Novel Xenografting Model to Explore the Mechanisms Mediating Acute and Chronic Fibrosis in Human Skin Fibroblasts

Abstract

INTRODUCTION: Human scar formation and fibrosis are difficult to accurately recapitulate using mouse models, given the significant anatomical and physiologic differences between mouse and human skin. Xenografts of human skin on immunodeficient mice provide an accessible means of assessing human skin’s physiology and response to wounding or fibrosis-inducing conditions in vivo. However, current xenograft models are limited by poor engraftment rates and inability to specifically explore the mechanisms mediating fibrosis in human fibroblasts. OBJECTIVE: We describe a novel skin xenografting model to investigate the response of human dermal fibroblasts to different fibrosis-promoting conditions. METHODS: Full-thickness circular 8-mm human foreskin samples were sutured into the dorsum of P2 immunocompromised (NSG) mice as subcutaneous grafts (n = 30), and surgically exposed after 7 days to produce cutaneous grafts. Successful engraftment and preservation of normal skin physiology were confirmed by histology. Machine learning–based assessment of collagen fiber networks from stained skin histology specimens was achieved using a novel computational algorithm developed by our laboratory. To study the acute fibrotic response, 4-mm partial-thickness wounds were created within the xenografts using a biopsy punch; wounds were monitored until closure. To explore the chronic fibrotic response, xenografted skin was irradiated with 30 Gy fractionated into six 5 Gy doses delivered every other day for a total of 12 days. Following radiation, chronic fibrotic changes were allowed to develop over an interval of 4 weeks. At the respective endpoints, xenografted skin was harvested for histology. Human fibroblasts were isolated using flow cytometry with a negative gating strategy to exclude mouse and human hematopoietic cells (CD45−/Ter-119−/CD235a−), endothelial cells (CD31−), and epithelial cells (CD326−[mEpCam]/mCD324−[E-Cadherin]), and a positive gate to include only human fibroblasts (CD90+). Microarray analyses were used to compare gene expression of human fibroblasts isolated from scarred/irradiated xenografts to those from unwounded/nonirradiated xenografts. RESULTS: Xenografted foreskin was structurally similar to native (ungrafted) foreskin on histology. Collagen fiber network analysis confirmed that xenografted skin was more similar to foreskin than to scarred adult human skin. Wounds created in xenografted skin exhibited slower wound closure compared to stented mouse wounds, indicating healing primarily via formation of granulation tissue (akin to human skin) rather than contraction (typical of mouse skin). Irradiated skin was significantly indurated on histologic assessment, consistent with chronic irradiation damage/fibrosis. Immunofluorescence staining confirmed successful xenograft vascularization and survival of human skin cells and human origin of granulation tissue. Gene expression analysis of fibroblasts isolated from acutely and chronically fibrosed xenografts revealed upregulation of the Wnt and FAK pathways and increased expression of the CD26 surface antigen. CONCLUSIONS: We present a novel foreskin xenografting model and demonstrate its utility in specifically investigating the in vivo human fibroblast response in acute and chronic fibrosis. This model provides an accessible and informative tool to aid in elucidation of fibroblast-driven mechanisms responsible for scarring and fibrosis. Ultimately, this work may enable the discovery of novel cellular and molecular targets to reduce skin scarring and fibrosis.