Related Experiment Video
Updated: Jun 27, 2026
In Vitro Modeling of Fat Deposition in Metabolic Dysfunction-Associated Steatotic Liver Disease
Published on: July 19, 2024
Preliminary In Vitro Screening of Structure-Dependent β-Hydroxybutyrate Responses to Dietary Fatty Acids in
Xiaojing Liu1,2, Fei Pan3, Yandan Wang1,4
1State Key Laboratory of Food Science and Resources, School of Food Science and Technology, Jiangnan University, Wuxi 214122, China.
Background/Objectives:
Controlled comparisons of extracellular β-hydroxybutyrate (β-HB) responses induced by individual dietary fatty acids (FAs) remain limited. This study established a preliminary hepatocyte-derived in vitro assay for comparing FA-associated β-HB responses and used exploratory descriptor-based modeling for hypothesis-generating FA ranking.
Methods:
Dose- and time-dependent extracellular β-HB accumulation induced by nineteen dietary FAs was quantified in HepG2 and AML12 cells under nutrient-deprived assay conditions. For exploratory descriptor-based modeling, the 19-FA HepG2 dataset was split into 15 training compounds and four held-out compounds. Model stability was further assessed using repeated random splits, LOOCV, permutation testing, MCFA sensitivity analyses, and simple structural baseline models. Four additional structurally related FAs were tested only for preliminary experimental ranking assessment, not as an independent external test set.
Results:
C8:0 and C10:0 consistently induced the highest extracellular β-HB accumulation, whereas most long-chain saturated FAs and very-long-chain monounsaturated FAs showed lower responses. The single four-compound held-out subset yielded an apparent R2 of 0.875, but repeated random-split assessment showed substantial split-dependent variability, and simple baseline models performed similarly to GradientBoosting.
Conclusions:
This study provides a preliminary in vitro dataset for comparing extracellular β-HB responses to selected dietary FAs under defined nutrient-deprived hepatocyte assay conditions. The descriptor-based analysis should be interpreted only as a small-sample, exploratory, hypothesis-generating structure-response framework.
