FASEB J.
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
 QUICK SEARCH:   [advanced]


     


FJ EXPRESS SUMMARY ARTICLE
The
Full-length version of this article is also available, published online May 26, 2005 as doi:10.1096/fj.04-3137fje.
Published as doi: 10.1096/fj.04-3137fje.
This Article
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
19/10/1344
04-3137fjev1    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Arimon, M.
Right arrow Articles by Fernàndez-Busquets, X.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Arimon, M.
Right arrow Articles by Fernàndez-Busquets, X.
(The FASEB Journal. 2005;19:1344-1346.)
© 2005 FASEB

Fine structure study of Aß1-42 fibrillogenesis with atomic force microscopy

Muriel Arimon*, Ismael Díez-Pérez{dagger}, Marcelo J. Kogan{ddagger}, Núria Durany§, Ernest Giralt{ddagger}, Fausto Sanz{dagger} and Xavier Fernàndez-Busquets*,1

* Laboratori de Recerca en Nanobioenginyeria, Parc Científic de Barcelona (PCB), Universitat de Barcelona (UB), Spain;
{dagger} Departament de Química Física, UB;
{ddagger} Institut de Recerca Biomèdica de Barcelona, PCB;
§ Facultat de Ciències de la Salut, Universitat Internacional de Catalunya, Barcelona; and
Departament de Química Orgànica, UB

1Correspondence: E-mail: xfernandez_busquets{at}ub.edu

SPECIFIC AIMS

Pathogenesis in Alzheimer’s disease (AD) is linked to the accumulation of the highly amyloidogenic self-associating amyloid ß1-42 peptide (Aß1-42). We have used atomic force microscopy (AFM) visualization on hydrophilic and hydrophobic surfaces in liquid environment to study in vitro the fibrillogenesis process of synthetic Aß1-42, with the aim of obtaining nanometric resolution of the main structures characteristic of the several steps of fibril formation from monomeric Aß1-42 to mature fibrils.

PRINCIPAL FINDINGS

1. Globular oligomers of Aß1-42 are the dominant species at early stages of aggregation and deposit best on hydrophilic surfaces
1-42 was dissolved in phosphate buffer, pH 7.4, to a final concentration of 10 µM and immediately visualized in liquid medium on mica substrate by tapping mode AFM. At this initial point of incubation (t0), the sample contained homogeneously distributed globular aggregates with a diameter of ~5 nm. When imaged on a hydrophobic substrate such as graphite, Aß1-42 forms a layer 1 to 2 nm in height. Another aliquot of the same sample was analyzed by Western blot after electrophoresis in SDS-containing polyacrylamide gels. The band with highest electrophoretic mobility corresponded to a molecular mass roughly coincident with that of the Aß1-42 monomer (4.5 kDa). Three minor bands of lower electrophoretic mobility were detected, with apparent approximate masses of 7, 16, and 20 kDa that most likely correspond, respectively, to the dimer, trimer, and tetramer species of Aß1-42 that have resisted dissociation by SDS. The most abundant species detected at t0 in immunoblots was a wide band with an apparent mass of ~60 kDa, which probably corresponds to larger heterogeneous SDS-resistant oligomers.

2. Ribbon-shaped Aß1-42 protofibrils deposit best on hydrophobic surfaces and do not grow longer than ~100 nm
The next step in the process of Aß aggregation is the appearance of protofibrils, the first fibrillar species. 1-42 samples incubated for 24 h and deposited on freshly cleaved mica or graphite immediately before AFM imaging in liquid revealed the presence of ribbon-shaped protofibrils ~1.5 nm high and ~5.5 nm wide (see below), with an average length of ~64.4 nm and seldom exceeding 100 nm. Protofibrils are visualized in much greater numbers when deposited on graphite, suggesting they have a hydrophobic surface and therefore settle down more efficiently on hydrophobic graphite rather than on hydrophilic mica. Protofibrils were observed to align in parallel, forming bundles that in turn can assemble end-to-end to form fibrils.

3. Most Aß1-42 fibrils exhibit a nodular structure with ~100 nm periodicity
Abundant fibrils with a width of ~11.4 nm and up to some microns long are observed in Aß1-42 preparations after 2 days of incubation (Fig. 1 A). Although protofibril bundle formation might be the initial step for fibril assembly, it is likely that individual protofibrils can also be incorporated into existing fibrils, as suggested by the observation of contacts between protofibril ends and fibrils (Fig. 1B ). We have identified two main types of Aß1-42 fibrils according to their contour: smooth and nodular (Fig. 1C ), the latter being ~10-fold more abundant than the former. Nodular fibrils (Fig. 1C , 1D ) have regularly spaced constrictions every ~93.5 nm (Fig. 1E ) that are not present on smooth fibrils (Fig. 1F ). Smooth fibrils have a height of ~5 nm, whereas the height of nodular fibrils varies between ~11 nm in the center of the nodules and ~5 nm at internodal points. A higher resolution scanning of the nodular fibril from Fig. 1C reveals a grooved surface (Fig. 1D ) that delineates six elongated structures within each nodule (Fig. 1G ). Based on images of intermediate species, we deduce that the elongated structures are protofibrils, which have a width of ~5.5 nm.



