|Home | About | Journals | Submit | Contact Us | Français|
Autophagy is one of the major immune mechanisms engaged to clear intracellular infectious agents. However, several pathogens have evolved strategies to evade autophagy. Here, we demonstrated that Mycobacteria, Shigella, and Listeria but not Klebsiella, Staphylococcus, and Escherichia inhibit IFNG-induced autophagy in macrophages by evoking selective and robust activation of WNT and SHH pathways via MTOR. Utilization of gain- or loss-of-function analyses as well as mir155-null macrophages emphasized the role of MTOR-responsive epigenetic modifications in the induction of Mir155 and Mir31. Importantly, cellular levels of PP2A, a phosphatase, were regulated by Mir155 and Mir31 to fine-tune autophagy. Diminished expression of PP2A led to inhibition of GSK3B, thus facilitating the prolonged activation of WNT and SHH signaling pathways. Sustained WNT and SHH signaling effectuated the expression of anti-inflammatory lipoxygenases, which in tandem inhibited IFNG-induced JAK-STAT signaling and contributed to evasion of autophagy. Altogether, these results established a role for new host factors and inhibitory mechanisms employed by the pathogens to limit autophagy, which could be targeted for therapeutic interventions.
Autophagy is primarily a homeostatic mechanism that enables recycling of the cellular cargo or damaged or excess organelles by sequestering them to characteristic double-membrane autophagosomes, which in turn fuse with lysosomes. 1 Mechanistic target of rapamycin (MTOR) is a critical regulator of autophagy, which inhibits the activation of the ULK (unc-51 like autophagy activating kinase) complex required for initiation of autophagy. Stimulation of the ULK complex leads to a cascade of events including activation of class III phosphatidylinositol 3-kinase (PtdIns3K) complexes, ATG9, MAP1LC3 (microtubule-associated protein 1 light chain 3) conjugation systems to form the functional autophagosome. 2 , 3 Apart from the cues such as nutrient limitation, stress, and starvation, infection-induced autophagy has recently garnered attention. 4 , 5
Autophagy contributes to both innate and adaptive immune responses to infections and plays an essential role in restricting intracellular pathogens and delivering pathogen-derived antigens for major histocompatibility complex class II presentation. 6 - 9 Thus, autophagy acts as one of the major regulatory mechanisms to curtail the infectious agents. However, intracellular pathogens, especially viruses such as herpes simplex virus, human immunodeficiency virus, influenza; and bacteria like Mycobacteria, Shigella, Listeria, exhibit multiple mechanisms to evade autophagy. 10 - 13 Escape from autophagy includes not only the ability of the pathogen or pathogen-derived antigens such as Shigella T3SS effector IcsB 14 and the Listeria protein ActA 15 to directly modulate the components of autophagy but also the utilization of this machinery for their survival as in the case of several viruses. 12 Further, genome-wide screening studies have indicated the existence of pathogen-induced host macromolecules that could inhibit autophagy during mycobacterial infections. 16 However, the identities of such host signaling molecules and mechanisms by which pathogens modulate autophagy have not been explored in depth.
Recognition of pathogens by pattern-recognition receptors (PRR) such as toll-like receptor (TLR) and subsequent cytokine responses lead to numerous signaling pathways that could modulate the various molecular regulators of autophagy. Interestingly, signaling via TLR4, as well as TLR7 and TLR8 but not TLR2, an important PRR utilized by pathogens such as Mycobacteria, Shigella, and Listeria, induce autophagy, 17 suggesting that these pathogens might use the TLR2 pathway to escape autophagy. Several reports have shown that TLR2 signaling by Mycobacteria leads to activation of a gamut of pathways like NOTCH1, 18 WNT, 19 SHH (sonic hedgehog) 20 contributing to the differential regulation of the innate immune response and cytokine milieu. 21 Of note, a recent study reveals the inhibitory effects of HH signaling on autophagy. 22 Further, the Th1 cytokines TNF 23 and IFNG (interferon, gamma) 24 , 25 that are implicated in protective responses to mycobacteria, could augment autophagy while the Th2 cytokines IL4 and IL13 could inhibit autophagy by modulating the class I phosphoinositide 3-kinase (PI3K) and MTOR signaling. 26 Based on these observations, we hypothesized that pathogens have the ability to differentially regulate the signaling pathways to regulate autophagy.
Here, we demonstrate that bacterial pathogens such as Mycobacteria, Shigella, and Listeria inhibit IFNG-induced autophagy and stimulate robust activation of MTOR-responsive WNT and SHH signaling. GSK3B, a negative regulator and a nodal link between WNT and SHH pathways was ascertained to be regulated by a protein phosphatase, PP2A. The activity of PP2A was marked by MTOR-dependent expression of microRNAs, Mir155 and Mir31 as validated by utilization of macrophages derived from mir155-null mice or by conventional siRNA or miRNA mimics or overexpression strategies. Importantly, we found that infection-induced WNT and SHH pathways stimulated the production of ALOXs (arachidonate lipoxygenases) that facilitated the ability of the pathogens to inhibit autophagy. Together, we have identified novel molecular mechanisms and host factors that are crucial to control autophagy and help the bacterial pathogens to evade the host immune responses.
To investigate the molecular mechanisms of regulation of autophagy by pathogens, we chose a panel of bacteria. Klebsiella pneumoniae, Staphylococcus aureus, and Escherichia coli were previously reported to induce autophagy, whereas, Mycobacterium bovis BCG, Shigella flexneri, and Listeria monocytogenes are known to inhibit autophagy. First, we assessed the ability of these pathogens to modulate IFNG-mediated autophagy in macrophages. As expected, IFNG-induced high levels of autophagy in macrophages as analyzed by the characteristic MAP1LC3 puncta formation and MAP1LC3-II lipidation (Fig. 1A–C). Interestingly, IFNG-induced autophagy was significantly inhibited by M. bovis BCG, S. flexneri, and L. monocytogenes whereas the other 3 bacteria did not modulate the process. For further investigations, M. bovis BCG was used as a model pathogen while other bacteria were used only for key experiments.
