Current and Future Perspectives of Stem Cell Therapy in Dermatology

1) Epidermal stem cells (EpSC)

Advantages of using EpSC for research, diagnostic and therapeutic purposes include their readily accessibility and relatively simple isolation from (bioptic) skin tissues in comparison to ESCs. EpSC are further considered to be less “artificial” than iPSCs. Immune rejection following autologous transplantation is not expected and the tumorigenicity of these cells is considered to be low, due to their lesser degree of potency and absence of (epi-)genetic manipulations. In contrast, iPSCs reprogramming with tumorigenic c-Myc increases the frequency of transformed cells during iPSC generation. Tumor formation risk increases when the c-Myc transgene remains in established iPSCs and becomes reactivated42.

EpSCs demonstrate further favorable features, such as their high proliferation rate with the ability to double their number within 3~4 days of culture. At the same time they are able to keep their potency and differentiation potential for longer periods, although progressive aneuploidy (a state in which cells have abnormal numbers of chromosomes) and polyploidy (a state in which cells have one or more extra fully duplicated sets of chromosomes) as well as accumulation of mutations occur after several passages in cell culture. Notably, ethical issues do not restrict their use. This is in contrast to the serious ethical concerns that arise in ESCs research when referring to human dignity and ideas of personhood along the creation as well as destruction of embryos as the earliest forms of human life specifically for research purposes. All these characteristics make skin derived adult SCs an ideal population for the use in SC-based therapies24, 43.

Grafts generated from autologous epithelial cultures that encase an appropriate number of EpSCs as holoclones were shown to permanently recover massive epithelial defects (e.g., in skin and ocular burns or epidermolysis bullosa). Therewith, EpSCs also prove to provide both, a cellular environment and normal ECM to mediate restoration of a normal dermal-epidermal junction44, 45, 46, 47.

2) Multipotent mesenchymal stromal cells (MSCs)

Since their first identification as fibroblast precursors in bone marrow in the 1950s, mesenchymal (stem) stromal cells (MSCs) have been obtained from several tissues, including adipose tissue, skin, umbilical cord blood, placenta, peripheral blood, endometrium, dental pulp, dermis, amniotic fluid, as well as from tumors48, 49. MSC of different origin share similar features but are not identical. Even in the skin several MSC subtypes exist. Regardless of their origin, MSCs possess a broad differentiating potential and some degree of plasticity, since they generate cells of not only mesodermal origin (i.e., osteocytes, adipocytes, chondrocytes, myoblasts, and tenocytes) but also of ectodermal (e.g., neurocytes, melanocytes) and endodermal lineages (e.g., hepatocytes, thyroid cells)50.

In 2006, the International Society for Cellular Therapy established guidelines for MSC characterization to counteract controversies concerning its name, definition, isolation and characterization criteria. The name “multipotent mesenchymal stromal cells” was favoured and three minimal criteria were delineated: i) adherence to plastic in culture; ii) expressing a combination of surface antigens (CD73⁺, CD90⁺, CD105⁺, CD34⁻, CD45⁻, CD11b⁻, CD14⁻, CD19⁻, CD79a⁻ and HLA-DR⁻); and iii) in vitro differentiation-capability into adipocytes, osteoblasts and chondrocytes51. However, MSC populations isolated from different tissues significantly differ in their proliferation, differentiation and molecular phenotype.

Besides the differentiation-capability in vitro, the trophic, paracrine and immunomodulatory functions of MSCs are those that hitherto may have the biggest therapeutic implication in vivo6, 52, 53, 54.

One of the main functions of MSCs is to support repair of damaged tissues. In response to inflammation MSCs migrate towards injured sites, differentiate into cells (mainly fibroblasts) and operate through the release of molecules participating in tissue regeneration such as cytokines (i.e., PGE2, GM-CSF, interleukin [IL]-1, RA, IL-7, IL-8, IL-10, and IL-11), growth factors and chemokines. In addition, MSCs modify tissue healing through pro-angiogenic, anti-fibrotic, and anti-apoptotic pathways. In stromal vascularized tissues, their perivascular amount correlates with the blood vessel density and the number of pericytes as mesenchymal progeny50, 55.

The immunomodulatory abilities of MSCs reside on the secretion of anti-inflammatory cytokines and the inhibition of CD4+ and CD8+ T cell, B-cell, and natural killer (NK) cell proliferation. These features depend on the microenvironmental milieu that MSCs encounter after their application. Thus, MSCs have been shown to exert even opposite effects in response to different inflammatory cues50, 56.

