The rLrp of Mycobacterium tuberculosis inhibits proinflammatory cytokine production and downregulates APC function in mouse macrophages via a TLR2-mediated PI3K/Akt pathway activation-dependent mechanism

Abstract

We demonstrate that Mycobacterium tuberculosis recombinant leucine-responsive regulatory protein (rLrp) inhibits lipopolysaccharide (LPS)-induced tumor necrosis factor alpha (TNF-α), interleukin-6, and interleukin-12 production and blocks the nuclear translocation of subunits of the nuclear-receptor transcription factor NF-κB (Nuclear factor-kappa B). Moreover, rLrp attenuated LPS-induced DNA binding and NF-κB transcriptional activity, which was accompanied by the degradation of inhibitory IκBα and a consequent decrease in the nuclear translocation of the NF-κB p65 subunit. RLrp interfered with the LPS-induced clustering of TNF receptor-associated factor 6 and with interleukin-1 receptor-associated kinase 1 binding to TAK1. Furthermore, rLrp did not attenuate proinflammatory cytokines or the expression of CD86 and major histocompatibility complex class-II induced by interferon-gamma in the macrophages of Toll-like receptor 2 deletion (TLR2^−/−) mice and in protein kinase b (Akt)-depleted mouse cells, indicating that the inhibitory effects of rLrp were dependent on TLR2-mediated activation of the phosphatidylinositol 3-OH kinase (PI3K)/Akt pathway. RLrp could also activate the PI3K/Akt pathway by stimulating the rapid phosphorylation of PI3K, Akt, and glycogen synthase kinase 3 beta in macrophages. In addition, 19 amino acid residues in the N-terminus of rLrp were determined to be important and required for the inhibitory effects mediated by TLR2. The inhibitory function of these 19 amino acids of rLrp raises the possibility that mimetic inhibitory peptides could be used to restrict innate immune responses in situations in which prolonged TLR signaling has deleterious effects. Our study offers new insight into the inhibitory mechanisms by which the TLR2-mediated PI3K/Akt pathway ensures the transient expression of potent inflammatory mediators.

Introduction

Tuberculosis remains a serious global health problem because we lack a detailed understanding of its pathogenesis. During the early stages of Mycobacterium tuberculosis infection, the extent of bacterial survival and proliferation primarily depends on the efficacy of the innate immune response, for which macrophages are the main effector cells.^1,2 Upon phagocytosis of M. tuberculosis, macrophages may activate several antimicrobial mechanisms to control intracellular replication of the bacilli.^3,4 Elicitation of these antimycobacterial responses by macrophages during the innate phase of activation plays a crucial role in determining the outcome of M. tuberculosis infection.⁵ Among these responses, the production of interleukin-12 (IL-12) and tumor necrosis factor alpha (TNF-α), which are proinflammatory cytokines, is critical for defense against mycobacterial infection.^6,7,8,9 TNF-α is important in combatting the bacilli because it can induce granuloma formation and cytotoxicity.^10,11 In contrast, IL-12 can activate a protective immune response involving the Th1 helper T-cells.^12,13 Regulation of proinflammatory responses often involves different signaling cascades, and the pathogenic M. tuberculosis bacterium has evolved several mechanisms to suppress IL-12 and TNF-α production by modulating these signaling pathways to support its long-term survival and persistence inside the host.^14,15,16

Pathogens have conserved molecular patterns termed pathogen-associated molecular patterns (PAMPs).^17,18 Many PAMPs induce signaling through the Toll-like receptors (TLRs) expressed on innate immune cells. Lipopolysaccharide (LPS), which is a component of the Gram-negative bacterial wall, is recognized by TLR4, whereas peptidoglycan (PGN), which is another bacterial wall component, stimulates TLR2. Humans have 10 different TLRs; TLR4 and TLR2 are expressed on the extracellular membrane, but TLR3, 7, and 9 localize inside the cell.^19,20,21,22 There is a short intracellular Toll/interleukin-1 receptor (IL-1R) (TIR) domain in the TLRs that recruits adapter molecules and induces downstream activation.²³ TLR2 and TLR4 both interact with MyD88, an adapter molecule that subsequently recruits members of the interleukin-1 receptor-associated kinase (IRAK) family.²⁴ IRAK1 and IRAK4 are serine threonine kinases that can phosphorylate and activate TNF receptor-associated factor 6 (TRAF6). TRAF6 activates the mitogen-activated protein kinase (MAPK) kinase, which in turn activates the inhibitor of kappa B kinase (IKK) complex, resulting in NF-κB (Nuclear factor-kappa B) activation.²⁵

Until now, the underlying cellular mechanisms that directly control anti-inflammatory versus pro-inflammatory cytokine production after TLR stimulation have been unknown, however, the TLR-signaling pathway can activate phosphatidylinositol 3-OH kinase (PI3K) to limit the production of TNF-α and IL-12.^26,27 Those findings suggest the critical involvement of the PI3K pathway in the control of pro- and anti-inflammatory cytokine production. Here, we describe how TLR signaling affects the PI3K pathway to determine whether a central effector molecule can mediate the ability of this pathway to differentially dictate the host inflammatory response.

It is known that macrophages, which phagocytose and degrade invading microorganisms to actively participate in innate immunity, are an early barrier against M. tuberculosis. In addition, they present microorganism-derived peptides, which are processed by intracellular compartments, through the major histocompatibility complex (MHC) to T lymphocytes. This promotes the adaptive immune response. In protective immunity, interferon-gamma (IFN-γ) plays a critical role against M. tuberculosis by enhancing both the antigen (Ag)-presenting and microbicidal functions of macrophages.^6,28,29,30 M. tuberculosis can survive and replicate within compartments of macrophages called phagocytic vacuoles.^31,32 Once inside macrophages, pathogens use many strategies to counteract host immune responses. For example, they can reduce T-cell-mediated immune responses by diminishing or abrogating their Ag presentation capacity.^33,34,35 Recently, it was demonstrated that prolonged exposure to the M. tuberculosis lipoproteins LpqH (19 kDa; Rv3763),^36,37 LprG (Rv1411c),³⁸ and LprA³⁹ inhibits Ag processing and presentation and MHC class II (MHC-II) protein expression, which may allow certain pathogens to evade immune surveillance and promote chronic infection.^40,41

In infected macrophages, IL-12p40 gene transcription is inhibited by M. tuberculosis,¹³ and this affects T cell responses to M. tuberculosis infection. Recent studies suggest that ESAT-6 inhibits IL-12p40 expression and secretion in macrophages., Recombinant ESAT-6 has also been shown to inhibit the secretion of IL-12 induced by multiple TLR agonists via binding to TLR2 and blocking the TLR signaling pathway.⁴² Lrp (Rv2779c) is a feast/famine regulatory protein from M. tuberculosis that is known to be a member of the leucine-responsive regulatory protein/asparagine synthase C family (Lrp/AsnC) of transcriptional regulators. In addition, Lrp/AsnC proteins take part in various metabolic processes in fungi and bacteria.^43,44 Here, our data show that recombinant leucine-responsive regulatory protein (rLrp) dampens TLR signaling by interfering with the assembly of the MyD88-dependent signaling scaffold. RLrp binding to TLR2-activated PI3K-protein kinase b (Akt)-dependent signaling and inhibited LPS-induced cytokine production and NF-κB activation by blocking IRAK1, TRAF6, and TAK1 signalosome formation. We demonstrate that direct interaction between TLR2 and a 19-peptide region of rLrp on the extracellular side of the plasma cell membrane inhibits the assembly of TLR signaling molecules that are inside the cell and required for the activation of innate immune responses.

