Báo cáo khoa học: Structures for amyloid fibrils doc

MINIREVIEW Structures for amyloid fibrils O. Sumner Makin and Louise C. Serpell Department of Biochemistry, John Maynard Smith Building, School of Life Sciences, University of Sussex, Falmer, East Sussex, UK Amyloidoses comprise over 20 diseases, including Alz- heimer’s disease, Creutzfeldt–Jakob Disease and type II diabetes [1–5]. Although amyloid is known to be toxic [2,6], there is considerable discussion as to its role in disease. It is suggested that the oligomeric precur- sors to amyloid may be substantially more toxic than amyloid itself [7,8]. Even if this is the case, fibrils are likely to play an important role, either as reservoirs or sinks of toxic oligomers. In some amyloidoses, no such doubt exists because the mass of amyloid may exceed that of the undiseased organ [9,10]. Once the amyloid structure has been determined, the rational design of new drugs may be possible (e.g. peptide mimetics) [11,12]. Amyloid in disease is generally defined to be extracellular, although intracellular structures sharing the same core structure are also observed (e.g. a-synuc- lein in Lewy bodies in Parkinson’s disease) [13]. As amyloid-like fibrils can also be formed in vitro from protein unconnected to the amyloidoses, the distinction between amyloid and amyloid-like fibrils is blurred [14]. It has been suggested that nearly all proteins have the ability to form amyloid under certain conditions, which has implications for the understanding of pro- tein folding [8]. Amyloid precursor proteins do not share a common size, sequence or secondary structure, yet the mature fibrils appear to share similar highly organized multimolecular morphology and mechanisms of toxicity [15]. Amyloid is defined in terms of empirical observa- tions from X-ray fibre diffraction, electron microscopy (EM) and specific chemical staining (Fig. 1). The cross-b diffraction pattern has two characteristic sig- nals, a sharp reflection at 4.7 A ˚ along the same direc- tion as the fibre and a more diffuse reflection at between 10 and 11 A ˚ perpendicular to the fibre direc- tion (Fig. 1B) [16]. EM shows the fibrils to be straight, unbranching, 70–120 A ˚ in diameter and of indeterminate Keywords Alzheimer’s disease; b-helix; cross-b structure; electron microscopy; mature amyloid fibril; model; solid state nuclear magnetic resonance; X-ray fibre diffraction Correspondence L. C. Serpell, Department of Biochemistry, John Maynard Smith Building, School of Life Sciences, University of Sussex, Falmer, East Sussex BN1 9QG, UK Fax: +44 1273678433 Tel: +44 1273877363 E-mail: L.C.Serpell@sussex.ac.uk (Received 24 June 2005, accepted 7 October 2005) doi:10.1111/j.1742-4658.2005.05025.x Alzheimer’s disease and Creutzfeldt–Jakob disease are the best-known examples of a group of diseases known as the amyloidoses. They are char- acterized by the extracellular deposition of toxic, insoluble amyloid fibrils. Knowledge of the structure of these fibrils is essential for understanding the process of pathology of the amyloidoses and for the rational design of drugs to inhibit or reverse amyloid formation. Structural models have been built using information from a wide variety of techniques, including X-ray diffraction, electron microscopy, solid state NMR and EPR. Recent advan- ces have been made in understanding the architecture of the amyloid fibril. Here, we describe and compare postulated structural models for the mature amyloid fibril and discuss how the ordered structure of amyloid contributes to its stability. Abbreviations Ab, amyloid b-peptide; EM, electron microscopy; EPR, electron-paramagnetic resonance; FTIR, Fourier transform infra red; IAPP, islet amyloid polypeptide; PrP, prion protein; TTR, transthyretin; STEM, scanning transmission electron microscopy; ssNMR, solid state NMR. 5950 FEBS Journal 272 (2005) 5950–5961 ª 2005 The Authors Journal compilation ª 2005 FEBS length (Fig. 1A) [17]. An apple-green colour is observed in the light microscope through cross-polaris- ers after staining with Congo red dye [18,19]. A shift in fluorescence after staining with Thioflavine T is also observed [20]. Both CD and Fourier transform infra red (FTIR) spectroscopy support a high b-sheet con- tent for amyloid fibrils. Structural studies have led to a better understanding of the mechanisms by which normally soluble proteins undergo a conformational change, associated with aggregation, to form amyloid. Amyloid may have bio- nanotechnology applications [21,22]. These include roles in catalysis, in electronics, as a plastic, for sup- porting cells or as a therapy for treating animals and humans [23]. It is only by understanding the detailed structure of fibrils that their properties can be improved and further applications developed. Much of the high-resolution, detailed structural data has been obtained from nonphysiological amyloid-like fibrils assembled from short peptides (either homolog- ous to regions of disease-related peptides or designed). These assemblies have been found to be extremely valuable for yielding high-quality data enabling detailed structures to be solved (see ‘Recent Advances’ below). These assemblies can serve as model systems that give greater insight into the internal arrangements within amyloid fibrils and are likely to be highly rele- vant to the amyloid core structure. Debate on the structure of the toxic oligomer (or protofibril) and structural intermediates on the fibril formation pathway are outside the scope of this review. Therefore, we limit our discussion to the struc- ture of the mature amyloid fibril. Macromolecular structure of amyloid: protofilaments EM and atomic-force microscopy have revealed much about the macromolecular structure of amyloid. Proto- filaments are fibrillar subunits comprising amyloid fibrils and are clearly visible in micrographs, even before image processing [17,24–26]. Type II diabetes- related amyloid fibrils composed of islet amyloid polypeptide (IAPP) are able to form different morpho- logies, depending on the in vitro conditions. Transmis- sion EM and atomic force microscopy were used to study fibril formation and showed that the