Proteolysis on the cell surface and in the extracellular matrix is essential for normal cellular functions during development and in the adult organism, but it may also have undesirable consequences by promoting diseases, such as cancer, inflammatory and autoimmune reactions, arthritis, cardiovascular disease and neurodegeneration. A particularly interesting function of proteolysis on the cell surface is to release ectodomains of membrane proteins from the plasma membrane. This process, which is referred to as protein ectodomain shedding, affects a variety of transmembrane proteins with important roles in development and disease, such as growth factors, cytokines, proteinases, receptors and adhesion proteins. Inhibitor studies have implicated two large families of zinc-dependent proteinases, the matrix metalloproteinases (MMPs) and the disintegrin-type metalloproteinases (ADAMs), in protein ectodomain shedding. In collaboration with Gillian Murphy, in Cambridge, we are studying the role of different metalloproteinases in syndecan-1 and erbB-2 ectodomain shedding. Syndecan-1 is a multifunctional cell surface co-receptor whose serum levels are elevated in diseases, such as multiple myeloma, lung cancer and acute graft-versus-host disease. Soluble syndecan-1 promotes growth of tumours in multiple myeloma and lung cancer patients, and strongly correlates with disease state and outcome. ErbB-2 is a member of the EGF receptor family whose serum levels are frequently elevated in breast cancer. Patients with high levels of soluble erbB-2 are more resistant to Herceptin therapy.