A Cell‐Intrinsic Role for RANK in Regulatory T Cells of the Human Thymus

Abstract

We identified RANK (TNFRSF11a) expression in human thymic FOXP3⁺ Tregs, revealing a previously unknown role for the RANK‐RANKL axis in human tTreg development. We show that IL‐2 and IL‐15 can drive RANK expression in CD4⁺ tTregs, which in turn may modulate their maturation and proliferation within the thymic environment. Created in BioRender. Biorender, I. (2026) https://BioRender.com/gbrjlpa.

Abbreviations

DP

double‐positive

SP

single‐positive

tTreg

thymic regulatory T cells

Thymic Regulatory T cells (tTreg) are essential players in T cell homeostasis, keeping autoimmunity in check [1]. Signals active during the commitment steps, particularly mediated by the TCR and/or IL‐2 and IL‐15, shape their quality, translating into specific tTreg functions [1, 2, 3]. tTreg selection also involves interactions with thymic epithelial cells (TECs), which occur both ways, as thymocytes also provide signals for TEC maturation [4]. The TNF receptor superfamily member RANKL, produced by developing thymocytes, binds RANK on TECs and induces their differentiation towards AIRE‐expressing selecting mTECs [5]. We propose here a cell‐intrinsic role for RANK in CD4 single‐positive (SP) Tregs within the human thymus. RANK–RANKL abrogation has recently been shown to result in pregnancy‐associated diabetes, demonstrating its importance for tolerance [6]. We previously performed bulk RNAseq analysis on sorted CD4SP tTreg and T conventional (Tconv) from human thymic tissue and found TNFRSF11A (RANK) expression in thymic CD4SP Tregs, while it was absent in CD4SP Tconv [7] (Figure S1A). Following these findings, we evaluated RANK (TNFRSF11A), RANKL (TNFSF11), FOXP3, and OPG (TNFRSF11B) expression in human sorted tTreg and Tconv counterparts from CD4⁺CD8⁺ double positive (DP) and CD4SP or CD8SP thymocytes (Figure S1B), using RT‐qPCR. RANK expression was found not only in CD4SP tTregs (Figure 1), but also in DP and in CD8SP tTregs (Figure S1C), while it was barely detectable in Tconv thymocytes (Figures 1 and S1C). Notably, RANK and RANKL featured similar expression levels in DP and CD4SP tTregs (Figures 1 and S1C). FOXP3 expression was found on the sorted tTregs, and OPG was virtually undetectable in all thymocyte populations (Figures 1 and S1D). Our findings concur with available datasets from human thymus in Park et al. (Figure S12 of that report) [8], and align with initial descriptions of RANK, using human stimulated PBMCs [9]. RANK expression across DP precursors, CD4SP, and CD8SP tTregs points to a range of stage‐specific roles. Additionally, RANK‐expressing tTregs may also compete with TEC precursors for RANKL, influencing mTEC maturation. However, our findings diverge from what has been described in reporter‐based mouse models, where RANK expression was not detected [5], in addition to other differences between mouse and human thymocytes, including the CD4⁺ Immature SP developmental stage or Class‐II expression in human thymocytes [10].

FIGURE 1.

RANK is expressed in human thymic CD4⁺ Tregs and is upregulated upon IL‐2 and IL‐15 stimulation. (A) RANK (TNFRSF11A), RANKL (TNFSF11), and FOXP3 expression levels in sorted CD4SP Tregs and Tconv (n = 5 thymi). (B) RANK relative expression levels in sorted CD4SP Tregs and Tconv (n = 3 thymi) stimulated for 24 h in the presence of IL‐2 (10U/mL), IL‐7 (10 ng/mL), and IL‐15 (12.5 ng/mL). Genes were quantified using TaqMan RT‐PCR (reference genes ACTB and 18S). Each dot represents a different biological replicate; Mean and SEM are depicted; comparisons performed using the Mann‐Whitney test or using ANOVA and Dunnett's post‐test; significant values shown as *p < 0.05; **p < 0.01; ***p < 0.001.

