Poly(ADP-ribosyl)ation of FOXP3 Protein Mediated by PARP-1 Protein Regulates the Function of Regulatory T Cells*

Background: PARP-1 has importance in the immune system. Results: Poly(ADP-ribosyl)ation of FOXP3 mediated by PARP-1 destabilizes FOXP3 and negatively regulates the suppressive activity of Treg cells. Conclusion: PARP-1 negatively regulates Treg function via FOXP3 poly(ADP-ribosyl)ation. Significance: This study helps the development of PARP-1 inhibitors to prevent autoimmune diseases. Poly(ADP-ribose) polymerase 1 (PARP-1) is an ADP-ribosylating enzyme participating in diverse cellular functions. The roles of PARP-1 in the immune system, however, have not been well understood. Here we find that PARP-1 interacts with FOXP3 and induces its poly(ADP-ribosyl)ation. By using PARP-1 inhibitors, we show that reduced poly(ADP-ribosyl)ation of FOXP3 results in not only FOXP3 stabilization and increased FOXP3 downstream genes but also enhanced suppressive function of regulatory T cells. Our results suggest that PARP-1 negatively regulates the suppressive function of Treg cells at the posttranslational level via FOXP3 poly(ADP-ribosyl)ation. This finding has implications for developing PARP-1 inhibitors as potential agents for the prevention and treatment of autoimmune diseases.

and DNA repair through the poly(ADP-ribosyl)ation of histones and the recruitment of single-strand DNA repair enzymes (5)(6)(7). In addition, PARP-1 has also been reported to regulate gene transcription in tumor generation, metabolism, and immune responses (8 -10). Because of its diverse and important roles, PARP-1 inhibitors have been used in the treatment of cancer and inflammation (11,12).
Regulatory T (Treg) cells are a subpopulation of CD4 ϩ T cells and induce and maintain peripheral tolerance and modulate immune responses. Dysregulation of the development and function of Treg cells has been shown to be associated with many immune-related diseases, including autoimmune diseases, chronic inflammation, acute and chronic infection, and transplant rejection (13). Forkhead box P3 (FOXP3), known as a key transcription factor of Treg cells, is required for their development, maintenance, and function (14,15). FOXP3 abnormality leads to the severe autoimmune disease IPEX (immune dysregulation, polyendocrinopathy, enteropathy, X-linked syndrome) in humans and the fatal lymphoproliferative disorder in mice (16,17). Studies have shown that the transcriptional activity, stability, and localization of FOXP3 are regulated at posttranslational levels, including acetylation, ubiquitination, and phosphorylation (18 -20). Recently, it has been shown that mice deficient in PARP-1 display increased numbers of regulatory T cells, indicating that PARP-1 may affect Treg cell differentiation (21). However, how PARP-1 may regulate the suppressive function of Treg cells has not been clear (22).
Here we provide evidence demonstrating that PARP-1 binds to FOXP3 and induces poly(ADP-ribosyl)ation of FOXP3. More importantly, we find that, after treatment with PARP-1specific inhibitors, decreased poly(ADP-ribosyl)ation of FOXP3 in Treg cells leads to an increased level of FOXP3 by preventing proteasome-mediated degradation, resulting in an increased suppressive function of Treg cells. Our results suggest that PARP-1 negatively regulates the suppressive function of Treg cells at the posttranslational level through FOXP3 poly(ADP-ribosyl)ation.
Cells and Transfection-HEK293T cells were cultured in DMEM supplemented with 10% FCS and 1% penicillin/streptomycin. Human Jurkat T cells were grown in RPMI 1640 medium supplemented with 10% FCS, 1% penicillin/streptomycin, 1% sodium pyruvate, and 1% non-essential amino acids. All cells were maintained at 37°C in a 5% CO 2 incubator. HEK293T cells were transfected with the indicated plasmids using polyethylenimine (Polyscience, catalog no. 23966-2) reagent according to the instructions of the manufacturer.
Immunoprecipitation and Immunoblot Analysis-Cells were lysed in radioimmune precipitation assay buffer containing 50 mM Tris/HCl (pH 7.4), 0.5% Nonidet P-40, 0.25% sodium deoxycholate, 150 mM NaCl, 1 mM EDTA with 1 mM PMSF, 1 mM Na 3 VO 4 , 1 mM NaF, and protease inhibitor (Sigma). The supernatants were immunoprecipitated with 1 g of the indicated antibodies and 10 l of protein A/G Plus-agarose (Santa Cruz Biotechnology), followed by separation in SDS/PAGE and analysis with Western blotting.
