Dissecting the dilemma between in vitro and in vivo drug screening: treating HepG2 cells with Artesunate as a model

Abstract

AbstractThe dilemma between in vitro and in vivo drug screening results persisted to hinder preclinical anti-cancer drug development. In this study, drug screening was initially performed on monolayer cultures of HepG2 cells, whereas in vivo drug testing was performed on the subcutaneous xenograft mouse model of HepG2 cells in athymic nude mice. Results showed that Artesunate inhibited HepG2 cell growth in vitro, but was ineffective in vivo. Hence, we investigated the difference between in vitro and in vivo settings using this cell-drug combination as a model. Immuno-staining of hepatocellular carcinoma (HCC) biomarkers showed that α-fetoprotein (AFP) was unaffected by Artesunate treatment in vitro or in vivo, suggesting that AFP was neither a biomarker nor response indicator. Alternatively, high albumin was detected in both monolayer and organoid cultures; contrariwise, the xenograft tumors prevailed low albumin, in consistency to human HCC with poor prognosis. Artesunate treatment reduced intracellular albumin expression in vitro; Artesunate did not alter tissue and serum albumin in xenograft mice, in coincidence with its irresponsiveness in vivo. However, Artesunate binding to albumin was undetectable. Instead, we observed a transient stimulation of Erk1/2 phosphorylation followed by DAPK1 dephosphorylation and apoptosis. Combined treatment of Artesunate with U0126 revoked Erk1/2-DAPK1 phosphorylation and exerted modest proliferative advantage at early time points, but eventually did not rescue HepG2 cells from death. U0126 induced multiploid formation independent of Artesunate, resulting in cell cycle arrest within 24 h post-treatment.SignificanceSystematic investigation of HCC biomarkers AFP and albumin among the in vitro models of monolayer and organoid cultures, and the in vivo models of subcutaneous and liver implantation xenograft mice was conducted.