RIPK1 plays a crucial role in maintaining regulatory T-Cell homeostasis by inhibiting both RIPK3- and FADD-mediated cell death

Abstract

Regulatory T (T_reg) cells play an essential role in maintaining immune balance across various physiological and pathological conditions. However, the mechanisms underlying T_reg homeostasis remain incompletely understood. Here, we report that RIPK1 is crucial for T_reg cell survival and homeostasis. We generated mice with T_reg cell-specific ablation of Ripk1 and found that these mice developed fatal systemic autoimmunity due to a dramatic reduction in the T_reg cell compartment caused by excessive cell death. Unlike conventional T cells, T_reg cells with Ripk1 deficiency were only partially rescued from cell death by blocking FADD-dependent apoptosis. However, simultaneous removal of both Fadd and Ripk3 completely restored the homeostasis of Ripk1-deficient T_reg cells by blocking two cell death pathways. Thus, our study highlights the critical role of RIPK1 in regulating T_reg cell homeostasis by controlling both apoptosis and necroptosis, thereby providing novel insights into the mechanisms of T_reg cell homeostasis.

Introduction

Regulatory T (T_reg) cells, a subset of CD4⁺ T cells, are essential for maintaining immune homeostasis and self-tolerance [1]. While T_reg cells are mainly generated in the thymus, they can also be induced from naive T cells in the periphery. In the steady state, T_reg cells are a relatively stable population, and abnormalities in their homeostasis are often associated with autoimmune or immunodeficient diseases. To maintain this relative stability, T_reg cells need to coordinate survival and death signals from cytokines, TCR/costimulatory signaling and anatomical location [2]. Similar to the role of IL-7 in conventional T-cell homeostasis, IL-2 plays a pivotal role in promoting T_reg cell homeostasis by driving the upregulation of antiapoptotic proteins [3, 4]. However, not all T_reg cells depend on IL-2 for homeostatic maintenance. For example, CD62L^lowCD44^high activated T_reg (aT_reg) cells rely on TCR signals and the costimulatory molecule ICOS to maintain their phenotypical identity and homeostasis [5, 6]. Sensing and responding to diverse survival or death signals, many intracellular factors are also involved in the maintenance of T_reg population homeostasis. Some studies have shown that Bak- and Bax-mediated intrinsic apoptosis pathways help constrain the T_reg population, but Mcl-1, upregulated by IL-2 signals, inhibits Bak- and Bax-mediated T_reg cell apoptosis [3]. Compared to extrinsic signals, the intracellular factors that regulate T_reg homeostasis remain poorly understood.

Receptor-interacting protein kinase 1 (RIPK1) plays a vital role in regulating apoptosis, necroptosis and inflammation, depending on the cellular context. Among these pathways, the TNFR1-mediated RIPK1 activation pathway has been extensively studied. Upon activation by TNF-α, TNFR1 recruits TNFR-associated death domain protein (TRADD), RIPK1, TRAF2, and cIAP1/2 to form complex I [7]. Within complex I, RIPK1 undergoes polyubiquitination, activating the pro-survival NF-kappa B (NF-κB) pathway. If complex I is disturbed, a new cytosolic complex IIa is formed by internalized TNFR1 and deubiquitinated RIPK1, resulting in cell apoptosis [8, 9]. In cases where apoptosis is inhibited, RIPK1 interacts with RIPK3 through the RIP homotypic interaction motif (RHIM) to form complex IIb, which facilitates the phosphorylation and oligomerization of mixed lineage kinase domain-like protein (MLKL), resulting in necroptosis [10, 11]. Consistent with these findings, genetic studies in mice have demonstrated the crucial role of RIPK1 in maintaining tissue homeostasis by regulating apoptosis, necroptosis, and inflammation. Specifically, Ripk1-deficient mice die at birth due to systemic inflammation, which can be prevented by inhibiting both FADD/Caspase8-dependent apoptosis and RIPK3/MLKL-dependent necroptosis [12]. Furthermore, studies on conditional Ripk1 knockout mice have shown that RIPK1 plays a critical role in regulating skin and intestinal inflammation, autoimmunity, and tissue fibrosis [13–15].

In addition, recent studies have identified individuals with biallelic loss-of-function mutations in the RIPK1 gene who displayed a range of clinical manifestations with lymphopenia [16, 17]. Consistently, mice with targeted deletion of RIPK1 in T cells also showed severe T lymphopenia [18]. These findings strongly indicate the importance of RIPK1 in T-cell function and highlight its relevance in the immune system. However, considering that T_reg cells have unique homeostatic mechanisms distinct from those found in conventional T cells, there is still great interest in investigating whether RIPK1 plays a role in T_reg cell homeostasis and function. In this study, we generated mice with specific Ripk1 ablation in T_reg cells and found that these mice developed fatal systemic autoimmunity due to a significant reduction in the T_reg cell compartment caused by excessive cell death. Unlike conventional T cells, which undergo FADD-dependent apoptosis when RIPK1 is absent, T_reg cells with Ripk1 deficiency exhibit only partial rescue with Fadd deletion. Rather, we discovered that simultaneous inhibition of both FADD-dependent apoptosis and RIPK3-dependent necroptosis fully rescued Ripk1-deficient T_reg cells from excessive death. These results highlight the critical role of RIPK1 in regulating T_reg cell survival by controlling both apoptosis and necroptosis, providing new insights into the mechanisms underlying T_reg cell homeostasis.

