Abstract
Pancreatic injury plus oncogenic Kras mutation produced cancer-associated chromatin states in vivo.
Major Finding: Pancreatic injury plus oncogenic Kras mutation produced cancer-associated chromatin states in vivo.
Mechanism: Chromatin remodeling led to production of cancer-related factors, including the cytokine IL33.
Impact: This reveals a mechanism by which environmental insults can cooperate with mutations to cause cancer.
In normal pancreata, injury triggers a metaplastic transition involving conversion of damaged acinar cells such that they adopt a duct-like state (termed acinar-to-ductal metaplasia, or ADM), which are ultimately restored to their original acinar state once the injury is resolved. In the presence of oncogenic KRAS mutation, however, the pancreatic injury healing response progresses abnormally, leading to persistent metaplasia and rapid formation of pancreatic intraepithelial neoplasia, a precursor to pancreatic cancer. Alonso-Curbelo and colleagues showed that chemically induced pancreatic injury caused large-scale chromatin accessibility changes in pancreatic epithelial cells undergoing ADM in vivo. Notably, Kras-mutant cells acquired chromatin accessibility profiles distinct from those of injured Kras–wild-type pancreata and from noninjured Kras-mutant pancreata, with the chromatin accessibility changes arising in cells from injured Kras-mutant pancreata being similar to those of cells from advanced pancreatic ductal adenocarcinomas. Specifically, differentially accessible chromatin between injured Kras-mutant and injured normal pancreatic cells contained binding sites for several master transcription factors known to regulate pancreatic lineage commitment and carcinogenesis. Using mouse models enabling pancreas-specific silencing of the mRNA encoding the key transcriptional coactivator bromodomain-containing protein 4 (BRD4), selected because of its role in regulating cell-identity genes and enhancer-mediated transcription, showed that BRD4 sustains the expression of genes associated with injury-driven dynamic chromatin. Phenotypically, Brd4 silencing did not prevent injury-induced ADM but blunted both its regenerative and tumor-initiating outputs, with its suppression impairing the restoration of acinar state in Kras-wild-type damaged cells, and the formation of neoplastic lesions in the context of an oncogenic Kras mutation. Mechanistically, the cooperation between Kras mutation and tissue damage produced an “acinar-to-neoplasia” chromatin switch that diverted BRD4-mediated transcription from acinar-specifying to cancer-defining genetic loci. As a result of this switch, the gene encoding the cytokine IL33 rapidly gained chromatin accessibility in Kras-mutant cells after injury, leading to its hyperactivation in cells undergoing neoplastic transformation, in a BRD4-dependent manner. Indeed, among more than 40 cytokines analyzed, IL33 was the most abundant in Kras-mutant pancreata following injury. Correspondingly, administration of exogenous IL33 hastened the development of premalignant lesions in Kras-mutant but not Kras–wild-type mice, mimicking injury response in the Kras-mutant pancreatic cells. Taken together, these results show how the combination of tissue injury and Kras mutation—neither of which is sufficient to induce pancreatic oncogenesis alone—cooperate to produce a unique chromatin state that is not accessible during normal regeneration and which contributes to the development of pancreatic cancer.
Note: Research Watch is written by Cancer Discovery editorial staff. Readers are encouraged to consult the original articles for full details. For more Research Watch, visit Cancer Discovery online at http://cancerdiscovery.aacrjournals.org/CDNews.
