Fragment Library Research Articles

Machine learning optimization of candidate antibody yields highly diverse sub-nanomolar affinity antibody libraries

Therapeutic antibodies are an important and rapidly growing drug modality. However, the design and discovery of early-stage antibody therapeutics remain a time and cost-intensive endeavor. Here we present an end-to-end Bayesian, language model-based method for designing large and diverse libraries of high-affinity single-chain variable fragments (scFvs) that are then empirically measured. In a head-to-head comparison with a directed evolution approach, we show that the best scFv generated from our method represents a 28.7-fold improvement in binding over the best scFv from the directed evolution. Additionally, 99% of designed scFvs in our most successful library are improvements over the initial candidate scFv. By comparing a library’s predicted success to actual measurements, we demonstrate our method’s ability to explore tradeoffs between library success and diversity. Results of our work highlight the significant impact machine learning models can have on scFv development. We expect our method to be broadly applicable and provide value to other protein engineering tasks.

Synergistic Inhibition Guided Fragment-Linking Strategy and Quantitative Structure-Property Relationship Modeling To Design Inhalable Therapeutics for Asthma Targeting CSF1R.

The colony-stimulating factor-1 receptor (CSF1R) is a tyrosine-protein kinase that is a potential target for asthma therapeutics. We have applied a fragment-lead combination approach to identify small fragments that act synergistically with GW2580, a known inhibitor of CSF1R. Two fragment libraries were screened in combination with GW2580 by surface plasmon resonance (SPR). Binding affinity measurements confirmed that thirteen fragments bind specifically to the CSF1R, and a kinase activity assay further validated the inhibitory effect of these fragments. Several fragment compounds enhanced the inhibitory activity of the lead inhibitor. Computational solvent mapping, molecular docking, and modeling studies suggest that some of these fragments bind adjacent to the binding site of the lead inhibitor and further stabilize the inhibitor-bound state. Modeling results guided the computational fragment-linking approach to design potential next-generation compounds. The inhalability of these proposed compounds was predicted using quantitative structure-property relationships (QSPR) modeling based on an analysis of 71 drugs currently on the market. This work provides new insights into the development of inhalable small molecule therapeutics for asthma.

Nucleophilic covalent ligand discovery for the cysteine redoxome.

With an eye toward expanding chemistries used for covalent ligand discovery, we elaborated an umpolung strategy that exploits the 'polarity reversal' of sulfur when cysteine is oxidized to sulfenic acid, a widespread post-translational modification, for selective bioconjugation with C-nucleophiles. Here we present a global map of a human sulfenome that is susceptible to covalent modification by members of a nucleophilic fragment library. More than 500 liganded sulfenic acids were identified on proteins across diverse functional classes, and, of these, more than 80% were not targeted by electrophilic fragment analogs. We further show that members of our nucleophilic fragment library can impair functional protein-protein interactions involved in nuclear oncoprotein transport and DNA damage repair. Our findings reveal a vast expanse of ligandable sulfenic acids in the human proteome and highlight the utility of nucleophilic small molecules in the fragment-based covalent ligand discovery pipeline, presaging further opportunities using non-traditional chemistries for targeting proteins.

Ultrafast Fragment Screening Using Photo-Hyperpolarized (CIDNP) NMR.

While nuclear magnetic resonance (NMR) is regarded as a reference in fragment-based drug design, its implementation in a high-throughput manner is limited by its lack of sensitivity resulting in long acquisition times and high micromolar sample concentrations. Several hyperpolarization approaches could, in principle, improve the sensitivity of NMR also in drug research. However, photochemically induced dynamic nuclear polarization (photo-CIDNP) is the only method that is directly applicable in aqueous solution and agile for scalable implementation using off-the-shelf hardware. With the use of photo-CIDNP, this work demonstrates the detection of weak binders in the millimolar affinity range using low micromolar concentrations down to 5 μM of ligand and 2 μM of target, thereby exploiting the photo-CIDNP-induced polarization twice: (i) increasing the signal-to-noise by one to two orders in magnitude and (ii) polarization-only of the free non-bound molecule allowing identification of binding by polarization quenching, yielding another factor of hundred in time when compared with standard techniques. The interaction detection was performed with single-scan NMR experiments of a duration of 2 to 5 s. Taking advantage of the readiness of photo-CIDNP setup implementation, an automated flow-through platform was designed to screen samples at a screening rate of 1500 samples per day. Furthermore, a 212 compounds photo-CIDNP fragment library is presented, opening an avenue toward a comprehensive fragment-based screening method.