Results

RIPK1 is upregulated in activated T_reg cells

To gain a better understanding of the role of RIPK1 in T_reg cells, we first examined its expression in these cells. Based on the expression levels of the cell surface markers CD44 and CD62L, T_reg cells can be divided into two populations: a resting T_reg cell population (rT_reg) characterized by CD44^lowCD62L^high and an activated T_reg cell population (aT_reg) characterized by CD44^highCD62L^low [19, 20]. Both immunoblotting and intracellular staining results revealed that aT_reg cells expressed higher levels of RIPK1 than quiescent rT_reg cells (Fig. 1A, B). Interestingly, effector T_reg cells, identified as KLRG1⁺ICOS⁺ aT_reg cells, exhibited the highest level of RIPK1 among all T_reg cells (Fig. 1B). TCR signals are important for the phenotypical identity of aT_reg cells [5]. The high expression of RIPK1 in aT_reg cells led us to suspect that it may be regulated by TCR signals. We observed that TCR stimulation in vitro for 24 h significantly increased the protein level of RIPK1 in T_reg cells (Fig. 1C). Notably, IL-2 stimulation did not lead to an upregulation of RIPK1 expression in T_reg cells (Supplementary Fig. 1A). These results suggest that RIPK1 is differentially expressed in T_reg cells under different activation states and may have varying roles in different populations of T_reg cells.

Fig. 1.

T_reg cell-specific deletion of RIPK1 leads to fatal systemic autoimmunity. A Immunoblot analysis of RIPK1 expression in sorted YFP⁺CD4⁺CD44^lowCD62L^high T_reg cells (rT_reg cells) and YFP⁺CD4⁺CD44^highCD62L^low T_reg cells (aT_reg cells) from Foxp3^Cre mice. B Intracellular staining of RIPK1 expression in rT_reg cells, KLRG1^lowICOS^low aT_reg and KLRG1^highICOS^high aT_reg cells from wild-type (WT) mice and T_reg cells from Ripk1^f/fCD4^Cre mice. C Immunoblotting of RIPK1 in sorted YFP⁺ WT T_reg cells stimulated with plate-bound anti-CD3/CD28 for the indicated time points. D Representative images of 6-week-old Foxp3^Cre and Ripk1^f/fFoxp3^Cre mice and the lymph organs isolated from these mice. E Survival curve of Foxp3^Cre and Ripk1^f/fFoxp3^Cre mice at the indicated time points (n = 10). F Hematoxylin and eosin staining of the indicated organs from Foxp3^Cre and Ripk1^f/fFoxp3^Cre mice. Arrows indicate the areas of immune cell infiltration. The magnification is 100×. G Flow cytometry analysis of CD44 and CD62L expression on T cells in the lymph nodes of 6-week-old Foxp3^Cre and Ripk1^f/fFoxp3^Cre mice. Numbers adjacent to the outlined areas indicate the percentages of CD44^highCD62L^low activated T cells. H Flow cytometry analysis of IFN-γ and IL-4 expression in CD4⁺ T cells and IFN-γ and granzyme B (GzmB) expression in CD8⁺ T cells in the lymph nodes of 6-week-old Foxp3^Cre and Ripk1^f/fFoxp3^Cre mice after 6-h stimulation with PMA plus ionomycin. Numbers in quadrants indicate the percentage of cells in each respective quadrant. Data in (A–C) are representative of at least two independent experiments. Data in (D, F–H) are representative of at least three independent experiments

T_reg cell-specific ablation of Ripk1 results in fatal systemic autoimmunity

To investigate the role of RIPK1 in T_reg cells, we crossed mice with loxP-flanked Ripk1 alleles (Ripk1^f/f) with Foxp3^YFP-Cre knock-in mice (Foxp3^Cre) to delete RIPK1 specifically in T_reg cells (Ripk1^f/fFoxp3^Cre) (Supplementary Fig. 1B). The Ripk1^f/fFoxp3^Cre mice showed stunted growth and a markedly reduced lifespan, typically surviving only 3–12 weeks (Fig. 1D, E). These mice also manifested significant splenomegaly and lymphadenopathy (Fig. 1D). Histological analysis revealed extensive leukocyte infiltration in the skin, lung, liver, and colon of Ripk1^f/fFoxp3^Cre mice (Fig. 1F). Flow cytometry analysis consistently showed a substantially higher percentage of Gr-1⁺ myeloid cells in the lymph nodes of these mice (Supplementary Fig. 1C). Furthermore, there was a significant increase in CD44^highCD62L^high memory or CD44^highCD62L^low effector T cells in these mice, which produced increased levels of effector cytokines such as IFN-γ, IL-4 and IL-17A in CD4⁺ T cells, as well as IFN-γ and granzyme B in CD8⁺ T cells (Fig. 1G, H, Supplementary Fig. 1D). These phenotypes indicate severe defects in T_reg cells resulting from Ripk1 deletion.

RIPK1 is crucial for maintaining peripheral T_reg cell homeostasis

We proceeded to analyze the frequencies of T_reg cells in the thymus, as well as various lymphoid and nonlymphoid organs, in both Ripk1^f/fFoxp3^Cre mice and control Foxp3^Cre mice aged 2 to 6 weeks. Our findings revealed a significant reduction in the percentages of T_reg cells in the spleens, lymph nodes, lungs, livers, small intestines and colons of Ripk1^f/fFoxp3^Cre mice (Fig. 2A, B). Interestingly, these decreases in T_reg cell percentages were already observed in Ripk1^f/fFoxp3^Cre mice as early as 3 weeks of age (Fig. 2A, B). It is worth noting that these reductions in T_reg cells were not attributed to developmental defects since the frequencies of T_reg cells in the thymi of wild-type mice were comparable to those in the thymi of Ripk1^f/fFoxp3^Cre or Ripk1^f/fCD4^Cre mice (Fig. 2A, B, Supplementary Fig. 2A, B). These results strongly suggest that RIPK1 serves as a crucial regulator of T_reg cell homeostasis in the periphery.

