The pulmonary collectin surfactant protein (SP)-A has a pivotal role in anti-inflammatory modulation of lung immunity. The mechanisms underlying SP-A-mediated inhibition of LPS-induced NF-κB activation in vivo and in vitro are only partially understood. We previously demonstrated that SP-A stabilizes IκB-α, the primary regulator of NF-κB, in alveolar macrophages (AM) both constitutively and in the presence of LPS. In this study, we show that in AM and PBMC from IκB-α knockout/IκB-β knockin mice, SP-A fails to inhibit LPS-induced TNF-α production and p65 nuclear translocation, confirming a critical role for IκB-α in SP-A-mediated LPS inhibition. We identify atypical (a) protein kinase C (PKC) ζ as a pivotal upstream regulator of SP-A-mediated IκB-α/NF-κB pathway modulation deduced from blocking experiments and confirmed by using AM from PKCζ−/− mice. SP-A transiently triggers aPKCThr410/403 phosphorylation, aPKC kinase activity, and translocation in primary rat AM. Coimmunoprecipitation experiments reveal that SP-A induces aPKC/p65 binding under constitutive conditions. Together the data indicate that anti-inflammatory macrophage activation via IκB-α by SP-A critically depends on PKCζ activity, and thus attribute a novel, stimulus-specific signaling function to PKCζ in SP-A-modulated pulmonary immune response.
Pulmonary surfactant is a lipid-protein complex that constitutes the alveolar liquid layer of the lung. The unique composition of surfactant facilitates the functional combination of different biological effects such as preventing alveolar collapsing at expiration and immunomodulating innate pulmonary host defense responses (1, 2). The latter function is primarily mediated by surfactant proteins (SP)3-A and SP-D belonging to the collectin family. SP-A, the most abundant pulmonary collectin, binds and aggregates a variety of microorganisms and enhances their phagocytosis and killing both in vivo and in vitro (3). SP-A-deficient mice exhibit delayed microbial clearance and higher levels of bronchoalveolar inflammatory mediators after intratracheal inoculation with a variety of clinically relevant pathogens (4), as well as to isolated LPS (5).
Alveolar macrophages (AM), normally accounting for ∼95% of airspace leukocytes, are a unique class of professional phagocytes that represent the major effector cells of the pulmonary innate immune system (6). SP-A directly interacts with AM through binding to cell surface receptors, resulting in modulation of chemotaxis, phagocytosis, and modified pro- or anti-inflammatory immune responses (7).
LPS, the main component of the outer leaflet of the outer membrane of Gram-negative bacteria, substantially contributes to the pathophysiology of the most frequent lung diseases, including pneumonia, acute lung injury, and acute respiratory distress syndrome (8). Whereas LPS recognition benefits the host by sensing bacteria and mobilizing defense mechanisms, an exaggerated response to LPS contributes to the development of a local or systemic septic shock syndrome. The interaction of picomolar concentrations of LPS with a receptor complex including TLR4, MD-2, LPS-binding protein, and CD14 on host cells initiates the sequential activation of multiple signaling pathways, including NF-κB, a central regulator of LPS-mediated cell activation (9, 10, 11).
The most prominent NF-κB heterodimer p65/p50 is sequestered in the cytoplasm by a family of inhibitory proteins, among which IκB-α is the best characterized and assumed to function as the primary regulator of NF-κB in both stimulated and resting cells (12, 13). In response to many stimuli including LPS, IκB-α becomes degraded via ubiquitination in an IκB kinase (IKK)-dependent manner (12, 14, 15). The removal of IκB-α allows the nuclear translocation of NF-κB and subsequently the transcription of downstream target genes, including IκB-α itself (16). Secondary clues point to an IκB-independent pathway, and include the phosphorylation of p65 to activate NF-κB-dependent gene transcription (17).
SP-A has been shown to inhibit LPS-induced TNF-α production (18, 19, 20, 21), inducible NO synthase protein expression (22), NF-κB activity (20, 23, 24), and, upon diverse stimuli, NADPH oxidase (25) in immunocompetent cells. We recently demonstrated that SP-A exerts its anti-inflammatory effects on LPS-challenged AM via a mechanism involving a SP-A-mediated direct modulation of the basal and LPS-coupled IκB-α turnover in these cells (24). In that study, SP-A increased IκB-α protein expression in a dose- and time-dependent manner without inducing IκB-α phosphorylation or IκB-α mRNA levels both basal and in the presence of LPS. However, the signaling pathways controlling the SP-A-mediated attenuation of LPS-induced proinflammatory signals via IκB-α remain unknown.
A role for individual members of the protein kinase C (PKC) family in both SP-A immune modulation (26) and IκB-α turnover (27) has been suggested previously. PKC family members are expressed in many different cell types, in which they regulate a wide variety of cellular processes such as cell differentiation, cytoskeletal remodeling, and gene expression in response to diverse stimuli (28, 29). Based on sequence homology and the mechanism of their regulation, individual members of the PKC family are subdivided into three classes as follows: classical PKCs (cPKC: α, βI, βII, and γ) are diacylglycerol (DAG) and Ca2+ dependent, whereas novel PKCs (δ, ε, η, and θ) are responsive to DAG, but are insensitive to Ca2+. The atypical (a) PKCs (ζ and λ/ι) do not bind to DAG and are not responsive to Ca2+ (30, 31).
