Letter | Published:

Single-cell transcriptomes of the regenerating intestine reveal a revival stem cell

Abstract

The turnover of the intestinal epithelium is driven by multipotent LGR5+ crypt-base columnar cells (CBCs) located at the bottom of crypt zones1. However, CBCs are lost following injury, such as irradiation2, but the intestinal epithelium is nevertheless able to recover3. Thus, a second population of quiescent ‘+4’ cells, or reserve stem cells (RSCs), has previously been proposed to regenerate the damaged intestine4,5,6,7. Although CBCs and RSCs were thought to be mutually exclusive4,8, subsequent studies have found that LGR5+ CBCs express RSC markers9 and that RSCs were dispensable—whereas LGR5+ cells were essential—for repair of the damaged intestine3. In addition, progenitors of absorptive enterocytes10, secretory cells11,12,13,14,15 and slow cycling LGR5+ cells16 have been shown to contribute to regeneration whereas the transcriptional regulator YAP1, which is important for intestinal regeneration, was suggested to induce a pro-survival phenotype in LGR5+ cells17. Thus, whether cellular plasticity or distinct cell populations are critical for intestinal regeneration remains unknown. Here we applied single-cell RNA sequencing to profile the regenerating mouse intestine and identified a distinct, damage-induced quiescent cell type that we term the revival stem cell (revSC). revSCs are marked by high clusterin expression and are extremely rare under homoeostatic conditions, yet give rise—in a temporal hierarchy—to all the major cell types of the intestine, including LGR5+ CBCs. After intestinal damage by irradiation, targeted ablation of LGR5+ CBCs, or treatment with dextran sodium sulfate, revSCs undergo a YAP1-dependent transient expansion, reconstitute the LGR5+ CBC compartment and are required to regenerate a functional intestine. These studies thus define a unique stem cell that is mobilized by damage to revive the homoeostatic stem cell compartment and regenerate the intestinal epithelium.

Access optionsAccess options

Rent or Buy article

Get time limited or full article access on ReadCube.

from$8.99

All prices are NET prices.

Data availability

The scRNA-seq data reported in this study is available from Gene Expression Omnibus with accession code GSE123516. All other data used in this study are provided within the article as Source Data or are available from the corresponding authors upon reasonable request.

Code availability

Citations for all previously described R packages used in this study are provided above, and the step-by-step scRNA-seq data analysis is described in graphical form in Extended Data Fig. 1a. Custom codes used for gene correlation analysis and pCreode-based pseudotime analysis are available at https://github.com/sandeep13712/wranalab-revscs.

Additional information

Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

  1. 1.

    Barker, N. et al. Identification of stem cells in small intestine and colon by marker gene Lgr5. Nature 449, 1003–1007 (2007).

  2. 2.

    Tao, S. et al. Wnt activity and basal niche position sensitize intestinal stem and progenitor cells to DNA damage. EMBO J. 36, 2920–2921 (2017).

  3. 3.

    Metcalfe, C., Kljavin, N. M., Ybarra, R. & de Sauvage, F. J. Lgr5 + stem cells are indispensable for radiation-induced intestinal regeneration. Cell Stem Cell 14, 149–159 (2014).

  4. 4.

    Yan, K. S. et al. The intestinal stem cell markers Bmi1 and Lgr5 identify two functionally distinct populations. Proc. Natl Acad. Sci. USA 109, 466–471 (2012).

  5. 5.

    Powell, A. E. et al. The pan-ErbB negative regulator Lrig1 is an intestinal stem cell marker that functions as a tumor suppressor. Cell 149, 146–158 (2012).

  6. 6.

    Takeda, N. et al. Interconversion between intestinal stem cell populations in distinct niches. Science 334, 1420–1424 (2011).

  7. 7.

    Montgomery, R. K. et al. Mouse telomerase reverse transcriptase (mTert) expression marks slowly cycling intestinal stem cells. Proc. Natl Acad. Sci. USA 108, 179–184 (2011).

  8. 8.

    Li, N. et al. Single-cell analysis of proxy reporter allele-marked epithelial cells establishes intestinal stem cell hierarchy. Stem Cell Reports 3, 876–891 (2014).

  9. 9.

    Muñoz, J. et al. The Lgr5 intestinal stem cell signature: robust expression of proposed quiescent’ +4′ cell markers. EMBO J. 31, 3079–3091 (2012).

  10. 10.

    Tetteh, P. W. et al. Replacement of lost Lgr5-positive stem cells through plasticity of their enterocyte-lineage daughters. Cell Stem Cell 18, 203–213 (2016).

  11. 11.

    Buczacki, S. J. A. et al. Intestinal label-retaining cells are secretory precursors expressing Lgr5. Nature 495, 65–69 (2013).

  12. 12.

    Yan, K. S. et al. Intestinal enteroendocrine lineage cells possess homeostatic and injury-inducible stem cell activity. Cell Stem Cell 21, 78–90.e6 (2017).

  13. 13.

    Jadhav, U. et al. Dynamic reorganization of chromatin accessibility signatures during dedifferentiation of secretory precursors into Lgr5+ intestinal stem cells. Cell Stem Cell 21, 65–77.e5 (2017).

  14. 14.

    van Es, J. H. et al. Dll1+ secretory progenitor cells revert to stem cells upon crypt damage. Nat. Cell Biol. 14, 1099–1104 (2012).

  15. 15.

    Yousefi, M., Li, L. & Lengner, C. J. Hierarchy and plasticity in the intestinal stem cell compartment. Trends Cell Biol. 27, 753–764 (2017).

  16. 16.

    Barriga, F. M. et al. Mex3a marks a slowly dividing subpopulation of Lgr5+ intestinal stem cells. Cell Stem Cell 20, 801–816.e7 (2017).

  17. 17.

    Gregorieff, A., Liu, Y., Inanlou, M. R., Khomchuk, Y. & Wrana, J. L. Yap-dependent reprogramming of Lgr5+ stem cells drives intestinal regeneration and cancer. Nature 526, 715–718 (2015).

  18. 18.

    Moor, A. E. et al. Spatial reconstruction of single enterocytes uncovers broad zonation along the intestinal villus axis. Cell 175, 1156–1167.e15 (2018).

  19. 19.

    Yui, S. et al. YAP/TAZ-dependent reprogramming of colonic epithelium links ECM remodeling to tissue regeneration. Cell Stem Cell 22, 35–49.e7 (2018).

  20. 20.

    Nusse, Y. M. et al. Parasitic helminths induce fetal-like reversion in the intestinal stem cell niche. Nature 559, 109–113 (2018).

  21. 21.

    Zhang, F. et al. Clusterin facilitates stress-induced lipidation of LC3 and autophagosome biogenesis to enhance cancer cell survival. Nat. Commun. 5, 5775 (2014).

  22. 22.

    Park, J.-J., Lim, K.-H. & Baek, K.-H. Annexin-1 regulated by HAUSP is essential for UV-induced damage response. Cell Death Dis. 6, e1654 (2015).

  23. 23.

    Wang, Y. et al. A nuclease that mediates cell death induced by DNA damage and poly(ADP-ribose) polymerase-1. Science 354, aad6872 (2016).

  24. 24.

    Poletto, M., Lirussi, L., Wilson, D. M. III & Tell, G. Nucleophosmin modulates stability, activity, and nucleolar accumulation of base excision repair proteins. Mol. Biol. Cell 25, 1641–1652 (2014).

  25. 25.

    Jirawatnotai, S. et al. A function for cyclin D1 in DNA repair uncovered by protein interactome analyses in human cancers. Nature 474, 230–234 (2011).

  26. 26.

    Picelli, S. et al. Full-length RNA-seq from single cells using Smart-seq2. Nat. Protoc. 9, 171–181 (2014).

  27. 27.

    Herring, C. A. et al. Unsupervised trajectory analysis of single-cell RNA-seq and imaging data reveals alternative tuft cell origins in the gut. Cell Syst. 6, 37–51.e9 (2018).

  28. 28.

    Oh, S. Y., Cho, K. A., Kang, J. L., Kim, K. H. & Woo, S. Y. Comparison of experimental mouse models of inflammatory bowel disease. Int. J. Mol. Med. 33, 333–340 (2014).

  29. 29.

    Gong, S. et al. A gene expression atlas of the central nervous system based on bacterial artificial chromosomes. Nature 425, 917–925 (2003).

  30. 30.

    Wu, S., Wu, Y. & Capecchi, M. R. Motoneurons and oligodendrocytes are sequentially generated from neural stem cells but do not appear to share common lineage-restricted progenitors in vivo. Development 133, 581–590 (2006).

  31. 31.

    Reginensi, A. et al. Yap- and Cdc42-dependent nephrogenesis and morphogenesis during mouse kidney development. PLoS Genet. 9, e1003380 (2013).

  32. 32.

    Yi, J. et al. Large tumor suppressor homologs 1 and 2 regulate mouse liver progenitor cell proliferation and maturation through antagonism of the coactivators YAP and TAZ. Hepatology 64, 1757–1772 (2016).

  33. 33.

