Caerulomycin A Enhances Transforming Growth Factor-β (TGF-β)-Smad3 Protein Signaling by Suppressing Interferon-γ (IFN-γ)-Signal Transducer and Activator of Transcription 1 (STAT1) Protein Signaling to Expand Regulatory T Cells (Tregs)*

Background: Caerulomycin A, a known antifungal agent, was explored for its novel immunomodulatory activity. Results: Caerulomycin A supports the generation of Tregs by augmenting the TGF-β-Smad3 and suppressing the IFN-γ-STAT1 signaling pathways by enhancing SOCS1 expression. Conclusion: Caerulomycin A induces and enhances the Treg population. Significance: Caerulomycin A can be a potent future drug for treating autoimmune diseases by eliciting the generation of Tregs. Cytokines play a very important role in the regulation of immune homeostasis. Regulatory T cells (Tregs) responsible for the generation of peripheral tolerance are under the tight regulation of the cytokine milieu. In this study, we report a novel role of a bipyridyl compound, Caerulomycin A (CaeA), in inducing the generation of Tregs. It was observed that CaeA substantially up-regulated the pool of Tregs, as evidenced by an increased frequency of CD4+ Foxp3+ cells. In addition, CaeA significantly suppressed the number of Th1 and Th17 cells, as supported by a decreased percentage of CD4+/IFN-γ+ and CD4+/IL-17+ cells, respectively. Furthermore, we established the mechanism and observed that CaeA interfered with IFN-γ-induced STAT1 signaling by augmenting SOCS1 expression. An increase in the TGF-β-mediated Smad3 activity was also noted. Furthermore, CaeA rescued Tregs from IFN-γ-induced inhibition. These results were corroborated by blocking Smad3 activity, which abolished the CaeA-facilitated generation of Tregs. In essence, our results indicate a novel role of CaeA in inducing the generation of Tregs. This finding suggests that CaeA has enough potential to be considered as a potent future drug for the treatment of autoimmunity.

The immune system of the body is its artillery against pathogens. The immune response is very precisely tuned and tamed to discriminate self from non-self. Although the body does generate autoreactive T cells, the immune system has an accurate mechanism to eliminate such cells during thymic selection (1). Paradoxically, some autoreactive T cells can still escape dele-tion, establish themselves in the body, and become a potential threat to the host by generating autoimmune diseases. The body has developed additional defense mechanisms against these treacherous cells through the induction of peripheral tolerance, anergy/unresponsiveness, release of immunosuppressive molecules (TGF-␤, IL-10, IL-35, etc.), and generation of regulatory T cells (Tregs) 3 (2)(3)(4). Despite extremely careful mechanisms operating in the body to safeguard against autoreactive T cells, these cells may still become activated and can be terrifically devastating in inflicting injury by causing autoimmune diseases (5).
Autoimmune diseases can be treated by non-steroidal antiinflammatory drugs, corticosteroids, disease-modifying antirheumatic drugs, and immunosuppressive agents (6). However, the limitations and side effects associated with the use of these drugs pave the way toward better and safer remedial measures (7)(8)(9). Recently, the generation of antigen-specific tolerance by Tregs has gained considerable impetus following the observation that these cells are endowed with enough potential to suppress autoimmune reactions (10,11).
It is a well known fact that most of the revolutionary medicines, like Rapa, cyclosporin A, and FK506, exhibiting antibiotic or antifungal properties were later proven to possess potent immunosuppressive activity (21)(22)(23). CaeA is a bipyridyl compound and has been reported in the literature to show antifungal and antibiotic properties (24,25). Hence, it encouraged us to study whether CaeA could also exhibit immunosuppression, in particular by evoking the generation of Tregs, as observed in the case of Rapa (18).
In this study, we have shown that CaeA independently induced Tregs (cTregs) and synergistically supported the TGF-␤-mediated expansion of Tregs (c␤Tregs). We also revealed the mechanism of action involved in the generation of Tregs by CaeA. Finally, the proof of concept was established by successfully treating animals suffering from arthritis with CaeA.

