Discovery of potent inhibitors of α-synuclein aggregation using structure-based iterative learning.

Machine learning methods hold the promise to reduce the costs and the failure rates of conventional drug discovery pipelines. This issue is especially pressing for neurodegenerative diseases, where the development of disease-modifying drugs has been particularly challenging. To address this problem, we describe here a machine learning approach to identify small molecule inhibitors of α-synuclein aggregation, a process implicated in Parkinson's disease and other synucleinopathies. Because the proliferation of α-synuclein aggregates takes place through autocatalytic secondary nucleation, we aim to identify compounds that bind the catalytic sites on the surface of the aggregates. To achieve this goal, we use structure-based machine learning in an iterative manner to first identify and then progressively optimize secondary nucleation inhibitors. Our results demonstrate that this approach leads to the facile identification of compounds two orders of magnitude more potent than previously reported ones.

Sequence-based prediction of pH-dependent protein solubility using CamSol.

Solubility is a property of central importance for the use of proteins in research in molecular and cell biology and in applications in biotechnology and medicine. Since experimental methods for measuring protein solubility are material intensive and time consuming, computational methods have recently emerged to enable the rapid and inexpensive screening of solubility for large libraries of proteins, as it is routinely required in development pipelines. Here, we describe the development of one such method to include in the predictions the effect of the pH on solubility. We illustrate the resulting pH-dependent predictions on a variety of antibodies and other proteins to demonstrate that these predictions achieve an accuracy comparable with that of experimental methods. We make this method publicly available at https://www-cohsoftware.ch.cam.ac.uk/index.php/camsolph, as the version 3.0 of CamSol.

An aggregation inhibitor specific to oligomeric intermediates of Aß42 derived from phage display libraries of stable, small proteins.

The self-assembly of amyloid ß peptide (Aß) to fibrillar and oligomeric aggregates is linked to Alzheimer's disease. Aß binders may serve as inhibitors of aggregation to prevent the generation of neurotoxic species and for the detection of Aß species. A particular challenge involves finding binders to on-pathway oligomers given their transient nature. Here we construct two phage­display libraries built on the highly inert and stable protein scaffold S100G, one containing a six-residue variable surface patch and one harboring a seven-residue variable loop insertion. Monomers and fibrils of Aß40 and Aß42 were separately coupled to silica nanoparticles, using a coupling strategy leading to the presence of oligomers on the monomer beads, and they were used in three rounds of affinity selection. Next-generation sequencing revealed sequence clusters and candidate binding proteins (SXkmers). Two SXkmers were expressed as soluble proteins and tested in terms of aggregation inhibition via thioflavin T fluorescence. We identified an SXkmer with loop­insertion YLTIRLM as an inhibitor of the secondary nucleation of Aß42 and binding analyses using surface plasmon resonance technology, Förster resonance energy transfer, and microfluidics diffusional sizing imply an interaction with intermediate oligomeric species. A linear peptide with the YLTIRLM sequence was found inhibitory but at a lower potency than the more constrained SXkmer loop. We identified an SXkmer with side-patch VI-WI-DD as an inhibitor of Aß40 aggregation. Remarkably, our data imply that SXkmer-YLTIRLM blocks secondary nucleation through an interaction with oligomeric intermediates in solution or at the fibril surface, which is a unique inhibitory mechanism for a library-derived inhibitor.

Correlation between the binding affinity and the conformational entropy of nanobody SARS-CoV-2 spike protein complexes.

Camelid single-domain antibodies, also known as nanobodies, can be readily isolated from naïve libraries for specific targets but often bind too weakly to their targets to be immediately useful. Laboratory-based genetic engineering methods to enhance their affinity, termed maturation, can deliver useful reagents for different areas of biology and potentially medicine. Using the receptor binding domain (RBD) of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) spike protein and a naïve library, we generated closely related nanobodies with micromolar to nanomolar binding affinities. By analyzing the structure-activity relationship using X-ray crystallography, cryoelectron microscopy, and biophysical methods, we observed that higher conformational entropy losses in the formation of the spike protein-nanobody complex are associated with tighter binding. To investigate this, we generated structural ensembles of the different complexes from electron microscopy maps and correlated the conformational fluctuations with binding affinity. This insight guided the engineering of a nanobody with improved affinity for the spike protein.

