Pericytes promote skin regeneration by inducing epidermal cell polarity and planar cell divisions

(A) Schematic of OC technique illustrating establishment of 3D skin regeneration within a six-well insert, where cells are grown initially submerged in culture media to establish a DE consisting of fibroblasts, pericytes, or both within a type 1 collagen gel and then maintained at an air–liquid interface to promote epidermal stratification and differentiation. (B) Histological analysis of OCs seeded with either fibroblasts, pericytes, or both on paraffin sections stained with H&E from two independent experiments—first and second row (supporting data shown from a third experiment in Fig 1A) to verify reproducibility of the observation that pericytes induce basal cell polarization in the basal layer of OCs. (C) Transmission electron microscopy analysis of dermo-epidermal junctional region of OCs from two independent OC experiments seeded with either fibroblasts, pericytes, or both, revealing that hemi-desmosome and basement membrane assembly are consistently observed in pericyte cocultures—images are additional data supporting transmission electron microscopy results shown in Fig 2B. Results in this figure and Fig 2B are representative of 18 random fields from two independent biological replicates. Scale bar = 0.5 μm.

(A, C, E, G, H) Human keratinocytes placed in OCs with DEs made up of fibroblasts, pericytes, or a combination of both exhibited differences in morphology. Hematoxylin and eosin staining revealed a polarized basal layer (A) and immunostaining showed increased expression of the proliferative markers Ki67 (C), ΔN-p63 (E), basal layer K15 expression (G), and ordered suprabasal K10 expression (H) in the presence of pericytes. Images representative of four independent OC experiments. (B, D, F) Quantitation of increased thickness of the epithelium (B), increased number of Ki67⁺ cells (12 random fields/group) (D), and % basal/total ΔN-p63⁺ (24 random fields/group) (F) in the presence of pericytes from four independent OC experiments. Error bars are mean ± SD. Unpaired t test. **P < 0.01, ***P < 0.001, n.s. not significant. Scale bar = 100 μm.

(A) Immunofluorescent staining for LAMA5 (red) and DAPI (blue, nuclei) in OCs populated with fibroblasts, pericytes, or a combination of both demonstrating its uniform deposition in pericyte-populated OCs. Images representative of three independent OC experiments. Scale bar = 100 μm. (B) Transmission electron micrographs of the dermo-epidermal junction of OCs revealing absence of a basement membrane in fibroblast OCs, sporadic initiation of hemi-desmosomes in fibroblast and pericyte-combined OCs, and complete basement membrane assembly with a continuous lamina lucida (LL) and lamina densa (LD) (enlarged boxed area) and regular hemi-desmosomes (white arrows) in pericyte OCs. The higher magnification pericyte OC transmission electron micrograph illustrates the inner plaque (ip) and outer plaque (op) of the hemi-desmosomes. Data representative of 18 random fields from two independent OCs. epi, epidermis; der, dermis. Scale bar = 0.5 μm.

(A–F) Whole mount confocal microscopy of OCs populated with fibroblasts (A–C) or pericytes (D–F) stained with rhodamine-conjugated phalloidin (red, actin) and DAPI (blue, nuclear); z-stacks (C, F) reveal irregular dermo-epidermal junctions in fibroblast-populated OCs (A, C) and their bipolar morphology (B). Pericyte-populated OCs show stellate pericyte morphology (E) and their proximity at the dermo-epidermal interface (D); z-stack reveals a well-organized and even basal layer and close contact with pericytes (F). Images representative of six random fields from one of three OC experiments. White dotted line = interface between the epidermis (epi) and dermis (der). Scale bar = 100 μm. (G–I) Triple immunofluorescent staining for DAPI (G: blue, DNA), pH 3 (H: green, mitotic cells), and γ-tubulin (I: red, centrosomes) in OCs populated with fibroblasts (F), pericytes (P), or a combination of both (F + P), showing representative cell divisions. Cell division orientation relative to the basement membrane was determined by measuring the angle between the dermo-epidermal junction and the centrosomal axis in pH 3⁺ cells. J. Quantitation of the angle of cell division (0^o–90^o) as % of total mitotic events in OCs populated with fibroblasts (F), pericytes (P), or both (F + P), from 43, 23, and 23 mitoses per group from three independent experiments displayed as a radial bar chart showing a tendency for mitoses parallel to the basement membrane in the presence of pericytes. Scale bar = 20 μm.

(A) Immunofluorescence analysis for GFP in vector-transduced control fibroblasts (F_GFP) and BMP-2–transduced fibroblasts (F_BMP2) in culture. Scale bar = 50 μm. (B) qPCR analysis of BMP-2 expression in control GFP (F_GFP) and BMP-2 (F_BMP2)–transduced fibroblasts, confirming expression of BMP-2 transcripts in the latter. Data expressed as mean ± SD from triplicate experiments. Statistical analysis performed using unpaired t test, **P < 0.01. (C–F) Immunofluorescence staining for BMP-2 in brefeldin-treated pericytes (C, D) or untreated pericytes (E, F), stained with anti-BMP-2 polyclonal antibody (C, E) or isotype control polyclonal antibody (D, F), confirming BMP-2 expression in pericytes (C). (G–J) Immunofluorescence staining with anti-BMP-2 (G, I) or isotype control antibody (H, J) in brefeldin A–treated F_GFP fibroblasts (G, H) and F_BMP2 fibroblasts (I, J). (K) Quantitation of BMP-2 immunofluorescence signal in F_BMP2 (n = 43 cells) and F_GFP (n = 83 cells) fibroblasts versus pericytes (n = 28 cells) in a representative experiment from a total of three experiments. Statistical significance was established by one-way ANOVA Tukey's multiple comparison test **P < 0.01, ****P < 0.0001. (L) OCs generated with F_GFP control or F_BMP2 fibroblasts stained for K15, demonstrating its absence. Dotted line indicates dermo-epidermal junction. Staining representative of three independent OC experiments. Scale bar = 100 μm.