The Route from the Folded to the Amyloid State: Exploring the Potential Energy Surface of a Drug-Like Miniprotein

Opening Amyloid-Windows to the secondary structure of proteins: The amyloidogenecity increases tenfold inside beta-sheets

Methods from artificial intelligence (AI), in general, and machine learning, in particular, have kept conquering new territories in numerous areas of science. Most of the applications of these techniques are restricted to the classification of large data sets, but new scientific knowledge can seldom be inferred from these tools. Here we show that an AI-based amyloidogenecity predictor can strongly differentiate the border- and the internal hexamers of β-pleated sheets when screening all the Protein Data Bank-deposited homology-filtered protein structures. Our main result shows that more than 30% of internal hexamers of β sheets are predicted to be amyloidogenic, while just outside the border regions, only 3% are predicted as such. This result may elucidate a general protection mechanism of proteins against turning into amyloids: if the borders of β-sheets were amyloidogenic, then the whole β sheet could turn more easily into an insoluble amyloid-structure, characterized by periodically repeated parallel β-sheets. We also present that no analogous phenomenon exists on the borders of α-helices or randomly chosen subsequences of the studied protein structures.

LoCoHD: a metric for comparing local environments of proteins

Protein folds and the local environments they create can be compared using a variety of differently designed measures, such as the root mean squared deviation, the global distance test, the template modeling score or the local distance difference test. Although these measures have proven to be useful for a variety of tasks, each fails to fully incorporate the valuable chemical information inherent to atoms and residues, and considers these only partially and indirectly. Here, we develop the highly flexible local composition Hellinger distance (LoCoHD) metric, which is based on the chemical composition of local residue environments. Using LoCoHD, we analyze the chemical heterogeneity of amino acid environments and identify valines having the most conserved-, and arginines having the most variable chemical environments. We use LoCoHD to investigate structural ensembles, to evaluate critical assessment of structure prediction (CASP) competitors, to compare the results with the local distance difference test (lDDT) scoring system, and to evaluate a molecular dynamics simulation. We show that LoCoHD measurements provide unique information about protein structures that is distinct from, for example, those derived using the alignment-based RMSD metric, or the similarly distance matrix-based but alignment-free lDDT metric.

Mechanisms of Biomolecular Self-Assembly Investigated Through In Situ Observations of Structures and Dynamics

Biomolecular self-assembly of hierarchical materials is a precise and adaptable bottom-up approach to synthesizing across scales with considerable energy, health, environment, sustainability, and information technology applications. To achieve desired functions in biomaterials, it is essential to directly observe assembly dynamics and structural evolutions that reflect the underlying energy landscape and the assembly mechanism. This review will summarize the current understanding of biomolecular assembly mechanisms based on in situ characterization and discuss the broader significance and achievements of newly gained insights. In addition, we will also introduce how emerging deep learning/machine learning-based approaches, multiparametric characterization, and high-throughput methods can boost the development of biomolecular self-assembly. The objective of this review is to accelerate the development of in situ characterization approaches for biomolecular self-assembly and to inspire the next generation of biomimetic materials.

Polymorphic amyloid nanostructures of hormone peptides involved in glucose homeostasis display reversible amyloid formation

A large group of hormones are stored as amyloid fibrils in acidic secretion vesicles before they are released into the bloodstream and readopt their functional state. Here, we identify an evolutionarily conserved hexapeptide sequence as the major aggregation-prone region (APR) of gastrointestinal peptides of the glucagon family: xFxxWL. We determine nine polymorphic crystal structures of the APR segments of glucagon-like peptides 1 and 2, and exendin and its derivatives. We follow amyloid formation by CD, FTIR, ThT assays, and AFM. We propose that the pH-dependent changes of the protonation states of glutamate/aspartate residues of APRs initiate switching between the amyloid and the folded, monomeric forms of the hormones. We find that pH sensitivity diminishes in the absence of acidic gatekeepers and amyloid formation progresses over a broad pH range. Our results highlight the dual role of short aggregation core motifs in reversible amyloid formation and receptor binding.

