As concerns FOXP3 expression by T-Regs, originally Sakaguchi’s group reported that activation of peripheral murine T cells via TCR/CD28 did not result in de novo expression of FOXP3 by FOXP3-negative cells, nor did it enhance or suppress the expression of FOXP3-positive cells (Science 299:1057, 2003).  Subsequently, Morgan and co-workers reported that CD25- human CD4+ T cells express  readily detectable FOXP3 mRNA after activation with either PHA or a combination of anti-CD3 and anti-CD28 mAbs (Human Immunol. 66:13, 2005). Moreover, these investigators found that CD8+ T cells which did not express FOXP3 mRNA after isolation, did so 24-40 hours after activation with anti-CD3/28.  On the basis of their data, these investigators speculated “that the production of human FOXP3 in an activated T cell could act in part as a natural negative feedback loop that would prevent unrestricted cytokine production and inflammatory reactions in humans”. 

Subsequently, Rudensky’s group reported a dramatic increase in the percentage of FOXP3+ cells detectable by flow cytometry in both CD4+ and CD8+ T cells, with as many as 25% of CD4+ T cells and 27% of CD8+ T cells positive for FOXP3 intracellular protein on day 3 after activation with  anti-CD3/28 (Proc Nat Acad Sci 103:6659, 2006).  Even more recently,  Rao’s group reported that NF-AT is capable of binding to FOXP3, forming a transcription factor complex that is capable of repressing the expression of the IL2 gene, thereby providing a molecular confirmation of the  negative feedback speculation of Morgan’s group (Cell 126:375, 2006). However, where does that leave Regulatory T cells? Since there is as yet no molecular mechanism that explains the T-Reg suppression of cytokine production by effector T cells, could it be that antigen-activated T cells that express FOXP3, whether they are CD4+ or CD8+, are suppressed by their own FOXP3, rather than by T-Regs?