Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Phospho-regulation of ATOH1 Is Required for Plasticity of Secretory Progenitors and Tissue Regeneration

Phospho-regulation of ATOH1 Is Required for Plasticity of Secretory Progenitors and Tissue... Short Article Phospho-regulation of ATOH1 Is Required for Plasticity of Secretory Progenitors and Tissue Regeneration Graphical Abstract Authors Goran Tomic, Edward Morrissey, Sarah Kozar, ..., Shalev Itzkovitz, Anna Philpott, Douglas J. Winton Correspondence [email protected] (A.P.), [email protected] (D.J.W.) In Brief Tomic et al. report that multisite phosphorylation of ATOH1 regulates the contribution of secretory progenitors to stem cell self-renewal in the small intestine and colon. With damage, the enhanced role of Atoh1 progenitors in mediating tissue repair is ablated in mice expressing phosphomutant ATOH1 and overall tissue regeneration is impaired. Highlights d Atoh1 progenitors contribute to the stem cell pool in homeostasis and regeneration d Multisite phosphorylation of ATOH1 regulates the plasticity of secretory progenitors d Loss of phosphorylation of ATOH1 reduces clonogenic capacity of Atoh1 cells d Phosphomutant ATOH1 mice are more susceptible to chemical colitis Tomic et al., 2018, Cell Stem Cell 23, 436–443 September 6, 2018 ª 2018 The Authors. Published by Elsevier Inc. https://doi.org/10.1016/j.stem.2018.07.002 Cell Stem Cell Short Article Phospho-regulation of ATOH1 Is Required for Plasticity of Secretory Progenitors and Tissue Regeneration 1 2 1 3 1 4,5 1 Goran Tomic, Edward Morrissey, Sarah Kozar, Shani Ben-Moshe, Alice Hoyle, Roberta Azzarelli, Richard Kemp, 1 3 4,5, 1,6, Chandra Sekhar Reddy Chilamakuri, Shalev Itzkovitz, Anna Philpott, * and Douglas J. Winton * Cancer Research UK Cambridge Institute, University of Cambridge, Cambridge CB2 0RE, UK Weatherall Institute of Molecular Medicine, University of Oxford, John Radcliffe Hospital, Oxford OX3 9DS, UK Department of Molecular Cell Biology, Weizmann Institute of Science, Rehovot, Israel Department of Oncology, Hutchison/Medical Research Council (MRC) Research Centre, University of Cambridge, Cambridge CB2 0XZ, UK Wellcome-MRC Cambridge Stem Cell Institute, University of Cambridge, Cambridge CB2 1QR, UK Lead Contact *Correspondence: [email protected] (A.P.), [email protected] (D.J.W.) https://doi.org/10.1016/j.stem.2018.07.002 SUMMARY Lgr5 cells (Barriga et al., 2017; Buczacki et al., 2013), progeni- tors committed to different intestinal lineages (van Es et al., The intestinal epithelium is largely maintained 2012; Tetteh et al., 2016), and cells dependent on alternate path- ways for stem cell maintenance (Takeda et al., 2011; Tian by self-renewing stem cells but with apparently et al., 2011). committed progenitors also contributing, particularly It has been demonstrated previously that cells of the secretory following tissue damage. However, the mechanism lineage possess reserve stem cell function in the small intestine of, and requirement for, progenitor plasticity in medi- (SI) epithelium in homeostasis and following tissue damage (van ating pathological response remain unknown. Here Es et al., 2012; Ishibashi et al., 2018; Yan et al., 2017; Yu et al., we show that phosphorylation of the transcription 2018). Subsequent to Delta-like expression (from Dll1 or Dll4), factor Atoh1 is required for both the contribution of the basic helix-loop-helix (bHLH) transcription factor Atoh1 is secretory progenitors to the stem cell pool and for upregulated, an event required for the creation of all secretory a robust regenerative response. As confirmed by lineages within the epithelium (Yang et al., 2001). Atoh1 pro- + (WT)CreERT2 lineage tracing, Atoh1 cells (Atoh1 mice) genitors exhibit self-renewal and give rise to multilineage clones give rise to multilineage intestinal clones both in the with higher frequency in homeostasis (Ishibashi et al., 2018) compared with previously described secretory Dll1 progenitors steady state and after tissue damage. In a phospho- (9S/T-A)CreERT2 (van Es et al., 2012). This observation highlights a significant mutant Atoh1 line, preventing phos- contribution of Atoh1 cells to the stem cell pool in the SI and phorylation of ATOH1 protein acts to promote secre- colon. However, the mechanisms regulating intestinal plasticity tory differentiation and inhibit the contribution of and the nature of the relationship linking it to self-renewal remain progenitors to self-renewal. Following chemical coli- unknown. + (9S/T-A)CreERT2 tis, Atoh1 cells of Atoh1 mice have ATOH1 can be phosphorylated on multiple sites by cyclin- reduced clonogenicity that affects overall regenera- dependent kinases. Here we demonstrate that maintenance of tion. Progenitor plasticity maintains robust self- the plasticity of committed secretory precursors allowing return renewal in the intestinal epithelium, and the balance to the stem compartment is dependent on the multisite phos- between stem and progenitor fate is directly coordi- phorylation of ATOH1, prevention of which inhibits Atoh1-medi- nated by ATOH1 multisite phosphorylation. ated self-renewal and results in compromised regeneration following damage. We conclude that reversibility of the commit- ment to differentiate is dependent on post-translational control of ATOH1 and is required to maintain a robust stem cell INTRODUCTION population. Within the intestinal epithelium, cell generation occurs from phenotypically heterogenous stem cells residing at the base RESULTS of glandular crypts (Vermeulen and Snippert, 2014). There is broad consensus that this heterogeneity reflects the combined Atoh1 Cells Show Stem Cell Activity behavior of active and reserve stem cells. The former dominates Initially, to determine the extent to which Atoh1-expressing in homeostatic self-renewal and the latter following tissue cells support stem cell maintenance in homeostasis, we gener- (WT)CreERT2 T2 damage. In homeostasis, rapidly cycling stem cells express ated a mouse (Atoh1 ) with an inducible CreER the R-spondin receptor Lgr5. Reserve stem cell function is less downstream of the Atoh1 coding sequence (Figure S1A). Acute defined and has been ascribed variously to a subset of quiescent lineage tracing demonstrated that tdTomato (tdTom) reporter 436 Cell Stem Cell 23, 436–443, September 6, 2018 ª 2018 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). + Figure 1. Lineage Tracing of Atoh1 Cells in Homeostasis and after Injury (A–D) The tdTom reporter is detected in Muc2 goblet cells in the SI (A), colon (B), and Lyz Paneth cells (C) but not in ChgA enteroendocrine cells 24 hr post-tamoxifen (D). Muc2, Mucin 2; Lyz, Lysozyme; ChgA, Chromogranin A. (E) ChgA cells labeled with tdTom on day 4 after induction. (F) Dclk1 tuft cells are not labeled with tdTom at 24 hr. (G and H) Reporter-positive clone in the SI (G) and colon (H) 30 days following tamoxifen. (I–L) tdTom clones at 30 days are composed of alkaline phosphatase (Alpi ) enterocytes (I), Paneth cells (J), goblet cells (K), and enteroendocrine cells (L). (M, P, and S) Schematic of induction and injury protocol: irradiation (M), azoxymethane (AOM) (P), and dextran sodium sulfate (DSS) (S). (N) Representative pictures of SI whole-mounts containing labeled crypts (arrowheads) 30 days post-induction. (O) Quantification of tdTom crypts in the SI (n = 4 for 0 Gy, n = 6 for 6 Gy [day 1], n = 4 for 6 Gy [day 5]). (Q and T) Representative images of colonic crypts on day 30 post-tamoxifen and AOM (Q) or DSS treatment (T). Note the large tdTom regenerative multicrypt patches (MCPs) associated with 2% DSS treatment (T). (R) Quantification of reporter-positive crypts in the colon (n = 6 for untreated, n = 5 for AOM-treated). (U) Quantification of tdTom MCPs in untreated and DSS-treated colons (n = 3 for both groups). Welch’s t test was used in (O) (mean ± SEM, ****p < 0.0001) and Mann-Whitney test in (R) (mean ± SEM, **p = 0.0087). Scale bars, 50 mm (A–L) and 100 mm (N, Q, and T). See also Figure S1. cells were positive for the reporter whereas enteroendocrine cells (EECs) were not; the latter observation confirms that Atoh1 expression is not maintained in mature enteroendocrine cells (Bjerknes et al., 2012; Sommer and Mostoslavsky, 2014). However, by 4 days post-tamox- ifen, enteroendocrine cells were also labeled (Figure 1E), indicating an origin from a secretory precursor that ex- pressed Atoh1. Tuft cells were also not labeled with tdTom (Figure 1F). Individual Paneth cells remained labeled 4 weeks post-induction, reflecting their longevity (Figure S1H). Similar results were found in the colon, and long-lived secretory cells were also identified (Figure S1I). By 30 days post-induction, cohesive patches of reporter-positive cells that occupied all or a significant portion of entire crypts expression 24 hr following a single pulse of tamoxifen was were present (Figures 1G and 1H) and continued to be observed restricted to secretory cells within the SI and colonic epithelium after several months (Figure S1J). Immunostaining established (Figures 1A–1D; Figures S1B–S1G). Mature Paneth and goblet the presence of goblet, Paneth, enteroendocrine, and absorptive Cell Stem Cell 23, 436–443, September 6, 2018 437 Figure 2. Identification of a Hyperactive Phosphomutant ATOH1 (A) Diagram depicting the location of proline- directed kinase motifs (serine-proline [SP] or threonine-proline [TP]) in Atoh1 protein and mutations of these sites into alanine in ATOH1 phosphomutants. (B) In vitro-translated Atoh1 protein band-shift following incubation with different cyclin-depen- dent kinases (CDKs). Ngn3 was used as a positive control. (C) WT ATOH1 bands (arrows) collapse following l phosphatase treatment, demonstrating phos- phorylation. (D) DLD-1 cell proliferation following doxycycline (Dox)-induced expression of WT or phosphomu- tant Atoh1 (n = 3 biological replicates, 2 technical replicates, mean ± SEM). (E) Cell cycle profile of uninduced and Dox-treated cells showing increased G1 and decreased S/M populations upon induction of 9S/T-A Atoh1. (F) Gene expression of Atoh1 and its target and secretory differentiation genes 72 hr after Dox in- duction of DLD-1 cells (n = 3 biological replicates, 2 technical replicates; Gapdh-normalized, mean ± SEM). Two-way ANOVA was used for statistical analysis; *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. approach (Figures S1P and S1Q), there was a 30-fold increase in the number of clones observed (Figures S1R and S1S). cells within reporter-positive epithelium, confirming their multili- Adapting the assay to perform a similar analysis for the colonic neage composition (Figures 1I–1L). These patterns are identical epithelium and to circumvent that tissue’s known radio-resistance to those arising from individual marked intestinal stem cells (Ver- (Cai et al., 1997), mice were treated with the colon-specific carcin- meulen et al., 2013) and demonstrate a clonal origin from Atoh1 ogen azoxymethane (AOM) 1 day after tamoxifen treatment. (WT)CreERT2 TdTom + precursors. Atoh1 ; Rosa26 mice were then Again, an increase in the frequency of tdTom crypts was Gfp crossed onto Lgr5 reporter mice to investigate co-expression observed (Figures 1P–1R). Following dextran sodium sulfate of Atoh1 and the intestinal stem cell marker Lgr5. The expression (DSS)-induced colitis, multicrypt tdTom patches (MCPs) were of Atoh1 and the tdTom reporter was identified in 1%–2% of detected at the margins of regions of damage (Figures 1S–1U; Fig- + + + Lgr5 (GFP ) cells (Figures S1K–S1O), representing a likely inter- ures S1T and S1U). Together, these results suggest that Atoh1 mediate state in the commitment process and candidate clono- cells directly contribute to regeneration following damage. genic population. Together, these results confirm that Atoh1 is appropriately expressed in mature Paneth and goblet cells but Creating a Pro-secretory Phosphomutant ATOH1 not enteroendocrine cells and that a proportion of Atoh1 Previous studies have indicated that multisite phosphorylation of progenitors are acting as long-term multipotential stem cells bHLH proteins restrains cell cycle exit and limits differentiation, whereas, conversely, un(der)phosphorylation promotes these (Bjerknes et al., 2012; Sommer and Mostoslavsky, 2014; Ishiba- shi et al., 2018). processes in the developing nervous system and pancreas (Ali et al., 2011, 2014; Azzarelli et al., 2017). However, a role for multi- Atoh1 Cells Contribute Directly to Epithelial site phospho-regulation of bHLH proteins in adult homeostasis Regeneration or tissue repair has not been reported. Hence, we hypothesized The extent of reversibility of Atoh1 cell commitment was studied a potential role for ATOH1 phosphorylation in controlling the in the context of irradiation-induced tissue damage. Irradiation transition between stem and progenitor compartments both in given 1 day after tamoxifen generated an increased number of homeostasis and under conditions of heightened proliferation tdTom crypts at 30 days in the SI compared with unirradiated following tissue damage. Cyclin-dependent kinases phosphory- controls (16-fold increase, 2.37% versus 0.15%). This effect was late on serine-proline (SP) or threonine-proline (TP) residues. abrogated when irradiation was given 5 days after tamoxifen (Fig- ATOH1 has 9 S/T-P sites available for phosphorylation (Figures ures 1M–1O), suggesting that regenerative potential is a property 2A–2C). ATOH1 can be phosphorylated on many sites; we of progenitors arising de novo from the stem cell compartment and observed at least 5 distinct phospho-forms of ATOH1 after phos- not of more mature secretory cells. Similarly, after targeted dele- phorylation by different Cyclin and Cdk combinations (Figure 2B). tion of the bulk of Lgr5 stem cells using a diphtheria toxin We expressed forms of ATOH1 where S/T-P sites were mutated 438 Cell Stem Cell 23, 436–443, September 6, 2018 to alanine-proline (AP) in colorectal cancer cells to determine the Comparing proliferation between the two lines demonstrated effect of ATOH1 phosphorylation on cell proliferation and on a slight overall decrease in the total proliferative index of the (9S/T-A)CreERT2 expression of markers of differentiation. The phosphorylation crypts of Atoh1 mice in both the SI and colon, of two SP sites has previously been shown to destabilize the but that did not reach significance in the latter. More detailed ATOH1 protein in the context of neuronal precursors (Forget spatial analysis within the crypt epithelium demonstrated that et al., 2014). Although mutation of these two phospho-sites this effect was largely accounted for by a decrease in the pro- had a modest effect on ATOH1 activity, mutation of all 9 S/T-P portion of cells in S phase in the epithelium of 9S/T-A mutants sites was more effective at promoting enhanced cell cycle exit in cell positions above the very base of the crypt and a reduc- (Figures 2D and 2E). Additionally, the expression of secretory tion in the frequency of proliferative goblet cells (Figures 3F–3H; genes (Figure 2F) was enhanced after mutation of all 9 potential Figure S3A). This supports the interpretation that the phosphor- phosphorylation sites compared with both wild-type ATOH1, ylation of ATOH1 in cells immediately arising from the stem cell 2S-A ATOH1, and 7S/T-A ATOH1. These observations are population limits Atoh1-dependent cell cycle exit to allow main- consistent with multisite phospho-regulation of ATOH1 playing tenance of proliferation in progenitors. Reciprocally, preventing a significant role in controlling the balance between proliferation this phosphorylation limits the ability to return to a proliferative and differentiation, as described for other bHLH family members stem and progenitor compartment. We next tested this hypoth- (Ali et al., 2011, 2014; Azzarelli et al., 2017). esis using a lineage tracing approach. Lineage tracing and fluorescence-activated cell sorting 9S/T-A Phosphomutant ATOH1 Promotes Secretory (FACS) analysis established that the acute pattern of reporter expression and absolute number of tdTom cells were the Maturation In Vivo (9S/T-A)CreERT2 To investigate how preventing phosphorylation of ATOH1 same in Atoh1 and controls (Figures S3B–S3I). affects progenitor-mediated self-renewal in homeostasis and However, lineage tracing at 30 days identified fewer epithelial (WT)CreERT2 repair, we substituted 9S/T-A ATOH1 for the wild-type form in clones in both the SI and colon than in Atoh1 mice its endogenous locus, generating a knockin mouse identical (Figures 3I–3M). The 9S/T-A Atoh1 cells were also impaired in (WT)CreERT2 + in design to Atoh1 but with the hyperactive phospho- their ability