EXPERIMENTAL PROCEDURES
Mice-Female C3He and BALB/c mice and male DBA/1 mice 6 -8 weeks of age were procured from the experimental animal facility of the Institute of Microbial Technology. The study was approved by the Institutional Animal Ethical Committee.
The modulation in the frequency of Tregs, Th1, and Th17 cells was analyzed by flow cytometry.
Treg Functional Assay-CD4 T cell activation was performed as described elsewhere (26). Briefly, CD4 T cells (2 ϫ 10 5 cells/ 200 l) were stimulated with anti-CD3 and CD28 Abs. Tregs were added to the cultures and incubated for 48 h. Later, cells were harvested and studied for activation by flow cytometry.
An alloresponse experiment was performed using splenocytes (2 ϫ 10 5 ) of MHC-mismatched strains of mice. The C3He (H-2 k ) cells served as the responder and ␥-irradiated DBA/1 (H-2 q ) cells as the stimulator. Alloresponse cultures were set in the presence or absence of Tregs for 4 days. Later, cultures were pulsed with 1 Ci of [methyl-3 H] thymidine for 16 h. The cells were harvested onto glass filter mats, and incorporated radioactivity was measured as counts per minutes by scintillation counting.
Generation of the Experimental Arthritis Model and Disease Assessment-The induction of arthritis and disease symptoms were monitored as described previously (27). Briefly, male DBA/1 mice (6 mice/group) were immunized intradermally on day 0 with bovine collagen type II (100 g) emulsified in complete Freund adjuvant (4 mg/ml desiccated Mtb H37Ra). A booster dose of bovine collagen type II in incomplete Freund adjuvant was injected on day 21. Later, CaeA (1 and 10 mg/kg body weight) and 0.5% carboxyl methyl cellulose emulsion in water was administered daily for 50 days, and animals were monitored every day for arthritis symptoms. A group treated with carboxyl methyl cellulose (0.5%) was taken as a vehicle control (placebo). Disease progression was assessed for a clinical score using following criteria: 0, no symptoms; 1, mild erythema and inflammation in the limb digits; 2, severe inflammation up to the paw; and 3, severe inflammation up to the ankle region. These score norms were adopted for each limb.
For immunological assays, animals were sacrificed on day 30. The optimum response was reported during that period (28). The cells isolated from the lymph nodes were pooled and in vitro-challenged with bovine collagen type II (50 g) for 72 h. Later, proinflammatory cytokines (IFN-␥ and TNF-␣) in the culture supernatants were measured by ELISA according to the instructions of the manufacturer. The proinflammatory markers matrix metalloproteinase-3 in the serum and IL-6, TNF-␣, and IFN-␥ in the knee joints were estimated on day 50. For histopathological analysis, the knee joints were fixed in buffered formalin (10%) and decalcified in formic acid (1%). Microtome sections of paraffin-embedded knee joints were stained with H&E to observe the extent of inflammation.
Fluorescence Imaging-The fluorescence imaging of arthritis-induced mice was performed with the help of the in vivo imager FMT 2500 Lx (PerkinElmer Life Sciences, Waltham, MA). ProSense 680 and 750 (cathepsin-specific activatable probes) and OsteoSense 680 and 800 (specific for bone degeneration) were used for visualizing inflammatory responses. These reagents were injected intravenously 24 h prior to imaging. Later on, hairs were removed by hair clipper and depilatory cream. The animals were imaged under a given laser wave length for excitation (680, 750, and 780 nm) and emission fluorescence (700, 780, and 805 nm). All procedures were per-formed under gas anesthesia (isoflurane). The intensity of fluorescence was directly proportional to the severity of the disease. Image processing and analysis was performed by TrueQuant software.
Isolation of Naïve CD4 ϩ T Cells-CD4 ϩ T cells were isolated by magnetic activated cell sorting according to the instructions of the manufacturer (CD4 T cell enrichment kit, BD Biosciences). Briefly, splenocytes obtained from two mice were pooled and RBC-depleted by ACK (ammonium-chloride-potassium) lysis buffer. Later, the cells were incubated with CD4 T cell enrichment mixture (50 l/10 7 cells) and biotin anti-CD25 Ab (5 l/10 7 cells) for 30 min at 4°C. The unbound Abs were removed by washing with RPMI 1640 (400 g, 5 min). The residual pellet was incubated with BD TM IMag streptavidin particles plus-DM (50 l/10 7 cells) for 30 min at 4°C. The cells were suspended in 4 ml of RPMI and placed in contact with iMagnet for 8 min. The CD4 ϩ T cells obtained by negative selection were of 92% purity, as confirmed by flow cytometry.
