Curdlan Limits Mycobacterium tuberculosis Survival Through STAT-1 Regulated Nitric Oxide Production

Host-directed therapies have emerged as an innovative and promising approach in tuberculosis (TB) treatment due to the observed limitations of current TB regimen such as lengthy duration and emergence of drug resistance. Thus, we explored the role of curdlan (beta glucan polysaccharide) as a novel strategy to activate macrophages against Mycobacterium tuberculosis (Mtb). The aim of the study was to investigate the role of curdlan in restricting the Mtb growth both in vitro and in vivo. Further, the immunomodulatory potential of curdlan against Mtb and the underlying mechanism is largely unknown. We found that curdlan treatment enhanced the antigen presentation, pro-inflammatory cytokines, Mtb uptake and killing activity of macrophages. In vivo studies showed that curdlan therapy significantly reduced the Mtb burden in lung and spleen of mice. Administration of curdlan triggered the protective Th1 and Th17 immunity while boosting the central and effector memory response in Mtb infected mice. Curdlan mediated anti-Mtb activity is through signal transducer and activator of transcription-1 (STAT-1), which regulates nitric oxide (NO) production through inducible NO synthase (iNOS) induction; along with this activation of nuclear factor kappa B (NF-κB) was also evident in Mtb infected macrophages. Thus, we demonstrate that curdlan exerts effective anti-tuberculous activity anti-tuberculous activity. It can be used as a potential host-directed therapy against Mtb.


INTRODUCTION
Tuberculosis (TB) continues to be the devastating infectious disease with the highest mortality and morbidity after HIV (Glaziou et al., 2018). Owing to its poor diagnosis and long regimen of drug treatment, resistant strains of Mycobacterium tuberculosis (Mtb) are likely to develop in the host (Gandhi et al., 2010;Lawn and Zumla, 2011;Prabowo et al., 2013). Among Mtb infected individuals, only 10-15% develops the disease during their lifetime, rest remains protected until the resurgence in their immunity (Alene et al., 2018). Thus, there is a need to understand the intricate interaction between pathogen and host immune factors, which regulates the disease pathogenesis and outcome. This would aid to design novel host-directed therapeutic interventions to control TB. An effective approach is to stimulate the host immune system against Mtb.
Beta glucans are known to bind PRRs and elicit host defense responses (Vannucci et al., 2013). Curdlan is a high molecular weight beta 1-3 glucan polysaccharide, extracted from nonpathogenic strains of the bacterium Alcaligenes faecalis (Zhang and Edgar, 2014). It has been approved by FDA owing to its safety as food additive and low cost of production (Spicer et al., 1999). It is known to stimulate dectin-1, a C-type lectin receptor that signals through Syk (Xie, 2012). Dectin-1 is expressed on myeloid cells including M s and DCs (Taylor et al., 2002). Mtb interaction with dectin-1 has been implicated in induction of Th1/Th17 immunity and higher production of IL-12 cytokine by infected DCs (Rothfuchs et al., 2007;van de Veerdonk et al., 2010). Although the beneficiary effects of curdlan have been reported in many diseased conditions such as cancer and leishmaniasis (Leibundgut-Landmann et al., 2008;Ghosh et al., 2013), the anti-Mtb activity of curdlan and its mechanism has not been extensively studied.
Induction of nitric oxide synthase (iNOS) is implicated in the control of diverse pathogens including Mtb (Chan et al., 1992). Immune cells produce NO through conversion of L-arginine to L-citrulline in the presence of iNOS (Bogdan et al., 2000). Antimicrobial effect of NO has been well established in both murine and human studies (Chan et al., 2001). Few reports have revealed that iNOS activity is controlled by signal transducer and activator of transcription (STAT) signaling pathways; however, the action of STAT-1 in regulating curdlan induced iNOS expression in Mtb infected M s is not known. In this study, we show that curdlan boosts the functionality of M s against Mtb and in vivo administration of curdlan in Mtb infected mice triggered the protective T cell response and reduced Mtb burden. Furthermore, curdlan mediated control of Mtb survival via NO release involved STAT-1 activation. This study demonstrates an important immunotherapeutic role of curdlan against Mtb infection.

