Ferulic acid induces Th1 responses by modulating the function of 1 dendritic cells and ameliorates Th2-mediated allergic airway inflammation in mice
Chen-Chen Lee,1¶ Ching-Chiung Wang,2¶ Huei Huei-Mei Huang Mei Huang Mei HuangMei Huang,3 Chu-Lun Lin,3 Sy-Jye 5 Leu,4 and Yueh-Lun Lee4*
1Department of Microbiology and Immunology, School of Medicine, China Medical University, Taichung, Taiwan
2School of Pharmacy, College of Pharmacy, Taipei Medical University, Taipei, Taiwan 3Graduate Institute of Medical Sciences, College Medicine, Taipei graduate Institute of Medical Sciences, College Medicine, Taipei graduate Institute of Medical Sciences, College Medicine, Taipei raduate Institute of Medical Sciences,
¶Chen-Chen Lee and and Ching-Chiung Wang contributed equally to this work. *Corresponding Author: Yueh-Lun Lee E-mail: [email protected]
Running title: Immunomodulatory effects of ferulic acid on allergic asthma Abstract
This study investigated the immunomodulatory effects of ferulic acid (FA) on antigen-presenting dendritic cells (DCs) in vitro and its antiallergic effects against ovalbumin (OVA)-induced Th2-mediated allergic asthma in mice. The activation of FA-treated bone marrow-derived DCs by lipopolysaccharide (LPS) stimulation induced a high level of interleukin (IL)-12 but reduced the expression levels of the
proinflammatory cytokines IL-1β, IL-6, and tumor necrosis factor (TNF)-α. Compared to control-treated DCs, FA significantly enhanced the expressions of Notch ligand Delta-like 4 (Dll4), MHC class II, and CD40 molecules by these DCs. 9 Furthermore, these FA-treated DCs enhanced T cell proliferation and Th1 cell 10 polarization. In animal experiments, oral administration of FA reduced the levels of OVA-specific immunoglobulin E (IgE) and IgG1, and enhanced IgG2a antibody production in serum. It also ameliorated airway hyperresponsiveness, and attenuated 13 eosinophilic pulmonary infiltration in dose-dependent manners. In addition, FA 14 treatment inhibited the production of eotaxin, Th2 cytokines (IL-4, IL-5, IL-13), and 15 proinflammatory cytokines, but promoted the Th1 cytokine interferon (IFN)-γ 16 production in bronchoalveolar lavage fluid (BALF) and the culture supernatant of 17 spleen cells. These findings suggest that FA exhibits an antiallergic effect via 18 restoring Th1/Th2 imbalance by modulating DCs function in an asthmatic mouse 19 model.20
Introduction
Asthma is a heterogeneous chronic inflammatory lung disease that is characterized 2 by various airway obstructions, bronchial hyperresponsiveness, and airway 3
inflammation. It is recognized that Th2 cells and their cytokines (interleukin (IL)-4, 4 IL-5, and IL-13) are responsible for initiating and maintaining Th2-associated asthma
5 [1]. These Th2 cytokines induce an inflammatory cascade that comprises 6
allergen-specific immunoglobulin (Ig)E production, mast cell activation, eosinophil 7 recruitment, and airway hyperresponsiveness (AHR) [2]. In addition to Th2 cell 8 effects, dendritic cells (DCs), as professional antigen-presenting cells (APCs), play an 9 important role in antigen presentation in the airways, and the expression of 10 costimulatory molecules and cytokine profile by DCs can determine whether T cells 11 differentiate into type 1 T-helper (Th1) cells, Th2 cells, or regulatory T cells (Tregs) 12 [3,4]. In addition, the ability of DCs to polarize Th2 responses may be enhanced by 13 engagement of Notch receptors at the surface of T cells with ligands Jagged on DCs 14 [5]. Therefore, inhibition of Th2 effector responses by modulating DCs
maturation 15 and function is considered a promising immunomodulatory strategy to treat 16 Th2-associated allergic asthma.
Ferulic acid (trans-4-hydroxy-3-methoxycinnamic acid; FA; molecular weight 19 194.18) belongs to the family of phenolic acids and is widely found in vegetables, 20 fruits, and some beverages such as coffee and beer [6]. Moreover, FA is also a 21 component of Chinese medicinal herbs, such as Angelica sinensis, Cimicifuga 22 racemosa, and Ligusticum chuangxiong. The daily amount of FA ingested was 23 calculated to be around 150~250 mg [7]. After oral ingestion, FA is quickly absorbed 24 and reaches a peak plasma concentration within 30 min [8]. Because of its 25 antioxidative and anti-inflammatory activities, FA was shown to have therapeutic 1 effects in various diseases such as Alzheimer’s disease, cancer, diabetes mellitus, and 2 arthritis [9-12]. Therefore, the abundant dietary sources, the relatively low toxicity, 3 and potential bioactivities of FA may be capable of influencing immune cell
functions 4 in allergic immune responses and provide possible alternative options for relieving 5 allergic asthma-associated symptoms. However, to date, the protective effects of FA 6 against chronic inflammatory lung diseases, such as asthma, and the underlying 7 mechanisms are still unavailable.
In the current study, we attempted to clarify whether treatment with FA can alter 10 the phenotype and regulatory ability of DCs on T cells, and achieve antiallergic 11 effects to alleviate the development of airway symptoms in a mouse model of allergic 12 asthma. Our data revealed that FA modulates DC function to promote interferon 13 (IFN)-γ production by activated T cells and convert T cells with Th1 activity in 14 Th2-driven allergic diseases. These findings provide insights into how FA affects the 15 Th2-biased immune response and provide guidance on the use of FA as an antiallergic 16 adjuvant in treating Th2-mediated allergic asthma. 17
5
2.1 Mice
Female BALB/c and C57BL/6 mice were purchased from the National Laboratory Animal Center (Taipei, Taiwan) and maintained in the of Taipei Medical Animal Center (Taipei, Taiwan)
2.2 Preparation of bone marrow-derived DCs (BMDCs) DCs were obtained by culturing BALB/c bone marrow cells in RPMI-1640 12 containing 5% fetal bovine serum (FBS), glutamine, penicillin/streptomycin, murine IL-4 (1000 U/ml), and GM-CSF (500 U/ml) for 6 days. Non-adherent cells were harvested and their purity was identified by flow cytometry, gated on CD11c+ cells. The FACS analysis showed that there were 70%~80% of DCs in this cell population. 16
2.3 Determination of cytokine and chemokine levels 18
On day 6, 106 cells/ml of BMDCs were stimulated with lipopolysaccharide (LPS; 1 19 μg/ml) plus various concentrations of FA (20, 100, and 400 μM) which was purchased 20 from Sigma-Aldrich (St Louis, MO, USA), or LPS alone for 24 h to achieve activation 21 before use in the cytokine or phenotypical assays. In the experiments, untreated DCs 22 were cultured with medium alone as the control. Concentrations of 1β, IL-6, IL-10, 23 IL-12, and tumor necrosis factor (TNF)-α in culture supernatants were evaluated by 24 enzyme-linked immunosorbent assay (ELISA) kits (IL-1β, IL-6, IL-12, and TNF-α from eBioscience, San Diego, CA, USA; IL-10 from Duoset, R&D Systems, Minneapolis, MN, USA). The concentration of cytokines was measured by converting the OD values of the samples to pg from the standard curve. 3
The levels of eotaxin, IL-1β, IL-4, IL-5, IL-6, IL-10, IL-13, TNF-α, and IFN-γ in
bronchoalveolar lavage fluid (BALF) and culture supernatants of splenocytes were also 5 determined by commercial ELISA kits (eotaxin, IL-4, IL-5, IL-13 and IFN-γ from 6 Duoset, R&D Systems).