Fig. 2.

RIPK1 is critical for the maintenance of peripheral T_reg homeostasis. A Flow cytometry analysis of YFP⁺ T_reg cells in CD4SP thymocytes in the thymus or in CD4⁺ T cells in the lymph nodes (LNs), spleen, lungs, liver, small intestine and colon of 3-week-old Foxp3^Cre and Ripk1^f/fFoxp3^Cre mice. Numbers adjacent to the outlined areas indicate the percentages of YFP⁺ T_reg cells. B Kinetics of the frequencies of YFP⁺ T_reg cells in the indicated organs from Foxp3^Cre and Ripk1^f/fFoxp3^Cre mice at the indicated ages (n = 3–5). C Flow cytometry analysis of YFP⁺Foxp3⁺ T_reg cells in the lymph nodes of female Foxp3^Cre/+ and Ripk1^f/fFoxp3^Cre/+ mice. The left panel shows a representative flow cytometry plot highlighting YFP⁺Foxp3⁺ T_reg cells in CD4⁺ T cells in the lymph nodes of 6-week-old female Foxp3^Cre/+ and Ripk1^f/fFoxp3^Cre/+ mice, with numbers adjacent to the outlined areas indicating the percentages of YFP⁺Foxp3⁺ T_reg cells. The right panel summarizes the frequencies and numbers of YFP⁺Foxp3⁺ T_reg cells in the lymph nodes of female Foxp3^Cre/+ and Ripk1^f/fFoxp3^Cre/+ mice at the indicated ages (n = 4). D Flow cytometry analysis of CD44^highCD62L^low activated T_reg (aT_reg) cells among the YFP⁺ T_reg cells in the lymph nodes of 6-week-old female Foxp3^Cre/+ and Ripk1^f/fFoxp3^Cre/+ mice. The numbers adjacent to the outlined areas indicate the percentages of aT_reg cells. Data in (A–D) are representative of at least two independent experiments. Data in (B, C: right panel) are mean ± s.e.m., ns not significant, *p < 0.05, **p < 0.01, ***p < 0.001 (two-tailed Student’s t test)

To investigate whether RIPK1 intrinsically regulates T_reg cell homeostasis, we analyzed T_reg cells in female Ripk1^f/fFoxp3^Cre/+ and Foxp3^Cre/+ mice. In female Ripk1^f/fFoxp3^Cre/+ mice, a portion of T_reg cells retained RIPK1 due to random X-chromosome inactivation, which prevented the development of autoimmune diseases. Our analysis revealed a gradual decrease in RIPK1-deficient YFP⁺ T_reg cells starting at 4 weeks of age in female Ripk1^f/fFoxp3^Cre/+ mice compared to their counterparts in female Foxp3^Cre/+ mice (Fig. 2C). Interestingly, the loss of CD44^highCD62L^low aT_reg cells was more pronounced in female Ripk1^f/fFoxp3^Cre/+ mice than the loss of CD44^lowCD62L^high rT_reg cells (Fig. 2D), which aligns with the higher RIPK1 protein expression in aT_reg cells. Thus, our findings suggest that RIPK1 plays a crucial role in regulating T_reg cell homeostasis in the periphery through an intrinsic cellular mechanism.

T_reg cells lacking RIPK1 are more susceptible to cell death

RIPK1 plays a dual role in regulating cell survival and death under different conditions. To investigate the primary cause of T_reg cell reduction in Ripk1^f/fFoxp3^Cre mice, we first examined cell death in T_reg cells, and we found that ex vivo RIPK1-deficient T_reg cells exhibited a higher rate of apoptotic cell death, as indicated by the Annexin V⁺7-AAD⁻ cell population (Fig. 3A). Similarly, both rT_reg and aT_reg cells from Ripk1^f/fFoxp3^Cre mice expressed elevated levels of cleaved caspase-3 protein compared with their wild-type counterparts (Fig. 3B, Supplementary Fig. 3A), indicating that RIPK1-deficient T_reg cells are more susceptible to apoptosis. Although IL-2 and TCR signals are essential for T_reg cell homeostasis, we observed no significant differences in cell death between RIPK1-sufficient and RIPK1-deficient cells when treated with IL-2 (Fig. 3C, Supplementary Fig. 3B). This suggests that RIPK1 may not be needed for IL-2-mediated T_reg cell survival. However, TCR stimulation resulted in significantly increased cell death in RIPK1-deficient cells (Fig. 3C, D, Supplementary Fig. 3B), indicating that RIPK1 potentially offers crucial protection against activation-induced T_reg cell death. Notably, we found no obvious defects in the proliferation of RIPK1-deficient T_reg cells through Ki-67 protein staining and CellTrace dilution analysis (Supplementary Fig. 3C, D).

Fig. 3.