    Gregorieff, A. et al. Expression pattern of Wnt signaling components in the adult intestine. Gastroenterology 129, 626–638 (2005).

  34. 34.

    Butler, A., Hoffman, P., Smibert, P., Papalexi, E. & Satija, R. Integrating single-cell transcriptomic data across different conditions, technologies, and species. Nat. Biotechnol. 36, 411–420 (2018).

  35. 35.

    McCarthy, D. J., Campbell, K. R., Lun, A. T. L. & Wills, Q. F. Scater: pre-processing, quality control, normalization and visualization of single-cell RNA-seq data in R. Bioinformatics 33, 1179–1186 (2017).

  36. 36.

    Waltman, L. & Van Eck, N. J. A smart local moving algorithm for large-scale modularity-based community detection. Eur. Phys. J. B 86, 471 (2013).

  37. 37.

    Zappia, L. & Oshlack, A. Clustering trees: a visualization for evaluating clusterings at multiple resolutions. Gigascience 7, 274035 (2018).

  38. 38.

    Van Der Maaten, L. & Hinton, G. Visualizing Data using t-SNE. J. Mach. Learn. Res. 9, 2579–2605 (2008).

  39. 39.

    McGinnis, C. S., Murrow, L. M. & Gartner, Z. J. DoubletFinder: Doublet detection in single-cell RNA sequencing data using artificial nearest neighbors. Preprint at https://doi.org/10.1101/352484 (2018).

  40. 40.

    Tringe, S. G. & Hugenholtz, P. A renaissance for the pioneering 16S rRNA gene. Curr. Opin. Microbiol. 11, 442–446 (2008).

  41. 41.

    Takara Bio USA. SMART-Seq® v4 Ultra® Low Input RNA Kit for Sequencing User Manual (Takara Bio USA, 2017).

  42. 42.

    Dobin, A. et al. STAR: ultrafast universal RNA-seq aligner. Bioinformatics 29, 15–21 (2013).

  43. 43.

    Liao, Y., Smyth, G. K. & Shi, W. featureCounts: an efficient general purpose program for assigning sequence reads to genomic features. Bioinformatics 30, 923–930 (2014).

  44. 44.

    Xue, Z. et al. Genetic programs in human and mouse early embryos revealed by single-cell RNA sequencing. Nature 500, 593–597 (2013).

  45. 45.

    Horvath, S. & Dong, J. Geometric interpretation of gene coexpression network analysis. PLoS Comput. Biol. 4, e1000117 (2008).

  46. 46.

    Langfelder, P. & Horvath, S. WGCNA: an R package for weighted correlation network analysis. BMC Bioinformatics 9, 559 (2008).

Download references

Acknowledgements

We thank L. Attisano (University of Toronto) for critical comments. This work is supported by the Canadian Institutes of Health Research (CIHR) Foundation grant number FDN143252 (J.L.W.), the Terry Fox Research Institute (TFRI) New Frontiers program (J.L.W.), the University of Toronto’s Medicine by Design (MbD) initiative, which receives funding from the Canada First Research Excellence Fund (CFREF) (J.L.W., and a postdoctoral fellowship to A.A.), and a CIHR Research Project grant and a Cancer Research Society Operating Grant awarded to A.G. This research was enabled in part by support provided by the Niagara Cluster in Scinet (www.scinethpc.ca) and Compute Canada (www.computecanada.ca).

Reviewer information

Nature thanks Toshiro Sato, Fuchou Tang and the other anonymous reviewer(s) for their contribution to the peer review of this work.

Author information

A.G., J.L.W. and A.A. co-conceived the study; A.A. and D.T. designed and executed the scRNA-seq experiment. A.A. identified t-SNE clusters, performed pulse chase, tissue regeneration, smFISH, cell ablation, survival assays, immunofluorescence for lineage identification and flow cytometry assays; K.C. performed 10x genomics’ and Smart-Seq2 scRNA-seq protocols; B.G. created the bioinformatics workflow and performed all computational analysis and data visualization pertaining to scRNA-seq and Smart-Seq2 studies; A.G. generated Clu-cre-ERT2 transgenic mice and performed in vivo Yap1 knockout, Lats1/2 knockout and regeneration experiments; S.O. did smFISH in YAP1 mutants; J.G. performed cleaved caspase-3 and Edu staining and analysed YAP1 mutants; S.K. performed correlation, minimum spanning tree and pCreode analysis; B.S. ablated CLU+ cells in vivo, performed smFISH and measured alterations in crypts and colon, M.F. genotyped mice and performed functional characterization of revSCs; J.S. performed sequence alignment for individual Smart-Seq2 samples; S.B. assisted in development of single-cell profiling strategies; A.A., S.K., J.L.W. and A.G. wrote the manuscript.