2.4 Quantitative real-time polymerase chain reaction 9
On day 6 of culture, BMDCs were collected and 106 cells/ml were treated with LPS 10 (1 μg/ml) and various concentrations of FA (20, 100, and 400 μM) or LPS alone for 6 h. 11 Untreated DCs were cultured with medium alone as a control. After incubation, cells 12 were collected, and total RNA was isolated according to the TRIzol method 13 (Sigma-Aldrich, St. Louis, MO, USA). Complementary (c)DNA was synthesized from 14 total RNA using the cDNA RT kit (Applied Biosystems, Foster City, CA, USA). 15 Real-time PCR was performed with cDNA using the PCR master mix and TaqMan 16 assays (Applied Biosystems). qPCR detection of mouse GAPDH, Delta1, Delta4, 17 Jagged1, and Jagged2 was conducted in triplicate using an Applied Biosystems 7900 18 PCR system. Quantification was carried out using GAPDH gene expression as an 19 internal normalization control, and the fold change in transcription was calculated using 20 the comparative threshold cycle (CT) method, 2-△△CT.
2.5 Flow cytometric analysis
Cytometric analysis was performed with a BD FACSort cell analyzer and CellQuest 24 (BD Biosciences) software. DCs were stained with antibodies in phosphate-buffered 25 saline (PBS) for 30 min at 4 °C and washed with 2 ml of PBS. Then, cells were 26 suspended in 0.5 mL of PBS with 0.1% sodium azide. Staining with isotype control 1 antibodies was performed in all experiments. DCs were gated according to the standard 2 forward- and side-scatter profiles of CD11c+ large cells. Anti-CD11c-fluorescein 3 isothiocyanate (FITC), anti-CD40-FITC, anti-CD86-FITC, anti-CD11c-phycoerythrin 4 (PE), anti-I-A/I-E (MHC class II)-PE, anti-CD80-PE, and anti-Jagged1-PE monoclonal 5 antibodies (mAbs) were purchased from eBioscience; anti-Delta1-PE, anti-Delta4-anti-Delta1-PE, 6 and anti-Jagged2-PE mAbs were purchased from BioLegend. 2.6 Mixed lymphocyte reaction (MLR)
This assay can be used to evaluate the stimulatory ability of DCs to alter T-cell 10 proliferation and cytokine production. Approximately 106 day-6 BMDCs derived from 11 BALB/c mice were treated with LPS (1 μg/ml), LPS plus FA (400 μM), or medium 12 only. After 24 h of culture, DCs were irradiated with 3000 rad (137Cs source). 13 Allogeneic naïve CD4+ T cells from female C57BL/6 mice were purified by using 14 CD4+ T cell isolation kit (Miltenyi Biotec, Auburn, CA, USA). Then CD4+ T cells (2 × 15 105 cells/well) were cultured with different proportions of irradiated DCs (0.5 × 104, 16 104, 2 × 104, and 2 × 104cells/well) in 96-well round-bottom plates for 2 days. Tritiated 17 thymidine (1 μCi/well, New England Nuclear, Boston, MA, USA) incorporation for 16 18 hours was determined with a dry scintillation counter (Packard Instrument, Meriden, 19 CT, USA). Additionally, levels of IL-5 and IFN-γ production in the culture supernatant 20 were assayed by ELISA kits.
2.7 Experimental protocol for the Th2 cell-mediated allergic asthma model 23 Female of BALB/c mice at 6 weeks of age (weight range 19~20 g) were 24
intraperitoneally (i.p.) sensitized with 50 μg OVA (grade V; Sigma-Aldrich, St. Louis, MO) plus 2 mg of aluminum hydroxide (Thermo Scientific, Rockford, IL, USA) on day 1, and boosted with 25 μg OVA in the same dosage of adjuvant on days 14, and 28. Subsequently, mice were intranasally (i.n.) challenged with 100 μg OVA on days 43 3 and 44. Finally, mice were exposed to an OVA aerosol challenge on days 45, 46, and 4 47 by inhalation 5% OVA in a normal saline using an atomizer (DeVilbiss Health Care, 5 Somerset, Pa., USA) for 30 min, and AHR was measured 1 d after the final challenge 6 (on day 48). To examine the effects of various dosages (25, 50, and 100 mg/kg of body 7 weight) of FA, three groups of mice (FA (25), FA (50) and FA (100)) were respectively 8 orally fed three doses of FA on days 21 to 47. sensitized and OVA-challenged 9 mice were fed sterilized water instead of FA as a positive control (PC). Negative 10 control (NC) mice were injected with PBS and challenged with OVA.
During the 11 experiment, no abnormal symptoms or toxic effects were observed in these groups. 12 Serum samples were collected on days 1, 7, 28, 35, and 48. 13 14
2.8 Serum antibody assay 15
OVA-specific IgE, IgG1, and IgG2a antibody titers in sera were determined by 16 ELISA (Becton Dickinson Biosciences, San Jose, CA, USA) as described previously 17 [13]. In brief, 96-well microtiter plates were coated with OVA (10 μg/ml) in NaHCO3 18 buffer (pH 9.6) at 4°C overnight, and then treated with mouse sera followed by 19 biotin-conjugated anti-mouse IgE, IgG1 or IgG2a (PharMingen, San Diego, CA) for 1 h 20 at 37°C. Then, avidin-horseradish peroxidase (1:5000, Pierce Biotechnology, Rockford, 21 IL, USA) was added to each well for another 30 min at room temperature. Finally, the 22 reaction was developed by 2,2’-azino-bis (3-ethylbenz-thiazoline-6-sulfonic acid) 23 (Sigma-Aldrich, St. Louis, MO) and the absorbance was determined at 405 nm. The 24 levels of antibody were compared with OVA-specific IgE, IgG1 and IgG2a standard 25 serum with predetermined concentrations (immunoglobulin
concentrations: IgE= 1.5 26
μg/ml, Ig G1= 32.5 μg/ml IgG2a= 15.2 μg/ml). The concentration of standard serum was 1 arbitrarily assigned as 1 ELISA unit (1 EU).
2.9 Measurement of airway responsiveness
At 24 h after the final aerosol OVA exposure, in-vivo lung function was measured by 5 detecting changes in airway resistance in response to increasing doses of aerosolized 6 methacholine (MCh, Sigma-Aldrich) in anesthetized mice as described previously [14]. 7 MCh aerosol was generated with a nebulizer and administrated directly through the 8 ventilator for 3 minutes. Airway reactivity was then monitored and data were expressed 9 as the pulmonary resistance (RL) to estimate airway resistance.
2.10 Analysis of the cellular composition Analysis of the cellular composition Analysis of the cellular composition Analysis of the cellular composition inBALF.
After measuring the pulmonary function parameters, lungs of mice were lavaged 13 three times with 1 mL of HBSS and the lavage fluids were centrifuged. The BALF 14 supernatants were further assayed by the ELISA method, and cell pellets were 15 resuspended in 1 mL HBSS. Total cells counts and differential counts of cytospin 16 preparations stained by Liu’s stain solution (Chi I Pao, Taipei, Taiwan) were 17 determined by microscopy. At least 200 cells were counted and identify as 18 macrophages, eosinophils, neutrophils, and lymphocytes under ×200 magnification. 19 Because macrophages and lymphocytes have similar staining profile, the different cell 20 types were identified based on their morphology and size. The lymphocyte has a round 21 nucleus and a high nuclear to cytoplasmic (N: C) ratio. Macrophages
have a relatively 22 low N: C ratio, and the cell size is larger (18-30 μm in diameter) than lymphocytes (6 23 μm in diameter).
