FIGURE 4.
Constitutive collagen XVII shedding is strongly diminished in Adam9−/− keratinocytes and can be further reduced by the ADAM10-selective GI254023X, whereas ionomycin-stimulated shedding is not affected in the absence of ADAM9. A, shedding in primary murine keratinocytes was analyzed as described under “Experimental Procedures.” Keratinocytes derived from at least four different mice for each Adam−/− deficient strain and their wild type littermate controls were used for each experiment. All experiments were performed in duplicate, and cell lysates and concentrated media were probed for full-length collagen XVII (CXVII) and/or the ectodomain (ECTO) by Western blot. The percentage of ectodomain release was calculated by densitometric quantification. Shedding of collagen XVII was strongly diminished in Adam9−/− keratinocytes (46.0 ± 11.3%; n = 5; *, p < 0.05) but not in keratinocytes from Adam8−/− (95.5 ± 14.9%; n = 4) or Adam15−/− mice (86.5 ± 19.1%; n = 4). Representative immunoblots of membrane-bound collagen XVII in the lysate and the ectodomain in the medium are shown. B, analysis of the ionomycin stimulation (1 μm; 20 min) of Adam9−/− keratinocytes (AD9−/−) revealed comparable responses to IM and sensitivity to batimastat (BB; #, p < 0.05; ##, p < 0.01) as in wild type keratinocytes (WT). The graph represents the densitometric analysis of collagen XVII ectodomain immunoblot signals from three experiments (mean ± S.D.; *, p < 0.05; **, p < 0.01). The DMSO vehicle control activities (Co) are normalized to 100% to be able to compare activated shedding between different experiments. C, confluent cultures of Adam9−/− keratinocytes were washed, and fresh medium with DMSO vehicle as control (Co) or 0.5 μm GI was added and incubated for 3 h. The cell lysates and concentrated media were probed for full-length collagen XVII and/or the ectodomain by Western blot, and the inhibition of shedding was determined by densitometric comparison of the blots of supernatants from control- or GI-treated keratinocytes from three experiments, each with keratinocytes from different mice (mean ± S.D.; #, p < 0.05). Collagen XVII shedding in Adam9−/− keratinocytes was further reduced by the GI, which is selective for ADAM10 over ADAM17 at 0.5 μm, indicating that ADAM9 and ADAM10 are both involved in the constitutive shedding of collagen XVII.