ADAP Y571 Phosphorylation Is Required to Prime STAT3 for Activation in TLR4-Stimulated Macrophages

The macrophage, a versatile immune cell present in most tissues, plays a key role in both innate and adaptive immunity (1–3). In response to microenvironmental cues, macrophages undergo dynamic switches in phenotype and function. Although an imperfect schematization, macrophage activation phenotypes are often broadly classified into two distinct subsets: classically activated (M1) and alternatively activated (M2) categories (4, 5). Functionally, M1-like–polarized macrophages display a high level of phagocytic activity, produce proinflammatory cytokines, and are associated with efficient pathogen clearance and inflammation; M2-like–polarized macrophages manifest anti-inflammatory feature and are linked to tissue remodeling and resolution of inflammation (3, 6). M1/M2 polarization mirrors the Th1-Th2 polarization of naive CD4⁺ T cells (2). Balance of macrophage M1/M2 polarization governs the resolution of inflammation and clearance of pathogens, and an imbalance of macrophage M1/M2 polarization is often associated with aberrant inflammatory conditions or immune disorders (7).

Signaling via TLR4 plays a central role in the control of the balance of M1/M2 macrophage polarization. TLR4 signaling, induced by LPS skews macrophage function toward the M1 phenotype via activation of a panel of transcription factors including NF-κB, AP-1, and the STAT1 (8). Furthermore, IFN-γ stimulates the JAK/STAT pathway directly, leading to STAT1 phosphorylation, nuclear translocation, and binding to the promoter of the hallmark M1-associated genes such as IL-12 and inducible NO-synthase (8). In contrast, macrophage function is skewed toward the M2 phenotype through activation of STAT3 and STAT6 (9). STAT3 is a major mediator of TLR4-dependent M2-skewed macrophage polarization (10, 11). TLR4 ligation induces the production of IL-6 and IL-10 in an autocrine/paracrine manner, which then bind to their cognate receptors primarily to evoke the phosphorylation of STAT3 via JAKs. Activated STAT3 then represses the expression of the M1-like proinflammatory cytokines such as TNF-α, IL-1β, IL-12, and IFN-γ (12–15); M1 macrophage surface markers CD80/CD86; and MHC class II (MHC-II) (16, 17). Activation of STAT3 by TLR4-triggered autocrine IL-6/IL-10 signaling limits M1 and favors M2 macrophage polarization, thus exerting an anti-inflammatory effect (18–20). STAT3-deficient mice display M1 phenotype with increased inflammation and are prone to the development of SCID (21–25), indicating that STAT3 acts as a checkpoint molecule in the control of the level of LPS-induced M1 polarization and inflammatory responses in macrophages. Although TLR4 engagement potentiates the M1 polarization, it also activates STAT3 to skew macrophage function toward the M2 phenotype via autocrine IL-6 and IL-10 pathways. However, to date, the molecular mechanism in directing LPS-TLR4 signaling toward the proinflammatory M1 polarization, rather than the STAT3-mediated anti-inflammatory M2 polarization, is largely unknown.

Adhesion and degranulation–promoting adaptor protein (ADAP), also known as FYB or SLAP-120/130, is encoded by the Fyb gene (26). ADAP is originally identified as a T cell immune adaptor involved in regulation of both integrin-mediated T cell adhesion and T cell signaling (27–29). ADAP–SLP-76 (SH2 domain–containing leukocyte protein of 76 kDa) module is critical for TCR-engaged signal transduction, formation of persistent microclusters and stabilization of T cell contacts via modulating integrin-dependent and -independent adhesion (28–30). ADAP-SKAP1 (Src kinase–associated phosphoprotein 1) module strengthens integrin lymphocyte function–associated Ag 1 (LFA-1) activation and LFA-1–mediated adhesion via inside-out and outside-in pathways (31–33). ADAP also regulates TCR-induced NF-κB activation via interaction with caspase recruitment domain-containing membrane-associated guanylate kinase protein 1 (CARMA1) (34). Recently, we also showed that ADAP-Ubc9 module facilitates inside-out signaling via converging of Rap1-RapL module and upregulating Rac1 activation (35).

The immune adaptor ADAP is widely expressed in various hematopoietic cells including T cells, mast cells, platelets, NK cells, dendritic cells, and macrophages (35–41). Apart from fine-tuning T cell signaling and adhesion in T cells, ADAP participates in inside-out and outside-in integrin signaling and actin remodeling via a model of the ADAP/SKAP2/Sirpα complex in macrophages (39), adhesion and FcεRI-induced degranulation in mast cells (40), and CD11c-triggered cellular responses in dendritic cells (37). In addition, ADAP is involved in the megakaryocyte polarization and αIIbβ3 integrin–mediated signaling in platelets (41–43).

Despite these qualities, the functional role of ADAP in macrophage polarization and inflammatory responses remains elusive. In this study, we show that ADAP-deficient macrophages display an enhanced M1 polarization in response to TLR4 stimulation by LPS. Mechanically, overexpression of ADAP enhances, whereas loss of ADAP reduces, LPS-mediated phosphorylation and activity of STAT3. This relies on the LPS-induced Y571 phosphorylation of ADAP. Together, our results indicate a novel role for ADAP as a molecular switch that is required to prime STAT3 phosphorylation in skewing the M1/M2 balance toward M2 phenotype.

The Fyb^−/− (or Adap^−/−) mice, a kind of gift from C. E. Rudd (University of Cambridge, Cambridge, U.K.), was generated as described previously (32), and genotyped by allele-specific PCR using a 5′-primer (5′-CCGTGGGGCCAAAGTCAGGAGAA-3′), a 3′-wt primer (5′-CCCACCCCAAGGTCCTTTCTTAC-3′), and a neo-3′-ko primer (5′-GCGCTACCGGTGGATGTGGAATGT-3′). Wild-type (WT) C57BL/6 mice were purchased from Joinn Laboratories (Suzhou, China). All mice were bred in a specific pathogen-free condition and maintained under constant temperature and humidity in the Animal Centre of Suzhou Institute of Systems Medicine. Age-matched (6- to 8-wk-old) and weight-matched female mice were used in this study. All animal experiments were carried out with approval of Jiangsu Association for Laboratory Animal Science in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the Ministry of Science and Technology of the People’s Republic of China.

