MolOptimizer: A Molecular Optimization Toolkit for Fragment-Based Drug Design.

MolOptimizer is a user-friendly computational toolkit designed to streamline the hit-to-lead optimization process in drug discovery. MolOptimizer extracts features and trains machine learning models using a user-provided, labeled, and small-molecule dataset to accurately predict the binding values of new small molecules that share similar scaffolds with the target in focus. Hosted on the Azure web-based server, MolOptimizer emerges as a vital resource, accelerating the discovery and development of novel drug candidates with improved binding properties.

Inferring primase-DNA specific recognition using a data driven approach.

DNA-protein interactions play essential roles in all living cells. Understanding of how features embedded in the DNA sequence affect specific interactions with proteins is both challenging and important, since it may contribute to finding the means to regulate metabolic pathways involving DNA-protein interactions. Using a massive experimental benchmark dataset of binding scores for DNA sequences and a machine learning workflow, we describe the binding to DNA of T7 primase, as a model system for specific DNA-protein interactions. Effective binding of T7 primase to its specific DNA recognition sequences triggers the formation of RNA primers that serve as Okazaki fragment start sites during DNA replication.

Dual-Acting Small-Molecule Inhibitors Targeting Mycobacterial DNA Replication.

Mycobacterium tuberculosis (Mtb) is a pathogenic bacterium and a causative agent of tuberculosis (TB), a disease that kills more than 1.5 million people worldwide annually. One of the main reasons for this high mortality rate is the evolution of new Mtb strains that are resistant to available antibiotics. Therefore, new therapeutics for TB are in constant demand. Here, we report the development of small-molecule inhibitors that target two DNA replication enzymes of Mtb, namely DnaG primase and DNA gyrase (Gyr), which share a conserved TOPRIM fold near the inhibitors' binding site. The molecules were developed on the basis of previously reported inhibitors for T7 DNA primase that bind near the TOPRIM fold. To improve the physicochemical properties of the molecules as well as their inhibitory effect on primase and gyrase, 49 novel compounds have been synthesized as potential drug candidates in three stages of optimization. The last stage of chemical optimization yielded two novel inhibitors for both Mtb DnaG and Gyr that also showed inhibitory activity toward the fast-growing non-pathogenic model Mycobacterium smegmatis (Msmg).

The Amuvatinib Derivative, N-(2H-1,3-Benzodioxol-5-yl)-4-{thieno[3,2-d]pyrimidin-4-yl}piperazine-1-carboxamide, Inhibits Mitochondria and Kills Tumor Cells under Glucose Starvation.

Glucose levels inside solid tumors are low as compared with normal surrounding tissue, forcing tumor cells to reprogram their metabolism to adapt to such low glucose conditions. Unlike normal tissue, tumor cells experience glucose starvation, making the targeting of pathways supporting survival during glucose starvation an interesting therapeutic strategy in oncology. Using high-throughput screening, we previously identified small molecules that selectively kill cells exposed to glucose starvation. One of the identified compounds was the kinase inhibitor amuvatinib. To identify new molecules with potential antineoplastic activity, we procured 12 amuvatinib derivatives and tested their selective toxicity towards glucose-starved tumor cells. One of the amuvatinib derivatives, N-(2H-1,3-benzodioxol-5-yl)-4-{thieno[3,2-d]pyrimidin-4-yl}piperazine-1-carboxamide, termed compound 6, was found to be efficacious in tumor cells experiencing glucose starvation. In line with the known dependence of glucose-starved cells on the mitochondria, compound 6 inhibits mitochondrial membrane potential. These findings support the concept that tumor cells are dependent on mitochondria under glucose starvation, and bring forth compound 6 as a new molecule with potential antitumor activity for the treatment of glucose-starved tumors.

Engineering Stem Cell Factor Ligands with Different c-Kit Agonistic Potencies.

