Peter Kempen
Master Biomedical sciences thesis
For more info or word document: peterkempen @ hotmail.com (remove spaces)
Master Biomedical sciences thesis
Chapter 1
Introduction. 4
Amyloid-beta. 4
Hbp. 5
Common properties. 7
Chapter 2 beta-structures. 8
Chapter 3 the beta solenoid Hbp. 11
Coil shape. 11
Oligomerisation. 11
Beta-arcs. 13
Internal environment 16
Handedness. 17
Twist 17
Beta-structure formation. 18
Chaper 4. Alzheimer beta-amyloid. 19
Model for Ab40. 19
Ab42 model 22
Fibrillisation. 25
Steric zipper 26
Interacting residues. 27
Chapter 5. Common properties. 28
Cross sectional shape. 29
Handedness and Twist 30
Arches. 31
Amino acid composition. 31
Formation the beta-structure. 34
Abbreviations. 35
References. 36
Introduction. 4
Amyloid-beta. 4
Hbp. 5
Common properties. 7
Chapter 2 beta-structures. 8
Chapter 3 the beta solenoid Hbp. 11
Coil shape. 11
Oligomerisation. 11
Beta-arcs. 13
Internal environment 16
Handedness. 17
Twist 17
Beta-structure formation. 18
Chaper 4. Alzheimer beta-amyloid. 19
Model for Ab40. 19
Ab42 model 22
Fibrillisation. 25
Steric zipper 26
Interacting residues. 27
Chapter 5. Common properties. 28
Cross sectional shape. 29
Handedness and Twist 30
Arches. 31
Amino acid composition. 31
Formation the beta-structure. 34
Abbreviations. 35
References. 36
Abstract
Peritonitis and Alzheimer do not seem to have any overlap. On a molecular basis, where Hemoglobin protease (Hbp) and Amyloid-beta (Ab) form the main problem, they are related by forming beta-structures. Both were found to have cross-beta structures. They contain parallel beta-sheets formed by stacks of beta-strands. In Hbp this is a helical beta-structure, in Ab this is stacks of monomers that form the sheets.
The structure of Hbp has been determined for many years now. More exiting are the systematic analyses that have been done on the protein family of beta-solenoids. With this the functions and specifications of the beta-helical structure can be determined.
The structure of Hbp has been determined for many years now. More exiting are the systematic analyses that have been done on the protein family of beta-solenoids. With this the functions and specifications of the beta-helical structure can be determined.
The molecular model for Ab has been unclear for many years due to the nature of a fibril. It has been difficult to determine the structural model with standard techniques. Recently, Japanese research showed progress by use of proline mutagenesis. This gave a more detailed construction of the monomers in the fibrillated form.
A combination of this information and systematic research on beta-solenoids can be used to predict structures and gain new research questions for investigation. Several structural elements can be used to describe the properties of fibrillating proteins; parallel or antiparallel beta-sheets, cross sectional shape, arc (turn) type, amino acid ladders, twist, handedness and type of steric zippers. These will be discussed in this review.
A combination of this information and systematic research on beta-solenoids can be used to predict structures and gain new research questions for investigation. Several structural elements can be used to describe the properties of fibrillating proteins; parallel or antiparallel beta-sheets, cross sectional shape, arc (turn) type, amino acid ladders, twist, handedness and type of steric zippers. These will be discussed in this review.
Chapter 1 Introduction
The beta conformation of amino acids is one of the earliest secondary protein structures described. These structures have been found by observations in studies of silk [5]. In the last decade or so it has become clear that these conformations play a crucial role in disease causing protein structures. These include amyloidogenic proteins and other beta-fibrous proteins.
Amyloids are associated with Alzheimer’s disease, the transmissible spongiform encephalopathies, including bovine spongiform encephalopathy, Creutzfeldt Jakob disease and Kuru, late-onset, or type II, diabetes and a number of systemic polyneuropathies [6]. Later, the involvement of amyloid fibrils in Huntington's disease was demonstrated [7]. The origin of the amyloidosis is dependent on the protein involved. The development of amyloid deposits is linked to either genetic mutation, incorrect processing, or the abnormal accumulation of wild-type proteins.
Amyloid-beta
The Amyloid-beta (Ab) polypeptide may be the best studied of all amyloid forming proteins because of the deposits it forms in the brains of patients with Alzheimer’s disease [8]. Ab appears in a 39 to 43 amino acid sequences which is a product of the beta- and gamma-secretase. They are defined as Ab39 till Ab43 where is Ab42 most amylodogenic and Ab40 represents the most abundant Ab isoform in the brain [9, 10].
The Amyloid-beta (Ab) polypeptide may be the best studied of all amyloid forming proteins because of the deposits it forms in the brains of patients with Alzheimer’s disease [8]. Ab appears in a 39 to 43 amino acid sequences which is a product of the beta- and gamma-secretase. They are defined as Ab39 till Ab43 where is Ab42 most amylodogenic and Ab40 represents the most abundant Ab isoform in the brain [9, 10].
 |
| Figure 1.1 The core of a beta-helical model of the amyloid-beta fibril [1]Stacking of monomers form a beta-fibrous structure. |
In addition to studies of full-length Ab, there has been a myriad of studies examining the fibril forming and structural properties of shorter segments of Ab. They reveal that peptide sequence, fragment length, and fibrillization conditions (e.g., pH) all affect the structure of the fibril. They may be composed of parallel or antiparallel sheets, contain turns, and multiple registers of the stacked peptides are possible.
First used to identify Ab was the characteristic green birefringence when stained with the dye Congo Red [11]. Later, the particular appearance under the electron microscope [12] and its particular X-ray diffraction pattern were used [13]. High-resolution methods structure (protein crystallography and NMR spectroscopy) eventually failed to determine the molecular structure because of the polymeric character and insolubility of the fibrils involved. Therefore, X-ray fiber diffraction, electron microscopy (EM), optical spectroscopy, and other biophysical approaches are the main sources of data for the models of Ab presented to date. In the last decade major progress has been made to solve the atomic structure of fibrous beta-proteins. Especially, the last few years were important for new structural information. Several new experimental techniques, including solid state NMR, scanning transmission EM mass measurements, electron paramagnetic resonance spectroscopy of spin-labeled derivatives [4, 14, 15] and scanning mutagenesis (newest models: [16-19], have emerged and provided new insights in the unfolding structural model.
It became clear that amyloid fibrils possess a structural spine that is formed by a cross-beta structure. This structure consists of oriented beta-sheets with interstrand hydrogen bonds aligned parallel with the main fibril axis. This was possible to see by X-ray [20] by comparing X-ray diffraction patterns of homologous beta-fibrous proteins known. It remained the only knowledge of the structure for a long time. The exact molecular structure remained to be explored. Lately, scanning mutagenesis gained new insight in the model. This will be discussed in more detail in chapter 4.
Hbp
Other beta-fibrous proteins that are related to amyloid by their large beta-structures are some virulence factors of bacteria [21]. Hemoglobin binding protease is such a pathogenic virulence factor. This is one of the key elements in severe peritonitis and systematic shock. This beta-structure is a protease secreted by E. coli that is involved in a bacterial synergy with B. fragilis. It binds hemoglobin and cuts out the heme group to supply B. fragilis with iron. This opportunistic pathogen can help both species in their growth. B. fragilis can help both species escape an attack by the immune system. This causes a life-threatening situation with mortality rates in excess of 50% [22, 23].
Hbp has been crystallized to gain its atomic structure by x-ray crystallography. The model shows an active site, a right-handed beta-helical C-terminal tail, forming three parallel beta sheets [3] and a capping structure to close the barrel [24]. The barrel extends from residue 260 onwards and includes 24 turns of right-handed beta-helix. This beta-helix is part of a large family of beta-solenoids. Furthermore, it also has a large N-terminal globular domain and a smaller domain of 75 residues (481-556) about half-way along the beta-helix, as well as decorative loops where the polypeptide chain departs from the helical stem. An alpha helix (Leu245-Asp255) connects this domain to the beta helical stem.
