An Accord of Nuclear Receptor Expression in M. tuberculosis Infected Macrophages and Dendritic Cells

Mycobacterium tuberculosis instigates interactions with host factors to promote its survival within the host inimical conditions. Among such factors, nuclear receptors (NRs) seem to be promising candidates owing to their role in bacterial pathogenesis. However, only few members of NR superfamily have been implicated in M. tuberculosis infection and there is a dearth of comprehensive knowledge about expression or function of the entire superfamily. In this study, we performed detailed expression analysis and identified key NRs getting differentially regulated in murine macrophages and dendritic cells (DC) upon infection with H37Rv. The murine macrophages and DCs infected with H37Rv entailed overlapping changes in the expression of certain NRs which reflect upon the possibility that both cells might utilize similar transcriptional programs upon M. tuberculosis infection. We identified Nr4a3 and Rora, which have not been implicated in M. tuberculosis pathogenesis, undergo similar changes in expression in macrophages and DCs upon H37Rv infection. Interestingly, a similar pattern in their expression was also observed in infected human monocyte derived macrophages and the findings corroborated well with PBMCs obtained from TB patients. This all-inclusive analysis provides the basis for a precise approach in identifying NRs that can be targeted therapeutically in intracellular bacterial infections.

The cells of innate immune system particularly macrophages and dendritic cells (DCs) provide the first line of defence against Mycobacterium tuberculosis and they are among the first few cells to interact with the bacteria upon infection in the host. Macrophages and DCs utilize a wide variety of surface receptors including toll-like receptors, opsonic receptors, mannose receptors, etc. that recognize antigenic molecules on the bacterial surfaces [1][2][3][4][5] . Engagement of bacterial pathogen-associated molecular patterns (PAMPs) with these host cell receptors results in the bacterial uptake and entrapment of bacteria in the phagosomes. The members of nuclear receptor (NR) superfamily not only regulate the signalling through such pathogen recognition receptors but have also been shown to sense the lipid components of mycobacteria by their direct binding to such lipids 6,7 . The NR superfamily that comprises a total of 49 members has been categorized into three families namely, endocrine, adopted orphan and orphan NRs on the basis of the nature of the ligand they bind [8][9][10][11] .
A growing literature about NRs in infectious diseases highlights that they are important in shaping up the innate immune responses and several pathogens, for their own benefits, have evolved strategies to modulate NRs either by changing their expression or by interfering with their transcriptional activity. For example, Listeria monocytogenes and M. tuberculosis induce the expression of peroxisome proliferator-activated receptor gamma (Pparg) in macrophages [12][13][14] . Expression of Pparg is also elevated in response to a PAMP called mannose-capped lipoarabinomannan obtained from the cell wall of M. tuberculosis 15 . Interestingly, silencing of Pparg in macrophages significantly reduces the bacterial burden supporting the pro-bacterial role for Pparg 12 . In addition to Pparg, another NR named testicular receptor 4 (Tr4) also acts as a molecular switch and by regulating the macrophage polarization towards M2 phenotype, augments the bacterial survival 12,16 . The infection with intracellular bacteria like M. tuberculosis, L. monocytogenes and Shigella flexneri also elevates the levels of liver X receptor alpha (Lxra) in macrophages 17,18 . Moreover, infected macrophages have elevated levels of Lxra target genes suggesting that transcriptional programs induced by Lxra promote the clearance of bacteria inside the macrophages 17 . In agreement with these findings, Lxra deficient mice showed increased susceptibility to bacterial infections 17,18 . Also, along with Lxr the binding sites of retinoic acid receptor (Rar) are enriched in H3K4me1 regions (a marker for active or poised enhancers) in M. tuberculosis infected THP-1 cells 19 . In accordance with these findings, studies performed on human DCs infected with M. tuberculosis also revealed enrichment of Lxr and Rar binding sites 20 . Furthermore, the transcriptomic analysis of blood samples from active pulmonary TB patients revealed an altered expression in vitamin D receptor (Vdr) associated genes compared to the individuals with latent bacterial infection. Vdr-regulated genes are also enriched in M. tuberculosis infected THP-1 cell line 21,22 . Many other members of NR superfamily which includes glucocorticoid receptor (Gr), pregnane X receptor (Pxr), Nr1d1, and Vdr have also been shown to be involved in the pathogenesis of bacterial infection [23][24][25][26][27][28] . Furthermore, NRs are amicable to modulation by small molecules (which are either available or can be synthesized easily) and hence numerous endogenous and synthetic NR ligands have been evaluated in various infection models as potential therapies 29 .
M. tuberculosis is unique in its ability as it successfully parasitizes both macrophages and DCs, which in traditional view are considered as host sentinels for initiating protective immunity against bacterial infections. With the emergence of knowledge about the cross-talk between host NRs and M. tuberculosis, it is compelling to identify all those NRs which have a bearing on the persistence and pathogenesis of mycobacteria in macrophages and DCs. A composite gene expression analysis of NRs performed in macrophages and DCs has suggested a regulatory role of NR superfamily in macrophages and DC functionality 8,30 . However, these studies have examined the expression of these NRs only in the presence of a PAMP or an immunological modulator and we still need to unravel a composite list of all NRs being modulated by mycobacteria in its innate immune cell niches. In this study, we performed a detailed expression profiling of NR superfamily in murine macrophages and DCs upon M. tuberculosis infection. In conjunction with identifying NRs whose role in M. tuberculosis pathogenesis has not yet been established we also recapitulated some of the findings which have been previously published. Intriguingly, we observed that NRs such as Vdr, Lxra, Pparg, Rarg, Nr4a's, RAR-related orphan receptor alpha (Rora), androgen receptor (Ar), and thyroid hormone receptor beta (Thrb) were undergoing differential regulation in both macrophages and DCs upon M. tuberculosis infection suggesting a plausible cross-talk between these NRs and M. tuberculosis. Interestingly, our analysis also revealed Rora and Nr4a3, both of which have not been previously implicated in TB infection, undergo changes in expression upon M. tuberculosis infection. We identified that H37Rv induces the expression of Rora while the expression of Nr4a3 was significantly reduced in murine and human macrophages. We also observed a similar pattern in the expression of Rora and Nr4a3 in murine DCs and PBMCs obtained from patients with active TB. Therefore, examining the potential of ligands for such NRs as an approach to anti-TB therapy could be promising.

