c-FLIP Expression in Foxp3-Expressing Cells Is Essential for Survival of Regulatory T Cells and Prevention of Autoimmunity

c-FLIP Expression in Foxp3-Expressing Cells Is Essential for Survival of Regulatory T Cells and Prevention of Autoimmunity

Report c-FLIP Expression in Foxp3-Expressing Cells Is Essential for Survival of Regulatory T Cells and Prevention of Autoimmunity Graphical Abstract ...

3MB Sizes 0 Downloads 3 Views

Report

c-FLIP Expression in Foxp3-Expressing Cells Is Essential for Survival of Regulatory T Cells and Prevention of Autoimmunity Graphical Abstract

Authors Carlos Plaza-Sirvent, Marc Schuster, Yvonne Neumann, Ulrike Heise, Marina C. Pils, Klaus Schulze-Osthoff, Ingo Schmitz

Correspondence [email protected]

In Brief The mechanisms regulating homeostasis of regulatory T (Treg) cells are incompletely understood. Plaza-Sirvent et al. demonstrate that expression of the caspase-8-inhibitor c-FLIP is essential for Treg survival and prevention of autoimmunity because mice lacking cFLIP in this cell type die early on due to fatal autoimmune disease.

Highlights d

Treg cells exhibit a high rate of apoptosis ex vivo

d

Treg cells express lower levels of c-FLIPL than Tcon

d

d

Treg-specific deletion of Cflar/c-FLIP results in a scurfy-like phenotype c-FLIP has additional functions in Treg cells than inhibition of CD95

Plaza-Sirvent et al., 2017, Cell Reports 18, 12–22 January 3, 2017 ª 2017 The Author(s). http://dx.doi.org/10.1016/j.celrep.2016.12.022

Cell Reports

Report c-FLIP Expression in Foxp3-Expressing Cells Is Essential for Survival of Regulatory T Cells and Prevention of Autoimmunity Carlos Plaza-Sirvent,1,2 Marc Schuster,1,2 Yvonne Neumann,1,2 Ulrike Heise,3 Marina C. Pils,3 Klaus Schulze-Osthoff,4,5 and Ingo Schmitz1,2,6,* 1Systems-Oriented Immunology and Inflammation Research Group, Department of Immune Control, Helmholtz Center for Infection Research, Inhoffenstraße 7, 38124 Braunschweig, Germany 2Institute of Molecular and Clinical Immunology, Otto-von-Guericke University, Leipziger Straße 44, 39120 Magdeburg, Germany 3Mouse Pathology Platform, Helmholtz Center for Infection Research, Inhoffenstraße 7, 38124 Braunschweig, Germany 4Interfaculty Institute for Biochemistry, University of Tu €bingen, Hoppe-Seyler-Str. 4, 72076 Tu €bingen, Germany 5German Cancer Consortium (DKTK) and German Cancer Research Center (DKFZ), Im Neuenheimer Feld 280, 69120 Heidelberg, Germany 6Lead Contact *Correspondence: [email protected] http://dx.doi.org/10.1016/j.celrep.2016.12.022

SUMMARY

Regulatory T (Treg) cells are critical for the shutdown of immune responses and have emerged as valuable targets of immunotherapies. Treg cells can rapidly proliferate; however, the homeostatic processes that limit excessive Treg cell numbers are poorly understood. Here, we show that, compared to conventional T cells, Treg cells have a high apoptosis rate ex vivo correlating with low c-FLIP expression. Treg-specific deletion of c-FLIP in mice resulted in fatal autoimmune disease of a scurfy-like phenotype characterized by absent peripheral Treg cells, activation of effector cells, multi-organ immune cell infiltration, and premature death. Surprisingly, blocking CD95L did not rescue Treg survival in vivo, suggesting additional survival functions of c-FLIP in Treg cells in addition to its classical role in the inhibition of death receptor signaling. Thus, our data reveal a central role for c-FLIP in Treg cell homeostasis and prevention of autoimmunity. INTRODUCTION Apoptosis is a highly regulated form of cell death and a major regulator of hematopoietic cell homeostasis (Bouillet and O’Reilly, 2009; Krammer et al., 2007). For example, apoptosis is an effective mechanism for the removal of excess lymphocytes during the contraction phase of an immune response. Apoptosis can be mediated by two major pathways. The intrinsic or mitochondrial pathway is controlled by anti- and pro-apoptotic Bcl-2 family members (Youle and Strasser, 2008). This pathway is activated upon cytokine deprivation, thereby acting as a mechanism of immune homeostasis (Bouillet and O’Reilly, 2009). On the other hand, the extrinsic pathway is activated by surface receptors of the tumor necrosis factor receptor super-

family, such as CD95 (Krammer et al., 2007). Binding of a death ligand to its corresponding receptor triggers the formation of the death-inducing signaling complex (DISC), which consists of oligomerized receptor subunits, the adaptor protein FADD and pro-caspase-8 (Green and Llambi, 2015). As a result, pro-caspase-8 is cleaved to its active form that activates effector caspases, such as caspase-3, leading to cellular demise (Green and Llambi, 2015). Inhibition of the extrinsic pathway can occur through the DISC recruitment of cellular FLICE-like inhibitory proteins (c-FLIP; encoded by the gene Cflar). Three c-FLIP isoforms have been described in human, the long isoform c-FLIPL and the short forms c-FLIPS and c-FLIPR, whereas only FLIPL and FLIPR are found in mice (Golks et al., 2005; Irmler et al., 1997; Ueffing et al., 2008a). c-FLIP proteins lack a functional caspase domain and prevent caspase-8 activation at the DISC (Hughes et al., 2016; Krueger et al., 2001; Scaffidi et al., 1999). The effector and suppressive components of the immune system need to be tightly controlled in order to effectively fight against pathogens, tumor and autoreactive cells. Key players in the suppression of aberrant immune responses are regulatory T (Treg) cells that are crucial for peripheral tolerance (Sakaguchi et al., 2008). The transcription factor Foxp3 is essential for the development and suppressive capacity of Treg cells (Fontenot et al., 2003; Hori et al., 2003). Thus, mutation of Foxp3 results in the complete absence of Treg cells and development of immunodysregulation polyendocrinopathy enteropathy X-linked (IPEX) syndrome in humans or the scurfy phenotype in mice, which is characterized by a fatal multi-organ autoimmune syndrome (Bennett et al., 2001; Brunkow et al., 2001; Wildin et al., 2001). Given the importance of Treg cells for prevention of autoimmune disease but also the therapeutic potential of targeting Treg cells in cancer patients, understanding the processes that maintain a proper Treg cell pool is of major importance. Recent studies identified IL-2-mediated induction of the anti-apoptotic Bcl-2 protein Mcl-1 as crucial for Treg cell survival (Pierson et al., 2013). Furthermore, Foxp3-dependent phosphorylation of Bim, an antagonist of Mcl-1, was determined as a proapoptotic