Bone marrow–derived macrophages (BMMs) were generated as previously described (44). Briefly, bone marrow cells were flushed from the femurs of mice and cultured at a density of 2 × 10⁶ cells per milliliter in RPMI 1640 medium containing 10% FBS (Serana Europe, Pessin, Germany), 100 U/ml penicillin-streptomycin (GE Healthcare Hyclone Laboratories), and 50 μM 2-ME and in the presence of 100 ng/ml rM-CSF (Sino Biological, Beijing, China) at 37°C and 5% CO₂. On day 3, fresh RPMI 1640 complete medium supplemented with 100 ng/ml M-CSF was added, cells were cultured for another 4 d to fully differentiate into macrophages. Adherent cells were then used as BMMs, and cell purity was verified by FACS analysis for F4/80 and CD11b (purity >96%). Immortalized BMMs (iBMMs) were established as previously described with some modifications (45). Briefly, bone marrow cells were isolated and cultured with concentrated J2-infected media containing 5% FBS, 5 μg/ml polybrene (Sigma-Aldrich), and 100 ng/ml M-CSF for 18 h. Adherent cells were then separated from nonadherent cells and incubated in RPMI 1640 complete medium supplemented with J2-infected media, polybrene, and M-CSF for additional 6 d. Cells were then maintained in RPMI 1640 complete medium supplemented with 10 mM HEPES (pH 7.8) and 20% L929 cell culture supernatant and propagated for more than 20 passages until immortalization. Peritoneal macrophages were isolated from the peritoneal cavity of mice as described previously (46). Briefly, mice were i.p. injected with 2 ml of 3% thioglycolate (MilliporeSigma). Three days after injection, thioglycolate-elicited peritoneal macrophages were harvested via peritoneal lavage by PBS. Cells were then cultured in RPMI 1640 complete medium for at least 2 h, and the adherent cells were ready for subsequent experimental use.

RAW264.7 and THP-1 cell lines were obtained from American Type Culture Collection (Manassas, VA). RAW264.7 cells were maintained in DMEM supplemented with 5% FBS and 100 U/ml penicillin-streptomycin. THP-1 cell line was cultured in RPMI 1640 complete medium. Prior to each experiment, THP-1 cells were seeded into six-well or 12-well plates at a cell density of 1 × 10⁶ cells per milliliter and differentiated into macrophages in the presence of 100 ng/ml PMA (Sigma-Aldrich) for 24 h. Stable ADAP knockdown, THP-1, and the control enhanced GFP (EGFP) knockdown THP-1 cells were established by lentivirus infection as described previously (35). Briefly, small hairpin RNA (shRNA) of ADAP (5′-GCAAAGGCCAGACGTCTTA-3′) were cloned into a pLKO.1 vector (47). Lentivirus with package of the plasmids were freshly prepared in 293FT cells, filtered through a 0.45-μm filter, and pelleted through ultracentrifugation at 25,000 rpm for 120 min. The lentiviral plasmids were resuspended in fresh medium and employed to infect THP-1 cells with spinoculation. Both THP-1 shRNA against ADAP (shADAP) and shRNA against EGFP (shEGFP) cell lines were selected with 10 μg/ml puromycin (Sigma-Aldrich) in parallel for at least 4 wk.

Anti-ADAP rabbit polyclonal Ab was obtained from Merck Millipore (Burlington, MA). Rabbit mAbs against phospho-STAT3 (p-Tyr⁷⁰⁵), STAT3, and histone H3; mouse mAb against phosphotyrosine (p-Tyr¹⁰⁰) and streptavidin-HRP; and anti-rabbit Alexa Fluor 555–conjugated Ab were all purchased from Cell Signaling Technology (Danvers, MA). Anti–α-tubulin rabbit mAb was from Abcam (Cambridge, U.K.). Anti-hemagglutinin (HA) mouse mAb was from Sigma-Aldrich (St. Louis, MO). Anti-rabbit and anti-mouse goat IRDye Abs were from LI-COR Biosciences (Lincoln, NE). Anti–IL-6 and anti–IL-10–neutralizing Abs were purchased from BD Biosciences (San Jose, CA). Fluorescence-conjugated FACS Abs against TLR4, F4/80, CD11b, CD86, CD80, MHC-II, CD3, CD4, and corresponding IgG isotypes were all from BD Biosciences except for anti-CD206 and anti–T-bet, which were from BioLegend (San Diego, CA).

For macrophage differentiation and stimulation, LPS derived from Escherichia coli strain O111:B4 was purchased from Sigma-Aldrich, and mouse rM-CSF, IL-6, and IL-10 were from Sino Biological (Beijing, China). The JAK/STAT3 chemical inhibitor ruxolitinib, as well as kinase inhibitors PP2, ibrutinib, and entospletinib, was purchased from Selleck Chemicals (Houston, TX).

Full-length human ADAP was cloned into pSRα vectors with HA tagged at N terminus. Based on the templates, point mutant constructs in vitro were generated using a Hieff Mut Site-Directed Mutagenesis Kit (Yeasen Biotechnology, Shanghai, China), following a standard PCR-based method. Primers used for ADAP mutagenesis were as follows: ADAP Y462F forward, 5′-GACAGTGAAGGAGAAACATTTGAAGACATAGAAGCATCC-3′ and reverse, 5′-GGATGCTTCTATGTCTTCAAATGTTTCTCCTTCACTGTC-3′; ADAP Y559F forward, 5′-CAGAACAGCAAGGGGTTCATTTGGCTATATTAAAACAACTGC-3′ and reverse, 5′-GCAGTTGTTTTAATATAGCCAAATGAACCCCTTGCTGTTCTG-3′; and ADAP Y571F forward, 5′-CTGCTGTAGAGATTGACTTTGATTCTTTG AAACTG-3′ and reverse, 5′-CAGTTTCAAAGAATCAAAGTCAATCTCTA CAGCAG-3′. Transient transfection of RAW264.7 cell line was described previously (48). Briefly, 2 × 10⁷ RAW264.7 cells suspended in Opti-MEM medium (Life Technologies) were mixed with 20 μg of DNA and electroporated by exponential pulse at 250 V and 950 μF with Gene Pulser Xcell (Bio-Rad Laboratories). After transfection, cells were recovered in room temperature for 10 min, washed with PBS, and cultured in antibiotic-free DMEM containing 10% FBS for 48 h before harvest.