RIPK1-deficient T_reg cells have a higher susceptibility to cell death. A Flow cytometry analysis of apoptotic and dead YFP⁺ T_reg cells from 6-week-old Foxp3^Cre and Ripk1^f/fFoxp3^Cre mice using Annexin V and 7-AAD staining. B Immunoblotting of cleaved Caspase-3 protein in sorted aT_reg and rT_reg cells from 3-week-old Foxp3^Cre and Ripk1^f/fFoxp3^Cre mice. C Cell viability analysis of sorted T_reg cells from Foxp3^Cre and Ripk1^f/fFoxp3^Cre mice after stimulation with IL-2 (500 U/mL) or plate-bound anti-CD3/CD28 plus IL-2 for 48 h (n = 4). D Flow cytometry analysis of apoptotic and dead T_reg cells sorted from Foxp3^Cre and Ripk1^f/fFoxp3^Cre mice after stimulation with plate-bound anti-CD3/CD28 for 6 h using Annexin V and 7-AAD staining. E Volcano plot of RNA-seq data, with representative survival, activation/cytotoxic and T_reg signature genes shown. The horizontal dotted line indicates a p value of 0.05, and the vertical dotted lines indicate a fold change of 1.5. F Heatmap analysis of representative gene expression in Foxp3^Cre and Ripk1^f/fFoxp3^Cre T_reg cells. G Flow cytometry analysis of IFN-γ⁺ or GzmB⁺ T_reg cells in the spleen and LNs of 6-week-old Foxp3^Cre and Ripk1^f/fFoxp3^Cre mice. The left panel shows representative flow cytometry plots of IFN-γ⁺ and GzmB⁺ T_reg cells in the spleen and LNs. Numbers in quadrants indicate the percentage of cells in each respective quadrant. The right panel summarizes the percentages of IFN-γ⁺ and GzmB⁺ T_reg cells in the spleen or LNs (n = 3). Data in (A–G) are representative of at least two independent experiments. Data in (C) are mean ± s.e.m., ns not significant, *p < 0.05, and ***p < 0.001 (two-tailed Student’s t test)

To gain further insights into the regulation of T_reg cell homeostasis and functions by RIPK1, we performed a comprehensive analysis of the gene expression profiles of rT_reg and aT_reg cells from Ripk1^f/fFoxp3^Cre and wild-type mice using RNA-Seq and flow cytometry analysis. Our findings showed that the expression levels of T_reg cell core signature genes and proteins such as Foxp3, Ctla4, Il10 and Tgfb1 in Ripk1^f/fFoxp3^Cre T_reg cells were not lower than those in wild-type T_reg cells (Fig. 3E, Supplementary Fig. 4A–D), indicating that the loss of RIPK1 does not disrupt T_reg cell development. However, we observed that Ripk1^f/fFoxp3^Cre T_reg cells exhibited some phenotypic and transcriptomic features of cytotoxic T cells. Specifically, the expression levels of Tbx21, Cxcr3, Ifng, Gzma and Gzmb were substantially higher in Ripk1^f/fFoxp3^Cre T_reg cells than in wild-type T_reg cells (Fig. 3E–G), suggesting that RIPK1 may play a critical role in maintaining T_reg cell identity and function. As expected, Ripk1^f/fFoxp3^Cre T_reg cells showed reduced expression levels of antiapoptotic genes, such as Bcl2 and Bcl2l2, while the levels of proapoptotic genes, such as Bak1 and Casp3, were elevated (Fig. 3E, F). However, we found no significant changes or only slight increases in the expression levels of apoptosis-related and necroptosis-related receptors such as TNFR1, CD95, and PD-1 (Supplementary Fig. 4A, B). This suggests that RIPK1 promotes T_reg cell survival independent of the regulation of these cell death-related receptors. Interestingly, we also observed elevated expression levels of Mlkl and Zbp1, key molecules involved in the necroptosis pathway, in Ripk1^f/fFoxp3^Cre T_reg cells (Fig. 3E, F). This suggests that, in addition to apoptosis, RIPK1 may also participate in regulating necroptosis in T_reg cells.

Ablation of Fadd partially rescues RIPK1-deficient T_reg cells from cell death

RIPK1 participates in multiple signaling pathways, including FADD/Caspase8-dependent apoptosis and RIPK3/MLKL-dependent necroptosis. To investigate the involvement of the FADD/Caspase8 apoptotic pathway in the functions of RIPK1 in T_reg cells, we utilized Ripk1^f/fFadd^−/−Fadd:gfp^f/fCD4^Cre mice to evaluate the elimination of Fadd in RIPK1-deficient T cells (Supplementary Fig. 5). Previous studies have reported that RIPK1 is essential for maintaining homeostasis in conventional T cells by inhibiting apoptosis [18]. Consistent with these findings, we observed a significant reduction in CD4⁺ and CD8⁺ T cells in Ripk1^{f /f}CD4^Cre mice. Under lymphopenic conditions, there was an increase in the proportion of activated CD8⁺ T cells but not activated CD4⁺ T cells in the lymph nodes of Ripk1^{f/ f}CD4^Cre mice (Supplementary Fig. 6A). Despite the increase in activated CD8⁺ T cells, there were no evident signs of autoimmune diseases in young Ripk1^f/fCD4^Cre mice (Supplementary Fig. 6B). This might be attributed to the overall reduction in effector T cells, particularly CD4⁺ effector T cells, resulting from the deficiency of RIPK1. As expected, we observed that both CD4⁺ and CD8⁺ T cells in various organs of Ripk1^f/fFadd^−/−Fadd:gfp^f/fCD4^Cre mice were restored to the levels in wild-type mice (Fig. 4A). Furthermore, we observed that the proportions of naive (CD44^lowCD62L^high) and activated (CD44^highCD62L^low) T-cell subsets in Ripk1^f/fFadd^−/−Fadd:gfp^f/fCD4^Cre mice were similar to those in wild-type CD4⁺ and CD8⁺ T cells, respectively (Supplementary Fig. 6A). These findings indicate that the RIPK1-FADD pathway is vital for maintaining the homeostasis of conventional T cells.