TLR2 is one of the major innate PRRs that mediates the immune response to mycobacteria. We analyzed the role for TLR2 in inhibition of IFNG-induced autophagy. M. bovis BCG failed to inhibit IFNG-induced autophagy both in TLR2 dominant negative (DN)-expressing and tlr2-deficient macrophages (Fig. S1A–S1C). Further, MTOR, an essential regulator of autophagy was phosphorylated and markedly active in M. bovis BCG-, S. flexneri- and L. monocytogenes-infected macrophages as assayed by phosphorylation status of ribosomal protein S6 kinase, 70 kDa, polypeptide 2 (RPS6KB2/p70S6K) at an MTOR-specific phosphorylation site (Thr389) (Fig. 1D). Corroborating these results, MTOR-immunoprecipitates from M. bovis BCG-, S. flexneri- and L. monocytogenes-infected macrophages displayed increased MTOR kinase activity as analyzed by ELISA-based activity assays (data not shown). These results indicate MTOR-dependent regulation of autophagy by these pathogens.
Previous reports suggest that mycobacteria activate the TLR2-dependent RAS pathway. 27 , 28 Further, phosphatidic acid, a product of PLD1 (phospholipase D1, phosphatidylcholine-specific), is a known inducer of MTOR signaling 29 and a role for RAS-induced, PLD1-mediated MTOR activity is also established. 30 Therefore, we examined the role of RAS-, PLD1- and PI3K-mediated MTOR activation during mycobacterial infection. We saw that, pharmacological inhibition of RAS-PLD1-PI3K significantly abrogated M. bovis BCG-induced activation of MTOR as assayed by phosphorylation status of RPS6KB2 (Thr389) (Fig. S1D). Interestingly, M. bovis BCG failed to induce MTOR activation in tlr2-deficient macrophages (Fig. S1D). S. flexneri and L. monocytogenes also exhibited similar TLR2 dependency for MTOR activation (data not shown). Further, M. bovis BCG-infection of macrophages triggered marked activating phosphorylation of ULK1 (Ser555) while MTOR-dependent inhibitory phosphorylation of ULK1 (Ser757) was marginal (Fig. S1E). These data thus suggest that regulation of autophagy by mycobacteria via MTOR is ULK-independent. Therefore, we aimed at identifying other MTOR-responsive mediators that could play a role in abrogating IFNG-induced autophagy during mycobacterial infection.
Previously, we have demonstrated that TLR2-mediated signaling by M. bovis BCG triggers NOTCH, 18 WNT 19 and SHH 20 pathways in macrophages. Interestingly, MTOR can inhibit GSK3B, 31 , 32 a negative regulator of WNT and SHH pathways suggesting that WNT and SHH might be implicated in MTOR-dependent regulation of autophagy by Mycobacteria and other pathogens. In line with this hypothesis, M. bovis BCG, S. flexneri, and L. monocytogenes induced increased activation of both WNT and SHH pathways (Fig. 1E; Fig. S1F). WNT pathway activation was analyzed by transcript levels of WNT-responsive genes Wnt5a, Fzd4, and Lrp5 and phosphorylation status of CTNNB1 and GSK3B. Phosphorylated CTNNB1 (Ser33 and Ser37, as well as Thr41) indicates inactivation of WNT pathway whereas phosphorylation at Ser9 renders GSK3B inactive and activates WNT signaling. Transcript analysis of Shh, Ptch1, and Gli1 and protein level expression of SHH, GLI1, NUMB, GSK3B were utilized as read out for activation of canonical SHH signaling. Increased SHH, GLI1, p-GSK3B (Ser9) and decreased NUMB indicates active SHH signaling. MTOR-mediated activation of these pathways was further confirmed by the inability of M. bovis BCG to induce the activation WNT and SHH pathways in the presence of Mtor-specific siRNA or DN (Fig. 1F and G; Fig. S1G) and that overexpression of MTOR in macrophages induced WNT and SHH pathways (Fig. 1G; Fig. S1G).
We further corroborated the inhibitory effects of WNT and SHH signaling on IFNG-induced autophagy by overexpression and knockdown experiments. Ectopic expression of WNT5A and SHH proteins drastically abrogated IFNG-induced MAP1LC3 lipidation and puncta formation (Fig. 2A and B; Fig. S2A and S2B). Substantiating this observation, inhibition of WNT and SHH signaling either by pharmacological inhibitors or by RNA interference of Wnt5a, Ctnnb1, Shh, or Gli1 significantly reduced the ability of M. bovis BCG to inhibit IFNG-induced autophagy (Fig. 2C–F; Fig. S2C–S2F).
