# Bad Blood

Enrichment Map used to identify signaling pathways that enable regulatory T cells to prevent runaway activation of the immune system

## Quick Summary

• Failure to repress immune responses can lead to widespread autoimmune disorders
• Tregs repress immune responses by sequestering IL-2 as well as down-regulating dendritic cells
• Enrichment analysis results displayed using an Enrichment Map linked IL-2 signaling in Tregs with their ability to complex with dendritic cells

## Author Profile

Takatoshi Chinen and Alexander Y Rudensky are at Memorial Sloan Kettering Cancer Center, New York, NY. Arun K Kannan is in the immunology discovery division of Biogen in Cambridge MA.

## Context

Therapies that aim to harness the immune system to target and disable cancer cells have stoked tremendous hope and in have shown promise. One success story is ipilimumab (Yervoy; Bristol-Myers Squibb) which was approved for the treatment of metastatic melanoma in 2011. Ipilimumab is a monoclonal antibody that acts by disrupting endogenous mechanisms within T cells that otherwise reduce their level of activation. In clinical trials, a significant increase in survival time could be attributed to ipilimumab for those whose disease was unresponsive to standard chemotherapy. Nevertheless, trial participants suffered complications including gastrointestinal, hepatic and skin inflammation. These observations reinforce the call for a better understanding of immune reactions induced by immunotherapy, and more generally, how a proper balance between immune activation and inhibition is achieved (Nature Med. 2017).

## Question

What are the signaling mechanisms that immune cells use to mediate system-wide immune repression?

## Goals

Regulatory T cells (Tregs) play an important role in restraining the immune system by maintaining peripheral tolerance, preventing autoimmune disorders and limiting chronic inflammation (Vignali 2008). Accordingly, quantitative deficits in Treg cells in humans and experimental mouse models leads to multiple organ-specific autoimmune diseases. This is consistent with the notion that the balance between Tregs and effector T cells (i.e. CD4+ helper and CD8+ cytotoxic) is dysregulated in autoimmune and inflammatory disorders. A major challenge is to develop a comprehensive account of the mechanisms through which Tregs repress immune responses and the relative importance of each in varying contexts.

Tregs have been shown to act through mechanisms that involve direct cell-contact, such as modulating antigen presenting cell (APC) competency, as well as by modulating the extracellular milieu of effector T cells (Figure 1). Signalling by IL-2 has been the focus of much research as it is essential for the development, homeostasis and function of both Tregs and effector T cells (Klatzmann 2015) (Figure 1c).

Figure 1. Basic mechanisms used by Treg cells. Depiction of the various regulatory T (Treg)-cell mechanisms centred around four basic modes of action. a. Inhibitory cytokines include interleukin-10 (IL-10), IL-35 and transforming growth factor-beta (TGFbeta). b. Cytolysis includes granzyme-A- and granzyme-B-dependent and perforin-dependent killing mechanisms. c. Metabolic disruption includes high-affinity CD25 (also known as IL-2 receptor alpha)-dependent cytokine-deprivation-mediated apoptosis, cyclic AMP (cAMP)-mediated inhibition, and CD39- and/or CD73-generated, adenosine receptor 2A (A2AR)-mediated immunosuppression. d. Targeting dendritic cells (DCs) includes mechanisms that modulate DC maturation and/or function such as lymphocyte-activation gene 3 (LAG3; also known as CD223)–MHC-class-II-mediated suppression of DC maturation, and cytotoxic T-lymphocyte antigen-4 (CTLA4)–CD80/CD86-mediated induction of indoleamine 2,3-dioxygenase (IDO), which is an immunosuppressive molecule made by DCs.. Adapted from Vignali et al., Figure 1 (Vignali 2008).

Here we describe a study by Chinen et al. (Chinen 2016) who dissected the IL-2 signalling pathway in Tregs in order to establish the means by which its components impact effector T cell suppression. The authors suggested at least three (not necessarily mutually exclusive) ways in which the IL-2 receptor might mediate its effects: Facilitate homing of Tregs towards self-reactive T cells which actually produce IL-2 ligand; IL-2 sequestration and deprivations; and intracellular signalling to support maintenance, proliferation and function via JAK-STAT5, PI3K-Akt or Ras-ERK pathways (Li 2016) (Figure 2).

Figure 2. Intracellular signalling in Tregs. The activation of Tregs is mediated by combined signalling through the T cell receptor (TCR) and co-stimulatory molecules such as CD28, and through the interleukin-2 receptor (IL-2R). This receptor is a complex of an alpha (CD25), beta (CD122) and a common gamma-chain (CD132) of polypeptides encoded by the genes il2ra, il2rb and il2rg, respectively. It is hypothesized that Treg cells have evolved to depend more heavily on signal transducer and activator of transcription 5 (STAT5)-dependent IL-2R signalling (thick arrow) than on TCR and CD28 signalling compared with effector T cells. Adapted from Klatzmann et al., Figure 2 (Klatzmann 2015).

## Approach

In Tregs, IL-2 binds to the extracellular domain of IL-2R which is comprised of three polypeptide chains: An $\alpha$-chain, a $\beta$-chain and a common $\gamma$-chain (Figure 2). Cytosolic membrane signals are transmitted via JAK kinase which in turn phosphorylates the cytoplasmic mediator Signal Transducer and Activator of Transcription 5 (STAT5), consisting of a heterodimer of STAT5A and STAT5B. A master regulator of Treg cells is the forkhead box 3 (Foxp3) gene, a member of the forkhead/winged-helix family of transcription factors (Rudensky 2011). The consensus view of Foxp3 function is to amplify and stabilize features of Treg cells that are previously conferred by T cell receptor (TCR) and cytokine signaling (Figure 2).