Although it is not fully determined whether MSCs are immunoprivileged or immunoevasive, they are specified as hypo-immunogenic due to their menial expression of major histocompatibility complex (MHC) class I molecules, as well as lack of MHC class II and co-stimulatory molecules, inclusive CD80, CD86, and CD40. These characteristics reduce the risk of immune rejection so that MSCs are considered to be safe when used in an allogeneic environment57, 58, 59.

Up to date more than 600 clinical trials are listed in the U. S. National Institutes of Health database (www.clinicaltrials. gov) dealing with MSC therapy for various diseases such as different forms of cancer, spinal cord injury, multiple sclerosis, Parkinson's disease, myocardial infarction, rheumatoid arthritis and graft-versus-host-disease (GVHD)60. Current research strategies try to increase the efficiency of therapeutically administered MSCs by mainly encountering their limited persistence in vivo50, 59. Various methods have been applied to generate “optimized” MSCs, including genetic modification through viral and non-viral modifications, bioengineering of surface receptors, and priming with biological agents. For example, MSCs activated by nucleotide oligomerization domain 2 (NOD2; involved in the regulation of differentiation of umbilical cord derived MSCs and able to modulate inflammatory responses) or MSCs overexpressing SOD3 (a powerful antioxidant molecule) have been shown to exert much higher therapeutic efficacy than naïve MSCs in experimental immune modulatory models of atopic eczema and psoriasis, respectively. Although a confirmation of these results in the clinic is still missing, development of exceedingly efficient MSCs with augmented benefit and minimum risk along genetic modifications gives promising therapeutic perspectives59, 61.

3) Bone marrow stem cells

Bone marrow comprises at least two different lineages of cells: hematopoietic and associated supporting stroma with mesenchymal cells. Hematopoietic cells are produced by hematopoietic stem cells (HSCs), which are situated in the bone marrow SC niche. The mesenchymal compartment contains a subset of cells (1 in 10⁷ to 10⁸) with probably (pluri-)multipotent differentiation capacity, referred to as MSCs.

4) Bone marrow derived mesenchymal stromal cells (BM-MSCs)

The BM-MSCs are similar but somewhat different to mesenchymal stromal cells isolated from other tissues. The former can be isolated, enriched and transfused into allogeneic or autologous recipients along bone marrow transplantation (BMT) and exert a substantial role in producing erythrocytes, leukocytes, and platelets. They show also plasticity with their ability to differentiate into tissues of mesodermal, endodermal, and ectodermal origin, including skin62, 63, 64 and have been implicated to contribute to skin development65. Nevertheless, the nature and function of these cells is still beeing controversial discussed.

BMT has been introduced as a kind of SC therapy over four decades ago. Although the initial implementation were for leukemia, lymphoma, hemoglobinopathies and aplastic anemia, the medical indications for BMT have extended and roughly 50,000 people worldwide receive this form of therapy annually63. Nevertheless, safety issues limit the application of BMT, since high morbidity and significant mortality is associated with BMT, even aggravated in patients with increased susceptibility to infections and a higher risk of tumor development66.

This therapeutic application has increasing relevance for skin diseases since several studies have reported that the BM poses a source of fibroblast-like cells in the dermis (along hematopoietic and mesenchymal lineages) and that their number augments in skin after wounding67, 68.

In addition, BM SCs may also serve as a reservoir for skin epithelial cells67. After BMT, donor cells differentiating into keratinocytes were detected in human epidermis of recipients for at least 3 years before vanishing. However, such BM-derived keratinocytes seem to be an extremely rare finding, perhaps contributing to only ~0.0001%~0.0003% of all epidermal cells in this setting69. Since BM-derived epithelial cells are sparse, the physiological role of BM cells in regeneration of the skin has been called into question.

Potential drawbacks of BM-MSC therapy refer to immune modulating abilities in context of a tumor microenvironment leading to an unfavourably alteration of anti-tumor response and angiogenesis. Furthermor MSCs may serve as precursors of tumor-associated fibroblasts and possess the capability to skew neutrophils and inflammatory monocytes or tissue macrophages into an immunosuppressive and tumor-promoting phenotype20, 70.

5) Induced pluripotent stem cells

Reprogramming of somatic cells to iPSCs provides an important (and ex vivo infinitely expandable) cell source to develop customized, patient-specific cells with a broad spectrum of cellular phenotypes for potential therapeutic applications7.

Skin cells like dermal fibroblasts, keratinocytes, dermal papilla cells or melanocytes are preferentially used for this technique, since they are easily accessible in the patient via isolation from punch biopsies. Especially fibroblasts further have plain culture conditions. Adult adipose SCs, yielded via lipoaspiration, pose another source for iPSCs71. The differentiation of both, mouse and human iPSCs into keratinocytes72, melanocytes73, and fibroblasts74 has already been successfully shown. This thus opens the possibility of extending iPSC technology into the field of dermatology17.

