Research may lead to reversal of acinar cell reprogramming in pancreatic tumors

Epithelial metaplasia — the replacement of one cell type in a tissue with a different but developmentally related one — is a hallmark of neoplasia. Intestinal metaplasia, for example, is a strong risk factor for esophageal adenocarcinoma in Barrett's esophagus and is also recognized as a premalignant lesion for gastric cancer.

In the exocrine pancreas, mature acinar cells may be replaced by pancreatic duct cells, leading to the formation of metaplastic ducts. Yet these ducts, thought to be a source of premalignant pancreatic intraepithelial neoplasia (PanIN), don't seem to mimic the epithelia of related tissues.

To understand why metaplastic pancreas ducts are the exception to the rule, a research team led by Howard C. Crawford, Ph.D., a cancer biologist at Mayo Clinic's campus in Florida, analyzed epithelial growth factor signaling activity in PanIN lesions in mice.

What they found was unexpected — a subset of PanIN cells with an apical bundle of filaments connected to a tuft of long microvilli that showed unusually high levels of epidermal growth factor receptor signaling. "They were very odd-looking cells, rather like Shmoos, the cartoon characters, and when stained, they lit up very strongly compared with other metaplastic cells associated with mouse pancreas tumors," Dr. Crawford says.

Using the markers DCLK1 and acetylated alpha-tubulin, researchers identified them as tuft cells, chemosensory cells that are common in gastrointestinal and pancreatobiliary epithelium, but absent in normal pancreas tissue. Tuft cells sample the environment, identifying where tissue is stressed and enabling it to respond appropriately.

"What we wanted to know," Dr. Crawford says, "is whether the cells were coming from the bile duct, where tuft cells are common, or were part of a process where acinar cells were reprogrammed to take on the character of tuft cells."

The investigators began by analyzing murine pancreatic tissue and tumors that express a form of activated KRAS. They also analyzed normal murine pancreatobiliary ductal systems and pancreatic metaplasia for tuft cell and biliary progenitor markers, including the transcription factor SOX17, which controls biliary development and differentiation. Finally, they looked at pancreatic tissue from mice engineered to express SOX17 alone or in combination with activated KRAS.

The results, published in the January 2014 issue of Gastroenterology, showed that pancreatic metaplasia was co-expressing SOX17 and PDX1, a combination of transcription factors that defined the cells as pancreatobiliary progenitor cells capable of forming both the bile duct and pancreas during embryonic development.

Genetic lineage labeling also showed that pancreatic tuft cells transdifferentiate from adult acinar cells and contain the same markers found in tuft cells in other tissues, further suggesting a common pancreatobiliary progenitor.

"We were trying to figure out what the consequences of this progenitor cell were," Dr. Crawford explains. "Another researcher, Jennifer Bailey, had identified it independently of us and showed that tuft cells act as tumor-initiating cells for pancreatic neoplasia. (Gastroenterology, January 2014). So we decided to see what would happen if SOX17 were overexpressed in the acinar cells of the pancreas. What we found is that with SOX17 expression alone, without any oncogene, the pancreas would undergo extensive replacement of normal pancreas tissue with metaplasia, including tuft cells. SOX17 combined with the KRAS oncogene greatly increased this tumor-initiating ability."

Dr. Crawford says these discoveries have great clinical potential, explaining, "If we know what is involved in reprogramming acinar cells to create tumors, we can reverse that process. We think cells remember what they used to be and can be reprogrammed back. But we need to find the signals that are controlling SOX17; we suspect there is some microenvironmental influence — fibrosis or inflammation — that is directing things, and we have to get a handle on that. Then, potentially, even in mature tumors, we may be able to inhibit cells, so they are no longer aggressive or growing."

For more information

DelGiorno K, et al. Identification and manipulation of biliary metaplasia in pancreatic tumors. Gastroenterology. 2014;146:233.

Bailey J, et al. DCLK1 marks a morphologically distinct subpopulation of cells with stem cell properties in preinvasive pancreatic cancer. Gastroenterology. 2014;146:245.