Receptor tyrosine kinases (RTKs) are major players in signal transduction, regulating cellular activities in both normal regeneration and malignancy. Thus, many RTKs, c-Kit among them, play key roles in the function of both normal and neoplastic cells, and as such constitute attractive targets for therapeutic intervention. We thus sought to manipulate the self-association of stem cell factor (SCF), the cognate ligand of c-Kit, and hence its suboptimal affinity and activation potency for c-Kit. To this end, we used directed evolution to engineer SCF variants having different c-Kit activation potencies. Our yeast-displayed SCF mutant (SCFM) library screens identified altered dimerization potential and increased affinity for c-Kit by specific SCF-variants. We demonstrated the delicate balance between SCF homo-dimerization, c-Kit binding, and agonistic potencies by structural studies, in vitro binding assays and a functional angiogenesis assay. Importantly, our findings showed that a monomeric SCF variant exhibited superior agonistic potency vs. the wild-type SCF protein and vs. other high-affinity dimeric SCF variants. Our data showed that action of the monomeric ligands in binding to the RTK monomers and inducing receptor dimerization and hence activation was superior to that of the wild-type dimeric ligand, which has a higher affinity to RTK dimers but a lower activation potential. The findings of this study on the binding and c-Kit activation of engineered SCF variants thus provides insights into the structure-function dynamics of ligands and RTKs.

Engineering a monomeric variant of macrophage colony-stimulating factor (M-CSF) that antagonizes the c-FMS receptor.

Enhanced activation of the signaling pathways that mediate the differentiation of mononuclear monocytes into osteoclasts is an underlying cause of several bone diseases and bone metastasis. In particular, dysregulation and overexpression of macrophage colony-stimulating factor (M-CSF) and its c-FMS tyrosine kinase receptor, proteins that are essential for osteoclast differentiation, are known to promote bone metastasis and osteoporosis, making both the ligand and its receptor attractive targets for therapeutic intervention. With this aim in mind, our starting point was the previously held concept that the potential of the M-CSFC31S mutant as a therapeutic is derived from its inability to dimerize and hence to act as an agonist. The current study showed, however, that dimerization is not abolished in M-CSFC31S and that the protein retains agonistic activity toward osteoclasts. To design an M-CSF mutant with diminished dimerization capabilities, we solved the crystal structure of the M-CSFC31S dimer complex and used structure-based energy calculations to identify the residues responsible for its dimeric form. We then used that analysis to develop M-CSFC31S,M27R, a ligand-based, high-affinity antagonist for c-FMS that retained its binding ability but prevented the ligand dimerization that leads to receptor dimerization and activation. The monomeric properties of M-CSFC31S,M27R were validated using dynamic light scattering and small-angle X-ray scattering analyses. It was shown that this mutant is a functional inhibitor of M-CSF-dependent c-FMS activation and osteoclast differentiation in vitro Our study, therefore, provided insights into the sequence-structure-function relationships of the M-CSF/c-FMS interaction and of ligand/receptor tyrosine kinase interactions in general.

NMR-Fragment Based Virtual Screening: A Brief Overview.

Fragment-based drug discovery (FBDD) using NMR has become a central approach over the last twenty years for development of small molecule inhibitors against biological macromolecules, to control a variety of cellular processes. Yet, several considerations should be taken into account for obtaining a therapeutically relevant agent. In this review, we aim to list the considerations that make NMR fragment screening a successful process for yielding potent inhibitors. Factors that may govern the competence of NMR in fragment based drug discovery are discussed, as well as later steps that involve optimization of hits obtained by NMR-FBDD.

Human translation initiation factor eIF4G1 possesses a low-affinity ATP binding site facing the ATP-binding cleft of eIF4A in the eIF4G/eIF4A complex.

