Name of paper: “A common mechanism of proteasome impairment by neurodegenerative disease-associated oligomers”
From our discussions regarding ubiquitation in BCM441 a couple weeks ago, we began to understand its importance for the degradation of misfolded proteins, along with its role in the removal of aggregates and damaged organelles through processes such as autophagy and mitophagy. We also discussed the role of E3 ubiquitin-protein ligase (Parkin) mutations in disrupting this necessary ubiquitination process, which led to the accumulation of mitochondrial damage and protein aggregates directly linked to neurodegenerative diseases such as Parkinson’s. Therefore, these discussions, along with previous studies on ubiquitation, reason that neurodegenerative diseases mainly result from the idea that proteins that normally should be degraded do not do so after misfolding occurs, causing neuronal pathologies. Moreover, it has been demonstrated that the aggregated proteins themselves become capable of interacting and interfering with proteasomes, which in turn further impairs ubiquitation and contributes to these pathologies (Bence et al., 2001). Though, despite these findings, it has not been demonstrated what types of aggregates specifically impair proteasome function, and a canonical understanding of why and how proteasomes are widely disrupted in neurodegenerative diseases has not been established (Thibaudeau et al., 2018). As a result, Thibaudeau et al. attempt to address these issues and introduce a mechanism that characterizes proteasome dysfunction caused by misfolded oligomers in Parkinson, Alzheimer and Huntington neurodegenerative diseases.
Ubiquitinated proteins are recognized for degradation by the 26S proteasome (Thibaudeau et al., 2018). It is made up of a 20S core particle capped on one or both ends by the 19S ATPase regulatory particle (Thibaudeau et al., 2018). The multistep process in which proteins get degraded starts with the 19S regulatory particle binding ubiquitinated substrates and opening a substrate entry gate in the 20S (Thibaudeau et al., 2018). It then unfolds the substrates by translocating them into a 20S catalytic chamber composed of four heteroheptameric rings arranged in a α7-β7-β7-α7 fashion where they get degraded (Groll et al., 2000). Most notably, the opening of the gate requires the C-terminal HbYX motif of the 19S ATPases to bind on top of the 20S, which alludes to the importance of 20S proteasome gate regulation for proper functionality (Thibaudeau et al., 2018). Therefore, the authors hypothesize that toxic oligomers specifically impair HbYX motif-dependent gate opening through a shared mechanism, which in turn blocks protein degradation and disrupts proteasome functionality.
Considering the inconsistency in the literature regarding the effects of aggregates on proteasome functionality, Thibaudeau et al. begin by generating populations of protein aggregates from either amyloid-β 1–42 (Aβ), α-Syn, or huntingtin exon 1 with a polyQ-expansion (Htt-53Q) (which correspond to those found in Alzheimer’s, Parkinson’s and Huntington’s respectively) and observed their impacts on proteasome functionality. Indeed, they found that the 20S proteasome ability to hydrolyze fluorogenic peptide substrates was impaired (Thibaudeau et al., 2018). Moreover, upon separation of soluble oligomers from insoluble ones, the authors found that only the former had an impact on proteasome function (in a concentration-depedent manner) (Thibaudeau et al., 2018). Considering this finding, along with the fact that Aβ, α-Syn, and Htt-53Q monomers were in fact degraded by the proteasomes, the authors inferred that proteasome impairment might be due to the structure of the soluble oligomers. Therefore, the authors replicated the generation of a poly-clonal anti-oligomer antibody (A11) that specifically recognizes some types of protein oligomers regardless of primary structure, initially developed by Kayed et al. Following the incubation of these antibodies with the oligomers, they were surprised that all three soluble oligomer types displayed strong A11 antibody binding (Thibaudeau et al., 2018), which hints to structural similarity between the oligomers.
The authors then went on to address whether proteasome impairment was due to oligomer size rather than structure and if shared A11 reactivity was merely a coincidence (Thibaudeau et al., 2018). Thus, they generated both 200–400 kDa A11+ Aβ oligomers and larger 700 kDa ones, and found that both retained the ability to impair substrate hydrolysis by all three active sites of the proteasome but to a lesser degree by the latter. This was expected based on results from a previous study that demonstrate how protofibrils (elongated clusters of cells that grow into fibrils) form when oligomers bind to one another in forming a chain of oligomers, which sterically block internal oligomers but not terminal ones from interacting with the proteasome (Yang et al., 2017). Before moving on, the authors also wanted to check whether the structural epitope of the A11 antibody on the Aβ oligomer was necessary for proteasome inhibition (Thibaudeau et al., 2018). And indeed, when the antibody was bound to the oligomers, proteasome function was rescued; however, when an antibody was bound to the N-terminus instead (as a control), functionality was disrupted. These results suggested that a specific A11 epitope site is essential for proteasome dysfunction induced by the soluble oligomers (Thibaudeau et al., 2018).
