News & Views | Published:


In the right place at the right time

Nature volume 528, pages 205206 (10 December 2015) | Download Citation


Regulatory T cells help to prevent autoimmune responses. A new imaging technique reveals that activation of these cells requires clustering with self-reactive effector T cells and sensing of the signalling protein interleukin-2. See Article p.225

The T cells of our immune system produce receptor proteins that specifically bind to molecular structures called antigens. Each T cell expresses one type of receptor, and the immense diversity of the T-cell population ensures that our immune system can detect and respond in a highly specific manner to almost any pathogen we encounter. But a fundamental problem of this sensing strategy is that it also generates T cells that bear self-reactive receptors, which can trigger dangerous autoimmune attacks. Specialized regulatory T cells are needed to keep such autoreactive effector T cells in check, but exactly how they achieve this task is incompletely understood. On page 225 of this issue, Liu et al.1 provide information on how the spatial organization of regulatory and effector T cells helps to maintain immune homeostasis.

For many years, conventional wisdom held that, as T cells develop in the thymus and become equipped with T-cell receptors (TCRs) of unpredictable specificity, they are subjected to a negative-selection process in which cells bearing receptors that react strongly with self antigens are removed. Recently, however, it has become evident2 that numerous autoreactive T cells continually leave the thymus and enter secondary lymphoid organs, such as the spleen and lymph nodes, where immune responses are initiated. Keeping these autoreactive T cells in check is achieved in part by regulatory T (Treg) cells, which mostly bear TCRs that also bind self antigens. When Treg cells lose their function, potentially lethal autoimmune diseases develop in both mice and humans3.

What is it that guides Treg cells to focus their immune-regulatory activities on autoreactive effector T cells in the vast expanse of the secondary lymphoid organs, where they are outnumbered 10 to 1 by their non-regulatory counterparts? Liu et al. observed in healthy mice that most Treg cells are evenly dispersed in the areas of lymphoid tissues that are densely packed with T cells, but some small clusters of Treg cells form in the lymph nodes' superficial paracortex. This region has the highest density of migratory dendritic cells (DCs) — immune cells that survey tissues and take up both self and foreign antigens before migrating through the lymphatic system to the tissue-draining lymph nodes to present these antigens to T cells.

The observation that clustered Treg cells are in preferential contact with migratory DCs even in the absence of foreign antigens suggests that the clustering is driven by the recognition of self antigens (Fig. 1). Indeed, the authors observed that Treg cells lost the ability to form clusters when their TCRs were deleted. Two previous studies4,5 showed that TCR signalling is required for mature Treg cells to acquire and sustain their full suppressive activity, which includes elevated expression of regulatory molecules such as CD73 and CTLA4. Intriguingly, clustered Treg cells also expressed the largest quantities of these molecules, and this property was lost after TCR deletion, indicating that TCR triggering not only positions Treg cells in clusters, but also increases their suppressive activity.

Figure 1: Regulatory T-cell control of autoreactive T cells.
Figure 1

Some types of the immune system's dendritic cells (DCs) collect molecular samples of our tissue environments and migrate to draining lymph nodes. Here, they present their samples to T cells, which express unique surface receptors that recognize specific molecular structures (antigens). Some of these T-cell receptors will bind self antigens; if such binding occurs and the T cell is activated, an autoimmune response could ensue. However, a specialized class of T cell — regulatory T (Treg) cells, which also recognize self antigens — can be activated to suppress these effector T-cell responses. Liu et al.1 suggest that Treg and autoreactive effector T cells encounter their specific antigens on migratory DCs, leading to initial activation and clustering of Treg cells close to effector T cells that produce the cell-signalling molecule IL-2. This cytokine then fuels further Treg-cell proliferation and/or function, and the fully activated Treg cells eventually suppress effector T-cell activation, possibly by acting directly on the effector T cells or indirectly by suppressing the stimulatory capacity of the DCs.

