Infergen Stimulated Macrophages Restrict Mycobacterium tuberculosis Growth by Autophagy and Release of Nitric Oxide

IFN alfacon-1 (Infergen) is a synthetic form of Interferon (IFN)-α2b. Infergen has immunomodulatory activity and is effective against hepatitis C virus. However, the effect of Infergen (IFG) on Mycobacterium tuberculosis (Mtb) has not yet been reported. Therefore, for the first time, we have studied the influence of IFG in constraining the survival of Mtb in human macrophages. We observed that IFG significantly enhanced the maturation and activation of macrophages. Further, it substantially augmented the secretion of IL-6, nitric oxide (NO) and antigen uptake. Moreover, macrophages exhibited remarkably higher bactericidal activity, as evidenced by reduction in the Mtb growth. Infergen-mediated mechanism was different from the type-1 interferons; since it worked through the activation of NF-κB, phosphorylation of STAT-3 and Akt-PI3K that improved the bactericidal activity through autophagy and NO release. In future, IFG immunotherapy can be a novel strategy for treating patients and controlling TB.

growth of Mtb 22 . Taking into account these evidences, we were interested to monitor the influence of IFG on the activation of Mtb infected macrophages, thereby restricting the survival of the Mtb. It was quite fascinating to observe that IFG stimulated macrophages displayed significantly improved skill to restrict the intracellular growth of Mtb. The mechanism involved was through the augmentation of autophagy and release of NO. Both these pathways are considered crucial in curbing the growth of Mtb. Therefore, this study illustrates that IFG may have enough potential to work as an immunotherapeutic agent against TB.

IFG restricts the intracellular survival of Mtb. Mtb not only infects macrophages but also it can survive
in the hostile environment of these cells 23,24 . Thus, in the initial phase of the study, we investigated the impact of IFG on avirulent strain of Mtb H37Ra infected THP-1 macrophages (H37Ra-Mϕ ). It was quite exciting to note that IFG stimulated macrophages (Mϕ IFG) showed significant (p < 0.0001) reduction in the intracellular survival of Mycobacterium, as compared to unstimulated control [ Fig. 1A]. The decrease in the CFUs was observed in a dose dependent fashion. Likewise, substantial (p < 0.0008) decline in the growth of virulent strain H37Rv (Mtb) was noticed [ Fig. 1B]. Furthermore, we proved the above event through time-point kinetics by demonstrating a extensive (p < 0.0001) increase in Mtb killing (GFP-H37Ra) by flow cytometry assay [ IFG augments the secretion of cytokines. IL-6 is a major cytokine produced by macrophages in response to intracellular pathogens 25 . This cytokine plays an important role in T cell activation and inhibition of the growth of Mtb [25][26][27][28][29][30][31] . IFG stimulated H37Ra infected Mϕ produced considerably (p < 0.001) higher level of IL-6, as compared to unstimulated (no Infergen) macrophages infected with H37Ra [ Fig. 1D]. The increase in IL-6 was observed in a dose dependent manner. Further, these results were corroborated by IL-6 gene expression by RT-qPCR [ Fig. 1E]. Additionally, we confirmed these results by observing a significant (p < 0.01) increase in the level of IL-6 and IFN-γ by IFG stimulated H37Ra infected macrophages isolated from human PBMCs [ Fig. 1F,G]. Infergen is reported to promote Th1 polarization 32  IFG upregulates the expression of CD80, CD86 and HLA-DR. It has been reported that signaling through IFN-α /β can activate macrophages and other cells of the immune system during viral infection [31][32][33][34][35] . Further, MHC and costimulatory molecules expressed on the surface of the macrophages are crucial for the optimum activation of T cells. It was noticed that Mtb-infected IFG stimulated THP-1 macrophages (Mtb-Mϕ -IFG) exhibited substantially (p < 0.02) higher expression of costimulatory molecules CD40 (p < 0.0004), CD80 (p < 0.0001), CD86 (p < 0.02) and MHC-II (HLA-DR) (p < 0.003) [ Fig. 2A-D and Fig. S3A-D]. These results were substantiated using CD11b + human macrophages. We observed that IFG induces significant upregulation of CD80 (p < 0.01), CD86 (p < 0.05), compared to unstimulated controls [ Fig. 2E,F and Fig. S3E,F]. Thus, our data suggest that IFG not only augments the activation and maturation of Mtb infected Mϕ differentiated from THP-1, but also macrophages isolated from human peripheral blood.
