Abstract DNA 5’ cytosine methylation at CpG dinucleotides, particularly at gene promoters and regulatory elements, can mediate epigenetic silencing of key tumor suppressor and caretaker genes in human cancers. This is largely mediated through the activity of methyl binding domain (MBD) “reader” proteins that bind methylated DNA and recruit chromatin repression complexes. Of the MBD family of proteins, MBD2 is a particularly attractive therapeutic target since targeted genetic disruption of this protein is well tolerated and can greatly attenuate tumor formation in cancer prone mice and growth of cancerous cells lines. Therefore development of pharmacological approaches for targeting MBD2 may have utility in cancer therapeutics. In this work we describe a time resolved fluorescence energy transfer (TR-FRET) based assay to find small molecule inhibitors of MBD2. First we optimized conditions to establish an assay that was amenable for high throughput screening (HTS) in 384 well plates, obtaining a Z’ factor of 0.58 (HTS target >0.5). To validate our assay we applied it to the 1280 compound Sigma® LOPAC library. Among 10 primary hits, four (Mitoxantrone, idarubicin, aurintricarboxylic acid, and NF449) showed dose responsive inhibition of MBD2 binding to methylated DNA. However, a counter screen showed these 4 compounds also inhibited interaction of the transcription factor SP-1 to DNA, suggesting their mode of action was not specific to MBD2. This pilot screen provided proof principle that our assay format could identify MBD2 inhibitors and that a counter screen strategy could evaluate the specificity of identified hit compounds. In collaboration with the Scripps Institute in Florida, we further miniaturized, refined, and deployed the MBD2 TR-FRET assay to screen the NCI small molecule library of 370,276 compounds. The primary screening effort identified 1,149 active compounds (able to disrupt MBD2-DNA interaction) at 4.4µM. In the secondary screen each of these were run in triplicate at 4.4µM, with 271 (26.2%) of them validating as active hits. Next, a TR-FRET assay to evaluate inhibition of binding of the ubiquitin-like with PHD and ring finger domains 1 (UHRF1) protein to methylated DNA was utilized to evaluate specificity of the 271 active compounds that passed secondary screening. This yielded 48 small molecules that appeared to specifically inhibit the MBD2-DNA binding interaction. These 48 were subjected to the TR-FRET assay in a dose response from ∼10nM to 50µM on both MBD2 and UHRF1, yielding 18 active compounds which specifically inhibit MBD2-DNA interactions in a dose responsive manner with IC50's between 1µM and 17µM. Work is currently underway to test these compounds for biological activity. Acknowledgments/Funding- The authors would like to thank Peter Hodder, Virneliz Fernandez-Vega and Tom Bannister at the Scripps Institute-Florida for their efforts with this project. Funding provided by Department of Defense award W81XWH-11-1-0618 (to NW), NIH/NCI grants CA70196 and CA58236 (to WGN and SY), and the Prostate Cancer Foundation. Citation Format: Nicolas A. Wyhs, Hugh Giovinazzo, David Walker, Srinivasan Yegnasubramanian, William G. Nelson. Small molecular library screening to find methyl binding domain 2 protein inhibitors. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr LB-129. doi:10.1158/1538-7445.AM2014-LB-129
Read full abstract