(en) Regulatory T cells (Tregs) are a subset of CD4+ T cells specialized in the inhibition of immune responses. Tregs are required for the maintenance of self-tolerance, but an excess of Treg function may inhibit anti-tumor immunity. Human Tregs are difficult to study due to the lack of a specific marker. To circumvent that issue, we derived stable human Treg clones from cancerous and non-cancerous patients. Treg clones were defined on the basis of in vitro suppressive function and were characterized by a stable epigenetic mark not found in other types of T cells. To gain insight into the suppression mechanisms by which Tregs inhibit immune responses, we performed gene expression profiling. We compared transcriptomes of Treg and T helper (Th) clones after stimulation through the T cell receptor (TCR). The only feature we found to distinguish stimulated Treg clones from Th clones was a signature suggestive of autocrine Transforming Growth Factor-beta (TGF-b) signaling. We showed that although both Treg and Th clones produce the latent, inactive form of TGF-b, only Treg clones produce active TGF-b after TCR stimulation. Furthermore, we provided evidence that Treg-derived TGF-b has paracrine actions on neighboring T cells, and that this effect requires cell-cell contact. Finally, we demonstrated that the in vitro suppressive function of Tregs relies at least in part on the production of active TGF-b. TGF-b is initially produced as a latent, inactive form, in which the mature cytokine is associated to the so-called Latency-Associated Peptide (LAP), preventing binding of the cytokine to its receptor. A subsequent step of activation is required to release active TGF-b from the LAP. How Tregs activate latent TGF-b is presently unknown. We showed that stimulated Treg clones, but not Th clones, bear LAP at their surface. Searching for a LAP receptor on Tregs, we observed that transmembrane protein GARP is expressed in stimulated Treg clones, but not in Th clones. We also found that GARP binds LAP and mature TGF-b, i.e. latent TGF-b. Forced GARP expression induced by lentiviral transduction in Th cells is sufficient to induce the binding of latent TGF-b at the cell surface, confirming that GARP is a receptor for latent TGF-b. However, GARP-transduced Th cells do not produce active TGFb, indicating that GARP expression is not sufficient to induce the transformation of latent TGF-b into the active cytokine. In summary, our results show that human Tregs, like other types of T cells, produce latent TGF-b. Latent TGF-b localizes at the Treg surface through binding to GARP, a receptor that is expressed upon TCR stimulation in Tregs, but not in other T cells. Stimulated Tregs transform latent TGF-b into the active cytokine, which exerts paracrine actions that require cell-cell contact. We propose therefore that a unique functional feature of human Tregs, by comparison to other T cells, consists in the ability to activate latent TGF-b, and this occurs close to the cell surface. GARP may be implicated in the activation of latent TGF-b by Tregs, but it is not sufficient. Altogether, our findings open the way for the discovery of the mechanism by which Tregs activate latent TGF-b.
Stockis, J. (2011). Activation of the latent form of Transforming Growth Factor-beta occurs at the surface of human regulatory T cells. https://hdl.handle.net/2078.5/216617