predomin- ant fibrillar structure was composed of two 5 nm diameter protofilaments wound in a left-handed direction [24,27]. Cross-sections of ex vivo amyloid fibrils taken from many sources and analysed by single-particle methods resulted in averaged images showing several protofilaments [25]. Improved images of the protofilaments have been revealed by single par- ticle averaging of clearly helical fibrils by the Saibil group [28–30] (Fig. 2). The SH3 domain of phosphoti- dylinositol-3¢-kinase forms twisted fibrils in which four protofilaments twist slowly around one another [28]. Differing numbers and arrangements of protofilaments may be present under the same experimental condi- 100 nm 4.7 Å ~10 Å Fibre axis A B Fig. 1. The characteristics of amyloid fibrils include their appearance in the electron microscope and the cross-b diffraction pattern. (A) Elec- tron microscopy (EM) of negatively stained amyloid fibrils formed by islet amyloid polypeptide (IAPP), showing long, unbranching fibrils of  100 A ˚ in diameter. (B) X-ray fibre diffraction pattern from aligned IAPP amyloid fibrils, showing the positions of the 4.7 A ˚ meridional and  10 A ˚ equatorial reflections in a cross-b pattern. O. S. Makin and L. C. Serpell Structures for amyloid fibrils FEBS Journal 272 (2005) 5950–5961 ª 2005 The Authors Journal compilation ª 2005 FEBS 5951 tions [26,28]. For example, the 3D reconstruction of insulin fibrils revealed fibrils formed with two, four (Fig. 2) and six protofilaments [29], although the size and shape of the individual protofilaments was the same. Cryo-EM images of ex vivo amyloid fibrils of Asp67His variant lysozyme showed wavy fibrils, and image analysis indicated the presence of six protofila- ments [30]. Internal structure of amyloid protofilaments Cross-b structure models for amyloid The cross-b pattern was first observed by X-ray dif- fraction from the egg-stalk of the lacewing Chrysopa [31]. The protein chains run orthogonal to the fibril direction and are hydrogen-bonded, 4.7 A ˚ apart, to form a b-sheet. A pseudo-repeat of 6.9 A ˚ is evident along the pleated b-chain (i.e. with an axial advance per peptide unit of 3.4 A ˚ arranged with a twofold heli- cal repeat). The spacing between the b-sheets depends on the size of the side-chain groups [32,33]. An early model of the protofilament structure arose from ana- lysis of X-ray fibre diffraction patterns from ex vivo transthyretin (TTR) Met30 variant amyloid fibrils [34] in which the b-strands were hydrogen bonded to form a continuous b-sheet structure. In this model, four b-sheets twisted around a central axis. X-ray fibre dif- fraction from a gallery of ex vivo and synthetic amy- loid fibrils suggested that they may share this generic cross-b structure [3,35]. X-ray diffraction studies have commonly examined amyloid formed by peptides corresponding to frag- ments of amyloid b-peptide (Ab) [36–43]. These serve as valuable model systems that give information about the core amyloid structure. X-ray diffractograms from magnetically aligned Ab(11–25) amyloid fibrils were recorded for three mutually orthogonal beam direc- tions [43]. These showed three different patterns, indi- cating that the structure was highly ordered with a preferred orientation. A structure was built in which the 15mer formed an extended b-strand associating to form cross-b ribbons. These sheets were 10.6 A ˚ apart, associated via side-chain contacts. The high order of the fibrils assembled from the short, central peptide of Ab enabled the collection of high-quality data and led to a detailed structural model [43]. Solid state NMR (ssNMR) studies of Ab(11–25) fibrils at two pH values resulted in models illustrating that the b-strands are able to slip within the protofilament structure, depend- ing on the fibril formation conditions [44]. X-ray dif- fraction data from full-length Ab is less detailed than that of Ab(11–25) fibrils. A model for Ab(1–40) amy- loid fibrils comprised five or six cross-b cylinders, 28 A ˚ wide, and spaced 55 A ˚ apart [42]. Cross-b models also appeared to fit data collected from synthetic amyloid fibrils formed by other short, disease-related peptides. These peptides included the first predicted a-helical region, residues (109–122) of cellular prion protein (PrPc) (H1) [41]; short PrP fragments [45]; 11-residue N-termini of the apoSAA family [46] and IAPP [26]. Recent advances in ssNMR have led to the creation of amyloid structural models [47]. This allows the measurement of distances between 13 C labels up to 6A ˚ apart, with standard error values of 0.1–0.2 A ˚ . Additional information is revealed about torsion angles, orientations relative to the applied magnetic field and the amount of order in the structure [48]. Recent ssNMR and EPR experiments on synthetic amyloid fibrils formed by full-length Ab have found the b-strands to be parallel and in-register [49–53]. A parallel, in-register structure was also found for Ab(10–35) fibrils [50,54–58], whilst studies on other peptides found antiparallel arrangements of b-strands [44,59–63]. A recent study, incorporating acylation with octanoic acid into fibrils of Ab(6–22), showed that the amphilicity of a peptide may be associated with the preference to form parallel or antiparallel b-sheet structures [64]. Information derived from ssNMR can be comple- mented by information from scanning transmission EM (STEM), which allows the determination of mass per unit length of a fibril by comparison with a stand- ard, such as the tobacco mosaic virus [65]. ssNMR and STEM measurements for Ab(1–40) amyloid fibrils were consistent with a structure in which the Ab(1–40) peptide is folded once and then these units stacked to form two b-sheets [58] (Fig. 3C). The b-sheets are in contact via side-chains. This model is consistent with Fig. 2. Helical image reconstruction using single particle analysis of amyloid fibrils composed of insulin shows four individual protofilaments that twist around one another. Adapted from [29]. The image was generated using PYMOL (http://www.pymol.org). It shows a high-density contour of the four protofilaments in solid white and a lower threshold contour of the fibril as a transparent