In addition to these TCR-dependent events, Liu et al. noted that the transcription factor STAT5 was phosphorylated (indicating that it was activated) in most clustering Treg cells. This phosphorylation was driven by a cytokine (a secreted cell-signalling protein) called interleukin (IL)-2, which was produced by effector T cells at the centre of Treg-cell clusters. When the researchers treated their mice with an antibody that inhibits IL-2 function, they observed increased activation of autoreactive effector T cells, indicating a loss of suppression by Treg cells. This finding also suggests that the initial activation-driven proliferation of effector T cells does not require the IL-2 that they themselves produce, as previously reported6. Instead, the cytokine primarily activates a negative feedback loop involving IL-2-responsive Treg cells that are recruited to sites of autoreactive effector T-cell activation, to stop these dangerous responses in their tracks (Fig. 1).

Liu and colleagues' study suggests that the activation of autoreactive effector T cells is a surprisingly frequent event that seemingly brings us to the verge of autoimmune disease, but that rapid recruitment of Treg cells into functional niches containing self-antigen-presenting DCs ensures that these self antigens can be tolerated. Further research is needed to determine whether clustering is essential for suppression or merely a side effect of the TCR-driven Treg-cell activation process.

The precise mechanism by which Treg-cell clusters are formed is also still unclear. The most straightforward explanation is that they arise when Treg cells stabilize their interactions with those DCs that were activated either through previous interactions with autoreactive T effector cells or by other means (which could have triggered the autoreactive effector T-cell response in the first place). Another, but not mutually exclusive, possibility is that TCR signalling in Treg cells renders them responsive to chemoattractant molecules produced by activated DCs, rather like the spatial organization that allows one class of effector T cell (CD4+ T cells) to provide 'help' to another (CD8+ T cells)7. These considerations also raise the question of whether clustering Treg and autoreactive effector T cells respond to related self antigens, and whether they may encounter them on different DCs or must interact with the same DC.

It will also be interesting to explore the fates of autoreactive effector T cells once they are brought under control, and to investigate whether this interplay between Treg and effector T cells applies only in lymphoid tissues, or also at effector sites of immune responses, where regulatory function requires that both Treg cells and effector T cells re-encounter their specific antigens8. Furthermore, how can the barrier to complete T-cell activation described in this study be overcome during appropriate responses against pathogens, or break down where autoimmunity occurs?

Finally, it is appropriate to point out the tremendous potential of the integrative tissue-imaging technique used in this study, which the authors have named histocytometry9. In contrast to other multiplexed single-cell- analysis techniques, this method records the position of multiple cell types in intact tissues, along with quantitative information on their activation states and gene-expression patterns. As illustrated here, this can reveal minor functional populations that participate in local cellular networks but would probably be overlooked in any type of global analysis. In times of increasingly fine-grained molecular analyses and definitions of subsets of immune cells, detailed information on their microenvironmental contexts promises to eventually produce a true systems-level understanding of the immune system.



  1. 1.

    et al. Nature 528, 225–230 (2015).

  2. 2.

    & Nature Immunol. 15, 815–823 (2014).

  3. 3.

    & Cold Spring Harb. Perspect. Biol. 4, a007021 (2012).

  4. 4.

    et al. Immunity 41, 722–736 (2014).

  5. 5.

    , , & Nature Immunol. 15, 1070–1078 (2014).

  6. 6.

    et al. J. Immunol. 183, 332–339 (2009).

  7. 7.

    et al. Nature 440, 890–895 (2006).

  8. 8.

    et al. J. Clin. Invest. 124, 2425–2440 (2014).

  9. 9.

    , , , & Immunity 37, 364–376 (2012).

Download references

Author information


  1. Esteban Carrizosa and Thorsten R. Mempel are at Harvard Medical School and the Center for Immunology and Inflammatory Diseases, Massachusetts General Hospital, Boston, Massachusetts 02114, USA.

    • Esteban Carrizosa
    •  & Thorsten R. Mempel


  1. Search for Esteban Carrizosa in:

  2. Search for Thorsten R. Mempel in:

Corresponding author

Correspondence to Thorsten R. Mempel.

About this article

Publication history



Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Newsletter Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing