Published online before print September 27, 2006
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Laboratório de Biologia do Reconhecer,
Laboratório de Biologia Celular e Tecidual, Universidade Estadual do Norte Fluminense, Rio de Janeiro, Brazil; and
Instituto de Biofísica Carlos Chagas Filho, Laboratório de Físico-Química Biológica, Universidade Federal do Rio de Janeiro, Rio de Janeiro, Brazil
1 Correspondence: LBR, CBB, UENF, Av. Alberto Lamego, 2000 Campos/RJ, Rio de Janeiro 28013-600, Brazil. E-mail: elena@uenf.br
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
B activation were important for the early cytoskeletal rearrangements observed, although suppression of these pathways is known to inhibit chemokine secretion by activated macrophages. B2-integrins blockade with a corresponding antibody inhibited macrophage spreading and polarization but had no effect on pseudopodia protrusions demonstrating the downstream position of integrin-mediated adhesion in PI3K- dependent signaling pathway leading to the motility phenotype. The obtained data demonstrate that the direct effect of mycobacteria on macrophage shape might be mediated through TLR2-dependent PI3K activation.
Key Words: BCG • RAW264.7 macrophages • signal transduction
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
B, MAPK, and class 1 PI3K pathways rapidly activated after binding of mycobacterial agonists with macrophage Toll-like receptors, mainly TLR2. Locomotor morphological changes in response to mycobacteria have not been characterized in detail nor have the signaling events leading to MF spreading and adhesion. The actin cytoskeleton is a primary determinant of cell shape and motility. Reorganization of the actin cytoskeleton is an early cellular response to a variety of extracellular signals. Dissection of pathways leading to actin rearrangement induced by a variety of extracellular factors has focused largely on those initiated by tyrosine kinase receptors for growth factors, integrins, and G protein-coupled receptors for chemokines [4
], whereas the role of TLRs is only beginning to be noticed. Recently, TLR- mediated actin remodeling was demonstrated to enhance antigen capture by dendritic cells [5
]. Additionally, CD14- dependent TLR2 activation by mycobacterial lipoarabinomannan (LAM) was proven to promote phagocytosis of bacillus Calmette-Guerin (BCG) mycobacteria by MF [6
]. Whether TLRs are involved in mycobacteria-induced MF spreading and motility has not been well elucidated. To study morphological changes induced in MF with mycobacteria, we treated RAW 264.7 MF with either viable BCG, BCG whole cell lysate (BCG-WCL), or with purified mycobacterial cell wall lipoprotein p19, a known TLR2 agonist [7
], and demonstrated the similar pattern of morphological changes: rapid radial spreading leading to subsequent cell polarization typical for the migratory phenotype, suggesting that TLR2 engagement could contribute to morphological rearrangements induced by BCG. Mycobacteria are a potent activator of PI3K/Akt and MAPK pathways in MF [3 , 6 , 8 ]. Growing evidence now supports the view that these kinase pathways could play a significant role in the determination of internal polarity of migrating leukocytes [9 ]. The role of PI3K [10 ], as well as ERK and p38 MAP kinases [11 , 12 ], in the regulation of MF spreading in response to oxidant stress and LPS was well demonstrated. The main pathway leading to MAPK activation by mycobacteria is mediated by TLR2/MyD88 engagement [13 ]. The role of TLR2 in PI3K activation by mycobacteria has not been verified, although it was established for the cells stimulated with S. aureus [14 ].
Activation of class I group of PI3K is characterized by the kinase translocation to the cytoplasmic membrane where the 85-kDa regulatory subunit through its Src homology 2 (SH2) domain interacts with phosphotyrosine residues of a variety of phosphorylated proteins, including that of the cytoplasmic end of membrane receptors leading to activation of the catalytic p110 subunit. Activity of p110 is directed to the phosphorylation of phosphatidylinositide lipids at the D3 position of the inositol ring, leading to the formation of phosphatidylinositol-3,4,5-triphosphate (PIP3) from phosphatidylinositol-4,5-diphosphate (PIP2) [15 ]. PI3K lipid products (PIP2 and PIP3) were demonstrated to bind Rac1/guanine-nucleotide exchange factors (GEFs), such as Vav1 and Vav2 and promote activation of Rac1 [16 ], which leads to actin polymerization and protrusion of leading lamellipodia [17 ].
In this study, we verified whether BCG treatment induces p85 recruitment to TLR2 and demonstrated that PI3K activation could be mediated by this receptor. Additionally, suppression of PI3K activity by the specific inhibitor LY294002 or through the TLR2 blockade with corresponding antibody, dramatically inhibited mycobacteria-induced cell spreading and polarization through the suppression of Rac-1 activation, F-actin assembly, and MF pseudopodia protrusion. In summary, our data demonstrate that TLR2-dependent PI3K pathway activation induced by BCG mycobacteria is essential for the MF morphological rearrangements leading to spreading and polarization.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
B RelA, rabbit anti- p65 Ab was a kind gift from Dr. N. Rice, NCI-Frederic Cancer Research and Development Centre, USA. Antibodies against TLR2, PI3-kinase regulatory subunit p85, Rac-1 and phospho-specific Ab against p-Akt1 (Ser 473)-R, p-ERK, p-p38, and p-Rac-1, as well as Protein A/G Plus agarose, were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). TLR2-neutralizing Ab (T2.5) was obtained from HyCult Biotechnology (Uden, The Netherlands), anti-β2 integrins (GAME-46) was from PharMingen (San Diego, CA), and anti-phosphotyrosine rabbit polyclonal Ab - from BD Bioscience (Franklin Lakes, NY). Mouse IgG1 Isotype Control was purchased from R&D Systems (Minneapolis, MN).
