Impairment of Drug Response Cascades, by DNA Methylation, in Platinum-Resistant Ovarian Cancer.
Curtis Balch,Dhaval Desai,Meng Li,John S. Montgomery,Henry Paik,Pearlly S. Yan,Tim H-M. Huang,Kenneth P. Nephew
IF: 11.2
2007-01-01
Cancer Research
Abstract:AACR Annual Meeting-- Apr 14-18, 2007; Los Angeles, CA 2867 To examine mechanisms of ovarian cancer development of platinum resistance, we developed a model culture system demonstrating stepwise adaptation of increasingly chemoresistant clones of the initially drug-sensitive cell line A2780. Specifically, DNA was isolated for genome-wide assessment of promoter DNA methylation, using a microarray approach known as differential methylation hybridization (DMH), to determine the relationship of this epigenetic phenomenon to drug resistance. Using DMH, global methylation of A2780 genomic DNA, following each selection cycle, was compared to DNA methylation of initially untreated (and chemosensitive) single clones. The total number of methylated (defined by a 1.5-fold hybridization difference) loci was found to increase following 1, 3, and 5 cycles of cisplatin treatment (590, 870, and 1176 loci, respectively). Moreover, a linear correlation between those increases in methylated loci and cisplatin resistance (as determined by IC50 dose) was also observed (R=0.98; P<0.001, Spearman analysis). Using the bioinformatics tool Pathway Express (Onto-Tools), we identified specific cascades altered by DNA methylation following 1, 3, and 5 cycles of cisplatin treatment. Notably, methylation of apoptosis (0, 2, and 7 genes, respectively, after 1, 3, and 5 cycles) and cell cycle (5, 7, and 11 genes) pathway components increased linearly with drug resistance, strongly suggesting that DNA methylation, and subsequent disruption, of those signaling cascades plays a prominent role in the development of insensitivity to platinum drugs. Finally, based on sequence information of chemoresistance-associated methylated loci, we used additional bioinformatics approaches to identify transcription factor-binding sites associated with those loci, examining the possibility of methylation-induced direct obstruction of transcription machinery as a means of gene silencing. In summary, we believe that these computational approaches will allow the generation of “epigenetic models” of ovarian cancer drug resistance that could provide for novel strategies for managing this devastating clinical phenomenon.