For some years now scientists have been trying to elucidate the physics behind amyloid fibrils, the multimeric and insoluble strands that cause a variety of amyloidoses, including the plaques of Alzheimer’s disease (see related news item). Today in Nature Structural Biology, researchers in Todd Yeates’ lab at the University of California, Los Angeles, report that the fibrils may be formed by a sucessive head-to-head and tail-to-tail arrangement of individual subunits.

This lab had previously shown that in fibrils of transthyretin, a protein studied as a model for amyloid formation, monomers can bind each other through their respective F β-strands, one of eight present in the molecule (the β-strands are labeled with consecutive letters of the alphabet). But this head-to-head arrangement would limit to just two the number of monomers that can be strung together. Clearly, some other forces must be at work to facilitate fibril growth.

First author Ahmed Serag et al. used site-directed spin labeling, which measures the distance between amino acids, to probe fibrils of transthyretin for likely spots where the monomers may be binding to each other. They found that amino acids on the B β-strands of individual subunits draw near as fibrils form—the gap between two cysteine31s, for example, shortens to 8 Angstroms—indicating that two subunits can bind via their B strands. As the B and F strands are at opposite ends of the molecule it suggests that fibrils form when subunits alternately stack head-to-head and tail-to-tail.

But why does transthyretin only form amyloid fibrils in certain individuals? The authors suggest that for a stable fibril to form, the B β-strands, normally protected from each other by the C and D β-strands, must be exposed. This could happen if the C or D strands are mutated or exposed to an acidic environment, scenarios that have been shown to be fulfilled in vivo.

Serag et al’s model for fibril formation is consistent with work from Jane and David Richardson’s lab at Duke University, Durham, North Carolina, which suggests several mechanisms that have evolved over time to protect proteins with extensive β-strand structure from aggregating to each other. In many cases a single charged amino acid in the middle of an exposed β-strand would suffice.—Tom Fagan


  1. Serag et al. report the results of site-directed spin labeling (SDSL) studies of inter-residue distances in native and amyloid-associated forms of transthyretin (TTR). TTR is associated with a number of amyloidoses, including senile systemic amyloidosis and familial amyloid polyneuropathy. The findings are of interest both for understanding the formation of amyloid fibrils by TTR and for understanding basic features of protein folding and design. An important question in this latter area is why proteins with high native b-sheet content do not assemble into amyloid. A systematic investigation into this question (Richardson and Richardson, 2002) revealed that protein evolution has resulted in the inclusion of structural elements in natively folded proteins that "protect" edge β-strands from interactions with neighboring strands that might lead to adventitious and pathologic β-sheet formation. Typically, this type of protection is provided in cis by amino acid sequence elements in the edge strand that inhibit or preclude inter-strand interactions. A second major design is in transprotection, i.e., the allosteric protection of edge strands through their sequestration "under" associated helices, loops, or other structural elements.

    In TTR, the studies of Serag et al. suggest that amyloid fibril assembly occurs when one edge strand (the C/C'-strand) moves, exposing the penultimate strand (B/B'-strand) "below" and allowing its homotypic interaction with a similarly exposed neighboring strand in the TTR tetramer. This finding suggests that strategies designed to provide exogenous small-molecule proxies for the erstwhile protective C/C'-strands might have clinical merit in the treatment of TTR amyloidoses. In addition, the Serag study attests to the value and power of SDSL for examination of the spatial arrangement of amino acids within peptides and proteins with propensities for amyloid formation.

    One area not addressed in this work is the mechanism of formation of amyloids by peptides that normally exist as portions of larger proteins, e.g., the Aβ peptide and the BRI proteins ABri and ADan. The mechanism preventing these proteins from assembling is quite simple, as they exist natively as parts of the polypeptide chains of their precursor proteins. Understanding the conformational changes necessary for amyloid formation in these cases may be a little harder than for TTR and other natively folded proteins. Nevertheless, the SDSL approach offers a viable method for improving our understanding of the topological features of amyloid formation for these proteins. It should also be noted that because all amyloid proteins share the property of forming extended β-sheets, an "edge-directed" therapeutic approach may also be of value for proteins like Aβ et al.


    . Natural beta-sheet proteins use negative design to avoid edge-to-edge aggregation. Proc Natl Acad Sci U S A. 2002 Mar 5;99(5):2754-9. PubMed.

Make a Comment

To make a comment you must login or register.


News Citations

  1. What Makes Amyloid Sticky?

Further Reading

Primary Papers

  1. . Natural beta-sheet proteins use negative design to avoid edge-to-edge aggregation. Proc Natl Acad Sci U S A. 2002 Mar 5;99(5):2754-9. PubMed.
  2. . Arrangement of subunits and ordering of beta-strands in an amyloid sheet. Nat Struct Biol. 2002 Oct;9(10):734-9. PubMed.