Fig. 4.

Ablation of Fadd partially rescues RIPK1-deficient T_reg cells from cell death. A Flow cytometry analysis of CD4SP and CD8SP thymocytes in the thymus and CD4⁺ and CD8⁺ T cells in the lymph nodes (LNs), spleen, small intestine (SI), and colon from 6-week-old WT, Ripk1^f/fCD4^Cre and Ripk1^f/fFadd^-/-Fadd:gfp^f/fCD4^Cre mice (upper panel); numbers adjacent to the outlined areas indicate the percentages of CD4SP and CD8SP in the thymus and CD4⁺ T and CD8⁺ T cells in the LNs, spleen, SI, and colon. The percentages of CD4SP or CD4⁺ T cells are summarized (n = 4, lower panel). B Flow cytometry analysis of Foxp3⁺ T_reg cells in the thymus, LNs, spleen, SI and colon from 6-week-old WT, Ripk1^f/fCD4^Cre and Ripk1^f/fFadd^-/-Fadd:gfp^f/fCD4^Cre mice (upper panel). Numbers adjacent to the outlined areas indicate the percentages of Foxp3⁺ T_reg cells in the thymus, LNs, spleen, SI and colon. The percentages of Foxp3⁺ T_reg cells are summarized (n = 4, lower panel). Data in (A, B) are representative of at least two independent experiments. Data in (A: lower panel, B: lower panel) are mean ± s.e.m., ns, not significant, *p < 0.05, **p < 0.01, ***p < 0.001 (two-tailed Student’s t test)

Surprisingly, unlike conventional T cells, T_reg cells were only partially restored in the spleens and lymph nodes in Ripk1^f/fFadd^−/−Fadd:gfp^f/fCD4^Cre mice, although T_reg cells in the gut were rescued to wild-type levels (Fig. 4B). As the absence of RIPK1 and FADD in conventional T cells potentially impacts T_reg homeostasis in these mice, we conducted cotransfer experiments to confirm whether the reduction in T_reg cell number was intrinsic to T_reg cells themselves. We cotransferred T_reg cells from Ripk1^f/fFadd^−/−Fadd:gfp^f/fCD4^Cre mice and CD45.1⁺ wild-type T_reg cells into CD45.1⁺CD45.2⁺ wild-type recipient mice and analyzed their proportions in recipient mice on Day 7 and Day 14 after transfer. During these time points, we observed a gradual decrease in the proportion of T_reg cells from Ripk1^f/fFadd^−/−Fadd:gfp^f/fCD4^Cre mice (Supplementary Fig. 6C), suggesting that T_reg cells deficient in RIPK1 and FADD do have inherent survival issues. Notably, despite only partial restoration of T_reg cells in Ripk1^f/fFadd^−/−Fadd:gfp^f/fCD4^Cre mice, the extent of T_reg cell restoration was sufficient to prevent autoimmune diseases (Supplementary Fig. 6B). Taken together, our findings demonstrate the critical role of the RIPK1-FADD pathway in maintaining T_reg cell homeostasis and function, with other RIPK1-mediated pathways also being involved.

Pro-apoptotic BH3-only proteins have previously been reported to play a role in the regulation of T_reg cell apoptosis [3]. Therefore, we proceeded to investigate whether Bim, one such protein, contributes to the cell death observed in RIPK1-deficient T_reg cells using Bcl2l11^−/−Ripk1^f/fFoxp3^Cre mice. We found that Bim ablation did not facilitate the restoration of the T_reg population (Supplementary Fig. 7), which suggests that Bim may not be directly involved in the cell death of RIPK1-deficient T_reg cells.

We subsequently investigated whether blocking RIPK3/MLKL can facilitate the restoration of the T_reg population by utilizing Ripk3^−/−Ripk1^f/fFoxp3^Cre mice, which have both Ripk3 and Ripk1 deletion in T_reg cells. Compared to Ripk1^f/fFoxp3^Cre mice, Ripk3^−/−Ripk1^f/fFoxp3^Cre mice displayed a significant reduction in the percentages of T_reg cells in the spleen and lymph nodes (Supplementary Fig. 8A). As anticipated, Ripk3^−/−Ripk1^f/fFoxp3^Cre mice showed severe inflammatory symptoms characterized by the activation of conventional CD4⁺ T cells and CD8⁺ T cells, as well as lymphocyte infiltration in various organs (Supplementary Fig. 8B, C). These findings suggest that solely blocking necroptosis is insufficient to rescue T_reg cells from the death caused by RIPK1 deficiency.

Deletion of Fadd and Ripk3 completely rescues cell death in RIPK1-deficient T_reg cells