Interestingly, fibroblasts differentiated from iPSCs may display specific properties that exceed those of the parental fibroblasts from which these iPSCs were originally reprogrammed, such as an increased production and assembly of ECM75. Acquisition of an augmented biological potency of modified cells when compared to their parental origin is probably related to a modified epigenetic signature following differentiation of iPSCs and is an important functional feature for using these cells in regenerative therapies.

Fibroblasts are essential in maintaining normal tissue homeostasis and wound repair through their synthesis of ECM proteins and secretion of growth factors. Their incorporation into tissue-engineered biomaterials seems promising for the use in repairing damaged or diseased tissues by fabricating dermal substitutes74. In this context, iPSC-derived fibroblasts offer a novel source of autologous cells for dermal regeneration.

Although iPSCs have enormous potential for cell-based drug designs, cell therapy, and disease modeling, their transition into the clinic is still hindered by the missing evidence of safety and reliability of the reprogramming technology.

Although cell identity can be modified by the exogenous expression of transcription factors, the efficiency of nuclear reprogramming remains low (0.1% to 3%). This low outcome is probably associated with residual epigenetic memory of the tissue from which iPSCs were derived, detected via gene profiling studies in iPSCs. It is known that differentiated somatic cells have distinctive epigenetic patterns to maintain their cell identity. Cellular reprogramming works to change this epigenetic status of differentiated cells back to an undifferentiated state. Further, there is evidence that through the reprogramming process a restructuring of the existing somatic epigenetic memory takes place, followed by the generation of a new “epigenetic signature” adapted to the type of cell to be differentiated76, 77, 78, 79.

In addition, currently available cell purification technologies may not fully succeed in separating the differentiated cells from undifferentiated iPSCs. Undifferentiated or partly differentiated iPSC could consequently be transplanted into the patient, carrying an increased risk of tumor/teratoma formation80.

Furthermore, it remains unclear to what extent the reprogramming process affects the genomic integrity of a cell. Several recent genomic analyses have signified that genomic abnormalities such as the accumulation of mutations and aberrant DNA methylation of distinct single bases emerge in iPSCs, either by the reprogramming process or following culture conditions81, 82. To address this issue, genome integration-free approaches are already widely used aiming at the reduction of the tumorigenic risk of insertion mutagenesis83, 84. However, it is necessary to perform more extensive and thorough genomic and epigenetic studies before using iPSCs in the clinic.

Interestingly not only iPSCs but also dermal fibroblasts themselves were demonstrated to have features of in vitro pluripotency without the necessity to be reprogrammed back to immaturity via activation of embryonic stage genes. Canadian researchers yielded a hematopoietic progenitor cell from a fibroblast through the in vitro implementation of specific cytokines. Consequently this hematopoietic precursor cell was able to give rise to granulocytic, monocytic, megakaryocytic and erythroid lineages, besides exhibiting the capability to repopulate the bone marrow by grafting85.

A recent study shows that terminally differentiated cells descending from mouse iPSCs do not provoke an immune response in syngeneic recipients86. This suggests that iPSC-derived cells might be well tolerated by the immune system. Again, more studies will be needed to definitively exclude a iPSC-mediated immune response in patients.

Immune rejection related to iPSC-based genetic correction is another problematic aspect, especially in skin diseases with homozygous null mutations of relevant genes. If a protein, that is unfamiliar to the host, is reintroduced, it may provoke an immune response and rejection of corrected iPSC-derived cells. This problem might be mitigated via prescreening for patients with compound heterozygous mutations or expression of nonfunctional, truncated variants of the protein of interest80, 87.

1) Epidermal stem cells (EpSCs)

EpSCs are a convenient target to use in wound therapies because they already reside within the skin and participate in the normal healing response. They have been shown to support healing by increasing proliferation and migration of fibroblasts and keratinocytes as well as enhancing angiogenesis by human vascular endothelial cells. In a type 2 diabetic nude mouse model, subcutaneous injection of allogeneic EpSC into acute full-thickness wounds have significantly shortened the healing time89. In a clinical study, a significant decline in wound size with increased reepithelialisation and vascularisation signs on histology taken after 4 months has been observed in 10 patients with non-healing leg wounds, after receiving autologous scalp-end terminal hair follicular grafts90. Terminal HFs (as a major SC niche) seem to play a decisive role in epidermal regeneration. This assumption is supported by a randomised controlled trial that compared the implantation of grafts containing scalp HFs with non-hairy skin grafts on chronic wounds in 12 patients. A significant reduction in wound size in the terminal HF-treated group was described91, 92. Notably, outer hair shafts diminish in most cases after a couple of weeks, hampering the development of HFs and subsequent cosmetic impairment by growing hair in engrafted transplants. Bulge SCs are highly relevant for epidermal renewal. In response to skin injury, these cells mobilize, leave their SC niche and contribute to repopulation of the epidermis while their progenies behave as transient amplifying cells with short lifespans37.

