Bullous pemphigoid (BP) is a major autoimmune blistering skin disorder, in which a majority of the autoantibodies (autoAbs) target the juxtamembranous extracellular noncollagenous 16A domain (NC16A) domain of hemidesmosomal collagen XVII. BP-autoAbs may target regions of collagen XVII other than the NC16A domain; however, correlations between epitopes of BP-autoAbs and clinical features have not been fully elucidated. To address correlations between the clinical features and specific epitopes of BP-autoAbs, we evaluated the epitope profiles of BP-autoAbs in 121 patients. A total of 87 patients showed a typical inflammatory phenotype with erythema and autoAbs targeting the anti-NC16A domain, whereas 14 patients showed a distinct noninflammatory phenotype, in which autoAbs specifically targeted the midportion of collagen XVII, but not NC16A. Interestingly, this group clinically showed significantly reduced erythema associated with scant lesional infiltration of eosinophils. Surprisingly, 7 of the 14 cases (50.0%) received dipeptidyl peptidase-IV inhibitors for the treatment of diabetes. Dipeptidyl peptidase-IV inhibitors were used in 3 of 76 (3.9%) typical cases of BP with autoAbs targeting NC16A; thus, dipeptidyl peptidase-IV inhibitors are thought to be involved in the development of atypical noninflammatory BP. This study shows that the autoAb profile differentiates between inflammatory and noninflammatory BP, and that noninflammatory BP may be associated with dipeptidyl peptidase-IV inhibitors.

It has long been recognized that the hair follicle growth cycle and oscillation in the thickness of the underlying adipocyte layer are synchronized. Although factors secreted by adipocytes are known to regulate the hair growth cycle, it is unclear whether the epidermis can regulate adipogenesis. We show that inhibition of epidermal Wnt/β-catenin signaling reduced adipocyte differentiation in developing and adult mouse dermis. Conversely, ectopic activation of epidermal Wnt signaling promoted adipocyte differentiation and hair growth. When the Wnt pathway was activated in the embryonic epidermis, there was a dramatic and premature increase in adipocytes in the absence of hair follicle formation, demonstrating that Wnt activation, rather than mature hair follicles, is required for adipocyte generation. Epidermal and dermal gene expression profiling identified keratinocyte-derived adipogenic factors that are induced by β-catenin activation. Wnt/β-catenin signaling-dependent secreted factors from keratinocytes promoted adipocyte differentiation in vitro, and we identified ligands for the bone morphogenetic protein and insulin pathways as proadipogenic factors. Our results indicate epidermal Wnt/β-catenin as a critical initiator of a signaling cascade that induces adipogenesis and highlight the role of epidermal Wnt signaling in synchronizing adipocyte differentiation with the hair growth cycle.skin | niche cross-talk | stem cells M ammalian skin is a complex organ composed of a variety of cell and tissue types, including interfollicular epidermis, hair follicles (HFs), melanocytes, nerves, blood vessels, muscles, fibroblasts, and adipocytes. The development and patterning of these cells and tissues are governed by intercellular communication (1-3).One well-known example of this communication is the link between the HF growth cycle and the oscillation in thickness of the dermal adipocyte layer (4-6). When HFs grow deep into the dermal adipocyte layer in the anagen (growth) phase of the cycle, the adipocyte layer dramatically increases in thickness. This event reflects both increased adipogenesis and hypertrophy of individual adipocytes (7). When HFs regress (catagen phase) and enter the resting (telogen) phase, the adipocyte layer becomes thinner. The growth cycle of rodent HFs is coordinated to form waves of hair growth that traverse the body and the thickness of the skin adipocyte layer oscillates in synchrony with these waves (3).The synchronized patterns of HF growth and expansion of dermal fat correlate with the activation of the canonical Wnt pathway, which is well established to positively regulate anagen (3,8). Expression of bone morphogenetic protein 2 (Bmp2) in the mature dermal adipocyte layer is inversely correlated with Wnt activity and inhibits HF growth (3). There is also evidence that immature dermal adipocytes activate HF stem cells to initiate the hair growth cycle (7). These reports suggest that the adipocyte differentiation process is a natural on-off cycling switch for the regulation of hair grow...

Complement activation and subsequent recruitment of inflammatory cells at the dermal/epidermal junction are thought to be essential for blister formation in bullous pemphigoid (BP), an autoimmune blistering disease induced by autoantibodies against type XVII collagen (COL17); however, this theory does not fully explain the pathological features of BP. Recently, the involvement of complement-independent pathways has been proposed. To directly address the question of the necessity of the complement activation in blister formation, we generated C3-deficient COL17-humanized mice. First, we show that passive transfer of autoantibodies from BP patients induced blister formation in neonatal C3-deficient COL17-humanized mice without complement activation. By using newly generated human and murine mAbs against the pathogenic noncollagenous 16A domain of COL17 with high (human IgG1, murine IgG2), low (murine IgG1), or no (human IgG4) complement activation abilities, we demonstrate that the deposition of Abs, and not complements, is relevant to the induction of blister formation in neonatal and adult mice. Notably, passive transfer of BP autoantibodies reduced the amount of COL17 in lesional mice skin, as observed in cultured normal human keratinocytes treated with the same Abs. Moreover, the COL17 depletion was associated with a ubiquitin/proteasome pathway. In conclusion, the COL17 depletion induced by BP autoantibodies, and not complement activation, is essential for the blister formation under our experimental system.