Competing interests

The authors declare no competing interests.

Correspondence to Jeffrey L. Wrana or Alex Gregorieff.

Extended data figures and tables

  1. Extended Data Fig. 1 Analysis of scRNA-seq data.

    a, Summary of bioinformatics workflow. b, Cell doublets are overlaid on t-SNE graphs using DoubletFinder. n = 8,710 and n = 6,644 single-cell transcriptomes for whole epithelium (samples 04 and 06) and enriched crypts (samples 05 and 07), respectively; see Methods. c, d, Unsupervised clustering of scRNA-seq data from whole epithelia (c; n = 8,710 single-cell transcriptomes) and isolated crypts (d; n = 6,644 single-cell transcriptomes). Unsupervised clusters are overlaid on the t-SNE map and are indicated by different colours and labelled according to their identity, on the basis of expression of cell-type-specific marker-gene expression. Cells from untreated controls (blue dots) or irradiated intestines (red dots) are also plotted as indicated. In c, median expression of epithelial cell markers (bottom left) and lymphocyte markers (bottom middle) are also shown. Heat maps (right) showing expression of indicated marker genes (c (right), d (right)) that identify distinct cell-types are shown. Note that the cluster 18 in d, highlighted by the question mark, does not express any lineage-specific markers. Gene-expression values are shown as median of log10(counts per million (CPM)). IM, immune cells.

  2. Extended Data Fig. 2 Global changes in the cellular composition of irradiated intestinal epithelium.

    a, Irradiation-induced changes in the cellular composition of the indicated intestinal epithelial cell types are plotted as per cent of untreated controls. b, Frequency plot of Lgr5 expression levels in individual cells of the indicated samples. Note that radiation depletes Lgr5high cells in both whole epithelia and enriched crypt samples. c, Representative images of untreated and irradiated small intestines stained for CD45 (red), EPCAM (green), and counterstained for DAPI (blue). Note that CD45+ immune cells infiltrate crypts upon irradiation (n = 3 mice each). d, Number of cells in the EC and CBC clusters from untreated (blue) or irradiated (red) intestines. e, Regional distribution of ECs and CBCs. scRNA-seq data from the indicated cell clusters were compared to zonation markers and transcriptional overlap between marker genes of the indicated individual CBC and EC clusters, and zone-specific gene signatures (see Methods) plotted as percentage. f, Regional distribution of intestinal epithelial cells. Number of CBCs (red) and all ECs (blue) that display the indicated zonation markers18 without (darker) or with (lighter) irradiation. Scale bar, 75 μm. Source data

  3. Extended Data Fig. 3 Identification of a cell type generated in the irradiated intestinal epithelium.

    a, Analysis of SSC1 and SSC2 showing expression of proliferation markers (left two panels), and the CBC signature (third from left), and individual CBC markers are overlaid on the crypt t-SNE graph (right three panels). n = 6,644 single-cell transcriptomes. b, Expression of indicated genes is plotted on a t-SNE graph of enriched crypts (n = 6,644 single-cell transcriptomes). c, Expression of Ly6a (also known as Sca-1) is shown in normal (top) and irradiated (bottom) cells of various enriched crypt clusters (n = 6,644 single-cell transcriptomes). d, Expression of indicated cell-type-specific genes is plotted across all enriched crypt clusters. Note that none of the differentiated lineage markers are enriched in SSC2. pct.exp, percentage of cells that express the gene; avg.exp.scale, scaled average expression of the gene; Pro., proliferating cells. e, Heat map showing differentially expressed genes in SSC2a–c. f, Expression of proliferation markers is overlaid on the SSC2 t-SNE map (n = 313 single-cell transcriptomes). g, Overlap between the top-50 SSC1 genes with SSC2a versus SSC2c is shown with a list of genes common to SSC1 and SSC2a. h, Analysis of zonation marker expression (as in Extended Data Fig. 2e). Note that SSC2a contains almost all the proliferative cells, overlaps extensively with SSC1 and CBC markers, and expresses crypt-specific genes. i, Expression of indicated genes identifying SSC2b, is overlaid on the crypt t-SNE plot (n = 6,644 single-cell transcriptomes). j, Clu expression is overlaid on the crypt t-SNE map (n = 6,644 single-cell transcriptomes). k, Dot plot showing scaled expression of Clu, Lgr5 and Ascl2 in the indicated SSC2 populations. Gene-expression values are shown as absolute (a, b, c, j, for individual genes) or median (a, f, i, for multiple genes) of log10(CPM).

