Mycobacterium tuberculosis LprE enhances bacterial persistence by inhibiting cathelicidin and autophagy in macrophages

Mycobacterium tuberculosis(Mtb) lipoproteins are known to facilitate bacterial survival by manipulating the host immune responses. Here, we have characterized a novel Mtb lipoprotein LprE(LprEMtb), and demonstrated its role in mycobacterial survival. LprEMtb acts by down-regulating the expression of cathelicidin, Cyp27B1, VDR and p38-MAPK via TLR-2 signaling pathway. Deletion of lprEMtb resulted in induction of cathelicidin and decreased survival in the host. Interestingly, LprEMtb was also found to inhibit autophagy mechanism to dampen host immune response. Episomal expression of LprEMtb in non-pathogenic Mycobacterium smegmatis(Msm) increased bacillary persistence by down-regulating the expression of cathelicidin and autophagy, while deletion of LprEMtb orthologue in Msm, had no effect on cathelicidin and autophagy expression. Moreover, LprEMtb blocked phago-lysosome fusion by suppressing the expression of EEA1, Rab7 and LAMP-1 endosomal markers by down-regulating IL-12 and IL-22 cytokines. Our results indicate that LprEMtb plays an important role in mycobacterial pathogenesis in the context of innate immunity.


INTRODUCTION
3 Tuberculosis (TB), caused by Mycobacterium tuberculosis (Mtb), is a widespread disease that kills more than 1.5 million people every year worldwide [1]. Macrophages, the primary resident cells of Mtb, are important components of innate defense mechanisms of the host.
Macrophages play a crucial role in recognition, phagocytosis, and killing of invaders. Nonpathogenic mycobacteria such as M. smegmatis (Msm) are readily killed by the macrophages, whereas pathogenic mycobacteria (Mtb) employ various strategies to survive inside the macrophages like prevention of phago-lysosome fusion [2], autophagy inhibition [3], modulation of host cytokine production [4], inhibition of reactive oxygen and nitrogen species [5] and the manipulation of antigen presentation to prevent or alter the quality of Tcell responses [6].
Among the key effector molecules responsible for bacterial killing are antimicrobial peptides such as cathelicidin and defensins that are expressed in different cells such as neutrophils, macrophages, monocytes and epithelial cells [7]. In contrast to the multiple defensins, only one cathelicidin gene, CAMP (cathelicidin antimicrobial peptide), has been found in humans [8,9]. The gene product human cationic antimicrobial peptide- 18 (hCAP18) is transcribed from the CAMP gene that contains vitamin D response elements (VDRE) in its promoter and requires enzymatic digestion by human neutrophil proteinase 3 (PR3) to produce mature LL-37 that exhibit broad antimicrobial activity [10]. LL-37 is an amphipathic -helical peptide that binds to negatively charged groups of the bacterial outer membrane causing disruption of the bacterial cell wall [11]. Previously, we and other studies have shown that LL-37 is able to kill both non-pathogenic and pathogenic mycobacteria both in vitro and in vivo conditions [12,13,14].
Bacterial lipoproteins are known to perform diverse functions such as host cell adhesion, cell invasion, nutrient transport, drug resistance and evasion of host defense mechanisms [18]. For example, Streptococcus pneumoniae PsaA lipoprotein is involved in the colonization of host cells and antibiotic resistance [19,20]. Borrelia burgdorfori VlsE lipoprotein is involved in bacterial persistence in host cells [21] and Haemophilus influenzae P6 lipoprotein activate host immune cells by inducing the secretion of pro-inflammatory cytokines through TLR-2 binding [22]. Similarly, bacterial lipoproteins have been shown to stimulate proliferation of B cells leading to increased immunoglobulin secretion [23]. Thus, bacterial lipoproteins are able to activate both the innate and adaptive wings of the immune system. Mycobacterium genome encodes for 99 lipoproteins, however, the functions of many of them are still unknown. Several of them have been shown to play a major role in virulence [16, 24, and 25]. Mtb PstS-1, a 38 kDa lipoprotein, is involved in phosphate transport as well as in inducing apoptosis [25]. Deletion mutants of Mtb lgt and lsp showed a significant reduction in virulence [26]. Similarly, an extensively studied  has been shown to strongly induce the innate immunity by activating cathelicidin mediated autophagy in a TLR-2 dependent manner [24,16]. LpqH binds to TLR-2 that leads to activation of cytokines such as interleukin-12 (IL-12), interleukin-1 beta (IL-1β) and tumor necrosis factor-alpha (TNF-α). Moreover, LpqH also plays a pivotal role in bacterial survival by inhibiting interferon-gamma (IFN-) response genes such as CIITA (Class II trans activator) that leads to reduced antigen presentation [24]. Another study has shown that a synthetic 19-kDa Mtb derived lipopeptide enhances the antimicrobial capacity of monocytes via TLR2/1 signaling, vitamin D and VDR-dependent pathway [27]. This involved induction of the CAMP gene and its protein [28]. Some of the lesser studied mycobacterial lipoprotein genes include LprG, LprA, LpqB, and LpqM.
In the present study, we have shown that one of the Mtb lipoproteins, LprE (LprE Mtb ), is involved in intracellular bacterial survival and evasion of host immune responses. We demonstrate that LprE Mtb enhances bacillary survival by inhibiting the expression of CAMP via TLR-2-p38-Cyp27B1-VDR signaling pathway in human macrophages. We also demonstrate that LprE Mtb downregulates the pathogen induced IL-1β pro-inflammatory response. Furthermore, mechanistic studies showed that LprE Mtb inhibits autophagy and phago-lysosome fusion by down-regulating the expression of several endosomal markers such as Early Endosome Antigen 1 (EEA1), Rab7 and lysosomal-associated membrane protein 1 (LAMP-1), and IL-12 and IL-22 cytokines, thus aid bacterial survival in human macrophages. In summary, we have characterized a novel Mtb lipoprotein that aids bacterial survival in host macrophages by down-regulating the expression of antimicrobial peptide cathelicidin.

Genetic organization of LprE in M. tuberculosis genome
Initial annotation from the Tuberculist Web server suggested that Mtb H37Rv LprE (LprE Mtb ), encoded by a 609 bp Rv1252c gene, belongs to a previously uncharacterized Rv_dir301 operon (webTB.org database). This operon consists of yet another uncharacterized Rv1251c gene (Fig 1A). In Mtb CDC1551, a clinical isolate, LprE is encoded by an MT1291 gene (Fig 1B). Multiple sequence alignment and BLAST results showed 100% sequence homology between Mtb H37Rv and Mtb CDC1551 LprE genes (Suppl Fig 1).The tertiary structure of LprE is not available. Therefore, we predicted its structure using a Modeller program. The LprE sequence showed high structural homology (94.5%) with human mitochondrial RNA polymerase (PDB ID: 3SPA) (Fig 1C) (Fig 2A). These results suggest a pivotal role for LprE Mtb  Several studies including our previous work have used Msm as a surrogate model to study the role of various Mtb proteins in pathogenesis [30][31][32][33][34]. The qRT-PCR analysis confirmed that LprE is ectopically expressed in Msm strain grown under in-vitro condition (Fig 2B) Msm genome analysis showed presence of MSMEG_5043, an ortholog of LprE Mtb [35]. To preclude the effect of MSMEG_5043, we first constructed Msm5043 mutant by allelic exchange method (Fig 2C). Deletion of MSMEG_5043 was confirmed by PCR using gene specific and flanking region primers ( Suppl Fig 2A) Msm::pSMT3 and Msm∆5043 strains (Fig 2F). We did not observe any difference in the phagocytosis rate of Msm::pSMT3, Msm::LprE, Msm∆5043 and Msm∆5043::LprE strains ( Fig 2G) indicating that the increased survival of Msm::LprE and Msm∆5043::LprE strains is not due to the difference in the uptake of bacteria by macrophages.