In the present study, we confirmed the role of IκB-α in SP-A anti-inflammatory effects by using IκB-α knockout/IκB-β knockin (AKBI) AM and PBMC. We further investigated the role of PKC in SP-A-specific anti-inflammatory signaling in primary AM and identified the aPKCζ isoform as a central regulator of SP-A-mediated IκB-α/NF-κB modulation.
Materials and Methods
Mice and reagents
Primary cells were obtained from pathogen-free male Sprague-Dawley rats (Charles River Laboratories), from AKBI mice (32), or from PKCζ−/− mice generated as described before (33). SV129 and CD1 wild-type (wt) control mice were from Charles River Laboratories. Animal care and experiments were conducted according to protocols approved by the Schleswig-Holstein Ministry of Environment, Nature, and Forestation. All mice used were between 6 and 12 wk of age and were maintained at the Research Center Borstel animal facility under specific pathogen-free conditions.
The smooth LPS from Salmonella friedenau strain H909 was extracted by the phenol/water method, purified, lyophilized, and transformed into the triethylamine salt form. RPMI 1640 medium, DMEM, and Dulbecco’s PBS were from Invitrogen Life Technologies. Ham’s-F12 medium and FCS were from BioWhittaker. Poly(dI/dC) was purchased from Pharmacia. [γ-32P]ATP was supplied by Hartmann Analytics; T4 polynucleotide kinase was purchased from Roche. Rabbit polyclonal anti-IκB-α, anti-PKCζ (in the manuscript referred to as anti-aPKC), anti-p65, and HRP-conjugated goat anti-rabbit IgG were from Santa Cruz Biotechnology; a specific Ab against PKCζ was generated as described (34); anti-phospho PKCζ/λ (Thr410/403) Ab was from Cell Signaling Technology. Gö-6850 was from Calbiochem; chelerythrine chloride was from Tocris Bioscience; Gö-6976 was from Alexis Biochemicals; apigenin and wortmannin were from Sigma-Aldrich; and aPKC pseudosubstrate (ps) peptide was from BioSource International. Myelin basic protein and ATP/magnesium mixture were from Upstate Biotechnology; C1q was from Advanced Research Technologies. All other reagents (except as noted) were obtained from Sigma-Aldrich.
Human SP-A was purified from the bronchoalveolar lavage of patients with alveolar proteinosis, as described in detail (35). Briefly, the lavage fluid was treated with butanol to extract SP-A, and the resulting pellet was sequentially solubilized in octylglucoside and 5 mM Tris (pH 7.4). SP-A was treated with polymyxin B agarose beads to reduce endotoxin contamination. SP-A preparations were tested for the presence of bacterial endotoxin using a Limulus amebocyte lysate assay (BioWhittaker); all SP-A preparations used contained <0.2 pg endotoxin/μg SP-A.
Stimulation of rat AM
AM were isolated, as described previously (22). Cells were plated at 0.8 × 106/500 μl in 24-well plates (Nunc) and allowed to attach for 90 min at 37°C in a 5% CO2 atmosphere. Then the medium was changed and the cells were treated with SP-A (20–60 μg/ml) and/or LPS (10–100 ng/ml) for indicated times at 37°C in the presence of 0.2% heat-inactivated (HI) FCS. In separate experiments, AM were treated with a panel of kinase inhibitors to determine their possible effect on IκB-α protein expression as follows: Gö-6976 (5 μM), Gö-6850 (5 nM), chelerythrine chloride (6 and 12 μM), or aPKCps peptides (2, 5, and 10 μM) were used to inhibit distinct subsets of cPKC, novel PKC, or aPKCs. Wortmannin (50 nM) was used as PI3K inhibitor, and apigenin (30 μM) to inhibit protein kinase CKII.
Nuclear protein extraction
After treatment, cells were scraped off with 500 μl of cold PBS and spun at 4,500 × g, 5 min, 4°C. The resulting pellet was resuspended in 400 μl of ice-cold buffer A (10 mM Tris, 5 mM MgCl2, 10 mM KCl, 1 mM ethyleneglycol-bis-(2-aminoethyl ether)-N,N,N′,N′-tetraacetic acid, 0.3 M sucrose, 1 mM DTT, 10 mM β-glycerol phosphate, 0.5 mM PMSF, and 1.5 μl of protease inhibitor mixture (Complete; Roche)) and incubated on ice for 15 min; 25 μl of 10% Nonidet P-40 was then added and vortexed for 10 s. The nuclei were pelleted by centrifugation at 4,500 × g, 5 min, 4°C. The supernatant was taken to represent the cytosolic fraction. The pellet was resuspended in 30 μl of buffer B (20 mM Tris, 5 mM MgCl2, 320 mM KCl, 0.2 mM ethyleneglycol-bis-(2-aminoethyl ether)-N,N,N′,N′-tetraacetic acid, 25% glycerol, 1 mM DTT, and the mixture of protease inhibitors mentioned above) and incubated on ice for 15 min, followed by centrifugation at 16,100 × g, 10 min, 4°C. The supernatant containing the nuclear fraction was transferred to a new vial. Cytosolic cell fractions (30–40 μg of protein) were immunoblotted for IκB-α. Nuclear extracts (2 μg of protein) of the cells were analyzed by EMSA for NF-κB DNA-binding activity.
