Computational drug design relies on the calculation of binding strength between two biological counterparts especially a chemical compound, i.e., a ligand, and a protein. Predicting the affinity of protein-ligand binding with reasonable accuracy is crucial for drug discovery, and enables the optimization of compounds to achieve better interaction with their target protein. In this paper, we propose a data-driven framework named DeepAtom to accurately predict the protein-ligand binding affinity. With 3D Convolutional Neural Network (3D-CNN) architecture, DeepAtom could automatically extract binding related atomic interaction patterns from the voxelized complex structure. Compared with the other CNN based approaches, our light-weight model design effectively improves the model representational capacity, even with the limited available training data. We carried out validation experiments on the PDBbind v.2016 benchmark and the independent Astex Diverse Set. We demonstrate that the less feature engineering dependent DeepAtom approach consistently outperforms the other baseline scoring methods. We also compile and propose a new benchmark dataset to further improve the model performances. With the new dataset as training input, DeepAtom achieves Pearson's R=0.83 and RMSE=1.23 $pK$ units on the PDBbind v.2016 core set. The promising results demonstrate that DeepAtom models can be potentially adopted in computational drug development protocols such as molecular docking and virtual screening.

In the autoimmune disease multiple sclerosis and its animal model, experimental autoimmune encephalomyelitis (EAE), expansion of pathogenic, myelin-specific Th1 cell populations drives active disease; selectively targeting this process may be the basis for a new therapeutic approach. Previous studies have hinted at a role for protein arginine methylation in immune responses, including T cell–mediated autoimmunity and EAE. However, a conclusive role for the protein arginine methyltransferase (PRMT) enzymes that catalyze these reactions has been lacking. PRMT5 is the main PRMT responsible for symmetric dimethylation of arginine residues of histones and other proteins. PRMT5 drives embryonic development and cancer, but its role in T cells, if any, has not been investigated. In this article, we show that PRMT5 is an important modulator of CD4⁺ T cell expansion. PRMT5 was transiently upregulated during maximal proliferation of mouse and human memory Th cells. PRMT5 expression was regulated upstream by the NF-κB pathway, and it promoted IL-2 production and proliferation. Blocking PRMT5 with novel, highly selective small molecule PRMT5 inhibitors severely blunted memory Th expansion, with preferential suppression of Th1 cells over Th2 cells. In vivo, PRMT5 blockade efficiently suppressed recall T cell responses and reduced inflammation in delayed-type hypersensitivity and clinical disease in EAE mouse models. These data implicate PRMT5 in the regulation of adaptive memory Th cell responses and suggest that PRMT5 inhibitors may be a novel therapeutic approach for T cell–mediated inflammatory disease.

Morphological profiling is a valuable tool in phenotypic drug discovery. The advent of high-throughput automated imaging has enabled the capturing of a wide range of morphological features of cells or organisms in response to perturbations at the single-cell resolution. Concurrently, significant advances in machine learning and deep learning, especially in computer vision, have led to substantial improvements in analyzing large-scale high-content images at high throughput. These efforts have facilitated understanding of compound mechanism of action, drug repurposing, characterization of cell morphodynamics under perturbation, and ultimately contributing to the development of novel therapeutics. In this review, we provide a comprehensive overview of the recent advances in the field of morphological profiling. We summarize the image profiling analysis workflow, survey a broad spectrum of analysis strategies encompassing feature engineering– and deep learning–based approaches, and introduce publicly available benchmark datasets. We place a particular emphasis on the application of deep learning in this pipeline, covering cell segmentation, image representation learning, and multimodal learning. Additionally, we illuminate the application of morphological profiling in phenotypic drug discovery and highlight potential challenges and opportunities in this field.

Interleukin-11 (IL-11), a dominant IL-6 family cytokine, is involved in tumorigenesis, tumor progression and differentiation in colon cancer cells. IL-11 signaling has been recently identified as a potential therapeutic target in colon cancer. Bazedoxifene, a third- generation selective estrogen modulator approved by the Food and Drug Administration (FDA), is a novel inhibitor of IL-11/GP130 signaling discovered by docking modeling. In this study, the inhibition efficacy of bazedoxifene in colon cancer cells and its potential mechanism were investigated in vitro and in vivo by using MTT cell viability assay, BrdU cell proliferation assay, colony formation assay, wound-healing/cell migration assay, immunofluorescence, western blot assay and the mouse xenograft tumor model. Bazedoxifene inhibits phosphorylation of signal transducer and activator of transcription 3 (p-STAT3) and its nuclear translocation induced by IL-11 in colon cancer cells. It also inhibits p-STAT3 induced by IL-6 and IL-11 but not by OSM or STAT1 phosphorylation induced by INF-γ in human colon cancer cells. In addition, bazedoxifene can significantly inhibit phosphorylation of AKT and STAT3 downstream targets. Furthermore, bazedoxifene alone or together with oxaliplatin can significantly induce apoptosis, inhibit cell viability, cell colony formation and cell migration in colon cancer cells. Knock-down of IL-11R can reduce the sensitivity of colon cancer cells to bazedoxifene. IL-11 can reduce the efficacy of oxaliplatin-mediated inhibition of cell viability. Consistent with in vitro findings, bazedoxifene alone also attenuated HCT-15 xenograft tumor burden and reduced p-STAT3, p-AKT and p-ERK in vivo. Its combination with oxaliplatin attenuated DLD-1 xenograft tumor burden and reduced p-STAT3 in vivo. Taken together, these results support bazedoxifene as a novel and effective therapeutic agent targeting IL-11/GP130 signaling for human colorectal cancer therapy.

Purpose: An ongoing challenge in cancer is the management of primary and metastatic brain malignancies. This is partly due to restrictions of the blood-brain barrier and their unique microenvironment. These challenges are most evident in cancers such as lymphoma and melanoma, which are typically responsive to treatment in systemic locations but resistant when established in the brain. We propose interleukin-1 receptor-associated kinase-4 (IRAK-4) as a potential target across these diseases and describe the activity and mechanism of oral IRAK-4 inhibitor CA-4948. Experimental Design: Human primary central nervous system lymphoma (PCNSL) and melanoma brain metastases (MBM) samples were analyzed for expression of IRAK-4 and downstream transcription pathways. We next determined the central nervous system (CNS) applicability of CA-4948 in naïve and tumor-bearing mice using models of PCNSL and MBM. The mechanistic effect on tumors and the tumor microenvironment was then analyzed. Results: Human PCNSL and MBM have high expression of IRAK-4, IRAK-1, and nuclear factor kappa B (NF-κB). This increase in inflammation results in reflexive inhibitory signaling. Similar profiles are observed in immunocompetent murine models. Treatment of tumor-bearing animals with CA-4948 results in the downregulation of mitogen-activated protein kinase (MAPK) signaling in addition to decreased NF-κB. These intracellular changes are associated with a survival advantage. Conclusions: IRAK-4 is an attractive target in PCNSL and MBM. The inhibition of IRAK-4 with CA-4948 downregulates the expression of important transcription factors involved in tumor growth and proliferation. CA-4948 is currently being investigated in clinical trials for relapsed and refractory lymphoma and warrants further translation into PCNSL and MBM.