A biomechanical switch regulates the transition towards homeostasis in oesophageal epithelium

Abstract

Epithelial cells rapidly adapt their behaviour in response to increasing tissue demands. However, the processes that finely control these cell decisions remain largely unknown. The postnatal period covering the transition between early tissue expansion and the establishment of adult homeostasis provides a convenient model with which to explore this question. Here, we demonstrate that the onset of homeostasis in the epithelium of the mouse oesophagus is guided by the progressive build-up of mechanical strain at the organ level. Single-cell RNA sequencing and whole-organ stretching experiments revealed that the mechanical stress experienced by the growing oesophagus triggers the emergence of a bright Krüppel-like factor 4 (KLF4) committed basal population, which balances cell proliferation and marks the transition towards homeostasis in a yes-associated protein (YAP)-dependent manner. Our results point to a simple mechanism whereby mechanical changes experienced at the whole-tissue level are integrated with those sensed at the cellular level to control epithelial cell fate.

Fig. 1. The postnatal mouse oesophagus as a model of rapid tissue expansion transitioning towards homeostasis.

a, H&E section of adult oesophagus. Inset showing tissue layers. Epi, epithelium. b, Images of whole oesophagus throughout postnatal development. c, Tissue area after longitudinally opening the oesophageal tube. n=103 mice; Millimeters, mm. d, Whole body growth in length excluding tail. n=26 mice; Centimeter, cm. e, In vivo protocol. Mice were treated with a single EdU injection 24h prior tissue collection at the time points indicated. f, Basal confocal view of EdU+ cells in typical OE wholemounts from (e). g, Quantification of EdU+ basal cells per field (left), and relative to the number of DAPI+ basal cells (right) from (e). Presented as mean ± SD. Points show individual measurements, greyscale indicates values from each of 3 mice. n=3 mice. h, Fraction of cell fate imbalance, showing the degree to which basal cell fate is biased towards duplication over differentiation and loss (left axis; see Methods and model in Fig. 2n). Total cell production throughout postnatal development (right axis; see Methods). Over time OE cells tilt towards balanced cell fate, in line with model of oesophageal homeostasis by Doupe et. al. (inset; P, progenitor cell; D, differentiating cell). n=3 mice. See Supplementary Note 1. i, Representative 3D rendered confocal z-stacks showing side views of OE wholemounts throughout postnatal development. j, Epithelial thickness, including enucleated terminally differentiated layers. n=3 mice; Micrometre, μm. All data derived from wild-type C57BL/6J mice, expressed as mean values ± SEM, unless otherwise stated. Data analysis was performed using one-way ANOVA with Tukey’s multiple comparisons test (^#p relative to P70; *p relative to P7). Scale bars. 1a(200 μm), inset(100 μm); 1b(2mm); 1f,i(20μm). Stainings. Blue, 4’,6-diamidino-2-phenylindole (DAPI); cyan, EdU; green, basal marker KRT14; greyscale, differentiation marker KRT4. Dashed lines indicate basement membrane. Dotted lines in graphs indicate P28. Orange diamonds depict the longitudinal orientation of the oesophagus where indicated (outlined in Extended Data Fig. 1a). Parts of (e) were drawn by using and/or adapting diagrams from Servier Medical Art. Source data are provided. See also Extended Data Fig. 1.

Fig. 2. KLF4 marks the emergence of an early committed population in the OE.

