CD8 regulatory T cells: What's old is now new

Abstract

CD8 regulatory T cells have had a long history in immunology. They were the first described subset of cells capable of inducing immune suppression as shown initially by Gershon and Kondo in the 1970s.1, 2 However, difficulties in characterization of these CD8 T cells and potential markers limited this area considerably. In the current issue, Cone et al.3 describe CD8 Tregs specific for antigen delivered to an immune privileged site—the eye—and their ability to limit effector T-cell responses in an ‘unconventional’ major histocompatibility complex (MHC) class Ib-restricted manner. Besides the initial difficulties in identifying CD8 Tregs, these regulatory cells were also outshone by the emergence of Tregs in the CD4 T-cell compartment. Hall et al.4 and Sakaguchi et al.5 showed that a subset of CD4 T cells expressing CD25 (the interleukin-2 receptor) was capable of inducing immune suppression. Initially described as CD4+ CD25+ T cells, the recent identification of the transcription factor foxp3 has defined a lineage of T cells that are generated in the thymus or induced by peripheral conversion, and can inhibit the autoimmune disease by suppression of autoreactive T cells. The importance of CD8 Tregs has been recently revived by observations in both T-cell vaccination and in knockout mice.6, 7 Cone et al.3 add to this re-emerging field the description of CD8 Tregs that can limit the delayed type hypersensitivity T-cell responses against antigen delivered to the eye, and show that this is restricted not by MHC Class Ia molecules but rather by Qa-1, an MHC Class Ib molecule expressed on effector T cells. Activated T cells can be used to vaccinate mice, and this induces a regulatory response that limits disease in a number of models including experimental autoimmune encephalomyelitis (EAE), the mouse model of multiple sclerosis. This has been shown to be due to the induction of CD8 T cells that interact with effector T cells limiting immune responses (CD8 Tregs).6, 7 Further, this response occurs not through the classical Class Ia MHC molecules, but through a Class Ib molecule Qa-1 (HLA-E in man).6, 7 The importance of this molecule's interaction with CD8 Tregs was shown in Qa-1 knockout mice, which in the absence of CD8 Treg function develop autoimmune disease.6 CD8 Tregs appear to act not only through direct lysis of effector T cells via perforin but also through immunosuppressive cytokines. Qa-1 is a very interesting molecule and is expressed on a range of cells in particular activated T cells, B cells and dendritic cells. It can bind β-2 globulin, such as a classical MHC Class Ia molecule and can present a range of peptides.8 The peptides it presents are derived either from the leader sequence of MHC molecules termed Qdm molecules or from self peptides including preproinsulin and heat-shock protein, and this is a transporter associated with antigen processing (TAP)-dependent process. In addition to acting as a restriction element for CD8 T cells, it is also a ligand for the inhibitory receptor NKG2A on natural killer (NK) cells. Thus, it can both inhibit NK cell lysis of T cells or can lead to CD8 Treg inhibition of T cells. Distinct binding sites on Qa-1 for CD8 Tregs and NK cell, NKG2A, have been defined, and mutations of each lead to opposing effects. The lack of activation of CD8 Tregs leads to worsening disease, whereas loss of NK inhibition leads to less severe disease (Figure 1).9 The numbers of CD8 Tregs are relatively small: <1% in peripheral circulation, though much higher among intestinal epithelial lymphocytes. A number of markers of CD8 Tregs have been studied including CD28low, CD122+ and CD8αα expression compared with the normal CD8αβ heterodimer.10 Additionally, foxp3 expression has been identified in some CD8s though this may reflect activation-induced foxp3 rather than a true regulatory phenotype. The use of Qa-1 tetramers while initially promising is limited by the relatively low binding efficiency of peptides to the molecule and also by the tetramer binding to NKG2A on large numbers of CD8 T cells. The restricted nature of CD8 Tregs has also been shown by T cell receptor (TCR) repertoire analysis. In summary, Qa-1-restricted CD8 Tregs appear to represent a unique regulatory subset that has a role in limiting autoimmune disease. The study by Cone et al.3 adds to the current literature by showing local immune suppression by CD8 Tregs and their restriction by Qa-1 rather than Class Ia molecules. Other data in a parallel study by these authors also show the role of the cytokines transforming growth factor-β and interferon-γ in the function of CD8 Tregs in suppressing the delayed type hypersensitivity.11 Qa-1-restricted CD8+ Tregs limit the effector T cells compared with the NK cells, which are inhibited by Qa-1 through NKG2A.