Eukaryotic translation initiation factor 4G (eIF4G) plays a crucial role in translation initiation, serving as a scaffolding protein binding several other initiation factors, other proteins, and RNA. Binding of eIF4G to the ATP-dependent RNA helicase eukaryotic translation initiation factor 4A (eIF4A) enhances the activity of eIF4A in solution and in crowded environments. Previously, this activity enhancement was solely attributed to eIF4G, conferring a closed, active conformation upon eIF4A. Here we show that eIF4G contains a low-affinity binding site at the entrance to the ATP-binding cleft on eIF4A, suggesting that regulation of the local ATP concentration may be an additional reason for the enhancement in activity.

Isolation, characterization, and aggregation of a structured bacterial matrix precursor.

Biofilms are surface-associated groups of microbial cells that are embedded in an extracellular matrix (ECM). The ECM is a network of biopolymers, mainly polysaccharides, proteins, and nucleic acids. ECM proteins serve a variety of structural roles and often form amyloid-like fibers. Despite the extensive study of the formation of amyloid fibers from their constituent subunits in humans, much less is known about the assembly of bacterial functional amyloid-like precursors into fibers. Using dynamic light scattering, atomic force microscopy, circular dichroism, and infrared spectroscopy, we show that our unique purification method of a Bacillus subtilis major matrix protein component results in stable oligomers that retain their native α-helical structure. The stability of these oligomers enabled us to control the external conditions that triggered their aggregation. In particular, we show that stretched fibers are formed on a hydrophobic surface, whereas plaque-like aggregates are formed in solution under acidic pH conditions. TasA is also shown to change conformation upon aggregation and gain some ß-sheet structure. Our studies of the aggregation of a bacterial matrix protein from its subunits shed new light on assembly processes of the ECM within bacterial biofilms.

NAGase sensing in 3% milk: FET-based specific and label-free sensing in ultra-small samples of high ionic strength and high concentration of non-specific proteins.

Biosensing with biological field-effect transistors (bioFETs) is a promising technology toward specific, label-free, and multiplexed sensing in ultra-small samples. The current study employs the field-effect meta-nano-channel biosensor (MNC biosensor) for the detection of the enzyme N-acetyl-beta-D-glucosaminidase (NAGase), a biomarker for milk cow infections. The measurements are performed in a 0.5 µL drops of 3% commercial milk spiked with NAGase concentrations in the range of 30.3 aM-3.03 µM (Note that there is no background NAGase concentration in commercial milk). Specific and label-free sensing of NAGase is demonstrated with a limit-of-detection of 30.3 aM, a dynamic range of 11 orders of magnitude and with excellent linearity and sensitivity. Additional two important research outcomes are reported. First, the ionic strength of the examined milk is ∼120 mM which implies a bulk Debye screening length <1 nm. Conventionally, a 1 nm Debye length excludes the possibility of sensing with a recognition layer composed of surface bound anti-NAGase antibodies with a size of ∼10 nm. This apparent contradiction is removed considering the ample literature reporting antibody adsorption in a predominantly surface tilted configuration (side-on, flat-on, etc.). Secondly, milk contains a non-specific background protein concentration of 33 mg/ml, in addition to considerable amounts of micron-size heterogeneous fat structures. The reported sensing was performed without the customarily exercised surface blocking and without washing of the non-specific signal. This suggests that the role of non-specific adsorption to the BioFET sensing signal needs to be further evaluated. Control measurements are reported.

From sensing interactions to controlling the interactions: a novel approach to obtain biological transistors for specific and label-free immunosensing.

