These patterns were observed regardless of treatment protocol (anti-GITR mAb and anti-CD25 mAb), strain (BALB/c check details and C57BL/6) and antigen (SRBC, IAV and PE). Importantly, these findings provide a basis to explain the marked increase in serum antibodies, especially switched isotypes, upon in vivo Treg-cell disruption or depletion. These data are also consistent with reports showing the ability of adoptively transferred Treg cells to suppress in vivo B-cell responses,21,30–42 including GC reactions32,41 and the generation of antibody-forming cells.33,34,36

Although it is clear that Treg cells participate in the control of GC reactions, the target and site of Treg-cell action are currently unknown. Two likely targets are Tfh cells and GC B cells. The Tfh cells are critical in the induction and maintenance of GCs because they provide key co-stimulatory signals through inducible T-cell costimulator (ICOS) and CD154, as well as key cytokines, especially IL-21.75 In

addition, it has been shown that the magnitude of the GC response is directly linked to the size of the induced Tfh-cell pool.76 While Treg-cell suppression of CD4+ T-cell activity is well established,11–13 few investigators have focused on whether Treg cells can specifically alter Tfh function. In a recent study by Erikson and co-workers, however, adoptive transfer of antigen-specific Treg cells was found to down-modulate Cisplatin manufacturer the expression of ICOS on Tfh cells.41 In addition, Weiner and colleagues reported that induction of Treg cells in vivo compromised the ability of Tfh cells to produce optimal levels of IL-21.39 As ICOS expression77 and IL-21 production78–80 by Tfh cells are crucial for optimal B-cell differentiation and switching, influencing these molecules would serve as an effective means by which Treg cells could

control the GC response. In preliminary experiments, PIK3C2G we tested whether total numbers of splenic Tfh cells were altered by anti-GITR treatment in SRBC-immunized mice. However, when examining days 8 and 12 (the peak of splenic Tfh-cell induction after antigen challenge), no differences were observed (see Supplementary material, Fig. S4). Germinal centre B cells are also a potential target because a number of studies have demonstrated that Treg cells directly suppress activated B cells in vitro.32,40,42–46 In these experiments, Treg–B-cell contact was required and in several reports, Treg cells effected suppression by killing B cells in either a Fas-dependent43 or granzyme B-dependent40,46 manner. Although Treg cells may indeed directly suppress GC B cells, it is uncertain whether they use a cytotoxic mechanism in vivo. Studies in our laboratory found that both Fas-mutated lpr mice and granzyme B-deficient mice generated normal GC responses after SRBC challenge (data not shown).