Analysis of gene expression profiles of cancer cells with acquired mitoxantrone resistance

Abstract

Cancer patients with similarly diagnosed tumours can have markedly different treatment responses. So far, individual markers have failed to correctly predict resistance against anticancer agents. We hypothesized that gene expression patterns attributable to chemotherapy resistant cells can help to identify key regulatory genes and predict response. We contrasted the expression profiles of four different human tumor cell lines of gastric (EPG85–257), pancreatic (EPP85–181), colon (HT29) and breast (MDA-MB-231) origin and their counterparts resistant to the topoisomerase inhibitor mitoxantrone. Mitoxantron is widely used in the treatment of breast- and prostata-carcinoma, NHL, AL and CML. Gene expression signatures for all cell lines were obtained by interrogating Affymetrix HGU95Av2 and HGU133A arrays. Normalization was performed using VSN and RMA. Prediction Analysis of Microarrays was applied for feature selection. The analysis using VSN revealed 20 genes in the HGU95Av2 chips and 392 genes in the HGU133A chips best correlated with mitoxantrone resistance at PAM threshold=2. The analysis using RMA revealed 36 genes in the HGU95Av2 chips and 413 genes in the HGU133A chips best correlated with resistance at PAM threshold=2. The differential expressions of the top selected genes were also confirmed by quantitative RT-PCR using a TaqMan micro fluidic card system. The gene lists selected in our study can be used to construct a custom chip to predict mitoxantrone resistance. Understanding the functional roles of individual genes will require the targeted over-expression or silencing of candidates in conjunction with drug-resistance modulation experiments.