Animals
Six-to 7-week old C57BL/6 female mice were obtained from the Animal Facility of CSIR-IMTECH and approved by the Institutional Animal Ethics Committee (IAEC) of CSIR-IMTECH. All the animal experiments and protocols used in the study were approved by the Institutional Animal Ethics Committee (IAEC) of CSIR-IMTECH. The experiments were done in accordance with the National Regulatory Guidelines released by Committee for the Purpose of Control and Supervision of Experiments on Animals (No. 55/1999/CPCSEA), Ministry of Environment and Forest, Government of India.

Culture, Infection, and Stimulation of Macrophages (M s)
Cells isolated from femurs and tibia of mouse bone marrow were cultured in complete medium consisting of RPMI-1640 (GIBCO Invitrogen Corporation, Grand Island, NY, United States) and 10%-FBS with L929 SN (20%), as a source of M-CSF. The cell cultures were incubated at 5% CO 2 /37 • C. The media was replaced on day 3. After 7 days, M s were harvested and used for experiments.
For Mtb infection experiments, 2 × 10 5 cells per well (in triplicates) in 48-well tissue culture grade plate were infected with Mtb at a MOI of 5 Mtb/cell. After 4 h, the cells were extensively washed with PBS (1×) 3-4 times and resuspended in the complete media (antibiotics free RPMI+ 10%-FBS) with amikacin (2 µg/ml) to remove the extracellular bacteria. Amikacin was kept throughout the experiment to kill extracellular bacteria . Thereafter, cells were stimulated with curdlan (50 µg/ml) for the indicated time points.
For alveolar M s, lungs of mice were washed with 1 ml PBS by flushing several times and the broncho-alveolar lavage was collected. The cells from the solution were pelleted down and were resuspended in DMEM containing 10% FBS for the subsequent infection experiment (Jung et al., 2017).

Analysis of M s by Flow Cytometry
M s were treated with Fc block (anti-CD16/32 antibody) for 30 min followed by incubation with fluorochrome-labeled antibodies (F4/80 + /CD11b + /CD86 + /CD40 + /MHC-II + ) and their isotype-matched control antibodies for another 30 min at 4 • C. Later, cells were washed and fixed in PFA (1%). The flow cytometry data were acquired using FACS Aria-II (BD Biosciences, Ashland, OR, United States)  and analyzed using DIVA software (BD Biosciences, Ashland, OR, United States). Gating strategy for assessing M s phenotype and expression markers through flow cytometry is mentioned in Supplementary Figure S1A. Integrated MFI (iMFI) in the flow cytometry analysis is calculated by multiplying the relative frequency (% positive) of cells expressing a particular marker with the mean fluorescence intensity (MFI) of that population.

Annexin V and Propidium Iodide (PI) Assay
Mtb infected M s were stimulated with curdlan (50 µg/ml) for 48 h, and then stained with Annexin V-FITC (5 µl/tube/10 6 cells) in the dark for 15 min at RT. Thereafter, cells were incubated with 2 µl of PI (50 mg/ml) for 10 min at RT . Cells were immediately assessed through flow cytometry using FACS Aria-II (BD Biosciences, Ashland, OR, United States).

Intracellular Killing Assay
M s, 2 × 10 5 cells per well (in triplicates) in 48-well plate were infected with Mtb (MOI of 5) for 4 h followed by extensive washings (3-4 times) with PBS. Later, the cells were suspended in antibiotic-free complete media (RPMI + FBS-10%) with amikacin (2 µg/ml) and treated with curdlan (50 µg/ml) for 48 h. Thereafter, cell culture SNs were removed and cells were made to lyse with saponin (0.1%) and the lysate (100 µl) was plated on mycobacterial media 7H11 agar plates. Bacterial (Mtb) colonies were counted after 21 days .
In case of experiments with Syk inhibitor, cells were pretreated with Syk inhibitor, piceatannol (20 µg/ml) (Shin et al., 2008) for 45 min prior to Mtb infection.