Protein phosphatases such as PP2A, a Ser/Thr protein phosphatase, have been implicated in the process of autophagy. 2 PP2A dephosphorylates the inactive GSK3B to active form. 33 However, the specificity of PP2A is attributed to the numerous regulatory subunits, among which, B56 (PPP2R5A)-containing PP2A has been proposed to have an inhibitory role on WNT 34 - 36 and SHH 37 , 38 pathways. As GSK3B inhibition was found to be MTOR-responsive (Fig. 1G), we investigated the repercussion of regulation of GSK3B by MTOR and PPP2R5A on autophagy. Interestingly, by RNA interference of either MTOR (Mtor siRNA) or individual molecular complexes (MTORC1-Rptor siRNA and MTORC2-Rictor siRNA), we found that M. bovis BCG downregulates the expression of PPP2R5A protein, but not transcripts, in an MTOR-dependent manner (Fig. 3A and B; Fig. S2G and S2H). Corroborating this result, both MTORC1 and MTORC2 seem to possess the ability to inhibit IFNG-induced autophagy, though the latter was comparatively less effective (Fig. 3C). Coimmunoprecipitation of PPP2R5A and GSK3B further confirmed the PP2A-GSK3B regulatory interactions (Fig. 3D). Importantly, macrophages that were transfected with Ppp2r5a-specific siRNA remained refractory to IFNG-triggered autophagy thus underscoring the importance of PPP2R5A in autophagy (Fig. 3E and F; Fig. S2I). In accordance, overexpression of PPP2R5A in macrophages strongly repressed M. bovis BCG-driven inhibitory effects on IFNG-induced autophagy (Fig. 3G and H; Fig. S2J) and failed to induce WNT and SHH pathway in response to M. bovis BCG (Fig. 3I). Together, these results demonstrate that pathogens could evade host response such as autophagy by modulating PP2A levels.
M. bovis BCG-induced decrease in PPP2R5A protein but not the transcripts (Fig. 3A and B), strongly advocates a role for posttranscriptional regulatory mechanisms like those mediated by miRNAs. Data from a miRNA microarray 20 together with extensive bioinformatic analysis (TargetScan, miRanda, miRWalk and RNAhybrid) identified Ppp2r5a as a potential target for Mir155 and Mir31. The target sites located at the residues spanning from 1022 to 1028 (for Mir155) and 960 to 967 (for Mir31) in Ppp2r5a mRNA were identified as critical for miRNA-3′UTR interactions (Fig. 4A).
To explore their possible role in the current study, macrophages infected with the panel of bacteria were analyzed for induced expression of Mir155 and Mir31. Infection with M. bovis BCG, S. flexneri, and L. monocytogenes but not with K. pneumoniae, S. aureus, and E. coli elicited robust expression of Mir155 and Mir31 (Fig. 4B). Importantly, M. bovis BCG infection or transfection with Mir155 or Mir31 mimics markedly reduced WT Ppp2r5a 3′UTR luciferase activity. However, the reduction was not significant when single mutants (for Mir155 or Mir31 binding sites) or double mutants (for Mir155 and Mir31 binding sites) Ppp2r5a 3′UTR constructs were utilized. Interestingly, Mir155 or Mir31 mutant reporters did exhibit a significant decrease in control experiments in presence of Mir31 or Mir155 mimics respectively (Fig. 4C). These results thus validate that Ppp2r5a is a direct target of Mir155 and Mir31.
To further corroborate the direct interaction of Ppp2r5a with Mir155 and Mir31, RNA immunoprecipitation of AGO2 (argonaute RISC catalytic component 2) was performed. As shown in Figure 4D, Mir155, Mir31 and Ppp2r5a mRNA were significantly enriched in AGO2 immunoprecipitates in WT bone marrow-derived macrophages (BMDMs). However, enrichment of Mir155 and Ppp2r5a mRNA in the AGO2 immunoprecipitates was severely reduced in mir155-deficient BMDMs. Mir106a was used as a nonspecific control that did not vary with M. bovis BCG infection. Together, these results suggest that Ppp2r5a is a bonafide target of M. bovis BCG-induced Mir155 and Mir31.
In agreement with the above results, infection-induced Mir155 and Mir31 were found to be dependent on MTOR (Fig. 4E; Fig. S3A) and TLR2 (Fig. S3B). Upon M. bovis BCG infection of macrophages, MTOR-dependent increased H3K4me3 and decreased H3K27me3 methylations, that mark active transcription, were observed at the Mir155 and Mir31 promoters. (Fig. 4F). Interestingly, a previous investigation suggested a role for MTOR in modulating H3K4 methylations. 39 In this perspective, we speculated that MTOR mediates these epigenetic changes at Mir155 and Mir31 promoters by regulating the histone-modifying enzyme ASH2L, a component of the KMT2A/MLL-SETD1A/SET1 family of histone methyltransferases that mediates H3K4 methylation. Consistent with the H3K4me3 chromatin immunoprecipitation (ChIP) data, recruitment of ASH2L to the promoters of Mir155 and Mir31 was found to be MTOR-dependent as pharmacological inhibition of MTOR by rapamycin severely abrogated the M. bovis BCG-induced ASH2L recruitment to the Mir155 and Mir31 promoters (Fig. 4F).
Inhibition of PPP2R5A activity by Mir155 and Mir31 was further confirmed by evaluating the activation status of GSK3B. Upon M. bovis BCG infection, BMDMs from mir155-null mice as well as Mir155 or Mir31 siRNA-transfected macrophages failed to abrogate PPP2R5A levels resulting in decreased inactive GSK3B (p-GSK3B) (Fig. 5A). Concomitantly, enforced expression of Mir155 and Mir31 resulted in markedly reduced PPP2R5A levels and increased inactive form of GSK3B (Fig. 5A; Fig. S3C). Substantiating these results, we also observed that BMDMs from mir155-null mice or Mir155 or Mir31 siRNA-transfected macrophages were unable to activate M. bovis BCG-induced WNT and SHH signaling, whereas, Mir155 or Mir31 mimics alone could activate WNT and SHH signaling as indicated by the increased p-CTNNB1, p-GSK3B, SHH, GLI1, and decreased NUMB expression (Fig. 5B).