View larger version (163K):
[in this window]
[in a new window]
 
Figure 1. Amplitude AFM images of Aß1-42 fibrils on graphite. A) Low magnification image of fibrils. Arrows point steps in the graphite surface. B) Fibrils and protofibrils coexisting. The arrow shows contacts between a fibril and protofibril ends. C) Nodular and smooth fibrils. D) Nodular fibril from panel C scanned at higher resolution. The black line indicates the section corresponding to the topographic profile in panel G. E, F) Longitudinal topographic profiles of the nodular fibril (E) and smooth fibril (F) from panel C. G) Transversal topographic profile of the nodular fibril from panel D. All samples were incubated for 2 days. Bar: 1 µm for panel A, 100 nm for all others.

4. Aß1-42 fibrillogenesis generates helical structures with a period of ~100 nm
Left-handed helical structures formed by two or more strands are frequently observed in our in vitro preparations visualized by AFM and transmission electron microscopy (TEM). Each helix strand observed by TEM was ~3.7 nm across, and the whole structure ranged from this narrowest width to ~7.6 nm at the widest point of the helix. The period of half a helix turn (i.e., the distance between two consecutive narrow spots of the helix) differed between individual fibrils, varying from 39.0 nm to 149.4 nm, with a mean value of ~92.5 nm. Each helix strand has a width close to that of protofibrils, but it can reach a length of several microns, and we will thus refer to each strand as protofilament.

5. Aß1-42 fibrils have structurally weak points every ~100 nm
Figure 2 A shows an AFM image of partially assembled/disassembled fibrils, where the fibril nodules are hinted by the presence of apparently softer material that can be significantly pushed aside when applying greater forces with the AFM tip by increasing the working amplitude (Fig. 2B ). This manipulation of fibrils permitted to reveal the existence of an underlying protofilament helix with a period that coincides precisely with the length of nodules in the fibril. Eventually, AFM and TEM images show fibrils that have been fractured in sections (Fig. 2C-E ) with a mean fragment length of ~107 nm, close to the measured periodicities of fibril nodules (~93 nm), and of the helical repeat of intertwined protofilaments (~93 nm) (Fig. 2F ). In these segmented fibrils the two protofilaments run parallel and do not present any twist (Fig. 2C ), and the protofilament sections strongly resemble protofibrils ~100 nm long. Such fragmentation events suggest the existence of a structural weakness related to the joining points between the constituent ~100 nm protofibril subunits.



View larger version (120K):
[in this window]
[in a new window]
 
Figure 2. AFM manipulation of Aß1-42 fibrils and visualization of segmented fibrils. A, B) Amplitude AFM images taken on graphite of the same fibrils scanned with (A) low and (B) high amplitude to increase force between tip and sample in the latter case. C) Amplitude AFM image of a segmented fibril. D, E) TEM images of segmented fibrils. All samples were incubated for 2 days. F) Histogram representations of the lengths of a statistically significant number (n) of isolated protofibrils (from AFM data), fibril nodules (AFM), helical periodicities (TEM), and fibril fragments (AFM and TEM). Bar: 100 nm.

CONCLUSIONS

It is likely that the most abundant features revealed at the onset of Aß1-42 fibrillogenesis by AFM on mica (globular aggregates) and by Western blots (~60 kDa band) correspond to the same molecular entity. This agrees with existing data indicating that, in a hydrophilic environment, Aß1-42 has a strong tendency to self-associate forming highly stable oligomeric structures termed ADDLs (amyloid-derived diffusible ligands), which may be the key effectors of cytotoxicity in AD. Small ADDLs include associations of four to six Aß1-42 units that have been described to form in the membrane of neurons the calcium-permeable channels that are suspected to be one of the causes of cell death associated with Aß. As an alternative mechanism recently described, ADDLs between 10 and 100 kDa have been shown to bind to certain synapses. The ensuing functional disruption may provide a molecular basis for synapse failure in memory loss.