ELISA-The cytokines IFN-␥, TNF-␣, and IL-17 secreted in the culture supernatants were determined by sandwich ELISA. Briefly, ELISA plates were coated with the appropriate concentration of specific Ab in phosphate buffer (pH 9.2) for 12 h at 4°C. Later, the plates were blocked with BSA (1%) to eliminate nonspecific binding. ELISA plates were incubated with culture supernatants and appropriate standards for 12 h at 4°C. The plates were treated with respective biotin-conjugated secondary Abs, followed by streptavidin-HRP. The plates were developed with the help of the substrate H 2 O 2 and the chromogenic agent O-phenylenediamine. The optical density of color development was measured at 495 nm. Each step was followed by washing five times with 1ϫ PBS-Tween 20 (0.05%) and regular incubation steps. The level of cytokines was estimated by plotting a standard curve using recombinant cytokines and expressed as picograms/milliliter. Later, cells were pulsed with TGF-␤ (2 ng/ml) for 1 h. As a positive control for STAT1 suppression, cells were also cultured with the JAK inhibitor pyridone 6. Immediately thereafter, the nuclear extract was prepared, and EMSA was performed, as described elsewhere (29). Briefly, the nuclei of cells were isolated by suspending cells in hypotonic conditions in buffer A (20 mM HEPES (pH 7.4), 0.42 M NaCl, 1.5 mM MgCl 2 , 0.2 mM EDTA, and 25% glycerol) containing the following proteinase inhibitors: 0.5 mM dithiothreitol, 1 mM PMSF, 2 g/ml leupeptin, and 10 g/ml aprotinin. Nuclear proteins were extracted in buffer B (20 mM HEPES (pH 7.4), 50 mM KCl, 0.2 mM EDTA, 20% glycerol, and proteinase inhibitors) and used for analysis of DNA binding. The double-stranded oligonucleotide of the consensus sequences 5Ј-CATGTTATGCATATT-CCTGTAAGTG-3Ј (STAT1) and 5Ј-TCGAGAGCCAGA-CAAAAAGCCAGACATTTAGCCAGACAC-3Ј (Smad3) was used for EMSA. The double-stranded oligo was end-labeled with [␥-32 P]ATP with the help of T4 polynucleotide kinase. The binding reaction was carried out for 30 min at room temperature using nuclear extract (5 g) along with a labeled probe in a final volume of 20 l of buffer (10 mM Tris (pH 7.5), 0.1 mM EDTA, 5 mM MgCl 2 , 80 mM KCl, 0.8 mM dithiothreitol, 2.5% glycerol, and 1 g of poly(dI-dC)). The protein with the DNA complex was separated in polyacrylamide gel (4%) using Tris borate-EDTA buffer. After electrophoresis, the gel was dried and exposed to the imager screen at room temperature for 12 h and scanned using a PhosphorImager (Fujifilm, Tokyo, Japan).

EMSA-The
Immunoprecipitation-CD4 ϩ T cells pulsed with IFN-␥ for 15 min were lysed in radioimmune precipitation assay buffer. The whole cell lysate was precleared using normal rabbit serum and protein G-agarose beads for 1 h at 4°C. Later, protein G-agarose beads were separated from the lysate by centrifugation (5000 ϫ g, 5 min). The cell lysate was then incubated with either anti-JAK1 or JAK2 Abs overnight at 4°C. Protein G-agarose beads were added and incubated for 1 h at 4°C. The immune complexes were isolated by centrifugation (5000 ϫ g, 5 min). Purified complexes were boiled in SDS sample buffer and used to analyze the phosphorylation of JAK1 and JAK2 proteins by Western blotting.