12 Cell Reports 18, 12–22, January 3, 2017 ª 2017 The Author(s). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

A

B

C

D

E

G

F

H

I

Figure 1. Treg Cells Display Higher Apoptosis Susceptibility Than Tcon Cells (A and B) Representative histograms (A) and statistical summary (B) of TMRE staining assessing ex vivo mitochondrial membrane potential of Tcon and Treg cells from thymus, spleen, peripheral (pLN), and mesenteric lymph nodes (mLN) of GFP-Foxp3 reporter mice (n = 20 for thymus and spleen; n = 19 for pLN and mLN).

(legend continued on next page)

Cell Reports 18, 12–22, January 3, 2017 13

mechanism, whereas the Treg-specific deletion of Bim resulted in elevated Treg cell numbers in the thymus (Chougnet et al., 2011; Tai et al., 2013). While human Treg cells are sensitive toward CD95-mediated apoptosis (Fritzsching et al., 2005), conflicting results have been reported for murine Treg cells (Banz et al., 2002; Weiss et al., 2011). Therefore, the function of the death receptor pathway in Treg cell homeostasis is not well understood. Here, we evaluated the role of the extrinsic apoptosis pathway in Treg cell homeostasis. We show that, compared to conventional T (Tcon) cells, Treg cells are highly sensitive to apoptosis, correlating with lower expression levels of c-FLIPL. In mice conditionally lacking c-FLIP in Treg cells, we detected fatal autoimmunity characterized by massive lymphadenopathy, splenomegaly, immune cell infiltration in multiple organs, as well as increased production of autoantibodies and proinflammatory cytokines. Our results, therefore, reveal a critical role for c-FLIP in Treg cell homeostasis and prevention of autoimmunity. RESULTS Cell Death Is More Pronounced in Treg Than in Tcon Cells To compare the cell death rate of Treg and Tcon cells, we first measured the apoptotic loss of the mitochondrial membrane potential using tetramethylrhodamine ethyl ester (TMRE). Treg cells from lymphoid organs analyzed directly ex vivo exhibited a greater frequency of TMRElow cells compared to Tcon cells, suggesting higher apoptosis susceptibility (Figures 1A and 1B). In addition, flow cytometric analysis with annexin V and 7-aminoactinomycin D (7AAD) revealed higher frequencies of apoptotic Treg cells than of Tcon cells (Figures 1C and 1D). We further analyzed caspase-3/7 activity with a fluorogenic substrate and confocal microscopy, which also indicated a higher apoptosis rate in Treg than in Tcon cells (Figures 1E and 1F). Furthermore, the increased apoptosis sensitivity was largely restricted to CD62L CD44+-activated Treg cells compared to CD62L+ CD44 naive Treg cells (Figures 1G and 1H). Pan-Caspase Inhibitor Improves the Viability of Treg Cells To further characterize Treg apoptosis, we analyzed phosphatidylserine exposure in the absence or presence of the pan-caspase inhibitor Q-VD-OPh (QVD), the homeostatic regulator IL2, or, as a positive apoptosis control, dexamethasone at different

time points. While, as expected, IL-2 impaired the spontaneous cell death of ex vivo analyzed Treg cells to some extent, QVD was even more efficient, demonstrating that Treg cell death is mediated by apoptosis (Figures S1A and S1B). Similarly, QVD promoted cell survival also in the presence of anti-CD3 (Figures S1C and S1D). While both QVD and IL-2 protected against spontaneous cell death of non-activated Tcon cells (Figures S2A and S2B), QVD, but not IL-2, improved the viability of anti-CD3-stimulated Tcon cells (Figures S2C and S2D). In all cases, dexamethasone induced massive cell death. In line, we detected active caspase-3 in short-term cultured Treg, but not in Tcon cells, which was abolished by treatment with PMA, ionomycin, and IL-2 (Figure 1I). Hence, multiple lines of evidence reveal a high apoptosis rate in Treg cells. Treg Cells Express Lower Amounts of c-FLIPL To explore the reason for the increased cell death susceptibility of Treg cells, we first determined the expression of death receptors. Strikingly, expression of CD95 was much higher than TNFR1 and the TRAIL receptor DR5 in both cell types. However, Tcon and Treg cells expressed similar levels of CD95 (Figure 2A), ruling out varying receptor levels as a cause for the differential apoptosis sensitivity. Because c-FLIP is known to regulate T cell apoptosis, we next analyzed the mRNA expression of cFLIP isoforms. While the levels of c-FLIPR were comparable, the expression of c-FLIPL was much lower in Treg cells compared to Tcon cells (Figure 2B), which might explain the higher apoptosis. The lower expression of c-FLIPL in Treg cells was confirmed on the protein level (Figure 2C). Furthermore, TCR triggering and stimulation with TGF-b increased c-FLIP expression in Treg cells in vitro, whereas addition of IL-2 dampened TCR-mediated upregulation of c-FLIP (Figure 2D). Treg Cell-Specific Deletion of c-FLIP Results in a scurfylike Phenotype To study the impact of c-FLIP on Treg survival, we deleted Cflar, the gene encoding the inhibitor of death receptor-mediated cell death c-FLIP, specifically in Treg cells of mice by using the Cre/loxP system. Strikingly, CflarDFoxp3 mice manifested growth retardation compared to control mice (Figures 3A and S3A) and exhibited runting, crusting, and increased thickness of the skin of the tail, ears, and eyelids (Figure S3B). Examining the main lymphoid organs, we found that thymus size was considerably reduced, whereas the lymph nodes were markedly larger in CflarDFoxp3 mice (Figure S3C). The spleens were comparable

