6533b823fe1ef96bd127f7de

RESEARCH PRODUCT

Oxidative stress response of tumor cells: microarray-based comparison between artemisinins and anthracyclines

Franz OeschThomas Efferth

subject

ArtemisininsDaunorubicinAntineoplastic AgentsPharmacologyBiologymedicine.disease_causeBiochemistryAntimalarialsInhibitory Concentration 50parasitic diseasesTumor Cells CulturedmedicineAnimalsCluster AnalysisHumansIdarubicinAnthracyclinesDoxorubicinRNA MessengerArtemisininOligonucleotide Array Sequence AnalysisPharmacologyCardiotoxicityGene Expression ProfilingArtemisininsGene expression profilingOxidative StressDrug Screening Assays AntitumorOxidation-ReductionSesquiterpenesOxidative stressmedicine.drug

description

The antimalarial artemisinins also reveal profound cytotoxic activity against tumor cells. Artemisinins harbor an endoperoxide bridge whose cleavage results in the generation of reactive oxygen species (ROS) and/or artemisinin carbon-centered free radicals. Established cancer drugs such as anthracyclines also form ROS and free radicals that are responsible for the cardiotoxicity of anthracyclines. In contrast, artemisinins do not reveal cardiotoxicity. In the present investigation, we compared the cytotoxic activities of different artemisinins (artemisinin, artesunate, arteether, artemether, artemisitene, dihydroartemisinylester stereoisomers) in 60 cell lines of the National Cancer Institute (NCI), USA, with those of anthracyclines (doxorubicin, daunorubicin, 4'-epirubicin, idarubicin, deoxydoxorubicin, trifluoroacetyl-doxorubicin-14-valerate). The inhibition concentration 50% (IC(50)) values of artemisinins and anthracyclines were correlated with the mRNA expression of 170 genes involved in oxygen stress response and metabolism as recently determined by microarray analysis and deposited in the NCI database (http://dtp.nci.nih.gov). The genes whose expression was significantly linked to cellular drug response in Kendall's tau tests were subjected to hierarchical cluster analysis and cluster image mapping. Mathematical correction for false-positive correlations was done by a false discovery rate algorithm. One cluster contained predominantly genes with a relationship to artemisinins and another one genes with a relationship to anthracyclines. In a third cluster, genes correlating to both drug classes were assembled. This indicates that different sets of genes involved in oxidative stress response and metabolism may contribute to the cytotoxic and differing toxic side effects of these drug classes.

https://doi.org/10.1016/j.bcp.2004.03.003