Methods: We transferred self-reactive CD4+ cells from KRN-Tg mice into recipient mice, which induced autoantibodies and autoinflammatory arthritis. This model allowed manipulation of environmental effects, such as inflammation, and use of transferred cells that were genetically deficient in important Tfh cell–associated molecules.
Results: A deficiency of signaling lymphocytic activation molecule–associated protein (SAP) in CD4+ cells from KRN-Tg mice completely protected against arthritis, indicating that stable T cell–B cell interactions are required for GC formation, autoantibody production, and arthritis induction. In contrast, a CXCR5 deficiency in CD4+ cells from KRN-Tg mice still induced disease when these cells were transferred into wild-type mice, suggesting that T cell help for B cells could rely on other migration mechanisms. However, various manipulations influenced this system, including elimination of bystander effects through use of CD28−/− recipient mice (reduced disease) or use of inflammation-inducing Freund's complete adjuvant (progression to arthritis). We also examined the capacity of preexisting GCs with a nonautoimmune specificity to co-opt autoimmune T cells and observed no evidence for any influence.
Conclusion: In addition to the quality and quantity of cognate CD4+ cell help, external factors such as inflammation and noncognate CD4+ cell bystander activation trigger autoimmunity by shaping events within autoimmune GC responses. SAP is an essential molecule for autoimmune antibody production, whereas the importance of CXCR5 varies depending on the circumstances.