The determination of factors that influence protein conformational changes is very important for the identification of potentially amyloidogenic and disordered regions in polypeptide chains. In our work we introduce a new parameter, mean packing density, to detect both amyloidogenic and disordered regions in a protein sequence. It has been shown that regions with strong expected packing density are responsible for amyloid formation. Our predictions are consistent with known disease-related amyloidogenic regions for eight of 12 amyloid-forming proteins and peptides in which the positions of amyloidogenic regions have been revealed experimentally. Our findings support the concept that the mechanism of amyloid fibril formation is similar for different peptides and proteins. Moreover, we have demonstrated that regions with weak expected packing density are responsible for the appearance of disordered regions. Our method has been tested on datasets of globular proteins and long disordered protein segments, and it shows improved performance over other widely used methods. Thus, we demonstrate that the expected packing density is a useful value with which one can predict both intrinsically disordered and amyloidogenic regions of a protein based on sequence alone. Our results are important for understanding the structural characteristics of protein folding and misfolding.
Protein folding is one of the most challenging issues in biophysical science. During the past few years it has been shown that some diseases are connected with protein misfolding and the formation of insoluble aggregates called amyloid plaques. These processes may be associated with several diseases such as Alzheimer disease, Parkinson disease, Creutzfeldt-Jacob disease, and even certain forms of cancer. It has been shown that proteins with intrinsically disordered regions are involved in protein–protein or protein–nucleic acid interactions. The main objective of this paper is to report insights into the molecular mechanisms of amyloid aggregation. This has been done using the parameter of the observed number of contacts for each amino acid residue in globular state, further called expected packing density. By analysis of sequences alone, the authors have demonstrated that regions that possess strong expected packing density can be responsible for amyloidogenic properties of a protein, while regions with weak expected packing density correspond to disordered regions. A new concept is proposed that could aid in understanding protein folding, misfolding, and amyloidosis. The results help to explain that the nature of the amyloidogenic propensity of proteins is connected to their amino acid sequences that are able to form a large number of contacts.
Citation: Galzitskaya OV, Garbuzynskiy SO, Lobanov MY (2006) Prediction of Amyloidogenic and Disordered Regions in Protein Chains. PLoS Comput Biol 2(12): e177. https://doi.org/10.1371/journal.pcbi.0020177
Editor: Eugene I. Shakhnovich, Harvard University, United States of America
Received: July 31, 2006; Accepted: November 6, 2006; Published: December 29, 2006
Copyright: © 2006 Galzitskaya et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported by the program Molecular and Cell Biology of the Russian Academy of Sciences, by the Russian Foundation for Basic Research (grant 05-04-48750), by the Howard Hughes Medical Institute (55005607), and by INTAS grant 05-1000004-7747.
Competing interests: The authors have declared that no competing interests exist.
Abbreviations: Aβ, amyloid β peptide; NAC, non-Aβ component of Alzheimer disease amyloid; ROC, receiver operating characteristic
Amyloid fibril formation is associated with an increase of β structure content, which leads to fibrillar aggregation . However, it should be noted that an increased level of the beta structure is a characteristic property of several different types of protein aggregates (amyloid fibrils, amorphous aggregates) [2,3]. In addition to proteins observed in amyloid diseases, recent studies have shown that diverse proteins not related to any amyloid disease can aggregate into fibrils under destabilizing conditions [4−6]. Normal proteins can become toxic when they undergo fibrillation . There is no consensus about toxicity of the different states: small oligomers, large oligomers, protofilaments, protofibrils, filaments, mature fibrils, or amorphous aggregates. Significant advancements in recent research have led to the discovery that the toxic species in the amyloid diseases may not be the fibrils themselves, but rather the pre-fibrillar aggregates . A possible mechanism for toxicity of α-synuclein protofibrils has been demonstrated . It has been shown that protofibrils can form elliptical pores, like bacterial toxins, which can puncture cell membranes, resulting in cell death . Therefore, the mechanism of amyloid formation is under intensive investigation. Recognition of the factors that influence protein conformational changes and misfolding is one of the general fundamental problems, the solution to which will help us find effective treatments for amyloid illnesses.
The experimental observation that not all proteins are amyloidogenic (or at least that some proteins are less amyloidogenic than others) and that specific continuous regions of amyloid-forming proteins are more amyloidogenic than others suggests that there is a sequence propensity for amyloid formation. Moreover, the observation that some short peptides also can form amyloids implies that these segments, which usually are exposed to the environment, can nucleate the transition of native proteins into the amyloid state, and suggests that fibril formation is sequence-specific . In the mechanism of amyloidogenesis for natively folded proteins such as β2-microglobulin and transthyretin, the partial unfolding observed is believed to be a prerequisite for the proteins' assembly into amyloid fibrils both in vitro and in vivo . It has been suggested that residues with enhanced flexibility and solvent accessibility are important for the initiation of fibrillation . This means that partial unfolding of the rigid native structure can provide a specific interface for the beginning of fibrillation. Thus, to understand the molecular mechanism of amyloidosis, it is necessary to find factors that induce partial unfolding of proteins and subsequent amyloid fibril formation at or near physiological conditions.
Some intrinsically disordered proteins are involved in amyloid diseases (type II diabetes, Alzheimer disease, and Parkinson disease). This fact may indicate that disorder is a necessary condition for aggregation. It has been shown that a very small change in the environment of such proteins often might cause their partial folding and aggregation . Knowledge of characteristics that control the process of amyloid fibril formation is important for finding effective drugs for treatment of amyloid diseases.
Uversky and Fink in their review  illustrate that protein fibrillogenesis requires a partially folded conformation (originated from partial unfolding of intrinsically structured proteins or partial folding of intrinsically disordered proteins).
The first high-resolution (1 Å) crystal of an amyloid fiber formed by a sequence-designed polypeptide has been obtained . Recently, the atomic structure of the cross-β spine  for a seven-residue peptide segment from Sup35 (GNNQQNY) was determined. It is a double β sheet, in which each sheet is formed from parallel segments stacked in register. Side chains protruding from the two sheets form a dry, tightly self-complementing steric zipper that bonds the sheets. Within each sheet, every segment is bound to two neighboring segments through stacks of both backbone and side-chain hydrogen bonds.
There are several computational methods for predicting a protein's propensity for amyloid fibril formation. In the work of Fernandez et al.  it was shown that a concentration of such defects as insufficient shielding of hydrogen bonds from water attack might yield an aggregation-induced nucleus. But the analysis of these defects revealed that the extensive exposure of hydrogen bonds to water attack might be a necessary but not sufficient condition to imply a propensity for organized aggregation .
A computational algorithm has been suggested that detects the nonnative (hidden) β strand propensity of sequences by consideration of the relationships between protein local sequence and secondary structure in terms of tertiary contacts . This algorithm detects sequences within the protein that are favorable for triggering amyloid fibril formation. It is worthwhile to emphasize here that both algorithms for prediction of amyloidogenic properties of polypeptide chains that are considered above can be applied only to those proteins for which the three-dimensional structure is known.
Based on the physico–chemical properties of β aggregation sequences and a computational algorithm, a model was developed for predicting the aggregation rate for a broad range of polypeptide chains . The model identifies aggregation sites within a protein and predicts the parallel or antiparallel organization of β sheets in a fibril. It should be noted, however, that the overpredictions of aggregation sites were not analyzed statistically.
