From theory to experiment: transformer-based generation enables rapid discovery of novel reactions.

Deep learning methods, such as reaction prediction and retrosynthesis analysis, have demonstrated their significance in the chemical field. However, the de novo generation of novel reactions using artificial intelligence technology requires further exploration. Inspired by molecular generation, we proposed a novel task of reaction generation. Herein, Heck reactions were applied to train the transformer model, a state-of-art natural language process model, to generate 4717 reactions after sampling and processing. Then, 2253 novel Heck reactions were confirmed by organizing chemists to judge the generated reactions. More importantly, further organic synthesis experiments were performed to verify the accuracy and feasibility of representative reactions. The total process, from Heck reaction generation to experimental verification, required only 15 days, demonstrating that our model has well-learned reaction rules in-depth and can contribute to novel reaction discovery and chemical space exploration.

Discovery of small-molecule compounds and natural products against Parkinson's disease: Pathological mechanism and structural modification.

Parkinson's disease (PD) is the second common neurodegenerative disease characterized by movement disorder. The symptoms of PD harm both the physical and mental health of patients. However, the current treatment strategies for PD only alleviate the symptoms but cannot recover the degenerative process of dopaminergic neurons. Therefore, it is necessary to develop novel and safe drugs for the treatment of PD. In this review, we comprehensively summarized the detailed pathological mechanisms and potential drugable targets of PD. The approved anti-PD drugs in clinical use and the drug candidates under clinical trials were also listed. More importantly, the compounds in the drug discovery phase with in vivo anti-PD activities in the recent two decades (2000-2020) were summarized. The structure-activity relationships (SARs) were also analyzed. Additionally, we predicted all the reviewed compounds' blood-brain barrier (BBB) permeability and statistically analyzed their pharmacological targets and in vivo anti-PD testing models. It is hoped that this review can provide practical information for researchers in the field of anti-PD drug discovery and promote their research work.

HDAC1/MAO-B dual inhibitors against Alzheimer's disease: Design, synthesis and biological evaluation of N-propargylamine-hydroxamic acid/o-aminobenzamide hybrids.

A series of N-propargylamine-hydroxamic acid/o-aminobenzamide hybrids inhibitors combining the typical pharmacophores of hydroxamic acid/o-aminobenzamide and propargylamine were designed and synthesized as HDAC1/MAO-B dual inhibitors for the treatment of Alzheimer's disease. Most of the hybrids displayed moderate to good MAO-B inhibitory activities. Among them, Hybrid If exhibited the most potent activity against MAO-B and HDAC1 (MAO-B, IC50 = 99.0 nM; HDAC1, IC50 = 21.4 nM) and excellent MAO selectively (MAO-A, IC50 = 9923.0 nM; SI = 100.2). Moreover, compound If significantly reversed Aß1-42-induced PC12 cell damage and decreased the production of intracellular ROS, exhibiting favorable antioxidant activity. More importantly, hybrid If instantly penetrated the BBB and accumulated in brain tissue as well as markedly ameliorated cognitive dysfunction in a Morris water maze ICR mice model. In summary, HDAC1/MAO-B dual inhibitor If is a promising potential agent for the therapy of Alzheimer's disease.

N-Propargylamine-hydroxypyridinone hybrids as multitarget agents for the treatment of Alzheimer's disease.

AD is a progressive brain disorder. Because of the lack of remarkable single-target drugs against neurodegenerative disorders, the multitarget-directed ligand strategy has received attention as a promising therapeutic approach. Herein, we rationally designed twenty-nine hybrids of N-propargylamine-hydroxypyridinone. The designed hybrids possessed excellent iron-chelating activity (pFe3+ = 17.09-22.02) and potent monoamine oxidase B inhibitory effects. Various biological evaluations of the optimal compound 6b were performed step by step, including inhibition screening of monoamine oxidase (hMAO-B IC50 = 0.083 ± 0.001 µM, hMAO-A IC50 = 6.11 ± 0.08 µM; SI = 73.5), prediction of blood-brain barrier permeability and mouse behavioral research. All of these favorable results proved that the N-propargylamine-hydroxypyridinone scaffold is a promising structure for the discovery of multitargeted ligands for AD therapy.