Materials and methods

Mice and cell lines

C57BL/6 mice were purchased from Vital River Company. TLR2^−/− and TLR4^−/− mice were purchased from The Jackson Laboratory. All mice were raised in the Animal Center of Peking University (Beijing, China). All animal experiments were performed in accordance with protocols approved by the Institutional Animal Care and Use Committee at Peking University. A mouse macrophage cell line, RAW264.7, and a human embryonic kidney cell line, HEK293, were originally obtained from the American Type Culture Collection. HEK293-Vector (transfected with empty pEGFP-N3 plasmid), HEK293-TLR2 (transfected with pEGFP-N3-TLR2 plasmid), and HEK293-TLR4 (transfected with pEGFP-N3-TLR4 plasmid) stable cell lines were maintained in our laboratory. Cell lines were cultured in DMEM (Thermo Fisher Scientific, Waltham, MA, USA) containing 10% fetal bovine serum (FBS) (Gibco BRL, Gaithersburg, MD, USA), 100 U mL⁻¹ penicillin, and 100 μg mL⁻¹ streptomycin. Primary cells were cultured in RPMI 1640 medium (Thermo Fisher Scientific) containing 10% FBS, 100 U mL⁻¹ penicillin, and 100 μg mL⁻¹ streptomycin at 37 °C in a humidified incubator containing 5% CO₂.

Cloning, expression, and purification of rLrp

The gene encoding rLrp was amplified by PCR based on the genomic DNA sequence of M. tuberculosis H37Rv. M. tuberculosis H37Rv was obtained from the Beijing TB Therapeutic Tumor Institute. PCR was carried out using SuperStar PCR Mix (Genstar Biosolutions Company, Beijing, China) in a C1000 Thermal Cycler (Bio-Rad Laboratories, Hercules, CA, USA) The PCR product was then cloned into the prokaryotic expression plasmid pET28a(+) (Novagen, Darmstadt, Germany) as previously described.⁴⁵ rLrp was purified by affinity chromatography using a His-Trap column (GE Healthcare Bio-Science AB, Uppsala, Sweden) and dialyzed against phosphate-buffered saline (PBS) for use in subsequent testing of its ability to induce cytokine production by macrophages.

Measurement of cytokine levels

Cytokines were measured in the supernatants of RAW264.7 cells, peritoneal macrophages, and T lymphocytes using the ELISA MAX Deluxe kit (Biolegend, San Diego, CA, USA). RAW264.7 cells were pretreated with wortmannin, an inhibitor of PI3K (IPA1003; Gene Operation, Ann Arbor, Michigan, USA), or an Akt inhibitor (A6730; Sigma-Aldrich, St Louis, MO, USA) for 30 min at 37 °C followed by treatment with 5 μg mL⁻¹ rLrp.

Isolation of mouse peritoneal macrophages

C57BL/6, TLR2^−/−, and TLR4^−/− mice were killed with ether anesthesia and cervical dislocation. Cells were aseptically isolated and cultured as previously described.⁴⁶ Non-adherent cells were removed by washing twice with medium, and then the adherent cells were treated with rLrp (5 μg mL⁻¹), LPS (1 μg mL⁻¹) (L4391; Sigma-Aldrich), or PGN (10 μg mL⁻¹) (69554; Sigma-Aldrich).

Cellular protein preparation and western blot analysis

RAW264.7 cells were stimulated for various periods of time and washed twice with cold PBS. The cells were lysed with cell lysis buffer (AidLab, Beijing, China) supplemented with a proteinase inhibitor cocktail and phosSTOP (Roche Molecular Biochemicals, Indianapolis, IN, USA) and centrifuged for 15 min at 4 °C at 12 000g. Protein concentration was measured using the Bradford method. For subcellular localization analysis of NF-κB, cell pellets were processed using the NE-PER Nuclear and Cytoplasmic Extraction kit (Pierce, Rockford, IL, USA). Equal amounts of protein were resolved via sodium dodecyl sulfate–polyacrylamide gel electrophoresis and then transferred to an Immobilon-P membrane (Millipore, Bedford, MA, USA). After blocking with 5% nonfat milk or bovine serum albumin (BSA) in Tris-buffered saline Tween (TBST) (pH 7.4, 0.05% v/v Tween-20), each membrane was incubated overnight at 4 °C with a primary antibody: anti-FLAG (F1804; Sigma-Aldrich), rabbit anti-NF-κB p65 (sc-109; Santa Cruz Biotechnology, Dallas, TX, USA), protein A/G PLUS-Agarose immunoprecipitation reagent (sc-2003), rabbit anti-hemagglutinin (HA)-probe (sc-805), rabbit anti-TRAF6 (sc-7221), rabbit anti-TAK1(sc-7162), rabbit anti-histone H3 (9715; Cell Signaling Technology, Beverly, MA, USA), rabbit anti-Akt (#4685; Cell Signaling Technology), rabbit anti- phospho-Akt (#4060), rabbit anti-PI3K (#4257), rabbit anti-phospho-PI3K (#4228), rabbit anti-phospho-IκBα (#2859), rabbit anti-phospho-GSK-3β (#5558), rabbit anti-LC3A/B (#12741), rabbit anti-IRAK1 (#4504), rabbit anti-phospho-TAK1 (#4508), rabbit anti-IκKα (9242), rabbit anti-phospho-IκKα (S32; 2859; Cell Signaling Technology), or mouse anti-β-actin (M20010; Abmart, Arlinton, MA, USA). After washing with TBST, the membranes were incubated for 1–2 h at 37 °C with the appropriate HRP-conjugated anti-mouse IgG or anti-rabbit IgG secondary Ab (1:10 000; Santa Cruz Biotechnology) followed by detection with ECL (CWBIO) and X-ray film (XAR5; Kodak).

Pull-down assay

Lysates from cells transfected with HEK293-Vector, -TLR2 or -TLR4 were prepared as described above. His-tagged rLrp (50 µg) immobilized on His-Trap beads was incubated with 500 μg of total cellular protein for 4 h at 4 °C in cell lysis buffer with gentle rotation. The beads were washed vigorously and boiled in 5× Laemmli buffer for 10 min. Pulled-down proteins were detected by western blotting with rabbit anti-GFP (E022200; EarthOx, San Francisco, CA, USA).

TLR binding assay

Cell lines expressing HEK293-vector, HEK293-TLR2, and HEK293-TLR4 were seeded overnight on poly-L-lysine-treated coverslips and then incubated for 30 min at 37°C with rLrp (5 μg mL⁻¹). The cells were washed with PBS, fixed for 15 min in 4% paraformaldehyde, and then stained during sequential 1-h incubations with mouse anti-His (Santa Cruz Biotechnology) and PE-conjugated goat anti-mouse IgG (BioLegend). Between each staining step, the cells were washed three times with PBS. After staining, the cells were mounted on slides using Mowiol solution (Sigma-Aldrich) containing 100 ng mL⁻¹ 4,6-diamidino-2-phenylindole (DAPI) and observed using a 63× oil objective on an Axiovert 200M microscope (Carl Zeiss, Inc., Munich, Germany). For flow cytometry, the cells were collected after staining, and 10 000 total events per sample were analyzed using a FACScan calibrator cytometer (BD Corporation, San Jose, CA, USA).

