Abstract 1061: High resolution DNA recognition by the Snail zinc finger protein: Testing of a molecular dynamics based model defines the atomic level interactions required for high affinity binding, E-box specificity and reveals potential strategies for small molecule control of EMT transcriptional programs

Abstract

Abstract In most epithelial-derived tumors, the early developmental program known as “Epithelial to Mesenchymal Transition” (EMT), is reactivated and plays an important role in cancer metastasis and recurrence. The EMT program directs cells to become non-adherent, motile and highly invasive thus leading to metastatic spread. Thus, strategies to shutdown the EMT program might be useful in controlling EMT mediated tumor progression. EMT is largely controlled by a handful of transcription factors, including Snail and Slug: these are the canonical master regulators of the EMT and function as transcriptional repressors of E-cadherin and other cell adhesion molecules. A rate-limiting step in EMT regulation by Snail occurs when its four tandem C2-H2 zinc finger (ZF) motifs bind the E-box DNA sequence (CANNTG) in target promoters. The atomic structure of Snail-DNA complex remains undefined. We derived a high resolution model for DNA recognition by Snail using Molecular Dynamics (MD) based threading of known ZF-DNA structures and energy minimization. The model was tested using purified recombinant Snail protein, and truncations/single amino acid substitutions in in vitro DNA binding assays. A Snail monomer binds with nanomolar affinities (Kd=22.1 nM) in the major groove of DNA. Disrupting cysteine residues in each finger shows that fingers 2 and 3 contribute most to the specificity for DNA recognition. Alanine scanning mutagenesis in the recognition helices revealed that R191, W193, S221, D219, N222 and R247 are invloved in direct DNA contact. Among them, W193 in finger 2, S221 in finger 3 and R247 in the finger 4 are predicted to mediate critical base-pair contacts and therefore determine the sequence specificity for Snail DNA-binding, while R191 in finger 2 and N222 in finger 3 stabilize the complex via phosphate backbone contacts. A comparison of the base-pair contacts required for binding by Snail vs. the bHLH protein Twist was highly informative. The Twist dimer does not utilize the central 2 base pairs within the E-box (ie. NN in the CANNTG sequence) for proper binding. In contrast, E-box binding by Snail requires a guanine as the second base in the central 2 base pairs (ie. NG in the CANGTG sequence). However, if the central 2 base pairs is CG, CpG methylation of this motif strongly inhibits Twist binding but does not affect Snail binding. Thus, remarkably, depending on the exact sequence of the E-box present in a promoter, CpG methylation of that site can differentially effect binding and thus regulation by Snail or Twist. These data provide the first examples of how specificity can be generated for target gene regulation by either Twist or Snail. Citation Format: Yuanjie Liu, Jeremy W. Prokop, Hongzhuang Peng, Frank J. Rauscher. High resolution DNA recognition by the Snail zinc finger protein: Testing of a molecular dynamics based model defines the atomic level interactions required for high affinity binding, E-box specificity and reveals potential strategies for small molecule control of EMT transcriptional programs. [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 1061. doi:10.1158/1538-7445.AM2014-1061