Isolation and stimulation of mouse AM and PBMC
Gene-deficient mice and respective controls were killed by i.p. injection of pentobarbital, followed by exsanguination by cardiac puncture for PBMC isolation. The lungs were lavaged with 1 ml of PBS containing 0.2 mM EDTA. AM were plated at a density of 2 × 105/ml in 24-well plates (Nunc) in the presence of 0.2% HI-FCS. PBMC were isolated by density gradient using Ficoll-Histopaque-1083 (Sigma-Aldrich). Heparinized blood was mixed with an equal amount of HBSS and centrifuged at 420 × g. The interphase layer containing the PBMC fraction was collected and washed twice with HBSS and once in serum-free RPMI 1640. Cells were plated at a density of 1 × 106/ml in 96-well plates. AM and PBMC were left untreated or treated with SP-A (20–60 μg/ml) for 1 h and/or stimulated with LPS (0.1–100 ng/ml) for 4 h at 37°C in the presence of 0.2 and 10% HI-FCS, respectively. After stimulation, cells were centrifuged at 200 × g, and cell-free supernatants were collected for TNF-α determination. In separate experiments, AM were left untreated or treated with SP-A (40 μg/ml) for 1 h, and/or stimulated with LPS (100 ng/ml) for 1 h at 37°C in the presence of 0.2% HI-FCS. Cytosolic cell fractions (30–40 μg of protein/lane) were immunoblotted for IκB-α. Nuclear extracts (2 μg of protein/lane) of the cells were analyzed by EMSA for NF-κB DNA-binding activity.
TNF-α was determined in pooled cell-free supernatants of stimulated AKBI and wt control cells by sandwich DuoSet ELISA using goat anti-mouse TNF-α Ab and biotinylated goat anti-mouse TNF-α Ab (R&D Systems), according to the manufacturer’s protocol.
Rat AM (2 × 106/well) were stimulated for the indicated times with SP-A, LPS, or, for some experiments, the complement protein C1q that shares structural and functional homology with SP-A. After culture, cells were lysed on ice for 30 min in 500 μl of lysis buffer (50 mM Tris-HCl (pH 8.0), 150 mM NaCl, 1% Nonidet P-40). The lysates were spun at 9300 × g for 15 min, and the supernatants were precleared by adding protein A-agarose (50 μl) and incubated at 4°C for 45 min, followed by centrifugation at 9300 × g for 10 min. The precleared supernatant was incubated with anti-aPKC Ab, anti-p65 Ab, control IgG, or no Ab for 2 h at 4°C, after which 50 μl of protein A-agarose was added for 2 h at 4°C with gentle rotation. The immune complexes were collected by centrifugation at 9300 × g for 5 min at 4°C, washed three times with cold lysis buffer, and released by boiling with 5× sample buffer. Samples were used to determine kinase activity or for Western analysis.
Western analysis was performed on cytosolic extracts, membrane fractions, and immunoprecipitated samples from rat and mouse cells. Experiments using the inhibitors above were performed under four different conditions, as follows: 1) in the absence of SP-A and LPS (basal); 2) in the presence of SP-A (40 μg/ml, 1 h) (constitutive); 3) in the presence of LPS (10 ng/ml, 30 min) (induced); and 4) in the presence of SP-A (40 μg/ml, 1 h) before LPS (10 ng/ml, 30 min) (modulated). After treatment, cytosolic fractions were assayed for protein content by the bicinchoninic acid reagent (Pierce Biotechnology), separated on SDS-PAGE, and transferred to nitrocellulose membrane. The membranes were then incubated with anti-IκB-α, anti-aPKC, anti-PKCThr410/403, anti-p65, or β-actin (mouse monoclonal) at a 1/700, 1/200, 1/700, 1/200, and 1/1000 dilution, respectively. Goat anti-rabbit IgG-HRP or rabbit anti-mouse IgG-peroxidase conjugate served as secondary Abs. Immunoreactive proteins were visualized using the ECL Western blotting detection system (Amersham).
In vitro aPKC kinase assay
Immunoprecipitated aPKC was subjected to the kinase assay in 100 μl of reaction mixture (20 mM MOPS (pH 7.2), 25 mM β-glycerol phosphate, 1 mM DTT, 5 mM EGTA, 1 mM Na3VO4, 75 mM MgCl2, 0.5 mM ATP, 100 μCi of [γ-32P]ATP) and 10 μl of myelin basic protein substrate at 37°C with gentle shaking. Reaction was stopped by adding 50 μl of Laemmli buffer. Immunocomplexes were released from agarose beads by boiling and centrifugation. The supernatant was transferred to a new vial and subjected to 12% SDS-PAGE. The gel was analyzed with a PhosphorImager (Amersham Biosciences) to quantitate band intensities.