a, Confocal z-stack side views of typical OE wholemounts. White arrow, basal KLF4+ cells; Dashed lines, basement membrane; Dotted lines, upper limit of the OE. b, Representative basal views of OE wholemounts from (a). c, 3D confocal side views of basal insets from adult OE show KLF4+ cells detaching from basement membrane (dotted line) and transitioning from basal to suprabasal layers. Basement membrane (dotted line). d, Illustration of transitioning KLF4+ (red) cells as shown in (c). e, In vivo protocol. 2h EdU chase at indicated time points. f, Basal EdU and KLF4 intensity quantification (a.u.) from (e) (see Methods). n=3 mice. g, Percentage of KLF4+ basal cells expressed as mean values ± SD. One-way ANOVA with Tukey’s multiple comparisons test (*p, relative to P7). Points show individual measurements, greyscale indicates values from different mice, n=3 mice. h, Basal views of adult OE from (e). Dotted lines, EdU+ basal cells. i, Basal views of adult OE wholemounts showing colocalization of KLF4 with basal (KRT14, green) and suprabasal (KRT4, greyscale) markers. White arrows, cell of interest. j, Side view illustration depicting KLF4+ (red), KRT14 (green) and KRT4 (grey) basal colocalization combinations as shown in (i). k, Percentage of KRT4+ cells within KLF4+/- basal population (mean ± SEM, Two-way ANOVA with Tukey’s multiple comparisons test, ns, not significant, n=3 mice). l, In vivo protocol. Adult mice were administered a single dose of EdU at 6pm, and sampled after 12 hours to identify post-mitotic EdU+ basal pairs. m, Representative basal views of adult OE from (l). Dotted lines, EdU+ pairs. n, Schematic showing quantitative data on KLF4 expression in EdU+ pairs from (l). One-way ANOVA, with Tukey’s multiple comparisons test; n=3 mice. Estimates from oesophageal homeostasis model by Doupe et. al. shown between parentheses. See Supplementary Note 1. o, Basal confocal views of typical adult OE wholemounts from (m). Dotted lines, representative pairs. Scale bars. 2a,b,h, and m(20 μm); 2c,i, and o(5 μm). Orange diamonds, longitudinal orientation of the oesophagus (Extended Data Fig. 1a). Parts of (e, l) were drawn by using and/or adapting diagrams from Servier Medical Art. Source data are provided. See also Extended Data Fig. 2.

Fig. 3. Single-cell transcriptional profiling defines transition towards homeostasis.

a, Schematic of single-cell RNA-seq data generation from OE sorted cells (EpCam⁺/Cd45^-) using 10X Genomics platform. Cells were isolated form 15 mice (P7), 12 mice (P28), and 9 (adults); 3 libraries (x10, Chromium) for P7 and P28, and 6 libraries for Adult. b, Cell distribution in the dimension reduction space UMAP at different postnatal time points. c, UMAPs representing annotated cell types. CB, cycling basal; RB, resting basal; D, differentiated. d, UMAPs representing resting basal cell clusters from Extended Data Fig. 3b. e, Fraction of resting basal cell clusters per time point. Fraction, expressed as mean values ± SEM. n=3 for P7 and P28 and n=6 for Adult. f, Heatmap representing expression of individual genes belonging to distinctive patterns of gene expression in resting basal cells as defined in (c) and (d), number of genes provided in brackets. For expression values, log2-transformed normalized UMIs were scaled and averaged across all cells belonging to each group of resting basal cell cluster(s). Scale bar denotes expression range. Inset shows the expression of selected genes related to KLF4, YAP and response to mechanical stimuli. The colours of gene names denotes the relevancy to KLF4, YAP and response to mechanical stimuli. g, Violin plots showing the expression distribution from resting basal through to differentiated clusters of selected genes in (f) related to KLF4 and tissue mechanics. The expression is log2-transformed normalized UMIs. Dotted lines, mean. h, Expression of relevant genes along the pseudotime trajectory from resting basal to differentiated cells for P28 and Adult. Left panel, YAP target genes (Cyr61, Ctgf, Thbs1) and genes associated with a response to mechanical stimuli (Cav1, Klf2, Dcn). Right panel, depicts KLF4 target genes (Krt4, Krt13, Cdnk1a, Cebpb). Gene expression is represented as auto-scaled, log2-transformed normalized UMIs smoothed using a rolling mean along its trajectory with a window size of 5% of cells. The three bars on the top denote the arrangement of cells according to real time points, pseudotime and clusters in Extended Data Fig. 3b, respectively. C, Cluster. Parts of (a) were drawn by using and/or adapting diagrams from Servier Medical Art. See also Extended Data Fig. 3,4.

Fig. 4. Homeostatic transition coincides with tissue spatial reorganisation.

a, Typical basal views of confocal z-stacks showing changes in cell shape, alignment and density in OE wholemounts. Blue, DAPI; green, KRT14; scale bar 20 μm. b, Schematic exemplifying changes seen in (a). c, Quantification of basal cell density. Expressed as mean values ± SEM; n=3 mice. d, Cell shape anisotropy tensor represented as an ellipse calculated from the nuclear centroid position of each basal cell. Long axis of each ellipse is proportional to the dominant eigen value of the tensor. Orientation is colour-coded. Results from representative experiment are shown; n=3 mice. e, Bidimensional structure factor quantifying basal cell spatial organisation. Changes in the dashed white outline (from ellipse to circle) depict a transition from anisotropic to isotropic spatial cell distribution over time; n=3 mice. f, Nematic order parameter indicative of the orientational order of cells in the tissue. Box-and-whisker plots: Box plots show median and quartiles; and whiskers, minima/maxima. n=3 mice Nuclei at the edge of the image frame were discarded in (d) and (e) to avoid confounding effects from partially captured cells. All data derived from wild-type C57BL/6J mice. Data analysis was performed using one-way ANOVA with Tukey’s multiple comparisons test (*p, relative to P7; ns, not significant). Dashed line in graphs indicates P28. Orange diamonds depict the longitudinal orientation of the oesophagus where indicated (outlined in Extended Data Fig. 1a). Source data are provided. See also Supplementary Note 1 and Extended Data Fig. 5.

Fig. 5. Differential growth generates longitudinal tissue strain sensed at the cellular level.

a, Method for oesophageal length measurements in situ and ex vivo (l). b, In situ and immediate ex vivo images of oesophageal tubes captured in P7 and adult mice. Dashed lines delineate oesophageal tube. c, Percentage increase in oesophageal length (in situ and ex vivo) compared to body length (excluding tail) throughout postnatal development (P7 set to 100). Dashed line indicates P28. Data expressed as mean ± SEM. Each time point was analysed using two tailed unpaired t-test (ns, not significant; grey and blue indicate difference between body and ex vivo, body and in vivo, respectively.); n=6 mice. d, Longitudinal tissue strain in vivo represented as percentage. Data expressed as mean ± SEM; n=6 mice. e and f, Schematic representing changes in F-actin (grey) and pMLC2 (green) levels in OE cells throughout postnatal development. g, Basal views of typical OE wholemount showing Phalloidin staining for F-actin at indicated time points. Blue, DAPI; greyscale, Phalloidin. h, Representative side views of tissue sections showing pMLC2 staining at indicated time points. Blue, DAPI; green, pMLC2; Dotted lines, basement membrane; ep, epithelium; strm, stroma. i and j, Quantification of basal F-actin and pMLC2 staining, respectively (see Methods). Box plots show median and quartiles; and whiskers, minima/maxima; (i) n=3 animals, (j) n=4 mice. k, Schematic representation of changes in YAP (green) localization during postnatal development. l, Basal view of representative OE wholemounts showing progressive translocation of YAP to the nucleus as tissue matures. Blue, DAPI; green, YAP; greyscale, B-Catenin (BCat). m, Quantification of basal nuclear and cytoplasmic YAP staining (see Methods). A total of 60 cells per time point were measured from 3 animals. Localization pattern shows the average intensity for one representative animal per time point. Full quantification in Extended Data Fig. 6 j, k. Scale bars. 5b(1 cm); 5g,l(20 μm); 5h (10 μm). Data analysis for d, i, and j was performed using one-way ANOVA with Tukey’s multiple comparisons test (^*p; relative to P7; ns, not significant). Orange diamonds, longitudinal orientation of the oesophagus (Extended Data Fig. 1a). Parts of (a) were drawn by using and/or adapting diagrams from Servier Medical Art. Source data are provided. See also Supplementary Note 1 and Extended Data Fig. 6.

Fig. 6. YAP nuclear localization promotes increased levels of basal commitment through KLF4.