(C and D) Representative dot plots (C) and statistical summary (D) of annexin V/7AAD ex vivo staining of Tcon and Treg cells from thymus, spleen, peripheral and mesenteric lymph nodes of GFP-Foxp3 reporter mice (n = 12 for thymus, spleen and pLN; n = 11 for mLN). Annexin V/7AAD double-negative cells are living, whereas double-positive cells are late apoptotic and necrotic cells. Annexin V-single positive cells are early apoptotic cells. (E) Representative images of anti-CD4-PacificBlue, anti-human CD2-APC, and active-caspase-3/7 ex vivo staining. A minimum of 100 Tcon and Treg cells per experiment randomly captured with confocal microscopic images, was counted. (F) Statistical summary of caspase-3/7-positive Treg and Tcon cells from lymph nodes of Foxp3hCD2 reporter mice (n = 5). Treg cells were distinguished from Tcon cells by hCD2 staining. Scale bars, 19 mm. (G and H) Representative dot plots (G) and statistical summary (H) of ex vivo caspase-3/7-positive resting (CD62L+/CD44 ) and activated (CD62L+/ CD44+) Treg cells from spleen and peripheral lymph nodes (pLN) of Foxp3hCD2 reporter mice (n = 6). (I) Immunoblot analysis of active caspase-3 in Tcon and Treg cells after 2 hr of culture in the presence or absence of IL-2, PMA, and ionomycin (Iono). Cells were purified from GFP-Foxp3 reporter mice using FACS sorting. One representative of two independent experiments is shown. Bar graphs (B, D, F, and H) represent the mean ± SEM. Statistical analyses were performed by two-tailed Mann-Whitney tests; *p < 0.05, **p < 0.01. See also Figures S1 and S2.

14 Cell Reports 18, 12–22, January 3, 2017

B

A

C

D

Figure 2. Lower Expression of c-FLIPL in Treg Cells (A) Surface expression of the death receptors CD95, TRAIL receptor DR5, and TNFR1 in Tcon and Treg cells. Cells were purified from Foxp3hCD2 reporter mice and assessed for death receptor expression by flow cytometry. Representative histograms of two independent experiments are shown. (B) Transcript expression of c-FLIPL (left panel) and c-FLIPR (right panel) in Tcon and Treg cells purified from Foxp3hCD2 reporter mice by FACS sorting and analyzed by qPCR (n = 17 each). Bar graphs represent the mean ± SEM. Statistical analyses were performed by two-tailed Mann-Whitney tests; *p < 0.05; n.s., not significant. (C) Immunoblot analysis of cFLIPL in MACS-purified Tcon and Treg cells from Foxp3hCD2 reporter mice. Numbers show normalized quantification of c-FLIPL relative to GAPDH band intensity. Quantification was performed using ImageJ 1.44p software. (D) Transcript expression of c-FLIPL (left panel) and c-FLIPR (right panel) as analyzed by qPCR in Treg cells purified from Foxp3hCD2 reporter mice by FACS sorting and cultured in the absence or presence of anti-CD3/anti-CD28, IL-2, TGF-b, and indicated combinations (n = 3 each). Bar graphs in (B) and (D) represent the mean ± SEM. Statistical analyses were performed by two-tailed Mann-Whitney tests; *p < 0.05; n.s., not significant.

between wild-type and CflarDFoxp3 mice, but considering the reduced body size, CflarDFoxp3 mice apparently developed splenomegaly (Figure 3B). All CflarDFoxp3 mice died within 25 days (Figure 3C). Thus, this phenotype is reminiscent of the scurfy mouse strain, which has a mutation in Foxp3 (Brunkow et al., 2001; Wildin et al., 2001). Next, we analyzed the activation status of T cells via staining of CD44 and CD62L. This analysis revealed a clear shift from a naive (CD62L+, CD44 ) to a memory-activated-like phenotype (CD62L , CD44+) for CD4+ T cells (Figures 3D and 3E) and CD8+ T cells in CflarDFoxp3 mice (Figures S3D and S3E). Additionally, CD4+ and CD8+ T cells showed higher frequencies in the thymus and spleen, whereas the frequencies in peripheral lymph nodes were comparable to controls (Figures S3F and S3G). In the same line, the absolute CD4+ and CD8+ T cell numbers of CflarDFoxp3 mice were higher in spleen and comparable in peripheral lymph nodes. Conversely, absolute T cell numbers were markedly lower in the thymus of CflarDFoxp3 mice, consistent with the strongly reduced organ size (Figure S3H). Moreover,

the serum levels of proinflammatory cytokines, such as IL-4, IL-5, IL-6, interferon g (IFN-g), and tumor necrosis factor alpha (TNF-a), were highly elevated in these mice compared to the control group (Figure 3F). We also tested the presence of autoantibodies and used sera of CflarDFoxp3 and control mice for immunoblotting on liver and kidney extracts from RAG-2 / mice. Sera from CflarDFoxp3 mice, but not control mice, reacted strongly with the tissue extracts, indicating the production of autoantibodies in the CflarDFoxp3 mice (Figure S3I). Finally, histological analysis revealed massive leukocyte infiltration and chronic inflammation of liver, lung, pancreas, and skin of CflarDFoxp3 animals (Figure 3G). Thus, loss of c-FLIP in Treg cells leads to hyperactivation of the immune system and a fatal scurfy-like autoimmune phenotype. Deficiency of c-FLIP Drastically Affects the Viability of Treg Cells in Secondary Lymphoid Organs Given that autoimmune disease of scurfy mice is caused by lack of Treg cells (Wildin et al., 2001), we then examined the Treg

Cell Reports 18, 12–22, January 3, 2017 15

A

B

D

C

E

F

G

Figure 3. Deletion of c-FLIP in Treg Cells in Mice Causes Abnormalities in Lymphoid Organs and Premature Death (A) Body weight of CflarDFoxp3 mice and littermate control mice (n = 8 each). Each symbol indicates an individual mouse. (B) Spleen weight relative to body weight of CflarDFoxp3 and littermate control mice (n = 4 each). Each symbol indicates an individual mouse. (C) Survival curve of CflarDFoxp3 mice and littermate control mice (n = 16 each).

