A small-molecule TNIK inhibitor targets fibrosis in preclinical and clinical models.

Idiopathic pulmonary fibrosis (IPF) is an aggressive interstitial lung disease with a high mortality rate. Putative drug targets in IPF have failed to translate into effective therapies at the clinical level. We identify TRAF2- and NCK-interacting kinase (TNIK) as an anti-fibrotic target using a predictive artificial intelligence (AI) approach. Using AI-driven methodology, we generated INS018_055, a small-molecule TNIK inhibitor, which exhibits desirable drug-like properties and anti-fibrotic activity across different organs in vivo through oral, inhaled or topical administration. INS018_055 possesses anti-inflammatory effects in addition to its anti-fibrotic profile, validated in multiple in vivo studies. Its safety and tolerability as well as pharmacokinetics were validated in a randomized, double-blinded, placebo-controlled phase I clinical trial (NCT05154240) involving 78 healthy participants. A separate phase I trial in China, CTR20221542, also demonstrated comparable safety and pharmacokinetic profiles. This work was completed in roughly 18 months from target discovery to preclinical candidate nomination and demonstrates the capabilities of our generative AI-driven drug-discovery pipeline.

AlphaFold accelerates artificial intelligence powered drug discovery: efficient discovery of a novel CDK20 small molecule inhibitor.

The application of artificial intelligence (AI) has been considered a revolutionary change in drug discovery and development. In 2020, the AlphaFold computer program predicted protein structures for the whole human genome, which has been considered a remarkable breakthrough in both AI applications and structural biology. Despite the varying confidence levels, these predicted structures could still significantly contribute to structure-based drug design of novel targets, especially the ones with no or limited structural information. In this work, we successfully applied AlphaFold to our end-to-end AI-powered drug discovery engines, including a biocomputational platform PandaOmics and a generative chemistry platform Chemistry42. A novel hit molecule against a novel target without an experimental structure was identified, starting from target selection towards hit identification, in a cost- and time-efficient manner. PandaOmics provided the protein of interest for the treatment of hepatocellular carcinoma (HCC) and Chemistry42 generated the molecules based on the structure predicted by AlphaFold, and the selected molecules were synthesized and tested in biological assays. Through this approach, we identified a small molecule hit compound for cyclin-dependent kinase 20 (CDK20) with a binding constant Kd value of 9.2 ± 0.5 µM (n = 3) within 30 days from target selection and after only synthesizing 7 compounds. Based on the available data, a second round of AI-powered compound generation was conducted and through this, a more potent hit molecule, ISM042-2-048, was discovered with an average Kd value of 566.7 ± 256.2 nM (n = 3). Compound ISM042-2-048 also showed good CDK20 inhibitory activity with an IC50 value of 33.4 ± 22.6 nM (n = 3). In addition, ISM042-2-048 demonstrated selective anti-proliferation activity in an HCC cell line with CDK20 overexpression, Huh7, with an IC50 of 208.7 ± 3.3 nM, compared to a counter screen cell line HEK293 (IC50 = 1706.7 ± 670.0 nM). This work is the first demonstration of applying AlphaFold to the hit identification process in drug discovery.

Chemistry42: An AI-Driven Platform for Molecular Design and Optimization.

Chemistry42 is a software platform for de novo small molecule design and optimization that integrates Artificial Intelligence (AI) techniques with computational and medicinal chemistry methodologies. Chemistry42 efficiently generates novel molecular structures with optimized properties validated in both in vitro and in vivo studies and is available through licensing or collaboration. Chemistry42 is the core component of Insilico Medicine's Pharma.ai drug discovery suite. Pharma.ai also includes PandaOmics for target discovery and multiomics data analysis, and inClinico─a data-driven multimodal forecast of a clinical trial's probability of success (PoS). In this paper, we demonstrate how the platform can be used to efficiently find novel molecular structures against DDR1 and CDK20.

Synthesis, inhibitory activity and oral dosing formulation of AV5124, the structural analogue of influenza virus endonuclease inhibitor baloxavir.

