Longitudinal analysis of human pancreatic adenocarcinoma development reveals transient gene expression signatures

Abstract

Previous transcriptome studies of human pancreatic ductal adenocarcinoma (PDAC) compare non-cancerous pancreatic intraepithelial neoplasias (PanIN) with late-stage PDAC obtained from different patients, thus have limited ability to discern network dynamics that contribute to the disease progression. We demonstrated previously that the 10-22 cell line, an induced pluripotent stem cell-like line reprogrammed from late-stage human PDAC cells, recapitulated the progression from PanINs to PDAC upon transplantation into NOD/LtSz-scid/IL2R-gammanull mice. Herein, we investigated the transition from precursor to PDAC using the isogenic model. We analyzed transcriptomes of genetically tagged 10-22 cells progressing from PanINs toPDAC in mice and validated the results using The Cancer Genome Atlas PDAC dataset, human clinical PanIN and PDAC tissues, and a well-established murine PDAC model. We functionally studied candidate proteins using human normal (H6C7) and cancerous (Miapaca2, Aspc1) pancreatic ductal epithelial cell lines. 10-22 cell-derived PDAC displayed the molecular signature of clinical human PDAC. Expression changes of many genes were transient during PDAC progression. Pathways for extracellular vesicle transport and neuronal cell differentiation were derepressed in the progression of PanINs to PDAC. HMG-box transcription factor 1 (HBP1) and BTB domain and CNC homolog 1 (BACH1) were implicated in regulating dynamically expressed genes during PDAC progression, and their expressions inversely correlated with PDAC patients' prognosis. Ectopic expression of HBP1 increased proliferation and migration of normal and cancerous pancreatic cells, indicating that HBP1 may confer the cell dissemination capacity in early PDAC progression. This unique longitudinal analysis provides insights into networks underlying human PDAC progression and pathogenesis. Implications: Manipulation of HBP1, BACH1, and RUN3 networks during PDAC progression can be harnessed to develop new targets for treating PDAC.