Amyloid fibrils

Abstract

Amyloid refers to the abnormal fibrous, extracellular, proteinaceous deposits found in organs and tissues. Amyloid is insoluble and is structurally dominated by β-sheet structure. Unlike other fibrous proteins it does not commonly have a structural, supportive or motility role but is associated with the pathology seen in a range of diseases known as the amyloidoses. These diseases include Alzheimer's, the spongiform encephalopathies and type II diabetes, all of which are progressive disorders with associated high morbidity and mortality. Not surprisingly, research into the physicochemical properties of amyloid and its formation is currently intensely pursued. In this chapter we will highlight the key scientific findings and discuss how the stability of amyloid fibrils impacts on bionanotechnology.

Introduction

Amyloid fibrils are formed by normally soluble proteins, which assemble to form insoluble fibers that are resistant to degradation. Their formation can accompany disease and each disease is characterized by a specfic protein or peptide that aggregates. Well known examples of amyloid diseases include Alzheimer's disease, Diabetes type 2 and the spongiform encephalopathies (e.g., Mad cow disease). The amyloid fibrils are deposited extracellularly in the tissues and are thought to have a pathogenic effect.¹ The fibrillar assemblies are inherently stable and structural studies have revealed that they are composed predominantly of β-sheet structure in a characteristic cross-β conformation. Recently, a number of examples of functional amyloid have been identified including a constituent of melanosomes, curli and hydrophobins.²

The History of Amyloid

The term ‘amyloid’ was coined initially by Schleiden and then by Virchow in the mid-19th century to describe the iodine stained deposits seen in liver at autopsy. Consequently the deposits were thought to be carbohydrate in nature until their high nitrogen content was later established.³ Nevertheless, the inaccurately descriptive name was retained for these highly proteinaceous deposits. Further tinctorial properties included the specific binding of amyloid to the dye Congo Red which produced an apple green birefringence when examined between cross polarisers in a light microscope⁴ (Fig. 1). This finding suggested that amyloid was fibrillar in structure and subsequent transmission electron micrographs of amyloid confirmed this.⁵ Further progress in biochemical and biophysical techniques enabled the isolation of amyloid fibrils from tissues⁶ and their now characteristic cross-β structure was interpreted from X-ray fiber diffraction patterns.⁷

Figure 1.

Isolated amyloid fibrils composed of Aα chain fragment of fibrinogen (a) stained with Congo red and visualized by light microscopy and (b) between crossed polars, showing characteristic apple-green birefringence. Figure adapted from reference ⁸³.

Amino acid composition and sequence analysis of the proteins comprising a range of ex vivo amyloid fibrils revealed that each amyloid disorder was associated with a particular protein or peptide.⁸ To date, more than 20 plasma proteins have been identified that form amyloid (Table 1). Interestingly, despite the obvious differences in amino acid sequences and native structure, these amyloidogenic peptides all appear to share a common β-sheet conformation of their polypeptide backbone^7,9 and it is likely that this characteristic confers the fibrillar, proteolytic resistant and insoluble characteristics to all forms of amyloid.

Table 1.

Some amyloidoses and their respective precursors and amyloid-ogenic proteins

Disease	Precursor protein	Amyloid protein
Alzheimer's disease	Amyloid precursor protein	Aβ peptides
Atrial amyloidosis	Atrial natriuretic factor (ANF)	Amyloid ANF
Spongiform encephalopathies	Prion protein (PrPc)	PrPsc
Primary systemic amyloidosis	Immunoglobulin light and heavy chains	AL and AH
Senile systemic amyloidosis	Wild-type transthyretin	ATTR
Haemodialysis-related amyloidosis	β2-microglobulin	Aβ2M
Hereditary nonneuropathic systemic amyloidosis	Lysozyme	ALys
Type II diabetes	Pro-IAPP	IAPP or “amylin”
Injection-localized amyloidosis	Insulin	AIns
Secondary systemic amyloidosis	(Apo) serum amyloid A	Serum amyloid A
Hereditary cerebral amyloid angiopathy	Cystatin C	ACys
Finnish hereditary systemic amyloidosis	Gelsolin	AGel
Familial amyloid polyneuropathy I	Transthyretin variants	ATTR
Familial amyloid polyneuropathy II	Apolipoprotein A1	AApoA1
Ageing pituitary, prolactinomas	Prolactin	APro
Familial amyloidosis	Fibrinogen αA-chain	AFib
British familial dementia	Amyloid Bri Precursor Protein	ABri