Multidimensional Protein Solubility Optimization with an Ultrahigh-Throughput Microfluidic Platform.

Protein-based biologics are highly suitable for drug development as they exhibit low toxicity and high specificity for their targets. However, for therapeutic applications, biologics must often be formulated to elevated concentrations, making insufficient solubility a critical bottleneck in the drug development pipeline. Here, we report an ultrahigh-throughput microfluidic platform for protein solubility screening. In comparison with previous methods, this microfluidic platform can make, incubate, and measure samples in a few minutes, uses just 20 µg of protein (>10-fold improvement), and yields 10,000 data points (1000-fold improvement). This allows quantitative comparison of formulation excipients, such as sodium chloride, polysorbate, histidine, arginine, and sucrose. Additionally, we can measure how solubility is affected by the combinatorial effect of multiple additives, find a suitable pH for the formulation, and measure the impact of mutations on solubility, thus enabling the screening of large libraries. By reducing material and time costs, this approach makes detailed multidimensional solubility optimization experiments possible, streamlining drug development and increasing our understanding of biotherapeutic solubility and the effects of excipients.

Proteome-wide observation of the phenomenon of life on the edge of solubility.

To function effectively proteins must avoid aberrant aggregation, and hence they are expected to be expressed at concentrations safely below their solubility limits. By analyzing proteome-wide mass spectrometry data of Caenorhabditis elegans, however, we show that the levels of about three-quarters of the nearly 4,000 proteins analyzed in adult animals are close to their intrinsic solubility limits, indeed exceeding them by about 10% on average. We next asked how aging and functional self-assembly influence these solubility limits. We found that despite the fact that the total quantity of proteins within the cellular environment remains approximately constant during aging, protein aggregation sharply increases between days 6 and 12 of adulthood, after the worms have reproduced, as individual proteins lose their stoichiometric balances and the cellular machinery that maintains solubility undergoes functional decline. These findings reveal that these proteins are highly prone to undergoing concentration-dependent phase separation, which on aging is rationalized in a decrease of their effective solubilities, in particular for proteins associated with translation, growth, reproduction, and the chaperone system.

Rational design of a conformation-specific antibody for the quantification of Aß oligomers.

Protein misfolding and aggregation is the hallmark of numerous human disorders, including Alzheimer's disease. This process involves the formation of transient and heterogeneous soluble oligomers, some of which are highly cytotoxic. A major challenge for the development of effective diagnostic and therapeutic tools is thus the detection and quantification of these elusive oligomers. Here, to address this problem, we develop a two-step rational design method for the discovery of oligomer-specific antibodies. The first step consists of an "antigen scanning" phase in which an initial panel of antibodies is designed to bind different epitopes covering the entire sequence of a target protein. This procedure enables the determination through in vitro assays of the regions exposed in the oligomers but not in the fibrillar deposits. The second step involves an "epitope mining" phase, in which a second panel of antibodies is designed to specifically target the regions identified during the scanning step. We illustrate this method in the case of the amyloid ß (Aß) peptide, whose oligomers are associated with Alzheimer's disease. Our results show that this approach enables the accurate detection and quantification of Aß oligomers in vitro, and in Caenorhabditis elegans and mouse hippocampal tissues.

π-Clamp-Mediated Homo- and Heterodimerization of Single-Domain Antibodies via Site-Specific Homobifunctional Conjugation.

Post-translational protein-protein conjugation produces bioconjugates that are unavailable via genetic fusion approaches. A method for preparing protein-protein conjugates using π-clamp-mediated cysteine arylation with pentafluorophenyl sulfonamide functional groups is described. Two computationally designed antibodies targeting the SARS-CoV-2 receptor binding domain were produced (KD = 146, 581 nM) with a π-clamp sequence near the C-terminus and dimerized using this method to provide a 10-60-fold increase in binding (KD = 8-15 nM). When two solvent-exposed cysteine residues were present on the second protein domain, the π-clamp cysteine residue was selectively modified over an Asp-Cys-Glu cysteine residue, allowing for subsequent small-molecule conjugation. With this strategy, we build molecule-protein-protein conjugates with complete chemical control over the sites of modification.