On the other hand, there is a method for the prediction of amyloidogenic regions from amino acid sequence alone . After the experimental investigation of the amyloidogenic properties of a model six-residue peptide and its mutants, the authors obtained a six-residue amyloidogenic pattern (STVIIE) and used this pattern for the identification of amyloidogenic fragments in proteins . This amyloidogenic pattern has been used to validate the premise that the amyloidogenicity of a protein is indeed localized in short protein stretches (amyloid stretch hypothesis ). It has been demonstrated that the conversion of a soluble, non-amyloidogenic protein (SH3 domain of α-spectrin) into an amyloidogenic-prone molecule can be triggered by a non-destabilizing six-residue amyloidogenic insertion in a particular structural environment.
Recently, a new method for identifying fibril-forming segments of proteins has been suggested . This method is based on the threading of six-residue peptides through the known crystal structure of an amyloid fiber  formed by the peptide from Sup35. The putative prediction is accepted as a prediction if its energy evaluated with RosettaDesign (http://www.rosettacommons.org) is lower than the threshold energy.
It should be added that molecular dynamics can yield valuable information about the structural changes that arise at the atomic level upon the formation of amyloid fibrils [23−24], while such information is difficult to obtain experimentally.
Another interesting new method (named PASTA) is based on sequence-specific interaction energies between pairs of protein fragments calculated from statistical analysis of the native folds of globular proteins . This algorithm correctly predicts the positions of most aggregation-prone portions of some polypeptide chains.
The formation of a sufficient number of interactions is necessary to compensate for the loss of conformational entropy during the protein folding process. Therefore, the structural uniqueness of native proteins is a result of the balance between the conformational entropy and the energy of residue interactions. It seems that disordered regions in a protein chain do not have a sufficient number of interactions to compensate for the loss of conformational entropy that results from the formation of a globular state. On the other hand, a large increase in the energy of interactions will lead to a loss of the unique structure because the strengthening of contact energy will speed up folding, but it is also likely to lead to erroneous folds (for example, to amyloid fibrils).
It has been suggested that the lack of a rigid globular structure under physiological conditions might represent a considerable functional advantage for intrinsically disordered proteins, as their large plasticity allows them to interact efficiently with several different targets, as compared with a folded protein with limited conformational flexibility [25−29]. It has been shown that disordered regions are involved in DNA binding and other types of molecular recognition . A large portion of the sequences of intrinsically disordered proteins contain segments of low complexity and high predicted flexibility [31−38]. It also has been indicated that a combination of low overall hydrophobicity and a large net charge represent a structural feature of intrinsically disordered proteins in comparison with small globular proteins [39,40]. There are currently several widely used methods for prediction of disordered regions: GlobPlot , a simple propensity-based approach for evaluating the tendency of residues to be in a regular secondary structure; PONDR VL3H , which is able to distinguish experimentally verified disordered proteins from globular proteins by various machine learning approaches; DISOPRED , in which the definition of disorder is restrained to regions that are missing from X-ray structures but are specifically recognized by a support vector machine in the DISOPRED model; and IUPred , which assigns the order/disorder status to residues on the basis of their ability to form favorable pairwise contacts. We were the first to our knowledge who used the number of contacts per residue as a parameter to distinguish folded and intrinsically disordered proteins . We have extended our method to predict disordered regions and have made comparisons with the above-mentioned methods . It has been demonstrated that our method is the best among widely used methods for the sets of proteins considered here.
Despite considerable efforts to understand the mechanism, it is still unclear what is responsible for amyloidogenic and disordered regions. The goal of this work is to test our hypothesis about whether protein regions that possess expected strong packing density are responsible for the amyloidogenic properties of proteins, while regions with weak packing density simultaneously are responsible for the appearance of disordered regions. We introduce a new parameter, namely mean packing density (number of residues within the given distance from the considered residue), which enables the prediction of both amyloidogenic and intrinsically disordered regions from protein sequence. These findings support the concept that the occurrence of amyloidogenic and intrinsically disordered regions has a similar basis in different peptides and proteins.
Observed Mean Packing Density for 20 Types of Amino Acid Residues and Expected Packing Density Profiles
To calculate the packing density observed in protein structures, we have constructed two databases of protein structures. The first one [45,46] includes proteins with sequence identity less than 80% (database 80%). The second database consists of proteins with sequence identity less than 25% (database 25%). The average packing density observed in protein structures (database 25%) for each of 20 types of amino acid residues is shown in Table 1. For database 80%, the 20 values were not identical but very similar (they can be found in ), so that the correlation coefficient between the two sets of values was as large as 99.95%. These values were considered to be the expected packing density for the residues in each protein or peptide sequence studied further. It is worth noting here that three aromatic residues (tryptophane, tyrosine, and phenylalanine) have the highest observed packing density among the 20 amino acids in both databases. Among the many parameters that have been proposed to promote amyloid fibril formation is the π-stacking of aromatic residues [47,48]. Many amyloidogenic regions of proteins have high content of aromatic residues. From experimental works [49,50], one can suggest that aromatic residues favor aggregation because of hydrophobicity, size, and intrinsic β sheet propensity rather than aromaticity. The specific nature of the side-chain interactions for each protein will drive the rate of fibril formation as well as the resulting stability.
Mean Observed Packing Density for 20 Amino Acid Residues (and Errors in the Determination of the Average) Obtained Using a Contact Radius of 8.0 Å from Database 25%
The expected packing densities were averaged over a sliding window, and a packing density profile was produced (see Materials and Methods). Similarly, the other types of profiles were built using other scales instead of the scale from Table 1 (for example, hydrophobicity profile basing on hydrophobicity scale, etc.).
Searching for Peptides That Are Fibril Formers and Fibril Nonformers
To obtain a threshold for our predictions, we took a database of six-residue peptides, some of which were fibril formers and some of which were fibril nonformers . The receiver operator characteristic (ROC) curves for our method are shown in Figure 1. The four ROC curves correspond to four scales: packing density for database 25% (Table 1), packing density for database 80% , hydrophobicity , and β sheet propensity . For further investigations, we considered the following values the thresholds for predicting amyloidogenic regions (which gave rather a high level of true predictions, about 80%, as well as a rather low level of false predictions, about 25%): packing density greater than 21.5 and 21.4 for the two scales obtained from database 25% and database 80%, correspondingly; hydrophobicity less than −0.75, and β sheet propensity less than −0.46 (the corresponding points on the ROC curves [Figure 1] are marked with symbols). It should be mentioned that when we consider the packing density scale for database 80%, the ROC curve is slightly better; the threshold is 21.4.
Searching for Optimal-Residue Long Sliding Window for Prediction of Amyloidogenic Regions
We collected a database of all known proteins and peptides that are associated with amyloid diseases, and for which the position of amyloidogenic regions is now experimentally examined (see Table 2). Amyloids are elongated fibrils that bind the aromatic dyes Congo red and Thioflavin-T have a common cross-β X-ray diffraction pattern .
Comparison of Prediction of Amyloidogenic Regions Using Contact Density Scale and Varying Size of Sliding Window (Scale Obtained from Database 25%)
Varying the size of the sliding window (three, five, seven, and nine residues), we constructed a packing density profile for each of these proteins and peptides. We predicted a region as amyloidogenic if expected packing density for the region (with size equal or greater than size of the window) is above the considered threshold. Our hypothesis is that regions with strong expected packing density should correspond to aggregation regions, which presumably intersect with amyloidogenic regions of proteins. The number of predicted amyloidogenic regions are presented in Table 2. One can see that the window size of seven residues is optimal for the prediction of amyloidogenic regions. The result was very similar for the scale obtained from the 80% database (cutoff is 21.4 for this scale).
