Immunoglobulin kappa light chain and its amyloidogenic mutants: A molecular dynamics study

Cu(II) binding to the λ6aJL2-R24G antibody light chain protein associated with light chain amyloidosis disease: The role of histidines

Light chain amyloidosis is a conformational disease caused by the abnormal proliferation and deposition of antibody light chains as amyloid fibers in organs and tissues. The effect of Cu(II) binding to the model recombinant protein 6aJL2-R24G was previously characterized in our group, and we found an acceleration of the aggregation kinetics of the protein. In this study, in order to confirm the Cu(II) binding sites, histidine variants of 6aJL2-R24G were prepared and the effects of their interaction with Cu(II) were analyzed by circular dichroism, fluorescence spectroscopy, isothermal calorimetry titrations, and molecular dynamics simulations. Confirming our earlier work, we found that His8 and His99 are the highest affinity Cu(II) binding sites, and that Cu(II) binding to both sites is a cooperative event.

Widespread amyloidogenicity potential of multiple myeloma patient-derived immunoglobulin light chains

BACKGROUND

In a range of human disorders such as multiple myeloma (MM), immunoglobulin light chains (IgLCs) can be produced at very high concentrations. This can lead to pathological aggregation and deposition of IgLCs in different tissues, which in turn leads to severe and potentially fatal organ damage. However, IgLCs can also be highly soluble and non-toxic. It is generally thought that the cause for this differential solubility behaviour is solely found within the IgLC amino acid sequences, and a variety of individual sequence-related biophysical properties (e.g. thermal stability, dimerisation) have been proposed in different studies as major determinants of the aggregation in vivo. Here, we investigate biophysical properties underlying IgLC amyloidogenicity.

RESULTS

We introduce a novel and systematic workflow, Thermodynamic and Aggregation Fingerprinting (ThAgg-Fip), for detailed biophysical characterisation, and apply it to nine different MM patient-derived IgLCs. Our set of pathogenic IgLCs spans the entire range of values in those parameters previously proposed to define in vivo amyloidogenicity; however, none actually forms amyloid in patients. Even more surprisingly, we were able to show that all our IgLCs are able to form amyloid fibrils readily in vitro under the influence of proteolytic cleavage by co-purified cathepsins.

CONCLUSIONS

We show that (I) in vivo aggregation behaviour is unlikely to be mechanistically linked to any single biophysical or biochemical parameter and (II) amyloidogenic potential is widespread in IgLC sequences and is not confined to those sequences that form amyloid fibrils in patients. Our findings suggest that protein sequence, environmental conditions and presence and action of proteases all determine the ability of light chains to form amyloid fibrils in patients.

Effect of Single Amino Acid Substitutions by Asn and Gln on Aggregation Properties of Bence-Jones Protein BIF

The nature of renal amyloidosis involving Bence-Jones proteins in multiple myeloma is still unclear. The development of amyloidosis in neurodegenerative diseases is often associated with a high content of asparagine and glutamine residues in proteins forming amyloid deposits. To estimate the influence of Asn and Gln residues on the aggregation of Bence-Jones protein BIF, we obtained recombinant BIF and its mutants with the substitution of Tyr187→Asn (Y187N) in α-helix of C_L domain, Lys170→Asn (K170N) and Ser157→Gln (S157Q) in C_L domain loops, Arg109→Asn in V_L-C_L linker (R109N) and Asp29→Gln in V_L domain loop (D29Q). The morphology of protein aggregates was studied at pH corresponding to the conditions in bloodstream (pH 7.2), distal (pH 6.5) and proximal renal tubules (pH 4.5) by atomic force microscopy (AFM) and small-angle X-ray scattering (SAXS). The Lys170→Asn replacement almost completely inhibits amyloidogenic activity. The Y187N forms fibril-like aggregates at all pH values. The Arg109→Asn replacement resulted in formation of fibril-like structures at pH 7.2 and 6.5 while the substitutions by Gln provoked formation of those structures only at pH 7.2. Therefore, the amyloidogenic properties are highly dependent on the location of Asn or Gln.

