Target-specific novel molecules with their recipe: Incorporating synthesizability in the design process.

Application of Artificial intelligence (AI) in drug discovery has led to several success stories in recent times. While traditional methods mostly relied upon screening large chemical libraries for early-stage drug-design, de novo design can help identify novel target-specific molecules by sampling from a much larger chemical space. Although this has increased the possibility of finding diverse and novel molecules from previously unexplored chemical space, this has also posed a great challenge for medicinal chemists to synthesize at least some of the de novo designed novel molecules for experimental validation. To address this challenge, in this work, we propose a novel forward synthesis-based generative AI method, which is used to explore the synthesizable chemical space. The method uses a structure-based drug design framework, where the target protein structure and a target-specific seed fragment from co-crystal structures can be the initial inputs. A random fragment from a purchasable fragment library can also be the input if a target-specific fragment is unavailable. Then a template-based forward synthesis route prediction and molecule generation is performed in parallel using the Monte Carlo Tree Search (MCTS) method where, the subsequent fragments for molecule growth can again be obtained from a purchasable fragment library. The rewards for each iteration of MCTS are computed using a drug-target affinity (DTA) model based on the docking pose of the generated reaction intermediates at the binding site of the target protein of interest. With the help of the proposed method, it is now possible to overcome one of the major obstacles posed to the AI-based drug design approaches through the ability of the method to design novel target-specific synthesizable molecules.

pBRICS: A Novel Fragmentation Method for Explainable Property Prediction of Drug-Like Small Molecules.

Generative artificial intelligence algorithms have shown to be successful in exploring large chemical spaces and designing novel and diverse molecules. There has been considerable interest in developing predictive models using artificial intelligence for drug-like properties, which can potentially reduce the late-stage attrition of drug candidates or predict the properties of novel AI-designed molecules. Concurrently, it is important to understand the contribution of functional groups toward these properties and modify them to obtain property-optimized lead compounds. As a result, there is an increasing interest in the development of explainable property prediction models. However, current explainable approaches are mostly atom-based, where, often, only a fraction of a fragment is shown to be significant. To address the above challenges, we have developed a novel domain-aware molecular fragmentation approach termed post-processing of BRICS (pBRICS), which can fragment small molecules into their functional groups. Multitask models were developed to predict various properties, including the absorption, distribution, metabolism, excretion, and toxicity (ADMET) properties. The fragment importance was explained using the gradient-weighted class activation mapping (Grad-CAM) approach. The method was validated on data sets of experimentally available matched molecular pairs (MMPs). The explanations from the model can be useful for medicinal chemists to identify the fragments responsible for poor drug-like properties and optimize the molecule. The explainability approach was also used to identify the reason behind false positive and false negative MMP predictions. Based on evidence from the existing literature and our analysis, some of these mispredictions were justified. We propose that the quantity, quality, and diversity of the training data will improve the accuracy of property prediction algorithms for novel molecules.

De novo design of anti-tuberculosis agents using a structure-based deep learning method.

Mycobacterium tuberculosis (Mtb) is a pathogen of major concern due to its ability to withstand both first- and second-line antibiotics, leading to drug resistance. Thus, there is a critical need for identification of novel anti-tuberculosis agents targeting Mtb-specific proteins. The ceaseless search for novel antimicrobial agents to combat drug-resistant bacteria can be accelerated by the development of advanced deep learning methods, to explore both existing and uncharted regions of the chemical space. The adaptation of deep learning methods to under-explored pathogens such as Mtb is a challenging aspect, as most of the existing methods rely on the availability of sufficient target-specific ligand data to design novel small molecules with optimized bioactivity. In this work, we report the design of novel anti-tuberculosis agents targeting the Mtb chorismate mutase protein using a structure-based drug design algorithm. The structure-based deep learning method relies on the knowledge of the target protein's binding site structure alone for conditional generation of novel small molecules. The method eliminates the need for curation of a high-quality target-specific small molecule dataset, which remains a challenge even for many druggable targets, including Mtb chorismate mutase. Novel molecules are proposed, that show high complementarity to the target binding site. The graph attention model could identify the probable key binding site residues, which influenced the conditional molecule generator to design new molecules with pharmacophoric features similar to the known inhibitors.