to form tdTom clones after radiation (Figure S3J). (9S/T-A)CreERT2 mutant Atoh1 allele (Figure S2A). Homozygous Together, the observations demonstrate that preventing phos- (9S/T-A)CreERT2 (WT)CreERT2 + Atoh1 and control Atoh1 mice were phorylation of ATOH1 impairs the return of Atoh1 cells to the generated. Phenotype analysis identified no gross differences stem cell compartment and confirm a role for ATOH1 phosphor- between the two lines. Mice developed normally, and the overall ylation in maintenance of progenitor plasticity. morphological appearance of the epithelium remained un- Previously, we and others have described that only a subset of changed. More detailed analysis found no difference in the num- competing stem cells drive increases in clone sizes that lead to ber or distribution of the different secretory lineages or in the surviving clones populating entire crypts (Kozar et al., 2013; frequency of apoptotic cells (Figures S2B–S2F). Ritsma et al., 2014). To determine the net contribution of To investigate whether the 9S/T-A mutations affect secretory Atoh1 cells to this population, mathematical modeling was maturation after lineage specification, transcriptional profiling used to infer the proportion of the clonogenic fraction that is (WT)CreERT2 (9S/T-A)CreERT2 of secretory cells in the two lines was performed. First the initially marked in Atoh1 and Atoh1 + (WT)CreERT2 + expression profile of Atoh1 cells from Atoh1 mice mice. In both the SI and colon, the contribution of Atoh1 pro- was determined by comparing tdTom (secretory) and tdTom genitors to the stem cell pool is reduced in 9S/T-A animals (absorptive) cells to define the baseline pro-secretory signature (Figures 3N–3Q). Between 1% and 2% of SI crypts in (WT)CreERT2 for both the colon and SI (Table S1). Next, the transcription pro- Atoh1 mice contain a single clonogenic stem cell + + files of tdTom cells from wild-type and mutant mice (Table S2) derived from an Atoh1 progenitor, and this is reduced 5-fold + (9S/T-A)CreERT2 were compared against this Atoh1 baseline and a published in Atoh1 mice (Figures 3N and 3O). In the colon, secretory signature (Lo et al., 2016). These gene set enrichment values are higher, with the observed 4% wholly populated crypts analyses (GSEAs) demonstrated a major pro-secretory shift in (WPCs) and 5% partly populated crypts (PPCs) identified in (9S/T-A)CreERT2 (WT)CreERT2 Atoh1 mice 30 days post-induction requiring that Atoh1 mice in both tissues and a strongly reduced intestinal stem cell signature compared with those from controls initially 44% of crypts (1 in 15 active stem cells) contained an (Figures 3A–3E). Atoh1 -derived stem cell. This is reduced to 11% in 9S/T-A The pro-secretory nature of 9S/T-A-expressing cells arose mutant mice (Figures 3P and 3Q). Notably, these rates reflect from an overall elevation in pro-secretory transcripts for goblet the contribution of a single cohort of transient progenitors arising and Paneth cell lineages (SI only). Atoh1 cells isolated from from the stem cell pool that are produced over 1 or 2 days. (9S/T-A)CreERT2 the SI of Atoh1 mice had a reduction in a subset of transcripts associated with enteroendocrine cells, indi- Compromised Epithelial Regeneration in 9S/T-A cating that ATOH1 phosphorylation influences their maturation Atoh1 Mice (Figure S2G). Although phosphorylation of ATOH1 clearly regulates reversion of secretory progenitors to the stem cell compartment, the Epithelial Proliferation and Clonogenicity Are Inhibited absence of any other apparent phenotype in 9S/T-A Atoh1 9S/T-A in Atoh1 Mice mice suggests a limited requirement for such plasticity in ho- We next investigated whether the enhanced pro-secretory meostasis. We next investigated the role of ATOH1 phosphory- signature induced by prevention of multisite phosphorylation lation in mounting a robust regenerative response following of ATOH1 is accompanied by changes in proliferation. tissue damage. In the DSS-induced chemical colitis model, Cell Stem Cell 23, 436–443, September 6, 2018 439 Figure 3. 9S/T-A ATOH1 Promotes Secre- tory Maturation and Reduces Proliferation and the Number of Clonogenic Atoh1 Cells (A and B) Gene set enrichment analysis (GSEA) of the Atoh1 SI secretory signature (A) and colon (B) shows enrichment of secretory genes in 9S/T-A Atoh1 tdTom cells. (C and D) GSEA utilizing a published secretory transcriptome reveals an increase in the secretory gene signature in phosphomutant-expressing tdTom cells in the SI (C) and colon (D). (E) GSEA using a published intestinal stem cell (ISC) gene signature (Mun˜ oz et al., 2012) shows a de-enrichment of ISC genes in the mutant SI progenitors. (F and G) Bromodeoxyuridine (BrdU) labeling index for a range of cell positions in SI crypts (F) shows a reduction in proliferation above the crypt base (n = 100 crypts, 4 mice per genotype; mean ± SEM; **p = 0.0061 and 0.0015). Shown in (G) is the fre- quency of crypt-villus units containing at least one BrdU goblet cell after a 24-hr BrdU pulse (n = 5 for both groups, *p = 0.0317). (H) Representative image of a crypt-villus unit with a BrdU Alcian blue (AB) and periodic acid-Schiff (PAS) cell. (I) Quantification of tdTom clonal events in the SI (n = 6 [WT], n = 7 [9S/T-A], mean ± SEM, **p = 0.0012). (J) Partly populated tdTom crypts (PPCs) in the colon (n = 6 [WT], n = 7 [9S/T-A], mean ± SEM, **p = 0.0023). (K) Wholly populated tdTom crypts (WPCs) in WT and 9S/T-A colons (n = 6 [WT], n = 7 [9S/T-A], mean ± SEM, **p = 0.0012; the same WT data are shown in Figure 1R because the experiment was done in parallel). All samples were collected 30 days after tamoxifen. Mann-Whitney test was used for all comparisons. (L and M) Representative images of WT (L) and 9S/T-A (M) colons scored in (J) and (K). (N–Q) Inference of the proportion of the clonogenic fraction of labeled Atoh1 cells in the proximal SI (N), distal SI (O), colon PPCs (P), and colon WPCs (Q). The numbers next to the dotted lines indicate the inferred proportion of crypts that had one labeled stem cell. Scale bars, 75 mm. 9S/T-A mutant mice showed a greater sensitivity after treatment, S4D). Together, these results demonstrate that mice lacking with increased weight loss and slowed recovery (Figures 4A, the ability to phospho-regulate ATOH1 have compromised S4A, and S4B). Analysis of this phenotype at the start of the regenerative capacity following damage and that the contribu- regenerative phase (9 days after the start of DSS treatment) tion of Atoh1 progenitors is required for robust tissue repair. showed areas of ulceration that were more extensive in mice car- rying the 9S/T-A mutant (Figures 4B and 4C). At both 5 and DISCUSSION 9 days, the proportion of secretory cells was identical for the two lines, and cell death was restricted to a few cells on the It is now accepted that cells with the capacity for self-renewal luminal surface, suggesting that the greater sensitivity does not arise from a larger population whose members all have the arise from enhanced damage or deletion of secretory cells in same self-renewal potential subject to occupying available 9S/T-A mutant mice (Figures 4D–4F). However, lineage tracing niches (Farin et al., 2016; Ritsma et al., 2014). Here we show 30 days following DSS treatment identified a reduced number that Atoh1 cells make a more substantial contribution to stem and size of tdTom regenerative patches in 9S/T-A colons cell maintenance from cells committing to secretory differentia- compared with the wild-type (WT) (Figures 4G, 4H, S4C, and tion than has been recognized so far (van Es et al., 2012; 440 Cell Stem Cell 23, 436–443, September 6, 2018 (9S/T-A)CreERT2 Figure 4. Atoh1 Mice Are Sensitive to Chemical Colitis (A) Change in mouse body weight during and after DSS treatment (n = 5 [WT], n = 6 [9S/T-A]; two 9S/T-A mice were euthanized on day 9 for health reasons, and one WT animal was taken for comparison [arrowhead]). (B) Representative pictures and schematics of the colon on day 9, showing extensive loss of crypts in 9S/T-A but not in the WT. Scale bars, 1 mm. (C) Total length of colon ulceration on day 9 (n = 6 [WT], n = 4 [9S/T-A], mean ± SEM, **p = 0.0095). (D and E) Representative images of the distal colon on day 5 of DSS treatment, showing apoptosis (D) and AB and PAS staining (E) in WT and 9S/T-A animals. (F) FACS analysis of the number of tdTom cells during DSS-induced colitis. (G and H) Analysis of the number (G) and total area (H) of tdTom MCPs following 1.5% DSS (n = 4 [WT], n = 7 [9S/T-A], mean ± SEM, **p = 0.0061, *p = 0.0424). See also Figure S4. Ishibashi et al., 2018). Self-renewal is therefore not solely a the balance between stem and progenitor fate behavior in the feature driven from a fixed pool of stem cells but, rather, involves intestine can be controlled by Atoh1 multisite phosphorylation dynamic interchange between progenitors and stem cells in the under normal homeostatic conditions. steady state. Control of proliferation and differentiation by modulation of Transcription factors of the bHLH family have been extensively bHLH protein phosphorylation is emerging as an important studied as master regulators of cell fate commitment and differ- mechanism in development of the nervous system and the entiation in a wide variety of tissues, including the nervous pancreas (Cleaver, 2017; Guillemot and Hassan, 2017). We system and intestine (Ali et al., 2011, 2014; Yang et al., 2001). now demonstrate that multisite phosphorylation is also required However, in recent years, additional roles for these proteins to restrain irreversible commitment of secretory precursors in the are emerging in direct co-ordination of cell cycle and differentia- adult homeostatic gut and so to maintain their ability to repopu- tion events, particularly during embryonic development (Castro late the stem cell compartment. Consistent with this, a phospho- and Guillemot, 2011). Intestinal homeostasis in many ways rep- mutant form of ATOH1 enhances the expression of gene sets resents an ongoing development-like hierarchical process where associated with a more mature secretory phenotype in colo- crypts are maintained by stem cells feeding a proliferating pro- rectal carcinoma cells. Interestingly, in the homeostatic gut, genitor compartment that gives rise to a variety of mature cell despite Atoh1 cells normally supplying up to 1 in 15 cells in types. What is now also emerging is a picture of significant plas- the stem cell compartment, the phosphomutant Atoh1-express- ticity where cells expressing Atoh1, previously thought to repre- ing intestine is essentially phenotypically normal, indicating that sent a population that has undergone secretory commitment, plasticity from the secretory to the stem compartment is not can nevertheless revert to ‘‘stemness’’ and repopulate the entire essential in normal homeostasis. However, intestinal regenera- crypt with surprisingly high frequency. The mechanisms control- tion after damage is substantially compromised by an inability ling this plasticity have been unclear. Here we determine that to phosphorylate ATOH1. Cell Stem Cell 23, 436–443, September 6, 2018 441 B Western blotting Taken together, our results indicate that multisite phosphory- B In vitro kinase assay lation of ATOH1 is used to dynamically regulate the return of B Cell proliferation and cell cycle analysis secretory precursors to the stem cell compartment, which facil- B RNA sequencing analysis itates the capacity of the epithelium as a whole to respond B Secretory signature gene list rapidly to changes in the cellular environment. Damaging the in- B Gene Set Enrichment Analysis (GSEA) testine using irradiation or DSS (van Es et al., 2012; Ishibashi d QUANTIFICATION AND STATISTICAL ANALYSIS et al., 2018) leads to acute cell damage and death, followed by B Computational analysis proliferative regeneration that produces new cells for tissue B Model fitting repair. Activation of cyclin-dependent kinases (CDKs) and B Statistical analysis mitogen-activated protein kinases (MAPKs) in rapidly prolifer- d DATA AND SOFTWARE AVAILABILITY ating cells undergoing regeneration would result in enhanced phosphorylation of ATOH1, restraining further progression down the secretory lineage and supporting re-entry of Atoh1-ex- SUPPLEMENTAL INFORMATION pressing cells into a stem-like state. The post-translational regu- Supplemental Information includes four figures and three tables and can be lation of ATOH1 by proline-directed kinases to modulate the found with this article online at https://doi.org/10.1016/j.stem.2018.07.002. balance between proliferation and differentiation in response to changing tissue demands in the adult intestinal epithelium ACKNOWLEDGMENTS echoes the regulation and effect of other bHLH proteins as development progresses (Hardwick et al., 2015). This work was supported by Cancer Research UK (to G.T., E.M., A.H., R.K., The secretory fate choice mediated by ATOH1, a master regu- and D.J.W.) and Wellcome Trust grant 103805 (to G.T. and D.J.W.). A.P. and lator, is not irreversible differentiation; rather, it is entry into a R.A. were funded by MRC research grant MR/K018329/1 and grant MR/ plastic state through which progression is regulated by post- L021129/1, the Rosetrees Trust, and the Stoneygate Trust and received core support from the Wellcome-MRC Cambridge Stem Cell Institute. We translational modifications. Functionally, the implications are thank S. Taylor for DLD-1 Flp-In T-Rex cells and D. Perera for plasmids. We likely to be that post-translational modifications facilitate rapid thank the Histopathology core and Biological Resources Unit at the CRUK cellular responses by allowing reversal of commitment or varying Cambridge Institute for technical support and the members of the Winton its extent or rate. Progenitor plasticity is not merely an incidental and Philpott labs for their help. acquired behavior following damage but plays an integral part in tissue restoration and requires post-translational regulation AUTHOR CONTRIBUTIONS of ATOH1. Conceptualization, D.J.W., A.P., G.T., and R.K.; Methodology, G.T., D.J.W., and A.P.; Formal Analysis, G.T., E.M., and C.S.R.C.; Investigation, G.T., STAR+METHODS S.K., S.B.-M., A.H., and R.A.; Writing – Original Draft, D.J.W., A.P., and G.T.; Writing – Review & Editing, G.T., D.J.W., and A.P.; Visualization, G.T.; Super- vision, D.J.W., A.P., and S.I.; Project Administration, D.J.W. and A.P.; Funding Detailed methods are provided in the online version of this paper Acquisition, D.J.W. and A.P. and include the following: DECLARATION OF INTERESTS d KEY RESOURCES TABLE d CONTACT FOR REAGENT AND RESOURCE SHARING The authors declare no competing interests. d EXPERIMENTAL MODEL AND SUBJECT DETAILS B Mice Received: February 2, 2018 B Cell Lines Revised: May 25, 2018 Accepted: July 6, 2018 d METHOD DETAILS Published: August 9, 2018 B Cloning of mouse knock-in constructs B ES cell targeting REFERENCES B Mouse genotyping B Creation of doxycycline inducible DLD-1 cells Ali, F., Hindley, C., McDowell, G., Deibler, R., Jones, A., Kirschner, M., B Treatment of animals Guillemot, F., and Philpott, A. (2011). Cell cycle-regulated multi-site phosphor- B Crypt fractionation and single cell preparation ylation of Neurogenin 2 coordinates cell cycling with differentiation during neu- B Flow cytometry rogenesis. Development 138, 4267–4277. B Whole-mount preparation Ali, F.R., Cheng, K., Kirwan, P., Metcalfe, S., Livesey, F.J., Barker, R.A., and Philpott, A. (2014). The phosphorylation status of Ascl1 is a key determinant B Whole-mount scanning and quantification of neuronal differentiation and maturation in vivo and in vitro. Development B Antibody staining 141, 2216–2224. B Quantification of crypts in whole-mounts Azzarelli, R., Hurley, C., Sznurkowska, M.K., Rulands, S., Hardwick, L., B Immunostainings Gamper, I., Ali, F., McCracken, L., Hindley, C., McDuff, F., et al. (2017). B Single molecule FISH Multi-site Neurogenin3 Phosphorylation Controls Pancreatic Endocrine B Analysis of gut sections Differentiation. Dev. Cell 41, 274–286.e5. B Colon ulceration scoring Barriga, F.M., Montagni, E., Mana, M., Mendez-Lago, M., Hernando- B RNA isolation Momblona, X., Sevillano, M., Guillaumet-Adkins, A., Rodriguez-Esteban, G., B Gene expression analysis Buczacki, S.J.A., Gut, M., et al. (2017). Mex3a Marks a Slowly Dividing B RNA sequencing Subpopulation of Lgr5+ Intestinal Stem Cells. Cell Stem Cell 20, 801–816.e7. 