Cell culture and mycobacteria
The RAW 264.7 murine macrophage line (ATCC TIB-71; American Type Culture Collection, Mannasas, VA) was cultivated in DMEM-F12 medium supplemented with 10% heat-inactivated FBS, 2 mM L-glutamine, and 50 µg/ml gentamicin (CM). Bone marrow-derived macrophages (BMM) were obtained from the femur bones of TLR2–/– or control TLR2+/+ C57BL/6 mice (kindly provided by S. Akira, Osaka, Japan). Bone marrow was flushed out of the bones and cultivated in the CM-supplemented with 20% L-929 fibroblast-conditioned medium for 6 days to obtain monolayers of adhered MF.
M. bovis (BCG vaccine, Onco-BCG, Moreau strain, Copenhagen SEED# July, 1978) was kindly donated by Instituto Butantan (São-Paulo, Brazil), where the vaccine was produced for bladder cancer treatment. The vaccine suspension was stored at 4°C and used in the experiments before the expiration date recommended by manufacturer. To obtain BCG-WCL, the vaccine suspension was washed twice with sterile PBS, pH 7.4, and sonicated in a lysing 10 mM Tris-HCl, 100 mM NaCl, 0.5% SDS, and 25 mM EDTA buffer, pH 7.8. BCG lysate preparations were passed through a Detoxi-Gel column using sterile pyrogen-free water, stored in pyrogen-free vials, and reconstituted with sterile pyrogen-free PBS. Evaluation of bacterial endotoxin was performed with the Limulus amebocyte lysate assay (E toxate kit; Sigma) and showed less than 1 pg/ml final concentration. Purified M. tuberculosis 19-kDa lipoprotein (p19) was provided by Dr. J. Belisle (Colorado State University, Fort Collins, CO) through a TB Research Materials and Vaccine Testing contract (National Institutes of Health, National Institute of Allergy and Infectious Diseases NO1-A1-40091). Dry p19 was resuspended to 26 mM in DMSO, and stored at –80°C.
For the experiments, RAW264.7 or BMM were grown in 24-well plates with glass coverslips at a concentration of 3 x 105 cells/ml, in CM, at 37°C in 5% CO2 for 24 h. RAW264.7 cells were then washed and incubated for additional 18 h in the medium with 0.1% serum. Resulting MF monolayers were infected with viable BCG at a multiplicity of infection (MOI) of 1:10 macrophage/bacteria, or treated with 1 µg/ml BCG-WCL or 5 µg/ml p19, and incubated at 37°C for various time periods before analysis. In some experiments, cells were pretreated with specific kinase inhibitors for 60 min, or neutralizing Ab, anti-TLR2 (15 µg/ml), anti-β2 integrins (10 µg/ml), or mouse IgG1 isotype control, for 30 min before BCG stimulation. IB Kinases (IKKs) inhibitor BAY11-7085 (10 µM) was used to suppress NF-B activation, PD98059 (50 µM)—for MEK1/ERK, SB203580 (10 µM), for p38 MAPK, and LY294002 (25 µM)—for PI3K inhibition. To serve as a control, the same volume of the diluent DMSO contained in kinase inhibitors was added to the cell culture.
For the intracellular protein analysis, cells were grown in 75 cm2 culture flasks, treated as indicated, washed in ice-cold sterile PBS, harvested by scraping, and used for further immunoprecipitation, kinase assay, and Western blot analysis.
Immunofluorescence and differential interference contrast light microscopy
For immunofluorescence studies, cells were grown on glass coverslips and fixed with 3% formaldehyde in PBS for 20 min at room temperature. Cells were permeabilized with 0.5% Triton X-100 in PBS for 10 min. For immunolocalization of p85 PI3K regulatory subunit, the cells were blocked with 1.5% BSA for 30 min and treated with the corresponding primary Ab and anti-rabbit IgG conjugated with PE. Coverslips were mounted over N-propyl gallate and observed in a confocal laser-scanning microscope (LSM-310, Zeiss).
For localization of F-actin filaments and quantification of the MF spreading and polarization, fixed and permeabilized cells were incubated with 0.1 µg/ml TRITC-labeled phalloidin for 30 min, washed, mounted over N-propyl gallate, and observed using a Zeiss Axioplan Microscope equipped with differential interference contrast. Each cell sample was examined first under DIC and then under fluorescence microscope, and images were recorded using a Kodak Tris X (ASA 400). Dimensions of the cells were measured using the ImageJ 1.33k software (Rasband, W.S., ImageJ, National Institutes of Health, Bethesda, MD, USA, http://rsb.info.nih.gov/ij/, 1997-2004). Cell spreading was estimated by manually tracing the cell perimeter and measuring the cell footprint area, including protrusions. Spreading index was estimated as a relation of a mean cell area of BCG-treated cells to the mean area of control untreated cells. Cell length and width, including filamentous protrusions were determined by manual tracing of randomly selected cells. A mean length/width ratio was calculated for each sample. Cell elongation index was estimated as a relation between the length/width ratio of BCG- treated cells and the length/width ratio of control cells. At least 80 randomly selected cells of each sample were quantified.