The finding that lysosomal extracts digested immunoglobulin light chain precursor protein into amyloidogenic fragments¹⁰ led to the realisation that many amyloid forming peptides were produced by the proteolytic processing of a precursor protein (Table 1). By 1982, Prusiner put forward the ‘prion’ hypothesis that described an infectious protein particle capable of causing scrapie (a fatal neurodegenerative disease) in sheep.¹¹ This protein was amyloidogenic but unlike other amyloid forming proteins it was infective and was termed: prion. Two years later, Aβ, the peptide forming the bulk of amyloid plaques in Alzheimer's diseased brain specimens, was identified and biochemically characterized.¹² In 1986, attention was refocused on prions when a devastating epidemic affected cattle in the UK. These animals were dying of Bovine spongiform encephalopathy (BSE), a prion disease with links to scrapie prion from sheep. Ten years later, a new fatal, neurodegenerative prion disease emerged in humans, variant Creutzfeldt-Jacob disease (vCJD) and the transmissibility of prions from species to species could not be denied as vCJD was strongly linked with exposure to the BSE agent. Interest in prion diseases and amyloid formation (believed to be important to the pathogenesis of the diseases) now escalated.

Earlier biochemical analysis of amyloid deposits had demonstrated that amyloid was not merely composed of fibrils but also contained the serum amyloid protein (SAP) and proteoglycans. Although SAP is believed to have a scavenging function whereby it may stabilize amyloid formation,¹³ the role of these proteins in amyloid formation and deposition is yet to be discerned. In 1988, the Pepys laboratory radiolabelled SAP and used it as an in vivo tracer to follow the progress of amyloid disorders in patients.¹⁴

Three years later, the propensity of a peptide to form amyloid was found to be sequence dependent. Polypeptides containing glutamine repeats were found to be associated with some inherited, neurodegenerative amyloidoses.¹⁵ Subsequently, Kelly and coworkers found that purified transthyretin formed fibrils in vitro when its native state was partially unfolded under acidic conditions.¹⁶ This finding suggested that induced conformational changes were sufficient to trigger aggregation into amyloid fibrils and implied that erroneous protein folding or misfolding underpinned the mechanism of amyloid formation, fuelling a new area of intense research in protein chemistry.

Amyloidogenesis: Assembly of Amyloidogenic Proteins

With advances in solid phase protein synthesis techniques in the early nineties it was possible to use synthetic Alzheimer's related peptide, Aβ(1–42), its truncated homologues and other short amyloid forming peptides to produce amyloid-like fibrils.¹⁷ These were physically, morphologically and tinctorially similar to ex vivo amyloid fibrils and enabled the study of amyloid ultrastructure and formation. An extensive range of biophysical tools and techniques including nuclear magnetic resonance (NMR), Circular Dichroism (CD), X-ray fiber diffraction, atomic force and electron microscopy (AFM and EM), Fourier Transform Infrared Spectroscopy (FTIR) have all since contributed to a better understanding of the structure of amyloid fibrils at a molecular level.

Studying the mechanism of formation of amyloid fibrils is of major importance since insights into the mechanisms underlying polymerisation of soluble, monomeric peptide into mature insoluble fibrils may provide researchers with possible therapeutic approaches to halting, reversing or avoiding fibril formation. From milestone studies, it is known that conversion of soluble peptide to insoluble amyloid often involves the production of a partially unfolded intermediate.^18–20 This thermodynamically unfavorable state rapidly progresses to the stable amyloidogenic form and the kinetics of this transition have been studied in vitro using a range of biophysical methods including light scattering, size exclusion chromatography, fluorimetry and ultracentrifugation. The results of such studies propose a nucleation dependent polymerisation model to describe fibril formation, likening the process to crystallization.²¹ A heterogenous nucleus (‘seed') or peptide micelle form above a critical ‘threshold’ concentration and fibrils nucleate within these, elongating by irreversibly binding monomers to their free ends.^22,23 Fibril growth may be represented diagrammatically as a lag exponential growth curve where the phase is considerably shortened in the presence of seeds.