Immunoblotting and immunoprecipitation were performed as described (49). For whole-cell lysates, cells were harvested and washed twice with cold PBS, and then lysed in the lysis buffer containing 1% Triton X-100, 20 mM Tris (pH 8.0), and 150 mM NaCl and supplemented with 1 mM sodium orthovanadate and 1% protease inhibitors (Roche Diagnostics, Rotkreuz, Switzerland). Lysates were incubated on ice for 30 min with vortex every 10 min, and insoluble materials were removed by centrifugation at 15,000 × g for 15 min at 4°C. For immunoprecipitation, cells (5 × 10⁶–5 × 10⁷) were lysed in 500 μl of lysis buffer as described. Cell lysates were incubated with 1 μg of the indicated Abs and immunoprecipitated overnight at 4°C with 30 μl of 50% (v/v) protein G Sepharose beads (GE Healthcare). Immunoprecipitates were washed three times with cold lysis buffer and dissolved by boiling in SDS sample buffer for 10 min. Proteins were separated by 8% SDS-PAGE and transferred onto a nitrocellulose membrane (Pall Life Sciences). The membranes were blocked with 5% skimmed milk and immunoblotted with primary Abs (1:2000) in 0.1% Tween-20 in PBS overnight at 4°C. After washing three times with 0.1% Tween-20 in PBS and incubating with IRDye secondary Abs (1:10,000) for 1 h at room temperature, proteins were visualized with Odyssey Infrared Imager (LI-COR Biosciences).

Nuclear and cytosolic lysate fractions were extracted from untreated or LPS-stimulated iBMMs, using a Nuclear and Cytoplasmic Protein Extraction Kit (Yeasen Biotechnology), following the manufacturer’s instructions. Purity of nuclear and cytoplasmic fractions was examined by analyzing cytoplasmic marker tubulin and nuclear marker histone H3, respectively, by Western blotting for each independent experiment.

One million iBMMs were seeded on coverslip slides in a six-well plate 24 h prior to LPS stimulation. Cells were fixed with 4% paraformaldehyde in PBS at room temperature for 15 min and precold methanol at −20°C for additional 10 min. Upon fixation, cells were blocked and permeabilized with 5% BSA containing 0.5% Triton X-100 for 1 h at room temperature, and subsequently stained with primary Ab for STAT3 at a dilution of 1:1000 in 5% BSA supplemented with 0.3% Triton X-100 overnight at 4°C. After incubation, slides were rinsed with PBS and incubated with Alexa Fluro 555–conjugated donkey anti-rabbit Ab (1:5000) at room temperature for 1 h. Nuclei were counterstained with DAPI (Beyotime Biotechnology, Shanghai, China) for 10 min. The cells were mounted with antifade mounting medium (Beyotime Biotechnology) and imaged using a confocal Zeiss LSM 880 laser scanning microscope (Carl Zeiss, Oberkochen, Germany). Twenty independent fields with a minimum of 100 cells were recorded for each condition. Nuclear or total STAT3 fluorescence intensity of each sample was quantified by ImageJ Software.

For identification of LPS-induced phosphorylation sites in ADAP, peritoneal macrophages were harvested and seeded in a six-well plate, followed by 6 h stimulation of LPS, and then lysed. Total ADAP was pulled down by immunoprecipitation and separated by SDS-PAGE as described above. To extract the peptides, the Coomassie-stained protein gel bands corresponding to ADAP were excised, destained, reduced, and alkylated, followed by digestion with 150 ng of trypsin as described (50). Tryptic extracts were collected, lyophilized, and resuspended in 0.1% formic acid for liquid chromatography–tandem mass spectrometry (LC-MS/MS) analysis (51). All samples were analyzed on an Easy-nLC 1000, coupled to an LTQ Orbitrap Elite Mass Spectrometer (Thermo Fisher Scientific) equipped with a nanoelectrospray source. Peptides were separated on a 15-cm analytical rapid-separation liquid chromatography column (Acclaim PepMap, 100 Å, C18 2-μm pore size, 150-mm length, 50-μm inner diameter) in a 180-min gradient of 0–95% buffer B (99.9% acetonitrile and 0.1% formic acid) at a flow rate of 300 nl/min. The LTQ Orbitrap Mass Spectrometer was operated in a data-dependent acquisition mode. Spray voltage was set to 2.1 kV, S-lens radio frequency level was set at 50%, and heated capillary temperature was set at 300°C. Scans with an m/z range of 150–2000 were collected in positive polarity mode. Full scans were analyzed with 60,000 resolution at m/z = 400, and a predicted automatic gain control target of 1 million. Top 10 most intense multiply-charged ions were measured at the resolution of 15,000, and the collision-induced dissociation fragmentation was enabled using normalized collision energy of 32. Tandem mass spectra were extracted by Mascot Distiller 2.7 (version 2.4.1; Matrix Science, London, U.K.). Data were searched against the mouse subset of the UniProt protein database with a tandem mass spectrometry mass tolerance of 0.05 Da and a precursor ion tolerance of 10 parts per million. Scaffold (version Scaffold_4.8.9; Proteome Software, Portland, OR) was used to validate peptide and protein identification. Peptide identification false discovery rate was set at 5%, whereas protein identification false discovery rate was set at 1% with identified peptides ≥2. Peptide spectra were assigned by pLabel software (version 2.4.0.5; pFind Group, Beijing, China) (52).

