Close

HSP110 dependent HSP70 disaggregation machinery mediates prion-like propagation of amyloidogenic proteins in metazoa

Jessica Tittelmeier, Carl Alexander Sandhof, Heidrun Maja Ries, Silke Druffel-Augustin, Axel Mogk, Bernd Bukau, Carmen Nussbaum-Krammer

Posted on: 27 October 2019 , updated on: 16 December 2019

Preprint posted on 9 October 2019

Article now published in The EMBO Journal at http://dx.doi.org/10.15252/embj.2019103954

Two sides of the same coin: Hsp110 activity protects against amorphous protein aggregation, but promotes amyloid aggregation and toxicity in C. elegans models

Selected by Tessa Sinnige

Introduction

Living cells and organisms depend on the function of molecular chaperones to ensure correct protein folding. In a variety of disease conditions and even during normal ageing, however, proteins can misfold and aggregate, posing a challenge to cellular protein homeostasis and leading to toxicity. For example, Alzheimer’s, Parkinson’s and Huntington’s diseases are all characterised by the deposition of protein aggregates in the brain, as are prion diseases. Yeast cells contain a particular chaperone, the Hsp104 disaggregase, which can solubilise existing aggregates. This activity may be beneficial, yet has also been shown to stimulate the propagation of yeast prions by generating smaller fibril fragments that can spread from cell to cell. Metazoa do not carry a Hsp104 homologue, but instead employ a complex of three chaperones, Hsp70/Hsp40/Hsp110, as a disaggregase that was shown to act on amorphous and amyloid-like aggregates in vitro (1, 2). In this preprint, the role of this disaggregase machinery in amyloid aggregation and the associated toxicity in vivo is investigated in the animal model C. elegans.

 

Key results

The authors of this preprint make use of well-established C. elegans models expressing the Parkinson’s disease related protein α-synuclein, and a fragment of the Huntingtin’s disease protein, polyglutamine, in the body wall muscle cells. Aggregation of the proteins in this tissue is associated with cellular toxicity that can be measured as a decline in the worms’ motility. The authors use an elegant system to achieve tissue-specific knock-down of Hsp110, which is part of the disaggregase machinery, to investigate the effects on protein aggregation and toxicity without causing widespread side effects.

First, they test the role of Hsp110 in amorphous aggregation using a luciferase reporter. The expression of luciferase in the worms leads to the formation of amorphous aggregates at higher temperatures, which disappear when the worms are returned to ambient temperatures. However, when Hsp110 is knocked down, a significant number of luciferase aggregates persists, confirming the role of Hsp110 in disaggregation. On the other hand, the knock-down of Hsp110 in the α-synuclein and polyglutamine worms results in a reduction in the number of foci and restores the motility defect. This result, although perhaps surprising at first sight, can be explained by the generation of smaller aggregate seeds catalysed by Hsp110, which overall leads to an amplification of aggregate formation by secondary processes. In support of this notion, the authors furthermore show that Hsp110 knock-down reduces the spread of α-synuclein to neighbouring tissue in a recently established C. elegans model (3), which is also thought to be mediated by small aggregate fragments.

Figure 1A and C from the preprint, showing that Hsp110 knock-down (HPI and HPII) reduces α-synuclein foci formation in C. elegans muscle cells, and restores the motility. Reproduced under a CC BY-NC-ND 4.0 license.

 

Although knocking down Hsp110 might sound like a good idea to halt disease-associated protein aggregation, this protein is critical to stimulate the ATPase activity of Hsp70, which in turn ensures the proper folding of a wide variety of client proteins. As such, a decrease in Hsp110 levels may interfere with protein folding and cellular homeostasis in general. Indeed, the authors find increased misfolding of a temperature-sensitive mutant muscle protein, accompanied by decreased motility, upon Hsp110 knock-down. Moreover, they find a faster decline in motility during ageing of wild-type animals when Hsp110 levels are lowered, consistent with an increased burden on the protein homeostasis capacity with age. In the case of the α-synuclein worms, this results in a loss of the protective effect of the Hsp110 knock-down beyond a certain age. Altogether, the ‘good’ and ‘bad’ activities of Hsp110 are thus two sides of the same coin, and the outcome will depend on the type of proteins that put the most pressure on the protein homeostasis system at a given time.