Inhibitor Design Strategy for Myostatin: Dynamics and Interaction Networks Define the Affinity and Release Mechanisms of the Inhibited Complexes

Myostatin, an important negative regulator of muscle mass, is a therapeutic target for muscle atrophic disorders such as muscular dystrophy. Thus, the inhibition of myostatin presents a strategy to treat these disorders. It has long been established that the myostatin prodomain is a strong inhibitor of the mature myostatin, and the minimum peptide of the prodomain-corresponding to the α1-helix of its lasso-region-responsible for the inhibitory efficiency was defined and characterized as well. Here we show that the minimum peptide segment based on the growth differentiation factor 11 (GDF11), which we found to be more helical in its stand-alone solvated stfate than the similar segment of myostatin, is a promising new base scaffold for inhibitor design. The proposed inhibitory peptides in their solvated state and in complex with the mature myostatin were analyzed by in silico molecule modeling supplemented with the electronic circular dichroism spectroscopy measurements. We defined the Gaussian-Mahalanobis mean score to measure the fraction of dihedral angle-pairs close to the desired helical region of the Ramachandran-plot, carried out RING analysis of the peptide-protein interaction networks and characterized the internal motions of the complexes using our rigid-body segmentation protocol. We identified a variant-11m2-that is sufficiently ordered both in solvent and within the inhibitory complex, forms a high number of contacts with the binding-pocket and induces such changes in its internal dynamics that lead to a rigidified, permanently locked conformation that traps this peptide in the binding site. We also showed that the naturally evolved α1-helix has been optimized to simultaneously fulfill two very different roles: to function as a strong binder as well as a good leaving group. It forms an outstanding number of non-covalent interactions with the mature core of myostatin and maintains the most ordered conformation within the complex, while it induces independent movement of the gate-keeper β-hairpin segment assisting the dissociation and also results in the least-ordered solvated form which provides extra stability for the dissociated state and discourages rebinding.

Comparative Study of the Solid-Liquid Interfacial Adsorption of Proteins in Their Native and Amyloid Forms

The adhesive properties of amyloid fibers are thought to play a crucial role in various negative and positive aggregation processes, the study of which might help in their understanding and control. Amyloids have been prepared from two proteins, lysozyme and β-lactoglobulin, as well as an Exendin-4 derivative miniprotein (E5). Thermal treatment was applied to form amyloids and their structure was verified by thioflavin T (ThT), 8-Anilino-1-naphthalenesulfonic acid (ANS) dye tests and electronic circular dichroism spectroscopy (ECD). Adsorption properties of the native and amyloid forms of the three proteins were investigated and compared using the mass-sensitive quartz crystal microbalance (QCM) technique. Due to the possible electrostatic and hydrophobic interactions, similar adsorbed amounts were found for the native or amyloid forms, while the structures of the adsorbed layers differed significantly. Native proteins formed smooth and dense adsorption layers. On the contrary, a viscoelastic, highly loose layer was formed in the presence of the amyloid forms, shown by increased motional resistance values determined by the QCM technique and also indicated by atomic force microscopy (AFM) and wettability measurements. The elongated structure and increased hydrophobicity of amyloids might contribute to this kind of aggregation.

Succinct Amyloid and Nonamyloid Patterns in Hexapeptides

Hexapeptides are widely applied as a model system for studying the amyloid-forming properties of polypeptides, including proteins. Recently, large experimental databases have become publicly available with amyloidogenic labels. Using these data sets for training and testing purposes, one may build artificial intelligence (AI)-based classifiers for predicting the amyloid state of peptides. In our previous work (Biomolecules 2021, 11, 500), we described the Support Vector Machine (SVM)-based Budapest Amyloid Predictor (https://pitgroup.org/bap). Here, we apply the Budapest Amyloid Predictor for discovering numerous amyloidogenic and nonamyloidogenic hexapeptide patterns with accuracy between 80% and 84%, as surprising and succinct novel rules for further understanding the amyloid state of peptides. For example, we have shown that for any independently mutated residue (position marked by "x"), the patterns CxFLWx, FxFLFx, or xxIVIV are predicted to be amyloidogenic, while those of PxDxxx, xxKxEx, and xxPQxx are nonamyloidogenic. We note that each amyloidogenic pattern with two x's (e.g.,CxFLWx) describes succinctly 20² = 400 hexapeptides, while the nonamyloidogenic patterns comprising four point mutations (e.g.,PxDxxx) give 20⁴ = 160 000 hexapeptides in total. We also examine the restricted substitutions for positions "x" from subclasses of proteinogenic amino acid residues; for example, if "x" is substituted with hydrophobic amino acids, then there exist patterns containing three x's, like MxVVxx, predicted to be amyloidogenic. If we can choose for the x positions any hydrophobic amino acids, except the "structure breaker" proline, then we get amyloid patterns with five x positions, for example, xxxFxx, each corresponding to 32 768 hexapeptides. To our knowledge, no similar applications of artificial intelligence tools or succinct amyloid patterns were described before the present work.

Dissection of the amyloid formation pathway in AL amyloidosis

In antibody light chain (AL) amyloidosis, overproduced light chain (LC) fragments accumulate as fibrils in organs and tissues of patients. In vitro, AL fibril formation is a slow process, characterized by a pronounced lag phase. The events occurring during this lag phase are largely unknown. We have dissected the lag phase of a patient-derived LC truncation and identified structural transitions that precede fibril formation. The process starts with partial unfolding of the V_L domain and the formation of small amounts of dimers. This is a prerequisite for the formation of an ensemble of oligomers, which are the precursors of fibrils. During oligomerization, the hydrophobic core of the LC domain rearranges which leads to changes in solvent accessibility and rigidity. Structural transitions from an anti-parallel to a parallel β-sheet secondary structure occur in the oligomers prior to amyloid formation. Together, our results reveal a rate-limiting multi-step mechanism of structural transitions prior to fibril formation in AL amyloidosis, which offers, in the long run, opportunities for therapeutic intervention.

AL amyloidosis is caused by the accumulation of overproduced light chain (LC) fragments as fibrils in patient organs and it is the most prevalent systemic amyloidosis. Here, the authors combine biochemical and biophysical experiments to characterise the lag phase of a patient-derived truncated LC and they identify structural transitions that precede fibril formation.

On the border of the amyloidogenic sequences: prefix analysis of the parallel beta sheets in the PDB_Amyloid collection

The Protein Data Bank (PDB) today contains more than 174,000 entries with the 3-dimensional structures of biological macromolecules. Using the rich resources of this repository, it is possible identifying subsets with specific, interesting properties for different applications. Our research group prepared an automatically updated list of amyloid- and probably amyloidogenic molecules, the PDB_Amyloid collection, which is freely available at the address http://pitgroup.org/amyloid. This resource applies exclusively the geometric properties of the steric structures for identifying amyloids. In the present contribution, we analyze the starting (i.e., prefix) subsequences of the characteristic, parallel beta-sheets of the structures in the PDB_Amyloid collection, and identify further appearances of these length-5 prefix subsequences in the whole PDB data set. We have identified this way numerous proteins, whose normal or irregular functions involve amyloid formation, structural misfolding, or anti-coagulant properties, simply by containing these prefixes: including the T-cell receptor (TCR), bound with the major histocompatibility complexes MHC-1 and MHC-2; the p53 tumor suppressor protein; a mycobacterial RNA polymerase transcription initialization complex; the human bridging integrator protein BIN-1; and the tick anti-coagulant peptide TAP.

Bacterial fermentation and isotope labelling optimized for amyloidogenic proteins

We developed a cost sensitive isotope labelling procedure using a fed-batch fermentation method and tested its efficiency producing the 15 N-, 13 C- and 15 N/13 C-labelled variants of an amyloidogenic miniprotein (E5: EEEAVRLYIQWLKEGGPSSGRPPPS). E5 is a surface active protein, which forms amyloids in solution. Here, we confirm, using both PM-IRRAS and AFM measurements, that the air-water interface triggers structural rearrangement and promotes the amyloid formation of E5, and thus it is a suitable test protein to work out efficient isotope labelling schemes even for such difficult sequences. E. coli cells expressing the recombinant, ubiquitin-fused miniprotein were grown in minimal media containing either unlabelled nutrients, or 15 N-NH4 Cl and/or 13 C-D-Glc. The consumption rates of NH4 Cl and D-Glc were quantitatively monitored during fermentation and their ratio was established to be 1:5 (for NH4 Cl: D-Glc). One- and two-step feeding schemes were custom-optimized to enhance isotope incorporation expressing five different E5 miniprotein variants. With the currently optimized protocols we could achieve a 1.5- to 5-fold increase of yields of several miniproteins coupled to a similar magnitude of cost reduction as compared to flask labelling protocols.