Searching for Amyloidogenic Regions in Proteins with Known Disease-Related Regions
We constructed a packing density profile using a sliding window of seven residues for each of the proteins and peptides considered here. The experimentally observed amyloidogenic regions and the predicted ones are presented in Table 3 (25% database). One can see that for eight of 12 examined proteins and peptides the predictions are consistent with the experimentally found amyloidogenic regions.
Predicted versus Experimentally Observed Amyloid-Forming Regions in Amyloidogenic Proteins and Peptides
In Alzheimer disease, τ-protein forms neurofibrillary tangles, which are bundles of paired helical filaments. A single region (amino acid residues 306−311), which is shown experimentally to be amyloidogenic , is correctly predicted by our method when we use a sliding window of five residues.
Despite a large body of experimental data related to the search for amyloidogenic regions in human prion protein, it is difficult to determine which regions these are. It has been shown that helix 1 (residues 144–153) of human prion protein (PrP) plays a critical role in the amyloidogenic process [55,56]. Peptides corresponding to three helical regions (residues 144−154, helical region one; residues 178–193, helical region two; and residues 198–218, helical region three) have been synthesized and studied . The peptides corresponding to the second helical region, residues 180–193 and residues 178–193, are the only ones that form an amyloid structure, according to data obtained by electron microscopy and Congo red birefringence . By using two intrinsic fluorescent variants of this protein (Y150W and F141W), conformational changes confined to the 132–160 segment were monitored . Our predicted fragments intersect with all helices.
Most mutations described in apolipoprotein A (ApoA) are within the N-terminal portion of the protein (residues 1–93), which represents the proteolysis fragment that is incorporated into amyloid deposits . We predict as amyloidogenic one region (residues 15–21) within the N-terminal portion as well as one additional region in the C-terminal part of apolipoprotein A, which has strong expected packing density.
The experimentally found amyloidogenic fragment of lysozyme (residues 49–64), which has been specifically implicated in amyloidogenic conversion [60,61], is a part of the β domain in the native structure of the protein. Our predictions for lysozyme are consistent with experimental results; however, three additional fragments (25–33, 76–82, and 107–114) are also predicted.
The most amyloidogenic peptide fragments from transthyretin (TTR) have been demonstrated in two regions: residues 10–19, which encompass the A strand of the inner β sheet structure that readily forms amyloid fibrils when dissolved in water at low pH [62,63]; and residues 105–115, which adopt an extended β strand conformation that is similar to that found in the native protein . We predicted correctly these important regions (11–17 and 105–113) and one additional region with strong expected packing density.
It has been found experimentally that the following sequences play a dominant role in the amyloidogenesis of β2-microglobulin: residues 20–41 , residues 59–71 , and residues 83–89 . All predicted regions are consistent with the experimental data except for fragment 83–89.
Reactive (or secondary) amyloidosis is characterized by the extracellular deposition of amyloid fibrils containing predominantly amyloid A protein (AA), which is a proteolytically derived fragment of serum amyloid A (SAA) protein. The N-terminus of amyloid A protein (residues 1–11 of AA protein) was shown to be the amyloidogenic part of the molecule . We predicted this region correctly (residues 1–9).
Medin is the main constituent of the aortic medial amyloid. It is derived from a proteolytic fragment of lactadherin, a mammary epithelial cell–expressed glycoprotein that is secreted as part of the milk fat globule membrane. It was previously demonstrated that an octapeptide fragment of medin (residues 42–49, NFGSVQFV) forms typical well-ordered amyloid fibrils . The last four residues (residues 47–50) have a large expected packing density, yet this region is not predicted by the rules of our algorithm (a region must be at least seven residues).
It has been shown that residues 16–20 in amyloid β (Aβ) peptide are essential for the peptide's polymerization . Also, solid-state NMR and site-directed spin labeling experiments suggest that residues 30–38  form a β strand in the fibrils. Our predictions (residues 15–22) are consistent with the first region.
It has been shown that a fragment (residues 20–27) from amylin (also called human islet amyloid protein or hIAPP) is amyloidogenic and cytotoxic . Other than this one, the shortest active fragments capable of self-assembly were found to be pentapeptides FLVHS (residues 15–19) and NFLVH (residues 14–18) . One of the fragments (residues 12–18) is correctly predicted by our method, but the second amyloidogenic region (residues 20–27) has expected packing density below the threshold.
Alpha-synuclein is a major component of Lewy bodies in Parkinson disease and is found to be associated with several other forms of dementia. The central fragment of α-synuclein (35 residues long), which has been isolated from purified amyloid of Alzheimer disease brains,  is called the non-Aβ component of Alzheimer disease amyloid (NAC). It has been shown that the N-terminal fragment of NAC (residues 3–18) forms aggregates and displays a transition from random coil to β sheet structure . On the contrary, the C-terminal fragment of NAC (residues 19–35) remains in solution with random coil conformation under the same conditions . No regions with expected packing density over 20.5 are observed. The predicted region (residues 9–13) appears only if the threshold is 20.3. Thus, we consider this prediction as a failure.
It has been shown that a peptide consisting of residues 15–19 of the human hormone calcitonin forms highly ordered fibrils, which are similar to those formed by the entire hormone sequence . This region is not predicted by the rules of our algorithm.
Our predicted regions are consistent with known disease-related regions for eight of 12 experimentally well-studied amyloidogenic peptides and proteins (transthyretin, β2-microglobulin, lysozyme, prion protein, and others). This result strongly indicates that the aggregation capability of a protein chain is one of the common properties of amyloid fibrils. Moreover, it should be noted that regions with high packing density are often surrounded by amino acids that disrupt their amyloidogenic capability, regions with weak expected packing density, that is, amyloid breakers.
Here we also tested the ability of two other scales, hydrophobicity  and β sheet propensity , to predict amyloidogenic regions and compared these results with our method of expected packing density. The choice of the thresholds (Figure 1) for these predictions was made in the similar way. On the one hand, from 18 experimentally determined amyloidogenic regions, the expected packing density scale finds 14 regions (see Tables 2 and 4), while the hydrophobicity scale finds nine, and the β sheet propensity scale finds two regions (in other words, the packing density scale misses four amyloidogenic fragments while the hydrophobicity scale misses nine fragments and the β sheet propensity scale misses 16). On the other hand, the scale of expected packing density finds seven additional regions while the scale of hydrophobicity finds seven extra regions, and the scale of β sheet propensity finds six additional regions, the amyloidogenic role of which is not confirmed by experiment. Therefore, here we suggest a new property of peptides and proteins that can be used to predict the formation of amyloid fibrils: regions with strong expected packing density.
Searching for Intrinsically Disordered Regions
To test the quality of our predictions of intrinsically disordered regions in proteins, we have used two databases, of which one has 427 intrinsically disordered proteins and regions  and the other has 559 fully folded proteins . The ROC curves obtained with different sizes of the sliding window are shown in Figure 2. The best result corresponds to the case where we construct the packing density profile smoothed over the sliding window of 41 residues; we chose 20.4 (the corresponding point is marked as a large circle) as the threshold when we use the scale from database 80% (true positives 0.74 and false positives 0.03) and 20.5 when we use the scale from database 25% (true positives 0.74 and false positives 0.05).
