Could a Faulty Ubiquitin Trigger Amyloid and Tau Deposits?

Scientists believe that in late-onset Alzheimer’s disease, Aβ clearance slows, causing the protein to accumulate in plaques. In the September 22 Nature Communications, researchers led by Michael Glickman and Yaron Fuchs at the Technion Israel Institute of Technology, Haifa, laid the blame for this sluggish clearance on a mis-transcribed version of the ubiquitin B protein. UBB+1 results from a frameshift reading error, and contains an extra 19 amino acids in its tail, rendering it unable to tag proteins for degradation. It is known to accumulate in AD brains.

  • Mis-transcribed ubiquitin UBB+1 appears early in AD.
  • In human three-dimensional neuronal cultures, UBB+1 alone triggers plaques and tangles.
  • In three-dimensional cultures expressing AD mutations, silencing UBB+1 curbs plaques and tangles.

The authors found that UBB+1 competed with normal ubiquitin for binding to an enzyme that controls disposal of the amyloid precursor protein APP. As a result, UBB+1 slowed APP degradation, allowing Aβ to build up. In human three-dimensional neuronal cultures, overexpressing UBB+1 was sufficient to trigger plaques and tangles. Conversely, silencing spontaneous UBB+1 in cultures that expressed familial AD mutations prevented amyloid and tau pathology. “Our findings provide direct evidence that the ubiquitin proteasome system is a crucial player in Alzheimer’s disease. This opens up a plethora of new targets for therapeutic intervention,” Glickman told Alzforum.

Nico Dantuma at the Karolinska Institute, Stockholm, agreed. “The most exciting part of the study [is] that they were able to show that selective depletion of endogenously produced UBB+1 reduces pathology in a cellular model. That is a major achievement,” he wrote to Alzforum (comment below).

Plaque Promoter. In cultured human neurons (green, left), the presence of UBB+1 alone (right) spurs nearly as many plaques (red) as do familial AD mutations (middle; quantification at right). [Courtesy of Maniv et al., Nature Communications.]

UBB+1 was first identified by the late Fred van Leeuwen at Maastricht University in The Netherlands, a co-author on the current paper (May 1998 webinar; van Leeuwen et al., 1998; Dec 2004 conference news). The aberrant transcript is caused by RNA polymerase slipping on the underlying DNA template, adding a single base at the end of the 76 amino acid coding region. This eliminates the normal stop codon and lengthens ubiquitin B. Van Leeuwen found UBB+1 in all AD brains he examined, but it was unclear if it contributed to the disease. Curiously, expressing UBB+1 in an amyloidosis mouse model resulted in less plaque buildup (Verheijen et al., 2018). With no good way to model the effect of UBB+1 in human cells, interest in the peptide waned.

Glickman and colleagues set out to explore UBB+1’s effects in a human system. Joint first authors Inbal Maniv, Mahasen Sarji, and Anwar Bdarneh established three-dimensional Matrigel neuronal cultures following the protocol developed by Doo Yeon Kim and Rudolph Tanzi at Harvard Medical School (Oct 2014 news; Kim et al., 2015). When they transfected wild-type neurons with a viral vector carrying UBB+1, the cultures accumulated twice as much aggregated Aβ and phosphorylated tau as control cultures over the next six weeks. Tau deposits contained the paired helical filaments characteristic of neurofibrillary tangles, as seen by immunostaining and electron microscopy. The findings demonstrated that UBB+1 alone was able to kick off AD pathology.

Was UBB+1 necessary for pathology? To test this, the authors made three-dimensional cultures of human neurons carrying familial AD mutations in APP and presenilin 1. Over six weeks, these generated amyloid deposits and tangles. UBB+1 also arose spontaneously, at threefold higher levels than in control cultures. Silencing UBB+1 via RNAi lowered Aβ and p-tau aggregates by 60 percent, demonstrating that the ubiquitin plays a role in protein aggregation.

Next, the authors investigated how. In co-immunoprecipitations, UBB+1 bound ubiquitin C-terminal hydrolase L1 (UCHL1), an enzyme known to promote the ubiquitination and disposal of APP (Zhang et al., 2014). Adding normal ubiquitin to the mix interfered with UBB+1 binding to UCHLI, indicating a competitive relationship. In keeping with this, overexpressing UBB+1 in HEK293 cells caused APP to accumulate. On the other hand, overexpressing UCHL1 in three-dimensional UBB+1 cultures slashed Aβ deposits in half. The results supported the idea that UBB+1 acts through UCHL1.

Early Appearance. In postmortem entorhinal cortex, UBB+1 (red) is barely present at Braak stage 0 (left), goes up at stage 1 (middle), and is abundant at stage 2 (right). [Courtesy of Maniv et al., Nature Communications.]