We next evaluated whether γ_c cytokines would regulate the expression of RANK on tTregs. We sorted thymic tTreg and Tconv CD4SP, immediately cultured them in the presence or absence of IL‐2, IL‐7, or IL‐15 for 24 h, and quantified RANK and RANKL expression by RT‐qPCR (Figure 1B and Figure S2). We observed that the expression of RANK was increased in tTregs in the presence of IL‐2, and, although to a minor extent, also in the presence of IL‐15, suggesting that this upregulation is part of the response to these cytokines (Figure 1B). Interestingly, RANKL expression levels were not altered upon the cytokine treatment in Tregs, showing that the γ_c cytokine regulation is restricted to RANK. Conversely, IL‐7 and IL‐15 appeared to modulate RANKL expression in Tconvs, while RANK was still not expressed (Figure S2). Taking into consideration the segregation of IL‐2 and IL‐15 production within the thymic tissue [3], these results also suggest that Tregs may respond to RANKL according to their spatial localization.

We next investigated the functional impacts of RANK expression in human tTregs, culturing freshly sorted CD4SP Tregs with or without exogenous RANKL (Figure 2). Cultures were performed in the presence of IL‐2 or IL‐15, taking into consideration the RANK upregulation in their presence. We evaluated immunophenotype, survival, and proliferation readouts on recovered Tregs, using flow cytometry (Figure S3). At earlier time points (48 h), RANKL impacted the phenotype of the recovered Tregs, with lower CD45RA and ICOS expression levels and a higher fraction of CD69^pos cells observed (Figure 2). Effects appeared milder in the presence of IL‐15, although only two thymi could be analyzed (Figure S4A). RANKL stimulation also promoted proliferation, as assessed by the expression of Ki‐67 (MKI67), while BCL2 expression levels were unchanged, suggesting no influence on survival (Figure 2). These results suggest an impact of the RANK/RANKL pathway on the modulation of the maturation kinetics of tTregs. Also, as CD69 and Ki‐67 expressions segregate (Figure 2), it may occur that different subpopulations respond differently to RANK‐mediated signals. The physiological effects of RANKL may be underestimated in our system, as the thymic 3D structure and other cell types, not present in culture, are important factors in thymocyte development, and endogenous RANKL could not be accounted for in these experiments. Our preliminary results using OPG from culture onset to hinder endogenous RANKL availability showed, at a later time‐point experiment (96 h), that RANKL and OPG tend to lead to opposite direction changes, namely in the expressions of CD45RA and ICOS in the presence of IL‐2, although the mild phenotype impacts impose future time‐course studies, as peak RANKL effects might occur earlier (Figure S4B).

FIGURE 2.

RANKL impacts maturation and proliferation of human thymic Tregs. Sorted human thymic Tregs (n = 3 thymi) were stimulated with RANKL (2 µg/mL) or cultured with medium+IL2 for 48 h; Histograms from a representative thymus and graph showing ratio of the Median Fluorescence Intensity of the indicated marker in the presence of RANKL versus the medium+IL‐2 condition; illustrative dot‐plots of CD69 and Ki67 expression and graph showing the ratio of the each marker frequency; each dot represents an individual; paired ratio t‐test was used and significant p‐values are shown *p < 0.05.

The observation of RANK expression across tTreg development, including in DP and CD8 SP tTregs, raises important questions concerning its possible role in Treg commitment and how it may modulate a Treg impact on TEC maturation. The precise characterization of RANK expression in tTregs is, however, still hostage to unreliable antibody detection, and it is thus still unknown whether given subpopulations or developmental windows are primary responders to RANK‐RANKL modulation.

Altogether, our results highlight the need to study Treg development in the human thymus. We identified a new signaling axis acting in human tTregs that is absent in mouse thymocytes, pointing to the importance of further studies to fully elucidate the role of RANK in human tTregs, which may translate into other tissues and need to be taken into consideration in the design and interpretation of RANK‐RANKL therapeutic modulation.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File: eji70132‐sup‐0001‐SuppMat.pdf.

Junginger Z.‐I., Adameck R., Raposo A. A. S. F., Rosmaninho P., Sousa A. E., and Almeida A. R. M., “A Cell‐Intrinsic Role for RANK in Regulatory T Cells of the Human Thymus.” European Journal of Immunology 56, no. 2 (2026): e70132. 10.1002/eji.70132

Data Availability Statement

The datasets referred to in this study can be found in an online repository: https://www.ebi.ac.uk/arrayexpress/E‐MTAB‐11211.