Total RNA was extracted from unstimulated or LPS-stimulated cells with TRIzol Reagent (Sigma-Aldrich), following the manufacturer’s instructions. Two hundred nanograms of total RNA was reverse transcribed by use of a First Strand cDNA Synthesis Kit (Thermo Fisher Scientific) according to the manufacturer’s instructions. Quantitative real-time PCRs were performed using Hieff Quantitative PCR (qPCR) SYBR Green Master Mix (Yeasen Biotechnology), and analyzed in triplicate by QuantStudio Design and Analysis System (Applied Biosystems, Foster City, CA). Relative expression levels were calculated by a standard ΔΔC_T method and normalized to housekeeping gene GAPDH values. Primers were synthesized by Sangon Biotech (Shanghai, China), and primer sequences are available on request.

Nuclear proteins were extracted from untreated or LPS-stimulated iBMMs as described above. A total of 10 μg of nuclear extracts were premixed with 24 μl of reaction buffer (0.5 M HEPES [pH 7.0], 0.5 M Tris [pH 8.0], 0.25 M EDTA, 50 mM DTT, and 25% glycerol) on ice for 10 min and incubated with 3 pmol of biotin-labeled probe-containing consensus STAT3 DNA–binding sequence (5′-GATCCTTCTGGGAA TTCCTAGATC-3′) at room temperature for 20 min. A probe in lysis buffer lacking nuclear extract was used as a negative control. Samples were separated on an 8% nondenaturing gel in TBE buffer and transferred onto a 0.45-μm PVDF membrane. The biotin-labeled STAT3 protein/DNA complex was then visualized with streptavidin-HRP and detected by ChemiDoc MP Imaging System (Bio-Rad Laboratories).

One day prior to transfection, THP-1 cells were cultured in a six-well plate at 4 × 10⁶ cells per well in the presence of 100 ng/ml PMA. Adherent PMA-primed THP-1 cells were then maintained in antibiotic-free RPMI 1640 medium with 10% FBS and cotransfected with 2.5 μg of STAT3-dependent pGM-luc reporter plasmid (Yeasen Biotechnology) and vector control by using Lipofectamine LTX and PLUS reagent (Invitrogen) according to the manufacturer’s instructions. Eighteen hours after transfection, cells were treated with 1 μg/ml LPS for 6 h, harvested, and lysed in 220 μl of lysis buffer provided in a Luciferase Reporter Gene Assay Kit (Yeasen Biotechnology). Lysates were then analyzed for luciferase activity with a Varioskan Lux Multimode Microplate Reader (Thermo Fisher Scientific). The relative STAT3-luciferase value was normalized to the level of vector control in each sample.

WT-recipient mice were i.p. injected with 200 μl of clodronate liposome (Vrije Universiteit, Amsterdam, the Netherlands) for 4 d for macrophage depletion. For BMMs adoptive transfer, WT or Adap^−/− BMMs (5 × 10⁶) were untreated or pretreated with 1 μg/ml LPS for 24 h in vitro and injected i.v. into recipient mice. Saline was used as a negative control for clodronate or adoptive cells. After 48 h of adoptive transfer, lymphocytes were isolated by grinding freshly harvested lymph nodes of WT recipients and subjected to flow cytometric analysis. Each experiment was performed with n = 4 of WT-recipient and n = 4 of WT- or Adap^−/−–donor mice for adoptive macrophage transfer. Data were obtained from at least three independent experiments.

For macrophage surface expression analysis, untreated or LPS-stimulated macrophages were harvested, washed with FACS buffer (PBS containing 2% BSA), and stained for 1 h on ice with fluorescence-conjugated Abs at a concentration of 1:150 in FACS buffer. Abs used include PE/allophycocyanin anti-F4/80, FITC anti-CD11b, allophycocyanin anti-TLR4, allophycocyanin anti–MHC-II, allophycocyanin anti-CD80, PE-Cy7 anti-CD86, allophycocyanin anti-CD206, and their isotype-matched negative control Abs. Cells were then washed twice and resuspended in FACS buffer for flow cytometric analysis. Macrophages defined as viable F4/80⁺ or F4/80⁺CD11b⁺ cells were subjected to subsequent analysis. For T cell analysis, lymphocytes were first stained with FITC anti-CD3 and PE anti-CD4 Abs for 1 h on ice for surface label. Allophycocyanin-conjugated anti–T-bet intracellular staining was then performed after fixation and permeabilization according to the manufacturer’s protocols (BD Biosciences). Data were all acquired on an FACSCalibur Flow Cytometer (BD Biosciences) and analyzed with FlowJo 10 (FlowJo Software).

A flow cytometry–based cytometric bead array assay was employed to measure cytokine levels in cell culture. BMM (5 × 10⁵ cells per milliliter) or RAW264.7 cell (1 × 10⁶ cells per milliliter) supernatants were collected, aliquoted, and analyzed by Cytometric Bead Array Mouse Inflammation Kit (BD Biosciences) for IL-6, IL-12p70, TNF-α, and IFN-γ, following the manufacturer’s protocols. Data were acquired on an FACSCalibur Flow Cytometer (BD Biosciences) and analyzed with FCAP Array software (v3.0 version; BD Biosciences).

All data were represented as mean ± SEM of at least three independent experiments and analyzed with GraphPad Prism 8 (GraphPad Software, San Diego, CA). Significance of difference between two groups was evaluated with unpaired and paired two-tailed Student t tests. A p value <0.05 was considered statistically significant.