2) Multipotent mesenchymal stromal cell (MSC) therapy

MSC therapy is another emerging option to treat acute and chronic non-healing wounds. Beneficial effects are accomplished through structural repair via cellular differentiation, immunomodulatory responses, direct secretion of growth factors, advanced neovascularization and reepithelialization, as well as mobilization of resident SCs. Thereby, MSC play a pivotal role in all three healing phases. At the wound margins they stimulate the formation of granulation tissue by enhancing epidermal cell proliferation and growth of new blood capillaries. Further, endothelial cell recruitment is stimulated through the release of pro-angiogenic factors and growth factors such as vascular endothelial growth factor and angiopoietin-1. MSCs modify tumor necrosis factor-α production and lower NK cell function in the inflammatory phase, thereby reducing interferon-γ activity. In the last healing phase, scar formation is reduced through PGE2 secretion and lowering of TGF-β1 to TGF-β3 ratio, IL-10 up-regulation as well as IL-6 and IL-8 down-regulation. These effects are accompanied by a decline in collagen production and fibrosis50, 93. In addition, MSCs exert also antimicrobial activity via secretion of antimicrobial proteins or immune-modulating factors50, 94, 95.

Several studies have already confirmed the positive effects of MSC (in particular MSC isolated from the skin, fat and bone marrow) on healing of acute and chronic wounds via induction and acceleration of regenerative processes50. Preclinical data demonstrated that local injection of BM-MSCs into an incisional full-thickness wound significantly shortens the healing time while stimulating angiogenesis, reepithelialization, and granulation96. Accelerated wound healing of diabetic ulcers has been also shown in preclinical and early human trials when BM-MSCs were used91, 94, 97.

In an additional clinical study, autologous BM-MSCs were topically delivered via a fibrin spray to acute surgical wounds and chronic lower extremity wounds98. When assessed 20 weeks later, an accelerated healing of acute wounds and a significant reduction in size or complete healing of chronic wounds was observed. Efficacy of autologous BM-MSCs was further provided in the treatment of chronic non-healing ulcers of the lower extremities (diabetic foot ulcers, Buerger disease), reflected by a significant decrease in ulcer size compared to standard wound care99.

Several studies investigated the wound-healing effect of human adipose-derived mesenchymal stem cells (AD-MSCs) both in vitro and in vivo. AD-MSCs are clinically attractive as they can be easily isolated in relatively high quantity and possess satisfactory recovery potency100. In addition, it has been shown that the proliferative phase of healing involves the repopulation of adipocytes within skin wounds. Likewise, an in vivo mouse study suggested that immature adipocytes are activated during the proliferative phase alongside with mature adipocytes and fibroblast migration. Interestingly, lipoatrophic mice display impaired wound healing in comparison to controls, supposing adipocytes to be key elements of the intercellular communication during wound repair that mediate fibroblast migration and function101. The wound-healing effect of AD-MSCs by reducing wound size and accelerating reepithelialisation has been confirmed in several studies102, 103.

However, limitations to this data include short durations of in vivo experiments, small sample sizes, short follow-up periods, lack of randomized control trials and the use of animal model, as it is not always possible to directly extrapolate findings to the human wound physiology104.

3) Cell- and collagen-derived dermal scaffolds

Because the delivery of MSCs through direct injection or topically through gel matrices is detrimental for cell survival and usually causes significant rapid cell death, new strategies have been developed to improve MSC cell adhesion, proliferation and migration. These techniques are based on the use of MSC-seeded micro-or nanostructured scaffolds with natural biomaterials, such as collagen and cellulose derivatives93, 94, 105, 106. Thereby, pronounced to complete regeneration of non-healing wounds (burns, decubitus ulcers, diabetic ulcers) has been reached in preclinical and clinical studies107, 108.

For example, electrospun nanofibers of collagen and poly (d,l)-lactic-co-glycolic acid (PLGA) containing BM-MSCs were used to mimick the multilayer structure of the skin in the treatment of full-thickness skin wounds109. While the electrospun scaffold (i.e., highly porous) mesh constructs (generated by electospinning techniques using high-voltage electrostatic fields) provided mechanical support and protection of the wounds against external stresses, the hydrogel offered a physiological environment for MSCs proliferation. After implantation of collagen-PLGA scaffolds in vivo, MSCs promoted collagen synthesis and reepithelialization of the injured skin, while the biomaterial scaffolds were slowly degraded93, 110.