Electrophoretic mobility shift assay (EMSA) for NF-κB

Nuclear extracts (10 mg) prepared from various experimental groups were incubated for 30 min at room temperature with 1 ng of digoxigenin (No. 03353575910; Roche) and end-labeled duplex oligodeoxyribonucleotides containing the NF-κB binding region (5′-TTGTTACAAGGGACTTTCCGCTGGGGACTTTCCAGGGAGGCGTGG-3′ and 5′-CCACGCCTCCCTGGAAAGTCCCCAGCGGAAAGTCCCTTGTAACAA-3′), and EMSA was performed as previously described.⁴⁷

Reporter assay

To evaluate NF-κB activity in various experimental groups, we used a luciferase NF-κB reporter construct. The pSV-β-galactosidase plasmid (Promega, Madison, WI, USA) was used to normalize transfection efficiency. RAW264.7 cells were cotransfected with 3xIgk-ConA-Luc (NF-κB-luciferase reporter plasmid) and either a pSV-β-galactosidase plasmid or pConA (vector control) using Lipofectamine 2000 reagent (Invitrogen, Carlsbad, CA, USA). After 24 h of transfection, cells were either left untreated or pretreated with rLrp (3 mg mL⁻¹) for 1 h followed by stimulation with LPS (3 mg mL⁻¹) for another 1 h. Cells were lysed and assayed for luciferase activity using a luciferase reporter assay system (Promega), and luciferase activity was monitored with a luminometer (Turner Designs, Sunnyvale, CA, USA). β-galactosidase expression was measured in a β-galactosidase ELISA (Roche Diagnostics, Indianapolis, IN, USA). Luciferase activity values were normalized to the transfection efficiency as measured by the β-galactosidase assay (Roche Diagnostics). The results are expressed as relative NF-κB activity of the test samples compared with activity of the pConA-transfected samples after normalizing for β-galactosidase activity and protein concentration.

Short hairpin (shRNA) interference

A lentivirus-based shRNA system (Dr. Chen Zhang of Peking University) was used to functionally silence genes in RAW246.7 cells. To generate lentiviruses, 293T cells (5 × 10⁵/well) were seeded in 6-well culture plates and cotransfected with 1.5 μg of pLKO.1-shRNA or mock vector together with packing vectors (1 μg of RRE, 1 μg of Rev, and 0.5 μg of VSVG) using polyethylenimine reagent (P3143; Sigma-Aldrich). Medium was refreshed 18 h after transfection. Supernatants were collected 48 h after transfection, filtered through a 0.45-μm membrane, and used to infect cells. RAW264.7 cells (4 × 10⁵/well) were seeded in 6-well culture plates, infected with the supernatants in the presence of Polybrene (TR-1003, Millipore) (8 μg mL⁻¹), and selected using puromycin (2 μg mL⁻¹) to generate stable transformants.

Subcellular localization of NF-κB

RAW264.7 cells (2 × 10⁵/well) were seeded in 12-well culture plates with glass cover slips, treated directly with rLrp (5 μg mL⁻¹) for 0, 15, 30, or 60 min after 30-min pretreatment with LPS, and washed twice with PBS. The cells were then fixed for 5 min on ice in cold methanol. To reduce background staining, the cells were incubated for 30 min in PBS containing 5% BSA before incubation with rabbit anti-NF-κB p65. After incubation for 1.5 h, the cells were washed and incubated with FITC-conjugated goat anti-rabbit IgG for 1 h at room temperature in the dark. The cells were washed and mounted on slides using Mowiol solution containing 100 ng mL⁻¹ DAPI. NF-κB localization was observed using a Zeiss LSM 710 confocal microscope equipped with a 63×, 1.4-numerical aperture, oil immersion objective (Carl Zeiss, Inc.).

Flow cytometry

RAW264.7 cells were incubated in 24-well plates (2 × 10⁵ cells/well) with IFN-γ (0 or 20 ng mL⁻¹)(Peprotech, London, UK) or IFN-γ plus rLrp (0 or 5 μg mL⁻¹) for 12, 24, or 48 h and washed with cold PBS. Peritoneal macrophages were incubated in 24-well plates (2 × 10⁵ cells/well) with IFN-γ (0 or 20 ng mL⁻¹) or IFN-γ plus rLrp (0 or 5 μg mL⁻¹) for 48 h and washed with cold PBS. The cells were stained with PE-CY5-conjugated anti-mouse MHC-II (eBioscience, San Diego, CA, USA), FITC-conjugated anti-mouse CD86 (BioLegend), or the corresponding isotype control antibody (BioLegend) in the dark for 1 h. The stained cells were then washed three times with PBS and fixed with 1% paraformaldehyde. Signals from 10 000 cells were acquired on a FACScan cytometer, and data were analyzed using Summit software (DakoCytomation, Fort Collins, CO, USA).

Generation of an Lrp-deletion M. tuberculosis mutant

To generate an allelic exchange construct to replace the Lrp gene with a hygromycin-resistance cassette, 500-bp sequences flanking to the left and right of the M. tuberculosis Lrp gene were PCR-amplified using H37Rv genomic DNA as the template and using the following primer pairs: LF-LL (5′-CCATAAATTGGGTTCGCATCAGGTGGTAAGC-3′) and LF-LR (5′-CCATTTCTTGGGCGAGTCTTGTGTTCCGTTG-3′), and LF-RL (5′-CCATAGATTGGTCCCTGATCTTCGAGCATC-3′) and LF-RR (5′-CCATCTTTTGGTGTCCCGCCAGCACTTTC-3′). After cloning into the pEASY-T vector (TransGen Biotech Company, Beijing, China) and sequencing, the cloned PCR fragments were removed via primer-introduced restriction sites and cloned into the allelic exchange vector p0004S. The obtained plasmid, p0004S-LF, was then packaged into the temperature-sensitive phage phAE159, as previously described,¹ to yield the Lrp-knockout phage phAE-LF. Specialized transduction was performed as previously described.¹ The p0004S vector and phAE159 were kindly provided by Professor William R. Jacobs Jr. (Howard Hughes Medical Institute, Chevy Chase, MD, USA).

Statistical analysis

The results are presented as the mean ± SD of triplicate samples. Statistical testing was conducted using Student's t-test or analysis of variance (ANOVA) followed by Tukey's test using GraphPad Prism 5 (GraphPad Software, Inc., La Jolla, CA, USA). The protection results were evaluated using one-way ANOVA followed by Dunnett's test. For all tests, p < 0.05 was considered statistically significant.

Results

rLrp attenuates proinflammatory cytokines in LPS-treated macrophages in a TLR2-dependent manner

Given that Lrp-deficient M. tuberculosis does not attenuate macrophage activation like wild type M. tuberculosis (Supplementary Figure 1a and b), we tested the possibility that Lrp exerts attenuating effects on TLR signaling. First, we expressed epitope-tagged M. tuberculosis rLrp in Escherichia coli. Then, we used LPS to stimulate the mouse macrophage cell line RAW264.7 in the presence or absence of rLrp. The rLrp protein of M. tuberculosis was found to inhibit LPS-induced expression of the proinflammatory cytokines TNF-α, interleukin-6 (IL-6), and IL-12p40 in a dose-dependent manner (Figure 1a–c). Although the degree of inhibition by rLrp was maximal in the range of 5–20 μg mL⁻¹, for all of our subsequent experiments we used a concentration of 5 μg mL⁻¹ to take into account batch-to-batch qualitative variations in the purified recombinant protein. Following silencing of TLR2 with a short hairpin RNA (shTLR2; Figure 1d), TNF-α, IL-12p40, and IL-6 production were similar in LPS-stimulated RAW264.7 cells both with and without rLrp (Figure 1e–g). Next, we used peritoneal macrophages isolated from C57BL/6, TLR2^−/−, and TLR4^−/− mice to further analyze whether TLR2 mediates a reduction in the production of cytokines in LPS-stimulated macrophages treated with rLrp. The macrophages were stimulated with LPS with or without rLrp. RLrp significantly lowered proinflammatory cytokine expression in LPS-treated macrophages from C57BL/6 and TLR4^−/− mice (p < 0.001) but did not attenuate proinflammatory cytokine expression in LPS-treated macrophages from TLR2^−/− mice, indicating that the inhibitory effects of rLrp are dependent on TLR2 (Figure 1e–h).