Antigen and Mtb Uptake Assay
For Mtb uptake experiments through CFU, M s (2 × 10 5 cells per well in triplicates) in 48-well plate were first stimulated with curdlan (50 µg/ml) for 48 h. It was followed by Mtb infection of cells (MOI = 5). After 4 h, cells were washed with PBS and treated for 1 h with amikacin (2 µg/ml) to kill extracellular bacteria. Thereafter, cell SNs were removed and cells were lysed with the saponin (0.1%) and lysate (100 µl) was plated on 7H11 agar plates. Bacterial colonies were enumerated on day 21 .
For confocal microscopy experiment, M s (3 × 10 5 cells/well) were stimulated with curdlan for 48 h and then infected with GFP + Mtb for 4 h. Thereafter, cells were treated with amikacin for next 1 h followed by washings with cold PBS (1×) and then fixed with PFA (2%) for 15 min. Cells were then kept on poly-L lysine coated coverslips . The imaging and analysis of intracellular uptake of Mtb by M s was performed with Nikon A1 confocal laser microscope system (Nikon, Tokyo, Japan). Also, Z-stacks were acquired to exclude the possibility of extracellular bacteria.
For dextran antigen uptake test, curdlan stimulated M s [4 × 10 5 cells per well (in triplicates) in 24-well plate] were incubated with dextran antigen labeled with FITC for 30 min at 37 • C, or on ice as the control. Dextran uptake was then monitored by flow cytometry.

Nitric Oxide (NO) Estimation
Supernatants from cell culture were collected and NO was quantified by Griess assay. Equal volume of SNs (50 µl) and Griess reagent (50 µl) was incubated for 5 min/RT . Later, absorbance was recorded at 550 nm using NanoDrop spectrophotometer (BioTek, Winooski, VT, United States).

CD4 T Cells Isolation and Co-culture With M s
Mice were aerosol challenged with Mtb (∼100 CFU). After 21 days of infection, CD4 T cells were isolated from single cell suspension of spleen using BD IMag TM mouse CD4 T lymphocyte enrichment set-DM (BD Biosciences, San Diego, CA, United States) according to the mentioned instruction's. For co-culture, Mtb infected curdlan stimulated M s (2 × 10 4 cells/well) in 96-well plate were co-cultured with CD4 T cells (2 × 10 5 cells/well) (M s: CD4 T cells ratio of 1:10). CD4 T cells were labeled with CFSE (2 µM) prior to co-culture (Rai et al., 2016). PPD (25 µg/ml) was added to the culture to assess Mtb-specific T cell response. After 72 h, cells were stained with PE-labeled anti-CD4 antibody. It was followed by analysis of CFSE low CD4 T cell population through flow cytometry. CD4 T cells gating strategy is depicted in Figure 3.

Western Blotting
M s (1 × 10 6 cells/ml) were harvested and lysed by cytosolic extraction lysis buffer (along with PMSF, protease and phosphatase inhibitor cocktail). Equal concentration of protein in lysates was run on SDS-PAGE, transferred to the PVDF membrane followed by blocking with 5% BSA and probed with iNOS, phospho and non-phospho form of STAT-1, STAT-3, STAT-6, and β-actin (loading control) antibodies . Blots were developed by chemiluminescence kit (ECL; Pharmacia-Amersham, Freiburg, Germany) and visualized using ImageQuant LAS 4000 (GE Healthcare, Pittsburgh, PA, United States). Further, images were analyzed by ImageJ analysis software (Fujifilm, New York, NY, United States).

Therapeutic Strategy
Mice were aerosol challenged with Mtb (∼100 CFU). After 21 days of infection, animals were administered curdlan (20 mg/kg of body weight) subcutaneously (s.c.) twice with an interval of 2 weeks. The control group (placebo) was administered with PBS. Anti-TB drug, isoniazid (25 mg/kg body wt. of mice) was orally administrated twice with 0.1% CMC (carboxymethylcellulose) along with the curdlan. After 7 weeks of infection, animals were sacrificed and Mtb burden was assessed in lung and spleen by CFU assay. The weight of the respective organ was taken into consideration while calculating Mtb burden . Bacterial burden was calculated as CFU/0.1 ml × dilution factor × ml/organ = CFU/organ and results are expressed as log 10 .

