Female BALB/c mice (6–8 weeks old) were obtained from the National Laboratory Animal Center in Taiwan. All mice were housed in air-conventional animal housing with food and water ad libitum. Animal care and experimental procedures were performed in accordance with the guidelines of the Laboratory Animal Care Committee of Chang Gung University of Science and Technology (IACUC approval number: 2014-023).

Phloretin (extracted from apple wood; ≥99% purity by HPLC; purchased from Sigma-Aldrich, St. Louis, MO, USA) was dissolved in DMSO. Mice were sensitized with intraperitoneal injections (200 µl) that contained 50 µg ovalbumin (OVA; Sigma, St. Louis, MO, USA) mixed with 2 mg aluminum hydroxide (Thermo, Rockford, IL, USA) in normal saline. Injections were performed on days 1–3 and 14, as described previously (13). On days 14, 17, 20, 23, and 27, mice were challenged with inhalations of 2% OVA for 30 min, delivered with an ultrasonic nebulizer (DeVilbiss Pulmo-Aide 5650D, USA). One hour before the OVA inhalation challenge, mice were treated with or without intraperitoneal injections of PT. Then, the OVA challenge was followed by an AHR assay. Mice were randomly divided into five groups (12 mice each), as follows: healthy mice sensitized with normal saline and were given equal volume of DMSO by intraperitoneal injection (N group) (2); mice sensitized with OVA and were given equal volume of DMSO by intraperitoneal injection (OVA group); OVA-sensitized mice treated with 5, 10, or 20 mg/kg PT (PT5, PT10, and PT20 groups, respectively).

Airway hyperresponsiveness was measured to assess airway function after inhaling aerosolized methacholine, as described previously (14). All mice inhaled 0–40 mg/ml methacholine for 3 min; then, mice were placed in a single-chamber, whole-body plethysmograph (Buxco Electronics, Troy, NY, USA) to record the enhanced pause (Penh), a variable for estimating the AHR.

Splenocytes (5 × 10⁶ cells/ml) were isolated and cultured in RPMI 1640 medium containing 10% FBS (Biological Industries, Haemek, Israel), 100 U/ml penicillin and streptomycin, and 100 µg/ml OVA for five continuous days. The supernatants were collected, and cytokine concentrations were measured, as previously described (13).

Blood collected from the orbital vascular plexus and centrifuged at 6,000 rpm for 5 min. The serum collected and stored at −80°C as previously described. Serum was measure OVA-specific antibodies with an enzyme-linked immunosorbent assay (ELISA).

Lung tissues were fixed in 10% formalin, embedded in paraffin, and cut into sections 6 μm thick. Sections were stained with hematoxylin and eosin (HE), to examine eosinophil infiltration Inflammatory index calculated and evaluated score using fivepoint scoring system as described previously (15). The score were considered by the inflammatory cell infiltration on perivascular regions and peribronchial of lungs. Furthermore, tracheal sections were stained with the periodic acid-Schiff (PAS) staining system (Sigma), to measure goblet cell hyperplasia, as described previously (14).

We measured MDA activity in lung tissues with the lipid peroxidation assay kit, according to the manufacturer’s instructions (Sigma). MDA activity was measured as nanomolars of MDA in milligram weight of wet tissue with a Multi-Mode Microplate Reader (BioTek SynergyHT, Bedfordshire, UK).

To detect glutathione levels in lung tissues, we employed a glutathione assay kit, according to the manufacturer’s instructions (Sigma). The kit measured total glutathione, glutathione disulfide, and reduced glutathione. The glutathione reaction was measured spectrophotometrically at 412 nm with a microplate reader (Multiskan FC, Thermo, Waltham, MA, USA). Glutathione expressed as nanomolars of MDA in milligram weight of wet tissue.

Mice were anesthetized and sacrificed to collect BALF, as described previously (16). Mouse tracheas were intubated with an indwelling needle to wash the lungs with normal saline. The supernatants would assay cytokine and chemokine levels. We used Liu stain solution (Polysciences, Inc., Taipei, Taiwan) to differentiate cell morphology and determine cells counts.

Serum OVA-specific antibodies, including IgE and IgG1, were measured with a specific ELISA kit (BD Biosciences). Serum of OVA-sensitized mice makes OVA-IgG1 standard curves for determining the concentrations of IgG1. Serum was diluted fivefold to detect the absorbance of OVA-IgE at an optical density of 450 nm, as previously described (14). Furthermore, the cell culture supernatants and BALF were used to measure with specific ELISA kits to detect the concentrations of CCL5, CCL11, CCL24, CCL26, intercellular adhesion molecule 1 (ICAM-1), IL-4, IL-5, IL-6, IL-8, IL-13, MCP-1, and tumor necrosis factor-α (TNF-α), according to the manufacturer’s instructions (R&D Systems, Minneapolis, MN, USA), as previously described (16).