LPS, a potent TLR4 agonist, stimulates macrophage polarization toward an M1 profile via TLR4 signaling (53). To determine the role of ADAP in TLR4-mediated macrophage polarization, phenotypic expression of CD86 and CD206, the unique surface markers of M1 and M2 macrophages, respectively, were determined by FACS analysis in WT and Adap^−/− BMMs with or without LPS stimulation. Without LPS treatment, both WT and Adap^−/− BMMs exhibited a similar pattern in the proportions of M1/M2 population. The frequency of CD86⁺CD206⁻ M1-like macrophages was induced in WT BMMs exposed to LPS and the effect was significantly increased in ADAP-deficient BMMs (26.9 versus 16.4%) (Fig. 1A). On the contrary, the frequency of CD86⁻CD206⁺ M2-like phenotype was lower in Adap^−/− BMMs than that in WT BMMs in the presence of LPS stimulation (14.9 versus 21.4%). Keeping on this, the frequency of the LPS-induced MHC-II^hi M1-like peritoneal macrophages deficient in ADAP was also significantly higher than that of WT peritoneal macrophages (Fig. 1B). Thus, these data clearly indicate a regulatory role for ADAP in the LPS-induced macrophage M1-like polarization. Furthermore, we assessed typical M1 proinflammatory cytokine expression pattern at both mRNA and protein levels in WT and Adap^−/− BMMs. The levels of a panel of M1 proinflammatory cytokines, including IL-6, IL-12, TNF-α, and IFN-γ, were markedly higher upon LPS stimulation in Adap^−/− BMMs than in WT BMMs, as measured at mRNA level by qPCR at 6 and 24 h post-LPS stimulation (Fig. 1C, upper panels) and at protein level by cytometric bead array at 24 h post-LPS stimulation (Fig. 1C, lower panels). On the contrary, the mRNA expression level of M2 signature genes, including IL-10, PPAR-γ, and Arg-1, was significantly decreased at 24 h post-LPS stimulation in Adap^−/− BMMs compared with that in WT BMMs (Fig. 1D). As a control, FACS analysis revealed comparable expression profiles of surface receptors F4/80 and TLR4 in both WT and Adap^−/− BMMs regardless of LPS treatment (Fig. 1E).

M1 macrophages have a capacity in potentiation of Th1 immune response (11). We next assessed the effect of ADAP deficiency on the ability of macrophages in LPS-induced Th1-priming. Adoptive transfer of ex vivo LPS-primed WT or ADAP-deficient BMMs to the macrophage-depleted WT mice showed a higher frequency of Th1 lymphocytes from recipient mice with adoptive transfer of ADAP-deficient BMMs than WT macrophages (17.3 versus 8.5%) (Fig. 1F), further supporting ADAP deficiency enhances macrophage M1-like polarization, consequently potentiating higher Th1 responses in vivo.

That Adap^−/− mice have an enhanced LPS-induced macrophage M1-like polarization and Th1 response is similar to the phenotype observed in STAT3-deficient mice (21–23), which prompted us to investigate whether ADAP deficiency may target STAT3 signaling to affect the LPS-induced macrophages M1-like polarization. As shown in Fig. 1G, inhibition of STAT3 by treatment with the JAK/STAT3 inhibitor ruxolitinib increased the frequencies of the LPS-induced CD86⁺CD206⁻ M1-like polarization from WT macrophages (Fig. 1G, left panel, bar 5 versus bar 3), an effect similar to that caused by ADAP deficiency. Interestingly, ADAP-deficient macrophages failed to further upregulate the frequencies of the LPS-induced CD86⁺CD206⁻ M1-like polarization in response to ruxolitinib treatment (Fig. 1G, left panel, bar 6 versus bar 4). Consistent observation was also found in respect to the CD80⁺ M1-like polarization (Fig. 1G, right panel). Thus, inhibition of STAT3 activity augmented LPS-induced macrophage M1-like polarization to a similar extent as observed with ADAP deficiency.

Together, these data demonstrate ADAP deficiency potentiates LPS-induced macrophage M1-skewed polarization, M1-like proinflammatory cytokine production and Th1 response, which is STAT3 activity dependent.

Next, we investigated the relationship between ADAP and STAT3 during TLR4 signaling in macrophages. We first examined the effect of downregulation of ADAP on TLR4-mediated STAT3 phosphorylation in macrophages. ADAP was knocked down in the human macrophage THP-1 cells by stably expressing shADAP or shEGFP as a control. Western blot analysis was performed to measure the levels of LPS-induced STAT3 tyrosine phosphorylation on Y705, a site required for functional activity of STAT3 (54), using anti–phospho-specific Ab. In both ADAP knockdown and control cells, LPS treatment induced STAT3 tyrosine phosphorylation on Y705. However, the level of LPS-induced Y705 phosphorylation of STAT3 was remarkedly decreased with slower kinetics in the ADAP knockdown cells as compared with that in the control shEGFP cells (Fig. 2A, lanes 6–10 versus lanes 1–5). A similar decrease of LPS-induced tyrosine phosphorylation of STAT3 was also observed in the Adap^−/− iBMMs (Fig. 2B, lanes 5–8), compared with that in WT iBMMs (Fig. 2B, lanes 1–4). In support, in contrast, overexpression of ADAP remarkably enhanced LPS-induced STAT3 phosphorylation compared with the vector control (Fig. 2C, lane 4 versus lane 2). These data indicate ADAP is required for activating phosphorylation of STAT3 in macrophages in response to LPS stimulation.

We further assessed the effect of ADAP deficiency on LPS-induced STAT3 nuclear translocation, a process dependent on its tyrosine phosphorylation. Nuclear and cytosolic fractions were extracted from WT and Adap^−/− iBMMs with or without LPS stimulation, and subjected to Western blotting with anti-STAT3. As shown in Fig. 2D, less STAT3 in nuclear fraction and more cytosolic STAT3 in lysates of Adap^−/− macrophages stimulated by LPS compared with WT control cells suggest ADAP is required for LPS-induced STAT3 nuclear translocation. The ADAP dependence of LPS-induced STAT3 nuclear translocation was further confirmed by anti-STAT3 immunofluorescence staining assay. As shown in Fig. 2E, STAT3 was predominantly localized in the cytosol of both WT and Adap^−/− iBMMs under resting condition. Upon LPS stimulation for 1 h, a substantial amount of STAT3 translocated to the nucleus of WT iBMMs, whereas in contrast, STAT3 was distributed in both the nucleus and the cytoplasm of Adap^−/− iBMMs. Furthermore, gel electrophoresis mobility shift assay showed that the nuclear fraction from WT iBMMs formed a retarded DNA-protein complex with the biotin-labeled STAT3-specific probe, which, as expected, was significantly increased upon LPS stimulation (Fig. 2F, lanes 2 and 3). In contrast, the complex formation between the nuclear fraction from Adap^−/− iBMMs and the STAT3-specific probe was significantly decreased, and LPS stimulation failed to increase the complex formation (Fig. 2F, lanes 4 and 5). Moreover, we evaluated the effect of ADAP deficiency on STAT3-mediated gene expression. SOCS3 and BCL3 are STAT3 target genes downstream of TLR4 pathway (55–57). Quantitative real-time PCR analysis showed the mRNA level of both SOCS3 and BCL3 induced by LPS stimulation was much lower in ADAP-deficient iBMMs than that in WT iBMMs (Fig. 2G). Finally, luciferase reporter assay also confirmed that LPS stimulation resulted in a reduced transcriptional response on the STAT3 reporter gene in stably shADAP-expressing THP-1 cells compared with the control cells expressing shEGFP (Fig. 2H). Thus, these data indicate that ADAP is required for the LPS-induced STAT3 phosphorylation and activity.

