Amyloidoses are a varied and complex set of syndromes caused by aggregation of specific proteins such as the amyloid-β, found in the brains of Alzheimer's patients, and transthyretin, which is found in a variety tissues damaged by systemic amyloidosis. In this week's early edition of PNAS, two papers explore why these amyloidogenic proteins form such stable and deadly aggregates.

Michel Steinmetz and Chris Dobson, from the Paul Scherrer Institute, Switzerland, and the University of Cambridge, U.K., respectively, together with colleagues elsewhere, report the synthesis of a novel peptide that may be extremely useful for testing the whys and wherefores of amyloid formation. Dubbed ccβ, this 17-residue peptide exists as a coiled-coil under ambient conditions, but converts into fibrils when the temperature is raised.

One of the beauties of ccβ is that it readily forms crystals which can be studied by x-ray diffraction. Such crystal structures have been difficult to obtain using natural amyloid peptides. Analysis of the crystals by Richard Kammerer and colleagues shows that the molecule exists as a coil of three parallel α-helices. When Kammerer and colleagues elevated the temperature of these peptides, the helical structures broke down and circular dichroism spectra revealed the emergence of β-sheet formations. By playing with the parent compound (S I R E L E X R I R E L E X R I G, where X-7 and X-14 are alanine and leucine, respectively), the authors were able to find substitutions that dramatically accelerate the formation of β fibrils; methionines at the 7 and 14 positions, for example, enhanced fibril formation by about 60-fold.

Further biophysical measurements showed that the β fibrils form as laminated cross-β sheets, a structure that has been proposed for a variety of amyloid fibrils, including the amyloid-β of Alzheimer's disease. The authors state that the "ccβ system provides an exceptionally favorable platform from which to develop a more detailed understanding of the origin and progression of the increasing number of human disorders associated with amyloid formation."

In the second paper, Joel Buxbaum and colleagues at The Scripps Research Institute, La Jolla, California, used tissue cultures to test the toxicity of various oligomeric forms of the transthyretin protein (TTR). Amyloidogenic peptides generally pass through several intermediary forms on their way from free soluble protein to amyloid aggregates, and evidence has mounted that these various protofibrillary forms are actually the most toxic entities (see ARF related news story).

First author Natalia Reixach found that native TTR, which exists as a homotetramer, has little effect on neuroblastoma cells, whereas the amyloidogenic mutant V30M TTR (in which valine at position 30 is substituted by methionine) induced apoptosis. To determine if the monomers play any role in the toxicity of the methionine mutant, Reixach carried out the same experiment in the presence of either resveratrol or chlorinated biaryl amine, two chemicals which have been shown to stabilize the tetrameric form of the protein. Both chemicals protected the cells against the V30M mutant. To confirm that the monomers are the toxic entities, Reixach tested variants of TTR that cannot form tetramers. All of these monomeric TTRs (M-TTRs) proved toxic, even those without the V30M mutation. The results suggest that it is the propensity to form monomers that confers toxicity on TTR mutants.

The monomeric TTRs also had a propensity to aggregate. Could the aggregates, therefore, be causing the toxicity? To test this, Reixach incubated M-TTR (without the V30M mutation) for five days, recovered the aggregates by gel filtration, and tested them in the neuroblastoma cultures. Aggregates over 100 KDa in size did not cause cytotoxicity. The authors conclude that monomers, or possibly aggregates of two or six monomers (as judged by their size on gel filtration columns) may be the toxic species. In the case of the multimers, they must be very active because Reixach found that the numbers of these being formed were very low.—Tom Fagan


  1. Reixach et al. study the aggregation and cytotoxicity of mutant and wild-type (WT)
    forms of Transthyretin (TTR), one of the 23 known human proteins involved in sporadic
    and familial amyloidoses. It was previously found by Kelly and coworkers that
    aggregation of the homotetrameric TTR involves a monomeric intermediate in vitro,
    mutants that cause monomerization of the protein form amyloid fibrils more readily than
    WT TTR, and small molecules that stabilize the native homotetramer can prevent
    aggregation. In this work, the authors used a human neuroblastoma cell line and four
    previously designed variants of TTR (WT TTR, V30M TTR, WT-M TTR and V30M-M
    TTR, the latter two being multi-point mutants that cause monomerization of the protein)
    to correlate the toxicity (measured using a MTT assay) with the aggregation state of the
    protein (characterized by size exclusion chromatography). They incubated known
    concentrations (100
    kDa) aggregates of the monomeric mutant WT-M TTR did not cause any cell death.
    Compounds that prevented fibrillization of TTR in vitro were found to inhibit
    cytotoxicity. These observations indicate that monomers and/or small soluble oligomers
    of TTR may induce cell death, while fibrils may be relatively benign.
    The TTR species appearing in solution under cell culture conditions were monitored
    by size exclusion. They found that the monomeric mutants heavily aggregate on the time
    scales of the cytotoxicity assay, while, curiously, the toxic mutant V30M TTR retains its
    tetrameric state. These studies provide solid evidence for a general model of aggregation-induced cytotoxicity in which misfolded proteins rapidly oligomerize into small prefibrillar species, which are responsible for cell death, by as yet unknown mechanism.

    View all comments by Sagar Khare
  2. Comment by B. Edward Bondo and Nikolay V. Dokholyan
    Kammerer et al. have developed a simplified protein-like model system to study the molecular details of the assembly of protein amyloid structures. Their system consists of a 17-residue peptide, ccβ, which forms native-like coiled-coil structures under ambient conditions, but can be induced to form amyloid fibrils when these conditions are altered. They find that the lag phase and growth phase of amyloid fibril assembly is temperature and concentration dependent and that the addition of preformed aggregates containing β-sheets eliminates the lag phase, suggesting that these peptides aggregate with a nucleation-dependent self-assembly process. By mutating their peptide, they find that interactions between the hydrophobic side chains in the peptides are very important in the rate of assembly of the fibrils. Structural studies allow the authors to attribute this effect to hydrophobic packing of the side chains. The system the authors design appears to be a promising model system to determine some of the driving forces of amyloid fibril formation.

    View all comments by Nikolay Dokholyan

Make a Comment

To make a comment you must login or register.


Other Citations

  1. ARF related news story

Further Reading


  1. . Stimulation of beta-amyloid precursor protein trafficking by insulin reduces intraneuronal beta-amyloid and requires mitogen-activated protein kinase signaling. J Neurosci. 2001 Apr 15;21(8):2561-70. PubMed.
  2. . Endoplasmic reticulum and trans-Golgi network generate distinct populations of Alzheimer beta-amyloid peptides. Proc Natl Acad Sci U S A. 1999 Jan 19;96(2):742-7. PubMed.

Primary Papers

  1. . Tissue damage in the amyloidoses: Transthyretin monomers and nonnative oligomers are the major cytotoxic species in tissue culture. Proc Natl Acad Sci U S A. 2004 Mar 2;101(9):2817-22. PubMed.
  2. . Exploring amyloid formation by a de novo design. Proc Natl Acad Sci U S A. 2004 Feb 26; PubMed.