Figure 1.2 A structural model of the hemoglobin protease from E. coli [3]. The beta-solenoid domains are shown in blue and the remaining regions in dark yellow. (A) Ribbon diagram of the 3D structure and (B) linear map of the domain distribution within the amino acid sequence. [24]
The beta-helix is part of a protein family, the beta-solenoids. These are structures that contain stackings of beta-strands. The number of solved structures with such a fold was large enough to support a systematic analysis [24]. Here, the structures are classified based on handedness, twist, oligomerization state, and coil shape. These properties are all described in chapter 3.
Common properties
Both proteins form cross-beta structures, in which the beta-strands interact with opposite and neighboring strands [13, 25-27]. The relevance of these proteins in important human diseases has led to considerable efforts to solve their structures. This may help research for targets in the development of new medicine. However, the determination of the structures of the beta-amyloid fibril has been a difficult task. Not all functions of the individual domains are always clear. Despite the difference in function of both beta-amyloid and the Hbp autotransporter, there is enough overlap that we can study. In this paper we try to explain the similarity between Amyloid-beta of Alzheimer disease and Hemoglobin protease that causes Peritonitis. The common properties of both will be discussed in chapter 5. First, an introduction to beta-structures will be given in chapter 2.
Chapter 2 beta-structures
There are two basic beta-structural arrangements found in proteins: the parallel and antiparallel beta-sheet structures [28, 29]. Here, the backbone of the amino acids makes hydrogen bonds with another strand. Only the directions of the interacting strands determine the type of beta-sheet (figure 2.1). The side chain groups are projected outwards of the sheet and may form interaction sites for other structures or other proteins. To permit these side chains to project out of the sheet, the backbone is pleated and not planar. The other relatively constant dimension in beta-structures is the distance of the hydrogen bonding between adjacent chains, this is about 4.8 Ǻ.
Figure 2.1 A Schematic diagrams of (a) beta-sheets that are parallel and antiparralel, respectively[1].
Some proteins were found to have these beta-strands perpendicular to the axis of the protein. This was called the cross-beta structure. It was also found to exist in a number of fibrous proteins [30]. Cross-beta structures can form in two ways. The first class (figure 2.2a) has two adjacent strands with a beta-turn between them that form an antiparallel beta-sheet called a beta-hairpin [31]. Two such folded chains form intramolecular H-bonds and their side chains interact with the other antiparallel beta-sheet. In the second class (figure 2.2b), the polypeptide chain also folds back on itself, but the connected beta-strands make contact via their side chains rather than interacting via H-bonds of the backbone. As a result, the two linked strands, find themselves in two different parallel beta-sheets with the other strand.
Figure 2.2 Schematic diagrams of a stack of two beta-strands. Two interacting hairpins (a) and two beta-arches (b), forming a short arcade. Arrows show b-strands; broken lines are H-bonds[2].
During the past few years, a new set of beta-structural fibrous folds has emerged [1], the first of them was a beta-helix. It was found in the bacterial pectate lyase in 1993 [32]. Since then hundreds of similar fibrous structures have been solved. These have been categorized in five different groups (figure 2.3). They include beta-solenoids, triple-stranded beta-solenoids, triangular cross-beta prisms, triple beta-spirals, and spiral beta-hairpin stacks [1]. These structures do not form fibrils in general. Nevertheless, these structures are built of axially stacked repetitive structural blocks. This arrangement, in principle, allows ready elongation to form fibrils by the simple addition of recurrent blocks. The other common property of the fibrous beta-proteins is the repetitive character of their amino acid sequences. The majority of these proteins are located on the surfaces of either bacteria or viruses. A significant portion of these proteins forms homotrimers.
Figure 2.3 Five beta-structural fibrous protein families. Beta-strands in the same strand have the same color [1].
Chapter 3 the beta solenoid Hbp
Hemoglobin binding protease is a coiled beta-helix that has several functional domains hanging on its structure. Here, we are discussing the beta-structure that is part of a protein superfamily, the beta-solenoids. They can be classified based on handedness, twist, oligomerization state, and coil shape, beta-arcs (turns), and stacking of amino acids.
 |
| Figure 3.1 A slice through the beta-helix. The backbone is shown as a green ribbon, and hydrophobic side chains are shown in pink [3] |
Coil shape
One of the most obvious properties is the shape of each coil. This helix is just like the first type beta-solenoid that revealed its molecular models, a ‘‘parallel beta-helix’’, which are three beta-strands per coil. This same shape is maintained throughout the whole solenoid with a length per coil of about 18-20 amino acids [33]. Conform the nomenclature, it has a cross section of a triangular T-type shape. This fairly large group, mainly consists of trimeric bacterial transferases, antifreeze protein from Choristoneura fumiferana, and several bacterial virulence factors. Besides these there are amyloid fibrils of the HET prion that form same shaped coils.
Oligomerisation
The composition of amino acids in each beta-arch determines the way it oligomerises and the amount of interactions it can make. This gives the molecule a shape and rigidity. It was found that beta-solenoids with sequence identity above 18% tend to have similar structures [2]. Here, the plasticity of the beta-helix is shown [34]. Especially a lot of hydrophilic residues can be stored internally without changing the structure.
The composition of beta-solenoid domains shows a higher incidence of glycine, valine, asparagine, threonine, and aspartic acid residues while less leucine, glutamic acid, lysine, arginine, and proline than average are present [24]. The amino acid compositions of single-stranded and triple-stranded beta-solenoids are similar. The differences can be explained: due to the large amount of beta-sheets there is an increased valine and threonine content [35], asparagine ladders increase the amount of Asn residues [3], glycine is frequently found in beta-arcs (turns in beta-solenoids) [2], and proline destabilizes beta-sheets and has therefore a lower incidence [36]. In general, there are more acidic residues (mainly because of asparagine) than of basic residues in beta-solenoids.
Figure 3.2 Histogram of amino acid composition in beta-solenois compared to the average one (filled bars) (Gasteiger et al., 2003). The amino acids are grouped by the following categories, respectively: amino acid residues whose incidence is respectively unchanged, increased, and decreased in beta- solenoids [24].
A survey has revealed some relationships of the amino acid sequences with their structures in beta-solenoids. In particular, the repetitive character of the coil sequences and conformations that recur in tracts of tandem repeats. For Hbp the sequence of the average beta-arch (coil) can be described by sxOxOxx-axOx-xxOxaxx (or visualized by figure 3.3). Lower letters stand for their amino acids (serine and alanine). An x represents any residue, circles indicate a bulky non-polar residue, and “−” is a site of deletion/insertion of any residue. All underlined symbols are within a beta-strand.
Figure 3.3 Schematic representation of an average strand of Hbp [33]. This is a visuallisation of the sxOxOxx-axOx-xxOxaxx sequence. Arrows represent beta-strands, filled circles are hydrophobic residues, a is alanine and s is serine.
Beta-arcs
Each cross section of the coil contains three beta-strands and tree turns. Beta-strands are visual by their alternately pattern of internal and external projecting residues. The turns are an irregularity of this pattern and are only projected to the external environment. These turns interact intermolecular by H-bonding and are therefore called beta-arches. A systemic analysis [2] revealed information about these turns that will be used here to describe Hbp.
 |
| Figure 3.4 Nomenclature used to describe residue positions in b-arches. Open and filled circles denote side-chains directed outside and inside the arch. Thick arrows denote beta-strands. Thin denote arc residues which number can vary over the arc type. [2]. |
Beta-arcs are here described as starting after the last interior projected amino acid (1B3 in figure 3.4) in beta-conformation. The arc ends prior to the first interior amino acid (2B1 in figure 3.4) in beta conformation. Therefore, a beta-strand is a continuous run of residues in beta-conformation, which starts and finishes with residues whose side-chains are buried inside the structure, and a beta-arc as the region between those two strands.