Tyrosine phosphorylation plays a vital role in ADAP-mediated signal transduction for protein assembly and cellular responses in T cells (58). To examine whether tyrosine phosphorylation is required for ADAP functional activity in TLR4-activated macrophages, whole-cell lysates from peritoneal macrophages with or without LPS stimulation for a time course of 6 h were subjected to immunoprecipitation by anti-ADAP, followed by immunoblotting with phosphotyrosine-specific mAb anti–p-Tyr¹⁰⁰. Although the tyrosine phosphorylation of ADAP was marginally detectable in the absence of LPS stimulation (Fig. 3A, lane 2), ADAP became notably tyrosine-phosphorylated in cells at 2 h post-LPS stimulation, which was gradually increased within 4–6 h of LPS stimulation (Fig. 3A, lanes 3–5). Likewise, the tyrosine phosphorylation of ADAP was detected in iBMMs (Fig. 3B) and in human PMA-primed THP-1 cells (Fig. 3C) at 1 h and peaked at 4 h post-LPS stimulation.

To determine the major LPS-induced tyrosine phosphorylation sites in ADAP in macrophages, ADAP was immunoprecipitated from mouse peritoneal macrophages with or without LPS stimulation, trypsin digested, and subjected to LC-MS/MS analysis. Analysis of mass spectra data identified a tryptic peptide containing a Y⁵⁷¹DSL motif of ADAP was phosphorylated in macrophages stimulated with LPS, given a +80 Da shifts found on fragment y6 ions (Y571), as well as corresponding shifts on peaks of y7–y10 ions (Fig. 3D, lower panel). Compared with unstimulated ADAP, 2-fold relative quantities of pYDSL peptide spectra were mapped in LPS-stimulated ADAP, in which unphosphorylated YDSL peptides were hardly identified. Comparative sequence alignment revealed the phosphorylated Y⁵⁷¹DSL motif and its surrounding sequence were highly conserved across various species (Fig. 3E).

To further verify the LPS-induced tyrosine phosphorylation of ADAP at Y571 identified by mass spectrometry, we generated HA-tagged ADAP constructs containing single phenylalanine substitution at tyrosine residues 571 (Y571F) as well as residues Y462 (Y462F) and Y559 (Y559F). These tyrosine residues act as major phosphorylation sites in response to TCR stimulation in T cells (58–60). RAW264.7 cells were transfected with ADAP WT and mutant plasmids. At 2 d posttransfection, whole-cell lysates from transfected cells with or without LPS stimulation were prepared and subjected to immunoprecipitation with an anti-HA mAb, followed by Western blotting with anti–p-Tyr¹⁰⁰. As shown in Fig. 3F, whereas both WT and mutant ADAP were hardly phosphorylated at resting condition (lanes 2–5 versus lane 1), LPS stimulation significantly induced tyrosine phosphorylation of ADAP WT (lane 7 versus lane 2). Consistent with the results obtained by LC-MS/MS analysis, phenylalanine substitution at residue Y571, but not at residues Y462 and Y559, resulted in a striking decrease in the level of LPS-induced ADAP tyrosine phosphorylation (lane 10 versus lanes 7–9), suggesting that Y571 is the predominant tyrosine phosphorylation site of ADAP induced by LPS in macrophages.

Moreover, we attempted to figure out the kinase(s) responsible for the LPS-induced ADAP tyrosine phosphorylation. To this end, we assessed the effect of treatment with inhibitor of Fyn, the putative kinase of ADAP phosphorylation in T cells (26), as well as the kinase inhibitors of Syk and Bruton tyrosine kinase (Btk) on LPS-induced ADAP tyrosine phosphorylation in iBMMs. As shown in Fig. 3G, whereas treatment with Syk kinase inhibitor entospletinib had little effect, treatment with Btk inhibitor ibrutinib substantially decreased the LPS-induced ADAP tyrosine phosphorylation, indicating that LPS-induced ADAP phosphorylation at Y571 required the Btk, which acts downstream of TLR4 signaling (61).

To attempt to study the functional impact of Y571 phosphorylation on ADAP function in the regulation of LPS-induced STAT3 activation, RAW264.7 macrophages were transfected with ADAP WT, or ADAP-Y571F, and ADAP-Y559F mutants, followed by Western blot analysis of LPS-induced phosphorylation of STAT3 at Y705. Overexpression of WT-ADAP, as well as Y559F-ADAP mutant, stimulated a marked increase in the LPS-induced STAT3 phosphorylation compared with the vector control (Fig. 4A, lanes 4 and 6 versus lane 2). In contrast, overexpression of Y571F mutant failed to induce the STAT3 phosphorylation in response to LPS stimulation, compared with that of overexpression of WT-ADAP (Fig. 4A, lane 8 versus lane 4), an effect similar to that caused by deficiency of ADAP (Fig. 2A, 2B). These data suggest that phosphorylation of Y571 in ADAP is required for LPS-induced STAT3 phosphorylation.

Given Y571 site in ADAP is required for LPS-induced STAT3 phosphorylation, which limits M1 and favors M2-like macrophage polarization, we thus assessed the impact of mutation of Y571 on LPS/TLR4-mediated macrophage M1-like polarization. Overexpression of WT-ADAP or Y559F mutant significantly inhibited the M1-like polarization as indicated by the reduced expression of TLR4-activated M1 marker CD80 and CD86, whereas overexpression of the Y571F mutant had no significant effect (Fig. 4B), suggesting that phosphorylation of Y571 is required for the inhibitory effect of ADAP on M1-like polarization. In line with the surface expression pattern, overexpression of the Y571F mutant failed to downregulate M1 cytokine IL-6 and TNF-α production by contrast to overexpressed WT-ADAP and Y559F mutant (Fig. 4C). Together, these data decisively demonstrate that Y571 tyrosine phosphorylation site mediates ADAP’s functions in the enhancement of LPS-induced STAT3 phosphorylation and in the inhibition of M1-like polarization.

