Though aggregates of misfolded proteins occur in numerous neurodegenerative disorders, it is not always clear how these aggregates contribute to pathology or why they form at different rates in different patients. In yesterday’s Science online, Richard Morimoto and colleagues at Northwestern University, Evanston, Illinois, proffer a new theory. They suggest that the prevalence of marginally stable proteins encoded in the genome can explain the variability in age of onset and progression of protein folding diseases. Normally, these folding misfits are innocuous, but Morimoto and colleagues report that they can significantly exacerbate the toxicity of polyglutamine-expanded proteins. This could apply in Alzheimer, Parkinson, and Huntington diseases, for example.

The theory is a variation on the old theme that aggregation-prone proteins overtax the cell’s protein folding machinery and send the cell into a destructive spiral (see ARF related Live Discussion). But first authors Tali Gidalevitz, Anat Ben-Zvi, and colleagues turn the tables on this idea. Using the roundworm Caenorhabditis elegans as a model, they show that not only do polyglutamine (polyQ)-expanded proteins turn metastable proteins into flagrantly unstable ones, but that it also works the other way around—the metastable proteins enhance aggregation of proteins with polyQ expansions.

The metastable proteins in question are temperature-sensitive (ts) mutants conferring distinct phenotypes that only emerge at “restrictive” temperatures. At “permissive” temperatures, worms with the mutations appear normal. The ts mutants are good indicators of how healthy a worm’s protein folding machinery is, because the mutated proteins rely heavily on that machinery for proper folding.

To test how polyQ affects folding of ts mutants, the researchers expressed yellow fluorescent protein (YFP) containing various polyQ expansions in worms harboring a paramyosin ts mutant. At the restrictive temperature, this mutation disrupts muscle formation, leading to premature death of both embryos and larvae. Though at the permissive temperature (15 degrees C) these worms normally survive as well as the wild-type, Gidalevitz, Ben-Zvi, and colleagues found that in the presence of YFP with a 40-glutamine stretch (PolyQ40m), embryos behaved as if they were grown at the restrictive temperature (25 degrees C). That is, almost half of the paramyosin ts embryos failed to hatch or even move at 15 degrees when the polyQ was present. Interestingly, expression of the PolyQ40m in a wild-type background had no effect on either embryos or larvae, indicating that the metastable protein and aggregation-prone protein together make a lethal mix.

To show that this is a general phenomenon, and not just a quirk of paramyosin, the researchers tested various other mutants. They found that in the presence of polyQ40, ts mutants of the neuronal protein dynamin-1 caused severe paralysis at normally permissive temperatures, while phenotypes evoked by mutations in homologs of human myosin, perlecan, and ras-1 were also evident at permissive temperatures in the presence of the polyQ YFP.

Given that polyQ40m, which has no effect on wild-type worms, has such a drastic impact on the ts mutants, Gidalevitz and colleagues next asked what effect the ts proteins might have on aggregation-prone polyglutamine proteins. They found that the number of visible polyQ40m aggregates dramatically increased from around 10 or fewer to around 60 if larvae also expressed the temperature-sensitive paramyosin. Importantly, the expression of loss-of-function paramyosin mutants had no effect on polyQ aggregation, indicating that it is not loss of paramyosin activity per se that exacerbates the process, but rather the instability of the ts protein. “Our data identify the presence of marginally stable or folding defective protein in the genetic background of conformational disease as potent extrinsic factors that modify aggregation and toxicity. Given the prevalence of polymorphisms in the human genome, they could contribute to variability of disease onset and progression,” write the authors.

The authors did not test any non-polyQ proteins in their model. It would be interesting to see what effect ts mutants might have on aggregation of amyloid-β, α-synuclein, tau, and other proteins that contribute to neurodegenerative pathologies.—Tom Fagan.

Gidalevitz T, Ben-Zvi A, Ho KH, Brignull HR, Morimoto RI. Progressive disruption of cellular protein folding in models of polyglutamine diseases. Science Express February 9, 2006. Abstract


Make a Comment

To make a comment you must login or register.

Comments on this content

  1. Cellular homeostasis is an exceedingly complex process. Conceptually, one may
    consider two regimes within which the phenomenon operates, extra- and intracellular.
    The extracellular regime requires dynamic responses of the cell to
    external stimuli. The intracellular regime involves metabolic processes that neurologists
    might refer to as “activities of daily living,” those processes that the
    cell must execute continuously to function normally. One of these activities is
    the synthesis and folding of proteins. This activity is highly efficient overall,
    but imperfect. A significant percentage of nascent proteins fold improperly,
    even with the help of folding chaperones, and thus must be “recycled” through
    proteolysis in the proteasome system. What happens if the capacity of the
    protein folding and degradation machinery is exceeded?

    In a paper published on 9 February in Sciencexpress, Morimoto and colleagues
    at Northwestern University address the general question raised above from the
    perspective of diseases of protein aggregation that cause neurodegenerative disorders.
    These diseases include Alzheimer’s, Parkinson’s, prion, and Huntington’s.
    The last disease is an archetypal member of a family of diseases caused
    by an increase in the number of contiguous glutamine residues within normal
    proteins, the “polyglutamine diseases.” A number of pathogenetic mechanisms
    have been postulated to explain the cellular and organismal effects of polyglutamine
    expansion. Morimoto et al. sought mechanistic insights through an
    examination of the effects of simultaneous expression in the worm C. elegans
    of temperature-sensitive (ts) protein mutants and fluorescent proteins bearing
    varying numbers of polyglutamine tracts. The key strategic kernel was the
    study of phenotype at permissive temperatures. In a visually and intellectually
    beautiful set of experiments, effects were observed in sarcomere morphology,
    movement, body shape, egg-laying, and early development. Affected animals
    displayed large numbers of protein aggregates, and evidence was presented that
    proteins expressed in the presence of a polyglutamine-containing protein acquired
    increased protease resistance (one characteristic of amyloids and other
    protein accretions).