Western Blotting-The expression of Smad7 and T-bet and the phosphorylation of STAT1, STAT3, STAT4, Smad3, JAK1, and JAK2 were analyzed by Western blotting. Briefly, the whole cell lysate was prepared by suspending cells in radioimmune precipitation assay buffer. The protein concentration was estimated by Bradford assay. The equal concentration of protein lysate of all samples was separated on SDS-PAGE gels (12%) followed by transfer to a nitrocellulose membrane at 4°C for 12 h. Membrane transfer was followed by blocking with BSA (5%) in PBS and then probing with their specific Abs, followed by HRP-conjugated secondary Abs. The blots were visualized by ECL Plus Western blotting substrate. The blots were developed onto x-ray film (Amersham Biosciences Hyperfilm ECL, GE Healthcare). Internal loading was analyzed by stripping off the primary Ab and reprobing with a suitable Ab.
Mitochondrial Membrane Potential Analysis-CD4 T cells stimulated with anti-CD3 (2 g/ml) and CD28 (1 g/ml) Abs were cultured with CaeA (0 -0.31 M) for 72 h. Later, cells were pelleted and resuspended in JC-1 (5,5Ј,6,6Ј-tetrachloro-1,1Ј,3,3Ј-tetraethylbenzimidazolylcarbocyanine iodide) medium (2.5 g/ml). The cultures were incubated at room temperature for 15-20 min in the dark. Later, cells were washed with flow cytometry staining buffer (2% FBS in PBS) and acquired immediately using a flow cytometer (FACSCalibur), and then the analysis was done using FACSDiva software. The increase in the mitochondrial membrane potential was measured by the change in the fluorescence intensity of JC-1 from a green to a red wavelength.
Flow Cytometry-The cells were harvested from the cultures and washed with flow cytometry buffer (PBS and 2% FBS). The Fc receptor was blocked by using anti-CD16 Ab, followed by staining with fluorochrome-conjugated anti-CD4 and CD69 Abs and their respective isotype-matched control Abs for 30 min at 4°C. For staining with biotinylated Abs, cells were incubated with biotin-conjugated Abs for 30 min at 4°C, followed by incubation with fluorochrome-conjugated streptavidin. Finally, cells were washed and fixed in paraformaldehyde (1%). Intracellular cytokine staining was performed by first staining for surface markers followed by fixing with paraformaldehyde (4%) for 20 min at 4°C. Later, the cells were washed twice in staining buffer and resuspended in permeabilizing buffer (1ϫ PBS/0.1% saponin/1% FBS) at room temperature for 20 min. This was followed by incubating with appropriate Abs or their isotype-matched controls for 30 min. Cells were washed twice with permeabilizing buffer and resuspended in staining buffer. Nuclear protein Foxp3 staining was performed with a Foxp3 buffer staining kit according to the instructions of the manufacturer (eBioscience, San Diego, CA). Cells were acquired using FACSCalibur or FACSAria II. Analysis was performed by FACSDiva software.
Statistical Analysis-Statistical analysis was performed with the help of GraphPad Prism software. The differences between groups were compared by statistical analysis (two-tailed, unpaired Student's t test).

Induction of the Enhancement of Foxp3 ϩ CD4 T Cells by
CaeA-The expression of Foxp3 is crucial for the generation of Tregs. Rapa is a known immunosuppressive drug that has been reported to induce Treg formation (18). Hence, in the preliminary phase of the study, we tested whether CaeA can elicit the generation of Tregs. Interestingly, we observed that anti-CD3 and CD28 Ab-stimulated naïve CD4 ϩ T cells cultured with CaeA showed a significant (p Ͻ 0.05) increase in Foxp3 ϩ T cells (cTregs) (Fig. 1A). Further, supplementing cultures with TGF-␤ significantly (p Ͻ 0.005) promoted the pool of Foxp3 ϩ CD4 T cells (c␤Tregs) (Fig. 1, A and B). The enhancement in the frequency of Foxp3 ϩ CD4 T cells was observed in a dose-dependent fashion. The data were reproduced by both flow cytometry and RT-qPCR (Fig. 1 C). The results were comparable with Rapa, which was used as a positive control (Fig. 1,  A-C). We also noticed an up-regulation of CD25 (IL-2 receptor) expression on CD4 T cells on CaeA treatment, which was augmented further on complementing the cultures with TGF-␤ (data not shown). Intriguingly, naïve CD4 T cells cultured with CaeA under Th1 and Th17 polarizing conditions exhibited a significant decline in the expression (p Ͻ 0.005) and secretion (p Ͻ 0.0001) of IFN-␥ and IL-17 by CD4 T cells, respectively (Fig. 1, D-F). Furthermore, down-regulation of CD44 and an increased mitochondrial membrane potential was noticed in CD4 T cells on treatment with CaeA (Fig. 1, G and H). T cell activation (CD44 hi ) is inversely proportional to the mitochondrial membrane potential (30). We observed a similar effect with CaeA.