Given that activation of TLR4 signaling by LPS upregulates the expression of IL-6 and IL-10, which in turn activate STAT3 phosphorylation in an autocrine manner in macrophages (20), we thus assessed the effect of IL-6 and IL-10 stimulation on the regulation of STAT3 phosphorylation by ADAP in macrophages. ADAP deficiency substantially dampened the level of STAT3 tyrosine phosphorylation induced by IL-6 (Fig. 5A, lanes 5 and 6 versus lanes 3 and 4) but not by IL-10 (Fig. 5A, lanes 9 and 10 versus lanes 7 and 8). As shown in Fig. 5B, both IL-6 and IL-10 stimulation enhanced the STAT3 phosphorylation in RAW264.7 macrophages (lanes 4 and 7 versus lane 1). Interestingly, whereas overexpression of WT-ADAP further increased the level of IL-6–induced STAT3 tyrosine phosphorylation (lane 5 versus lane 4), Y571F-ADAP mutant failed to do so (lane 6 versus lane 4). In contrast, both WT-ADAP and Y571F-ADAP mutant failed to further increase the IL-10–stimulated STAT3 tyrosine phosphorylation (lanes 8 and 9 versus lane 7). Moreover, we assessed the impact of ADAP loss on LPS-mediated STAT3 nuclear translocation in a background without IL-6 or IL-10 signaling. As shown in Fig. 5C, in WT iBMMs, inhibition of either IL-6 or IL-10 signaling with neutralizing Abs robustly decreased the level of LPS-induced phosphorylation and nuclear translocation of STAT3 in WT iBMMs (lanes 3 and 4 versus lane 2). Loss of ADAP also decreased the LPS-induced nuclear translocation of STAT3 (lane 6 versus lane 2). In contrast, interestingly, when ADAP was absent, treatment with IL-10 blocking Ab but not IL-6 blocking Ab gave rise to a further decrease in LPS-induced phosphorylation and STAT3 nuclear translocation (lane 8 versus lane 6), which is in line with our observation shown in Fig. 5A, 5B that overexpression of ADAP Y571F mutant or loss of ADAP decreased IL-6– rather than IL-10–induced STAT3 phosphorylation. These observations suggest the involvement of ADAP in the regulation of STAT3 phosphorylation is mainly via IL-6 rather than IL-10 pathway in the TLR4-activated macrophages.

Next, we assessed if ADAP has a physical interaction with STAT3 in macrophages. The lysates of iBMMs with or without LPS stimulation were subjected to immunoprecipitation with anti-ADAP Ab, followed by immunoblotting with anti-STAT3. As shown in Fig. 5D, endogenous STAT3 coprecipitated with anti-ADAP Ab in resting iBMMs (lane 2 versus lane 1). Surprisingly, the level of coprecipitated STAT3 by anti-ADAP was significantly decreased in iBMMs upon LPS stimulation for 1 and 4 h (lanes 3 and 4 versus lane 2), whereas the ADAP-STAT3 complex formation was not affected upon stimulation of iBMMs with IL-6 or IL-10 (lane 5 or 6 versus lane 2). Thus, ADAP associates with STAT3 in resting cells, and LPS stimulation leads to dissociation of the complex.

These data indicate ADAP and its Y571 site is required for IL-6– but not IL-10–induced STAT3 activation during TLR4 signaling.

To this end, we demonstrate that macrophages from ADAP-deficient mice display an enhanced M1 polarization in response to LPS stimulation. Overexpression of ADAP enhances, whereas loss of ADAP attenuates, TLR4-mediated STAT3 activity, suggesting ADAP is required and acts as a coactivator of STAT3 in TLR4-activated macrophages. The potential effect of ADAP on STAT3 activity requires the phosphorylation of ADAP on Y571 induced by TLR4 activation. Although LPS stimulation leads to partial disassembly of ADAP-STAT3 binding, phosphorylation of Y571 is nevertheless required for the TLR4-activated phosphorylation of STAT3. Collectively, our findings have thus expanded the new immunological roles for ADAP in the control of macrophage inflammation and polarization via modulating STAT3 activity.

We found that in response to LPS stimulation, macrophages from ADAP-deficient mice exhibited robustly enhanced M1 feature and polarization with a higher expression of M1 markers and proinflammatory cytokines, including TNF-α, IL-6, IL-12, and IFN-γ, but a lower expression of M2 markers along with decreased levels of anti-inflammatory cytokine IL-10. Furthermore, ADAP-deficient macrophages displayed an increased capacity in potentiation of Th1 responses, suggesting ADAP is inhibitory to TLR4-mediated M1-like polarization. Interestingly, this is consistent with the previous reports that ADAP-deficient mice displayed enhanced inflammation along with exaggerated production of proinflammatory cytokines including IL-1, IL-6, TNF-α, and IFN-γ during influenza virus infection (62).

Signaling in macrophages via TLR4 plays a key role in orchestrating macrophage polarization and inflammatory responses mainly via STAT1 and STAT3 (11). STAT3 controls the level of LPS-induced M1 polarization and inflammatory responses in macrophages by inhibiting M1 macrophage polarization and promoting an M2-like phenotype (11). STAT3-deficient mice also display enhanced macrophage M1 skewing with increased inflammation (21–23), a phenotype similar to that in ADAP-deficient mice (62). Consistent with this known function of STAT3 that inhibits M1 polarization, inhibition of STAT3 also increased the frequency of M1 polarization in WT macrophages; however, interestingly, this increase was marginal in ADAP-deficient macrophages (Fig. 1G), suggesting the stimulatory effect of ADAP depletion on LPS-induced macrophage M1-like polarization is via suppression of STAT3 action, and ADAP acts upstream of STAT3. Indeed, ADAP deficiency caused a significant deduction of LPS-induced STAT3 phosphorylation, nuclear translocation, and expression of downstream target genes, including SOCS3 and BCL3, whereas overexpression of ADAP enhanced LPS-induced STAT3 phosphorylation, further supporting that ADAP is required for TLR4-activated full activation of STAT3 in macrophages. Thus, ADAP deficiency orientates LPS-induced M1-toward polarization via an attenuation of the M2-skewed functions of STAT3 in macrophages. Interestingly, in contrast to our finding that ADAP deficiency decreases STAT3 phosphorylation/activation, it has been reported that loss of ADAP leads to an upregulation of IL-15–induced STAT5 phosphorylation in T cells (63), suggesting ADAP plays distinct roles in regulation of different members of the STAT transcription factor family in different immune cell context.