Identification of NRs modulated by M. tuberculosis H37RV in mBMDCs.
Being an intracellular pathogen, M. tuberculosis resides in cell types such as macrophages and DCs. Among these, macrophages have been largely the focus for studying cross-talk of M. tuberculosis with the host. However, given the unique ability of DCs to initiate T cell responses, DCs could possibly regulate the adaptive immune responses against the bacteria. Recently DCs have been shown to express certain repertoire of NRs, few of which have also been shown to regulate DC functions 8,31,32 . So, in order to address whether this pathogen has the ability to modulate any of these NRs in DCs and thereby DC function, in this study, we infected mouse bone marrow-derived DCs (mBMDCs) with H37Rv and monitored the expression of NRs relative to uninfected control cells. We selected 0 h, 12 h, and 48 h post infection as the time points which allowed us to identify the changes in the transcript levels of NRs during early and later stages of infection. Interestingly, we found that the levels of some NRs were being altered upon infection (Fig. 1A). Among the endocrine NRs, Ar, Rarg, Thrb, and Vdr underwent significant downregulation of greater than 2-folds (Fig. 1B). While Ar and Vdr remained constantly repressed during later time points of infection, the expression of Rarg and Thrb started to rescue at 48 h post infection. The adopted orphan NRs, Lxra and Pparg were downregulated at 12 h (Fig. 1C) followed by an enhanced expression of Lxra and recovery of Pparg's basal expression at 48 h after infection. In the orphan NR family, all three members of Nurr family (Nr4a1, Nr4a2 and Nr4a3) exhibited repression while Rora was upregulated during later stages of infection (Fig. 1D). Taken together, our data identifies the members of NR superfamily which are being significantly modulated in mBMDCs upon M. tuberculosis infection. Of these, Vdr, Lxra and Pparg have already been shown to have an explicit role in M. tuberculosis pathogenesis thereby indicating plausible implication of other altered NRs in M. tuberculosis infection.