NF-κB activation assay
After exposing the cells to the experimental conditions, nuclear extracts were prepared and analyzed, as described above. The activity of NF-κB in the nuclear extracts was determined by EMSA. NF-κB oligonucleotides were end labeled with [γ-32P]ATP using T4 kinase. Two micrograms of crude nuclear extract was incubated for 20 min in binding buffer containing 50 μg of poly(dI/dC)/ml with 7.5 fmol of the 32P-labeled oligonucleotides encoding the consensus NF-κB site 5′-AGCTCAGAGGGGACTTTCCGAGAGAGC-3′. Samples were separated by electrophoresis in 5% polyacrylamide gels for 2 h at 180 V, after which gels were analyzed with a PhosphorImager.
Assay for aPKC membrane translocation
Rat AM were treated with SP-A (40 μg/ml) for the times indicated. Incubation was stopped with ice-cold PBS, and the cells were resuspended in 300 μl of hypotonic fractionation buffer (10 mM Tris (pH 7.4), 4.5 mM EDTA, 2.5 mM EGTA, 2.3 mM 2-ME, 1 mM PMSF, 10 μg/ml aprotinin, 10 μg/ml leupeptin, 10 μg/ml pepstatin, and 0.1 mM Na3VO4), incubated for 20 min at 4°C while rotating, and sonicated. Total cell lysates were then centrifuged at 100,000 × g for 30 min at 4°C to separate cytosolic from particulate fractions. The resulting pellets were extracted in 150 μl of hypotonic fractionation buffer containing 0.5% Triton X-100 and centrifuged at 16,000 × g for 20 min at 4°C. The resulting supernatant was taken to represent the membrane fraction. Equal amount of membrane protein for each sample (30–40 μg) was separated by SDS-PAGE and blotted with anti-aPKC Ab.
AM were plated at a density of 1 × 105 cells on 8-well Lab Tek II chamber slides (Nunc). Cells were allowed to attach for 90 min at 37°C in a 5% CO2 atmosphere in the presence of 0.2% HI-FCS. After incubation with the indicated stimuli, cells were fixed with ice-cold (−20°C) methanol, washed with PBS, and permeabilized with 0.25% Triton X-100. After a repeated washing, the cells were blocked with 10% BSA/PBS, and incubated with isoform-specific anti-PKCζ Ab, anti-p65 Ab, or control IgG at a 1/250, 1/50, or 1/100 dilution, respectively. Goat anti-rabbit IgG conjugated to Alexa Fluor 488 served as secondary Ab. Cell nuclei were counterstained with propidium iodide. Samples were analyzed using a Leica TCS SP confocal laser scanning microscope (Leica Microsystems). All images were acquired under identical settings with Leica TCSNT software and assembled using Adobe Photoshop 6.0. An average of 120 (Fig. 1,F) or 100 (Fig. 5, C and E) cells was counted per condition and experiment, which was repeated three times.
Data were statistically analyzed, as indicated in the figure legends using GraphPad Prism (version 4.0; GraphPad). Values were considered significant when p < 0.05. Data are presented as SEM.
SP-A fails to inhibit LPS-induced TNF-α release by AKBI cells
To confirm our previous findings on the role of IκB-α in anti-inflammatory activity of SP-A, in this study, we used primary immune cells from IκB-α-deficient mice. In AKBI mice, the integrated IκB-β gene is under the control of the IκB-α promoter, and at the same time a null mutation in IκB-α is introduced (32). Unlike IκB-α-deficient mice, which are postnatal lethal (36), AKBI mice have a normal phenotype, and NF-κB induction by PMA or TNF-α in AKBI mouse fibroblasts and thymocytes is identical compared with wt cells (32). In support of this notion, AKBI mice reveal a functional redundancy of IκB-α and IκB-β upon LPS challenge in vivo (37). In line with these data, we show in this study that PBMC (Fig. 1,A) and AM (Fig. 1,C) from AKBI mice released similar amounts of TNF-α in response to LPS (0.1–100 ng/ml) compared with wt cells (Fig. 1, B and D). However, whereas SP-A (20–60 μg/ml) significantly (p < 0.05 to p < 0.001) and in a dose-dependent manner inhibited LPS-induced TNF-α release by wt cells (Fig. 1, B and D), SP-A failed to inhibit LPS-induced TNF-α release by PBMC (Fig. 1,A) or AM (Fig. 1 C) from AKBI mice. These data confirm that SP-A-mediated inhibition of LPS-induced TNF-α production in PBMC and AM critically depends on the presence of IκB-α, and that IκB-β could not compensate the lack of IκB-α under these conditions.
SP-A fails to inhibit LPS-induced p65 nuclear translocation in AKBI AM
Because p65 nuclear translocation is tightly linked to NF-κB-dependent TNF-α gene induction (15), we next investigated the effect of SP-A on LPS-induced p65 localization in AKBI and wt AM by confocal microscopy. LPS-induced p65 nuclear accumulation in wt AM (Fig. 1,Ec) was almost completely prevented (60 ± 5%) by pretreatment of the cells with SP-A (Fig. 1, Ed and F). In contrast LPS-induced p65 nuclear accumulation in AKBI AM (Fig. 1,Eg) was not inhibitable by SP-A (Fig. 1, Eh and F). As expected, comparable amounts of nuclear p65 are seen under resting or SP-A conditions in wt and AKBI AM. Together the data suggest that SP-A-mediated inhibition of LPS-induced p65 nuclear translocation critically depends on the presence of IκB-α.
