(484.8) To ChIP, or to CUT, that is the question: Comparative Evaluation of NextGen Methodologies for Studying the genome-wide distribution of Histone H3 Lysine 9 di-methyl mark in pancreatic cells

Poster Board Number: A20

Guillermo Urrutia (Medical College of Wisconsin), Jenica Abrudan (Medical College of Wisconsin), Meijun Du (Medical College of Wisconsin), Thiago Milech de Assuncao (Medical College of Wisconsin), Angela Mathison (Medical College of Wisconsin), Michael Zimmermann (Medical College of Wisconsin), Gwen Lomberk (Medical College of Wisconsin), Raul Urrutia (Medical College of Wisconsin)

Our laboratories study the role of the G9a-H3K9me2 pathway in PDAC initiation and development in response to oncogenic KRASG12D mutation. Understanding how the epigenomic landscape responds to this mutation remains fundamental for the identification of potential mechanisms and targets for pancreatic oncogenesis. Unfortunately, because of its low levels in human cells and its discrete “block-like” genome-wide deposition, the H3K9me2 mark is more difficult to study than more abundant, sharp peak-forming marks (e.g. H3K27ac). The aim of this study was to perform a comparison among multiple Next Generation Sequencing (NGS) technologies to identify the methodology with the best technical performance for detection of H3K9me2 at a genome-wide scale. To do so, we performed Chromatin Immunoprecipitation (ChIP), CUTamp;RUN and CUTamp;TAG assays for the Histone H3 Lysine 9 di-methyl histone mark (H3K9me2) followed by NGS on two pancreatic ductal epithelial-like cells: hTERT-HPNE E6/E7 and the KRASG12D mutant-expressing hTERT-HPNE E6/E7 KRASG12D cells. As expected, ChIP-Seq required the largest number of total reads for peak detection. Both ChIP-Seq and CUTamp;TAG had the highest percentage of high-quality reads (gt;75%), while CUTamp;RUN performed below the 50% mark. No significant differences in library size were observed among all three techniques. Overall, CUTamp;TAG allowed identification of the largest number of H3K9me2-marked genes in parental HPNE and KRASG12D-expressing HPNE cells. Spearman correlation values above 0.8 for inter-sample comparison indicated a high degree of repeatability for CUTamp;TAG replicates. In addition, CUTamp;TAG showed the largest occupation of the H3K9me2 mark near the transcription start site (TSS) of identified genes. Therefore, we have compared three methodologies for the evaluation of H3K9me2 occupation at the genome-wide scale and identified CUTamp;TAG as the method with the highest quality, lowest noise-to-signal ratio, and highest repeatability. In conclusion, CUTamp;TAG represents the best option for NGS of the H3K9me2 histone mark in pancreatic cells.  Notably, because this methodology is the best among the three tested for miniaturization of epigenomic studies, this information bears relevance to translational research in human, where sample availability is limited.

This work was supported by NIH Grants R01DK52913 (RU and GL) and R01CA247898 (GL), Advancing a Healthier Wisconsin Endowment (GL and RU), and The Linda T. and John A. Mellowes Endowed Innovation and Discovery Fund (RU).