Generation and characterization of avian single chain variable fragment against human Alpha-Enolase

Overexpression of human alpha-enolase (hEno1)has been reported in a wide range of cancers and is tightly associated with poor prognosis, making it a remarkable biomarker and therapeutic target. In this study, polyclonal yolk-immunoglobulin (IgY) antibodies purified from hEno1-immunized chickens showed a noticeable specific humoral response. Phage display technology was used to construct two antibody libraries of IgY gene-derived single-chain variable fragments (scFvs) containing 7.8 × 107 and 5.4 × 107 transformants, respectively. Phage-based ELISA indicated that specific anti-hEno1 clones were significantly enriched. The nucleotide sequences of scFv-expressing clones were determined and classified into seven groups either in the short linker or the long linker. Moreover, higher mutation rates were revealed in the CDR regions, especially in the CDR3. Three distinguish antigenic epitopes were identified on the hEno1 protein. The binding activities of selected anti-hEno1 scFv on hEno1-positive PE089 lung cancer cells were confirmed using Western blot, flow cytometry, and immunofluorescence assay. In particular, hEnS7 and hEnS8 scFv antibodies significantly suppressed the growth and migration of PE089 cells. Taken together, these chicken-derived anti-hEno1 IgY and scFv antibodies have great potential to develop diagnostic and therapeutic agents for the treatment of lung cancer patients with high expression levels of hEno1 protein.

Identification of 4-hydroxyphenylpyruvate dioxygenase inhibitors by virtual screening, molecular docking, molecular dynamic simulation.

4-Hydroxyphenylpyruvate dioxygenase (HPPD) herbicides control broadleaf and gramineous weeds with better crop safety for corn, sorghum and wheat. Multiple screening models in silico have been established to obtain novel lead compounds as HPPD inhibition herbicides. Topomer comparative molecular field analysis (CoMFA) combined with topomer search technology and Bayesian, genetic approximation functions (GFA) and multiple linear regression (MLR) models generated by calculating different descriptors were constructed for the quinazolindione derivatives of HPPD inhibitors. The coefficient of determination (r2 ) of topomer CoMFA, MLR and GFA were 0.975, 0.970 and 0.968, respectively; all the models established displayed excellent accuracy and high predictive capacity. Five compounds with potential HPPD inhibition were obtained via screening fragment library combined with the validation of the above models and molecular docking studies. After molecular dynamics (MD) validation and absorption, distribution, metabolism, excretion and toxicity (ADMET) prediction, the compound 2-(2-amino-4-(4H-1,2,4-triazol-4-yl) benzoyl)-3-hydroxycyclohex-2-en-1-one not only exhibited stable interactions with the protein but also high solubility and low toxicity, and has potential as a novel HPPD inhibition herbicide. In this study, five compounds were obtained through multiple quantitative structure-activity relationship screening. Molecular docking and MD experiments showed that the constructed approach had good screening ability for HPPD inhibitors. This work provided molecular structural information for developing novel, highly efficient and low-toxicity HPPD inhibitors. © 2023 Society of Chemical Industry.

An Automated, Open-Source Workflow for the Generation of (3D) Fragment Libraries.