In the evolution of mature fibrils, several metastable intermediates have been identified and isolated.^22,24–26 These include very early species: dimers, trimers, tetramers (collectively known as oligomers²⁷), Aβ derived diffusible ligands or ADDLs²⁸ and the later bead-like structures up to 200 nm in length called protofibrils.²⁹ Alternatively, fibril formation may follow an ‘offset pathway’ without the production of fibrils but instead involving conversion of the intermediates into amorphous deposits.

The dynamics of protein folding/misfolding appear to play a key role in in vitro fibril formation and consequently several molecular mechanisms to explain amyloid formation have been proposed; including the polar zipper³⁰ and domain swapping models.^31,32 It has been proposed that amyloid formation is a generic property of all peptides³³ since under denaturing conditions many normally globular, nondisease related proteins, have been shown to assemble to form fibrils.^34,35 However, the propensity of a peptide to form amyloid is dependent on several factors including: polypeptide charge, sequence, hydrophobicity and secondary structure. Consequently several algorithms have been developed to predict the propensity and the rate at which different sequences will aggregate and which mutations will result in an increase or decrease of aggregation rate.^36–40 These algorithms will prove useful in determining which regions of a polypeptide chain is specifically involved in fibril formation or the amyloid core structure.

The Structure of the Amyloid Fibril

Amyloid fibrils are insoluble and heterogeneous, so commonly used methods of structure determination are difficult. Therefore, most studies have involved X-ray fiber diffraction, electron microscopy⁴¹ and more recently, solid state nuclear magnetic resonance (ssNMR)42 and electroparamagnetic resonance.^43–45

Electron and atomic force microscopy have given insights into the macromolecular structure of amyloid fibrils and have shown that fibrils are long, straight and unbranching (Fig. 2A) and are made up of individual subunits named “protoflaments.”^46–48 These may vary in number and are often observed to twist around one another to form the mature fibril.^46,48–50 Synthetic amyloid-like fibrils, may vary in morphology and this may depend upon the assembly conditions.^46,51

Figure 2.

Synthetic amyloid fibrils made from Aβ peptide (A) electron micrograph showing long, straight, unbranching fibrils. (B) X-ray fiber diffraction pattern from partially aligned amyloid fibrils showing the characteristic “cross-β” diffraction pattern. (C) The structure of the Aβ amyloid fibril interpreted from ssNMR data,⁶⁷ showing the top view of the fiber (i and ii) with side chains (i), showing the importance of side chain packing with in the fiber and as a cartoon (ii). The side view (iii) revealing the β-strands running perpendicular to the fiber axis.

Cryo-electron microscopy and single particle processing of mature amyloid fibrils composed of the SH3 domain of phosphotidylinositol-3′-kinase indicated that a single fibril was comprised of four protofilaments wound around a central core.² Consequently, a molecular model of a SH3 amyloid fibril has been generated showing the protofilaments each composed of continuous β-sheet structure.⁴⁸ Further cryo-electron microscopy studies revealed that synthetic amyloid fibrils formed by insulin,⁵⁰ lysozyme⁴⁹ and Aβ(1–42)⁵² are also composed of several protofilaments wound around one another. The numbers of protofilaments can differ from 2 to 6.⁵⁰

