Human Dermal Fibroblast Subpopulations Display Distinct Gene Signatures Related to Cell Behaviors and Matrisome

The dermis is divided into the (upper) papillary dermis and the (lower) reticular dermis that differ structurally and functionally. The isolation of the resident fibroblasts from papillary and reticular dermis revealed that the papillary (Fp) and reticular (Fr) fibroblasts have distinct morphology and behavior (Harper and Grove, 1979Harper R.A. Grove G. Human skin fibroblasts derived from papillary and reticular dermis: differences in growth potential in vitro.Science. 1979; 204: 526-527Crossref PubMed Scopus (223) Google Scholar). This supported the concept that subpopulations of fibroblasts are present within the dermis and prompted the search for selective cell markers (Driskell and Watt, 2015Driskell R.R. Watt F.M. Understanding fibroblast heterogeneity in the skin.Trends Cell Biol. 2015; 25: 92-99Abstract Full Text Full Text PDF PubMed Scopus (216) Google Scholar). However, the identified markers cannot explain per se the structural and morphological signatures of these two compartments. Notably, although dermis is primarily composed of extracellular matrix (ECM), few ECM genes were identified in the previous transcriptome comparing Fp and Fr subpopulations (Janson et al., 2012Janson D.G. Saintigny G. van Adrichem A. Mahé C. El Ghalbzouri A. Different gene expression patterns in human papillary and reticular fibroblasts.J Invest Dermatol. 2012; 132: 2565-2572Abstract Full Text Full Text PDF PubMed Scopus (93) Google Scholar). We performed a transcriptomic analysis of dermal fibroblast subpopulations using Affymetrix GeneChip human genome arrays, using stringent criteria: (i) selecting young donors at a time fibroblasts are active and produce high levels of ECM, (ii) taking skin samples from a unique anatomic site (i.e., breast), (iii) performing the analysis soon after fibroblast isolation. Samples were obtained from female donors during plastic surgery procedures, after written informed consent and institutional approval. Site-matched Fp and Fr fibroblasts were isolated from dermatomed dermis, and explant-derived cells were cultured at passage 1. The purity of the isolated fibroblast subpopulations was checked based on cell morphology, cell proliferation, and qRT-PCR of the Fp marker PDPN (Supplementary Figure S1 online). The absence of contaminant cells was checked by immunostaining of Fp and Fr using endothelial cells or keratinocytes markers (Supplementary Figure S2 online). K14-positive cells were absent in Fr and only occasionally observed in Fp. A total of 25,223 probes (45% of the total probes) were hybridized with cRNA from the dermal fibroblasts of three independent donors aged 20 to 22 (Supplementary Table S1 online). Principal component analysis showed that Fp and Fr grouped separately (Supplementary Figure S1e). Bioinformatics analyses delivered 230 upregulated and 139 downregulated genes in Fp versus Fr. We found markers identified in previous studies in our list of genes: the Fp markers MMP1 (Mine et al., 2008Mine S. Fortunel N.O. Pageon H. Asselineau D. Aging alters functionally human dermal papillary fibroblasts but not reticular fibroblasts: a new view of skin morphogenesis and aging.PloS One. 2008; 3: e4066Crossref PubMed Scopus (141) Google Scholar) and PDPN (Janson et al., 2012Janson D.G. Saintigny G. van Adrichem A. Mahé C. El Ghalbzouri A. Different gene expression patterns in human papillary and reticular fibroblasts.J Invest Dermatol. 2012; 132: 2565-2572Abstract Full Text Full Text PDF PubMed Scopus (93) Google Scholar), and the Fr markers TGM2 and MGP (Janson et al., 2012Janson D.G. Saintigny G. van Adrichem A. Mahé C. El Ghalbzouri A. Different gene expression patterns in human papillary and reticular fibroblasts.J Invest Dermatol. 2012; 132: 2565-2572Abstract Full Text Full Text PDF PubMed Scopus (93) Google Scholar) (Supplementary Table S2 online). We next performed gene ontology and Kyoto Encyclopedia of Genes and Genomes pathway enrichment analyses of Fp versus Fr (Supplementary Figure S3 online). Gene ontology terms related to defense function, response to cytokines, regulation of cell motility, and proliferation were highly represented in Fp, whereas Fr showed higher expression of genes related to “development of connective tissues” (Supplementary Figure S3a). Kyoto Encyclopedia of Genes and Genomes analysis identified five main pathways (Supplementary Figure S3b). Notably, cell proliferation related to the key mitogen-activated protein kinase signaling pathway was appropriately assigned to Fp. Altogether, these data indicated that Fp are more prone to produce and to respond to chemical signals than Fr as reported (Sorrell et al., 2004Sorrell J.M. Baber M.A. Caplan A.I. Site-matched papillary and reticular human dermal fibroblasts differ in their release of specific growth factors/cytokines and in their interaction with keratinocytes.J Cell Physiol. 