Expanding the drug discovery space with predicted metabolite-target interactions

Abstract

Metabolites produced in the human gut are known modulators of host immunity. However, large-scale identification of metabolite-host receptor interactions remains a daunting challenge. Here, we employed computational approaches to identify 983 potential metabolite-target interactions using the Inflammatory Bowel Disease (IBD) cohort dataset of the Human Microbiome Project 2 (HMP2). Using a consensus of multiple machine learning methods, we ranked metabolites based on importance to IBD, followed by virtual ligand-based screening to identify possible human targets and adding evidence from compound assay, differential gene expression, pathway enrichment, and genome-wide association studies. We confirmed known metabolite-target pairs such as nicotinic acid-GPR109a or linoleoyl ethanolamide-GPR119 and inferred interactions of interest including oleanolic acid-GABRG2 and alpha-CEHC-THRB. Eleven metabolites were tested for bioactivity in vitro using human primary cell-types. By expanding the universe of possible microbial metabolite-host protein interactions, we provide multiple drug targets for potential immune-therapies.

Fig. 1. Metabolomics results and comparisons to the original HMP2 IBD study (HMP2).

a UMAP analysis of the metabolomics sample distribution by Crohn’s disease (CD) and Ulcerative Colitis (UC) patients and controls (nonIBD). b Volcano plot showing the differential abundance of each metabolite per disease state against the consensus scoring of each state. c Number of metabolites considered relevant in HMP2 and current study per disease state, subdivided into overlapping and non-overlapping subsets. d Total number of metabolites selected relevant in each study. e Intersection matrix between metabolites selected each study. f Correlation plot between the bootstrapped power estimation method used to determine metabolite differential abundance between CD and UC patients results. g Correlation between the consensus scoring used in this study and HMP2 FDR-adjusted p values for each metabolite (refer to Table 1 for samples composition).

Fig. 2. Overview of the connected metabolites with highest ranking scores and the perspective targets.

a Distribution of target drug classes per each metabolite class (numbers represent unique metabolites per metabolite class). b Alluvial plot describing the distribution of connections between metabolite classes, modulation type and drug target classes (numbers represent unique targets per drug target class).

Fig. 3. Overview of the transcriptomics analysis results.

a UMAP analysis of the transcriptomic samples by biopsy location. b Volcano plots representing target differential expressions in Crohn’s disease (CD) and Ulcerative Colitis (UC) states by FDR-adjusted p value. c Cross-plot showing possible interactions of interest between targets with expression on the vertical axis and perspective modulator metabolite differential abundances on the horizontal axis, per each disease type (refer to Table 1 for samples composition).

Fig. 4. Overview of identified for putative target proteins.

Metabolites were connected through similar ChEMBL compounds where similarity is classified as medium (0.8 < Tanimoto score < 0.9) or strong (Tanimoto score > 0.9) for (a) CXCR1/2 and NOS2, (b) HTR4, (c) GABRG2 and (d) SLC22A3. Direction and affinity of analog binding to target was parsed from ChEMBL assay databases and represented as medium (5.5 < pxC₅₀ < 7.0) and strong (pxC₅₀ > 7.0). Direction (i.e., up- or down-) of differential expression for targets or differential abundance for metabolites are represented by colors. Target significance (log q value) is based on differential expression of the gene or GWAS association. Complete results for targets with genetic evidence are shown in Supplementary Fig. 5.

Fig. 5. Examples of biomarker readouts from in vitro cell assays for four of eleven tested metabolites.

a Butyrate, b Nicotinic acid, c Alpha-CEHC and d Oleanolic acid. Metabolites were administered at different concentrations, here ranked from higher to lower (concentrations in Supplementary Data 8). Readouts graph show the differential abundance vs baseline for the B and T cell system (BT), arterial smooth muscle cells (CASM3C) and wound healing (HDF3CGF) (full results for all 11 metabolites are shown in Supplementary Fig. 6). Knowledge-based graphs on the right represent possible pathway connections between the proposed targets for each metabolite and the most significant biomarker readouts. Interactions are color-coded for positive (green), negative (red) and unknown (gray) modulation.