Resveratrol suppresses NTHi-induced inflammation via up-regulation of the negative regulator MyD88 short

Upper respiratory tract inflammatory diseases such as asthma and chronic obstructive pulmonary diseases (COPD) affect more than one-half billion people globally and are characterized by chronic inflammation that is often exacerbated by respiratory pathogens such as nontypeable Haemophilus influenzae (NTHi). The increasing numbers of antibiotic-resistant bacterial strains and the limited success of currently available pharmaceuticals used to manage the symptoms of these diseases present an urgent need for the development of novel anti-inflammatory therapeutic agents. Resveratrol has long been thought as an interesting therapeutic agent for various diseases including inflammatory diseases. However, the molecular mechanisms underlying its anti-inflammatory properties remain largely unknown. Here we show for the first time that resveratrol decreases expression of pro-inflammatory mediators in airway epithelial cells and in the lung of mice by enhancing NTHi-induced MyD88 short, a negative regulator of inflammation, via inhibition of ERK1/2 activation. Furthermore, resveratrol inhibits NTHi-induced ERK1/2 phosphorylation by increasing MKP-1 expression via a cAMP-PKA-dependent signaling pathway. Finally, we show that resveratrol has anti-inflammatory effects post NTHi infection, thereby demonstrating its therapeutic potential. Together these data reveal a novel mechanism by which resveratrol alleviates NTHi-induced inflammation in airway disease by up-regulating the negative regulator of inflammation MyD88s.


Resveratrol suppresses NTHi-induced inflammation in airway epithelial cells in vitro and in the lung of mice.
Previous studies have shown that resveratrol reduces expression of proinflammatory cytokines 29 . Human bronchial smooth muscle cells treated with resveratrol showed a decrease in pro-inflammatory cytokine release in response to lipoteichoic acid stimulation 26 . Therefore, we first sought to determine if resveratrol can reduce NTHi-induced inflammation in bronchial epithelial (BEAS-2B) cells. As shown in (Fig. 1a-d), resveratrol significantly suppresses NTHi-induced IL-1β , IL-6, CCL-2 and GM-CSF mRNA expression in a dose-dependent manner. Resveratrol also has been shown to reduce lung tissue inflammation in a murine model of allergic airway disease 25 . Similar to the in vitro findings, resveratrol suppresses NTHi-induced IL-1β , IL-6, CCL-2 and GM-CSF mRNA expression in the lung of mice ( Fig. 1e-h). A histological analysis was performed from lung tissue to verify the anti-inflammatory effects of resveratrol. Resveratrol reduces NTHiinduced leukocyte infiltration (Fig. 1i). To ensure these findings are not a result of cytotoxicity, lactate dehydrogenase (LDH) release was measured and no significant cytotoxicity was observed (Fig. 1j).