Structure and energetic basis of overrepresented λ light chain in systemic light chain amyloidosis patients

Amyloid formation and deposition of immunoglobulin light-chain proteins in systemic amyloidosis (AL) cause major organ failures. While the κ light-chain is dominant (λ/κ=1:2) in healthy individuals, λ is highly overrepresented (λ/κ=3:1) in AL patients. The structural basis of the amyloid formation and the sequence preference are unknown. We examined the correlation between sequence and structural stability of dimeric variable domains of immunoglobulin light chains using molecular dynamics simulations of 24 representative dimer interfaces, followed by energy evaluation of conformational ensembles for 20 AL patients' light chain sequences. We identified a stable interface with displaced N-terminal residues, provides the structural basis for AL protein fibrils formation. Proline isomerization may cause the N-terminus to adopt amyloid-prone conformations. We found that λ light-chains prefer misfolded dimer conformation, while κ chain structures are stabilized by a natively folded dimer. Our study may facilitate structure-based small molecule and antibody design to inhibit AL. This article is part of a Special Issue entitled: Accelerating Precision Medicine through Genetic and Genomic Big Data Analysis edited by Yudong Cai & Tao Huang.

Localized conformational changes trigger the pH-induced fibrillogenesis of an amyloidogenic λ light chain protein

Solvent conditions modulate the expression of the amyloidogenic potential of proteins. In this work the effect of pH on the fibrillogenic behavior and the conformational properties of 6aJL2, a model protein of the highly amyloidogenic variable light chain λ6a gene segment, was examined. Ordered aggregates showing the ultrastructural and spectroscopic properties observed in amyloid fibrils were formed in the 2.0-8.0 pH range. At pH <3.0 a drastic decrease in lag time and an increase in fibril formation rate were found. In the 4.0-8.0 pH range there was no spectroscopic evidence for significant conformational changes in the native state. Likewise, heat capacity measurements showed no evidence for residual structure in the unfolded state. However, at pH <3.0 stability is severely decreased and the protein suffers conformational changes as detected by circular dichroism, tryptophan and ANS fluorescence, as well as by NMR spectroscopy. Molecular dynamics simulations indicate that acid-induced conformational changes involve the exposure of the loop connecting strands E and F. These results are compatible with pH-induced changes in the NMR spectra. Overall, the results indicate that the mechanism involved in the acid-induced increase in the fibrillogenic potential of 6aJL2 is profoundly different to that observed in κ light chains, and is promoted by localized conformational changes in a region of the protein that was previously not known to be involved in acid-induced light chain fibril formation. The identification of this region opens the potential for the design of specific inhibitors.

Convergent mechanisms favor fast amyloid formation in two lambda 6a Ig light chain mutants

Extracellular deposition as amyloids of immunoglobulin light chains causes light chain amyloidosis. Among the light chain families, lambda 6a is one of the most frequent in light chain amyloidosis patients. Its germline protein, 6aJL2, and point mutants, R24G and P7S, are good models to study fibrillogenesis, because their stability and fibril formation characteristics have been described. Both mutations make the germline protein unstable and speed up its ability to aggregate. To date, there is no molecular mechanism that explains how these differences in amyloidogenesis can arise from a single mutation. To look into the structural and dynamical differences in the native state of these proteins, we carried out molecular dynamics simulations at room temperature. Despite the structural similarity of the germline protein and the mutants, we found differences in their dynamical signatures that explain the mutants' increased tendency to form amyloids. The contact network alterations caused by the mutations, though different, converge in affecting two anti-aggregation motifs present in light chain variable domains, suggesting a different starting point for aggregation in lambda chains compared to kappa chains.

Effect of single point mutations in a form of systemic amyloidosis

Amyloid deposits of light-chain proteins are associated with the most common form of systemic amyloidosis. We have studied the effects of single point mutations on amyloid formation of these proteins using explicit solvent model molecular dynamics simulations. For this purpose, we compare the stability of the wild-type immunoglobulin light-chain protein REI in its native and amyloid forms with that of four mutants: R61N, G68D, D82I, and A84T. We argue that the experimentally observed differences in the propensity for amyloid formation result from two effects. First, the mutant dimers have a lower stability than the wild-type dimer due to increase exposure of certain hydrophobic residues. The second effect is a shift in equilibrium between monomers with amyloid-like structure and such with native structures. Hence, when developing drugs against light-chain associated systemic amyloidosis, one should look for components that either stabilize the dimer by binding to the dimer interface or reduce for the monomers the probability of the amyloid form.