Sonochemical synthesis and biological evaluation of isoquinolin-1(2H)-one/isoindolin-1-one derivatives: Discovery of a positive ago-allosteric modulator (PAAM) of 5HT2CR.

Efforts have been devoted for the discovery and development of positive allosteric modulators (PAMs) of 5-HT2CR because of their potential advantages over the orthosteric agonist like Lorcaserin that was withdrawn from the market. On the other hand, pursuing a positive ago-allosteric modulator (PAAM) is considered as beneficial particularly when an agonist is not capable of affecting the potency of the endogenous agonist sufficiently. In search of a suitable PAAM of 5-HT2CR we adopted an in silico based approach that indicated the potential of the 3-(1-hydroxycycloalkyl) substituted isoquinolin-1-one derivatives against the 5-HT2CR as majority of these molecules interacted with the site other than that of Lorcaserin with superior docking scores. These compounds along with the regioisomeric 3-methyleneisoindolin-1-one derivatives were prepared via the Cu(OAc)2 catalyzed coupling of 2-iodobenzamide with 1-ethynylcycloalkanol under ultrasound irradiation. According to the in vitro studies, most of these compounds were not only found to be potent and selective agonists but also emerged as PAAM of 5-HT2CR whereas Lorcaserin did not show PAAM activities. According to the SAR study the isoquinolin-1(2H)-ones appeared as better PAAM than isoindolin-1-ones whereas the presence of hydroxyl group appeared to be crucial for the activity. With the potent PAAM activity for 5-HT2CR (EC50 = 1 nM) and 107 and 86-fold selectivity towards 5-HT2C over 5-HT2A and 5-HT2B the compound 4i was identified as a hit molecule. The compound showed good stability in male BALB/c mice brain homogenate (∼85 % remaining after 2 h), moderate stability in the presence of rat liver microsomes (42 % remaining after 1 h) and acceptable PK properties with fast reaching in the brain maintaining âˆ¼ 1:1 brain/plasma concentration ratio. The compound at a dose of 50 mg/kg exhibited decreased trend in the food intake starting from day 3 in S.D. rats, which reached significant by 5th day, and the effect was comparable to Lorcaserin (10 mg/kg) on day 5. Thus, being the first example of PAAM of 5-HT2CR the compound 4i is of further medicinal interest.

An interpretable machine learning model for selectivity of small-molecules against homologous protein family.

Aim: In the early stages of drug discovery, various experimental and computational methods are used to measure the specificity of small molecules against a target protein. The selectivity of small molecules remains a challenge leading to off-target side effects. Methods: We have developed a multitask deep learning model for predicting the selectivity on closely related homologs of the target protein. The model has been tested on the Janus-activated kinase and dopamine receptor families of proteins. Results & conclusion: The feature-based representation (extended connectivity fingerprint 4) with Extreme Gradient Boosting performed better when compared with deep neural network models in most of the evaluation metrics. Both the Extreme Gradient Boosting and deep neural network models outperformed the graph-based models. Furthermore, to decipher the model decision on selectivity, the important fragments associated with each homologous protein were identified.

An In Silico Explainable Multiparameter Optimization Approach for De Novo Drug Design against Proteins from the Central Nervous System.

The aim of drug design and development is to produce a drug that can inhibit the target protein and possess a balanced physicochemical and toxicity profile. Traditionally, this is a multistep process where different parameters such as activity and physicochemical and pharmacokinetic properties are optimized sequentially, which often leads to high attrition rate during later stages of drug design and development. We have developed a deep learning-based de novo drug design method that can design novel small molecules by optimizing target specificity as well as multiple parameters (including late-stage parameters) in a single step. All possible combinations of parameters were optimized to understand the effect of each parameter over the other parameters. An explainable predictive model was used to identify the molecular fragments responsible for the property being optimized. The proposed method was applied against the human 5-hydroxy tryptamine receptor 1B (5-HT1B), a protein from the central nervous system (CNS). Various physicochemical properties specific to CNS drugs were considered along with the target specificity and blood-brain barrier permeability (BBBP), which act as an additional challenge for CNS drug delivery. The contribution of each parameter toward molecule design was identified by analyzing the properties of generated small molecules from optimization of all possible parameter combinations. The final optimized generative model was able to design similar inhibitors compared to known inhibitors of 5-HT1B. In addition, the functional groups of the generated small molecules that guide the BBBP predictive model were identified through feature attribution techniques.