LprE Mtb inhibits cathelicidin expression in human macrophages
Macrophages exhibit antimicrobial response by inducing the expression of cationic antimicrobial peptide cathelicidin (hCAP18/LL-37) [36]. However, Mtb is known to down regulate the expression of cathelicidin to avoid killing by macrophages [37]. Previously, it has been shown that 19-kDa lipoprotein from Mtb interacts with TLR-2, which subsequently up-regulates the expression of Cyp27B1 hydroxylase, VDR translocation into the nucleus and finally the induction of cathelicidin expression [14]. Our previous study also showed that cathelicidin is able to kill both pathogenic and non-pathogenic mycobacteria under in vitro and ex vivo conditions [13]. Therefore, we first evaluated the expression of CAMP, which encodes human cathelicidin, in Mtb, Mtb∆LprE, and Mtb∆LprE::LprE infected THP-1 cells.
We observed an ~4-fold increase in CAMP expression in cells infected with Mtb∆LprE compared with Mtb and Mtb∆LprE::LprE infected cells (Fig 3A). In contrast, Msm::LprE conditions without exhibiting any toxic effect on macrophages [13]. Furthermore we established that Msm elimination is dependent on cathelicidin expression [13]. In agreement with our previous data, we observed increased killing of Msm::LprE and Msm∆5043::LprE under both pre and post-treated conditions in comparison with untreated cells (Fig 3C).
However, more bacterial killing was observed under post-treated condition. As expected treatment with 50 µg/ml LL-37 did not show any cytotoxic effect on THP-1 cells (data not shown).
Vitamin D3 is a known inducer of cathelicidin, which results in increased bacterial killing in human macrophages [38]. As shown in Figure 3D, treatment with 20 nM Vitamin D3 further reduced Msm::pSMT3 and Msm∆5043 survival compared with untreated cells (Fig   3D), whereas we observed no difference in the cfu's of Msm::LprE and Msm∆5043::LprE (Fig 3D), suggesting that LprE Mtb impedes the vitamin D3 mediated response.

LpqH Mtb and LprE Mtb antagonostically regulate CAMP expression
Mycobacterium genome encodes for 90 lipoproteins. Among them, 19-KDa lipoprotein LpqH Mtb has been shown to induce cathelicidin expression to restrict intracellular mycobacterial growth [16]. Therefore, we investigated the impact of LpqH Mtb 3F).

LprE Mtb inhibits cathelicidin via p38 MAPK pathway in macrophages
Based on above observations, our subsequent studies were focused on the investigation of the molecular mechanism(s) that are responsible for the down-regulation of CAMP expression by LprE Mtb . It has been reported that LpqH Mtb and TLR-2 interaction regulate CAMP expression via activation of downstream p38 MAPK signaling pathway [14]. Mycobacterial antigen-  Fig 3A). This is in agreement with the previous study that Mtb infection down-regulates phospho-p38 [39].

LprE Mtb down-regulates CYP27B1 and VDR expression in macrophages
Next, we studied the expression of cathelicidin regulation pathway intermediates CYP27B1 and the VDR. Increased expression of CYP27B1 was observed at both transcriptional ( Fig   11   3I) and translational levels (Fig 3J) in Mtb∆LprE infected macrophages as compared to Mtb and Mtb∆LprE::LprE infected cells. Of note, CYP27B1 expression was found to be increased both at 12 and 24 h post-infection (Fig 3J). On the other hand, the expression of CYP27B1 was found to be significantly down-regulated at both transcriptional (Suppl Fig 3B) and translational (Suppl Fig 3C)

LprE Mtb down-regulate the expression of CAMP via TLR-2
Previous studies have shown that lipoproteins regulate cathelicidin expression through TLRs [40,41]. To identify the involvement of specific TLR in the regulation of CAMP expression, we first determined in-silico binding efficiency of LprE Mtb with human TLR-1,2,4 and 6. For this, we first performed docking analysis of energy minimized LprE Mtb protein with human TLR's tertiary structure obtained from PDB database. LprE Mtb showed strong binding efficiency with TLR-2 (-582.21) as compared to TLR-1 (-223.5), TLR-4 (2.89) and TLR-6 (- Fig 4A), as determined by atomic contact energy (ACE) score obtained from PatchDock analysis [42].

LprE Mtb inhibits the production of pro-inflammatory cytokine IL-1β
IL-1β cytokine induces the expression of antimicrobial peptides through TLR signaling to clear the bacterial infection [43]. Active TB patients showed reduced levels of IL-1β suggesting that it may have a protective role in TB infection [44]. Next, we investigated the  Fig 5B).
In macrophages, processing and release of active IL-1β are dependent on caspase-1 activation [45]. Therefore, we investigated whether LprE Mtb mediated IL-1β down-regulation is dependent on caspase activation. Indeed, western blot analysis showed an increased level of cleaved caspase-1 in Mtb∆LprE infected macrophages when compared with Mtb and Mtb∆LprE::LprE infected cells after 12 h of infection (Fig 5C). Together data suggests that LprE Mtb suppresses caspase-1 dependent IL-1β production to facilitate bacterial persistence in macrophages.

LprE Mtb inhibits autophagy to dampen host immune response
Autophagy is a known host defense mechanism that plays an important role in the restriction Western blot analysis showed the increased conversion of LC3-I to a characteristic autophagic induction marker LC3-II in MtbLprE infected cells (Fig 6A).We also examined the expression of an autophagic flux marker p62(SQSTM1) [47]. As shown in Fig 6A, p62 level decreased in case of Mtb∆LprE infected macrophages (Fig 6A). Moreover, the expression of other autophagic markers such as Atg-5 and Beclin-1, which are recruited to the phagosomal compartments during autophagic vesicle formation [48], were also found to be increased in Mtb∆LprE infected cells as compared to Mtb and Mtb∆LprE::LprE infected cells (Fig 6B). 14 To further confirm the role of LprE Mtb  Msm∆5043::LprE, which otherwise showed more survival by down-regulating the expression of autophagy, as compared to untreated cells (Fig 6C). We confirmed that rapamycin treatment indeed induced autophagy by determining the expression of LC3-II and Atg-5 by western blots analysis (Fig 6D).

LprE Mtb blocks phago-lysosome fusion by down-regulating the expression of endosomal markers
Non-pathogenic mycobacteria containing phagosomes readily fuse with lysosomes leading to the elimination of pathogen; however, pathogenic mycobacteria inhibit the fusion of phagosomes with lysosomes to survive in macrophages for an extended period of time.
Previously, we have shown that exogenous administration of cathelicidin increased the colocalization of M. bovis-BCG containing phagosomes with lysosomes resulting in increased bacterial killing [11]. Based on these observations, we extended our study to investigate whether LprE Mtb  expression as compared to vector control Msm::pSMT3 infected cells (Suppl Fig 6B). The IL-22 level also increased by more than 2-fold in macrophages infected with the Mtb∆LprE 16 (Fig 7D), while Msm::LprE and Msm∆LprE::LprE infected condition significantly reduced IL-22 transcript levels (Suppl Fig 6C). These results indicate that LprE Mtb probably blocks phagosomal maturation by down-regulating the expression of IL-12 and IL-22 cytokines.