2.11 Histological examination of lung tissues
After being lavaged, the lungs were immediately inflated with 10% formalin, fixed 2 for 24 hours, and embedded in paraffin. For histological examination, 5-μm-thick 3 sections were cut, placed on glass slides, deparaffinized, and sequentially stained with 4 hematoxylin-eosin (H&E) to evaluate inflammatory cell infiltration, or with periodic 5 acid-Schiff (PAS) to identify goblet cells. Light microscopy was used for the 6 histopathological assessment.
2.12 Statistical analysis
Experimental r Experimental r Experimental r Experimental r Experimental results are esults are esults are esults are esults are esults are presented presentedpresented presented as the mean ± standard error of (SEM).
3. Results
3.1 FA treatment enhanced Th1-polarizing cytokine production and reduced proinflammatory cytokine secretion by LPS-stimulated DCs
Before testing the effects of FA on BMDCs, we examined the effect of FA on cell viability. After 72 h of treatment with different doses of FA (0~5000 μM) in BMDCs, cell numbers were counted and cell viability was determined using the MTT assay. Results showed that the highest concentration (400 μM) of FA used in the following experiments was not cytotoxic (Figure S1). We then addressed the impact of FA on the 8 production of cytokines by DCs. LPS has been described as an inducer of DC activation and maturation. We used LPS as a positive control in this study. When BMDCs were treated with FA alone, there was no effect on cytokine production npublished data). Then, BMDCs were exposed to FA (20, 100, and 400 μM)
stimulation. In the presence of FA, dose-dependent increased 13 levels of the Th1-polarizing cytokine IL-12 were observed in LPS-stimulated DCs 14 (Figure 1). However, a dose-dependent decrease in IL-10 production was detected in 15 DCs treated with FA plus LPS. We then examined the impact of FA on the production 16 of proinflammatory cytokines, and as depicted in Figure 1, FA obviously inhibited 17 TNF-α, IL-β, and IL-6 release by LPS-stimulated DCs.
3.2 FA enhanced Notch ligand Delta expression by LPS-stimulated DCs 20
Variations in expressions of the Notch ligands, Delta (like 1 [Dll1] and 21 Delta-like 4 [Dll4]) and Jagged families (Jagged1 and Jagged2), by mature DCs were 22 shown to be crucial for T cell differentiation. It was established that the Dll group 23 directs T cell polarization toward Th1, whereas the Jagged group promotes Th2 or Treg 24
interfered with their expressions, as assessed using a real-time RT-PCR. After LPS 2 stimulation, DCs expressed higher levels of Dll1, Dll4, and Jagged1 mRNA synthesis 3 compared to untreated DCs (Figure 2A). Additionally, compared to LPS-stimulated 4 DCs, LPS combined with FA (400 μM) treatment induced higher amounts of Dll1 and 5 Dll4 transcripts in DCs. In contrast, exposure of DCs to LPS plus FA stimulation 6 significantly reduced Jagged1 mRNA synthesis, although there was no interruption in 7 Jagged2 transcripts. We further examined the surface expressions of these Notch 8 ligands on DCs by performing FACS analyses. Compared to untreated DCs (medium 9 group), a higher level of Dll1 was detected in LPS/FA-stimulated DCs (LPS+FA group), 10 although there was no significant difference between the groups (Figure 2B, C). 11 Notably, our data showed that the pattern of protein expression of Dll4 was
significantly 12 correlated with mRNA expression. Expressions of Jagged1 and Jagged2 showed no 13 difference among the groups.
3.3 Effects of FA on the maturation and activation of LPS-stimulated DCs
To investigate whether FA modulated the maturation and activation of mouse 17 BMDCs in vitro, we compared the phenotype of mouse DCs treated with medium, LPS 18 alone, or LPS plus FA (400 μM) for 24 h. As shown in Figure 3A and 3B,
expressions 19 of MHC class II and CD40 molecules during maturation were markedly upregulated in 20 LPS/FA-treated DCs compared to LPS-treated DCs or untreated DCs, respectively 21 (Figure 3). Furthermore, expressions of CD80 and CD86 molecules in LPS/FA-treated 22 DCs were higher than those of untreated DCs, although the difference was not 23 significant. 24
3.4 FA- and LPS-treated DCs enhanced T cell proliferation and Th1 polarization 2 Mature DCs have the capacity to induce proliferation of activated T cells at much 3 higher levels than immature DCs. To determine the ability of DCs to induce T cell 4 proliferation and cytokine production, an allogeneic MLR assay was used. DCs from 5 BALB/c mice were treated with medium, LPS alone, or LPS plus FA and then 6 co-cultured with C57BL/6 mouse-derived naive CD4+ T cells at various ratios 7 (DC/T-cell ratios of 1:5, 1:10, 1:20, and 1:40) in vitro. As shown in Figure 4A, 8 treatment with FA of DCs cultured in the presence of LPS stimulated T-cell 9 proliferative responses more effectively than those of LPS-treated DCs (Figure 4A). 10 Additionally, we found that FA modulated the secretion of IFN-γ and IL-5 by activated 11 CD4+ T cells. FA- and LPS-treated DCs markedly enhanced T cell secretion of IFN-γ 12 in the
supernatant compared to treated DCs (Figure 4B). In contrast, FA plus 13 LPS-treated DCs significantly suppressed IL-5 release by T-cells in a dose-dependent 14 manner. These results showed that FA can enhance the Th1 response through DCs. 15
cells. At the first At the first At the firstAt the first At the first , we evaluated the possible cytotoxic effect of FA on CD4+ T cells. 17 After treatment of 1 to 50 μM of FA for 72 h, T cells did not show any change on cell 18 viability as detected by MTT assay (Figure S2A). To evaluate the direct effect of FA on 19 T cells, CD4+ T cells treated with FA (5, 10, and 50 μM) and activated by 20 anti-CD3/anti-CD28 antibodies. Compared to control T cells (anti-CD3/anti-CD28 21 treatment group), activated T cells cultured in the presence of FA did not show any 22 apparent differences in cytokine production (IFN-γ, IL-10, and IL-5) (Figure S2B). 23 These results indicated that FA could not directly inhibit or promote cytokine 24 production by T cells. 25 3.5 FA reduced serum anti-OVA IgE and IgG1 and enhanced anti-OVA IgG2a 2 levels in an animal model of asthma 3
To investigate the in vivo impacts of FA on regulating immune responses of allergic 4 diseases, groups of mice were exposed to OVA sensitization and challenge (Figure 5A). 5 These OVA-sensitized mice were orally fed different doses (25, 50, and 100 mg/kg of 6 body weight (BW)) of FA (FA (25), FA (50) and FA (100), respectively). PC mice 7 received distilled water instead of FA. NC mice were neither sensitized with OVA nor administered FA treatment, although they received an OVA challenge. Serum samples were collected on indicated days, and OVA-specific IgE, IgG1, and IgG2a serum 10 antibody titers were determined by an ELISA. In mice, the Th2 cytokine, IL-4, 11 promotes IgG1 and IgE production. However, the Th1 cytokine, IFN-γ, induces the 12 production of IgG2a. Therefore, expression levels of subclasses (IgG1 and IgG2a) or 13 classes (IgE) of antibodies can be used as markers to identify what type of Th response 14 was induced in these mice. In the PC group, strong IgE and IgG1 production and low 15 IgG2a expression were induced (Figure 5B). In contrast, FA treatment efficiently 16 inhibited OVA-specific IgE and IgG1 production. Notably, the administration of FA 17 markedly upregulated IgG2a synthesis compared to the PC group.