In order to further delineate the mechanism of Mir155 and Mir31-mediated inhibition of autophagy, we performed conventional gain-of-function and loss-of-function analyses. Knockdown of Mir155 and Mir31 using specific siRNAs in macrophages significantly reduced the ability of M. bovis BCG to inhibit IFNG-induced autophagy as shown by MAP1LC3 puncta (Fig. 5C). MAP1LC3-II lipidation further confirmed the compromised ability of M. bovis BCG to inhibit autophagy in Mir155 or Mir31 siRNA-transfected macrophages as well as in mir155-deficient BMDMs (Fig. 5D and E). In agreement with these observations, Mir155 or Mir31 mimic-transfected macrophages could significantly reduce IFNG-induced autophagy (Fig. 5F–H). These results thus strongly implicate Mir155 and Mir31 in the modulation of PPP2R5A and downstream signaling pathways that regulate autophagy.
Pathogens such as mycobacteria successfully evade the host immune response by triggering antiinflammatory pathways. 40 , 41 ALOX/LO (arachidonate lipoxygenase)-mediated production of lipoxins (LXs), which are chalones in inflammatory responses, represents one such mechanism. 42 , 43 Therefore, we examined the expression of ALOXs such as ALOX5 and ALOX15 in response to the tested microbes. Notably, microbes that inhibited IFNG-induced autophagy, M. bovis BCG, S. flexneri, and L. monocytogenes, induced marked levels of both ALOX5 and ALOX15 (Fig. 6A; Fig. S4A). Further, by using macrophages from tlr2-null mice and macrophages transfected with Mtor siRNA or the MTOR overexpression construct, we show that M. bovis BCG-mediated ALOX induction was TLR2- and MTOR-dependent (Fig. 6B; Fig. S4B and S4C).
We then explored molecular mechanisms that promote infection-induced ALOX expression. As M. bovis BCG infection induced robust expression of Mir155 and Mir31, we investigated the role of these miRNAs in the regulation of ALOX. We show that mir155-null BMDMs or Mir31 siRNA-transfected macrophages failed to induce ALOX5 and ALOX15 upon M. bovis BCG infection (Fig. 6C and D). In agreement with this, enforced expression of Mir155 and Mir31 augmented the expression of ALOX5 and ALOX15 (Fig. 6D). Also, PPP2R5A negatively regulated the expression of infection-induced ALOXs (Fig. 6E).
Next, contribution of induced WNT and SHH signaling pathways in driving ALOX expression was verified. Overexpression of WNT5A and SHH (Fig. 6F and G) in macrophages instigated the expression of ALOX5 and ALOX15 whereas pharmacological inhibition (Fig. 6H; S4D and S4E) or siRNA-mediated interference (Fig. 6I) of WNT and SHH signaling mediators (Wnt5a, Ctnnb1, Shh, or Gli1) significantly impaired the ability of M. bovis BCG to induce ALOXs. Finally, ChIP assays were performed to confirm the participation of TLR2-responsive WNT and SHH signaling effectors, CTNNB1 and GLI1 respectively, in regulating the expression of both ALOX5 and ALOX15. By using MatInspector program, a total of 8 and 5 CTNNB1 binding sites and 2 and 3 GLI binding sites were identified in promoters (2 kb upstream from the start site) of Alox5 and Alox15 genes respectively. Interestingly, TLR2-dependent 2- to 3-fold recruitment of CTNNB1 and GLI1 to the promoters of Alox5 and Alox15 was observed (Fig. 6J). Altogether, these results demonstrate that Mir155-Mir31-WNT-SHH arbitrate infection-induced ALOX expression in a TLR2-dependent manner.
As Mir155-Mir31-WNT-SHH coordinated the M. bovis BCG-induced ALOXs, it raised the possibility that ALOXs play a critical role in autophagy. Accordingly, knockdown of ALOXs using specific siRNA (Fig. 7A–D) or utilization of ALOXs-specific pharmacological inhibitors (Fig. 7E and F) significantly repressed the ability of M. bovis BCG to inhibit IFNG-induced autophagy. Conversely, macrophages overexpressing ALOX5 or ALOX15 reduced the IFNG-induced autophagy (Fig. 7G and H).
IFNG activates the downstream STAT1 (signal transducer and activator of transcription 1) signaling to induce antimicrobial activity of macrophages such as autophagy. 44 In this regard, we examined the effect of ALOXs on the IFNG-induced STAT pathway. Suppression of JAK (janus kinase)-STAT signaling using pharmacological inhibitor AG490 severely compromised IFNG-induced autophagy (Fig. 7I and J). Interestingly, enforced expression of ALOX5 and ALOX15 significantly reduced the IFNG-mediated activation of STAT signaling (Fig. 7K).
In addition to M. bovis BCG, virulent M. tuberculosis also exhibited the ability to inhibit IFNG-induced autophagy by arbitrating the infection-induced ALOX expression via induction of Mir155-Mir31-WNT-SHH signaling (Fig. 8). Further, to analyze the functional consequence of altered IFNG-dependent autophagy in terms of bacterial replication, we performed intracellular CFU-based analysis of M. bovis BCG, S. flexneri, and L. monocytogenes in presence of inhibitors of the proposed pathway. As illustrated in Figure 9A, pharmacological perturbations of WNT (FH535), SHH (Cyclopamine), ALOX5 (AA-861), ALOX15 (PD146176) or MTOR (rapamycin) significantly reduced the ability of tested bacteria to resist IFNG-mediated bacterial clearance.
