Inhibition of Immunological Suppression

Suppressor/regulatory T cells have a long and checkered immunological history. T cells with immunosuppressive activity were first described by Gershon and Kondo (1, 2) more than 50 years ago. Serological studies of congenic mice led to the hypothesis that suppressor T cells were restricted by an immune response (Ir) gene in a MHC locus called I-J (3). When the advent of molecular biology enabled the cloning of the MHC, no genetic locus corresponding to the putative I-J region was identified (4, 5). This finding cast a pall over the study of suppressive T cells (6). What we now call regulatory T cells (Tregs), a thymic-derived T cell lineage specified by the transcription factor Foxp3, were not identified until 2003 (7, 8). Interestingly, much of the phenomenology originally ascribed to suppressor T cells was rediscovered in the CD4⁺Foxp3⁺ Treg compartment (9). TGF-β–secreting suppressive T cells were first reported in the gut (10) as the mediators of peripheral nonresponsiveness induced by orally administered Ag. They were dubbed Th3 cells (11), implying a third lineage of T cells (following Th1 and Th2). Their assignment to a distinct lineage was, however, premature; secretion of TGF-β is now known to be one of many mechanisms by which CD4⁺Foxp3⁺ Tregs exert their immunosuppressive function (12).

Before the discovery of Foxp3, the identification of a cell surface marker helped to reinvigorate interest in suppressor cells/Tregs and their function. CD4⁺ suppressor T cells that prevented autoimmunity were characterized by their expression of the IL-2Rα chain, CD25 (13–15). Powrie et al. (16, 17) showed that the Tregs that restrained intestinal inflammation had the same phenotype as those controlling autoimmunity. They examined Treg function in a model they had previously described in which transfer of CD4⁺CD45RB^hi effector T cells to lymphopenic mice induced colitis that was rescued by transfer of CD4⁺CD45RB^lo T cells. T cell–mediated suppression of inflammation was attributed to the secretion of TGF-β (17) and IL-10 (18). Importantly, the colitic inflammation was not autoreactive, but was directed against intestinal microbes (now known to comprise the microbiome); colitis was not induced when CD4⁺CD45RB^hi T effectors were transferred into antibiotic-treated or germ-free SCID recipients (19). In the highlighted Pillars of Immunology article, Read et al. (20) fractionated the CD4⁺CD45RB^lo compartment into CD25⁺ and CD25⁻ subsets and showed that only the CD25⁺ subset ameliorated intestinal inflammation when cotransferred with CD4⁺CD45RB^hi T cells into lymphopenic mice.

T cells in the CD4⁺CD45RB^loCD25⁺ fraction also expressed CTLA-4 constitutively. It was not at all clear, however, whether CTLA-4 was a memory/activation marker or whether it contributed to suppressor T cell function. The costimulatory molecules CTLA-4 and CD28 interact with B7.1 and B7.2 (CD80/CD86) on APCs and provide the second signal for T cell activation, following TCR-mediated recognition of Ag presented by MHC. CD28 is expressed on naive T cells; signaling through this costimulatory receptor leads to T cell proliferation and activation. Cell surface expression of CTLA-4 is induced in response to T cell activation. Concomitant signaling through CTLA-4 and the TCR blocks proliferation and inhibits IL-2 gene expression (21–23). The affinity of CTLA-4 for B7.1/B7.2 is higher than that of CD28; ligation of CTLA-4 imparts an inhibitory signal that attenuates T cell proliferation. Like much in this field, the role for CTLA-4 proposed in this Pillars of Immunology article was controversial at the time this work was published. Earlier work from Wahl and colleagues (24) reported that engagement of CTLA-4 induced CD4⁺ T cells to produce TGF-β. An inhibitory CTLA-4 signal could block the inflammatory response mediated by the CD4⁺CD45RB^hi T effector cells or promote the suppressive function of CD4⁺CD45RB^loCD25⁺ Tregs. To distinguish between these possibilities, Read et al. (20) treated transfer recipients with a mAb to CTLA-4 every other day for 6 wk. No differences were detected in colitis induced by the transfer of CD4⁺CD45RB^hi T effectors in the presence or absence of treatment with anti–CTLA-4. The suppressive activity of cotransferred CD4⁺CD45RB^loCD25⁺ Tregs was, however, blocked by treatment of the transfer recipients with anti–CTLA-4. The ability of cotransferred CD4⁺CD45RB^loCD25⁺ Tregs to suppress intestinal inflammation was also attenuated by treatment of the transfer recipients with a mAb to TGF-β. By separating effector and regulatory cell populations, both phenotypically and functionally, in a reproducible in vivo model, Read et al. (20) were able to convincingly demonstrate that expression of CTLA-4 and secretion of TGF-β were required for CD4⁺CD45RB^loCD25⁺ T cell–mediated control of intestinal inflammation.

The findings described in this Pillars of Immunology article therefore provided foundational information on the identity of Treg populations and the effector molecules contributing to immune suppression. Other work showed that administration of mAbs to block inhibitory signals from CTLA-4 enhanced antitumor immunity (25). Translational studies led ultimately to successful clinical trials and the establishment of “checkpoint inhibitor” therapies for cancer with Abs to CTLA-4 as well as PD-1/PD-L1. James Allison and Tasuku Honjo won the 2018 Nobel Prize in Physiology and Medicine for “their discovery of cancer therapy by inhibition of negative immune regulation.”