The Budapest Amyloid Predictor and Its Applications

The amyloid state of proteins is widely studied with relevance to neurology, biochemistry, and biotechnology. In contrast with nearly amorphous aggregation, the amyloid state has a well-defined structure, consisting of parallel and antiparallel β-sheets in a periodically repeated formation. The understanding of the amyloid state is growing with the development of novel molecular imaging tools, like cryogenic electron microscopy. Sequence-based amyloid predictors were developed, mainly using artificial neural networks (ANNs) as the underlying computational technique. From a good neural-network-based predictor, it is a very difficult task to identify the attributes of the input amino acid sequence, which imply the decision of the network. Here, we present a linear Support Vector Machine (SVM)-based predictor for hexapeptides with correctness higher than 84%, i.e., it is at least as good as the best published ANN-based tools. Unlike artificial neural networks, the decisions of the linear SVMs are much easier to analyze and, from a good predictor, we can infer rich biochemical knowledge. In the Budapest Amyloid Predictor webserver the user needs to input a hexapeptide, and the server outputs a prediction for the input plus the 6 × 19 = 114 distance-1 neighbors of the input hexapeptide.

Amyloid-Like Aggregation in Diseases and Biomaterials: Osmosis of Structural Information

The discovery that the polypeptide chain has a remarkable and intrinsic propensity to form amyloid-like aggregates endowed with an extraordinary stability is one of the most relevant breakthroughs of the last decades in both protein/peptide chemistry and structural biology. This observation has fundamental implications, as the formation of these assemblies is systematically associated with the insurgence of severe neurodegenerative diseases. Although the ability of proteins to form aggregates rich in cross-β structure has been highlighted by recent studies of structural biology, the determination of the underlying atomic models has required immense efforts and inventiveness. Interestingly, the progressive molecular and structural characterization of these assemblies has opened new perspectives in apparently unrelated fields. Indeed, the self-assembling through the cross-β structure has been exploited to generate innovative biomaterials endowed with promising mechanical and spectroscopic properties. Therefore, this structural motif has become the fil rouge connecting these diversified research areas. In the present review, we report a chronological recapitulation, also performing a survey of the structural content of the Protein Data Bank, of the milestones achieved over the years in the characterization of cross-β assemblies involved in the insurgence of neurodegenerative diseases. A particular emphasis is given to the very recent successful elucidation of amyloid-like aggregates characterized by remarkable molecular and structural complexities. We also review the state of the art of the structural characterization of cross-β based biomaterials by highlighting the benefits of the osmosis of information between these two research areas. Finally, we underline the new promising perspectives that recent successful characterizations of disease-related amyloid-like assemblies can open in the biomaterial field.

The Route from the Folded to the Amyloid State: Exploring the Potential Energy Surface of a Drug-Like Miniprotein

The amyloid formation of the folded segment of a variant of Exenatide (a marketed drug for type‐2 diabetes mellitus) was studied by electronic circular dichroism (ECD) and NMR spectroscopy. We found that the optimum temperature for E5 protein amyloidosis coincides with body temperature and requires well below physiological salt concentration. Decomposition of the ECD spectra and its barycentric representation on the folded‐unfolded‐amyloid potential energy surface allowed us to monitor the full range of molecular transformation of amyloidogenesis. We identified points of no return (e.g.; T=37 °C, pH 4.1, c_E5=250 μm, c_NaCl=50 mm, t>4–6 h) that will inevitably gravitate into the amyloid state. The strong B‐type far ultraviolet (FUV)‐ECD spectra and an unexpectedly strong near ultraviolet (NUV)‐ECD signal (Θ_{≈275–285  nm}) indicate that the amyloid phase of E5 is built from monomers of quasi‐elongated backbone structure (φ≈−145°, ψ≈+145°) with strong interstrand Tyr↔Trp interaction. Misfolded intermediates and the buildup of “toxic” early‐stage oligomers leading to self‐association were identified and monitored as a function of time. Results indicate that the amyloid transition is triggered by subtle misfolding of the α‐helix, exposing aromatic and hydrophobic side chains that may provide the first centers for an intermolecular reorganization. These initial clusters provide the spatial closeness and sufficient time for a transition to the β‐structured amyloid nucleus, thus the process follows a nucleated growth mechanism.