  4. Extended Data Fig. 4 CLU+ cells are distinct from CBCs.

    a, SSC2 t-SNE plots are overlaid with expression of the indicated genes (n = 313 single-cell transcriptomes). Cluhigh cells are circled in each map (black circles) and cells co-expressing either Lgr5 (middle) or Ascl2 (right) are shown in green. No CLU+ cells co-expressing Lgr5 were observed, and CLU+ cells co-expressing Ascl2 were rare (green circles with green arrows, right). Gene-expression values of the indicated genes are shown as log10(CPM). b, Representative images and close-up views (yellow box and bottom panels) are shown of smFISH analyses of endogenous Clu (green) and Lgr5 (red) mRNA expression in the small intestine, counterstained with DAPI (blue) at the indicated times after irradiation. Note that CLU+ cells (green arrowheads) are distinct from LGR5+ cells (red arrowheads) with occasional examples of CLU+ cells at 3 dpi that contain Lgr5 transcripts highlighted with yellow arrowheads (n = 2 mice each). Analyses of non-irradiated intestines is shown in Fig. 2a. c, Endogenous Clu mRNA identified by smFISH (yellow) followed by BAC-Clu-GFP transgene visualization by immunostaining (green) is shown in irradiated small intestine. White arrowheads indicate cells co-expressing Clu and GFP (n = 1 mouse). d, Immunohistochemistry analysis of Clu-GFP+ cells in small intestine of untreated (0 Gy) or irradiated (2–4 dpi) mice (n = 2 mice each). e, Either Lgr5-GFP+ or Clu-GFP+ cells within the EPCAM+ epithelial cell fraction of small intestine were quantified by flow cytometry (see Extended Data Fig. 5b) and are plotted as percentage at the indicated dpi (n ≥ 2 mice each). f, Immunostaining of Clu-GFP (green, yellow arrowheads) and OLFM4 (red, white arrowheads) expression in the crypt regions of small intestine. (n = 2 mice). Scale bars, 35 μm (b); 25 μm (c); 40 μm (d); 25 μm (f). Source data

  5. Extended Data Fig. 5 CLU+ cells are transiently generated following irradiation.

    a, Plots of Clu-GFP+ and Lgr5-GFP+ quantification by FACS of single cells isolated from whole epithelium of the indicated mice analysed at different times after irradiation (n = 2 independent experiments with similar results). b, Gating strategy used to first select EPCAM+ epithelial cells and calculate percentage of Clu-GFP+ and Lgr5-GFP+ population within EPCAM+ epithelial cell fraction shown in Extended Data Fig. 4e. c, Gating strategy used to select healthy single cells for the above flow cytometry analyses. (related to a and b; n ≥ 3 mice each). d, Gating strategy used for flow cytometry-based purification of GFP+ and GFP cells collected from enriched crypts of irradiated BAC-Clu-GFP mice for the Smart-Seq2 experiment shown in e−g (n = 2 mice). e, Violin plot of Clu expression, which represents distribution of expression intensity of individual Clu mRNA values in single GFP+ and GFP epithelial cells purified in d and profiled by Smart-Seq2. f, t-SNE map of individual transcriptomes of Clu-GFP+ and Clu-GFP (red and blue dots, respectively, in middle panel) cells purified as in d. Two distinct clusters identified by unsupervised analysis are marked (left, orange and light blue), and presence of cells in GFP+ versus GFP populations are marked (middle). Expression of the indicated SSC2c signature is overlaid on the t-SNE plot (right). Gene-expression values (right) are shown as median of log10(CPM). g, Dot plot showing scaled expression of indicated SSC2c, proliferation, SSC2a, CBC and differentiated lineage markers in the clusters identified in f.

  6. Extended Data Fig. 6 CLU+ cells are produced in a YAP1-dependent manner and are quiescent.

    a, Expression of the indicated YAP1-signature genes is overlaid on the crypt t-SNE plot. Gene expression is shown as median of log10(CPM). n = 6,644 single-cell transcriptomes. b, Endogenous Clu mRNA identified by smFISH (red) followed by visualization of E-cadherin-expressing epithelial cells by immunostaining (green) is shown in small intestine of wild-type or Yap1−/− mice at 3 dpi (n = 3 mice). c, Representative immunohistochemistry images of Clu-GFP transgene expression in 2 dpi wild-type or Yap1−/− small intestines, or in untreated Lats1−/−Lats2−/− (Lats1/2 KO) mutant small intestines (n = 2 mice). d, Analysis of proliferation in CLU+ cells. Representative images (left) of Clu-GFP expression (green) and EdU incorporation (red) at 3 dpi. Green arrowheads indicate GFP+EdU cells, red arrowheads indicate GFPEdU+ cells and yellow arrowheads indicate GFP+EdU+ cells (n = 65 images analysed). Pie graphs (right) show quantification of EdU+GFP cells (top) and EdU+ cells in the GFP+ population (bottom). Note that GFP+ cells rarely display EdU incorporation. Scale bars, 35 μm (bd).