Cytoplasm, Nucleus, and Chromatin Extraction-1 ϫ 10 6 cells were harvested in 1 ml of ice-cold 1ϫ PBS buffer and then suspended in 300 l of cytoplasm buffer (10 mM Tris/HCl (pH 7.5), 10 mM KCl, 0.1 mM EDTA, 1 mM DTT, and 0.5 mM PMSF). Nonidet P-40 was added to a final concentration of 0.6% after 15 min incubation on ice. Another 15 min later, the lysate was centrifuged, and the supernatant was cytoplasm fraction. The pellet was resuspended in 200 l of nucleus extract buffer (20 mM Tris/HCl (pH 8.0), 400 mM NaCl, 1 mM EDTA, 1 mM DTT, 1 mM PMSF, 1 mM Na 3 VO 4 , and 1 mM NaF) and then incubated on ice for another 30 min. After centrifugation, the supernatant was nucleus fraction, and the pellet was chromatin fraction.
In Vitro Suppression Assay-Before incubation, Treg cells were pretreated with PARP-1 inhibitors for 6 or 12 h. After this treatment, human PBMC cells were labeled with 5 M CFSE (Invitrogen) for 10 min at 37°C, followed by incubation with anti-CD3/CD28 beads and pretreated Treg cells at a different ratio in a U-bottom 96-well plate. After 4 days, cells were harvested to stain with viability dye (fixable viability dye, eFluor 780, eBioscience) and anti-CD8-APC antibody (eBioscience) and then analyzed on a Fortessa cytometer (BD Biosciences).
Statistics-Two-paired Student's t tests were used for the calculation of p values.

PARP-1 Interacts with and Poly(ADP-ribosyl)ates FOXP3-
The role of PARP-1 in regulating Treg cell function has been controversial (21,22). A previous study has found that PARP-1 knockout mice displayed an increased frequency of Treg cells but normal suppression function, indicating that PARP-1 might affect Treg cell differentiation but not function. However, another study has shown that PARP-1 knockout Treg cells exhibited stronger suppressive activity than the WT. Therefore, whether and how PARP-1 regulates the suppressive function of Treg cells is still not clear.
Studies have shown that PARP-1, as a ADP-ribose transferase, poly(ADP-ribosyl)ates not only histones but also several transcription factors, including Smad2/3 and NFAT (10,23). Therefore, we first set out to examine whether FOXP3 could be a substrate of PARP-1 in Treg cells. In a co-immunoprecipitation assay in which Myc-tagged PARP-1 and FLAG-tagged FOXP3 were co-transfected into HEK293T cells, we found that PARP-1 interacted with FOXP3 in reciprocal immunoprecipitation (Fig. 1A). Such an interaction of PARP-1 and FOXP3 was also confirmed in Jurkat cells stably expressing HA-tagged FOXP3 and in human primary iTreg cells (Fig. 1, B and C).
To further determine which domains in PARP-1 and FOXP3 were responsible for the interaction between these two proteins, we generated a panel of PARP-1 and FOXP3 truncation expression constructs. Human PARP-1 has been shown to contain a DNA-binding domain, a BRCA1 C terminus, a tryptophan-glycine-arginine domain (WGR), and a catalytic domain (CA). The FOXP3 protein is composed of several functional domains, including proline-rich, zinc finger, Leu zipper, coiledcoil (CC), and forkhead domains (FK). In the co-transfection experiments in HEK293T cells followed by immunoprecipitation, we found that the zinc finger and Leu zipper domains of FOXP3 and the DNA-binding domain and BRCA1 C terminus of PARP-1 were crucial for the interaction between PARP-1 and FOXP3 (Fig. 2, A and B).
To examine whether PARP-1 promoted the poly(ADP-ribosyl)ation of FOXP3, we used anti-PAR-specific antibody to detect the poly(ADP-ribosyl)ation level of FOXP3 in immunoprecipitates. We found that this modification of FOXP3 was increased with anti-PAR-specific antibody when FOXP3 and PARP-1 were co-transfected into HEK293T cells (Fig. 3A). Furthermore, in an in vitro poly(ADP-ribosyl)ation assay, we also examined the poly(ADP-ribosyl)ation of FOXP3 mediated by recombinant His-tagged PARP-1 protein (Fig. 3B).