blue surface. Structures for amyloid fibrils O. S. Makin and L. C. Serpell 5952 FEBS Journal 272 (2005) 5950–5961 ª 2005 The Authors Journal compilation ª 2005 FEBS A B C DE Fig. 3. Structures showing the arrangement of polypeptide chains within amyloid-like fibrils, solved from X-ray or solid state NMR (ssNMR) data. Amyloid fibril structures viewed down the fibre axis. (A) A 15Q peptide [75] constructed from X-ray fibre diffraction data. The structure shows interdigitation of the glutamine side-chains allowing very close packing of the b-sheets. (B) KFFEAAAKKFFE [101] constructed from X-ray and electron diffraction data. One pair of antiparallel b-strands is shown (two layers, one above the other, into the page). The structure shows interactions between the phenylalanine residues between the b-sheets and also between b -strands within a b-sheet. (C) Amyloid b-peptide (Ab) 1-40 molecule [Ab(1–40)] constructed from ssNMR data [58]. The structure shows the Ab(1–40) molecule folding into two b-strands and joined by a turn. Many molecules stack to form a pair of parallel b-sheets. (D) GNNQQNY [103] solved by X-ray crystallography. The crystal structure shows two pairs of sheets interacting via interdigitating side-chains with water excluded. The adjacent sheets interact via water molecules and a single interaction between the tyrosine side-chain. (E) A schematic, side-on view of the b-sheets, showing b-strands that are hydrogen bonded over the length of the fibre. This scheme shows parallel b-sheets, although the sheets can also be com- posed of antiparallel b-sheets (as in the models shown in panels A and B). The figure was prepared using Pymol (http://www.pymol.org). O. S. Makin and L. C. Serpell Structures for amyloid fibrils FEBS Journal 272 (2005) 5950–5961 ª 2005 The Authors Journal compilation ª 2005 FEBS 5953 protofilament width measurements from electron micrographs of Ab(1–40) amyloid fibrils. The b-structure within a single fibril has been visu- alized using cryo-EM of Ab(11–25) fibrils, showing a strong meridional signal at 4.7 A ˚ in the Fourier transform and striations 4.7 A ˚ apart in the image [66]. This clearly indicates that the b-strands run per- pendicular to the fibre axis. The observation of stria- tions within the image suggests an in-register arrangement of more than one b-sheet and highlights the stability of the amyloid fibril. A strong reflection at 4.7 A ˚ was also observed in Fourier transforms of cryo-EM images of full-length IAPP fibrils in ice [26], supporting the view that this is common to amyloid fibrils. Modelling studies have suggested that Ure2p [67] and IAPP [68] amyloid fibrils can be modelled as ‘par- allel superpleated beta’ structures, in which the protein or peptide folds into ‘serpentines’ linking b-strands between adjacent b-sheets. These units then stack to form a several b-sheets that gradually twist. Alternative models to the cross-b structure b-helix and nanotube A b-helical or nanotube structure has been suggested as a possible generic structure for the amyloid fibril. In these models, one or more extended b-sheets wrap around a hollow core in a helical manner. This leads to an rise of the helix-per-turn. The first b-helical model was a cylindrical antiparallel b-helix with a radius of 10 A ˚ , suggested as a model for amyloid formed from the peptide with sequence KLKLKLE- LELELG [69]. The model was based on X-ray diffrac- tion, EM and, particularly, the CD spectrum analyzed by comparing it with that of pectate lyase E [70]. A later study on this peptide using FTIR concluded that the structure might be an extended b-strand [71]. Anti- parallel b-helix models have also been inferred for amyloid formed from TTR, Ab and immunoglobulin light chain [72]. A water-filled nanotube b-helical model was pro- posed from analysis of data collected from fibrils assembled by a polyglutamine peptide [73]. Twenty res- idues per turn formed a hollow tube with internal and external radii of 6 and 16 A ˚ , respectively [74] This model was based on the absence of a 10–11 A ˚ spacing on the equator of X-ray diffraction patterns. The apparent similarities between such a structure and that of carbon nanotubes suggest an underlying elegance, simplicity and perhaps an understanding of the generic nature of amyloid. However, re-evaluation of the dif- fraction data showed that it was actually consistent with a cross-b arrangement of antiparallel b-hairpins [75]. The diffraction data showed an 8.3 A ˚ diffraction signal, which was shown to be consistent with closely packed b-sheets involving hydrogen bonding of the glutamine side-chains (Fig. 3A). This model shows interdigitation of side-chains as well as hydrogen bond- ing between the glutamines along the fibre axis, result- ing in a highly stable, rigid structure. Some b-helical models are based on the b-helical fold of globular proteins with a triangular rather than a circular projection along the fibre axis [76,77]. These structures show a rise-per-turn consistent with a helix. The b-helical protein family includes the pectate lyases, P22 tailspike protein and UDP-N-acetylglucosamine acyltransferase [78–81]. Electron crystallography of 2D crystals of scrapie prion fragment used image analysis to yield low-resolution projection maps [82]. The resulting average was compared with calculated projec- tion maps of trimers of left-handed parallel b-helices [83]. Studies based on hydrogen–deuterium exchange and a proline scan on full-length Ab also proposed ‘b-heli- cal models’ [77,84–87]. Exchange protection resulting from hydrogen bonding was present along substantial lengths of the peptide; this differs from the pattern for most globular proteins, which generally have regions several residues long that are exchange-protected and interspersed with shorter unprotected regions. This implies either a close-to-ideal b-helix or a very wide b-sheet. Further evidence for b-helical models comes from the suggestion that the predicted intersheet distance signal (usually at 10–11 A ˚ ) may be a dehydration artefact illustrated in X-ray diffraction experiments involving Sup35 fibrils [88]. However, this may also be explained by the reduced quality of diffraction data from fibrils in solution. These fibrils will be dis- persed in solution (and not packed as in a dried sample), leading to a low coherence length. This, coupled with water scatter, will probably lead to the intersheet reflection being very weak and therefore unobserved. NMR data are inconclusive on this length scale (more than 6 A ˚ ) and unable to clearly differentiate between in-register parallel b-helix (although this does not show a rise-per-turn) and an in-register parallel cross-b structure. However, the collapse of such a ‘b-helix’ structure would result in a conformation very similar to the cross-b structure [58]. 