Figure 1.

rLrp inhibits LPS-induced cytokine secretion in mouse macrophages through TLR2. (a, b, c) RAW264.7 cells were pretreated with the indicated concentrations of rLrp for 2 h at 37°C and then were treated with LPS (1 μg mL⁻¹) alone or with rLrp. Levels of TNF-α, IL-6, and IL-12p40 in the supernatants were measured by ELISA. (d) TLR depletion in RAW264.7 cells stably expressing shTLR2. β-actin was used as a loading control (e, f, g). The shTLR2 RAW264.7 cells were pretreated with rLrp (5 μg mL⁻¹) for 2 h at 37°C, and then the cells were incubated for 4 or 24 h with LPS (1 μg mL⁻¹). After incubation, supernatants were harvested and levels of TNF-α, IL-6, and IL-12p40 were measured by ELISA. (h) Peritoneal macrophages from C57BL/6, TLR2^−/−, or TLR4^−/− mice were treated with rLrp (5 μg mL⁻¹), LPS (1 μg mL⁻¹), or PGN (10 μg mL⁻¹), and levels of TNF-α, IL-6, and IL-12p40 in the supernatants were measured by ELISA. The data are representative of three different experiments. Significant differences were calculated using Student's t-test between different treatments and the medium (**p < 0.01; ***p < 0.001).

rLrp attenuates NF-κB activity in LPS-treated macrophages

It is known that NF-κB transcription factors can induce proinflammatory cytokines.^48,49 NF-κB/rel transcription factors bind to the NF-κB binding sites localized in the promoters of the IL-12p40⁵⁰ and TNF-α⁵¹ genes. Therefore, we measured NF-κB activity using a luciferase reporter construct in LPS-treated macrophages. At 24 h post-transfection with the NF-κB luciferase plasmid, we used LPS to stimulate macrophages with or without rLrp. As predicted, rLrp was found to attenuate LPS-stimulated NF-κB activity (Figure 2a) by three-fold. To further verify these observations, we measured the cognate DNA-binding activity of NF-κB by EMSA. The macrophages were pre-incubated with either LPS alone or LPS and rLrp, and then nuclear extracts were obtained and the DNA-binding activity of NF-κB was tested by EMSA using duplex oligonucleotides containing NF-κB binding sites. The DNA-binding activity of NF-κB was low in unstimulated macrophages (Figure 2b, lanes 1 and 2) and highest in macrophages stimulated with LPS at 30 min; however, complex formation was markedly reduced in LPS-stimulated cells treated with rLrp (Figure 2b, lanes 3 and 4 and lanes 5 and 6).

Figure 2.

LPS-induced NF-κB activity is inhibited by rLrp in macrophages. (a) RAW264.7 cells were transfected with either the vector control alone or the NF-κB luciferase reporter plasmid along with a pSV-β-galactosidase plasmid. After 24 h, the macrophages were treated for 1 h with either medium or rLrp (5 μg mL⁻¹) and then stimulated with LPS (1 μg mL⁻¹) for another 1 h. The transcriptional activity of NF-κB was measured by luciferase assays. The data are shown as fold inductions over basal activity and represent three independent experiments performed in triplicate. (b) Nuclear extracts prepared from RAW264.7 cells activated with LPS (1 μg mL⁻¹) in the absence or presence of rLrp (5 μg mL⁻¹) were subjected to EMSA. The data shown are representative of three independent experiments with similar results.

rLrp prevents the translocation of NF-κB subunits to the nucleus in LPS-treated macrophages in a TLR2-dependent manner

The p50 and p65 NF-κB/rel proteins are important transcription factors that are known to regulate IL-12p40 gene transcription^49,50 as well as the induction of TNF-α.⁵¹ Because rLrp inhibited NF-κB activity in LPS-stimulated macrophages (Figure 2), we next examined the nuclear levels of NF-κB p65 in these macrophages by western blotting using a specific antibody against the transcription factor. The western blots indicated that nuclear levels of NF-κB p65 were reduced in LPS-stimulated macrophages treated with rLrp compared with macrophages treated with LPS alone at 30 min (Figure 3a). In contrast, levels of the NF-κB p65 subunit were higher in the cytoplasm in the groups treated with rLrp and LPS than in the group treated with LPS alone (Figure 3a), indicating that rLrp may inhibit the translocation of the NF-κB p65 transcription factor from the cytoplasm to the nucleus. In LPS-treated cells, we use confocal microscopy to demonstrate the localization of NF-κB p65 to the nucleus. After 15 min of treatment with LPS, the nuclear levels of NF-κB p65 were very high, but after 1 h the levels decreased. In LPS-stimulated cells treated with rLrp, we found that NF-κB p65 was primarily located in the cytoplasm (Figure 3b).

Figure 3.

rLrp affects NF-κB subcellular location through TLR2. (a) RAW264.7 cells were pre-incubated with rLrp (5 μg mL⁻¹) for 2 h and then were treated with LPS (1 μg mL⁻¹) for 15 or 30 min. Cell lysates were prepared and used for western blot analysis to determine the effects of rLrp on the translocation of NF-κB from the cytoplasm to the nucleus. Histone H3 and β-actin were used as loading controls. Similar data were obtained in two independent experiments. (b) RAW264.7 cells were plated onto glass coverslips and incubated at 37 °C for 2 h with rLrp (5 μg mL⁻¹), and then the cells were treated with LPS (1 μg mL⁻¹) for 15 or 30 min. After washing, the cells were conjugated with anti-NF-κB for 1 h and then incubated for 30 min with FITC-conjugated anti-rabbit IgG. The cells were fixed with 4% paraformaldehyde and photographed with a fluorescence microscope (original magnification 63×). (c) RAW264.7 cells were incubated with LPS (1 μg mL⁻¹) or LPS with rLrp (5 μg mL⁻¹) for 0–60 min, cell lysates were prepared, and western blot analysis with anti-phospho-IκBα (p-IκBα) and anti-IκBα was performed to examine the phosphorylation of IκBα. Similar data were obtained in two independent experiments. (d) Confocal microscopy (original magnification 63×, 1.4-NA, oil-immersion objective) was used to determine the effects of rLrp on the translocation of NF-κB from the cytoplasm to the nucleus in shTLR2 RAW264.7 cells. The data are representative of three independent experiments.