Autophagy is one of the major host effector mechanisms to clear intracellular microbes. However, pathogens utilize several pathways to evade autophagy although microbes display differential abilities to counter autophagy. 10 In the present study, by using a panel of bacterial species, we elucidated the rationale and molecular signatures involved in autophagic evasion. We show that M. bovis BCG, S. flexneri, and L. monocytogenes but not K. pneumoniae, S. aureus, and E. coli inhibit IFNG-induced autophagy of macrophages via MTOR by Mir155- and Mir31-responsive WNT and SHH signaling. Though certain reports suggest the refractoriness of mycobacteria-infected macrophages toward IFNG, 18 , 45 - 51 we present molecular evidence that demonstrates repression of IFNG-induced JAK-STAT pathway during autophagy. Supporting our observation, previous reports suggest that Mycobacteria and its cell wall components, like TDM, downregulate IFNG-mediated responses by inhibiting STAT phosphorylation including p-STAT1. 52 - 54 Of note, individuals with STAT1 deficiency or defects in IFN-JAK-STAT pathway exhibit increased susceptibility to mycobacterial infections. 55 , 56 Impaired activation of STAT1 or reduced STAT1 phosphorylation in patients with active tuberculosis further underscores our data. 57 , 58
Interestingly, in the current study, both MTORC1 and MTORC2 contributed to the mycobacteria-mediated inhibition of IFNG-induced autophagy. This could be attributed to either the feed-forward loop from MTORC2 to MTORC1 or presence of MTORC2-dependent inhibitory effects on autophagy. 59
The decisive role for TLR2 in shaping the outcome of infection with several pathogens, including mycobacteria, is well established. 20 , 60 - 62 However, contribution of TLR2 responses in modulating autophagy has not been addressed. Present observations could add new insights to the infection-induced TLR2 signaling during inhibition of autophagy. Interestingly, in agreement with our results, ligand-dependent activation of TLR4, as well as TLR7 and TLR8 but not TLR2 induces autophagy. 17 However, other TLR2 ligands like Pam3CSK4 or PIM2 failed to abrogate IFNG-induced autophagy when compared with M. bovis BCG (Fig. S5A and S5B). We would like to emphasize that whole viable mycobacteria do exhibit various immune evasion mechanisms including inhibition of IFNG responses by utilizing a complex network of signaling events. However, in our view, treatment with purified, TLR2 ligands may not reflect the infection scenario. Likewise, crucial signaling pathway determinants have been implicated to regulate autophagy. Recently, studies in Drosophila and mammalian cells have identified HH signaling to inhibit autophagy. 22 Herein, along with SHH signaling, our results attribute novel role for TLR2-dependent WNT signaling pathways in suppressing autophagy.
Post-transcriptional regulation of gene expression by miRNAs is fundamental for regulation of signaling pathways and consequently for calibrating the immune responses. 63 , 64 In the current context, Mir155 and Mir31 activate WNT-SHH pathway by targeting PP2A, a Ser/Thr phosphatase that determines GSK3B activity. We propose that infection-responsive, miRNA-mediated augmentation of WNT and SHH pathways are necessary for the sustained and long-term inhibition of autophagy of host cells that are constantly influenced by the dynamic cytokine milieu in vivo. In support of our proposition, B56-containing PP2A are implicated in the regulation of WNT and HH signaling in various cellular contexts. 34 - 38 Notably, pharmacological perturbation of PP2A inhibits autophagy in rat hepatocytes 2 and MTOR possesses inhibitory effects on PP2A. 65 , 66 We have previously demonstrated that Mir155 induces apoptosis during M. bovis BCG infection of macrophages. 60 Apoptosis of the pathogen-infected cell, in general, is assumed to help cross-priming and thus activation of adaptive immune response. However, Mycobacteria-mediated apoptosis of the macrophages is known to assist in dissemination of the bacteria and thus act as a means for bacterial replication. 67 Fitting well into the known concepts of interplay between autophagy and apoptosis, 68 our data suggest that M. bovis BCG induces Mir155 to inhibit autophagy and meanwhile promote macrophage apoptosis. Interestingly, activation of calpain during apoptosis could induce the cleavage of ATG5 and thus lead to inhibition of autophagy. 69 However, in the current study, inhibition of IFNG-induced autophagy by M. bovis BCG was significant even in the presence of ALLN, a calpain inhibitor and Z-VAD, a broad-spectrum caspase inhibitor ruling out such possibilities (Fig. S6A–S6C). Further, corroborating these results, Mir155-transfected macrophages markedly reduced IFNG-induced autophagy in presence of ALLN and Z-VAD (Fig. S6D and S6E). This suggests that Mir155-mediated apoptosis may not contribute to inhibition of autophagy in case of macrophages infected with M. bovis BCG. Previously, we have shown that SHH-responsive Mir31 acts as a negative regulator of TLR2 responses to mycobacteria. 20 However, current evidence has identified a novel function of Mir31 to inhibit autophagy by inducing SHH signaling, thus highlighting the ability of individual miRNAs to affect the expression of many proteins that could even constitute the same pathway. 70 Thus, our data advocates novel functions for Mir155 and Mir31 during the immune response to pathogens.
ALOXs play a significant role in determining the outcome of ensued immunity during infection. ALOXs inhibit Th1 cytokines and promote Th2 responses. Of note, ALOXs are targeted by many pathogens to evade immune surveillance. 71 , 72 A report also suggests that alox5-deficient mice are less prone to mycobacterial infections and exhibit rapid clearance of pathogens from the lungs. 73 We observed an indispensable role for the infection-induced ALOX5 and ALOX15 in controlling IFNG-induced autophagy. Albeit the precise mechanism involved in ALOX-LX-mediated inhibition of autophagy requires further study, we found that ALOXs mediate the downregulation of JAK-STAT signaling. STAT1-induced cytokine signaling in macrophages induces antibacterial effects including autophagy. 44 Interestingly, infection of macrophages with M. bovis BCG induces the expression of many inhibitors of cytokine signaling such as SOCS and PIAS, which may support the current observations. 18 , 74 , 75 Alternatively, ALOX-LX signaling might activate PI3Ks, 76 veritable inhibitors of autophagy.
Interestingly, similar to M. bovis BCG-induced responses, we found that virulent M. tuberculosis could inhibit IFNG-induced autophagy via the MTOR-Mir155-Mir31-WNT-SHH-ALOX pathway. However, M. tuberculosis-mediated inhibition was less effective when compared with M. bovis BCG. These results are in agreement with the recent reports, which suggest that M. tuberculosis can inhibit autophagy. 13 , 16 , 77
In summary, our results have implicated multiple host factors such as Mir155-Mir31 and WNT-SHH-ALOXs to orchestrate the microbe-induced evasion of autophagy (Fig. 9B). These molecular regulators bear potential importance in disease control by aiding the search for effective drugs and therapeutics.
Primary macrophages were isolated from peritoneal exudates of C57BL/6J (000664), C3H/HeJ (000659) and tlr2-KO (004650) mice that were purchased from The Jackson Laboratory and maintained in the Central Animal Facility, Indian Institute of Science. Briefly, mice were intraperitoneally injected with 1 ml of 8% Brewer thioglycollate. After 4 d of injection, mice were sacrificed and peritoneal cells were harvested by lavage from peritoneal cavity with ice-cold PBS. The cells were cultured in DMEM (Gibco-Invitrogen, 12100-061) containing 10% FBS (Gibco-Invitrogen, 10270-106) for 6 to 8 h and adherent cells were used as peritoneal macrophages. All studies involving mice were performed after the approval from the Institutional Ethics Committee for animal experimentation as well as from Institutional Biosafety Committee. mir155-KO mice and littermate controls were generated as described earlier. 78 , 79 BMDMs were isolated from femurs and tibias obtained from WT or mir155-KO mice and cultured in DMEM medium containing 30% L-929 fibroblast-conditioned medium for 7 d. The purity of these cells was confirmed by F4/80 staining using FACS and was > 95%. All transfection studies were performed in murine RAW 264.7 macrophage-like cells obtained from the National Center for Cell Sciences, Pune, India. M. bovis BCG Pasteur 1173P2 was obtained from Pasteur Institute, Paris, France and S. aureus, K. pneumoniae, S. flexneri, L. monocytogenes, and E. coli were obtained from The Microbial Type Culture Collection and Gene Bank (MTCC), Chandigarh, India. Bacteria were grown to mid-log phase and used at 10 multiplicity of infection (MOI) in all the experiments unless mentioned otherwise.
General laboratory chemicals were obtained from Sigma-Aldrich or Merck. IFNG was purchased from PeproTech Inc. (315-05). Anti-ACTB (A3854) and anti-AGO2 (SAB4200085) antibodies was purchased from Sigma-Aldrich. Anti-MAP1LC3I/II (4108), anti-Ser-33/37/Thr-41 phospho CTNNB1 (9561), anti-CTNNB1 (9562), anti-WNT5A (2392), anti-SHH (2207), anti-GLI1 (2534), anti-NUMB (2756), anti-Ser9 phospho GSK3B (9322), anti- GSK3B (5195), anti-ALOX5 (3289), anti-Ser2448 phospho MTOR (2971), anti-MTOR (2972), anti-Thr389 phospho RPS6KB2 (9205), anti-Thr37/46 phospho EIF4EBP1 (9455), anti-RPTOR (2280), anti-RICTOR (2114), anti-Ser555 phospho ULK1 (5869), anti-Ser757 phospho ULK1 (6888), anti-H3K27me3 (9733), anti-H3K4me3 (9751), anti-ASH2L (5019), anti-Tyr705 phospho STAT3 (9131) and anti-Tyr701 phospho STAT1 (9171) were purchased from Cell Signaling Technology. Anti-PPP2R5A (sc-136045) and anti-ALOX15 (sc-32940) were purchased from Santa Cruz Biotechnology, Inc. HRP conjugated anti-rabbit IgG (111-035-045) and anti-mouse IgG (115-001-003) were obtained from Jackson ImmunoResearch.
In all experiments, cells were treated with the given inhibitor for 1 h before experimental treatments at the following concentrations: Rapamycin (100 nM) (Calbiochem, 553210), FH535 CTNNB1 and TCF (transcription factor family) inhibtor (15 μM) (Calbiochem, 219330), IWP-2 (5 μM) (Calbiochem, 681671), cyclopamine (10 μM) (Calbiochem, 239803), betulinic Acid (10 μM) (Calbiochem, 200498), AA-861 (10 μM) (Sigma-Aldrich, A3711), PD146176 (10 μM) (Sigma-Aldrich, P4620), AG490 (10 μM) (Calbiochem, 658401), manumycin A (10 μM) (Calbiochem, 444170), propranolol (10 μM) (Calbiochem, 537075), LY294002 (50 μM) (Calbiochem, 440202), ALLN, calpain inhibitor I (10 μM) (Sigma, A6185) and Z-VAD (20 μM) (Calbiochem, 620610). DMSO at 0.1% concentration was used as the vehicle control. In all experiments involving pharmacological reagents, a tested concentration was used after careful titration experiments assessing the viability of the macrophages using the MTT (3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) assay. Pam3CSK4 (1 µg/ml) (Imgenex, IMG 2201) and PIM2 (4 µg/ml) (Colorado State University) were also used.
The full-length 3′UTR of Ppp2r5a was PCR amplified and cloned into a pmirGLO vector using the restriction enzyme pair SacI and XbaI. The Mir155 or Mir31 binding sites in the 3′UTR of Ppp2r5a were mutated by nucleotide replacements through site-directed mutagenesis using the megaprimer inverse PCR method. The forward/reverse primer comprised the desired mutation and respective reverse primers were used to generate the megaprimer (listed in Table S1). The megaprimer was in turn used to amplify the miRGLO Ppp2r5a plasmid and generate the Mir155Δ and Mir31Δ 3′UTR mutants. The double mutant plasmid was generated utilizing a megaprimer mutant for Mir31-binding sites on the Mir155Δ mutant Ppp2r5a plasmid background.
Murine RAW 264.7 macrophage cells were transfected with 5 µg of pEGFP-LC3 using low m.w. polyethylenimine (Sigma-Aldrich, 40872-7) in all the experiments involving immunofluorescence. Transiently transfection of RAW 264.7 macrophages with dominant negative (DN) mutant forms of TLR2 and MTOR or overexpression constructs of MTOR, WNT5A, SHH, PPP2R5A, ALOX5, ALOX15, WT Ppp2r5a 3′UTR, Mir155Δ Ppp2r5a 3′UTR, Mir31Δ Ppp2r5a 3′UTR, Mir155Δ Mir31Δ Ppp2r5a 3′UTR, β-Galactosidase (GLB1) was performed using low m.w. polyethylenimine. In case of experiments involving siRNA and mimics, RAW 264.7 macrophage cells were transfected with 100 nM siRNA or miRNA mimic unless mentioned otherwise. Mtor (M-065427-00), Wnt5a (M-065584-00), Ctnnb1 (M-040628-00), Shh (M-042008-01), Gli1 (M-047917-01), Ppp2r5a (M-040213-00), Alox5 (M-065427-00), Alox15 (M-061509-01), Rptor (M-058754-01), Rictor (M-064598-01), nontargeting siRNA (D-001210-01-20) and siGLO Lamin A/C (D-001620-03-05) were obtained from Dharmacon as siGENOME™ SMARTpool reagents, which contain a pool of four different double-stranded RNA oligonucleotides. Mir155 (C-310430-07), Mir31 (C-300507-05) and control mimics (CN-001000-01-05) were purchased from Dharmacon and Mir155 and Mir31 siRNAs (4390771) and control siRNA (4611G) were purchased from Ambion. Transfection efficiency was more than 50% in all the experiments as determined by counting the number of siGLO Lamin A/C positive cells in a microscopic field using fluorescent microscope. In all cases, 48 h post-transfection, the cells were treated or infected as indicated and processed for analysis.
RAW 264.7 macrophages were seeded on to coverslips and treated as indicated. The cells were fixed with 3.7% paraformaldehyde for 15 min at room temperature and the coverslips were mounted on a slide with glycerol. Confocal images were taken on a Zeiss LSM 710 Meta confocal laser scanning microscope (Carl Zeiss AG) using a plan-Apochromat 63×/1.4 Oil DIC objective (Carl Zeiss AG) and images were analyzed using ZEN 2009 software.
Macrophages were treated or infected as indicated and total RNA from macrophages was isolated by TRI reagent (Sigma-Aldrich, T9424). Two micrograms of total RNA was converted into cDNA using a First Strand cDNA synthesis kit (Bioline, BIO-65050). Quantitative real-time RT-PCR was performed using SYBR Green PCR mixture (KAPA Biosystems, KM4101) for quantification of the target gene expression. All the experiments were repeated at least 3 times independently to ensure the reproducibility of the results. Gapdh was used as internal control. The primers used for quantitative real-time RT-PCR amplification are summarized in Table S1.
For detection of Mir155 and Mir31, total RNA was isolated from infected or treated macrophages using the TRI reagent. Quantitative real-time RT-PCR for Mir155 (assay ID 002571), Mir31 (assay ID 000185) and MIR106a (assay ID 002459) was performed using TaqMan miRNA assays (Life Technologies, 4427975) as per manufacturer’s instructions. U6 snRNA was used for normalization.
Infected or treated macrophages were lysed in RIPA buffer constituting 50 mM TRIS-HCl (pH 7.4), 1% NP-40, 0.25% Sodium deoxycholate, 150 mM NaCl, 1 mM EDTA, 1 mM PMSF, 1 μg/ml of each aprotinin, leupeptin, pepstatin, 1 mM Na3VO4 and 1 mM NaF. An equal amount of protein from each cell lysate was resolved on a 12% SDS-polyacrylamide gel and transferred to polyvinylidene difluoride membranes (Millipore, IPVH00010) by the semidry transfer (Bio-Rad, 170-3940) method. Nonspecific binding was blocked with 5% nonfat dry milk powder in TBST [20 mM TRIS-HCl (pH 7.4), 137 mM NaCl, and 0.1% Tween 20] for 60 min. The blots were incubated overnight at 4 °C with primary antibody followed by incubation with anti-rabbit-HRP or anti-mouse-HRP secondary antibody in 5% BSA for 2 h. After washing in TBST, the immunoblots were developed with enhanced chemiluminescence detection system (Perkin Elmer, NEL105001EA) as per the manufacturer’s instructions. ACTB was used as loading control. For probing another protein in the same region of the PVDF membrane, the blots were stripped in the stripping buffer (62.5 mM TRIS-HCl (pH 6.8), 2% SDS and 0.7% β-mercaptoethanol) at 60 °C on a shaker, blocked with 5% nonfat dry milk powder and probed with antibodies as mentioned above.
Cells were lysed in Reporter lysis buffer (Promega, E4030) and assayed for luciferase activity using Luciferase Assay Reagent (Promega, E4030) as per the manufacturer’s instructions. The results were normalized for transfection efficiencies measured by GLB1 activity. O-nitrophenol β-D-galactopyranoside (Hi-Media, RM582) was utilized for the GLB1 assay.
