Pain in the brain

In the best-case scenario, self-attacking T cells are ousted before they have a chance to leave the thymus. That essential process, which eliminates autoreactive thymocytes from the get-go, is known to occur in the thymic medulla. Now, on page 2575, McCaughtry et al. provide solid evidence that this elimination also occurs in the thymic cortex. Although immunologists have reported T cell deletion in the cortex before, their results were made ambiguous by the transgenic mice models they used. These mice expressed T cell receptors specific for a self-peptide earlier in development than they are normally expressed. One fear was that those T cells were being deleted before they had a chance to migrate into the medulla. To quell that concern, McCaughtry's team generated a transgenic mouse in which the receptors are turned on at the appropriate time. Turns out, the earlier models were correct. T cells reacting to ubiquitous self-antigens were deleted in the cortex alone, before entering the medulla. Furthermore, the medulla-based transcription factor AIRE, which mediates T cell deletion, did not seem to be required in the cortical deletions. Local dendritic cells, however, were needed. Without them, a large fraction of autoreactive T cell clones survived. The locale of T cell deletion may depend on the antigen involved. Tissue-specific self-antigens are produced only in the medulla, whereas ubiquitous self-antigens, such as housekeeping peptides and the antigen recognized in the authors' mouse model, are widespread and can be handled in the cortex, at the site of first encounter. Now that the question of location has been settled, future studies can focus on learning how cortical dendritic cells control T cells gone awry. Arthritis sufferers might find some relief in a new compound found to ease bone damage in mice. Bull et al. (page 2457) prevented joint erosion when they crippled a member of the TNF receptor superfamily, Death Receptor 3 (DR3), by blocking its ligand, TL1A. Patients with rheumatoid arthritis are more likely to have extra copies of the gene that codes for DR3. A causative role for DR3 in disease pathology, however, had not been shown. Bull et al. now find that mice lacking DR3 are freed from the bone damage usually caused by antigen-induced arthritis. And blocking TL1A curtailed the bone erosion typical of collagen-induced arthritis in mice. Back in the dish, the team investigated how this receptor–ligand pair might induce bone destruction in humans. They found …