Early studies on amyloid fibrils revealed that they shared a common X-ray diffraction fingerprint⁷ and this was later confirmed in a fiber diffraction study of a number of different ex-vivo and synthetic amyloid fibrils.^9,53,54 The X-ray diffraction pattern given by amyloid fibrils is “cross-β” (Fig. 2B), a diffraction fingerprint first identified for silk from the egg stalk of the lacewing, Chrysopa.⁵⁵ The pattern indicates that these fibrous molecules share a particular core structure consisting of β-sheet conformation in which the hydrogen bonding direction runs parallel to the fiber axis and the β-strands are perpendicular, much like the rungs of a ladder (Fig. 2). The diffraction pattern consists of two major reflections at 4.7 Å and 10 Å found on orthogonal axes and arising from the hydrogen bonding distances between β-strands and side chain packing between the sheets respectively (see Fig. 2).^9,56 More detailed X-ray diffraction patterns have been obtained from synthetic amyloid fibrils and the additional information obtained has enabled molecular models to be proposed.⁵⁷ Generally these models are cross-β in nature and the β-sheet conformation forms the core of the structure. The β-strands are hydrogen bonded along the length of the β-sheet structure and parallel to the fiber axis.⁵⁸ The β-sheet ribbons are associated via side chain interactions⁵⁷ that serve to stabilize the structure. Very recently, a number of short amyloidogenic peptides have been crystallized this has enabled their atomic structure to be elucidated by X-ray crystallography.^59,60 This work has given insights into the nature of the interactions that drive the association of the β-sheets via the amino acid side chains.

SsNMR structural studies have significantly added to our knowledge of the structure of the amyloid fibril⁶¹ and this work has been complemented by results from EPR on amyloid formed by a number of different disease associated peptides.⁴⁵ SsNMR of amyloid fibrils formed by various peptides homologous to regions of the Alzheimer's peptide, Aβ have shown that β-sheet may be arranged parallel or anti-parallel within the protofilaments depending on the properties of the precursor polypeptide.^62–66 Studies of fibrils formed from full-length Aβ, have shown that the peptide folds into a β-bend structure that then associates with other molecules to form parallel, in register β-structure^67,68 (Fig. 2C).

Small angle X-ray scattering has recently contributed insights into both the structure and the assembly of amyloid fibrils⁶⁹ revealing for the first time, the oligomeric assembly of insulin fibrils. The SAX study suggested that insulin fibrils are assembled from a helical fibrillation precursor composed of five to six insulin monomers. The mature amyloid fibril is composed of three individual filaments that wrap around one another⁶⁹ (Fig. 3B). Small angle X-ray scattering from fibrils formed by alpha-synuclein revealed a contrasting structure in which a single filament appears to make up the mature fiber (Vestergaard, personal communication) (Fig. 3A).

Figure 3.

Models of mature protein fibrils based on Small-Angle X-ray scattering solution data. (A) Human alpha-synuclein fibrils and (B) human insulin fibrils.⁶⁹ The results suggest that insulin fibrils (B) are formed of three intertwining protofibrils, whereas a-synuclein fibril (A) consist of only one protofibril. Each protofibril is assumed to consist of two intertwining protofilaments. Four and three repeating units are shown for alpha-synuclein and insulin respectively.

Toxicity of Amyloid: The Search for the Toxic Species and Its Mode of Action

It is clear that amyloid deposition is a consequence of the amyloidoses; what is uncertain is whether it is a causative agent in its pathogenesis or a secondary event. The amyloid cascade hypothesis⁷⁰ proposed that altered metabolism of amyloid precursor protein (APP) initiates the pathogenesis of Alzheimer's disease (AD) leading to aggregation of Aβ and formation of neuritic plaques. These plaques would cause further pathological changes including the formation of neurofibrillary tangles and compromised synaptic connections ultimately resulting in neuronal cell loss and dementia. However, there was no correlation between the density of plaques and tangles and the severity of AD. Instead, the concentration of soluble Aβ appeared to correlate with cognitive impairment in other studies.^71–73 This finding set the premise for studies suggesting that soluble nonfibrillar intermediates, such as oligomers (20 to >50 kDa globular aggregates, including ADDLs) and protofibrils (curvilinear structures 4–11 nm in diameter and ≤200 nm long) are the actual initiators of AD pathogenesis and that mature fibril formation represents the end point of the disease.