3D reconstructions of SH3 fibrils revealed protofilaments with flat cross- sections, inconsistent with a three-sided b-helix [28]. Structures for amyloid fibrils O. S. Makin and L. C. Serpell 5954 FEBS Journal 272 (2005) 5950–5961 ª 2005 The Authors Journal compilation ª 2005 FEBS Amyloid models retaining native structure Models composed largely of native structure (i.e. retaining structure that is present in the monomeric, native form) have been proposed for filaments of the yeast prion, Ure2p. Biochemical evidence suggested that Ure2p retained much of its native structure with the filaments and that they may be composed of asso- ciated monomers [89]. An opposing model concludes that the fibrils have a cross-b core [90,91]. In the case of Ure2p filaments, the experimental conditions are extremely important because a cross-b structure is evi- dent after heating [92]. It remains unclear whether Ure2p filaments fit the criteria for amyloid. X-ray fibre diffraction data of TTR fibrils led to the construction of a model composed of axially arrayed monomers [93]. The crystal structure of a highly amy- loidogenic TTR triple mutant has been solved and shows a three-residue slip in one of the b-strands [94]. From this data, a model was constructed in which the b-slip allows the construction of an infinite b-sheet in which existing b-strands, present in the native struc- ture, align. The resulting structure is a double helical arrangement of monomers [94]. There is an overlap repeat of 114.5 A ˚ , close to the meridional repeat dis- tance, 115.5 A ˚ , calculated from fibre diffraction [95]. The diameter of model is 120 A ˚ , which is close to 130 A ˚ , measured using EM [96]. However, TTR mono- mers have a pair of b-sheets at an angle to one another. In the largely native TTR fibril model, the b-strands stack in a more complicated manner than purely along the fibre axis. This is not consistent with the cross-b model or with X-ray diffraction data. A 29 A ˚ meridional reflection might be expected from stacked monomers, but is not observed [93,95]. Prox- imity information from site-directed spin labelling of TTR fibrils permitted the building of a head-to-head and tail-to-tail model [97], where the edge-strands of the TTR monomer are displaced. However, although the interfaces in one sheet were revealed, the arrange- ment of the second sheet could not be visualized. Therefore, it may be that the TTR conformation is altered in the fibrils. Recent advances in amyloid structure As described in the introduction, amyloid has long been known to be composed of a ladder of b-strands in a hydrogen-bonded b-sheet. It is clear that many proteins (involved both in disease and in vitro) are able to access this rather simple and repetitive structure, indicating that perhaps primary sequence plays a minor role. However, it has become increasingly clear that primary sequence is important from fibril forma- tion experiments using very short peptides [98] and lar- ger proteins [99]. Recent advances in the elucidation of structure of amyloid have enabled a better understand- ing of why this might be. Examination of sequences in disease-related, amy- loidogenic proteins revealed a preponderance of aro- matic groups [100], and suggested the importance of phenyalanine side-chains in p–p stacking. This was highlighted in a subsequent structural study in which b-sheets were zipped together via p-stacking and salt bridges [101]. A 12mer peptide containing two KFFE motifs separated by an AAAK motif (AAAK) formed amyloid nanocrystals that yielded high-resolution X-ray and electron diffraction data. These data were indexed to a unit cell and space group, revealing the symmetry arrangements of the monomeric molecules. The peptide associates to form antiparallel b-sheets, and the sheets are associated via a staggered arrange- ment allowing contacts between the side-chains (Fig. 3B). Modelling revealed that very few arrange- ments of the phenylalanine residues were possible within the tightly constrained cell, and the structure showed p–p stacking of the phenylalanine residues both between the b-sheets and also between hydrogen- bonded strands. The peptide is almost palindromic and has almost identical faces, meaning that the structure was propagated in both the hydrogen bonding and sheet directions [101]. This structure presented a key step in understanding the nature of the intersheet association in amyloid fibrils and revealed how side- chains might enable the sheets to zip together, exclu- ding water (Fig. 3B). Similarly, fibrous crystals were grown from an amy- loid fibril forming peptide from Sup35, GNNQQNY [102,103]. Electron diffraction and X-ray fibre diffrac- tion from nanocrystals yielded high-resolution data [102]. Subsquent growth of microcrystals allowed col- lection of X-ray crystallography data at the synchro- tron, enabling the structure of this amyloid-like assembly to be solved [103]. In contrast to ‘AAAK’, this peptide has different faces and stacks into pairs of parallel b-sheets. The crystal structure showed very close packing between two sheets involving interdigi- tated side-chains that exclude water, termed a ‘steric zipper’ (Fig. 3D). The other face of the sheet was packed against another sheet with water molecules, and contacts occurred only via a pairing of the tyro- sine side-chain in an edge-to-face arrangement (similar to the arrangement of Phe in the ‘AAAK’ peptide). ssNMR of fibrils formed by another peptide corres- ponding to a fragment of a yeast