The protofibril dimensions deduced from our AFM images (ribbon-like structures ~1.5 nm high and ~5.5 nm wide) are similar to those reported in previous studies of Aß1-42 fibrillogenesis. In these models, protofibrils are proposed to be single ß-sheets formed by a stacking of the peptide in a hairpin conformation perpendicular to the protofibril axis (Fig. 3 A), with a spacing of ~5 Å between consecutive hydrogen bond-linked ß-strands. The mean length of isolated protofibrils is ~64 nm and they seldom grow longer than 100 nm. This observation could be explained by a selective channeling of longer protofibrils approaching 100 nm toward their incorporation into bundles (Fig. 3Biii ), as suggested by the left-skewed histogram in Fig. 2F corresponding to the distribution of protofibril lengths. Protofibril association is likely driven by the thermodynamically unfavorable exposure of the constituent hydrophobic ß-sheets to the aqueous environment. We have identified three other entities in the 1-42 fibrillogenetic pathway that share a ~100 nm structural unit, namely, fibril nodules, helical protofilaments, and fibril fragments. Taken together, our data are consistent with the existence of a ~100 nm long protofibril species with a role as short-lived assembly intermediate whose elongation rate is slower than its rate of association to form fibrils. Protofibril bundles stack end-to-end (Fig. 3Biv ) and appear to be capable of undergoing internal rearrangements leading to the generation of nodular fibrils (Fig. 3Bv ) and of a helical structure (Fig. 3Bvi ) that might play an important role in fibrillogenesis. We have observed a variety of helical periods that probably correspond to different points of coiling in the dynamic process leading from protofibrils to protofilaments and fibrils. Conformational changes experienced by protofibrils turning into protofilaments could be responsible for the different dimensions of isolated protofibrils (short ribbon-like structures ~5.5 nm wide and ~1.5 nm high) and of protofilaments (long cylinders ~4 nm in diameter). Although Aß1-42 and a number of other amyloidogenic proteins and peptides are not sequence-related, they form twisted protofilaments and nodular fibrils remarkably similar in periodicity, dimensions, and number of constituent subunits. This is in agreement with the current view that, despite the different nature of precursor proteins and peptides, amyloid fibrils represent a structural superfamily and share a common protofilament structure.



View larger version (42K):
[in this window]
[in a new window]
 
Figure 3. Proposed model for Aß1-42 fibrillogenesis in vitro. A) Cartoon illustrating two possible models for the formation of an Aß1-42 protofibril on graphite. The height of the protofibril (~1.5 nm) is provided by the thickness of the peptide adopting a hairpin (for each model, 4 peptide molecules are represented), and the width of the protofibril (~5.5 nm) corresponds to the length of the hairpin. On the left scheme, both strands of the hairpin are in contact with the graphite, whereas on the right scheme only one strand of the hairpin is facing the surface. B) Schematic model proposal of the different stages in the amyloid fibril formation process: (i) monomeric Aß1-42 associates to form protofibrils and globular oligomers; the data presented here do not provide evidence for the incorporation of oligomers into growing protofibrils (ii); protofibrils that reach ~100 nm can form bundles (iii) that through end-to-end stacking (iv) generate the nodular type fibrils (v), which may be related to a helical structure formed by intertwined protofilaments (vi); finally, nodular fibrils might evolve further to yield smooth fibrils (vii).

FOOTNOTES

To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.04-3137fje;




This article has been cited by other articles:


Home page
J. Neurosci.Home page
E. Sturchler, A. Galichet, M. Weibel, E. Leclerc, and C. W. Heizmann
Site-Specific Blockade of RAGE-Vd Prevents Amyloid-{beta} Oligomer Neurotoxicity
J. Neurosci., May 14, 2008; 28(20): 5149 - 5158.
[Abstract] [Full Text] [PDF]


Home page
Proc. Natl. Acad. Sci. USAHome page
M. S. Z. Kellermayer, A. Karsai, M. Benke, K. Soos, and B. Penke
Stepwise dynamics of epitaxially growing single amyloid fibrils
PNAS, January 8, 2008; 105(1): 141 - 144.
[Abstract] [Full Text] [PDF]


Home page
Biophys. JHome page
M. J. O. Widenbrant, J. Rajadas, C. Sutardja, and G. G. Fuller
Lipid-Induced {beta}-Amyloid Peptide Assemblage Fragmentation
Biophys. J., December 1, 2006; 91(11): 4071 - 4080.
[Abstract] [Full Text] [PDF]


This Article
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
19/10/1344
04-3137fjev1    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Arimon, M.
Right arrow Articles by Fernàndez-Busquets, X.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Arimon, M.
Right arrow Articles by Fernàndez-Busquets, X.


HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS