Notch lineages and activity in intestinal stem cells determined by a new set of knock-in mice

Abstract

The conserved role of Notch signaling in controlling intestinal cell fate specification and homeostasis has been extensively studied. Nevertheless, the precise identity of the cells in which Notch signaling is active and the role of different Notch receptor paralogues in the intestine remain ambiguous, due to the lack of reliable tools to investigate Notch expression and function in vivo. We generated a new series of transgenic mice that allowed us, by lineage analysis, to formally prove that Notch1 and Notch2 are specifically expressed in crypt stem cells. In addition, a novel Notch reporter mouse, Hes1-EmGFP(SAT), demonstrated exclusive Notch activity in crypt stem cells and absorptive progenitors. This roster of knock-in and reporter mice represents a valuable resource to functionally explore the Notch pathway in vivo in virtually all tissues.

Figure 1. Notch1 and Notch2 expression in the small intestine.

X-gal staining (blue) of thin sections of proximal small intestine of N1-CreERT2^SAT/+;R26R/+ (A) and N2-CreERT2^SAT/+;R26R/+ (B) 24 hours after tamoxifen administration shows that Notch1 and Notch2 are expressed only in the intestinal crypts. C–D) In situ hybridization for Notch1 (C) and Notch2 (D) demonstrates specific expression in intestinal crypts. E–F) Whole mounts duodenum of N1-CreERT2^SAT/+;R26R/+ (E) and N2-CreERT2^SAT/+;R26R/+ (F) 7 months after tamoxifen administration demonstrates long-term labeling of X-gal marked Notch lineages. Counterstain: hematoxylin/eosin in A. Scale bar: 50 µm in A–D and 1 mm in E–F.

Figure 2. Tracing of Notch1 and Notch2 lineages shows stem cell labeling.

Four week-old N1-CreERT2^SAT/+;R26R/+ (A–C) and N2-CreERT2^SAT/+;R26R/+ (D–F) mice were injected with tamoxifen and analyzed for β-galactosidase expression (blue) after 3 days (A,D), 7 days (B,E) and 60 days (C,F). The lineage of Notch1-expressing cells is shown in thin longitudinal sections in A and C and in whole mounts in B to better visualize continuous streams of labeled cells. Sections are counterstained with hematoxylin/eosin in C–F. Scale bar: 40 µm in A,D,E; 50 µm in B and 80 µm in C,F.

Figure 3. Kinetics of stem cell replacement and monoclonal conversion of Notch1-expressing cells.

To define the localization of Notch1-expressing crypt cells, only crypts harboring a single marked cell were scored, as shown in A. Cells were classified as “CBCs” (marked by white arrowheads in B), “+4” (indicated by a white arrow in B) or “other” when it was not possible to clearly assign their position. In B, N1-CreERT2^SAT/+;R26^mTmG/+ mice were injected with tamoxifen at 4 weeks and analyzed 24 h after induction. Genetically marked cells show membrane-bound green fluorescence (GFP), while membrane-anchored Tomato (red) fluorescence is present in non-recombined cells. C: graphic view of the quantification of labeled cells. At least five fields from a single intestine were counted and at least three mice per time point (12 h, 18 h, 24 h) were scored independently by two researchers. The total number of crypts counted was 1095. D) Marked crypts were counted at different time points after tamoxifen injection (x axis). The Y axis represents the percentage of labeled crypts over the total crypt population. E) Graphic representation of the percentage of fully labeled (dark blue) or mosaic crypts (light blue) among the population of marked crypts. F) Data of the monoclonal conversion of marked crypts (dots), compared with a fit (straight line), applying the neutral drift theory . The only fitting parameter is the stem cell renewal rate λ = 1.32±0.05/day. G) Representative cross section of 60 days chase showing fully labeled (all blue), partially labeled and non-recombined (pink) crypts used to quantitatively evaluate the monoclonal conversion kinetics. Counterstain: hematoxylin and eosin. Scale bar: 10 µm in A, B.

Figure 4. Notch1-expressing cells in the intestine are multipotent.

N1-CreERT2^SAT/+;R26^mTmG/+ mice were analyzed 30 days after tamoxifen administration. Notch1-expressing crypt cells give rise to all four differentiated cell types of the intestine, as illustrated by the concomitant labeling of membrane-tagged GFP (Notch progeny in green) and A-A′) UEA1 staining (red) marking Goblet cells indicated by the arrows; B-B′) anti-lysozyme staining (red) recognizing Paneth cells; C-C′) anti-villin staining (red) marking the apical brush border of enterocytes; and D-D′) anti-chromogranin immuno-labeling (red) revealing the position of enteroendocrine cells (indicated by a white arrowhead in D). A′–D′ show higher magnifications of the four cell types presenting both membrane-tethered green and red fluorescence. Nuclei are marked in blue with DAPI. Scale bar: 50 µm in A–D; 10 µm in B′ and 5 µm in A′, C′ and D′.

Figure 5. Expression of the Notch transcriptional target Hes1 in the intestine.

A) Schematic representation of the targeted Hes1 locus in the Hes1-EmGFP^SAT reporter line. Hes1-EmGFP^SAT mice were analyzed for GFP expression in frozen sections (B–G). DBZ-treated animals (D) show absence of Hes1-EmGFP expression in crypt cells, while the signal in the villus compartment is unaltered by Notch inhibition, as compared to untreated mice (B and C). Immunofluorescent labeling of secretory cells reveals the complete absence of GFP label from differentiated secretory cells, detected by anti-mucin antibodies for Goblet cells (red) in E, anti-lysozyme staining for Paneth cells (red) in F (see also the absence of green fluorescence in Paneth cells in the inset B′) and anti-chromogranin staining for enteroendocrine cells and (red) in G. Nuclei are marked in blue with DAPI in C–G. Scale bar: 100 µm in B; 10 µm in B′; 80 µm in C–D and 40 µm in E–G.