Each ROC curve corresponds to predictions with specified (on the legend) size of the sliding window. The open circle corresponds to the value of packing density that is chosen as a threshold, 20.5 for database 25% (A) and 20.4 for database 80% (B).
To test the quality of predictions obtained by our method compared with other methods of prediction of disordered regions such as IUPred , DISOPRED2 , PONDR VL3H , and GlobPlot , we examined the same proteins that were used by Dosztanyi et al. , who compared the quality of predictions obtained by their method IUPred with DISOPRED, PONDR VL3H, and GlobPlot (the data on these methods were taken from ). These were a dataset of globular proteins (559 proteins) and long disordered protein segments (129 proteins). Table 5 demonstrates that our method (FoldUnfold) showed improved performance over these widely used methods on these sets of proteins (the averaging for our method is done in the same two ways as for the other methods —over amino acid residues and over proteins).
We demonstrate that expected packing density is a useful value for the prediction of both intrinsically disordered and amyloidogenic regions of a protein based only on its sequence. In Figure 3, a distribution of average packing densities of globular proteins, is presented. The determined thresholds (21.4 for amyloidogenic regions and 20.4 for intrinsically disordered ones) correspond to the ends of this distribution.
Structures of peptides such as NNQQNY (derived from Sup35 protein ), KFFEAAAKKFFE (a designed 12-mer peptide ), and YTIAALLSPYS (derived from transthyretin ) confirm that the peptides adopt an extended β-strand conformation in amyloid fibrils. These fibrils achieve their stability through optimal values of main-chain and dihedral angles, as well as through extensive hydrophobic packing of side chains (hydrophobic template, Serrano's pattern—STVIIE) and salt bridge formation from polar side chains (polar template, Eisenberg's pattern—NNQQNY). It should be emphasized that between these two templates there probably exist many different intermediate variants. Our approach finds amyloidogenic regions closer to the hydrophobic template than to the polar one.
If amyloid fibril formation is a generic feature of proteins , some common properties of amino acid sequences possessing amyloidogenic propensities should be observed. Experimental data as well as theoretical analyses can help reveal the common structural and chemical properties for this process, one of which is the tight packing density.
We tried to collect all known amyloidogenic proteins and peptides for which disease-related regions are experimentally localized. By analysis of primary structure alone, we have demonstrated that regions that possess strong expected packing density can be responsible for the amyloidogenic properties of a protein, while regions with weak expected packing density correspond to disordered regions. A new concept is proposed that could aid in the understanding of protein folding, misfolding, and amyloidosis.
Our study provides new insights into the process of amyloid formation. The results help to explain that the nature of the amyloidogenic propensity of proteins is related to their amino-acid sequences that are able to form a large number of contacts. Our results can help determine the amyloidogenic propensity of amyloidogenic proteins for which the position of amyloidogenic regions now remains unexplored experimentally.
Materials and Methods
Observed packing density for 20 types of amino acid residues.
The set of protein structures used for calculation of the packing density observed in protein structures was obtained by inspection of the SCOP (Structural Classification of Proteins)  database 1.61 release (for database 80%) and 1.65 release (for database 25%). For the first database, 5,829 domains from four general classes (a–d) with less than 80% sequence identity values were found: 1,133 all-α proteins from class a, 1,644 all-β proteins from class b, 1,617 α/β proteins from class c, and 1,435 α + β proteins from class d. A total of 3,769 domains from four general classes (a–d) with less than 25% sequence identity values were found (database 25%): 794 all-α proteins from class a, 928 all-β proteins from class b, 1,089 α/β proteins from class c, and 958 α + β proteins from class d. The observed packing density for each amino acid residue from this database was calculated as the number of close residues (within the given distance). In our case a residue is considered close to the given residue if any pair of their heavy atoms is at distance of less than 8 Å. The neighboring residues bound with peptide bonds (which are close in any case) are not taken into account. The mean observed packing density for each of 20 types of amino acid residues is presented in Table 1. These 20 values were used for prediction of packing density from protein sequences, that is, the expected packing density (we consider the expected packing density of a residue equal to the mean observed packing density of the corresponding residue in a globular state).
Calculation of the expected packing density profile.
It is worthwhile to emphasize that the order of the residues may play an important role in protein folding and may account for regions with weak and strong packing density in a protein structure. To predict such regions in a protein, we construct a profile of the expected packing density for the protein sequence. The calculations are based on a sliding window-averaging technique. For each peptide and protein, in the prediction of amyloidogenic regions the sliding window size is varied from three to nine residues while the sliding window size is 11 (or 41) residues in the case of intrinsically disordered regions prediction. The packing density profile is calculated as follows. First, the expected packing density is determined for each residue (see Table 1); then, these numbers are averaged for five residues inside the window and assigned to the central residue of the window. Therefore, the influence of residues along the sequence flanking each window is included in our calculation. The value of the average expected packing density for every position of the polypeptide chain provides the packing density profile. If more than five residues in a row have values over a specified threshold, this region is predicted to be amyloidogenic. On the other hand, any region having more than 11 (or 41) residues with values below a specified threshold is predicted to be intrinsically disordered.
Databases used to test our method.
To evaluate the accuracy of, and confidence in, our method of predicting amyloidogenic regions, a database of 67 peptides that are six-residue fibril formers and 91 peptides that are six-residue fibril nonformers was used . To test our method, we also used the amino acid sequences of 12 disease-related amyloidogenic proteins and peptides (for which the position of amyloidogenic regions is localized experimentally); the sequences were taken from the SWISS-PROT database  (http://us.expasy.org/sprot/). To test our method for predicting intrinsically disordered regions, we used three databases. Two of them were downloaded from the Database of Protein Disorder DISPROT . The first one consists of sequences of 427 completely intrinsically disordered proteins and intrinsically disordered fragments. The second database contains 129 intrinsically disordered proteins. The third database consists of 559 globular proteins without intrinsically disordered fragments . This database was constructed using Protein Data Bank (PDB) entries from the above work.
Evaluation of the quality of predictions.
To obtain the quality of predictions and to determine thresholds, we calculated true positive and false positive rates and made so-called receiver operator characteristic (ROC) curves. In predictions of intrinsically disordered regions, the true positive rate was calculated as the fraction of residues predicted as intrinsically disordered over the intrinsically disordered set of residues; the false positive rate was the fraction of predicted intrinsically disordered residues over the set of folded residues. Similarly, in the case of six-residue peptides that were fibril formers, the true positive rate was calculated as the fraction of peptides predicted as fibril formers in the fibril formers set of peptides while the false positive rate was the fraction of peptides predicted as fibril formers in the fibril nonformers set of peptides.
The other scales for prediction of amyloidogenic regions.
Using hydrophobicity and β sheet propensity scales, we predicted the amyloidogenic regions of the considered proteins and peptides and evaluated the obtained results in a similar way to how we predicted these regions using packing density scales. The hydrophobicity scale of 20 types of amino acid residues was taken from the work of Fauchere and Pliska . The β sheet propensities of the 20 types of amino acid residues in an internal β sheet position were taken from the work of Minor and Kim . The original hydrophobicity and β sheet propensity scales were taken with reversed sign since the most hydrophobic and β sheet–predisposed amino acid residues have the largest negative values.