1) MSC-cell therapies

SC therapies are increasingly established in the experimental treatment of genetic diseases, recently also in patients with recessive dystrophic EB (RDEB), a rare genetic blistering disease. RDEB patients lack genes for synthesis of Collagen VII. These symptoms mainly account for more severe disease complications, such as mitten deformities of hands and feet and aggressive epithelial cancers.

Therapeutic approaches, either with intravenous infusion or direct local administration of MSCs to chronic wounds have started however, to provide novel insight into key BM cells and mechanisms germane to repair and regeneration of the epithelium.

Initially, intradermal injections of human BM-MSCs showed a dose-dependent, significant higher production and in loco deposition of type VII collagen associated with restoration of immature anchoring fibrils and superior dermal-epidermal integrity compared to controls with intradermal phosphate-buffered saline-injections in DEB mouse models123, 124, 125.

In line with this preclinical data, Conget et al.126 described the replenishment of type VII collagen at the dermal–epidermal junction upon intradermal BM-MSC injection provided from healthy donors into chronic wounds of two RDEB individuals. The administration led to a reduced blister formation (up to 6 months) and an increased reepithelialisation of chronic ulcers. Tissue remodeling activity of the transplanted MSCs, owing to both, their integration into the skin and their secretion of growth factors and cytokines participating in the regulation of tissue regeneration, might activate self-healing mechanisms in RDEB skin.

Subsequent studies questioned a predominant impact of SC on phenotypic amelioration. Petrof et al.127 showed that a single intradermal injection of allogeneic fibroblasts accelerated the initial rate of wound healing in patients with RDEB within the first 28 days although this effect diminished thereafter. In another study, both the injection of allogeneic cultured fibroblasts in suspension solution as well as of the suspension solution alone led to a similar increase in type VII collagen expression and improved wound healing in chronic non-healing ulcers of RDEB patients, independently of type VII collagen regeneration128.

The mechanical stimulus in the course of intradermal injection itself was thus supposed as a potential cause of improved wound healing, reflecting an elevated expression of heparin binding-EGF-like growth factor in response to subclinical inflammations that occur after injections into the human skin127.

Providing a clinically feasible approach of systemic treatment, also the infusion of allogeneic BM-derived SCs led to a phenotypic improvement in RDEB patients, showing reduced blistering and tissue fragility. Transplanted SCs from healthy donors are believed to engraft into the skin, differentiate to fibroblasts and keratinocytes, synthesize the missing type VII collagen anchoring fibers and thereby improve the fragile skin and shearing off of the epidermis63. In this respect, it has been shown that low levels of restored expressions of collagen VII (below 30%) are sufficient to significantly improve the RDEB phenotype and even a correction of about 3% COL7A1 mRNA seemed to be adequate for phenotypic reversion in a mouse model129, 130.

Ten RDEB children who received systemic (intravenous) allogeneic (wild-type) BM-MSCs showed improved wound healing, reduced skin redness, a subjective improvement in quality of life and high tolerability. As skin biopsies at 2 months post-treatment revealed no increase in type VII collagen nor new anchoring fibrils, phenotypic improvement possibly reflected a predominant immune suppressive and immune modulating effect of MSCs131. In another study with RDEB patients fewer blistering and significantly shortened healing time after treatment with BM-MSCs was demonstrated, either with or without concomitant cyclosporine132. In this cohort, electron microscopic examination additionally confirmed an increase in anchoring fibrils after treatment.

In summary, intradermal or intravenous injections of MSCs have shown some clinical benefits for RDEB patients. However, feasible application techniques as well as optimal cell dosage and frequency schedules for administration of allogeneic MSCs still have to be established and evaluated in future clinical trials. Further, biological and pratical limitations, like the ephemerality of the transplanted cells and potential immune rejection towards neo-antigens hitherto hinder the clinical applicability.

2) Bone-marrow transplantation

Following the demonstration of successful BM SC transplantation in murine RDEB, a clinical trial of whole BMT was performed in children with RDEB133. The first step included a high-dose chemotherapy to immunoablate patients to ensure more dependable lymphohematopoietic engraftment. Unfiltered cell populations of the donor BM were used based on previous observations that both hematopoietic and mesenchymal BM-derived SCs have the potential to increase the production of C7134. All six patients, who underwent BMT had some clinical improvement and five of the six showed increased C7 at the DEJ at the time biopsies where taken between day 100 and 200. Three of the six individuals showed an immense clinical improvement, with a reduction from 50% blistering area of the body surface to less than 10%. The remaining three patients exhibited a moderate improvement with less than 25% of the body surface area affected. However, toxicities occurred, as one patient died before the BMT due to heart failure possibly related to cyclophosphamide toxicity and pre-existing renal failure. Another individual died 6 months after transplantation because of an infection related to graft failure. Therefore, BMT protocols are currently refined by considering the use of reduced intensity conditioning and targeting of distinct subpopulations of BM cells implicated to be more effective, such as BM-derived Circulating PDGFRα⁺ mesenchymal cells)63, 134.