Lymphocytes Isolation, T Cells Activation, and Phenotypic Analysis
Mice were perfused with cold PBS containing heparin (100 U/ml) followed by isolation of lungs and spleen. The single cell suspension of tissue was made. Later, cells were treated with ACK lysis buffer to lyse RBCs, washed with PBS (3×) and then resuspended in complete media. Further, cells (2.5 × 10 5 cells per well in 96-well round bottom plate) were in vitro cultured with PPD (25 µg/ml) for 48-72 h to assess the phenotypic markers and intracellular cytokines. For intracellular staining, cells were stimulated with phorbol 12-myristate 13acetate (PMA) (50 ng/ml) and ionomycin (1 µg/ml) for 4 h followed by the incubation with brefeldin A (5 mg/ml) for another 2 h to block protein transport to the Golgi. Later, cells were stained for the surface marker CD4 and fixed with PFA (4%) for 10 min. The cells were then permeabilized by treatment with saponin buffer (0.2% in 1× PBS) for 15 min/RT. Further, cells were incubated with anti-mouse IFN-γ, and antimouse IL-17 antibodies along with their respective isotypematched controls. Regular washings were carried out after each and every step (Rai et al., 2016). CD4 T cells gating strategy for the intracellular cytokine analysis is depicted in Supplementary Figure S6.
For surface markers analysis, cells were incubated with Fc block for 30 min/4 • C and then stained with fluorochromelabeled antibodies and their matched isotype control antibodies to check the surface expression of CD4, CD44, CD62L, and CCR7 for 30 min at 4 • C. Then the stained cells were fixed in PFA (1%) and monitored via flow cytometry (Rai et al., 2016). The CD4 T cells gating strategy for the memory phenotype analysis is depicted in Supplementary Figure S7.
Cells were acquired in FACS Aria II and analysis was performed with BD FACS DIVA software.

Histopathology
Lung specimens were fixed in 10% formalin buffer followed by staining of histological sections with hematoxylin and eosin dye. The microscopic photographs were captured on Olympus IX71 microscope (Olympus, Tokyo, Japan). The images are shown at 40× magnifications.

NF-κB Activation Analysis
For confocal microscopy, Mtb infected M s were placed on poly-L-lysine coated coverslips for 15-20 min. The cells were then stimulated with the curdlan (50 µg/ml) for 30 min at 37 • C. Later, cells were fixed with PFA (2%) for 10 min, followed by Triton X-100 (0.1%) treatment for 2 min. The samples were then incubated with BSA (2%) for 2 h to block the non-specific sites. Later, cells were incubated with anti-mouse NF-κB p65 Ab (1:400) for 2 h. Subsequently, cells were incubated with Alexa fluor 633-anti-rabbit Ab for 1 h, followed by staining with DAPI dye. Regular washings were performed at each step . The cells were imaged through Nikon A1 confocal laser microscope (Nikon, Tokyo, Japan). Data were analyzed using image analysis software, Nikon NIS-AR 4.1 (Nikon, Melville, NY, United States).
For electrophoretic mobility shift assay (EMSA), nuclear extract was prepared from Mtb infected and curdlan stimulated M s. After 30 min of stimulation, an equal amount of nuclear extract (3 µg) from each sample was incubated for 20 min/37 • C in water bath with [P 32 ] end labeled duplex oligonucleotides that contains binding site for NF-κB. The DNA-protein complexes were resolved on a native PAGE-gel (7%) by electrophoresis . The gel was dried and exposed to screen at RT for 6-12 h and scanned by phosphor-imager scanning screen (Fujifilm, FLA-5000, Tokyo, Japan).

Statistical Analysis
One-way non-parametric ANOVA test and Student's t-test were used for analysis of the multiple and two groups, respectively. Data were analyzed with Graph Pad Prism 6 software. P-values of <0.05 were considered significant.

Curdlan Restricts Mycobacterial Growth in M s
Initially, we examined the effect of curdlan on Mtb survival in bone marrow-derived M s. Dose titration experiment showed optimum secretion of IL-6 by M s at a dose of 50 µg/ml of curdlan and was found to be non-cytotoxic (Supplementary Figures S1B, S2). Mtb infected M s treated with curdlan in comparison to untreated control, exhibited significant (p ≤ 0.01) decrease in intracellular Mtb growth as seen in low CFU counts concomitant with the increase in IL-6 cytokine (p ≤ 0.01) release in cell SNs (Figures 1A,B). Further, to define the specificity of the involved pathway, we inhibited the Syk kinase, as curdlan signals in a dectin-1/Syk dependent manner (Brown, 2006). Inhibitor of Syk, piceatannol substantially reduced the stimulatory effect of curdlan in infected M s as seen in the impaired clearance of Mtb (p ≤ 0.01) and IL-6 (p ≤ 0.001) production (Figures 1A,B). The ability of curdlan to restrict Mtb growth was also confirmed using THP-1 derived M s and alveolar macrophages through CFU assay (Supplementary Figures S3A,B). It is interesting to note that curdlan exerts its Mtb killing effect optimally at 48 h of treatment (Supplementary Figure S3C). Curdlan induces elevated levels of proinflammatory cytokines IL-12 (p ≤ 0.001), TNF-α (p ≤ 0.001) and IL-1β (p ≤ 0.05) (Figures 1C-E), while anti-inflammatory IL-10 secretion was found to be reduced (p ≤ 0.01) in culture SNs of Mtb infected M s versus untreated control ( Figure 1F). Additionally, we found upregulated expression of activation markers such as CD86, CD40, and MHC-II on M s after curdlan treatment through flow cytometry ( Figure 1G). The gating strategy for assessing macrophage population is depicted in Supplementary  Figure S1A. Moreover, curdlan elicits higher expression of genes such as dectin-1 (receptor for curdlan) along with IL-6 and TNF-α that are involved in protective immune response while suppressive Arg-1 was found to be downregulated in Mtb infected M s (Supplementary Figure S4).

