Intensive international efforts are being made to devise new molecular therapies for genetic skin diseases.1 There are three major reasons for these efforts. First, the skin is the largest organ in the human body, with an area of almost 2 m2 and a weight of 2–3 kg. Therefore, diseases of the skin have a severe negative impact on the daily functions and quality of life of affected individuals. Second, a large number of individuals worldwide are affected by at least one of the nearly 400 known genodermatoses. Third, the skin is an ideal organ with which to study new therapeutic approaches, because it is easily accessible both for treatment and for macroscopic, microscopic, cellular, and molecular analyses. Its advantages have already been proven in practice by the first pilot study on treatment of human junctional epidermolysis bullosa with keratinocyte gene therapy.2 In this issue of Molecular Therapy, Remington et al.3 report on an interesting approach to treating dystrophic epidermolysis bullosa (DEB) in a mouse model using protein therapy. DEB is a genetic skin disorder that manifests with mechanically induced skin blistering and scarring of varying clinical severity. All individuals with DEB are confronted with lifelong skin fragility and impaired resistance of the skin and mucous membranes to external shearing forces, a situation that can lead to secondary symptoms including joint contractures, mutilating deformities of hands and feet, malnutrition, anemia, and growth retardation.4 A feared complication—caused by chronic blistering, massive scarring, and perturbed epidermal–dermal interactions5—is a drastically increased risk of skin cancer.4 DEB is caused by mutations in the COL7A1 gene, which encodes type VII collagen, a major component of the anchoring fibrils and an important adhesion molecule at the dermal-epidermal junction.6 Thus far, nearly 500 COL7A1 mutations have been reported, and homozygous, heterozygous, and compound heterozygous mutation constellations generate a broad spectrum of phenotypes.7 The clinical and laboratory findings in affected individuals, together with the effects of complete or partial inactivation of the Col7a1 gene in mice,5,8 clearly show that the main determinant of the blistering phenotype in DEB is the absence or reduced expression of type VII collagen due to gene mutations. Therefore, therapeutic approaches for DEB based on the restoration of type VII collagen expression, including gene, cell, and protein therapies, are at the center of interest for both researchers and patients (see refs. 2, 3, 9, and 10 and references therein). Remington et al. treated severely affected Col7a1 null mice with repeated intradermal injections of recombinant full-length human type VII collagen (type VII procollagen). The authors report that the injected collagen was incorporated into the dermal-epidermal junction in the Col7a1 null mouse skin but also diffused into distant sites and persisted in the skin for about 2 months. Restoration of this collagen at the epidermal basement membrane significantly improved the phenotype of experimental DEB, as shown by decreased skin blistering and prolonged survival of the animals. Remarkably, the treatment of the DEB mice with heterologous type VII collagen induced development of antibodies to this antigen, but these neither bound to the dermal-epidermal basement membrane nor prevented the formation of anchoring fibrils. Moreover, this specific antibody response could be inhibited pharmacologically by treatment with a monoclonal antibody to CD40L. The findings of this study are intriguing in that they show that injections into the skin of a monomeric type VII collagen—a component of highly specific supramolecular polymers, the anchoring fibrils—can result in phenotypic improvement, and they suggest that intradermal injections of recombinant type VII collagen could represent a straightforward treatment rationale for human DEB. At the same time, the results raise several new interesting questions, a few of which are discussed below. The first is how the injected recombinant type VII procollagen matures to collagen and polymerizes into functional anchoring fibrils. Type VII collagen is the largest member of the collagen superfamily.11 Under physiological conditions, it is synthesized by keratinocytes and fibroblasts as a procollagen polypeptide of about 300 kDa, which first forms a triple-helical trimer of 900 kDa and then undergoes a complex supramolecular aggregation process. During fibrillogenesis, the nascent type VII procollagen molecules dimerize in an antiparallel manner, the C-propeptides are removed by bone morphogenetic protein 1 (ref. 12), and the processed antiparallel dimers then laterally aggregate into anchoring fibrils (Figure 1). The adhesive function of the fibrils is strengthened by ligand interactions with basement membrane components and crosslink formation with fibrillar suprastructural elements in the upper dermis.13 It will be interesting to learn whether the injected human type VII procollagen is processed in this manner in mouse skin or whether other, hitherto-unknown molecular and/or cellular mechanisms contribute to its effects as dermal-epidermal glue and to increased epidermal adherence in Col7a1 null mice receiving type VII collagen injections. Figure 1 Schematic representation of the polymerization of anchoring fibrils. The type VII procollagen monomer consists of three α1(VII) polypeptide chains folded into a triple helix. Two monomers form an antiparallel dimer, from which the NC-2 propeptides ... The second question is what amount of recombinant type VII procollagen would be needed for effective treatment of DEB in humans. If a neonatal mouse weighing 1–2 g required weekly injections of approximately 50 μg of type VII collagen for a therapeutic effect, an infant with DEB who weighed 4 kg might need 100 mg of recombinant type VII collagen per injection. Under the conditions used in the mouse study, 100 mg would need to be dissolved in about 300 ml of buffer solution and administered repeatedly during the course of the disease. This poses a challenge for good manufacturing practice production of functionally active recombinant human type VII collagen in sufficiently high concentrations, given that all collagens are known to be insoluble by their very nature. The third important question is whether repeated injections of type VII collagen will result in significant adverse effects. In the study by Remington et al., circulating antibodies to human type VII collagen were generated, but these were not pathogenic. This observation is in line with the findings in a transgenic mouse coexpressing murine and human type VII collagen.14 In this mouse, coexistence of the two type VII collagens in the skin was well tolerated and did not produce a pathological phenotype. However, it remains unclear why the antibodies to human type VII collagen are not pathogenic in the Col7a1 knockout mice. Furthermore, would this necessarily imply that antibodies induced by protein therapy of human DEB will also lack pathogenic effects? It is well known that autoantibodies specific to type VII collagen cause a blistering phenotype similar to DEB both in individuals with EB acquisita (a human autoimmune bullous disorder) and in corresponding mouse models.15 A further challenge will be to find the best immunomodulatory approach to effectively curbing a putative unwanted immune response induced by protein therapy in DEB patients. Taken together, advances in designing effective therapy for DEB such as those presented in Remington and colleagues' report may decisively contribute to achievement of the goals for treatment of genetic skin diseases. However, several aspects of feasibility and potential unwanted adverse effects will need to be rigorously tested and optimized. Further studies should address the conditions defining the production of recombinant type VII procollagen in amounts and concentrations suitable for an effective protein substitution therapy in DEB patients. In addition, it will be necessary to analyze the supramolecular aggregation and ligand interactions of exogenously administered recombinant type VII collagen in the skin in order to understand the molecular mechanisms of protein therapy and to predict the duration of increased dermal–epidermal adhesion after treatment. Finally, the immune response triggered by protein therapy should be more closely investigated to exclude potential pathogenic effects and to develop effective ways to modulate autoantibody production. These studies should greatly facilitate translating and establishing protein therapy for the management of DEB and related genetic skin diseases.