Correlation analysis of genome-wide gene expression profiles by DNA microarray. Homo sapiens

We previously isolated a new angiogenesis inhibitor, terpestacin, from natural products. To investigate molecular mechanism of terpestacin for its anti-angiogenic activity, we identified a cellular binding protein of terpestacin using phage display biopanning analysis. Terpestacin specifically bound to the small molecular ubiquinone binding protein (QP-C), a 13.4 kD subunit of the mitochondrial Complex III. To explore the functional specificity of terpestacin at QP-C, we performed cDNA microarray analysis. The genome-wide gene expression profiling showed a significant phenotype correlation between terpestacin and QP-C genetic knock-down. These results demonstrate that QP-C is a biologically relevant target of terpestacin. Keywords: Sample comparison Overall design: We performed cDNA microarray analysis of transcriptional changes in HT1080 cells treated with terpestacin, antimycin A, or siRNA directed against human QP-C (siQP-C). The probe preparation and microarray hybridization was performed using the TwinChip Human-8K cDNA microarray. The signature correlation analysis was performed using both hierarchical clustering of gene expression profiles and signature correlation coefficient value (p). The comparison of gene expression profiles between different experimental conditions revealed that the signatures of siQP-C versus terpestacin correlated significantly (p = 0.62), while signatures of siQP-C versus antimycin A (p = -0.01) and terpestacin vs antimycin A (p < 0.1) exhibit no significant correlation. These results demonstrate that the cellular phenotype induced by terpestacin is significantly correlated with that of QP-C knock-down cells.