Fig. 2. ADAM17 is required for PMA- and PV-stimulated shedding of Kitl1 in mEFs, whereas ADAMs 8, 9, 10, 12 and 15 are dispensable. (A) Kitl1 shedding was analyzed in primary mEFs isolated from E13.5 wild-type controls (lanes 1, 2, 5, 6, 9 and 10) or from corresponding Adam8^–/–, Adam9/12/15^–/–, Adam17^–/– embryos (lanes 3, 4, 7, 8, 11 and 12). Adam10^–/– cells (lanes 3, 4, 7, 8, 11 and 12) and Adam10^+/– controls (shown under wild type in lanes 1, 2, 5, 6, 9 and 10) were immortalized since Adam10^–/– embryos die at E9.5. Whereas PMA- and PV-induced shedding was comparable between Adam8^–/– and Adam9/12/15^–/– mEFs and cells isolated from their respective wild-type controls, as well as between Adam10^–/– and Adam10^+/– cells, PMA-dependent increase in shedding of Kitl1 was abolished (lane 4) and PV-induced shedding was significantly reduced in Adam17^–/– cells (lane 8) compared with cells from wild-type controls (lanes 2 and 6). In all Adam^–/– and wild-type cells analyzed here, the BB94-sensitive component of constitutive Kitl1 shedding was unaffected (lanes 10 and 12 compared with lanes 9 and 11, respectively), suggesting that other metalloproteases besides the ADAMs tested here are responsible for constitutive Kitl1 shedding. (B) PMA-stimulated Kitl1 shedding in Adam17^–/– cells (lane 2) could be rescued by coexpression of mouse ADAM17 (lane 4). A western blot of the cell lysates of Adam17^–/– cells expressing Kitl1 (lane 5) or Kitl1 and ADAM17 (lane 6) is shown on the right. Overexpression of ADAM17 reduces the slowest migrating form of Kitl1 in Adam17^–/– cells. All experiments were repeated four times with essentially identical results.