De Novo Structure-Based Drug Design Using Deep Learning.

In recent years, deep learning-based methods have emerged as promising tools for de novo drug design. Most of these methods are ligand-based, where an initial target-specific ligand data set is necessary to design potent molecules with optimized properties. Although there have been attempts to develop alternative ways to design target-specific ligand data sets, availability of such data sets remains a challenge while designing molecules against novel target proteins. In this work, we propose a deep learning-based method, where the knowledge of the active site structure of the target protein is sufficient to design new molecules. First, a graph attention model was used to learn the structure and features of the amino acids in the active site of proteins that are experimentally known to form protein-ligand complexes. Next, the learned active site features were used along with a pretrained generative model for conditional generation of new molecules. A bioactivity prediction model was then used in a reinforcement learning framework to optimize the conditional generative model. We validated our method against two well-studied proteins, Janus kinase 2 (JAK2) and dopamine receptor D2 (DRD2), where we produce molecules similar to the known inhibitors. The graph attention model could identify the probable key active site residues, which influenced the conditional molecule generator to design new molecules with pharmacophoric features similar to the known inhibitors.

De novo design of new chemical entities for SARS-CoV-2 using artificial intelligence.

Background: The novel coronavirus SARS-CoV-2 has severely affected the health and economy of several countries. Multiple studies are in progress to design novel therapeutics against the potential target proteins in SARS-CoV-2, including 3CL protease, an essential protein for virus replication. Materials & methods: In this study we employed deep neural network-based generative and predictive models for de novo design of small molecules capable of inhibiting the 3CL protease. The generative model was optimized using transfer learning and reinforcement learning to focus around the chemical space corresponding to the protease inhibitors. Multiple physicochemical property filters and virtual screening score were used for the final screening. Conclusion: We have identified 33 potential compounds as ideal candidates for further synthesis and testing against SARS-CoV-2.

Accelerating De Novo Drug Design against Novel Proteins Using Deep Learning.

In the world plagued by the emergence of new diseases, it is essential that we accelerate the drug design process to develop new therapeutics against them. In recent years, deep learning-based methods have shown some success in ligand-based drug design. Yet, these methods face the problem of data scarcity while designing drugs against a novel target. In this work, the potential of deep learning and molecular modeling approaches was leveraged to develop a drug design pipeline, which can be useful for cases where there is limited or no availability of target-specific ligand datasets. Inhibitors of the homologues of the target protein were screened at the active site of the target protein to create an initial target-specific dataset. Transfer learning was used to learn the features of the target-specific dataset. A deep predictive model was utilized to predict the docking scores of newly designed molecules. Both these models were combined using reinforcement learning to design new chemical entities with an optimized docking score. The pipeline was validated by designing inhibitors against the human JAK2 protein, where none of the existing JAK2 inhibitors were used for training. The ability of the method to reproduce existing molecules from the validation dataset and design molecules with better binding energy demonstrates the potential of the proposed approach.

Elucidation of an essential function of the unique charged domain of Plasmodium topoisomerase III.