OVG conceived and designed the experiments. SOG and MYL performed the experiments. OVG, SOG, and MYL analyzed the data. OVG wrote the paper.
- 1. Jimenez JL, Guijarro JI, Orlova E, Zurdo J, Dobson CM, et al. (1999) Cryo-electron microscopy structure of an SH3 amyloid fibril and model of the molecular packing. EMBO J 18: 815–821.JL JimenezJI GuijarroE. OrlovaJ. ZurdoCM Dobson1999Cryo-electron microscopy structure of an SH3 amyloid fibril and model of the molecular packing.EMBO J18815821
- 2. Ohnishi S, Takano K (2004) Amyloid fibrils from the viewpoint of protein folding. Cell Mol Life Sci 61: 511–524.S. OhnishiK. Takano2004Amyloid fibrils from the viewpoint of protein folding.Cell Mol Life Sci61511524
- 3. Rousseau F, Schymkowitz J, Serrano L (2006) Protein aggregation and amyloidosis: Confusion of the kinds? Curr Opin Struct Biol 16: 118–126.F. RousseauJ. SchymkowitzL. Serrano2006Protein aggregation and amyloidosis: Confusion of the kinds?Curr Opin Struct Biol16118126
- 4. Guijarro JI, Sunde M, Jones JA, Campbell ID, Dobson CM (1998) Amyloid fibril formation by an SH3 domain. Proc Natl Acad Sci U S A 95: 4224–4228.JI GuijarroM. SundeJA JonesID CampbellCM Dobson1998Amyloid fibril formation by an SH3 domain.Proc Natl Acad Sci U S A9542244228
- 5. Chiti F, Webster P, Taddei N, Clark A, Stefani M, et al. (1999) Designing conditions for in vitro formation of amyloid protofilaments and fibrils. Proc Natl Acad Sci U S A 96: 3590–3594.F. ChitiP. WebsterN. TaddeiA. ClarkM. Stefani1999Designing conditions for in vitro formation of amyloid protofilaments and fibrils.Proc Natl Acad Sci U S A9635903594
- 6. Fandrich M, Fletcher MA, Dobson CM (2001) Amyloid fibrils from muscle myoglobin. Nature 410: 165–166.M. FandrichMA FletcherCM Dobson2001Amyloid fibrils from muscle myoglobin.Nature410165166
- 7. Bucciantini M, Calloni G, Chiti F, Formigli L, Nosi D, et al. (2004) Prefibrillar amyloid protein aggregates share common features of cytotoxicity. J Biol Chem 279: 31374–31382.M. BucciantiniG. CalloniF. ChitiL. FormigliD. Nosi2004Prefibrillar amyloid protein aggregates share common features of cytotoxicity.J Biol Chem2793137431382
- 8. Maries E, Dass B, Collier TJ, Kordower JH, Steece-Collier K (2003) The role of alpha-synuclein in Parkinson's disease: Insights from animal models. Nat Rev Neurosci 4: 727–738.E. MariesB. DassTJ CollierJH KordowerK. Steece-Collier2003The role of alpha-synuclein in Parkinson's disease: Insights from animal models.Nat Rev Neurosci4727738
- 9. Ivanova MI, Sawaya MR, Gingery M, Attinger A, Eisenberg D (2004) An amyloid-forming segment of beta2-microglobulin suggests a molecular model for the fibril. Proc Natl Acad Sci U S A 101: 10584–10589.MI IvanovaMR SawayaM. GingeryA. AttingerD. Eisenberg2004An amyloid-forming segment of beta2-microglobulin suggests a molecular model for the fibril.Proc Natl Acad Sci U S A1011058410589
- 10. Yamamoto S, Hasegawa K, Yamaguchi I, Tsutsumi S, Kardos J, et al. (2004) Low concentrations of sodium dodecyl sulfate induce the extension of beta 2-microglobulin-related amyloid fibrils at a neutral pH. Biochemistry 43: 11075–11082.S. YamamotoK. HasegawaI. YamaguchiS. TsutsumiJ. Kardos2004Low concentrations of sodium dodecyl sulfate induce the extension of beta 2-microglobulin-related amyloid fibrils at a neutral pH.Biochemistry431107511082
- 11. Pedersen JS, Christensen G, Otzen DE (2004) Modulation of S6 fibrillation by unfolding rates and gatekeeper residues. J Mol Biol 341: 575–588.JS PedersenG. ChristensenDE Otzen2004Modulation of S6 fibrillation by unfolding rates and gatekeeper residues.J Mol Biol341575588
- 12. Yamin G, Munishkina LA, Karymov MA, Lyubchenko YL, Uversky VN, et al. (2005) Forcing nonamyloidogenic beta-synuclein to fibrillate. Biochemistry 44: 9096–9107.G. YaminLA MunishkinaMA KarymovYL LyubchenkoVN Uversky2005Forcing nonamyloidogenic beta-synuclein to fibrillate.Biochemistry4490969107
- 13. Uversky VN, Fink AL (2004) Conformational constraints for amyloid fibrillation: The importance of being unfolded. Biochim Biophys Acta 1698: 131–153.VN UverskyAL Fink2004Conformational constraints for amyloid fibrillation: The importance of being unfolded.Biochim Biophys Acta1698131153
- 14. Makin OS, Atkins E, Sikorski P, Johansson J, Serpell LC (2005) Molecular basis for amyloid fibril formation and stability. Proc Natl Acad Sci U S A 102: 315–320.OS MakinE. AtkinsP. SikorskiJ. JohanssonLC Serpell2005Molecular basis for amyloid fibril formation and stability.Proc Natl Acad Sci U S A102315320
- 15. Nelson R, Sawaya MR, Balbirnie M, Madsen AO, Riekel C, et al. (2005) Structure of the cross-beta spine of amyloid-like fibrils. Nature 435: 747–749.R. NelsonMR SawayaM. BalbirnieAO MadsenC. Riekel2005Structure of the cross-beta spine of amyloid-like fibrils.Nature435747749
- 16. Fernandez A, Kardos J, Scott LR, Goto Y, Berry RS (2003) Structural defects and the diagnosis of amyloidogenic propensity. Proc Natl Acad Sci U S A 100: 6446–6451.A. FernandezJ. KardosLR ScottY. GotoRS Berry2003Structural defects and the diagnosis of amyloidogenic propensity.Proc Natl Acad Sci U S A10064466451
- 17. Yoon S, Welsh WJ (2004) Detecting hidden sequence propensity for amyloid fibril formation. Protein Sci 13: 2149–2160.S. YoonWJ Welsh2004Detecting hidden sequence propensity for amyloid fibril formation.Protein Sci1321492160
- 18. Tartaglia GG, Cavalli A, Pellarin R, Caflisch A (2005) Prediction of aggregation rate and aggregation-prone segments in polypeptide sequences. Protein Sci 14: 2723–2734.GG TartagliaA. CavalliR. PellarinA. Caflisch2005Prediction of aggregation rate and aggregation-prone segments in polypeptide sequences.Protein Sci1427232734
- 19. Lopez de la Paz M, Serrano L (2004) Sequence determinants of amyloid fibril formation. Proc Natl Acad Sci U S A 101: 87–92.M. Lopez de la PazL. Serrano2004Sequence determinants of amyloid fibril formation.Proc Natl Acad Sci U S A1018792
- 20. Esteras-Chopo A, Serrano L, Lopez de la Paz M (2005) The amyloid stretch hypothesis: Recruiting proteins toward the dark side. Proc Natl Acad Sci U S A 102: 16672–16677.A. Esteras-ChopoL. SerranoM. Lopez de la Paz2005The amyloid stretch hypothesis: Recruiting proteins toward the dark side.Proc Natl Acad Sci U S A1021667216677
- 21. Thompson MJ, Sievers SA, Karanicolas J, Ivanova MI, Baker D, et al. (2006) The 3D profile method for identifying fibril-forming segments of proteins. Proc Natl Acad Sci U S A 103: 4074–4078.MJ ThompsonSA SieversJ. KaranicolasMI IvanovaD. Baker2006The 3D profile method for identifying fibril-forming segments of proteins.Proc Natl Acad Sci U S A10340744078
- 22. Trovato A, Chiti F, Maritan A, Seno F (2006) Insight into the structure of amyloid fibrils from the analysis of globular proteins . PLoS Comput Biol 2(12): e170.. A. TrovatoF. ChitiA. MaritanF. Seno2006Insight into the structure of amyloid fibrils from the analysis of globular proteins .PLoS Comput Biol212e170.