The angles that each residue makes can be sorted in a Ramachandran map (figure 3.5). Here, amino acids can be plotted by their Phi (x-axis) and Psi (y-axis) angles. In the graph, areas have been assigned with a letter. This way arcs can be given a name by the. The main information that this analysis revealed are the conformations most common found (figure 3.6). The arcs that can be found in Hbp will be discussed below with all specific information known about these turns. The average beta-arch (sxOxOxx-axOx-xxOxaxx) described two 2 residue arcs, 2 three residue arcs and one 4 residue arc.
Figure 3.5 A Ramachandran plot has the two angles that an amino acid make on its x and y axis. Here, the plot is divided into areas that give name to the conformations of residues. This way, a beta-arc can get a name by the conformation of its amino acids.
2 residue arcs
Most 2 residue arcs are of the bl conformation (84%) in beta-solenoids [2]. A subtype is an N-bl configuration that has an internal Asn (or Ser, Thr, Cys and Val) stacking at the 2B1 position. Additionally, there is a V-bl subtype of the arcs that have apolar side chains, mostly Val, inside the structure at the same position.
Most 2 residue arcs are of the bl conformation (84%) in beta-solenoids [2]. A subtype is an N-bl configuration that has an internal Asn (or Ser, Thr, Cys and Val) stacking at the 2B1 position. Additionally, there is a V-bl subtype of the arcs that have apolar side chains, mostly Val, inside the structure at the same position.
In Hbp, 2 residue arcs can be found where one has a serine in the 2B1 position. That makes this an N-bl due to its serine residue. The second arc has an alanine in the 2B1 position and is therefore a V-bl configuration.
Other 2 residue conformations like ab, xp, and ag are very rare. That makes these less likely to occur in Hbp, although the difference between them and bl configuration is the absence of hydrophilic side-chains (Asn) in interior positions. An internal Asn residue in the (1wxr) model could not have been found (own findings).
Figure 3.6 Diagrams showing percentages of different types of arc conformations in beta-solenoid proteins. Each pie accounts for a certain amount of residues within an arc. The arcade-forming arcs are in bold and underlined [2].
3 residue arcs
 |
| Figure 3.7 The ppl arch that has a intramolecular H-bond. |
The majority (55%) of 3 residue arcs have the ppl configuration in beta-solenoids. The next most frequent are xbl/xpl arcs (7%); bed/beb arcs (7%); gbp/app/gbb arcs (6%) and bll arcs (4%). The xbl/xpl arches form their own arcades or are inserted into ppl arcades. The bed/beb and gbp/app/gbb arches occur as isolated elements, while bll arches can form arcades. Moreover, pag/pgg arches (3%) frequently have Pro in the second position. In triangular beta-solenoids ppl arches are most common arc where they often form arcades.
Except for the tree reside arcs that can be found in the average arch, there is no information in literature about the Hbp arcs. The information of the article is not sufficient to connect an arc type to the molecule. A quick search with Swiss PDB viewer helped to find some of these arcs. Position 767-769 is an example which has a ppl configuration. This only gives information of the existence of these arcs not their configuration.
4 residue arcs
 |
| Figure 3.8 The gbeb arch about 120°. |
These arcs have a larger variety of potential configurations. Gbpl, bepl and gbeb are most common found arcs. These conformations turn the chain through 120o to 160o. For gbeb, only isolated cases were observed in triangular beta-solenoids. Usually these turns are located at the end of a coil because of the large angle turn they make. These arches do not have H-bonds within the arc but have inter-coil H-bonds. Here, the A3 residue is always Gly.
Between the second and third beta-strand is a possible 4 residue arc. The appendix of the systematic review delivers evidence of a gbeb-arc (figure 3.8) at positions 391-396 in Hbp. The location and the Gly at the A3 fit the data for a gbeb configuration. Only the proportion of these arcs remains unclear.
Internal environment
The inside of these beta-helixes are tightly packed with amino acid side chains, in most cases they are hydrophobic. The extended hydrophobic core is largely but not exclusively filled with aliphatic side chains. The N-terminal region close to the serine protease domain has a higher proportion of buried aromatic residues. The stacking of aromatic rings give face-to-face packing [37]. This makes
it water-excluding caused by the tightly packing of residues. Very few polar residues point into the hydrophobic core, but Ser 780 is one exception [3]. The other proteins in the SPATE (serine protease autotransporters of the enterobacteria) family generally have threonine at this position.
Stacks
Interacting molecules can have local effect on rigidity of a polypeptide structure. In 1993, stacking of similar residues in coils was first discussed [32]. These residues tend to appear at the corresponding positions in neighboring strands.
Interacting molecules can have local effect on rigidity of a polypeptide structure. In 1993, stacking of similar residues in coils was first discussed [32]. These residues tend to appear at the corresponding positions in neighboring strands.
The relationship between stackings and beta-structures is one of the most basic features of their architecture, no beta-helix has been seen without. However, less known is about the interaction of the residues because of the distance. The spacing between coils is about 4.8Å which is slightly longer than the ideal packing of valines or isoleucines [41].
 |
| Figure 3.9 A stack of Asn residues in Hbp. This is called an “Asn ladder.” On the left side, part of ladder Asn729, 746, 765, 785, 815, 836, 878is shown, right Asn669, 717, 735. |
Unlike some other beta-solenoid, Hbp has no regular “stacks” of identical residues packed in layers within the core except for some Asn ladders [3]. The asparagine stack (Asn729, 746, 765, 785, 815, 836, 878) begins and ends with a threonine reside (Thr680, Thr898) which is peculiar. Not mentioned in literature are two aligned Asn ladders (Asn600, 657, 676, 724, 742 and Asn669, 717, 735). Besides these, a stack of internal indole and aromatic residues can be found (Trp873, 891 and Phe716, 734) residues.
Handedness
The hand of the molecule is one of the features where beta-solenoids can be divided on. This is the direction with which the amino acids winds around the solenoid axis. Most of the known beta-solenoids (70%), are right handed [42, 43] and are mostly longer than 18 residues [24]. Hbp belongs to the larger group of right-handed beta-solenoids with about 574 residues (own findings).
Twist
Beta-strands are rarely perfectly extended but exhibit a twist due to the lack of an internal plane of symmetry in amino acids. The preferred angles are –135°, 135° (φ, ψ) which significantly different from the fully extended conformation (φ, ψ) = (–180°, 180°). The twist prevents β strands in a larger sheet from splaying apart.
Their left or right handedness can be found by the relation between reference points at each individual cross-section. The coils are not stacked exactly above one another but with a small offset. These reference points form a helix in a beta-solenoid. The sense of the twist is given by the hand of this helix as it winds around the solenoid axis. In the case of Hbp this forms a left-handed twist (figure 3.3). This is suggested to be the most energetically favorable [24].
Beta-strands are rarely perfectly extended but exhibit a twist due to the lack of an internal plane of symmetry in amino acids. The preferred angles are –135°, 135° (φ, ψ) which significantly different from the fully extended conformation (φ, ψ) = (–180°, 180°). The twist prevents β strands in a larger sheet from splaying apart.
Their left or right handedness can be found by the relation between reference points at each individual cross-section. The coils are not stacked exactly above one another but with a small offset. These reference points form a helix in a beta-solenoid. The sense of the twist is given by the hand of this helix as it winds around the solenoid axis. In the case of Hbp this forms a left-handed twist (figure 3.3). This is suggested to be the most energetically favorable [24].
