TNF and receptors in organ-specific autoimmune disease: multi-layered functioning mirrored in animal models

George Kassiotis’s present address is: Division of Immunology, National Institute for Medical Research, Mill Hill, London, United Kingdom. Inhibition of TNF in rheumatoid arthritis (RA) is proving efficacious for a large number of patients (1). Despite this exciting outcome, however, there is still little knowledge of the actual biology that is targeted by this therapy. Potential effector activities of TNF that may be relevant to the observed therapeutic impact are numerous and include its well-known proinflammatory activities, the potential activating effects it may have on important cell types (e.g., macrophages or synoviocytes), and its functions in shaping innate and adaptive immunity (2). TNF is usually overproduced in chronic inflammatory/autoimmune diseases, such as RA and multiple sclerosis (MS), and this overproduction leads to strikingly similar pathologies when reproduced in transgenic animal models (see ref. 2 and references therein). Moreover, pharmacological inhibition of TNF in collagen-induced arthritis (CIA) and experimental autoimmune encephalomyelitis (EAE), autoimmune animal models for RA and MS, respectively, has yielded encouraging results. These notions led the way to the use of anti-TNF therapies in the clinic. Surprisingly, however, systemic blockade of TNF in MS patients led to immune activation and increased disease activity (3, 4). Further insight into the mechanisms of TNF function is therefore required, and more work with the animal models is necessary.