Pharmacogenomic identification of c-Myc/Max-regulated genes associated with cytotoxicity of artesunate towards human colon, ovarian and lung cancer cell lines

Abstract

Development of novel therapy strategies is one of the major pressing topics of clinical oncology to overcome drug resistance of tumors. Artesunate (ART) is an anti-malarial drug, which also exerts profound cytotoxic activity towards cancer cells. We applied a gene-hunting approach using microarray-based transcriptome-wide mRNA expression profiling and COMPARE analyses. We identified a set of genes, whose expression was associated either with high IC50 values or low IC50 values for ART. Therefore, these genes may function as resistance or sensitivity factors for response of tumor cells towards ART. This viewpoint is conceivable for genes involved in ribosomal activity, drug transport, cellular antioxidant defense, apoptosis, cell proliferation, cell cycle progression etc. An investigation of underlying signal transduction by pathway analysis suggested a role of the signaling pathways related to tumor necrosis factor (TNF) and the tumor suppressor p53. On the other hand, there were genes without obvious functional link to cellular response to ART, such as genes involved in the survival of cochlear outer and inner hair cells etc. We proved the hypothesis that ART influences the activity of transcription factors regulating downstream genes involved or not involved in response of cancer cells towards ART. This would explain the identification of genes with and without obvious relation to the cytotoxic activity of ART by microarray and COMPARE analyses. By analysis of the binding motifs for the transcription factors c-Myc and Max, we indeed found that 53 of 56 genes contained one or more binding sites for c-Myc/Max upstream of the gene-location. We conclude that c-Myc and Max-mediated transcriptional control of gene expression might contribute to the therapeutic effects of ART in cancer cells, but may also confer unwanted side effects by affecting therapy-unrelated genes.

Figure 1

Ranked order of IC₅₀ values for ART in 39 human cell lines of three different anatomical locations. (A) Colon cancer cell lines, (B) non-small cell lung cancer (NSCLC) cell lines, (C) ovarian cancer cell lines, and (D) sensitive ovarian cell lines and sublines resistant to cisplatin, adriamycin, or paclitaxel.

Figure 2

Network view of associated proteins created within Ingenuity Pathway Analysis software for visualizing molecular interrelationships. (A) TNF-α and TP53, (B) SKAP and GFAP associated proteins. Coloring is based on the expression values that were uploaded with the dataset. Gray indicates the molecule was part of the dataset. Lines connecting molecules indicate molecular relationships. Dashed lines indicate indirect interactions; solid lines indicate direct interactions. The style of the arrows indicates specific molecular relationships and the directionality of the interaction.

Figure 3

Linear regression of IC₅₀ values for ART and mRNA expression of c-Myc (A) and Max (C) in our cell lines and linear regression of log₁₀ IC₅₀ values for ART and mRNA expression of c-Myc (B) and Max (D) in the NCI cell line panel. Significance levels were calculated using Pearson’s correlation test.