442 Cell Stem Cell 23, 436–443, September 6, 2018 Bjerknes, M., Khandanpour, C., Mo¨ ro¨ y, T., Fujiyama, T., Hoshino, M., Klisch, Love, M.I., Huber, W., and Anders, S. (2014). Moderated estimation of fold T.J., Ding, Q., Gan, L., Wang, J., Martı´n, M.G., and Cheng, H. (2012). Origin change and dispersion for RNA-seq data with DESeq2. Genome Biol. 15, 550. of the brush cell lineage in the mouse intestinal epithelium. Dev. Biol. 362, Lyubimova, A., Itzkovitz, S., Junker, J.P., Fan, Z.P., Wu, X., and van 194–218. Oudenaarden, A. (2013). Single-molecule mRNA detection and counting in Buczacki, S.J.A., Zecchini, H.I., Nicholson, A.M., Russell, R., Vermeulen, L., mammalian tissue. Nat. Protoc. 8, 1743–1758. Kemp, R., and Winton, D.J. (2013). Intestinal label-retaining cells are secretory Mun˜ oz, J., Stange, D.E., Schepers, A.G., van de Wetering, M., Koo, B.-K., precursors expressing Lgr5. Nature 495, 65–69. Itzkovitz, S., Volckmann, R., Kung, K.S., Koster, J., Radulescu, S., et al. Cai, W.B., Roberts, S.A., Bowley, E., Hendry, J.H., and Potten, C.S. (1997). (2012). The Lgr5 intestinal stem cell signature: robust expression of proposed Differential survival of murine small and large intestinal crypts following ionizing quiescent ’+4 cell markers. EMBO J. 31, 3079–3091. radiation. Int. J. Radiat. Biol. 71, 145–155. Ritsma, L., Ellenbroek, S.I.J., Zomer, A., Snippert, H.J., de Sauvage, F.J., Carpenter, B., Gelman, A., Hoffman, M.D., Lee, D., Goodrich, B., Betancourt, Simons, B.D., Clevers, H., and van Rheenen, J. (2014). Intestinal crypt homeo- M., Brubaker, M., Guo, J., Li, P., and Riddell, A. (2017). Stan : A Probabilistic stasis revealed at single-stem-cell level by in vivo live imaging. Nature 507, Programming Language. J. Stat. Softw. 76, 1–32. 362–365. Castro, D.S., and Guillemot, F. (2011). Old and new functions of proneural fac- Snippert, H.J., van der Flier, L.G., Sato, T., van Es, J.H., van den Born, M., tors revealed by the genome-wide characterization of their transcriptional tar- Kroon-Veenboer, C., Barker, N., Klein, A.M., van Rheenen, J., Simons, B.D., gets. Cell Cycle 10, 4026–4031. and Clevers, H. (2010). Intestinal crypt homeostasis results from neutral competition between symmetrically dividing Lgr5 stem cells. Cell 143, Cleaver, O. (2017). b Cell Renewal versus Differentiation: Slow and Steady 134–144. Wins the Race. Dev. Cell 41, 223–225. Sommer, C.A., and Mostoslavsky, G. (2014). RNA-Seq analysis of enteroendo- Farin, H.F., Jordens, I., Mosa, M.H., Basak, O., Korving, J., Tauriello, D.V.F., de crine cells reveals a role for FABP5 in the control of GIP secretion. Mol. Punder, K., Angers, S., Peters, P.J., Maurice, M.M., and Clevers, H. (2016). Endocrinol. 28, 1855–1865. Visualization of a short-range Wnt gradient in the intestinal stem-cell niche. Nature 530, 340–343. Subramanian, A., Tamayo, P., Mootha, V.K., Mukherjee, S., Ebert, B.L., Gillette, M.A., Paulovich, A., Pomeroy, S.L., Golub, T.R., Lander, E.S., and Forget, A., Bihannic, L., Cigna, S.M., Lefevre, C., Remke, M., Barnat, M., Mesirov, J.P. (2005). Gene set enrichment analysis: a knowledge-based Dodier, S., Shirvani, H., Mercier, A., Mensah, A., et al. (2014). Shh signaling approach for interpreting genome-wide expression profiles. Proc. Natl. protects Atoh1 from degradation mediated by the E3 ubiquitin ligase Huwe1 Acad. Sci. USA 102, 15545–15550. in neural precursors. Dev. Cell 29, 649–661. Takeda, N., Jain, R., LeBoeuf, M.R., Wang, Q., Lu, M.M., and Epstein, J.A. Guillemot, F., and Hassan, B.A. (2017). Beyond proneural: emerging functions (2011). Interconversion between intestinal stem cell populations in distinct and regulations of proneural proteins. Curr. Opin. Neurobiol. 42, 93–101. niches. Science 334, 1420–1424. Hardwick, L.J.A., Ali, F.R., Azzarelli, R., and Philpott, A. (2015). Cell cycle regu- Tetteh, P.W., Basak, O., Farin, H.F., Wiebrands, K., Kretzschmar, K., Begthel, lation of proliferation versus differentiation in the central nervous system. Cell H., van den Born, M., Korving, J., de Sauvage, F., van Es, J.H., et al. (2016). Tissue Res. 359, 187–200. Replacement of Lost Lgr5-Positive Stem Cells through Plasticity of Their Ishibashi, F., Shimizu, H., Nakata, T., Fujii, S., Suzuki, K., Kawamoto, A., Anzai, + Enterocyte-Lineage Daughters. Cell Stem Cell 18, 203–213. S., Kuno, R., Nagata, S., Ito, G., et al. (2018). Contribution of ATOH1 Cells to the Homeostasis, Repair, and Tumorigenesis of the Colonic Epithelium. Stem Tian, H., Biehs, B., Warming, S., Leong, K.G., Rangell, L., Klein, O.D., and de Sauvage, F.J. (2011). A reserve stem cell population in small intestine renders Cell Reports 10, 27–42. Lgr5-positive cells dispensable. Nature 478, 255–259. Itzkovitz, S., Lyubimova, A., Blat, I.C., Maynard, M., van Es, J., Lees, J., Jacks, van Es, J.H., Sato, T., van de Wetering, M., Lyubimova, A., Yee Nee, A.N., T., Clevers, H., and van Oudenaarden, A. (2011). Single-molecule transcript counting of stem-cell markers in the mouse intestine. Nat. Cell Biol. 14, Gregorieff, A., Sasaki, N., Zeinstra, L., van den Born, M., Korving, J., et al. (2012). Dll1+ secretory progenitor cells revert to stem cells upon crypt dam- 106–114. age. Nat. Cell Biol. 14, 1099–1104. Kim, D., Pertea, G., Trapnell, C., Pimentel, H., Kelley, R., and Salzberg, S.L. Vermeulen, L., and Snippert, H.J. (2014). Stem cell dynamics in homeostasis (2013). TopHat2: accurate alignment of transcriptomes in the presence of in- sertions, deletions and gene fusions. Genome Biol. 14, R36. and cancer of the intestine. Nat. Rev. Cancer 14, 468–480. Vermeulen, L., Morrissey, E., van der Heijden, M., Nicholson, A.M., Sottoriva, Kozar, S., Morrissey, E., Nicholson, A.M., van der Heijden, M., Zecchini, H.I., Kemp, R., Tavare´ , S., Vermeulen, L., and Winton, D.J. (2013). Continuous A., Buczacki, S., Kemp, R., Tavare, S., and Winton, D.J. (2013). Defining stem clonal labeling reveals small numbers of functional stem cells in intestinal cell dynamics in models of intestinal tumor initiation. Science 342, 995–998. crypts and adenomas. Cell Stem Cell 13, 626–633. Yan, K.S., Gevaert, O., Zheng, G.X.Y., Anchang, B., Probert, C.S., Larkin, K.A., Lallemand, Y., Luria, V., Haffner-Krausz, R., and Lonai, P. (1998). Maternally Davies, P.S., Cheng, Z.F., Kaddis, J.S., Han, A., et al. (2017). Intestinal expressed PGK-Cre transgene as a tool for early and uniform activation of Enteroendocrine Lineage Cells Possess Homeostatic and Injury-Inducible the Cre site-specific recombinase. Transgenic Res. 7, 105–112. Stem Cell Activity. Cell Stem Cell 21, 78–90.e6. Lo, Y.-H., Chung, E., Li, Z., Wan, Y.-W., Mahe, M.M., Chen, M.-S., Noah, T.K., Yang, Q., Bermingham, N.A., Finegold, M.J., and Zoghbi, H.Y. (2001). Bell, K.N., Yalamanchili, H.K., Klisch, T.J., et al. (2016). Transcriptional Requirement of Math1 for secretory cell lineage commitment in the mouse in- Regulation by ATOH1 and its Target SPDEF in the Intestine. Cell. Mol. testine. Science 294, 2155–2158. Gastroenterol. Hepatol. 3, 51–71. Yu, S., Tong, K., Zhao, Y., Balasubramanian, I., Yap, G.S., Ferraris, R.P., Lopez-Garcia, C., Klein, A.M., Simons, B.D., and Winton, D.J. (2010). Intestinal Bonder, E.M., Verzi, M.P., and Gao, N. (2018). Paneth Cell Multipotency stem cell replacement follows a pattern of neutral drift. Science 330, 822–825. Induced by Notch Activation following Injury. Cell Stem Cell 23, 46–59.e5. Cell Stem Cell 23, 436–443, September 6, 2018 443 STAR+METHODS KEY RESOURCES TABLE REAGENT or RESOURCE SOURCE IDENTIFIER Antibodies Mouse Monoclonal anti-Atoh1 Developmental Studies Hybridoma Bank Cat# Math1 (Atoh1): RRID:AB_10805299 Rabbit Polyclonal anti-b-actin Abcam Cat# ab8227: RRID:AB_2305186 IRDye 800CW Goat anti-Mouse IgG (H + L) LI-COR Biosciences Cat# P/N 925-32210: RRID:AB_2687825 IRDye 680LT Goat anti-Rabbit IgG (H + L) LI-COR Biosciences Cat# P/N 925-68021: RRID:AB_2713919 Rat Anti-Mouse CD326 (Ep-CAM) BioLegend Cat# 118210: RRID:AB_1134099 Monoclonal Antibody, Alexa Fluor 488 Conjugated, Clone G8.8 Sheep Polyclonal BrdU Antibody Abcam Cat# ab1893: RRID:AB_302659 Rabbit Polyclonal Anti Human Lysozyme Dako Cat# A0099: RRID:AB_2341230 Biotin-SP-AffiniPure Donkey Anti-Sheep Jackson ImmunoResearch Labs Cat# 713-066-147: RRID:AB_2340717 IgG (H+L) Biotin-SP-AffiniPure Donkey Anti-Rabbit Jackson ImmunoResearch Labs Cat# 711-065-152: RRID:AB_2340593 IgG (H+L) Rabbit Anti-Chromogranin A Polyclonal Abcam Cat# ab15160: RRID:AB_301704 Antibody Rabbit Polyclonal Anti-Synaptophysin Millipore Cat# AB9272: RRID:AB_570874 Antibody Rabbit Polyclonal anti-DCAMKL1 Antibody Abcam Cat# ab31704: RRID:AB_873537 Rabbit Anti-Human Lysozyme Polyclonal Dako Cat# F037201: RRID:AB_578661 Antibody, FITC Conjugated Rabbit Anti-Mucin 2 Polyclonal Antibody Santa Cruz Biotechnology Cat# sc-15334: RRID:AB_2146667 Donkey anti-Rabbit IgG (H+L) Secondary Thermo Fisher Scientific Cat# A-21206: RRID:AB_2535792 Antibody, Alexa Fluor 488 Bacterial and Virus Strains RP24-77K22 Bacterial Artificial BACPAC Resources Center N/A Chromosome Chemicals, Peptides, and Recombinant Proteins Lambda Protein Phosphatase (Lambda PP) New England Biolabs Cat# P0753S Doxycycline Hydrochloride, Ready Made Sigma-Aldrich Cat# D3072 Solution Tet Approved FBS Clontech Laboratories Cat# 631101 Tamoxifen Sigma-Aldrich Cat# T5648 Dextran Sulfate Sodium Salt MP Biomedicals Cat# 02160110 Critical Commercial Assays In-Fusion HD Cloning Kit Clontech Laboratories Cat# 639648 TruSeq Stranded mRNA Library Prep Kit Illumina Cat# 20020595 Deposited Data RNA sequencing data This paper GEO: GSE115416 Mendeley Data This paper https://doi.org/10.17632/vgvdv5b949.1 Experimental Models: Cell Lines DLD-1 Flp-In T-Rex cell line Laboratory of Stephen Taylor N/A Experimental Models: Organisms/Strains (WT)CreERT2 Atoh1 This paper N/A (9S/T-A)CreERT2 Atoh1 This paper N/A (Continued on next page) e1 Cell Stem Cell 23, 436–443.e1–e7, September 6, 2018 Continued REAGENT or RESOURCE SOURCE IDENTIFIER Oligonucleotides Left integration arm, Atoh1 locus, forward This paper N/A primer GGACAGGCGGGAACCACAGA Left integration arm, Atoh1 locus, reverse This paper N/A primer TTGTCAACACGAGCTGGTCGAA Right integration arm, Atoh1 locus, forward This paper N/A primer CAACACAACCCTGACCTGTG Right integration arm, Atoh1 locus, reverse This paper N/A primer CCCTAACCAGTGTGCCCTTA Left integration arm, DLD-1, forward primer This paper N/A AGTCAG CAACCATAGTCCCG Left integration arm, DLD-1, reverse primer This paper N/A TTCTGCGGGCGATTTGTGTA Right integration arm, DLD-1, forward This paper N/A primer TAAACGGCCACAAGTTCAGC Left integration arm, DLD-1, reverse primer This paper N/A CGGGCCTCTTCGCTATTACG Atoh1 genotyping forward primer This paper N/A TTTGTTGTTGTTGTTCGGGG Atoh1 genotyping reverse primer This paper N/A TCTTTTACCTCAGCCCACTCTT Software and Algorithms TopHat2 Kim et al., 2013 N/A DESeq2 Love et al., 2014 N/A GSEA Subramanian et al., 2005 http://software.broadinstitute.org/ gsea/index.jsp CONTACT FOR REAGENT AND RESOURCE SHARING Further information and requests for resources and reagents should be directed to and will be fulfilled by the Lead Contact, Douglas J. Winton ([email protected]). EXPERIMENTAL MODEL AND SUBJECT DETAILS Mice Mice used in this study were 8-16 weeks old males and females of C57BL/6 background. The mice were housed under controlled conditions (temperature (21 ± 2 C), humidity (55 ± 10%), 12 h light/dark cycle) in a specific-pathogen-free (SPF) facility (tested ac- cording to the recommendations for health monitoring by the Federation of European Laboratory Animal Science Associations). The animals had unrestricted access to food and water, were not involved in any previous procedures and were test naive. All experi- (WT)CreERT2 (9S/T-A)CreERT2 ments were carried out on homozygous Atoh1 and Atoh1 lines. For lineage tracing experiments, the mice tdTom/+ were heterozygous for the reporter gene (Rosa26 ). All animal experiments were carried out in accord with the guidelines of the UK Home Office, under the authority of a Home Office project license approved by the Animal Welfare and Ethical Review Body at the CRUK Cambridge Institute, University of Cambridge. Cell Lines DLD-1 (human colon adenocarcinoma, male) cells, modified with the Flp-In T-Rex system (Thermo Fisher), were used in the study. The cell line authentication was carried out using Single Tandem Repeat (STR) genotyping. Tests were performed routinely to confirm mycoplasma-negative status of the cells. The cells were maintained in Dulbecco’s Modified Eagle Medium (DMEM) with L-glutamine. Medium was supplemented with 10% Tet System-approved fetal bovine serum (FBS, Clontech). The cells were cultured under stan- dard conditions (5% CO ,37 C). Cell Stem Cell 23, 436–443.e1–e7, September 6, 2018 e2 METHOD DETAILS Cloning of mouse knock-in constructs For generation of mouse knock-ins Atoh1 locus and homology arms were amplified from a baterial artificial chromosome (BAC) RP24-77K22 (BACPAC Resources Centre). The targeting construct was assembled by a combination of seamless cloning (In-Fusion, Clontech) and restriction digest and ligation. For this a loxP site was introduced into 5 UTR of Atoh1 via PCR amplification. T2 0 0 A neomycin cassette was inserted such that the 3 UTR was not disrupted. The CreER -hCD2-3 UTR was generated via gene syn- (WT) (9S/T-A) thesis service (Integrated DNA Technologies). The Atoh1 sequence (Atoh1 or Atoh1 ) was merged with this construct, and then ligated with Atoh1 vector containing the homology arms. The targeting vector sequence was verified by Sanger sequencing and linearized by SwaI enzyme before transfecting into ES cells. The final inserted sequence is available on request. ES cell targeting Electroporation of the targeting construct into mouse ES cells was conducted by the CRUK CI Transgenic Core. ES cells were posi- tively selected with G418. Correct integration of the construct was verified by long range PCR (SequalPrep, Thermo Fisher) according 0 0 to the manufacturer’s instructions. Left integration arm was detected using a forward primer 5 -GGA CAG GCG GGA ACC ACA GA-3 0 0 and a reverse primer 5 -TTG TCA ACA CGA GCT GGT CGA A-3 . Right integration arm was amplified using the following set of 0 0 0 0 primers: forward 5 - CAA CAC AAC CCT GAC CTG TG-3 , and reverse 5 -CCC TAA CCA GTG TGC CCT TA-3 . Copy number of the clones was determined by qPCR of the neomycin selection cassette via a commercial genotyping service provider (Transnetyx). Single copy ES cell clones were taken forward for blastocyst injection, and chimeric mice were generated. Following successful germline transmission, the mice heterozygous for the targeting construct were crossed onto PGK-Cre line (Lallemand et al., 1998) in order to remove both the neo selection cassette and the endogenous Atoh1 locus at the same time. A constitutively active T2 Atoh1-P2A-CreER allele was generated in this process. Mouse genotyping Genotyping was carried out by Transnetyx. Manual genotyping by PCR was used to distinguish between homozygous and hetero- 0 0 0 zygous Atoh1 animals. The following primers were used: forward 5 -TTT GTT GTT GTT GTT CGG GG-3 ; reverse 5 -TCT TTT ACC TCA GCC CAC TCT T-3 . Creation of doxycycline inducible DLD-1 cells To generate an inducible stable cell line, a DLD-1 Flp-In T-Rex cell line containing a single Frt site was obtained (a generous gift from Prof Stephen Taylor, University of Manchester). Atoh1 construct in a pcDNA 5/FRT/TO vector (Thermo Fisher) was co-transfected with pOG44 (Flp recombinase-expressing plasmid) in a 1:9 ratio (JetPrime, Polyplus transfection). Cells were washed 24 h after trans- fection, and fresh medium was added. Two days after transfection, the cells were split at a low confluence (less than 25%), and hygromycin (400 mg/mL) was added to the trypsinised cells. Fresh medium was added to the cells every 3-4 days, until the non-trans- fected cells died off, and foci of surviving cells could be visualized. Doxycycline (100 ng/mL, Sigma) was added to the culture 24 h after seeding, to induce expression of the gene of interest. Validation of the correct recombination of the construct was carried out by PCR. Left integration arm was detected by using the 0 0 0 0 following set of primers: forward 5 -AGT CAG CAA CCA TAG TCC CG-3 ; reverse 5 - TTC TGC GGG CGA TTT GTG TA-3 . Correct 0 0 0 integration on the 3 end of the construct was done using a forward primer 5 -TAA ACG GCC ACA AGT TCA GC-3 , and a reverse 0 0 primer 5 -CGG GCC TCT TCG CTA TTA CG-3 . Parental DLD-1 Flp-In T-Rex cell line was used as a negative control. The expected loss of b-galactosidase activity on targeting was verified by X-Gal (5-bromo-4-chloro-3-indolyl-b-D-galactopyrano- side) staining of fixed cells. To validate that the constructs were integrated as a single copy in the genome, copy number qPCR was employed. Copy number TaqMan probes detected HygR (Mr00661678_cn) and used a reference copy number assay for RNase P detection. Treatment of animals T2 Induction of