Oral Presentation 32nd Lorne Cancer 2020

DNA methyltransferase 3a (DNMT3A) promotes inflammation‑associated gastric cancer growth (#20)

Anne Huber 1 , Christine Dijkstra 1 , Merridee Wouters 2 , Michael Buchert 1 , Matthias Ernst 1 , Moritz Eissmann 1
  1. Olivia Newton-John Cancer Research Institute, Melbourne/Yallambie, VICTORIA, Australia
  2. Cancer Data Science Group, Children's Medical Research Institute, Sydney, New South Wales, Australia

Gastric cancer (GC) remains the third leading cause of cancer‑related death worldwide. While inflammation is a well‑established driver of gastric tumorigenesis, only a small subset of GC patients respond to immune checkpoint therapy. The understanding of the pro‑inflammatory tumour microenvironment, which suppresses anti‑tumour immune responses, is a field of great interest. One proposed mechanism of immune escape is silencing tumour‑antigen expression and thereby avoiding immune recognition. DNA methyltransferases (DNMTs) are the enzymes responsible for epigenetic silencing of gene expression and are often overexpressed in tumours. Here we are studying the role of DNMT3A in gastric cancer mouse models and its potential as a therapy target.

 

We identified Dnmt3a as a top ten cancer‑driver gene in a stomach‑specific Sleeping Beauty transposon mutagenesis screen, where transposon‑mediated mutation of Dnmt3a in inflammation‑sensitized mice (gp130F/+) led to spontaneous gastric adenoma formation. We and others detected DNMT3A overexpression in human gastric cancer specimen. To study the functional and mechanistic contribution of DNMT3A to gastric tumour growth, we established a Dnmt3a‑overexpressing inflammation‑driven gastric cancer mouse model (gp130FF, A33Dnmt3a), where gp130FF mutant gastric tumours have a 10-fold elevated Dnmt3a expression. Importantly, tumour‑specific Dnmt3a overexpression significantly increased gastric tumour burden. In a second mouse model, where mutant Kras, Pi3kca and Tp53 expression results in highly advanced invasive gastric carcinoma formation, we have identified DNMT3A as being highly expressed in the invasive front of tumours as well as in their liver metastases. We are currently establishing DNMT3A‑deficient (Kras-, Pi3kca-, Tp53-mutant) tumour organoid models to study the effects of DNA methylation on tumour progression, immune escape and therapy responses.

 

Taken together, we provide evidence for a gastric cancer‑driver function of Dnmt3a, which encourages further studies to investigate the potential of DNMT3A as a novel therapy target in gastric cancer.