Studies have shown that poly(ADP-ribose) glycohydrase (PARG) rapidly degrades the poly(ADP) ribose chain (24). Therefore, we wanted to test whether PARG would inhibit the poly(ADP-ribosyl)ation of FOXP3 by PARP-1. HEK293T cells were simultaneously transfected with HA-tagged PARP-1, FLAG-tagged FOXP3, and Myc-tagged PARG. In the subsequent immunoprecipitation assays, we found that PARG did not affect the association between PARP-1 and FOXP3 but, rather, the opposite. The presence of PARG seemed to promote this interaction (Fig. 3C). Nevertheless, by using anti-PAR antibody, we found that the presence of PARG resulted in decreased poly(ADP-ribosyl)ation of FOXP3 (Fig. 3D). All of these results suggest that PARP-1 poly(ADP-ribosyl)ates FOXP3 and that FOXP3 is a substrate of PARP-1.
Poly(ADP-ribosyl)ation, as a crucial posttranslational modification, has been reported to regulate protein stabilization, cel- lular localization, and transcriptional activity (4). To further investigate how PARP-1 inhibitors regulate FOXP3 function by inhibiting its poly(ADP-ribosyl)ation, we first set out to exam-ine FOXP3 protein and mRNA levels in Treg cells treated with 3-AB or AG14361. We found that, 12 h after human primary Treg cells had been treated with 3-AB or AG14361, the protein  level of FOXP3 was up-regulated, but the mRNA level did not change (Fig. 4, B and C), indicating that PARP-1 inhibitors stabilized FOXP3 protein level in Treg cells through inhibiting poly(ADP-ribosyl)ation of FOXP3.
Also, we detected the cellular localization of FOXP3 in PARP-1 inhibitor-treated Treg cells and found increased FOXP3 after PARP-1 inhibitor treatment. However, FOXP3 localization still did not change (Fig. 4D), suggesting that the suppression of PARP-1 enzymatic activity stabilized FOXP3 but not localization in human Treg cells.
PARP-1 Promotes the Poly-ubiquitination and Degradation of FOXP3-The E3 ubiquitin ligase Stub1 has already been identified to negatively regulate the suppressive function of Treg cells by promoting the poly-ubiquitination and degradation of FOXP3 (20). Given the fact that the inhibition of PARP-1 enzymatic activity not only inhibited the poly(ADP-ribosyl)ation of FOXP3 but also stabilized FOXP3 in Treg cells, we decided to further examine whether PARP-1 promoted the poly-ubiquitination and degradation of FOXP3. First, Histagged Ub, HA-tagged FOXP3, Myc-tagged PARP-1, and FLAG-tagged Stub1 were co-transfected into HEK293T cells. As expected, the FOXP3 level was reduced with addition of Stub1, and, importantly, even less FOXP3 was found in the Stub1, PARP-1, and FOXP3 co-expressed group, suggesting that PARP-1 may promote FOXP3 degradation mediated by Stub1 (Fig. 5A). To further examine whether PARP-1 affected FOXP3 poly-ubiquitination to promote FOXP3 degradation, a His pulldown assay was used to detect the endogenous polyubiquitination level of FOXP3 after 4 h of MG132 treatment. The addition of the proteasome inhibitor MG132 prevented FOXP3 loss, suggesting that this process was proteasome-dependent. Moreover, the poly-ubiquitination level of FOXP3 was up-regulated significantly by addition of PARP-1, indicating that PARP-1 promoted FOXP3 poly-ubiquitination and degradation (Fig. 5B).
To determine whether the enzymatic activity of PARP-1 affected FOXP3 poly-ubiquitination, we detected the endogenous poly-ubiquitination level of FOXP3 in PARP-1 inhibitortreated Treg cells. After 12 h of treatment, we found decreased FOXP3 in Ub immunoprecipitates after PARP-1 inhibitor treatment (Fig. 5C), suggesting that the inhibition of PARP-1 enzymatic activity suppressed the poly-ubiquitination and degradation of FOXP3.