CreER in animals was carried out using the free base tamoxifen (Sigma) dissolved in ethanol/oil (1:9). The animals received 3 mg tamoxifen via an intra-peritoneal injection in all experiments. To define Atoh1 secretory signature, the mice were in- jected with 1 mg tamoxifen per day on 3 consecutive days for maximal labeling of all secretory lineages. SI injury was induced by exposing animals to whole-body irradiation (6 Gy). To induce colon-specific injury, mice were given 1.5% DSS (MP Biomedicals) in drinking water for 5 days. DSS was replaced every two days during the treatment. To induce lineage + DTR tracing and ablate Lgr5 cells in Lgr5 mice, the animals first received 3 mg tamoxifen i.p., followed by an i.p. injection of DT in saline (50 mg/kg) 6 h later. Crypt fractionation and single cell preparation SI (proximal 15 cm) and colon were dissected, flushed with PBS, everted and fed onto a glass rod spiral. They were incubated at 37 C +2 +2 in Hank’s Balanced Salt Solution (HBSS) without Ca and Mg , containing 10 mM EDTA and 10 mM NaOH. Crypt release was facil- itated using a vibrating stirrer (Chemap). Samples were incubated for 1 h and pulsed every 10 min. Fractions were collected after each pulse, and fresh solution added. Crypt-enriched fractions were pooled and washed in cold 2% FBS/PBS. Fraction 1 (villus-enriched) e3 Cell Stem Cell 23, 436–443.e1–e7, September 6, 2018 was discarded. Pooled fractions were resuspended in 0.05% trypsin and incubated for 7 min at 37 C, shaking every 1 min. Single cells were then filtered through a 70 mm mesh, and washed twice in 2% FBS/PBS. Flow cytometry Single cell suspension obtained by trypsin treatment was washed and incubated with an anti-mouse CD326 (EpCAM) AlexaFluor 647 antibody (1:2,000, clone G8.8, Biolegend). DAPI (10 mg/mL) was added to distinguish between live and dead cells. Flow sorting was carried out on a BD FACS Aria SORP (BD Biosciences), using appropriate single-stained and unstained controls. Whole-mount preparation Tissue was cut open, pinned out luminal side up, and fixed for 3 h at room temperature in ice-cold 4% PFA in PBS (pH 7.4). Whole- mounts were washed with PBS, and incubated with demucifying solution (3 mg/mL dithiothreitol (DTT), 20% ethanol, 10% glycerol, 0.6% NaCl, 10 mM Tris, pH 8.2) for 20 min, and mucus removed by washing with PBS. Whole-mount scanning and quantification The tdTom fluorescence in colon whole-mounts was detected using Amersham Typhoon 5 laser scanner (GE Healthcare) at a 10 mm resolution. The tdTom foci were scored manually in Fiji. Mid and distal colon were scored only as the shape of the proximal colon prevented confident assessment of tdTom patches. Antibody staining For staining whole-mount sections of 2 cm in length were excised, washed in 0.1% PBS-T for 2 days, and blocked in 10% donkey serum in PBS overnight at 4 C, protected from light. Samples were then incubated with an anti-mouse CD326 (EpCAM) AlexaFluor 647 antibody (1:100, clone G8.8, Biolegend) in 10% donkey serum in PBS for 3 days. Finally, the tissue was washed with PBS-T for 1 day. Quantification of crypts in whole-mounts Imaging was done on a TCS SP5 confocal microscope (Leica). Images were analyzed using Fiji. For SI, a minimum of 2,500 crypts per animal was scored. For colon, at least 900 crypts per mouse were scored. For the low-power analysis of clonal events, tdTom clones were scored across the entire length of the SI whole-mounts using a stereomicroscope (Nikon). Immunostainings For immunohistochemistry SI and colon were opened and fixed for 24 h in 10% neutral buffered formaldehyde in PBS. The tissue was paraffin embedded and sectioned by the CRUK CI Histopathology core. Haematoxylin and eosin staining was performed using an automated ST5020 Multistainer (Leica Biosystems). Alcian Blue and Periodic Acid/Schiff staining was carried out by the CI Histopa- thology Core. Briefly, slides were incubated in Alcian Blue for 10 min, and washed in water. They were then incubated in 0.5% pe- riodic acid for 5 min, and washed 3 times. Slides were incubated in Schiff’s reagent for 15 min, washed 3 times, and counterstained with Mayer’s Haematoxylin. BrdU and lysozyme immunohistochemistry was carried out using a Bond Max autostainer (Leica), with a proteinase K antigen retrieval. Slides were blocked with 3% hydrogen peroxide, followed by incubation in Avidin/Biotin Blocking Kit (Vector Laboratories). BrdU was detected using a sheep anti-BrdU antibody (1:500, Abcam ab1893). Rabbit anti-lysozyme antibody (1:500, Dako A0099) was used for lysozyme staining. Secondary antibodies in the two cases were biotinylated donkey anti-sheep (1:250, Jackson ImmunoResearch 713-066-147) and biotinylated donkey anti-rabbit (1:250, Jackson ImmunoResearch 711-065-152), respectively. Slides were incubated with Streptavidin coupled with horseradish peroxidase (HRP), and color developed using diaminobenzidine (DAB) and DAB Enhancer (Leica). Synaptophysin and Chromogranin A detection was carried out by manual IHC. Antigen retrieval was performed with 10 mM citrate buffer (pH 6.0) in a pressurised heating chamber. Tissue sections were incubated with rabbit anti-Chromogranin A antibody (1:500, Abcam ab15160), rabbit anti-Synaptophysin antibody (1:300, Millipore AB9272), overnight at 4 C. Slides were incubated with biotinylated donkey anti-rabbit secondary antibody (1:500, Jackson ImmunoResearch 711-065-152). Streptavidin-HRP conjugate (Vector Laboratories) was added onto the slides and incubated for 30 min. DAB Chromogen substrate (Dako) was added for dye development. Counterstaining and dehydration was performed on the ST5020 Multistainer (Leica) followed by coverslipping. For immunofluorescence tissue was excised and fixed for 48 h in 4% PFA in PBS at 4 C, after which it was transferred to 20% sucrose solution. After cryosectioning antigen retrieval where needed was accomplished by incubating the slides in 1% SDS for 5 min. Blocking was performed with 5% donkey serum. Following a wash primary antibodies were added and incubated overnight at 4 C. The following primary antibodies were used: rabbit FITC-anti-Lyz (1:400, Dako, F037201), rabbit anti-Muc2 (1:50, Santa Cruz, sc-15334), rabbit anti-ChgA (1:100, Abcam, ab15160), and rabbit anti-Dclk1 antibody (1:1000, Abcam, ab31704). Secondary detec- tion was with AlexaFluor 488 donkey anti-rabbit secondary antibody (1:500, Thermo Fisher, A-21206). Alkaline phosphatase activity was detected using Blue AP kit (Vector Laboratories). Sections were covered with Prolong Gold with DAPI (Life Technologies). Fluo- rescent imaging was carried out on a TCS SP5 confocal microscope (Leica). Cell Stem Cell 23, 436–443.e1–e7, September 6, 2018 e4 Single molecule FISH Harvested SI and colon tissues were flushed with cold 4% formaldehyde (FA) in PBS and incubated first in 4% FA/PBS for 3 hours, then in 30% sucrose in 4% FA/PBS overnight at 4 C with constant agitation. Fixed tissues were embedded in OCT. Quantification of co-expression was achieved by smFISH. Probe library design, hybridization procedures, and imaging settings were carried out ac- cording to published methods (Itzkovitz et al., 2011; Lyubimova et al., 2013). A Nikon-Ti-E inverted fluorescence microscope equip- ped with a Photometrics Pixis 1024 CCD camera was used to image a 10 mm cryo-section. A stack of 30 frames with 0.3 mm intervals was acquired to allow 3D cell imaging. FITC-conjugated antibody for E-cadherin was added to the hybridization mix and used to visualize cell borders. Detection of cells that were positive for Lgr5 transcripts, Atoh1 transcripts or both was performed manually with Fiji. Analysis of gut sections Stained longitudinal sections of the SI and colon were visualized and positive cells scored manually. BrdU and negative nuclei were + + scored in complete half-crypt sections. Lysozyme cells were counted per whole crypt section. Alcian Blue and PAS cells were counted in complete half-villus sections, between the crypt neck and the tip of the villus. Cells in which the stain was clearly asso- + + ciated with a corresponding nucleus were marked as positive. Chromogranin A and synaptophysin cells were scored per complete half-crypt-villus section. Positive and negative crypts were scored, and results expressed as a frequency of positive cells. Colon ulceration scoring H&E-stained sections of colons were scanned on Aperio slide scanner (Leica Biosystems), and analyzed using eSlide Manager (Leica Biosystems). Ulceration was defined as a region of a complete loss of crypt architecture and high cellularity. RNA isolation For gene expression analysis by qPCR, cells were lysed and RNA isolated using RNeasy Mini Plus kit (QIAGEN). For sequencing, total RNA was isolated from flow-sorted cells using RNeasy Micro Plus kit (QIAGEN). Gene expression analysis RNA was converted into cDNA (iScript cDNA synthesis kit, BioRad), and gene expression was analyzed using TaqMan gene expression probes (Thermo Fisher). The following probes were used: Atoh1 (Mm00476035_s1), Muc2 (Hs00894053_g1), Tff3 (Hs00902278_m1), Spdef (Hs01026050_m1), Dll4 (Hs00184092_m1), Rassf4 (Hs00604698_m1), Gapdh (Hs02758991_g1). All TaqMan assays are listed in Table S3. RNA sequencing Samples for RNA sequencing were collected 24 h post-tamoxifen induction (3 mg i.p. injection). The tissue was fractionated as described above and cells prepared for flow cytometry. The cells were stained and sorted in the same way as for other experiments, + + as noted above. EpCAM tdTom live cells were collected directly into the lysis buffer and RNA was extracted immediately following the sort (RNeasy Micro Plus Kit, QIAGEN). RNA quality was assessed on a 2100 Bioanalyser instrument (Agilent), according to the manufacturer’s instructions. The libraries were prepared using TruSeq Stranded mRNA Library Prep Kit (Illumina) and sequenced as 50 bp single-end reads on the Illumina HiSeq 4000 system. Western blotting Protein extracts for SDS-PAGE were prepared by lysing the cells with RIPA buffer containing protease and phosphatase inhibitor cocktail (Thermo Fisher). Mouse anti-ATOH1 antibody (1:100, Developmental Studies Hybridoma Bank) and a rabbit anti-b-A anti- body (1:5,000, ab8227, Abcam) were used. Fluorescent secondary antibodies were used (Li-Cor, goat anti-mouse 800LT (1:5,000), goat anti-rabbit 680LT (1:20,000)). For some experiments, protein extracts were incubated with l phosphatase (New En- gland Biolabs) prior to western blotting, according to the manufacturer’s instructions. In vitro kinase assay The assay was performed as previously described (Azzarelli et al., 2017), with minor modifications. HA-tagged WT and mutant ATOH1 were in vitro translated (TNT Quick Coupled Transcription/Translation Systems, Promega) in the presence of LiCl (800 mM) to reduce potential phosphorylation in reticulocyte lysate. Samples were incubated with human recombinant CDK/Cyclins (0.25 mM final concentration) in the presence of 10 mM ATP for 1 h at 30 C. Proteins were separated on Phos-tag gels (Alpha Laboratories, 7.5% acrylamide, 50 mM phos-tag PAGE, Wako) and immunoblotted with rat anti-HA-Peroxidase (1:5000, Roche). Cell proliferation and cell cycle analysis Cell proliferation was assessed by an automated live-cell imaging system (IncuCyte ZOOM, Essen Bioscience). For cell cycle anal- ysis, the cells were trypsinised, washed, fixed with ethanol, and stained with propidium iodide prior to flow cytometry. e5 Cell Stem Cell 23, 436–443.e1–e7, September 6, 2018 RNA sequencing analysis The reads were aligned to the mouse reference genome [GRCm38] using TopHat2 aligner (Kim et al., 2013). Differentially expressed gene lists were generated using DESeq2 package from Bioconductor (Love et al., 2014). Secretory signature gene list + - The list of differentially expressed genes (p < 0.01) was generated by comparing the transcripts from tdTom and tdTom cells of (WT)CreERT2 tdTom/+ + Atoh1 Rosa26 mice following tamoxifen. Upregulated genes in tdTom cells were selected to define a secretory signature in the small intestine and colon (Table S1). The top 500 upregulated, differentially expressed genes were used to perform the Gene Set Enrichment Analysis (GSEA). Gene Set Enrichment Analysis (GSEA) This analysis was performed using the GSEA software from the Broad Institute (http://software.broadinstitute.org/gsea/index.jsp) (Subramanian et al., 2005). The list comprised all differentially expressed and non-differentially expressed genes from the 9S/T-A v WT comparison in SI and colon, respectively. This gene list was probed with the previously generated secretory signatures (top 500 upregulated genes), and the published Atoh1 gene signatures for ileum, colon (Lo et al., 2016), and intestinal stem cells (Mun˜ oz et al., 2012). QUANTIFICATION AND STATISTICAL ANALYSIS Computational analysis The process by which crypt stem cells replace each occurs in a random though predictable manner. This behavior can be modeled via a stochastic birth-death process (Lopez-Garcia et al., 2010; Snippert et al., 2010). The model was derived to model experiments where a single stem cell is labeled in a handful of crypts. As the number of initially labeled crypts was not of interest and to bypass any variability coming from the initial induction, make the different time points comparable, the equations were rescaled to account for only the surviving clones. Here we know the parameters of the stem cell dynamics (Kozar et al., 2013; Vermeulen et al., 2013), and would like to know the starting number of labeled stem cells per crypt and the number of labeled crypts. For this analysis we use the equations described previously (Lopez-Garcia et al., 2010; Snippert et al., 2010), reproduced below. The probability of a crypt having clone of size n (for 0 < n < N) at time t is: N1 2 pm pmn 2 pm 4l sin p ðtÞ sin sin e t (Equation 1) N N N 2N m = 1 Here n is the number of labeled stem cells, N is the total number of stem cells, l is the rate of stem cell replacement. And for the probability of all stem cells labeled we have: N1 2 pm 2 pm m=+ 1 2 4l sin p ðtÞ ð1Þ cos 1  e t (Equation 2) N 2N 2N m = 1 These equations assume the initial conditions of one labeled stem cell at t = 0. The starting labeled stem cells were chosen randomly at the beginning of each simulation. The values we observe for the clonal frequencies are substantially lower than what the model would predict, suggesting that not all crypts have labeled stem cells. In order to find out the fraction of labeled crypts v we use a mixture model: Q ðtÞ = ð1  vÞd + vP ðtÞ (Equation 3) n 0;n n Where Q ðtÞ is the probability that a randomly selected crypt has a clone of size n labeled stem cells at time t. We use the values of N, l and t from Kozar et al. (2013) and Vermeulen et al. (2013) and estimate v. Model fitting For every mouse, at day 30 we count the number of clones (k ) and the number of crypts (C ). We use a hierarchical model to capture i i the mouse to mouse variability. The statistical model is a follows k  BinomialðC ; R ð30  tÞÞ (Equation 4) i i i R ð30  tÞ Student tðh; Qð30  tÞ; sÞ (Equation 5) Here R is truncated to [0, 1]. For the SI no distinction is made in clone size, so Q is the sum of all Q and for the colon we use only the i n full clones for fitting Q = Q N. The priors on the population parameters are: h  Gammað2:0:1Þ (Equation 6) Cell Stem Cell 23, 436–443.e1–e7, September 6, 2018 e6 s  Gammað0:01; 0:01Þ (Equation 7) The prior on the mixing coefficient is h  Betað1=2; 1=2Þ (Equation 8) The posterior was derived via MCMC using Rstan (Carpenter et al., 2017). For the proximal and distal SI we used is t = 5 as the clones were measured in ribbons coming out of the crypt, which take a few days to emerge from the crypt base. Whereas for the colon we used t = 1. The parameters used were t = 0:1, N = 5 for proximal SI, t = 0:2, N = 6 for distal and t = 0:3, N = 7 for colon. Statistical analysis Statistical tests were not used to predetermine sample size. Randomization was not performed to allocate samples/animals to exper- imental groups. Blinding was performed for quantifications in Figures 3F and 3G, as well as Figures S2B–S2F. Data analysis was per- formed using GraphPad Prism software or R package. DATA AND SOFTWARE AVAILABILITY The accession number for the RNA sequencing data reported in this paper is GEO: GSE115416. Mendeley Dataset of original data can be accessed at https://doi.org/10.17632/vgvdv5b949.1. e7 Cell Stem Cell 23, 436–443.e1–e7, September 6, 2018 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Cell Stem Cell Unpaywall

Phospho-regulation of ATOH1 Is Required for Plasticity of Secretory Progenitors and Tissue Regeneration

Download PDF
16 pages

Loading next page...
