Assessing the translatability of in vivo cardiotoxicity mechanisms to in vitro models using causal reasoning

Area of Science:

• Pharmacology and Toxicology
• Computational Biology
• Cardiovascular Research

Background:

• Drug-induced cardiac toxicity is a major cause of drug withdrawals, with preclinical studies often failing to detect risks.
• Translating findings from simple cell systems to complex in vivo environments remains a challenge for accurate toxicity prediction.
• Improved in vitro safety screens are needed early in drug discovery to identify cardiotoxic potential.

Purpose of the Study:

• To analyze the translatability of cardiotoxic effects from rodent in vivo models to two in vitro cell systems (H9C2 and primary rat cardiomyocytes) using transcriptional response.
• To apply a novel computational systems biology approach, the Causal Reasoning Engine (CRE), to infer upstream molecular events driving gene expression changes.
• To identify common molecular mechanisms of cardiotoxicity and validate computational predictions experimentally.

Main Methods:

• Analysis of transcriptional response in H9C2 cells and primary rat cardiomyocytes exposed to cardiotoxic drugs.
• Application of the Causal Reasoning Engine (CRE) for systems biology pathway analysis to infer molecular events.
• Experimental verification of predicted molecular hypotheses, including Kruppel-like factor 4 (KLF4) and Transforming growth factor beta 1 (TGFB1).

Main Results:

• Evidence of mechanistic convergence towards common molecular pathways underlying cardiotoxicity, irrespective of the specific toxic phenotype.
• Successful experimental validation of KLF4 and TGFB1 as key molecular mediators translating from in vivo to in vitro models.
• Demonstration that simpler in vitro models (H9C2) can be as effective as more complex ones (primary rat cardiomyocytes) for translatability if appropriate endpoints are used.

Conclusions:

• A novel systems biology approach effectively predicts cardiotoxicity mechanisms (e.g., KLF4, TGFB1) that translate from in vivo to in vitro.
• The choice of in vitro model complexity may be less critical than selecting the right endpoints for accurate translatability assessment.
• Further validation of predicted molecular hypotheses is essential for developing robust predictive in vitro cardiotoxicity assays.