BMT has been also investigated in other forms of EB including severe generalized junctional epidermolysis bullosa (JEB)135. An important consideration in this form is, however, that the defective protein (laminin-332) is not synthesized by fibroblasts but keratinocytes. Thus, whether rectified donor BM-derived cells in the dermis are able/sufficient to regenerate the missing basement membrane protein, or a concurrent correction of epithelial cells is essential, still remains to be determined.

BM-derived MSCs were previously shown to differentiate into keratinocytes. Likewise, BM-derived hematopoietic and MSCs were demonstrated to be able to produce laminin- 332. Former clinical studies have further indicated that BMT increases laminin-332 expression at the dermoepidermal junction in some individuals136. Also other epithelial synthesized proteins involved in forms of EB associated with severe morbidity may be therapeutically targeted by BM-derived stem-cells. Thus, phenotypic improvement of type XVII collagen-deficient mice after BM-SC therapy suggests human studies in generalized intermediate junctional EB with deficient type XVII collagen63, 137, 138, 139.

3) Gene therapy

A low worldwide incidence of rare (or “orphan”) genetic skin diseases limits the accrual of data for research and suspends advancements in developing therapeutic concepts for these skin conditions. But since all of the most devastating forms of such inherited skin diseases like EB are mostly resulting from monogenic defects, in theory, their remediation at the genetic level should be more feasible. Sustained gene correction of continuously renewing tissue like skin, however, relies on the efficient molecular targeting of SCs.

A proof-of-principle study with somatic gene therapy in EB has been published44. Via a retroviral vector, a LAMB3 transgene has been inserted into autologous keratinocytes that were then grafted back as a confluent sheet onto the thigh of an adult patient with generalized intermediate junctional EB. In the following years, the graft has continued to express laminin-332 at the dermoepidermal junction, leading to a clinical improvement and long-term epidermal stability. This remarkable benefit was achieved although the number of holoclone SCs was reduced in the patient, probably a repercussion of long-term skin blistering resulting in niche destruction and SC depletion or exhaustion63, 140.

In the future, optimized protocols, with the goal to effectively isolate a sufficient amount of autologous EpSC, before being corrected by gene transfer, might facilitate the procedure. These cells, after building grafts and being transplanted, have the potential to induce long-term (if not permanent) regeneration of wounded skin. So far, promising observations of the use of genetically corrected skin grafts include (1) a total engraftment, yielding a morphologically and functionally normal, non-blistering skin that is able to resist mechanical stress, (2) continuing laminin beta-3-protein and laminin 332 expression that consequently strengthens the epidermal-dermal junction over a period of at least 8 years, (3) the ability to persistently and effectively restore the epidermis with only a few transgenic EpSC, which accounted only for a small subpopulation of transduced cells (most transduced keratinoctes have been shorter-living transit-amplifying progenitors)141.

Additional gene therapy studies, notably for RDEB and junctional EB, are currently at preclinical or early clinical stages. They mostly include keratinocyte grafting following ex vivo the introduction of a transgene or correction of both keratinocytes and fibroblasts with subsequent generation of a skin equivalent. Approaches to local gene therapy also include the injection of genetically corrected autologous fibroblasts harbouring the reparative transgene. Thereby potential pitfalls of wound infection (along the transplantation of skin equivalents) and graft loss (due to lower immunogenicity) in RDEB should be avoided. In contrast, intravenous gene therapeutic (systemic) approaches may be more feasible, tolerable and of systemic impact since the systemic nature of the disorder can be addressed63.

With regard to hitherto limited and/or transient efficacy as well as safety/tolerability issues, , optimal therapy for EB is doubtful to involve just allogeneic cells. More promise may hold a combination of gene, protein, drug, and cell therapies.

4) iPSCs

iPSCs therapies are therapeutically promising for genetic skin disease, because they can be rather easily exploited to be genetically manipulated. In addition, this approach makes modeling of skin diseases via targeted mutagenesis of the relevant genes possible forgoing the usage of ESCs142.

The successful establishment of iPSC-based therapies for hereditary skin diseases mainly relies on four steps: First, via a patient's skin biopsy cells have to be isolated. Second, these cells have to be transformed into iPSCs via genetic reprogramming. Third, genetic aberrations in obtained iPSCs have to be (safely) corrected, preferably through homologous recombination (HR). Fourth, these genetically corrected patient-specific iPSCs need to be differentiated into the desired cell type, followed by transplantation onto the same patient as an autograft.