 
/lp/unpaywall/phospho-regulation-of-atoh1-is-required-for-plasticity-of-secretory-DaSDmpJ7jh

References (58)

Publisher
Unpaywall
ISSN
1875-9777
DOI
10.1016/j.stem.2018.07.002
Publisher site
See Article on Publisher Site

Abstract

Short Article Phospho-regulation of ATOH1 Is Required for Plasticity of Secretory Progenitors and Tissue Regeneration Graphical Abstract Authors Goran Tomic, Edward Morrissey, Sarah Kozar, ..., Shalev Itzkovitz, Anna Philpott, Douglas J. Winton Correspondence [email protected] (A.P.), [email protected] (D.J.W.) In Brief Tomic et al. report that multisite phosphorylation of ATOH1 regulates the contribution of secretory progenitors to stem cell self-renewal in the small intestine and colon. With damage, the enhanced role of Atoh1 progenitors in mediating tissue repair is ablated in mice expressing phosphomutant ATOH1 and overall tissue regeneration is impaired. Highlights d Atoh1 progenitors contribute to the stem cell pool in homeostasis and regeneration d Multisite phosphorylation of ATOH1 regulates the plasticity of secretory progenitors d Loss of phosphorylation of ATOH1 reduces clonogenic capacity of Atoh1 cells d Phosphomutant ATOH1 mice are more susceptible to chemical colitis Tomic et al., 2018, Cell Stem Cell 23, 436–443 September 6, 2018 ª 2018 The Authors. Published by Elsevier Inc. https://doi.org/10.1016/j.stem.2018.07.002 Cell Stem Cell Short Article Phospho-regulation of ATOH1 Is Required for Plasticity of Secretory Progenitors and Tissue Regeneration 1 2 1 3 1 4,5 1 Goran Tomic, Edward Morrissey, Sarah Kozar, Shani Ben-Moshe, Alice Hoyle, Roberta Azzarelli, Richard Kemp, 1 3 4,5, 1,6, Chandra Sekhar Reddy Chilamakuri, Shalev Itzkovitz, Anna Philpott, * and Douglas J. Winton * Cancer Research UK Cambridge Institute, University of Cambridge, Cambridge CB2 0RE, UK Weatherall Institute of Molecular Medicine, University of Oxford, John Radcliffe Hospital, Oxford OX3 9DS, UK Department of Molecular Cell Biology, Weizmann Institute of Science, Rehovot, Israel Department of Oncology, Hutchison/Medical Research Council (MRC) Research Centre, University of Cambridge, Cambridge CB2 0XZ, UK Wellcome-MRC Cambridge Stem Cell Institute, University of Cambridge, Cambridge CB2 1QR, UK Lead Contact *Correspondence: [email protected] (A.P.), [email protected] (D.J.W.) https://doi.org/10.1016/j.stem.2018.07.002 SUMMARY Lgr5 cells (Barriga et al., 2017; Buczacki et al., 2013), progeni- tors committed to different intestinal lineages (van Es et al., The intestinal epithelium is largely maintained 2012; Tetteh et al., 2016), and cells dependent on alternate path- ways for stem cell maintenance (Takeda et al., 2011; Tian by self-renewing stem cells but with apparently et al., 2011). committed progenitors also contributing, particularly It has been demonstrated previously that cells of the secretory following tissue damage. However, the mechanism lineage possess reserve stem cell function in the small intestine of, and requirement for, progenitor plasticity in medi- (SI) epithelium in homeostasis and following tissue damage (van ating pathological response remain unknown. Here Es et al., 2012; Ishibashi et al., 2018; Yan et al., 2017; Yu et al., we show that phosphorylation of the transcription 2018). Subsequent to Delta-like expression (from Dll1 or Dll4), factor Atoh1 is required for both the contribution of the basic helix-loop-helix (bHLH) transcription factor Atoh1 is secretory progenitors to the stem cell pool and for upregulated, an event required for the creation of all secretory a robust regenerative response. As confirmed by lineages within the epithelium (Yang et al., 2001). Atoh1 pro- + (WT)CreERT2 lineage tracing, Atoh1 cells (Atoh1 mice) genitors exhibit self-renewal and give rise to multilineage clones give rise to multilineage intestinal clones both in the with higher frequency in homeostasis (Ishibashi et al., 2018) compared with previously described secretory Dll1 progenitors steady state and after tissue damage. In a phospho- (9S/T-A)CreERT2 (van Es et al., 2012). This observation highlights a significant mutant Atoh1 line, preventing phos- contribution of Atoh1 cells to the stem cell pool in the SI and phorylation of ATOH1 protein acts to promote secre- colon. However, the mechanisms regulating intestinal plasticity tory differentiation and inhibit the contribution of and the nature of the relationship linking it to self-renewal remain progenitors to self-renewal. Following chemical coli- unknown. + (9S/T-A)CreERT2 tis, Atoh1 cells of Atoh1 mice have ATOH1 can be phosphorylated on multiple sites by cyclin- reduced clonogenicity that affects overall regenera- dependent kinases. Here we demonstrate that maintenance of tion. Progenitor plasticity maintains robust self- the plasticity of committed secretory precursors allowing return renewal in the intestinal epithelium, and the balance to the stem compartment is dependent on the multisite phos- between stem and progenitor fate is directly coordi- phorylation of ATOH1, prevention of which inhibits Atoh1-medi- nated by ATOH1 multisite phosphorylation. ated self-renewal and results in compromised regeneration following damage. We conclude that reversibility of the commit- ment to differentiate is dependent on post-translational control of ATOH1 and is required to maintain a robust stem cell INTRODUCTION population. Within the intestinal epithelium, cell generation occurs from phenotypically heterogenous stem cells residing at the base RESULTS of glandular crypts (Vermeulen and Snippert, 2014). There is broad consensus that this heterogeneity reflects the combined Atoh1 Cells Show Stem Cell Activity behavior of active and reserve stem cells. The former dominates Initially, to determine the extent to which Atoh1-expressing in homeostatic self-renewal and the latter following tissue cells support stem cell maintenance in homeostasis, we gener- (WT)CreERT2 T2 damage. In homeostasis, rapidly cycling stem cells express ated a mouse (Atoh1 ) with an inducible CreER the R-spondin receptor Lgr5. Reserve stem cell function is less downstream of the Atoh1 coding sequence (Figure S1A). Acute defined and has been ascribed variously to a subset of quiescent lineage tracing demonstrated that tdTomato (tdTom) reporter 436 Cell Stem Cell 23, 436–443, September 6, 2018 ª 2018 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). + Figure 1. Lineage Tracing of Atoh1 Cells in Homeostasis and after Injury (A–D) The tdTom reporter is detected in Muc2 goblet cells in the SI (A), colon (B), and Lyz Paneth cells (C) but not in ChgA enteroendocrine cells 24 hr post-tamoxifen (D). Muc2, Mucin 2; Lyz, Lysozyme; ChgA, Chromogranin A. (E) ChgA cells labeled with tdTom on day 4 after induction. (F) Dclk1 tuft cells are not labeled with tdTom at 24 hr. (G and H) Reporter-positive clone in the SI (G) and colon (H) 30 days following tamoxifen. (I–L) tdTom clones at 30 days are composed of alkaline phosphatase (Alpi ) enterocytes (I), Paneth cells (J), goblet cells (K), and enteroendocrine cells (L). (M, P, and S) Schematic of induction and injury protocol: irradiation (M), azoxymethane (AOM) (P), and dextran sodium sulfate (DSS) (S). (N) Representative pictures of SI whole-mounts containing labeled crypts (arrowheads) 30 days post-induction. (O) Quantification of tdTom crypts in the SI (n = 4 for 0 Gy, n = 6 for 6 Gy [day 1], n = 4 for 6 Gy [day 5]). (Q and T) Representative images of colonic crypts on day 30 post-tamoxifen and AOM (Q) or DSS treatment (T). Note the large tdTom regenerative multicrypt patches (MCPs) associated with 2% DSS treatment (T). (R) Quantification of reporter-positive crypts in the colon (n = 6 for untreated, n = 5 for AOM-treated). (U) Quantification of tdTom MCPs in untreated and DSS-treated colons (n = 3 for both groups). Welch’s t test was used in (O) (mean ± SEM, ****p < 0.0001) and Mann-Whitney test in (R) (mean ± SEM, **p = 0.0087). Scale bars, 50 mm (A–L) and 100 mm (N, Q, and T). See also Figure S1. cells were positive for the reporter whereas enteroendocrine cells (EECs) were not; the latter observation confirms that Atoh1 expression is not maintained in mature enteroendocrine cells (Bjerknes et al., 2012; Sommer and Mostoslavsky, 2014). However, by 4 days post-tamox- ifen, enteroendocrine cells were also labeled (Figure 1E), indicating an origin from a secretory precursor that ex- pressed Atoh1. Tuft cells were also not labeled with tdTom (Figure 1F). Individual Paneth cells remained labeled 4 weeks post-induction, reflecting their longevity (Figure S1H). Similar results were found in the colon, and long-lived secretory cells were also identified (Figure S1I). By 30 days post-induction, cohesive patches of reporter-positive cells that occupied all or a significant portion of entire crypts expression 24 hr following a single pulse of tamoxifen was were present (Figures 1G and 1H) and continued to be observed restricted to secretory cells within the SI and colonic epithelium after several months (Figure S1J). Immunostaining established (Figures 1A–1D; Figures S1B–S1G). Mature Paneth and goblet the presence of goblet, Paneth, enteroendocrine, and absorptive Cell Stem Cell 23, 436–443, September 6, 2018 437 Figure 2. Identification of a Hyperactive Phosphomutant ATOH1 (A) Diagram depicting the location of proline- directed kinase motifs (serine-proline [SP] or threonine-proline [TP]) in Atoh1 protein and mutations of these sites into alanine in ATOH1 phosphomutants. (B) In vitro-translated Atoh1 protein band-shift following incubation with different cyclin-depen- dent kinases (CDKs). Ngn3 was used as a positive control. (C) WT ATOH1 bands (arrows) collapse following l phosphatase treatment, demonstrating phos- phorylation. (D) DLD-1 cell proliferation following doxycycline (Dox)-induced expression of WT or phosphomu- tant Atoh1 (n = 3 biological replicates, 2 technical replicates, mean ± SEM). (E) Cell cycle profile of uninduced and Dox-treated cells showing increased G1 and decreased S/M populations upon induction of 9S/T-A Atoh1. (F) Gene expression of Atoh1 and its target and secretory differentiation genes 72 hr after Dox in- duction of DLD-1 cells (n = 3 biological replicates, 2 technical replicates; Gapdh-normalized, mean ± SEM). Two-way ANOVA was used for statistical analysis; *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. approach (Figures S1P and S1Q), there was a 30-fold increase in the number of clones observed (Figures S1R and S1S). cells within reporter-positive epithelium, confirming their multili- Adapting the assay to perform a similar analysis for the colonic neage composition (Figures 1I–1L). These patterns are identical epithelium and to circumvent that tissue’s known radio-resistance to those arising from individual marked intestinal stem cells (Ver- (Cai et al., 1997), mice were treated with the colon-specific carcin- meulen et al., 2013) and demonstrate a clonal origin from Atoh1 ogen azoxymethane (AOM) 1 day after tamoxifen treatment. (WT)CreERT2 TdTom + precursors. Atoh1 ; Rosa26 mice were then Again, an increase in the frequency of tdTom crypts was Gfp crossed onto Lgr5 reporter mice to investigate co-expression observed (Figures 1P–1R). Following dextran sodium sulfate of Atoh1 and the intestinal stem cell marker Lgr5. The expression (DSS)-induced colitis, multicrypt tdTom patches (MCPs) were of Atoh1 and the tdTom reporter was identified in 1%–2% of detected at the margins of regions of damage (Figures 1S–1U; Fig- + + + Lgr5 (GFP ) cells (Figures S1K–S1O), representing a likely inter- ures S1T and S1U). Together, these results suggest that Atoh1 mediate state in the commitment process and candidate clono- cells directly contribute to regeneration following damage. genic population. Together, these results confirm that Atoh1 is appropriately expressed in mature Paneth and goblet cells but Creating a Pro-secretory Phosphomutant ATOH1 not enteroendocrine cells and that a proportion of Atoh1 Previous studies have indicated that multisite phosphorylation of progenitors are acting as long-term multipotential stem cells bHLH proteins restrains cell cycle exit and limits differentiation, whereas, conversely, un(der)phosphorylation promotes these (Bjerknes et al., 2012; Sommer and Mostoslavsky, 2014; Ishiba- shi et al., 2018). processes in the developing nervous system and pancreas (Ali et al., 2011, 2014; Azzarelli et al., 2017). However, a role for multi- Atoh1 Cells Contribute Directly to Epithelial site phospho-regulation of bHLH proteins in adult homeostasis Regeneration or tissue repair has not been reported. Hence, we hypothesized The extent of reversibility of Atoh1 cell commitment was studied a potential role for ATOH1 phosphorylation in controlling the in the context of irradiation-induced tissue damage. Irradiation transition between stem and progenitor compartments both in given 1 day after tamoxifen generated an increased number of homeostasis and under conditions of heightened proliferation tdTom crypts at 30 days in the SI compared with unirradiated following tissue damage. Cyclin-dependent kinases phosphory- controls (16-fold increase, 2.37% versus 0.15%). This effect was late on serine-proline (SP) or threonine-proline (TP) residues. abrogated when irradiation was given 5 days after tamoxifen (Fig- ATOH1 has 9 S/T-P sites available for phosphorylation (Figures ures 1M–1O), suggesting that regenerative potential is a property 2A–2C). ATOH1 can be phosphorylated on many sites; we of progenitors arising de novo from the stem cell compartment and observed at least 5 distinct phospho-forms of ATOH1 after phos- not of more mature secretory cells. Similarly, after targeted dele- phorylation by different Cyclin and Cdk combinations (Figure 2B). tion of the bulk of Lgr5 stem cells using a diphtheria toxin We expressed forms of ATOH1 where S/T-P sites were mutated 438 Cell Stem Cell 23, 436–443, September 6, 2018 to alanine-proline (AP) in colorectal cancer cells to determine the Comparing proliferation between the two lines demonstrated effect of ATOH1 phosphorylation on cell proliferation and on a slight overall decrease in the total proliferative index of the (9S/T-A)CreERT2 expression of markers of differentiation. The phosphorylation crypts of Atoh1 mice in both the SI and colon, of two SP sites has previously been shown to destabilize the but that did not reach significance in the latter. More detailed ATOH1 protein in the context of neuronal precursors (Forget spatial analysis within the crypt epithelium demonstrated that et al., 2014). Although mutation of these two phospho-sites this effect was largely accounted for by a decrease in the pro- had a modest effect on ATOH1 activity, mutation of all 9 S/T-P portion of cells in S phase in the epithelium of 9S/T-A mutants sites was more effective at promoting enhanced cell cycle exit in cell positions above the very base of the crypt and a reduc- (Figures 2D and 2E). Additionally, the expression of secretory tion in the frequency of proliferative goblet cells (Figures 3F–3H; genes (Figure 2F) was enhanced after mutation of all 9 potential Figure S3A). This supports the interpretation that the phosphor- phosphorylation sites compared with both wild-type ATOH1, ylation of ATOH1 in cells immediately arising from the stem cell 2S-A ATOH1, and 7S/T-A ATOH1. These observations are population limits Atoh1-dependent cell cycle exit to allow main- consistent with multisite phospho-regulation of ATOH1 playing tenance of proliferation in progenitors. Reciprocally, preventing a significant role in controlling the balance between proliferation this phosphorylation limits the ability to return to a proliferative and differentiation, as described for other bHLH family members stem and progenitor compartment. We next tested this hypoth- (Ali et al., 2011, 2014; Azzarelli et al., 2017). esis using a lineage tracing approach. Lineage tracing and fluorescence-activated cell sorting 9S/T-A Phosphomutant ATOH1 Promotes Secretory (FACS) analysis established that the acute pattern of reporter expression and absolute number of tdTom cells were the Maturation In Vivo (9S/T-A)CreERT2 To investigate how preventing phosphorylation of ATOH1 same in Atoh1 and controls (Figures S3B–S3I). affects progenitor-mediated self-renewal in homeostasis and However, lineage tracing at 30 days identified fewer epithelial (WT)CreERT2 repair, we substituted 9S/T-A ATOH1 for the wild-type form in clones in both the SI and colon than in Atoh1 mice its endogenous locus, generating a knockin mouse identical (Figures 3I–3M). The 9S/T-A Atoh1 cells were also impaired in (WT)CreERT2 + in design to Atoh1 but with the hyperactive