Site-directed mutagenesis reveals regions implicated in the stability and fiber formation of human λ3r light chains

Light chain amyloidosis (AL) is a disease that affects vital organs by the fibrillar aggregation of monoclonal light chains. λ3r germ line is significantly implicated in this disease. In this work, we contrasted the thermodynamic stability and aggregation propensity of 3mJL2 (nonamyloidogenic) and 3rJL2 (amyloidogenic) λ3 germ lines. Because of an inherent limitation (extremely low expression), Cys at position 34 of the 3r germ line was replaced by Tyr reaching a good expression yield. A second substitution (W91A) was introduced in 3r to obtain a better template to incorporate additional mutations. Although the single mutant (C34Y) was not fibrillogenic, the second mutation located at CDR3 (W91A) induced fibrillogenesis. We propose, for the first time, that CDR3 (position 91) affects the stability and fiber formation of human λ3r light chains. Using the double mutant (3rJL2/YA) as template, other variants were constructed to evaluate the importance of those substitutions into the stability and aggregation propensity of λ3 light chains. A change in position 7 (P7D) boosted 3rJL2/YA fibrillogenic properties. Modification of position 48 (I48M) partially reverted 3rJL2/YA fibril aggregation. Finally, changes at positions 8 (P8S) or 40 (P40S) completely reverted fibril formation. These results confirm the influential roles of N-terminal region (positions 7 and 8) and the loop 40-60 (positions 40 and 48) on AL. X-ray crystallography revealed that the three-dimensional topology of the single and double λ3r mutants was not significantly altered. This mutagenic approach helped to identify key regions implicated in λ3 AL.

A residue-specific shift in stability and amyloidogenicity of antibody variable domains

Variable (V) domains of antibodies are essential for antigen recognition by our adaptive immune system. However, some variants of the light chain V domains (VL) form pathogenic amyloid fibrils in patients. It is so far unclear which residues play a key role in governing these processes. Here, we show that the conserved residue 2 of VL domains is crucial for controlling its thermodynamic stability and fibril formation. Hydrophobic side chains at position 2 stabilize the domain, whereas charged residues destabilize and lead to amyloid fibril formation. NMR experiments identified several segments within the core of the VL domain to be affected by changes in residue 2. Furthermore, molecular dynamic simulations showed that hydrophobic side chains at position 2 remain buried in a hydrophobic pocket, and charged side chains show a high flexibility. This results in a predicted difference in the dissociation free energy of ∼10 kJ mol(-1), which is in excellent agreement with our experimental values. Interestingly, this switch point is found only in VL domains of the κ family and not in VLλ or in VH domains, despite a highly similar domain architecture. Our results reveal novel insight into the architecture of variable domains and the prerequisites for formation of amyloid fibrils. This might also contribute to the rational design of stable variable antibody domains.

Understanding the relevance of local conformational stability and dynamics to the aggregation propensity of an IgG1 and IgG2 monoclonal antibodies

Aggregation of monoclonal antibodies is often a multi-step process involving structural alterations in monomeric proteins and subsequent formation of soluble or insoluble oligomers. The role of local conformational stability and dynamics of native and/or partially altered structures in determining the aggregation propensity of monoclonal antibodies, however, is not well understood. Here, we investigate the role of conformational stability and dynamics of regions with distinct solvent exposure in determining the aggregation propensity of an IgG1 and IgG2 monoclonal antibody. The temperatures employed span the pre-unfolding range (10-40°C) and the onset temperatures (T onset ) for exposure of apolar residues (≈ 50°C), alterations in secondary structures (≈ 60°C) and initiation of visible aggregate formation (≈ 60°C). Solvent-exposed regions were found to precede solvent-shielded regions in an initiation of aggregation for both proteins. Such a process was observed upon alterations in overall tertiary structure while retaining the secondary structures in both the proteins. In addition, a greater dynamic nature of solvent-shielded regions in potential intermediates of IgG1 and the improved conformational stability increased its resistance to aggregation when compared to IgG2. These results suggest that local conformational stability and fluctuations of partially altered structures can influence the aggregation propensity of immunoglobulins.