FACS assay for F-actin
To quantify relative content of F-actin in cells stimulated with BCG, the cells were grown in 6-well plates and treated according to the experimental design. Cells were then stained with TRITC-phalloidin, washed, scraped from the plate, and analyzed by flow cytometry (Coulter Elite/ESP). Data were collected as histograms of distribution of cell fluorescence intensities, and geometrical means of each histogram were used for further analysis. Experiments were performed in duplicates and repeated at least two times.
Scanning electron microscopy
Cells were grown on glass coverslips to 50–70% confluence and then treated according to the experimental design. The coverslips were rinsed with PBS, and the cells were fixed quickly with 1% osmium tetroxide in 0.1 M cacodylate buffer for 5 s at room temperature. This was followed by two consecutive 60-min fixations with 2.5% glutaraldehyde in 0.1 M cacodylate buffer at room temperature and a final wash in 0.1 M cacodylate buffer for 10 min. The fixed cells were then dehydrated in ethanol, critical point dried in CO2, covered with a layer of gold and observed using 5 kV in a Zeiss 964 Scanning Electron Microscope.
Western blot analysis and immunoprecipitation
After stimulations, cells were lysed for 15 min in lysis buffer (0,5% Nonidet P-40, 3 mM sodium pyrophosphate, 50 mM Tris-HCl pH 7.6, 150 mM NaCl, 0.1 mM EDTA, 50 mM NaF, 1 mM Na3VO4, 1 mM PMSF, 10 µg/ml leupeptin, 1 µg/ml pepstatin, and 1 µg/ml aprotinin). Lysates were clarified by centrifugation at 14,000 g for 10 min and used either for Western blot analysis of distinct proteins or for immunoprecipitations. The protein concentration of the cell extracts was determined using the Bradford method. For immunoprecipitation, the whole cell lysate proteins, 600 µg, were then incubated with specific Ab followed by protein A/G Plus agarose at 4°C for 1h each. Immunoprecipitations were performed using rabbit anti-PI3K and rabbit or goat anti-TLR2 Abs. The immunoprecipitates were then washed in lysis buffer before addition of 2x gel sample buffer, heated to 95°C for 3 min, and subsequently analyzed by SDS-PAGE, followed by electrophoretic transfer to polyvinylidene difluoride membranes (Immobilon-P; Millipore, Bedford, MA). Membranes were blocked in 5% nonfat skim milk and probed with anti-TLR2, anti-p85 subunit PI3K, anti-Rac-1 or anti-p-Tyr Ab. Blots were visualized with HRP-conjugated anti-goat or anti-rabbit Ab (1/3000). (Amersham) and developed using ECL kit.
PI3 kinase activity determination
PI3K activity was measured by lipid kinase assay, as described previously [18
]. Briefly, cell lysate proteins, 400 µg, were resuspended in 0.16 ml of the PI3K assay buffer containing 200 µg phosphoinositides and [
-32P]ATP (20 µCi at final concentration of 0.05 mM). The reaction was performed for 30 min at room temperature and then quenched with 3 ml ice-cold CHCl3:CH3OH:HCl (200:100:0.075), starting the lipid extraction. The lipid extraction and separation by TLC was performed according to the previously described method [19
]. Autoradiography of the TLC plates was carried out using X-ray film (Kodak T-Mat) and an intensifier screen. The cassette was stored at –70°C for 
48 h, and then the film was developed following the manufacturer’s specifications. The [32P]-labeled phosphoinositide 3-phosphate band was visualized by autoradiography, measured by densitometry and liquid scintillation counting with comparable results, and PI3K activation was calculated as a percent to the untreated control cells.
Statistical analysis
Results are expressed as means ± SE. The differences between means were evaluated by ANOVA and considered significant at P < 0.05.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
 View larger version (154K): [in a new window] |
Figure 1. BCG-induced morphological changes and actin reorganization in RAW264.7 macrophages. Cells were seeded on glass coverslips and incubated for 24 h; the culture medium was changed to the medium with 0.1% FCS and incubated for additional 18 h. Cells were then left unstimulated (C) or infected with viable BCG mycobacteria at MOI = 10:1 for the indicated time periods. Cells were then stained with TRITC-labeled phalloidin to reveal F-actin. Images were obtained using differential interference contrast light microscopy (A–D) or fluorescence microscopy (E–H). Arrowheads indicate numerous actin-rich filopodia at spreading cells observed 0.5–3 h after treatment (B, C, F, G) and F-actin-positive leading edge of polarized cells observed 24 h after treatment (D, H). Arrows indicate uropods of polarized cells lacking F-actin. Scale bar = 10 µM.