Antibody-antigen interactions are shaped by the solution pH level, ionic strength, and electric fields, if present. In biological field-effect transistors (BioFETs), the interactions take place at the sensing area in which the pH level, ionic strength and electric fields are determined by the Poisson-Boltzmann equation and the boundary conditions at the solid-solution interface and the potential applied at the solution electrode. The present study demonstrates how a BioFET solution electrode potential affects the sensing area double layer pH level, ionic strength, and electric fields and in this way shapes the biological interactions at the sensing area. We refer to this as 'active sensing'. To this end, we employed the meta-nano-channel (MNC) BioFET and demonstrate how the solution electrode can determine the antibody-antigen equilibrium constant and allows the control and tuning of the sensing performance in terms of the dynamic range and limit-of-detection. In the current work, we employed this method to demonstrate the specific and label-free sensing of Alpha-Fetoprotein (AFP) molecules from 0.5 µL drops of 1 : 100 diluted serum. AFP was measured during pregnancy as part of the prenatal screening program for fetal anomalies, chromosomal abnormalities, and abnormal placentation. We demonstrate AFP sensing with a limit-of-detection of 10.5 aM and a dynamic range of 6 orders of magnitude in concentration. Extensive control measurements are reported.

The interaction between eukaryotic initiation factor 1A and eIF5 retains eIF1 within scanning preinitiation complexes.

Scanning of the mRNA transcript by the preinitiation complex (PIC) requires a panel of eukaryotic initiation factors, which includes eIF1 and eIF1A, the main transducers of stringent AUG selection. eIF1A plays an important role in start codon recognition; however, its molecular contacts with eIF5 are unknown. Using nuclear magnetic resonance, we unveil eIF1A's binding surface on the carboxyl-terminal domain of eIF5 (eIF5-CTD). We validated this interaction by observing that eIF1A does not bind to an eIF5-CTD mutant, altering the revealed eIF1A interaction site. We also found that the interaction between eIF1A and eIF5-CTD is conserved between humans and yeast. Using glutathione S-transferase pull-down assays of purified proteins, we showed that the N-terminal tail (NTT) of eIF1A mediates the interaction with eIF5-CTD and eIF1. Genetic evidence indicates that overexpressing eIF1 or eIF5 suppresses the slow growth phenotype of eIF1A-NTT mutants. These results suggest that the eIF1A-eIF5-CTD interaction during scanning PICs contributes to the maintenance of eIF1 within the open PIC.

Zinc-binding domain of the bacteriophage T7 DNA primase modulates binding to the DNA template.

The zinc-binding domain (ZBD) of prokaryotic DNA primases has been postulated to be crucial for recognition of specific sequences in the single-stranded DNA template. To determine the molecular basis for this role in recognition, we carried out homolog-scanning mutagenesis of the zinc-binding domain of DNA primase of bacteriophage T7 using a bacterial homolog from Geobacillus stearothermophilus. The ability of T7 DNA primase to catalyze template-directed oligoribonucleotide synthesis is eliminated by substitution of any five-amino acid residue-long segment within the ZBD. The most significant defect occurs upon substitution of a region (Pro-16 to Cys-20) spanning two cysteines that coordinate the zinc ion. The role of this region in primase function was further investigated by generating a protein library composed of multiple amino acid substitutions for Pro-16, Asp-18, and Asn-19 followed by genetic screening for functional proteins. Examination of proteins selected from the screening reveals no change in sequence-specific recognition. However, the more positively charged residues in the region facilitate DNA binding, leading to more efficient oligoribonucleotide synthesis on short templates. The results suggest that the zinc-binding mode alone is not responsible for sequence recognition, but rather its interaction with the RNA polymerase domain is critical for DNA binding and for sequence recognition. Consequently, any alteration in the ZBD that disturbs its conformation leads to loss of DNA-dependent oligoribonucleotide synthesis.

Thioredoxin, the processivity factor, sequesters an exposed cysteine in the thumb domain of bacteriophage T7 DNA polymerase.