Beta-structure formation
 |
| Figure 3.10 The c-terminal autochaperone domain. Here the beta-hairpin is most known part of it. |
The formation of a beta-helix requires a domain which is called the intra molecular autochaperone domain [44]. It shows similarities with other intramolecular chaperone sequences and has a folding-associated function. This might increase the translocation efficiency, either by stabilizing the beta-barrel, or by promoting the folding of the passenger domain. The beta-barrel is a structure in the outer membrane that helps proteins translocation. A stable structure increases translocation speed. Promoting the folding process might have a pulling effect on the internal polypeptide strand.
The autochaperone domain is usually located between the beta-barrel and the passenger domain [45]. When the passenger domain is translocated, starting with its C terminus, the autochaperone domain is first out. This would result in the formation of a hairpin structure [46].
Chaper 4. Alzheimer beta-amyloid
As is explained earlier in the introduction, Alzheimer has two major molecules; Ab40 which is most abundant molecule and Ab42 (ratio 1 Ab42 : 10 Ab40 under physiological conditions) which is the most toxic and aggregative of all. Until recently there was only data available of solid state NMR, scanning transmission EM mass measurements, electron paramagnetic resonance spectroscopy of spin-labeled derivatives that could not give fully detailed information about the molecular structure of both aggregated and non aggregated protofibrils.
First the models had individual amyloid molecules that stacked to form a beta-helix. Most thought of an antiparralel beta-helix [47] where thermodynamically most unfavorable domains are shielded. Two Ab molecules interacted with each other to form an internal hydrophobic core and hydrophilic external. Recent biophysical research using solid state NMR [14] showed that fibrils are rich in beta-sheet structure and have the segments lie perpendicular to the axis of the fibril.
Lately, research was done with proline replacement. They showed to have even more complex molecular models than was thought before. Those models will be discussed in detail for both Ab40 and Ab42.
Model for Ab40
The first proper molecular model for Ab40 was based on solid state NMR (ssNMR) experiments [14]. They found that residues 1–10 are disordered, residues 9–21 and 30–36 consistently adopt beta-strand conformations, and residues 23, 24, 25, and 29 may adopt non-beta-strand conformations. Earlier ssNMR experiments [48, 49] and site-directed spin-labeling [50] found all Ab peptides stack parallel and in-register. Therefore, a model was proposed with a parallel-stacked hairpin-like structure [14]. Positions 12–24 and 30–40 are 2 beta-strands, where residues 25–29 form a turn in the chain. The fibril would then be a cross-beta spine with hairpins stacked in a parallel register. The side chains of both strands are into proximity to interact with one another.
Figure 4.1 a) Structural model of Aβ1–40 fibrils by data of solid state NMR experiments. The yellow arrow indicates the direction of thefibril axis. The cross-β structure is double-layered, with in-register parallel β-sheets formed by positions 12–24 (orange) and 30–40 (blue). (b) structured part of the Aβ1–40 molecule Residues are colored according to their side chains as hydrophobic (green), polar (magenta), positive (blue), or negative (red).[14]
 |
| Figure 4.2 The Ab40 model by proline replacement studies |
A different fibril model for Ab1–40 was made based on scanning proline mutagenesis [51] and molecular modeling [52]. Here, the model was made by template proteins and experimental data of proline substitution. They found some residues relatively insensitive to proline replacement (Proline is rarely found in beta-sheets). Residues 1–14 and 37–40 are probably unstructured, flexible elements projected outwards from the core that consists out of beta-strands. Residues 22, 23, 29 and 30 occupy turn positions between these three beta-structured elements. [51].
Probably a left-handed parallel model is favored, not only hydrogen bondings, but also stabilizing hydrophobic interactions, are then maximized. Residues 17 and 34 are placed in close proximity, because double cysteine mutants at these positions form disulfide bonds after fibrillization [53].
Furthermore, the distance between Tyr10 and Met35 in was measured using pulsed ESR and was found to be c. 30 Å [19]. This is not exlusively in favor of the new model but is at least no counterargument (figure 4.3).
Figure 4.3 (a)The structural model of Ab40 by using systematic proline replacement and by (b) using solid-state nuclear magnetic resonance spectroscopy for fibril. In both, the distance between Tyr10 and Met35 is suspected to be far. In (b), the intramolecular salt bridge between Asp23 and Lys28 is formed [19].
There is a good correlation between the aggregation and neurotoxicity of Ab (r = -0.815). The mutations at positions 22 and 23 that enhanced the aggregative potency and neurotoxicity suggest that these structures are more often in a different structure than wild-type Ab. These Ab mutations are involved in the pathogenesis of CAA.
Ab40 fibrillises easily when a salt bridge is formed between D23 and K28 [14]. This might include that these residues are in close proximity. This might reject the proline replacement model but might also be an artifact.
 |
| Figure 4.4 The 3D structure of a 35MoxAb42 fibril. These are ribbon diagrams of the core structure of residues 17–42 illustrating the intermolecular nature of the inter-beta-strand interactions. The beta-sheets are shown by cyan arrows. The hydrophobic, polar, negatively charged, and positively charged amino acid side chains are shown in yellow, green, red, and blue, respectively. Positively and negatively charged surfaces are shown in blue and red, respectively, and all others(van der Waals) are shown in white. The direction of the fibril axis is indicated by an arrow pointing from even to odd [4]. |
NMR studies found beta-amyloid to have two beta-strands (18-26 and 31-42) that form intermolecular sheets. Their mainly hydrophobic side chains would interact to form the close hairpin loop. This model was based on a 4 amino acid turn where lysine is directed inside and made the turn be m-shaped. [4, 14]
Later proline replacement studies found some new turn positions [17]. Prolines are never found in beta-sheets although they can easily accommodate in a beta-turn as a Pro-X corner. Their replacement can disrupt beta-sheets that can be examined by their aggregative velocity and toxicity.
Later proline replacement studies found some new turn positions [17]. Prolines are never found in beta-sheets although they can easily accommodate in a beta-turn as a Pro-X corner. Their replacement can disrupt beta-sheets that can be examined by their aggregative velocity and toxicity.
Four beta-sheets were identified in Ab42. Only the E22P-Ab42 mutant was found to display stronger neurotoxicity than wild-type Ab42. This suggests that the residues at positions 15–21 and 24–32 (mutation positions 15-32) are involved in the beta-sheet. Furthermore, the turn at positions 22 and 23 plays an important role in the aggregation and neurotoxicity of Ab42. The C-terminal proline mutants (A42P-, I41P-, and V40P-Ab42) hardly aggregated with extremely weak cytotoxicity. C-terminal threonine mutants (A42T- and I41T-Ab42) aggregated potently with significant cytotoxicity. This means that the hydrophobicity of the C-terminal two residues of Ab42 is not related to its aggregative ability and neurotoxicity, rather the C-terminal three residues adopt the beta-sheet. Also the 39 and 40 positions were sensitive to replacement for fibril formation. It destabilizes the N-terminal beta-sheets (35-37 and 40-42) and looses the toxicity.
Three turn positions became clear. The 22nd position mutation was found faster aggregating and more toxic than wild-type Ab. This explains the high ability of aggregation in cerebral amyloid angiopathy (CAA) where there is a mutation found in this position. Residue 33, 34 and 38, 39 were found to be turn positions that can enhance aggregation in case of proline mutation. A triple proline mutant was made on positions 22, 24 and 38 that confirmed the findings of three turns.
Just like NMR [4, 14] studies, proline substitution studies showed the N-terminus not adopting a secondary structure. The neurotoxicity and aggregation speed were quite similar to wild-type Ab42.