Since blocking either the Fadd pathway or the Ripk3 pathway alone cannot fully rescue T_reg cells from the death caused by RIPK1 deficiency, we hypothesized that RIPK1 deficiency may lead to both necroptosis and apoptosis in T_reg cells. To investigate this hypothesis, we utilized Ripk3^−/−Ripk1^−/−Fadd^−/− mice, which lack both FADD-mediated apoptosis and RIPK3-mediated necroptosis [12]. As expected, these mice exhibited normal conventional CD4⁺ and CD8⁺ T compartments (Fig. 5A), similar to Ripk1 and Fadd double knockout mice. However, unlike Ripk1 and Fadd double knockout mice, Ripk3^−/−Ripk1^−/−Fadd^−/− mice showed normal percentages of T_reg cells in all examined organs, including the thymus, spleen, lymph nodes, small intestine and colon (Fig. 5B). Additionally, we observed that T_reg cells from Ripk3^−/−Ripk1^−/−Fadd^−/− mice recovered from cell death under in vitro TCR stimulation (Supplementary Fig. 9A) and exhibited normal suppressive function in vitro compared to wild-type T_reg cells (Supplementary Fig. 9B). Moreover, no autoimmune diseases were detected in these mice (Supplementary Fig. 9C, D). These findings strongly suggest that the removal of Fadd and Ripk3 greatly improved the cell survival and functions of RIPK1-deficient T_reg cells. Since triple knockout mice are not conditional knockout mice, we further analyzed T_reg cells in chimeric mice reconstituted with Ripk3^−/−Ripk1^−/−Fadd^−/− bone marrow to validate our observations in a more controlled model. Compared to wild-type bone marrow chimeras, Ripk3^−/−Ripk1^−/−Fadd^−/− bone marrow chimeras also exhibited normal T_reg cell percentages and numbers (Supplementary Fig. 10). Together, these results demonstrate that RIPK1 functions to inhibit both FADD-mediated apoptosis and RIPK3-mediated necroptosis in T_reg cells.

Fig. 5.

Deletion of Fadd and Ripk3 completely rescues cell death in RIPK1-deficient T_reg cells. A Flow cytometry analysis of CD4SP and CD8SP thymocytes in the thymus and CD4⁺ and CD8⁺ T cells in the lymph nodes (LNs), spleen, small intestine (SI), and colon from 6-week-old WT, Ripk1^f/fFoxp3^Cre and Ripk3^-/-Ripk1^-/-Fadd^-/- mice (upper panel); numbers adjacent to the outlined areas indicate the percentages of CD4SP and CD8SP in the thymus and CD4⁺ T and CD8⁺ T cells in the LNs, spleen, SI, and colon. The percentages of CD4SP or CD4⁺ T cells are summarized in the lower panel (n = 5 in thymus; n = 3 in LNs, spleen, SI, colon). B Flow cytometry analysis of Foxp3⁺ T_reg cells in CD4SP cells in the thymus and CD4⁺ T cells in the LNs, spleen, SI and colon from 6-week-old WT, Ripk1^f/fFoxp3^Cre and Ripk3^-/-Ripk1^-/-Fadd^-/- mice (upper panel). Numbers adjacent to the outlined areas indicate the percentages of Foxp3⁺ T_reg cells in CD4SP in the thymus and CD4⁺ T cells in the LNs, spleen, SI and colon. The percentages of Foxp3⁺ T_reg cells are summarized (n = 3, lower panel). Data in (A, B) are representative of at least two independent experiments. Data in (A: lower panel, B: lower panel) are mean ± s.e.m., ns not significant, *p < 0.05, **p < 0.01, and ***p < 0.001 (two-tailed Student’s t test)

Discussion

T-cell homeostasis, achieved by a well-orchestrated balance of T-cell survival and death, is crucial for a functional immune system. Receptor-interacting protein kinase 1 (RIPK1) has been identified as a master regulator of apoptosis, necroptosis and inflammation, depending on cell type and context [14, 21–23]. Although RIPK1 is not essential for T-cell development in the thymus, it is crucial for conventional T-cell homeostasis because it inhibits apoptosis [18]. In our study, we observed an increase in the percentage of CD4SP thymocytes in 6-week-old Ripk1^f/fFoxp3^Cre mice compared to WT mice (Fig. 5A), which is likely due to a secondary effect of the severe autoimmune diseases present in Ripk1^f/fFoxp3^Cre mice. As a special subset of CD4⁺ T cells, T_reg cells have quite different homeostasis mechanisms from conventional CD4⁺ T cells. It remains unclear whether RIPK1 has similar functions in T_reg cell homeostasis. In this study, we found that T_reg cell-specific ablation of Ripk1 leads to fatal systemic autoimmunity due to excessive T_reg cell death. Unlike conventional T cells, T_reg cells with RIPK1 deficiency are only partially rescued by Fadd deletion. However, the deletion of both Fadd and Ripk3 completely rescued T_reg cells from death in Ripk1-deficient mice. Our results demonstrate that RIPK1 regulates T_reg cell survival by controlling both necroptosis and apoptosis.

Mature T cells, including T_reg cells, rely on TCR signaling to maintain their homeostasis. However, the mechanisms by which TCR regulates T-cell homeostasis remain to be elucidated. Our study reveals that TCR stimulation can induce the upregulation of RIPK1, which plays a vital role in T_reg cell survival upon TCR stimulation. This indicates that RIPK1 function is essential for TCR-mediated T-cell homeostasis. Although there is currently no direct evidence implicating RIPK1 in TCR signal transduction, it has been demonstrated that MALT1, downstream of TCR signals, regulates the activation of caspase-8, which can form a complex with RIPK1 and c-FLIPs [24, 25]. Therefore, it is plausible that RIPK1 promotes T_reg cell survival through its interaction with MALT1 and caspase-8, which warrants further investigation.