Gene 5 protein (gp5) of bacteriophage T7 is a non-processive DNA polymerase. It achieves processivity by binding to Escherichia coli thioredoxin (trx). gp5/trx complex binds tightly to a primer-DNA template enabling the polymerization of hundreds of nucleotides per binding event. gp5 contains 10 cysteines. Under non-reducing condition, exposed cysteines form intermolecular disulfide linkages resulting in the loss of polymerase activity. No disulfide linkage is detected when Cys-275 and Cys-313 are replaced with serines. Cys-275 and Cys-313 are located on loop A and loop B of the thioredoxin binding domain, respectively. Replacement of either cysteine with serine (gp5-C275S, gp5-C313S) drastically decreases polymerase activity of gp5 on dA(350)/dT(25). On this primer-template gp5/trx in which Cys-313 or Cys-275 is replaced with serine have 50 and 90%, respectively, of the polymerase activity observed with wild-type gp5/trx. With single-stranded M13 DNA as a template gp5-C275S/trx retains 60% of the polymerase activity of wild-type gp5/trx. In contrast, gp5-C313S/trx has only one-tenth of the polymerase activity of wild-type gp5/trx on M13 DNA. Both wild-type gp5/trx and gp5-C275S/trx catalyze the synthesis of the entire complementary strand of M13 DNA, whereas gp5-C313S/trx has difficulty in synthesizing DNA through sites of secondary structure. gp5-C313S fails to form a functional complex with trx as measured by the apparent binding affinity as well as by the lack of a physical interaction with thioredoxin during hydroxyapatite-phosphate chromatography. Small angle x-ray scattering reveals an elongated conformation of gp5-C313S in comparison to a compact and spherical conformation of wild-type gp5.

An interaction between DNA polymerase and helicase is essential for the high processivity of the bacteriophage T7 replisome.

Synthesis of the leading DNA strand requires the coordinated activity of DNA polymerase and DNA helicase, whereas synthesis of the lagging strand involves interactions of these proteins with DNA primase. We present the first structural model of a bacteriophage T7 DNA helicase-DNA polymerase complex using a combination of small angle x-ray scattering, single-molecule, and biochemical methods. We propose that the protein-protein interface stabilizing the leading strand synthesis involves two distinct interactions: a stable binding of the helicase to the palm domain of the polymerase and an electrostatic binding of the carboxyl-terminal tail of the helicase to a basic patch on the polymerase. DNA primase facilitates binding of DNA helicase to ssDNA and contributes to formation of the DNA helicase-DNA polymerase complex by stabilizing DNA helicase.

Molecular crowding enhanced ATPase activity of the RNA helicase eIF4A correlates with compaction of its quaternary structure and association with eIF4G.

Enzymatic reactions occur in a crowded and confined environment in vivo, containing proteins, RNA and DNA. Previous reports have shown that interactions between macromolecules, and reactions rates differ significantly between crowded environments and dilute buffers. However, the direct effect of crowding on the level of high-resolution structures of macromolecules has not been extensively analyzed and is not well understood. Here we analyze the effect of macromolecular crowding on structure and function of the human translation initiation factors eIF4A, a two-domain DEAD-Box helicase, the HEAT-1 domain of eIF4G, and their complex. We find that crowding enhances the ATPase activity of eIF4A, which correlates with a shift to a more compact structure as revealed with small-angle X-ray scattering. However, the individual domains of eIF4A, or the eIF4G-HEAT-1 domain alone show little structural changes due to crowding except for flexible regions. Thus, the effect of macromolecular crowding on activity and structure need to be taken into account when evaluating enzyme activities and structures of multidomain proteins, proteins with flexible regions, or protein complexes obtained by X-ray crystallography, NMR, or other structural methods.

Conformational dynamics of bacteriophage T7 DNA polymerase and its processivity factor, Escherichia coli thioredoxin.