Figure 4.7 (a) The physiological and toxic conformations of Ab42 identified by previous studies using systematic proline replacement studies . The first turn position is different in both stages and the distance between Tyr10 and Met35 becomes close only in the toxic conformer. (b) The mechanism of neurotoxicity of Ab42 and the structure for fibrillization. The toxic conformer in (a) where the 22, 23 positions turn bring the Tyr10 close to Met35. Here, the sulfur atom of Met35 gets oxidized in collaboration with metal ions (probably CuII). The resultant S-oxidized radial cation can be stabilized by the C-terminal carboxylate anion that forms the C-terminal. This hydrophobic core accelerates oligomerization, resulting in long-term oxidative stress [19].
Major and minor conformer
Wild-type Ab42 has two chemical shifts that indicate existence of two conformations. We describe these as a major and a minor conformer. This was found by analyzing the wild-type Ab42 by physiological experiments [19]. The properties of the molecule are different in both configurations.
Wild-type Ab42 has two chemical shifts that indicate existence of two conformations. We describe these as a major and a minor conformer. This was found by analyzing the wild-type Ab42 by physiological experiments [19]. The properties of the molecule are different in both configurations.
In the minor form, the 22 and 23 position make a turn structure in a way that Tyr10 and Met35 are close [54]. Then the sulfur atom can become oxidized which includes the toxic properties. The mutant E22K-Ab42 was found to have a chemical shift that indicated that the Lys22 and Asp23 could be close to each other and located at the same side [55]. This supports the theory about a 22, 23 turn structure where there is a repulsion of the two residues in wild-type Ab42. Therefore, there is a decreased ratio of the minor conformer. A mutation E22K (Italian mutation) is proof of this because it showed a stronger aggregative and neurotoxic ability; this might have been increased due to the minor conformer by the ionic interaction in the turn formation.
Figure 4.8 A structural model of the first turn in Ab42 found by solid-state NMR and proline replacement studies. (a) The physiological conformer with turn position at residue 25, 26. (b) The first turn in the toxic conformer at position 22, 23. The electrostatic interactions between Glu22 and Asp23 would suppress the turn formation at this position and therefore have a decreased ratio of this conformer [56].
The major conformation has a turn in the position 25 and 26. This was found by thermodynamic analysis based on systematic proline replacement, and its presence was verified by solid-state NMR using DARR [57]. A salt bridge between Asp23 and Lys28 has been found in Ab40 [14, 58]. This does not have to contradict the earlier findings. This might be an interaction in the major conformer that is not present in the minor conformer.
Fibrillisation
Before amyloid assembles into fibrils, they stay in solution for an extended period [59]. This is called the lag period which is the major energy barrier from unstructured monomers to well-ordered fibrils. After this period the fibrils form rapidly. For short peptides ( 6 residues), it has been suggested that the transition-state structures, which determine the kinetic barrier to fibril formation, are either 2 molecules in a single sheet or this pair interacting with 1 molecule from the apposing sheet, leading to a nucleus consisting of 3–4 molecules, 2 in one sheet interdigitated with 1–2 in the apposing sheet [60]. A different model has been suggested by co-crystallisation of Ab40 with an affibody protein ZAb3 (a binding protein) experiments. Here, a structural mechanism for amyloid formation based on soluble oligomeric hairpin intermediates is proposed. Intramolecular beta-sheets are then preferred to
intermolecular ones because the fibril core structure is not fully established. Oligomers of beta-hairpins might form by hydrophobic stackings that remain soluble by the hydrogen bonding of exposed peptide backbones [61]. The formation of the fibril is a joint conformational change of all beta-hairpins.
Figure 4.9 Hypothetical aggregation mechanism of Ab in a hairpin structure. Beta-hairpins form with interior hydrophobic core side chains (orange) and exterior side chains (yellow). Soluble oligomers form by hydrophobic stacking of beta-hairpins. A joint conformational change forms the beta-structure.
Besides aggregation by formation of beta sheets, formation can occur through interarc H-bonds. Tight standard beta-arcs [62] can form interarc H-bonds of the same strength as those in beta strands. This makes it reasonable that nuclei for fibrillation can be stacks of beta arcs. Some mutations in turn regions have been found to affect amyloid fibril formation [63] although there can be different reasons for such a mutation as is explained in chapter 3[17, 19]. There, it has been said that the angle of a turn depends on its amino acids. They decide the conformation of the secondary structure. In some structures there is an interaction between residues that enhance aggregation, not the interaction that turns make with other turns.
Two beta-strands can mate tightly and interdigitate to form a water-tight interface. Here, the residue side chains intermesh with close complementarity, in what we call a steric zipper [64]. The structures found have been classified into 5 groups. They are distinguished by whether their sheets are parallel or antiparallel, whether sheets pack ‘face-to-face’ or ‘face-to-back’ to one another, and whether the sheets are oriented parallel or antiparallel. Combinations of these three structural arrangements give eight theoretically possible classes of steric zippers. All have 4-12 residues that can form fibrils by its self. Because every fibril-forming protein contains a steric zipper suggests that it may drive fibril formation.
 |
| Figure 4.10 MVGGVV steric zipper. Two beta-sheets meet and mate tightly with a dry interface. |
It seems likely that fibrillisation starts by the unmasking of zipper-domains in several identical protein molecules, permitting them to interdigitate. Recruitment of monomers into pre-formed fibrils is expected to be more rapid than nucleation. The recruitment for fibrillization requires one molecule to unmask its fibril-forming sequence, the other interacting unit is already showing its domain. For nucleus formation several molecules have to unmask their zipper-forming segments simultaneously, which is less likely to occur.
Ab has sequences GGVVIA (37-42) and MVGGVV (35-40) that form steric zippers. The first, GGVVIA is a class 4, parallel face-to-back, sequence. The second, MVGGVV is a class 8, antiparalel face=back, sequence. These antiparallel beta-sheets do not seem to fit the data of earlier mentioned NMR and proline replacement studies although the direction of the fibril is in agreement.
Interacting residues
Stabilization of the arches may come partially from Asn residues that are stacked axial. Usually, they are aligned in a row and form H-bonds by their side chains. Therefore they are called “Asn ladders” [62]. Small proteins have been used to determine the effect of stacking Asn residues [65]. They found the interactions of Asn–Asn stacking include both backbone and side-chain hydrogen bonds. The interacting Asn residues would therefore stick the two beta-strands together and assist the “on-pathway” oligomerization of the arches.
Glutamine residues form H-bonds with the backbone of the apposing beta sheet. The interior Gln side chains knit the double beta sheets together via networks of H bonds, both axial (within the sheet) and lateral (intersheet). In terms of its spatial network of H bonds, this structure resembles polyglycine, which also forms stable insoluble aggregates [66].
Models of Ab fibrils also suggest that oppositely charged side chains can stabilize the double beta-sheet by intersheet ionic bonds [4, 67]. In this scenario, “knobs-into-holes” packing of the sheets, supplemented by H bonds and ionic bonds formed by interior polar
side chains, imparts the high stability characteristic of amyloid fibrils.
Although there are no hydrogen bonds bridging two tightly complementing sheets across the dry interface, each GNNQQNY molecule forms 11 hydrogen bonds to its two neighbouring molecules in the same sheet (Fig. 2e).Five of these are backbone C=O . H-N hydrogen bonds, and four are ‘amide stacks’; that is, amide–amide hydrogen bonds between pairs of identical Asn or Gln residues in adjacent molecules within a sheet. It is these hydrogen-bonded amide stacks that force the GNNQQNY and NNQQNY molecules to stack parallel and in register in their respective sheets [60].
Chapter 5. Common properties
An important additional insight has been the realization that there is a class of monomeric and oligomeric proteins, called beta-solenoids [68], whose structural properties closely resemble those of amyloids in that they also have parallel, in-register, cross-beta structures. The beta strands stacked in register in beta-solenoids tend to have similar rather than identical amino acid sequences, as they are in most amyloid fibrils. Although E. coli’s Hbp and Ab of Alzheimers disease do not have their phenotype as a disease in common, they have their beta-fibrous molecular structure that overlap.