(legend continued on next page)

16 Cell Reports 18, 12–22, January 3, 2017

compartment in CflarDFoxp3 mice. Indeed, Foxp3+ cells were barely detectable in spleen and lymph nodes in CflarDFoxp3 mice, while they were readily detectable in the control littermates (Figure 4A). Interestingly, we found relatively normal frequencies of CD4+ Foxp3+ cells in the thymus of these animals, indicating that Treg development was not impaired (Figures 4A and 4B). A substantial drop in peripheral Treg cell frequencies became apparent from day 8 after birth onward, while frequencies stayed constant in the thymus (Figure S4A). Coinciding with reduced Treg cell frequencies, the activation status of CD4+ and CD8+ Tcon cells was elevated from day 8 post birth onward (Figure S4B). Similar to the Treg frequency, the absolute numbers of Treg cells were highly decreased in spleen and lymph nodes. In contrast to the unaltered frequencies, the absolute number of Treg cells in the thymus was much lower, most likely due to the reduced organ size (Figure 4C). Consistent with normal T cell development, we observed unaltered frequencies of CD4+ CD25+ GITR+ Foxp3 Treg precursors and mature Treg cells in the thymus of 6- to 8-day-old CflarDFoxp3 mice (Figures S4C and S4D). In addition, overall T cell development appeared normal at this age, i.e., in the absence of inflammation, as analyzed by expression of CD4, CD8, CD25, CD44, and Foxp3 in different thymocyte subsets (Figures S4E and S4F). In line, immunohistochemistry revealed a strong reduction of Treg cells in the periphery but not in the thymus of CflarDFoxp3 mice (Figure 4D). c-FLIP Regulates CD95-Independent Survival Functions in Treg Cells In Vivo To analyze whether CD95L-induced apoptosis is also a major cause of Treg cell loss in CflarDFoxp3 mice in vivo, we injected CD95L-neutralizing antibodies into CflarDFoxp3 mice starting at day 6, a time point when Treg cell generation had started (Asano et al., 1996) but disease symptoms were not yet developed (see Figure S4A). Unexpectedly and despite several attempts, we were unable to increase Treg cell frequencies in CflarDFoxp3 mice to control levels using two different antiCD95L antibodies (Figures 5A, 5B, and S5A). Similarly, in vivo injection of the caspase inhibitor QVD did not result in Treg cell frequencies comparable to control mice (Figure S5B). To exclude that the failure to rescue the scurfy phenotype of the neonatal CflarDFoxp3 mice was due to the high inflammation, we also injected CD95L antibodies into adult Foxp3hCD2 reporter mice. In this setting, Treg cell frequencies remained unchanged (Figures 5C and 5D). Therefore, our data reveal an essential role for c-FLIP in Treg homeostasis and prevention of autoimmunity but also suggest that c-FLIP has additional survival functions beyond blocking CD95-mediated apoptosis.

DISCUSSION Although Treg cells were first regarded as a rather anergic cell population (Itoh et al., 1999), further studies revealed a high proliferation rate in Treg compared to Tcon cells (Fisson et al., 2003; Klein et al., 2003; Schuster et al., 2012). Despite previous reports that analyzed Treg apoptosis in vitro (Chougnet et al., 2011; Wang et al., 2012) or caspase activation ex vivo (Pierson et al., 2013), the importance of cell death for Treg homeostasis remains poorly understood. By directly measuring several apoptosis parameters, we show that Treg cells exhibit a high apoptosis sensitivity compared to Tcon cells and express reduced levels of c-FLIP suggesting a pivotal role of this molecule in Treg survival. Previous work on Treg cell survival has largely concentrated on the intrinsic pathway. In this regard, the anti-apoptotic Bcl-2 protein Mcl-1, but neither Bcl-2 nor Bcl-xL, was found to be crucial for Treg cell survival (Pierson et al., 2013). In turn, Treg-specific deletion of Bcl2l11 (encoding Bim) or Bax and Bak1 resulted in increased frequencies of Treg cells in peripheral lymphoid organs (Chougnet et al., 2011; Pierson et al., 2013; Wang et al., 2012). We have observed a similar increase in Treg frequencies in aged transgenic mice expressing the short isoform c-FLIPR in hematopoietic cells (Ewald et al., 2014), suggesting that c-FLIP, by interfering with the death receptor pathway, is involved in Treg homeostasis. However, Treg cell accumulation in this system might have been also indirectly caused by Treg cell-extrinsic effects of the Cflar transgene and the autoimmunity in these mice. To unambiguously define a role of c-FLIP for Treg survival, we generated CflarDFoxp3 mice that specifically lack c-FLIP in Treg cells. Treg-specific deletion of Cflar resulted in severe autoimmune disease of a scurfy-like phenotype and premature death of the mice. Interestingly, all CflarDFoxp3 mice died very early (i.e., within 25 days), similar to scurfy mice, whereas a considerable proportion of mice lacking Mcl-1 in Treg cells survived more than 50 days (Pierson et al., 2013). Therefore, it is possible that Mcl-1 and c-FLIP have distinct roles in Treg cell survival or that c-FLIP exerts additional functions beyond that of apoptosis inhibition. c-FLIP is known to inhibit CD95-mediated apoptosis in T cells (Schmitz et al., 2003; Telieps et al., 2013; Ueffing et al., 2008b), but not cell death induced by perforin/granzyme B, g-irradiation, or chemotherapeutic drugs (Kataoka et al., 1998). Interestingly, however, caspase inhibitors and CD95L antagonists were unable to rescue the scurfy phenotype and to increase Treg cell frequencies in vivo, neither under inflammatory conditions in CflarDFoxp3 mice nor in healthy mice. Therefore, under in vivo conditions c-FLIP has presumably additional survival effects. In this context, c-FLIP was shown

(D and E) Representative pseudocolor dot plots (D) and frequencies (E) of T cell activation markers CD44 and CD62L in CD4+ cells from spleen and lymph nodes of CflarDFoxp3 and littermate control mice. CD62L+/CD44 cells are naive T cells, CD62L /CD44+ cells are activated and memory T cells, and CD62L+/CD44+ cells are central memory T cells (n = 7 each). (E) Bar graphs represent the mean ± SEM. Statistical analyses were performed by two-tailed Mann-Whitney tests; *p < 0.05, **p < 0.01, ***p < 0.001; n.s., not significant. (F) Expression of IL-4, IL-5, IL-6, IFN-g, IL-12p70, and TNF-a in sera from CflarDFoxp3 and littermate control mice determined by Luminex assay (n = 11 each). Bar graphs represent the mean ± SEM. Statistical analyses were performed by two-tailed Mann-Whitney tests; *p < 0.05, **p < 0.01, ***p < 0.001; n.s., not significant. (G) Representative H&E-stained liver, lung, pancreas, and skin sections from CflarDFoxp3 and littermate control mice. Scale bar, 50 mm. See also Figure S3.

