Molecular profiles indicative for response of cancer cells to artemisinin derivatives

T Efferth, PhD, a, E Gebhart, PhD, A Olbrich, PhD, and JO Funk, MD

a Virtual Campus Rhineland-Palatinate, Rodeneck-Platz 2, 55126 Mainz, Germany; e-mail:

Aim: In a search for novel antineoplastic agents, we investigated the artemisinin derivatives artesunate (ART), artemether (ARM), and arteether (ARE). We correlated the basal mRNA expression levels of 465 genes obtained by micro-array biochip assays (Scherf et al., 2000) with the inhibition concentration 50 % (IC50) values. Methods: The anticancer activity was measured under the auspices of the National Cancer Institute, U.S.A. We correlated the microarray data by Kendall’s t test and hierarchical cluster analysis (implements of WinStat, Kalmia, CA) with drug response data. Results: 117/465 genes correlated with the IC50 values of ART. These genes were from different gene categories with a high percentage among proliferation-regulating genes. Indeed, ART correlated significantly with proliferation parameters (cell doubling times, G0/G1 and S cell cycle phases). By hierarchical cluster analysis and clustered image mapping, we found three distinct clusters which significantly correlated with IC50 values of ART (very sensitive, sensitive, less sensitive). Further cluster analyses with ARM and ARE showed that one cluster of genes correlated with all 3 drugs. Another cluster contained genes correlating specifically with one of the 3 drugs. Conclusions: Proliferation-regulating genes play important role for the anticancer activity of ART, ARM and ARE. Correlation to different genes may explain differing anticancer activities and side effects on normal tissues, respectively.

KEY WORDS: Chemotherapy, Natural products, Traditional Chinese Medicine.

For more information, contact

Paper presented at the International Symposium on Predictive Oncology and Intervention Strategies; Paris, France; February 9 - 12, 2002; in the section on Gene Expression, Part 2.