A chemical kinetic basis for measuring translation initiation and elongation rates from ribosome profiling data.

Analysis methods based on simulations and optimization have been previously developed to estimate relative translation rates from next-generation sequencing data. Translation involves molecules and chemical reactions, hence bioinformatics methods consistent with the laws of chemistry and physics are more likely to produce accurate results. Here, we derive simple equations based on chemical kinetic principles to measure the translation-initiation rate, transcriptome-wide elongation rate, and individual codon translation rates from ribosome profiling experiments. Our methods reproduce the known rates from ribosome profiles generated from detailed simulations of translation. By applying our methods to data from S. cerevisiae and mouse embryonic stem cells, we find that the extracted rates reproduce expected correlations with various molecular properties, and we also find that mouse embryonic stem cells have a global translation speed of 5.2 AA/s, in agreement with previous reports that used other approaches. Our analysis further reveals that a codon can exhibit up to 26-fold variability in its translation rate depending upon its context within a transcript. This broad distribution means that the average translation rate of a codon is not representative of the rate at which most instances of that codon are translated, and it suggests that translational regulation might be used by cells to a greater degree than previously thought.

MobiDB 3.0: more annotations for intrinsic disorder, conformational diversity and interactions in proteins.

The MobiDB (URL: mobidb.bio.unipd.it) database of protein disorder and mobility annotations has been significantly updated and upgraded since its last major renewal in 2014. Several curated datasets for intrinsic disorder and folding upon binding have been integrated from specialized databases. The indirect evidence has also been expanded to better capture information available in the PDB, such as high temperature residues in X-ray structures and overall conformational diversity. Novel nuclear magnetic resonance chemical shift data provides an additional experimental information layer on conformational dynamics. Predictions have been expanded to provide new types of annotation on backbone rigidity, secondary structure preference and disordered binding regions. MobiDB 3.0 contains information for the complete UniProt protein set and synchronization has been improved by covering all UniParc sequences. An advanced search function allows the creation of a wide array of custom-made datasets for download and further analysis. A large amount of information and cross-links to more specialized databases are intended to make MobiDB the central resource for the scientific community working on protein intrinsic disorder and mobility.

A natural product inhibits the initiation of α-synuclein aggregation and suppresses its toxicity.

The self-assembly of α-synuclein is closely associated with Parkinson's disease and related syndromes. We show that squalamine, a natural product with known anticancer and antiviral activity, dramatically affects α-synuclein aggregation in vitro and in vivo. We elucidate the mechanism of action of squalamine by investigating its interaction with lipid vesicles, which are known to stimulate nucleation, and find that this compound displaces α-synuclein from the surfaces of such vesicles, thereby blocking the first steps in its aggregation process. We also show that squalamine almost completely suppresses the toxicity of α-synuclein oligomers in human neuroblastoma cells by inhibiting their interactions with lipid membranes. We further examine the effects of squalamine in a Caenorhabditis elegans strain overexpressing α-synuclein, observing a dramatic reduction of α-synuclein aggregation and an almost complete elimination of muscle paralysis. These findings suggest that squalamine could be a means of therapeutic intervention in Parkinson's disease and related conditions.

Rationally Designed Antibodies as Research Tools to Study the Structure-Toxicity Relationship of Amyloid-ß Oligomers.

Alzheimer's disease is associated with the aggregation of the amyloid-ß peptide (Aß), resulting in the deposition of amyloid plaques in brain tissue. Recent scrutiny of the mechanisms by which Aß aggregates induce neuronal dysfunction has highlighted the importance of the Aß oligomers of this protein fragment. Because of the transient and heterogeneous nature of these oligomers, however, it has been challenging to investigate the detailed mechanisms by which these species exert cytotoxicity. To address this problem, we demonstrate here the use of rationally designed single-domain antibodies (DesAbs) to characterize the structure-toxicity relationship of Aß oligomers. For this purpose, we use Zn2+-stabilized oligomers of the 40-residue form of Aß (Aß40) as models of brain Aß oligomers and two single-domain antibodies (DesAb18-24 and DesAb34-40), designed to bind to epitopes at residues 18-24 and 34-40 of Aß40, respectively. We found that the DesAbs induce a change in structure of the Zn2+-stabilized Aß40 oligomers, generating a simultaneous increase in their size and solvent-exposed hydrophobicity. We then observed that these increments in both the size and hydrophobicity of the oligomers neutralize each other in terms of their effects on cytotoxicity, as predicted by a recently proposed general structure-toxicity relationship, and observed experimentally. These results illustrate the use of the DesAbs as research tools to investigate the biophysical and cytotoxicity properties of Aß oligomers.

