Synthesis and Structural Characterisation of Selective Non-Carbohydrate-Based Inhibitors of Bacterial Sialidases

DeepFrag: a deep convolutional neural network for fragment-based lead optimization

Machine learning has been increasingly applied to the field of computer-aided drug discovery in recent years, leading to notable advances in binding-affinity prediction, virtual screening, and QSAR. Surprisingly, it is less often applied to lead optimization, the process of identifying chemical fragments that might be added to a known ligand to improve its binding affinity. We here describe a deep convolutional neural network that predicts appropriate fragments given the structure of a receptor/ligand complex. In an independent benchmark of known ligands with missing (deleted) fragments, our DeepFrag model selected the known (correct) fragment from a set over 6500 about 58% of the time. Even when the known/correct fragment was not selected, the top fragment was often chemically similar and may well represent a valid substitution. We release our trained DeepFrag model and associated software under the terms of the Apache License, Version 2.0.

Sialic acid and biology of life: An introduction

Sialic acids are important molecule with high structural diversity. They are known to occur in higher animals such as Echinoderms, Hemichordata, Cephalochorda, and Vertebrata and also in other animals such as Platyhelminthes, Cephalopoda, and Crustaceae. Plants are known to lack sialic acid. But they are reported to occur in viruses, bacteria, protozoa, and fungi. Deaminated neuraminic acid although occurs in vertebrates and bacteria, is reported to occur in abundance in the lower vertebrates. Sialic acids are mostly located in terminal ends of glycoproteins and glycolipids, capsular and tissue polysialic acids, bacterial lipooligosaccharides/polysaccharides, and in different forms that dictate their role in biology. Sialic acid play important roles in human physiology of cell-cell interaction, communication, cell-cell signaling, carbohydrate-protein interactions, cellular aggregation, development processes, immune reactions, reproduction, and in neurobiology and human diseases in enabling the infection process by bacteria and virus, tumor growth and metastasis, microbiome biology, and pathology. It enables molecular mimicry in pathogens that allows them to escape host immune responses. Recently sialic acid has found role in therapeutics. In this chapter we have highlighted the (i) diversity of sialic acid, (ii) their occurrence in the diverse life forms, (iii) sialylation and disease, and (iv) sialic acid and therapeutics.

9-Azido-9-deoxy-2,3-difluorosialic Acid as a Subnanomolar Inhibitor against Bacterial Sialidases

A library of 2(a),3(a/e)-difluorosialic acids and their C-5 and/or C-9 derivatives were chemoenzymatically synthesized. Pasteurella multocida sialic acid aldolase (PmAldolase), but not its Escherichia coli homologue (EcAldolase), was found to catalyze the formation of C5-azido analogue of 3-fluoro(a)-sialic acid. In comparison, both PmAldolase and EcAldolase could catalyze the synthesis of 3-fluoro(a/e)-sialic acids and their C-9 analogues although PmAldolase was generally more efficient. The chemoenzymatically synthesized 3-fluoro(a/e)-sialic acid analogues were purified and chemically derivatized to form the desired difluorosialic acids and derivatives. Inhibition studies against several bacterial sialidases and a recombinant human cytosolic sialidase hNEU2 indicated that sialidase inhibition was affected by the C-3 fluorine stereochemistry and derivatization at C-5 and/or C-9 of the inhibitor. Opposite to that observed for influenza A virus sialidases and hNEU2, compounds with axial fluorine at C-3 were better inhibitors (up to 100-fold) against bacterial sialidases compared to their 3F-equatorial counterparts. While C-5-modified compounds were less-efficient antibacterial sialidase inhibitors, 9-N₃-modified 2,3-difluoro-Neu5Ac showed increased inhibitory activity against bacterial sialidases. 9-Azido-9-deoxy-2-(e)-3-(a)-difluoro- N-acetylneuraminic acid [2(e)3(a)DFNeu5Ac9N₃] was identified as an effective inhibitor with a long effective duration selectively against pathogenic bacterial sialidases from Clostridium perfringens (CpNanI) and Vibrio cholerae.

Triazole-linked transition state analogs as selective inhibitors against V. cholerae sialidase

Sialidases or neuraminidases are enzymes that catalyze the cleavage of terminal sialic acids from oligosaccharides and glycoconjugates. They play important roles in bacterial and viral infection and have been attractive targets for drug development. Structure-based drug design has led to potent inhibitors against neuraminidases of influenza A viruses that have been used successfully as approved therapeutics. However, selective and effective inhibitors against bacterial and human sialidases are still being actively pursued. Guided by crystal structural analysis, several derivatives of 2-deoxy-2,3-didehydro-N-acetylneuraminic acid (Neu5Ac2en or DANA) were designed and synthesized as triazole-linked transition state analogs. Inhibition studies revealed that glycopeptide analog E-(TriazoleNeu5Ac2en)-AKE and compound (TriazoleNeu5Ac2en)-A were selective inhibitors against Vibrio cholerae sialidase, while glycopeptide analog (TriazoleNeu5Ac2en)-AdE selectively inhibited Vibrio cholerae and A. ureafaciens sialidases.

Streptococcus pneumoniae Sialidase SpNanB-Catalyzed One-Pot Multienzyme (OPME) Synthesis of 2,7-Anhydro-Sialic Acids as Selective Sialidase Inhibitors

Streptococcus pneumoniae sialidase SpNanB is an intramolecular trans-sialidase (IT-sialidase) and a virulence factor that is essential for streptococcal infection of the upper and lower respiratory tract. SpNanB catalyzes the formation of 2,7-anhydro- N-acetylneuraminic acid (2,7-anhydro-Neu5Ac), a potential prebiotic that can be used as the sole carbon source of a common human gut commensal anaerobic bacterium. We report here the development of an efficient one-pot multienzyme (OPME) system for synthesizing 2,7-anhydro-Neu5Ac and its derivatives. Based on a crystal structure analysis, an N-cyclohexyl derivative of 2,7-anhydro-neuraminic acid was designed, synthesized, and shown to be a selective inhibitor against SpNanB and another Streptococcus pneumoniae sialidase SpNanC. This study demonstrates a new strategy of synthesizing 2,7-anhydro-sialic acids in a gram scale and the potential application of their derivatives as selective sialidase inhibitors.