We further checked the activation status of other immune cells like B cells and T cells. As compared to untreated B cells, IFG treated B cells displayed a higher expression of CD80, CD86 and HLA-DR, [ Fig. S4A-F]. We also noted significant (p < 0.01) upregulation of the activation marker CD25 on CD8 T cells treated with IFG [ Fig. S5A]. Moreover, IFG incubated CD8 T cells showed substantial improvement (p < 0.01) in their ability to lyse the target cells [ Fig. S5B]. Furthermore, IFG induced significant (p < 0.05) proliferation of PHA elicited human PBMCs [ Fig. S5C]. These results designate that IFG not only stimulates the macrophages, but also B cells and CD8 T cells.
IFG activated macrophages show enhancement in antigen uptake. Macrophages are extremely efficient APCs responsible for phagocytosis. Therefore, we were curious to check whether IFG stimulated macrophages acquire an enhanced capacity to phagocytose Mtb. Consequently, we stimulated macrophages with IFG and then infected them with Mtb. The extracellular bacteria were removed by extensive washings and subsequently treated with amikacin. Later, the cells were lysed and bacterial growth was enumerated by CFU plating. Remarkably, IFG significantly (p < 0.05) improved the ability of macrophages to phagocytose H37Ra, compared to unstimulated control [   . We also showed that there was downregulation in the autophagy marker LC3, when iNOS was inhibited by NM [ Fig. 4E]. Thus, establishing that iNOS activity is required for autophagy. Furthermore, autophagy inhibitor 3 methyl adenine (3MA) was used to unambiguously establish the induction of autophagy in Mtb infected Mϕ treated with IFG [ Fig. 4E]. Rapamycin was used as a positive control. Moreover, blocking of Akt-PI3K and NF-кB pathways by their respective inhibitors, substantially reduced the secretion of NO [ Fig. 4F].
Autophagy plays a fundamental role in inhibiting the intracellular survival of Mtb 37 . Interestingly, IFG stimulation exhibited higher conversion of LC3I to LC3II, a hallmark of autophagy. We observed further amplification of autophagy by rapamycin treatment of the infected macrophages stimulated with IFG [ Fig. 5A]. Thus corroborating the IFG induced autophagy. Additionally, blocking of Akt and PI3K pathways by their respective inhibitors inhibited the induction of autophagy [ Fig. 5B]. Furthermore, augmentation (p < 0.0001) of the autophagy genes ATG5, ATG7 and BECLIN1, categorically established the involvement of autophagy [ Fig. 5C-E]. Finally, we verified this finding by observing the over expression of autophagy influx by LC3 puncta formation in GFP-H37Ra infected macrophages through confocal microscopy after IFG and rapamycin treatment [ Fig. 5F,G]. We observed that virulent and avirulent mycobacteria have similar responses with respect to autophagy. Interestingly, we also observed that IFG itself has an ability to induce autophagy influx in macrophages (Fig. S7A,B). Likewise, the acidic vacuoles were stained with acridine orange dye to show the induction of autophagy [ Fig. 5H,I]. We also monitored autophagy by treating cells (infected and IFG stimulated) with rapamycin and correlated this with NO IFG induces the phosphorylation of STAT-3 and activates Akt-PI3K pathway. It has been well established that IFN-α /β interacts with its receptor IFNAR and activates tyrosine kinase Jak-1 along with infected macrophages (not stimulated with IFG); DIC: differential interference contrast. Image magnification: 60x. Data are representative of two independent experiments. *p < 0.05, **p < 0.003, ***p < 0.0004, ****p < 0.0001. Tyk-2 38,39 . Jak-1 stimulates and phosphorylates STAT-1/STAT-3. As a result, Akt is activated through STAT-3 along with PI3K and translocates into the nucleus. Further, it helps in delivering survival signal through the activation of NF-κ B. It has been reported that like IFN-α /β , IFG exhibits affinity for IFNAR and activates tyrosine kinase Jak-1 along with Tyk-2. Therefore, we investigated the mechanism involved in IFG-mediated signalling events responsible for activating macrophages to restrict the intracellular growth of Mtb. Mtb infected macrophages were stimulated with IFG and phosphorylation of STAT-3, PI3K and Akt molecules were monitored by Western blotting in the cytosolic extract. Noticeable augmentation in the phosphorylation of STAT-3, PI3K and Akt was observed in the Mϕ IFG [ Fig. 6A-C]. The decrease in the phosphorylation of STAT-3 and PI3K was checked in Mϕ -IFG by treating with their respective inhibitors [ Fig. 6A,B]. Further, augmentation in the

Discussion
There is an urgent need and a challenge to discover new drugs to treat TB because the disease continues to kill millions of people annually. Further, despite of the availability of potent drugs, the current regime remains quite complicated due to lengthy treatment, side effects and emergence of drug-resistant strains of Mtb.
