Interfollicular epidermal stem-like cells for the recreation of the hair follicle epithelial compartment

Abstract

Background: Hair follicle (HF) development and growth are dependent on epithelial-mesenchymal interactions (EMIs). Dermal papilla (DP) cells are recognized as the key inductive mesenchymal player, but the ideal source of receptive keratinocytes for human HF regeneration is yet to be defined. We herein investigated whether human interfollicular epidermal keratinocytes with stem-like features (EpSlKCs), characterized by a α6^bri/CD71^dim expression, can replace human hair follicular keratinocytes (HHFKCs) for the recreation of the HF epithelium and respective EMIs.

Methods: The α6^bri/CD71^dim cellular fraction was selected from the whole interfollicular keratinocyte population through fluorescence-activated cell sorting and directly compared with follicular keratinocytes in terms of their proliferative capacity and phenotype. The crosstalk with DP cells was studied in an indirect co-culture system, and EpSlKC hair forming capacity tested in a hair reconstitution assay when combined with DP cells.

Results: EpSlKCs exhibited a phenotypic profile similar to follicular keratinocytes and were capable of increasing DP cell proliferation and, for short co-culture times, the number of alkaline phosphatase-active cells, suggesting an improvement of their inductivity. Moreover, the recreation of immature HFs and sebaceous glands was observed after EpSlKC and DP cell co-grafting in nude mice.

Conclusions: Our results suggest that EpSlKCs are akin to follicular keratinocytes and can crosstalk with DP cells, contributing to HF morphogenesis in vivo, thus representing an attractive epithelial cell source for hair regeneration strategies.

Keywords: Dermal papilla cells; Epidermal keratinocytes with stem-like features; Epithelial-mesenchymal interactions; Hair follicle; Sebaceous gland.

Fig. 1

Morphology, proliferation, and phenotype of EpSlKCs and HHFKCs. a Representative light microscopy images of human epidermal stem-like keratinocyte (EpSlKC) and human hair follicular keratinocyte (HHFKC) culture. b DNA quantification of the cells along the culture time (n = 5, EpSlKCs; n = 4, HHFKCs). c Representative flow cytometry histograms and respective quantification regarding the percentage of α6^bri/CD71^dim cells, and those positive for integrin β1 (CD29), keratin 19 (K19), and keratin 14 (K14) in EpSlKC (n = 4) and HHFKC (n = 3) cultures after 1 week. d Immunofluorescence staining of β-actin filaments (phalloidin), keratin 14 (K14), keratin 10 (K10), keratin 6 (K6), keratin 15 (K15), and the proliferation-associated marker Ki67 in EpSlKCs and HHFKCs. DAPI was used as a nuclear counterstainer. Data shown are mean ± SEM. Scale bars are 100 μm for a and 50 μm for d. ***p < 0.001; ··p < 0.01 and ····p < 0.0001 vs. day 3; ^###p < 0.001 and ^####p < 0.0001 vs. day 5

Fig. 2

Characterization of EpSlKCs co-cultured with DP cells. a Schematic representation and b light microscopy images of EpSlKCs in co-culture with dermal papilla (DP) cells and in the respective monoculture controls. c DNA quantification of EpSlKCs after 5, 9, and 13 days of culture (n = 5). d Representative images of immunofluorescence staining against the epithelial markers K14 and K10, K6, and K15 after 5 days in culture. e Immunostaining against K14 staining in EpSlKCs after 13 days in culture and the respective flow cytometry quantitative data (n = 3). DAPI was used as nuclear counterstaining. Data shown are mean ± SEM. Scale bars are 100 μm for b and 50 μm for d and e. ****p < 0.0001; ····p < 0.0001 vs. day 5; ^##p < 0.01, ^###p < 0.001, and ^####p < 0.0001 vs. day

Fig. 3

Characterization of DP cells co-cultured with EpSlKCs. a Schematic representation of the co-culture established with DP cells and EpSlKCs and the respective monoculture controls. b Representative images of DP cell morphology after staining against the mesenchymal cytoskeleton marker vimentin. c DNA quantification of DP cells along the culture time (n = 5). d Expression of the versican V1 and V2 isoforms by DP cells in the different culture conditions at day 5 of culture. The presence of versican V2-positive cells was observed when DP cells were co-cultured with EpSlKCs (zoomed-in image). Images were counterstained with DAPI. e The inductivity of DP cells was assessed through the quantification of active alkaline phosphatase (ALP, n = 3). f Quantitative flow cytometry data (n = 3). *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001; ··p < 0.01 and ····p < 0.0001 vs. day 5; ^#p < 0.05, ^##p < 0.01, and ^####p < 0.0001 vs. day 9. Scale bars are 50 μm

Fig. 4

Quantification of the amount of a PDGF-A, b VEGF, and c BMP2 released by both co-cultured cells and monocultured EpSlKCs and DP cells by ELISA (n = 3). Data shown are mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001; ∙ p < 0.05, ·· p < 0.01, ··· p < 0.001, ···· p < 0.0001 vs. day 5; ^#p < 0.05 vs. day 9. bd, bellow detection level; BMP2, bone morphogenetic protein 2; PDGF-A, platelet-derived growth factor A; VEGF, vascular endothelial growth factor

Fig. 5

HF and SG induction in mice. a Representative hematoxylin and eosin (H&E) images of the area where EpSlKCs and DP cells, b DP cells alone, or c vehicle were transplanted (dashed lines indicating the wound area). Immunostaining of the recreated hair follicle (HF)- and sebaceous gland (SG)-like structures against the epithelial markers i.a K14, i.b K15 (dashed circle), or i.c both (white arrow). i.d Expression analysis of the differentiation marker K10 within the wound area (ii.a). H&E higher magnification image evidencing an apparently differentiated core (black arrowheads) within the formed structures, and the presence of small vessels around them (yellow arrowheads). ii.b Immunodetection of the fatty acid-binding protein 4 (FABP4) within the structures resembling SGs. Scale bars are 500 μm for a–c and 50 μm for i.a–i.d and ii.a and ii.b