Shedding of Collagen XVII/BP180 in Skin Depends on Both ADAM10 and ADAM9

Collagen XVII is a transmembrane collagen and the major autoantigen of the autoimmune skin blistering disease bullous pemphigoid. Collagen XVII is proteolytically released from the membrane, and the pathogenic epitope harbors the cleavage site for its ectodomain shedding, suggesting that proteolysis has an important role in regulating the function of collagen XVII in skin homeostasis. Previous studies identified ADAMs 9, 10, and 17 as candidate collagen XVII sheddases and suggested that ADAM17 is a major sheddase. Here we show that ADAM17 only indirectly affects collagen XVII shedding and that ADAMs 9 and 10 are the most prominent collagen XVII sheddases in primary keratinocytes because (a) collagen XVII shedding was not stimulated by phorbol esters, known activators of ADAM17, (b) constitutive and calcium influx-stimulated shedding was sensitive to the ADAM10-selective inhibitor GI254023X and was strongly reduced in Adam10^−/− cells, (c) there was a 55% decrease in constitutive collagen XVII ectodomain shedding from Adam9^−/− keratinocytes, and (d) H₂O₂ enhanced ADAM9 expression and stimulated collagen XVII shedding in skin and keratinocytes of wild type mice but not of Adam9^−/− mice. We conclude that ADAM9 and ADAM10 can both contribute to collagen XVII shedding in skin with an enhanced relative contribution of ADAM9 in the presence of reactive oxygen species. These results provide critical new insights into the identity and regulation of the major sheddases for collagen XVII in keratinocytes and skin and have implications for the treatment of blistering diseases of the skin.Collagen XVII (also called BP180 or BPAG2) is a hemidesmosomal adhesion component in the skin and mucosa and belongs to the emerging group of collagenous transmembrane proteins (1). This type II oriented transmembrane protein is involved in the molecular pathology of human skin diseases. Mutations in the COL17A1 gene are associated with junctional epidermolysis bullosa, a genetic skin blistering disease (2). Patients with bullous pemphigoid and related autoimmune bullous dermatoses have tissue-bound and circulating autoantibodies targeting collagen XVII (3). Structural and functional changes of collagen XVII play an important role in these diseases, although the molecular pathology is not yet fully understood. The collagen XVII consists of three 180-kDa α1 (XVII) chains, each with an intracellular N-terminal domain, a short transmembrane stretch, and a flexible extracellular C-terminal ectodomain with collagenous (Col)² subdomains that are interrupted by short non-collagenous (NC) sequences. The human and murine collagen XVII molecules differ in size and in the number of the Col and NC domains. Human collagen XVII consists of 1497 amino acid residues with 15 Col and 16 NC domains, whereas the murine form, which is 86% identical (4), consists of 1433 amino acid residues with 13 Col and 14 NC domains. In humans the extracellular linker domain NC16A between the plasma membrane and the Col15 domain is functionally important because it is believed to play a role in both ectodomain shedding and in the proper folding of the triple helical structure of collagen XVII (5–7).Our previous studies revealed two forms of collagen XVII, the 180-kDa membrane-anchored form and the soluble 120-kDa form. The latter represents the extracellular collagenous ectodomain, which is released by cleavage by membrane-anchored metalloproteinases of the a disintegrin and metalloproteinase (ADAM) family (8). The shed ectodomain of collagen XVII is very stable in vivo and in vitro. In wound scratch assays, both addition of the purified soluble ectodomain or overexpression of ADAMs suppressed cell motility (8), indicating that the ectodomain has a role in regulating keratinocyte-matrix interactions. In the context of the known functions of collagen XVII as an adhesion molecule, its shedding could therefore regulate its functions in keratinocyte migration, differentiation, and proliferation.ADAMs are also involved in the release of several other type I or type II transmembrane proteins and are considered to be critical regulators of epidermal growth factor receptor signaling, tumor necrosis factor α release, and Notch signaling to name a few examples (9, 10). Previously ADAM9, ADAM10, and ADAM17 had been identified as potential sheddases for collagen XVII in keratinocytes by overexpression in cell-based assays (8). Moreover Adam17^−/− keratinocytes had 50% diminished collagen XVII shedding, which was interpreted to suggest that ADAM17 represents an important, if not the major, physiological collagen XVII sheddase (8). The major goal of the current study was to further explore the contribution of ADAM17 and other candidate sheddases to the release of collagen XVII from primary keratinocytes and mouse skin. The identification of the major collagen XVII sheddases and their regulation is critical for understanding the role of collagen XVII shedding in the pathogenesis of skin diseases.