In unstimulated macrophages, NF-κB is inhibited by its inhibitor IκBα in the cytosol. When cells are stimulated with LPS, the inhibitor of kappa B kinase (IKK) complex phosphorylates IκBα, resulting in the ubiquitination of IκBα. Then, IκBα is degraded rapidly through the 26S proteasome, which leads to the release of NF-κB. Therefore, we used western blotting to examine the effects of rLrp on the phosphorylation and degradation of IκBα induced by LPS. After stimulation for 5–10 min with LPS alone, IκBα was phosphorylated and degraded, whereas rLrp significantly blocked LPS-induced IκBα phosphorylation and degradation (Figure 3c). These data suggest that rLrp can inhibit LPS-induced NF-κB activation by preventing the activation of IKK. Next, we used shTLR2 to determine whether TLR2 was necessary for NF-κB activation in RAW264.7 cells. We found that NF-κB p65 localized to the nucleus in LPS-stimulated shTLR2-transfected cells with or without rLrp, indicating that the inhibitory effects of rLrp are dependent on TLR2 (Figure 3d).

rLrp is a ligand for mouse TLR2

Because rLrp could inhibit LPS-induced proinflammatory cytokine expression via TLR2, we performed a pull-down experiment to test the interactions between rLrp and different TLRs. HEK293 cells lack TLRs but retain downstream components of TLR-signaling. To determine whether rLrp can signal through TLR2, we transfected HEK293 cells with mouse TLR2 (HEK293-TLR2) or TLR4 (HEK293-TLR4) or a control vector (HEK293-vector) and then incubated the cells with rLrp for 24 h. The result showed that immobilized rLrp pulled down TLR2 but not TLR4 according to western blots probed with anti-GFP (Figure 4a). Immunofluorescence microscopy likewise showed strong fluorescence on the cell surface of HEK293-TLR2 cells that were exposed to rLrp but little fluorescence on the cell surface of HEK293-vector or HEK293-TLR4 cells (Figure 4b). We also used flow cytometry and found that the percentage of HEK293-TLR2 cells bound to rLrp was significantly higher than the percentage of HEK293-vector or HEK293-TLR4 cells (Figure 4c). These observations indicated that rLrp interacts with TLR2 predominantly and specifically.

Figure 4.

rLrp interacts with TLR2. (a) Lysates (10% of the total lysate) from HEK293-TLR2-GFP, HEK293-TLR4-GFP, and HEK293-vector-GFP cells were incubated with Ni-NTA-bead-conjugated rLrp, and the proteins pulled down were identified by western blotting using anti-GFP. (b) HEK293-TLR2, HEK293-TLR4, and HEK293-vector cells were plated on glass coverslips and incubated with rLrp (10 μg mL⁻¹) at 37 °C for 30 min. After washing, the cells were fixed with 4% paraformaldehyde, sequentially incubated with PE-conjugated goat anti-mouse, and photographed with a confocal fluorescence microscope (original magnification, 63×). (c) HEK293-TLR2, HEK293-TLR4, and HEK293-vector cells were incubated with rLrp (10 μg mL⁻¹) for 30 min. After washing, cells were stained as described in b, and flow cytometry was used to analyze rLrp binding. The data are representative of three independent experiments.

The N-terminal region of rLrp is important for downregulation activity

Several mutant rLrp constructs that included only the N-terminal region (from amino acid 1 to 90; rLrpΔN), deletion of amino acids 46 to 65 (rLrpΔD46-C65) or only the C-terminal region (from amino acid 91 to I179; rLrpΔC) were generated (Figure 5a). To determine which region of rLrp inhibits cytokine expression, RAW264.7 cells were stimulated with LPS with or without these constructs for 24 h, and ELISA was performed to assess IL-12p40 levels in the supernatants. We found that the rLrpΔN mutant inhibited LPS-induced IL-12p40 expression, whereas the rLrpΔC mutant induced the production of IL-12p40, similar to LPS alone. Furthermore, we found that the region from amino acids 46 to 65 (rLrpΔD46-C65) appears to play an important role in the inhibition of cytokine expression (Figure 5b–d). To confirm this, we chemically synthesized a peptide containing 19 amino acids from this region and marked the 19-amino acid peptide with biotin, using TLR2-GFP to pull down the biotin-marked peptide. The results indicated that the peptide interacted with TLR2 (Figure 5e). These data provided sufficient evidence that the N-terminal 19-amino acid region of rLrp is important for its inhibition.

Figure 5.

The N-terminus of rLrp mediates its inhibitory effects on TLR signaling. (a) Illustration of deletion mutants. (b) RAW264.7 cells were incubated with LPS (1 μg mL⁻¹) in the presence or absence of the rLrp peptide M1-V90 (10 μg mL⁻¹), and levels of IL-12p40 in the supernatants were measured by ELISA. (c) RAW264.7 cells were incubated with LPS (1 μg mL⁻¹) alone or with rLrp peptide G91-I17 (10 μg mL⁻¹), and IL-12p40 levels in the supernatants were measured by ELISA. (d) RAW264.7 cells were incubated with LPS (1 μg mL⁻¹) alone or with the rLrp mutant protein ΔD46–C65 (10 μg mL⁻¹), and levels of IL-12p40 in the supernatants were measured by ELISA. (e) Lysates (10% of the total lysate) from HEK293-TLR2-GFP and HEK293-vector-GFP cells were incubated with biotin-conjugated Lrp peptide, and the proteins pulled down were identified via western blotting using avidin-HRP. The data are expressed as the mean ± SD of three experiments.

Effects of rLrp on the LPS-induced association of TRAF6 with IRAK1 and TAK1

The TLR4-mediated signaling pathway, which comprises the adaptors MyD88 and TIR-domain-containing adaptor-inducing IFN-β (TRIF), can be activated by LPS.⁵² Activated MyD88 molecules bind with IRAK4, IRAK1, and IRAK2 and interact with TRAF6 and TAK1, leading to the activation of NF-κB.^53,54

To determine how rLrp or LPS affects IRAK1, TRAF6, and TAK1 signalosome formation, which is important for the activation of NF-κB (and potentially MAPKs), TRAF6 was immunoprecipitated and its binding to TAK1 and IRAK1 was detected by western blotting. After 30 min of LPS treatment, the association of TRAF6 with IRAK1 and TAK1 was observed, but TRAF6 binding to TAK1 was significantly reduced after treatment with rLrp, whereas TRAF6 binding to IRAK1 was maintained (Figure 6a and b). These findings suggest that rLrp inhibits LPS-induced NF-κB activation by blocking TAK1 binding to IRAK1 and TRAF6, and therefore signalosome formation. As expected, rLrp also inhibited the phosphorylation of TAK1 and TRAF6 ubiquitination, which are normally stimulated by LPS (Figure 6c and d). We next examined whether TLR2 was necessary for TRAF6 binding to TAK1 using shTLR2-transfected RAW264.7 cells. We found that the inhibitory effects of rLrp on TAK1 and TRAF6 binding were not detected, indicating that this inhibition is dependent on TLR2 (Figure 6e).

Figure 6.

rLrp inhibits the TRAF6-TAK1 interaction, TAK1 activation and TRAF6 ubiquitination. Immunoassay of lysates of RAW264.7 cells treated with or without rLrp. (a) Immunoblotting of lysates of cells transfected with plasmids expressing HA-tagged IRAK1 and TAK1 and Flag-tagged TRAF6 with anti-HA or anti-Flag after immunoprecipitation (IP) with anti-Flag. (b) Immunoblotting with anti-IRAK1 and anti-TAK1 after the immunoprecipitation of endogenous TRAF6 with anti-TRAF6 and reprobing with anti-IRAK1, anti-TAK1, and anti-TRAF6 to evaluate protein loading. (c) Western blotting was used to determine the effects of rLrp on the phosphorylation of TAK1 in RAW264.7 cells after treatment with LPS or rLrp (5 μg mL⁻¹) for the indicated times. (d) Immunoblotting (IB) of cell lysates transfected with plasmids expressing His-tagged ubiquitin and Flag-tagged TRAF6 with anti-His or anti-Flag, assessed after immunoprecipitation with anti-Flag-agarose. (e) shTLR2-transfected RAW264.7 cells were treated with LPS or rLrp (5 μg mL⁻¹) for the indicated times, and then immunoblotting was performed with anti-TAK1 after the immunoprecipitation of endogenous TRAF6 with anti-TRAF6. The data are representative of three experiments.

