AIM2 safeguards Treg stability by suppressing the Akt-mTOR pathway via the protein phosphatase PP2A. Chou, WC, Guo, Z, Guo, H, et al. AIM2 in regulatory T cells restrains autoimmune diseases. Nature. 2021; 591: 300– 305. Regulatory T cells that express the transcription factor Foxp3 (Tregs) are a centerpiece in immune tolerance, including tolerance to transplants. However, their suppressive activities are often abrogated in an inflammatory milieu, conditions under which tolerance is destabilized or even lost. Thus, a better understanding of the interface between Tregs and inflammatory responses is therapeutically important. One of the key triggers in inflammation is activation of the inflammasome, a large molecular complex consisting of a cytosolic sensor, an adaptor molecule called ASC and pro-caspase-1/11, which traditionally results in the production of inflammatory cytokines and pyroptotic death of inflammatory cells. In a recent study in Nature, Chou et al. made the surprising finding that besides myeloid cells, Tregs also express AIM2, a cytosolic sensor for double-stranded (ds)DNA. Instead of activating the inflammasome, AIM2 instead safeguards Tregs, protecting them from losing their suppressive functions even under inflammatory conditions that usually subvert Treg functions. Importantly, Treg protection by AIM2 is completely independent of the inflammasome pathway. Thus, besides its classical role in inflammasome activation, AIM2 is also critical in immune tolerance that relies on Tregs. The authors initially took a genetic approach, comparing the induction of experimental autoimmune encephalitis (EAE) in mice deficient for AIM2 (Aim2−/−) and ASC (Asc−/−), and in mice deficient for both caspase-1 and caspase-11 (Casp-1/11 double knockouts). They found that the Aim2−/− mice developed more severe EAE than did the wild-type mice, while the Asc−/− and Casp-1/11 double knockouts were protected from the induction of EAE, suggesting that AIM2 has additional roles other than inflammasome activation. Adoptive transfer experiments involving transfer of wild type and Aim2−/− CD4+ T cells into Rag1−/− hosts recapitulated the enhanced induction of EAE by Aim2−/− CD4+ T cells, which was accompanied by prominent Th17 cells and a marked reduction of Tregs in the spinal cord. In a colitis model, where mucosal injury induced by naïve CD4+ T cells is susceptible to suppression by wild-type Tregs, they showed that the Aim2−/− Tregs failed to suppress the induction of colitis, thus identifying a functional defect of Tregs deficient in AIM2. In a series of ChIP-seq experiments, the authors further showed that the transcription factors RUNX1, EST1, BCL11B and CREB, which are known to regulate Foxp3 expression and Treg stability, also bound to the Aim2 promoter, thus demonstrating a Treg-intrinsic transcriptional network that also favors AIM2 expression in Tregs. The authors then performed sophisticated lineage-tracing experiments, in which Tregs are genetically marked by the expression of a red fluorescent protein (i.e., Foxp3Cre-Rosa26tdTomato mice), and demonstrated that, upon EAE induction, Aim2−/− Tregs readily lost the expression of Foxp3 and became interferon (IFN)-γ–producing T effector cells in the spinal cord (i.e., inflammatory sites), but not in host spleen and lymph nodes, demonstrating a key role for AIM2 in Treg stability. They also found that Aim2−/− Tregs displayed much higher glycolic activities compared with wild-type Tregs, with a concurrent reduction in oxygen consumption rate and fatty acid oxidation, suggesting drastically altered metabolic pathways in Aim2−/− Tregs. Moreover, global transcriptional profiling (RNA-seq) of wild-type and Aim2−/− Tregs, together with gene set enrichment analysis, revealed that the Aim2−/− Tregs showed a prominent IFN-responsive gene signature, which is known to antagonize Foxp3 expression as well as selective enrichment of genes in the Akt-mTOR pathway. In fact, Akt and mTOR were highly activated in Aim2−/− Tregs, as shown by increased protein phosphorylation of Akt, S6 and 4E-BP1 (targets of mTOR), a process that could be blocked by the mTOR inhibitor rapamycin. This heightened activation of Akt and mTOR also prevented the conversion of Aim2−/− T effector cells to inducible Tregs in vitro by transforming growth factor (TGF)-β. In a series of unbiased mass spectrometric experiments, the authors identified and validated RACK1 as a binding partner of AIM2. They further demonstrated that RACK1 recruited the protein phosphatase PP2A, which inhibits the Akt-mTOR pathway by dephosphorylating Akt and reducing mTOR activation. Apparently, in Aim2−/− Tregs, this feedback inhibitory loop is impaired, leading to hyperactivation of the Akt–mTOR pathway, and consequently destabilizing Tregs. Clearly, this study provides novel insights into the role of AIM2, a cytosolic sensor traditionally linked to inflammasome activation, in regulating Treg stability and peripheral tolerance. Xian C. Li, MD, PhD is professor and director at the Immunobiology and Transplant Science Center and Department of Surgery at Houston Methodist Hospital in Texas.