prion, HET-s, has supported a two-sheet structure [104]. Modelling O. S. Makin and L. C. Serpell Structures for amyloid fibrils FEBS Journal 272 (2005) 5950–5961 ª 2005 The Authors Journal compilation ª 2005 FEBS 5955 suggested that the peptide formed four b-strands, forming two b-strand ⁄ turn ⁄ b-strand structural motifs. Importantly, infectivity of the prion was seen to corre- late with the ability to form the b-sheet structure. This ability was examined by substituting residues within the predicted b-strand with Pro, which would be expected to disrupt the b-strand structure. Some of these Pro mutants were unable to form large aggre- gates and this correlated with their inability to infect. Again, this highlights the importance of the constituent side-chains for fibril formation. Structural studies have begun to explain the existence of strains that are particularly relevant to the prion diseases. These are self-perpetuating conformations that may underlie different phenotypes (e.g. the age of onset, disease progression, ability for cross-species infectivity) in the absence of primary sequence changes. ssNMR and STEM measurements of amyloid fibrils formed by Ab(1–40) showed that different morpholo- gies had different molecular structures [44,105]. Differ- ent morphologies can be influenced by fibril growth conditions, and these conditions (and different struc- tures) can yield samples with significantly different toxi- cities. All the fibrils, grown in quiescent or agitating conditions, contained the cross-b motif; however, they differed in the number of molecular layers and in the participation of particular residues in the b-strands [105]. These different ‘strains’ could be shown to per- petuate themselves in seeding experiments. A study involving the yeast prion, Sup35, from two different yeast species, showed that different fibril growth condi- tions could produce different ‘strains’ of fibrils with different abilities to cross the species barrier [106]. Structural examination of these different strains using FTIR and EPR suggested that they differed in the resi- dues which participated in the b-strand structure. Atomic force microscopy showed that PrP from differ- ent species form particular morphologies [107] that can cross-species seed and maintain their morphology. Biophysical studies using fluorescently labelled vari- ants of the NM region of Sup35 have revealed specific interactions between residues within the amyloid core [108]. Variations in length of the amyloid and nature of intermolecular interactions within it have been sug- gested to underlie different ‘strains’ of prion. A model structure was suggested, based on the b-helical struc- ture for a protofilament. However, the accumulated data for amyloid, described in this review, seems to support a two-sheet b-helix that would show all the structural features of the cross-b fold. Table 1. Summary of the major proposed structures for amyloid fibrils. Fibril forming protein or peptide Type of data collected Structure proposed Reference(s) IAPP EM ⁄ AFM, X-ray fibre diffraction, electron diffraction Twisted protofilaments, cross-b [24,27] SH3 domain EM, Cryo-EM and Four twisted protofilaments [28] Insulin single particle Two, four or six protofilaments [29] Lysozyme image analysis Six protofilaments, wavy fibrils [30] Insect silk X-ray fibre diffraction First cross-b structure [31] TTR(Met30) X-ray fibre diffraction Cross-b model (four sheets, twisted) [34] Double helical arrangement of axially arrayed monomers [93,94] Ab(11–25) X-ray diffraction ⁄ ssNMR Antiparallel cross-b ribbons [43,44] Ab(1–40) X-ray fibre diffraction Five or six cross-b cylinders [42] ssNMR and EPR Parallel, in-register b-structure [49–54,58] Ab(10–35) ssNMR [50] Ure2p and IAPP EM, X-ray diffraction, electron diffraction, modelling ‘b-serpentine’ [67,68] Polyglutamine peptide (D 2 Q 15 K 2 ) X-ray fibre diffraction b-helical nanotube [73] Cross-b arrangement of antiparallel b-hairpins [75] KFFEAAAKKFFE X-ray and electron diffraction Staggered, antiparallel cross-b sheets associated via p–p stacking of Phe groups [101] GNNQQNY X-ray crystallography Pairs of parallel, cross-b-sheets ‘steric zipper’ [103] Ab, amyloid b-peptide; AFM, atomic-force microscopy; EM, electron microscopy; IAPP, islet amyloid polypeptide; ssNMR, solid state NMR; TTR, transthyretin. Structures for amyloid fibrils O. S. Makin and L. C. Serpell 5956 FEBS Journal 272 (2005) 5950–5961 ª 2005 The Authors Journal compilation ª 2005 FEBS In the section describing models of b-helices, it is clear that a b-helix should have a rise-per-turn consis- tent with a helix (as suggested for the Perutz model for polyQ) [73]. The accumulated structural data presented here show rather a stacking of b-turn units, where the repeat distance of the ‘helix’ is no larger than the hydro- gen bonding distance (i.e. the helical pitch is zero). Conclusions Recent advances have highlighted the importance of residue composition and sequence on amyloid fibril formation. It is clear that certain simple primary sequence motifs have the ability to fit together in a complementary way, yielding highly ordered aggregates and thus crystalline arrangements [101,103]. In the prion diseases, it is postulated that this is responsible for the species barrier [103,105–108]. In disease, some peptides or proteins may have an enhanced ability to fit together in a complementary way, packing the side- chains to form highly stable structures. This is a prop- erty of a particular primary sequence. However, several studies [105–108] have now highlighted that dif- ferent fibril growth conditions can affect the internal architecture of the fibril, favouring different side-chain packing arrangements and allowing different parts of the sequence to participate in the b-structure. This can yield fibrils with outwardly different morphologies and with differing growth rates, stabilities and the ability to seed other, related, peptides (strains). High-resolu- tion structural studies have given some insight into the incredibly ordered arrangements of side-chains that could underlie this phenomenon [101,103]. Many models for amyloid structure have been