3.6 FA decreased the severity of AHR and airway inflammation 20
To assess the antiallergic effects of FA on allergic asthma, both the AHR and 21 accumulation of inflammatory cells in BALF were investigated. One day after the last OVA challenge, the airway responsiveness to aerosolized MCh of each group of mice was measured. PC mice, which were administered distilled water instead of FA and sensitized and challenged with OVA, developed markedly increased airway 1
responsiveness to MCh inhalation compared to that of non-allergic NC mice (Figure 2 6A). However, oral treatment with FA dose-dependently alleviated the development of 3 AHR compared to the PC group. Furthermore, in the NC group, cells in the BALF were 4 mostly composed of macrophages (Figure 6B). On the other hand, in the PC group, 5 exposure to aerosolized OVA induced marked increases in cell numbers of
eosinophils 6 in the BALF. In contrast, mice administered FA showed significantly reduced increases 7 in eosinophils. Furthermore, a pathological examination of lung tissue sections from the 8 PC group exhibited increased numbers of inflammatory cells around the 9 peribronchiolar region (Figure 6C), relative to the NC group. Of note, lungs of mice 10 administered FA demonstrated efficient inhibitory effects on peribronchial 11 inflammation. 12
3.7 The influence of FA on OVA-induced chemokine and cytokine secretions in BALF Eotaxin is the most potent chemokine for recruiting eosinophils. Evaluation of the eotaxin level in BALF showed that administration of FA significantly reduced eotaxin secretion in a dose-dependent manner (Figure 7). This result indicated that FA 18 suppressed eotaxin secretion, and subsequently decreased numbers of recruited 19 eosinophils. Furthermore, to determine the possible effect of FA on T-cell responses, 20 we assessed levels of the important Th2 cytokines, IL-4, IL-5, and IL-13, in BALF. The 21 data showed that FA treatment had dose-dependent suppressive effects on the 22 production of Th2 cytokines compared to the PC group. Similarly, the expression level 23 of TNF-α in the BALF also decreased in FA-treated mice. In contrast, we observed 24
significant enhancement of BALF IFN-γ expression in mice treated with FA compared 1 to the PC group. 2
3.8 Effects of FA on OVA-induced cytokine secretions by splenocytes 4
By studying the production of Th1 and Th2 cytokines in cell culture supernatants 5 from the spleen, we determined whether it was possible to affect Th2 immune 6 responses through inducing IFN-γ-producing T cells. Freshly isolated spleen cells from 7 different groups of mice were cultured with OVA. After 3 days of culture, cell 8 supernatants were collected and analyzed by an ELISA. Results showed that FA 9 administration markedly inhibited IL-4, IL-5, and IL-13 production by splenocytes 10 (Figure 8). Furthermore, the data demonstrated that OVA-sensitized mice treated with 11 FA exhibited an obviously enhanced level of IFN-γ production. We also checked 12 expressions of the proinflammatory cytokines, TNF-α, IL-1β, and IL-6, by splenocytes 13 and found that three dosages of FA treatment significantly reduced production of these 14 cytokines relative to mice in the PC group. Taken together, the above findings suggest 15 that FA treatment might induce protective immunity against established Th2-mediated 16 allergic asthma through induction of IFN-γ production by OVA-specific T cells. 17
4. Discussion
Asthma affects over 300 million people worldwide, and constitutes heavy medical, 2 social, and economic burdens, because its prevalence is continually increasing. Still, in many cases, although corticosteroid and β-agonists can improve asthma
symptoms, their long-term use has adverse effects. Therefore, it is imperative to find substances, preferably non-steroidal in nature and especially those derived from plants used as folk medicines or foods, which have therapeutic effects on allergic asthma and can be taken for a long time with no side effects. 8
FA is one of the most abundant phenolic acids in plants. It acts as a potent 10
antioxidant which effectively scavenges free radicals and inhibits lipid peroxidation 11 [19]. FA reduced hydrogen peroxide-induced lipid peroxidation in peripheral blood 12 mononuclear cells, and this effect was more evident than that of other polyphenols such 13 as caffeic acid and ellagic acid [20]. FA also increases the activity of enzymes that can 14 counteract free radical-induced damage. Based on this evidence, FA was shown to have 15 a protective role in many disorders, including cardiovascular dysfunctions, cancer, 16 diabetes, and neurodegenerative diseases [21-23]. On the other hand, FA exerts 17 anti-inflammatory properties by affecting the immune system. Hirabayashi et al. found 18 that FA reduced IL-8 production in the influenza virus-infected RAW264.7 19 macrophage cell line [24]. In the presence of FA, respiratory syncytial virus 20 (RSV)-infected RAW264.7 cells showed reduced macrophage inflammatory protein 21 (MIP)-2 production in a dose-dependent manner [25]. Further, FA inhibited the release 22 of TNF-α from phytohemagglutinin-stimulated splenocytes [26]. These data indicate 23 that FA inhibits inflammation by interfering with immune cells.
In the immune system, DCs are the most potent APCs and are critically involved in initiating primary immune responses and inducing T cell responses. Thus, elucidating the immunomodulatory effect of FA on DCs could be crucial for understanding the impacts of FA on immune responses. As mature APCs, fully activated mature DCs display high levels of peptide-MHC class II complexes on their surfaces and
upregulate surface levels of co-stimulatory molecules (CD40, CD80, and CD86). These mature 6 DCs are activated through CD40 and produce high levels of IL-12 and TNF-α which 7 drive Th1 cell differentiation [27]. In contrast, differentiation of Th2 cells depends on a 8 low-IL-12 and high-IL-4 environment. IL-4 is not produced by DCs but by other cells, 9 such as mast cells [28]. Similarly, differentiation of Th17 cells is also determined by 10 the cytokine milieu. IL-1β, IL-6, and TGF-β program Th17 cell differentiation [29]. In 11 our study, a FACS analysis of DCs cultured in the presence of FA plus LPS revealed 12 upregulation of MHC class II and CD40 molecules. Further, these FA- and LPS-treated 13 DCs exhibited increased production of IL-12. However, they did not release elevated 14 levels of proinflammatory cytokines, such as TNF-α, IL-1β, or IL-6. Thus, from the 15 above results, we predict that Th1 differentiation may preferably be driven by FA- and 16 LPS-treated DCs. Despite it is interesting to discuss which intracellular signaling 17 pathway was involved in modulating those
cytokines gene expression in DCs by FA, 18 further evidence is not available. MAP kinases (MAPKs) consist of three major groups, 19 p38 MAPK, the c-Jun N-terminal kinase 1/2 (JNK1/2) and extracellular 20 signal-regulated kinase 1/2 (ERK1/2), and previous reports indicate that these proteins 21 play a role in regulating cytokine production from APCs [30]. One study reported that 22 LPS instructs DCs to produce IL-12, which depends on the phosphorylation of p38 23 MAPK and JNK1/2 [31]. Thus, we predict that FA might selectively enhance IL-12 24 production via amplification of p38 MAPK or JNK1/2 pathway. In addition, FA can act 25 as an antioxidant to reduce proinflammatory cytokine production by scavenging 26
reactive oxygen species (ROSs) [21]. Thus, the inhibition of TNF-α, IL-1β, and IL-6 by 1 FA might be via a mechanism involving the blockage of ROS pathway. In contrast, it is 2 possible that the up-regulation of IL-12 production in DCs is through the inhibition of 3 ROS activity. This explanation is supported by one report that H2O2, a major 4 component of ROSs, down-regulates IL-12 production in murine macrophages in 5 response to stimulation with LPS plus IFN-γ [32].
