Balb/c mice were purchased from Guangdong Medical Laboratory Animal Center (Guangzhou, China). All mice were maintained at Shenzhen University under specific pathogen-free conditions. All experiments were conducted using age- and sex-matched 6- to 8-week-old male and female mice. The experimental protocols were reviewed and approved by the animal Care and Use Committee in Shenzhen University and were in accordance with the guidelines and regulations of the institution.

Mouse asthma model was constructed as previously described and also illustrated in Figure 1A (31). Briefly, mice were sensitized by intraperitoneal injection with 50 µg of house dust mite extract (HDM, Greer Laboratories, Lenoir, NC) supplemented with 2 mg of adjuvant aluminum hydroxide (Al(OH)₃) (Thermo Fisher Scientific, Cleveland, OH, USA) in 200 µl PBS at day 1, 3 and 7. During the sensitization, mice were also administered intranasally every day with or without 10 µg of vanadium pentoxide (V₂O₅, the most common form of vanadium; Sigma-Aldrich, Inc., St. Louis, Mo, USA) per mouse in 20 µl of PBS. The mice were challenged intranasally with 50 µg of HDM per mouse in 20 µl of PBS from day 15 to 21 with or without 10 µg of V₂O₅ per mouse. All mice were sacrificed 24 h after the last challenge. In some cases, the mice were pre-treated with 100 µg/kg cholecalciferol (vitamin D₃, Sigma-Aldrich, St. Louis, MO) dissolved in coconut oil (Sigma-Aldrich, St. Louis, MO) via oral administration 1 h before every single challenge. PBS-treated age- and gender-matched mice were taken as control.

Broncheoalveolar lavage fluids (BALFs) was collected by instillation of 0.8 ml of PBS through a tracheal cannula. BALF samples were centrifuged at 1500 rpm for 5 min at 4°C. Supernatants were collected and stored for cytokine measurements. The number of total cells, eosinophils, neutrophils, lymphocytes and macrophages was determined by flow cytometry as previously described (32). Briefly, cells were blocked with CD16/32 antibody for 15 min, then incubated for 30 min with antibodies as following: anti-Siglet F-PE (S17007L, BioLegend), anti-Mac-3- FITC (M3/84, BioLegend), anti-Gr-1- APC (RB6-8C5, BioLegend), anti-CD3ϵ PerCP-Cyanine5.5 (145-2C11, BioLegend). Cells were analyzed on a FACSCalibur cytometer (BD Biosystems). SSC^highSiglec-F⁺Mac-3^- and SSC^highSiglec-F⁺Mac-3⁺ cells were identified as eosinophils and alveolar macrophages, respectively. FSC^highSSC^high Gr-1⁺cells were recognized as neutrophils. FSC^lowSSC^lowCD3⁺ cells were identified as lymphocytes.

Lung tissues were fixed with 4% of formalin for 24 h. After dehydration, lung tissues were embedded in paraffin, and then cut into 4 µm sections. To evaluate lung inflammation, sections were deparaffinized and dehydrated, and then subjected to H&E and periodic acid–Schiff (PAS) staining. Masson trichrome staining was performed with a kit (Electron Microscopy Sciences, Halfield, PA, USA) under the manufacturer’s instructions.

Concentrations of cytokines (IL-4, IL-5, IL-13, IL-17A, IFN-ƴ, TSLP, IL-25 and IL-33) in cell-free BAL were measured by enzyme-linked immunosorbent assay (ELISA) under the manual instructions of ELISA kits (4A Biotech Co., Ltd., Beijing, China) and (Thermo Fisher Scientific, Cleveland, OH, USA). Serum HDM-specific IgE, IgG1 and IgG2a (sIgE, sIgG1 and sIgG2a) were detected by ELISA as previously described (13). Serum was diluted at 1:4 for IgE, 1:800 for IgG1, and 1:40 for IgG2a.

Human bronchial epithelial cells (HBECs) were cultured in medium which contains a 1:1 mixture of Dulbecco’s modified Eagle’s medium and Ham’s F-12 (DMEM/F-12; Gibco, Invitrogen corporation, Beijing, China) supplemented with 10% fetal bovine serum (FBS) (Gibco, Invitrogen corporation, Beijing, China) and 1% penicillin-streptomycin (Gibco, Invitrogen corporation, Beijing, China). The cells were maintained at 37°C with a humidified atmosphere containing 5% CO₂.

HBE cells were seeded into 96-well plate (5x10³/well) and cultured overnight. Then cell viability was measured using Cell Counting Kit-8 (CCK-8; Kumamoto, Japan) according to the protocol. Briefly, cells were treated with different doses of V₂O₅ with/without HDM (50 µg/ml) for 24 h. Discard the medium and wash the cells with PBS twice, and then add CCK-8 work solution (diluted with medium, 100 µl/well) to each well of the plate. After incubation at 37°C for 2 h, absorbance at 450 nm was measured with a microplate reader (Thermo Fisher Scientific, Cleveland, OH, USA).