An aPKC is involved in IκB-α stabilization
To elucidate the underlying molecular mechanisms of SP-A-mediated immune protection, we focused on identifying the upstream regulators involved in the stabilization of IκB-α by SP-A. Because members of the PKC family have been shown to be involved in IκB-α turnover (27) and SP-A immune functions (26), we used a panel of PKC inhibitors with different specificity to test the role of PKC in SP-A-mediated IκB-α stabilization in primary rat AM. Experiments were performed under four different conditions as follows: 1) distinct inhibitors alone (basal, Fig. 2,A); 2) inhibitors plus SP-A (40 μg/ml) (constitutive, Fig. 2,B); 3) inhibitors plus LPS (10 ng/ml) (induced, Fig. 2,C); and 4) inhibitors plus SP-A (40 μg/ml), followed by LPS (10 ng/ml) (modulated, Fig. 2 D). Cytosolic IκB-α protein expression was determined by Western analysis.
Neither of the inhibitors had a significant effect on IκB-α protein expression under basal conditions (Fig. 2,A). In line with our previous data (24), SP-A significantly increased IκB-α protein expression by 210 ± 44% (p < 0.05) under constitutive conditions (Fig. 2,B) and by 137 ± 10% (p < 0.05) in the presence of LPS (Fig. 2,D). cPKC isoforms (α, βI, βII, γ), which are diacylglycerol (DAG) sensitive and Ca2+ responsive, are specifically inhibited by Gö-6976. The effect of SP-A on IκB-α stabilization was still significant (149 ± 12% of control) in the presence of Gö-6976 (Fig. 2,B), but was abolished in the presence of LPS (Fig. 2,D). Whereas treatment of AM with Gö-6850, which inhibits the cPKCs α and γ as well as the DAG-sensitive but Ca2+-insensitive novel PKC isoforms δ and ε, did not inhibit the SP-A-mediated stabilization of IκB-α (Fig. 2,B), it inhibited SP-A’s effect in the presence of LPS (Fig. 2,D). In sharp contrast, treatment of the cells with chelerythrine chloride, an inhibitor of Ca2+- and DAG-dependent and DAG-independent PKCs including the DAG-insensitive atypical isoforms ζ and λ/ι, abolished the stabilizing effect of SP-A on IκB-α both constitutively and in the presence of LPS, suggesting that aPKCs might be involved in SP-A effects (Fig. 2, B and D).
Both SP-A and LPS can activate PI3K in human macrophages (38) and in THP-1 cells (39), respectively. In the present study, the PI3K inhibitor wortmannin significantly reduced IκB-α expression in the presence of LPS (Fig. 2,C). Wortmannin abrogated SP-A’s effect on IκB-α stabilization both constitutively and in the presence of LPS (Fig. 2, B and D). In mouse embryonic fibroblasts, CKII is critically involved in the basal turnover of IκB-α (12). Apigenin, a selective CKII inhibitor, inhibited IκB-α stabilization by SP-A constitutively and in the presence of LPS (Fig. 2, B and D).
Because, among aPKCs, PKCζ has been shown to modulate IκB-α turnover (27) and IKK/NF-κB activation in both nuclear and whole lung extracts (33), we wanted to elucidate the role of aPKC isoforms in SP-A’s AM immunomodulation more precisely. Therefore, we used the isoform-specific cell-permeable inhibitory myristoylated peptide derived from the ps motif of aPKCs. aPKCps mimics the substrate and maintains aPKC in its nonactive form. Of note, the ps peptides are not totally specific, because the ps sequences (SIYRRGARRWRKL) are identical in both isoforms PKCζ and PKCλ/ι (40).
aPKC inhibition suppresses SP-A-mediated IκB-α stabilization
Again, cytosolic IκB-α protein in AM was determined under four different conditions (basal, in the presence of SP-A, or LPS, or both). Treatment of the cells with aPKCps (2–10 μM) did not significantly affect IκB-α protein level under basal (Fig. 3,A) or LPS (Fig. 3,C) conditions. Compared with basal conditions, SP-A significantly increased IκB-α protein expression both constitutively (Fig. 3,B) and in the presence of LPS (Fig. 3,D). Treatment of the cells with aPKCps at 10 μM resulted in a significant inhibition of IκB-α by SP-A (Fig. 3,B). Interestingly, in the presence of LPS, aPKCps pretreatment at any concentration resulted in a significant inhibition (p < 0.02) of IκB-α protein levels by SP-A (Fig. 3 D) when compared with the corresponding aPKCps concentration in the absence of SP-A. As expected, LPS-induced NF-κB activity in AM was abolished after pretreatment with aPKCps (data not shown). Taken together, these results suggest that an aPKC isoform is involved in IκB-α stabilization by SP-A both constitutively and in the presence of LPS.