Resveratrol suppresses NTHi-induced inflammation by up-regulating MyD88 short expression.
We previously showed that NTHi induces expression of MyD88s in vitro and in the lung of mice. We also showed that MyD88s is a negative regulator of inflammation in airway epithelial cells 22 . An alternative strategy for the development of anti-inflammatory therapeutics is to up-regulate the negative regulators of inflammation. We thus hypothesized that the anti-inflammatory effects of resveratrol are a result of up-regulation of MyD88s. Interestingly, we found that resveratrol significantly enhances NTHi-induced MyD88s mRNA expression in BEAS-2B cells and in the lung of mice (Fig. 2a,b). Immunoprecipitation and western blot analysis were performed to confirm that resveratrol increases MyD88s protein expression in vitro (Fig. 2c). We then used MyD88s siRNA to further confirm that resveratrol decreases NTHi-induced proinflammatory cytokine expression by increasing MyD88s expression (Fig. 2d-j). Resveratrol no longer significantly reduces NTHi-induced IL-1β , IL-6, CCL-2 and GM-CSF mRNA expression in cells transfected with MyD88s siRNA (Fig. 2e-h). Furthermore, ELISA shows that resveratrol decreases NTHi-induced IL-6 and GM-CSF protein expression in BEAS-2B cells. However, resveratrol no longer significantly decreases NTHi-induced IL-6 and GM-CSF protein expression in cells transfected with MyD88s siRNA (Fig. 2i,j). Together these data suggest that resveratrol suppresses NTHi-induced lung inflammation by up-regulating MyD88s expression.
Resveratrol up-regulates NTHi-induced MyD88 short via suppressing ERK1/2 phosphorylation. We have previously shown that MyD88s is negatively regulated by the ERK1/2 signaling pathway.
Resveratrol enhances MyD88 short expression via up-regulating MKP-1. To further elucidate the molecular mechanisms by which resveratrol enhances NTHi-induced MyD88s, we investigated the role of MKP-1. MKP-1 is known to negatively regulate TLR-dependent immune responses by dephosphorylating the MAP (a-d) Relative quantities of human IL-1β (a) human IL-6 (b) human CCL-2 (c) and human GM-CSF (d) mRNA were measured by real-time QPCR analysis in human bronchial epithelial BEAS-2B cells. Cells were treated with 10 μ M, 25 μ M or 50 μ M of resveratrol followed by NTHi stimulation. (e-i) C57BL/6 mice were intraperitoneally inoculated with resveratrol 1 hour before intratracheal inoculation with NTHi or saline for the control. Relative quantities of murine IL-1β (e), murine IL-6 (f), murine CCL-2 (g) and murine GM-CSF (h) mRNA were measured by real-time QPCR analysis of RNA extracted from lung tissue. (i) Lung tissue was harvested for histological analysis (H&E stain, 400x). (j) Cells were treated with resveratrol and NTHi for 24 hours. Cytotoxicity was determined by measuring lactate dehydrogenase release from BEAS-2B cells. Data are mean ± SD (n = 3). *p < 0.05. Statistical analysis was performed using Student's t-test. n.s., nonsignificant. Data are representative of three or more independent experiments. kinase ERK 31,32 . An MKP-1 overexpression vector was used to determine that MKP-1 is a negative regulator of ERK1/2 phosphorylation in airway epithelial cells. As expected, MKP-1 overexpression reduces NTHi-induced ERK1/2 phosphorylation (Fig. 4a). Additionally, bacterial stimulation has been shown to induce MKP-1 mRNA expression in epithelial cells 33 . We thus postulated that resveratrol may mediate ERK1/2 phosphorylation by increasing NTHi-induced MKP-1 mRNA expression. As shown in Fig. 4b,c, NTHi induces MKP-1 mRNA expression, and interestingly, resveratrol significantly enhances MKP-1 mRNA expression in BEAS-2B cells and in the lung of mice. Western blot analysis shows that resveratrol increases MKP-1 protein expression in vitro (Fig. 4d).

NTHi-induced MyD88 short expression is mediated by MKP-1 up-regulation via a cAMP-PKA-dependent mechanism.
Resveratrol is known to directly inhibit PDE1, 3 and 4 25,26 . PDEs catalyze and degrade cAMP, an important second messenger with a critical role in regulating immune responses. PDE inhibitors therefore increase intracellular cAMP levels 34 . We therefore tested if cAMP plays a role in resveratrol-mediated MyD88s and MKP-1 enhancement using 8-bromo-cAMP, a cell-permeable cAMP analog resistant to PDE degradation. 8-bromo-cAMP treatment increases NTHi-induced MyD88s and MKP-1 mRNA expression (Fig. 5a,b). Forskolin, a cell permeable diterpenoid used to raise intracellular levels of cAMP, was then used to determine the role of cAMP in resveratrol-mediated MyD88s and MKP-1 expression. Forskolin treatment significantly increases NTHi-induced MyD88s and MKP-1 mRNA expression (Fig. 5c,d). To confirm the role of cAMP in resveratrol-mediated MyD88 and MKP-1 expression, we measured the effects of resveratrol on cAMP levels. Resveratrol significantly increases cAMP levels in BEAS-2B cells (Fig. 5e). Western blot analysis was performed to verify the effects of forskolin on NTHi-induced protein expression. As expected, forskolin increases NTHi-induced MKP-1 protein expression (Fig. 5f). Furthermore, forskolin no longer increases MyD88s mRNA expression in cells transfected with shMKP-1 confirming that cAMP acts upstream of MKP-1 (Fig. 5g). Together, these data suggest that cAMP is required for MKP-1 dependent up-regulation of NTHi-induced MyD88s.
A major downstream signaling effector of cAMP is protein kinase A (PKA) and the cAMP-PKA signaling pathway has been shown to regulate MKP-1 expression 32,33 . Therefore, we tested if PKA is involved in resveratrol-mediated MyD88s and MKP-1 expression. Cells treated with PKA inhibitor H89, followed by NTHi stimulation show a significant reduction of MyD88s and MKP-1 mRNA expression (Fig. 5h,i). PKI, a PKA-specific inhibitor, was further used to confirm PKA involvement. PKI treatment significantly decreases MyD88s and MKP-1 mRNA expression (Fig. 5j,k). Western blot analysis was then performed to confirm the  effects of PKI on NTHi-induced protein expression. As expected, PKI treatment increases NTHi-induced MKP-1 protein expression (Fig. 5l). Furthermore, PKI no longer increases MyD88s mRNA expression in cells transfected with shMKP-1, confirming that PKA acts upstream of MKP-1 (Fig. 5m). Collectively, these data suggest that resveratrol mediates MKP-1-dependent up-regulation of MyD88s expression via a cAMP-PKA-dependent mechanism.