ROS in the lung tissues and HBECs were measured by using flow cytometry and immunostaining, respectively. For ROS in lung tissues, lung tissues were minced and then ground slightly. Single-cell suspensions were obtained after filtration with 40 µm of cell strainers (Millipore). Single cells were stained with 5 µM of CM-H₂DCFDA (Thermo Fisher Scientific) at 37°C for 30 min, and then the ROS generation was measured by flow cytometry. For detection of intracellular superoxide in lung sections, frozen sections were incubated with 5 µM of dihydroethidium (DHE, Thermo Fisher Scientific) 37°C for 30 min, and then fluorescent signal was observed under microscope (Nikon, Chiyoda, Japan). For ROS in HBECs, these treated HBECs were incubated with 5 µM of CM-H₂DCFDA or 5 µM of mitoSOX (Thermo Fisher Scientific) at 37°C for 20 min. Total and mitochondrial ROS production were analyzed under microscope.

After de-paraffin, dehydration and antigen retrieval, the slides were blocked with blocking buffer at room temperature for 1h. Then tissues were incubated with primary antibodies against α-SMA (Sigma-Aldrich, St. Louis, MO) or vimentin (Cell Signaling Technology, Beverly, MA, USA) at 4°C overnight. After three washes with TBST, secondary antibody incubation was followed at room temperature for 1h. The nuclei were counterstained with 4′,6-Diamidino-2-phenylindole dihydrochloride at room temperature for 10 minutes. The sections were mounted with fluoromount aqueous mounting medium (Sigma-Aldrich, St. Louis, MO) and imaged under microscope (Nikon, Japan). To determine the fluorescent signal, four different high-power fields from each slide were quantified with Fiji (National Institutes of Health) and presented as mean fluorescence intensity per square micrometer.

Statistical significance among experimental groups were analyzed by One-way ANOVA with Student-Newman-Keuls test using GraphPad Prism version 5.1 software (GraphPad Software, San Diego, CA, USA). Data were represented as mean ± SEM. A p-value <0.05 were considered statistically significant for all analyses.

To explore whether vanadium co-exposure can aggravate allergic lung inflammation, we constructed a mouse model of asthma with either HDM or V₂O₅ or co-exposure following the protocol as illustrated in Figure 1A . Compared to HDM alone, histological analysis demonstrated that mice treated with both HDM and V₂O₅ co-exposure showed increased airway inflammation as characterized by dense peri-bronchial inflammatory cell infiltrates (H&E, upper panel) and goblet cell hyperplasia (PAS, lower panel) ( Figure 1B ). The increased airway inflammation caused by V₂O₅ co-exposure, but not V₂O₅ alone, was further supported by the increased inflammatory cells in BAL samples, including the number of total cells, eosinophils, neutrophils, lymphocytes, and macrophages ( Figure 1C ). Furthermore, serum levels of house dust mite specific IgE (sIgE), sIgG1, and sIgG2a were also increased in V₂O₅ co-exposure treated mice as compared to those treated with HDM alone ( Figure 1D ). The results suggest that V₂O₅ co-exposure can aggravate HDM-induced airway inflammation.

Next, we investigated whether vanadium co-exposure can aggravate HDM-induced cytokine release. As compared to HDM alone, V₂O₅ co-exposure showed significantly increased levels of IL-4, IL-13, and IFNγ in BALFs. No significant increase was noted for IL-5 and IL-17 ( Figure 2A ). Furthermore, we specifically investigated whether V₂O₅ co-exposure can aggravate HDM-induced the release of epithelial-derived cytokines. As expected, higher levels of IL-25 and TSLP were observed in BAL samples of V₂O₅ co-exposure-treated mice as compared to those treated with HDM alone ( Figure 2B ). While there is no difference in the level of IL-33 ( Figure 2B ). To further confirm whether V₂O₅ co-exposure can aggravate HDM-induced epithelial-derived cytokine release, we performed in vitro analyses by treating HBECs with either V₂O₅ or HDM or in combination. Cell viability was firstly examined by CCK-8 assay in HBECs treated with HDM (50 μg/ml) or HDM with different doses of V₂O₅ (0.125, 0.25, 0.5 and 1 µg/ml) for 24 h. As shown in Figure 2C , no significant effect on cell viability was noted for HDM-treated HBECs and in HBECs treated with HDM (50 μg/ml) and V₂O₅ at a dose from 0.125 to 0.5 µg/ml. A modest reduction in cell viability was observed for V₂O₅ co-exposure at a dose of 1 µg/ml. HBECs were then treated with either HDM (50 μg/ml) or V₂O₅ (0.5 µg/ml) for 24 h and pro-inflammatory cytokines and chemokines were measured in supernatants by ELISA. Compared to HDM, V₂O₅ co-exposure led to increased levels of IL-25, TSLP, IL-8, CCL2 and CCL20 in supernatants of HBECs ( Figure 2D ). Taken together, both in vivo and in vitro analyses demonstrate that vanadium co-exposure can aggravate cytokine release, particularly those epithelial-derived cytokines and chemokines.