phospho- their ability to form tdTom clones after radiation (Figure S3J). (9S/T-A)CreERT2 mutant Atoh1 allele (Figure S2A). Homozygous Together, the observations demonstrate that preventing phos- (9S/T-A)CreERT2 (WT)CreERT2 + Atoh1 and control Atoh1 mice were phorylation of ATOH1 impairs the return of Atoh1 cells to the generated. Phenotype analysis identified no gross differences stem cell compartment and confirm a role for ATOH1 phosphor- between the two lines. Mice developed normally, and the overall ylation in maintenance of progenitor plasticity. morphological appearance of the epithelium remained un- Previously, we and others have described that only a subset of changed. More detailed analysis found no difference in the num- competing stem cells drive increases in clone sizes that lead to ber or distribution of the different secretory lineages or in the surviving clones populating entire crypts (Kozar et al., 2013; frequency of apoptotic cells (Figures S2B–S2F). Ritsma et al., 2014). To determine the net contribution of To investigate whether the 9S/T-A mutations affect secretory Atoh1 cells to this population, mathematical modeling was maturation after lineage specification, transcriptional profiling used to infer the proportion of the clonogenic fraction that is (WT)CreERT2 (9S/T-A)CreERT2 of secretory cells in the two lines was performed. First the initially marked in Atoh1 and Atoh1 + (WT)CreERT2 + expression profile of Atoh1 cells from Atoh1 mice mice. In both the SI and colon, the contribution of Atoh1 pro- was determined by comparing tdTom (secretory) and tdTom genitors to the stem cell pool is reduced in 9S/T-A animals (absorptive) cells to define the baseline pro-secretory signature (Figures 3N–3Q). Between 1% and 2% of SI crypts in (WT)CreERT2 for both the colon and SI (Table S1). Next, the transcription pro- Atoh1 mice contain a single clonogenic stem cell + + files of tdTom cells from wild-type and mutant mice (Table S2) derived from an Atoh1 progenitor, and this is reduced 5-fold + (9S/T-A)CreERT2 were compared against this Atoh1 baseline and a published in Atoh1 mice (Figures 3N and 3O). In the colon, secretory signature (Lo et al., 2016). These gene set enrichment values are higher, with the observed 4% wholly populated crypts analyses (GSEAs) demonstrated a major pro-secretory shift in (WPCs) and 5% partly populated crypts (PPCs) identified in (9S/T-A)CreERT2 (WT)CreERT2 Atoh1 mice 30 days post-induction requiring that Atoh1 mice in both tissues and a strongly reduced intestinal stem cell signature compared with those from controls initially 44% of crypts (1 in 15 active stem cells) contained an (Figures 3A–3E). Atoh1 -derived stem cell. This is reduced to 11% in 9S/T-A The pro-secretory nature of 9S/T-A-expressing cells arose mutant mice (Figures 3P and 3Q). Notably, these rates reflect from an overall elevation in pro-secretory transcripts for goblet the contribution of a single cohort of transient progenitors arising and Paneth cell lineages (SI only). Atoh1 cells isolated from from the stem cell pool that are produced over 1 or 2 days. (9S/T-A)CreERT2 the SI of Atoh1 mice had a reduction in a subset of transcripts associated with enteroendocrine cells, indi- Compromised Epithelial Regeneration in 9S/T-A cating that ATOH1 phosphorylation influences their maturation Atoh1 Mice (Figure S2G). Although phosphorylation of ATOH1 clearly regulates reversion of secretory progenitors to the stem cell compartment, the Epithelial Proliferation and Clonogenicity Are Inhibited absence of any other apparent phenotype in 9S/T-A Atoh1 9S/T-A in Atoh1 Mice mice suggests a limited requirement for such plasticity in ho- We next investigated whether the enhanced pro-secretory meostasis. We next investigated the role of ATOH1 phosphory- signature induced by prevention of multisite phosphorylation lation in mounting a robust regenerative response following of ATOH1 is accompanied by changes in proliferation. tissue damage. In the DSS-induced chemical colitis model, Cell Stem Cell 23, 436–443, September 6, 2018 439 Figure 3. 9S/T-A ATOH1 Promotes Secre- tory Maturation and Reduces Proliferation and the Number of Clonogenic Atoh1 Cells (A and B) Gene set enrichment analysis (GSEA) of the Atoh1 SI secretory signature (A) and colon (B) shows enrichment of secretory genes in 9S/T-A Atoh1 tdTom cells. (C and D) GSEA utilizing a published secretory transcriptome reveals an increase in the secretory gene signature in phosphomutant-expressing tdTom cells in the SI (C) and colon (D). (E) GSEA using a published intestinal stem cell (ISC) gene signature (Mun˜ oz et al., 2012) shows a de-enrichment of ISC genes in the mutant SI progenitors. (F and G) Bromodeoxyuridine (BrdU) labeling index for a range of cell positions in SI crypts (F) shows a reduction in proliferation above the crypt base (n = 100 crypts, 4 mice per genotype; mean ± SEM; **p = 0.0061 and 0.0015). Shown in (G) is the fre- quency of crypt-villus units containing at least one BrdU goblet cell after a 24-hr BrdU pulse (n = 5 for both groups, *p = 0.0317). (H) Representative image of a crypt-villus unit with a BrdU Alcian blue (AB) and periodic acid-Schiff (PAS) cell. (I) Quantification of tdTom clonal events in the SI (n = 6 [WT], n = 7 [9S/T-A], mean ± SEM, **p = 0.0012). (J) Partly populated tdTom crypts (PPCs) in the colon (n = 6 [WT], n = 7 [9S/T-A], mean ± SEM, **p = 0.0023). (K) Wholly populated tdTom crypts (WPCs) in WT and 9S/T-A colons (n = 6 [WT], n = 7 [9S/T-A], mean ± SEM, **p = 0.0012; the same WT data are shown in Figure 1R because the experiment was done in parallel). All samples were collected 30 days after tamoxifen. Mann-Whitney test was used for all comparisons. (L and M) Representative images of WT (L) and 9S/T-A (M) colons scored in (J) and (K). (N–Q) Inference of the proportion of the clonogenic fraction of labeled Atoh1 cells in the proximal SI (N), distal SI (O), colon PPCs (P), and colon WPCs (Q). The numbers next to the dotted lines indicate the inferred proportion of crypts that had one labeled stem cell. Scale bars, 75 mm. 9S/T-A mutant mice showed a greater sensitivity after treatment, S4D). Together, these results demonstrate that mice lacking with increased weight loss and slowed recovery (Figures 4A, the ability to phospho-regulate ATOH1 have compromised S4A, and S4B). Analysis of this phenotype at the start of the regenerative capacity following damage and that the contribu- regenerative phase (9 days after the start of DSS treatment) tion of Atoh1 progenitors is required for robust tissue repair. showed areas of ulceration that were more extensive in mice car- rying the 9S/T-A mutant (Figures 4B and 4C). At both 5 and DISCUSSION 9 days, the proportion of secretory cells was identical for the two lines, and cell death was restricted to a few cells on the It is now accepted that cells with the capacity for self-renewal luminal surface, suggesting that the greater sensitivity does not arise from a larger population whose members all have the arise from enhanced damage or deletion of secretory cells in same self-renewal potential subject to occupying available 9S/T-A mutant mice (Figures 4D–4F). However, lineage tracing niches (Farin et al., 2016; Ritsma et al., 2014). Here we show 30 days following DSS treatment identified a reduced number that Atoh1 cells make a more substantial contribution to stem and size of tdTom regenerative patches in 9S/T-A colons cell maintenance from cells committing to secretory differentia- compared with the wild-type (WT) (Figures 4G, 4H, S4C, and tion than has been recognized so far (van Es et al., 2012; 440 Cell Stem Cell 23, 436–443, September 6, 2018 (9S/T-A)CreERT2 Figure 4. Atoh1 Mice Are Sensitive to Chemical Colitis (A) Change in mouse body weight during and after DSS treatment (n = 5 [WT], n = 6 [9S/T-A]; two 9S/T-A mice were euthanized on day 9 for health reasons, and one WT animal was taken for comparison [arrowhead]). (B) Representative pictures and schematics of the colon on day 9, showing extensive loss of crypts in 9S/T-A but not in the WT. Scale bars, 1 mm. (C) Total length of colon ulceration on day 9 (n = 6 [WT], n = 4 [9S/T-A], mean ± SEM, **p = 0.0095). (D and E) Representative images of the distal colon on day 5 of DSS treatment, showing apoptosis (D) and AB and PAS staining (E) in WT and 9S/T-A animals. (F) FACS analysis of the number of tdTom cells during DSS-induced colitis. (G and H) Analysis of the number (G) and total area (H) of tdTom MCPs following 1.5% DSS (n = 4 [WT], n = 7 [9S/T-A], mean ± SEM, **p = 0.0061, *p = 0.0424). See also Figure S4. Ishibashi et al., 2018). Self-renewal is therefore not solely a the balance between stem and progenitor fate behavior in the feature driven from a fixed pool of stem cells but, rather, involves intestine can be controlled by Atoh1 multisite phosphorylation dynamic interchange between progenitors and stem cells in the under normal homeostatic conditions. steady state. Control of proliferation and differentiation by modulation of Transcription factors of the bHLH family have been extensively bHLH protein phosphorylation is emerging as an important studied as master regulators of cell fate commitment and differ- mechanism in development of the nervous system and the entiation in a wide variety of tissues, including the nervous pancreas (Cleaver, 2017; Guillemot and Hassan, 2017). We system and intestine (Ali et al., 2011, 2014; Yang et al., 2001). now demonstrate that multisite phosphorylation is also required However, in recent years, additional roles for these proteins to restrain irreversible commitment of secretory precursors in the are emerging in direct co-ordination of cell cycle and differentia- adult homeostatic gut and so to maintain their ability to repopu- tion events, particularly during embryonic development (Castro late the stem cell compartment. Consistent with this, a phospho- and Guillemot, 2011). Intestinal homeostasis in many ways rep- mutant form of ATOH1 enhances the expression of gene sets resents an ongoing development-like hierarchical process where associated with a more mature secretory phenotype in colo- crypts are maintained by stem cells feeding a proliferating pro- rectal carcinoma cells. Interestingly, in the homeostatic gut, genitor compartment that gives rise to a variety of mature cell despite Atoh1 cells normally supplying up to 1 in 15 cells in types. What is now also emerging is a picture of significant plas- the stem cell compartment, the phosphomutant Atoh1-express- ticity where cells expressing Atoh1, previously thought to repre- ing intestine is essentially phenotypically normal, indicating that sent a population that has undergone secretory commitment, plasticity from the secretory to the stem compartment is not can nevertheless revert to ‘‘stemness’’ and repopulate the entire essential in normal homeostasis. However, intestinal regenera- crypt with surprisingly high frequency. The mechanisms control- tion after damage is substantially compromised by an inability ling this plasticity have been unclear. Here we determine that to phosphorylate ATOH1. Cell Stem Cell 23, 436–443, September 6, 2018 441 B Western blotting Taken together, our results indicate that multisite phosphory- B In vitro kinase assay lation of ATOH1 is used to dynamically regulate the return of B Cell proliferation and cell cycle analysis secretory precursors to the stem cell compartment, which facil- B RNA sequencing analysis itates the capacity of the epithelium as a whole to respond B Secretory signature gene list rapidly to changes in the cellular environment. Damaging the in- B Gene Set Enrichment Analysis (GSEA) testine using irradiation or DSS (van Es et al., 2012; Ishibashi d QUANTIFICATION AND STATISTICAL ANALYSIS et al., 2018) leads to acute cell damage and death, followed by B Computational analysis proliferative regeneration that produces new cells for tissue B Model fitting repair. Activation of cyclin-dependent kinases (CDKs) and B Statistical analysis mitogen-activated protein kinases (MAPKs) in rapidly prolifer- d DATA AND SOFTWARE AVAILABILITY ating cells undergoing regeneration would result in enhanced phosphorylation of ATOH1, restraining further progression down the secretory lineage and supporting re-entry of Atoh1-ex- SUPPLEMENTAL INFORMATION pressing cells into a stem-like state. The post-translational regu- Supplemental Information includes four figures and three tables and can be lation of ATOH1 by proline-directed kinases to modulate the found with this article online at https://doi.org/10.1016/j.stem.2018.07.002. balance between proliferation and differentiation in response to changing tissue demands in the adult intestinal epithelium ACKNOWLEDGMENTS echoes the regulation and effect of other bHLH proteins as development progresses (Hardwick et al., 2015). This work was supported by Cancer Research UK (to G.T., E.M., A.H., R.K., The secretory fate choice mediated by ATOH1, a master regu- and D.J.W.) and Wellcome Trust grant 103805 (to G.T. and D.J.W.). A.P. and lator, is not irreversible differentiation; rather, it is entry into a R.A. were funded by MRC research grant MR/K018329/1 and grant MR/ plastic state through which progression is regulated by post- L021129/1, the Rosetrees Trust, and the Stoneygate Trust and received core support from the Wellcome-MRC Cambridge Stem Cell Institute. We translational modifications. Functionally, the implications are thank S. Taylor for DLD-1 Flp-In T-Rex cells and D. Perera for plasmids. We likely to be that post-translational modifications facilitate rapid thank the Histopathology core and Biological Resources Unit at the CRUK cellular responses by allowing reversal of commitment or varying Cambridge Institute for technical support and the members of the Winton its extent or rate. Progenitor plasticity is not merely an incidental and Philpott labs for their help. acquired behavior following damage but plays an integral part in tissue restoration and requires post-translational regulation AUTHOR CONTRIBUTIONS of ATOH1. Conceptualization, D.J.W., A.P., G.T., and R.K.; Methodology, G.T., D.J.W., and A.P.; Formal Analysis, G.T., E.M., and C.S.R.C.; Investigation, G.T., STAR+METHODS S.K., S.B.-M., A.H., and R.A.; Writing – Original Draft, D.J.W., A.P., and G.T.; Writing – Review & Editing, G.T., D.J.W., and A.P.; Visualization, G.T.; Super- vision, D.J.W., A.P., and S.I.; Project Administration, D.J.W. and A.P.; Funding Detailed methods are provided in the online version of this paper Acquisition, D.J.W. and A.P. and include the following: DECLARATION OF INTERESTS d KEY RESOURCES TABLE d CONTACT FOR REAGENT AND RESOURCE SHARING The authors declare no competing interests. d EXPERIMENTAL MODEL AND SUBJECT DETAILS B Mice Received: February 2, 2018 B Cell Lines Revised: May 25, 2018 Accepted: July 6, 2018 d METHOD DETAILS Published: August 9, 2018 B Cloning of mouse knock-in constructs B ES cell targeting REFERENCES B Mouse genotyping B Creation of doxycycline inducible DLD-1 cells Ali, F., Hindley, C., McDowell, G., Deibler, R., Jones, A., Kirschner, M., B Treatment of animals Guillemot, F., and Philpott, A. (2011). Cell cycle-regulated multi-site phosphor- B Crypt fractionation and single cell preparation ylation of Neurogenin 2 coordinates cell cycling with differentiation during neu- B Flow cytometry rogenesis. Development 138, 4267–4277. B Whole-mount preparation Ali, F.R., Cheng, K., Kirwan, P., Metcalfe, S., Livesey, F.J., Barker, R.A., and Philpott, A. (2014). The phosphorylation status of Ascl1 is a key determinant B Whole-mount scanning and quantification of neuronal differentiation and maturation in vivo and in vitro. Development B Antibody staining 141, 2216–2224. B Quantification of crypts in whole-mounts Azzarelli, R., Hurley, C., Sznurkowska, M.K., Rulands, S., Hardwick, L., B Immunostainings Gamper, I., Ali, F., McCracken, L., Hindley, C., McDuff, F., et al. (2017). B Single molecule FISH Multi-site Neurogenin3 Phosphorylation Controls Pancreatic Endocrine B Analysis of gut sections Differentiation. Dev. Cell 41, 274–286.e5. B Colon ulceration scoring Barriga, F.M., Montagni, E., Mana, M., Mendez-Lago, M., Hernando- B RNA isolation Momblona, X., Sevillano, M., Guillaumet-Adkins, A., Rodriguez-Esteban, G., B Gene expression analysis Buczacki, S.J.A., Gut, M., et al. (2017). Mex3a Marks a Slowly Dividing B RNA sequencing Subpopulation of Lgr5+ Intestinal Stem Cells. Cell Stem Cell 20, 801–816.e7. 