Fragment-hopping-based discovery of a novel chemical series of proto-oncogene PIM-1 kinase inhibitors

Comprehensive Insights that Targeting PIM for Cancer Therapy: Prospects and Obstacles

Proviral integration sitea for Moloney-murine leukemia virus (PIM) kinases are a family of highly conserved serine/tyrosine kinases consisting of three members, PIM-1, PIM-2, and PIM-3. These kinases regulate a wide range of substrates through phosphorylation and affect key cellular processes such as transcription, translation, proliferation, apoptosis, and energy metabolism. Several PIM inhibitors are currently undergoing clinical trials, such as a phase I clinical trial of Uzanserti (5) for the treatment of relapsed diffuse large B-cell lymphoma that has been completed. The current focus encompasses the structural and biological characterization of PIM, ongoing research progress on small-molecule inhibitors undergoing clinical trials, and evaluation analysis of persisting challenges in this field. Additionally, the design and discovery of small-molecule inhibitors targeting PIM in recent years have been explored, with a particular emphasis on medicinal chemistry, aiming to provide valuable insights for the future development of PIM inhibitors.

On the ability of machine learning methods to discover novel scaffolds

The recent advances in the application of machine learning to drug discovery have made it a 'hot topic' for research, with hundreds of academic groups and companies integrating machine learning into their drug discovery projects. Nevertheless, there remains great uncertainty regarding the most appropriate ways to evaluate the relative performance of these powerful methods against more traditional cheminformatics approaches, and many pitfalls remain for the unwary. In 2020, researchers at MIT (Stokes et al., Cell 180(4), 688-702, 2020) reported the discovery of a new compound with antibacterial activity, halicin, through the use of a neural network machine learning method. A robust ability to identify new active chemotypes through computational methods would be very useful. In this study, we have used the Stokes et al. dataset to compare the performance of this method to two other approaches, Mapping of Activity Through Dichotomic Scores (MADS) by Todeschini et al. (J Chemom 32(4):e2994, 2018) and Random Matrix Theory (RMT) by Lee et al. (Proc Natl Acad Sci 116(9):3373-3378, 2019). Our results demonstrate that all three methods are capable of predicting halicin as an active antibacterial compound, but that this result is dependent on the dataset composition, pre-processing and the molecular fingerprint used. We have further assessed overall performance as determined by several performance metrics. We also investigated the scaffold hopping potential of the methods by modifying the dataset by removal of the β-lactam and fluoroquinolone chemotypes. MADS and RMT are able to identify actives in the test set that contained these substructures. This ability arises because of high scoring fragments of the withheld chemotypes that are in common with other active antibiotic classes. Interestingly, MADS is relatively better compared to the other two methods based on general predictive performance.

Fragment hopping protocol for the design of small-molecule protein-protein interaction inhibitors

Fragment-based ligand discovery (FBLD) is one of the most successful approaches to designing small-molecule protein-protein interaction (PPI) inhibitors. The incorporation of computational tools to FBLD allows the exploration of chemical space in a time- and cost-efficient manner. Herein, a computational protocol for the development of small-molecule PPI inhibitors using fragment hopping, a fragment-based de novo design approach, is described and a case study is presented to illustrate the efficiency of this protocol. Fragment hopping facilitates the design of PPI inhibitors from scratch solely based on key binding features in the PPI complex structure. This approach is an open system that enables the inclusion of different state-of-the-art programs and softwares to improve its performances.

FragExplorer: GRID-Based Fragment Growing and Replacement

Understanding which chemical modifications can be made to known ligands is a key aspect of structure-based drug design and one that was pioneered by the software GRID. We developed FragExplorer with the explicit aim of showing GRID users which fragments would best match the GRID molecular interaction fields in a protein binding site, given a bound ligand as a starting point. Users can grow ligands or replace existing moieties; the R-Group Exploration mode identifies all potential R-Groups and searches for replacements automatically; the Scaffold Exploration mode does the same for all potential scaffolds. For a ligand with three points of variation, R-Group Exploration will typically explore a chemical space of 10¹⁶ potential molecules; including Scaffold Exploration increases this to 10²². FragExplorer was designed to be integrated within an interactive 3D Editor/Designer; therefore, the speed of computation was an important consideration; a typical fragment search takes 20 seconds. In a fragment reprediction test, FragExplorer demonstrates an overall fragment retrieval rate of 55%, increasing to 69% for smaller fragments. At a 90% substructural match, the retrieval rate increases to ∼80%. We also show how the approach could have been used to hop from olmesartan to azilsartan or to optimize a p38 MAP kinase lead to a compound that bears similarity to a known nanomolar inhibitor.