Crystal structure of the catalytic domain of Clostridium perfringens neuraminidase in complex with a non-carbohydrate-based inhibitor, 2-(cyclohexylamino)ethanesulfonic acid

Anti-bacterial and anti-viral neuraminidase agents inhibit neuraminidase activity catalyzing the hydrolysis of terminal N-acetylneuraminic acid (Neu5Ac) from glycoconjugates and help to prevent the host pathogenesis that lead to fatal infectious diseases including influenza, bacteremia, sepsis, and cholera. Emerging antibiotic and drug resistances to commonly used anti-neuraminidase agents such as oseltamivir (Tamiflu) and zanamivir (Relenza) have highlighted the need to develop new anti-neuraminidase drugs. We obtained a serendipitous complex crystal of the catalytic domain of Clostridium perfringens neuraminidase (CpNanI_CD) with 2-(cyclohexylamino)ethanesulfonic acid (CHES) as a buffer. Here, we report the crystal structure of CpNanI_CD in complex with CHES at 1.24 Å resolution. Amphipathic CHES binds to the catalytic site of CpNanI_CD similar to the substrate (Neu5Ac) binding site. The 2-aminoethanesulfonic acid moiety and cyclohexyl groups of CHES interact with the cluster of three arginine residues and with the hydrophobic pocket of the CpNanI_CD catalytic site. In addition, a structural comparison with other bacterial and human neuraminidases suggests that CHES could serve as a scaffold for the development of new anti-neuraminidase agents targeting CpNanI.

•

We determined the crystal structure of CpNanI bound to CHES at 1.24 Å resolution.

•

CHES binds to the catalytic site of CpNanI similar to the substrate binding site.

•

We suggest strategies for modification of CHES for the development of anti-CpNanI agents.

Sequence diversity of NanA manifests in distinct enzyme kinetics and inhibitor susceptibility

Streptococcus pneumoniae is the leading pathogen causing bacterial pneumonia and meningitis. Its surface-associated virulence factor neuraminidase A (NanA) promotes the bacterial colonization by removing the terminal sialyl residues from glycoconjugates on eukaryotic cell surface. The predominant role of NanA in the pathogenesis of pneumococci renders it an attractive target for therapeutic intervention. Despite the highly conserved activity of NanA, our alignment of the 11 NanAs revealed the evolutionary diversity of this enzyme. The amino acid substitutions we identified, particularly those in the lectin domain and in the insertion domain next to the catalytic centre triggered our special interest. We synthesised the representative NanAs and the mutagenized derivatives from E. coli for enzyme kinetics study and neuraminidase inhibitor susceptibility test. Via molecular docking we got a deeper insight into the differences between the two major variants of NanA and their influence on the ligand-target interactions. In addition, our molecular dynamics simulations revealed a prominent intrinsic flexibility of the linker between the active site and the insertion domain, which influences the inhibitor binding. Our findings for the first time associated the primary sequence diversity of NanA with the biochemical properties of the enzyme and with the inhibitory efficiency of neuraminidase inhibitors.

Homologous ligands accommodated by discrete conformations of a buried cavity

Conformational change in protein-ligand complexes is widely modeled, but the protein accommodation expected on binding a congeneric series of ligands has received less attention. Given their use in medicinal chemistry, there are surprisingly few substantial series of congeneric ligand complexes in the Protein Data Bank (PDB). Here we determine the structures of eight alkyl benzenes, in single-methylene increases from benzene to n-hexylbenzene, bound to an enclosed cavity in T4 lysozyme. The volume of the apo cavity suffices to accommodate benzene but, even with toluene, larger cavity conformations become observable in the electron density, and over the series two other major conformations are observed. These involve discrete changes in main-chain conformation, expanding the site; few continuous changes in the site are observed. In most structures, two discrete protein conformations are observed simultaneously, and energetic considerations suggest that these conformations are low in energy relative to the ground state. An analysis of 121 lysozyme cavity structures in the PDB finds that these three conformations dominate the previously determined structures, largely modeled in a single conformation. An investigation of the few congeneric series in the PDB suggests that discrete changes are common adaptations to a series of growing ligands. The discrete, but relatively few, conformational states observed here, and their energetic accessibility, may have implications for anticipating protein conformational change in ligand design.

Exploitation of physiology and metabolomics to identify pneumococcal vaccine candidates

Streptococcus pneumoniae (the pneumococcus) is the etiologic agent of community-acquired pneumonia and invasive pneumococcal diseases such as septicemia and bacterial meningitis. The increasing antibiotic resistance and the suboptimal efficacy or limited serotype coverage of currently available vaccines urgently requires novel approaches in exploring new antimicrobials, therapeutic intervention strategies and vaccines. The current vaccine development strategies rely on the hypothesis that surface-exposed proteins, which are essential for pneumococcal virulence, are the most suitable candidates for future protein-based vaccines. Since virulence is closely linked with bacterial fitness, the potential of a pathogen to colonize and infect the host depends further on its physiology. This review summarizes the application of genome-wide techniques and their exploitation to decipher fundamental insights into bacterial factors associated with fitness, metabolism and virulence, leading to the discovery of vaccine candidates or antimicrobials.