  7. Extended Data Fig. 7 CLU+ cells contribute to intestinal regeneration.

    a, Schematic of strategy used to generate Clucre-ERT2 knockin mouse line (top; also see Methods). Clucre-ERT2/+;Rosa26lsl-tdTomato/+ mice (ClutdTomato) treated with or without TAM, and analysed at 2 and 6 dpi, as indicated, were stained using smFISH to simultaneously detect tdTomato and endogenous Clu mRNA transcripts (bottom left), or analysed by immunofluorescence at 6 dpi in non-TAM treated mice (bottom right). Note that tdTomato and Clu strongly overlap in TAM-treated mice whereas no tdTomato is detected without TAM (n = 2 mice). b, Representative images showing tdTomato expression detected in the indicated intestinal epithelia of ClutdTomato reporter mice in the presence or absence of irradiation (n = 2 independent experiments with similar results), that were treated with TAM at 1 dpi and analysed at 6 dpi. Scale bars, 35 μm (a); 70 μm (b).

  8. Extended Data Fig. 8 CLU+ cells are multipotent.

    a, Representative images showing examples of single-cell tdTomato+ clones produced at either the +3 (left; note LYZ1+ PCs in crypt base) or +4 (top middle) positions relative to the crypt base, or at the crypt–villus junction (bottom middle) one day after a single TAM injection. Position of LGR5+ CBCs is shown using non-TAM injected LGR5–GFP reporter mice (top left). Dotted lines indicate crypt boundaries (n = 3 mice). b, Images showing tdTomato+ clone size in the crypts of ClutdTomato mice analysed at the indicated time points (n ≥ 3 mice). c, Small intestinal crypts imaged for tdTomato+ clones (red) and Lgr5-GFP+ cells (green), are shown at different chase times, as indicated. White arrows indicate tdTomato+ cells that do not express GFP, whereas yellow arrows identify cells co-expressing tdTomato and GFP (n = 2 mice). d, Small intestines at the indicated chase times were stained and imaged for tdTomato and the indicated differentiated lineage marker: goblet cell, tuft cell, enteroendocrine cell (top; note white arrows that show lineage cells do not express tdTomato at 3 dpt and 45 dpt), or PC (bottom; note white arrows that indicate clonal tdTomato+ cells that do not express PC marker LYZ1 and yellow arrows identify cells co-expressing tdTomato and LYZ1) (n = 2 mice each, experiment repeated once revealing similar results). e, tdTomato+ ribbons (red) at 60 and 148 dpt were co-stained for markers (yellow) identifying the indicated differentiated cells. Yellow arrows indicate co-expressing cells and white arrows identify LYZ1+ PCs that do not express tdTomato (n = 2, mice). f, Schematic of the temporal hierarchy of lineage emergence within tdTomato+ clones. Scale bars, 35 μm (ac); 30 μm (d, e).

  9. Extended Data Fig. 9 Differentiated progeny transition through CLU+ cells to produce de novo LGR5+ CBCs.

    a, Gene correlation matrix for the indicated genes was calculated from the crypt scRNA-seq data (left) and used to generate a GCN (middle) with genes as nodes (labelled and coloured according to legend on the left) and correlations as edge colours (scale). The minimum spanning tree of the GCN is shown on the right. b, Illustration of experimental strategy and interpretation. (see Methods for detailed discussion). c, Related to b (subpanel (ii); top right), representative images of small intestines treated as indicated and co-stained for tdTomato (red) and GFP (green) (left: yellow arrows show crypts expressing Lgr5-GFP and tdTomato, and white arrows show crypts expressing Lgr5-GFP but not expressing tdTomato). Right, quantification of green and yellow crypt frequency (n = 32 crypts across two mice each, two-tailed Mann–Whitney test; data are mean ± s.e.m., two independent experiments performed with similar results). d, Related to b (subpanel (iii); bottom right), representative images of small intestines co-stained for tdTomato (red) and GFP (green) and labelled as in c. Flow cytometry plots to quantify co-expression are shown on the right. Note that tracing LGR5+ cells and their offspring at 7 and 6 dpt shows 98% of Lgr5-GFP+ cells also express tdTomato. Each data point represents a mouse; n = 18 mice analysed; two-tailed Mann–Whitney test; box plot shows the median, box edges represent the first and third quartiles, and the whiskers show minimum and maximum values; pooled data from two biologically independent experiments with similar results are shown. e, Representative images of Lgr5GFP-iDTR intestines stained for Lgr5-GFP (green), cleaved-caspase-3 (red) and DAPI (blue), 6 h after DT treatment, as indicated. White arrows show dying cells losing GFP expression and gaining cleaved caspase-3 activity (n = 2 mice). f, Experimental scheme for single versus continuous ablation (left). Cells expressing tdTomato (white arrowheads) or endogenous Clu (yellow arrows) were identified using smFISH (RNAScope) (n = 3 mice). Scale bars, 70 μm (c, d); 30 μm (e, f). Source data