The recent success of fragment-based drug discovery (FBDD) is inextricably linked to adequate library design. To guide the design of our fragment libraries, we have constructed an automated workflow in the open-source KNIME software. The workflow considers chemical diversity and novelty of the fragments, and can also take into account the three-dimensional (3D) character. This design tool can be used to create large and diverse libraries but also to select a small number of representative compounds as a focused set of unique screening compounds to enrich existing fragment libraries. To illustrate the procedures, the design and synthesis of a 10-membered focused library is reported based on the cyclopropane scaffold, which is underrepresented in our existing fragment screening library. Analysis of the focused compound set indicates significant shape diversity and a favorable overall physicochemical profile. By virtue of its modular setup, the workflow can be readily adjusted to design libraries that focus on properties other than 3D shape.

Fragtory: Pharmacophore-Focused Design, Synthesis, and Evaluation of an sp3-Enriched Fragment Library.

Fragment-based drug discovery has played an important role in medicinal chemistry and pharmaceutical research. Despite numerous demonstrated successes, the limited diversity and overrepresentation of planar, sp2-rich structures in commercial libraries often hamper the full potential of this approach. Hence, the thorough design of screening libraries inevitably determines the probability for meaningful hits and subsequent structural elaboration. Against this background, we present the generation of an exclusive fragment library based on iterative entry nomination by a specifically designed computational workflow: "Fragtory". Following a pharmacophore diversity-driven approach, we used Fragtory in an interdisciplinary academic setting to guide both tailored synthesis efforts and the implementation of in-house compounds to build a curated 288-member library of sp3-enriched fragments. Subsequent NMR screens against a model protein and hit validation by protein crystallography led to the identification of structurally novel ligands that were further characterized by isothermal titration calorimetry, demonstrating the applicability of our experimental approach.

Integrated Covalent Drug Design Workflow Using Site Identification by Ligand Competitive Saturation.

Covalent drug design is an important component in drug discovery. Traditional drugs interact with their target in a reversible equilibrium, while irreversible covalent drugs increase the drug-target interaction duration by forming a covalent bond with targeted residues and thus may offer a more effective therapeutic approach. To facilitate the design of this class of ligands, computational methods can be used to help identify reactive nucleophilic residues, frequently cysteines, on a target protein for covalent binding, to test various warhead groups for their potential reactivities, and to predict noncovalent contributions to binding that can facilitate drug-target interactions that are important for binding specificity. To further aid covalent drug design, we extended a functional group mapping approach based on explicit solvent all-atom molecular simulations (SILCS: site identification by ligand competitive saturation) that intrinsically considers protein flexibility, functional group, and protein desolvation along with functional group-protein interactions. Through docking of a library of representative warhead fragments using SILCS-Monte Carlo (SILCS-MC), reactive cysteines can be correctly identified for proteins being tested. Furthermore, a machine learning model was trained to quantify the effectiveness of various warhead groups for proteins using metrics from SILCS-MC as well as experimental model compound warhead reactivity data. The ability to rank covalent molecular binders with similar warheads using SILCS ligand grid free energy (LGFE) ranking was also tested for several proteins. Based on these tools, an integrated SILCS-based workflow was developed, named SILCS-Covalent, which can both qualitatively and quantitatively inform covalent drug discovery.

Photoselective sequencing: microscopically guided genomic measurements with subcellular resolution.

In biological systems, spatial organization and function are interconnected. Here we present photoselective sequencing, a new method for genomic and epigenomic profiling within morphologically distinct regions. Starting with an intact biological specimen, photoselective sequencing uses targeted illumination to selectively unblock a photocaged fragment library, restricting the sequencing-based readout to microscopically identified spatial regions. We validate photoselective sequencing by measuring the chromatin accessibility profiles of fluorescently labeled cell types within the mouse brain and comparing with published data. Furthermore, by combining photoselective sequencing with a computational strategy for decomposing bulk accessibility profiles, we find that the oligodendrocyte-lineage-cell population is relatively enriched for oligodendrocyte-progenitor cells in the cortex versus the corpus callosum. Finally, we leverage photoselective sequencing at the subcellular scale to identify features of chromatin that are correlated with positioning at the nuclear periphery. These results collectively demonstrate that photoselective sequencing is a flexible and generalizable platform for exploring the interplay of spatial structures with genomic and epigenomic properties.