Expression analysis of NRs in mBMDMs upon M. tuberculosis infection. Given the modulation
of NRs with an unidentified role in M. tuberculosis infection, in H37Rv infected mBMDCs, we next set out to profile entire NR superfamily in H37Rv infected mouse bone marrow-derived macrophages (mBMDMs) to further explore the members with a possible role in M. tuberculosis pathogenesis. A similar kind of analysis revealed that unlike mBMDCs, in mBMDMs certain NRs started responding during the early time point of infection (0 h). These included Esr1, Gr, Rar's, Thrb, Vdr, Pparg, Rxrb, Nr1d's, Tr4, Nr4a1, and Nr4a3 ( Fig. 2). At 48 h post infection, when the equilibrium between bacteria and macrophages was established, few NRs that did not respond during early infection (0 h) showed elevated levels (Ppard, Rxra, Nr4a2, and Rora). Furthermore, this analysis reflected that while some of the early responding NRs such as Esr1 and Nr1d2 had a reverse pattern of their expression at 48 h, others simply recovered their basal levels (Nr1d1, Tr4, Nr4a1, and Nr4a3). Thus, M. tuberculosis infection affects the levels of different NRs in its innate immune reservoirs, but whether these induced changes in an individual receptor favours pathogen or host needs to be established.

NRs commonly modulated in mBMDCs and mBMDMs by M. tuberculosis. Next, to additionally
identify the key members of NR superfamily which might be crucial for M. tuberculosis pathogenesis we followed a systematic approach. Intriguingly, further in depth analysis revealed that M. tuberculosis was altering certain repertoire of NRs which was shared by both mBMDCs and mBMDMs. We identified that among the 49 members of the NR superfamily there were 10 NRs which were being modulated in both cell types upon M. tuberculosis infection (Fig. 3). Interestingly, six of them showed a similar pattern of regulation of their expression in mBMDCs and mBMDMs ( Fig. 3A-F). Of these, Vdr, Lxra and Pparg have already been reported to have an implication in the tuberculosis disease. Whereas the role of others, which includes Rarg, Nr4a3 and Rora, has not yet been identified and remains to be established in M. tuberculosis infection. On the contrary, Ar (Fig. 3G), Thrb (Fig. 3H), Nr4a2 (Fig. 3I) and Nr4a1 (Fig. 3J) exhibited upregulation in mBMDMs whereas their expression was downregulated in mBMDCs. The probable explanation for this could be the cell specific responses against mycobacteria in the studied cell types. As shown in Fig. 3K, Rora and Nr4a3 have similar while Ar, Thrb and Nr4a2 have different expression pattern in mBMDCs and mBMDMs. In corroboration, we observed similar modulation of Rora and Nr4a3 at protein levels as well (Fig. 3L). Taken together, we speculated that these commonly modulated NRs could prove to be novel target molecules for developing drugs to control M. tuberculosis infection.