Insights into the potential aggregation liabilities of the b12 Fab fragment via elevated temperature molecular dynamics

Aggregation is a common hurdle faced during the development of antibody therapeutics. In this study, we explore the potential aggregation liabilities of the Fab (fragment antigen-binding) from a human IgG1κ antibody via multiple elevated temperature molecular dynamic simulations, analogous to accelerated stability studies performed during formulation development. Deformation and solvent exposure changes in response to thermal stress were monitored for individual structural domains (V(H), V(L), C(H)1 and C(L)), their interfaces (V(H):V(L) and C(H)1:C(L)), edge beta-strands and sequence-predicted aggregation-prone regions (APRs). During simulations, domain interfaces deformed prior to the unfolding of individual domains. However, interfacial beta-strands retained their secondary structure and remained solvent protected longer than all other strands or loops. Thus, APRs located in interfacial beta-strands are effectively blocked from self-association. Structural deformations were also observed in complementarity-determining regions, edge beta-strands and adjoining framework beta-strands, which increased their solvent-accessible surface area and exposed APRs in these regions. From the analysis of these structural changes, two potential aggregation liabilities were identified in the V(H) domain of this Fab. Insights gained from this investigation should be useful in devising a rational structure-based strategy for the design and selection of antibody candidates with high potency and improved developability.

Current perspectives on cardiac amyloidosis

Amyloidosis represents a group of diseases in which proteins undergo misfolding to form insoluble fibrils with subsequent tissue deposition. While almost all deposited amyloid fibers share a common nonbranched morphology, the affected end organs, clinical presentation, treatment strategies, and prognosis vary greatly among this group of diseases and are largely dependent on the specific amyloid precursor protein. To date, at least 27 precursor proteins have been identified to result in either local tissue or systemic amyloidosis, with nine of them manifesting in cardiac deposition and resulting in a syndrome termed "cardiac amyloidosis" or "amyloid cardiomyopathy." Although cardiac amyloidosis has been traditionally considered to be a rare disorder, as clinical appreciation and understanding continues to grow, so too has the prevalence, suggesting that this disease may be greatly underdiagnosed. The most common form of cardiac amyloidosis is associated with circulating amyloidogenic monoclonal immunoglobulin light chain proteins. Other major cardiac amyloidoses result from a misfolding of products of mutated or wild-type transthyretin protein. While the various cardiac amyloidoses share a common functional consequence, namely, an infiltrative cardiomyopathy with restrictive pathophysiology leading to progressive heart failure, the underlying pathophysiology and clinical syndrome varies with each precursor protein. Herein, we aim to provide an up-to-date overview of cardiac amyloidosis from nomenclature to molecular mechanisms and treatment options, with a particular focus on amyloidogenic immunoglobulin light chain protein cardiac amyloidosis.

Thermodynamic and kinetic characterization of a germ line human lambda6 light-chain protein: the relation between unfolding and fibrillogenesis

Proteins encoded by the gene segment 6a of the lambda variable light-chain repertoire are strongly associated with amyloid deposition. 6aJL2 is a model protein constructed with the predicted sequences encoded by the 6a and JL2 germ line genes. In this work, we characterized the urea- and temperature-induced unfolding of 6aJL2. In the short time scale, spectroscopic, hydrodynamic and calorimetric experiments were compatible with a two-state transition. Furthermore, DeltaG, m and the midpoint urea concentration obtained from equilibrium experiments were compatible with those obtained from kinetic experiments. Since fibril formation is a slow process, samples were also incubated for longer times. After incubation for several hours at 37 degrees C, spectroscopic, hydrodynamic and calorimetric experiments revealed the presence of a partially unfolded off-pathway intermediate around the midpoint urea concentration (1.5-3.0 M urea). In vitro fibrillogenesis assays show that the maximum growth rate for fibril formation and the minimum lag time were obtained at urea concentrations where the partially unfolded state was populated (2.5 M urea at 37 degrees C). This indicates that this partially unfolded state is critical for in vitro fibril formation. Concentration-dependent kinetics and hydrodynamic properties of the intermediate were consistent with a soluble oligomeric state. The intermediate is formed around the midpoint urea concentration, where the native and unfolded states are equally populated and their rate of interconversion is the slowest. This situation may promote the slow accumulation of an intermediate state that is prone to aggregate.