Chromatin immunoprecipitation (ChIP) assays were performed using a protocol provided by Upstate Biotechnology, USA, with certain modifications. Briefly, macrophages were fixed with 1.42% formaldehyde for 15 min at room temperature followed by inactivation of formaldehyde with addition of 125 mM glycine. Nuclei were isolated from macrophages using modified RIPA buffer containing 1% Triton X-100 and chromatin was sheared. Chromatin extracts containing DNA fragments with an average size of 500 bp were immunoprecipitated using anti-CTNNB1 or anti-GLI1 or anti-H3K27me3 or anti-H3K4me3 or anti-ASH2L or rabbit preimmune sera. Purified DNA was analyzed by quantitative PCR. All results were normalized to amplification of 28S rRNA. All ChIP experiments were repeated at least three times and the primers utilized are listed in Table S1.
Immunoprecipitation (IP) assays were performed using a protocol provided by Millipore, USA, with certain modifications. Briefly, macrophages were gently lysed in ice-cold RIPA buffer on an orbital shaker. The cell lysates were incubated with anti-PPP2R5A, anti-p-GSK3B, anti-GSK3B or rabbit preimmune sera at 4 °C for 2 h on an orbital shaker. The immunocomplexes were captured using Protein A (Bangalore Genei, 62211018005A) agarose at 4 °C for 2 h. The beads were harvested, washed, and boiled in 5× Laemmli buffer for 10 min. The samples were separated by SDS–PAGE and further subjected for immunoblotting.
Infected BMDMs were lysed in 300 µl of complete polysomal lysis buffer (5 mM MgCl2, 0.1 M KCl, 0.5% NP40, 0.01M HEPES pH 7.5). Total cell lysate (200 µg) was diluted to 500 µl using complete polysomal lysis buffer for IP. 50 µl of anti-mouse IgG precleared lysate was used as Input for the experiment and total RNA was isolated as described earlier. The rest of the precleared lysate was incubated with 1 µg anti-mouse IgG or anti-AGO2 prebound Protein A beads overnight. Further, beads were washed in complete polysomal lysis buffer and 25% of the beads were eluted in 5× Laemmli buffer for immunoblotting with anti-AGO2. The remaining beads were eluted in TRI reagent and processed for RNA isolation as described earlier. 500 ng of the RNA was converted into cDNA and quantitative real-time RT-PCR was performed to analyze Ppp2r5A. Gapdh from Input was utilized for normalization. TaqMan miRNA assays were used to quantify Mir155, Mir31, and Mir106a levels. U6 from Input was utilized for normalization.
Two million/well C3H/HeJ mouse macrophages were seeded in 12-well plates and treated with respective inhibitors for 1 h. After 2 h of the bacterial infection, the cells were thoroughly washed with PBS and treated with gentamycin (10 µg/ml) for 2 h to kill the extracellular bacteria. The cells remained infected for further 10 h and IFNG (200 U/ml) treatment was performed for next 18 h. Macrophages were thoroughly washed with PBS and lysed in 0.1% Triton X-100. For Ct based CFU analysis, intracellular bacteria were collected and boiled at 95 °C for 20 min and sonicated (40% duty, 20 pulses). The supernatant was utilized for quantitative real-time RT-PCR using bacterial specific primers as listed in Table S1. Known CFUs of bacteria were utilized to generate the standard curves for CFU vs Ct.
Levels of significance for comparison between samples were determined by the Student t test distribution and one-way ANOVA. The data in the graphs are expressed as the mean ± SE for 5 or 6 values from 3 independent experiments and P values < 0.05 were defined as significant. Graphpad Prism 5.0 software (Graphpad Software) was used for all the statistical analysis.
No potential conflicts of interest needed to be disclosed.
We thank the Central Animal facility, Indian Institute of Science (IISc) for providing mice for experimentation and Samrajyam Nara and Pannaga of the MCB, IISc confocal facility for generous help. We acknowledge Dr Devram Sampat Ghorpade, Dr Kushagra Bansal, Shambhuprasad TK, Jamma Trinath and Dr Yeddula Narayana for critical comments and timely help during the current course of investigation. We sincerely thank Dr Douglas Golenbock, University of Massachusetts Medical School, Worcester, MA USA for the TLR2 DN construct, Dr Peter Houghton, The Research Institute at Nationwide Children’s Hospital, OH USA for MTOR DN and overexpression constructs and Dr Gautam Das and Dr EH Baehrecke, University of Massachusetts Medical School, Worcester, MA USA for the pEGFP-LC3 construct. Dr Roel Nusse, Stanford University Medical Center, CA USA; Dr Andrew McMahon, Harvard University, MA USA; Dr Stefan Strack, University of Iowa College of Medicine, IA USA and Dr Colin D Funk, Queen’s University, Kingston, CA are acknowledged for kind gifts of WNT5A, SHH, PPP2R5A, ALOX5, and ALOX15 overexpression constructs respectively. The Alox5 promoter luciferase construct was provided by Dr Dieter Steinhilber, Institute of Pharmaceutical Chemistry, Frankfurt, Germany. We thank Dr Iain McInnes, University of Glasgow, Scotland UK, for sharing mir155-KO mice. We acknowledge Dr Usha Vijayraghavan, Dr Annapoorni Rangarajan, Dr Kumaravel Somasundaram, and Dr Saibal Chatterjee for their kind help.
This study is supported by funds from the Department of Biotechnology (DBT), Department of Science and Technology (DST), Council for Scientific and Industrial Research (CSIR) and Indian Council of Medical Research (ICMR) and Government of India and the Indo-French Center for Promotion of Advanced Research (IFCPAR/CEFIPRA). Infrastructure support from ICMR (Center for advanced study in Molecular Medicine), DST (FIST) and UGC (special assistance) (KNB); fellowship from IISc (SH) are acknowledged.