|
|
View this table: [in a new window] |
Table 1. Modulation of BCG-Induced Morphological Rearrangements of RAW264.7 Macrophages by the Specific Inhibitors of Distinct Signaling Pathways
|
 View larger version (31K): [in a new window] |
Figure 2. Signaling cascades activated by BCG mycobacteria. RAW 264.7 cells were pretreated or not with a specific kinase inhibitor, LY294002, 25 µM (A, B), PD98059, 50 µM (C), SB203580, 10 µM (D), or BAY11-7085, 10 µM (E), for 1 h at 37°C, and then infected with BCG at MOI = 10:1 for the indicated time periods. The cells were washed, lysed, and submitted to PI3 lipid kinase assay (A) or Western blot analysis (B–E), as described in Materials and Methods. The resulted membranes were treated with Ab against phosphorylated Akt (B), ERK (C), or p38MAPK (D) and then reprobed with Ab against the total protein. To study NF-B activation, nuclear (nucl) and cytoplasmatic (cyto) fractions of the cells were examined for the presence of p65 NF-B subunit, employing corresponding Ab (E).
|
95% viable by the Trypan blue exclusion test of cells treated with the drugs or vehicle). Previously, we demonstrated that BCG mycobacteria, viable or lysed, induce in MF phosphorylation of the p85 PI3K regulatory subunit [3 ], which subsequently binds and activates the p110 catalytic subunit of the kinase. A direct method of estimating PI3K activation is to assess the accumulation of one of its major lipid products, namely PIP3. Infection of RAW264.7 cells with BCG resulted in increased PIP3 levels measured by lipid kinase assay (Fig. 2A) . The kinase activation was sensitive to the PI3K inhibitor LY294002, which acts as a competitive inhibitor at the ATP binding site of the p110 PI3K subunit [20 ]. Macrophage pretreatment with this drug completely inhibited PIP3 accumulation (Fig. 2A) .
The major downstream effector of PI3K activity is Akt/PKB, a serine/threonine kinase, whose recruitment, phosphorylation, and subsequent activation is entirely dependent upon the lipid products of PI3K [21 ]. Consistent with these data, BCG induced sustained PKB phosphorylation on Ser473 that was observed up to 1 h after infection and PI3K inhibition abrogated the kinase activation in response to mycobacteria (Fig. 2B) .
Our data demonstrated that infection with BCG induced rapid activation of ERK within 15 min, whereas activation of p38MAP kinase was detectable 1 h after infection (Fig. 2C 2D) . Phosphorylation of both kinases was completely suppressed in the presence of specific kinase inhibitors, PD98059 and SB203580, correspondingly.
To suppress NF-B activation, we verified the inhibitory effect of BAY11-7085, IKKs inhibitor. Infection of RAW264.7 macrophages with viable BCG induced rapid translocation of activated NF-B p65 subunit to the nucleus that was abrogated in the cells pretreated with BAY11-7085 (Fig. 2E)
.
There were no significant differences in the kinetics of the kinase activation induced by viable (Fig. 2) or lysed mycobacteria (our previous data) [3 ], as well as in MF morphological rearrangements observed under the same experimental conditions (Fig. 1 , Table 1 ). Therefore, in our subsequent experiments, the cells were treated only with lysed bacteria.
Inhibitory analysis of the signaling pathways involved in regulation of RAW264.7 macrophage cytoskeletal rearrangements induced by BCG mycobacteria
To study the importance of the distinct kinase pathways for the cytoskeletal rearrangements observed in mycobacteria-treated cells, we evaluated morphological changes in the cells pretreated with the kinase pharmacological inhibitors.
The most pronounced effect on MF morphological changes was observed in the cells pretreated with inhibitor of PI3K (Table 1) . LY294002 dramatically suppressed the rate of BCG-induced filopodia protrusion and cell spreading at all time points of observation. The mean cell area of these cells was similar to that of control cells nontreated with BCG. Moreover, length/width ratio and the elongation index were significantly reduced, even in the control cells demonstrating the rounded-up cell shape and absence of polarization in the presence of the inhibitor. It should be noted that the effect of LY294002 was reversible since removal of the inhibitor restored the MF spreading induced by mycobacteria (data not shown). Scanning electron microscopy (SEM) analysis of cell morphology revealed that LY-treated cells were rounded or star-shaped, tended to be much less elongated, and, additionally, displayed rather few filopodia (or uropods) and many dorsal membrane projections presumed to be microvilli (Fig. 3 ). More detailed analysis demonstrated (Fig. 4B ) that the number of these projections was considerably reduced on the polarized cells treated with BCG in comparison with control cells, whereas pretreatment with LY significantly increased microvilli formation on the MF surface membrane. Flow cytometry analysis of the F-actin content in macrophages pretreated with different kinase inhibitors demonstrated that LY, but not any other used inhibitor, significantly reduced the amount of F-actin observed in the cells stimulated with BCG, suggesting suppression of actin polymerization (Fig. 4A) . The data suggest the reduced cell capacity to protrude filopodia in the presence of LY and/or contraction of the existing protrusions that lead to the ruffling of the cell surface through the recruitment of an adhered or latent intracellular membrane.
