Date of Award

Winter 1-15-2021

Author's School

Graduate School of Arts and Sciences

Author's Department

Biology & Biomedical Sciences (Molecular Cell Biology)

Degree Name

Doctor of Philosophy (PhD)

Degree Type



Differentiated cells exhibit the ability to adjust their cell fate and become more progenitor-like after wide-scale tissue injury. This inherent cell plasticity is shown across many tissues and organisms and is a conserved behavior that ensures organ function even in a chronic injury setting. At the tissue level, the change in cell fate from a differentiated cell to one with more progenitor properties can be identified as metaplasia. Importantly, metaplasias, like Spasmolytic Polypeptide-Expressing Metaplasia (SPEM) in the stomach and Acinar-to-Ductal Metaplasia (ADM) in the pancreas, are risk factors for the development of adenocarcinoma. Thus, understanding the cellular and molecular mechanisms of injury-induced cell plasticity will better inform our understanding into cancer initiation.

To identify the cellular contribution to SPEM in the stomach, we used a high dose tamoxifen injury model in mice, which phenocopies human gastric metaplasia. There is debate as to whether stem cells or the differentiated zymogenic chief cells (ZCs) are the major contributing cell type to SPEM. To address this question, we impeded stem cell contribution by administering the proliferation-blocking drug 5-fluorouracil during injury, allowing only non-stem cells to function. We found that SPEM can arise, even without stem cells, and that ZCs are the most likely cells of origin for SPEM. We also observed several cell transition stages as a ZC is en route to SPEM in both our mouse model and in human metaplasia. We had strong observational studies to support ZC involvement in metaplasia initiation, but we wanted to further interrogate what signals prompt ZCs to undergo this cell plasticity event.

We previously showed one mechanism that differentiated cell use to become more progenitor-like and metaplastic following damage is paligenosis. Paligenosis consists of three sequential stages: 1. Autodegradation of mature/differentiated cell structures via lysosomes and autophagy, 2. Expression of a progenitor/embryonic gene network, 3. Cell cycle re-entry. Paligenosis is governed by biphasic mTORC1 activity; however, because paligenosis is a conserved process, we hypothesized that there must be other genes that evolved to regulate the stages. We characterized Activating Transcription Factor 3 (Atf3) as a gene necessary for paligenosis. ATF3 is largely dispensable for development and homeostasis but is upregulated early after injury in secretory cells of the stomach and pancreas during paligenosis Stage 1. We found that ATF3 works to transcriptionally upregulate Rab7b, a vesicle trafficking protein important for lysosome maturation. To determine how ATF3 controls paligenosis progression, we used Atf3–/– mice in our stomach and pancreas injury models. We found that Atf3–/– mice fail to induce large-scale lysosomes and autophagy in Stage 1 and fail to form many RAB7-vesicles, overall signifying a defect in this critical stage of paligenosis. While ATF3 plays an important role early after injury, we wanted to determine how Atf3 loss effected tissue repair and regeneration. We found increased cell death and decreased proliferation in Atf3–/– stomach and pancreas after injury, indicating that Atf3 loss impairs progression through paligenosis, which has severe consequences for organ integrity and function. We also showed that ATF3 is expressed in human gastric metaplasia, ATF3 is induced in injury models across 5 organs, and ATF3 is induced during Axolotl limb regeneration. These data suggest ATF3 is not only important for the stomach and pancreas, but that ATF3 function is key for cell plasticity and paligenosis across many tissues and organisms.

ATF3 induction is the earliest molecular event in paligenosis described thus far, but we sought to distinguish some of the injury signals that begin paligenosis and activate ATF3. The integrated stress response (ISR) is a pathway, conserved through yeast, that cells use to sense stress. The ISR coordinates a response based on several stressors – like infection, amino acid deprivation, iron deprivation, and ER stress – and ultimately works to globally reduce translation until the stress is resolved. We found that the ISR is activated during paligenosis in the stomach by the indication of phosphorylated eIF2α, the protein important for reducing translation. We also observed activation of the ISR kinase, PERK (EIF2AK3), and performed studies in mice using pharmacological inhibitors and activators of ER stress and the ISR, which altogether suggest ER stress may be one of the inputs driving paligenosis.

Overall, these studies began with a tissue-level view of metaplasia and found that mature, secretory cells are the main cells of origin for metaplasia. We also found that paligenosis is one mechanism used by cells to contribute to regeneration after injury and do so by first activating the ISR to induce ATF3. ATF3, in turn, transcriptionally regulates Rab7b and other genes coordinating upregulation of lysosomes and autophagy to help degrade mature/specialized cell structures allowing for expression of a progenitor-like gene network. These studies illustrate how cell plasticity, regeneration, and cancer initiation are tightly linked fields that rely on many of the same molecular mechanisms.


English (en)

Chair and Committee

Jason C. Mills

Committee Members

Joshua B. Rubin, Charles K. Kaufman, Jieya Shao, David J. Kast,