How soon in AD does UBB+1 accumulate? The authors immunostained postmortem entorhinal cortex samples from brains at Braak stage 0, 1, and 2. Stage 2 entorhinal cortex had significantly more UBB+1 than did stage 0, demonstrating that the dysfunctional ubiquitin appears quite early. In amyloidosis model mice as well, UBB+1 was elevated in the hippocampus by 1 month of age.

The data imply that defects in this proteolytic system occur in AD independent of the underlying genetic cause, Dantuma noted. He thinks silencing UBB+1 could be beneficial. However, because UBB+1 has no catalytic activity, it cannot be pharmaceutically targeted with an inhibitor. “By understanding the reasons for its accumulation, which likely are related to problems in protein quality control and clearance, it may be possible to identify key players that will be more amenable to therapeutic targeting,” Dantuma suggested.

Glickman and colleagues are investigating those ideas. Glickman believes UBB+1 itself could be a therapeutic target. “The groundbreaking success of COVID-19 vaccines has shattered previous barriers associated with RNA-based drugs by addressing concerns about safety and efficacy. Given that UBB+1 fundamentally arises from defective mRNA, employing siRNA approaches emerges as the most obvious strategy to suppress UBB+1,” he wrote to Alzforum.—Madolyn Bowman Rogers

Comments

  1. This is a very interesting paper. It is good to see that the authors brought testing of the UBB+1 hypothesis to the next level. It was also good to see that Fred van Leeuwen, who discovered molecular misreading and UBB+1, was involved in this study before his passing (the paper is dedicated to him).

    The observations in that UBB+1 is already present in early stages of Alzheimer’s, and that in various genetic models for Alzheimer’s disease UBB+1 accumulates, strengthen the idea that accumulation of UBB+1 is a common feature in Alzheimer’s. Combined with earlier studies that showed UBB+1 is a substrate of the ubiquitin-proteasome system (UPS), and that under physiological conditions it is efficiently cleared by the UPS, this work suggests that defects in this proteolytic system may occur in Alzheimer’s independent of the underlying genetic cause, and that these defects may be apparent early in the pathology.

    It has been known for a long time that accumulated UBB+1 is not only cleared by the UPS, but also can impair the functionality of this system. In the current study it is proposed that its interaction with UCH-L1 plays an important role, but I would consider it likely that interference with other ubiquitin-interacting proteins may also contribute to its negative effect in the UPS.

    That accumulation of UBB+1 is a common phenomenon in Alzheimer’s, occurs early in the cellular pathology, and seems to contribute to cellular dysfunction, makes it interesting from a therapeutic point of view. It implies that preventing accumulation of UBB+1 may have a beneficial effect in Alzheimer’s. This is supported by what is, in my opinion, the most exciting part in the study, namely that the authors were able to show that selective depletion of endogenously produced UBB+1 reduces pathology in a cellular model. That is a major achievement.

    Therapeutic targeting of UBB+1 may be hard because it lacks catalytic activity, and therefore efforts should be aimed at reducing levels of UBB+1. An important issue to address is the role of UBB+1 in Alzheimer’s etiology in an animal model, as it has been shown previously that in mice UBB+1 overexpression reduced the load of Aβ, which seems to contrast the current finding (Verheijen et al., 2018). Unfortunately, the authors did not cite or comment on this earlier study, which seems highly relevant for the current finding. While I don’t think that the data necessarily conflict, they may be confusing at first sight, deserve some discussion.

    I think one of the most urgent issues is to understand why UBB+1 accumulates in Alzheimer’s disease. By understanding the reasons for its accumulation, which likely relate to problems in protein quality control and clearance, it may be possible to identify key players that will be more amenable to therapeutic targeting.

    References:

    Verheijen BM, Stevens JA, Gentier RJ, van 't Hekke CD, van den Hove DL, Hermes DJ, Steinbusch HW, Ruijter JM, Grimm MO, Haupenthal VJ, Annaert W, Hartmann T, van Leeuwen FW. Paradoxical effects of mutant ubiquitin on Aβ plaque formation in an Alzheimer mouse model. Neurobiol Aging. 2018 Dec;72:62-71. Epub 2018 Aug 18 PubMed.

    View all comments by Nico Dantuma

  2. I think it is quite intriguing that UBB+1 specifically targets AD pathologies (Aβ and p-tau), but not other pathologies, such as α-synuclein. It would be important to know how UBB+1 selectively modulates APP degradation, but not the degradation of other proteins, through proteasome pathways.

    The authors tested UBB+1 in 3-D iPSC cell models expressing human FAD mutations. I wonder if similar phenotypes can be seen in cell models expressing sporadic AD risk variants, such as ApoE4. I think it would be critical to know if the interaction between UBB+1 and UCHL is detectable in human AD brains during disease development and progression, and whether this interaction is key to the development of Ab and tau aggregates, before we consider UBB+1 as a therapeutic target.

