Single-cell genomic analysis has shown that mutations in the KRAS oncogene co-opt a proto-oncogenic enhancer network in inflammation-induced metaplastic progenitor cells, initiating pancreatic ductal adenocarcinoma (PDAC), Chinese researchers reported in the November 3, 2020, edition of Nature Cancer.

This study finding has important implications for the use of 'cancer-specific' transcriptional factor (TF) enhancers for treating PDAC, the most common form of pancreatic cancer, which is aggressive with poor survival.

"To our knowledge, this is the first study to show how KRAS oncogenic mutations initiate inflammation-induced [PDAC]," said study leader Charles David, an assistant professor at the Tsinghua University School of Medicine in Beijing.

"Cancer-specific enhancers are only found in cancer cells, but not in other cell types," David told BioWorld Science.

"In other malignancies, such enhancers can be targeted, if we know both the driver TFs for that malignancy and a drug with which to target them," he explained.

"For example, in breast cancer, tamoxifen targets estrogen receptors, the driving TF of that malignancy, but most TFs do not have available drugs to target them, and are generally thought to be poor drug targets."

However, that idea is beginning to change, especially for PDAC, for which there are limited treatment options.

Treatment options

"Currently, chemotherapy is the only option for most patients, with pancreatic cancer having lagged behind other malignancies in the development of targeted therapies and successful deployment of immunotherapy," noted David. "As a result, there has been relatively little progress in increasing the survival rates for this malignancy."

Cancer originates when a specific oncogene in tissues, stem or progenitor cells is activated and preferentially transformed by oncogenic signals.

Most pancreatic cells are normally resistant to tumorigenesis due to mutant KRAS, the oncogenic driver in more than 90% of PDACs.

Rather, inflammatory pancreatitis, which is strongly linked to PDAC risk, is critical in KRAS-driven tumorigenesis. But how inflammation is linked to mutant KRAS-driven tumor initiation is poorly understood.

Experimentally, pancreatitis can be induced in mice by the cholecystokinin analogue caerulein (CAE), with such models closely reiterating the situation in humans.

"Most of our experiments were conducted in CAE mouse models, but based on single-cell ATAC-sequencing data, we are confident the results in mice are relevant to human cancer, as the enhancer elements we identified in mice were evolutionarily conserved in human cancer cells," said David.

In mice, CAE induces a strong inflammatory response, which is accompanied by acinar-to-ductal cell metaplastic (ADM) differentiation.

However, ADM resolves rapidly in the KRAS wild-type (WT) pancreas and lesions are rapidly replaced with normal pancreatic tissue.

Although ductal ADM cells have been proposed to 're-differentiate' and regenerate acinar cells, to date this has not been directly demonstrated.

Mutant KRAS dramatically alters the fate of ADM lesions, rapidly producing the PDAC precursor lesion, pancreatic intraepithelial neoplasia (PanIN), but how mutant KRAS achieves this is unknown.

After initiation, depletion of mutant KRAS from PanINs, primary tumors and metastases leads to rapid lesion regression at all stages of PDAC progression.

Mutant KRAS co-ordinates control of gene transcription in the multiple metabolic pathways essential for PDAC proliferation, survival and oncogenicity, suggesting these genes comprise a transcriptional program in the WT pancreas, but how mutant KRAS controls this is also unknown.

The new Nature Cancer study demonstrated that pancreatic inflammation is coupled to the emergence of transient progenitor cells, which were readily transformed in the presence of mutant KRAS.

"The transient emergence of these progenitor cells went hand in hand with inflammation, as shown in knock-in mice using an activated fluorescent protein in the progenitor cells," said David. "This allowed us to perform lineage tracing experiments, which conclusively demonstrated that the cells function as acinar progenitor cells."

These transient progenitor cells were then shown to harbor a proto-oncogenic transcriptional program driven by a transient enhancer network.

The KRAS mutation was found to reinforce this network, providing sustained KRAS-dependent oncogenesis-driving tumor progression.

This enhancer co-option was demonstrated to occur through functional interactions between the KRAS-activated TFs, Jun-b and Fosl1, and pancreatic lineage transcription factors, potentially accounting for the observed inter-tissue specificity of oncogene transformation, which has implications for drug development.

Drug development

"We hope that these TFs can be targeted using small molecules, although a supplementary approach to targeting the TFs is to target chromatin factors associated with them," said David.

"TFs bind to DNA, but require the activities of co-factors that they recruit to the DNA, which usually act on chromatin, either remodeling or modifying the histone proteins that package chromatin, with these chromatin factors often being amenable to targeting with drugs."

The PDA cell of origin therefore provides an oncogenic transcriptional program driving tumor progression after initiation, accounting for the intra-tissue selectivity of KRAS transformation.

However, whether these findings may be relevant to malignancies other than PDAC remains unanswered.

"We are only now beginning to explore this, but there is significant context-specificity in transcription," David said.

"Having achieved a better understanding of tumor initiation through this research, we are now also interested in understanding how this transcriptional program changes during tumor progression," he added.

"Here, we have only looked at the very earliest stages [of tumorigenesis], when the oncogene promotes formation of a benign neoplasm, but it remains unknown what happens to this transcriptional program during progression, when the tumor acquires locally invasive or metastatic properties." (Li, Y. et al. Nat Canc 2020, Advanced publication).