2004; 200: 134-145Crossref PubMed Scopus (88) Google Scholar). The gene ontology terms related to “vasculature development” and “angiogenesis” were also highly represented in Fp suggesting its paracrine role in the vascularization of the dermis. This is consistent with the capillary density of the papillary dermis that is approximately 10 times that of the reticular dermis (Papp, 2012Papp A. Dermal wound healing and burn wounds. In: Larjava H, editor. Oral wound healing [Internet]. West Sussex, UK: John Wiley & Sons; 2012. p.333–45. http://onlinelibrary.wiley.com/doi/10.1002/9781118704509.ch14/summary (accessed 3 August 2016).Google Scholar). We exploited the database MatrisomeDB (Naba et al., 2016Naba A. Clauser K.R. Ding H. Whittaker C.A. Carr S.A. Hynes R.O. The extracellular matrix: tools and insights for the “omics” era.Matrix Biol J Int Soc Matrix Biol. 2016; 49: 10-24Crossref PubMed Scopus (507) Google Scholar) to build the matrisome of the dermal fibroblast subpopulations (Figure 1a, Supplementary Figure S4a online). The core matrisome was highly enriched in Fr (13 of 18 genes differentially regulated), whereas the matrisome-associated genes (38 regulated in our transcriptome) were both enriched in Fp and Fr. The upregulation of ECM regulator genes encoding matrix metalloproteinases and cathepsins (involved in matrix remodeling), and LOXL3 (an amine oxidase with activity toward elastin and collagens), indicates intensive matrix remodeling of the papillary dermis. The upregulation of COL7A1 and ELN in Fp and Fr, respectively, confirmed at the transcript and protein levels, was of particular interest (Figures 1b and 2). Fibroblasts and keratinocytes contribute to the development of the dermoepidermal junction that is predominantly composed of COLVII anchoring fibrils. Dermal fibroblasts are used for cell therapy in patients with skin blistering diseases (Nyström et al., 2013Nyström A. Bruckner-Tuderman L. Kern J.S. Cell- and protein-based therapy approaches for epidermolysis bullosa.Methods Mol Biol. 2013; 961: 425-440Crossref PubMed Scopus (11) Google Scholar). The use of Fp has the dual advantage of displaying elevated levels of COLVII and a high proliferation rate that could improve cell therapy. As for Fr, highly expressed ELN and MAFP5 both encode major components of the elastic fibrils. Elastin electron-dense deposits were observed in the ECM of Fr (Figure 2b). This is coherent with the presence of larger amounts of thick elastic fibers in the ECM of Fr compared with Fp. Most unexpected was the presence of COL11A1 as the first upregulated gene in Fr (Supplementary Table S2). COL11A1 encodes the COLXIα1 chain whose expression is chiefly restricted to cartilage (Ricard-Blum and Ruggiero, 2005Ricard-Blum S. Ruggiero F. The collagen superfamily: from the extracellular matrix to the cell membrane.Pathol Biol (Paris). 2005; 53: 430-442Crossref PubMed Scopus (268) Google Scholar). As such, the gene ontology term “Skeletal system development” was highly enriched in Fr (Supplementary Figure S3a). Five of the representative genes (COL11A1, MGP, FGF18, COMP, and ACAN) were in the top 10 genes upregulated in Fr (Supplementary Table S2). COL11A1 upregulation was confirmed in cultured Fr independently of the anatomic site (breast or abdominal), and in breast reticular dermis (Supplementary Figure S4b, Figure 2c). COLXI-containing ECM produced by cultivated cells is highly insoluble (Kleman et al., 1992Kleman J.P. Hartmann D.J. Ramirez F. van der Rest M. The human rhabdomyosarcoma cell line A204 lays down a highly insoluble matrix composed mainly of alpha 1 type-XI and alpha 2 type-V collagen chains.Eur J Biochem FEBS. 1992; 210: 329-335Crossref PubMed Scopus (67) Google Scholar). Therefore, the presence of COLXIα1 in reticular dermis was assayed in crude extracts of freshly dermatomed dermis. COLXIα1 (pro)-chains were significantly enriched in reticular dermis extracts (Figure 2c). COL11A1 was expressed in murine dermal fibroblasts in response to Wnt/β-catenin signaling and deposited at the dermis-hair follicle interface (Collins et al., 2011Collins C.A. Kretzschmar K. Watt F.M. Reprogramming adult dermis to a neonatal state through epidermal activation of β-catenin.Development. 2011; 138: 5189-5199Crossref PubMed Scopus (108) Google Scholar). ACAN, COMP, and/or COL11A1 expression were enhanced in tissues exposed to mechanical tension as fibrotic skin (Agarwal et al., 2013Agarwal P. Schulz J.-N. Blumbach K. Andreasson K. Heinegård D. Paulsson M. et al.Enhanced deposition of cartilage oligomeric matrix protein is a common feature in fibrotic skin pathologies.Matrix Biol J Int Soc Matrix Biol. 2013; 32: 325-331Crossref PubMed Scopus (43) Google Scholar, Giannoni et al., 2003Giannoni P. Siegrist M. Hunziker E.B. Wong M. The mechanosensitivity of cartilage oligomeric matrix protein (COMP).Biorheology. 2003; 40: 101-109PubMed Google Scholar, Naitoh et al., 2005Naitoh M. Kubota H. Ikeda M. Tanaka T. Shirane H. Suzuki S. et al.Gene expression in human keloids is altered from dermal to chondrocytic and osteogenic lineage.Genes Cells Devoted Mol Cell Mech. 2005; 10: 1081-1091Crossref PubMed Scopus (69) Google Scholar). The Fr might respond to the ECM density-induced mechanical tension of the reticular dermis by expressing cartilaginous ECM genes. COL11A1 expression was not accompanied by expression of the cartilage marker COL2A1, indicating that COLXIα1 may fulfill a specific function in skin. Our study identified a cell behavior and ECM signature of the fibroblast subpopulations that is coherent with the organization and function of the papillary and reticular dermis. This could help refine therapeutic strategies for skin aging and diseases and bioengineering new skin substitutes that better match skin physiology. VB, LM, SB, and BC are employees of SILAB. FR declares the receipt of a grant from SILAB. PN and SN-B state no conflict of interest. We thank M. Koch for providing us with collagen VII antibodies, M. Malbouyres (IGFL, Lyon) for assistance with transmission electron microscopy that was performed at the “Centre Technique des Microstructures” (Université de Lyon), and J. Burden for editing the manuscript for English. Download .pdf (.46 MB) Help with pdf files Supplementary Data