Imagine the scene. You are planning an invasion into enemy territory but know that the army waiting to defend against any attack is far bigger than your own small-scale operation. What you really need is a weapon that will wipe out a large number of the enemy in one go, without having to get involved in hand-to-hand combat. Staphylococcus aureus seems to have come up with an ideal solution. It secretes a protein that knocks out a large percentage of the B-cell repertoire in a supraclonal manner.

S. aureus protein A (SpA) is a B-cell superantigen that forms a complex with lymphocytes expressing B-cell receptors (BCRs) with clan-VHIII-encoded variable regions. These BCRs are displayed by 5–10% of mature mouse B cells, including marginal-zone and follicular B cells, and B1 cells with a BCR antigen-binding region of the T15 idiotype. Now, Goodyear and Silverman have characterized the deletion of B cells in response to this toxin in a study published in The Journal of Experimental Medicine.

T15i immunoglobulin knock-in (T15i+/+) mice, in which almost all B cells express an S107 (clan VHIII) VH transgene, were exposed to SpA and analysed after various time points. The initial response of B cells to SpA was shown to be similar to the normal response to antigen exposure. After 2 hours, levels of the targeted cell-surface BCR were decreased; after 16 hours, the BCR co-receptors CD19 and CD21 were downregulated, and the activation markers CD69 and CD86 were upregulated; and after further in vitro culture, levels of MHC class II molecules and other markers were increased.

Next, the authors looked at whether the decrease in the number of T15i+/+ B cells that follows this initial activation in response to SpA is owing to effects on trafficking or apoptosis. The adoptive transfer of labelled T15i+/+ splenocytes to C57BL/6 recipients treated with SpA showed that after 48 hours of in vivo exposure, despite certain B cells having undergone 2–3 rounds of proliferation, the number of T15i B cells was decreased by ∼36%. This indicates that splenic T15i+/+ B cells exposed to SpA undergo an increased rate of apoptosis, and this was also seen at other anatomical locations. After in vivo exposure to SpA, levels of caspase-3 were increased in parallel with the induction of DNA fragmentation. The inclusion of Z-VAD, a potent pan-caspase inhibitor, delayed the 'assisted suicide' of SpA-targeted T15i+/+ B cells. Bcl-2 overexpression, but not Fas (CD95) deficiency, was shown to rescue B cells of SpA-treated T15i+/+ mice from apoptosis. The authors suggest that apoptosis in response to SpA is similar to the activation-induced cell death of B cells in response to antigen.

The authors are now hoping to put this microbial battle plan to therapeutic use. Up to 50% of human B cells express clan-VHIII-encoded BCRs and are therefore susceptible to SpA. Silverman and colleagues are engineering variants of SpA with higher binding affinities, and also with specificities for different framework variable regions, which will be evaluated for the selective destruction of malignant B-cell populations in autoimmune diseases, such as systemic lupus erythematosus, and cancers, such as lymphoma and leukaemia.