In the current study, we found that FA treatment induced a higher expression level 8 of Dll4 by DCs than those of other Notch ligands. In addition, co-culture of CD4+ T 9 cells with LPS-pulsed Dll4-stimulated DCs enhanced the differentiation of Th1 cells 10 characterized by the secretion of increased levels of IFN-γ. Unstimulated DCs express 11 low levels of Notch ligands. Under different conditions, the expression of Notch 12 ligands can be induced towards preferential expression of either Jagged or Delta in 13 response to different stimuli. Amsen et al. demonstrated that expression of Dll4 by DCs 14 was correlated with the ability of LPS to polarize Th1 cells, whereas Jagged1 15 expression was correlated with the ability of LPS to polarize Th2 cells [5]. The 16 discovery of the contribution of Notch signaling in Th1/Th2 differentiation opens a 17 new pathway for regulating helper T-cell differentiation. Furthermore, Sun et al. 18 showed that ectopic expression of Dll4 by BMDCs promoted Th1-cell differentiation, 19 and allowed them to strongly inhibit Th2 development [33]. In addition, inhibition of 20 Th2-type cytokine IL-4 production allows Th cells activated by Dll4 to develop into 21 Th1 cells through a pathway that is IL-12 independent. In summary, our results showed 22 that stimulation via FA can modify the maturation profile and function of DCs, and 23 high expression of Dll4 by these DCs enhanced their ability to activate naïve Th cells 24 and promote Th1 cell development. 25 In the present study, for the first time, we explored the immunomodulatory effects of 1 FA in a murine asthma model. Our study demonstrated that oral administration 25 to 50 2 mg/kg of FA could get significantly protective effects against Th2
cell-mediated 3 allergic asthma in mice. FA is low toxicity with LD50 of 2445 mg/kg and 2113 mg/kg 4 body weight in male and female rats, respectively [34]. Although
human may consume 5 as much as 80 to165 mg FA per meal, low bioavailability after oral administration may 6 limit the clinical use of FA. Another limitation is that many studies on humans were 7 carried out by using foods containing FA rather than the purified FA. Thus, 8 unfortunately, we could not compare with these studies to evaluate the suitable dose of 9 FA applied in human.
Our results clearly demonstrated that orally administered FA has an antiallergic 12 effect via suppressing OVA-specific IgE production. Serum IgE is one of the most 13 important clinical markers for allergic responses. Such a reduction in serum 14 OVA-specific IgE would likely relieve the allergic syndrome mediated by mast cells 15 [35]. Further, our present results showed that FA treatment significantly induced a 16 reduction in the IL-4 level in mice. IL-4 is a switching factor for IgE production, and it 17 also induces the rolling and adhesion of circulating eosinophils to endothelial cells [36]. 18 Therefore, we speculated that FA may have an antiallergic effect on allergic asthma by 19 suppressing IL-4 secretion which consequently reduces IgE production and eosinophil 20 infiltration into the lungs. 21
Airway eosinophilic inflammation is one of the characteristics of bronchial allergic 23 asthma. It is believed that eosinophil transmigration into the airways is orchestrated by 24 cytokines, such as IL-4, IL-5, TNF-α, and IL-13, and is coordinated by specific 25 chemokines, such as eotaxin [37,38]. Specifically, IL-5 is the most important mediator that regulates eosinophilic inflammation through its effects on the proliferation, 1 differentiation, and activation of eosinophils and as a signal for mobilization of 2 eosinophils from the bone marrow into the lungs after an allergen challenge [39,40]. 3 Additionally, TNF-α and IL-1β are particularly important inflammatory mediators 4 because they play major roles in coordinating mechanisms that command 5
proinflammatory responses. Previous studies reported that levels of TNF-α, IL-1β, and 6 IL-6 released in the BALF of asthmatic patients increased [41]. TNF-α and IL-1β can 7 induce the synthesis of eotaxin in human lung epithelial cells [42]. The results of our 8 study clearly showed that FA attenuated the overproduction of
Th2-associated 9 cytokines and downregulated eotaxin expression. Likewise, levels of these 10 proinflammatory cytokines were markedly reduced in FA-treated asthmatic mice. As 11 expected, consistent with results of the histological analysis of lung tissues, FA 12 successfully alleviated lung inflammation and reduced infiltration of total inflammatory 13 cells in peribronchiolar regions. Based on these findings, we speculated that FA inhibits 14 the recruitment of inflammatory cells by reducing levels of Th2-type cytokines, 15 chemokines, and proinflammatory cytokines, providing a possible mechanistic 16 explanation for the observed profound
regulatory effects of FA on the development of 17 airway inflammation in this OVA-induced asthma model. 18
AHR is a major characteristic feature of asthma and a pathophysiologic consequence 20 of the effect of inflammatory processes. In particular, IL-13 can directly cause AHR by 21 affecting airway epithelial cells and airway smooth muscle cells [43,44]. Herein, we 22 found that BALF and spleen IL-13 levels decreased after oral administration of FA; 23 therefore, it caused attenuation of AHR sensitized by an allergen in mice. IL-10 was 24 shown to be produced by specialized subsets of Treg cells and the regulatory activities 25 of T cells were found to be closely related to their IL-10-producing ability [45,46]. In this murine asthma model, we detected the expression level of IL-10 in splenocytes. 1 The data showed that FA treatment slightly reduced IL-10 production in splenocytes 2 (Figure S3A). In addition, treatment with FA enhanced but not reduced the 3 OVA-specific T-cell proliferative response (Figure S3B). These results imply that FA 4 treatment did not induce Treg-cell response in this asthmatic animal model.
By administering FA, we demonstrated that FA-treated DCs stimulated T-cell 7 proliferation and enhanced the generation of Th1 cells in an allogeneic MLR. In an 8 animal study, we also showed that treatment of asthmatic mice with FA had
inductive 9 effects on Th1 responses, including anti-OVA IgG2a and IFN-γ production. IFN-γ, 10 which is secreted by Th1 cells, promotes the proliferation of Th1 cells and induces 11 IgG2a production by B cells. It inhibits Th2 cell proliferation in vitro. Many studies 12 demonstrated that the exogenous delivery of IFN-γ suppresses
recruitment of 13 eosinophils, and inhibits AHR and bronchial mucus production in asthmatic mice 14 [47-49]. These data support the ability of IFN-γ to block Th2 cell cytokine production 15 and CD4 T cell recruitment to the lungs. To examine the precise mechanism of 16 suppression, Nakagome et al. delivered IFN-γ to asthmatic mice using an 17 IFN-γ-producing plasmid and reported that IFN-γ attenuated the OVA-induced Th2 18 immune response by suppressing DC functions, such as antigen presentation, cytokine 19 production, and migration to the lungs [49]. Mitchell et al. used IFN-γ 20 receptor-expressing transgenic mice and showed that IFN-γ acting through the airway 21 epithelium blocked Th2-mediated mucus and inhibited eosinophil generation in the 22 bone marrow [50]. IFN-γ is also considered to play some role in host defense against 23 rhinovirus infection in asthmatic patients. It was speculated that a decrease in its 24 production by peripheral blood mononuclear cells or by lung cells may result in an 25 exacerbation of asthma [51,52]. These findings suggest that IFN-γ may be beneficial in
