Differentiated cells maintain the capacity to reprogram into other lineages, which is the origin for metaplasia and cancer in multiple organs. Differentiation had largely been thought irreversible until Yamanaka and colleagues dramatically showed that cells could reprogram back to the progenitor/stem cell state. In recent years, it has become clear that this reversal of differentiated fate is actually an evolutionarily conserved response to tissue injury, wherein mature cells are recruited to other fates and/or called back into the cell cycle to serve as reserve stem cells. We have been showing that gastric zymogenic chief cells have this capacity to reverse their fate (“dedifferentiate”) and help maintain tissue integrity during Helicobacter-like injury that results in loss (atrophy) of the acid-secreting parietal cells in the stomach. The change in chief cell differentiation is what underlies a substantial portion of the metaplasia that is seen in humans infected with Helicobacter. We have introduced the drug Tamoxifen to study how chief cells dedifferentiate, proliferate, and redifferentiate, because > 3mg/20 g mouse weight induces metaplasia in the stomach. Recently, we have shown that the process or reprogramming from a mature to a regenerative state may be similar across cells and species by comparing the stages of reprogramming in stomach and pancreatic metaplasia. Because cells go through a sequence of cellular-molecular steps that are genetically controlled which can be inhibited, we have proposed that this sort of reprogramming is a general feature of cells, like apoptosis or mitosis. We term it “paligenosis.”


Of all the stem cells constitutively active in adult mammals, the one fueling the quotidian regeneration of the epithelium of the stomach has been arguably the least characterized. Pioneering work from Karam and Leblond localized its niche within the isthmus between the funnel-mouth-shaped pit region (contiguous with the cells directly lining the lumen of the stomach) and the deeper glandular portion of the gastric unit. They also defined the morphological features of the stem cell. We have applied, over the years, early laser-capture microdissection techniques and early Affymetrix microarrays to develop gene expression profiles of these cells. In more recent work, we have shown that these cells require a ℗-ERK->CD44->STAT3 signaling cascade to respond to certain types of injury (e.g., those caused by infection with the bacterium Helicobacter pylori). We have also investigated how cytokines and drugs regulate proliferation of the stem cell in vivo and in organoid models (“gastroids”). Understanding how the stem cell responds to H. pylori, which greatly increases risk for gastric cancer, should allow us to modulate the risk of infected patients to developing gastric cancer. We are also interested in how mature cells undergo paligenosis to act as “reserve” stem cells or recruited progenitor cells upon severe injury. We have recently found that a specific type of gastric metaplasia may derive, independently of stem cells, through chief cell paligenosis.



We have shown that there are evolutionarily conserved transcription factor modules that invoke specific subcellular architecture programs. We call such transcription factors “Scaling Factors” and have worked to understand how they enhance efficiency of specific cellular functions. The prototypic Scaling Factor is the bHLH MIST1 (BHLHA15), whose ortholog in flies is DIMM. MIST1 and DIMM are necessary and sufficient to scale up the entire secretory granule apparatus during differentiation of long-lived, professional secretory cells. Loss of MIST1/DIMM does not affect what cells secrete (i.e., their lineage fate choices), just how effective they are at secreting what they have been specified to secrete. Mist1 expression depends on X-box binding protein 1 (XBP1), which directly upregulates Mist1 while also vastly expanding rough endoplasmic reticulum (rER) to produce the proteins that the MIST1-governed secretory granule apparatus then stores and secretes. Protein-secreting cells throughout the body use the XBP1->MIST1 cassette to develop and maintain high-level of secretion: gastric chief cells, antibody-secreting plasma cells, and pancreatic acinar cells. We have shown the transcription factor HNF4α directly upregulates XBP1. Recently, we have also shown that MIST1 universally establishes secretory morphology in cells that perform regulated secretion.