rLrp-mediated inhibition of TLR-driven proinflammatory cytokine expression in LPS-treated macrophages involves activation of the PI3K/Akt pathway

We next examined whether the inhibition of LPS-induced proinflammatory cytokine expression in RAW264.7 cells by rLrp requires the PI3K/Akt pathway. LPS induction of IL-12p40 and IL-6 in RAW264.7 cells was measured in the presence and absence of rLrp and PI3K activation inhibitor (wortmannin). RLrp inhibited the LPS-mediated induction of IL-12p40 and IL-6, whereas there was a significant reduction in this rLrp inhibition in the presence of wortmannin (p < 0.001; Figure 7a and b). Similar results were observed in RAW264.7 cells with Akt inhibitors (Figure 7c and d). When Akt was depleted from RAW264.7 cells by shAkt, rLrp inhibition of LPS-induced IL-12p40 and IL-6 production was also reduced (Figure 7e–g). In addition, confocal microscopy of shAkt-transfected cells also demonstrated that NF-κB p65 localized to the nuclei in both LPS- and LPS plus rLrp-treated cells after 15–30 min of stimulation (Figure 7h). These results indicate that the PI3K/Akt pathway is likely involved in the inhibition of LPS-induced IL-12p40 and IL-6 production and activation of NF-κB p65 by rLrp.

Figure 7.

rLrp-mediated inhibition of TLR-driven IL-6 and IL-12p40 production depends on the PI3K/Akt pathway. (a, b) RAW264.7 cells were pretreated for 30 min with or without the PI3K inhibitor wortmannin (5 μM) and stimulated with LPS (1 μg mL⁻¹) with or without rLrp (5 μg mL⁻¹). The levels of IL-6 and IL-12p40 in the supernatants were measured by ELISA. (c, d) RAW264.7 cells were pretreated for 30 min with or without an Akt inhibitor (20 μM) and stimulated with LPS (1 μg mL⁻¹) with or without rLrp (5 μg mL⁻¹). The levels of IL-6 and IL-12p40 in the supernatants were measured with ELISA. (e) Western blot of lysates from RAW264.7 cells stably expressing shAkt. (f, g) Wild type and shAkt RAW264.7 cells were treated with LPS (1 μg mL⁻¹) with or without rLrp (5 μg mL⁻¹), and the levels of IL-6 and IL-12p40 were measured by ELISA. (h) Confocal microscopy (original magnification 63×, 1.4-NA, oil-immersion objective) was used to determine the effects of rLrp on the translocation of NF-κB from the cytoplasm to the nucleus in shAkt RAW264.7 cells.

rLrp induces the phosphorylation of PI3K, Akt, and GSK-3β through TLR2 signaling

Phosphorylation is a critical prerequisite for PI3K, Akt, and glycogen synthase kinase 3 beta (GSK-3β) activation and is indicative of their enzymatic activity levels.⁵⁵ After treatment with rLrp (5 µg mL⁻¹), we used western blotting to examine the phosphorylation of PI3K, Akt, and GSK-3β in RAW264.7 cells at different time points. Strong phosphorylation of PI3K, Akt, and GSK-3β was detected after 15–60 min of treatment. However, little phosphorylation was observed in untreated cells. After 30–60 min of treatment with rLrp, phosphorylation peaked (Figure 8a–c). Furthermore, we showed that rLrp inhibition of LPS-induced IL-12p40 and IL-6 production required TLR2-dependent activation of the PI3K/Akt pathway. In shTLR2-transfected RAW264.7 cells, no phosphorylation of PI3K, Akt, and GSK-3β was observed, and the phosphorylation of Akt and GSK-3β was significantly more attenuated in TLR2^−/− mice than in C57BL/6 and TLR4^−/− mice (Figure 8d and e).

Figure 8.

rLrp activates the PI3K/Akt pathway through TLR2 singing. (a) RAW264.7 cells were incubated with rLrp (1 μg mL⁻¹) for 0–60 min, cell lysates were prepared, and western blot analysis with anti-phospho-PI3K (p-PI3K) was performed to examine the phosphorylation of PI3K. Similar data were obtained in two independent experiments. (b) Immunoblotting of Akt and phospho-Akt (p-Akt) from the lysates of RAW264.7 cells treated with rLrp for 0–60 min. (c) Immunoblotting of phospho-GSK-3β (p-GSK-3β) and β-actin from the lysates of RAW264.7 cells treated with rLrp for 0–60 min. (d) shTLR2 RAW264.7 cells were treated with rLrp (5 μg mL⁻¹) for 0–90 min, and then p-PI3K, p-Akt and p-GSK-3β were detected in lysates by western blotting. (e) Western blotting was used to determine the effects of rLrp on the phosphorylation of PI3K, Akt, and GSK-3β in peritoneal macrophages from C57BL/6, TLR2^−/−, and TLR4^−/− mice after treatment with medium alone or rLrp (5 μg mL⁻¹) for the indicated times (0–90 min).

rLrp attenuates IFN-γ-induced expression of MHC-II in mouse macrophages in a TLR2-mediated and PI3K/Akt pathway activation-dependent manner

To investigate whether rLrp has an effect on the expression of MHC-II, we used IFN-γ (20 ng mL⁻¹) to stimulate RAW264.7 cells for 12, 24, or 48 h with or without rLrp and used flow cytometry to the examine the expression of MHC-II. The level of MHC-II expression was low in untreated RAW264.7 cells for 0–48 h (Figure 9a–e). Incubation with IFN-γ led to a two- to three-fold increase in the expression of MHC-II at 24–48 h. However, MHC-II expression was attenuated significantly after 24 h when RAW264.7 cells were treated with rLrp and IFN-γ together (Figure 9e).

Figure 9.

rLrp reduces IFN-γ-induced expression of MHC-II in RAW264.7 cells through TLR2. (a, b, c, d) MHC-II expression on RAW264.7 cells at baseline (medium; a) and after stimulation with IFN-γ (20 ng mL⁻¹) or rLrp (5 μg mL⁻¹) in combination with IFN-γ for 12 (b), 24 (c), or 48 h (d). Cells were stained with isotype-control antibody or PE-CY5-conjugated anti-MHC-II. The graphs are representative of the results from three experiments. (e, f, g) Flow cytometry was used to quantitate the relative mean fluorescence value (ΔMFV; MFV of cells stained with anti-MHC-II minus the MFV of cells stained with the isotype-control antibody) of RAW264.7 (e), shTLR2-transfected (f) and shAkt-transfected (g) cells. ***p < 0.001, stimulated cells versus those cultured in medium alone (h, i). Macrophages from C57BL/6 and TLR2^−/− mice treated with IFN-γ (20 ng mL⁻¹) or IFN-γ in combination with rLrp (5 μg mL⁻¹) for the indicated times. The mean levels of MHC-II expression in the medium were set to 100, and the relative changes in the expression of MHC-II in the presence of different stimuli were calculated. The values are expressed as the mean ± SD from three experiments.

Because Ag presentation and T cell stimulation are influenced by changes in the expression of co-stimulatory molecules on the cell surface, we incubated RAW264.7 cells with IFN-γ alone or IFN-γ with rLrp for 48 h and evaluated the expression of surface markers with flow cytometry. Expression of the cell surface co-stimulatory molecule CD86 increased significantly after 48 h of incubation with IFN-γ (p < 0.001; Figure 10a–e). Incubation of cells with IFN-γ and rLrp significantly decreased the expression of CD86, suggesting that rLrp may prevent the innate immune response by downregulating MHC-II and co-stimulatory molecules.