Topoisomerase III (TopoIII) along with RecQ helicases are required for the resolution of abnormal DNA structures that result from the stalling of replication forks. Sequence analyses have identified a putative TopoIII in the Plasmodium falciparum genome (PfTopoIII). PfTopoIII shows dual nuclear and mitochondrial localization. The expression and association of PfTopoIII with mtDNA is tightly linked to the asexual replication of the parasite. In this study, we observed that PfTopoIII physically interacts with PfBlm and PfWrn. Sequence alignment and domain analyses have revealed that it contains a unique positively charged region, spanning 85 amino acids, within domain II. A molecular dynamics simulation study revealed that this unstructured domain communicates with DNA and attains a thermodynamically stable state upon DNA binding. Here, we found that the association between PfTopoIII and the mitochondrial genome is negatively affected by the absence of the charged domain. Our study shows that PfTOPOIII can completely rescue the slow growth phenotype of the ΔtopoIII strain in Saccharomyces cerevisiae, but neither PfY421FtopoIII (catalytic-active site mutant) nor Pf(Δ259-337)topoIII (charged region deletion mutant) can functionally complement ScTOPOIII. Hydroxyurea (HU) led to stalling of the replication fork during the S phase, caused moderate toxicity to the growth of P. falciparum, and was associated with concomitant transcriptional upregulation of PfTOPOIII. In addition, ectopic expression of PfTOPOIII reversed HU-induced toxicity. Interestingly, the expression of Pf(Δ259-337)topoIII failed to reverse HU-mediated toxicity. Taken together, our results establish the importance of TopoIII during Plasmodium replication and emphasize the essential requirement of the charged domain in PfTopoIII function.

Network analysis of hydroxymethylbilane synthase dynamics.

Hydroxymethylbilane synthase (HMBS) is one of the key enzymes of the heme biosynthetic pathway that catalyzes porphobilinogen to form the linear tetrapyrrole 1-hydroxymethylbilane through four intermediate steps. Mutations in the human HMBS (hHMBS) can lead to acute intermittent porphyria (AIP), a lethal metabolic disorder. The molecular basis of importance of the amino acid residues at the catalytic site of hHMBS has been well studied. However, the role of non-active site residues toward the activity of the enzyme and hence the association of their mutations with AIP is not known. Network-based analyses of protein structures provide a systems approach to understand the correlations of the residues through a series of inter-residue interactions. We analyzed the dynamic network representation of HMBS protein derived from five molecular dynamics trajectories corresponding to the five steps of pyrrole polymerization. We analyzed the network clusters for each stage and identified the amino acid residues and interactions responsible for the structural stability and catalytic function of the protein. The analysis of high betweenness nodes and interaction paths from the active site help in understanding the molecular basis of the effect of non-active site AIP-causing mutations on the catalytic activity.

Computational modeling of the catalytic mechanism of hydroxymethylbilane synthase.

Hydroxymethylbilane synthase (HMBS), the third enzyme in the heme biosynthesis pathway, catalyzes the formation of 1-hydroxymethylbilane (HMB) by a stepwise polymerization of four molecules of porphobilinogen (PBG) using the dipyrromethane (DPM) cofactor. The mechanism by which HMBS polymerizes four units of PBG has not been elucidated to date. In vitro and in silico studies on HMBS have suggested certain residues with catalytic importance, but their specific role in the catalysis is unclear. To understand the catalytic mechanism of HMBS, quantum mechanical (QM) calculations were performed on model systems obtained from the active site of the human HMBS enzyme. The addition of one molecule of PBG to the DPM cofactor is carried out in four steps: (1) protonation of the substrate, PBG; (2) deamination of PBG; (3) electrophilic addition of the deaminated substrate to the terminal pyrrole ring of the enzyme-bound DPM cofactor and (4) deprotonation of the carbon atom at the α-position of the second ring of DPM. Based on the energy profiles from the QM calculations on cluster models, R26 is proposed to be the best suitable proton donor to the PBG moiety, which aids in the deamination of the substrate. During the electrophilic addition step, the intermediate formed is stabilized by the carboxylate side chain of the D99 residue. In the final deprotonation step, an extra proton from the second ring of DPM is transferred to R26 via the carboxylate side chain of D99, thus completing one cycle of the catalytic mechanism. The residues in the cluster model seem to play an important role in obtaining accurate energy barriers. All the stationary points along the reaction pathway have been characterized using QM calculations. The rate limiting step for the complete mechanism is found to be the deamination of the PBG moiety. The results of this study provide a detailed understanding of the catalytic mechanism and would help design future studies aimed at modulating the activity of HMBS.

TPP riboswitch aptamer: Role of Mg2+ ions, ligand unbinding, and allostery.

