Drug discovery based on artificial intelligence has been in the spotlight recently as it significantly reduces the time and cost required for developing novel drugs. With the advancement of deep learning (DL) technology and the growth of drug-related data, numerous deep-learning-based methodologies are emerging at all steps of drug development processes. In particular, pharmaceutical chemists have faced significant issues with regard to selecting and designing potential drugs for a target of interest to enter preclinical testing. The two major challenges are prediction of interactions between drugs and druggable targets and generation of novel molecular structures suitable for a target of interest. Therefore, we reviewed recent deep-learning applications in drug–target interaction (DTI) prediction and de novo drug design. In addition, we introduce a comprehensive summary of a variety of drug and protein representations, DL models, and commonly used benchmark datasets or tools for model training and testing. Finally, we present the remaining challenges for the promising future of DL-based DTI prediction and de novo drug design.

Implantation of bone-marrow-derived MSCs (mesenchymal stem cells) has emerged as a potential treatment modality for liver failure, but in vivo differentiation of MSCs into functioning hepatocytes and its therapeutic effects have not yet been determined. We investigated MSC differentiation process in a rat model of TAA (thioacetamide)-induced liver cirrhosis. Male Sprague-Dawley rats were administered 0.04% TAA-containing water for 8 weeks, MSCs were injected into the spleen for transsplenic migration into the liver, and liver tissues were examined over 3 weeks. Ingestion of TAA for 8 weeks induced micronodular liver cirrhosis in 93% of rats. Injected MSCs were diffusely engrafted in the liver parenchyma, differentiated into CK19 (cytokeratin 19)- and thy1-positive oval cells and later into albumin-producing hepatocyte-like cells. MSC engraftment rate per slice was measured as 1.0-1.6%. MSC injection resulted in apoptosis of hepatic stellate cells and resultant resolution of fibrosis, but did not cause apoptosis of hepatocytes. Injection of MSCs treated with HGF (hepatocyte growth factor) in vitro for 2 weeks, which became CD90-negative and CK18-positive, resulted in chronological advancement of hepatogenic cellular differentiation by 2 weeks and decrease in anti-fibrotic activity. Early differentiation of MSCs to progenitor oval cells and hepatocytes results in various therapeutic effects, including repair of damaged hepatocytes, intracellular glycogen restoration and resolution of fibrosis. Thus, these results support that the in vivo hepatogenic differentiation of MSCs is related to the beneficial effects of MSCs rather than the differentiated hepatocytes themselves.

Drug discovery typically involves investigation of a set of compounds (e.g. drug screening hits) in terms of target, disease, and bioactivity. CSgator is a comprehensive analytic tool for set-wise interpretation of compounds. It has two unique analytic features of Compound Set Enrichment Analysis (CSEA) and Compound Cluster Analysis (CCA), which allows batch analysis of compound set in terms of (i) target, (ii) bioactivity, (iii) disease, and (iv) structure. CSEA and CCA present enriched profiles of targets and bioactivities in a compound set, which leads to novel insights on underlying drug mode-of-action, and potential targets. Notably, we propose a novel concept of ‘Hit Enriched Assays”, i.e. bioassays of which hits are enriched among a given set of compounds. As an example, we show its utility in revealing drug mode-of-action or identifying hidden targets for anti-lymphangiogenesis screening hits. CSgator is available at http://csgator.ewha.ac.kr, and most analytic results are downloadable.

N-α-acetyltransferase 20 (Naa20), which is a catalytic subunit of the N-terminal acetyltransferase B (NatB) complex, has recently been reported to be implicated in hepatocellular carcinoma (HCC) progression and autophagy, but the underlying mechanism remains unclear. Here, we report that based on bioinformatic analysis of Gene Expression Omnibus and The Cancer Genome Atlas data sets, Naa20 expression is much higher in HCC tumors than in normal tissues, promoting oncogenic properties in HCC cells. Mechanistically, Naa20 inhibits the activity of AMP-activated protein kinase (AMPK) to promote the mammalian target of rapamycin signaling pathway, which contributes to cell proliferation, as well as autophagy, through its N-terminal acetyltransferase (NAT) activity. We further show that liver kinase B1 (LKB1), a major regulator of AMPK activity, can be N-terminally acetylated by NatB in vitro, but also probably by NatB and/or other members of the NAT family in vivo, which may have a negative effect on AMPK activity through downregulation of LKB1 phosphorylation at S428. Indeed, p-LKB1 (S428) and p-AMPK levels are enhanced in Naa20-deficient cells, as well as in cells expressing the nonacetylated LKB1-MPE mutant; moreover, importantly, LKB1 deficiency reverses the molecular and cellular events driven by Naa20 knockdown. Taken together, our findings suggest that N-terminal acetylation of LKB1 by Naa20 may inhibit the LKB1–AMPK signaling pathway, which contributes to tumorigenesis and autophagy in HCC.