BACKGROUND: The development and clinical implementation of the cap-dependent endonuclease (CEN) inhibitor baloxavir marboxil was a breakthrough in influenza therapy, but it was associated with the emergence of drug-resistant variants. OBJECTIVES: To design and synthesize structural analogues of CEN inhibitors and evaluate their safety, pharmacokinetics and antiviral potency in vitro and in vivo. METHODS: The drug candidate AV5124 and its active metabolite AV5116 were synthesized based on pharmacophore modelling. Stability in plasma and microsomes, plasma protein binding, cytotoxicity and antiviral activities were assessed in vitro. Pharmacokinetics after IV or oral administration were analysed in CD-1 mice. Acute toxicity and protective efficacy against lethal A(H1N1)pdm09 influenza virus challenge were examined in BALB/c mice. RESULTS: Pharmacophore model-assisted, 3D molecular docking predicted key supramolecular interactions of the metal-binding group and bulky hydrophobic group of AV5116 with the CEN binding site (Protein Data Bank code: 6FS6) that are essential for high antiviral activity. AV5116 inhibited influenza virus polymerase complexes in cell-free assays and replication of oseltamivir-susceptible and -resistant influenza A and B viruses at nanomolar concentrations. Notably, AV5116 was equipotent or more potent than baloxavir acid (BXA) against WT (I38-WT) viruses and viruses with reduced BXA susceptibility carrying an I38T polymerase acidic (PA) substitution. AV5116 exhibited low cytotoxicity in Madin-Darby canine kidney cells and lacked mitochondrial toxicity, resulting in favourable selective indices. Treatment with 20 or 50 mg/kg AV5124 prevented death in 60% and 100% of animals, respectively. CONCLUSIONS: Overall, AV5124 and A5116 are promising inhibitors of the influenza virus CEN and warrant further development as potent anti-influenza agents.

Synthesis, biological evaluation and in silico modeling of novel pan-genotypic NS5A inhibitors.

A series of novel small-molecule pan-genotypic hepatitis C virus (HCV) NS5A inhibitors with picomolar activity containing 2-[(2S)-pyrrolidin-2-yl]-5-[4-(4-{2-[(2S)-pyrrolidin-2-yl]-1H-imidazol-5-yl}buta-1,3-diyn-1-yl)phenyl]-1H-imidazole core was designed based on molecular modeling study and SAR analysis. The constructed in silico model and docking study provide a deep insight into the binding mode of this type of NS5A inhibitors. Based on the predicted binding interface we have prioritized the most crucial diversity points responsible for improving antiviral activity. The synthesized molecules were tested in a cell-based assay, and compound 1.12 showed an EC50 value in the range of 2.9-34 pM against six genotypes of NS5A HCV, including gT3a, and demonstrated favorable pharmacokinetic profile in rats. This lead compound can be considered as an attractive candidate for further clinical evaluation.

Non-rigid Diarylmethyl Analogs of Baloxavir as Cap-Dependent Endonuclease Inhibitors of Influenza Viruses.

4-Substituted 2,4-dioxobutanoic acids inhibit influenza virus cap-dependent endonuclease (CEN) activity. Baloxavir marboxil, 4, is approved for treating influenza virus infections. We describe here the synthesis and biological evaluation of active compounds, 5a-5g, and their precursors (6a, 6b, 6d, and 6e) with flexible bulky hydrophobic groups instead of the rigid polyheterocyclic moieties. In silico docking confirmed the ability of 5a-5g to bind to the active site of influenza A CEN (PDB code: 6FS6) like baloxavir acid, 3. These novel compounds inhibited polymerase complex activity, inhibited virus replication in cells, prevented death in a lethal influenza A virus mouse challenge model, and dramatically lowered viral lung titers. 5a and 5e potently inhibited different influenza genera in vitro. Precursors 6a and 6d demonstrated impressive mouse oral bioavailability with 6a, providing effective in vivo protection. Thus, these novel compounds are potent CEN inhibitors with in vitro and in vivo activity comparable to baloxavir.

Synthesis and biological activity of 5-aryliden-2-thiohydantoin S-aryl derivatives.

Three new and complementary approaches to S-arylation of 2-thiohydantoins have been developed: copper-catalyzed cross coupling with either arylboronic acids or aryl iodides under mild conditions, or direct nucleophilic substitution in activated aryl halides. For 38 diverse compounds, reaction yields for all three methods have been determined. Selected by molecular docking, they have been tested on androgen receptor activation, and p53-Mdm2 regulation, and A549, MCF7, VA13, HEK293T, PC3, LnCAP cell lines for cytotoxicity, Two of them turned out to be promising as androgen receptor activators (likely by allosteric regulation), and another one is shown to activate the p53 cascade. It is hoped that 2-thiohydantoin S-arylidenes are worth further studies as biologically active compounds.

