF2 function. Toward this goal, identification of additional components of TRAF2-mediated mitophagy beyond p62, including K63-ubiquitinated targets on the mitochondrial membrane, may provide approaches to selectively manipulate this pathway and thereby establish its specific contribution to the protective effects of TRAF2.

In conclusion, Rawnsley et al. reported a novel observation that mitophagy-mediated disposal of cytosolic protein aggregates could be a critical mechanism maintaining both mitochondrial and cytosolic PQC during proteinopathy. Elucidating the specific underlying mechanisms of mitochondrial transfer of protein aggregate and the molecular mechanism of TRAF2-mediated mitophagy may lead to the development of novel interventions for proteinopathy in the heart. Although Rawnsley et al. utilized mouse models of proteinopathy to demonstrate the role of TRAF2-mediated mitophagy in PQC mechanisms, it is important to determine whether mitochondrial import of misfolded proteins and consequent removal by mitophagy are also activated under more pathophysiologically common conditions, including ischemia, metabolic stress, aging, and heart failure, in which cytosolic proteotoxicity and mitochondrial dysfunction frequently coexist ⁸. Although these abnormalities are generally considered parallel consequences of cellular stress, the current model suggests a more direct mechanistic connection. Specifically, mitochondrial dysfunction may develop as a consequence of a failed attempt by mitochondria to alleviate cytosolic proteotoxic stress, particularly when compensatory activation of mitophagy is insufficient. Both hemodynamic and metabolic stress activate mitophagy in the heart; however, this activation is often transient and is followed by the development of cardiac dysfunction. It would therefore be important to investigate whether cytosolic protein aggregates accumulate in mitochondria under these pathophysiological conditions and whether TRAF2-mediated mitophagy contributes to their clearance.