 

Why I chose this preprint

Disclaimer: I also work on protein aggregation using C. elegans models! I find this preprint a really great example of the beauty of this model organism to study the interplay between protein aggregation, toxicity and cellular quality control mechanisms. The findings are exciting because they shed more light on the function of Hsp110 and the metazoan disaggregase machinery. Furthermore, they show that amorphous and amyloid-like protein aggregation are two fundamentally distinct processes that need to be dealt with in different ways. Finally, the notion that chaperone activity can be ‘bad’, i.e. enhance the aggregation and spread of toxic proteins, is starting to gain more attention and will be very important when trying to target chaperones as a therapeutic strategy.

 

Questions

Can these results help us to understand the nature of the α-synuclein foci and the species that are capable of spreading? It has previously been shown that the majority of α-synuclein inclusions in one of these C. elegans models appear mobile and not amyloid-like (4, 5). Also the supposedly fibrillar nature of Lewy Bodies has recently come into question (6). Would the involvement of Hsp110 imply that α-synuclein inclusions must contain at least some amyloid fibrils, or could the disaggregase machinery also act on more disordered (intermediate) states?

Would an intervention to reduce Hsp110 activity be a promising therapeutic approach for amyloid-associated diseases? The effect on the general protein folding capacity may lead to negative consequences, but could this be a viable compromise in the context of a system challenged by amyloidogenic proteins?

 

References

  1. Gao X, Carroni M, Nussbaum-Krammer C, Mogk A, Nillegoda NB, Szlachcic A, Guilbride DL, Saibil HR, Mayer MP and Bukau B (2015) Human Hsp70 Disaggregase Reverses Parkinson’s-Linked α-Synuclein Amyloid Fibrils. Mol. Cell 59: 781–793
  2. Nillegoda NB, Kirstein J, Szlachcic A, Berynskyy M, Stank A, Stengel F, Arnsburg K, Gao X, Scior A, Aebersold R, Guilbride DL, Wade RC, Morimoto RI, Mayer MP and Bukau B (2015) Crucial HSP70 co-chaperone complex unlocks metazoan protein disaggregation. Nature 524: 247–251
  3. Sandhof CA, Hoppe SO, Druffel-Augustin S, Gallrein C, Kirstein J, Voisine C and Nussbaum-Krammer C (2019) Reducing INS-IGF1 signaling protects against non-cell autonomous vesicle rupture caused by SNCA spreading. Autophagy 0: 1–22
  4. Van Ham TJ, Thijssen KL, Breitling R, Hofstra RMW, Plasterk RHA and Nollen EAA (2008) C. elegans model identifies genetic modifiers of α-synuclein inclusion formation during aging. PLoS Genet. 4: e1000027
  5. Laine RF, Sinnige T, Ma KY, Haack AJ, Poudel C, Gaida P, Curry N, Perni M, Nollen EAA, Dobson CM, Vendruscolo M, Kaminski Schierle GS and Kaminski CF (2019) Fast Fluorescence Lifetime Imaging Reveals the Aggregation Processes of α-Synuclein and Polyglutamine in Aging Caenorhabditis elegans. ACS Chem. Biol. 14: 1628–1636
  6. Shahmoradian SH, Genoud C, Graff-Meyer A, Hench J, Moors T, Schweighauser G, Wang J, Goldie KN, Suetterlin R, Castano-Diez D, Perez-Navarro P, Huisman E, Ipsen S, Ingrassia A, de Gier Y, Rozemuller AJM, Da Paepe A, Erny J, Staempfli A, et al. (2019) Lewy pathology in Parkinson’s disease consists of a crowded organellar membranous medley. bioRxiv 2019: 10.1101/137976

 

doi: https://doi.org/10.1242/prelights.14845

Read preprint (No Ratings Yet)

Author's response

Carmen Nussbaum shared

  1. Can these results help us to understand the nature of the α-synuclein foci and the species that are capable of spreading? It has previously been shown that the majority of α-synuclein inclusions in one of these C. elegans models appear mobile and not amyloid-like (4, 5). Also the supposedly fibrillar nature of Lewy Bodies has recently come into question (6). Would the involvement of Hsp110 imply that α-synuclein inclusions must contain at least some amyloid fibrils, or could the disaggregase machinery also act on more disordered (intermediate) states?