Synthesis, biological evaluation and in silico modeling of novel integrase strand transfer inhibitors (INSTIs).

Although a relatively wide range of therapeutic options is currently available for the treatment of HIV/AIDS, it is still among the most serious and virulent diseases and is associated with a high mortality rate. Integrase strand transfer inhibitors (INSTIs), e.g., FDA-approved dolutegravir (DTG), bictegravir (BIC) and cabotegravir (CAB), have recently been included in standard highly active antiretroviral therapy (HAART) schemes as one of the five major components responsible for the most beneficial clinical outcome. In this paper, we describe a combinatorial amide synthesis, biological evaluation and in silico modeling of new INSTIs containing heteroaromatic bioisosteric substitution instead of the well-studied halogen-substituted benzyl fragment. With the focus on the mentioned diversity point, a medium-sized library of compounds was selected for synthesis. A biological study revealed that many molecules were highly active INSTIs (EC50 < 10 nM). Two compounds 1{4} and 1{26} demonstrated picomolar antiviral activity that was comparable with CAB and were more active than DTG and BIC. Molecular docking study was performed to evaluate the binding mode of compounds in the active site of HIV-1 IN. In rats, lead compound 1{26} showed two-fold greater bioavailability than CAB and had a similar half-life. Compound 1{26} and its sodium salt were considerably more soluble in water than the parent drugs. Both molecules were very stable in human liver microsomes and plasma, demonstrated high affinity towards plasma proteins and did not show cytochrome (CYP) inhibition. This benefit profile indicates the great potential of these molecules as attractive candidates for subsequent evaluation as oral long-acting drugs and long-acting nanosuspension formulations for intramuscular injection.

Molecular Sets (MOSES): A Benchmarking Platform for Molecular Generation Models.

Generative models are becoming a tool of choice for exploring the molecular space. These models learn on a large training dataset and produce novel molecular structures with similar properties. Generated structures can be utilized for virtual screening or training semi-supervized predictive models in the downstream tasks. While there are plenty of generative models, it is unclear how to compare and rank them. In this work, we introduce a benchmarking platform called Molecular Sets (MOSES) to standardize training and comparison of molecular generative models. MOSES provides training and testing datasets, and a set of metrics to evaluate the quality and diversity of generated structures. We have implemented and compared several molecular generation models and suggest to use our results as reference points for further advancements in generative chemistry research. The platform and source code are available at https://github.com/molecularsets/moses.

Large-scale High-throughput Screening Revealed 5'-(carbonylamino)-2,3'- bithiophene-4'-carboxylate as Novel Template for Antibacterial Agents.

BACKGROUND: The key issue in the development of novel antimicrobials is a rapid expansion of new bacterial strains resistant to current antibiotics. Indeed, World Health Organization has reported that bacteria commonly causing infections in hospitals and in the community, e.g. E. Coli, K. pneumoniae and S. aureus, have high resistance vs the last generations of cephalosporins, carbapenems and fluoroquinolones. During the past decades, only few successful efforts to develop and launch new antibacterial medications have been performed. This study aims to identify new class of antibacterial agents using novel high-throughput screening technique. METHODS: We have designed library containing 125K compounds not similar in structure (Tanimoto coeff.< 0.7) to that published previously as antibiotics. The HTS platform based on double reporter system pDualrep2 was used to distinguish between molecules able to block translational machinery or induce SOS-response in a model E. coli system. MICs for most active chemicals in LB and M9 medium were determined using broth microdilution assay. RESULTS: In an attempt to discover novel classes of antibacterials, we performed HTS of a large-scale small molecule library using our unique screening platform. This approach permitted us to quickly and robustly evaluate a lot of compounds as well as to determine the mechanism of action in the case of compounds being either translational machinery inhibitors or DNA-damaging agents/replication blockers. HTS has resulted in several new structural classes of molecules exhibiting an attractive antibacterial activity. Herein, we report as promising antibacterials. Two most active compounds from this series showed MIC value of 1.2 (5) and 1.8 µg/mL (6) and good selectivity index. Compound 6 caused RFP induction and low SOS response. In vitro luciferase assay has revealed that it is able to slightly inhibit protein biosynthesis. Compound 5 was tested on several archival strains and exhibited slight activity against gram-negative bacteria and outstanding activity against S. aureus. The key structural requirements for antibacterial potency were also explored. We found, that the unsubstituted carboxylic group is crucial for antibacterial activity as well as the presence of bulky hydrophobic substituents at phenyl fragment. CONCLUSION: The obtained results provide a solid background for further characterization of the 5'- (carbonylamino)-2,3'-bithiophene-4'-carboxylate derivatives discussed herein as new class of antibacterials and their optimization campaign.