Efficient Ligand Discovery Using Sulfur(VI) Fluoride Reactive Fragments.

Sulfur(VI) fluorides (SFs) have emerged as valuable electrophiles for the design of "beyond-cysteine" covalent inhibitors and offer potential for expansion of the liganded proteome. Since SFs target a broad range of nucleophilic amino acids, they deliver an approach for the covalent modification of proteins without requirement for a proximal cysteine residue. Further to this, libraries of reactive fragments present an innovative approach for the discovery of ligands and tools for proteins of interest by leveraging a breadth of mass spectrometry analytical approaches. Herein, we report a screening approach that exploits the unique properties of SFs for this purpose. Libraries of SF-containing reactive fragments were synthesized, and a direct-to-biology workflow was taken to efficiently identify hit compounds for CAII and BCL6. The most promising hits were further characterized to establish the site(s) of covalent modification, modification kinetics, and target engagement in cells. Crystallography was used to gain a detailed molecular understanding of how these reactive fragments bind to their target. It is anticipated that this screening protocol can be used for the accelerated discovery of "beyond-cysteine" covalent inhibitors.

Abstract 3429: Beyond cereblon IMIDs - biophysics-based discovery of novel molecular glue chemotypes

Abstract Molecular glue degraders are compact, low molecular weight molecules that can efficiently induce specific and potent degradation of a target protein. This class of degraders function by inducing interactions between a target of interest and a ubiquitin-ligase, either by stabilization of weak pre-existing interactions, or by generation of entirely novel interactions. These molecules offer significant opportunity beyond heterobifunctional degraders such as PROTACs, not least in terms of improved molecular properties. However, beyond the IMID molecular glues, typified by thalidomide, pomalidomide and lenalidomide, novel molecular glue chemotypes remain scarce. To address this need, we have developed biophysics-based molecular glue screening platform, exploiting our internal, high quality fragment library and proximity-based screening platforms to rapidly identify promising new molecular glues for further optimization. A potential advantage of utilizing cell-free biophysical systems is the opportunity to select both the target and the desired ligase, opening up for development of degraders that capitalize upon differential expression of ligases in different tissues. As proof of concept, we have applied this platform to find new molecular glues to degrade CK1α. This Ser/Thr kinase has been found to be over-expressed in metastatic colorectal cancer, and this over-expression correlates with poor overall survival. The kinase has also been implicated as an oncogenic driver in tumors such as B-Cell lymphomas and non-Hodgkin lymphomas, suggesting a potential therapeutic application for novel CK1α molecular glues. Utilizing known IMID-derived molecular glues between CK1α and CRBN as benchmark controls, we identified several non-IMID derived chemotypes as tentative stabilizers of the CRBN/CK1α interaction. Further studies on these novel candidate degrader templates are now underway. Citation Format: Yujia Zhang, Jessica Bates, Benoit Gourdet, Louise Birch, Philip Addis, Roland Hjerpe, Allan M. Jordan. Beyond cereblon IMIDs - biophysics-based discovery of novel molecular glue chemotypes [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 3429.

Going beyond Binary: Rapid Identification of Protein-Protein Interaction Modulators Using a Multifragment Kinetic Target-Guided Synthesis Approach.

Kinetic target-guided synthesis (KTGS) is a powerful screening approach that enables identification of small molecule modulators for biomolecules. While many KTGS variants have emerged, a majority of the examples suffer from limited throughput and a poor signal/noise ratio, hampering reliable hit detection. Herein, we present our optimized multifragment KTGS screening strategy that tackles these limitations. This approach utilizes selected reaction monitoring liquid chromatography tandem mass spectrometry for hit detection, enabling the incubation of 190 fragment combinations per screening well. Consequentially, our fragment library was expanded from 81 possible combinations to 1710, representing the largest KTGS screening library assembled to date. The expanded library was screened against Mcl-1, leading to the discovery of 24 inhibitors. This work unveils the true potential of KTGS with respect to the rapid and reliable identification of hits, further highlighting its utility as a complement to the existing repertoire of screening methods used in drug discovery.