pos- tulated, including cross-b, b-helix and predominantly native structures (summarized in Table 1). The cross-b structure has considerable support, including highly detailed structures for amyloid fibrils formed from cer- tain peptides [43,75,101,103] (Fig. 3). The evidence for these structures appears to clearly exclude other mod- els. It may be that some fibrils, which have been referred to as amyloid-like, are in fact not amyloid but simply have a sufficiently high b-sheet content to fall within an over-broad definition. It is only by combi- ning data from many sources, such as X-ray diffrac- tion, ssNMR and EM, for a wide range of peptides, that an improved understanding of amyloid structure can be developed. Acknowledgements LCS is supported by a Wellcome Trust Research Career Development fellowship. OSM is funded by BBSRC. The authors would like to thank Prof. H. Saibil for providing Fig. 2, Dr P. Sikorski for providing coordinates for the Poly Q structure for Fig. 3A and Dr R. Tycko for coordinates for Ab(9–40) for Fig. 3C. References 1 Husby G, Stenstad T, Magnus JH, Sletten K, Nordvag BY & Marhaug G (1994) Interaction between circulat- ing amyloid fibril protein precursors and extracellular tissue matrix components in the pathogenesis of sys- temic amyloidosis. Clin Immunol Immunopathol 70, 2–9. 2 Kelly JW (1996) Alternative conformations of amy- loidogenic proteins govern their behavior. Curr Opin Struct Biol 6, 11–17. 3 Sunde M, Serpell LC, Bartlam M, Fraser PE, Pepys MB & Blake CC (1997) Common core structure of amyloid fibrils by synchrotron X-ray diffraction. J Mol Biol 273, 729–739. 4 Xing Y & Higuchi K (2002) Amyloid fibril proteins. Mech Ageing Dev 123, 1625–1636. 5 Blumenthal HT (2004) Amyloidosis: a universal disease of aging? J Gerontol A Biol Sci Med Sci 59, M361– M369. 6 Kelly JW (1998) The environmental dependency of protein folding best explains prion and amyloid dis- eases. Proc Natl Acad Sci USA 95, 930–932. 7 Kirkitadze MD, Bitan G & Teplow DB (2002) Para- digm shifts in Alzheimer’s disease and other neuro- degenerative disorders: the emerging role of oligomeric assemblies. J Neurosci Res 69, 567–577. 8 Stefani M & Dobson CM (2003) Protein aggregation and aggregate toxicity: new insights into protein fold- ing, misfolding diseases and biological evolution. J Mol Med 81, 678–699. 9 Tan SY & Pepys MB (1994) Amyloidosis. Histopathol- ogy 25, 403–414. 10 Pepys MB (1996) Amyloidosis. In The Oxford Text- book of Medicine (Weatherall DJ, Ledingham JGG & Warell DA, eds), pp. 1512–1524. Oxford University Press, Oxford, UK. 11 Lashuel HA, LaBrenz SR, Woo L, Serpell LC & Kelly JW (2000) Protofilaments, filaments, ribbons, and fibrils from peptidomimetic self-assembly: Implications for amyloid fibril formation and materials science. J Am Chem Soc 122, 5262–5277. 12 Kolstoe S & Wood S (2004) Perspectives for drug intervention in amyloid diseases. Curr Drug Targets 5, 151–158. 13 Serpell LC, Berriman J, Jakes R, Goedert M & Crowther RA (2000) Fiber diffraction of synthetic alpha-synuclein filaments shows amyloid-like cross-beta conformation. Proc Natl Acad Sci USA 97, 4897– 4902. O. S. Makin and L. C. Serpell Structures for amyloid fibrils FEBS Journal 272 (2005) 5950–5961 ª 2005 The Authors Journal compilation ª 2005 FEBS 5957 14 de la Paz ML, Goldie K, Zurdo J, Lacroix E, Dobson CM, Hoenger A & Serrano L (2002) De novo designed peptide-based amyloid fibrils. Proc Natl Acad Sci USA 99, 16052–16057. 15 Dobson CM (2004) Principles of protein folding, mis- folding and aggregation. Semin Cell Dev Biol 15, 3–16. 16 Eanes ED & Glenner GG (1968) X-ray diffraction stu- dies on amyloid filaments. J Histochem Cytochem 16, 673–677. 17 Cohen AS, Shirahama T & Skinner M (1982) Electron microscopy of amyloid. In Electron Microscopy of Protein (Harris I, ed.), pp. 165–205. Academic Press, London, UK. 18 Puchtler H, Sweat F & Levine M (1961) On the bind- ing of Congo red by amyloid. J Histochem Cytochem 10, 355–364. 19 Puchtler H & Sweat F (1965) Congo red as a stain for fluorescence microscopy of amyloid. J Histochem Cyto- chem 13, 693–694. 20 LeVine H III (1993) Thioflavine T interaction with synthetic Alzheimer’s disease beta-amyloid peptides: detection of amyloid aggregation in solution. Protein Sci 2, 404–410. 21 Zhang S (2003) Fabrication of novel biomaterials through molecular self-assembly. Nat Biotechnol 21, 1171–1178. 22 Waterhouse SH & Gerrard JA (2004) Amyloid fibrils in bionanotechnology. Curr Chem 57, 519–523. 23 Dobson CM & MacPhee CE (2004) Mixed Fibrils. United States patent application PCT ⁄ GB01 ⁄ 05190. 24 Goldsbury CS, Cooper GJ, Goldie KN, Muller SA, Saafi EL, Gruijters WT, Misur MP, Engel A, Aebi U & Kistler J (1997) Polymorphic fibrillar assembly of human amylin. J Struct Biol 119, 17–27. 25 Serpell LC, Sunde M, Benson MD, Tennent GA, Pepys MB & Fraser PE (2000) The protofilament sub- structure of amyloid fibrils. J Mol Biol 300, 1033–1039. 26 Makin OS & Serpell LC (2004) Structural characterisa- tion of islet amyloid polypeptide fibrils. J Mol Biol 335, 1279–1288. 27 Goldsbury C, Kistler J, Aebi U, Arvinte T & Cooper GJ (1999) Watching amyloid fibrils grow by time-lapse atomic force microscopy. J Mol Biol 285, 33–39. 28 Jimenez JL, Guijarro JI, Orlova E, Zurdo J, Dobson CM, Sunde M & Saibil HR (1999) Cryo-electron microscopy structure of an SH3 amyloid fibril and model of the molecular packing. Embo J 18, 815–821. 29 Jimenez JL, Nettleton EJ, Bouchard M, Robinson CV, Dobson CM & Saibil HR (2002) The protofilament structure of insulin amyloid fibrils. Proc Natl Acad Sci USA 99, 9196–9201. 30 Jimenez JL, Tennent G, Pepys M & Saibil HR (2001) Structural diversity of ex vivo amyloid fibrils studied by cryo-electron microscopy. J Mol Biol 311, 241–247. 31 Geddes AJ, Parker KD, Atkins ED & Beighton E (1968) Cross-beta’ conformation in proteins. J Mol Biol 32, 343–358. 32 Krejchi MT, Cooper SJ, Deguehi Y, Atkins EDT, Fournier MJ, Mason TL & Tirrell DA (1997) Crystal structures of chain-folded antiparallel beta-sheet assem- blies from sequence-designed periodic polypeptides. Macromolecules 30, 5012–5024. 33 Fandrich M & Dobson CM (2002) The behaviour of polyamino acids reveals an inverse side chain effect in amyloid structure formation. Embo J 21, 5682–5690. 