Chaperones have a reputation for keeping the cell neat and tidy, but recent reports suggest the cell’s cleaning crew resort to littering when intracellular trash piles up. According to a study published in The EMBO Journal on June 3, the chaperone complex Hsc70/DnaJC5 latches onto a host of miscreant proteins implicated in neurodegenerative disease, including tau and α-synuclein, and ushers them out of neurons. The study, led by Chad Dickey at the University of South Florida in Tampa, adds neuronal intrigue to recent findings implicating another chaperone, USP19, in exporting misfolded proteins out of non-neuronal cells (see Jun 2016 news). Whether the chaperone-mediated offload could help propagate misfolded proteins throughout the brain in neurodegenerative diseases remains unexplored.

Tau Triage.

DnaJ proteins decide the fate of Hsc70 cargo proteins, including tau (black squiggles). The authors propose that DnaJC5 (also known as CS) funnels the cargo to vesicles for secretion via SNAP-23. [Courtesy of Fontaine et al., EMBO 2016.]

The idea that proteins associated with neurodegenerative disease spread across the brain in a prion-like manner is building steam, but solid mechanisms for the cell-to-cell transfer remain scarce. Dickey and colleagues speculated that chaperones—owing to their intimate relations with misfolded proteins—could play a role. One such chaperone, the ubiquitously expressed Hsc70, is known to associate with proteins implicated in neurodegenerative disease (see Sarkar et al., 2008Pemberton et al., 2011; and Monsellier et al., 2015). The chaperone carries out its myriad functions with the help of various co-chaperones, which include several DnaJ proteins with which the chaperone forms complexes. These co-chaperones dictate whether Hsc70 refolds, disaggregates, degrades, or exports its protein cargo. The researchers speculated that DnaJC5, identified as part of the exocytosis machinery, could play a role in releasing proteins from the cell. 

To test this idea, first author Sarah Fontaine, now at the University of Kentucky in Lexington, and colleagues overexpressed DnaJC5 and tau in HEK293 cells, washed them to remove traces of serum proteins floating in the medium, and then measured levels of tau released into serum-free medium after six hours. They found that overexpression of the co-chaperone boosted tau release approximately fivefold, while intracellular tau levels dropped. DnaJC5 upped the release of endogenous tau from M17 neuroblastoma cells as well as brain slice cultures made to overexpress the co-chaperone. Conversely, brain slice cultures from mice lacking DnaJC5 released less tau than wild-type slices, indicating that DnaJC5 played a role in the normal export of tau from neurons. In addition, the researchers found that DnaJC5 ousted the normal and disease-associated mutants of tau, α-synuclein, and TDP-43, but not the polyQ-25 fragment of the Huntingtin protein.

Using confocal microscopy of primary neurons overexpressing DnaJC5, the researchers observed the co-chaperone comingling with both Hsc70 and tau in the presynaptic region. The researchers used immunoprecipitation to confirm that DnaJC5 formed a tripartite complex with tau and Hsc70, and treatment of cells with an inhibitor that blocked the association between chaperone and co-chaperone greatly reduced the release of tau. Tau export was also stymied when the researchers knocked down Hsc70 expression, or expressed a mutant form of tau that cannot bind to Hsc70. Interestingly, overexpression of other DnaJ proteins failed to facilitate tau secretion, indicating that DnaJC5 serves a specific role in ushering Hsc70 cargoes out of the cell. Overexpression of DnaJC7 actually blocked tau export and promoted its degradation instead.

How were DnaJC5-Hsc70 cargo exiting the cells? The researchers speculated that the SNARE protein SNAP-23, which is ubiquitously expressed and facilitates exocytosis, could play a role. Indeed, when they knocked down this SNARE in HEK293 or neuroblastoma cells, the DnaJC5-mediated release of tau, α-synuclein, and TDP-43 plummeted.

Unlike the classic neuronal specific SNARE SNAP-25, which facilitates neurotransmitter release from presynaptic vesicles, SNAP-23 is expressed in all cells and cannot completely rescue SNAP-25 function. The researchers speculated that release of proteins via SNAP-23 represents an alternative secretion pathway that serves to relieve the cell of accumulating protein aggregates that overwhelm the lysosomal degradation machinery. Rather than degrading such proteins, perhaps the cell reroutes trash-filled vesicles to the cell surface, they proposed.

How and where in the cell the protein cargos enter this export pathway is unknown. However, a recent study led by Yihong Ye at the National Institutes of Health in Bethesda, Maryland, described a pathway called MAPS (misfolding associated protein secretion), in which the deubiquitinase and chaperone USP19 bound to misfolded proteins at the ER membrane, which were then whisked away in late endosomes bound for the cell surface (see Jun 2016 news). The study was carried out in non-neuronal cells. Whether this and the DnaJC5-Hsc70 pathway are somehow connected remains to be tested, although Ye speculated that they are likely one and the same. “If so, it would be interesting to dissect the functional interplays between USP19 and Hsc70 in this novel and intriguing protein quality control pathway,” he wrote to Alzforum.  