Lung tissues were homogenized, and RNA was extracted with TRIzol reagent (Life Technologies, Carlsbad, CA, USA); complementary DNA (cDNA) was generated from 1 µg total RNA using cDNA synthesis kit (Life Technologies). cDNA gene expression assayed by real-time PCR performed with the SYBR Green system (Fermentas, Thermo, Waltham, MA, USA) and specific primers (Table 1) for the genes by the spectrofluorometric thermal cycler (iCycler; Bio-Rad Laboratories, Hercules, CA, USA). The average of gene cycle threshold (C_t) was measured for each experiment. Relative cDNA expressions (2−ΔΔCt) for the specific genes were determined by the compared C_t method, which generates ΔΔC_t as the discrepancy between the housekeeping genes β-actin and the specific gene for each sample.

Phloretin was dissolved in DMSO at a concentration of 100 mM to produce a stock solution. After dilution, the DMSO component was ≤0.1% of the experimental culture medium. We seeded immortalized human bronchial epithelial cells (BEAS-2B) into 24-well plates with DMEM/F12 medium. Cell cultures were pretreated with PT (3–30 µM) for 1 h, then treated with 10 ng/ml TNF-α for 24 h, or 10 ng/ml TNF-α and 20 ng/ml IL-4 for 24 h. The supernatants were collected, and the levels of cytokines or chemokines were determined with specific ELISA kits.

BEAS-2B cells were treated with PT and stimulated with TNF-α for 24 h. The human monocytic cell line, THP-1, was cultured and stained with calcein-AM solution (Sigma) for 0.5 h, then cocultured with BEAS-2B cells for 1 h. Adherent cells were observed and evaluated with fluorescence microscopy (Olympus, Tokyo, Japan).

TNF-α-stimulated BEAS-2B cells treated with PT were seeded in 96-well plates for 24 h. Next, cells were stained with 20 µM 2′,7′-dichlorofluorescin diacetate (DCFH-DA) for 30 min. Then, cells were lysed and analyzed with a Multi-Mode Microplate Reader (BioTek synergy HT); fluorescence was evaluated by exciting at 485 nm and measuring emission at 528 nm. Furthermore, intracellular ROS was visualized with a fluorescence microscope (Olympus).

Lung tissue proteins were quantified and separated on 10% SDS polyacrylamide gels. The proteins were transferred to polyvinylidene fluoride (PVDF) membranes (Millipore, Billerica, MA, USA) and incubated with primary antibodies overnight at 4°C. Then, membranes were washed and incubated with secondary antibodies. Finally, PVDF membranes were treated with Luminol/Enhancer Solution (Millipore) to detect antibody signals with the BioSpectrum 600 system (UVP, Upland, CA, USA). Primary antibodies included anti-HO-1, anti-Nrf2, and anti-Lamin B1 (Santa Cruz, CA, USA); β-actin expression was evaluated as a loading control (Sigma).

Data were assessed with one-way analysis of variance, followed by the Tukey–Kramer post hoc test for multiple comparisons. All values represent the mean ± SEM. Values of p < 0.05 were considered significant.

We evaluated whether PT could improve shortness of breath and abnormal airflow in the airways of asthmatic mice. Airway function was evaluated as the AHR, and the Penh value was calculated to determine the AHR. Mice were placed into the single-chamber whole-body plethysmograph to record the Penh, during inhalation of various methacholine doses (0–40 mg/ml). We found that, upon inhaling methacholine, the Penh value increased in a dose-dependent manner. The Penh values were greater in OVA-sensitized mice compared to normal mice (Figure 1A). At 40 mg/ml of inhaled methacholine, PT-treated asthmatic mice showed significantly lower Penh values (PT5, 5.9 ± 0.66, p = 0.32; PT10, 4.41 ± 0.58, p < 0.05; PT20, 3.94 ± 0.48, p < 0.01) compared to asthmatic mice in the OVA-sensitized group (OVA, 6.96 ± 0.78). Hence, PT could significantly diminish AHR in asthmatic mice.