Expression of NRs in CD11c positive cells isolated from lungs of M. tuberculosis infected mice.
To determine whether above identified commonly regulated NRs were also significant in an in vivo setting, we tested their levels in lungs isolated from the most commonly used mice model of M. tuberculosis infection. We selected day 7, 15 and 30 post aerosol infection for our analysis in order to study their kinetics over the early and later stages of infection. CD11c positive cells were isolated from the lungs of these mice which represented a cellular population consisting of alveolar macrophages and DCs. Interestingly, alteration in the expression of Nr4a3 (Fig. 4A) and Rora (Fig. 4B) was in concordance with our in vitro findings. In addition, Ar (Fig. 4C), Thrb (Fig. 4D) and Lxra (Fig. 4E) reflected an expression pattern which was similar to mBMDCs, while Nr4a2 (Fig. 4F) exhibited that of mBMDMs.

Expression of Nr4a3 and Rora in hMDMs and clinical samples.
To further extend our in vitro findings we monitored the expression of Nr4a3 and Rora in human monocyte derived macrophages (hMDMs) (obtained from healthy individuals). The hMDMs were infected with H37Rv and the levels of Nr4a3 and Rora were measured at different time points (0, 12, and 48 h) (Fig. 5A). Consistent with our in vitro findings, we In addition, we also monitored relative mRNA abundance of Nr4a3 and Rora in human PBMCs isolated from freshly diagnosed TB patients (Fig. 5B). We observed a relatively lesser expression of Nr4a3 and higher expression of Rora in PBMCs isolated from TB patients compared to PBMCs isolated from healthy individuals. Altogether, these findings suggest that Nr4a3 and Rora might have an important role in M. tuberculosis pathogenesis.

