Mechanism of Action Identified in 30 Days: a Systems Biology Approach

Understanding how emerging threat agents alter cellular function and identifying drug candidates for effective response require the rapid identification of their mechanisms of action (MoA). To capture molecular changes due to xenobiotic exposure, protocols for cell exposure, dose finding, microarray‐based transcriptomics, high‐throughput proteomics, and rapid untargeted metabolomics were developed and applied to hepatocellular carcinoma (HepG2/C3A) cell lines. The cells were exposed to a test compound with a known canonical MoA and the cellular response was followed in time. The MoA behind the observed changes was explored by knowledge‐based pathway analysis and Bayesian network inference to identify known features and discover new attributes, respectively. Conventional confirmatory assays were carried out to validate the MoA discerned from the multi‐omics data. A 30‐day test was conducted to follow responses to forskolin exposures between 2 min and 48 h. Among the 67,528 probed transcripts, 79 exhibited statistically significant ( p ≤ 0.05) fold changes of at least 1.5. Based on 300,000 peptide matches in tandem mass spectra, 3,531 protein groups were identified (FDR < 1%) with an additional 1,173 protein groups identified in the phosphoprotein fraction. Novel untargeted metabolomic technologies, based on laser ablation electrospray ionization (LAESI) mass spectrometry (MS) in combination with ion mobility separation (IMS), and high resolution MS using laser desorption ionization from nanopost arrays (NAPA), revealed the identity of 59 and 75 metabolites, respectively. Fold changes for the quantified transcripts, proteins, and metabolites were used to perform knowledge‐based pathway analysis. To gauge the efficacy of our approach, results were compared to the canonical MoA of forskolin, a known stimulant of the cAMP signaling pathway. Pathway analysis of the multi‐omics data identified forskolin as the top upstream regulator (out of 3,502 compounds), and cAMP‐mediated signaling ranked as sixth out of the 642 canonical pathways in the knowledge base. De novo Bayesian network inference discovered multiple molecular events downstream from the canonical MoA. To confirm the identified MoA, conventional phenotypic, qRT‐PCR, immunofluorescence, and intracellular signaling assays were conducted. High‐throughput technologies to capture molecular changes induced by a test compound in the transcripts, proteins, and metabolites of the exposed cells enabled the recovery of the canonical MoA in 30 days with significant fidelity. Support or Funding Information Research was sponsored by the U.S. Army Research Office and the Defense Advanced Research Projects Agency and was accomplished under Cooperative Agreement Number W911NF‐14‐2‐0020. The views and conclusions contained in this document are those of the authors and should not be interpreted as representing the official policies, either expressed or implied, of the Army Research Office, DARPA, or the U.S. Government. The U.S. Government is authorized to reproduce and distribute reprints for Government purposes notwithstanding any copyright notation hereon.