Identification of the taxonomic affiliation of Cryptosporidium spp. in piglets in the conditions of the north-west of the Russian Federation using molecular genetic methods

The purpose of the research is to determine taxa of the genus Cryptosporidium species in pigs using molecular genetic methods in the north-west of the Russian Federation.Materials and methods. Fecal samples were taken from pigs of different age groups on farms of different types of incorporation that differ in climatic and geographical zones and animal keeping technologies in the Vologda Region. Microscopic research methods identified “positive” samples in which Cryptosporidium species were present; they were sorted out and deep-frozen. Then the samples were examined using the equipment of the resource center «Genomic Technologies, Proteomics and Cell Biology» of ARRIAM. The Cryptosporidium species in the fecal samples from farm animals were identified using high-throughput sequencing of 18S rRNA gene amplicon libraries obtained by a nested polymerase chain reaction (PCR) assay.Results and discussion. A primer system was designed for the nested PCR to amplify a potentially species-specific 393 bp fragment of the 18S rRNA gene. The sequence of the ILL_R2_ Zheng primer was modified with included degenerated positions to make the primer more versatile. As a result of sequencing of the libraries of 18S rRNA gene fragments obtained with the selected primers and subsequent taxonomic analysis of the nucleotide sequences, it was shown that all the studied samples included representatives of only one species, Cryptosporidium scrofarum.

Generation and Next-Generation Sequencing-Based Characterization of a Large Human Combinatorial Antibody Library.

Antibody phage display is a key technology for the discovery and development of target-specific monoclonal antibodies (mAbs) for use in research, diagnostics, and therapy. The construction of a high-quality antibody library, with larger and more diverse antibody repertoires, is essential for the successful development of phage display-derived mAbs. In this study, a large human combinatorial single-chain variable fragment library (1.5 × 1011 colonies) was constructed from Epstein-Barr virus-infected human peripheral blood mononuclear cells stimulated with a combination of two of the activators of human B cells, the Toll-like receptor 7/8 agonist R848 and interleukin-2. Next-generation sequencing analysis with approximately 1.9 × 106 and 2.7 × 106 full-length sequences of heavy chain variable (VH) and κ light chain variable (Vκ) domains, respectively, revealed that the library consists of unique VH (approximately 94%) and Vκ (approximately 91%) sequences with greater diversity than germline sequences. Lastly, multiple unique mAbs with high affinity and broad cross-species reactivity could be isolated from the library against two therapeutically relevant target antigens, validating the library quality. These findings suggest that the novel antibody library we have developed may be useful for the rapid development of target-specific phage display-derived recombinant human mAbs for use in therapeutic and diagnostic applications.

Fragment-Based Ab Initio Phasing of Peptidic Nanocrystals by MicroED.

Electron diffraction (MicroED/3DED) can render the three-dimensional atomic structures of molecules from previously unamenable samples. The approach has been particularly transformative for peptidic structures, where MicroED has revealed novel structures of naturally occurring peptides, synthetic protein fragments, and peptide-based natural products. Despite its transformative potential, MicroED is beholden to the crystallographic phase problem, which challenges its de novo determination of structures. ARCIMBOLDO, an automated, fragment-based approach to structure determination, eliminates the need for atomic resolution, instead enforcing stereochemical constraints through libraries of small model fragments, and discerning congruent motifs in solution space to ensure validation. This approach expands the reach of MicroED to presently inaccessible peptide structures including fragments of human amyloids, and yeast and mammalian prions. For electron diffraction, fragment-based phasing portends a more general phasing solution with limited model bias for a wider set of chemical structures.

Genetic Algorithms for the Discovery of Homogeneous Catalysts.