Figure 10.

rLrp reduces IFN-γ–induced CD86 surface expression on RAW264.7 cells through TLR2 (a, b, c, d). CD86 expression on RAW264.7 cells at baseline (medium; a) and after stimulation with IFN-γ (20 ng mL⁻¹) or rLrp (5 μg mL⁻¹) in combination with IFN-γ for 12 (b), 24 (c), or 48 h (d). Cells were stained with isotype-control antibody or FITC-conjugated anti-CD86. The graphs are representative of the results from three experiments (e, f, g). Flow cytometry was used to quantitate the relative mean fluorescence value (ΔMFV; MFV of cells stained with anti-CD86 minus the MFV of cells stained with the isotype-control antibody) of RAW264.7 (e), shTLR2-transfected (f) and shAkt-transfected (g) cells. **p < 0.01; ***p < 0.001, stimulated cells versus those cultured in medium alone (h, i). Macrophages from C57BL/6 and TLR2^−/− mice treated with IFN-γ (20 ng mL⁻¹) or IFN-γ in combination with rLrp (5 μg mL⁻¹) for the indicated times. The mean levels of MHC-II expression in the medium were set to 100, and the relative changes in the expression of MHC-II in the presence of different stimuli were calculated. The values are expressed as the mean ± SD from three experiments.

We also detected rLrp-mediated inhibition of IFN-γ-induced MHC-II and CD86 expression in peritoneal macrophages from C57BL/6 and TLR2^−/− mice and in TLR2-depleted RAW264.7 cells. The cells were incubated with IFN-γ with or without rLrp for 48 h, and flow cytometry was used to examine the expression of MHC-II proteins and CD86. The same levels of MHC-II and CD86 were observed in both the presence and absence of rLrp in peritoneal macrophages from TLR2^−/− mice and shTLR2-transfected RAW264.7 cells (Figure 9f, h, and i and Figure 10f, h, and I), suggesting a role for TLR2 in the rLrp-mediated inhibition of IFN-γ-induced expression of MHC-II and co-stimulatory molecules. In addition, rLrp-mediated inhibition of the IFN-γ-induced expression of MHC-II and CD86 was also reduced in shAkt-transfected RAW264.7 cells (Figures 9g and 10g), indicating that rLrp-mediated inhibition of MHC-II and co-stimulatory molecules in IFN-γ-treated macrophages involves activation of the PI3K/Akt pathway.

rLrp inhibits IFN-γ induced formation of autophagosomes

Treatment of macrophages with IFN-γ promotes the formation of autophagosomes.^56,57 We tested whether this effect could be inhibited by treating RAW264.7 cells with IFN-γ in combination with rLrp. As expected, the induction of autophagy by IFN-γ increased both LC3-I and LC3-II band intensity in western blots of RAW264.7 cell extracts (Figure 11a, lanes 2 and 3) due to the autophagic consumption of LC3. When rLrp was added to the macrophages, the intensity of the LC3-I and LC3-II bands decreased (Figure 11a, lanes 5 and 6). In addition, we found that IFN-γ treatment increased the percentage of RAW264.7 cells with large vacuoles that stained positively for monodansylcadaverine, a marker of autophagic vacuoles, and this effect was inhibited by rLrp (Figure 11b). Furthermore, rLrp significantly reduced the number of IFN-γ-induced pEGFP-LC3+ puncta per cell (Figure 11c).

Figure 11.

Inhibition of IFN-γ-induced autophagy by rLrp in macrophages. (a) Immunoblot analysis of the LC3 lipidation state (I or II) in RAW264.7 cells induced by IFN-γ and incubated without or with rLrp. β-actin was used as the loading control. (b, c) RAW264.7 cells were transiently transfected with pEGFP-LC3, induced with IFN-γ for 8 h with or without rLrp (5 μg mL⁻¹), plated onto glass coverslips and incubated at 37 °C. After washing, the cells were fixed with 4% paraformaldehyde and photographed with a fluorescence microscope (b; original magnification, 63×). (c) The numbers of large (<$>\ge<$>0.5 μm) LC3 puncta per cell were quantified. The data presented are the mean ± SEM; ***p < 0.001.

Discussion

During the innate immune response, macrophages secrete cytokines such as IL-12 and TNF-α, which are important for the promotion of the Th1 response and cell-mediated immune responses.^11,58 The host defense response to intracellular infections such as tuberculosis consists of these proinflammatory cytokines, and therefore they are targeted by pathogenic mycobacteria to avoid immune surveillance.^17,59 Thus, identification of the interrupted immune signaling components that block the induction of these cytokines is necessary to develop a proper treatment to control mycobacterial infection. In our present study, we found that the production of IL-12 and TNF-α stimulated by LPS was inhibited by rLrp. In particular, the production of IL-12 was more strongly inhibited by rLrp than TNF-α production. It is well known that the levels of TNF-α and IL-12 production are inversely correlated with the virulence of pathogenic mycobacteria.^15,16,60,61 Holscher et al. demonstrated that human monocyte-derived macrophages exposed to M. tuberculosis produce significantly lower amounts of IL-12 than those exposed to other bacteria.⁵⁸ In our studies, RAW264.7 cells and peritoneal macrophages from C57BL/6 mice stimulated with H37Rv produced significantly lower amounts of IL-12 and TNF-α than macrophages from C57BL/6 mice stimulated with Lrp^−/− H37Rv (Supplementary Figure 1a and b), indicating that rLrp specifically targets the signaling cascades that lead to the production of this cytokine.

As a mammalian transcription factor, NF-κB controls a number of genes such as TNF-α, IL-6, and IL-12. These cytokines are very important for inflammation and immunity.^49,50,51 Based on this, we used EMSA and reporter gene assays to test whether rLrp inhibited NF-κB activity in RAW 264.7 macrophages. The results show that rLrp suppresses the DNA-binding and transcriptional activities of NF-κB induced by LPS. To investigate how rLrp inhibits NF-κB activity, we sought to determine the influence of rLrp on the activation of signaling targeting NF-κB. After LPS stimulation, the phosphorylation and degradation of IκBα leads to NF-κB activation and translocation to the nucleus. We found that rLrp inhibited the translocation of NF-κB and the phosphorylation and degradation of IκBα. The phosphorylation and degradation of IκBα is a critical process that regulates NF-κB activity.^62,63 NF-κB is a homo- or heterodimer of the five Rel family proteins in mammals, and it is the primary transcription factor activated by TLR signaling.⁶⁴ In a given cell, the combination of Rel proteins comprising the NF-κB dimer affects whether NF-κB positively or negatively regulates promoter sites. In dendritic cells, inactivation of NF-κB decreases the expression of MHC-II and co-stimulatory molecules, leading to decreased priming of T cells.⁶⁵ In promonocytic U937 cells, p50 NF-κB homodimers negatively regulate NF-κB binding sites contained within the MHC-II invariant chain promoter.⁶⁶