a, Confocal basal views of typical adult OE from C57BL/6J WT mice. Yellow arrows indicate colocalization of KLF4 (high) and YAP (high) staining. b, Correlation plot for KLF4 and YAP intensities in the basal layer the adult OE (a). n=3 mice. (r, Pearson correlation coefficient) c, Quantification of KLF4 intensity (a.u) within the lowest and highest YAP expressing cells (0-20^th and 80-100^th percentile respectively). n=296 cells per percentile from 3 mice. d, In vivo protocol. Adult Krt5-rtTA;TEADi mice were doxycycline (DOX) treated for 3 weeks to block YAP and TAZ-induced TEAD activity in basal cells. TEADi expressing cells are marked by GFP expression (TEADi⁺). e, Representative basal views of OE wholemounts from (d). TEADi expressing cells are marked by GFP expression. Dotted circles indicate GFP⁺ cells. f, Percentage of KLF4+/- and Ki67+/- cells in TEADi+/- basal cells from (d, as marked by GFP). A total of 800 TEADi+/-basal cells were quantified from 72 fields out of 2 mice. Analysed by one sided Chi-squared test. g, In vivo protocol. P6 rtTA/tetOYAP pups were doxycycline treated for 3 days to induce an active form of YAP (S127A), and culled at P9. h, Staining in OCT embedded OE cryosections (10 μm thick) of YAP activated mice from (g). Nuclear KLF4 expression colocalizes with YAP+ cells. Dotted white lines, YAP+ cells. i, KLF4 intensity quantification (a.u) in control and YAP overexpressing mice as shown in (j). n=1362-1779 cells from 3 animals. j, Representative basal views of OE wholemounts from (g). k, Representative 3D rendered side views of OE wholemounts from (g), showing expanded KLF4 suprabasal compartment in rtTA/tetOYAP mice compared to controls. Scale bars. 6a,e and k(10 μm), 6h,j(20μm). Stainings. Blue, DAPI. 6A, 6H, 6J, 6K (green, YAP; red, KLF4; greyscale, BCat). 6A (green, TEADi; red, KLF4; greyscale, Ki67) Dotted line, basement membrane (h, k). Violin plots (c, i) show median (solid line) and quartiles (dotted lines). Data from (c) and (i) were analysed using two-tailed unpaired t test. Orange diamonds depict the longitudinal orientation of the oesophagus where indicated (outlined in Extended Data Fig. 1a). Parts of (d, g) were drawn by using and/or adapting diagrams from Servier Medical Art. Source data are provided. See also Supplementary Note 1.

Fig. 7. Changes in tissue mechanics influence basal KLF4 expression.

a, In vitro protocol. Oesophageal tubes were exposed to a 40% stretch using 3D-printed stretcher and kept in vitro as whole-organ cultures for 48h. b, 3D model of stretching device (left), in use (right). c, Confocal basal views of stretched and unstreched control samples after 48h. d, Basal quantification of KLF4+ cells expressed as percentage of DAPI+ basal cells from (c). P7 and P28 n=5 mice, Adult n=6 mice. e, Threshold response. Basal quantification of KLF4+ cells in P7 whole-organ cultures after exposing to 20%, 40% or 100% longitudinal stretch, or unstreched control conditions for 24h (as per a). n=3 mice. f, In vitro protocol. Adult epithelial-stromal composites were kept in vitro and treated with 25μM blebbistatin (BLEBB) for 48h and EdU for the final 2h. g, Confocal basal views after a 48 hour BLEBB treatment in vitro (f). h, Percentage of KLF4+ basal cells relative to DAPI+ basal cells from (f). n=3 mice. i, In vitro protocol. Oesophagi were exposed to 40% stretch and treated with DMSO/Verteporfin (VP; 1 μM) for 24 hours. j, Confocal basal views after 24 hour VP treatment in vitro (i). k, KLF4 intensity quantification (a.u) in OE basal cells from (i). A total of n=794-1398 cells from 3 animals per condition. l, Suggested model. The physiological stretching experienced by the postnatal oesophagus promotes the expression of KLF4 in the basal progenitor cell compartment in a YAP dependent manner. The onset of basal KLF4 expression marks the emergence of an early committed population, which balances proliferation, and defines the transition towards adult homeostasis. Scale bars. 7b(1cm), inset (5mm) c,g,j(20μm). Stainings. Blue, DAPI; red, KLF4; greyscale, Ecad. Data expressed as mean±SD. Analysis for (e) and (k) one-way ANOVA with Tukey’s multiple comparisons test (ns, not significant). (e) relative to Control. Analysis for (d) and (g) was performed using two-tailed unpaired t test (ns, not significant). Individual points show individual measurements, greyscale indicates values from each of 3 mice. Orange diamonds, longitudinal orientation of the oesophagus (Extended Data Fig. 1a). Parts of (a, f, i, l) were drawn by using and/or adapting diagrams from Servier Medical Art. Source data are provided. See also Extended Data Fig. 7.