Understanding structural/functional properties of immunoconjugates for cancer therapy by computational approaches

Monoclonal antibodies coupled to highly toxic molecules (immunoconjugates) are currently being developed for cancer therapy. We have used an in silico procedure for evaluating some physicochemical properties of two tumor-targeting anti-HER2 immunoconjugates: (a) the single-chain antibody scFv(FRP5) linked to a bacterial toxin, that has been recently progressed to phase I clinical trial in human cancer; (b) the putative molecule formed by the intrinsically stable scFv(800E6), which has been proposed as toxin carrier to cancer cells in human therapy, joined to the same toxin of (a). Structural models of the immunoconjugates have been built by homology modeling and assessed by molecular dynamics simulations. The trajectories have been analyzed to extract some biochemical properties and to assess the potential effects of the toxin on the structure and dynamics of the anti-HER2 antibodies. The results of the computational approach indicate that the antibodies maintain their correct folding even in presence of the toxin, whereas a certain stiffness in correspondence of some structural regions is observed. Furthermore, the toxin does not seem to affect the antibody solubility, whereas it enhances the structural stability. The proposed computational approach represent a promising tool for analyzing some physicochemical properties of immunoconjugates and for predicting the effects of the linked toxin on structure, dynamics, and functionality of the antibodies.

Structure and dynamics of the anti-AMCV scFv(F8): effects of selected mutations on the antigen combining site

The recombinant antibody fragment scFv(F8), which recognizes the coat protein of the plant virus AMCV, is characterized by peculiar high in vitro stability and functional folding even in reducing environments, making it fit for designing stable antibodies with desired properties. Mutagenesis and functional analysis evidenced two residues, at positions 47 and 58 of the V(H) chain, playing a crucial role in the antigen binding recognition. Here, we used a computational procedure to assess the effects of these mutations on the stability, structure and dynamics of the antigen-binding site. Structural models of the wild type scFv(F8) and of its H47 and H58 mutants were built by homology modelling and assessed by multiple 15.5ns of molecular dynamics simulations. Computational results indicate that the 47H substitution strongly affects the CDR-H(2) conformation, destabilizes the V(H)/V(L) interface and confers high conformational flexibility to the antigen-binding site, leading the mutant to functional loss. The mutation at position H58 strenghtens the binding site, bestowing a high antigen specificity on the mutant. The essential dynamics and the analysis of the protein-solvent interface further corroborate the correspondence between the extent of the structurally-determined flexibility of the binding site with the different functional behaviours proved by the wild-type and its mutants. These results may have useful implications for structure-based design of antibody combining site.

Insights into unbinding mechanisms upon two mutations investigated by molecular dynamics study of GSK3β-axin complex: Role of packing hydrophobic residues

Glycogen synthase kinase 3beta (GSK 3beta) is a key component of several cellular processes including Wnt and insulin signalling pathways. The interaction of GSK3beta with scaffolding peptide axin is thought to be responsible for the effective phosphorylation of beta-catenin, the core effector of Wnt signaling, which has been linked with the occurrence of colon cancer and melanoma. It has been demonstrated that the binding of axin to GSK3beta is abolished by the single-point mutation of Val267 to Gly (V267G) in GSK3beta or Leu392 to Pro (L392P) in axin. Molecular dynamics (MD) simulations were performed on wild type (WT), V267G mutant and L392P one to elucidate the two unbinding mechanisms that occur through different pathways. Besides, rough energy and residue-based energy decomposition were calculated by MM_GBSA (molecular mechanical Generalized_Born surface area) approach to illuminate the instability of the two mutants. The MD simulations of the two mutants and WT reveal that the structure of GSK3beta remains unchanged, while axin moves away from the interfacial hydrophobic pockets in both two mutants. Axin exhibits positional shift in V267G mutant, whereas, losing the hydrogen bonds that are indispensable for stabilizing the helix structure of wild type axin, the helix of axin is distorted in L392P mutant. To conclude, both two mutants destroy the hydrophobic interaction that is essential to the stability of GSK3beta-axin complex.

Where disease pathogenesis meets protein formulation: Renal deposition of immunoglobulin aggregates

Aggregation is one of the important issues encountered during the development of immunoglobulin-based drugs. The aim of the current review is to discuss the causes and consequences of immunoglobulin aggregation as well as the relevance of immunoglobulin aggregation to disease pathogenesis. Extracellular deposition of immunoglobulins, either monoclonal light chains or intact polyclonal antibodies, induces renal failure in various nephropathies. The aggregates can present fibrillar or amorphous structures. In this review, factors known to influence protein aggregation, such as the primary structure of the protein, local environment and glycosylation are assessed, as well as the subsequent altered clearance, fibril formation and toxicity. The role of the protein local environment is emphasized. Even if the local environment causes only minor perturbations in the protein structure, these perturbations might be sufficient to trigger aggregate formation. This fact underlines the importance of choosing appropriate formulations for protein drugs. If the formulation provides a slightly destabilizing environment to the protein, the long-term stability of the drug may be compromised by aggregate formation.

