High autophagic vesicle content marks facultative stem cells of the gut

Abstract Understanding how macroautophagy/autophagy contributes to tissue homeostasis is essential for understanding organismal health. The intestinal epithelium is an ideal model to define mechanisms that regulate tissue homeostasis because it houses well-defined populations of intestinal stem cells. Active intestinal stem cells (a-ISCs) are defined by their active cycling and self-renewal during homeostasis, which supports continual tissue turnover in vivo. In vitro, this is observed as long-term organoid formation capacity. A second population of stem cells, called “facultative intestinal stem cells” (f-ISCs), are defined by their ability to 1) survive tissue damage that depletes the injury-sensitive a-ISCs and 2) reenter the cell cycle to repopulate the a-ISC compartment and regenerate the epithelium. The prospective identification of f-ISCs has been challenging, as cells expressing markers of multiple differentiated lineages, particularly secretory lineages, appear to function as f-ISCs in diverse injury contexts. We evaluated cell age (defined as time elapsed after cell cycle exit) and autophagic state (marked by autophagic vesicle content) as molecular features that may be related to f-ISC capacity. We found that autophagic state, but not cell age, prospectively identifies f-ISCs within multiple lineages. As such, we describe autophagy as a lineage-agnostic marker of f-ISC capacity in the mammalian intestine.

One of the leading hypotheses in the field is that f-ISC capacity is retained within early progenitor cells derived from cycling a-ISCs and that this capacity is progressively lost as cells differentiate. This idea purports that early progenitor cells may retain a degree of cellular plasticity enabling them to be recruited back to the stem cell compartment as a-ISCs following tissue damage. To address this hypothesis, we utilized a doxycycline-inducible histone H2B-GFP mouse model that labels all cells with GFP [1]. Following doxycycline withdrawal, cells progressively lose their GFP label as they divide and segregate their histones during replication. By combining this model with intestinal organoid culture, which provides the essential cytokine growth niche for a-ISCs, we asked whether GFP + cells shortly after doxycycline withdrawal exhibit greater organoid formation than cells that retain their GFP label for increasingly long chase periods (e.g., progressively older cells). Surprisingly, we observed that organoid formation capacity remains similar across timepoints, suggesting that cells in the intestine do not reach an epigenetic point of no return after which they can no longer function as f-ISCs and reenter the a-ISC state.
Our parallel hypothesis was that f-ISCs are highly autophagic. This was based on the following prior observations: 1) Caloric restriction, a dietary regimen characterized by elevated autophagic activity, enhances f-ISC frequency and intestinal regeneration after injury; 2) genetic impairment of autophagy in the intestinal epithelium compromises organoid-formation capacity and increases sensitivity to radiation-induced DNA damage; and 3) studies in the stomach and pancreas report that autophagy is essential for de-differentiation and cell cycle reentry in response to acute tissue injury. We therefore asked whether autophagic state correlates with f-ISC capacity in the intestinal epithelium. To address this question, we utilized an amphiphilic tracer dye, CytoID, which fluoresces when incorporated into autophagosomes and autolysosomes, but not lysosomes. Using fluorescenceactivated cell sorting (FACS) we demonstrated that autophagic vesicle content in intestinal epithelial cells correlates with organoid-formation capacity and that this advantage is mitigated in the presence of the lysosomal inhibitor bafilomycin A 1 , suggesting a functional role for autophagy in the acquisition of stem cell activity. Singlecell RNA-sequencing of CytoID-low and -high cells revealed that the CytoID-high population is composed primarily of post-mitotic secretory cells, supporting the notion that f-ISC capacity is an attribute of secretory cell lineages.
Prior literature suggests that f-ISC activity may be exclusive to secretory cells; however, studies evaluating the regenerative contribution of secretory lineages consistently report only a small subset of cells with f-ISC capacity, raising an interesting question: what distinguishes cells that behave as f-ISCs from those that do not (and perhaps cannot) within a lineage? Our findings suggest that the answer may lie in the autophagic state of the cell at the time of injury. By combining our CytoID assay with mouse models and antibody approaches that allow the identification of cells within enteroendocrine and Paneth secretory lineages, we show that autophagic vesicle content can be used to isolate cells with organoid-formation capacity from each lineage. Last, to determine whether autophagic state also correlates with the second hallmark of f-ISCs -resistance to DNA damaging injury -we stained mouse intestinal tissue for markers of autophagy (RFP-LC3) and DNA damage following whole body irradiation. We observed that intestinal epithelial cells with a high number of RFP + puncta exhibit decreased DNA damage compared to cells with little or no RFP + puncta, indicating that autophagy is correlated with and may be functionally important for resistance to DNA damage (Figure1). Our study supports the conceptual advance that f-ISCs may be holistically identified by their state -in this case, high autophagic vesicle content -rather than by lineage or differentiation state. Newer and ongoing work suggests that mechanisms regulating autophagy in intestinal stem cells are, in part, post-transcriptional. Indeed, autophagic gene signatures are not found to be differentially expressed in intestinal epithelial cells at homeostasis versus post-injury. We postulate that post-transcriptional regulation of autophagy will continue to emerge as an important factor in cell plasticity that may span tissue types or contexts. In this regard, there is significant opportunity to apply our findings to understand human diseases where autophagy dysregulation is implicated, including in diseases of the human gastrointestinal tract such as Crohn disease.

Disclosure statement
No potential conflict of interest was reported by the author(s).

Funding
This work was funded by the HHS ¦ National Institutes of Health (NIH).