Extended Data Fig. 1. Postnatal characterisation. Related to Fig. 1.

a, Diagram illustrating longitudinal oesophageal orientation from proximal to distal as marked by orange diamond. b, and c, Oesophageal tissue growth in length and width over time, respectively. Data expressed as mean ± SEM and analysed using one-way ANOVA with Tukey’s multiple comparisons test (n = 103 mice; ^#p relative to P70; *p relative to P7). d, Images showing animal body growth throughout postnatal development. e, Representative images showing EdU+ basal cells 24 hours post-labelling in P14 and P49 from Fig. 1e-g. f, Graphical representation of differential basal cell production rate throughout postnatal development. See Methods; data expressed as mean ± SEM; n=3 mice. g, 3D rendered z-stacks showing split confocal channels from Fig. 1i. h, Typical 3D rendered confocal z-stacks showing tilted side views from Fig. 1i. Yellow arrows indicate immature epithelial barrier. i, Representative side views of confocal z-stacks showing the thickening of the oesophageal epithelium (OE). j, Representative H&E sections of the oesophagus showing increasing cornification, as delimited by dotted lines. k, Quantification of the cornified thickness. n=3 mice; Micrometer, µm. Data expressed as mean ± SEM and analysed using one-way ANOVA with Tukey’s multiple comparisons test (^#p relative to P70; *p relative to P7). Scale bars. S1d(2 cm); S1e,g-j(20 µm). Stainings. Blue, DAPI; cyan, EdU; green, KRT14; greyscale, KRT4. All data derived from wild-type C57BL/6J mice. Dashed lines indicate basement membrane. Dotted lines in graphs indicate P28. Orange diamonds depict longitudinal orientation of the oesophagus where indicated. Source data are provided.

Extended Data Fig. 2. KLF4 basal cell profile in FUCCI2a mice. Related to Fig. 2.

a, Representative confocal z-stacks showing side views of OE wholemounts from Fig. 2a. Dashed lines indicate basement membrane; dotted lines mark the upper limit of the OE; white arrows indicate basal KLF4+ cells. Red, KLF4. b, In vivo protocol. Oesophagi from FUCCI2a mice were collected at time points indicated. Schematic indicating expression pattern of fluorescent proteins in FUCCI2a mouse model. c, Confocal images showing basal views of typical FUCCI2a OE wholemounts in (b). Orange diamonds indicate longitudinal orientation of the oesophagus. White dashed lines indicate mVenus+ cells; green, mVenus; red, mCherry. d and e, Correlation between KLF4 protein expression and reporter fluorescent proteins mCherry (d)/mVenus (e) in the basal layer from (b) and (c). n=3 mice; Scale bars 20 µm. Parts of (b) were drawn by using and/or adapting diagrams from Servier Medical Art.

Extended Data Fig. 3. Single-cell RNA sequencing annotation. Related to Fig. 3.

a, Flow cytometry gating strategy for isolation of OE cells. Oesophageal cell suspensions (i) were gated to sort the single (ii) viable (iii) population, enriched for epithelial cells (iv; EpCam+/CD45-). Cells were isolated from 15 mice (P7), 12 mice (P28), and 9 (adults). Representative plots from adult sample are shown. b, UMAP representing cell clusters based on louvain clustering. c, UMAP representing the distribution of cell clusters in (b) after integrating data from different time points (see Methods). d, UMAP showing expression of representative makers for basal (left panel) and differentiated cells (right panel) in OE. e, Heatmap showing expression of representative marker genes for basal cells, cell cycle, and differentiation for the 17 clusters shown in (cluster number from b in upper bar). Expression values were log2-transformed normalized UMIs followed by scaling and averaging across cells in the same clusters. f, Violin plots showing expression of representative epidermal (basal vs. differentiated) and cell cycle markers at different postnatal stages split by annotated cell cohorts in Fig. 3c. g, Violin plots showing expression of genes associated with regeneration and homeostasis for basal and differentiated cell types at distinct postnatal stages. Basal cells include both cycling and resting cells. h, UMAP showing spatial distribution of distinct cell cycle phases, annotated using cell cycle analysis by R package scran (v 1.12.1) combined with manual curation based on genes in (e). i, Violin plots showing expression of representative cell cycle genes at postnatal stages split by cell cycle cohorts identified in (h). Colour scheme for cell cycle phases as in (h). j, In vivo protocol. Mice were treated with EdU 2 hours prior culling at indicated time points. k, Typical 3D rendered confocal side views showing basal EdU population (see Methods). Dashed white lines, basement membrane. Dotted white lines, upper OE limit. Blue, DAPI; cyan, EdU; scale bar 10 µm. For violin plots in (f), (g) and (i), expression level means log2-transformed normalized UMIs, dotted lines indicate the median of the distribution. Colour bars of UMAPs in (d) denotes expression range. Parts of (j) were drawn by using and/or adapting diagrams from Servier Medical Art.