If we compare Ab fibrils with all different beta-fibrous proteins described in chapter 2 (figure 2.3) it fits best with beta-solenoids. Both Ab and beta-solenoids shows an axially stacking of similar blocks [1]. Another characteristic that they share is their parallel and in-register nature [69].
To form a beta structure like there need to be a few stabilizing factors in order to connect the stacks of beta-strands. In both structures a few properties can be found that might help stabilize the structure. In this chapter we try to compare those. They are: cross sectional shape, handedness and twist, Arches (turns), interacting residues, steric zippers, stacking of side chains and formation of the beta-fibril.
Cross sectional shape
A slice through the beta-helix of Hbp shows a triangular beta-arch. Each turn creates three beta-strands that form beta-sheets with neighboring turns. As is explained in chapter 3, this is called a T-shaped beta-solenoid.
Figure 5.1 The cross sectional shapes of respectively, Hbp, Ab42 and Ab40.
For Ab this is more complicated to determine. The model is not in such a detail that all angles of amino acids are known. Especially for the first two turns of Ab42 this would help. The models proposed for both fibrillized monomers suspect a similar shape as Hbp. They both have the triangular T-shape. The physiological conformer of Ab42 has no stable interaction between position 35 and 42 and has therefore no fixed fourth strand. It now looks like the L-type beta-solenoids. Other proteins with this cross section are bacterial, plant, and fungal enzymes for the degradation and modification of polysaccharides, phage P22 tailspike endorhamnosidase and the P.69 pertactin virulence factor from B. pertussis.
Handedness and Twist
The angular offset that every coil has compared to the next stacked coil, is called the twist. As with many known beta-helices, the model derived for Hbp has a left-handed twist [3]. For Ab fibrils, there is large probability of twist within the structure.
Several articles propose left-handed twist for Ab40. Petkova et al. (2002) proposed a left-handed twist just like Hbp in the Ab40 model. Another article wrote about different fibrils that can form under the same conditions [70]. There, no directional twist has been determined although a left-handed angle is most preferred. In 2008 a model with a threefold symmetry has been proposed [71]. Here, three hairpin structures form a triangular shape with a left-handed twist. Finally, a model was proposed by help of proline replacement studies on Ab40. Again, a left-handed parallel model is proposed, not only hydrogen bondings, but also stabilizing hydrophobic interactions are then maximized [51].
Also, Monomers in an Ab 42 fibril are suspected to have an angular twist. For the fibril model a dipole study revealed a 6Å distance between A21 residues in different monomers[56]. That is more than the exact beta-strand stacking of 4.8Å. Therefore, there must be a small angular twist. Besides that, several conformers have been found for the Ab42 model. Both might be in the same fibril, the positional relationship between the two conformers remain unclear. A mosaic pattern cannot exist with a 6Å distance between A21 residues but nuclei of both conformers could interact within a fibril. This then would give a small offset between the blocks of different nuclei.
The direction with which the amino acids winds around the solenoid axis is called the handedness. Hbp has a right-handed model. Ab fibrils on the other hand, do not have handedness. There is no coiling of the structure in Ab fibrils, just stacking of the individual monomers. Therefore, no handedness direction can be given.
Arches
In beta structures there are stacked strand-loop-strand motifs. The turns that connect the beta-strands interact by H-bonding. Therefore, these are termed beta-arcs. A systemic analysis on crystal structures of beta-solenoids revealed that beta-arcs assume a limited number of preferred conformations. Although not that much is known about the arcs in Ab knowledge of the systemic analysis of arcs in beta-solenoids might be used. The arcs of Ab cannot be defined by the nomenclature because the proline replacement studies did not give us the angles of the amino acids. On the other hand, the analysis revealed that there are a limited number of preferred conformations that can be specified by more than just the angles [2].
The proline replacement studies on Ab40 and Ab42 showed to have only two residue turns. There, an 84% certainty could be given on a bl-arc. In this conformation usually an Asn at the 2B1 position forms an internal stack. That can also be a Ser, Thr or Cys or Val residue. Ab40 has this Val at the 2B1 of the first arc. Ab42 has a Val at both the first and third turn 2B1 position. That is a v-bl arc because it has an apolar internal residue. The second turn in Ab42 and Ab40 has not such residues at this position and is therefore less certain.
Amino acid composition
Both structures rely on their amino acid compositions to make beta-strands, turns (or arcs) interaction sites and gain stability. Earlier has been explained what the compositions of the beta-solenoids are. They shows to be enriched in glycine, valine, asparagine, threonine, and aspartic acid while the incidence of leucine, glutamic acid, lysine, arginine, and proline is lower than average [72]. Overall, there is a higher incidence of acidic residues (mainly asparagine) than of basic residues in beta-solenoids. Moreover, it is rare to observe charged residues of both signs in these proteins.
The Ab sequences (figure 5.4) contain mostly Glycine and Valine. Even when we look at the central structures which are turns and beta-strands, we find these residues most frequently. This is in line with the enriched residues in Hbp or beta-solenoids.
Ab40 DAEFRHDSGYEVHHQKLVFFAEDVGSNKGAIIGLMVGGVV
Ab42 DAEFRHDSGYEVHHQKLVFFAEDVGSNKGAIIGLMVGGVVIA
Ab42 DAEFRHDSGYEVHHQKLVFFAEDVGSNKGAIIGLMVGGVVIA
Figure 5.4 Both AB40 and Ab42amino acid sequences are given. Here, the bold letters represent the residues in beta-conformation.
Stacks
“Aligned” and “stacked” residues are ubiquitous in beta-sheet containing proteins. These are same type residues that form ladders over different coils. These proteins may contain three different types of stacks: Aliphatic stacks, aromatic stacks and polar stacks. Aligning of Asparagine, tryptohan and phenylalanine residues have been found in Hbp.
“Aligned” and “stacked” residues are ubiquitous in beta-sheet containing proteins. These are same type residues that form ladders over different coils. These proteins may contain three different types of stacks: Aliphatic stacks, aromatic stacks and polar stacks. Aligning of Asparagine, tryptohan and phenylalanine residues have been found in Hbp.
Although it is not known for sure, the formation of these ladders in amyloid fibrils is likely to occur. Their contribution to the stability of such a structure must be essential because no parallel beta-helix has been observed without. Because both Ab40 and Ab42 form fibrils by stacking the exact sequences on top of one another, stacks should occur. Especially in rigid structures like beta-sheets and beta-arcs.
Aliphatic stacks can be formed by leucine, isoleucine and valine. All these can be found in the Ab40 and Ab42 structures. Especially “Cupped stacks” of valines and isoleucines [73] can repeatedly be found. Of all aromatic residues (phenylanaline, tryptophan, tyroksine, proline and histidine) only phenylanaline is found in the structured part of Ab.
Asparagine is found in a beta-strand position of Ab where it might assist the oligomerization. This Asn–Asn stacking is similar to the Asn ladder commonly found in beta-helix proteins [65] like Hbp.
Stacks can be a mix of valines or isoleucines with leucines or methionines. The average spacing between coils is about 4.8 Å which means that the distances between the equivalent atoms are slightly longer than would give ideal packing of valines or isoleucines [41]. When the fibril structure has a twist, a neighboring amino acid in a successive strand might be the closest to reach. In the Ab structure this might be I31, I32 and V39, V40, V41 and L17 V18.
Steric zipper
Some strands mate like “knobs-into-holes” at a completely dry interface. These residue side chains between two tightly interdigitated beta-sheets are called a steric zipper. These interactions are found in lots of fibrillising proteins [64] like Ab.
The sequence GGVVIA (37-42) and MVGGVV (35-40) were found to form steric zippers in Amyloid-beta. They could only be found in Ab42 because Ab40 does not have beta-strands at these positions. All other sequences for steric zippers were found in different fibrous proteins. These involve Sup35, RNAse, Tau and insulin.