SP-A stimulates aPKCThr410/403 phosphorylation and kinase activity
To investigate whether SP-A can stimulate aPKC activation, the phosphorylation status of aPKC was determined. To activate aPKC, phosphorylation of the activation loop consensus threonine residue Thr410 by PI3K-dependent PDK1 is substantial (41). In fact, treatment of AM with SP-A significantly increased aPKCThr410/403 phosphorylation after 1–10 min (Fig. 4,A) of incubation and then declined to baseline (Fig. 4 B).
We then investigated whether the observed SP-A-mediated increase in aPKCThr410/403 phosphorylation correlates with an increased kinase activity. SP-A, but not C1q, a structural homologue of SP-A, stimulates aPKC activity in a time-dependent manner, reaching a maximum at 1 h (p < 0.05) (Fig. 4, C and D). SP-A-induced aPKC activity (Fig. 4,E, lane 2) was comparable to that induced by LPS (Fig. 4,E, lane 3), used as a positive control (42). Incubation of AM with aPKCps peptides (Fig. 4 E, lane 4) abolished the aPKC activity in the presence of SP-A. Together, the results support the idea that SP-A mediates IκB-α stabilization in primary AM via activation of aPKC.
Even though phosphorylation at Thr410 facilitates kinase activity, it is not sufficient to provide full activity. A rapid autophosphorylation is necessary for a concomitant translocation of PKCζ (43, 44). Therefore, we examined the effect of SP-A on aPKC translocation by two approaches, i.e., membrane fractionation and confocal microscopy.
SP-A favors the accumulation of PKCζ in the plasma membrane
Cell fractionation demonstrated that aPKC membrane translocation is stimulated by SP-A in a time-dependent manner, reaching a maximum at 1 h (Fig. 5,A). Using confocal microscopy, we confirmed that SP-A, compared with basal conditions, favors a translocation of PKCζ to the plasma membrane in primary AM, as detected by an isoform-specific Ab (Fig. 5,B, l and m). In contrast, LPS stimulation of AM induced a translocation of PKCζ (Fig. 5,Bn) toward the nucleus. Pretreatment of the cells with SP-A, however, largely reduced LPS-mediated nuclear translocation of PKCζ (Fig. 5 Bo).
SP-A modulates p65 localization and p65/PKCζ coimmunoprecipitation
Stimulus-induced PKCζ associates with and phosphorylates trans-activating p65 (33, 45). We next asked whether SP-A-favored PKCζ membrane localization is associated with the distribution of p65. We confirmed our initial observation in wt mouse AM (Fig. 1,E) and in rat AM (Fig. 5,D), and indeed revealed that p65 distribution strongly parallels that of PKCζ (Fig. 5,B) under SP-A or LPS conditions (Fig. 5, B, Dc, Dd, and E). Furthermore, as in wt mouse AM, pretreatment of the cells with SP-A almost abolished LPS-induced p65 nuclear translocation and induced an accumulation of p65 at the plasma membrane (Fig. 5, De and E). Because SP-A, alone or in the presence of LPS, favored both PKCζ and p65 membrane localization, the effect of SP-A on PKCζ/p65 interaction was examined next. SP-A, significantly and in a concentration-dependent manner, increased PKCζ/p65 coimmunoprecipitation under constitutive conditions (Fig. 5, F and G). Because the SP-A-enhanced interaction obviously has no NF-κB trans-activating potency, we hypothesized that this effect of SP-A directly or indirectly prevents p65 nuclear translocation.
PKCζ is essential for SP-A-mediated IκB-α stabilization and inhibition of LPS-induced NF-κB activity
To establish the selective role of PKCζ in mediating the SP-A effect described, we used mice deficient for the PKCζ isoform (33). Whereas PKCλ/ι−/− mice are embryonic lethal (46), PKCζ−/− mice are grossly normal, but exhibit an impairment of IKK activation in whole lung extracts after LPS challenge (33). In freshly isolated AM from PKCζ−/− mice, however, NF-κB activation was apparent both basal and LPS induced. SP-A failed to inhibit basal and only slightly reduced LPS-induced NF-κB DNA binding compared with wt AM (Fig. 6, A and B, lower panel). Compared with wt AM, SP-A-mediated IκB-α stabilization was almost abolished in PKCζ−/−AM (Fig. 6, A and B, upper panel). Surprisingly, in the presence of LPS, IκB-α was enhanced in PKCζ−/− AM compared with wt AM, and this effect was abolished when cells had been pretreated with SP-A (Fig. 6 B). Taken together, these data indicate that PKCζ, constitutively and in the presence of LPS is required for SP-A-mediated anti-inflammatory signaling in AM.