442 Cell Stem Cell 23, 436–443, September 6, 2018 Bjerknes, M., Khandanpour, C., Mo¨ ro¨ y, T., Fujiyama, T., Hoshino, M., Klisch, Love, M.I., Huber, W., and Anders, S. (2014). Moderated estimation of fold T.J., Ding, Q., Gan, L., Wang, J., Martı´n, M.G., and Cheng, H. (2012). Origin change and dispersion for RNA-seq data with DESeq2. Genome Biol. 15, 550. of the brush cell lineage in the mouse intestinal epithelium. Dev. Biol. 362, Lyubimova, A., Itzkovitz, S., Junker, J.P., Fan, Z.P., Wu, X., and van 194–218. Oudenaarden, A. (2013). Single-molecule mRNA detection and counting in Buczacki, S.J.A., Zecchini, H.I., Nicholson, A.M., Russell, R., Vermeulen, L., mammalian tissue. Nat. Protoc. 8, 1743–1758. Kemp, R., and Winton, D.J. (2013). Intestinal label-retaining cells are secretory Mun˜ oz, J., Stange, D.E., Schepers, A.G., van de Wetering, M., Koo, B.-K., precursors expressing Lgr5. Nature 495, 65–69. Itzkovitz, S., Volckmann, R., Kung, K.S., Koster, J., Radulescu, S., et al. Cai, W.B., Roberts, S.A., Bowley, E., Hendry, J.H., and Potten, C.S. (1997). (2012). The Lgr5 intestinal stem cell signature: robust expression of proposed Differential survival of murine small and large intestinal crypts following ionizing quiescent ’+4 cell markers. EMBO J. 31, 3079–3091. radiation. Int. J. Radiat. Biol. 71, 145–155. Ritsma, L., Ellenbroek, S.I.J., Zomer, A., Snippert, H.J., de Sauvage, F.J., Carpenter, B., Gelman, A., Hoffman, M.D., Lee, D., Goodrich, B., Betancourt, Simons, B.D., Clevers, H., and van Rheenen, J. (2014). Intestinal crypt homeo- M., Brubaker, M., Guo, J., Li, P., and Riddell, A. (2017). Stan : A Probabilistic stasis revealed at single-stem-cell level by in vivo live imaging. Nature 507, Programming Language. J. Stat. Softw. 76, 1–32. 362–365. Castro, D.S., and Guillemot, F. (2011). Old and new functions of proneural fac- Snippert, H.J., van der Flier, L.G., Sato, T., van Es, J.H., van den Born, M., tors revealed by the genome-wide characterization of their transcriptional tar- Kroon-Veenboer, C., Barker, N., Klein, A.M., van Rheenen, J., Simons, B.D., gets. Cell Cycle 10, 4026–4031. and Clevers, H. (2010). Intestinal crypt homeostasis results from neutral competition between symmetrically dividing Lgr5 stem cells. Cell 143, Cleaver, O. (2017). b Cell Renewal versus Differentiation: Slow and Steady 134–144. Wins the Race. Dev. Cell 41, 223–225. Sommer, C.A., and Mostoslavsky, G. (2014). RNA-Seq analysis of enteroendo- Farin, H.F., Jordens, I., Mosa, M.H., Basak, O., Korving, J., Tauriello, D.V.F., de crine cells reveals a role for FABP5 in the control of GIP secretion. Mol. Punder, K., Angers, S., Peters, P.J., Maurice, M.M., and Clevers, H. (2016). Endocrinol. 28, 1855–1865. Visualization of a short-range Wnt gradient in the intestinal stem-cell niche. Nature 530, 340–343. Subramanian, A., Tamayo, P., Mootha, V.K., Mukherjee, S., Ebert, B.L., Gillette, M.A., Paulovich, A., Pomeroy, S.L., Golub, T.R., Lander, E.S., and Forget, A., Bihannic, L., Cigna, S.M., Lefevre, C., Remke, M., Barnat, M., Mesirov, J.P. (2005). Gene set enrichment analysis: a knowledge-based Dodier, S., Shirvani, H., Mercier, A., Mensah, A., et al. (2014). Shh signaling approach for interpreting genome-wide expression profiles. Proc. Natl. protects Atoh1 from degradation mediated by the E3 ubiquitin ligase Huwe1 Acad. Sci. USA 102, 15545–15550. in neural precursors. Dev. Cell 29, 649–661. Takeda, N., Jain, R., LeBoeuf, M.R., Wang, Q., Lu, M.M., and Epstein, J.A. Guillemot, F., and Hassan, B.A. (2017). Beyond proneural: emerging functions (2011). Interconversion between intestinal stem cell populations in distinct and regulations of proneural proteins. Curr. Opin. Neurobiol. 42, 93–101. niches. Science 334, 1420–1424. Hardwick, L.J.A., Ali, F.R., Azzarelli, R., and Philpott, A. (2015). Cell cycle regu- Tetteh, P.W., Basak, O., Farin, H.F., Wiebrands, K., Kretzschmar, K., Begthel, lation of proliferation versus differentiation in the central nervous system. Cell H., van den Born, M., Korving, J., de Sauvage, F., van Es, J.H., et al. (2016). Tissue Res. 359, 187–200. Replacement of Lost Lgr5-Positive Stem Cells through Plasticity of Their Ishibashi, F., Shimizu, H., Nakata, T., Fujii, S., Suzuki, K., Kawamoto, A., Anzai, + Enterocyte-Lineage Daughters. Cell Stem Cell 18, 203–213. S., Kuno, R., Nagata, S., Ito, G., et al. (2018). Contribution of ATOH1 Cells to the Homeostasis, Repair, and Tumorigenesis of the Colonic Epithelium. Stem Tian, H., Biehs, B., Warming, S., Leong, K.G., Rangell, L., Klein, O.D., and de Sauvage, F.J. (2011). A reserve stem cell population in small intestine renders Cell Reports 10, 27–42. Lgr5-positive cells dispensable. Nature 478, 255–259. Itzkovitz, S., Lyubimova, A., Blat, I.C., Maynard, M., van Es, J., Lees, J., Jacks, van Es, J.H., Sato, T., van de Wetering, M., Lyubimova, A., Yee Nee, A.N., T., Clevers, H., and van Oudenaarden, A. (2011). Single-molecule transcript counting of stem-cell markers in the mouse intestine. Nat. Cell Biol. 14, Gregorieff, A., Sasaki, N., Zeinstra, L., van den Born, M., Korving, J., et al. (2012). Dll1+ secretory progenitor cells revert to stem cells upon crypt dam- 106–114. age. Nat. Cell Biol. 14, 1099–1104. Kim, D., Pertea, G., Trapnell, C., Pimentel, H., Kelley, R., and Salzberg, S.L. Vermeulen, L., and Snippert, H.J. (2014). Stem cell dynamics in homeostasis (2013). TopHat2: accurate alignment of transcriptomes in the presence of in- sertions, deletions and gene fusions. Genome Biol. 14, R36. and cancer of the intestine. Nat. Rev. Cancer 14, 468–480. Vermeulen, L., Morrissey, E., van der Heijden, M., Nicholson, A.M., Sottoriva, Kozar, S., Morrissey, E., Nicholson, A.M., van der Heijden, M., Zecchini, H.I., Kemp, R., Tavare´ , S., Vermeulen, L., and Winton, D.J. (2013). Continuous A., Buczacki, S., Kemp, R., Tavare, S., and Winton, D.J. (2013). Defining stem clonal labeling reveals small numbers of functional stem cells in intestinal cell dynamics in models of intestinal tumor initiation. Science 342, 995–998. crypts and adenomas. Cell Stem Cell 13, 626–633. Yan, K.S., Gevaert, O., Zheng, G.X.Y., Anchang, B., Probert, C.S., Larkin, K.A., Lallemand, Y., Luria, V., Haffner-Krausz, R., and Lonai, P. (1998). Maternally Davies, P.S., Cheng, Z.F., Kaddis, J.S., Han, A., et al. (2017). Intestinal expressed PGK-Cre transgene as a tool for early and uniform activation of Enteroendocrine Lineage Cells Possess Homeostatic and Injury-Inducible the Cre site-specific recombinase. Transgenic Res. 7, 105–112. Stem Cell Activity. Cell Stem Cell 21, 78–90.e6. Lo, Y.-H., Chung, E., Li, Z., Wan, Y.-W., Mahe, M.M., Chen, M.-S., Noah, T.K., Yang, Q., Bermingham, N.A., Finegold, M.J., and Zoghbi, H.Y. (2001). Bell, K.N., Yalamanchili, H.K., Klisch, T.J., et al. (2016). Transcriptional Requirement of Math1 for secretory cell lineage commitment in the mouse in- Regulation by ATOH1 and its Target SPDEF in the Intestine. Cell. Mol. testine. Science 294, 2155–2158. Gastroenterol. Hepatol. 3, 51–71. Yu, S., Tong, K., Zhao, Y., Balasubramanian, I., Yap, G.S., Ferraris, R.P., Lopez-Garcia, C., Klein, A.M., Simons, B.D., and Winton, D.J. (2010). Intestinal Bonder, E.M., Verzi, M.P., and Gao, N. (2018). Paneth Cell Multipotency stem cell replacement follows a pattern of neutral drift. Science 330, 822–825. Induced by Notch Activation following Injury. Cell Stem Cell 23, 46–59.e5. Cell Stem Cell 23, 436–443, September 6, 2018 443 STAR+METHODS KEY RESOURCES TABLE REAGENT or RESOURCE SOURCE IDENTIFIER Antibodies Mouse Monoclonal anti-Atoh1 Developmental Studies Hybridoma Bank Cat# Math1 (Atoh1): RRID:AB_10805299 Rabbit Polyclonal anti-b-actin Abcam Cat# ab8227: RRID:AB_2305186 IRDye 800CW Goat anti-Mouse IgG (H + L) LI-COR Biosciences Cat# P/N 925-32210: RRID:AB_2687825 IRDye 680LT Goat anti-Rabbit IgG (H + L) LI-COR Biosciences Cat# P/N 925-68021: RRID:AB_2713919 Rat Anti-Mouse CD326 (Ep-CAM) BioLegend Cat# 118210: RRID:AB_1134099 Monoclonal Antibody, Alexa Fluor 488 Conjugated, Clone G8.8 Sheep Polyclonal BrdU Antibody Abcam Cat# ab1893: RRID:AB_302659 Rabbit Polyclonal Anti Human Lysozyme Dako Cat# A0099: RRID:AB_2341230 Biotin-SP-AffiniPure Donkey Anti-Sheep Jackson ImmunoResearch Labs Cat# 713-066-147: RRID:AB_2340717 IgG (H+L) Biotin-SP-AffiniPure Donkey Anti-Rabbit Jackson ImmunoResearch Labs Cat# 711-065-152: RRID:AB_2340593 IgG (H+L) Rabbit Anti-Chromogranin A Polyclonal Abcam Cat# ab15160: RRID:AB_301704 Antibody Rabbit Polyclonal Anti-Synaptophysin Millipore Cat# AB9272: RRID:AB_570874 Antibody Rabbit Polyclonal anti-DCAMKL1 Antibody Abcam Cat# ab31704: RRID:AB_873537 Rabbit Anti-Human Lysozyme Polyclonal Dako Cat# F037201: RRID:AB_578661 Antibody, FITC Conjugated Rabbit Anti-Mucin 2 Polyclonal Antibody Santa Cruz Biotechnology Cat# sc-15334: RRID:AB_2146667 Donkey anti-Rabbit IgG (H+L) Secondary Thermo Fisher Scientific Cat# A-21206: RRID:AB_2535792 Antibody, Alexa Fluor 488 Bacterial and Virus Strains RP24-77K22 Bacterial Artificial BACPAC Resources Center N/A Chromosome Chemicals, Peptides, and Recombinant Proteins Lambda Protein Phosphatase (Lambda PP) New England Biolabs Cat# P0753S Doxycycline Hydrochloride, Ready Made Sigma-Aldrich Cat# D3072 Solution Tet Approved FBS Clontech Laboratories Cat# 631101 Tamoxifen Sigma-Aldrich Cat# T5648 Dextran Sulfate Sodium Salt MP Biomedicals Cat# 02160110 Critical Commercial Assays In-Fusion HD Cloning Kit Clontech Laboratories Cat# 639648 TruSeq Stranded mRNA Library Prep Kit Illumina Cat# 20020595 Deposited Data RNA sequencing data This paper GEO: GSE115416 Mendeley Data This paper https://doi.org/10.17632/vgvdv5b949.1 Experimental Models: Cell Lines DLD-1 Flp-In T-Rex cell line Laboratory of Stephen Taylor N/A Experimental Models: Organisms/Strains (WT)CreERT2 Atoh1 This paper N/A (9S/T-A)CreERT2 Atoh1 This paper N/A (Continued on next page) e1 Cell Stem Cell 23, 436–443.e1–e7, September 6, 2018 Continued REAGENT or RESOURCE SOURCE IDENTIFIER Oligonucleotides Left integration arm, Atoh1 locus, forward This paper N/A primer GGACAGGCGGGAACCACAGA Left integration arm, Atoh1 locus, reverse This paper N/A primer TTGTCAACACGAGCTGGTCGAA Right integration arm, Atoh1 locus, forward This paper N/A primer CAACACAACCCTGACCTGTG Right integration arm, Atoh1 locus, reverse This paper N/A primer CCCTAACCAGTGTGCCCTTA Left integration arm, DLD-1, forward primer This paper N/A AGTCAG CAACCATAGTCCCG Left integration arm, DLD-1, reverse primer This paper N/A TTCTGCGGGCGATTTGTGTA Right integration arm, DLD-1, forward This paper N/A primer TAAACGGCCACAAGTTCAGC Left integration arm, DLD-1, reverse primer This paper N/A CGGGCCTCTTCGCTATTACG Atoh1 genotyping forward primer This paper N/A TTTGTTGTTGTTGTTCGGGG Atoh1 genotyping reverse primer This paper N/A TCTTTTACCTCAGCCCACTCTT Software and Algorithms TopHat2 Kim et al., 2013 N/A DESeq2 Love et al., 2014 N/A GSEA Subramanian et al., 2005 http://software.broadinstitute.org/ gsea/index.jsp CONTACT FOR REAGENT AND RESOURCE SHARING Further information and requests for resources and reagents should be directed to and will be fulfilled by the Lead Contact, Douglas J. Winton ([email protected]). EXPERIMENTAL MODEL AND SUBJECT DETAILS Mice Mice used in this study were 8-16 weeks old males and females of C57BL/6 background. The mice were housed under controlled conditions (temperature (21 ± 2 C), humidity (55 ± 10%), 12 h light/dark cycle) in a specific-pathogen-free (SPF) facility (tested ac- cording to the recommendations for health monitoring by the Federation of European Laboratory Animal Science Associations). The animals had unrestricted access to food and water, were not involved in any previous procedures and were test naive. All experi- (WT)CreERT2 (9S/T-A)CreERT2 ments were carried out on homozygous Atoh1 and Atoh1 lines. For lineage tracing experiments, the mice tdTom/+ were heterozygous for the reporter gene (Rosa26 ). All animal experiments were carried out in accord with the guidelines of the UK Home Office, under the authority of a Home Office project license approved by the Animal Welfare and Ethical Review Body at the CRUK Cambridge Institute, University of Cambridge. Cell Lines DLD-1 (human colon adenocarcinoma, male) cells, modified with the Flp-In T-Rex system (Thermo Fisher), were used in the study. The cell line authentication was carried out using Single Tandem Repeat (STR) genotyping. Tests were performed routinely to confirm mycoplasma-negative status of the cells. The cells were maintained in Dulbecco’s Modified Eagle Medium (DMEM) with L-glutamine. Medium was supplemented with 10% Tet System-approved fetal bovine serum (FBS, Clontech). The cells were cultured under stan- dard conditions (5% CO ,37 C). Cell Stem Cell 23, 436–443.e1–e7, September 6, 2018 e2 METHOD DETAILS Cloning of mouse knock-in constructs For generation of mouse knock-ins Atoh1 locus and homology arms were amplified from a baterial artificial chromosome (BAC) RP24-77K22 (BACPAC Resources Centre). The targeting construct was assembled by a combination of seamless cloning (In-Fusion, Clontech) and restriction digest and ligation. For this a loxP site was introduced into 5 UTR of Atoh1 via PCR amplification. T2 0 0 A neomycin cassette was inserted such that the 3 UTR was not disrupted. The CreER -hCD2-3 UTR was generated via gene syn- (WT) (9S/T-A) thesis service (Integrated DNA Technologies). The Atoh1 sequence (Atoh1 or Atoh1 ) was merged with this construct, and then ligated with Atoh1 vector containing the homology arms. The targeting vector sequence was verified by Sanger sequencing and linearized by SwaI enzyme before transfecting into ES cells. The final inserted sequence is available on request. ES cell targeting Electroporation of the targeting construct into mouse ES cells was conducted by the CRUK CI Transgenic Core. ES cells were posi- tively selected with G418. Correct integration of the construct was verified by long range PCR (SequalPrep, Thermo Fisher) according 0 0 to the manufacturer’s instructions. Left integration arm was detected using a forward primer 5 -GGA CAG GCG GGA ACC ACA GA-3 0 0 and a reverse primer 5 -TTG TCA ACA CGA GCT GGT CGA A-3 . Right integration arm was amplified using the following set of 0 0 0 0 primers: forward 5 - CAA CAC AAC CCT GAC CTG TG-3 , and reverse 5 -CCC TAA CCA GTG TGC CCT TA-3 . Copy number of the clones was determined by qPCR of the neomycin selection cassette via a commercial genotyping service provider (Transnetyx). Single copy ES cell clones were taken forward for blastocyst injection, and chimeric mice were generated. Following successful germline transmission, the mice heterozygous for the targeting construct were crossed onto PGK-Cre line (Lallemand et al., 1998) in order to remove both the neo selection cassette and the endogenous Atoh1 locus at the same time. A constitutively active T2 Atoh1-P2A-CreER allele was generated in this process. Mouse genotyping Genotyping was carried out by Transnetyx. Manual genotyping by PCR was used to distinguish between homozygous and hetero- 0 0 0 zygous Atoh1 animals. The following primers were used: forward 5 -TTT GTT GTT GTT GTT CGG GG-3 ; reverse 5 -TCT TTT ACC TCA GCC CAC TCT T-3 . Creation of doxycycline inducible DLD-1 cells To generate an inducible stable cell line, a DLD-1 Flp-In T-Rex cell line containing a single Frt site was obtained (a generous gift from Prof Stephen Taylor, University of Manchester). Atoh1 construct in a pcDNA 5/FRT/TO vector (Thermo Fisher) was co-transfected with pOG44 (Flp recombinase-expressing plasmid) in a 1:9 ratio (JetPrime, Polyplus transfection). Cells were washed 24 h after trans- fection, and fresh medium was added. Two days after transfection, the cells were split at a low confluence (less than 25%), and hygromycin (400 mg/mL) was added to the trypsinised cells. Fresh medium was added to the cells every 3-4 days, until the non-trans- fected cells died off, and foci of surviving cells could be visualized. Doxycycline (100 ng/mL, Sigma) was added to the culture 24 h after seeding, to induce expression of the gene of interest. Validation of the correct recombination of the construct was carried out by PCR. Left integration arm was detected by using the 0 0 0 0 following set of primers: forward 5 -AGT CAG CAA CCA TAG TCC CG-3 ; reverse 5 - TTC TGC GGG CGA TTT GTG TA-3 . Correct 0 0 0 integration on the 3 end of the construct was done using a forward primer 5 -TAA ACG GCC ACA AGT TCA GC-3 , and a reverse 0 0 primer 5 -CGG GCC TCT TCG CTA TTA CG-3 . Parental DLD-1 Flp-In T-Rex cell line was used as a negative control. The expected loss of b-galactosidase activity on targeting was verified by X-Gal (5-bromo-4-chloro-3-indolyl-b-D-galactopyrano- side) staining of fixed cells. To validate that the constructs were integrated as a single copy in the genome, copy number qPCR was employed. Copy number TaqMan probes detected HygR (Mr00661678_cn) and used