PTENtiating autoimmunity through T_reg cell deregulation

Regulatory T (T_reg) cells dampen and promote resolution of the effector immune response, reducing both immunopathology and autoimmunity via secretion of the anti-inflammatory cytokine interleukin-10 (IL-10) and competition with effector T cells for interleukin-2 (IL-2) ¹. In this issue of Nature Immunology, Huynh, et al. and Shrestha, et al. demonstrate that the phosphatase PTEN protein maintains suppressive function of T_reg cells in homeostatic conditions through preserved expression of the high affinity IL-2 receptor (IL-2Rα, also known as CD25) and their canonical transcription factor, forkhead box P3 (Foxp3) (Huynh, et al, 2014; Shrestha, et al 2014).

IL-2 is a known inducer of T_reg cell differentiation and maintenance, signaling through Signal Transducer and Activator of Transcription 5 (STAT5) to maintain Foxp3 expression ². IL-2 signaling is also mediated by phosphoinositide 3-kinase (PI3K) and activation of the metabolic checkpoint kinase mTOR in effector T cells, such as T helper 1 (T_H1) and T_H17 cells ³; however, differentiation of T_reg cells is not reliant upon signaling via this IL-2-dependent pathway (Fig. 1a). T_reg cells also do not upregulate glycolytic metabolism, instead preferentially utilizing fatty acids to promote their regulatory functions ⁴. PTEN is a potent negative regulator of PI3K signaling, utilizing its phosphatase activity to oppose PI3K-mediated conversion of the membrane-associated signaling molecule phosphatidylinositol-4,5-bisphosphate (PIP2) to PIP3 ³. In line with its role in immunoregulation, deletion of PTEN in CD4⁺ T cells leads to thymus-derived lymphoma and autoimmunity, although the T cell subset-intrinsic role for PTEN in regulating autoimmunity is unclear ⁵. PTEN is highly expressed in T_reg cells, and has been previously found to suppress T_reg cell expansion ⁶. Huynh, et al. and Shrestha, et al. now demonstrate that PTEN in T_reg cells is critical for control of autoimmunity, finding it essential for the maintenance of CD25 and Foxp3 expression (Fig. 1b). As a consequence, T_reg cell-intrinsic expression of PTEN is necessary for the suppression of autoimmune responses mediated by T_H1 and follicular B helper T (T_FH) cells, drivers of B cell maturation, via attenuation of interferon-γ (IFN-γ) production.

Both studies take a genetic deletion approach, generating T_reg cells that are selectively deficient in PTEN via breeding mice carrying a floxed Pten allele (Pten^fl/fl) to animals expressing Cre recombinase driven by the Foxp3 promoter (Foxp3-Cre). T_reg cells lacking PTEN were increased in number, as earlier demonstrated ⁶; however, these cells also had a marked reduction in cell surface CD25 expression, leading both sets of authors to conclude that PTEN-deficient T_reg cells converted to ‘ex-T_reg cells’, or cells that have lost Foxp3 expression and suppressor function. Lineage tracing experiments ⁷ confirmed this idea, with the demonstration that PTEN-deficient T_reg cells did indeed lose Foxp3 expression, while also losing the capacity to suppress effector cells. Instead, PTEN-deficient ex-T_reg cells upregulated expression of the inflammatory cytokine IFN-γ ⁷. This increase in T_H1-cell effector function was a result of mTOR complex 2 (mTORc2) activation, as Shrestha, et al. found that deletion of the RICTOR subunit of this complex restored the phenotype of PTEN-deficient T_reg cells.

In line with changes in their suppressor function, qualitative differences in T_reg cell metabolism in the absence of PTEN were found by both sets of authors. Unlike effector cells, T_reg cells have a reduced reliance on glucose and anabolic metabolism for their development and maintenance, instead utilizing fatty acids to promote their suppressor capability ⁴. PI3K induces anabolic metabolism through mTOR activation, consistent with the higher PTEN activity in T_reg cells compared to their effector counterparts ³. PTEN deletion in the former led to upregulated glycolytic metabolism, further promoting the effector cell phenotype of these cells. As the effector counterparts require glucose for function, it would be interesting to know if the PTEN-deficient ex-T_reg cells required glucose for their autoimmune potential, and whether blockade of glucose uptake could rescue the T_reg cell phenotype and suppressive functions.

