The diversity of neuroendocrine tumors (NETs) – which includes variation in location, mutational profile, and response to therapy – may be due to divergent cellular origins in different tissue sites, according to a new study.
The pathogenesis of gastroenteropancreatic neoplasms (GEP-NENs) is poorly understood, in part because of a lack of modeling systems, according to Suzann Duan, PhD, and colleagues. They are a heterogeneous group of tumors that are increasingly prevalent in the United States. GEP-NENs arise from endocrine-producing cells and include gastric carcinoids, gastrinomas, and pancreatic NETs.
Despite the general mystery surrounding GEP-NENs, there is at least one clue in the form of the MEN1 gene. Both inherited and sporadic mutations of this gene are associated with GEP-NENs. Menin is a tumor suppressor protein, and previous studies have shown that inactivation of MEN1 leads to loss of that protein and is associated with endocrine tumors in the pancreas, pituitary, and upper GI tract.
In new research published in Cellular and Molecular Gastroenterology and Hepatology, researchers investigated the role of MEN1 in neuroendocrine cell development and traced it to a potential role in the development of NETs.
Patients with MEN1 mutations are at increased risk of gastrinomas, which lead to increased production of the peptide hormone gastrin. Gastrin increases acid production and can lead to hyperplasia in parietal and enterochromaffin cells. These generally develop in Brunner’s glands within the submucosa of the duodenum. At time of diagnosis, more than half of such tumors have developed lymph node metastases.
It remains unclear how loss of MEN1 suppresses gastrin production. Previous research showed that homozygous MEN1 deletion in mice is lethal to embryos, while leaving one copy intact leads to heightened risk of endocrine tumors in the pancreas and pituitary gland, but not in the GI tract. The studies did not reveal the tumor’s origin cell.
The researchers developed a novel mouse model in which MEN1 is conditionally deleted from the GI tract epithelium. This led to hyperplasia of gastrin-producing cells (G cells) in the antrum, as well as hypergastrinemia and development of gastric NETs. Exposure to a proton pump inhibitor accelerated gastric NET development, and the researchers identified expansion of enteric glial cells that expressed gastrin and GFAP. Glial cells that differentiated into endocrine phenotype were associated with a reversible loss of menin. “Taken together, these observations suggest that hyperplastic G cells might emerge from reprogrammed neural crest–derived cells in addition to endoderm-derived enteroendocrine cells,” the authors wrote.
That idea is supported by previous research indicating that multipotent glial cells expressing GFAP or SOX10 may play a developmental role in formation of neuroendocrine cells.
With this in mind, the researchers deleted MEN1 in GFAP-expressing cells to see if it would promote neuroendocrine cell development.
The result was hyperplasia in the gastric antrum and NETs in the pituitary and pancreas. To the researchers’ surprise, NET development was associated with loss of GFAP expression as well as activation of neuronal and neuroendocrine-related genes in the stomach, pancreas, and pituitary. There was universal reduction of GFAP protein expression in pituitary and pancreatic NETs, but GFAP transcript levels stayed steady in the gastric antra despite a reduction in GFAP-reporter expression. This could indicate that the menin protein interacts with GFAP. If so, eliminating menin in GFAP-positive cells could change the localization of GFAP, which may in turn lead to changes in glial cell identity.
When the researchers compared transcriptomes of hyperplastic antral tissues to well-differentiated NETs, they found that NETs exhibited a greater loss of glial-restricted progenitor lineage–associated genes as well as more downregulation of gliogenesis-directing factors. “Thus, the transition from a glial-to-neuronal cell phenotype appears to promote the progression from neuroendocrine cell hyperplasia to tumor development,” the authors wrote. They also found that NETs have higher levels of expression of genes associated with neural stem and progenitor cells, as well as upregulation of factors secreted from neural crest cells that promote neurogenesis and restrict the glial cell fate. Many of these factors are part of the Hedgehog signaling pathway, and menin is known to repress Hedgehog signaling.
Intestinal glial cells have a high degree of plasticity. They can become neuronal progenitor cells and yet they can dedifferentiate to differentiate again into other cell lineages.
The research could eventually lead to identification of unique cells-of-origin for these tumors. The authors say that the diversity of the tumors – which includes variation in location, mutational profile, and response to therapy – may be due to divergent cellular origins in different tissue sites. “Defining the cells-of-origin and the events preceding neoplastic transformation will be critical to informing molecular signaling pathways that can then be targeted therapeutically,” the authors wrote.
The authors disclosed no conflicts of interest.