 View larger version (117K): [in a new window] |
Figure 3. Specific kinase inhibitor effects on the morphological changes of BCG-stimulated RAW264.7 macrophages. (A) The cells were pretreated with LY294002 (LY), PD98059 (PD), SB203580 (SB), BAY11-7085 (BAY) or DMSO (C) for 60 min and then stimulated with BCG-WCL for 24 h (BCG). The resulted cell cultures were fixed and processed for scanning electron microscopy. Scale bars = 20 µm.
|
 View larger version (84K): [in a new window] |
Figure 4. PI3K inhibitor reduced F-actin content and cell protrusions but not membrane ruffling in BCG-stimulated RAW264.7 macrophages. (A) The cells were grown in 6-well tissue plates, pretreated with the kinase inhibitors, and then stimulated with BCG-WCL as indicated in Fig. 3
. The resulting cells were stained for F-actin, scraped from the plate, and analyzed by flow cytometry. The results are presented as a geometrical means of the FACS histograms that were normalized to untreated, control samples. (B) The cells were grown and treated with LY294002 for 1 h and BCG-WCL for 24 h. Surface membrane morphology was evaluated by scanning electron microscopy. Scale bar = 10 µm. Note intensive membrane ruffling in LY-treated cells.
|
In contrast to ERK, p38MAPK inhibition by the specific inhibitor SB203580, had no effect on MF spreading and polarization induced by BCG (Fig. 3 , Table 1 ), suggesting the minor role of p38MAP kinase in MF cytoskeletal rearrangements induced by mycobacteria.
Macrophage activation by BCG leads to NF-B activation and to subsequent transcription of NF-B-dependent genes for chemokines and colony-stimulating factors that could induce MF migratory activity in an autocrine manner. To verify the role of NF-B- dependent factors in BCG-induced changes of MF shape, NF-B activation was suppressed with BAY11-7085. NF-B inhibition had no influence on early MF spreading and a slight inhibitory effect on MF polarization that was 20% reduced in comparison with cells treated with BCG for 24h (Table 1)
. Nevertheless, the mean cell area of the cells treated with BAY was 20% larger, the cells continued to be more spread, less polarized, than the cells treated solely with BCG (Fig. 3)
. The data suggest that NF-B-dependent factors could promote MF polarization induced by BCG but have no effect on early MF spreading, thus demonstrating the importance of the direct effect of mycobacteria both on spreading and subsequent polarization.
To study which MF receptors could mediate the mycobacterial effect on cell shape, we verified the role of β2-integrins known to mediate cell adhesion (CD18/LFA-1), as well as to bind mycobacteria (complement receptor, CR3). B2-integrins were blocked by the treatment of MF with neutralizing Ab, and the cells were stimulated with BCG-WCL. In spite of antibody treatment, BCG induced numerous radial protrusions of F-actin-rich filopodia around the cell margin within 15 min of stimulation with lysed bacteria (Fig. 5A-B ). Similar changes were observed in MF cultures treated solely with BCG (data not shown), demonstrating that blockade of β2-integrins had no effect on filopodia protrusion. Subsequent MF radial spreading that was clearly observed 30 min after stimulation with BCG (Fig. 1) was significantly inhibited in the cultures pretreated with anti-β2 integrin antibody (Fig. 5C 5D) , but not with isotype control Ab (data not shown). Filopodia contraction could be observed, suggesting that β2 integrins blockade reduced filopodia adhesion, thus preventing cell spreading.
 View larger version (150K): [in a new window] |
Figure 5. Blockade of β2- integrins with Ab inhibits BCG-induced RAW264.7 spreading through the inhibition of adhesion of pseudopodia protrusions. Macrophages were pretreated with anti-β2 integrin Ab, 10 µg/ml, for 30 min, and then treated with BCG-WCL for 15 min (A, B) and 3 h (C, D). The cells were fixed, stained with phalloidin-TRITC and examined under DIC (A, C) and confocal microscopy (B, D). The images show a basal section of the cell samples. Arrows indicate numerous filopodia (A, B) that disappeared 3 h after BCG stimulation reducing cell spreading (C, D).
|
To verify the induction of MF spreading through the TLR2 activation, we treated RAW264.7 with purified mycobacterial p19 lipoprotein, a known TLR2 agonist [7 ] or BCG-WCL and observed morphological changes of MF shape using phase contrast microscopy to quantify spreading and elongation index (Fig. 6A ). The changes observed in p19- treated cells resembled those induced by whole bacteria. The spreading index increased within 30 min after the treatment with p19, reaching sevenfold enlargement after 3 h. Cell polarization started 3 h posttreatment. After 24 h, the majority of the cells in culture presented an elongated form, providing a fourfold rise of the elongation index.
 View larger version (63K): [in a new window] |
Figure 6. BCG, as well as mycobacterial lipoprotein p19, induce cell spreading, PI3K, and Rac-1 activation through TLR2. RAW264.7 macrophages were treated with BCG-WCL, 1 µg/ml, or p19, 5 µg/ml, and the cells were examined for spreading kinetics during the indicated period of time (A). PI3K activation in p19- stimulated cells, pretreated or not with anti-TLR2 neutralizing (Ab1) and isotype control Ab (Ab2), was evaluated either through analysis of tyrosine phosphorylation of immunoprecipitated p85 PI3K regulatory subunit by Western blot analysis (B), or through immunolocalization of p85 PI3K in the cells treated with anti-p85 Ab using fluorescent microscopy (C). Note perinuclear localization of the kinase in untreated cells and its diffuse redistribution corresponding to membrane localization in p19-stimulated cells that was abrogated in the presence of anti-TLR2. (D) Cells treated with BCG-WCL for 10–30 min were lysed and immunoprecipitated using either anti- p85 PI3K or anti-TLR2 Ab. The immune complexes were submitted to 7.5% SDS-PAGE. The proteins were then blotted to membrane, treated with either anti-PI3K or anti-TLR2 Ab, and corresponding secondary Ab, and visualized by ECL cocktail. Note association of PI3K and TLR2 in obtained immunoprecipitates. The p85 PI3-K subunit bounds to 83kDa TLR2 protein. H chain, Ab heavy chain. (E) The cells were pretreated with LY, 25 µM, for 1 h and then stimulated with BCG-WCL, 1 µg/ml, for 10–30 min. Cell lysates were examined for the Rac-1 phosphorylation. Each sample of proteins, 100 µg, was submitted to 15% SDS-PAGE, blotted to membrane, and treated with anti-p-Rac-1 Ab. The resulting membrane was srtipped and reprobed with anti-Rac-1.