- 23. Gsponer J, Haberthur U, Caflisch A (2003) The role of side-chain interactions in the early steps of aggregation: Molecular dynamics simulations of an amyloid-forming peptide from the yeast prion Sup35. Proc Natl Acad Sci U S A 100: 5154–5159.J. GsponerU. HaberthurA. Caflisch2003The role of side-chain interactions in the early steps of aggregation: Molecular dynamics simulations of an amyloid-forming peptide from the yeast prion Sup35.Proc Natl Acad Sci U S A10051545159
- 24. Khare SD, Wilcox KC, Gong P, Dokholyan NV (2005) Sequence and structural determinants of Cu, Zn superoxide dismutase aggregation. Proteins 61: 617–632.SD KhareKC WilcoxP. GongNV Dokholyan2005Sequence and structural determinants of Cu, Zn superoxide dismutase aggregation.Proteins61617632
- 25. Wright PE, Dyson HJ (1999) Intrinsically unstructured proteins: Re-assessing the protein structure–function paradigm. J Mol Biol 293: 321–331.PE WrightHJ Dyson1999Intrinsically unstructured proteins: Re-assessing the protein structure–function paradigm.J Mol Biol293321331
- 26. Dyson HJ, Wright PE (2002) Insights into the structure and dynamics of unfolded proteins from nuclear magnetic resonance. Adv Protein Chem 62: 311–340.HJ DysonPE Wright2002Insights into the structure and dynamics of unfolded proteins from nuclear magnetic resonance.Adv Protein Chem62311340
- 27. Romero P, Obradovic Z, Dunker AK (2004) Natively disordered proteins: Functions and predictions. Appl Bioinformatics 3: 105–113.P. RomeroZ. ObradovicAK Dunker2004Natively disordered proteins: Functions and predictions.Appl Bioinformatics3105113
- 28. Fink AL (2005) Natively unfolded proteins. Curr Opin Struct Biol 15: 35–41.AL Fink2005Natively unfolded proteins.Curr Opin Struct Biol153541
- 29. Dunker AK, Cortese MS, Romero P, Iakoucheva LM, Uversky VN (2005) Flexible nets. The roles of intrinsic disorder in protein interaction networks. FEBS J 272: 5129–5148.AK DunkerMS CorteseP. RomeroLM IakouchevaVN Uversky2005Flexible nets. The roles of intrinsic disorder in protein interaction networks.FEBS J27251295148
- 30. Dyson HJ, Wright PE (2005) Intrinsically unstructured proteins and their functions. Nat Rev Mol Cell Biol 6: 197–208.HJ DysonPE Wright2005Intrinsically unstructured proteins and their functions.Nat Rev Mol Cell Biol6197208
- 31. Wootton JC (1994) Non-globular domains in protein sequences: Automated segmentation using complexity measures. Comput Chem 18: 269–285.JC Wootton1994Non-globular domains in protein sequences: Automated segmentation using complexity measures.Comput Chem18269285
- 32. Dunker AK, Garner E, Guilliot S, Romero P, Albrecht K, et al. (1998) Protein disorder and the evolution of molecular recognition: Theory, predictions and observations. Pac Symp Biocomput. pp. 473–484.AK DunkerE. GarnerS. GuilliotP. RomeroK. Albrecht1998Protein disorder and the evolution of molecular recognition: Theory, predictions and observations.Pac Symp Biocomput473484
- 33. Romero P, Obradovic Z, Kissinger CR, Villafranca JE, Garner E, et al. (1998) Thousands of proteins likely to have long disordered regions. Pac Symp Biocomput 3: 437–448.P. RomeroZ. ObradovicCR KissingerJE VillafrancaE. Garner1998Thousands of proteins likely to have long disordered regions.Pac Symp Biocomput3437448
- 34. Romero P, Obradovic Z, Dunker AK (1999) Folding minimal sequences: The lower bound for sequence complexity of globular proteins. FEBS Lett 462: 363–367.P. RomeroZ. ObradovicAK Dunker1999Folding minimal sequences: The lower bound for sequence complexity of globular proteins.FEBS Lett462363367
- 35. Galzitskaya OV, Surin AK, Nakamura H (2000) Optimal region of average side-chain entropy for fast protein folding. Protein Sci 9: 580–586.OV GalzitskayaAK SurinH. Nakamura2000Optimal region of average side-chain entropy for fast protein folding.Protein Sci9580586
- 36. Vucetic S, Brown CJ, Dunker AK, Obradovic Z (2003) Flavors of protein disorder. Proteins 52: 573–584.S. VuceticCJ BrownAK DunkerZ. Obradovic2003Flavors of protein disorder.Proteins52573584
- 37. Obradovic Z, Peng K, Vucetic S, Radivojac P, Brown CJ, et al. (2003) Predicting intrinsic disorder from amino acid sequence. Proteins 53: 566–572.Z. ObradovicK. PengS. VuceticP. RadivojacCJ Brown2003Predicting intrinsic disorder from amino acid sequence.Proteins53566572
- 38. Radivojac P, Obradovic Z, Smith DK, Zhu G, Vucetic S, et al. (2004) Protein flexibility and intrinsic disorder. Protein Sci 13: 71–80.P. RadivojacZ. ObradovicDK SmithG. ZhuS. Vucetic2004Protein flexibility and intrinsic disorder.Protein Sci137180
- 39. Uversky VN, Gillespie JR, Fink AL (2000) Why are “natively unfolded” proteins unstructured under physiologic conditions? Proteins 41: 415–427.VN UverskyJR GillespieAL Fink2000Why are “natively unfolded” proteins unstructured under physiologic conditions?Proteins41415427
- 40. Uversky VN (2002) What does it mean to be natively unfolded? Eur J Biochem 269: 2–12.VN Uversky2002What does it mean to be natively unfolded?Eur J Biochem269212
- 41. Linding R, Jensen LJ, Diella F, Bork P, Gibson TJ, et al. (2003) Protein disorder prediction: Implications for structural proteomics. Structure 11: 1453–1459.R. LindingLJ JensenF. DiellaP. BorkTJ Gibson2003Protein disorder prediction: Implications for structural proteomics.Structure1114531459
- 42. Ward JJ, McGuffin LJ, Bryson K, Buxton BF, Jones DT (2004) The DISOPRED server for the prediction of protein disorder. Bioinformatics 20: 2138–2139.JJ WardLJ McGuffinK. BrysonBF BuxtonDT Jones2004The DISOPRED server for the prediction of protein disorder.Bioinformatics2021382139
- 43. Dosztanyi Z, Csizmok V, Tompa P, Simon I (2005) The pairwise energy content estimated from amino acid composition discriminates between folded and intrinsically unstructured proteins. J Mol Biol 347: 827–839.Z. DosztanyiV. CsizmokP. TompaI. Simon2005The pairwise energy content estimated from amino acid composition discriminates between folded and intrinsically unstructured proteins.J Mol Biol347827839
