Weak Microbial Metabolites: a Treasure Trove for Using Biomimicry to Discover and Optimize Drugs

Microbial Metabolites as Ligands to Xenobiotic Receptors: Chemical Mimicry as Potential Drugs of the Future

Xenobiotic receptors, such as the pregnane X receptor, regulate multiple host physiologic pathways including xenobiotic metabolism, certain aspects of cellular metabolism, and innate immunity. These ligand-dependent nuclear factors regulate gene expression via genomic recognition of specific promoters and transcriptional activation of the gene. Natural or endogenous ligands are not commonly associated with this class of receptors; however, since these receptors are expressed in a cell-type specific manner in the liver and intestines, there has been significant recent effort to characterize microbially derived metabolites as ligands for these receptors. In general, these metabolites are thought to be weak micromolar affinity ligands. This journal anniversary minireview focuses on recent efforts to derive potentially nontoxic microbial metabolite chemical mimics that could one day be developed as drugs combating xenobiotic receptor-modifying pathophysiology. The review will include our perspective on the field and recommend certain directions for future research. SIGNIFICANCE STATEMENT: Xenobiotic receptors (XRs) regulate host drug metabolism, cellular metabolism, and immunity. Their presence in host intestines allows them to function not only as xenosensors but also as a response to the complex metabolic environment present in the intestines. Specifically, this review focuses on describing microbial metabolite-XR interactions and the translation of these findings toward discovery of novel chemical mimics as potential drugs of the future for diseases such as inflammatory bowel disease.

Large-scale comparison of machine learning algorithms for target prediction of natural products

Natural products (NPs) and their derivatives are important resources for drug discovery. There are many in silico target prediction methods that have been reported, however, very few of them distinguish NPs from synthetic molecules. Considering the fact that NPs and synthetic molecules are very different in many characteristics, it is necessary to build specific target prediction models of NPs. Therefore, we collected the activity data of NPs and their derivatives from the public databases and constructed four datasets, including the NP dataset, the NPs and its first-class derivatives dataset, the NPs and all its derivatives and the ChEMBL26 compounds dataset. Conditions, including activity thresholds and input features, were explored to access the performance of eight machine learning methods of target prediction of NPs, including support vector machines (SVM), extreme gradient boosting, random forests, K-nearest neighbor, naive Bayes, feedforward neural networks (FNN), convolutional neural networks and recurrent neural networks. As a result, the NPs and all their derivatives datasets were selected to build the best NP-specific models. Furthermore, the consensus models, as well as the voting models, were additionally applied to improve the prediction performance. More evaluations were made on the external validation set and the results demonstrated that (1) the NP-specific model performed better on the target prediction of NPs than the traditional models training on the whole compounds of ChEMBL26. (2) The consensus model of FNN + SVM possessed the best overall performance, and the voting model can significantly improve recall and specificity.

Targeting the Aryl Hydrocarbon Receptor with Microbial Metabolite Mimics Alleviates Experimental Colitis in Mice

Targeting the aryl hydrocarbon receptor (AhR) is an emerging therapeutic strategy for multiple diseases (e.g., inflammatory bowel disease). Thermosporothrix hazakensis microbial metabolite 2-(1'H-indole-3'-carbonyl)-thiazole-4-carboxylic acid methyl ester (ITE) is a putative AhR endogenous ligand. To improve the chemical stability, we synthesized a series of ITE chemical mimics. Using a series of in vitro assays, we identified 2-(1H-indole-3-carbonyl)-N-methyl thiazole-4-carboxamide (ITE-CONHCH₃) as a highly potent (EC₅₀ = 1.6 nM) AhR agonist with high affinity (K_i = 88 nM). ITE-CONHCH₃ triggered AhR nuclear translocation and dimerization of AhR-ARNT, enhanced AhR binding in the CYP1A1 promoter, and induced AhR-regulated genes in an AhR-dependent manner. The metabolic stability of ITE-CONHCH₃ in a cell culture was 10 times higher than that of ITE. Finally, we observed protective effects of ITE-CONHCH₃ in mice with DSS-induced colitis. Overall, we demonstrate and validate a concept of microbial metabolite mimicry in the therapeutic targeting of AhR.

Deciphering structural bases of intestinal and hepatic selectivity in targeting pregnane X receptor with indole-based microbial mimics

Microbial metabolite mimicry is a new concept that promises to deliver compounds that have minimal liabilities and enhanced therapeutic effects in a host. In a previous publication, we have shown that microbial metabolites of L-tryptophan, indoles, when chemically altered, yielded potent anti-inflammatory pregnane X Receptor (PXR)-targeting lead compounds, FKK5 and FKK6, targeting intestinal inflammation. Our aim in this study was to further define structure-activity relationships between indole analogs and PXR, we removed the phenyl-sulfonyl group or replaced the pyridyl residue with imidazolopyridyl of FKK6. Our results showed that while removal of the phenyl-sulfonyl group from FKK6 (now called CVK003) shifts agonist activity away from PXR towards the aryl hydrocarbon receptor (AhR), the imidazolopyridyl addition preserves PXR activity in vitro. However, when these compounds are administered to mice, that unlike the parent molecule, FKK6, they exhibit poor induction of PXR target genes in the intestines and the liver. These data suggest that modifications of FKK6 specifically in the pyridyl moiety can result in compounds with weak PXR activity in vivo. These observations are a significant step forward for understanding the structure-activity relationships (SAR) between indole mimics and receptors, PXR and AhR.

Indole scaffolds as a promising class of the aryl hydrocarbon receptor ligands

The aryl hydrocarbon receptor (AhR), deemed initially as a xenobiotic sensor, plays multiple physiological roles and is involved in various pathophysiological processes and many diseases' etiology. Therefore, the therapeutic and chemopreventive targeting of AhR is a fundamental issue. To date, thousands of structurally diverse ligands of AhR have been identified. The bottleneck in targeting the AhR is that it is a Janus-faced player with beneficial vs. harmful effects in the ligand-specific context. A distinct structural class of the AhR ligands is those with indole-based scaffolds. The present review summarizes the knowledge on the existing indole-derived AhR ligands, comprising natural and dietary compounds, synthetic compounds including clinically used drugs, endogenous intermediary metabolites, and catabolites produced by human microbiota. The examples of novel, indole ring containing, rational design based AhR ligands are presented. The molecular, in vitro, and in vivo effects are described.