|
induction by p19 (data not shown) that was used subsequently. The data presented in Fig. 6B
demonstrate that the tyrosine phosphorylation of the p85 PI3K regulatory subunit observed in the MF treated with p19 for 5-30 min was inhibited in the cells pretreated with anti-TLR2 but not with unrelated isotype control Ab. Activation of PI3K was associated with redistribution of the p85 PI3K within the cells (Fig. 6C)
. In control cells the p85 protein was localized predominantly in the perinuclear zone of the cytoplasm. After stimulation with p19 for 10–30 min, the protein translocated to the cytoplasmic membrane, which could be observed as diffuse localization of the protein in a more superficial (apical) region of cells, reflecting the kinase recruitment to the superficial membrane along with activation. The cell pretreatment with anti-TLR2 suppressed p85 redistribution, as well as abrogated the MF spreading (Fig. 6C)
. To confirm PI3K engagement through the TLR2 binding, we treated RAW264.7 cells with BCG-WCL for 10 and 30 min and examined association of TLR2 with the p85 regulatory subunit in a coimmunoprecipitation experiment. The antibody against TLR2 (Santa Cruz) precipitated two proteins of 83 and 110 kDa that were recognized by the same Ab in Western blot analysis (Fig. 6D) . Treatment of the same precipitated samples with anti-p85 PI3K revealed the presence of a 85-kDa protein. Immunoprecipitation by Ab against PI3K demonstrated that p85 PI3K subunit was associated with TLR2 protein of 83 kDa. Thus we have shown that p85 and cytosolic domain of TLR2 physically associate in a signaling complex dependent on the stimulus.
Additionally, we verified activation of Rac-1 small GTPase in BCG-treated MF and the presence of this protein in TLR2/PI3K complex, as Rac1 was demonstrated previously to bind TLR2 and PI3K in cells stimulated with another TLR2 agonist [14 ].
Treatment with BCG for 10 and 30 min induced phosphorylation of Rac-1 in RAW264.7 cells that was inhibited in the cells pretreated with LY (Fig. 6E) , demonstrating PI3K-dependent character of Rac-1 activation by mycobacteria. Nevertheless, we failed to find Rac-1 protein in TLR2 immunoprecipitates prepared under the same conditions (data not shown).
The data obtained provide the evidence that mycobacterial TLR2 agonist induces in MF PI3K activation and morphological changes resembling those induced by BCG mycobacteria, and that PI3K activation by mycobacteria could be triggered through the TLR2.
Macrophages from TLR2-deficient mice display less morphological changes in response to BCG
To study the effect of BCG on morphology of primary MF expressing or not expressing TLR2, we examined the cytoskeletal rearrangements in bone marrow MF (BMM) derived from TLR2 –/– and TLR2 +/+ mice. To reduce cell polarization of BMM normally observed in the presence of M-CSF, the cell cultures were incubated for 24 h in CM without L-929-conditioned medium supplement containing M-CSF and then treated with BCG-WCL or p19 for additional 24 h. The data demonstrate that the average length of the cells in the BCG- or p19- stimulated cultures of TLR2 –/– BMM was at least twofold less than that in TLR2 +/+ cultures (Fig. 7
), confirming the importance of TLR2 expression for the MF cytoskeletal rearrangements in response to BCG.
 View larger version (97K): [in a new window] |
Figure 7. TLR2 –/– BMM are less sensitive to induction of cell polarization by BCG or p19 lipoprotein. (A) Macrophages derived from bone marrow of TLR2 –/– and control TLR2 +/+ C57Bl/6 mice were treated or not (C) with BCG-WCL, 1 µg/ml, or p19, 5 µg/ml, for 24 h, and then stained with phalloidin-TRITC. Images were obtained using DIC (A–C) or fluorescence microscopy (D–F). Scale bar = 10 µM. (B) Cell length was estimated by manual tracing and measuring of the cell footprint area, including protrusions using Image J software. The data are presented as a cell length fold induction in comparison with control, unstimulated, cells (±SD). Asterisks indicate values significantly different (**, P<0.01, Student's t test) from those obtained from the corresponding control cells.
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Since cell polarization is not absolutely dependent on the presence of a gradient of external stimulus and can occur stochastically [23 ], we studied the response of cultured RAW264.7 MF to mycobacteria by the registration of the change from spherical to polarized shapes either in the presence of viable BCG or when exposed to a uniform concentration of mycobacterial components. To evaluate a direct effect of mycobacteria on MF, all experiments were performed in serum-starved medium to reduce the effect of serum on cell adhesion, intracellular signaling, and proliferation, as well as bacterial opsonization.