Anti-Tumor Drug Discovery Based on Natural Product ß-Elemene: Anti-Tumor Mechanisms and Structural Modification.

Natural products are important sources for drug discovery, especially anti-tumor drugs. ß-Elemene, the prominent active ingredient extract from the rhizome of Curcuma wenyujin, is a representative natural product with broad anti-tumor activities. The main molecular mechanism of ß-elemene is to inhibit tumor growth and proliferation, induce apoptosis, inhibit tumor cell invasion and metastasis, enhance the sensitivity of chemoradiotherapy, regulate the immune system, and reverse multidrug resistance (MDR). Elemene oral emulsion and elemene injection were approved by the China Food and Drug Administration (CFDA) for the treatment of various cancers and bone metastasis in 1994. However, the lipophilicity and low bioavailability limit its application. To discover better ß-elemene-derived anti-tumor drugs with satisfying drug-like properties, researchers have modified its structure under the premise of not damaging the basic scaffold structure. In this review, we comprehensively discuss and summarize the potential anti-tumor mechanisms and the progress of structural modifications of ß-elemene.

Discovery of natural anti-inflammatory alkaloids: Potential leads for the drug discovery for the treatment of inflammation.

Inflammation is an adaptive response of the immune system to tissue malfunction or homeostatic imbalance. Corticosteroids and non-steroidal anti-inflammatory drugs (NSAIDs) are frequently applied to treat varieties of inflammatory diseases but are associated with gastrointestinal, cardiovascular, and kidney side effects. Developing more effective and less toxic agents remain a challenge for pharmaceutical chemist due to the complexity of the different inflammatory processes. Alkaloids are widely distributed in plants with diverse anti-inflammatory activities, providing various potential lead compounds or candidates for the design and discovery of new anti-inflammatory drug candidates. Therefore, re-examining the anti-inflammatory alkaloid natural products is advisable, bringing more opportunities. In this review, we summarized and described the recent advances of natural alkaloids with anti-inflammatory activities and possible mechanisms in the period from 2009 to 2020. It is hoped that this review of anti-inflammatory alkaloids can provide new ideas for researchers engaged in the related fields and potential lead compounds for the discovery of anti-inflammatory drugs.

Rational design, synthesis and biological evaluation of novel multitargeting anti-AD iron chelators with potent MAO-B inhibitory and antioxidant activity.

A series of (3-hydroxypyridin-4-one)-coumarin hybrids were developed and investigated as potential multitargeting candidates for the treatment of Alzheimer's disease (AD) through the incorporation of iron-chelating and monoamine oxidase B (MAO-B) inhibition. This combination endowed the hybrids with good capacity to inhibit MAO-B as well as excellent iron-chelating effects. The pFe3+ values of the compounds were ranging from 16.91 to 20.16, comparable to more potent than the reference drug deferiprone (DFP). Among them, compound 18d exhibited the most promising activity against MAO-B, with an IC50 value of 87.9 nM. Moreover, compound 18d exerted favorable antioxidant activity, significantly reversed the amyloid-ß1-42 (Aß1-42) induced PC12 cell damage. More importantly, 18d remarkably ameliorated the cognitive dysfunction in a scopolamine-induced mice AD model. In brief, a series of hybrids with potential anti-AD effect were successfully obtained, indicating that the design of iron chelators with MAO-B inhibitory and antioxidant activities is an attractive strategy against AD progression.

Transfer Learning: Making Retrosynthetic Predictions Based on a Small Chemical Reaction Dataset Scale to a New Level.