As further evidence of the reduced suppressive capacity of PTEN-deficient T_reg cells, mice bearing these mutant cells developed an autoimmune syndrome resembling systemic lupus erythematosus (SLE, lupus), with production of antinuclear autoantibodies (ANAs) leading to immune-complex glomerulonephritis. Pathogenic autoantibody production in murine and human lupus is mediated by exuberant activation of T_FH cells followed by aberrant germinal center (GC) B-cell responses, in line with the findings of Huynh, et al. and Shrestha, et al. Indeed, GCs are regulated by a specific T_reg-cell subset (follicular regulatory T cells, or T_FR cells), and the autoimmune phenotypes seen in the absence of T_reg PTEN are in line with reduced T_FR-mediated suppression. Additionally, Shrestha, et al. showed that IFN-γ is increased in all effector cells in animals with PTEN-deficient T_reg cells, reminiscent of several lupus-prone mouse strains, including the Roquin^san/san mouse, in which this cytokine is required for aberrant T_FH cell and GC B cell expansion and autoantibody production ⁸. Accordingly, deletion of IFN-γ in the Pten^fl/flFoxp3-Cre mice by Shrestha, et al. rescued the aberrant increase in T_FH and GC B cells and production of pathogen-specific antibodies. In addition to ex-T_reg cells, CD4⁺ and CD8⁺ T effector cells have an increase in IFN-γ secretion in the absence of PTEN in T_reg cells. While its increased secretion by CD4⁺ T cells is likely an autoimmune driver, it would be important to know the contribution of IFN-γ secretion to disease by all three lineages – T_reg, T_FH and effector T cells – as this has implications for understanding and treating autoimmunity resulting from defective suppression by T_reg cells.

The novel insights from the work of Huynh, et al. and Shrestha, et al. raise additional questions. Is T_H1-mediated autoimmunity mTORc2 regulated? The finding that mTORc2 upregulation enables T_reg cells to adopt a T_H1 phenotype was surprising, given that mTORc1 has been found to promote T_H1 differentiation, with mTORc2 promoting that of T_H2 cells ³. Thus, it is possible that mTORc2 is a critical regulator of T_H1-cell function, particularly via IFN-γ upregulation. It was also interesting that T_reg cell dysfunction led to selective expansion of T_H1 and T_FH cells through the induction of IFN-γ, and not to the proliferation of other T cell subsets. How is IFN-γ pathogenically induced in these contexts? Could reduced T_reg cell suppression be the key to loss of immune tolerance? Indeed, these authors find that deletion of PTEN from T_reg cells can recapitulate many of the autoimmune phenotypes of mice globally deficient in PTEN. Future studies in human autoimmune syndromes, particularly those driven by IFN-γ signaling, are required to determine whether T_reg cell deregulation is a principal mediator of disease pathogenesis. How PI3K prevents T_reg cell function is also unknown; however, Shrestha, et al. found that Blimp-1 is upregulated in the absence of PTEN, suggesting that IL-2-signaling, known to induce this transcriptional regulator, does so via induction of PI3K signaling. The authors speculate that PTEN reduces PI3K-mediated Blimp-1 upregulation, allowing for prolonged IL-2 signaling within these cells. However, further experiments are required to parse the requirements of PI3K and STAT5 in inducing Blimp-1, as STAT5 has also been found to be an important regulator of its expression, at least in part through maintaining cell surface expression of CD25 ⁹.

The studies conducted by Huynh et al. and Shrestha et al. support the rationale for IFN-γ blockade in humans with SLE. Furthermore, their studies emphasize that deregulation of T_reg cells might be a primary driver of systemic autoimmunity, and conversely that their modulation or adoptive transfer might be beneficial. For example, rapamycin, known to promote T_reg cell proliferation at the expense of that of effector cells, is therapeutically beneficial in SLE ¹⁰, while transfers of T_reg cells is efficacious in murine lupus models ¹¹. Similar approaches are in clinical trials of type I diabetes mellitus (http://clinicaltrials.gov/ct2/show/NCT01210664?term=type+1+diabetes+and+tregcells&rank=1/). These clinical approaches highlight the broad implications of the findings of Huynh et al. and Shrestha et al. that PTEN is critical for T_reg cell suppression of potentially pathogenic T cells in non-infectious conditions.