Cell Reports 18, 12–22, January 3, 2017 17

A

B

C

D

Figure 4. Treg Cell Population Is Strongly Reduced in Spleen and Lymph Nodes, but Not in the Thymus of CflarDFoxp3 Mice (A–C) Representative flow cytometric dot plots (A), percentages (B), and absolute numbers (C) of Foxp3+ Treg cells within CD4+ CD8 cells in thymus, spleen, and peripheral lymph nodes of CflarDFoxp3 and littermate control mice are shown (n = 7 each; except thymus CflarDFoxp3, n = 4). Bar graphs represent the mean ± SEM. Statistical analyses were performed by two-tailed Mann-Whitney tests; **p < 0.01; n.s., not significant. (D) Immunohistochemistry of Foxp3 in thymus, spleen and lymph nodes of CflarDFoxp3 and littermate controls. Scale bars, 50 mm. See also Figure S4.

18 Cell Reports 18, 12–22, January 3, 2017

A

C

B

D

Figure 5. CD95L-Blocking Antibodies Do Not Increase Treg Cell Frequencies (A and B) Representative flow cytometric dot plots (A) and frequencies (B) of Foxp3+ Treg cells within CD4+ CD8 cells in thymus, spleen, and peripheral lymph nodes of neonatal CflarDFoxp3 and littermate control mice treated with 100 mg of anti-CD95L per injection. Mice were injected at days 6, 7, and 8 after birth and analyzed at day 9. Data is representative of two independent experiments. Bar graphs represent the mean ± SEM. Statistical analyses were performed by twotailed Mann-Whitney tests; n.s., not significant. (C and D) Representative flow cytometric dot plots (C) and frequencies (D) of Foxp3+ Treg cells within CD4+ CD8 cells in spleen and peripheral lymph nodes of adult Foxp3hCD2 reporter mice treated with 200 mg per injection of anti-CD95L (clone 3C82) or isotype control antibody (mouse IgG1). Mice were injected at three times in non-consecutive days and analyzed 1 day after the last injection. Data is representative of two independent experiments with a total of four mice analyzed per genotype. Bar graphs represent the mean ± SEM. Statistical analyses were performed by two-tailed Mann-Whitney tests; n.s., not significant. See also Figure S5.

to inhibit not only death receptor-mediated apoptosis but also necroptosis and autophagy in T cells (He and He, 2015). Thus, Treg cells might die by other cell death mechanisms in vivo under certain conditions. Of note, alternative cell death modes such as necroptosis or ferroptosis have been reported to execute physiological T cell death in other contexts (Matsushita et al., 2015; Osborn et al., 2010). Additionally, c-FLIP can activate survival pathways such as ERK and nuclear factor kB (NF-kB) (Golks et al., 2006; Kataoka et al., 2000; Koenig et al., 2014; Misra et al., 2007). Thus, apoptosis-independent functions of c-FLIP in Treg cells might be also relevant for the severe phenotype of CflarDFoxp3 mice.

Taken together, the results demonstrate that c-FLIP is essential for Treg survival and homeostasis to prevent autoimmunity. Given the capacity of Treg cells to control autoimmunity and to repress anti-cancer immunity, c-FLIP may represent a therapeutic target to modulate Treg cell abundance and immune responses in cancer or autoimmune disease. EXPERIMENTAL PROCEDURES Mice Foxp3tm2Ayr, Foxp3tm1(CD2/CD52)Shori, B6.129(Cg)-Foxp3tm4(YFP/icre)Ayr/J, and B6.129-Cflartm1Ywh/J mice were described (Fontenot et al., 2005; Komatsu

Cell Reports 18, 12–22, January 3, 2017 19

et al., 2009; Rubtsov et al., 2008; Zhang and He, 2005). CflarDFoxp3 mice were generated by crossing B6.129-Cflartm1Ywh/J with B6.129(Cg)Foxp3tm4(YFP/icre)Ayr/J mice. Mice were kept under pathogen-free conditions in the animal facility of the Helmholtz Center for Infection Research. Animal experiments were performed in accordance with the guidelines of the local and national authorities.

Immunohistochemistry was done after heat-mediated antigen retrieval using anti-FoxP3 (clone FJK-16 s, eBioscience).

Flow Cytometric Analyses For surface staining, cells were resuspended in 100 mL FACS buffer (2% BSA in PBS) and incubated with the respective antibodies for 15 min at 4 C in the dark. Cells were then washed twice with 500 mL FACS buffer and analyzed in a LSRII flow cytometer (BD Biosciences). In the case of cell population analyses (excluding viability determinations), dead cells were excluded by LIVE/ DEAD Fixable Blue Dead Cell Stain staining (Life Technologies) for 30 min at 4 C in the dark. After two washes in PBS, Fc receptors were blocked by 15 min incubation with Fc-block (CD16/32) in FACS buffer at 4 C. Intracellular proteins were stained using Foxp3 Staining Buffer Set (Miltenyi Biotec). For viability analysis, after the surface staining, the cells were incubated with annexin V and 7AAD or stained with TMRE in the dark, both for 20 min. Alternatively, cell viability was determined using CellEvent Caspase 3/7 Green (Life Technologies).

qPCR RNA was purified using RNAeasy Plus kit (QIAGEN). cDNA was synthesized from 100 ng of RNA template using cDNA kit (Thermo Scientific). qPCR was implemented in a Roche LightCycler using Roche FastStart SYBRgreen Master. Ubiquitin-conjugating enzyme E2D 2A (UCE) was used as reference gene for relative quantification.

