. Ion mobility-mass spectrometry reveals a conformational conversion from random assembly to β-sheet in amyloid fibril formation. Nat Chem. 2011 Feb;3(2):172-7. PubMed.


Please login to recommend the paper.


Make a Comment

To make a comment you must login or register.

Comments on this content

  1. Filling a Vacuum in Knowledge...With a Vacuum! Ion-Mobility Mass Spectrometry Provides Insight Into Amyloid Protein Assembly

    In an exciting new paper published in Nature Chemistry, Bleiholder et al. report that the technique of ion mobility-mass spectrometry can reveal structural details of individual amyloid protein oligomers. These oligomers were formed by penta- or hexapeptide segments of the amyloid proteins yeast Sup35, human insulin, and human islet amyloid polypeptide. This paper represents a significant advance in the study of peptide oligomerization and fibrillization because it allows one to pick out particular assemblies in a heterogeneous mixture of assemblies and to study these unique assemblies in isolation. Other techniques used to monitor amyloid protein assembly generally are "averaging methods." They reveal population-average information and thus cannot distinguish specific conformers within populations. Importantly, in the experiments of Bleiholder et al., monomers and higher-order assemblies are unmodified, i.e., the primary structure is native. The lack of “tags,” including fluorescent molecules or unnatural amino acids, ensures that the conformers thus studied are more likely to represent the native structures found in vitro or in vivo.

    The transition that occurs in amyloid assembly from the monomer state to fibrils is one in which an extended or globular monomer converts into a polymer with a large aspect ratio and a cross-core structure that is quite different from the native monomer structure. This process may be considered one of fibril "maturation." A particularly striking example of this is the observation of amyloid protein assemblies that initially appear as beaded chains, but then mature into straight fibrils. The proposed mechanisms that Bleiholder et al. show in Figs. 1 and 5 are the first visualizations of this maturation process, and they are intriguing.

    The new findings provide guidance for next-generation experiments to test the models. This testing is critical because the experimental system of Bleiholder et al. involves an in hydro to in vacuo conversion process. Are the structures observed in the gas phase identical to those observed in the liquid phase? Prior experimental and computational work by the groups of Mike Bowers and Joan-Emma Shea suggest that the answer is “yes.”

    A second area for future work is the extension of this exciting new method to the structural dynamics of full-length amyloid peptides and proteins. All known amyloid proteins are substantially longer than five or six amino acid, and the amino acids adjacent to the amyloid-forming peptides studied by Bleiholder et al. have significant effects on amyloid formation kinetics and morphology. The examination of the native proteins is likely to be more problematic because of the substantially increased conformational complexity and dynamics of the full-length protein population. Nevertheless, I think the field should be very excited about the prospects.

  2. Protein Assembly Pathways Revealed?
    The question of assembly pathways is central to understanding the formation of protein structures associated with amyloid diseases. The latest research by Bleiholder et al. from the lab of Mike Bowers, recently published in Nature Chemistry, addresses the assembly pathways of three fibril-forming amyloid peptide fragments, as well as a control peptide, the hormone enkephalin that only forms an isotropic crystal. The conversion of the four peptide fragments from monomeric to multimeric assemblies was probed by ion mobility-mass spectrometry, resulting in high-precision, collision cross-section measurements for each of the observed assembly sizes. This study is exciting because by utilizing the knowledge of scaling of the measured cross-sections with the assembly size for each of the four peptides, it elucidates, for the first time, a specific assembly pathway from monomers through oligomeric to fibril-like, steric zipper-like, or isotropic aggregates. The results of this study are consistent with atomic force microscope observations and prior studies of steric zippers done in the lab of David Eisenberg (Sawaya et al., 2007).

    The explored peptide sequences were the fibril-forming NNQQNY, VEALYL, SSTNVG, and the isotropic crystal-forming YGGFL. It is of importance that the three fibril-forming, six-residue fragments showed quite distinct assembly pathways, as illustrated in Fig.5 of this paper. Moreover, even though the authors did not discuss this aspect, this ion mobility-mass spectrometry approach seems to be capable of directly determining the size of the nucleus needed for fibril formation to become a "downhill polymerization'' process, for which a nucleation theory exists (Ferrone, 2006). It would be interesting to know whether this method could be applied to examine the existence of secondary nucleation processes that seem to be consistent with the observed fibril growth but have not been characterized experimentally (Knowles et al., 2009).