DISCUSSION
Mtb exhibits an extraordinary ability to survive inside the host cells. This is mainly attributed to the plethora of virulence factors produced by the bacterium. Many pathogenic bacteria, including Mtb, inhibit autophagy mechanism to facilitate its persistence in host cells [66]. Several bacterial virulence proteins were found to inhibit the autophagy. LpqH Mtb activates autophagy, whereas EIS Mtb (enhance intracellular protein) protein suppresses autophagy [67]. Our previous work also showed that a Mtb phosphoribosyltransferase suppressed autophagy expression to promote intracellular bacterial survival [32]. In the current study, we found that LprE Mtb  play an important role in the phagosomal escape of bacteria [69].
Stimulation of macrophages with cytokines restricts the growth of microorganisms within endocytic compartments [70]. It is shown that treatment of human macrophages with IL-12 restrict the growth of M. bovis-BCG by facilitating the phago-lysososome fusion [51].
Similarly, treatment with IL-22 was shown to increase the expression of endocytic marker proteins and increased bacterial killing by inducing phago-lysosomal fusion [52]. In our study, we observed decreased levels of IL-12 and IL-22 in cells infected with LprE Mtb expressing mycobacteria, while expression of these cytokines was found to be increased in Collectively, our findings identified a new Mtb lipoprotein that could be used by pathogenic mycobacteria to alter host immune responses thus presenting attractive targets for new drug therapies. Figure 8 shows a schematic representation of modulation of different host immune responses by LprE Mtb leading to increased bacillary survival in macrophages.

Ethics and Biosafety Statement
All experiments were approved by the Institutional Biosafety committee of KIIT University

Construction of M. tuberculosis CDC1551 LprE strain
For the construction of complemented strain, LprE was amplified from Mtb CDC1551 genomic DNA using gene specific primers ( Table 1) and was cloned into NdeI-Hind III sites of the pVV16 vector [72] and pNiT vector [72]. The pVV16-LprE and pNit-LprE constructs were transformed into MtbΔLprE mutant to generate MtbΔLprE::LprE complemented strain.

Construction of recombinant M.smegmatis strain expressing LprE Mtb and LpqH Mtb
LprE Mtb and LpqH Mtb were PCR amplified using gene specific primers ( Table 1) using genomic DNA as a template. The PCR amplified products were purified from the gel, double digested with PstI and HindIII and cloned into pSMT3 shuttle vector separately. The recombinant constructs were transformed into competent E. coli XL-10 gold. LB agar plates supplemented with 20 µg/ml tetracycline and 50 µg/ml hygromycin was used to select the positive colonies and were further confirmed by colony PCR and sequencing using gene specific primers. Finally, the recombinant constructs were transformed into electro competent Msm. The positive colonies were selected on 7H10 medium containing 50 µg/ml hygromycin. The positive transformants were confirmed by colony PCR and sequencing using gene specific primers ( Table 1).Generation of recombinant Msm::LprE-LpqH strain was confirmed by sequencing using gene specific primers.

Construction of M. smegmatis∆5043 mutant
The allelic exchange substrate (AES) for the generation of Msm 5043 mutant (Msm∆5043) was produced as described previously [72]. Briefly, approximately 800 kb regions upstream and downstream of MSMEG_5043 loci were PCR amplified using specific primers ( Table 1).

Flow cytometry analysis
THP-1 cells (2x10 5 cells/well) were seeded on a 24-well plate and treated with PMA as described above. Then cells were infected with Msm::pSMT3, Msm::LprE, Msm∆5043, and Msm∆5043::LprE at an MOI 10 as described above. After infection, cells were lysed by treating with 0.5% Triton X-100 for 2 min to release the intracellular bacteria. The cell suspension was centrifuged at 1000 g for 10 min to pellet down the lysed macrophage cells.
Supernatants that contained released bacteria were again centrifuged at 12000 g for 10 min. plates to count live bacteria. Bacterial preparations were plated before infection to ensure that an equal number of bacteria were used for infection assays.

Isolation of human peripheral blood mononuclear cells (PBMCs)
Blood was collected from healthy individuals, diluted ( h before (pre-treated) and after (post-treated) infection as described previously [11].
Scrambled siRNA was used as a negative control. The electroporated cells were treated with 20 nM PMA and seeded onto 6 well plates for infection assays. Silencing efficiency was determined using gene specific primers.

Western blotting
All the antibodies except LC3 and CYP27B1 were purchased from Cell Signaling. The Image J2 (NIH, USA) was used for computer analysis of pixel intensity of bands on films.
Relative band densities relative to respective loading control were determined.