Tregs Generated in the Presence of CaeA Significantly Suppressed the Activity of Effector T Cells-We next checked whether cTregs and c␤Tregs were capable of inhibiting the function of effector T cells. It is noteworthy that a diminished expression of CD69 was observed on effector CD4 T cells cocultured with either cTregs or c␤Tregs (Fig. 2, A and B). Further, cTregs and c␤Tregs substantially (p Ͻ 0.0001) retarded the alloresponse in a mixed lymphocyte reaction (Fig. 2 C). However, the extent of suppression was better in the case of c␤Tregs than cTregs. These data clearly signify that cTregs and c␤Tregs are functionally active and have the potential to retard the activity of effector T cells.
CaeA Sufficiently Expands Foxp3 ϩ Tregs under Th1 and Th17 Polarization Conditions-Loss of Tregs is one of the main reasons for the breakdown of peripheral tolerance. The instability of Tregs under proinflammatory conditions is a major hurdle for the restoration of tolerance by reprogrammed Tregs (31). Hence, proinflammatory conditions reciprocally regulate Treg generation by favoring pathogenic Th17 cell development (32). Hence it was crucial for us to monitor whether the presence of Th1 and Th17 polarizing cytokines can influence the differentiation of CaeA-induced Tregs. It is of interest to note that the proinflammatory milieu could not change the generation of Tregs (Fig. 3, A-D). In contrast, a significant reduction was observed in the frequency of IFN-␥and IL-17-producing T cells (Fig. 3, E and F). The response was observed to be in a dose-dependent manner (Fig. 3, A-F). The data denote that the presence of the Th1 and Th17 differentiation cytokines IFN-␥ and IL-17, respectively, does not hinder the CaeA-induced formation of Tregs.
CaeA Ameliorates Clinical Symptoms in an Experimental Model of Arthritis-Th1 and Th17 cells play a potential role in the progression of autoimmune diseases (33). In contrast, Tregs are known to suppress autoimmunity (34). Because cTregs suppress both Th1 and Th17 cells, it was obvious to check whether the administration of CaeA can restrain autoimmunity. In this study, we used a collagen-induced animal model of arthritis. Fascinatingly, there was a regression in the clinical symptoms of arthritis upon treatment with CaeA (Fig. 4A). The animals were monitored for a clinical score over a period of 50 days (Fig. 4B). Furthermore, this observation was supported by the histopathological examination, which revealed decreased infiltration and less synovitis of the joint capsule of animals treated with CaeA (Fig. 4, C and D). Furthermore, we substantiated our finding by examining decreased inflammation by employing a whole body fluorescence imager and probing with biomarkers specific for cathepsins (ProSense) and bone remodelling (Oste-oSense). The inflammatory response was decreased considerably (p Ͻ 0.005) with an increased dose of CaeA (Fig. 4, E and F). Further, compared with control mice, the level of the proinflammatory cytokines IL-6, TNF-␣, and IFN-␥ was lower in the joint capsule of animals treated with CaeA (Fig. 4G).
During the early onset of disease (30 days), we also enumerated the frequency of Tregs in the draining lymph nodes. The results showed that animals treated with CaeA exhibited, in vivo, an increased percentage of Tregs (Fig. 4H). In contrast, a decline in the yield of IFN-␥ and TNF-␣ was noticed when challenging cells with collagen in vitro (Fig. 4I). We also observed a reduction in the serum levels of matrix metalloproteinase-3 in mice that received CaeA (Fig. 4J). Secretion of the proinflammatory marker matrix metalloproteinase-3 has been reported to be inhibited by Tregs (35). In essence, these results (Fig. 4, A-J) indicate categorically that CaeA restrains arthritis symptoms by inducing the generation of Tregs and suppressing proinflammatory factors.
