Article Figures & Data
Figures
- Figure 1. Murine scratch wounding model follows physiological phases of wound healing.
(A) Mouse model, R26-CreERT2:R26-mTmG, before Cre recombination cells ubiquitously express a membrane-localized tdTomato. Upon Tamoxifen injection, Cre is activated and randomly and dose-dependently recombines the reporter cassette from tdTomato to GFP. Therefore, some cells start to express membrane-localized GFP. (B) Schematic murine epidermis upon scratch wounding. (C) Immunofluorescence stainings immediately (0), 8, 16, and 24 h after wounding. Fibronectin (FN, magenta) indicates dermal remodeling in scratch wounds (left panel, scale bar 100 μm) or full-thickness punch biopsies (right panel, scale bar 100 μm) of E-cad-CFP mice. White bar indicates the wound site. DAPI, blue. (D) Immunofluorescence stainings detecting keratin 14 (basal cell marker, magenta), and keratin 10 (suprabasal cell marker, yellow) (top panel), with arrow pointing to a fibrin clot. Keratin 6 (stress response marker, magenta) and CD45 (leukocyte marker, yellow) (bottom panel) in the skin of E-cad-CFP mice (cyan) (representative images from n = 3 mice). Scale bar, 100 μm.
- Figure S1. Scratch wound distribution on back skin.
(A) Immunofluorescence staining for LN332 (basement membrane) in R26-CreERT2:R26-mTmG mice before wounding, immediately after wounding (0), 8 and 16 h post wounding. Arrows indicating scratch wounds. Scale bar 500 μm. (B) Frequency distribution of induces scratch wounds sizes.
- Figure S2. Determination of cell areas.
(A) Representative x-z projections of immunofluorescence staining of skin whole mounts in E-cad-CFP mice immediately after (0), 8, 16, and 24 h post wounding (upper panel). Representative images of E-cad-CFP signals in the basal or suprabasal layer, respectively (middle panel). This signal was submitted to unbiased segmentation and cell size measurement using ImageJ (lower panel). All scale bars 50 μm. (B) Measurements of cell areas (μm2) of suprabasal cells based on E-cad-CFP expression in different distances towards the wound, directly after wounding (0), 8, 16 and 24 h post wounding. (n = 2 individual mice, with four wounds each per time point).
- Figure 2. Basal keratinocytes migrate as a sheet between two static layers.
(A) Measurements of cell areas (μm2) of basal cells based on E-cad-CFP expression in different distances towards the wound, directly after wounding (0), 8, 16, and 24 h post wounding. (n = 2 individual mice, with four wounds each per time point). (B) Measurements of cell areas of basal (B, magenta) and suprabasal cells (SB, yellow), directly after wounding (0), 8, 16 and 24 h post wounding. (n = 2 individual mice, with four wounds each per time point). (C) Schematic representation of three proposed models of modes of migration during re-epithelization. (D) Representative sequential images of intravital microscopy of the suprabasal layer (top panel) and the basal layer (bottom panel) after scratch wounding. Scale bar, 100 μm. See also Video 1. (E) Representative images of intravital microscopy of a defined area in proximity (290 μm away from the wound edge) to the wound bed starting 16 h after wound induction; suprabasal keratinocytes (top panel) and basal keratinocytes (bottom layer). Individual basal keratinocytes highlighted with a yellow outline. Scale bar 100 μm. See also Video 2. (F) Quantification of mean migration velocity in relation to cell size 16 h after induction of scratch wounding. (n = 5 individual mice). Dashed line indicating 380 μm2, separating basal from suprabasal keratinocytes. (G) Quantification of mean migration velocity of migratory (>10 μm/8 h) basal keratinocytes in relation to the distance of the wound 16 h after induction of scratch wounding. Median and interquartile distances are plotted for the distance bins <0.5 mm; 0.5–1 mm; >1 mm. (n = 5 individual mice). ****P > 0.0001, Two-tailed Mann–Whitney U-tests were performed for all statistical analyses. (H) Rose plot showing migration directionality in respect to the wound site of representative basal (left) and suprabasal cells (right). Inner rose plot in suprabasal cells represents a zoom of the initial rose plot. Value 0 indicates the relative starting point of migration 16 h post wounding.
Source data are available for this figure.
Source Data for Figure 2[LSA-2020-00765_SdataF2.xlsx]
- Figure 3. Single keratinocytes move within the collectively migrating sheet.
(A) Representative images of a migrating pairs at different distances to the scratch of recombined GFP-expressing cells in a R26-CreERT2:R26-mTmG mouse 16 h after scratch wound induction. Scale bar, 50 μm. (B) Reconstruction of migrating cell pairs and their neighboring cells, followed over minimal 2 h 30 min, 16 h after scratch wounding. The reconstructions depict three examples of cell group with different distances to the scratch wound. (C) The fraction of original neighbors that are remained is plotted over time. Basal keratinocytes are grouped based on initial distance to scratch wound (>1 mm; 0.5–1 mm; <0.5 mm). Imaging was performed 16 h post scratch wounding for 8 h. Plotted is the mean ± SEM for 25 imaging positions in three mice. ANOVA multiple comparison with *P > 0.1, **P > 0.01, and ****P > 0.0001.
- Figure S3. Velocity of exchanging neighboring cells.