None of the sequences that form steric zippers could be identified in Hbp. The protein sequences of Hbp found at the NCBI protein database (Escherichia coli 83972, Escherichia coli 83972) were analyzed for the steric zipper domains. This implies that Hbp might not form steric zippers.
Parallel | Class 1 | VQIVYK (Tau) | NNQQNY (Sup35) |
GNNQQNY (Form 1 Sup35) | GNNQQNY (Form 2 Sup35) | ||
Class 2 | SSTSAA (RNase) | NNQQ (Form 2 Sup35) | |
Class 4 | SNQNNF Prion protein | GGVVIA (Amyloid-β) | |
Antiparallel | Class 7 | LYQLEN (Insulin) | VEALYL (Insulin) |
Class 8 | MVGGVV (Form 1 Amyloid-β) | MVGGVV (Form 2 Amyloid-β) |
Figure 5.5 All known stericzipper sequences in beta-structures. Here, shown by class and type of beta-sheet they are in. [64]
Formation the beta-structure
The induction of fibrillisation or assembly for both Ab and Hbp remains unclear. Although both have different theories there is still a lot of doubt about them. Hbp has a possible autochaperone at the C-terminal. Ab has several theories of nucleation and changing structures that help form fibrils. These theories do not show any overlap.
Energetically, Ab fibrils are likely to be harder to form than a Hbp molecule. Beta arches with fewer residues in their beta arcs pose lower entropic barriers [62] and therefore form faster fibril structures. Ab fibrils has more amino acids in each arch than Hbp (Ab has 22 or 27, Hbp has about 18) which might form easier.
Abbreviations
Hbp: Hemoglobin binding protease
Beta-strand: polypeptide chain in almost fully extended conformation
Beta-sheet: stacked beta-strands interacting by their backbones
Beta-arcades: stacked columns of identical beta arches
Beta-arc: a turn between beta strands within a beta-fibrous protein
Protofilaments: Fibrils with minimal mass-per-length
Intramolecular: within the molecule
Intermolecular: between two or more molecules
CAA: cerebral amyloid angiopathy
NMR: Nuclear Magnetic Resonance
Pulsed ESR: pulsed electron spin resonance
References
1. Kajava, A.V., J.M. Squire, and D.A.D. Parry, [beta]-Structures in Fibrous Proteins. 2006, Elsevier. p. 1-15.
2. Hennetin, J., et al., Standard Conformations of [beta]-Arches in [beta]-Solenoid Proteins. 2006, Elsevier. p. 1094-1105.
3. Otto, B.R., et al., Crystal structure of hemoglobin protease, a heme binding autotransporter protein from pathogenic Escherichia coli. 2005, ASBMB. p. 17339.
4. Lührs, T., et al., 3D structure of Alzheimer's amyloid- (1–42) fibrils. 2005, National Acad Sciences. p. 17342.
5. Astbury, W.T. and H.J. Woods, X-ray studies of the structure of hair, wool, and related fibres. II. The molecular structure and elastic properties of hair keratin. 1934, JSTOR. p. 333-394.
6. M.B. Pepys (M. M. Frank, K.F.A., H. N. Claman & E. R. Unanue), Amyloidosis, in Samter's Immunologic Diseases. 1994. p. 637-655.
7. Scherzinger, E., et al., Huntingtin-encoded polyglutamine expansions form amyloid-like protein aggregates in vitro and in vivo. Cell, 1997. 90(3): p. 549-58.
8. Nelson, R. and D. Eisenberg, Structural models of amyloid-like fibrils. Adv Protein Chem, 2006. 73: p. 235-82.
9. Mori, H., et al., Mass spectrometry of purified amyloid beta protein in Alzheimer's disease. 1992, ASBMB. p. 17082.
10. Näslund, J., et al., Relative abundance of Alzheimer A beta amyloid peptide variants in Alzheimer disease and normal aging. 1994, National Acad Sciences. p. 8378.
11. Missmahl, H.P. and M. Hartwig, Polarisationsoptische Untersuchungen an der Amyloidsubstanz. 1953, Springer. p. 489-508.
12. Cohen, A.S. and E. Calkins, Electron microscopic observations on a fibrous component in amyloid of diverse origins. 1959, Nature Publishing Group.
13. Eanes, E.D. and G.G. Glenner, X-ray diffraction studies on amyloid filaments. 1968, Histochemical Soc. p. 673.
14. At, P., et al., A structural model for Alzheimer's beta-amyloid fibrils based on experimental constraints from solid state NMR. 2002. p. 26.
15. Popova, L.A., et al., Structural Variations in the Cross- Core of Amyloid Fibrils Revealed by Deep UV Resonance Raman Spectroscopy, ACS Publications. p. 6324-6328.
16. Williams, A.D., S. Shivaprasad, and R. Wetzel, Alanine Scanning Mutagenesis of A [beta](1-40) Amyloid Fibril Stability. 2006, Elsevier. p. 1283-1294.
17. Morimoto, A., et al., Analysis of the Secondary Structure of {beta}-Amyloid (A {beta} 42) Fibrils by Systematic Proline Replacement. 2004, ASBMB. p. 52781.
18. Morimoto, A., et al., Aggregation and neurotoxicity of mutant amyloid beta (Abeta) peptides with proline replacement: importance of turn formation at positions 22 and 23. 2002, Academic Press. p. 306-311.
19. Murakami, K., et al., The turn formation at positions 22 and 23 in the 42-mer amyloid peptide: The emerging role in the pathogenesis of Alzheimer's disease, John Wiley & Sons. p. S169-S179.
20. Inouye, H., P.E. Fraser, and D.A. Kirschner, Structure of beta-crystallite assemblies formed by Alzheimer beta-amyloid protein analogues: analysis by x-ray diffraction. 1993, Elsevier. p. 502-519.
21. Kajava, A.V., J.M. Squire, and D.A. Parry, Beta-structures in fibrous proteins. Adv Protein Chem, 2006. 73: p. 1-15.
22. Farthmann, E.H. and U. Schöffel, Epidemiology and pathophysiology of intraabdominal infections (IAI). 1998, Springer. p. 329-334.
23. Aldridge, K.E., The occurrence, virulence, and antimicrobial resistance of anaerobes in polymicrobial infections. 1995. p. 2-7.
24. Kajava, A.V. and A.C. Steven, [beta]-Rolls,[beta]-Helices, and Other [beta]-Solenoid Proteins. 2006, Elsevier. p. 55-96.
25. Blake, C.C., et al., A molecular model of the amyloid fibril. 1996. p. 6.
26. Caughey, B.W., et al., Secondary structure analysis of the scrapie-associated protein PrP 27-30 in water by infrared spectroscopy. Biochemistry, 1991. 30(31): p. 7672-80.
27. Kirschner, M. and T. Mitchison, Beyond self-assembly: from microtubules to morphogenesis. 1986. p. 329-342.
28. Fraser, I.E., Proteins of keratin and their synthesis. I. Proteins of prekeratin and keratin. Aust J Biol Sci, 1969. 22(1): p. 213-29.
29. Pauling, L. and R.B. Corey, The structure of hair, muscle, and related proteins. 1951, National Academy of Sciences. p. 261.
30. Fraser, R.D.B. and T.P. MacRae, Conformation in fibrous proteins and related synthetic polypeptides. 1973: Academic Press.
31. Hennetin, J., et al., Standard conformations of beta-arches in beta-solenoid proteins. J Mol Biol, 2006. 358(4): p. 1094-105.
32. Yoder, M.D., S.E. Lietzke, and F. Jurnak, Unusual structural features in the parallel [beta]-helix in pectate lyases. 1993, Elsevier. p. 241-251.