- 44. Garbuzynskiy SO, Lobanov MY, Galzitskaya OV (2004) To be folded or to be unfolded? Protein Sci 13: 2871–2877.SO GarbuzynskiyMY LobanovOV Galzitskaya2004To be folded or to be unfolded?Protein Sci1328712877
- 45. Galzitskaya OV, Garbuzynskiy SO, Lobanov MY (2006) Prediction of natively unfolded regions in protein chain. Mol Biol (Moscow) 40: 341–348.OV GalzitskayaSO GarbuzynskiyMY Lobanov2006Prediction of natively unfolded regions in protein chain.Mol Biol (Moscow)40341348
- 46. Galzitskaya OV, Garbuzynskiy SO, Lobanov MY (2006) FoldUnfold: Web server for the prediction of disordered regions in protein chain. Bioinformatics 22: 2948–2949.OV GalzitskayaSO GarbuzynskiyMY Lobanov2006FoldUnfold: Web server for the prediction of disordered regions in protein chain.Bioinformatics2229482949
- 47. Gazit E (2002) A possible role for pi-stacking in the self-assembly of amyloid fibrils. FASEB J 16: 77–83.E. Gazit2002A possible role for pi-stacking in the self-assembly of amyloid fibrils.FASEB J167783
- 48. Tartaglia GG, Cavalli A, Pellarin R, Caflisch A (2004) The role of aromaticity, exposed surface, and dipole moment in determining protein aggregation rates. Protein Sci 13: 1939–1941.GG TartagliaA. CavalliR. PellarinA. Caflisch2004The role of aromaticity, exposed surface, and dipole moment in determining protein aggregation rates.Protein Sci1319391941
- 49. Tracz SM, Abedini A, Driscoll M, Raleigh DP (2004) Role of aromatic interactions in amyloid formation by peptides derived from human amylin. Biochemistry 43: 15901–15908.SM TraczA. AbediniM. DriscollDP Raleigh2004Role of aromatic interactions in amyloid formation by peptides derived from human amylin.Biochemistry431590115908
- 50. Bemporad F, Taddei N, Stefani M, Chiti F (2006) Assessing the role of aromatic residues in the amyloid aggregation of human muscle acylphosphatase. Protein Sci 15: 862–870.F. BemporadN. TaddeiM. StefaniF. Chiti2006Assessing the role of aromatic residues in the amyloid aggregation of human muscle acylphosphatase.Protein Sci15862870
- 51. Fauchere II, Pliska V (1983) Hydrophobic parameters amino-acid side chains from partitioning of N-acetyl-amino-acid amides. Eur J Med Chem-Chim Ther 18: 369–375.II FauchereV. Pliska1983Hydrophobic parameters amino-acid side chains from partitioning of N-acetyl-amino-acid amides.Eur J Med Chem-Chim Ther18369375
- 52. Minor DL Jr, Kim PS (1994) Context is a major determinant of beta-sheet propensity. Nature 371: 264–267.DL Minor JrPS Kim1994Context is a major determinant of beta-sheet propensity.Nature371264267
- 53. Rudall KM (1952) The proteins of the mammalian epidermis. Adv Protein Chem 7: 253–290.KM Rudall1952The proteins of the mammalian epidermis.Adv Protein Chem7253290
- 54. Von Bergen M, Friedhoff P, Biernat J, Heberle J, Mandelkow EM, et al. (2000) Assembly of tau protein into Alzheimer paired helical filaments depends on a local sequence motif ((306)VQIVYK(311)) forming beta structure. Proc Natl Acad Sci U S A 97: 5129–5134.M. Von BergenP. FriedhoffJ. BiernatJ. HeberleEM Mandelkow2000Assembly of tau protein into Alzheimer paired helical filaments depends on a local sequence motif ((306)VQIVYK(311)) forming beta structure.Proc Natl Acad Sci U S A9751295134
- 55. Morrissey MP, Shakhnovich EI (1999) Evidence for the role of PrP(C) helix 1 in the hydrophilic seeding of prion aggregates. Proc Natl Acad Sci U S A 96: 11293–11298.MP MorrisseyEI Shakhnovich1999Evidence for the role of PrP(C) helix 1 in the hydrophilic seeding of prion aggregates.Proc Natl Acad Sci U S A961129311298
- 56. Speare JO, Rush IIITS, Bloom ME, Caughey B (2003) The role of helix 1 aspartates and salt bridges in the stability and conversion of prion protein. J Biol Chem 278: 12522–12529.JO SpeareIIITS RushME BloomB. Caughey2003The role of helix 1 aspartates and salt bridges in the stability and conversion of prion protein.J Biol Chem2781252212529
- 57. Thompson A, White AR, McLean C, Masters CL, Cappai R, et al. (2000) Amyloidogenicity and neurotoxicity of peptides corresponding to the helical regions of PrP(C). J Neurosci Res 62: 293–301.A. ThompsonAR WhiteC. McLeanCL MastersR. Cappai2000Amyloidogenicity and neurotoxicity of peptides corresponding to the helical regions of PrP(C).J Neurosci Res62293301
- 58. Torrent J, Alvarez-Martinez MT, Liautard JP, Balny C, Lange R (2005) The role of the 132–160 region in prion protein conformational transitions. Protein Sci 14: 956–967.J. TorrentMT Alvarez-MartinezJP LiautardC. BalnyR. Lange2005The role of the 132–160 region in prion protein conformational transitions.Protein Sci14956967
- 59. Hamidi AK, Liepnieks JJ, Nakamura M, Parker F, Benson MD (1999) A novel apolipoprotein A-1 variant, Arg173Pro, associated with cardiac and cutaneous amyloidosis. Biochem Biophys Res Commun 257: 584–588.AK HamidiJJ LiepnieksM. NakamuraF. ParkerMD Benson1999A novel apolipoprotein A-1 variant, Arg173Pro, associated with cardiac and cutaneous amyloidosis.Biochem Biophys Res Commun257584588
- 60. Krebs MR, Wilkins DK, Chung EW, Pitkeathly MC, Chamberlain AK, et al. (2000) Formation and seeding of amyloid fibrils from wild-type hen lysozyme and a peptide fragment from the beta-domain. J Mol Biol 300: 541–549.MR KrebsDK WilkinsEW ChungMC PitkeathlyAK Chamberlain2000Formation and seeding of amyloid fibrils from wild-type hen lysozyme and a peptide fragment from the beta-domain.J Mol Biol300541549
- 61. Frare E, Polverino de Laureto P, Zurdo J, Dobson CM, et al. (2004) A highly amyloidogenic region of hen lysozyme. J Mol Biol 340: 1153–1165.E. FrareP. Polverino de LauretoJ. ZurdoCM Dobson2004A highly amyloidogenic region of hen lysozyme.J Mol Biol34011531165
- 62. Chamberlain AK, MacPhee CE, Zurdo J, Morozova-Roche LA, Hill HA, et al. (2000) Ultrastructural organization of amyloid fibrils by atomic force microscopy. Biophys J 79: 3282–3293.AK ChamberlainCE MacPheeJ. ZurdoLA Morozova-RocheHA Hill2000Ultrastructural organization of amyloid fibrils by atomic force microscopy.Biophys J7932823293