Emerging evidence demonstrates a pivotal role of the pulmonary collectin SP-A in anti-inflammatory immunomodulation of innate immune responses of the lung to LPS (1, 2). Airway inflammation associated with local or systemic LPS release from Gram-negative bacteria is still a major cause of life-threatening pulmonary diseases (8). Efficient negative signaling cascades to prevent autotoxic mediator release by AM, the major effector cell of the pulmonary innate immune system, in response to LPS, and, in particular, the modulation of NF-κB activation threshold by SP-A have been studied intensively in recent years. Inhibition of LPS-induced NF-κB activity by SP-A has been suggested to occur via direct interaction of SP-A with components of the LPS receptor complex, including LPS-binding protein, CD14, TLR-4, and MD-2, but also independently of LPS-specific signal transduction pathways (19, 20, 21, 23, 24, 47, 48). Intracellularly, the pivotal role of IκB-α in NF-κB regulation has recently been highlighted by Tergaonkar et al. (13), demonstrating that IκB-α (as well as β and ε) is essential for both preventing NF-κB DNA binding and NF-κB-dependent gene expression in the basal state of cell activation and for making p65/IκB complexes responsive to diverse stimuli. We previously suggested that SP-A inhibits LPS-induced NF-κB activation by posttranscriptionally slowing the basal turnover of IκB-α in primary AM (24). However, the in vivo proof and the intracellular signaling pathways involved have not yet been defined.
The present study confirms that IκB-α is critically involved in SP-A anti-inflammatory effects. In AKBI mice, the integrated IκB-β gene is under the control of the IκB-α promoter, and at the same time a null mutation in IκB-α is introduced (32). AKBI mice have a normal phenotype and reveal a functional redundancy of IκB-α and IκB-β upon LPS challenge in vivo (37). In line with this, we found that LPS-induced TNF-α release by AKBI cells was comparable to that by wt cells, confirming the described in vivo redundancy of IκB-α and IκB-β in response to LPS (37). However, whereas SP-A significantly inhibited LPS-induced TNF-α release (Fig. 1, B and D) and p65 nuclear translocation in wt cells (Fig. 1,Ed), it failed to inhibit TNF-α production (Fig. 1, A and C) as well as p65 nuclear accumulation in AKBI cells (Fig. 1 Eh). These data strongly suggest that IκB-β could not compensate the lack of IκB-α in SP-A-mediated inhibition of LPS-induced cell activation.
The present study further identifies aPKCζ as an essential upstream mediator of IκB-α/NF-κB regulation by SP-A in primary cells. In PKCζ−/− AM, SP-A-mediated IκB-α stabilization was almost completely abrogated (Fig. 6,B), strongly supporting our experiments using the inhibitory aPKCps peptides in rat AM (Fig. 3, A and B). The combined data confirm that the PKCζ isoform is critically involved in SP-A-mediated IκB-α stabilization and suggest that the inhibition or the lack of PKCζ potentiates the rate of IκB-α turnover in resting AM.
In contrast to the well-characterized IκB-α turnover under stimulus-induced conditions, little is known about the mechanisms that regulate the basal turnover of IκB-α. In mouse embryonic fibroblasts, an efficient basal turnover of IκB-α requires the C-terminal CKII phosphorylation (Ser293), whereas IKK phosphorylation (Ser32 and Ser36) plays no role under basal conditions (12). Even though IκB-α is not a direct substrate of PKCζ (49), PKCζ-associated CKII preferentially phosphorylates Ser293 compared with nonassociated CKII in transfected NIH3T3 cells, thereby potentially accelerating the basal turnover of IκB-α (27). In the present study, however, the inhibition or the lack of PKCζ accelerated IκB-α turnover under basal conditions in AM. In addition, apigenin, a selective CKII inhibitor, abolished the SP-A-mediated stabilization of IκB-α, suggesting that in AM both PKCζ and CKII activity contribute to this process.
One central route controlling IκB-α turnover is the proteasome-mediated degradation of ubiquitinated IκB-α. PKCζ has been shown to influence proteasome-mediated protein degradation (50, 51). However, when we investigated the effect of SP-A on IκB-α ubiquitination in AM, SP-A enhanced the presence of ubiquitin-conjugated IκB-α (and other proteins) alone and in combination with a proteasome inhibitor and/or aPKCps, suggesting that SP-A-activated PKCζ does not stabilize IκB-α by inhibiting its degradation via the ubiquitin-dependent pathway (data not shown).
The finding that this PKC isoenzyme is involved in SP-A-mediated anti-inflammation was initially surprising because LPS- or cytokine-induced PKCζ activity was previously shown to induce NF-κB activation by distinct mechanisms, as follows: first, through its activation of IKK (52, 53), and second, by phosphorylating p65 (45, 54). Besides PKCζ, kinases implicated in p65 phosphorylation include CKII, protein kinase A, IKK2, Akt, p38, p42, and p44 (17). Among these kinases, only PKCζ has been shown to directly interact with p65 (33) and phosphorylates p65 at Ser311 in TNF-α-stimulated mouse fibroblasts (45). In the present study, SP-A enhanced the interaction of PKCζ with p65 in primary AM under constitutive conditions (Fig. 5 D). Because SP-A alone has no NF-κB trans-activating potency, we thus hypothesized that SP-A, directly or indirectly, prevents p65 nuclear translocation. Confocal and cell fractionation analysis of p65 localization in rat AM revealed a substantial inhibition of LPS-induced p65 nuclear translocation by SP-A. Based on the combined results, we propose a model in which SP-A-mediated PKCζ activation stabilizes IκB-α by preventing p65 nuclear translocation in primary AM.