In this account, we discuss the use of genetic algorithms in the inverse design process of homogeneous catalysts for chemical transformations. We describe the main components of evolutionary experiments, specifically the nature of the fitness function to optimize, the library of molecular fragments from which potential catalysts are assembled, and the settings of the genetic algorithm itself. While not exhaustive, this review summarizes the key challenges and characteristics of our own (i.e., NaviCatGA) and other GAs for the discovery of new catalysts.

A specific anti-cyclin D1 intrabody represses breast cancer cell proliferation by interrupting the cyclin D1-CDK4 interaction.

Cyclin D1 overexpression may contribute to development of various cancers, including breast cancer, and thus may serve as a key cancer diagnostic marker and therapeutic target. In our previous study, we generated a cyclin D1-specific single-chain variable fragment antibody (ADκ) from a human semi-synthetic single-chain variable fragment library. ADκ specifically interacted with recombinant and endogenous cyclin D1 proteins through an unknown molecular basis to inhibit HepG2 cell growth and proliferation. Here, using phage display and in silico protein structure modeling methods combined with cyclin D1 mutational analysis, key residues that bind to ADκ were identified. Notably, residue K112 within the cyclin box was required for cyclin D1-ADκ binding. In order to elucidate the molecular mechanism underlying ADκ anti-tumor effects, a cyclin D1-specific nuclear localization signal-containing intrabody (NLS-ADκ) was constructed. When expressed within cells, NLS-ADκ interacted specifically with cyclin D1 to significantly inhibit cell proliferation, induce G1-phase arrest, and trigger apoptosis of MCF-7 and MDA-MB-231 breast cancer cells. Moreover, the NLS-ADκ-cyclin D1 interaction blocked binding of cyclin D1 to CDK4 and inhibited RB protein phosphorylation, resulting in altered expression of downstream cell proliferation-related target genes. We identified amino acid residues in cyclin D1 that may play key roles in the ADκ-cyclin D1 interaction. A nuclear localization antibody against cyclin D1 (NLS-ADκ) was constructed and successfully expressed in breast cancer cells. NLS-ADκ exerted tumor suppressor effects via blockingthe binding of CDK4 to cyclin D1 and inhibiting phosphorylation of RB. The results presented here demonstrate anti-tumor potential of intrabody-based cyclin D1-targeted breast cancer therapy.

Covalent fragment mapping of KRasG12C revealed novel chemotypes with in vivo potency

G12C mutant KRas is considered druggable by allele-specific covalent inhibitors due to the nucleophilic character of the oncogenic mutant cysteine at position 12. Discovery of these inhibitors requires the optimization of both covalent and noncovalent interactions. Here, we report covalent fragment screening of our electrophilic fragment library of diverse non-covalent scaffolds equipped with 40 different electrophilic functionalities to identify fragments as suitable starting points targeting Cys12. Screening the library against KRasG12C using Ellman's free thiol assay, followed by protein NMR and cell viability assays, resulted in two potential inhibitor chemotypes. Characterization of these scaffolds in in vitro cellular- and in vivo xenograft models revealed them as promising starting points for covalent drug discovery programs.

Green Drug Discovery: Novel Fragment Space from the Biomass-Derived Molecule Dihydrolevoglucosenone (CyreneTM)

Biomass-derived molecules can provide a basis for sustainable drug discovery. However, their full exploration is hampered by the dominance of millions of old-fashioned screening compounds in classical high-throughput screening (HTS) libraries frequently utilized. We propose a fragment-based drug discovery (FBDD) approach as an efficient method to navigate biomass-derived drug space. Here, we perform a proof-of-concept study with dihydrolevoglucosenone (CyreneTM), a pyrolysis product of cellulose. Diverse synthetic routes afforded a 100-membered fragment library with a diversity in functional groups appended. The library overall performs well in terms of novelty, physicochemical properties, aqueous solubility, stability, and three-dimensionality. Our study suggests that Cyrene-based fragments are a valuable green addition to the drug discovery toolbox. Our findings can help in paving the way for new hit drug candidates that are based on renewable resources.