Molecular dynamics simulations and free energy analyses on the dimer formation of an amyloidogenic heptapeptide from human beta2-microglobulin: implication for the protofibril structure

Amyloid formation is associated with many neurodegenerative diseases. Recent findings suggest that early oligomeric aggregates could be major sources of toxicity. We present a computational investigation of the first step of amyloid initiation-dimer formation of a seven residue peptide (NHVTLSQ) from human beta2-microglobulin at pH 2.0, which renders +2.0 units charges to each peptide. A total of over 1.2 micros of simulations with explicit solvent and 1.0 micros of simulations with implicit solvent were conducted. Main-chain conformational restraint was applied to facilitate the formation of ordered dimers. An antiparallel beta-sheet with six main-chain hydrogen bonds was dominant in the implicit solvent simulations. In contrast, no stable dimers were observed in the two negative controls, the mouse heptapeptide (KHDSMAE, +3.0 units charges) and the scrambled human heptapeptide (QVLHTSN). Explicit solvent simulations presented a more complex scenario. The wild-type human heptapeptide formed predominantly antiparallel beta-sheets ( approximately 38%) although parallel ones ( approximately 12%) were also observed. Hydrophobic contacts preceded hydrogen bond saturation in the majority of the association events in the explicit solvent simulations, highlighting the important role of hydrophobic interaction in amyloid initiation. The fact that the mouse dimer dissociated immediately after the removal of conformational restraint suggests that the higher conformational entropy barrier, along with the stronger charge repulsion and weaker hydrophobic interaction, contributed to its inability to form amyloid fibril. The closeness of positive charge pairs in the dimers of the scrambled human heptapeptide may prohibit further beta-sheet extension and fibril growth. Combining the results from simulations and free energy analyses, we propose that the building block for this amyloid fibril is an antiparallel dimer with a two-residue register shift and six main-chain hydrogen bonds. A double-layer protofibril structure is also proposed in which two antiparallel beta-sheets face each other and are held together by hydrophobic staples and hydrogen bonds of the polar side-chains.

Simulation of pH-dependent edge strand rearrangement in human beta-2 microglobulin

Amyloid fibrils formed from unrelated proteins often share morphological similarities, suggesting common biophysical mechanisms for amyloidogenesis. Biochemical studies of human beta-2 microglobulin (beta2M) have shown that its transition from a water-soluble protein to insoluble aggregates can be triggered by low pH. Additionally, biophysical measurements of beta2M using NMR have identified residues of the protein that participate in the formation of amyloid fibrils. The crystal structure of monomeric human beta2M determined at pH 5.7 shows that one of its edge beta-strands (strand D) adopts a conformation that differs from other structures of the same protein obtained at higher pH. This alternate beta-strand arrangement lacks a beta-bulge, which may facilitate protein aggregation through intermolecular beta-sheet association. To explore whether the pH change may yield the observed conformational difference, molecular dynamics simulations of beta2M were performed. The effects of pH were modeled by specifying the protonation states of Asp, Glu, and His, as well as the C terminus of the main chain. The bulged conformation of strand D is preferred at medium pH (pH 5-7), whereas at low pH (pH < 4) the straight conformation is observed. Therefore, low pH may stabilize the straight conformation of edge strand D and thus increase the amyloidogenicity of beta2M.

Protein folding and unfolding simulations: a new challenge for data mining

One of the unsolved paradigms in molecular biology is the protein folding problem. In recent years, with the identification of several diseases as protein folding disorders and with the explosion of genome information and the need for efficient ways to predict protein structure, protein folding became a central issue in molecular sciences research. Using molecular dynamics unfolding simulations of an amyloidogenic protein--transthyretin--as an example, we put forward a series of ideas on how simulations of this type may be used to infer rules and unfolding behavior in amyloidogenic proteins, and to extrapolate rules for protein folding in different structural classes of proteins. These, in turn, could help in the development of protein structure prediction methods. The need to analyse different proteins and to run multiple simulations creates a huge amount of data which has to be stored, managed, analyzed and shared (database and Grid technology; data mining). Once the data is captured, the next challenge is to find meaningful patterns (associations, correlations, clusters, rules, relationships) among molecular properties, or their relative importance at different stages of the folding or unfolding processes. This clearly puts new and interesting challenges to the bioinformatics community.