In order to definitively establish the role of IL-2R in Treg cell function in vivo, the authors employed a transgenic mouse that disabled il2ra or il2rb only within cells that express Foxp3, that is, the Treg lineage. As expected, deletion of either receptor chain resulted in an early onset of fatal autoimmune inflammatory disease and lymphoproliferation. These phenotypes were accompanied by a drastic reduction in tyrosine-phosphorylation of STAT5 (Figure 2). Accordingly, ablation of stat5a and stat5b in peripheral Fox3p-expressing cells mimicked the autoimmune phenotype observed in IL-2R mutants. Finally, constitutive expression of stat5b in an il2ra-deficient strain (i.e. unable to sequester IL-2) was sufficient to rescue the systemic inflammation and early fatal disease. These observations are consistent with a requirement for cell-intrinsic IL-2R signalling in Tregs.

In contrast to the rescue of early inflammatory disease, constitutive activation of stat5b only marginally restrained expansion of CD8+ T cells in later weeks (3-12) and these mice eventually succumb to disease, suggesting a STAT5 signalling-independent mechanism of suppression. Indeed, application of IL-2-neutralizing antibody in il2rb-deficient mice with constitutive expression of stat5b (i.e. stat5bCA) at 5 - 7 days of age completely suppressed CD8+ T cell activation and proliferation. These observations are consistent with the notion that Tregs suppress CD8+ T cells via IL-2 depletion.

Next, Chinen et al. turned their attention to the question of what mechanisms downstream of STAT5 signaling potentiated Treg suppression. Interestingly, the authors demonstrated that constitutive stat5b expression did not increase suppression by augmenting the number of Tregs but rather, it enhanced the suppressive potential of each individual cell. Moreover, previous work showed that large increases in STAT5 did not translate into increased genome binding of FOX3P (Arvey 2014).

To gain more insight into the qualitative differences conferred by persistent STAT5 signalling, the authors compared gene expression in Tregs with constitutive stat5b expression relative to controls. Amongst ~11 000 genes analyzed, there were 342 genes up-regulated and 314 down-regulated. Analysis of signaling pathways was performed to better understand the cellular functions potentiated by activated STAT5: BiNGO was used to identify Gene Ontology (GO) terms enriched in the gene expression data (10 589 genes total). The authors chose to display the enriched gene sets using an Enrichment Map in which gene sets are represented by nodes in a network and shared genes via edges connecting nodes. The visual nature of the Enrichment Map aids in the interpretability of enrichment analyses by providing a visual ‘landscape’ of the gene sets and reducing complexity by clustering redundant gene sets.

The authors chose to display the enriched gene sets using an Enrichment Map in which gene sets are represented by nodes in a network and shared genes via edges connecting nodes.

Figure 3. RNA-seq analysis of Treg cells expressing an active form of STAT5. Network analysis of gene-ontology term enrichment among genes significantly up-regulated in STAT5b-CA+ Treg cells relative to their expression in Foxp3Cre-ERT2 Treg cells; dashed outlines (added manually) indicate groups of similar gene- ontology terms (Supplementary Table 1); line thickness and color are proportional to the similarity coefficient between connected nodes; node color is proportional to the FDR-adjusted P value of the enrichment and node size is proportional to gene-set size. Adapted from Chinen et al., Figure 5f (Chinen 2016).

The pathway enrichment analysis and resulting Enrichment Map enabled authors to quickly narrow in on gene set products involved in cell-cell and cell-extracellular matrix interactions, cell adhesion and cell locomotion (Figure 3). This finding led authors to speculate that STAT5 could potentiate interactions with target cells. Indeed, this notion is supported by previous imaging studies in vivo demonstrating Tregs in stable contact with dendritic cells (DC) and the fact that Treg-DC interactions have been shown to modulate their ability of DCs to potentiate effector T cell activation (Figure 1). To this end, Chinen et al. demonstrated that constitutive expression of stat5b in Fox3P-expressing cells promoted the formation of conjugates between Tregs and DCs (Figure 4). Thus, the Enrichment Map helped to unify two bodies of knowledge: The ability of Tregs to modulate T cell activation via DCs and IL-2R signaling via STAT5.

… the Enrichment Map helped to unify two bodies of knowledge: The ability of Tregs to modulate T cell activation via DCs and IL-2R signaling via STAT5.

Figure 4. Augmented STAT5 signaling in Treg cells increases the formation of conjugates of Treg cells and DCs. In vitro formation of conjugates of non-Treg cells (Foxp3Cre-ERT2) or Tregs with a constitutively active stat5b (Rosa26Stat5bCAFoxp3Cre-ERT2) or without (Foxp3Cre-ERT2). Conjugates were assessed by flow cytometry after culture together for 720 min at various ratios (horizontal axes) in the absence (left) or presence (right) of recombinant mouse IL-2 (100 IU/ml), analyzed by flow cytometry; each symbol represents a single well.Adapted from Chinen et al., Figure 6a (Chinen 2016).

## Summary

Here, a pathway enrichment analysis coupled to an Enrichment Map presented groups of altered pathways drawn from the underlying patterns of gene expression. This approach provided a link between cytokine signalling in Tregs and their ability to interact with DCs that represents a plausible mechanistic basis for how Tregs negatively regulate system-wide repression of immune activation. This detailed understanding provides the basis of rational approaches towards optimal design of Treg-cell based therapies for a variety of autoimmune and inflammatory disorders. Conversely, this knowledge may contribute to efforts to counteract the potential side-effects of activating immunotherapies.