Type 1 IFNs (IFN-α /β ) are produced by macrophages, plasmacytoid DCs, fibroblasts and endothelial cells and play crucial effector function in viral infections 34,40 . Undeniably, previous references have indicated the efficiency of the IFNs therapy in treating yellow fever, chronic hepatitis B (HBV), hepatitis C (HCV) and ebola virus infections 41,42 . They augment the synthesis of several key antiviral mediators, including 2′ -5′ oligoadenylate synthetase (2′ -5′ A synthetase) and protein kinase R 43,44 . Further, type 1 IFNs upregulate CXCL10 expression in Mtb infected dendritic cells, which could be important to induce the homing and recruitment of cells to the granulomas 45 .
Considering the importance of IFN-α /β in activating cells of the immune system and bolstering immunity, a synthetic interferon IFG-α 2b, which is known as Infergen (IFG) was synthesized [46][47][48] . Infergen has been reported to upregulate MHC-I expression and enhance the presentation of viral peptides by APCs. As a result, it enhances the activation and cytolytic activity of CD8 + cytotoxic T lymphocytes (CTLs) [49][50][51] . Further, like IFN-α /β , IFG exhibits an optimized affinity for IFNAR and augments the survival and activation of neutrophils, macrophages, We propose that pSTAT-3 binds to the promoter sequence of INOS gene to produce ions in the cytosol that generates NO, a potent antimicrobial agent that may ultimately restrict the growth of Mtb. The data were validated by blocking the Akt-PI3K, iNOS and autophagy with their respective inhibitors to establish the role of autophagy in restricting the growth of Mtb by IFG stimulated macrophages.
[7] Blocking of the Akt-PI3K, iNOS (NM), autophagy (3MA|CQ) with their specific inhibitors and therefore substantial reduction in the secretion of NO and autophagy, as well as increase in the survival of Mtb signifies that autophagy is directly involved in the killing of Mtb. Signifying that the function of iNOS and autophagy are interlinked. Overall, this mechanism illustrates that IFG plays a key role in controlling the survival of Mtb inside macrophages.
In the past, regular efforts have been made to investigate the impact of new molecules or delineate novel role of previously established molecules in the treatment of TB. Therefore, keeping in view the above-mentioned reports, we thought to evaluate the impact of IFG on the activation of macrophages and inhibition of the intracellular survival of Mtb. Following major findings have emerged from our study. IFG significantly i) inhibited the growth of Mtb in the macrophages; ii) augmented the secretion of IL-6 and IFN-γ ; iii) upregulated the expression of HLA-DR and costimulatory molecules CD80 and CD86; iv) enhanced phagocytic potential of macrophages; v) enhanced CD8 T cell mediated target lysis; vi) the mechanism elucidated in curbing the survival of Mtb was through the involvement of nitric oxide and autophagy; vii) signaling pathway induced by IFG was through phosphorylation of STAT-3, Akt and activation of NF-κ B.