Curdlan Enhances M s Phagocytosis Ability
Next, we sought to determine the uptake ability of curdlan stimulated M s. Thus, we stimulated the M s with curdlan followed by their incubation with Dextran-FITC. Flow cytometry data showed enhanced (p ≤ 0.01) dextran antigen uptake, relative to unstimulated M s (Figures 2A,B). The gating strategy is same as shown in Supplementary Figure S1A. For Mtb phagocytic assay, M s were first stimulated with curdlan followed by Mtb infection. Thereafter, GFP-Mtb uptake was assessed by confocal microscopy ( Figure 2C). Further, these results were confirmed by significant increase (p ≤ 0.01) of Mtb uptake as observed through CFU assay ( Figure 2D). This data suggests that curdlan has the capacity to improve phagocytosis of Mtb by M s.

Curdlan Stimulated M s Induce Better Mtb Antigen-Specific CD4 T Cell Proliferation and Th1/Th17 Immune Response
One of the important attributes of antigen presenting cell is to activate T cells (Khan et al., 2012). Thus, we investigated the stimulatory potential of Mtb infected curdlan stimulated M s to activate CD4 T cells (isolated from spleen of Mtb infected mice) in the presence of PPD (Mtb antigen). Curdlan treatment of infected M s resulted in the better proliferation of CD4 T cells in comparison to untreated macrophages as determined by flow cytometry analysis (Figures 3A,B). In addition, these CD4 T cells secreted higher levels of IFN-γ (p ≤ 0.01) and IL-17 (p ≤ 0.001) cytokines as detected in culture SNs by ELISA. This represents the induction of Th1 and Th17 response, respectively, by curdlan activated M s (Figures 3C,D).

In vivo Therapeutic Efficacy of Curdlan Against TB
Next, we assessed the therapeutic effect of curdlan in the experimental model of TB. After 21 days of Mtb infection, curdlan was administered subcutaneously (s.c.) in mice twice after the gap of 2 weeks ( Figure 4A). Strikingly, curdlan therapy significantly reduced the Mtb burden in the lungs (p ≤ 0.01) and spleen (p ≤ 0.05) as evident by the decline in CFU counts (Figures 4B,C). Further, histological analysis of lung tissue also showed improved pathology and less granulomatous  lesions as compared to placebo control ( Figure 4D). Interestingly, we also observed that immunization of Mtb infected mice with curdlan in combination with isoniazid (INH) considerably (p < 0.01) reduced the Mtb burden in lungs. Thus, curdlan administration along with INH imparted better protection against Mtb (Figure 4E). In addition, curdlan treatment led to higher levels of IFN-γ (p ≤ 0.01), IL-12 (p ≤ 0.05), and NO (p ≤ 0.001) whereas IL-10 (p ≤ 0.01) secretion was found to be reduced in culture SNs of lung cells (Figures 4F-I). We further validated the cytokines and NO production at mRNA level through qRT-PCR (Supplementary Figure S5).