CaeA Specifically Interferes with STAT1 Signaling-The cytokine milieu plays an important role in the lineage commitment of CD4 T cells. Furthermore, cytokines exert signaling through a group of conserved proteins known as STATs. Hence, we next examined whether CaeA influences cytokinemediated signaling of STATs. Interestingly, CD4 T cells preincubated with different concentrations of CaeA significantly (p Ͻ 0.005) inhibited the phosphorylation of STAT1 signaling mediated by IFN-␥ and IL-6 (Fig. 5, A and B). To further corroborate the effect of CaeA on STAT1, we quantified the binding of STAT1 to its DNA with an enriched nuclear protein lysate of IFN-␥-treated CD4 T cells. We observed decreased STAT1 binding to DNA, substantiating the observation (Fig.  5C) that CaeA specifically antagonizes STAT1 signaling. The inhibitor of JAK pyradone 6, used as a positive control, also showed inhibition of STAT1 binding to DNA. No change was observed in the phosphorylation of JAK1/2 by CaeA (Fig. 5D). The suppressor of cytokine signaling 1 (SOCS1) expression lev-els were augmented with CaeA (Fig. 5E). SOCS1 has been reported to antagonize interferon signaling by preventing the phosphorylation of STAT1 by JAK1/2 (36,37). We also studied the expression of IFN-dependent genes like ISG15, IRF-1, IRF-3, and IRF-7. It was observed that the levels of ISG15, IRF-1, and IRF-7 were suppressed on CaeA treatment (Fig. 5,  F-H). No change was observed in the case of IRF-3 (data not shown). Furthermore, CaeA treatment failed to alter IL-6-and IL-12-mediated STAT3 and STAT4 signaling, respectively (Fig.  6, A and B), thus confirming the specificity of CaeA and its interference in STAT1 signaling.
CaeA Interferes with IFN-␥ Stimulation and Suppresses the Expression of FasL, Smad7, and T-bet-STAT1 is an important transcription factor for IFN-␥-mediated signaling. Because CaeA inhibits STAT1 phosphorylation, we next examined the status of IFN-␥-induced FasL, Smad7, and T-bet expression. The expression of FasL is under the control of IFN-␥-mediated STAT1 signaling (38,39). CaeA significantly (p Ͻ 0.05) suppressed FasL expression (Fig. 7, A and B). RT-qPCR data depict a dose-dependent decrease in the levels of Smad7 and T-bet with CaeA treatment (Fig. 7, C and D). We further supported these results with Western blotting experiments (Fig. 7, E and  F). Overall, the results of these experiments indicate that CaeA interferes with the IFN-␥-mediated signaling pathway and genes under the control of STAT1 (Fig. 7, A-F).
CaeA Enhances TGF-␤-mediated Smad3 Signaling-TGF-␤ differentiates Tregs through Smad3 signaling (40). CaeA synergizes with TGF-␤ in the lineage commitment of Tregs (Fig. 1,  A-C). Therefore, we studied the role of CaeA in bolstering TGF-␤-mediated Smad3 signaling. CaeA significantly enhanced TGF-␤ induced Smad3 phosphorylation of activated CD4 T cells (Fig. 8A). IFN-␥ inhibits TGF-␤-elicited phosphorylation of Smad3 (41). However, increased TGF-␤-induced Smad3 phosphorylation was noticed with CaeA treatment (Fig. 8B). This signifies categorically that CaeA neutralizes the IFN-␥ effect. These results were authenticated further by nuclear translocation of Smad3 (Fig. 8C). SIS3 is a Smad3 inhibitor, therefore we used it as a positive control to prove the specificity of Smad3 binding to DNA (42). IFN-␥ also impedes the generation of Tregs (43). Consequently, the role of CaeA in rescuing Tregs from IFN-␥-influenced suppression was studied. Remarkably, CaeA treatment significantly (p Ͻ 0.005) rescued the Tregs from the inhibitory effect of IFN-␥, as evidenced by an increased percentage of Foxp3 ϩ CD4 T cells (Fig. 8, D and E). We also explicitly ascertained the authenticity of CaeA-induced formation of Tregs by SIS3. The results establish categorically that the Tregs generated by CaeA are entirely dependent on Smad3 activity, as verified by a noteworthy (p Ͻ 0.005) retardation of Foxp3 ϩ cells (Fig. 8, F and G). Consequently, it may be concluded from the data that CaeA can neutralize the suppressive action of IFN-␥ in TGF-␤-mediated Smad3 signaling and, thus, helps in rescuing Tregs.

