|
Status |
Public on Nov 23, 2020 |
Title |
Aggressive PDACs show hypomethylation of repetitive elements and the execution of an intrinsic IFN program linked to a ductal cell-of-origin |
Organism |
Homo sapiens |
Experiment type |
Methylation profiling by high throughput sequencing
|
Summary |
Pancreatic ductal adenocarcinoma (PDAC) is a highly aggressive cancer embedded in an extensive desmoplastic stroma. Since this challenges the molecular analyses of bulk tumor samples, we FACS-purified epithelial cells from PDAC and healthy human pancreas and performed genome-wide transcriptome and DNA methylome analyses. Clustering based on DNA methylation revealed two distinct groups of PDAC with different methylation levels at genomic regions encoding repeat elements. Methylationlow tumors showed higher expression of endogenous retroviral (ERV) transcripts and a strong engagement of the dsRNA sensing machinery. This results in the cell intrinsic activation of an interferon response signature (IFNsign), leading to the reprogramming of stromal cells towards a pro-tumorigenic microenvironment and poor patient outcome. Methylationlow/IFNsignhigh and Methylationhigh/IFNsignlow PDAC cells harbored distinct lineage traits specific for normal ductal or acinar pancreatic epithelial cells at the methylation and transcriptional level. Moreover, ductal-cell-derived KrasG12D/Trp53-/- mutant mouse PDACs showed higher expression of IFNsign compared to tumors initiated by the same drivers in acinar cells. Collectively, our data point to two distinct origins and etiology of human PDACs, with the aggressive Methylationlow/IFNsignhigh tumor subtype potentially targetable by agents blocking cell intrinsic IFN-signaling.
|
|
|
Overall design |
WGBS analysis of the epithelial cells isolated from 7 human PDAC tumors and 6 normal distal adjacent pancreas tissue. Epithelial cells were isolated from the human tissue using flow cytometry and defined as EpCAM+/CD45-. Data were processed as previously described (Wang et al. Nat Protocol, 2013): Sequencing reads were aligned using BWA with human reference genome version hg19/GRCh37. Duplicate reads were discarded. Methylation calling was performed as previously (Delacher and Imbusch et al. Nat Immunology, 2017). The raw counts of methylated and unmethylated reads for each CpG site from different libraries were merged for each sample. *** Submitter declares that the raw data is deposited in the European Genome-phenome Archive (EGA) due to patient privacy concerns. EGA accession number is EGAS00001004660. ***
|
|
|
Contributor(s) |
Espinet E, Trumpp A, Imbusch CD, Gu Z |
Citation(s) |
33060108 |
|
Submission date |
Nov 22, 2020 |
Last update date |
Nov 25, 2020 |
Contact name |
Elisa Espinet |
Organization name |
German Cancer Research Center
|
Street address |
Im Neuenheimer Feld 280
|
City |
Heidelberg |
ZIP/Postal code |
69120 |
Country |
Germany |
|
|
Platforms (1) |
GPL11154 |
Illumina HiSeq 2000 (Homo sapiens) |
|
Samples (13)
|
|
Relations |
BioProject |
PRJNA680038 |