Effective computational prediction of complex or novel molecule syntheses can greatly help organic and medicinal chemistry. Retrosynthetic analysis is a method employed by chemists to predict synthetic routes to target compounds. The target compounds are incrementally converted into simpler compounds until the starting compounds are commercially available. However, predictions based on small chemical datasets often result in low accuracy due to an insufficient number of samples. To address this limitation, we introduced transfer learning to retrosynthetic analysis. Transfer learning is a machine learning approach that trains a model on one task and then applies the model to a related but different task; this approach can be used to solve the limitation of few data. The unclassified USPTO-380K large dataset was first applied to models for pretraining so that they gain a basic theoretical knowledge of chemistry, such as the chirality of compounds, reaction types and the SMILES form of chemical structure of compounds. The USPTO-380K and the USPTO-50K (which was also used by Liu et al.) were originally derived from Lowe's patent mining work. Liu et al. further processed these data and divided the reaction examples into 10 categories, but we did not. Subsequently, the acquired skills were transferred to be used on the classified USPTO-50K small dataset for continuous training and retrosynthetic reaction tests, and the pretrained accuracy data were simultaneously compared with the accuracy of results from models without pretraining. The transfer learning concept was combined with the sequence-to-sequence (seq2seq) or Transformer model for prediction and verification. The seq2seq and Transformer models, both of which are based on an encoder-decoder architecture, were originally constructed for language translation missions. The two algorithms translate SMILES form of structures of reactants to SMILES form of products, also taking into account other relevant chemical information (chirality, reaction types and conditions). The results demonstrated that the accuracy of the retrosynthetic analysis by the seq2seq and Transformer models after pretraining was significantly improved. The top-1 accuracy (which is the accuracy rate of the first prediction matching the actual result) of the Transformer-transfer-learning model increased from 52.4% to 60.7% with greatly improved prediction power. The model's top-20 prediction accuracy (which is the accuracy rate of the top 20 categories containing actual results) was 88.9%, which represents fairly good prediction in retrosynthetic analysis. In summary, this study proves that transferring learning between models working with different chemical datasets is feasible. The introduction of transfer learning to a model significantly improved prediction accuracy and, especially, assisted in small dataset based reaction prediction and retrosynthetic analysis.

Amide-sulfamide modulators as effective anti-tumor metastatic agents targeting CXCR4/CXCL12 axis.

Breast cancer is the most frequently diagnosed malignancy and the second common cause of death in women worldwide. High mortality in breast cancer is frequently associated with metastatic progression rather than the primary tumor itself. It has been recently identified that the CXCR4/CXCL12 axis plays a pivotal role in breast cancer metastasis, especially in directing metastatic cancer cells to CXCL12-riched organs and tissues. Herein, taking the amide-sulfamide as the lead structure, the second-round structural modifications to the sulfamide structure were performed to obtain more active CXCR4 modulators against tumor metastasis. Both in vivo and in vitro experiments illustrated that compound IIIe possessed potent CXCR4 binding affinity, excellent anti-metastatic and anti-angiogenetic activity against breast cancer. More importantly, in a mouse breast cancer lung metastasis model, compound IIIe exerted a significant inhibitory effect on breast cancer metastasis. Taken together, all these positive results demonstrated that developing of CXCR4 modulators is a promising strategy to mediate breast cancer metastasis.

Discovery of small-molecule candidates against inflammatory bowel disease.

Inflammatory bowel disease (IBD) is a chronic and recurrent inflammatory disease in the gastrointestinal tract emerged as a public health challenge worldwide. IBD exhibits a relapsing and remitting course results in negative impacts on both physical and psychological health of IBD patients. Great efforts have been made during the past few years, but relatively limited drugs are currently available for the management of IBD. Clinically, there is a strong demand for new drugs for the treatment of IBD with better efficacy and lower side effects. This review focuses on the drug discovery process of the anti-IBD agents, aiming to introduce the general characteristics of IBD, as well as systematically summarize the recent advances in the discovery of small-molecule candidates and natural products with promising in vivo potential for the treatment of IBD.