The data obtained demonstrate that viable BCG mycobacteria, BCG-WCL, and purified mycobacterial lipoprotein p19 had a similar effect on the MF shape, inducing numerous F-actin-rich filopodia protrusions as early as 15–30 min after treatment, leading to intensive cell spreading and, subsequently, to cell elongation and polarization. Cell elongation was accompanied by reduction of filopodia number and F-actin content. Actin polymerization was maintained predominantly in the region of the leading edge of the protrusive lamellipodia in bipolar cells. The morphological changes observed are typical for the migrating MF suggesting that some mycobacterial components, including p19 cell wall lipoprotein, could induce MF motility.
It should be noted that purified p19, a known TLR2 agonist [7 ] stimulated MF spreading at a fivefold higher concentration than mycobacterial lysate. TLR2, as well as other pattern recognition receptors, presents low affinity for the agonist molecules [24 ]. Ligand loading to TLR increases either in the presence of a higher number of agonist molecules or through additional binding to synergistic receptors, CD14, for example [25 ]. In our test system, p19 binding to CD14 is unlikely since no opsonization is possible in the serum-starved conditions we used, nevertheless, increasing the agonist concentration allowed to enhance TLR2 signaling.
To study mycobacteria-induced signal transduction leading to MF response, we performed an inhibitory analysis of kinase pathways possibly involved in regulation of cytoskeletal rearrangements employing specific pharmacological inhibitors.
Suppression of PI3K with LY294002 significantly inhibited cell protrusions induced by BCG leading to reduction of the number of filopodia and lamillopodia and abrogating MF spreading and polarization. Most likely, the effect was mediated by inhibition of actin polymerization since the content of F-actin was significantly reduced in the cells pretreated with this inhibitor.
Another factor that could limit pseudopod extension and spreading is availability of the cytoplasmatic membrane [26 ]. SEM analysis of the cell surface confirmed that PI3K inhibition led to suppression of MF protrusions, rounding of the cells, formation of occasional long, spindly uropods, and to intensive ruffling of the dorsal plasma membrane. The morphological changes observed suggest that not only extension of new filopodia was abrogated. Reduction of control cell adherence in the presence of LY294002 led to contraction of existing filopodia, liberation of cell membrane, and formation of numerous dorsal microvilli on the cell surface. Thus, the limitation of pseudopod extension in the presence of PI3K inhibitor occurred in spite of high membrane availability in these cells.
The role of class I PI3K activation in morphological changes of MF during phagocytosis of mycobacteria and other opsonized microorganisms was well elucidated, establishing the importance of PIP3 production for the rearrangements of cortical actin that are necessary for phagosome formation [27 ]. Our data demonstrate that mycobacteria-induced PI3K activation is essential for changes of MF shape as well, in a manner independent from bacterial internalization since lysed and viable BCG provided a similar effect on polarization of these cells.
Other kinase inhibitors employed in this study had less pronounced, if any, effect on MF cytoskeletal rearrangements. Suppression of MEK1/ERK with PD98059 led to moderate reduction of BCG-induced polarization. This effect could be explained by the known importance of ERK activation in generating a leading pseudopod via regulation of myosin light chain kinase and phosphorylation of myosin light chain leading to actin-myosin assembly during cell migration [28 ]. Nevertheless, ERK and p38MAPK inhibitors had no significant effect on early spreading of mycobacteria-treated MF in contrast to MF exposed to oxidative stress [11 ] or LPS [12 ] that were demonstrated to be sensitive to ERK and p38 inhibition.
Macrophage proinflammatory activation by mycobacteria is known to be mediated by Toll-like receptors, mainly TLR2. TLR2-dependent pathways leading to NF-B and MAPK activation and subsequent transcription of inflammatory genes were well established [2
]. Nevertheless, the role of TLR2 in regulation of MF spreading and polarization was not elucidated.
Our data demonstrating that TLR2 agonist, p19, as well as BCG, could induce cell spreading in TLR2 +/+, but not in TLR2 –/– or anti-TLR2- pretreated MF, suggest that some pathways triggered by this receptor could regulate MF cytoskeletal rearrangements. Because activation of PI3K was essential for the observed morphological changes, we verified whether TLR2 could be involved in PI3K activation by mycobacteria.
The results demonstrate that purified mycobacterial p19 lipoprotein induced PI3K activation in a manner similar to BCG. Cell treatment with p19 induced phosphorylation of p85 PI3K regulatory subunit and its translocation to the surface membrane. PI3K activation and MF spreading were abrogated in the cells pretreated with TLR2- neutralizing Ab. Additionally, we detected physical binding of TLR2 with activated PI3K in BCG-treated cells. These results clearly demonstrate that TLR2 could mediate mycobacteria-induced PI3K activation and subsequent cytoskeletal rearrangements leading to MF spreading and polarization.
The data obtained are in accordance with observation of Arbibe et al. demonstrating that heat-killed S. aureus, another TLR2 agonist, induces PI3K activation through the binding with TLR2 [14 ]. Recently, the role of CD14 has been revealed as essential for the BCG-induced PI3K activation through the TLR2 [6 ]. The authors demonstrated that BCG or purified Man-LAM were able to induce significant PI3K activation only after opsonization by serum or lipopolysaccharide binding protein (LBP), which enhanced mycobacteria binding to CD14-overexpressing THP-1 monocytes promoting signaling through the TLR2. Additionally, TLR4 engagement by LPS or minimally modified low-density lipoproteins, LDL [29 ], was demonstrated to induce PI3K activation and MF spreading in a CD14-dependent manner.