In order to characterize the mechanism of proteasome impairment, the authors adopt an established protocol from Barghorn et al. to generate physiologically relevant Aβ*56 oligomers in which the A11+ epitope was maintained (consistent with those purified from human brain tissue in Alzheimer’s) (Thibaudeau et al., 2018). The authors then observed whether these oligomers could directly interact with the 20S proteasome. They found that the oligomers were clearly co-migrating with the 20S proteasome visualized by Native-PAGE gel (Thibaudeau et al., 2018). Therefore, this confirmed that these oligomers physically interact with the proteasome which in turn leads to its dysfunction.
From here, the authors went on to address whether the impaired 20S proteasome function was due to impaired substrate entry through the gate or by allosterically impairing the active sites (Thibaudeau et al., 2018). To do so, they compared the impacts of the three A11+ oligomers (Aβ*56, α-Syn, and Htt-53Q) on WT proteasome and mutant proteasome α3ΔN which has a constitutively open gate. They found that only the WT proteasome was impaired, but not the mutant; suggesting that the oligomers require a functioning gate to impair proteasome activity (Thibaudeau et al., 2018). Further, they found that all three catalytic sites were impaired in the WT as opposed to only one or two for example, which also contributes to the suggestion that substrate access altogether might be inhibited. In addition, the authors perform a substrate saturation curve on the WT proteasome with and without the Aβ*56 oligomers. Through the use of nonlinear regression and Michaelis–Menten kinetic analysis, they found that the oligomers resulted in a decrease in Vmax and an increase in Km, which are characteristic of allosteric (non-active site) inhibition. Thus, they suggested that all three oligomers impair the proteasome with the same allosteric mechanism, since they all require the closable gate to impair function (Thibaudeau et al., 2018).
As previously mentioned, the opening of the 20S proteasome gate requires the C-terminal HbYX motif of the 19S ATPases to bind on top of the 20S, which induces a conformational change of the 20S α- subunits and stabilizes the open state of the N-terminal gating residues (Thibaudeau et al., 2018). However, the 11S family of proteasome activators (PA28α/β and PA26) lack the HbYX motif of the 19S ATPases and open the gate through an 11S activation loop, which is independent of the HbYX mechanism (Whitby et al., 2000). Thus, the authors elected to test which mechanism of these two is the one impaired by the disease-related oligomers (Thibaudeau et al., 2018). Upon addition of the A11 oligomers with saturating amounts of PA26 activator in one trial and PA28α/β in another, they found that none of the oligomers were able to impair proteasome function in either case. These results were further confirmed via kinetic analysis, suggesting that the A11 oligomers bind to the 20S proteasome at a location separate from the 11S proteasome activators (Thibaudeau et al., 2018).
Next, the authors tested if A11+ Aβ*56 oligomers could impair ubiquitin-dependent (Ub4(lin)-GFP-35) protein degradation by purified human 26S proteasomes (Thibaudeau et al., 2018). Indeed, they found that protein degradation was severely impaired. Considering that this process requires ATP-dependent unfolding and injection into the 20S core, this data suggested that the oligomers impair the HbYX mechanism of gate opening (Thibaudeau et al., 2018). Through native-PAGE analysis, the authors confirm that this impairment was not due to dissociation of the 19S and 20S, further strengthening their results of an allosteric (and not competitive) inhibition. To further validate their results, the authors prepared mouse brain lysates that resemble the complex heterogeneous cellular environment in vivo. Indeed, the oligomers nearly completely impaired proteasome function even in this complex environment (Thibaudeau et al., 2018).
Considering that the 26S proteasome adopts multiple conformations during the ATP hydrolysis cycle by 19S ATPase, the authors examine the impact of the oligomers on these conformations. It is known that hydrolysable ATP stabilizes the closed gate conformation, while non-hydrolyzing ATP analogues better stabilize the open gate form (Thibaudeau et al., 2018). Surprisingly, the oligomers impaired the normal physiological (with ATP) state of the 26S but not the synthetically opened state (using ATPγS analog). These results support the inhibition of the HbYX-dependent conformational changes that lead to gate opening, as opposed to inhibition of the proteasome after the gate is opened (Thibaudeau et al., 2018). Moreover, the authors tested this hypothesis by adding the A11 oligomers in the presence and absence of an established gate opening peptide Rpt5. The authors found that the oligomers did impair gate opening by Rpt5; however, they were not as effective when high concentrations of Rpt5 were added (1mM). These confirm the impairment of HbYX-dependent gate opening, but what was also important is that the HbYX peptide could overcome impairment by the oligomers at higher concentrations (Thibaudeau et al., 2018).