Curdlan Therapy in Mice Induces Mtb
Antigen-Specific IFN-γ and IL-17 Producing CD4 T Cells CD4 T cells secreting IFN-γ and IL-17 serve as main effector cells during Mtb infection (Flynn et al., 1993;Gopal et al., 2014). We next evaluated the impact of curdlan administration on Mtb antigen-specific T cell response against Mtb. As shown in Figures 5A-D, lungs cells of mice treated with curdlan exhibited both higher percentage and absolute numbers of IFN-γ + (p ≤ 0.001) and IL-17 + (p ≤ 0.01) CD4 T cells after in vitro stimulation with PPD as compared to placebo control indicating effective generation of Th1 and Th17 immunity. Further, the population of polyfunctional CD4 T cells co-expressing both IFN-γ and IL-17 (p ≤ 0.01) was also increased in these mice as determined by flow cytometry (Figures 5E,F). The gating strategy for flow cytometry analysis of IFN-γ and IL-17 producing CD4 T cells is depicted in Supplementary Figure S6.

Curdlan Treatment in vivo Enhances CD4 T Cell Proliferation and Memory Response
Mtb is known to impair the proliferation capability of antigenspecific CD4 T cells, crucial for host immunity (Canaday et al., 2001). After curdlan administration in Mtb infected mice, lung lymphocytes were isolated (7 weeks post Mtb infection), CFSE (2 µM) labeled and in vitro stimulated with PPD (25 µg/ml) for 72 h to assess the CD4 T cell proliferation through flow cytometry. Interestingly, these mice exhibited enhanced (p ≤ 0.01) ability of CD4 T cells to proliferate in comparison to placebo control (Figures 6A,B).
Another hallmark of effective and long-term protection is the generation of memory T cells. They provide an efficient immune response on pathogen re-exposure (Gourley et al., 2004). In vivo administration of curdlan resulted in a considerably increased frequency of central (CD44 hi /CD62L hi ) (p ≤ 0.01) and effector memory (CD44 hi /CD62L low ) (p ≤ 0.001) CD4 T cell pool (Figures 6C-E). Further, chemokine receptor CCR7 (C-C chemokine receptor type 7) is involved in the migration of memory T cells to peripheral organs (Ohl et al., 2004). Thus, we next evaluated the migratory potential of activated CD4 T cells in the lungs of animals. Intriguingly, we observed elevated (p ≤ 0.001) population of CD44 hi /CCR7 hi CD4 T cells in curdlan treated mice versus placebo control (Figures 6F,G). The gating strategy for memory phenotype analysis of CD4 T cells is depicted in Supplementary Figure S7.
Moreover, we also found substantial increase of central memory cells represented as CD44 hi CD62L hi CCR7 hi and effector memory (CD44 hi CD62L lo CCR7 lo ) CD4 T cells in lungs of mice with curdlan therapy (Supplementary Figure S7B).

Curdlan Elicits Generation of Nitric Oxide in Mtb Infected M s
Next, we were interested to decipher the mechanism involved in curdlan mediated killing of Mtb. The expression of iNOS and NO production has been shown to play a pivotal role in Mtb killing (Chan et al., 1992). Thus, we treated Mtb infected M s with curdlan and assessed the expression of iNOS by western blot. Interestingly, curdlan induced higher levels of iNOS relative to untreated M s ( Figure 7A). This observation is in line with the increased (p ≤ 0.01) NO production in curdlan stimulated M s ( Figure 7B). Further, the involvement of NO was confirmed by inhibiting its synthesis with iNOS inhibitor, N-monomethyl L-arginine (NM). The curdlan mediated increase in NO secretion and Mtb killing was impaired upon NM treatment (Figures 7B,C). These results indicate that NO plays a crucial role in curdlan induced clearance of Mtb in M s.

Curdlan Activates STAT-1 and NF-κB in Mtb Infected M s
We further assessed the activation of signaling pathways in curdlan stimulated M s during Mtb infection. Given that STAT-1 has been reported to be involved in the suppression of pathogens including Mtb (Najjar and Fagard, 2010). Additionally, Tyr701 phosphorylation of STAT-1 is related to the expression of iNOS (Heitmeier et al., 1999), thus we were interested to examine STAT pathway involvement in curdlan triggered iNOS expression in Mtb infected M s. We performed western blot analysis using cell lysate of Mtb infected curdlan treated and untreated M s. Strikingly, we found that the level of pSTAT-1 was increased in curdlan stimulated M s versus untreated control ( Figure 8A). Likewise, we observed a reduction of phosphorylated STAT-6 and STAT-3 in these M s. We ascertained the role of STAT-1 in curdlan mediated iNOS expression and NO generation with fludarabine (a specific inhibitor of STAT1; STAT1 i). As shown in Figures 8B,C, M s treated with STAT-1 inhibitor prior to curdlan stimulation displayed impaired iNOS expression and NO secretion.
Moreover, we observed considerable activation and translocation of NF-κB into the nucleus of Mtb infected and curdlan treated M s as seen by confocal microscopy and also corroborated this result by EMSA (Figures 8D,E). These data indicate the significant role of STAT-1 and NF-κB pathways in curdlan-mediated anti-Mtb effect.