  10. Extended Data Fig. 10 YAP1 and CLU+ cells are required for efficient regeneration after injury.

    a, Wild-type or Clucre-ERT2/+;Rosa26lsl-DTA/+ (Cluless) mice were irradiated and analysed at 3 and 5 dpi. Small intestinal crypts were stained for the proliferation marker KI67 (green), and counterstained with EPCAM (red) and DAPI (blue). Nuclear KI67 staining, crypt numbers and crypt length were quantified (n = 49 images analysed; two-tailed Mann–Whitney test; box edges show 25th and 75th percentile, the central point is the median and whiskers represent minimum and maximum values. n.s., non-significant (P > 0.05); *P < 0.05, ***P < 0.001. b, Representative images of intestinal epithelia from wild-type or YAP1-mutant mice at 3 dpi visualized with EPCAM (red) and DAPI (blue) staining (left). Crypts per unit length of small intestine for the indicated mice and conditions are quantified (right; n = 16 images analysed across two mice each; two-tailed Mann–Whitney test; box edges show 25th and 75th percentile, the central point is the median and whiskers represent minimum and maximum values; the experiment was repeated once with similar results). c, Colons from wild-type and YAP1 mutants at 3 dpi or untreated (normal) were analysed by immunohistochemistry (representative images; n = 2 mice each). d, Images of non-irradiated, TAM-treated (10 days) wild-type and Cluless small intestines and colons stained for epithelial marker EPCAM (n = 2 mice each). e, Colons dissected from non-irradiated, wild-type or Cluless mice (compare with Fig. 4b) (n = 2 mice each). f, CLU+ revSCs are required for recovery from DSS. Mice of the indicated genotypes were treated with TAM in the presence or absence of DSS, as indicated (schematic, top). Body weight was measured daily and is plotted as change relative to starting weight (percentage ± s.d., n = 62 total mice tested). g, Kaplan–Meier survival curves for the indicated cohort of mice that were either exposed to DSS (2.5%) treatment or left untreated (number of mice of the corresponding genotype is indicated; two-sided log-rank test was applied to compare survival curves). h, Average crypt length in the small intestine is quantified (n = 22 crypts, two-tailed Mann–Whitney Test, box edges show 25th and 75th percentile, central point is median and whiskers represent minimum and maximum values). i, Representative images are shown of small intestinal and colonic tissues in the absence or presence of DSS treatment stained for endogenous Clu mRNA (red; white arrows) using smFISH (RNAScope), (n = 2, mice each). j, Representative images showing that non-DSS treated control mice that received nine TAM injections developed rare ribbon formation in the small intestine and colon. Quantification of tdTomato ribbons formed in the presence of the indicated TAM treatment either without (blue), or with DSS (red) treatment is plotted as percent of crypt–villus axes (n = 10 images each across two treated mice each; two-tailed Mann–Whitney Test; box edges show 25th and 75th percentile, central point is median and error bars represent minimum and maximum values, experiment was repeated once with similar results). Scale bars, 120 μm (b, left, c); 70 μm (d); 1.5 cm (e); 30 μm (i); 75 μm (j). Source data

Supplementary information

  1. Reporting Summary

  2. Supplementary Table Supplementary Table S1: Analyses of raw scRNA-seq data. Detailed information of initial and final cell count, mapping rate, type of filters applied, number of genes per cell and UMI per cell are shown for individual samples.

