Methodology for the production of human hair follicles
- Autores: Cristina Quílez López
- Directores de la Tesis: José L. Jorcano Noval (dir. tes.), Diego Velasco Bayón (codir. tes.)
- Lectura: En la Universidad Carlos III de Madrid ( España ) en 2021
- Idioma: español
- Tribunal Calificador de la Tesis: Alvaro Meana Infiesta (presid.), Carlos León (secret.), Joan Xavier Fontdevila Font (voc.)
- Programa de doctorado: Programa de Doctorado en Ciencia y Tecnología Biomédica por la Universidad Carlos III de Madrid
- Materias:
- Enlaces
- Tesis en acceso abierto en: e-Archivo
- Resumen
Hair follicles are a signature in mammals and cover almost the entire surface of their skin. They are the most important skin-derived organs, as they are involved in diverse biological processes: they provide protection, thermal isolation and also comprise a reservoir of cells for skin regeneration and wound healing. Disorders associated with hair loss do not only compromise correct functioning of the human body, but also have associated psychological consequences. During adulthood, hair follicle structure is not able to regenerate, reason why most hair follicle disorders imply permanent hair loss. From all of them, Androgenetic Alopecia is probably the most socially relevant as it has a prevalence of 70% in men older than 70 years. It is a non-scaring pathology in which hair is progressively lost, following a pattern distribution by genetically predisposed hair follicles that are sensitive to androgens. The high cost of the available treatments and the low availability of hair follicles in severe cases of hair loss underline the need to develop a hair regeneration therapy in adults. Nonetheless, hair follicle is a very complex and specialized structure difficult to replicate. Additionally, the increasing need of skin in vitro models in pharmacological testing requires the generation of realistic skin models with all its associated appendages.
In this scenario, a big effort has been made by the scientific community to understand and characterize human hair morphogenesis in the embryo. Despite several approaches that successfully regenerate the hair follicle have been reported in the literature along the last 40 years, most of them imply the use of either embryonic or neonatal cells from murine origin or genetically modified cells in mouse models. This raises the question of whether these methods can be applied in humans, not only due to safety reason, but also to the differences found in hair morphogenesis in the two species. From all the proposed methodologies to regenerate the hair follicle, those based on the dermal papilla (DP) attract most of the attention for regenerative purposes. DP is a spherical structure, located at the base of the hair follicle that contains highly specialized fibroblasts. This structure is crucial to induce the differentiation of the hair follicle in the embryo and its maintenance throughout life in adults. Furthermore, dissected murine DPs were shown to induce the generation of new follicles in vivo upon transplantation to host recipients. Later experiments demonstrated that this property was maintained by cultured DP cells, revealing the potential of these cells to induce hair follicles formation. However, extrapolation of these results to human DP is not straightforward since human dermal papilla cells lose their inductive capacity as soon as the DP structure is broken and they are placed in 2D culture, contrary to rodent cells. Nonetheless, it has been demonstrated that human DP cells partially and inefficiently recover their in vivo inductive capacity if cultured as 3D aggregates (spheroids) emulating the in vivo microenvironment of a DP and transplanted to the dermo-epidermal interface of human skin transplanted to immunodeficient mice.
Based on these needs, this thesis aimed at: 1) finding out the reasons of the poor efficiency of the spheroids to reprogram cultured DP cells and improve it; 2) developing an in vitro system allowing to perform these studies in a quick and versatile way; 3) exploring of the clinical and/or industrial utility of this in vitro system. Here we propose the use of adult human DP cells in combination with adult skin epithelial cells to induce hair follicle formation in a plasma-derived in vitro skin model. For this purpose, two different systems were proposed to culture DP cells and mimic human DP structure: dermal papilla spheroids (DPS) and fibrin microgels (FM) with encapsulated cells. In that sense, this methodology can both solve the problem of hair regeneration in humans with autologous cells as well as provide an in vitro skin model with hair follicles, milestones not yet achieved.
In order to mimic human hair follicle DP structure, a series of experiments were performed to first establish the appropriate conditions to obtain human follicle dermal papilla cells (hFDPc) from dissected DP obtained from donors and to culture them. As a result, two different hFDPc cell lines were established from human scalp follicles, which showed the same morphology as hFDPc of commercial origin. Then, DP spheroids (DPS) containing 3000 cells were generated using hFDPc from both donors and commercial origin. To do so, different spheroid formation methodologies (hanging drop and low-adhesion plate) and cell culture condition (cell origin, cells passage or the presence of antibiotic in the culture media) were studied and compared in terms of spheroid formation efficiency, morphology, and size. Moreover, from this analysis two main conclusions were reached: 1) the presence of antibiotic had no effect on spheroid formation efficiency and 2) low adhesion plate is a more scalable and reproducible methodology to generate DPS when compared with hanging drop method. Positive expression of Alkaline Phosphatase, Versican and alpha-smooth muscle actin proved by immunostaining of the cells within the spheroid confirmed stemness recovery of the cells. These results led us to stablish a protocol to generate DPS that successfully reprogram the stem fate in cells from both commercial and patient origin. However, a conscientious analysis revealed a low cell viability in the surface of the spheroid. Morphometric analysis of the spheroid over time revealed that as a consequence of cell-to-cell interaction, the spheroid compacted and then stabilized. Simultaneously to this reduction in size, metabolic activity (ATP), viability (Live/Dead) and replication also declined, suggesting that the spheroid reached cell culture homeostasis. Nonetheless, cell viability in the spheroid is low and needs to be improved to optimize the cell number in the spheroid, something crucial for hFDPc due to its low proliferative capacity and limited expansion rate. In spheroid cultures, diffusion of nutrients and oxygen to the core of the spheroid is one of the limiting factors regarding cell viability. To characterize the effect of cell nourishment in cell viability, spheroids containing lower cell number (1500 and 750 cells/spheroid) and cultured in dynamic conditions were analyzed. Nonetheless, viability results revealed that these culture conditions did not improve cell viability, which is not surprising, as the radius of spheroids with 3000 cells (~150 µm) is below the diffusion limit (250 µm). These results highlighted that alternative factors to cell number and dynamic conditions need to be explored to enhance hFDPc viability in the spheroid. For example, extracellular matrix composition of the DP structure may be analyzed in detail as it is dynamic and plays an important role during the hair follicle cycle. Despite DPS successfully reprograming hFDPc stem-fate, the low viability exhibited by the cells questions whether spheroids are the suitable culture system for hFDPc.