34 Blake CCF & Serpell L (1996) Synchrotron X-ray stu- dies suggest that the core of the transthyretin amyloid fibril is a continuous b-helix. Structure 4, 989–998. 35 Sunde M & Blake CCF (1997) The structure of amy- loid fibrils by electron microscopy and X-ray diffrac- tion. Adv Protein Chem 50, 123–159. 36 Kirschner DA, Inouye H, Duffy LK, Sinclair A, Lind M & Selkoe DJ (1987) Synthetic peptide homologous to beta protein from Alzheimer disease forms amyloid-like fibrils in vitro. Proc Natl Acad Sci USA 84, 6953–6957. 37 Halverson K, Fraser PE, Kirschner DA & Lansbury PT Jr (1990) Molecular determinants of amyloid deposition in Alzheimer’s disease: conformational stud- ies of synthetic beta-protein fragments. Biochemistry 29, 2639–2644. 38 Fraser PE, Nguyen JT, Inouye H, Surewicz WK, Sel- koe DJ, Podlisny MB & Kirschner DA (1992) Fibril formation by primate, rodent, and Dutch-hemorrhagic analogues of Alzheimer amyloid beta-protein. Biochem- istry 31, 10716–10723. 39 Fraser PE, McLachlan DR, Surewicz WK, Mizzen CA, Snow AD, Nguyen JT & Kirschner DA (1994) Conformation and fibrillogenesis of Alzheimer A beta peptides with selected substitution of charged residues. J Mol Biol 244, 64–73. 40 Inouye H, Fraser PE & Kirschner DA (1993) Structure of beta-crystallite assemblies formed by Alzheimer beta-amyloid protein analogues: analysis by x-ray dif- fraction. Biophys J 64, 502–519. 41 Inouye H & Kirschner DA (1996) Refined fibril struc- tures: the hydrophobic core in Alzheimer’s amyloid beta-protein and prion as revealed by X-ray diffrac- tion. Ciba Found Symp 199, 22–35; discussion 35–9. 42 Malinchik SB, Inouye H, Szumowski KE & Kirschner DA (1998) Structural analysis of Alzheimer’s beta (1– 40) amyloid: protofilament assembly of tubular fibrils. Biophys J 74, 537–545. 43 Sikorski P, Atkins ED & Serpell LC (2003) Structure and texture of fibrous crystals formed by Alzheimer’s abeta (11–25) peptide fragment. Structure (Camb) 11, 915–926. 44 Petkova AT, Buntkowsky G, Dyda F, Leapman RD, Yau WM & Tycko R (2004) Solid state NMR reveals a pH-dependent antiparallel beta-sheet registry in Structures for amyloid fibrils O. S. Makin and L. C. Serpell 5958 FEBS Journal 272 (2005) 5950–5961 ª 2005 The Authors Journal compilation ª 2005 FEBS fibrils formed by a beta-amyloid peptide. J Mol Biol 335, 247–260. 45 Nguyen JT, Inouye H, Baldwin MA, Fletterick RJ, Cohen FE, Prusiner SB & Kirschner DA (1995) X-ray diffraction of scrapie prion rods and PrP peptides. J Mol Biol 252, 412–422. 46 Kirschner DA, Elliott-Bryant R, Szumowski KE, Gonnerman WA, Kindy MS, Sipe JD & Cathcart ES (1998) In vitro amyloid fibril formation by synthetic peptides corresponding to the amino terminus of apoSAA isoforms from amyloid-susceptible and amy- loid-resistant mice. J Struct Biol 124, 88–98. 47 Tycko R (2003) Insights into the amyloid folding problem from solid-state NMR. Biochemistry 42, 3151–3159. 48 Tycko R (2000) Solid-state NMR as a probe of amy- loid fibril structure. Curr Opin Chem Biol 4, 500–506. 49 Antzutkin ON, Balbach JJ & Tycko R (2003) Site- specific identification of non-beta-strand conformations in Alzheimer’s beta-amyloid fibrils by solid-state NMR. Biophys J 84, 3326–3335. 50 Antzutkin ON, Leapman RD, Balbach JJ & Tycko R (2002) Supramolecular structural constraints on Alzhei- mer’s beta-amyloid fibrils from electron microscopy and solid-state nuclear magnetic resonance. Biochemis- try 41, 15436–15450. 51 Balbach JJ, Petkova AT, Oyler NA, Antzutkin ON, Gordon DJ, Meredith SC & Tycko R (2002) Supramo- lecular structure in full-length Alzheimer’s beta-amy- loid fibrils: evidence for a parallel beta-sheet organization from solid-state nuclear magnetic reson- ance. Biophys J 83, 1205–1216. 52 Antzutkin ON, Balbach JJ, Leapman RD, Rizzo NW, Reed J & Tycko R (2000) Multiple quantum solid-state NMR indicates a parallel, not antiparallel, organiza- tion of beta-sheets in Alzheimer’s beta-amyloid fibrils. Proc Natl Acad Sci USA 97, 13045–13050. 53 Torok M, Milton S, Kayed R, Wu P, McIntire T, Glabe CG & Langen R (2002) Structural and dynamic features of Alzheimer’s Abeta peptide in amyloid fibrils studied by site-directed spin labeling. J Biol Chem 277, 40810–40815. 54 Benzinger TL, Gregory DM, Burkoth TS, Miller-Auer H, Lynn DG, Botto RE & Meredith SC (1998) Propa- gating structure of Alzheimer’s beta-amyloid (10–35) is parallel beta-sheet with residues in exact register. Proc Natl Acad Sci USA 95, 13407–13412. 55 Benzinger TL, Gregory DM, Burkoth TS, Miller-Auer H, Lynn DG, Botto RE & Meredith SC (2000) Two- dimensional structure of beta-amyloid (10–35) fibrils. Biochemistry 39, 3491–3499. 56 Burkoth TS, Benzinger TLS, Urban V, Morgan DM, Gregory DM, Thiyagarajan P, Botto RE, Meredith SC & Lynn DG (2000) Structure of the beta-amyloid (10–35) Fibril. J Am Chem Soc 122, 7883–7889. 57 Gregory DM, Benzinger TL, Burkoth TS, Miller-Auer H, Lynn DG, Meredith SC & Botto RE (1998) Dipolar recoupling NMR of biomolecular self-assemblies: determining inter- and intrastrand distances in fibri- lized Alzheimer’s beta-amyloid peptide. Solid State Nucl Magn Reson 13, 149–166. 58 Petkova AT, Ishii Y, Balbach JJ, Antzutkin ON, Leap- man RD, Delaglio F & Tycko R (2002) A structural model for Alzheimer’s beta-amyloid fibrils based on experimental constraints from solid state NMR. Proc Natl Acad Sci USA 99, 16742–16747. 59 Balbach JJ, Ishii Y, Antzutkin ON, Leapman RD, Rizzo NW, Dyda F, Reed J & Tycko R (2000) Amy- loid fibril formation by A beta 16–22, a seven-residue fragment of the Alzheimer’s beta-amyloid peptide, and structural characterization by solid state NMR. Biochemistry 39, 13748–13759. 60 Jarrett JT, Costa PR, Griffin RG & Lansbury PT (1994) Models of the beta protein C-terminus: differences in amyloid structure may lead to segregation of ‘long’ and ‘short’ fibrils. J Am Chem Soc 116, 9741–9742. 61 Lansbury PT Jr, Costa PR, Griffiths JM, Simon EJ, Auger M, Halverson KJ, Kocisko DA, Hendsch ZS, Ashburn TT, Spencer RG, et al. (1995) Structural model for the beta-amyloid fibril based on interstrand alignment of an antiparallel-sheet comprising a C-ter- minal peptide. Nat Struct Biol 2, 990–998. 