  3. Supplementary Table Supplementary Table S2: Clusters of whole epithelium. This table contains lists of genes differentially expressed in individual clusters identified in the whole epithelium using two-sided Wilcoxon Rank test, where cells in individual clusters (Clusters #1, n=594; Clusters #2, n=163; Clusters #3, n=304; Clusters #4, n=674; Clusters #5, n=855; Clusters #6, n=205; Clusters #7, n=290; Clusters #8, n=550; Clusters #9, n=845; Clusters #10, n=1662; Clusters #11, n=298; Clusters #12, n=833; Clusters #13, n=105; Clusters #14, n=401; Clusters #15, n=203; Clusters #16, n=135; Clusters #17, n=371; Clusters #18, n=151 and Clusters #19, n=71 single cell transcriptomes) were compared with total cells (n=8710 single cell transcriptomes) present in the t-SNE map. Adjusted p-values are calculated based on the FDR (false discovery rate) correction.

  4. Supplementary Table Supplementary Table S3: Clusters of enriched crypts. This table contains lists of genes differentially expressed in individual clusters identified in the enriched crypts samples using two-sided Wilcoxon Rank test, where cells in individual clusters (Clusters #1, n=519; Clusters #2, n=150; Clusters #3, n=519; Clusters #4, n=597; Clusters #5, n=88; Clusters #6, n=1037; Clusters #7, n=280; Clusters #8, n=405; Clusters #9, n=80; Clusters #10, n=70; Clusters #11, n=250; Clusters #12, n=193; Clusters #13, n=496; Clusters #14, n=354; Clusters #15, n=324; Clusters #16, n=238; Clusters #17, n=686; Clusters #18, n=313 and Clusters #19, n=45 single cell transcriptomes) were compared with total cells (n=6644 single cell transcriptomes) present in the t-SNE map. Adjusted p-values are calculated based on the FDR (false discovery rate) correction.

  5. Supplementary Table Supplementary Table S4: Overlap with zonation markers. Complete list of genes that overlap with individual clusters is provided.

  6. Supplementary Table Supplementary Table S5: Subclusters of SSC2. This table contains lists of genes differentially expressed in individual subclusters identified in SSC2 using two-sided Wilcoxon Rank test, where cells in individual clusters (SSC2a, n=117; SSC2b, n=40 and SSC2c, n=156 single cell transcriptomes) were compared with total cells (n=133 single cell transcriptomes) present in the t-SNE map. Adjusted p-values are calculated based on the FDR (false discovery rate) correction.

  7. Supplementary Table Supplementary Table S6: Smart-seq2 reads and mapping rate. This table provides a complete list of total number of reads and mapping rate of individual cells sequenced in SMART-seq2 experiment.

  8. Supplementary Table Supplementary Table S7: Smart-seq2 clusters. This table contains lists of genes differentially expressed in individual clusters identified in Smart-seq2 experiment using two-sided Wilcoxon Rank test, where cells in individual clusters (Cluster 1 (revSCs), n=59 and Cluster 2, n=107 single cell transcriptomes) were compared with total cells (n=166 single cell transcriptomes) present in the t-SNE map. Adjusted p-values are calculated based on the FDR (false discovery rate) correction.

  9. Supplementary Table Supplementary Table S8: Markers used in trajectory analysis. This table contains list of genes that mark individual cell populations used in trajectory analysis.

Source data

  1. Source Data Fig. 2

  2. Source Data Fig. 3

  3. Source Data Extended Data Fig. 2

  4. Source Data Extended Data Fig. 4

  5. Source Data Extended Data Fig. 9

  6. Source Data Extended Data Fig. 10

Rights and permissions

To obtain permission to re-use content from this article visit RightsLink.

About this article

Publication history

  • Received

  • Accepted

  • Published

  • Issue Date

DOI

https://doi.org/10.1038/s41586-019-1154-y

Fig. 1: Irradiation induces a unique cell type in the intestinal epithelium.
Fig. 2: CLU+ cells are rare but multipotent under homoeostasis.
Fig. 3: CLU+ cells reconstitute damaged crypts.
Fig. 4: CLU+ revSCs are required for intestinal regeneration.
Extended Data Fig. 1: Analysis of scRNA-seq data.
Extended Data Fig. 2: Global changes in the cellular composition of irradiated intestinal epithelium.
Extended Data Fig. 3: Identification of a cell type generated in the irradiated intestinal epithelium.
Extended Data Fig. 4: CLU+ cells are distinct from CBCs.
Extended Data Fig. 5: CLU+ cells are transiently generated following irradiation.
Extended Data Fig. 6: CLU+ cells are produced in a YAP1-dependent manner and are quiescent.
Extended Data Fig. 7: CLU+ cells contribute to intestinal regeneration.
Extended Data Fig. 8: CLU+ cells are multipotent.
Extended Data Fig. 9: Differentiated progeny transition through CLU+ cells to produce de novo LGR5+ CBCs.
Extended Data Fig. 10: YAP1 and CLU+ cells are required for efficient regeneration after injury.

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.