treating asthma by inhibiting virus-induced exacerbation. Although those studies 1 indicated a novel role for IFN-γ as a broad immune suppressor to modulate immune 2 responses in allergic inflammation, understanding the various mechanisms of IFN-γ regulation in Th2-mediated allergic asthma may define a pathway for new asthma 4
therapies to target in the future. 5. Conclusions
In conclusion, the current study demonstrated the ability of FA to induce maturation 2 of a distinct subset of DCs after in vitro activation with LPS. In addition, these DCs 3 induced differentiation of IFN-γ-secreting Th1 cells in vitro. Furthermore, oral 4 administration of OVA induced asthma in recipient mice with FA that could lead to 5 differentiation of a specific Th1 population able to downregulate reactivity mediated by 6 IFN-γ production. Therefore, we concluded that the antiallergic effects of FA was due 7 to the modulation of DCs, which promoted the differentiation of naïve T cells into Th1 8 cells thereby restoring the Th1/Th2 imbalance in a Th2-mediated allergic asthmatic 9 mouse model. Thus, FA might become a very attractive
immunomodulatory material 10 that can supplement the human diet. However, further research is necessary to focus on 11 its immunomodulatory activities on DCs. Acknowledgments
This study was supported by research grants from the Ministry of Science and 2 Technology, Taiwan (NSC 99-2314-B-038-002-MY3 and MOST 3 103-2320-B-038-032-MY3). 4
References
[1] C. M. Lloyd and S. Saglani, “T cells in asthma: influences of genetics, 3
environment, and T-cell plasticity,” The Journal of Allergy and Clinical 4 Immunology, vol. 131, no. 5, pp. 1267-1274; quiz 1275, 2013. 5
[2] S. J. Galli, M. Tsai, and A. M. Piliponsky, “The development of allergic 6 inflammation,” Nature, vol. 454, no. 7203, pp. 445-454, 2008. 7
[3] M. Moser and K. M. Murphy, “Dendritic cell regulation of TH1-TH2 8 development,” Nature Immunology, vol. 1, no. 3, pp. 199-205, 2000. 9
[4] O. Akbari and D. T. Umetsu, “Role of regulatory dendritic cells in allergy and 10 asthma,” Current Allergy and Asthma Reports, vol. 5, no. 1, pp. 56-61, 2005. 11 [5] D. Amsen, J. M. Blander, G. R. Lee, K. Tanigaki, T. Honjo, and R. A. Flavell, 12 “Instruction of distinct CD4 T helper cell fates by different notch ligands on 13 antigen-presenting cells,” Cell, vol. 117, no. 4, pp. 515-526, 2004. 14
[6] M. D'Archivio, C. Filesi, R. Di Benedetto, R. Gargiulo, C. Giovannini, and R. 15 Masella, “Polyphenols, dietary sources and bioavailability,” Annali 16 dell'Istituto Superiore di Sanita, vol. 43, no. 4, pp. 348-361, 2007. 17
[7] E. Barone, V. Calabrese, and C. Mancuso, “Ferulic acid and its therapeutic 18 potential as a hormetin for age-related diseases,” Biogerontology, vol. 10, no. 19 2, pp. 97-108, 2009. 20
[8] L. Rondini, M. N. Peyrat-Maillard, A. Marsset-Baglieri, and C. Berset, 21 “Sulfated ferulic acid is the main in vivo metabolite found after short-term 22 ingestion of free
ferulic acid in rats, ” J Agric Food Chem, vol. 50, no. 10, pp. 23 3037-3041, 2002. 24 [9] M. J. Kim, S. J. Choi, S. T. Lim et al., “Ferulic acid supplementation prevents 25 trimethyltin-induced cognitive deficits in mice,” Bioscience, Biotechnology, 26 and Biochemistry, vol. 71, no. 4, pp. 1063-1068, 2007. 27
[10] H. Zhu, Q. H. Liang, X. G. Xiong et al., “Anti-inflammatory effects of the 28 bioactive compound ferulic acid contained in Oldenlandia diffusa on 29 collagen-induced arthritis in rats,” Evidence-Based Complementary and 30 Alternative Medicine, vol. 2014, Article ID 573801, 2014. 31
[11] S. Roy, S. K. Metya, S. Sannigrahi, N. Rahaman, and F. Ahmed, “Treatment 32 with ferulic acid to rats with streptozotocin-induced diabetes: effects on 33 oxidative stress, pro-inflammatory cytokines, and apoptosis in the pancreatic 34 beta cell,” Endocrine, vol. 44, no. 2, pp. 369-379, 2013. 35
[12] N. Baskaran, S. Manoharan, S. Balakrishnan, and P. Pugalendhi, 36 “Chemopreventive potential of ferulic acid in 37
27 27
7,12-dimethylbenz[a]anthracene-induced mammary carcinogenesis in 1 Sprague-Dawley rats,” European Journal of Pharmacology, vol. 637, no. 1-3, 2 pp. 22-29, 2010. 3
[13] K. J. Liu, S. J. Leu, C. H. Su, B. L. Chiang, Y. L. Chen, and Y. L. Lee, 4 “Administration of polysaccharides from Antrodia camphorata modulates 5 dendritic cell function and alleviates allergen-induced T helper type 2 6 responses in a mouse model of asthma,” Immunology, vol. 129, no. 3, pp. 7 351-362, 2010. 8
[14] H. M. Huang, G. Hsiao, C. K. Fan et al., “Notch ligand delta-like 4-pretreated 9 dendritic cells alleviate allergic airway responses by enhancing IL-10 10 production,” PloS One, vol. 8, no. 5, e63613, 2013. 11
[15] D. Amsen, A. Antov, and R. A. Flavell, “The different faces of Notch in 12 T-helper-cell differentiation,” Nature Reviews Immunology, vol. 9, no. 2, pp. 13 116-124, 2009. 14
[16] J. S. Yuan, P. C. Kousis, S. Suliman, I. Visan, and C. J. Guidos, “Functions of 15 notch signaling in the immune system: consensus and controversies,” Annual 16 Review of Immunology, vol. 28, pp. 343-365, 2010. 17
[17] F. Radtke, N. Fasnacht, and H. R. Macdonald, “Notch signaling in the immune 18 system,” Immunity, vol. 32, no. 1, pp. 14-27, 2010. 19
[18] T. Ito, J. M. Connett, and S. L. Kunkel, and A. Matsukawa, “Notch system in 20 the linkage of innate and adaptive immunity,” Journal of Leukocyte Biology, 21 vol. 92, no. 1, pp. 59-65, 2012. 22
potential through its antioxidant property,” Journal of Clinical Biochemistry 24 and Nutrition, vol. 40, no. 2, pp. 92-100, 2007. 25
[20] K. L. Khanduja, P. K. Avti, S. Kumar, N. Mittal, K. K. Sohi, and C. M. Pathak, 26 “Anti-apoptotic activity of caffeic acid, ellagic acid and ferulic acid in normal 27 human peripheral blood mononuclear cells: a Bcl-2 independent mechanism,” 28 Biochimica et Biophysica Acta, vol. 1760, no. 2, pp. 283-289, 2006. 29
[21] C. Mancuso and R. Santangelo, “Ferulic acid: pharmacological and 30 toxicological aspects, ” Food Chem Toxicol, vol. 65, pp. 185-195, 2014. 31
[22] M. A. Alam, C. Sernia, and L. Brown, “Ferulic acid improves cardiovascular 32 and kidney structure and function in hypertensive rats,” Journal of 33 Cardiovascular Pharmacology, vol. 61, no. 3, pp. 240-249, 2013. 34
[23] E. Y. Kwon, G. M. Do, Y. Y. Cho, Y. B. Park, S. M. Jeon, and M. S. Choi, 35 “Anti-atherogenic property of ferulic acid in apolipoprotein E-deficient mice 36 fed
Western diet: comparison with clofibrate,” Food and Chemical Toxicology, 37 vol. 48, no. 8-9, pp. 2298-2303, 2010. 38
28 28
[24] T. Hirabayashi, H. Ochiai, S. Sakai, N. Nakajima, and K. Terasawa, 1 “Inhibitory effect of ferulic acid and isoferulic acid on murine interleukin-8 2 production in response to influenza virus infections in vitro and in vivo,” 3 Planta Medica, vol. 61, no. 3, pp. 221-226, 1995. 4
[25] S. Sakai, H. Kawamata, T. Kogure et al., “Inhibitory effect of ferulic acid and 5 isoferulic acid on the production of macrophage inflammatory protein-2 in 6 response to respiratory syncytial virus infection in RAW264.7 cells,” 7 Mediators of Inflammation, vol. 8, no. 3, pp. 173-175, 1999. 8
[26] L. Ou, L. Y. Kong, X. M. Zhang, and M. Niwa, “Oxidation of ferulic acid by 9 Momordica charantia peroxidase and related anti-inflammation activity 10 changes,” Biological & Pharmaceutical Bulletin, vol. 26, no. 11, pp. 11 1511-1516, 2003. 12 [27] N. Kadowaki, “Dendritic cells: a conductor of T cell differentiation,” 13 Allergology International, vol. 56, no. 3, pp. 193-199, 2007. 14
[28] M. Kool, H. Hammad, and B. N. Lambrecht, “Cellular networks controlling 15 Th2 polarization in allergy and immunity,” F1000 Biology Reports, vol. 4, pp. 16 6, 2012. 17
[29] H. Vroman, B. van den Blink, and M. Kool, “Mode of dendritic cell activation: 18 the decisive hand in Th2/Th17 cell differentiation. Implications in asthma 19 severity? ,” Immunobiology, vol. 220, no. 2, pp. 254-261, 2015. 20
[30] A. K. Yi, J. G. Yoon, S. J. Yeo, S. C. Hong, B. K. English, and A. M. Krieg, 21 “Role of mitogen-activated protein kinases in CpG DNA-mediated IL-10 and 22 IL-12
production: central role of extracellular signal-regulated kinase in the 23 negative feedback loop of the CpG DNA-mediated Th1 response, ” J Immunol, 24 vol. 168, no. 9, pp. 4711-4720. 25
[31] S. Agrawal, A. Agrawal, B. Doughty et al., “Cutting edge: different Toll-like 26 receptor agonists instruct dendritic cells to induce distinct Th responses via 27 differential modulation of extracellular signal-regulated 28 kinase-mitogen-activated protein kinase and c-Fos, ” J Immunol, vol. 171, no. 29 10, pp. 4984-4989, 2003. 30 [32] N. Khan, S. S. Rahim, C. S. Boddupalli et al., “Hydrogen peroxide inhibits 31 IL-12 p40 induction in macrophages by inhibiting c-rel translocation to the 32 nucleus through activation of calmodulin protein, ” Blood, vol. 107, no. 4, pp. 33 1513-1520. [33] J. Sun, C. J. Krawczyk, and E. J. Pearce, “Suppression of Th2 cell 35 development by Notch ligands Delta1 and Delta4,” Journal of Immunology, 36 vol. 180, no. 3, pp. 1655-1661, 2008. 37
[34] Y. Tada, K. Tayama, and N. Aoki, “Acute oral toxicity of ferulic acid, natural 1 food additive, in rats, ” Ann Rep Tokyo Metr Lab PH, vol. 50, pp. 311-313, 2 1999. 3
[35] Q. Hamid, and M. Tulic, “Immunobiology of asthma,” Annual Review of 4 Physiology, vol. 71, pp. 489-507, 2009. 5
[36] F. D. Finkelman, S. P. Hogan, G. K. Hershey, M. E. Rothenberg, and M. Wills-Karp, “Importance of cytokines in murine allergic airway disease and 7 human asthma,” Journal of Immunology, vol. 184, no. 4, pp. 1663-1674, 2010. 8
[37] M. Kelly, J. M. Hwang, and P. Kubes, “Modulating leukocyte recruitment in inflammation,” The Journal of Allergy and Clinical Immunology, vol. 120, no. 10 1, pp. 3-10, 2007. [38] L. Gallelli, M. T. Busceti, A. Vatrella, R. Maselli, and G. Pelaia, “Update on anticytokine treatment for asthma,” BioMed Research International, vol. 2013, Article ID 104315, 2013. 14
[39] N. I. Kang, H. Y. Yoon, Y. R. Lee et al., “A20 attenuates allergic airway
inflammation in mice,” Journal of Immunology, vol. 183, no. 2, pp. 1488-1495, 16 2009. [40] G. Garcia, C. Taille, P. Laveneziana, A. Bourdin, P. Chanez, and M. Humbert, 18 “Anti-interleukin-5 therapy in severe asthma,” European Respiratory Review, 19 vol. 22, no. 129, pp. 251-257, 2013. 20
[41] P. Gosset, A. Tsicopoulos, B. Wallaert, M. Joseph, A. Capron, and A. B. 21 Tonnel, “Tumor necrosis factor alpha and interleukin-6 production by human 22
mononuclear phagocytes from allergic asthmatics after IgE-dependent 23
stimulation,” The American Review of Respiratory Disease, vol. 146, no. 3, pp. 24 768-774, 1992. 25
[42] C. M. Lilly, H. Nakamura, H. Kesselman et al., “Expression of eotaxin by 26 human lung epithelial cells: induction by cytokines and inhibition by 27
1767-1773, 1997.
[43] D. A. Kuperman, X. Huang, L. L. Koth et al., “Direct effects of interleukin-13 30 on epithelial cells cause airway hyperreactivity and mucus overproduction in 31
asthma,” Nature Medicine, vol. 8, no. 8, pp. 885-889, 2002. 32
[44] J. Corren, “Role of interleukin-13 in asthma,” Current Allergy and Asthma 33 Reports, vol. 13, no. 5, pp. 415-420, 2013. 34
[45] P. L. Vieira, J. R. Christensen, S. Minaee et al., “IL-10-secreting regulatory T 35 cells do not express Foxp3 but have comparable regulatory function to 36 naturally occurring CD4+CD25+ regulatory T cells, ” J Immunol, vol. 172, no. 37 10, pp. 5986-5993, 2004.
[46] F. J. Barrat, D. J. Cua, A. Boonstra et al., “In vitro generation of interleukin 1 -producing regulatory CD4(+) T cells is induced by immunosuppressive 2 drugs and inhibited by T helper type 1 (Th1)- and Th2-inducing cytokines,” J 3 Exp Med, vol. 195, no. 5, pp. 603-616
[47] I. Iwamoto, H. Nakajima, H. Endo, and S. Yoshida, “Interferon gamma 5 regulates antigen-induced eosinophil recruitment into the mouse airways by 6 inhibiting the infiltration of CD4+ T cells,” The Journal of Experimental 7 Medicine, vol. 177, no. 2, pp. 573-576, 1993.
[48] X. M. Li, R. K. Chopra, T. Y. Chou, B. H. Schofield, M. Wills-Karp, and S. K. 9 Huang, “Mucosal IFN-gamma gene transfer inhibits pulmonary allergic 10 responses in mice,” Journal of Immunology, vol. 157, no. 8, pp. 3216-3219, 11 1996.
[49] K. Nakagome, K. Okunishi, M. Imamura et al., “IFN-gamma attenuates antigen-induced overall immune response in the airway as a Th1-type immune regulatory cytokine,” Journal of Immunology, vol. 183, no. 1, pp. 209-220, 15 2009. [50] C. Mitchell, K. Provost, N. Niu, R. Homer, and L. Cohn, “IFN-γ acts on the airway epithelium to inhibit local and systemic pathology in allergic airway 18 disease,” Journal of Immunology, vol. 187, no.7, pp. 3815-3820, 2011. 19
[51] N. G. Papadopoulos, L. A. Stanciu, A. Papi, S. T. Holgate, and S. L. Johnston, 20 “A defective type 1 response to rhinovirus in atopic asthma,” Thorax, vol. 57, 21 no. 4, pp. 328-332, 2002. [52] G. D. Brooks, K. A. Buchta, C. A .Swenson, J. E. Gern, and W. W. Busse, 23 “Rhinovirus-induced interferon-gamma and airway responsiveness in asthma,” 24 American Journal of Respiratory and Critical Care Medicine, vol. 168, no. 9, 25 pp. 1091-1094, 2003.
Figure 1: Ferulic acid modulates cytokine secretion in 3 lipopolysaccharide-stimulated dendritic cells. On day 6 of culture, bone 4 marrow-derived dendritic cells (BMDCs) were collected and treated with RPMI 5 medium 1640, lipopolysaccharide (LPS, 1 μg/ml) alone, or LPS plus various 6 concentrations of FA (20, 100, and 400 μM) for 24 h. Culture supernatants were 7 collected, and cytokine profiles of DCs were
measured by an ELISA. Results from 8 three independent experiments are shown, and results are expressed as the mean ± 9 SEM. * p < 0.05, ** p < 0.01, *** p < 0.001 vs. LPS-treated DCs.