The RIPK1 protein comprises a kinase domain in its N-terminus, an RHIM domain, and a death domain in its C-terminus, each with distinct functional sites where posttranslational modifications such as ubiquitination and phosphorylation can occur [26]. Recent studies have demonstrated the critical role of RIPK1 ubiquitination at K376 in both apoptosis and necroptosis in vitro and in vivo [27–29]. However, in Ripk1^f/K376RCD4^Cre mice, we did not observe any abnormalities in T_reg cell development or homeostasis (Supplementary Fig. 11). This suggests that while RIPK1 ubiquitination on the K376-mediated pathway is essential for suppressing cell death during embryogenesis and postnatal inflammation, it is not required for T_reg cell homeostasis. Additionally, research has shown that phosphorylated RIPK1 can translocate to the nucleus, where it regulates chromatin remodeling and transcriptional control of inflammatory factors [30]. Therefore, in T_reg cell homeostasis and during T_reg activation, RIPK1 may serve not only as a vital protein for cell survival but also as a regulator of relevant gene expression. Our study revealed alterations in the expression levels of multiple genes in response to RIPK1 deficiency. Further investigation is necessary to determine whether these genes are directly regulated by phosphorylated RIPK1.

T_reg cells are not a homogenous population but instead display some degree of heterogeneity in their phenotypes, functions, and lineage origins [4]. Our study has revealed intriguing insights into the differential roles played by RIPK1 in various subpopulations of T_reg cells. Specifically, we found that RIPK1 deficiency leads to greater loss of activated T_reg (aT_reg) cells, suggesting that aT_reg cells rely more on RIPK1 to maintain homeostasis than resting T_reg (rT_reg) cells. Within the gut mucosa, T_reg cells include a significant proportion of peripherally induced T_reg (pT_reg) cells that are derived from conventional CD4⁺ T cells [31, 32]. Strikingly, our research demonstrated that under steady-state conditions, RIPK1-deficient T_reg cells in the gut can be fully rescued through the deletion of the Fadd gene, whereas the same restorative effect is not observed in their counterparts within secondary lymphoid organs. These findings imply that while RIPK1 may have similar functions in pT_reg cells and conventional CD4⁺ T cells, it exerts distinct effects on thymus-derived T_reg (tT_reg) cells.

Studies have provided evidence that the responsiveness of tumor-infiltrating T_reg cells to microenvironmental signals can disrupt their stability and survival, thereby influencing the outcome of antitumor immune responses [33]. Our findings suggest that under steady-state conditions, the primary function of RIPK1 is to suppress T_reg cell death and maintain self-tolerance. However, it remains to be determined whether manipulating RIPK1 expression in T_reg cells within the tumor microenvironment or other pathological contexts could be leveraged as a strategic approach to modulate disease progression. Further investigations are warranted to shed light on this important aspect.

In summary, our study highlights the critical role of RIPK1 in preserving T_reg cell homeostasis by restraining both RIPK3- and FADD-mediated cell death pathways. These findings provide fresh insights into the underlying mechanisms that govern the balance and stability of T_reg cells.

Materials and methods

Mice

The Ripk1^f/f mice were generated by a homologous recombination strategy (Shanghai Model Organisms Center, Inc.). In brief, hybrid mouse embryonic stem (ES) cells were electroporated with a targeting vector containing floxed exon 3. The Ripk3^−/− mice were provided by Dr. Xiaodong Wang (National Institute of Biological Sciences, Beijing, China), and the Fadd^−/−Fadd:gfp^f/f mice were provided by Dr. Jianke Zhang (Thomas Jefferson University, Philadelphia, PA, USA), as previously described [34]. CD4^Cre and Foxp3^YFP-Cre (Foxp3^Cre) mice were provided by Dr. Bin Li (Shanghai Institute of Immunology, Shanghai Jiao Tong University, Shanghai, China). Ripk1^f/fCD4^Cre and Ripk1^f/fFoxp3^Cre mice were generated by crossing Ripk1^f/f mice with CD4^cre or Foxp3^Cre mice, respectively. Ripk1^f/fFadd^−/−Fadd:gfp^f/fCD4^Cre mice were obtained by breeding Fadd^−/−Fadd:gfp^f/f mice with Ripk1^f/fCD4^Cre mice. The CD45.1⁺ mice were provided by Dr. Qibin Leng (Institut Pasteur of Shanghai). All mice were maintained under specific pathogen-free conditions at the Institut Pasteur of Shanghai or Tongji University in China. All animal experiments were performed following the protocols approved by the Institutional Animal Care and Use Committee of the Institut Pasteur of Shanghai or Shanghai East Hospital, Tongji University.

Flow cytometry and cell sorting

For surface marker staining, cells were washed with staining buffer (PBS containing 1% FBS and 1 mM EDTA). To block nonspecific binding, anti-CD16/CD32 antibodies (2.4G2) were added, followed by a 30-minute incubation with surface antibodies on ice. The antibodies used for FACS analysis included APC-Cy7-anti-CD4 (GK1.5, Thermo Fisher Scientific), PE-Cy7-anti-CD4 (GK1.5, Thermo Fisher Scientific), PE-Cy7-anti-CD8 (53–6.7, Thermo Fisher Scientific), PerCP-Cy5.5-anti-CD8 (53–6.7, BioLegend), PE-anti-CD44 (IM7, Thermo Fisher Scientific), APC-anti-CD62L (MEL-14, Thermo Fisher Scientific), PE-Cy7-anti-CD25 (PC61.5, Thermo Fisher Scientific), PE-anti-TNFR1 (55R-286, BioLegend), PE-Cy7-anti-PD-1 (29 F.1A12, BioLegend), and AF647-anti-FAS (Jo2, BD Bioscience). Dead cells were excluded using a Live/Dead Fixable Aqua Dead Cell staining kit (Thermo Fisher Scientific). Samples were acquired using an LSRFortessa flow cytometer (BD PharMingen) and analyzed by FlowJo software (TreeStar).