Pathak et al. described that M. tuberculosis could subvert the innate immune response with a series of dampening signals.⁴² We must understand the attenuating mechanisms of innate immune signaling by mycobacterial effectors so that we can look for new strategies to control undesired dampened inflammatory signaling.^67,68,69 Because TLR signaling is very important to the innate immune response, we developed the hypothesis that rLrp attenuates TLR signaling. Inhibition of the MyD88-dependent or independent arms of TLR signaling may result in negative regulation of TLR signaling (Supplementary Figure 2). It is known that MyD88, IRAK-M, Tollip, suppressor of cytokine signalling 1 (SOCS1), suppressor of cytokine signalling 3 (SOCS3),⁷⁰ β arrestin 1, β-arrestin 2,⁷¹ and Smad6⁷² can negatively regulate the MyD88-IRAK-TRAF6 axis. Here, we found that rLrp also attenuates MyD88-dependent signaling. The receptor complex is formed by MyD88, IRAK4 or IRAK1 and TRAF6 during TLR stimulation. We demonstrated that rLrp interfered with this complex by inhibiting the recruitment of TAK1 to IRAK1 and TRAF6. As a result, TRAF6 ubiquitination was attenuated, similar to the effects observed with other bacteria and viruses that commonly target TLR signaling complexes upstream of IKK activation to evade the immune response. For example, both IRAK2 and TRAF6 are targeted by the poxvirus protein A52R.^42,73

Lipoproteins from M. tuberculosis such as the 19-kDa LpqH primarily activate innate immunity through TLR2. However, prolonged TLR2 signaling results in the attenuation of certain important immune functions such as MHC-II Ag processing in macrophages.^36,37,38,39 In this study, we purified and identified 19.8-kDa rLrp as a novel TLR-2 ligand from M. tuberculosis, and similar results were obtained following the exposure (12–48 h) of macrophages to rLrp, which led to significant inhibition of MHC-II protein expression induced by IFN-γ. Additionally, we found that rLrp inhibited TLR-2-dependent MHC-II expression. Because MHC-II-restricted CD4-T cells are important for host defense against M. tuberculosis, reduction of MHC-II expression is an effective immune evasion technique. MHC-II Ag presentation involves antigen-presenting cell (APC) internalization and Ag processing. Ag peptides are loaded onto MHC-II molecules, and then the peptide-MHC-II complexes are transported to the cell surface for presentation to CD4-T cells. Although certain other bacteria can harness IFN-γ-induced gene expression,^36,40,74 M. tuberculosis is quite successful in utilizing this strategy. First, M. tuberculosis is a intracellular pathogen targeting macrophages, the TLR-expressing cells that possess many IFN-γ-modulating genes involved in the host immune response. Second, the resistant nature of M. tuberculosis protects it from acute innate microbicidal responses and allows it to survive inside macrophages to facilitate chronic exposure to PAMPs. Third, intracellular bacteria pump out proteins,⁷⁵ stimulating TLR2 to recruit phagosomal compartments.^76,77 Through TLR2 signaling, rLrps may decrease Ag-presenting functions and MHC-II expression. The infected macrophages would then function as a niche in which M. tuberculosis could survive despite detection by CD4-T cells. Our findings support a model in which a successful chronic infection by M. tuberculosis requires at least three factors: invasion of macrophages and persistence in the replicative phagosome, interaction with TLR2 through rLrps, and inhibition of Ag presentation and MHC-II expression via modulatory cytokines produced by the invaded cell.

We found that rLrp interacted with TLR2 directly and exerted its inhibitory effects. The inhibition effects of rLrp were not evident in macrophages from TLR2^–/– mice; this finding supports the conclusion that interaction with TLR2 is necessary to reduce TLR-mediated signaling by rLrp. Deletion of 19 residues from the N-terminal region of rLrp (a region that likely participates in rLrp interactions with other partners) was sufficient to remove its effects. The deletion of 20 amino acids from the C-terminus of rLrp did not inhibit TLR signaling. In principle, TLR2 recognizes acylated peptides or proteins, but recently the unmodified protein Yersinia pestis V Ag has been reported to function through TLR2.⁷⁸ Here, our study raises the possibility that short peptides can dampen TLR signaling.

There are conflicting reports on whether PI3K-Akt signaling leads to the activation or inhibition of NF-κB, and these conflicting reports are mostly based on the use of pharmacological inhibitors.^79,80 We believe that PI3K-Akt signaling negatively impacts TLR-mediated NF-κB activation, and this is supported by studies that used shRNA to downregulate the p110b subunit of PI3K in RAW cells.^81,82 However, the effects of PI3K signaling likely rely on both the original trigger and the cell type. Here, we also confirmed that the attenuating effects of rLrp that are dependent on Akt extend to the inhibition of inflammatory cytokines, particularly TNF-α and IL-12, because we showed that rLrp can activate the PI3K/Akt pathway by stimulating the rapid phosphorylation of PI3K, Akt, and GSK-3β in macrophages. Arbibe et al. have shown direct evidence for the involvement of PI3K in TLR-signaling.⁸⁰ They also found that mutagenesis of TLR2 leads to a loss in the ability of p85 to interact with TLR2 and also a loss in the ability of TLR2 to induce NF-κB transcriptional activity. Subsequently, Guha et al.²⁷ noted that inhibition of PI3K led to the increased activity of NF-κB p65 and increased the production of certain pro-inflammatory cytokines. The impact of the PI3K pathway on the host inflammatory response was appreciated with the generation of PI3K knockout mice. Fukao et al.^26,83 found that deficiencies in the regulatory subunit of PI3K (p85a) enhanced Th1-like immune responses to an intestinal nematode in mice; however, the mice were unable to clear the infection. Additionally, p85a knockout mice were found to be resistant to Leishmania infection with concomitant augmented Th1-associated immunity. The alterations in Th1 and Th2 immunity in PI3K knockout mice led to increased IL-12 production by dendritic cells.^26,83 Subsequently, downstream of TLR2, activation of the PI3K pathway exerted differential effects on IL-10 and IL-12 production.⁸⁴ Indeed, these studies demonstrated that the inhibition of PI3K activity downregulated IL-10 production but upregulated the production of IL-12.⁸⁵

Several groups have reported that mycobacterial infection can increase the levels of SOCS3, which inhibits proinflammatory gene expression.⁸⁶ Furthermore, high expression of SOCS3 in macrophages also suppresses LPS-mediated TNF-α and CD40 expression.^87,88 We therefore believe that these microorganisms may have also evolved strategies to hijack the SOCS3 signaling system. In this study, we showed that rLrp can induce the phosphorylation of tyrosine in SOCS3, which was found to be essential for the suppression of proinflammatory cytokine production in LPS-activated macrophages. In addition, rLrp induced the expression of SOCS3, and this induction was dependent on the PI3K/Akt pathway (Supplementary Figure 3).

Many significant phenotypic features of Mycobacterium species are related to individual cell wall components, not only in terms of colony morphology but also for resistance to different environmental stresses and virulence. Furthermore, changes in cell wall components can modulate the signaling events initiated in infected cells, whereas changes in colony morphology may reflect differences in virulence as well as survival of a given Mycobacterium species.^89,90,91 When we assessed the effects of rLrp on the colony morphology of M. tuberculosis, we found that mutant rLrp resulted in colony morphology variants with a small, smooth, translucent phenotype compared to the rough, glossy, larger, strongly cohesive phenotype of the wild type M. tuberculosis (Supplementary Figure 4). This observation suggests a role for rLrp in the survival and virulence of M. tuberculosis in the context of adaptation to stress conditions.

Our findings suggest that the engagement of a TLR may induce negative rather than positive signaling that dampens TLR signaling. Inhibition of TLR signaling by rLrp required the kinase Akt. Direct binding of rLrp to TLR2 activated the PI3K/Akt signaling pathway and prevented interactions between TRAF6/IRAK1 and TAK1, thus abrogating NF-κB activation (Supplementary Figure 5).