a reference copy number assay for RNase P detection. Treatment of animals T2 Induction of CreER in animals was carried out using the free base tamoxifen (Sigma) dissolved in ethanol/oil (1:9). The animals received 3 mg tamoxifen via an intra-peritoneal injection in all experiments. To define Atoh1 secretory signature, the mice were in- jected with 1 mg tamoxifen per day on 3 consecutive days for maximal labeling of all secretory lineages. SI injury was induced by exposing animals to whole-body irradiation (6 Gy). To induce colon-specific injury, mice were given 1.5% DSS (MP Biomedicals) in drinking water for 5 days. DSS was replaced every two days during the treatment. To induce lineage + DTR tracing and ablate Lgr5 cells in Lgr5 mice, the animals first received 3 mg tamoxifen i.p., followed by an i.p. injection of DT in saline (50 mg/kg) 6 h later. Crypt fractionation and single cell preparation SI (proximal 15 cm) and colon were dissected, flushed with PBS, everted and fed onto a glass rod spiral. They were incubated at 37 C +2 +2 in Hank’s Balanced Salt Solution (HBSS) without Ca and Mg , containing 10 mM EDTA and 10 mM NaOH. Crypt release was facil- itated using a vibrating stirrer (Chemap). Samples were incubated for 1 h and pulsed every 10 min. Fractions were collected after each pulse, and fresh solution added. Crypt-enriched fractions were pooled and washed in cold 2% FBS/PBS. Fraction 1 (villus-enriched) e3 Cell Stem Cell 23, 436–443.e1–e7, September 6, 2018 was discarded. Pooled fractions were resuspended in 0.05% trypsin and incubated for 7 min at 37 C, shaking every 1 min. Single cells were then filtered through a 70 mm mesh, and washed twice in 2% FBS/PBS. Flow cytometry Single cell suspension obtained by trypsin treatment was washed and incubated with an anti-mouse CD326 (EpCAM) AlexaFluor 647 antibody (1:2,000, clone G8.8, Biolegend). DAPI (10 mg/mL) was added to distinguish between live and dead cells. Flow sorting was carried out on a BD FACS Aria SORP (BD Biosciences), using appropriate single-stained and unstained controls. Whole-mount preparation Tissue was cut open, pinned out luminal side up, and fixed for 3 h at room temperature in ice-cold 4% PFA in PBS (pH 7.4). Whole- mounts were washed with PBS, and incubated with demucifying solution (3 mg/mL dithiothreitol (DTT), 20% ethanol, 10% glycerol, 0.6% NaCl, 10 mM Tris, pH 8.2) for 20 min, and mucus removed by washing with PBS. Whole-mount scanning and quantification The tdTom fluorescence in colon whole-mounts was detected using Amersham Typhoon 5 laser scanner (GE Healthcare) at a 10 mm resolution. The tdTom foci were scored manually in Fiji. Mid and distal colon were scored only as the shape of the proximal colon prevented confident assessment of tdTom patches. Antibody staining For staining whole-mount sections of 2 cm in length were excised, washed in 0.1% PBS-T for 2 days, and blocked in 10% donkey serum in PBS overnight at 4 C, protected from light. Samples were then incubated with an anti-mouse CD326 (EpCAM) AlexaFluor 647 antibody (1:100, clone G8.8, Biolegend) in 10% donkey serum in PBS for 3 days. Finally, the tissue was washed with PBS-T for 1 day. Quantification of crypts in whole-mounts Imaging was done on a TCS SP5 confocal microscope (Leica). Images were analyzed using Fiji. For SI, a minimum of 2,500 crypts per animal was scored. For colon, at least 900 crypts per mouse were scored. For the low-power analysis of clonal events, tdTom clones were scored across the entire length of the SI whole-mounts using a stereomicroscope (Nikon). Immunostainings For immunohistochemistry SI and colon were opened and fixed for 24 h in 10% neutral buffered formaldehyde in PBS. The tissue was paraffin embedded and sectioned by the CRUK CI Histopathology core. Haematoxylin and eosin staining was performed using an automated ST5020 Multistainer (Leica Biosystems). Alcian Blue and Periodic Acid/Schiff staining was carried out by the CI Histopa- thology Core. Briefly, slides were incubated in Alcian Blue for 10 min, and washed in water. They were then incubated in 0.5% pe- riodic acid for 5 min, and washed 3 times. Slides were incubated in Schiff’s reagent for 15 min, washed 3 times, and counterstained with Mayer’s Haematoxylin. BrdU and lysozyme immunohistochemistry was carried out using a Bond Max autostainer (Leica), with a proteinase K antigen retrieval. Slides were blocked with 3% hydrogen peroxide, followed by incubation in Avidin/Biotin Blocking Kit (Vector Laboratories). BrdU was detected using a sheep anti-BrdU antibody (1:500, Abcam ab1893). Rabbit anti-lysozyme antibody (1:500, Dako A0099) was used for lysozyme staining. Secondary antibodies in the two cases were biotinylated donkey anti-sheep (1:250, Jackson ImmunoResearch 713-066-147) and biotinylated donkey anti-rabbit (1:250, Jackson ImmunoResearch 711-065-152), respectively. Slides were incubated with Streptavidin coupled with horseradish peroxidase (HRP), and color developed using diaminobenzidine (DAB) and DAB Enhancer (Leica). Synaptophysin and Chromogranin A detection was carried out by manual IHC. Antigen retrieval was performed with 10 mM citrate buffer (pH 6.0) in a pressurised heating chamber. Tissue sections were incubated with rabbit anti-Chromogranin A antibody (1:500, Abcam ab15160), rabbit anti-Synaptophysin antibody (1:300, Millipore AB9272), overnight at 4 C. Slides were incubated with biotinylated donkey anti-rabbit secondary antibody (1:500, Jackson ImmunoResearch 711-065-152). Streptavidin-HRP conjugate (Vector Laboratories) was added onto the slides and incubated for 30 min. DAB Chromogen substrate (Dako) was added for dye development. Counterstaining and dehydration was performed on the ST5020 Multistainer (Leica) followed by coverslipping. For immunofluorescence tissue was excised and fixed for 48 h in 4% PFA in PBS at 4 C, after which it was transferred to 20% sucrose solution. After cryosectioning antigen retrieval where needed was accomplished by incubating the slides in 1% SDS for 5 min. Blocking was performed with 5% donkey serum. Following a wash primary antibodies were added and incubated overnight at 4 C. The following primary antibodies were used: rabbit FITC-anti-Lyz (1:400, Dako, F037201), rabbit anti-Muc2 (1:50, Santa Cruz, sc-15334), rabbit anti-ChgA (1:100, Abcam, ab15160), and rabbit anti-Dclk1 antibody (1:1000, Abcam, ab31704). Secondary detec- tion was with AlexaFluor 488 donkey anti-rabbit secondary antibody (1:500, Thermo Fisher, A-21206). Alkaline phosphatase activity was detected using Blue AP kit (Vector Laboratories). Sections were covered with Prolong Gold with DAPI (Life Technologies). Fluo- rescent imaging was carried out on a TCS SP5 confocal microscope (Leica). Cell Stem Cell 23, 436–443.e1–e7, September 6, 2018 e4 Single molecule FISH Harvested SI and colon tissues were flushed with cold 4% formaldehyde (FA) in PBS and incubated first in 4% FA/PBS for 3 hours, then in 30% sucrose in 4% FA/PBS overnight at 4 C with constant agitation. Fixed tissues were embedded in OCT. Quantification of co-expression was achieved by smFISH. Probe library design, hybridization procedures, and imaging settings were carried out ac- cording to published methods (Itzkovitz et al., 2011; Lyubimova et al., 2013). A Nikon-Ti-E inverted fluorescence microscope equip- ped with a Photometrics Pixis 1024 CCD camera was used to image a 10 mm cryo-section. A stack of 30 frames with 0.3 mm intervals was acquired to allow 3D cell imaging. FITC-conjugated antibody for E-cadherin was added to the hybridization mix and used to visualize cell borders. Detection of cells that were positive for Lgr5 transcripts, Atoh1 transcripts or both was performed manually with Fiji. Analysis of gut sections Stained longitudinal sections of the SI and colon were visualized and positive cells scored manually. BrdU and negative nuclei were + + scored in complete half-crypt sections. Lysozyme cells were counted per whole crypt section. Alcian Blue and PAS cells were counted in complete half-villus sections, between the crypt neck and the tip of the villus. Cells in which the stain was clearly asso- + + ciated with a corresponding nucleus were marked as positive. Chromogranin A and synaptophysin cells were scored per complete half-crypt-villus section. Positive and negative crypts were scored, and results expressed as a frequency of positive cells. Colon ulceration scoring H&E-stained sections of colons were scanned on Aperio slide scanner (Leica Biosystems), and analyzed using eSlide Manager (Leica Biosystems). Ulceration was defined as a region of a complete loss of crypt architecture and high cellularity. RNA isolation For gene expression analysis by qPCR, cells were lysed and RNA isolated using RNeasy Mini Plus kit (QIAGEN). For sequencing, total RNA was isolated from flow-sorted cells using RNeasy Micro Plus kit (QIAGEN). Gene expression analysis RNA was converted into cDNA (iScript cDNA synthesis kit, BioRad), and gene expression was analyzed using TaqMan gene expression probes (Thermo Fisher). The following probes were used: Atoh1 (Mm00476035_s1), Muc2 (Hs00894053_g1), Tff3 (Hs00902278_m1), Spdef (Hs01026050_m1), Dll4 (Hs00184092_m1), Rassf4 (Hs00604698_m1), Gapdh (Hs02758991_g1). All TaqMan assays are listed in Table S3. RNA sequencing Samples for RNA sequencing were collected 24 h post-tamoxifen induction (3 mg i.p. injection). The tissue was fractionated as described above and cells prepared for flow cytometry. The cells were stained and sorted in the same way as for other experiments, + + as noted above. EpCAM tdTom live cells were collected directly into the lysis buffer and RNA was extracted immediately following the sort (RNeasy Micro Plus Kit, QIAGEN). RNA quality was assessed on a 2100 Bioanalyser instrument (Agilent), according to the manufacturer’s instructions. The libraries were prepared using TruSeq Stranded mRNA Library Prep Kit (Illumina) and sequenced as 50 bp single-end reads on the Illumina HiSeq 4000 system. Western blotting Protein extracts for SDS-PAGE were prepared by lysing the cells with RIPA buffer containing protease and phosphatase inhibitor cocktail (Thermo Fisher). Mouse anti-ATOH1 antibody (1:100, Developmental Studies Hybridoma Bank) and a rabbit anti-b-A anti- body (1:5,000, ab8227, Abcam) were used. Fluorescent secondary antibodies were used (Li-Cor, goat anti-mouse 800LT (1:5,000), goat anti-rabbit 680LT (1:20,000)). For some experiments, protein extracts were incubated with l phosphatase (New En- gland Biolabs) prior to western blotting, according to the manufacturer’s instructions. In vitro kinase assay The assay was performed as previously described (Azzarelli et al., 2017), with minor modifications. HA-tagged WT and mutant ATOH1 were in vitro translated (TNT Quick Coupled Transcription/Translation Systems, Promega) in the presence of LiCl (800 mM) to reduce potential phosphorylation in reticulocyte lysate. Samples were incubated with human recombinant CDK/Cyclins (0.25 mM final concentration) in the presence of 10 mM ATP for 1 h at 30 C. Proteins were separated on Phos-tag gels (Alpha Laboratories, 7.5% acrylamide, 50 mM phos-tag PAGE, Wako) and immunoblotted with rat anti-HA-Peroxidase (1:5000, Roche). Cell proliferation and cell cycle analysis Cell proliferation was assessed by an automated live-cell imaging system (IncuCyte ZOOM, Essen Bioscience). For cell cycle anal- ysis, the cells were trypsinised, washed, fixed with ethanol, and stained with propidium iodide prior to flow cytometry. e5 Cell Stem Cell 23, 436–443.e1–e7, September 6, 2018 RNA sequencing analysis The reads were aligned to the mouse reference genome [GRCm38] using TopHat2 aligner (Kim et al., 2013). Differentially expressed gene lists were generated using DESeq2 package from Bioconductor (Love et al., 2014). Secretory signature gene list + - The list of differentially expressed genes (p < 0.01) was generated by comparing the transcripts from tdTom and tdTom cells of (WT)CreERT2 tdTom/+ + Atoh1 Rosa26 mice following tamoxifen. Upregulated genes in tdTom cells were selected to define a secretory signature in the small intestine and colon (Table S1). The top 500 upregulated, differentially expressed genes were used to perform the Gene Set Enrichment Analysis (GSEA). Gene Set Enrichment Analysis (GSEA) This analysis was performed using the GSEA software from the Broad Institute (http://software.broadinstitute.org/gsea/index.jsp) (Subramanian et al., 2005). The list comprised all differentially expressed and non-differentially expressed genes from the 9S/T-A v WT comparison in SI and colon, respectively. This gene list was probed with the previously generated secretory signatures (top 500 upregulated genes), and the published Atoh1 gene signatures for ileum, colon (Lo et al., 2016), and intestinal stem cells (Mun˜ oz et al., 2012). QUANTIFICATION AND STATISTICAL ANALYSIS Computational analysis The process by which crypt stem cells replace each occurs in a random though predictable manner. This behavior can be modeled via a stochastic birth-death process (Lopez-Garcia et al., 2010; Snippert et al., 2010). The model was derived to model experiments where a single stem cell is labeled in a handful of crypts. As the number of initially labeled crypts was not of interest and to bypass any variability coming from the initial induction, make the different time points comparable, the equations were rescaled to account for only the surviving clones. Here we know the parameters of the stem cell dynamics (Kozar et al., 2013; Vermeulen et al., 2013), and would like to know the starting number of labeled stem cells per crypt and the number of labeled crypts. For this analysis we use the equations described previously (Lopez-Garcia et al., 2010; Snippert et al., 2010), reproduced below. The probability of a crypt having clone of size n (for 0 < n < N) at time t is: N1 2 pm pmn 2 pm 4l sin p ðtÞ sin sin e t (Equation 1) N N N 2N m = 1 Here n is the number of labeled stem cells, N is the total number of stem cells, l is the rate of stem cell replacement. And for the probability of all stem cells labeled we have: N1 2 pm 2 pm m=+ 1 2 4l sin p ðtÞ ð1Þ cos 1  e t (Equation 2) N 2N 2N m = 1 These equations assume the initial conditions of one labeled stem cell at t = 0. The starting labeled stem cells were chosen randomly at the beginning of each simulation. The values we observe for the clonal frequencies are substantially lower than what the model would predict, suggesting that not all crypts have labeled stem cells. In order to find out the fraction of labeled crypts v we use a mixture model: Q ðtÞ = ð1  vÞd + vP ðtÞ (Equation 3) n 0;n n Where Q ðtÞ is the probability that a randomly selected crypt has a clone of size n labeled stem cells at time t. We use the values of N, l and t from Kozar et al. (2013) and Vermeulen et al. (2013) and estimate v. Model fitting For every mouse, at day 30 we count the number of clones (k ) and the number of crypts (C ). We use a hierarchical model to capture i i the mouse to mouse variability. The statistical model is a follows k  BinomialðC ; R ð30  tÞÞ (Equation 4) i i i R ð30  tÞ Student tðh; Qð30  tÞ; sÞ (Equation 5) Here R is truncated to [0, 1]. For the SI no distinction is made in clone size, so Q is the sum of all Q and for the colon we use only the i n full clones for fitting Q = Q N. The priors on the population parameters are: h  Gammað2:0:1Þ (Equation 6) Cell Stem Cell 23, 436–443.e1–e7, September 6, 2018 e6 s  Gammað0:01; 0:01Þ (Equation 7) The prior on the mixing coefficient is h  Betað1=2; 1=2Þ (Equation 8) The posterior was derived via MCMC using Rstan (Carpenter et al., 2017). For the proximal and distal SI we used is t = 5 as the clones were measured in ribbons coming out of the crypt, which take a few days to emerge from the crypt base. Whereas for the colon we used t = 1. The parameters used were t = 0:1, N = 5 for proximal SI, t = 0:2, N = 6 for distal and t = 0:3, N = 7 for colon. Statistical analysis Statistical tests were not used to predetermine sample size. Randomization was not performed to allocate samples/animals to exper- imental groups. Blinding was performed for quantifications in Figures 3F and 3G, as well as Figures S2B–S2F. Data analysis was per- formed using GraphPad Prism software or R package. DATA AND SOFTWARE AVAILABILITY The accession number for the RNA sequencing data reported in this paper is GEO: GSE115416. Mendeley Dataset of original data can be accessed at https://doi.org/10.17632/vgvdv5b949.1. e7 Cell Stem Cell 23, 436–443.e1–e7, September 6, 2018

Journal

Cell Stem Cell – Unpaywall

Published: Sep 1, 2018

There are no references for this article.