Deep scaffold hopping with multimodal transformer neural networks

Scaffold hopping is a central task of modern medicinal chemistry for rational drug design, which aims to design molecules of novel scaffolds sharing similar target biological activities toward known hit molecules. Traditionally, scaffolding hopping depends on searching databases of available compounds that can't exploit vast chemical space. In this study, we have re-formulated this task as a supervised molecule-to-molecule translation to generate hopped molecules novel in 2D structure but similar in 3D structure, as inspired by the fact that candidate compounds bind with their targets through 3D conformations. To efficiently train the model, we curated over 50 thousand pairs of molecules with increased bioactivity, similar 3D structure, but different 2D structure from public bioactivity database, which spanned 40 kinases commonly investigated by medicinal chemists. Moreover, we have designed a multimodal molecular transformer architecture by integrating molecular 3D conformer through a spatial graph neural network and protein sequence information through Transformer. The trained DeepHop model was shown able to generate around 70% molecules having improved bioactivity together with high 3D similarity but low 2D scaffold similarity to the template molecules. This ratio was 1.9 times higher than other state-of-the-art deep learning methods and rule- and virtual screening-based methods. Furthermore, we demonstrated that the model could generalize to new target proteins through fine-tuning with a small set of active compounds. Case studies have also shown the advantages and usefulness of DeepHop in practical scaffold hopping scenarios.

Fragment hopping-based discovery of novel sulfinylacetamide-diarylpyrimidines (DAPYs) as HIV-1 nonnucleoside reverse transcriptase inhibitors

The fragment hopping approach is widely applied in drug development. A series of diarylpyrimidines (DAPYs) were obtained by hopping the thioacetamide scaffold to novel human immunodeficiency virus type 1 (HIV-1) nonnucleoside reverse transcriptase inhibitors (NNRTIs) to address the cytotoxicity issue of Etravirine and Rilpivirine. Although the new compounds (11a-l) in the first-round optimization possessed less potent anti-viral activity, they showed much lower cytotoxicity. Further optimization on the sulfur led to the sulfinylacetamide-DAPYs exhibiting improved anti-viral activity and a higher selectivity index especially toward the K103N mutant strain. The most potent compound 12a displayed EC₅₀ values of 0.0249 μM against WT and 0.0104 μM against the K103N mutant strain, low cytotoxicity (CC₅₀ > 221 μM) and a high selectivity index (SI _WT > 8873, SI _K103N > 21186). In addition, this compound showed a favorable in vitro microsomal stability across species. Computational study predicted the binding models of these potent compounds with HIV-1 reverse transcriptase thus providing further insights for new developments.

Impact of Scaffold Exploration on Novel Dual-Acting Histone Deacetylases and Phosphodiesterase 5 Inhibitors for the Treatment of Alzheimer's Disease

A novel systems therapeutics approach, involving simultaneous inhibition of phosphodiesterase 5 (PDE5) and histone deacetylase (HDAC), has been validated as a potentially novel therapeutic strategy for the treatment of Alzheimer's disease (AD). First-in-class dual inhibitors bearing a sildenafil core have been very recently reported, and the lead molecule 7 has proven this strategy in AD animal models. Because scaffolds may play a critical role in primary activities and ADME-Tox profiling as well as on intellectual property, we have explored alternative scaffolds (vardenafil- and tadalafil-based cores) and evaluated their impact on critical parameters such as primary activities, permeability, toxicity, and in vivo (pharmacokinetics and functional response in hippocampus) to identify a potential alternative lead molecule bearing a different chemotype for in vivo testing.

Protocol for fragment hopping

Fragment hopping is a fragment-based approach to designing biologically active small molecules. The key of this approach is the determination of the minimal pharmacophoric elements in the three-dimensional space. Based on the derived minimal pharmacophoric elements, new fragments with different chemotypes can be generated and positioned to the active site of the target protein. Herein, we detail a protocol for performing fragment hopping. This approach can not only explore a wide chemical space to produce new ligands with novel scaffolds but also characterize and utilize the delicate differences in the active sites between isofunctional proteins to produce new ligands with high target selectivity/specificity.

Intravital biobank and personalized cancer therapy: the correlation with omics

Biobanks have played a decisive role in all aspects of the field of cancer, including pathogenesis, diagnosis, prognosis and treatment. The significance of cancer biobanks is epitomized through the appropriate application of various "-omic" techniques (omics). The mutually motivated relationship between biobanks and omics has intensified the development of cancer research. Human cancer tissues that are maintained in intravital biobanks (or living tissue banks) retain native tumor microenvironment, tissue architecture, hormone responsiveness and cell-to-cell signalling properties. Intravital biobanks replicate the structural complexity and heterogeneity of human cancers, making them an ideal platform for preclinical studies. The application of omics with intravital biobanks renders them more active, which makes it possible for the cancer-related evaluations to be dynamically monitored on a real-time basis. Integrating intravital biobank and modern omics will provide a useful tool for the discovery and development of new drugs or novel therapeutic strategies. More importantly, intravital biobanks may play an essential role in the creation of meaningful patient-tailored therapies as for personalized medicine.