Notably, even uncorrected iPSCs are valuable as disease- relevant patient-specific cells. They are used to customize in vitro disease models and in vivo xenograft models to gain new insights into disease mechanisms and drug discovery72.

Up to now, human iPSCs have been generated from several differentiated cell types derived from patients with certain genetic skin disorders. Besides type VII collagen (Col7)-deficient RDEB143, 144,LAMB3 gene-deficient JEB145 and EB simplex (EBS) with a dominant R125C keratin 14 mutation, these include, for instance, p63 mutant ectodactyly, ectodermal dysplasia, and cleft lip/palate syndrome, epidermolytic hyperkeratosis with a dominant N188S keratin 1 mutation, and dyskeratosis congenita, a multisystemic telomere disorder with several pathogeneic gene mutations146, 147.

In cell culture models, successfully generated iPSc from either gene corrected (autologous) RDEB fibroblasts or healthy (allogenic) individuals are able to differentiate into hematopoietic SCs and MSCs that can home to mucocutaneous blistering areas where they differentiate into keratinocytes and fibroblasts69, 143. Additionally, autologous grafting of in vitro generated 3-dimensional (3D) skin equivalents by iPSCs shows generation of stratified epidermis in vitro and in vivo (animal models)148, 149.

Up to now, several groups were able to show the therapeutic potential using corrected patient derived iPS cells in RDEB therapy studies in vitro17. In one study the correction of JEB-keratinocytes by iPSCs was done through a LAMB3-encoding lentivirus145. Another study used transfection with an expression plasmid encoding the wild-type human COL7 gene to yield a transient gene correction of COL7A1 gene-deficient RDEB-iPSCs144. The manipulated RDEB-iPSCs differentiated into structures resembling skin as well as cells of a hematopoietic lineage for BMT. In summary, corrected iPSCs have the potential to pose an unlimited source of autologous cells of both epidermal and mesenchymal lineages for the treatment of RDEB and perhaps other hereditary skin diseases150.

Nevertheless, this molecular approach harbours the risk of mutagenesis due to the application of viral and non-viral mediated gene correction and long-term cultivation involving multiple passages. Extensive genetic analysis (including whole genome/epigenomic assessment) will be necessary to purge the concern of off-target events that may occur after correction of iPSCs with techniques like ZFNs, TALENs and CRISPR/Cas systems. These drawbacks significantly impair the current clinical feasibility of iPSc-based strategies of gene correction17.

To overcome these obstacles, spontaneous reverted skin cells, so-called patient-specific naturally gene-reverted induced pluripotent stem cells, could be an alternative source of genetically corrected cells suitable for transplantation in patients with RDEB. This approach relies on the spontaneous reversions of inherited mutations that can be seen in different forms of EB, like RDEB, JEB, EBS (and many other genetic diseases and inherited skin disorders)151. Revertant mosaicism leads to somatic mosaicism that shows up as areas of normal skin in patients otherwise overwhelmingly affected by the disease-specific phenotype. This pattern is caused by a clonal outgrowth of cells harbouring acquired secondary mutations that reverse the effects of the primary mutation so that the phenotype locally diminishes. Several mechanisms are known to underlie this second “correcting” gene mutation in revertant cells, including gene conversion, second-site mutation and intragenic crossovers. Thus, if spontaneously revertant skin cells in EB are used to generate iPSCs, the risk of tumorigenesis via e.g. insertional mutagenesis after differentiation of these cells into hematopoietic (mesenchymal) or epidermal cells is reduced. Moreover, an unlimited source of naturally corrected cells for a cell replacement therapy could be obtainable151, 152.

The first attempt to use revertant cell therapy in an individual with generalized intermediate junctional EB yielded no functional benefits after (successful) grafting of isolated revertant keratinocytes, which were expanded to epidermal sheets. Of note, cultured keratinocytes showed 30% reversion, whereas the number of reverted keratinocytes dropped to 3% in the graft, probably because of lacking holoclones153. An alternative approach using punch graft transplantation of revertant skin, however, has been used successfully to heal chronic erosions with enhanced expression of laminin-332 in a patient with a similar form of junctional EB and mutated LAMB3 gene. The improved skin integrity was maintained for at least 18 months154.

The future impact of this approach thus relies on methods to more efficiently expand revertant keratinocytes in culture and to generate grafts containing adequate numbers of revertant SC to yield functional repair and regeneration of the skin.