Western Blot Analysis Cells were lysed by incubation in TPNE buffer (13 PBS, 300 mM NaCl, 2 mM EDTA, 1% Triton X-100) supplemented with protease inhibitors for 20 min on ice. After centrifugation protein concentration was determined by BCA assay (Thermo Scientific). Protein lysates were loaded onto 12% SDS-polyacrylamide gels, blotted onto PVDF membrane (GE Healthcare) and detected by chemiluminescence. For autoantibody detection, liver and kidney extracts (20 mg) from RAG2-deficient mice were separated on SDS-polyacrylamide gels. After protein transfer, PVDF membranes were probed with sera from CflarDFoxp3 and control mice diluted to a protein concentration of 5 mg/mL.

Cytokine Determination Serum cytokine levels were determined using Th1/Th2 Mouse 6-Plex Panel (Life Technologies) in a Luminex instrument (Luminex Corporation).

Statistical Analysis Statistical analyses were performed by Mann-Whitney tests to determine statistical significance using GraphPad Prism (GraphPad Software). SEM is represented as error bars in the graphs. SUPPLEMENTAL INFORMATION Supplemental Information includes Supplemental Experimental Procedures and five figures and can be found with this article online at http://dx.doi.org/ 10.1016/j.celrep.2016.12.022. AUTHOR CONTRIBUTIONS C.P.-S., M.S., and I.S. conceived experiments. C.P.-S., M.S., Y.N., U.H., and M.C.P. performed experiments and analyzed the data. K.S.-O. provided essential reagents. C.P.-S., K.S.-O., and I.S. wrote the paper. I.S. supervised the study. ACKNOWLEDGMENTS

Cell Culture T cells were seeded in RPMI-1640 supplemented with 10% FCS (PAA Laboratories), 50 mM b-mercaptoethanol, 50 mg/mL each of penicillin and streptomycin, 1% non-essential amino acids, and 1 mM sodium pyruvate (all from Life Technologies). For T cell stimulation, PMA (10 ng/mL, Sigma Aldrich), ionomycin (1 mM, Sigma Aldrich), anti-CD3 (145-2C11, BioLegend), murine IL-2 (R&D Systems), and QVD (MP Biomedicals) were used. For cell viability studies, Tcon and Treg cells from Foxp3hCD2 mice were separated using CD4+ CD25+ regulatory T cell isolation kit for MACS magnetic separation (Miltenyi Biotec) according to manufacturer’s protocol, except that anti-hCD2-PE (RPA-2.10, Biolegend) was used to separate Tcon and Treg cells. To induce apoptosis, cells were stimulated with 20 ng/mL of CD95L (self-made), TRAIL (Novitec), TNF-a (R&D Systems), or 0.5 mM dexamethasone (Sigma-Aldrich) for 16 hr. FasFc fusion protein (50 mg/mL, self-made) was used to block CD95L-induced cell death. Thymocytes were cultured with 5 ng/mL CD95L, 20 ng/mL TRAIL, 20 ng/mL TNF-a, and 0.5 mM dexamethasone, respectively. To determine c-FLIP isoform levels, Treg cells were seeded in the presence or absence of 10 mg/mL anti-CD3, 5 mg/mL anti-CD28 (37.51, Biolegend), 50 ng/mL IL-2, and 50 ng/mL TGF-b (R&D Systems). Confocal Microscopy Lymph node cell suspensions from hCD2-Foxp3 mice were stained with antiCD4-PacificBlue, anti-hCD2-APC, and CellEvent Caspase 3/7 Green and directly visualized under a Nikon Ti Eclipse microscope equipped with an UltraViewVox Spinning Disc module (Perkin Elmer). For CD95L sensitivity determination, Tcon and Treg cells from Foxp3hCD2 mice were sorted (BD FACS Aria II) and incubated for 16 hr in the absence or presence of CD95L. After staining with CellEvent Caspase 3/7 Green, cells were visualized by confocal microscopy. Histology Liver, lung, skin, and pancreas from CflarDFoxp3 and control mice were fixed in formaldehyde and embedded in paraffin. Sections of 3 mm were stained with H&E and evaluated in a blinded manner as described (Lahl et al., 2007).

20 Cell Reports 18, 12–22, January 3, 2017

We thank S. Schumann and C. Kozowsky for excellent technical assistance, Drs. J. Huehn, M. Leverkus, and T. Sparwasser for various reagents, Dr. L. Gro¨be for cell sorting, J. Petzold for Luminex measurements, and all members of the HZI animal facility. We are grateful to Drs. Y.-W. He, S. Hori, and A. Rudensky for providing Cflar flox, hCD2-Foxp3, GFP-Foxp3, and Foxp3-Cre mice. C.P.S. was supported by the President’s Initiative and Networking Fund of the Helmholtz Association of German Research Centers under contract number VH-GS-202. I.S. received funding from the Fritz-Thyssen-Stiftung (Az. 10.13.1.200) and Deutsche Forschungsgemeinschaft (SCHM1586/ 6-1). Received: October 16, 2015 Revised: October 21, 2016 Accepted: December 7, 2016 Published: January 3, 2017 REFERENCES Asano, M., Toda, M., Sakaguchi, N., and Sakaguchi, S. (1996). Autoimmune disease as a consequence of developmental abnormality of a T cell subpopulation. J. Exp. Med. 184, 387–396. Banz, A., Pontoux, C., and Papiernik, M. (2002). Modulation of Fas-dependent apoptosis: a dynamic process controlling both the persistence and death of CD4 regulatory T cells and effector T cells. J. Immunol. 169, 750–757. Bennett, C.L., Christie, J., Ramsdell, F., Brunkow, M.E., Ferguson, P.J., Whitesell, L., Kelly, T.E., Saulsbury, F.T., Chance, P.F., and Ochs, H.D. (2001). The immune dysregulation, polyendocrinopathy, enteropathy, X-linked syndrome (IPEX) is caused by mutations of FOXP3. Nat. Genet. 27, 20–21. Bouillet, P., and O’Reilly, L.A. (2009). CD95, BIM and T cell homeostasis. Nat. Rev. Immunol. 9, 514–519. Brunkow, M.E., Jeffery, E.W., Hjerrild, K.A., Paeper, B., Clark, L.B., Yasayko, S.A., Wilkinson, J.E., Galas, D., Ziegler, S.F., and Ramsdell, F. (2001).