Targeting disordered proteins with small molecules using entropy.

The human proteome includes many disordered proteins. Although these proteins are closely linked with a range of human diseases, no clinically approved drug targets them in their monomeric forms. This situation arises, at least in part, from the current lack of understanding of the mechanisms by which small molecules bind proteins that do not fold into well-defined conformations. To explore possible solutions to this problem, we discuss quite generally how an overall decrease in the free energy associated with intermolecular binding can originate from different combinations of enthalpic and entropic contributions. We then consider more specifically a mechanism of binding by which small molecules can affect the conformational space of a disordered protein by creating an entropic expansion in which more conformations of the protein become populated.

Parapred: antibody paratope prediction using convolutional and recurrent neural networks.

Motivation: Antibodies play essential roles in the immune system of vertebrates and are powerful tools in research and diagnostics. While hypervariable regions of antibodies, which are responsible for binding, can be readily identified from their amino acid sequence, it remains challenging to accurately pinpoint which amino acids will be in contact with the antigen (the paratope). Results: In this work, we present a sequence-based probabilistic machine learning algorithm for paratope prediction, named Parapred. Parapred uses a deep-learning architecture to leverage features from both local residue neighbourhoods and across the entire sequence. The method significantly improves on the current state-of-the-art methodology, and only requires a stretch of amino acid sequence corresponding to a hypervariable region as an input, without any information about the antigen. We further show that our predictions can be used to improve both speed and accuracy of a rigid docking algorithm. Availability and implementation: The Parapred method is freely available as a webserver at http://www-mvsoftware.ch.cam.ac.uk/and for download at https://github.com/eliberis/parapred. Supplementary information: Supplementary information is available at Bioinformatics online.

Third generation antibody discovery methods: in silico rational design.

Owing to their outstanding performances in molecular recognition, antibodies are extensively used in research and applications in molecular biology, biotechnology and medicine. Recent advances in experimental and computational methods are making it possible to complement well-established in vivo (first generation) and in vitro (second generation) methods of antibody discovery with novel in silico (third generation) approaches. Here we describe the principles of computational antibody design and review the state of the art in this field. We then present Modular, a method that implements the rational design of antibodies in a modular manner, and describe the opportunities offered by this approach.

Rational design of antibodies targeting specific epitopes within intrinsically disordered proteins.

Antibodies are powerful tools in life sciences research, as well as in diagnostic and therapeutic applications, because of their ability to bind given molecules with high affinity and specificity. Using current methods, however, it is laborious and sometimes difficult to generate antibodies to target specific epitopes within a protein, in particular if these epitopes are not effective antigens. Here we present a method to rationally design antibodies to enable them to bind virtually any chosen disordered epitope in a protein. The procedure consists in the sequence-based design of one or more complementary peptides targeting a selected disordered epitope and the subsequent grafting of such peptides on an antibody scaffold. We illustrate the method by designing six single-domain antibodies to bind different epitopes within three disease-related intrinsically disordered proteins and peptides (α-synuclein, Aß42, and IAPP). Our results show that all these designed antibodies bind their targets with good affinity and specificity. As an example of an application, we show that one of these antibodies inhibits the aggregation of α-synuclein at substoichiometric concentrations and that binding occurs at the selected epitope. Taken together, these results indicate that the design strategy that we propose makes it possible to obtain antibodies targeting given epitopes in disordered proteins or protein regions.

A Rationally Designed Hsp70 Variant Rescues the Aggregation-Associated Toxicity of Human IAPP in Cultured Pancreatic Islet ß-Cells.