    View all comments by Dongming Cai

  3. This research by Maniv et al. provides further insights into the molecular connection between UBB+1 and the pathology of Alzheimer's disease (AD). UBB+1, a ubiquitin RNA frameshift mutation identified by van Leeuwen and colleagues more than two decades ago, has been the subject of numerous studies investigating its role in AD and other neurodegenerative conditions (van Leeuwen et al., 1998). 

    This study presents an intriguing perspective by exploring whether the initial sequence of events involving proteolysis (specifically, defective ubiquitination) may be a contributing cause, rather than a consequence, of AD. Addressing this question poses a considerable challenge, as dysregulated proteostasis is typically regarded as a downstream event in the early pathogenesis of AD.

    The findings convincingly demonstrate that viral overexpression of UBB+1 in 3D human cell cultures leads to the formation of pathological markers associated with AD. However, there remains some uncertainty regarding the reasons for the accumulation of UBB+1 in the early Braak stages of AD. Further investigation is needed to determine if mouse models exhibiting amyloid plaques and tau aggregation can faithfully replicate observations from human studies. Additionally, it is important to ascertain whether non-neuronal cells within the brain also accumulate UBB+1.

    In terms of mechanistic insights, the study elegantly proposes that UBB+1 competes with ubiquitin for binding to UCHL1, thereby hindering its role in facilitating the proteolysis of amyloid precursor protein (APP). However, questions persist regarding whether proteasome and autophagy activities are directly or indirectly impacted by UBB+1. Notably, the study did not detect an increase in polyubiquitin chains in three-dimensional cultures overexpressing UBB+1, despite known occurrences of ubiquitin inclusions and decreased proteolysis in AD and in AD models.

    This research presents a promising novel target for AD and underscores the multiple levels within the ubiquitin-proteasome system that can be potentially targeted, including ubiquitination, ligases, deubiquitinating enzymes, and proteasome complex levels.

    References:

    van Leeuwen FW, de Kleijn DP, van den Hurk HH, Neubauer A, Sonnemans MA, Sluijs JA, Köycü S, Ramdjielal RD, Salehi A, Martens GJ, Grosveld FG, Peter J, Burbach H, Hol EM. Frameshift mutants of beta amyloid precursor protein and ubiquitin-B in Alzheimer's and Down patients. Science. 1998 Jan 9;279(5348):242-7. PubMed.

    View all comments by Natura Myeku

Make a Comment

To make a comment you must login or register.

References

Webinar Citations

  1. Frameshift Mutants of β Amyloid Precursor Protein and Ubiquitin-B in Alzheimer's and Down's Syndrome Patients

News Citations

  1. Conformation Rules Part 1: News, Common Threads, Debate from San Diego Conference
  2. Alzheimer’s in a Dish? Aβ Stokes Tau Pathology in Third Dimension

Paper Citations

  1. van Leeuwen FW, de Kleijn DP, van den Hurk HH, Neubauer A, Sonnemans MA, Sluijs JA, Köycü S, Ramdjielal RD, Salehi A, Martens GJ, Grosveld FG, Peter J, Burbach H, Hol EM. Frameshift mutants of beta amyloid precursor protein and ubiquitin-B in Alzheimer's and Down patients. Science. 1998 Jan 9;279(5348):242-7. PubMed.
  2. Verheijen BM, Stevens JA, Gentier RJ, van 't Hekke CD, van den Hove DL, Hermes DJ, Steinbusch HW, Ruijter JM, Grimm MO, Haupenthal VJ, Annaert W, Hartmann T, van Leeuwen FW. Paradoxical effects of mutant ubiquitin on Aβ plaque formation in an Alzheimer mouse model. Neurobiol Aging. 2018 Dec;72:62-71. Epub 2018 Aug 18 PubMed.
  3. Kim YH, Choi SH, D'Avanzo C, Hebisch M, Sliwinski C, Bylykbashi E, Washicosky KJ, Klee JB, Brüstle O, Tanzi RE, Kim DY. A 3D human neural cell culture system for modeling Alzheimer's disease. Nat Protoc. 2015 Jul;10(7):985-1006. Epub 2015 Jun 11 PubMed.
  4. Zhang M, Cai F, Zhang S, Zhang S, Song W. Overexpression of ubiquitin carboxyl-terminal hydrolase L1 (UCHL1) delays Alzheimer's progression in vivo. Sci Rep. 2014 Dec 3;4:7298. PubMed.

Further Reading

Primary Papers

  1. Maniv I, Sarji M, Bdarneh A, Feldman A, Ankawa R, Koren E, Magid-Gold I, Reis N, Soteriou D, Salomon-Zimri S, Lavy T, Kesselman E, Koifman N, Kurz T, Kleifeld O, Michaelson D, van Leeuwen FW, Verheijen BM, Fuchs Y, Glickman MH. Altered ubiquitin signaling induces Alzheimer's disease-like hallmarks in a three-dimensional human neural cell culture model. Nat Commun. 2023 Sep 22;14(1):5922. PubMed.

Annotate

To make an annotation you must Login or Register.