Could these protein secretion pathways mediate the spread of misfolded proteins in the brain? Much work needs to be done to chip away at that question. Marc Diamond of the University of Texas Southwestern Medical Center in Dallas pointed out that the researchers looked at the release of tau monomers, despite findings from his lab indicating that tau trimers are the smallest unit of propagation (see Apr 2015 news). As with the MAPS study, they also did not examine aggregated proteins specifically, he added.

Lary Walker of Emory University in Atlanta found the data intriguing. “Fontaine and colleagues show that the DnaJ/Hsc70 chaperone complex is indeed yet another non-canonical pathway by which cells can expel disease-related proteins,” he wrote to Alzforum. “Along with the recent paper by Lee et al., the findings underscore the importance of pursuing the sometimes-unorthodox biology of cells that are threatened by proteopathic stress.”—Jessica Shugart


  1. This is another interesting paper to help us understand the extracellular release of tau protein from cells. In their report, expressed tau protein appears to be secreted from the cytoplasm in a chaperone complex- and SNAP-dependent manner. Although the majority of the experiments were performed using HEK293T cells, a similar mechanism may be at work in neurons and glia as the authors suggested. This is different from the recent report from Yihong Ye’sgroup, which reported direct secretion of GFP or GFP-tau from the cell in an endoplasmic reticulum-driven manner in HEK293T cells. Thus there appear to be multiple pathways of protein secretion. It is noteworthy that several heat shock protein molecules are consistently detected in exosomes, so it is possible that these also play a role in exosome secretory pathway. One caution is that transient expression of recombinant molecules in HEK293T cells will result in some cytotoxicity, which results in the non-specific release of recombinant molecules. I would like to see more discussion of DnaCJ5 for its clinical relevance to tauopathy, which is missing in the discussion. Overall these are carefully performed experiments in mostly non-neuronal cell lines and would require future validation in more physiological conditions.

  2. The study by Fontaine et al. reports an unconventional protein secretion mechanism that eliminates misfolded proteins from the cytosol. Intriguingly, several neurodegenerative disease-associated aberrant proteins, such as tau and α-synuclein, can be released into the cell exterior through this mechanism. This was observed in regular 293T cells as well as in neurons. It has been postulated previously that tau protein might exit the cell using an exosome-mediated unconventional protein secretion pathway, although this model lacks supporting evidence. By contrast, the authors here present strong evidence, demonstrating that most tau and α-synuclein are actually released from the cell by a pathway distinct from exosome-mediated secretion, and that it depends on the cytosolic heat shock protein Hsc70, its co-chaperone DnaJC5, and a SNARE protein. These results suggest the possibility that misfolded cargo proteins are initially recruited by the chaperone and then use a vesicle intermediate to exit the cell. Although the authors did not know exactly what kind of vesicle intermediate carries out this function, or the precise identity of this pathway, the working hypothesis proposed (Fig. 4) is surprisingly similar to the model reached by our recent study, published in Nature Cell Biology (Lee et al., 2016). In our study, we reported a pathway named MAPS that is dedicated for exporting misfolded proteins including α-synuclein from the cell. This pathway uses the ER-associated chaperone/deubiquitinase USP19 to recruit misfolded proteins to the ER surface, putting them into late endosomes that are parked on the ER for secretion. Although the experimental conditions used in these two studies are different (Fontaine et al. used serum-free medium to grow cells while analyzing secretion, whereas we used complete medium), it is very likely that Fontaine and colleagues are dealing with the same pathway as MAPS. If so, it would be interesting to dissect the functional interplays between USP19 and Hsc70 in this novel and intriguing protein quality control pathway.


    . Unconventional secretion of misfolded proteins promotes adaptation to proteasome dysfunction in mammalian cells. Nat Cell Biol. 2016 Jun 13; PubMed.

Make a Comment

To make a comment you must login or register.


News Citations

  1. Can’t Degrade That Pesky Misfolded Protein? Push It Off the MAPS
  2. Tau Triple Threat: Do Trimers Make Bad Seeds?

Paper Citations

  1. . Two motifs within the tau microtubule-binding domain mediate its association with the hsc70 molecular chaperone. J Neurosci Res. 2008 Sep;86(12):2763-73. PubMed.
  2. . Hsc70 protein interaction with soluble and fibrillar alpha-synuclein. J Biol Chem. 2011 Oct 7;286(40):34690-9. PubMed.
  3. . Molecular interaction between the chaperone Hsc70 and the N-terminal flank of huntingtin exon 1 modulates aggregation. J Biol Chem. 2015 Jan 30;290(5):2560-76. Epub 2014 Dec 10 PubMed.
  4. . Unconventional secretion of misfolded proteins promotes adaptation to proteasome dysfunction in mammalian cells. Nat Cell Biol. 2016 Jun 13; PubMed.

Further Reading

No Available Further Reading

Primary Papers

  1. . DnaJ/Hsc70 chaperone complexes control the extracellular release of neurodegenerative-associated proteins. EMBO J. 2016 Jul 15;35(14):1537-49. Epub 2016 Jun 3 PubMed.