1) Cancer stem cells

Several years ago, it was discovered that a small sub-population of acute myeloid leukemia cells could reestablish tumors in severe combined immunodeficiency mice, while the vast majority of the tumor cells could not158. This study underlies the cancer SC hypothesis, which implicates that cancer SCs have characteristics comparable to the SC population of their tissue of origin (i.e., self-renewal, differentiation potential). They are assumable very rare within the tumor and are thought to produce progenitor cells that can generate all types of cells comprising the tumor. CSCs pose a challenge for cancer therapies, because eradicating the bulk tumor usually does not include all CSCs, leaving enough of them at liberty to re-establish the complete heterogeneity of cancer tissue.

In addition, these SCs might be more resistant to chemotherapy, and even targeted molecular therapies via their relatively high expression of the multi-drug resistance genes (e.g., MDR-1, BRCP1), a common feature of many SCs. Furthermore, owing to their ability to rapidly induce DNA repair mechanisms, CSCs are often highly resistant to radiation therapy. Finally, they appear to be particularly adept in stimulating angiogenesis, nurturing tumor development22.

CSCs can switch between quiescence (tumor dormancy) and active cell division with subsequent varying chemosensitivity, a behavior that mainly depends on changes in the microenvironmental niche and involves complex signaling pathways regulating tumorigenic growth and dormant arrest159.

CSCs further possess the capability to create new niches during the metastatic process160, 161. These “metastatic niches” are defined by specific locations (e.g., metastatic cells occupying native SC and perivascular niches), signaling pathways (e.g., PI3K-AKT pathway as a critical survival input for metastatic cancer cells), incorporated stromal (e.g., endothelial) cell types and ECM proteins (e.g., tenascin C which strongly promotes SC functions). The components of this micromilieu support the survival, self-renewal and expansion of disseminated metastatic CSCs.

Beside melanocytic SCs or melanoma cancer SCs probably involved in the pathogenesis of melanoma, CSCs have also been demonstrated in non-melanoma skin cancer such as squamous cell carcinoma (SCC) in mice. The proliferation and expansion of these CSCs are markedly influenced by their ability to respond to TGF-β receptor II and integrin/focal adhesion kinase-mediated signaling at the tumor–stroma interface. This pathway is crucially important in human cancer. It acts initially tumor suppressive (inhibition of proliferation) but promotes metastasis in later stages in response to a tumor associated altered cellular context and variable environmental signaling profiles162. Studies have revealed that several distinct CSC populations coexist in SCC and that tumor initiation and metastatic potential of these populations can be uncoupled. Therefor understanding CSC biology it is critical to develop novel CSC-targeted therapies, especially for patients with cancer and a poor prognosis163.

New therapeutics may be designed to specifically target these cells to block cancer progression. At the moment many chemotherapeutics attack rapidly dividing cells, so that it is easy for slowly dividing cancer SCs to evade these therapies40. Whether skin tumors like melanoma follow a cancer SC model for tumor development or a hierarchical model of tumor growth and progression (or combinatory/ other models) remains to be determined. These features of tumor dynamics, however, have implications on drug development in order to increase the efficacy of CSCs targeting19.

Cancer SCs from solid tumors usually express organ-specific markers. However, many caveats impede the discovery and identification of cancer SC markers for diagnostic and therapeutic purposes, to include the potential that the expression of these cell surface markers is not stable, that daughter cells may express different markers, that markers may not be unique to the cancer SCs but expressed in other cell types as well and that these surface protein markers may not have any role in cancer SC biology.

Beside the isolation of CSCs by flow cytometry according to CSC-specific cell surface markers, CSCs can be identified by so-called “side population chains (SP)” within a tumor. The latters refer to a subpopulation of tumor cells that is highly conserved in human cancer cell lines and linked to SC characteristics (clonogenic). It further features drug transport property with multidrug resistance and might serve as a “evolutionary backup” to keep alive at least a sub-fraction of cells when exposure to cytotoxic compounds occurs. SP show differential efflux activity to the main cell population usually measured by efflux of the fluorescent DNA binding dye Hoechst 3334. Moreover, CSCs may be determined through sphere assays, since tumorigenic cells showing SC characteristics have the ability to grow as floating spheres in serum-free medium19.

The significant role of aberrant Wnt signaling in cancer and CSC has engendered substantial efforts into the development of therapeutic approaches to target this pathway. Several small molecules, involved in tumor signaling, have been identified that selectively block the p300/β-catenin interaction, thereby increasing the CBP/β-catenin interaction, which maintains long-term pluripotency in a variety of SC populations. Thus the therapeutic potential of CBP/β-catenin antagonists (e.g., ICG-001) has been studied in various preclinical tumor models, where it has demonstrated the ability to safely eliminate drug-resistant tumor-initiating cells164.