62 Kammerer RA, Kostrewa D, Zurdo J, Detken A, Garcia-Echeverria C, Green JD, Muller SA, Meier BH, Winkler FK, Dobson CM, et al. (2004) Exploring amy- loid formation by a de novo design. Proc Natl Acad Sci USA 101, 4435–4440. 63 Naito A, Kamihira M, Inoue R & Saito H (2004) Structural diversity of amyloid fibril formed in human calcitonin as revealed by site-directed 13C solid-state NMR spectroscopy. Magn Reson Chem 42, 247–257. 64 Gordon DJ, Balbach JJ, Tycko R & Meredith SC (2004) Increasing the amphiphilicity of an amyloido- genic peptide changes the beta-sheet structure in the fibrils from antiparallel to parallel. Biophys J 86, 428– 434. 65 Tycko R (2004) Progress towards a molecular-level structural understanding of amyloid fibrils. Curr Opin Struct Biol 14, 96–103. 66 Serpell LC & Smith JM (2000) Direct visualisation of the beta-sheet structure of synthetic Alzheimer’s amy- loid. J Mol Biol 299, 225–231. 67 Kajava AV, Baxa U, Wickner RB & Steven AC (2004) A model for Ure2p prion filaments and other amy- loids: the parallel superpleated beta-structure. Proc Natl Acad Sci USA 101, 7885–7890. 68 Kajava AV, Aebi U & Steven AC (2005) The parallel superpleated beta-structure as a model for amyloid fibrils of human amylin. J Mol Biol 348, 247–252. O. S. Makin and L. C. Serpell Structures for amyloid fibrils FEBS Journal 272 (2005) 5950–5961 ª 2005 The Authors Journal compilation ª 2005 FEBS 5959 [...]... amyloid- like fibrils Nature 435, 773–778 Structures for amyloid fibrils 104 Ritter C, Maddelein ML, Siemer AB, Luhrs T, Ernst M, Meier BH, Saupe SJ & Riek R (2005) Correlation of structural elements and infectivity of the HET-s prion Nature 435, 844–848 105 Petkova AT, Leapman RD, Guo Z, Yau WM, Mattson MP & Tycko R (2005) Self-propagating, molecular-level polymorphism in Alzheimer’s beta -amyloid fibrils Science... Doucet J & Melki R (2003) The native-like conformation of Ure2p in fibrils assembled under physiologically relevant conditions switches to an amyloid- like conformation upon heat-treatment of the fibrils J Struct Biol 141, 132–142 Inouye H, Domingues FS, Damas AM, Saraiva MJ, Lundgren E, Sandgren O & Kirschner DA (1998) Analysis of x-ray diffraction patterns from amyloid of biopsied vitreous humor and kidney... basis for amyloid fibril formation and stability Proc Natl Acad Sci USA 102, 315–320 102 Diaz-Avalos R, Long C, Fontano E, Balbirnie M, Grothe R, Eisenberg D & Caspar DL (2003) Crossbeta order and diversity in nanocrystals of an amyloidforming peptide J Mol Biol 330, 1165–1175 103 Nelson R, Sawaya MR, Balbirnie M, Madsen AO, Riekel C, Grothe R & Eisenberg D (2005) Structure of the cross-beta spine of amyloid- like.. .Structures for amyloid fibrils O S Makin and L C Serpell 69 Lazo ND & Downing DT (1997) Beta-helical fibrils from a model peptide Biochem Biophys Res Commun 235, 675–679 70 Sieber V, Jurnak F & Moe GR (1995) Circular dichroism of the parallel beta helical proteins pectate lyase C and E Proteins 23, 32–37 71 Khurana R & Fink AL (2000) Do parallel beta-helix proteins have a unique Fourier transform... ML & Serrano L (2004) Sequence determinants of amyloid fibril formation Proc Natl Acad Sci USA 101, 87–92 99 Chiti F, Stefani M, Taddei N, Ramponi G & Dobson CM (2003) Rationalization of the effects of mutations on peptide and protein aggregation rates Nature 424, 805–808 100 Gazit E (2002) A possible role for pi-stacking in the self-assembly of amyloid fibrils Faseb J 16, 77–83 101 Makin OS, Atkins E,... yeast prion Sup35 amyloid fibers Biochem Biophys Res Commun 315, 739–745 Bousset L, Thomson NH, Radford SE & Melki R (2002) The yeast prion Ure2p retains its native alphahelical conformation upon assembly into protein fibrils in vitro Embo J 21, 2903–2911 Baxa U, Taylor KL, Wall JS, Simon MN, Cheng N, Wickner RB & Steven AC (2003) Architecture of Ure2p prion filaments: the N-terminal domains form a central... Nature 412, 143–144 75 Sikorski P & Atkins E (2005) New model for crystalline polyglutamine assemblies and their connection with amyloid fibrils Biomacromolecules 6, 425–432 76 Jenkins J & Pickersgill R (2001) The architecture of parallel beta-helices and related folds Prog Biophys Mol Biol 77, 111–175 77 Wetzel R (2002) Ideas of order for amyloid fibril structure Structure (Camb) 10, 1031–1036 78 Yoder... Met30 familial amyloidotic polyneuropathy (FAP) patients: axially arrayed TTR monomers constitute the protofilament Amyloid 5, 163–174 Eneqvist T, Andersson K, Olofsson A, Lundgren E & Sauer-Eriksson AE (2000) The beta-slip: a novel concept in transthyretin amyloidosis Mol Cell 6, 1207–1218 Blake C & Serpell L (1996) Synchrotron X-ray studies suggest that the core of the transthyretin amyloid fibril is... 91 92 93 94 95 96 97 trometric studies of amyloid fibrils Protein Sci 12, 635–643 Kheterpal I, Lashuel HA, Hartley DM, Walz T, Lansbury PT Jr & Wetzel R (2003) Abeta protofibrils possess a stable core structure resistant to hydrogen exchange Biochemistry 42, 14092–14098 Williams AD, Portelius E, Kheterpal I, Guo JT, Cook KD, Xu Y & Wetzel R (2004) Mapping abeta amyloid fibril secondary structure using scanning... parallel beta-helix proteins have a unique Fourier transform infrared spectrum? Biophys J 78, 994–1000 72 Lazo ND & Downing DT (1998) Amyloid fibrils may be assembled from beta-helical protofibrils Biochemistry 37, 1731–1735 73 Perutz MF, Finch JT, Berriman J & Lesk A (2002) Amyloid fibers are water-filled nanotubes Proc Natl Acad Sci USA 99, 5591–5595 74 Perutz MF & Windle AH (2001) Cause of neural death . highly detailed structures for amyloid fibrils formed from cer- tain peptides [43,75,101,103] (Fig. 3). The evidence for these structures appears to clearly exclude other mod- els. It may be that some fibrils, . zipper’ [103] Ab, amyloid b-peptide; AFM, atomic-force microscopy; EM, electron microscopy; IAPP, islet amyloid polypeptide; ssNMR, solid state NMR; TTR, transthyretin. Structures for amyloid fibrils O Parkinson’s disease) [13]. As amyloid- like fibrils can also be formed in vitro from protein unconnected to the amyloidoses, the distinction between amyloid and amyloid- like fibrils is blurred [14]. It