Riboswitches are non-coding RNAs that regulate gene expression in response to the binding of metabolites. Their abundance in bacteria makes them ideal drug targets. The prokaryotic thiamine pyrophosphate (TPP) riboswitch regulates gene expression in a wide range of bacteria by undergoing conformational changes in response to the binding of TPP. Although an experimental structure for the aptamer domain of the riboswitch is now available, details of the conformational changes that occur during the binding of the ligand, and the factors that govern these conformational changes, are still not clear. This study employs microsecond-scale molecular dynamics simulations to provide insights into the functioning of the riboswitch aptamer in atomistic detail. A mechanism for the transmission of conformational changes from the ligand-binding site to the P1 switch helix is proposed. Mg2+ ions in the binding site play a critical role in anchoring the ligand to the riboswitch. Finally, modeling the egress of TPP from the binding site reveals a two-step mechanism for TPP unbinding. Findings from this study can motivate the design of future studies aimed at modulating the activity of this drug target.

2-[2-(4-(trifluoromethyl)phenylamino)thiazol-4-yl]acetic acid (Activator-3) is a potent activator of AMPK.

AMPK is considered as a potential high value target for metabolic disorders. Here, we present the molecular modeling, in vitro and in vivo characterization of Activator-3, 2-[2-(4-(trifluoromethyl)phenylamino)thiazol-4-yl]acetic acid, an AMP mimetic and a potent pan-AMPK activator. Activator-3 and AMP likely share common activation mode for AMPK activation. Activator-3 enhanced AMPK phosphorylation by upstream kinase LKB1 and protected AMPK complex against dephosphorylation by PP2C. Molecular modeling analyses followed by in vitro mutant AMPK enzyme assays demonstrate that Activator-3 interacts with R70 and R152 of the CBS1 domain on AMPK γ subunit near AMP binding site. Activator-3 and C2, a recently described AMPK mimetic, bind differently in the γ subunit of AMPK. Activator-3 unlike C2 does not show cooperativity of AMPK activity in the presence of physiological concentration of ATP (2 mM). Activator-3 displays good pharmacokinetic profile in rat blood plasma with minimal brain penetration property. Oral treatment of High Sucrose Diet (HSD) fed diabetic rats with 10 mg/kg dose of Activator-3 once in a day for 30 days significantly enhanced glucose utilization, improved lipid profiles and reduced body weight, demonstrating that Activator-3 is a potent AMPK activator that can alleviate the negative metabolic impact of high sucrose diet in rat model.

Human hydroxymethylbilane synthase: Molecular dynamics of the pyrrole chain elongation identifies step-specific residues that cause AIP.

Hydroxymethylbilane synthase (HMBS), the third enzyme in the heme biosynthetic pathway, catalyzes the head-to-tail condensation of four molecules of porphobilinogen (PBG) to form the linear tetrapyrrole 1-hydroxymethylbilane (HMB). Mutations in human HMBS (hHMBS) cause acute intermittent porphyria (AIP), an autosomal-dominant disorder characterized by life-threatening neurovisceral attacks. Although the 3D structure of hHMBS has been reported, the mechanism of the stepwise polymerization of four PBG molecules to form HMB remains unknown. Moreover, the specific roles of each of the critical active-site residues in the stepwise enzymatic mechanism and the dynamic behavior of hHMBS during catalysis have not been investigated. Here, we report atomistic studies of HMB stepwise synthesis by using molecular dynamics (MD) simulations, mutagenesis, and in vitro expression analyses. These studies revealed that the hHMBS active-site loop movement and cofactor turn created space for the elongating pyrrole chain. Twenty-seven residues around the active site and water molecules interacted to stabilize the large, negatively charged, elongating polypyrrole. Mutagenesis of these active-site residues altered the binding site, hindered cofactor binding, decreased catalysis, impaired ligand exit, and/or destabilized the enzyme. Based on intermediate stages of chain elongation, R26 and R167 were the strongest candidates for proton transfer to deaminate the incoming PBG molecules. Unbiased random acceleration MD simulations identified R167 as a gatekeeper and facilitator of HMB egress through the space between the enzyme's domains and the active-site loop. These studies identified the specific active-site residues involved in each step of pyrrole elongation, thereby providing the molecular bases of the active-site mutations causing AIP.

