Extracellular vesicle release reflects dose-dependent hepatotoxic responses in HepG2 cells

Introduction: Drug-induced liver injury (DILI) is a leading cause of acute liver failure and a major challenge in clinical practice and drug development. Extracellular vesicles (EVs) are membrane-bound nanoparticles emerging as sensitive biomarkers of cellular stress and damage. While most studies have focused on EV cargo, quantitative changes in particle number as a function of drug dose remain poorly characterized. Here, we evaluated whether EV concentration is modulated dose-dependently in HepG2 hepatocytes exposed to two hepatotoxic drugs affecting mitochondrial disruption.

Methods: HepG2 cells (1×10⁶/well) were treated with valproic acid (VPA) or amiodarone (AMI) at IC10 and IC50, as previously determined by Martínez-Sena et al. Conditioned medium was sequentially centrifuged to remove cells and debris, then EVs isolated by SEC yielding a vesicle-enriched (F1) and protein-enriched (F2) fraction. F1 particle concentration was quantified by TRPS and both fractions analysed by ATR-FTIR to estimate protein concentration based on amide I band absorbance. Comparisons used an unpaired t-test (n=1) between treated and untreated control cells. EV identity will be confirmed by Western blot using positive and negative markers and morphological characterization will be performed by TEM.

Results: Both drugs induced a dose-dependent increase in EV concentration versus untreated controls. No significant differences were observed at IC10 (VPA: 1.12-fold, p=0.778; AMI: 2.74-fold, p=0.093). At IC50, VPA showed a 2.5-fold increase (p=0.0183) and AMI a nearly 5-fold increase (p=0.0014). ATR-FTIR revealed higher protein in F2 of treated samples, while F1 remained low (~0.48–0.91 mg/mL), confirming effective SEC-based separation. EV counts will be normalized to viable cell number in subsequent analyses.

Conclusions: EV concentration increases dose-dependently in HepG2 cells at IC50 for both VPA and AMI, suggesting EV release reflects hepatocellular stress. Further validation including EV marker profiling, morphological characterization, and analysis of patient samples will be required to establish EVs as non-invasive DILI biomarkers.