For the optimum activation of T cells, two signals are required. First is the occupancy of T-cell receptor (TCR) by MHC-peptide complex. Second signal is also APCs driven in the form of costimulatory molecules like CD80 and CD86. In the absence of second signal, T cells undergo anergy, i.e. a state of unresponsiveness. Upregulation of HLA-DR, CD80 and CD86 by IFG indicate its potential role in the optimal activation of T cells. Further, it will help in counteracting Mtb-mediated downregulation of costimulatory molecules and thereby avoiding T cell anergy 23 . IFG predominantly promotes the generation of Th1 cells and CD8 T cells 32,[49][50][51] . Both these cells are responsible for cell mediated immunity and play a cardinal role in protection against Mtb. CD8 T cells lyse the infected cells and release the intracellular Mtb. The IFN-γ secreted by Th1 cells activate macrophages and consequently they phagocytose and eliminate the bacterium. It was an exciting observation that IFG elicited macrophages showed enhanced phagocytosis and killing of Mtb. These results reveal a novel role of IFG against Mtb.
IFN-α /β interacts with IFNAR receptor and induces the activation of Jak-1 and Tyk-2. Jak-1 stimulates and phosphorylates STAT-3 38,39 . Consequently, Akt-PI3K is activated through STAT-3 and translocates into the nucleus. Further, Akt-PI3K-STAT-3 signaling helps in delivering survival signal through the activation of NF-κ B 38,39 . Like IFN-α /β , IFG [(IFN)-α 2b] also exhibits a specific affinity for IFNAR and activates tyrosine kinase Jak-1 along with Tyk-2 38,39,52,53 . It is also important to mention that signaling through IL-6 also follows the gp130/ Jak/STAT pathway 54 . We investigated the signaling mechanism mediated by IFG in activating macrophages to restrict the intracellular survival of Mtb. We observed that Mtb infected macrophages when activated with IFG showed release of IL-6, but no noticeable change was seen in IL-1β, TGF-β, IL-10, TNF-α genes except IL-6 gene expression and subsequent phosphorylation of STAT-3 and Akt but not STAT-1. This indicates that IL-6 pathway is involved in IFG signaling. Phosphorylation of STAT-3 and Akt-PI3K delivers a survival signal to macrophages through upregulation of Bcl-xL and activation of NF-κ B, thereby ensuring cell survival, activation and induction of autophagy. We speculate that STAT-3 binds to a promoter sequence of INOS gene, enhances the synthesis of iNOS and therefore production of NO that ultimately leads to the killing of Mtb [Fig. 7]. The data were validated by blocking the Akt-PI3K, iNOS and autophagy by their respective inhibitors to establish the role of autophagy in restricting the growth of Mtb by IFG stimulated macrophages. Blocking of Akt-PI3K, iNOS and autophagy, and substantial reduction in the secretion of NO and autophagy, as well as increase in the survival of Mtb signifies that autophagy is directly involved in the killing of Mtb. Further, the function of iNOS and autophagy are interlinked with each other. This provides the unambiguous evidence of a possible mechanism operating in constraining the Mtb growth. It has been reported that NO suppresses mTOR expression, which leads to autophagy 55 . Corroborating with this observation, we hypothesize that NO directly suppresses the expression of mTOR, as shown by the enhancement of NO release by mTOR inhibitor rapamycin (autophagy inducer), which led to the induction of autophagy in Mtb infected and IFG stimulated human macrophages. In essence, this study provides a novel immunomodulatory and therapeutic role of IFG against intracellular pathogen Mtb, which may have enough potential in future to treat TB patients. Mycobacterial uptake by IFG stimulated macrophages. THP-1 monocytes (3 × 10 5 /well) were differentiated with PMA (25 ng/ml) for 16 h, followed by resting for additional 16 h in 24 well plate. Later, macrophages were stimulated with IFG (64 ng/ml) for 24 h in RPMI-FBS-10% at 37 °C/5% CO 2 . After 24 h, IFG stimulated cells were infected with Mtb strains H37Rv or H37Ra at MOI (1:5) for 4 h (3 h for M. smegmatis). This was followed by incubation with amikacin (2 μ g/ml) for 1 h to kill extracellular bacteria. After 1 h, cells were lysed with saponin (0.1%) and lysate (100 μ l) was plated on 7H11 agar plates. Bacterial colonies were enumerated on 21d. All subsequent experiments were performed with strain H37Rv (Mtb) or otherwise indicated.

Phagocytosis of GFP + Mtb by IFG stimulated macrophages by confocal microscopy.
