Task I: Troy+ stem cell dynamics during homeostasis and injury repair
I-1) Tracking Troy+ stem cells under homeostasis and conditions of specific cell lineage loss: Using Troy-eGFP-ires-CreER;Rosa-YFP (reporter) mice, we have collected clonal growth dynamics at the population level. This result is currently under mathematical model analysis with my collaborator – Prof. Ben Simons (University of Cambridge). In order to understand which part of the propagation dynamics is changed in response to the loss of a certain cell type, we have developed a cell type-specific diphtheria toxin Receptor (DTR) knockin mouse lines.
I-2/3) Investigating expression changes of Troy+ cells in quiescent and activated states: The Troy+ StSC showed a rapid response upon tissue damage, implying a molecular switch from quiescence to active cycling. We have uncovered the molecular switch regulating the behavior of Troy+ StSCs (Lee & Han et al. manuscript in preparation).
I-4) Determining the stem cell hierarchy in stomach corpus epithelium: We have performed unbiased random genetic labeling and lineage tracing in the stomach corpus epithelium and found the fundamental structure of gastric corpus glands. Gastric corpus glands are compartmentalized into two independent zones, harboring actively-cycling stem cells in the isthmus and slow-cycling reserve stem cells at the base. Independent lineage tracing based on Stmn1 and Ki67 expression confirmed that rapidly-cycling Isthmus stem cells (IsthSC) maintain the pit-isthmus-neck region of corpus glands. Finally, single-cell RNA-seq analysis was used to define the molecular identity and lineage relationship of a single, cycling IsthSC population (Han & Fink et al. Cell Stem Cell, 2019).
Task II: Unmasking the stomach stem cell program using in vitro and in vivo functional genetics
II-1) Investigating the tumor suppressor role of RNF43 and ZNRF3 in stomach epithelium: We have developed stomach epithelium-specific CreERT2 knock-in mouse. By crossing this new Cre driver with our RNF43;ZNRF3 conditional knockout mice, we have studied the role of RNF43 and ZNRF3 in the stomach tumorigenesis. RNF43 and ZNRF3 are two paralogues that function as a negative regulator of the Wnt signaling pathway, knockout of RNF43 and ZNRF3 induce the rapid formation of gastric hyperplasia (Kim & Merker et al. manuscript in preparation). In addition, we have found that RNF43 is under the regulation of CK1 kinase (Tsukiyama et al. Nat Comm, 2020).
II-2) Rapid gain- and loss-of-function analyses of Troy+ stem cell-specific genes in gastric organoid culture: Using PiggyBac-based dox-inducible gene expression system, we have tested the role of several candidate genes and found one regulator that clearly govern the quiescent and proliferative nature of Troy+ stomach stem cells (Lee & Han et al. manuscript in preparation).
II-3) Conditional knockout modeling of selected candidates: As mentioned in Task II-1, we have developed stomach epithelium-specific CreERT2 knock-in mouse (Anxa10-CreERT2) with Dr. Daniel Stange (TUD, Dresden). Several tumor suppressor models have been crossed with Anxa10-CreERT2 mouse in order to model human gastric cancer in the genetically engineered mouse models (Seidlitz et al. Gastroenterilogy, 2019).
Task III: Characterizing human stomach stem cells and human gastric cancer
III-1/2) Isolation and characterization of human stomach stem cells from human gastric organoids: In order to identify human stem cell markers we have taken advantage of the organoid cell culture. Normal human stomach from five patients has been successfully grown out into independent organoid lines. In normal gastric organoids, growth factor deprivation in the medium induces profound differentiation including downregulation of the Wnt signaling pathway. We have performed MassSpec of membrane enriched proteins in organoids (grown in normal medium) vs. organoids (grown in differentiation medium). 400 differentially regulated proteins have thus been identified.
III-3) Investigating the function of candidate genes using human organoids: We are testing the role of RNF43 and ZNRF3 in the human stomach organoid line. We have also set up human gastric cancer organoid models (Seidlitz et al. Gut, 2019).