Disruption of a new forkhead/winged-helix protein, scurfin, results in the fatal lymphoproliferative disorder of the scurfy mouse. Nat. Genet. 27, 68–73. Chougnet, C.A., Tripathi, P., Lages, C.S., Raynor, J., Sholl, A., Fink, P., Plas, D.R., and Hildeman, D.A. (2011). A major role for Bim in regulatory T cell homeostasis. J. Immunol. 186, 156–163. Ewald, F., Annemann, M., Pils, M.C., Plaza-Sirvent, C., Neff, F., Erck, C., Reinhold, D., and Schmitz, I. (2014). Constitutive expression of murine c-FLIPR causes autoimmunity in aged mice. Cell Death Dis. 5, e1168. Fisson, S., Darrasse-Je`ze, G., Litvinova, E., Septier, F., Klatzmann, D., Liblau, R., and Salomon, B.L. (2003). Continuous activation of autoreactive CD4+ CD25+ regulatory T cells in the steady state. J. Exp. Med. 198, 737–746. Fontenot, J.D., Gavin, M.A., and Rudensky, A.Y. (2003). Foxp3 programs the development and function of CD4+CD25+ regulatory T cells. Nat. Immunol. 4, 330–336. Fontenot, J.D., Rasmussen, J.P., Williams, L.M., Dooley, J.L., Farr, A.G., and Rudensky, A.Y. (2005). Regulatory T cell lineage specification by the forkhead transcription factor foxp3. Immunity 22, 329–341. Fritzsching, B., Oberle, N., Eberhardt, N., Quick, S., Haas, J., Wildemann, B., Krammer, P.H., and Suri-Payer, E. (2005). In contrast to effector T cells, CD4+CD25+FoxP3+ regulatory T cells are highly susceptible to CD95 ligandbut not to TCR-mediated cell death. J. Immunol. 175, 32–36. Golks, A., Brenner, D., Fritsch, C., Krammer, P.H., and Lavrik, I.N. (2005). c-FLIPR, a new regulator of death receptor-induced apoptosis. J. Biol. Chem. 280, 14507–14513. Golks, A., Brenner, D., Krammer, P.H., and Lavrik, I.N. (2006). The c-FLIP-NH2 terminus (p22-FLIP) induces NF-kappaB activation. J. Exp. Med. 203, 1295– 1305. Green, D.R., and Llambi, F. (2015). Cell death signaling. Cold Spring Harb. Perspect. Biol. 7, a006080. He, M.-X., and He, Y.-W. (2015). c-FLIP protects T lymphocytes from apoptosis in the intrinsic pathway. J. Immunol. 194, 3444–3451. Hori, S., Nomura, T., and Sakaguchi, S. (2003). Control of regulatory T cell development by the transcription factor Foxp3. Science 299, 1057–1061. Hughes, M.A., Powley, I.R., Jukes-Jones, R., Horn, S., Feoktistova, M., Fairall, L., Schwabe, J.W.R., Leverkus, M., Cain, K., and MacFarlane, M. (2016). Cooperative and hierarchical binding of c-FLIP and Caspase-8: a unified model defines how c-FLIP isoforms differentially control cell fate. Mol. Cell 61, 834–849. Irmler, M., Thome, M., Hahne, M., Schneider, P., Hofmann, K., Steiner, V., Bodmer, J.L., Schro¨ter, M., Burns, K., Mattmann, C., et al. (1997). Inhibition of death receptor signals by cellular FLIP. Nature 388, 190–195. Itoh, M., Takahashi, T., Sakaguchi, N., Kuniyasu, Y., Shimizu, J., Otsuka, F., and Sakaguchi, S. (1999). Thymus and autoimmunity: production of CD25+CD4+ naturally anergic and suppressive T cells as a key function of the thymus in maintaining immunologic self-tolerance. J. Immunol. 162, 5317–5326. Kataoka, T., Schro¨ter, M., Hahne, M., Schneider, P., Irmler, M., Thome, M., Froelich, C.J., and Tschopp, J. (1998). FLIP prevents apoptosis induced by death receptors but not by perforin/granzyme B, chemotherapeutic drugs, and gamma irradiation. J. Immunol. 161, 3936–3942. Kataoka, T., Budd, R.C., Holler, N., Thome, M., Martinon, F., Irmler, M., Burns, K., Hahne, M., Kennedy, N., Kovacsovics, M., and Tschopp, J. (2000). The caspase-8 inhibitor FLIP promotes activation of NF-kappaB and Erk signaling pathways. Curr. Biol. 10, 640–648. Klein, L., Khazaie, K., and von Boehmer, H. (2003). In vivo dynamics of antigenspecific regulatory T cells not predicted from behavior in vitro. Proc. Natl. Acad. Sci. USA 100, 8886–8891. Koenig, A., Buskiewicz, I.A., Fortner, K.A., Russell, J.Q., Asaoka, T., He, Y.-W., Hakem, R., Eriksson, J.E., and Budd, R.C. (2014). The c-FLIPL cleavage product p43FLIP promotes activation of extracellular signal-regulated kinase (ERK), nuclear factor kB (NF-kB), and caspase-8 and T cell survival. J. Biol. Chem. 289, 1183–1191.