PARP-1 Inhibitors Enhanced the Suppressive Function of Treg Cells through PARylated FOXP3-To further investigate whether PARP-1 inhibitors regulated Treg cell function by inhibiting FOXP3 poly(ADP-ribosyl)ation, we examined some of the FOXP3-associated genes in Treg cells treated with 3-AB. We found that, 6 h after human primary Treg cells had been treated with 3-AB, the expression levels of CD25, CTLA4, and Il10 in the cells were up-regulated (Fig. 6A), suggesting that PARP-1 inhibitor affected FOXP3 downstream genes. Given the fact that the inhibition of PARP-1 enzymatic activity not  only inhibited the poly(ADP-ribosyl)ation of FOXP3 but also regulated FOXP3 stabilization and its downstream genes, we decided to further examine the suppressive function of Treg cells in the standard Treg suppression cell co-culture system in vitro. For some of the cell culture groups, Treg cells were pre-treated with the PARP-1 inhibitor 3-AB for 6 h before the suppression assay. Human PBMCs from healthy donors were labeled with CFSE before being cultured with human primary Treg cells at different ratios and stimulated with anti-CD3/ CD28 beads for 4 days. The results showed that 3-AB-pre-  t test); NT, no treatment. Data represent at least three independent experiments. B, in vitro suppression assay. Human primary Treg cells were treated with or without 10 mM PARP inhibitor 3-AB for 6 h prior to the suppression assay. The Treg cells were then cultured with CFSE-labeled human PBMCs and anti-CD3/CD28 beads at the indicated ratio for 4 days. Proliferation of CD8 ϩ T effector cells was examined by flow cytometry. C, human primary Treg cells were treated with or without 5 M PARP-1 inhibitor AG14361 for 12 h. Then Treg cells were cultured with CFSE-labeled human PBMCs and anti-CD3/CD28 beads at the indicated ratio for 4 days in the in vitro suppression assay. The proliferation of CD8 ϩ T effector cells was examined by flow cytometry. Data represent at least two independent experiments. treated Treg cells exhibited a much stronger suppressive effect than untreated Treg cells on the proliferation of PBMCs (Fig.  6B), suggesting that the poly(ADP-ribosyl)ation of FOXP3 mediated by PARP-1 negatively regulates the suppressive function of Treg cells. We also used another PARP-1-specific inhibitor, AG14361, and found that AG14361 pretreated Treg cells had an increased suppressive function in vitro as well (Fig. 6C). All of these data strongly suggest that PARP-1 inhibitors promote the suppressive function of Treg cells through the inhibition of FOXP3 poly(ADP-ribosyl)ation.

Discussion
In this study, we demonstrated that PARP-1 interacted with human FOXP3 and promoted its poly(ADP-ribosyl)ation. We identified the zinc finger/Leu zipper domains of FOXP3 and the DNA-binding domain/BRCA1 C terminus of PARP-1 as crucial domains for the direct interaction between these two proteins. We also showed that the poly(ADP-ribosyl)ation of FOXP3 mediated by PARP-1 was inhibited by PARG or PARP-1 inhibitors. Most importantly, we found that reduced poly(ADP-ribosyl)ation of FOXP3 by PARP-1 inhibitor stabilized FOXP3, upregulated the expression levels of FOXP3 downstream genes in Treg cells, and enhanced the suppressive function of Treg cells. Our results suggest that the poly(ADP-ribosyl)ation of FOXP3 mediated by PARP-1 negatively regulates the suppressive function of Treg cells.
Studies have reported that PARylation does not act alone but in concert with other posttranslational modifications (4). We found PARP-1 promoted FOXP3 poly-ubiquitination and degradation mediated by Stub1. In addition, the inhibition of PARP-1 enzymatic activity suppressed the poly-ubiquitination and degradation of FOXP3, suggesting that poly(ADP-ribosyl)ation of FOXP3 mediated by PARP-1 affected FOXP3 poly-ubiquitination and degradation. However, the relationship and mechanism between these two important posttranslational modifications still needs to be investigated further.
PARP-1 is an abundant nuclear protein with low enzyme activity that can be activated by DNA break, reactive oxygen species, and inflammation. The activation of PARP-1 results in the poly(ADP-ribosyl)ation of its target proteins, which leads to the modulation of target protein transcriptional activity, localization, and the creation of new protein interaction scaffolds (25). Our data show that PARP-1 regulates the suppressive function of Treg cells through FOXP3 poly(ADP-ribosyl)ation. However, the regulation of PARP-1 activity and the signal induces FOXP3 poly(ADP-ribosyl)ation in Treg cells are still unclear. Besides, PARP-1 takes part in several cellular processes, including cell death, transcriptional regulation, and inflammation. Additional studies will be needed to figure out how these diverse functions of PARP-1 are integrated and controlled in different cells.
Because of its diverse and important roles, a number of PARP-1 inhibitors are under clinical development for the treatment of cancer, such as iniparib (BSI-201), Olaparib (AZD-2281, oral) and veliparib (ABT-888, oral) (11,12). Our study not only has implications in the developing PARP-1 inhibitors as potential agents for the treatment of autoimmune diseases but also shows the potential risk for cancer treatment.
In summary, here we uncovered the previously unrecognized molecular mechanism that PARP-1 regulated the suppressive function of Treg cells at a posttranslational modification level through poly(ADP-ribosyl)ation of FOXP3. Furthermore, more specific PARP-1 inhibitors will be required both as tools and therapeutics of autoimmune diseases on the basis of the role of PARP-1 in the immune system.