Gene 5 of bacteriophage T7 encodes a DNA polymerase (gp5) responsible for the replication of the phage DNA. Gp5 polymerizes nucleotides with low processivity, dissociating after the incorporation of 1 to 50 nucleotides. Thioredoxin (trx) of Escherichia coli binds tightly (Kd = 5 nM) to a unique segment in the thumb subdomain of gp5 and increases processivity. We have probed the molecular basis for the increase in processivity. A single-molecule experiment reveals differences in rates of enzymatic activity and processivity between gp5 and gp5/trx. Small angle X-ray scattering studies combined with nuclease footprinting reveal two conformations of gp5, one in the free state and one upon binding to trx. Comparative analysis of the DNA binding clefts of DNA polymerases and DNA binding proteins show that the binding surface contains more hydrophobic residues than other DNA binding proteins. The balanced composition between hydrophobic and charged residues of the binding site allows for efficient sliding of gp5/trx on the DNA. We propose a model for trx-induced conformational changes in gp5 that enhance the processivity by increasing the interaction of gp5 with DNA.

Pyrovanadolysis, a pyrophosphorolysis-like reaction mediated by pyrovanadate, Mn2+, and DNA polymerase of bacteriophage T7.

DNA polymerases catalyze the 3'-5'-pyrophosphorolysis of a DNA primer annealed to a DNA template in the presence of pyrophosphate (PP(i)). In this reversal of the polymerization reaction, deoxynucleotides in DNA are converted to deoxynucleoside 5'-triphosphates. Based on the charge, size, and geometry of the oxygen connecting the two phosphorus atoms of PP(i), a variety of compounds was examined for their ability to carry out a reaction similar to pyrophosphorolysis. We describe a manganese-mediated pyrophosphorolysis-like activity using pyrovanadate (VV) catalyzed by the DNA polymerase of bacteriophage T7. We designate this reaction pyrovanadolysis. X-ray absorption spectroscopy reveals a shorter Mn-V distance of the polymerase-VV complex than the Mn-P distance of the polymerase-PP(i) complex. This structural arrangement at the active site accounts for the enzymatic activation by Mn-VV. We propose that the Mn(2+), larger than Mg(2+), fits the polymerase active site to mediate binding of VV into the active site of the polymerase. Our results may be the first documentation that vanadium can substitute for phosphorus in biological processes.

Machine learning approaches to optimize small-molecule inhibitors for RNA targeting.

In the era of data science, data-driven algorithms have emerged as powerful platforms that can consolidate bioisosteric rules for preferential modifications on small molecules with a common molecular scaffold. Here we present complementary data-driven algorithms to minimize the search in chemical space for phenylthiazole-containing molecules that bind the RNA hairpin within the ribosomal peptidyl transferase center (PTC) of Mycobacterium tuberculosis. Our results indicate visual, geometrical, and chemical features that enhance the binding to the targeted RNA. Functional validation was conducted after synthesizing 10 small molecules pinpointed computationally. Four of the 10 were found to be potent inhibitors that target hairpin 91 in the ribosomal PTC of M. tuberculosis and, as a result, stop translation.

4-phenylbutyric acid-Identity crisis; can it act as a translation inhibitor?

Loss of proteostasis can occur due to mutations, the formation of aggregates, or general deficiency in the correct translation and folding of proteins. These phenomena are commonly observed in pathologies, but most significantly, loss of proteostasis characterizes aging. This loss leads to the chronic activation of stress responses and has a generally deleterious impact on the organism. While finding molecules that can alleviate these symptoms is an important step toward solutions for these conditions, some molecules might be mischaracterized on the way. 4-phenylbutyric acid (4PBA) is known for its role as a chemical chaperone that helps alleviate endoplasmic reticulum (ER) stress, yet a scan of the literature reveals that no biochemical or molecular experiments have shown any protein refolding capacity. Here, we show that 4PBA is a conserved weak inhibitor of mRNA translation, both in vitro and in cellular systems, and furthermore-it does not promote protein folding nor prevents aggregation. 4PBA possibly alleviates proteostatic or ER stress by inhibiting protein synthesis, allowing the cells to cope with misfolded proteins by reducing the protein load. Better understanding of 4PBA biochemical mechanisms will improve its usage in basic science and as a drug in different pathologies, also opening new venues for the treatment of different diseases.