Macrophages (3 × 10 5 /well) were stimulated as mentioned above with IFG for 24 h. Later, cells were infected with GFP + Mtb (H37Ra) [MOI:1:10] for 4 h. The amikacin (2 μ g/ml) treatment was given for 1 h to kill the extracellular bacteria, followed by extensive washing with cold PBS. Subsequently, cells were fixed with paraformaldehyde (1%) for 10 min. These cells were washed and placed in poly-L lysine pre-coated cover slips and imaged using Nikon A1 confocal laser microscope (Nikon, Tokyo, Japan). Z-stacks and extent of GFP + Mtb internalization were monitored among experimental and control samples. The analysis of the intensity of surface plots was performed using Nikon NIS-AR 4.1 image analysis software (Nikon, Melville, NY). The confocal results were further validated by flow cytometry and analysis was done by FACS Suite software (BD Biosciences, San Jose).

Intracellular killing of Mtb by IFG stimulated macrophages. Macrophages were infected with Mtb
(H37Rv or H37Ra) at MOI (1:10) for 4 h in 24 w plates, followed by extensive washings with PBS (1X) to remove extracellular bacteria. Later, cells were cultured in antibiotic free RPMI + FBS-10% along with IFG (0-64 ng/ml) for 24 h. Amikacin (2 μ g/ml) was added to the cultures to kill the extracellular Mtb. After 24 h, cells were lysed with saponin (0.1%) and lysate (100 μ l) was plated on 7H11 agar plate. Colonies were enumerated on 21d.
Cell culture and estimation of cytokines. THP-1 macrophages (2.5 × 10 5 /w) and macrophages (2.5 × 10 5 /w) isolated from the human PBMCs by adhering on tissue culture grade petri plates for 4-6 h were cultured with IFG (0-64 ng/ml) in 48 w plates for 48 h. The control cells were cultured in medium alone. The supernatants were collected after 24 h and IL-6 and IFN-γ were estimated. Briefly, 96 w plates were coated O.N/4 °C with Abs to human IL-6 (2 μ g/ml) and IFN-γ (2 μ g/ml) in phosphate buffer (pH 9.2, 0.01M). The unbound sites were blocked with BSA (1%) in PBS (1X) for 3 h/RT. Subsequently, respective recombinant cytokines as standard, and culture supernatants (50 μ l volume) were added and incubated ON/4 °C. Later, biotinylated anti-human IL-6 (2 μ g/ml) and IFN-γ (2 μ g/ml) Abs diluted (1:1) in PBS-Tween buffer were added into the plates and incubated for 2 h/RT. Next, avidin-HRP (1:10,000) was added and incubated at 37 °C/45 min. The regular procedures of washing and incubation were followed at each step. The color was developed using H 2 O 2 -OPD substrate-chromogen. The reaction was stopped by H 2 SO 4 (7%) after 20 min. The plates were read at 492 nm. The results expressed as pg/ml of secreted cytokines were calculated using serial log 2 dilutions of standard curve drawn with rIL-6 and rIFN-γ .