DISCUSSION
Treatment of TB is extremely challenging due to drug-associated side effects and the rise of resistant Mtb strains (Kaufmann, 2013). This imperatively demands to develop innovative therapies for better management and treatment of TB. Recently, host-directed therapies involving immunomodulators have gained attention in restricting Mtb infection Pahari et al., 2016Pahari et al., , 2017. Innate immunity receptors viz. TLRs, CLRs, NLRs, and RLRs plays a central role in host defense against the array of pathogens (Kawai and Akira, 2011). Immunomodulators that trigger signaling through innate receptors present on M s can be utilized against Mtb (Pahari et al., 2018). Curdlan is a biological response modifier that stimulates dectin-1 receptor on M s and other myeloid cells (Goodridge et al., 2009). It has been approved by FDA, with proven beneficial effects in leishmania, cancer, bacterial and viral infections (Ghosh et al., 2013). There is lack of studies depicting mechanistic role of curdlan in imparting immunity against Mtb. Thus, we investigated in vitro and in vivo efficacy of curdlan during Mtb infection. Further, the underlying mechanism and therapeutic potential of curdlan to be used as host-directed therapy in TB was demonstrated.
In order to generate protective immune response against Mtb, optimum expression of costimulatory molecules such as CD86, CD80, and CD40 on M s is very crucial for the activation of T cells (Aderem and Underhill, 1999;Harris and Ronchese, 1999;Reis e Sousa, 2004;Joffre et al., 2009) along with the production of pro-inflammatory cytokines (Domingo-Gonzalez et al., 2016). Curdlan triggered activation of M s against Mtb can be attributed to: (i) upregulation of MHC-II, CD40, and CD86 expression; (ii) augmented secretion of IL-12, IL-1β, and TNF-α, while the decline in suppressive IL-10; (iii) enhanced capacity to phagocytose and kill Mtb; (iv) improved ability to activate and proliferate CD4 T cells. Further, in vivo curdlan administration resulted in: (i) increased Th1 and Th17 immune response with the expanded pool of memory CD4 T cells in lungs; (ii) declined Mtb survival in lungs and spleen along with improved lung pathology in Mtb infected mice. Previously, studies have highlighted the beta glucans mediated activation of antigen presenting cells that cause expansion and differentiation of T cells against tumors (Leibundgut-Landmann et al., 2008). It is worth to mention here that our study also demonstrates the generation of Mtb specific CD4 T cell response by curdlan activated macrophages. The limitations of the study include lack of dectin-1 knockdown or knockout models. Dectin-1 agonist curdlan have been reported to work through the Syk pathway (Brown, 2006). Thus, we have examined the specificity of curdlan acting via Syk dependent dectin-1 pathway by piceatannol (Syk inhibitor) (Shin et al., 2008) that resulted in impairment of curdlan mediated Mtb killing in macrophages. However, the possibility of Sykindependent effects of piceatannol cannot be ruled out and impact of more specific Syk inhibitors such as R406 needs future investigation.
FIGURE 9 | Curdlan induced signaling pathways and events in Mtb infected M s. This proposed model illustrates the signaling events involved in restricting the Mtb growth. Curdlan binds to its receptor, dectin-1 and induces the STAT-1 phosphorylation in Mtb infected M s. This leads to the pSTAT-1 mediated activation of iNOS gene to generate NO, which has antimicrobial properties to restrict the Mtb survival. Subsequently, there is activation and translocation of NF-κB into the nucleus. These signaling pathways activate and boost the function of M s (increased proinflammatory cytokines, bacterial uptake, expression of costimulatory molecules) which further elicit the protective effector and memory T cell response against Mtb.