TLR4-mediated STAT3 phosphorylation is stimulated by secondary signals from autocrine produced cytokines, including IL-6 and IL-10 (56). The downregulation of LPS-induced STAT3 phosphorylation in ADAP-deficient cells is likely via an inhibitory effect on IL-6 signaling to STAT3. This hypothesis is supported by the observation that overexpression of ADAP potentiated IL-6–induced STAT3 phosphorylation (Fig. 5B). Furthermore, STAT3 phosphorylation was dampened in Adap^−/− macrophages in response to IL-6 but not IL-10 stimulation (Fig. 5A). Our findings point toward a loss of ADAP renders the macrophages insensitive to a second TLR4-induced autocrine stimulus of IL-6, leading to a decreased STAT3 tyrosine phosphorylation and LPS–IL-6–STAT3 axis–dominated M2 skew.

Mechanistically, we demonstrated that ADAP interacts with STAT3 in resting macrophages; however, surprisingly, LPS stimulation but not IL-6 or IL-10 stimulation results in decreased ADAP association with STAT3. Despite eliciting the dissociation of ADAP-STAT3 complex, LPS stimulation concurrently induces the increase in the tyrosine phosphorylation of STAT3 not only in WT macrophages but also in ADAP-deficient macrophages. This implies that before activation, STAT3 needs to be primed by ADAP to acquire its full activation during TLR4 signaling by LPS stimulation. Similarly, it was reported that ADAP is required for the assembly of the CARMA1–BCL-10–MALT1 complex in the TCR-induced activation of NF-κB (34). The role of the ADAP in the STAT3-containing complex could serve as a priming molecule that confers STAT3 in a conformation favoring initial STAT3 phosphorylation in response to LPS stimulation. Once the STAT3 tyrosine phosphorylation is initiated, ADAP is no longer needed anymore for sustained phosphorylation of STAT3.

ADAP bears multiple tyrosine phosphorylation sites within typical motifs at tyrosine residues 462 (Y462), 571 (Y571), and 559 (Y559), whose phosphorylation induced by TCR stimulation is essential for ADAP functions in T cell signaling (58–60). Our data showed that LPS stimulation induced a rapid increase of ADAP tyrosine phosphorylation in both macrophage cell lines and the primary peritoneal macrophages. Although, the putative kinase for ADAP tyrosine phosphorylation is Fyn in T cells (26), it was identified that Btk, a member of the Tec family of tyrosine kinases, mediated the ADAP tyrosine phosphorylation in TLR4-activated macrophages. In support of this claim, interestingly, the motif surrounded with Y571 of ADAP resembles an ITIM (58), whereas Btk can phosphorylate the ITIM-containing molecule such as PECAM-1 on the ITIMs (64, 65). Moreover, only Y571 site in ADAP was identified as the major tyrosine phosphorylation site in TLR4-activated macrophages, consistent with the previous finding that phospho-Y571 accounts for the most abundant ADAP phosphorylation in T cells (58). This suggests ADAP tyrosine phosphorylation is triggered on Y571 in a manner independent of immune cell type and receptor specificity. The LPS-induced STAT3 phosphorylation was substantially attenuated by ADAP deficiency. Mutation of the LPS-induced phosphorylation site Y571 led to a decrease in levels of phosphorylated STAT3 in responses to LPS stimulation. Functionally, while in T cells, Y571 mediates the binding of ADAP with the ZAP-70 kinase, and phosphorylation of Y571 of ADAP is involved in chemokine induced migration of T cells (60), tyrosine phosphorylation at Y571 is required for ADAP function in priming of TLR4-activated STAT3 phosphorylation, suggesting that same site plays multifunction with proteins in different cell context. Furthermore, the requirement of Y571 phosphorylation in ADAP-mediated regulation of STAT3-mediated M2 polarization was supported by the observation that whereas ex vivo overexpression of ADAP decreased LPS-induced M1-like polarization, Y571F-ADAP mutant lost its inhibitory activities, and by that, the decreased LPS-induced tyrosine phosphorylation of STAT3 observed in ADAP-deficient macrophages was recapitulated by overexpression of the Y571F-ADAP mutant. Thus, phosphorylation at Y571 of ADAP together with concomitant ADAP-STAT3 dissociation is essential for the regulation of TLR4-activated STAT3 phosphorylation in macrophages. However, it remains to be seen exactly how the phosphorylation at Y571 regulates the LPS-induced STAT3 phosphorylation. It could be that Y571 phosphorylation renders ADAP in a conformation that favors ADAP binding to STAT3 or provides a docking site that recruits a potential coactivator or kinase for phosphorylation of STAT3. Investigation of such an exact coactivator molecule recruited by Y571 phosphorylation for LPS-induced STAT3 phosphorylation would be of interest for future study.

In summary, our data identify a novel coactivator function of ADAP in TLR4-mediated STAT3 activity, modulating macrophage phenotypic polarization and inflammatory responses. In this study, initial TLR4-activated phosphorylation of ADAP at Y571 enables subsequent dissociation from STAT3. This, in turn, primes phosphorylation of STAT3, skewing the macrophage program toward an anti-inflammatory M2 phenotype. Our findings suggest that ADAP potentially serves as a novel target for therapeutic interventions in excessive inflammation or other immune disorders arose from macrophage polarization imbalance.

The authors have no financial conflicts of interest.