- 63. MacPhee CE, Dobson CM (2000) Chemical dissection and reassembly of amyloid fibrils formed by a peptide fragment of transthyretin. J Mol Biol 297: 1203–1215.CE MacPheeCM Dobson2000Chemical dissection and reassembly of amyloid fibrils formed by a peptide fragment of transthyretin.J Mol Biol29712031215
- 64. Jaroniec CP, MacPhee CE, Astrof NS, Dobson CM, Griffin RG (2002) Molecular conformation of a peptide fragment of transthyretin in an amyloid fibril. Proc Natl Acad Sci U S A 99: 16748–16753.CP JaroniecCE MacPheeNS AstrofCM DobsonRG Griffin2002Molecular conformation of a peptide fragment of transthyretin in an amyloid fibril.Proc Natl Acad Sci U S A991674816753
- 65. Kozhukh GV, Hagihara Y, Kawakami T, Hasegawa K, Naiki H, et al. (2002) Investigation of a peptide responsible for amyloid fibril formation of beta 2-microglobulin by Achromobacter protease I. J Biol Chem 277: 1310–1315.GV KozhukhY. HagiharaT. KawakamiK. HasegawaH. Naiki2002Investigation of a peptide responsible for amyloid fibril formation of beta 2-microglobulin by Achromobacter protease I.J Biol Chem27713101315
- 66. Jones S, Manning J, Kad NM, Radford SE (2003) Amyloid-forming peptides from beta2-microglobulin—Insights into the mechanism of fibril formation in vitro. J Mol Biol 325: 249–257.S. JonesJ. ManningNM KadSE Radford2003Amyloid-forming peptides from beta2-microglobulin—Insights into the mechanism of fibril formation in vitro.J Mol Biol325249257
- 67. Patel H, Bramall J, Waters H, De Beer MC, Woo P (1996) Expression of recombinant human serum amyloid A in mammalian cells and demonstration of the region necessary for high-density lipoprotein binding and amyloid fibril formation by site-directed mutagenesis. Biochem J 318: 1041–1049.H. PatelJ. BramallH. WatersMC De BeerP. Woo1996Expression of recombinant human serum amyloid A in mammalian cells and demonstration of the region necessary for high-density lipoprotein binding and amyloid fibril formation by site-directed mutagenesis.Biochem J31810411049
- 68. Reches M, Gazit E (2004) Amyloidogenic hexapeptide fragment of medin: Homology to functional islet amyloid polypeptide fragments. Amyloid 11: 81–89.M. RechesE. Gazit2004Amyloidogenic hexapeptide fragment of medin: Homology to functional islet amyloid polypeptide fragments.Amyloid118189
- 69. Tjernberg LO, Callaway DJ, Tjernberg A, Hahne S, Lilliehook C, et al. (1999) A molecular model of Alzheimer amyloid beta-peptide fibril formation. J Biol Chem 274: 12619–12625.LO TjernbergDJ CallawayA. TjernbergS. HahneC. Lilliehook1999A molecular model of Alzheimer amyloid beta-peptide fibril formation.J Biol Chem2741261912625
- 70. Torok M, Milton S, Kayed R, Wu P, McIntire T, et al. (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.M. TorokS. MiltonR. KayedP. WuT. McIntire2002Structural and dynamic features of Alzheimer's Abeta peptide in amyloid fibrils studied by site-directed spin labeling.J Biol Chem2774081040815
- 71. Azriel R, Gazit E (2001) Analysis of the minimal amyloid-forming fragment of the islet amyloid polypeptide. An experimental support for the key role of the phenylalanine residue in amyloid formation. J Biol Chem 276: 34156–34161.R. AzrielE. Gazit2001Analysis of the minimal amyloid-forming fragment of the islet amyloid polypeptide. An experimental support for the key role of the phenylalanine residue in amyloid formation.J Biol Chem2763415634161
- 72. Mazor Y, Gilead S, Benhar I, Gazit E (2002) Identification and characterization of a novel molecular-recognition and self-assembly domain within the islet amyloid polypeptide. J Mol Biol 322: 1013–1024.Y. MazorS. GileadI. BenharE. Gazit2002Identification and characterization of a novel molecular-recognition and self-assembly domain within the islet amyloid polypeptide.J Mol Biol32210131024
- 73. Ueda K, Fukushima H, Masliah E, Xia Y, Iwai A, et al. (1993) Molecular cloning of cDNA encoding an unrecognized component of amyloid in Alzheimer disease. Proc Natl Acad Sci U S A 90: 11282–11286.K. UedaH. FukushimaE. MasliahY. XiaA. Iwai1993Molecular cloning of cDNA encoding an unrecognized component of amyloid in Alzheimer disease.Proc Natl Acad Sci U S A901128211286
- 74. Bodles AM, Irvine GB (2004) Alpha-synuclein aggregation. Protein Pept Lett 11: 271–279.AM BodlesGB Irvine2004Alpha-synuclein aggregation.Protein Pept Lett11271279
- 75. Haspel N, Zanuy D, Ma B, Wolfson H, Nussinov R (2005) A comparative study of amyloid fibril formation by residues 15–19 of the human calcitonin hormone: A single beta-sheet model with a small hydrophobic core. J Mol Biol 345: 1213–1227.N. HaspelD. ZanuyB. MaH. WolfsonR. Nussinov2005A comparative study of amyloid fibril formation by residues 15–19 of the human calcitonin hormone: A single beta-sheet model with a small hydrophobic core.J Mol Biol34512131227
- 76. Vucetic S, Obradovic Z, Vacic V, Radivojac P, Peng K, et al. (2005) DisProt: A database of protein disorder. Bioinformatics 21: 137–140.S. VuceticZ. ObradovicV. VacicP. RadivojacK. Peng2005DisProt: A database of protein disorder.Bioinformatics21137140
- 77. Jaroniec CP, MacPhee CE, Bajaj VS, McMahon MT, Dobson CM, et al. (2004) High-resolution molecular structure of a peptide in an amyloid fibril determined by magic angle spinning NMR spectroscopy. Proc Natl Acad Sci U S A 101: 711–716.CP JaroniecCE MacPheeVS BajajMT McMahonCM Dobson2004High-resolution molecular structure of a peptide in an amyloid fibril determined by magic angle spinning NMR spectroscopy.Proc Natl Acad Sci U S A101711716
- 78. Murzin AG, Brenner SE, Hubbard T, Chothia C (1995) SCOP: A structural classification of proteins database for the investigation of sequences and structures. J Mol Biol 247: 536–540.AG MurzinSE BrennerT. HubbardC. Chothia1995SCOP: A structural classification of proteins database for the investigation of sequences and structures.J Mol Biol247536540
- 79. Bairoch A, Apweiler R (2000) The SWISS-PROT protein sequence database and its supplement TrEMBL in 2000. Nucleic Acids Res 28: 45–48.A. BairochR. Apweiler2000The SWISS-PROT protein sequence database and its supplement TrEMBL in 2000.Nucleic Acids Res284548