Molecular chaperones are key components of the protein homeostasis system against protein misfolding and aggregation. It has been recently shown that these molecules can be rationally modified to have an enhanced activity against specific amyloidogenic substrates. The resulting molecular chaperone variants can be effective inhibitors of protein aggregation in vitro, thus suggesting that they may provide novel opportunities in biomedical and biotechnological applications. Before such opportunities can be exploited, however, their effects on cell viability should be better characterised. Here, we employ a rational design method to specifically enhance the activity of the 70-kDa heat shock protein (Hsp70) against the aggregation of the human islet amyloid polypeptide (hIAPP, also known as amylin). We then show that the Hsp70 variant that we designed (grafted heat shock protein 70 kDa-human islet amyloid polypeptide, GHsp70-hIAPP) is significantly more effective than the wild type in recovering the viability of cultured pancreatic islet β-cells RIN-m5F upon hIAPP aggregation. These results indicate that a full recovery of the toxic effects of hIAPP aggregates on cultured pancreatic cells can be achieved by increasing the specificity and activity of Hsp70 towards hIAPP, thus providing evidence that the strategy presented here provides a possible route for rationally tailoring molecular chaperones for enhancing their effects in a target-dependent manner.

Understanding the frustration arising from the competition between function, misfolding, and aggregation in a globular protein.

Folding and function may impose different requirements on the amino acid sequences of proteins, thus potentially giving rise to conflict. Such a conflict, or frustration, can result in the formation of partially misfolded intermediates that can compromise folding and promote aggregation. We investigate this phenomenon by studying frataxin, a protein whose normal function is to facilitate the formation of iron-sulfur clusters but whose mutations are associated with Friedreich's ataxia. To characterize the folding pathway of this protein we carry out a Φ-value analysis and use the resulting structural information to determine the structure of the folding transition state, which we then validate by a second round of rationally designed mutagenesis. The analysis of the transition-state structure reveals that the regions involved in the folding process are highly aggregation-prone. By contrast, the regions that are functionally important are partially misfolded in the transition state but highly resistant to aggregation. Taken together, these results indicate that in frataxin the competition between folding and function creates the possibility of misfolding, and that to prevent aggregation the amino acid sequence of this protein is optimized to be highly resistant to aggregation in the regions involved in misfolding.

Oxetane Grafts Installed Site-Selectively on Native Disulfides to Enhance Protein Stability and Activity In†Vivo.

A four-membered oxygen ring (oxetane) can be readily grafted into native peptides and proteins through site-selective bis-alkylation of cysteine residues present as disulfides under mild and biocompatible conditions. The selective installation of the oxetane graft enhances stability and activity, as demonstrated for a range of biologically relevant cyclic peptides, including somatostatin, proteins, and antibodies, such as a Fab arm of the antibody Herceptin and a designed antibody DesAb-Aß against the human Amyloid-ß peptide. Oxetane grafting of the genetically detoxified diphtheria toxin CRM197 improves significantly the immunogenicity of this protein in mice, which illustrates the general utility of this strategy to modulate the stability and biological activity of therapeutic proteins containing disulfides in their structures.

A Rational Design Strategy for the Selective Activity Enhancement of a Molecular Chaperone toward a Target Substrate.

Molecular chaperones facilitate the folding and assembly of proteins and inhibit their aberrant aggregation. They thus offer several opportunities for biomedical and biotechnological applications, as for example they can often prevent protein aggregation more effectively than other therapeutic molecules, including small molecules and antibodies. Here we present a method of designing molecular chaperones with enhanced activity against specific amyloidogenic substrates while leaving unaltered their functions toward other substrates. The method consists of grafting onto a molecular chaperone a peptide designed to bind specifically an epitope in the target substrate. We illustrate this strategy by describing Hsp70 variants with increased affinities for α-synuclein and Aß42 but otherwise unaltered affinities for other substrates. These designed variants inhibit protein aggregation and disaggregate preformed fibrils significantly more effectively than wild-type Hsp70 indicating that the strategy presented here provides a possible route for tailoring rationally molecular chaperones for specific purposes.