For cytokine staining, cells were stimulated with 10 ng/mL PMA (phorbol 12-myristate 13-acetate, Sigma, St. Louis, MO) and 1 μg/mL ionomycin (Sigma) for 4 h. After treatment with 10 μg/mL brefeldin A (Sigma) for another 2 h, surface markers were stained, and cells were fixed with 3.7% formaldehyde for 15 min at room temperature. Subsequently, the cells were permeabilized with 0.2% saponin in PBS on ice for 20 min before intracellular staining. Samples were acquired using an LSRFortessa flow cytometer (BD PharMingen) and analyzed by FlowJo software (TreeStar).

To sort T_conv and T_reg cells, CD4⁺ T cells were enriched by negative selection using biotin-labeled CD8a, TER-119 and B220 antibodies and streptavidin magnetic beads (Magnet sort, Thermo). CD44^lowCD62L^high naive CD4⁺ T cells or CD4⁺YFP⁺ T_reg cells were sorted by a FACS Aria II cell sorter (BD Bioscience). The sorted cell populations were >98% pure.

Isolation of tissue-resident lymphocytes

To isolate lamina propria (LP) mononuclear cells, small intestines and colons were dissected, and fat tissues and Peyer’s patches were removed. Predigestion was performed using PBS containing 1 mM DTT (Sigma) and 30 mM EDTA (AMResco) at 37 °C for 30 min. Then, the pretreated tissues were digested with Liberase (250 μg/mL, Roche) and DNase I (200 μg/mL, Sigma) in DMEM supplemented with 5% FBS at 37 °C for 30 min in a shaker. The digested tissues were then meshed in a 70 μm cell strainer and subjected to Percoll gradient (40%/80%) centrifugation. Mononuclear cells present in the interphase were collected, washed with PBS and resuspended in T-cell medium.

Construction of bone marrow chimeras

To generate bone marrow chimeras, bone marrow cells from either wild-type or Ripk3^−/−Ripk1^−/−Fadd^−/− mice were transferred to 6- to 8-week-old TCRβ^−/− recipient mice that had been irradiated with 450 rads. After 2 weeks of neomycin treatment, the chimeric mice were analyzed following a 6-week reconstitution period.

T_reg cell cotransference

T_reg cells were isolated from the spleens and lymph nodes of Ripk1^f/fFadd^−/−Fadd:gfp^f/fCD4^Cre mice or CD45.1⁺ wild-type mice using a T_reg cell isolation kit (Miltenyi Biotec). T_reg cells from Ripk1^f/fFadd^−/−Fadd:gfp^f/fCD4^Cre mice were then mixed with CD45.1⁺ wild-type cells at a 1:1 ratio and subsequently transferred into CD45.1⁺CD45.2⁺ wild-type recipient mice via intravenous injection. The proportions of transferred T_reg cells in the recipient mice were analyzed 1 and 2 weeks after transfer.

In vitro T_reg suppression assay

The in vitro T_reg suppression assay was performed as previously described [35]. Briefly, CD4⁺CD25⁻CD45RB^high naive CD4⁺ T cells and CD4⁺CD25⁺CD45RB^low T_reg cells were sorted using fluorescence-activated cell sorting (FACS). Naive CD4⁺ T cells were labeled with CellTrace Violet (Thermo Fisher Scientific) and cocultured with T_reg cells at a 1:1 ratio on a plate coated with anti-CD3/anti-CD28 antibodies for 72 h. Subsequently, the dilution of CellTrace was analyzed using flow cytometry.

Cell viability assay

Cell viability was measured using the CellTiter-Glo Luminescent Cell Viability Assay kit (Promega) according to the manufacturer’s instructions. The luminescence was recorded with a microplate luminometer (Thermo Scientific).

Histopathology

For histological analysis, the samples were fixed in 3.7% formaldehyde, dehydrated in ethanol, cleared with xylene, and embedded in paraffin. Thin sections, approximately five micrometers thick, were cut and mounted onto glass slides that had been precoated with poly-l-lysine. These sections were then stained with hematoxylin and eosin using standard procedures.

RNA-Seq analysis

Total RNA was extracted using TRIzol reagent (Life Technologies) following the manufacturer’s instructions. The quality of the RNA was evaluated, and sequencing was performed using the Illumina HiSeq X platform at Novogene (Beijing, China). Raw data from the sequencer underwent preprocessing, alignment, and deduplication before being exported as raw counts. The differential expression gene analysis was conducted using the R package DESeq2 [36]. For the generation of the heatmap, the raw counts were transformed using the variance stabilizing transformation method and used as input. The RNA-seq data have been deposited in the Gene Expression Omnibus under the primary accession code GSE230483.

Statistical analysis

Statistical analysis were performed using Prism 9 (GraphPad). Two-tailed Student’s t tests and paired t tests were used to calculate p values (*p < 0.05, **p < 0.01 and ***p < 0.001). The data presented in this article are the representative results of at least two independent experiments.

Author contributions

HZ and HW conceived and supervised the research; XD, LW, YZ, ZY, YC, YS, YZ, HZ and HW contributed to the project design and discussions; XD, LW and YZ conducted the experiments; QL, YZ, WZ, YT, TW and JD helped with the mixed bone marrow chimera experiments and some phenotype analysis; FD and JR assisted with mouse models; PH did the bioinformatics analysis; XD, LW, YZ, ZY, YZ, HZ and HW wrote and edited the manuscript.

Competing interests

The authors declare no competing interests.

Supplementary information

The online version contains supplementary material available at 10.1038/s41423-023-01113-x.