To overcome the viability problems showed by DPS, the use of human fibrin microgels (FM) for hFDPc encapsulation were proposed as an alternative culture system to mimic DP structure. This system aimed to provide hFDPc with a transient fibrin extracellular matrix that can be remodeled by the cells into one more similar to that found in human DP. For this purpose, FM containing 6000, 3000, 1500 and 750 cells/microgel were generated and analyzed in terms of morphology, cell viability, stemness recovery and matrix remodeling mediated by cells. Microgel contraction and cell morphology inside the FM studies, together with cell viability results based on with Live/Dead and ATP generation assays suggest that high cell concentrations yield worse results due to an imbalance of cell number and microgel volume. On this basis, a higher number of encapsulated cells requires a proportional increase of FM volume to provide a cell/volume ratio that guarantees cell accommodation and hence, improves cell viability. Moreover, cell/volume ratio in DP structure plays a crucial role in vivo. In humans, depending on hair location the total cell number inside the DP varies from 1300 cells in the scalp to 3000 cells in the face. Based on these values, FM containing 1500 and 3000 cells seem to be the best choice to generate a system like the human DP structure. Positive expression of Alkaline Phosphatase, Versican and alpha-smooth muscle actin proved by immunostaining of the FM confirmed stemness recovery of the cells. Moreover, presence of Versican and Collagen IV in the FM after 15 days in culture demonstrated the capacity of the FM to support cell-mediated matrix remodeling into one similar to the native DP structure. Furthermore, for all cell numbers this system improved cell viability, metabolism and replication capacity when compared to that of DPS. This suggests that FM-like methods, providing an extracellular matrix to the hFDPc seem to be a better alternative for the in vitro generation of human DP-like structures. Nonetheless, the cell-mediated matrix contraction and remodeling process suggests that FM require longer culture times to obtain more mature DP-like structures. In addition to that, a thorough analysis and characterization of marker expression and extracellular matrix composition are needed to choose the optimal culture conditions in terms of cell number, microgel matrix composition and culture time required to obtain in vitro more mature DP-like structures able to induce hair follicle neogenesis.
Finally, to induce hair follicle neogenesis, a method was developed to culture these 3D (DPS and FM) structures with skin cells -keratinocytes and fibroblasts- in a plasma-derived in vitro 3D skin model to promote hair follicle formation. For this purpose, DPS and FM were introduced, by means of ad hoc produced combs, in the dermal compartment of plasma-derived fibrin 3D skin cultures and then covered with epidermal keratinocytes. Combs were designed to mimic hair follicle patterning and positioning in human skin and were then 3D printed using stereolithography. However, hair follicle-patterning disrupted dermal compartment stability, reason why hydrogels were stabilized by increasing final fibrin concentration to 2.4mg/mL. Despite the modification of dermal compartment composition, correct epidermal differentiation was proven by positive expression of as K5, K10, Filaggrin, Loricrin and Involucrin, typical epidermal markers present in human skin. Although the mechanical stability of the fibrin dermal matrix was improved by increasing fibrin concentration, culture times lasted only up to three weeks, whis was not long enough to allow an adequate follicle induction. Despite this limitation, co-culture of hFDPc (either in DPS or FM) with human epidermal keratinocyte showed presence of hair-follicle like structures in the organotypic cultures after 2 and 3 weeks in culture. A thorough analysis of the cultures by histological and immunofluorescence methods revealed the presence of hair follicle-like structures at early stages of development when using both DPS and FM. Moreover, these structures showed positive expression of K14, K71, K75 and K15 present in the hair follicle outer root sheath, inner root sheath, companion layer and hair germ, respectively. Contrary to organotypic cultures containing DPS, those with FM rapidly degraded and only lasted for 3 weeks. Formation of hair follicle-like structures in the first stages of development is achieved with an induction efficiency of 30 and 50% for DPS and FM respectively. Although these results are far away from the ones reported in the literature, to our knowledge this is the first work in which human hair follicle differentiation is induced in vitro using both adult hFDPc and keratinocytes, crucial when developing regenerative therapies for clinical purposes.
Whereas further work is needed to achieve the goal of complete hair follicle differentiation, the potential of our experimental setting for in vitro hair follicle neogenesis using wild adult hFDPc either in DPS or FM must be highlighted. Moving forward in hair follicle differentiation first involves dermal compartment stabilization because of two main limitations: 1) higher hair follicle densities are required and 2) longer culture times for FM cultures require more durable matrices. On behalf of exploiting the benefits that human blood plasma provides, new ways to complement plasma-derived hydrogel with additional biocompatible materials such as alginate of polyethylene-glycol need to be explored. While in the case of DPS longer times did not provide with more mature hair follicles, this must be evaluated for FM cultures. In any case, the use of a mouse model to transplant with our organotypic constructs containing DPS/FM can help to elucidate whether all the above-mentioned limitations are due to the in vitro culture itself or because of any other limitation of the DPS/FM systems.