Viability of IFG treated macrophages. Macrophages were stimulated with IFG (0-64 ng/ml) for 24 h at 37 °C in RPMI-FBS-10% (500 μ l) in 48 w plates. The stimulated cells were harvested, washed and resuspended in PBS and stained with 5 μ l/tube annexin V and 2.5 μ l/tube of propidium iodide (50 mg/ml). The cells were incubated in the dark for 15 min at RT. Later, PBS (200 μ l) was added to the cells and acquired immediately using BD FACS Calibur flow cytometer and analysis was done by DIVA software. IL-6, ATG5, ATG7, BECLIN-1, INOS, IL-1β, IL-10, TGF-β and TNF-α. Total RNA was isolated using trizol reagent from IFG (64 ng/ml) stimulated macrophages for 6 h, according to the manufacturer's instruction (Invitrogen, Carisbad, CA). Briefly, RNA was quantified with the help of NanoDrop spectrophotometer. A260/A280 ratio of all the samples was in the range of 1.90 to 2.00. DNA contamination from RNA samples was removed by amplification grade DNase. RNA samples (3 μ g) were incubated with DNase (3U) for 15 min in the reaction buffer. Later, DNase activity was terminated by adding stop solution. Further, the samples were heated at 70 °C/10 min to inactivate DNase activity. Analysis was done by comparative Ct method, whereas Ct values were normalized against house-keeping control actin. The results are represented as relative expression (fold change). The fold change was calculated considering the value of uninfected/unstimulated controls as 1. RT-qPCR and data analysis were done by Step One Plus RT-qPCR system (Applied Biosystems, Chromos, Singapore). The RT-qPCR primer sequences are as follows:  Cells stimulated with LPS (2 μ g/ml) were used as a positive control. These cells were then harvested, washed, and lysed in lysis buffer (Cytoplasmic protein extraction buffer, phosphatase and protease inhibitor cocktail). Later, protein was estimated in the lysate and SDS-PAGE was performed. After transfer to PVDF membranes and subsequent blocking, the membranes were immunoblotted using Abs to phosphorylated and non-phosphorylated form of STAT-3, Akt, STAT-1 and PI3K. Blots were also incubated with Abs against LC3B, Bcl-xL, iNOS and loading control actin. Blots were developed using a chemiluminescence kit (Amersham Pharmacia Biotech, Buckinghamshire, UK) (ECL; Pharmacia-Amersham, Freiburg, Germany). Scanning of the blots was done by ImageQuant LAS 4000 (GE Healthcare, Freiburg, Germany). Image analysis was performed with MultiGuage image and ImageJ analysis softwares.

Real time PCR for the quantification of
Scientific RepoRts | 6:39492 | DOI: 10.1038/srep39492 Quantification of autophagy influx by confocal microscopy. The THP-1 macrophages were infected with GFP-H37Ra for 4 h and stimulated with IFG (64 ng/ml) and rapamycin [1 μ M] for another 4 h. Mtb-Mϕ -IFG were fixed on poly-L-lysine coated cover slips for 15 min. Later, cells were fixed with paraformaldehyde (2x) for 10 min followed by treatment with triton X100 (0.1%) for 3 min. The non-specific sites on cells were blocked with BSA (2%) for 2 h. This was followed by incubation with rabbit anti-human LC3 Ab for 2 h. Subsequently, cells were incubated with Alexa Fluor 633-anti-rabbit Ab for 1 h and followed by staining with DAPI dye. Regular washings were followed at each and every step. The cells were imaged through confocal microscopy. Rapamycin was used as a positive control. Number of puncta formation was counted manually from 4-5 different microscopic field. Percent of cells with LC3 puncta were presented in bar diagram.
Demonstration of NF-κB by EMSA. Macrophages were cultured with IFG for 30 min. Later, the cells were harvested and nuclear extract was prepared for EMSA. Briefly, an equal amount of nuclear extract (3 μ g) from each sample was incubated for 20 min/37 °C in water bath with [ 32 P] end labeled duplex oligonucleotides containing binding site for NF-κ B. The DNA-protein complexes were resolved by electrophoresis on a native PAGE-gel (7%). After electrophoresis, the native gel was dried and exposed to screen at RT for 6-10 h and scanned by phospho-imager (Fujifilm, Tokyo, Japan).
The translocation of NF-κB by confocal microscopy. Mϕ IFG were fixed on poly-L-lysine (0.1%) coated cover slips for 15 min. Later, cells were fixed with paraformaldehyde (2x) for 10 min followed by treatment with triton X100 (0.1%) for 3 min. The non-specific sites were blocked with BSA (2%) for 2 h. It was followed by incubation with mouse anti-human NF-κ B p65 Ab for 2 h. Subsequently, cells were incubated with FITC-anti-mouse Ab for 1 h and followed by staining with Hoechst dye. Regular washings were followed at each step. The cells were imaged through Nikon A1 confocal laser microscope (Nikon, Tokyo, Japan) and analysis was performed using Nikon NIS-AR 4.1 image analysis software (Nikon, Melville, NY).

Statistics.
The data analysis was done by Student's 't' test, non-parametric Mann-Whitney two tailed and repeated measure ANOVA with post Student-Newman-Keuls multiple comparison test using Graph Pad InStat 3 and Graph Pad Prism 6 softwares.