Identification of pyrrolo-pyridine derivatives as novel class of antibacterials.

A series of 5-oxo-4H-pyrrolo[3,2-b]pyridine derivatives was identified as novel class of highly potent antibacterial agents during an extensive large-scale high-throughput screening (HTS) program utilizing a unique double-reporter system-pDualrep2. The construction of the reporter system allows us to perform visual inspection of the underlying mechanism of action due to two genes-Katushka2S and RFP-which encode the proteins with different imaging signatures. Antibacterial activity of the compounds was evaluated during the initial HTS round and subsequent rescreen procedure. The most active molecule demonstrated a MIC value of 3.35 µg/mL against E. coli with some signs of translation blockage (low Katushka2S signal) and no SOS response. The compound did not demonstrate cytotoxicity in standard cell viability assay. Subsequent structural morphing and follow-up synthesis may result in novel compounds with a meaningful antibacterial potency which can be reasonably regarded as an attractive starting point for further in vivo investigation and optimization.

Substituted Furanocoumarins as Novel Class of Antibacterial Translation Inhibitors.

INTRODUCTION: A variety of organic compounds has been reported to have antibacterial activity. However, antimicrobial resistance is one of the main problems of current anti-infective therapy, and the development of novel antibacterials is one of the main challenges of current drug discovery. METHODS: Using our previously developed dual-reporter High-Throughput Screening (HTS) platform, we identified a series of furanocoumarins as having high antibacterial activity. The construction of the reporter system allows us to differentiate three mechanisms of action for the active compounds: inhibition of protein synthesis (induction of Katushka2S), DNA damaging (induction of RFP) or other (inhibition of bacterial growth without reporter induction). RESULTS: Two primary hit-molecules of furanocoumarin series demonstrated relatively low MIC values comparable to that observed for Erythromycin (Ery) against E. coli and weakly induced both reporters. Dose-dependent translation inhibition was shown using in vitro luciferase assay, however it was not confirmed using C14-test. A series of close structure analogs of the identified hits was obtained and investigated using the same screening platform. Compound 19 was found to have slightly lower MIC value (15.18 µM) and higher induction of Katushka2S reporter in contrast to the parent structures. Moreover, translation blockage was clearly identified using both in vitro luciferase assay and C14 test. The standard cytotoxicity test revealed a relatively low cytotoxicity of the most active molecules. CONCLUSION: High antibacterial activity in combination with low cytotoxicity was demonstrated for a series of furanocoumarins. Further optimization of the described structures may result in novel and attractive lead compounds with promising antibacterial efficiency.

Deep learning enables rapid identification of potent DDR1 kinase inhibitors.

We have developed a deep generative model, generative tensorial reinforcement learning (GENTRL), for de novo small-molecule design. GENTRL optimizes synthetic feasibility, novelty, and biological activity. We used GENTRL to discover potent inhibitors of discoidin domain receptor 1 (DDR1), a kinase target implicated in fibrosis and other diseases, in 21 days. Four compounds were active in biochemical assays, and two were validated in cell-based assays. One lead candidate was tested and demonstrated favorable pharmacokinetics in mice.

Identification of Novel Antibacterials Using Machine Learning Techniques.