Extended Data Fig. 4. Single-cell RNA sequencing expression profile. Related to Fig. 3.

a, Distinctive patterns (Pt) of gene expression in basal cells as defined in (Fig. 3c). Grey, relative expression profiles of individual genes belonging to each pattern. Solid coloured lines, median values at each time point. To calculate the relative expression profiles, log2-transformed normalized UMIs were scaled and averaged across all basal cells at each time point and adjusted compared to the value at P7. b, Heatmap representing expression of individual genes belonging to the 4 patterns in (a). For expression values, log2-transformed normalised UMIs were scaled and averaged across all basal cells for each cluster and time point. The table on the right shows selected GO terms for major Pt2 and Pt4, corresponding p-values and representative genes. Closely related GO terms are grouped together. See Supplementary Table 3 for GO analysis result for all 4 expression patterns. c and d, UMAPs showing expression of genes related to key biological processes from Gene Ontology analysis for Patterns 2 (c) and 4 (d) in (b). e, Violin plots showing expression of Klf4 for cells in individual clusters at P7 (left), P28 (middle) and Adult (right). f, Expression of relevant genes along the pseudotime trajectory from basal resting to differentiated cells for P7. Left panel, YAP target genes (Cyr61, Ctgf, Thbs1) and genes associated with a response to mechanical stimuli (Cav1, Klf2, Dcn). Right panel, depicts KLF4 target genes (Krt4, Krt13, Cdnk1a, Cebpb). For violin plots in (e), expression level means log2-transformed normalized UMIs and dotted lines indicate the median of the distribution. Colour bars of UMAPs in (c) and (d) indicate log2-transformed normalized UMIs. Gene expression in (f) is represented as auto-scaled, log2-transformed normalized UMIs smoothed using a rolling mean along its trajectory with a window size of 5% of cells. Two bars on the top denotes the arrangement of cells according to pseudotime and clusters in Extended Data Fig. 3b, respectively.

Extended Data Fig. 5. Deep Learning based segmentation. Related to Fig. 4.

a, Protocol for in situ fixation of the oesophagus, and b, Images from in situ oesophagi compared with oesophagi fixed immediately after dissection confirm that basal cell shape was not affected by tissue harvesting (Supplementary to Fig. 4a). Blue, DAPI; green, KRT14. Scale bar, 20 µm. c, Schematic depicting deep learning based segmentation principle. Manually or semi-automatically annotated “ground truth” images were used to train a U-Net convolutional neural network. Training of the network was assessed on validation images and iteratively optimized until the achievement of satisfactory automated segmentation. d, Schematic of pipeline utilised for segmentation of single z-slice confocal images of OE basal layer. Nuclear segmentation was based on DAPI staining (blue). Mask overlay shows the match between the binary mask and the original fluorescence image. Scale bar, 20 µm. e, Schematic describing the computation of Voronoi diagrams of the tissue. Single z-slice confocal images of the OE basal layer are segmented as described in (d). Cell centroids are computed using the binary mask. Delaunay triangulation of cells was performed using cell centroids coordinates. Voronoi diagrams are calculated as the dual of Delaunay triangulation of cells in the tissue and overlaid onto the original fluorescence image. Scale bar, 20 µm. f, Cell shape anisotropy tensor at P14 and P49 (supplementary to Fig. 4d). n=3 mice. Orientation is colour-coded. Results from a representative experiment are shown; n=3 mice. g, Violin plots showing the distribution in cell shape anisotropy throughout postnatal development. n=2052-2594 cells from 3 animals per time point. Black dashed line, median. One-way ANOVA with Tukey’s multiple comparisons test (*p relative to P7). h, Bidimensional structure factor at P14 and P49 (supplementary to Fig. 4e). Changes in the dashed white outline (from ellipse to circle) depict a transition from anisotropic to isotropic spatial distribution over time; n=3 mice. i, Structure factor shape anisotropy distribution as shown in (h) and Fig. 4e. Box plots show median and quartiles; and whiskers, minima/maxima. Individual measurements from n=3 mice; (ns, not significant; *p relative to P7). All data derived from wild-type C57BL/6J mice. Parts of (a) were drawn by using and/or adapting diagrams from Servier Medical Art. Source data are provided.