Komatsu, N., Mariotti-Ferrandiz, M.E., Wang, Y., Malissen, B., Waldmann, H., and Hori, S. (2009). Heterogeneity of natural Foxp3+ T cells: a committed regulatory T-cell lineage and an uncommitted minor population retaining plasticity. Proc. Natl. Acad. Sci. USA 106, 1903–1908. Krammer, P.H., Arnold, R., and Lavrik, I.N. (2007). Life and death in peripheral T cells. Nat. Rev. Immunol. 7, 532–542. Krueger, A., Schmitz, I., Baumann, S., Krammer, P.H., and Kirchhoff, S. (2001). Cellular FLICE-inhibitory protein splice variants inhibit different steps of caspase-8 activation at the CD95 death-inducing signaling complex. J. Biol. Chem. 276, 20633–20640. Lahl, K., Loddenkemper, C., Drouin, C., Freyer, J., Arnason, J., Eberl, G., Hamann, A., Wagner, H., Huehn, J., and Sparwasser, T. (2007). Selective depletion of Foxp3+ regulatory T cells induces a scurfy-like disease. J. Exp. Med. 204, 57–63. Matsushita, M., Freigang, S., Schneider, C., Conrad, M., Bornkamm, G.W., and Kopf, M. (2015). T cell lipid peroxidation induces ferroptosis and prevents immunity to infection. J. Exp. Med. 212, 555–568. Misra, R.S., Russell, J.Q., Koenig, A., Hinshaw-Makepeace, J.A., Wen, R., Wang, D., Huo, H., Littman, D.R., Ferch, U., Ruland, J., et al. (2007). Caspase-8 and c-FLIPL associate in lipid rafts with NF-kappaB adaptors during T cell activation. J. Biol. Chem. 282, 19365–19374. Osborn, S.L., Diehl, G., Han, S.-J., Xue, L., Kurd, N., Hsieh, K., Cado, D., Robey, E.A., and Winoto, A. (2010). Fas-associated death domain (FADD) is a negative regulator of T-cell receptor-mediated necroptosis. Proc. Natl. Acad. Sci. USA 107, 13034–13039. Pierson, W., Cauwe, B., Policheni, A., Schlenner, S.M., Franckaert, D., Berges, J., Humblet-Baron, S., Scho¨nefeldt, S., Herold, M.J., Hildeman, D., et al. (2013). Antiapoptotic Mcl-1 is critical for the survival and niche-filling capacity of Foxp3+ regulatory T cells. Nat. Immunol. 14, 959–965. Rubtsov, Y.P., Rasmussen, J.P., Chi, E.Y., Fontenot, J., Castelli, L., Ye, X., Treuting, P., Siewe, L., Roers, A., Henderson, W.R., Jr., et al. (2008). Regulatory T cell-derived interleukin-10 limits inflammation at environmental interfaces. Immunity 28, 546–558. Sakaguchi, S., Yamaguchi, T., Nomura, T., and Ono, M. (2008). Regulatory T cells and immune tolerance. Cell 133, 775–787. Scaffidi, C., Schmitz, I., Krammer, P.H., and Peter, M.E. (1999). The role of c-FLIP in modulation of CD95-induced apoptosis. J. Biol. Chem. 274, 1541– 1548. Schmitz, I., Krueger, A., Baumann, S., Schulze-Bergkamen, H., Krammer, P.H., and Kirchhoff, S. (2003). An IL-2-dependent switch between CD95 signaling pathways sensitizes primary human T cells toward CD95-mediated activation-induced cell death. J. Immunol. 171, 2930–2936. Schuster, M., Glauben, R., Plaza-Sirvent, C., Schreiber, L., Annemann, M., €hl, A.A., Clayton, L.K., Sparwasser, T., Schulze-Osthoff, K., Floess, S., Ku et al. (2012). IkB(NS) protein mediates regulatory T cell development via induction of the Foxp3 transcription factor. Immunity 37, 998–1008. Tai, X., Erman, B., Alag, A., Mu, J., Kimura, M., Katz, G., Guinter, T., McCaughtry, T., Etzensperger, R., Feigenbaum, L., et al. (2013). Foxp3 transcription factor is proapoptotic and lethal to developing regulatory T cells unless counterbalanced by cytokine survival signals. Immunity 38, 1116–1128. Telieps, T., Ewald, F., Gereke, M., Annemann, M., Rauter, Y., Schuster, M., Ueffing, N., von Smolinski, D., Gruber, A.D., Bruder, D., and Schmitz, I. (2013). Cellular-FLIP, Raji isoform (c-FLIP R) modulates cell death induction upon T-cell activation and infection. Eur. J. Immunol. 43, 1499–1510. €hne, R., Schulze-Osthoff, K., and Schmitz, I. Ueffing, N., Keil, E., Freund, C., Ku (2008a). Mutational analyses of c-FLIPR, the only murine short FLIP isoform, reveal requirements for DISC recruitment. Cell Death Differ. 15, 773–782. Ueffing, N., Schuster, M., Keil, E., Schulze-Osthoff, K., and Schmitz, I. (2008b). Up-regulation of c-FLIP short by NFAT contributes to apoptosis resistance of short-term activated T cells. Blood 112, 690–698. Wang, X., Szymczak-Workman, A.L., Gravano, D.M., Workman, C.J., Green, D.R., and Vignali, D.A. (2012). Preferential control of induced regulatory T cell homeostasis via a Bim/Bcl-2 axis. Cell Death Dis. 3, e270.

Cell Reports 18, 12–22, January 3, 2017 21

Weiss, E.-M., Schmidt, A., Vobis, D., Garbi, N., Lahl, K., Mayer, C.T., Sparwasser, T., Ludwig, A., Suri-Payer, E., Oberle, N., and Krammer, P.H. (2011). Foxp3-mediated suppression of CD95L expression confers resistance to activation-induced cell death in regulatory T cells. J. Immunol. 187, 1684– 1691.

neonatal diabetes mellitus, enteropathy and endocrinopathy syndrome is the human equivalent of mouse scurfy. Nat. Genet. 27, 18–20.

Wildin, R.S., Ramsdell, F., Peake, J., Faravelli, F., Casanova, J.L., Buist, N., Levy-Lahad, E., Mazzella, M., Goulet, O., Perroni, L., et al. (2001). X-linked

Zhang, N., and He, Y.-W. (2005). An essential role for c-FLIP in the efficient development of mature T lymphocytes. J. Exp. Med. 202, 395–404.

22 Cell Reports 18, 12–22, January 3, 2017

Youle, R.J., and Strasser, A. (2008). The BCL-2 protein family: opposing activities that mediate cell death. Nat. Rev. Mol. Cell Biol. 9, 47–59.