To summarize, the authors propose the following mechanistic model of how A11 oligomers inhibit proteasome function: They bind to the outer surface of the proteasome (specifically the α-subunits along the C2 axis); this in turn leads to stabilization of the closed-gate conformation and prevention of HbYX-dependent gate opening (as opposed to no impairment of gate opening by the 11S proteasome activators). The model proposes opposing allosteric controls between two allosteric modulators (the HbYX motif and the oligomers) that bind to distinct sites on the 20S proteasome, with the motif having an activating effect on the proteasome vs. the oligomers having an impairing effect (Thibaudeau et al., 2018). Ultimately, the study demonstrates that proteasome impairment achieved by Alzheimer, Parkinson and Huntington oligomers are due to 3D shape, as opposed to primary structure and sequence.
Overall, the study is very important in that it proposes a mechanism in which soluble aggregates from three very dangerous diseases all share a common way to impair proteasome function. It is also high yield in the sense that results were consistent when mimicking physiological conditions. However, many questions remain with regard to future studies and therapeutic applications. For example, what levels of physiological oligomers are sufficient to impair activity? Also, considering that the HbYX motif and non-hydrolyzing ATP stabilize the open-gate form, would it be possible to prevent the hydrolysis of ATP to ADP and thus favor the open-gate form and retain normal function? If these questions are addressed, translational interventions may have the potential to restore normal degradation of misfolded proteins, and decrease the severity of the neurodegenerative diseases discussed in this study.
- Barghorn, Stefan, Volker Nimmrich, Andreas Striebinger, Carsten Krantz, Patrick Keller, Bodo Janson, Michael Bahr, et al. 2005. “Globular Amyloid Beta-Peptide Oligomer – a Homogenous and Stable Neuropathological Protein in Alzheimer’s Disease.” Journal of Neurochemistry 95 (3): 834–47. https://doi.org/10.1111/j.1471-4159.2005.03407.x.
- Bence, N. F., R. M. Sampat, and R. R. Kopito. 2001. “Impairment of the Ubiquitin-Proteasome System by Protein Aggregation.” Science (New York, N.Y.) 292 (5521): 1552–55. https://doi.org/10.1126/science.292.5521.1552.
- Groll, Michael, Monica Bajorek, Alwin Köhler, Luis Moroder, David M. Rubin, Robert Huber, Michael H. Glickman, and Daniel Finley. 2000. “A Gated Channel into the Proteasome Core Particle.” Nature Structural & Molecular Biology 7 (11): 1062–67. https://doi.org/10.1038/80992.
- Kayed, Rakez, Elizabeth Head, Jennifer L. Thompson, Theresa M. McIntire, Saskia C. Milton, Carl W. Cotman, and Charles G. Glabe. 2003. “Common Structure of Soluble Amyloid Oligomers Implies Common Mechanism of Pathogenesis.” Science (New York, N.Y.) 300 (5618): 486–89. https://doi.org/10.1126/science.1079469.
- Thibaudeau, Tiffany A., Raymond T. Anderson, and David M. Smith. 2018. “A Common Mechanism of Proteasome Impairment by Neurodegenerative Disease-Associated Oligomers.” Nature Communications 9 (1): 1097. https://doi.org/10.1038/s41467-018-03509-0.
- Whitby, F. G., E. I. Masters, L. Kramer, J. R. Knowlton, Y. Yao, C. C. Wang, and C. P. Hill. 2000. “Structural Basis for the Activation of 20S Proteasomes by 11S Regulators.” Nature 408 (6808): 115–20. https://doi.org/10.1038/35040607.
- Yang, Ting, Shaomin Li, Huixin Xu, Dominic M. Walsh, and Dennis J. Selkoe. 2017. “Large Soluble Oligomers of Amyloid β-Protein from Alzheimer Brain Are Far Less Neuroactive Than the Smaller Oligomers to Which They Dissociate.” The Journal of Neuroscience: The Official Journal of the Society for Neuroscience 37 (1): 152–63. https://doi.org/10.1523/JNEUROSCI.1698-16.2016.