Many pharmaceutical companies are avoiding the development of novel antibacterials due to a range of rational reasons and the high risk of failure. However, there is an urgent need for novel antibiotics especially against resistant bacterial strains. Available in silico models suffer from many drawbacks and, therefore, are not applicable for scoring novel molecules with high structural diversity by their antibacterial potency. Considering this, the overall aim of this study was to develop an efficient in silico model able to find compounds that have plenty of chances to exhibit antibacterial activity. Based on a proprietary screening campaign, we have accumulated a representative dataset of more than 140,000 molecules with antibacterial activity against Escherichia coli assessed in the same assay and under the same conditions. This intriguing set has no analogue in the scientific literature. We applied six in silico techniques to mine these data. For external validation, we used 5,000 compounds with low similarity towards training samples. The antibacterial activity of the selected molecules against E. coli was assessed using a comprehensive biological study. Kohonen-based nonlinear mapping was used for the first time and provided the best predictive power (av. 75.5%). Several compounds showed an outstanding antibacterial potency and were identified as translation machinery inhibitors in vitro and in vivo. For the best compounds, MIC and CC50 values were determined to allow us to estimate a selectivity index (SI). Many active compounds have a robust IP position.

2-Pyrazol-1-yl-thiazole derivatives as novel highly potent antibacterials.

The present report describes our efforts to identify new structural classes of compounds having promising antibacterial activity using previously published double-reporter system pDualrep2. This semi-automated high-throughput screening (HTS) platform has been applied to perform a large-scale screen of a diverse small-molecule compound library. We have selected a set of more than 125,000 molecules and evaluated them for their antibacterial activity. On the basis of HTS results, eight compounds containing 2-pyrazol-1-yl-thiazole scaffold exhibited moderate-to-high activity against ΔTolC Escherichia coli. Minimum inhibitory concentration (MIC) values for these molecules were in the range of 0.037-8 µg ml-1. The most active compound 8 demonstrated high antibacterial potency (MIC = 0.037 µg ml-1), that significantly exceed that measured for erythromycin (MIC = 2.5 µg ml-1) and was comparable with the activity of levofloxacin (MIC = 0.016 µg ml-1). Unfortunately, this compound showed only moderate selectivity toward HEK293 eukaryotic cell line. On the contrary, compound 7 was less potent (MIC = 0.8 µg ml-1) but displayed only slight cytotoxicity. Thus, 2-pyrazol-1-yl-thiazoles can be considered as a valuable starting point for subsequent optimization and morphing.

Are We Opening the Door to a New Era of Medicinal Chemistry or Being Collapsed to a Chemical Singularity?

The paradigm of "drug-like-ness" dramatically altered the behavior of the medicinal chemistry community for a long time. In recent years, scientists have empirically found a significant increase in key properties of drugs that have moved structures closer to the periphery or the outside of the rule-of-five "cage". Herein, we show that for the past decade, the number of molecules claimed in patent records by major pharmaceutical companies has dramatically decreased, which may lead to a "chemical singularity". New compounds containing fragments with increased 3D complexity are generally larger, slightly more lipophilic, and more polar. A core difference between this study and recently published papers is that we consider the nature and quality of sp3-rich frameworks rather than sp3 count. We introduce the original descriptor MCE-18, which stands for medicinal chemistry evolution, 2018, and this measure can effectively score molecules by novelty in terms of their cumulative sp3 complexity.

Identification of N-Substituted Triazolo-azetidines as Novel Antibacterials using pDualrep2 HTS Platform.

AIM AND OBJECTIVE: Antibiotic resistance is a serious constraint to the development of new effective antibacterials. Therefore, the discovery of the new antibacterials remains one of the main challenges in modern medicinal chemistry. This study was undertaken to identify novel molecules with antibacterial activity. MATERIALS AND METHODS: Using our unique double-reporter system, in-house large-scale HTS campaign was conducted for the identification of antibacterial potency of small-molecule compounds. The construction allows us to visually assess the underlying mechanism of action. After the initial HTS and rescreen procedure, luciferase assay, C14-test, determination of MIC value and PrestoBlue test were carried out. RESULTS: HTS rounds and rescreen campaign have revealed the antibacterial activity of a series of Nsubstituted triazolo-azetidines and their isosteric derivatives that has not been reported previously. Primary hit-molecule demonstrated a MIC value of 12.5 µg/mL against E. coli Δ tolC with signs of translation blockage and no SOS-response. Translation inhibition (26%, luciferase assay) was achieved at high concentrations up to 160 µg/mL, while no activity was found using C14-test. The compound did not demonstrate cytotoxicity in the PrestoBlue assay against a panel of eukaryotic cells. Within a series of direct structural analogues bearing the same or bioisosteric scaffold, compound 2 was found to have an improved antibacterial potency (MIC=6.25 µg/mL) close to Erythromycin (MIC=2.5-5 µg/mL) against the same strain. In contrast to the parent hit, this compound was more active and selective, and provided a robust IP position. CONCLUSION: N-substituted triazolo-azetidine scaffold may be used as a versatile starting point for the development of novel active and selective antibacterial compounds.