    Despite its clear relevance to understanding the basic mechanisms of assembly, the work poses questions that might be challenging to address. The first question, which has been partially addressed in the Supporting Information, is, What is the role of protonation in both structural characteristics and pathway assemblies? Does the protonation prevent certain peptide assemblies from forming? Does it change the relative abundances of different assembly sizes? If so, would that affect the assembly pathway(s)? The second question is whether this same approach can be applied to determine the assembly pathways of full-length amyloid proteins, which are directly relevant to amyloid diseases. Here, the scaling of the cross-section with oligomer size may not be as simple as in the case of short peptides. The reason is that full-length peptides within a quasi-spherical oligomer are characterized by several turns and might be more or less compact with increased oligomer size, as was shown for cross-linked Aβ(1-40) oligomers in a recent study by (Ono et al., 2009). In any case, the present method presents a leap forward and has the potential to highly impact our basic understanding of protein assembly. [Editor’s note: Michael Bowers and Brigita Urbanc receive funds from the same NIH Program Project Grant.]


    . Atomic structures of amyloid cross-beta spines reveal varied steric zippers. Nature. 2007 May 24;447(7143):453-7. PubMed.

    . Nucleation: the connections between equilibrium and kinetic behavior. Methods Enzymol. 2006;412:285-99. PubMed.

    . An analytical solution to the kinetics of breakable filament assembly. Science. 2009 Dec 11;326(5959):1533-7. PubMed.

    . Structure-neurotoxicity relationships of amyloid beta-protein oligomers. Proc Natl Acad Sci U S A. 2009 Sep 1;106(35):14745-50. PubMed.

  3. The toxic aggregates of the Aβ peptide that are thought to be responsible for at least the initial AD pathogenesis are either small soluble oligomers or larger, but still soluble, protofibrils that contain smaller aggregates as building blocks. A major problem for the field is that there is a multitude of oligomeric species of different sizes, of which some might be toxic and others perhaps not. This prohibits detailed studies of their structures, because measured chemical and physical properties are averages of contributions from different aggregates. This is a great pity, because knowledge of these structures might provide solid ground for discovering drugs aimed at inhibiting formation of neurotoxic aggregates. For instance, the initial phase of Aβ aggregation, which by many is believed to reflect the transition from harmless aggregates to toxic oligomers, is a structural conversion from a disordered state into a conformation rich in ordered β-sheet secondary structure. A drug inhibiting this interconversion in the brain would be potentially successful in combating AD. Hence, if there was a method to monitor the interconversion in oligomers of different sizes, then one could, for instance, use that method to read out effects of potential drug leads in a screening assay.

    Here, Bleiholder et al. in Michael T. Bowers’s lab present such a method. It is yet to be applied to Aβ itself, but the paper represents a big step forward. The authors use a rather advanced mass spectroscopic technique called IMS-MS (ion mobility spectrometry-mass spectrometry). Briefly, this technique allows for determination of both the masses and the so-called “collision cross sections” of different-sized oligomers in a sample. Cross-section values depend on aggregate shape, and the authors can conclude if a particular aggregate is disordered (isotropic) or looks like one or the other of two fibril-like structures—that is, if a particular oligomer has undergone the disordered-to-ordered conversion. They find that different peptides adopt ordered structures when they have aggregated to a certain size, for instance, five, nine, or 20 monomer units, depending on their sequence. The authors could also conclude if the structures formed were likely to be “single stranded” or “steric zippers." These results are very significant for research on amyloid-forming peptides.

    The measurements are carried out on short model peptides of five or six amino acid residues. This is inconsequential for the structural chemistry, as the nucleating sequences in the aggregation of amyloid-forming peptides indeed are believed to be that short. It would be extremely interesting to see a corresponding analysis of different fragments of Aβ and, in particular, to hear if the technique is applicable to peptides longer than six residues. (A problem might be that the correspondingly heavier oligomers cannot “fly” in the mass spectrometer.) This is because the ultimate targets for drug discovery against Aβ, or α-synuclein, should probably be longer fragments, even though one might not need the full-length peptides.