The HSP70 disaggregation machinery can act on different kind of protein aggregates and we have no evidence that its activity would differ between amorphous or amyloidogenic substrates. However, our study showed that the disaggregation activity of Hsp110 towards amyloidogenic substrates has fundamentally different consequences as compared to amorphous aggregates: amyloid aggregates disappear upon Hsp110 inhibition whereas amorphous aggregates are stabilized.

From our point of view, the most obvious explanation for these contrasting outcomes of the same activity is the self-templating nature of amyloidogenic substrates. Prion-like propagation is mediated by the recruitment and incorporation of monomers into the amyloid fibril. We previously showed that disaggregation of recombinant α-synuclein fibrils leads to smaller fragments [1], i.e. more fibril ends, which would allow a more efficient recruitment of subunits. In this respect, the Hsp70 disaggregation activity resembles the role of the Hsp104/Hsp70 chaperone system in the propagation of yeast prions [2]. It could also fragment larger fibrils and generate prion “propagons”, smaller sized units that are more seeding-competent and more efficiently transmitted.

We agree that the majority of α-synuclein certainly does not form large amyloid inclusions in C. elegans (nor in most other model systems or even Lewy bodies of PD patients). Nevertheless, α-synuclein is able to adopt multiple conformations and form amorphous aggregates, different fibrillar polymorphs, as well as smaller oligomeric structures. The nature of the toxic and spreading-competent α-synuclein species (which might be an even distinct protein species) is despite intensive research still highly controversial. Our study suggests that the disaggregation of amyloidogenic substrates produces protein species that are more toxic and transmissible. We do not know the exact size of these units, but they do not correlate with large immobile inclusions. This is an important question, of course, and we are currently characterizing the products of the disaggregation reaction in more detail.

 

Would an intervention to reduce Hsp110 activity be a promising therapeutic approach for amyloid-associated diseases? The effect on the general protein folding capacity may lead to negative consequences, but could this be a viable compromise in the context of a system challenged by amyloidogenic proteins?

We believe that the modulation of specific components of the Hsp70 disaggregation machinery could be a promising therapeutic approach for the treatment of amyloid diseases, despite the observed negative effects on cellular proteostasis. Negative side effects could be minimized in several ways, e.g. via a transient inhibition of disaggregation activity by temporary drug administration. This could be sufficient to breach the toxic replication cycle of amyloid aggregates. In C. elegans, there is only one HSP110-type co-chaperone, which is advantageous for investigating its function, whereas in humans, there are three HSP110 isoforms (HSP105 (HSPH1), APG2 (HSPH2), and APG1 (HSPH3)). Targeting only one member would probably reduce disaggregation activity rather than abolish it, leading to fewer side effects. And this may still be sufficient to shift the balance of amyloid propagation towards resolubilization or degradation. Another promising approach could be to block the binding of a specific J-domain protein (DNAJ), such as DNAJB1, to amyloid fibers. The DNAJ superfamily conveys the substrate specificity of the Hsp70 system and humans have 41 different J-domain proteins. This strategy could facilitate to specifically prevent the disaggregation of amyloids and at the same time leave the (re-)folding and disaggregation of other substrates unaffected.

 

  1. Gao X, Carroni M, Nussbaum-Krammer C, et al. Human Hsp70 Disaggregase Reverses Parkinson’s-Linked alpha-Synuclein Amyloid Fibrils. Mol Cell. 2015 Sep 3;59(5):781-93. doi: 10.1016/j.molcel.2015.07.012. PubMed PMID: 26300264.
  2. Chernoff YO, Lindquist SL, Ono B, et al. Role of the chaperone protein Hsp104 in propagation of the yeast prion-like factor [psi+]. Science. 1995 May 12;268(5212):880-4. doi: 10.1126/science.7754373. PubMed PMID: 7754373; eng.

Have your say

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Sign up to customise the site to your preferences and to receive alerts

Register here
Close