Figure 2: Effects of ferulic acid on expressions of Notch ligands in 3
lipopolysaccharide-stimulated dendritic cells. (A) On day 6 of culture, BMDCs 4 were left untreated or treated with LPS (1 μg/ml) alone or LPS plus various 5
concentrations of FA (20, 100, and 400 μM). Untreated immature DCs served as the 6 medium group. Cells were harvested after 5 h, and Jagged1, Jagged2, Dll1, and Dll4 7 mRNA accumulations were measured using a real-time RT-PCR. Results from three 8 independent experiments are shown, and results are expressed as the mean ± SEM. ## 9 p < 0.01, ### p < 0.001 vs. medium-treated DCs. * p < 0.05, ** p < 0.01, and *** p < 10 0.001 vs. LPS-treated DCs. (B) Medium-treated DCs, LPS-treated DCs, and FA 11 (400 μM) plus LPS-treated DCs were cultured for 24 h. Then cells were harvested, and surface Notch ligand expressions were analyzed by flow cytometry. Each 1 histogram shown is from one representative experiment performed. Values shown in the flow cytometric profiles are the mean fluorescence intensity (MFI) indexes. Cells 3 were gated on CD11c, and the incidence of CD11c+ cells expressing the Notch ligands is indicated within each histogram. (C) The quantification of surface Notch 5 ligand expressions is analyzed by flow cytometry. Results from three independent 6 experiments are shown and are expressed as the mean ± SEM. ## p < 0.01 vs. 7 medium-treated DCs.* p < 0.05 vs. LPS-treated DCs. 8
Figure 3: Ferulic acid treatment affects the maturation phenotype of 3
lipopolysaccharide-stimulated dendritic cells. DCs were treated with medium, LPS 4 (1 μg/ml) alone, or LPS plus FA (400 μM) for 24 h. After incubation, cells were 5 collected, and expressions of MHC class II, CD40, CD80, and CD86 by DCs were 6 analyzed using flow cytometry. (A) Values shown in the flow cytometric profiles are 7 the mean fluorescence intensity (MFI). Cells were gated on CD11c, and the incidence 8 of CD11c+ cells expressing the surface marker is indicated within each histogram. 9 Values shown are from one representative experiment of three independent 10 experiments performed. (B) The MFI was calculated, and results are expressed as the 11 mean ± SEM from three independent experiments. ## p < 0.01, ### p < 0.001 vs. 12 medium-treated DCs.* p < 0.05, *** p < 0.001 vs. LPS-treated DCs.
Figure 4: Ferulic acid-modulated dendritic cells enhanced T cell proliferation 3 and Th1 cell polarization. (A) Modulatory effects of FA on T cell proliferation. DCs 4 were treated with medium, LPS (1 μg/ml), or LPS plus FA (400 μM) for 24 h. 5 Allogeneic naïve CD4+ T cells (2 × 105 cells/well) were co-cultured with different 6 proportions of γ-irradiated immature DCs, LPS-pulsed DCs, or LPS plus FA-treated 7 DCs at DC/T-cell ratios of 1:5, 1:10, 1:20, and 1:40 in 96-well round-bottomed plates. 8 After 2
days of culture, cells were pulsed with 1 μCi/well of [3H]-thymidine for
16~18 h. Specific incorporation of [3H]-thymidine was determined with a β-counter, 1 and results are expressed as counts per minute (cpm). (B) Modulatory effects of FA 2 on T-cell cytokine secretion. Concentrations of IFN-γ and IL-5 secreted by naïve 3 CD4+ T cells, which were co-cultured with γ-irradiated DCs in 96-well plates for 2 4 days, were determined by an ELISA. Results in all panels are expressed as the mean 5 ± SEM of three independent experiments. # p < 0.05, ## p < 0.01, ### p < 0.001 vs. 6 medium-treated DCs. * p < 0.05, ** p < 0.01, *** p < 0.001 vs. LPS-treated DCs. Figure 5: Anti-OVA antibody expression levels of ferulic acid-treated 3 OVA-induced allergic asthmatic mice. (A) Brief description of the protocol of 4 animal sensitization and challenge. On days 1, 14, and 28, all groups of mice were 5 sensitized by an intraperitoneal injection of the OVA allergen. Three groups of mice 6 ((FA (25), FA (50), and FA (100)) were respectively orally fed 25, 50, and 100 mg/kg 7 of FA on days 21 to 47. Positive control (PC) mice were administered sterilized water 8 instead of FA. Then, mice were intranasally challenged with OVA on days 43 and 44. 9
Subsequently, mice were exposed to OVA aerosols for 3 consecutive days, and AHR was measured 1 day after the last challenge. BALF was collected after measuring the 1 AHR. Negative control (NC) mice were neither sensitized with OVA nor 2
administered FA treatment but were challenged with OVA. (B) IgE, IgG1, and IgG2a 3 anti-OVA antibody expressions of all groups of mice were measured by an ELISA. 4 Results are expressed as the mean ± SEM (n = 8 in each group). * p < 0.05, ** p < 5 0.01, *** p < 0.001 vs. the PC group.
Figure 6: Effects of ferulic acid on the airway hyperresponsiveness and 3 airway inflammation. (A) Oral administration of FA suppressed the development 4 of AHR in OVA-sensitized mice. One day after the last OVA challenge, airway 5 resistance was measured in response to increasing concentrations of methacholine 6 (0~32 mg/ml) by invasive body plethysmography. Results are expressed as the mean 7 ± SEM (n = 8 in each group) of the pulmonary resistance (RL) after PBS nebulization. 8 (B) Changes in the cellular composition of the BALF of mice exposed to an 9 aerosolized allergen. One day after measuring the pulmonary function parameters, 10 each group of mice was sacrificed, and the BALF was collected. Cells were counted 11 and classified as macrophages (Macro), eosinophils (Eos), neutrophils (Neu), and 12 lymphocytes (Lym). Results are expressed as the mean ± SEM (n = 8 in each group).
(C) Effect of FA on lung inflammatory cell infiltration in OVA-induced allergicasthmatic mice (H&E staining). Lung sections were obtained from NC mice, sterilized water-treated PC mice, and those water-treated with different doses of FA. Sections were stained with H&E for the morphological analysis. Tissues wereexamined by light microscopy (original magnification: x200). * p < 0.05, ** p < 0.01, 5 *** p < 0.001 vs. the PC
group.
Figure 7: Effect of ferulic acid on levels of chemokines and cytokines in the
bronchoalveolar lavage fluid. Immediately after measuring the pulmonary function parameters, each group of mice was sacrificed, and BALF was collected and analyzed for eotaxin, IL-4, IL-5, IL-13, IFN-γ, and TNF-α contents by an ELISA. Results are expressed as the mean ± SEM (n = 8 in each group). * p < 0.05, ** p < 0.01, *** p < 7 0.001 vs. the PC group.
Figure 8: Effect of ferulic acid on the cytokine profile of OVA-restimulated 3
splenocytes. Splenocytes (107 cells/ml) from FA-treated and control groups were 4 stimulated with 50 μg/ml OVA in 24-well plates, and culture supernatants were 5 collected after 72 h. Levels of cytokine production of IL-4, IL-5, IL-13, IFN-γ, 6 TNF-α, IL-1β, and IL-6 were analyzed by an ELISA. Results are expressed as the 7 mean ± SEM (n = 8 in each group). * p < 0.05, ** p < 0.01, *** p < 0.001 vs. the PC 8 group. 9