CD4 T cells play an indispensable role in conferring host defense against Mtb (Canaday et al., 2001). Th1 cells and Th17 cells play a crucial role in mounting a protective immune response against Mtb (Flynn et al., 1993;Gopal et al., 2014). Interestingly, in line with the previous studies (Hsieh et al., 1993;LeibundGut-Landmann et al., 2007), we observed that curdlan induced the secretion of Th1 and Th17-polarizing cytokines, such as IL-12 and IL-1β, respectively, in Mtb infected macrophages. Accordingly, we found expanded pool of Th1 cells and Th17 cells in the animals with curdlan therapy. This is consistent with the substantial decline in Mtb load in lungs, as well as prevented dissemination of Mtb to the spleen. Recently, polyfunctional T cells have been reported to impart better protection against the infection (Caccamo et al., 2010;Rai et al., 2016). Furthermore, the presence of polyfunctional IFNγ + /IL-17 + CD4 T cells was evident in Mtb infected animals with curdlan treatment. The establishment of effective memory T cell response is necessary to impart long-term protection (Lindenstrom et al., 2009;Singh et al., 2010). Our results suggested better memory T cell response against Mtb as seen by the enhanced pool of central and effector memory CD4 T cells. Thereby, indicating a crucial impact of curdlan in improving the immunological memory.
The mechanism involved in curdlan mediated clearance of Mtb is unknown. The role of NO in host defense against Mtb has been reported earlier (Chan et al., 1992). Here, we demonstrate that stimulation of M s with curdlan induced higher expression of iNOS (involved in NO production). Inhibition of NO generation impaired the curdlan mediated Mtb killing.
Further, STATs are crucial in regulating the functionality and polarization of M s (Ohmori and Hamilton, 2001;Kernbauer et al., 2012). Interestingly, previous reports suggest that STAT-1 deficiency is linked to impaired anti-Mtb immunity while its activation results in protective M phenotype that possesses anti-microbial activity. On the other hand, STAT-3 and STAT-6 phosphorylation are prominent in suppressive M s that are unable to restrict Mtb survival (Sica and Mantovani, 2012;Lim et al., 2016). Moreover, STAT-1 and STAT-3 reciprocally regulate each other (Costa-Pereira et al., 2002;Qing and Stark, 2004). Our study is consistent with these findings, we show that curdlan activated M s exhibit increased STAT-1 phosphorylation while pSTAT-3 and pSTAT-6 levels were reduced indicating sufficient ability of M s to restrict Mtb growth. This is in accordance with the elevated level of IL-12, TNF-α, iNOS (protective factors) and reduced expression of IL-10 and Arg-1 (immunosuppressive factors) upon curdlan treatment. Thus, STAT-1 phosphorylation sustains the activation of curdlan stimulated M s. Strikingly, studies have described the regulation of iNOS expression by phosphorylation of STAT-1 (Lowenstein et al., 1993;Heitmeier et al., 1999). We hypothesize that STAT-1 binds to the promoter sequences of Inos gene and enhances the expression of iNOS and therefore the generation of NO. In line with this, we validate our findings by blocking STAT-1 with the specific STAT-1 inhibitor fludarabine. Interestingly, it abrogated the iNOS expression and NO secretion induced by curdlan. Thus, curdlan induced STAT-1 activation is required for the iNOS expression and NO production. Further, another important transcription factor, NF-κB plays a crucial role in controlling Mtb infection by inducing expression of many proinflammatory cytokines and protective immune mediators (Davis et al., 2005;Fallahi-Sichani et al., 2012). We show that curdlan stimulation in Mtb infected M s resulted in NF-κB activation which is also consistent with the observed induction of IL-12, TNF-α, IL-6, and iNOS.
In summary, we have demonstrated that anti-Mtb activity of curdlan is through activation of STAT-1 and NF-κB pathway activation that resulted in NO production and thus restricting the Mtb survival (Figure 9). This approach of boosting the host immune system with immunomodulators such as curdlan may prove to be an important therapeutic approach to treat TB.

ETHICS STATEMENT
Mice were obtained from the Animal Facility of CSIR-IMTECH and approved by the Institutional Animal Ethics Committee (IAEC) of CSIR-IMTECH. All the animal experiments and protocols used in the study were approved by the Institutional Animal Ethics Committee (IAEC) of CSIR-IMTECH. The experiments were done in accordance with the National Regulatory Guidelines released by Committee for the Purpose of Control and Supervision of Experiments on Animals (No. 55/1999/CPCSEA), Ministry of Environment and Forest, Government of India.