Extended Data Fig. 6. Oesophageal tissue strain and second harmonic generation. Related to Fig. 5.

a, In situ and immediate ex vivo images of oesophageal tubes at P28 and P49 (supplementary to Fig. 5b). White dashed lines delineate oesophageal tube; scale bar 1 cm. b, Representative images showing the size of combined and separate oesophageal layers; Full oesophageal tube (Tube), epithelial composite (Epi) and muscle layer (Muscle); scale bar 5 mm. c, Longitudinal tissue strain relative to muscle, represented as percentage. Data expressed as mean ± SEM. One-way ANOVA with Tukey’s multiple comparisons test (P7 n=16, P28 and Adult n=9; *p relative to P7; ns, not significant). d, Measure of the stomach perimeter over time. Mean values ± SEM; n=3 mice; Millimetre, mm. e, Basal confocal view of EdU+ cells in wholemounts of typical squamous stomach epithelium from (f). f, In vivo protocol. Mice were treated with a single EdU injection 24h prior stomach collection at the time points indicated. g, Quantification of EdU+ basal cells per field from (f). Presented as mean ± SD; n = 3- h, Representative views of stroma underlying OE basement membrane using second harmonic generation (SHG). Left panels, collagen in magenta. Middle panels, colour map of SHG signal intensity. Right panels, colour-coded local orientation map of SHG signal. Scale bar 100 µm. i, Representative histograms depicting orientation distribution of collagen fibres in (h). n=3 mice. j, Basal view of representative OE wholemounts at P14. Green, YAP; greyscale, B-Catenin (BCat). Scale bar 10 µm. k and l, Quantification of basal nuclear and cytoplasmic staining of YAP and DAPI, respectively (see Methods; supplementary to Fig. 5m). A total of 20 cells for 3 different animals were measured. All data derived from wild-type C57BL/6J mice. Data analysis for c and d was performed using one- and two-way ANOVA, respectively, with Tukey’s multiple comparisons test (n=3-9; ns, not significant; (c) relative to P7). g, k and l was performed using two-tailed unpaired t test (ns, not significant). Box plots show median and quartiles; whiskers, 0.1 and 0.99 percentiles. Orange diamonds depict longitudinal orientation of the oesophagus where indicated. Parts of (f) were drawn by using and/or adapting diagrams from Servier Medical Art. Source data are provided.

Extended Data Fig. 7. Changes in tissue mechanics at cellular level. Related to Fig. 7.

a, Model of parts required for 3D printed stretcher. Scale bar, 1 cm. b, In vitro protocol. Oesophagi were exposed to 40% stretch using stretcher (Fig. 7a) and treated with/without 25 µM blebbistatin (BLEBB) for 48 hours. c, Individual basal cell areas. Data analysis was performed using Two-way ANOVA with Tukey’s multiple comparisons test (n=3 mice; ns, not significant; black indicates significance between control vs. BLEBB conditions; grey indicates statistical differences between stretching conditions). d, Confocal basal views of typical organ cultures after 48 hour BLEBB treatment in vitro from (b). Scale bar, 20 µm. e, In vitro protocol. Oesophagi stretched at the indicated levels, treated with EdU for 1 hour, and kept in vitro as whole-organ cultures for a 24 hours. f, Confocal basal views showing EdU and phosphohistone H3 (PhosphoH3) staining in cultures from (e). Scale bar, 20 µm. g and h, Basal quantification of EdU+ and PhosphoH3 cells expressed as percentage of DAPI+ cells from (e). Mean ± SD One-way ANOVA with Tukey’s multiple comparisons test (n=3 mice; *p relative to Control; ns, not significant;). i, In vivo protocol. Oesophageal wholemounts from Rosa26-mT/mG mice were mechanically peeled, separating cornified suprabasal layers or underlying stroma (see Methods). j, Individual area of basal cells. Data analysis was performed using One-way ANOVA with Tukey’s multiple comparisons test (n=3 mice; *p relative to full tissue). k, Confocal basal views of typical OE wholemounts after mechanical separation. Red, mTmG; scale bar, 10 µm. l, Basal cell density, expressed in number of cells per field after BLEBB treatment in Fig. 7e. Mean ± SD; n=3 mice. Unpaired t test (ns, not significant). m, Basal quantification of EdU+ cells expressed as percentage of DAPI+ cells after BLEBB treatment in Fig. 7e. Mean ± SD; n=3 mice. Unpaired t test (ns, not significant). Data derived from wild-type C57BL/6J mice, unless otherwise stated. Individual points show individual measurements, greyscale indicates values from each of 3 mice. Parts of (b, e) were drawn by using and/or adapting diagrams from Servier Medical Art. Source data are provided.