A novel SIRT1 inhibitor, 4bb induces apoptosis in HCT116 human colon carcinoma cells partially by activating p53.

The NAD+-dependent protein deacetylase SIRT1 has emerged as an important target for epigenetic therapeutics of colon cancer as its increased expression is associated with cancer progression. Additionally, SIRT1 represses p53 function via deacetylation, promoting tumor growth. Therefore, inhibition of SIRT1 is of great therapeutic interest for the treatment of colon cancer. Here, we report discovery of a novel quinoxaline based small molecule inhibitor of human SIRT1, 4bb, investigated its effect on viability of colon cancer cells and molecular mechanism of action. In vitro, 4bb is a significantly more potent SIRT1 inhibitor, compared to ß-naphthols such as sirtinol, cambinol. Increasing concentration of 4bb decrease viability of colon cancer cells but, does not affect the viability of normal dermal fibroblasts depicting cancer cell specificity. Further, 4bb treatment increased p53 acetylation, Bax expression and induced caspase 3 cleavage suggesting that the death of HCT116 colon cancer cells occur through intrinsic pathway of apoptosis. Overall, our results presents 4bb as a new class of human SIRT1 inhibitor and suggest that inhibition of SIRT1 by 4bb induces apoptosis of colon cancer cells at least in part via activating p53 by preventing p53 deacetylation, increasing Bax expression and inducing caspases. Therefore, this molecule provide an opportunity for lead optimization and may help in development of novel, non-toxic epigenetic therapeutics for colon cancer.

Structural insights into E. coli porphobilinogen deaminase during synthesis and exit of 1-hydroxymethylbilane.

Porphobilinogen deaminase (PBGD) catalyzes the formation of 1-hydroxymethylbilane (HMB), a crucial intermediate in tetrapyrrole biosynthesis, through a step-wise polymerization of four molecules of porphobilinogen (PBG), using a unique dipyrromethane (DPM) cofactor. Structural and biochemical studies have suggested residues with catalytic importance, but their specific role in the mechanism and the dynamic behavior of the protein with respect to the growing pyrrole chain remains unknown. Molecular dynamics simulations of the protein through the different stages of pyrrole chain elongation suggested that the compactness of the overall protein decreases progressively with addition of each pyrrole ring. Essential dynamics showed that domains move apart while the cofactor turn region moves towards the second domain, thus creating space for the pyrrole rings added at each stage. Residues of the flexible active site loop play a significant role in its modulation. Steered molecular dynamics was performed to predict the exit mechanism of HMB from PBGD at the end of the catalytic cycle. Based on the force profile and minimal structural changes the proposed path for the exit of HMB is through the space between the domains flanking the active site loop. Residues reported as catalytically important, also play an important role in the exit of HMB. Further, upon removal of HMB, the structure of PBGD gradually relaxes to resemble its initial stage structure, indicating its readiness to resume a new catalytic cycle.

Computational investigation of the effect of thermal perturbation on the mechanical unfolding of titin I27.

The emergence of single-molecule force measurement experiments has facilitated a better understanding of protein folding pathways and the thermodynamics involved. Computational methods such as steered molecular dynamics (SMD) simulations are helpful in providing atomistic level information on the unfolding pathways. Recent experimental studies have showed that combinations of single-molecule experiments with traditional methods such as chemical and/or thermal denaturation yield additional insights into the folding phenomenon. In this study, we report results from extensive computations (a total of about 60 SMD simulations with a total length of about 0.4 µs) that address the effect of thermal perturbation on the mechanical stability of the I27 domain of the protein titin. A wide range of temperatures (280-340 K) were considered for the pulling, which was done at both constant velocity and constant force using SMD simulations. Good agreement with experimental data, such as for the trends in changes in average force and the maximum force with respect to the temperature, was obtained. This study identifies two competing pathways for the mechanical unfolding of I27, and illustrates the significance of combining various techniques to examine protein folding.