We counted various types of inflammatory cells to assess whether PT decreased the inflammatory response in asthmatic mice by reducing the numbers of inflammatory cells in BALF (Figure 1B). Asthmatic mice treated with PT had significantly reduced numbers of eosinophils and total cells compared to the OVA group (eosinophils: PT5: 6.2 × 10⁵ ± 6.9 × 10⁴, p = 0.57; PT10: 4.3 × 10⁵ ± 9.3 × 10³, p < 0.05; PT20: 3.3 × 10⁵ ± 1.1 × 10⁴, p < 0.01 vs. OVA: 6.4 × 10⁵ ± 8.9 × 10⁴) (total cells: PT5: 1.0 × 10⁶ ± 1.3 × 10⁵, p = 0.84; PT10: 8.4 × 10⁵ ± 1.3 × 10⁴, p = 0.24; PT20: 6.3 × 10⁵ ± 1.2 × 10⁴, p < 0.01 vs. OVA: 1.1 × 10⁶ ± 1.0 × 10⁵). Furthermore, the proportion of eosinophils in BALF derived from the P10 or the P20 group was significantly reduced compared to the proportion in the OVA group (Figure 1C).

The levels of cytokines and chemokines in BALF were determined by ELISA (Figure 2). Our results showed that PT could significantly suppress IL-4 levels compared to OVA-sensitization alone (PT5: 32.4 ± 6.2 pg/ml, p = 0.33; PT10: 24.3 ± 5.2 pg/ml, p < 0.05; PT20: 18.5 ± 4.8 pg/ml, p < 0.05 vs. OVA: 41.3 ± 3.8 pg/ml). In addition, the PT groups had significantly decreased the levels of CCL11, CCL24, TNF-α, IL-6, IL-5, and IL-13 compared to the OVA group. A real-time PCR analysis of the expression of genes in lung tissue showed that PT could significantly decrease the levels of CCL11, CCL24, and ICAM-1 expression compared to the levels in OVA-sensitized asthmatic mice. PT also inhibited IL-4, IL-5, IL-13, MUC5AC, Gob5, iNOS, and COX-2 gene expression. Furthermore, PT increased the expression of IFN-γ compared to the expression in asthmatic mice (Figure 3).

Eosinophil infiltration in the lungs was evaluated with HE staining. Compared to normal mice, OVA-sensitized mice exhibited more infiltrating eosinophils between the bronchus and blood vessels (Figures 4A,B). PT reduced eosinophil infiltration in the lungs of asthmatic mice. We evaluated tracheal goblet cell hyperplasia with PAS staining. We found that PT could inhibit goblet cell hyperplasia compared to untreated, OVA-sensitized asthmatic mice (Figures 4C,D).

Acute asthma attacks can also cause oxidative stress. Previous studies showed that the expression of antioxidant HO-1 could protect and decrease lung damage during oxidative stress (17). We found that the lungs in PT-treated mice had increased HO-1 expression of lung compared to asthmatic mice. Nrf2, is a transcription factor, could translocate into the nucleus to promote HO-1 expression for antioxidant response. PT could increase nuclear Nrf2 expression of lung cells compared to OVA-sensitized asthmatic mice (Figure 5A). We also found that the OVA-sensitized asthmatic mice had significantly increased MDA activity and decreased GSH levels in lung tissues compared to the levels in normal mice (Figures 5B,C). However, PT significantly reduced MDA activity and promoted GSH production in lung tissues, compared to the levels in OVA-sensitized asthmatic mice.

Splenocyte culture supernatant analyses showed that PT significantly attenuated the levels of IL-4, IL-5, and IL-13, compared to untreated OVA-sensitized cells. PT also significantly decreased the levels of OVA-IgE and OVA-IgG1 in the serum of OVA-sensitized asthmatic mice (Figure 6).

Phloretin could decrease IL-6, IL-8, CCL5, and MCP-1 levels in TNF-α-activated BEAS-2B cells. When BEAS-2B cells were stimulated with TNF-α and IL-4, PT also significantly inhibited CCL11, CCL24, and CCL26 production (Figure 7).

Phloretin significantly decreased ICAM-1 expression in TNF-α-activated BEAS-2B cells (Figure 8A). We also evaluated whether PT could inhibit the attachment of THP-1 monocyte cells to inflammatory BEAS-2B cells. TNF-α-stimulated BEAS-2B cells were cocultured with THP-1 cells (stained with calcein AM). PT treatment significantly reduced THP-1 cell adherence to TNF-α-activated BEAS-2B cells (Figures 8B,C).

BEAS-2B cells were stained with DCFH-DA, the cells were lysed, and ROS production was quantified with a Multi-Mode Microplate Reader. We found that PT reduced ROS production in TNF-α-activated BEAS-2B cells (Figure 9A). Furthermore, we examined intracellular ROS in intact cells with a fluorescence microscope. We observed that PT attenuated intracellular ROS expression in TNF-α-activated BEAS-2B cells (Figures 9B,C).