33. Kajava, A.V. and A.C. Steven, The turn of the screw: Variations of the abundant [beta]-solenoid motif in passenger domains of Type V secretory proteins. 2006, Elsevier. p. 306-315.
34. Choi, J.H., et al., Site-Directed Mutagenesis Demonstrates the Plasticity of the [beta] Helix: Implications for the Structure of the Misfolded Prion Protein. 2009, Elsevier. p. 1014-1023.
35. Chou, P.Y. and G.D. Fasman, Conformational parameters for amino acids in helical, beta-sheet, and random coil regions calculated from proteins. Biochemistry, 1974. 13(2): p. 211-22.
36. Schulman, B.A. and P.S. Kim, Proline scanning mutagenesis of a molten globule reveals non-cooperative formation of a protein's overall topology. 1996, Nature Publishing Group. p. 682-687.
37. Yoder, M.D. and F. Jurnak, Protein motifs. 3. The parallel beta helix and other coiled folds. 1995, FASEB. p. 335.
38. Raetz, C.R.H. and S.L. Roderick, A left-handed parallel beta helix in the structure of UDP-N-acetylglucosamine acyltransferase. 1995, AAAS. p. 997.
39. Petersen, T.N., S. Kauppinen, and S. Larsen, The crystal structure of rhamnogalacturonase A from Aspergillus aculeatus: a right-handed parallel [beta] helix. 1997, Elsevier. p. 533-544.
40. HunterJuswinder, C.A. and J.M. Thornton, [pi]-[pi] interactions: the geometry and energetics of phenylalanine-phenylalanine interactions in proteins. 1991, Elsevier. p. 837-846.
41. Jenkins, J. and R. Pickersgill, The architecture of parallel [beta]-helices and related folds. 2001, Elsevier. p. 111-175.
42. Groves, M.R. and D. Barford, Topological characteristics of helical repeat protein. 1999, Elsevier. p. 383-389.
43. Kobe, B. and A.V. Kajava, When protein folding is simplified to protein coiling: the continuum of solenoid protein structures. 2000, Elsevier. p. 509-515.
44. Oliver, D.C., G. Huang, and R.C. Fernandez, Identification of secretion determinants of the Bordetella pertussis BrkA autotransporter. 2003, Am Soc Microbiol. p. 489.
45. Girard, V. and M. Mourez, Adhesion mediated by autotransporters of Gram-negative bacteria: structural and functional features. 2006, Elsevier. p. 407-416.
46. Nancy, R. and M. Michael, Surface display of proteins by Gram-negative bacterial autotransporters.
47. Chaney, M.O., et al., Molecular modeling of the Abeta1-42 peptide from Alzheimer's disease. 1998, Oxford Univ Press. p. 761.
48. Antzutkin, O.N., et al., Multiple quantum solid-state NMR indicates a parallel, not antiparallel, organization of -sheets in Alzheimer's -amyloid fibrils. 2000, National Acad Sciences. p. 13045.
49. Antzutkin, O.N., et al., Supramolecular Structural Constraints on Alzheimer's [beta]-Amyloid Fibrils from Electron Microscopy and Solid-State Nuclear Magnetic Resonance†. 2002, ACS Publications. p. 15436-15450.
50. Torok, M., et al., Structural and dynamic features of Alzheimer's Abeta peptide in amyloid fibrils studied by site-directed spin labeling. 2002, ASBMB. p. 40810.
51. Williams, A.D., et al., Mapping A [beta] amyloid fibril secondary structure using scanning proline mutagenesis. 2004, Elsevier. p. 833-842.
52. Guo, J., R. Wetzel, and Y. Xu, Molecular modeling of the core of A amyloid fibrils. 2004, John Wiley & Sons. p. 357-364.
53. Shivaprasad, S. and R. Wetzel, An Intersheet Packing Interaction in A [beta] Fibrils Mapped by Disulfide Cross-Linking†. 2004, ACS Publications. p. 15310-15317.
54. Takegoshi, K., S. Nakamura, and T. Terao, 13C-1H dipolar-driven 13C-13C recoupling without 13C rf irradiation in nuclear magnetic resonance of rotating solids. 2003. p. 2325-2341.
55. Masuda, Y., et al., Verification of the turn at positions 22 and 23 of the [beta]-amyloid fibrils with Italian mutation using solid-state NMR. 2005, Elsevier. p. 6803-6809.
56. Masuda, Y., et al., Verification of the intermolecular parallel -sheet in E22K-A 42 aggregates by solid-state NMR using rotational resonance: implications for the supramolecular arrangement of the toxic conformer of A 42. 2008, J-STAGE. p. 807040969.
57. Masuda, Y., et al., Identification of physiological and toxic conformations in A 42 aggregates. 2009, John Wiley & Sons. p. 287-295.
58. Hetényi, A., et al., Ligand-Induced Flocculation of Neurotoxic Fibrillar A (1-42) by Noncovalent Crosslinking. 2008, John Wiley & Sons. p. 748-757.
59. Chiti, F., et al., Rationalization of the effects of mutations on peptide andprotein aggregation rates. 2003, Nature Publishing Group. p. 805-808.
60. Nelson, R., et al., Structure of the cross- spine of amyloid-like fibrils. 2005, Nature Publishing Group. p. 773-778.
61. Hoyer, W., et al., Stabilization of a -hairpin in monomeric Alzheimer's amyloid- peptide inhibits amyloid formation. 2008, National Acad Sciences. p. 5099.
62. Kajava, A.V., U. Baxa, and A.C. Steven, {beta} arcades: recurring motifs in naturally occurring and disease-related amyloid fibrils, FASEB. p. 1311.
63. Frieden, C., Protein aggregation processes: in search of the mechanism. 2007, John Wiley & Sons. p. 2334-2344.
64. Sawaya, M.R., et al., Atomic structures of amyloid cross-&bgr; spines reveal varied steric zippers. 2007, Nature Publishing Group. p. 453-457.
65. Tsai, H.H.G., et al., Energy landscape of amyloidogenic peptide oligomerization by parallel-tempering molecular dynamics simulation: significant role of Asn ladder. 2005, National Acad Sciences. p. 8174.
66. Kajava, A.V., Dimorphism of polyglycine I: structural models for crystal modifications. 1999, International Union of Crystallography. p. 436-442.
67. Petkova, A.T., W.M. Yau, and R. Tycko, Experimental Constraints on Quaternary Structure in Alzheimer's -Amyloid Fibrils†. 2006, ACS Publications. p. 498-512.
68. Bateman, A., A.G. Murzin, and S.A. Teichmann, Structure and distribution of pentapeptide repeats in bacteria. 1998, John Wiley & Sons. p. 1477-1480.
69. Margittai, M. and R. Langen, Fibrils with parallel in-register structure constitute a major class of amyloid fibrils: molecular insights from electron paramagnetic resonance spectroscopy. 2008, Cambridge Univ Press. p. 265-297.
70. Meinhardt, J., et al., A [beta](1-40) Fibril Polymorphism Implies Diverse Interaction Patterns in Amyloid Fibrils. 2009, Elsevier. p. 869-877.
71. Paravastu, A.K., et al., Molecular structural basis for polymorphism in Alzheimer's -amyloid fibrils. 2008, National Acad Sciences. p. 18349.
72. Kajava, A.V. and A.C. Steven, Beta-rolls, beta-helices, and other beta-solenoid proteins. Adv Protein Chem, 2006. 73: p. 55-96.
73. Richardson, J.S., E.D. Getzoff, and D.C. Richardson, The beta bulge: a common small unit of nonrepetitive protein structure. 1978, National Acad Sciences. p. 2574.
Potent and selective amylin receptor antagonist that displays 38-fold and 400-fold selectivity over calcitonin and CGRP receptors respectively. Through attenuating the activation of initiator and effector caspases in vitro, AC 187
BeantwoordenVerwijderen