Notch signaling modulates proliferation and differentiation of intestinal crypt base columnar stem cells

Abstract

Notch signaling is known to regulate the proliferation and differentiation of intestinal stem and progenitor cells; however, direct cellular targets and specific functions of Notch signals had not been identified. We show here in mice that Notch directly targets the crypt base columnar (CBC) cell to maintain stem cell activity. Notch inhibition induced rapid CBC cell loss, with reduced proliferation, apoptotic cell death and reduced efficiency of organoid initiation. Furthermore, expression of the CBC stem cell-specific marker Olfm4 was directly dependent on Notch signaling, with transcription activated through RBP-Jκ binding sites in the promoter. Notch inhibition also led to precocious differentiation of epithelial progenitors into secretory cell types, including large numbers of cells that expressed both Paneth and goblet cell markers. Analysis of Notch function in Atoh1-deficient intestine demonstrated that the cellular changes were dependent on Atoh1, whereas Notch regulation of Olfm4 gene expression was Atoh1 independent. Our findings suggest that Notch targets distinct progenitor cell populations to maintain adult intestinal stem cells and to regulate cell fate choice to control epithelial cell homeostasis.

Fig. 1.

Loss of crypt base columnar (CBC) stem cells with Notch inhibition. Histological analysis of jejunum from Lgr5-GFP mice treated with vehicle (veh) or DBZ (n=4). (A,B) Co-immunostaining for GFP (green) and Ki67 (red) with DAPI nuclear counterstain (blue). Arrowheads, GFP-positive Ki67-negative cells; arrows, GFP-positive Ki67-positive cells. (C) Quantitative reverse transcriptase PCR (qRT-PCR) analysis of GFP mRNA abundance. Values were normalized to Gapdh expression and are reported as fold-change relative to vehicle control. (D,E) Morphometric analysis of GFP and Ki67 co-stained paraffin sections, showing the percentage of crypts that contained at least one GFP-positive cell (D) and the average number of GFP-positive cells per expressing crypt (E). (F) Co-immunostaining (arrows) for activated caspase 3 (red) and GFP (green). Insets show separate green and red channels of the boxed region. (G) Crypts were isolated from adult mice to establish organoids, which were pretreated for 2-5 days with 25 μM DAPT before dispersal and passage to initiate new organoids without further addition of γ-secretase inhibitor (GSI). The efficiency of organoid initiation was assessed 2 days post-passage by calculating the percentage of organoids relative to the initial number of plated crypts. Quantitative data are presented as mean ± s.e.m. ^*P<0.05, ^**P<0.01, by Student’s t-test compared with vehicle-treated control. Scale bars: 10 μm.

Fig. 2.

The CBC stem cell marker Olfm4 is a sensitive readout of intestinal Notch activity. (A,B) Stem/progenitor marker expression was determined by qRT-PCR analysis of intestinal RNAs from adult mice treated with vehicle (veh) or DBZ (n=6) (A) or from duodenum treated with neutralizing antibodies against Notch1 and Notch2 (αN1 + αN2) or an isotype control (αGD) (n=3) (B). Olfm4 is not expressed (N.E.) in the mouse colon. (C) Olfm4 expression levels in newborn Vil-Cre (Cre) and Vil-Cre; Rosa^NotchIC (NICD) intestine (n=5). (D,E) Olfm4 expression (D) or 427OLFM4-luciferase activity (E) in the human colon cancer cell line LS174T treated with vehicle or DAPT (40 μM) or transfected with empty vector (Mock) or NICD1 (n=3). Gene expression values were normalized to Gapdh. Luciferase activity values were normalized to total protein content. All data are expressed as mean ± s.e.m. of the fold-change relative to the corresponding control-treated region/group. ^*P<0.05, ^**P<0.01, ^***P<0.001.

Fig. 3.

Notch inhibition leads to secretory cell hyperplasia. Secretory cell populations were analyzed in the duodenum of vehicle-treated (A,C,E,G) and DBZ-treated (B,D,F,H) adult mice. (A,B) Goblet cells were stained with periodic acid-Schiff/Alcian Blue (PAS/AB). (C,D) Enteroendocrine cells were identified by immunostaining for chromogranin A (CgA; green, arrows). (E,F) Tuft cells were identified by DCAMKL1 immunostaining (green, arrows). (G,H) Paneth cells were identified by lysozyme immunostaining (green). Arrows indicate punctate staining in mature Paneth cells in vehicle-treated intestine. Insets show high-magnification views of the boxed regions. DAPI (blue) was used as nuclear counterstain in C-H. Scale bars: 100 μm.