Discussion
M. tuberculosis gets transmitted via aerosol thereby lodging the lungs where it inhabits cells of different phenotypes as macrophages, DCs, neutrophils, monocytes, endothelial and epithelial cells. Among these, the innate immune cells are the principal cell type harbouring the bacteria inside the host body. The scientific challenges in understanding the immunology of M. tuberculosis arise from the observation that an apparently appropriate immune response is developed upon infection however, it generally leads to the establishment of a latent infection rather than elimination of pathogen 33,34 . In addition, M. tuberculosis outsmarts host defence mechanisms, manipulates host signalling mechanisms and overturns them for its own survival within the host 35 . The pathogen has a unique repertoire of cell wall lipids which have been shown to possess immunomodulatory properties and are important for its pathogenesis 36 . Earlier studies have shown that these lipids help the bacteria to invade the host cells and in subverting immune responses by modulating phagosome maturation, cytokine response, and antigen presentation 37 . In addition, M. tuberculosis actively sheds its lipids in the host cell which then enter the endocytic network and are trafficked within as well as outside the infected cell [38][39][40][41] . It has been proposed that microvesicles released in the extracellular environment by the infected host cells are taken up by the uninfected bystander cells including macrophages and DCs 39,40 . Among these, DCs in particular can migrate from the lungs to the local draining lymph nodes and this trafficking might result in the leaching of mycobacterial antigens from their primary site of infection to the distant sites. Thus, providing the mechanism for dissemination of mycobacterial components and hence establishing a communicating network that may have whole body manifestation. The cell integrates the lipid signalling pathways via lipid sensors that include membrane receptors and certain lipid sensing NRs such as Tr4 16 . Owing to the fact that these NRs are amenable to pharmacological modulation makes them a good therapeutic drug target to combat TB. However, we still lack an in-depth knowledge of how M. tuberculosis manipulates the host NRs for its survival and pathogenesis.
In this study, we monitored and determined the levels of NRs in macrophages and DCs upon M. tuberculosis infection to elucidate the cross-talk between the pathogen and host NRs. We chose different time points for infection to mimic the conditions where early on the host cells are either trying to eliminate the infection or later time points where the pathogen might have established a persistent infection by exploiting the host intracellular machinery. It became apparent from our findings that M. tuberculosis infection leads to an alteration in the levels of several members of NR superfamily in innate immune cells. Interestingly, few of the receptors (Vdr, Lxra, Pparg, Rarg, Rora, Nr4a's, Ar, and Thrb) were commonly modulated in the studied cell types. Among these Lxra has been previously established to actively contribute towards the protective immune response against intracellular pathogen M. tuberculosis 17 . The epigenetic landscapes identified in M. tuberculosis infected human macrophages and human DCs also signifies the importance of Lxr in shaping the bactericidal response against intracellular bacteria 19,20 . In addition, 1,25-dihydroxyvitamin D that acts as a ligand for Vdr also aids in the better clearance of bacteria [25][26][27] . Vdr is also identified as a significant transcriptional modulator of the genes that are upregulated in human macrophages in response to M. tuberculosis infection 22 . Moreover, vitamin D deficiency has been correlated with increased TB susceptibility. On the contrary, Pparg increases the persistence of bacteria inside the host cells and our recent findings demonstrated that it does so by driving the alternative activation of macrophages thereby promoting the fitness of M. tuberculosis 12,15,29 . Furthermore, rosiglitazone, a ligand to Pparg, enhances the bacterial survival inside the macrophages 12 . In concordance with earlier studies, we found a dynamic change in the expression of Pparg upon M. tuberculosis infection. Interestingly, this study also identified changes in the expression of Nr4a3 and Rora, whose functions remained unknown in M. tuberculosis infection. While the expression of Nr4a3 was significantly decreased in macrophages and DCs, the expression of Rora was found to elevate in the infected cells. The PBMCs obtained from TB patients also showed significant changes in the expression of Nr4a3 and Rora pointing towards their plausible role in bacterial infection. Hence, it is conceivable that the dynamics of M. tuberculosis infection might be regulating the expression of those NRs that either enhance successful intracellular parasitisation by bacteria or are responsible for its clearance. Therefore, it will be interesting to decipher the function of other differentially regulated NRs with no reported role in M. tuberculosis pathogenesis.
On the other hand, we observed that few NRs are constitutively expressed at high levels without undergoing any significant changes upon M. tuberculosis infection but have a reported role in its pathogenesis such as Tr4. Tr4 acts as a molecular switch and by regulating the macrophage polarization towards M2 phenotype, augments the bacterial survival 12 . In a recent published study, it has been elucidated that the lipid components of mycobacterial cell wall binds to Tr4 16 . This leads to the transactivation of Tr4 thereby inducing foamy biogenesis and granuloma formation. It has now been appreciated that sometimes rather than the NR expression it is the changing repertoire of the host cellular lipids, which act as ligands for these NRs, that regulates their function. This led us to speculate that NRs with high expression might also be having a plausible role in the clearance or persistence of M. tuberculosis. However, detailed functional studies should be performed for validating their roles in M. tuberculosis infection. It has been revealed that Pxr, an adopted orphan NR, increases the survival of M. tuberculosis in its host niche (human macrophages) 23 . But in corroboration with other reports, we found insignificant expression of this NR in studied murine cell types 8,30 .
Given the fact that the current TB treatment is complex and often results in poor patient compliance, there is a compelling need for new drug regimens which are more effective with shorter treatment period. Most drugs developed till date to cure mycobacterial infections target crucial enzymatic processes in the bacterium such as transcription, translation or biosynthetic pathways like cell wall synthesis 42 . In addition to developing new novel TB-drugs that are directed against bacterial processes, there is also an increasing interest in developing therapeutics that are not directly bactericidal but target host pathways used by bacteria for either infecting, persisting or replicating inside the host 43,44 . In this connection, the findings that NRs modulate several cellular pathways that are important for bacterial survival inside the cell and their amenability to pharmacological modulation makes it interesting to explore the possibility of utilizing NRs as targets for anti-TB therapies. Adding small molecules that can regulate the activity of NRs in ways that strengthen the microbicidal properties of macrophages and DCs along with standard TB-drugs may significantly reduce the duration of TB treatment. Moreover, this adjunct therapy may also be effective in providing treatment against latent bacteria which is otherwise resistant to the anti-TB drugs.
Collectively, our study suggests the existence of a complex interplay between host NRs and M. tuberculosis. These NRs may provide immense opportunities as potential therapeutic targets that can regulate immune responses thus directing protective immunity against infectious diseases.