DISCUSSION
It has been reported recently that autoimmune diseases have increased considerably in the developed world (44). Furthermore, women suffer three times more than men from autoimmune diseases. Autoimmune diseases can inflict debilitating damage to body components. These diseases can influence virtually any part of the body, including the brain/nervous system (multiple sclerosis), heart (myocarditis), pancreas (type 1 diabetes), joints (rheumatoid arthritis), and multiple organs (systemic lupus erythematosus) (45). Several therapies are available for the treatment of autoimmune diseases. However, it is unfortunate that existing therapies have severe side effects (6). For example, more than 100,000 Americans are hospitalized annually because of the side effects of ulcers and gastrointestinal bleeding. As a result, 7000 -10,000 of them die because of druginflicted injuries (46). With limited options for drug therapy, it is an urgent need and a challenge for the scientific community to identify alternative therapies that can effectively and safely control and cure autoimmune diseases.
Recently, the role of Tregs has been linked successfully to immune suppression (47)(48). Further, Tregs can effectively impede the function of Th1 and Th17 cells. Th1 and Th17 cells play important role in provoking autoimmune diseases (49). Furthermore, Tregs can be generated successfully in vitro. Therefore, this strategy can be used as a possible remedial measure to treat autoimmune diseases. Unfortunately, such Tregs were unstable and lost their regulatory properties (32). Class-specific depletion of gut microbiota and cytophage-flavo- bacter-bacteroidetes by vancomycin treatment has also been known to suppress Th17 differentiation and induction of Tregs in vivo (50). The change in the gut microbiota also enhances the susceptibility to allergic asthma (51). Hence, a regular search is being undertaken to identify safer and better methods for generating stable Tregs (14). For this reason, we identified CaeA, which can successfully generate stable and functional Tregs in vitro and in vivo. Consequently, the following major observations were made during the study with CaeA: the induction of the generation of Tregs individually and in concert with TGF-␤, the suppression of Th1 and Th17 cells, the amelioration of experimental arthritis, the inhibition of the IL-6-and IFN-␥mediated STAT1 phosphorylation, and the augmentation of TGF-␤-induced signaling of Smad3.
In this study, we demonstrated the vis-à-vis impact of CaeA on the generation of Tregs. Interestingly, CaeA successfully induced the formation of Tregs. Additionally, CaeA exhibited a synergism with TGF-␤ and further substantiated the augmentation in the pool of Tregs. We further authenticated the role of CaeA in generating Tregs by increased mitochondrial membrane potential (lipid peroxidation). It has been reported that Tregs rely on lipid peroxidation (30). Further, the presence of CD44 hi -expressing effector T cells has a negative effect on Tregs (52). However, we noticed that CaeA down-regulated CD44 on activated CD4 T cells, eventually favoring Treg generation. Furthermore, we demonstrated that cTregs were functionally competent because they inhibited the activation of effector T cells and suppressed alloreactivity.