In our experimental model, nonopsonized BCG, as well as purified p19, was able to induce phosphorylation of the p85 regulatory subunit and kinase activity of PI3K, demonstrating that mature RAW264.7 MF could be activated by TLR2 agonists directly, probably because of a higher level of TLR2 expression than in monocytic cells. Nevertheless, our data do not exclude the possibility that CD14 expressed on RAW264.7 cells could increase mycobacterial binding in the presence of serum, thus promoting TLR2/PI3K signaling.
To verify the role of other receptors possibly involved in BCG-induced changes of MF shape, we treated the cells either with Ab against β2-integrins known to be activated by mycobacteria binding or adhesion to substrate, or with the specific NF-B inhibitor BAY11-7085 to suppress the NF-B-dependent synthesis of chemokines [3
], which could induce cytoskeletal rearrangements through the MF chemokine receptors in an autocrine manner.
B2 integrins are present in macrophage receptors known to bind mycobacteria (CR3; CD18) and to promote signaling leading to cell adhesion and phagocytosis. Treatment with β2-integrin- neutralizing Ab had no effect on filopodia extension but reduced BCG-induced MF spreading and polarization through the inhibition of adhesion of protruding filopodia that later underwent retraction. These data demonstrate the downstream position of β2-integrin activation in mycobacteria-induced PI3K-mediated signaling to morphological cell rearrangements. Our observation is in accordance with a recent study [6 ] demonstrating that BCG-induced PI3K pathway could lead to β2-integrin activation through the cytohesin-1 induction promoting CR3-mediated mycobacterial phagocytosis.
NF-B inhibition with BAY11-7085 had no effect on early filopodia extension and radial spreading but slightly reduced cell elongation 24 h after treatment with BCG. The cells continued to be more spread and less polarized in comparison with MF treated solely with mycobacteria. These data demonstrate that activation of NF-B by mycobacteria is not important for the early cytoskeletal rearrangements of MF, though it could contribute to cell polarization essential for migration, most likely through the production of MF- stimulating chemokines.
The most striking finding of this study is that PI3K activation by mycobacteria could be mediated by TLR2, and activity of this kinase is essential for actin assembly. Downstream molecules of the PI3K pathway leading to actin should be defined yet. The small GTPases, Rac, cdc42, and Rho previously have been shown to be involved downstream of PI3K in the reorganization of actin filaments [30 ]. In most cell types, Rac generates a protrusive force through the localized polymerization of actin and is responsible for the protrusion of lamellipodia.
It should be noted that Rac1, as well as p85 PI3K, were recruited to the TLR2 signaling complex in the cells treated with heat-inactivated S. aureus [14 ]. In our experimental model, we did not detect the presence of Rac1 in TLR2/PI3K complex immunoprecipitaed from BCG-treated RAW264.7 cells, although Rac-1 was found to be phosophorylated in these cells, and its activation was inhibited in the presence of PI3K inhibitor. The data suggest that Rac1 is activated downstream PI3K activation in BCG- stimulated cells, most likely, by PIP3 product of activated PI3K and could mediate actin polymerization and pseudopod extension.
The data obtained in this study suggest that TLR2 signaling triggered by BCG contributes to regulation of MF cytoskeletal rearrangements important for the cell motility. The recent data demonstrating that mycobacterial infection of TLR2 KO mice were characterized by reduced leukocyte influx into the pulmonary compartments, in contrast to their normal counterparts [31 ] support this hypothesis; nevertheless, it needs direct experimental verification.
The effect of TLR2 agonist on MF shape described in our manuscript is not limited with p19 and BCG. Induction of cell spreading was described in macrophage cultures infected with M. tuberculosis [32 ]. Moreover, TLR4 agonists, LPS and LDL, were able to induce MF spreading as well [29 , 33 ]. The effect was mediated by TLR4, since it was abrogated in macrophages from TLR4 –/– KO mice [33 ]. These data suggest that TLRs, at least TLR2 and TLR4, mediate transduction of signals leading to MF spreading and polarization known to be important for cell migration.
B Abs. We also thank V. R. Lanes and F. C. Lopes for excellent technical support. Received January 27, 2006; revised August 20, 2006; accepted August 22, 2006.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  L. A. Yeager, A. K. Chopra, and J. W. Peterson Bacillus anthracis Edema Toxin Suppresses Human Macrophage Phagocytosis and Cytoskeletal Remodeling via the Protein Kinase A and Exchange Protein Activated by Cyclic AMP Pathways Infect. Immun., June 1, 2009; 77(6): 2530 - 2543. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. He, W. Li, S.-Y. Chen, S. Zhang, Y.-T. Chen, Y. Hayashida, Y.-T. Zhu, and S. C. G. Tseng Suppression of Activation and Induction of Apoptosis in RAW264.7 Cells by Amniotic Membrane Extract Invest. Ophthalmol. Vis. Sci., October 1, 2008; 49(10): 4468 - 4475. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Narayana and K. N. Balaji NOTCH1 Up-regulation and Signaling Involved in Mycobacterium bovis BCG-induced SOCS3 Expression in Macrophages J. Biol. Chem., May 2, 2008; 283(18): 12501 - 12511. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