Treatment with resveratrol post NTHi infection enhances MyD88 short expression and decreases expression of proinflammatory mediators in vitro and in the lung of mice.
We have demonstrated that pre-administration of resveratrol significantly decreases NTHi-induced lung inflammation via up-regulation of MyD88s. Furthermore, we have previously shown that treatment with other PDE inhibitors post infection has therapeutic potential 32 . We thus sought to validate the physiological relevance of resveratrol to the clinical situation. BEAS-2B cells were treated with resveratrol at various times pre-and post-NTHi stimulation. Our data show that resveratrol increases NTHi-induced MyD88s mRNA expression and decreases NTHi-induced IL-1β and IL-6 mRNA expression (Fig. 6a-f). Furthermore, there is no significant change in MyD88s, IL-1β and IL-6 mRNA expression among cells treated with resveratrol before and after NTHi stimulation (Fig. 6a-c). Mice were also treated with resveratrol 3 hours post NTHi stimulation. Consistent with the in vitro data, treatment with resveratrol post NTHi stimulation increases MyD88s and decreased IL-1β and IL-6 mRNA levels in lung tissue of  (Fig. 6d-f). Together these data confirm the physiological relevance for the clinical use of resveratrol for the treatment of NTHi-induced lung inflammation.

Discussion
Asthma and COPD affect more than one-half billion people globally resulting in billions of dollars in healthcare costs, and there are no cures [1][2][3][4][5] . These diseases are characterized by excessive inflammation that is often exacerbated by bacterial respiratory infections, such as NTHi 6,7 . The increasing numbers of antibiotic-resistant NTHi strains, tolerance issues and adverse side effects of currently available therapies, together with the increasing prevalence of airway diseases requires the development of novel, non-antibiotic therapeutics 5,10,11 . Resveratrol is being widely studied for its therapeutic potential for a vast array of diseases [28][29][30] . However, the molecular mechanisms by which resveratrol exerts its anti-inflammatory effects are still largely unknown. We have previously shown that MyD88s is a negative regulator of NTHi-induced inflammation in airway epithelial cells and that MyD88s is negatively regulated by ERK1/2 signaling pathways in a negative feedback manner suggesting that MyD88s is essential for maintaining immune homeostasis 22 .
In this study we show that resveratrol decreases NTHi-induced expression of pro-inflammatory mediators in airway epithelial cells and in the lung of mice. We demonstrate for the first time that resveratrol suppresses NTHi-induced inflammation by up-regulating MyD88s, a negative regulator of inflammation, via inhibition of ERK1/2 phosphorylation. Moreover, resveratrol mediates ERK-dependent down-regulation of MyD88s via up-regulation of MKP-1, a critical negative regulator of ERK. Furthermore, resveratrol enhances MKP-1 expression via a cAMP-PKA-dependent mechanism. Finally, we show that treatment with resveratrol post NTHi stimulation supports the clinical relevance of the use of resveratrol for the treatment of NTHi-induced lung inflammation. Together these data reveal a novel mechanism of the anti-inflammatory effects of resveratrol by which resveratrol alleviates NTHi-induced inflammation partially by counteracting ERK-dependent down-regulation of MyD88s via a cAMP-PKA-MKP-1 pathway (Fig. 7).
Many ongoing studies focus on the therapeutic potential of resveratrol with promising anti-inflammatory and airway remodeling results in respiratory disease models [25][26][27] . Concurrently, our findings suggest that resveratrol usage for acute lung inflammation has clinical significance. Human clinical trials have been performed with patients for cancer, cardiovascular disease and metabolic diseases. However, the results of these studies are not as promising as in vitro and in vivo studies with many failing to demonstrate any anti-inflammatory effects with resveratrol. These discrepancies among clinical trials have been attributed to study populations or design with a key factor being the limited bioavailability of resveratrol 35,36 . Metabolism in humans is rapid with conversion to metabolites within 30 minutes 28 . Therefore, identifying the molecular mechanisms of resveratrol's promising anti-inflammatory effects may lead to the development of better therapeutics with improved clinical trial outcomes. Furthermore, MyD88s may be a critical therapeutic target with significant therapeutic potential for suppressing inflammation associated with chronic airway disease.

Materials and Methods
Cell Culture. Human bronchial epithelial cells BEAS-2B cells (ATCC) were maintained in RPMI 1640 medium (Gibco) supplemented with 10% (v/v) heat-inactivated FBS (Sigma-Aldrich) and 100 units/mL penicillin and 0.1 mg/mL streptomycin. Cells were cultured at 37 °C in a humidified atmosphere of 5% CO 2 with a passage number no greater than 25 23,35 . Cells were seeded in 12 well plates for all experiments.

Bacteria Strain and Culture Conditions.
A clinical isolate of NTHi strain 12 was grown on chocolate agar plates at 37 °C in 5% CO 2 overnight, harvested and incubated overnight in brain heart infusion (BHI) broth supplemented with 3.5 μ g/ml NAD and hemin. Bacteria were subcultured in fresh BHI broth to log phase growth, as measured by optical density, pelleted washed, and resuspended in DMEM for in vitro experiments or isotonic saline for in vivo experiments. BEAS-2B cells were stimulated with NTHi at a multiplicity of infection (MOI) of 50 unless otherwise specified for 6 hours or as indicated.

Reagents and Antibodies.
Resveratrol was purchased from Calbiochem. PD98059, 8-bromo-cAMP, and forskolin were purchased from Enzo Life Sciences. PKI and H89 were purchased from EMD Millipore. PKI and 8-br-cAMP were reconstituted in water and diluted with RPMI 1640 media. All other inhibitors were reconstituted in dimethyl sulfoxide (DMSO) and diluted with RPMI 1640 media to a final concentration of 0.1% DMSO. 0.1% DMSO in RPMI 1640 media was used as a control. Forty-eight hours after seeding, cells were treated with 25 μ M of resveratrol or as indicated, 10 μ M of PD98059, 50 μ M of 8-bromo-cAMP, 10 μ M of forskolin, 50 μ M of PKI, or 20 μ M of H89 in a final volume of 500 μ L per well of 12 well plates one hour before stimulation with NTHi or as indicated. Antibody against MyD88 (ab133739) was purchased from Abcam. Antibodies against phospho-ERK1/2 (# 9101), total ERK1/2 (# 9102) and HRP-conjugated rabbit (# 7074) or mouse IgG (# 7076) were purchased from Cell Signaling Technology. Antibodies against MKP-1(# 370), α -tubulin (# 69969), normal rabbit IgG (#2027) and FITC-conjugated goat anti-mouse IgG (# 2010) were purchased from Santa Cruz Biotechnology.

Enzyme-linked immunosorbent assay (ELISA) and cAMP enzyme immunoassay. BEAS-2B cells
were treated with resveratrol 1 hour prior to NTHi stimulation. Culture supernatants were collected 12 hours after NTHi stimulation and centrifuged to remove cell debris prior to the assay. Human IL-6 and GM-CSF protein were measured with LEGEND MAX Human IL-6 and Human GM-CSF ELISA kits (Biolegend, Inc.) according to the manufacturer's instructions. cAMP levels were quantitatively measured using the Cyclic AMP Enzyme Immunoassay kit (Alfa Aesar) according to the manufacturer's instructions.
Cytotoxicity Assay. The CytoTox 96 ® non-radioactive cytotoxicity assay (Promega) was performed according to the manufacturer's instructions. Twenty-four hours after stimulation with 25 μ M or 50 μ M of resveratrol and NTHi, the supernatant was collected and cells were lysed. The supernatant was diluted 1:2 and lysate was diluted 1:10 with RPMI in a final volume of 50 μ L in a 96 well assay plate in triplicate. 50 μ L of RPMI was added in triplicate to correct for background. Each well was incubated with 50 μ L of a substrate for 30 minutes at room temperature protected from light, and 50 μ L of stop solution was added to each well. The absorbance was read at 490 nm, and the background values were subtracted from the sample readings. The percent LDH release was Scientific RepoRts | 6:34445 | DOI: 10.1038/srep34445 calculated as the absorbance of the supernatant divided by absorbance of the supernatant plus the absorbance of the lysate (endogenous LDH release).
RNA-mediated Interference. siRNA for MyD88s, 5′ -CCCAGCATTGGGCATATGCCT-3′ and ON-TARGETplus non-targeting control pool were purchased from Dharmacon. At 80% confluence, cells were transfected with 20 nM MyD88s siRNA and 1 μ L of DharmaFECT (Dharmacon) in 1000 μ L of media per well according to the manufacturer's instructions. shMKP-1 was generated as previously described 39 . MKP-1 knockdown was performed using 0.8 μ g DNA and 1 μ L Lipofectamine 3000 (Invitrogen) following the manufacturer's instructions. Forty-eight hours after transfection, cells were treated with resveratrol and stimulated with NTHi. The cell lysate was used for mRNA analysis.
Western Blot Analysis, Immunoprecipitation and Immunofluorescence. Western blot analyses were performed as previously described 33 . Following treatment with resveratrol and NTHi stimulation, cells were lysed, incubated on ice for 30 minutes and centrifuged at 12,000 × g for 15 minutes. Supernatants were separated on a 10% SDS-PAGE gel, transferred to polyvinylidene fluoride membrane. The membranes were blocked with a solution of Tris-buffered saline (TBS) containing 0.1% Tween 20 (TBS-T) and 5% nonfat dry milk, then incubated with primary antibodies against phospho-ERK1/2 or total ERK1/2 at a 1:2000 dilution or MyD88, MKP-1 or α -tubulin at a 1:1000 dilution in 5% BSA-TBS-T overnight. After three washes in TBS-T, the membrane was incubated with secondary HRP-conjugated rabbit or mouse IgG antibody at a 1:5000 dilution in 5% nonfat dry milk-TBS-T for 1 hour. Proteins were visualized using Amersham ECL Prime Western Blotting Detection Reagent (GE Healthcare Biosciences). Images were cropped for presentation. Full-size images are presented in Supplementary Figs 1-3. For immunoprecipitation, cell lysates were incubated with 4 μ L of antibodies against MyD88 or normal IgG and conjugated to protein G plus agarose beads (Santa Cruz Biotechnology) overnight at 4 °C. Eluted protein was subjected to western blot analysis. For immunofluorescence, cells were cultured on four-chamber microscope slides. After treatment with resveratrol and NTHi, the cells were fixed in 4% paraformaldehyde solution and incubated with antibodies against phospho-ERK1/2 or total ERK for 1 hour. FITC-conjugated goat anti-rabbit IgG was used to detect primary antibody. Slides were viewed and photographed using a Zeiss Axiophot microscope.
Plasmid and Transfection. The wild-type expression plasmid MKP-1 was previously described 40 . Cells were seeded in 12 well plates and at 80% confluence transfected with 0.8 μ g using TransIT-LT-1 reagent (Mirus) according to the manufacturer's instructions. Forty-eight hours after transfection, cells were stimulated with or without NTHi. The cell lysate was used for mRNA analysis or immunoprecipitation and western blot analysis.
Mice and Animal Experiments. C57BL/6 mice were purchased from the Jackson Laboratory.
Anaesthetized mice were intraperitoneally inoculated with 20 mg/kg resveratrol 1 hour before or 3 hours post intratracheal inoculation with NTHi at a concentration of 5 × 10 7 CFU per mouse or saline as control. Mice were sacrificed by intraperitoneal injection with 100 mg/kg sodium pentobarbital 6 hours after bacterial inoculation. Lung tissue was harvested for total RNA extraction as previously described 41 . For histological analysis, harvested lung tissue was fixed with 10% buffered formaldehyde, embedded in paraffin and sectioned at 4-μ M thickness. Sections were then stained with hematoxylin and eosin (H&E) to visualize the inflammatory response. Stained sections were visualized and images were recorded under light microscopy systems (Axiovert 40 CFL, Axiocam MRC, and Axiovision LE Image system; Carl Zeiss) as previously described 42 . All animal studies were carried out in accordance with the policies of, and with approval from, the Institutional Animal Care and Use Committee of Georgia State University.