Fig. 4.

Accumulation of cells co-expressing Paneth and goblet cell markers with Notch inhibition. Intestines from vehicle-treated (A,C,E,G) and DBZ-treated (B,D,F,H) mice were analyzed for Paneth cell morphology and marker expression. (A,B) Electron microscopy analysis of jejunal sections. Blue outline (asterisk) marks ‘Paneth’ cells; orange outline (arrow) marks a CBC stem cell; arrowhead marks a goblet cell. (C-F) Co-immunostaining of paraffin sections from ileum (C,D) and colon (E,F) for the Paneth cell marker Mmp7 (green) and the goblet cell marker Muc2 (red) with DAPI nuclear stain (blue) to identify double-positive cells. Crypt height is indicated with brackets in C,D. Rare double-positive cells were observed in vehicle-treated control ileum (C, arrowhead). (G,H) Immunostaining jejunum for the transcription factor Sox9 (green) with DAPI nuclear stain (pseudo-colored red). Inset shows green channel of the boxed area. Scale bars: 10 μm in A,B; 50 μm in C-H.

Fig. 5.

Notch regulation of Olfm4 expression is Atoh1 independent. (A,B) Immunostaining of duodenum for Atoh1-positive cells (brown, arrowheads) in vehicle (veh)- and DBZ-treated mice. (C-F) Expression of the indicated transcription factors was determined by qRT-PCR analysis of duodenal RNAs from mice treated with vehicle or DBZ. Values were normalized to Gapdh and are expressed as fold-change relative to the vehicle control. n=4; ^*P<0.05, ^**P<0.01, ^***P<0.001. (G-I) Analysis of fetal intestine cultured from wild-type (WT) and Atoh1^lacZ/lacZ (KO) mice treated with vehicle or 40 μM DAPT for 3 days. (G) Goblet cells were visualized by PAS/AB. (H,I) Expression of the goblet cell marker Tff3 (H) and the CBC marker Olfm4 (I) was determined by qRT-PCR. Values were normalized to Gapdh and are presented as the mean ± s.e.m. n=3-6, ^*P<0.05, ^**P<0.01, compared with vehicle treatment of the same genotype. Scale bars: 50 μm.

Fig. 6.

OLFM4 transcription is regulated by canonical Notch signaling. (A) The proximal promoter sequence of the human OLFM4 gene, featuring three RBP-Jκ consensus binding sites, an N-box, an E-box, the TATA box (underlined), transcription start site (arrow, +1) and translation start site (underlined). The asterisks at –146 and –28 indicate the positions of primers used to amplify chromatin in the immunoprecipitation (ChIP) analysis. (B,C) Luciferase activity was measured in LS174T cells that were co-transfected with the full-length human 427OLFM4-luciferase construct (WT) or mutant OLFM4 luciferase constructs (Δ427-306, Δ290-161 or Δ145-56) together with either empty vector or the NICD1 expression construct (B). Mutant constructs with targeted deletions of the RBPJκ-b (Δb) or RBPJκ-a (Δa) consensus sites were transfected with NICD1, dominant-negative mastermind (dnMAML), or both (C). Bar colors correspond to the shaded sequence regions in A that were altered in each mutant construct. Transfection data are reported as fold-change relative to untreated WT controls. n=3 experiments performed in triplicate; ^**P<0.01, ^***P<0.001. (D) ChIP of the proximal region containing RBP-Jκ sites or of a distant upstream region without RBP-Jκ sites was performed in LS174T cells transfected with either vector control (CS2) or NICD1. (E) ChIP of a HES1 promoter region that contains known RBP-Jκ consensus sites. ChIP data were normalized by subtracting the values of the control IgG samples from those of the Notch antibody samples and are presented as the percentage of the initial chromatin input. n=4 independent experiments, qPCR performed in triplicate; ^**P<0.01. Values are presented as the mean ± s.e.m.

Fig. 7.

Notch signaling regulates several aspects of intestinal epithelial cell homeostasis. A summary of the major findings of this study, showing that Notch regulates multiple aspects of intestinal epithelial cell differentiation. (1) Notch signaling directly targets the CBC stem cell to activate Olfm4 transcription, maintain proliferation and promote cell survival. (2) Notch acts to promote differentiation to the absorptive lineage by repressing Atoh1 transcription. Although the specific cellular target for this function has not been identified, Notch is likely to act on a transit-amplifying progenitor cell to bias cell fate choice to absorptive rather than secretory cells. (3) Notch appears to act later, in an indirect manner, to segregate the Paneth/goblet cell lineage and/or maintain the mature Paneth cell phenotype.