Abstract 148: Development of a DNA methylation array normalization method for analyzing demethylating treatment effects in paired samples

Abstract

Abstract Introduction: Normalization of high throughput array data reduces the background noise between samples. However, most normalization techniques will correct differences between paired samples after treatment similarly as background noise and remove valuable information. We developed a new normalization technique for analyzing methylation arrays data after demethylating treatment effects in paired samples. Materials and methods: neuro-ectodermal tumor cell lines (18 NBL and 5 PNETs) were treated with methylation inhibitors and HDAC inhibitors. We used 30 nM Decitabine (DAC) for 72 hours and 25 nM Trichostatin A (TSA) for the last 48 hours of treatment. Samples before and after treatment were subjected to 244 K CpG island arrays (Agilent). Results: Conventional normalization: Normal Loess normalization and VSN two channel normalization were initially used and showed that the demethylation treatment induced an increase of methylation in 8% and a decrease in 10% of the CpGs after treatment, and suggested a Gaussian distribution with equal medians after treatment Novel normalization: The new normalization technique is based on the restriction enzyme based Differential Methylation Hybridization technique (Yan et al. 2002). Methylation data on the arrays are based on differences in methylation sensitive enzymatic DNA cuts. Our new normalization method is based on fragments without methylation sensitive restriction sites. These data points are not methylation and treatment independent and used as controls in a weighed Loess normalization. This resulted in a significant decrease of methylation in 69.1% of all CpGs and an increase in 1% of the CpGs after the combined epigenetic treatment. This well represents the expected effect of the treatment. To confirm this, we tested the effect of demethylation after treatment on known methylated genes (DcR1, DcR2, RASSF1A, DR3). The test used the array data by using the conventional normalization, the new normalization and compared these results to methylation specific PCR (MSP) changes for these genes. The new normalization showed significant demethylation and compared best to the MSP data. Additionally, we confirmed the 69% decrease in global CpG methylation in one cell line pair (62% less methylation) and is currently being validated in 5 more cell lines with the Methylated DNA Quantification kit (Sigma-Aldrich). Conclusion: We developed a new normalization technique for high resolution methylation arrays, based on the equal distribution of uncut DNA fragments after methylation specific enzymatic digestion in all samples. We showed that this technique will preserve the differences of demethylating effects in paired samples. Note: This abstract was not presented at the AACR 101st Annual Meeting 2010 because the presenter was unable to attend. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 101st Annual Meeting of the American Association for Cancer Research; 2010 Apr 17-21; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2010;70(8 Suppl):Abstract nr 148.