    All proteins constantly sample different areas of conformational space. In the ts
    mutants studied, amino acid substitutions alter the energy-dependence of this
    sampling process, allowing the protein to fold into pathologic conformations if
    sufficient activation energy can be obtained (at the non-permissive temperature).
    One intriguing part of the work of Morimoto et al. is the finding that
    polyglutamine tracts can decrease this activation energy, allowing the ts proteins
    to fold pathologically at relatively low temperatures. Importantly, this effect is
    bidirectional. The ts proteins themselves contribute to the self-assembly of
    the polyglutamine proteins, as evidenced by a synergistic increase in aggregate
    number when both proteins are expressed.

    These results suggest two answers to the question posed above, one obvious
    from first principles and one new and unexpected. The obvious answer is that
    the quality and quantity of pathologically folded proteins can overwhelm
    a cell’s homeostatic mechanisms and cause disease. The novel insight is that
    pathologically folded proteins may act relatively nonspecifically to alter folding
    pathways of many other nascent proteins—an effect opposite to that of folding
    chaperones—and thus affect many different physiologic processes.
    How does the work further our understanding of pathogenetic mechanisms of
    polyglutamine diseases in particular, vis-à-vis the beneficial or detrimental effect
    of aggregation (a question posed by Dr. Eddie Koo at UC, San Diego). In
    my opinion, the work increases the complexity of the problem and emphasizes
    the difficulty of ascribing a single pathologic process to a protein-linked disease.
    Specifically, macroscopic aggregates of polyglutamine-containing proteins
    may be “protective” through their ability to sequester toxic protein polymers.
    However, monomers, low-order oligomers, or protofibrils, prior to accretion on
    deposits, all may contribute to the effects observed by Morimoto et al. If so,
    strategies to eliminate deposits may fail because they do not target smaller,
    intracellular, toxic assemblies.

  2. The authors have elegantly demonstrated the importance of the presence of intracellular misfolded proteins in mediating cellular dysfunction in neurodegenerative disease. Coexpressing the temperature-sensitive (ts) mutants with polyQ in C. elegans at permissive conditions resulted in phenotypes similar to those exhibited by ts mutants under restrictive conditions. This conversion of relatively harmless ts mutants into those which exhibit mutant phenotypes under permissive conditions is a fascinating and enlightening observation. The experiments with various other strains of ts mutants make the case that the expression of aggregation-prone polyQ protein meddles with the structure and function of unrelated proteins. Specifically, the authors suggest that the levels of polyQ influence the folding of ts protein and that perhaps the opposite is also true, as though a positive feedback mechanism exists to augment the imbalance in cellular folding.

    In interpreting the results, the authors propose that marginally stable proteins do not in and of themselves cause disease; rather, they misfold and subsequently modify the aggregation and toxicity of aggregation-prone proteins associated with conformational disease, and this overwhelms the balance of cellular folding and clearing processes. However, if in the presence of aggregation-prone proteins these metastable proteins misfold, this hints at a possible interaction between the two that perpetuates their mutual misfolding and accumulation. To expand on the authors’ interpretation, perhaps the expression of the aggregation-prone protein somehow induces the misfolding of the metastable mutant proteins, contributing to the progressive disruption of cellular processes that maintain the folding environment.

    This imbalance of clearance and folding may be initially caused by the early accumulation of such aggregation-prone proteins as polyQ, which not only stress the overall folding capacity of the cell, but also play a role in stimulating the misfolding of metastable proteins. In fact, in our hands, many aggregation-prone proteins in vitro are capable of interacting with other marginally stable proteins and inducing subsequent conformational changes and aggregation.

    Alternatively, the expression of aggregation-prone proteins might in some way stimulate the synthesis of more of these metastable proteins, thus increasing the chances for even more misfolding and increasing their accumulation. If that is the case, again, the cellular clearance processes may become overwhelmed and tilt the delicate balance of protein homeostasis.

    Whatever the case may be, clearly these data suggest that disruption of cellular processes does not result from a single defect. Rather, in combination, aggregation-prone and metastable proteins can affect protein homeostasis and help to explain the gradual accumulation of damaged proteins observed in misfolding diseases. It would be interesting to try these same experiments with other aggregation-prone proteins.

Comments on Primary Papers for this Article

No Available Comments on Primary Papers for this Article


Paper Citations

  1. . Progressive disruption of cellular protein folding in models of polyglutamine diseases. Science. 2006 Mar 10;311(5766):1471-4. PubMed.

Other Citations

  1. ARF related Live Discussion

Further Reading


  1. . Suppression of polyglutamine-induced protein aggregation in Caenorhabditis elegans by torsin proteins. Hum Mol Genet. 2003 Feb 1;12(3):307-19. PubMed.
  2. . Progressive disruption of cellular protein folding in models of polyglutamine diseases. Science. 2006 Mar 10;311(5766):1471-4. PubMed.


  1. Yeast Teases Apart Huntington’s and Parkinson’s Protein Aggregation
  2. Conformation Rules Part 1: News, Common Threads, Debate from San Diego Conference

Primary Papers

  1. . Progressive disruption of cellular protein folding in models of polyglutamine diseases. Science. 2006 Mar 10;311(5766):1471-4. PubMed.