Materials and Methods
Human ethics statement. The  Mycobacterial strain and culture conditions. M. tuberculosis H37Rv was cultured in 7H9 broth medium containing 5% glycerol and 0.05% Tween 80. The broth was supplemented with 10% OADC and the culture was incubated at 37 °C on a shaker. Bacterial cultures in log phase were used for infection. mBMDCs and mBMDMs culture. For the generation of mBMDCs, bone marrow precursors were obtained from the femur and tibia of C57BL/6 mice and treated with RBC lysis buffer. The precursors were then cultured in ultra-low attachment plates (Corning Costar) in RPMI-1640 medium (Gibco) supplemented with 10% new born calf serum (Gibco), 1% Penicillin Streptomycin (Pen Strep; Gibco), 10 ng/ml of GM-CSF (eBioscience) and 10 ng/ml of IL-4 (eBioscience), for 7 days. Non-adherent cells were harvested on day 7 and they exhibit ~90% purity as determined by the expression of CD11c. mBMDMs were derived by culturing bone marrow precursors in RPMI-1640 medium supplemented with 10% new born calf serum, 1% Penicillin Streptomycin and 50 ng/ml of GM-CSF. The non-adherent cells were replated on day 3 and kept for another 4 days. On day 7, adherent cells were harvested and they present ~98% as determined by analyzing the expression of CD11b and F4/80 by flow cytometry. Immunoblot analysis. Whole cell lysates were prepared and standard procedures were followed for immunoblotting. Briefly, equal amounts of total proteins were separated by SDS-PAGE on a 10% acrylamide gel and then transferred onto polyvinylidene difluoride membranes. The membranes were then blocked with 5% BSA for 1 h at room temperature, followed by overnight incubation at 4 °C with primary antibodies against Nr4a3 (AV45646, Sigma-Aldrich), Rora (sc-28612, Santa Cruz), and β-actin (sc-47778, Santa Cruz). Membranes were then washed three times with 1× PBST (pH 7.4) and incubated with appropriate horseradish peroxidase (HRP)-conjugated secondary antibodies for an hour at room temperature and developed with chemiluminescent HRP substrate. PCR array. The isolation of total RNA was performed from cells using an RNeasy Mini Kit (Qiagen). Using an RT 2 First Strand Kit (Qiagen), 1 µg of total RNA was reverse transcribed into cDNA which was then subjected to RT-qPCR using RT 2 SYBR Green ROX qPCR Mastermix (Qiagen). Customized RT 2 Profile PCR Array (SA Bioscience) was used to determine the expression of 49 murine NRs as per manufacturer's instructions. The QC for the RT-qPCR arrays was performed using a RT 2 Profiler PCR Array Data Analysis version 3.5 software. The expression data which successfully passed the QC were further analysed to determine the relative fold regulation by using 2 −ΔΔCt method. The NRs with a Ct value greater than 32 were excluded from final analysis.

Quantitative real-time PCR. Total RNA was obtained from cells by extraction with RNeasy Mini Kit
(Qiagen) and subjected to cDNA synthesis with the Verso cDNA kit (Thermo Scientific). RT-qPCR was performed using the SYBR Green method (DyNAmoColorFlash SYBR Green qPCR Kit, Thermo Scientific). The analysis of the expressed transcripts was performed by employing 2 −ΔΔCt method.

Statistical analysis.
Results are represented as mean and sd. Statistical analysis was performed in Prism 7 (GraphPad). Statistical significance with multiple parameters was analyzed with one-way ANOVA followed by post-hoc Tukey's test. The correlation analysis of NR expression between different time points has been done by using JMP11. Comparison of all pairs was performed using Tukey-Kramer HSD.