(A) The fraction of remaining original neighbors is plotted over time. Basal keratinocytes are grouped based on migration velocity (<0.08; 0.08–0.18; > 0.18 μm/min) and imaging was performed from 16 h post scratch wounding for 8 h. Plotted is the mean ± SEM for 25 imaging positions in three mice. ANOVA multiple comparison with *P > 0.1, **P > 0.01, and ****P > 0.0001.
- Figure 4. Migrating keratinocytes bypass hair follicle obstacles.
(A) Representative images of migrating keratinocytes in a Fucci2 mouse 16 h after scratch wound induction bypassing a hair follicle (*). Scale bar, 50 μm. Individual keratinocytes are highlighted by their migration tracks. Also see Video 3. (B) Representative sequential images of labelled cells within mTmG mice 16–24 h post wounding within the infundibulum and upper hair follicle (upper panel) and cells within the basal layer (lower panel). Migration/no migration is indicated by the shifts in yellow circles over time. (C) Quantification of total displacement in different distances towards the wound of hair follicle (grey) and basal cells (blue), respectively. Data points represent single cells from two mice, with five different wounds per mouse. ****P < 0.0001; **P = 0.002 (unpaired t test with Welsh correction).
Source data are available for this figure.
Source Data for Figure 4[LSA-2020-00765_SdataF4.xlsx]
- Figure 5. Proliferation occurs in a distinct zone and does not affect overall migration.
(A) Left panel: fluorescence staining of EdU incorporation (green) immediately after wounding and 8, 16 and 24 h post wounding. LN332 (magenta) immunofluorescence counterstain marks the basement membrane. Scale bar 100 μm. Right panel: quantification of EdU incorporation, each dot represents one wound in n = 2 mice. (B) Representative image of a scratch wound in Fucci2 mice. Cells in G1-phase (magenta), in G2/S-phase (green) in their different distances towards the wound. Scale bar 100 μm. (C) Quantification of percentage of proliferating (G2/S, green) and non-proliferating cells (G1, magenta) grouped based on their distance to the wound (0–200, 200–400, 400–600, 600–800, and 800–1,000 μm). **P = 0.0086 (ANOVA multiple comparison). (D) Percentage of proliferating G2/S cells within a distance of 200–400 μm away from the wound bed is plotted between 16 and 24 h post wounding. Plots depict the min., max., and median, the lower and upper quantile, the plus indicates the mean. (E) Linear regression analysis between scratch size (μm) and percentage of G2/S-phase proliferating cells 200–600 μm away from the wound site. R2 = 0.8241. (F) Linear regression analysis between scratch size (μm) and average migration velocity (μm/h) of basal keratinocytes 0–200 μm away from the wound site. R2 = 0.6129. (G) Quantification of migration velocity of proliferating S/G2-phase cells (green) and non-proliferating G1-phase cells (magenta) within different distances towards the wound. ns, not significant (unpaired t test). (H) Quantification of persistence of migratory proliferating S/G2-phase cells (green) and non-proliferating G1-phase cells (magenta) within different distances towards the wound. ns, not significant (Kruskal–Wallis test).
Source data are available for this figure.
Source Data for Figure 5[LSA-2020-00765_SdataF5.xlsx]
- Figure 6. Model of scratch wound re-epithelialization.
Schematic representation of the model for re-epithelization in partial-thickness wounds. Upon the initiation of a scratch wound, basal keratinocytes immediately move towards the wound side without entering the wound bed. After a period of jamming, basal keratinocytes enter the wound bed, whereas suprabasal keratinocytes remain static. Basal keratinocytes migrate individually through a collectively moving basal layer directed towards the wound bed. Left panel, top view of the basal migrating keratinocytes. Dashed line representing the cross section shown in the right panel.
Supplementary Materials
- Video 1
Re-epithelization of the wound bed is initiated by basal keratinocytes ∼20 h after wound induction. Representative movie of suprabasal (left panel) and basal (right panel) keratinocyte dynamics in response to induced scratch wounds in the back skin of R26-mTmG; R26-ACTB-CreERT2 mice. These mice express mTom+ (magenta) in all cells and some cells are randomly labelled by GFP+ (green) after tamoxifen-induced recombination facilitating the study of individual cells. Keratinocytes were tracked shortly after wound induction and images were taken every 30 and 45 min–40 h after wounding. Download video
- Video 2
Basal keratinocytes migrate underneath the suprabasal layer. Representative movie of depth color coded z-maximum projection of random, individually labelled GFP+ migrating keratinocytes. Color code in the top right corner, with blue representing superficial suprabasal layer and red representing deeper basal layer. Tracking of keratinocytes was initiated 16 h post wounding and images were taken every 30 min over the course of minimally 8 h. Download video
- Video 3
Migrating keratinocytes bypass hair follicle obstacles. Representative movie of migrating keratinocytes in a Fucci2 mouse 16 h after scratch wound induction bypassing a hair follicle (*) at a distance of 200–400 μm away from the wound. Scale bar, 50 μm. Tracks of migrating keratinocytes are depicted. Download video
- Video 4
Uncoupled proliferation of migration of basal keratinocytes. Representative movie of migrating basal layer dynamics in response to induced scratch wounds in the back skin of Fucci2 mice. These mice express mCherry-hCdt1 (magenta) in cells that are in a G1-cell cycle state and mVenus-hGem (green) in proliferating cells in S/G2 phase. Keratinocytes were tracked 16–24 h after wound induction and images were taken every 30 min. Scale bar 100 μm. Download video
Related Data