Entangled Conditional Adversarial Autoencoder for de Novo Drug Discovery.

Modern computational approaches and machine learning techniques accelerate the invention of new drugs. Generative models can discover novel molecular structures within hours, while conventional drug discovery pipelines require months of work. In this article, we propose a new generative architecture, entangled conditional adversarial autoencoder, that generates molecular structures based on various properties, such as activity against a specific protein, solubility, or ease of synthesis. We apply the proposed model to generate a novel inhibitor of Janus kinase 3, implicated in rheumatoid arthritis, psoriasis, and vitiligo. The discovered molecule was tested in vitro and showed good activity and selectivity.

Reinforced Adversarial Neural Computer for de Novo Molecular Design.

In silico modeling is a crucial milestone in modern drug design and development. Although computer-aided approaches in this field are well-studied, the application of deep learning methods in this research area is at the beginning. In this work, we present an original deep neural network (DNN) architecture named RANC (Reinforced Adversarial Neural Computer) for the de novo design of novel small-molecule organic structures based on the generative adversarial network (GAN) paradigm and reinforcement learning (RL). As a generator RANC uses a differentiable neural computer (DNC), a category of neural networks, with increased generation capabilities due to the addition of an explicit memory bank, which can mitigate common problems found in adversarial settings. The comparative results have shown that RANC trained on the SMILES string representation of the molecules outperforms its first DNN-based counterpart ORGANIC by several metrics relevant to drug discovery: the number of unique structures, passing medicinal chemistry filters (MCFs), Muegge criteria, and high QED scores. RANC is able to generate structures that match the distributions of the key chemical features/descriptors (e.g., MW, logP, TPSA) and lengths of the SMILES strings in the training data set. Therefore, RANC can be reasonably regarded as a promising starting point to develop novel molecules with activity against different biological targets or pathways. In addition, this approach allows scientists to save time and covers a broad chemical space populated with novel and diverse compounds.

New copper(II) thiohydantoin complexes: Synthesis, characterization, and assessment of their interaction with bovine serum albumin and DNA.

New copper(II) complexes of 2-alkylthio-5-arylmethylene-4H-imidazolin-4-ones: (5Z)-2-(methylsulfanyl)-3-(prop-2-en-1-yl)-5-(pyridin-2-ylmethylidene)-3,5-dihydro-4H-imidazol-4-one) (1a), (5Z,5'Z)-2,2'-(ethan-1,2-diyldisulfanyldiyl)bis(5-(2-pyridilmethylen)-3-allyl-3,5-dihydo-4Н-imidazole-4-one) (2a) and (5Z,5'Z)-3,3'-hexan-1,6-diylbis[5-(2-pyridilmethylen)-2-methylthiotetrahydro-4Н-imidazole-4-one)] (3a) were synthesized as possible anticancer drugs. Their structures were characterized by 1H NMR spectroscopy, elemental analysis, and X-ray crystallography. The composition of the complexes were found for 1a (Cu:L=1:1), 2a (Cu:L=2:1), and 3a (Cu:L=2:1). The chelation constants were found by competitive complexation with ethylenediamine tetraacetate: 1a (6.7±0.6)×1015M-1, 2a=(4.9±0.4)×1019M-2, and 3a (5.7±0.5)×1019M-2. Supramolecular binding with calf thymus DNA by competitive ethidium bromide quenching was made for complex 2a as the most promising anticancer model, the Stern-Volmer constants were found to be KSV=(8.0±0.4)×106M-1, Kq=(6.5±0.4)×105M-1. The binding of the complex 2a to BSA was made by the Scatchard method, the value of the constant is Kb=(1.9±0.2)×106M-1.