A previous study (33) showed that NF-κB DNA-binding activity is reduced in nuclear lung extracts of PKCζ−/− mice treated with LPS or IL-1 i.p. In whole lung extracts, the lack of PKCζ resulted in an impairment of IKK activation upon LPS stimulation, suggesting a substantial role of PKCζ in the control of stimulus-induced IKK/NF-κB signaling cascade in the pulmonary compartment (33). In the present study, IκB-α expression in response to LPS was decreased in wt AM (Fig. 6,A), as expected. Surprisingly, LPS enhanced IκB-α in PKCζ−/− AM (Fig. 6,B). Possible explanations might include that the lack of PKCζ impairs the LPS-induced IKK activation and the subsequent IκB-α degradation, as previously suggested (33). Alternatively, the IκB-α increase observed in PKCζ−/− AM could result from de novo transcription and might be referred to a compensatory increase of PKCλ/ι. However, Western blot analysis of lung extracts from wt and PKCζ-deficient mice for PKCζ and PKCλ/ι protein expression with specific Abs revealed that the level of PKCλ/ι is not affected by the loss of PKCζ (55), rendering the latter explanation unlikely. The LPS-mediated IκB-α increase in PKCζ−/− AM was abolished by pretreatment of the cells with SP-A, proposing that SP-A not only fails to stabilize IκB-α, but provides a signal that triggers IκB-α degradation, presumably by activating IKK in a PKCζ-independent way. Of note, in PKCζ−/− AM, SP-A failed to inhibit basal and LPS-induced NF-κB activation compared with wt cells, suggesting that the down-modulation of constitutive and LPS-induced NF-κB activity by SP-A is dependent on PKCζ (Fig. 6 B).
The mechanisms underlying PKCζ activation are not completely understood and display substantial differences for different cell types, stimuli, and incubation conditions (30). PKCζ can be activated in vitro by phosphatidylinositol-3,4,5-triphosphat (42, 56), phosphatidic acid (57), ceramide, and arachidonic acid (58, 59), and/or by direct interaction with binding proteins (60). To activate PKCζ, phosphorylation of the activation loop consensus threonine residue Thr410 by PI3K-induced phosphoinositide-dependent kinase 1 is substantial (41). Besides different activation mechanisms, distinct intracellular localizations may have a major impact on cell-specific PKCζ function (29). We show that SP-A enhanced aPKCThr410/403 phosphorylation (Fig. 4, A and B), aPKC kinase activity (Fig. 4, C–E), and translocation (Fig. 5, A and B) in primary rat AM. In line with previous work (60), we found that LPS treatment of AM favored a nuclear accumulation of PKCζ, in contrast to the cell membrane accumulation of the kinase in the presence of SP-A. However, pretreatment of the cells with SP-A largely prevented LPS-induced subcellular localization of the kinase (Fig. 6 B), the biological role of which remains to be established. Interestingly, recent data suggest that LPS-induced and PI3K-dependent PKCζ activation can be modified by anti-inflammatory mediators. As shown in RAW 264.7 cells, LPS-induced PKCζ activation is potentiated by the lipid mediator 15-deoxy-Δ (12, 14)-PGJ2, resulting in an inhibition of IKK and NF-κB activity (61).
Fig. 7 depicts a model implicating SP-A-induced PKCζ as a p65-interacting kinase that reduces macrophage-inflammatory responsiveness by promoting IκB-α stabilization under resting conditions, subsequently leading to inhibition of LPS-induced NF-κB activation. However, the mechanisms of PKCζ activation as well as the PKCζ effectors operative in IκB-α-dependent anti-inflammatory modulation by SP-A in the resting AM were not identified in the present study, and will be the focus of future efforts.
In summary, the evidence presented in this work indicates that anti-inflammatory macrophage activation via IκB-α by SP-A critically involves PKCζ activity. In addition, the data imply a possible role of PKCζ as a positive or negative regulator in IκB-α/NF-κB signaling pathways in a stimulus-specific manner in the lung. Because uncontrolled NF-κB activation may well be deleterious, tonic activation of PKCζ by SP-A may serve as a physiological brake on induction of TNF-α in AM localized to sites of pulmonary inflammation. Further studies on the role of SP-A immunoregulatory mechanisms unique to the lung will potentially lead to specific therapeutic strategies in inflammatory and infectious human lung diseases.
We thank Professor John Engelhardt (University of Iowa, Iowa City, IA) for generously providing the AKBI mice, Professor Karl Dalhoff (University of Lübeck, Lübeck, Germany) for providing lung lavage fluid from alveolar proteinosis patients, and Dr. Prim B. Singh (Research Center Borstel, Borstel, Germany) for critical reading of the manuscript.
The authors have no financial conflict of interest.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
This work was supported by the Deutsche Forschungsgemeinschaft 609/1-3 and 609/1-4 (to C.S.) and 621/2-2 (to A.B.S.).
Abbreviations used in this paper: SP, surfactant protein; AKBI, IκB-α knockout/IκB-β knockin; AM, alveolar macrophage; aPKC, atypical protein kinase C; cPKC, classical protein kinase C; DAG, diacylglycerol; HI, heat inactivated; IKK, IκB kinase; PKC, protein kinase C; ps, pseudosubstrate; wt, wild type.