Usually small organic compounds are screened and selected for their binding and inhibitory effects on amyloid precursors or the enzymes involved in their proteolytic processing. These potential anti-amyloid drugs are often highly toxic and have a profound effect on the immune system. Biological ligands such as monoclonal antibodies have been raised against amyloid and are more successful at plaque clearance and reduction in soluble Aβ levels in the CNS.⁷⁴ However during phase II trials of an Aβ vaccine, 6% of subjects developed an inflammatory response (meningoencephalitis) which halted further clinical development. The search therefore for an effective and safe anti-amyloid drug continues. In the interim, several questions remain unanswered: (1) which intermediate species are responsible for the toxicity of amyloid; (2) what is its atomic structure and the exact mechanism of toxicity and (3) how can this toxicity be safely reversed.

“Exploiting Amyloid”

Functional Amyloid.

Amyloid fibrils are extremely stable and resistant to degradation. They have been described as having a similar tensile strength to steel,⁷⁵ a property that they share with their structural cousin, silk. Recently, a number of nonpathogenic, functional forms of amyloid have been identified in bacteria, fungi, insects and mammals.² Curli is a functional fiber found on the surface of bacteria such as E. coli. The fibers share structural similarities to amyloid and formed by a protein called CsgA. Fibrillization is carefully regulated by a protein machine complex.^76,77 Many fungi produce amphipathic proteins called hydrophobins.⁷⁸ These have the ability to assemble into β-sheet rich fibrils at air-water interfaces and these are thought to play a protective role in fungal structures such as spores and fruiting bodies. The yeast prions have been suggested to play a functional role, forming cytoplasmic fibrillar assemblies that may be associated with a method of heritable information transfer.⁷⁹

In mammals, a protein called Pmel17 is involved in the biosynthesis of melanin. Recently, the Pmel17 protein has been found to form fibrous structures with all the characteristics of amyloid fibrils and these structures are found at the surface of melanosomes. These are thought to be involved in the templating of melanin biosynthesis.⁸⁰

Bionanotechnological Applications.

The amyloid fibril has been increasingly examined for its potential role in forming nanotubular scaffolding for bionanotechnology. The fibrils themselves can be very strong, as previously discussed and these fibers can also be functionalized by assembling fusion proteins⁸¹ or used as a template for binding to metals.^80,82 Baldwin and co-workers⁸¹ assembled a fusion protein composed of a functional cytochrome b562 with an amyloidogenic SH3 sequence. The assemblies have the amyloid-like core, displaying functional, folded, globular cytochrome.⁸¹ Nanowires have been fabricated by assembling proteins such as the N-terminal region of the yeast prion, Sup35. Conjugate colloidal gold particles were associated along the fibers using exposed cysteine residues of a variant Sup35,⁸⁰ yielding wires around 100 nm in diameter (Fig. 4A). A very short peptide, composed of two phenylalanine residues, assembles to form amyloid-like nanotubes and these may be functionalised using ionic silver in the centre of the nanotube (Fig. 4B).⁸² This work yielded nanowires around 20 nm in diameter.

Figure 4.

Amyloid-like fibers for bionanotechnology. (A) Nanowires based on the N-terminal region of the yeast prion, Sup35. Nanogold was covalently linked to the engineered cysteine residues in the protein and conjugate colloidal gold and silver particles were associated along the fibers to form wires,⁸⁰ (B) assembly of diphenylalanine to form nanotubes that can be filled with silver to make nanowires.⁸²

Conclusions

Amyloid fibrils are extremely strong, highly ordered and organized fibers that can be formed by a large number of proteins and peptides. Their stability and insolubility means that they are useful in a large number of naturally occurring forms as well as bionanotechnology. However, it also means that they are destructive and have the ability to accumulate in the tissues in disease. Although recent key advances have been made in understanding the structure of amyloid-like fibrils, better understanding of the mechanism of assembly is essential for combating amyloid diseases as well as exploiting the potential benefits of protein/peptide design.

Note

This manuscript has been previously published: Rambaran RN, Serpell LC. Amyloid Fibrils: Abnormal Protein Assembly. In: Scheibel T, editor. Fibrous Proteins. Austin: Landes Bioscience; 2008. pp. 1–11.