It has been reported that gene expression elicited by cytokines can program T cell differentiation by their respective transcription factors. The cytokines IL-12ϩIFN-␥, TGF-␤, and IL-6ϩTGF-␤ are known to favor the differentiation of naïve CD4 T cells to Th1, Tregs, and Th17, respectively (16,53,54). IFN-␥ induces the activation of the transcription factor STAT1 through its receptor-associated protein tyrosine kinase JAK1 (55). The inhibitor for the JAK-STAT pathway, tofacitinib, is used to treat rheumatoid arthritis patients (56,57). Suppressors of cytokine signaling (SOCS) are a special class of proteins responsible for the inhibition of cytokine signaling by preventing the phosphorylation of STATs (36). SOCS1 specifically antagonizes interferon signaling by inhibiting the phosphorylation of STAT1 (37). T-bet is a transcription factor that is strongly induced by STAT1 and is essential for the differentiation of Th1 and Th17 cells (58). IL-6 transduces the signal through the transcription factor STAT3 and also shares the STAT1 signaling pathway (59), whereas TGF-␤ signals through a cascade of the transcription factors Smad2, Smad3, and Smad4 (40). Hence, on the basis of these earlier findings, we next established the mechanism of action operating in the CaeA-mediated generation of Tregs. Remarkably, our results demonstrate that CaeA antagonizes the phosphorylation of the transcription factor STAT1 induced by IFN-␥ and IL-6. Our results showed the enhanced expression of SOCS1 by CaeA. SOCS1 is responsible for the stabilization of Foxp3 by controlling Th1 and Th17 differentiation and suppressing STAT1 phosphorylation (60), therefore indicating that CaeA inhibits the differentiation of both Th1 cells and Th17 cells by increasing SOCS1. IFN-␥-mediated phosphorylation of STAT1 leads to the expression of T-bet, Smad7, and FasL (61,62). Intriguingly, CaeA treatment down-regulated the exhibition of T-bet, Smad7, and FasL. It signifies that CaeA controls the expression of T-bet, Smad7, and FasL by antagonizing STAT1. It has been reported that Smad7 is responsible for the inhibition of the regulatory function of Tregs (63). Interestingly, it was noted that CaeA decreased the level of Smad7. This may be correlated with the enhanced suppressive ability of c␤Tregs.
The signaling events triggered by TGF-␤ are regulated negatively by IFN-␥ (62). Phosphorylation of STAT1 induces transcription of Smad7, which is a negative regulator for the cascade of Smad3-mediated TGF-␤ signaling (64). Interestingly, CaeA suppressed IFN-␥ induced Smad7 expression. Further, CaeA rescued inhibition of the Smad3 cascade, induced by IFN-␥-mediated Smad7 signaling. This indicates that CaeA inhibits the neutralizing effect of IFN-␥ on TGF-␤-mediated Smad3 signaling and, thereby, promotes the generation of Tregs. It is also known that IFN-␥ induces reactive oxygen species generation, which interferes with the formation of Tregs (43). Remarkably, it was noticed that CaeA suppressed the release of reactive oxygen species and, therefore, may be supporting the generation of Tregs, even in the presence of IFN-␥ (data not shown). SIS3 is an inhibitor of Smad3-mediated signaling of TGF-␤. Finally, we established the specificity of CaeA-mediated Smad3 induction of the generation of Tregs by using SIS3. We noticed a significant decline in the frequency of Foxp3 ϩ Tregs. These data categorically authenticate the specificity of CaeA-induced generation of Tregs. We report here, for the first time, that suppression of IFN-␥-mediated STAT1 leads to enhanced Smad3 activity, thereby increasing the Foxp3 ϩ Treg population. In essence, our signaling data provide insights into the mechanism of action of induction of Tregs by CaeA by stimulating the Smad3 pathway.
Although Tregs are known to regulate the suppression of the immune system, this property is not very well retained in vivo (47,48,65). Furthermore, it is also known that the presence of the proinflammatory cytokines IL-6, IL-12, and IFN-␥ can influence the stability of Tregs. Therefore, this may limit their application for treating autoimmunity (31)(32). Hence, it was very important for us to monitor the in vivo stability of cTregs. Interestingly, our study revealed that cTregs were not affected by the IL-6, IL-12, and IFN-␥ milieu. These conditions favor the differentiation of Th17 and Th1 cells, respectively. We also demonstrated that CaeA can directly suppress the generation of Th1 and Th17 cells.
We proved the functionality of cTregs in vivo in an experimental model of arthritis. It is worth mentioning here that the animals treated with CaeA exhibited a significant inhibition of arthritis, as evidenced by decreased inflammation and clinical symptoms. Interestingly, these animals also showed an enhanced population of Foxp3 ϩ Tregs. Overall, our study indicates that CaeA may be a potent future immunotherapeutic agent for treating arthritis by inducing the generation of Tregs.