Studies have suggested that V₂O₅ exposure at 200 µg/animal can lead to the development of airway fibrosis and remodeling in a rat model (19). Thus, we examined whether V₂O₅ exposure can aggravate the HDM-induced airway fibrosis and remodeling. Masson trichrome staining was first performed in the lung tissues. Mice with V₂O₅ (10µg/mouse) co-exposure showed dramatic increase in airway thickening and collagen deposition/fibrosis as compared to those treated with either HDM or V₂O₅ alone ( Figures 3A, B ). The increased airway fibrosis was further confirmed by means of immunofluorescence staining with α-SMA and vimentin. Mice with V₂O₅ co-exposure showed a significant increase in the expression of both α-SMA ( Figures 3C, D ) and vimentin ( Figures 3E, F ) in the airways. Taken together, these data suggest that V₂O₅ co-exposure can aggravate HDM-induced airway fibrosis/remodeling.

ROS have been considered as important mediators contributing to oxidative damage and chronic inflammation in allergic diseases (32, 33). To explore the underlying mechanism by which V₂O₅ co-exposure aggravates HDM-induced airway inflammation and remodeling, we investigated whether V₂O₅ co-exposure could enhance the HDM-induced ROS generation. Firstly, single cells were prepared from lung tissues of asthma mouse model at the end of the treatment/challenge protocol and the relative level of ROS production was evaluated by flow cytometry with CM-H₂DCFDA. Compared to control, significantly higher level of ROS was found in the HDM group, which was further augmented by V₂O₅ co-exposure ( Figure 4A ). The enhanced ROS generation by V₂O₅ co-exposure was further supported by the in vivo analyses of ROS production in the lung tissues by dihydrothidium (DHE) immunostaining ( Figure 4B ). Lung tissues from V₂O₅ co-exposure treated mice had a significant increase in ROS levels as compared to those from HDM-treated mice. These findings were further confirmed by in vitro analysis in HBECs. Compared to untreated HBECs, treatment with either HDM or V₂O₅ at different doses induced ROS generation as determined by immunostaining with CM-H₂DCFDA ( Figure 4C ). Notably, the increased ROS were further potentiated by V₂O₅ co-exposure with HDM. The same pattern was also observed for mitochondrial ROS, one of the major sources of ROS (33), in those treated and un-treated HBECs ( Figure 4D ). Compared to HDM or V₂O₅ treatment, V₂O₅ co-exposure induced increased production of mitochondrial ROS as assessed by immunostaining with MitoSOX, a fluorescent mitochondrial ROS reporter dye. Collectively, our findings support the notion that V₂O₅ co-exposure potentiates HDM-induced ROS generation, particularly mitochondrial ROS production.

To examine the role of ROS in V₂O₅ co-exposure-induced airway inflammation, we specifically investigated whether ROS play a role in V₂O₅ co-exposure induced cytokine release from epithelial cells. HBECs were pre-treated with NAC, a widely used ROS inhibitor, and then treated with either V₂O₅ or HDM or co-exposure. Supernatants were harvested after 24 h and cytokines and chemokines in supernatants were measured. Similar to the previous findings that V₂O₅ co-exposure significantly upregulated HDM-induced IL-25, TSLP, IL-8, CCL2, and CCL20 production ( Figure 5 ). In contrast, pre-treatment with NAC almost completely blocked the V₂O₅ co-exposure-induced significant increase in those cytokines and chemokines. The same pattern was also seen when HBECs were pre-treated with mitoTEMPO, a mitochondria-targeted antioxidant (34). These results implicate that vanadium co-exposure induces epithelial cytokine and chemokine release through ROS generation.

Vitamin D₃ (VD₃) supplementation has been shown to protect against oxidative stress processes in the OVA-induced airway inflammation (35). Thus, we hypothesized that VD₃ could inhibit the V₂O₅ co-exposure-induced ROS generation and subsequent airway inflammation. Vitamin D₃ (VD₃, 100 μg/kg) was used to treat mice before every single challenge followed the protocol as illustrated in Figure 6A . To confirm whether pre-treatment with VD₃ as an antioxidant can inhibit V₂O₅ co-exposure-induced ROS generation, we detected ROS generation in single cell suspension from lung tissues of asthma mouse model by flow cytometry analysis ( Figure 6B ) and in lung tissues by immunofluorescence staining ( Figure 6C ). As expected, both flow cytometry analysis and immunofluorescence staining consistently showed increased ROS generation in V₂O₅ co-exposured mice but its level was significantly reduced when those mice were pre-treated with VD₃. H&E and PAS staining demonstrated that VD₃ treatment reduced V₂O₅ co-exposure induced lung inflammation and mucus production ( Figure 6D ). Significant reductions were noted for the numbers of total inflammatory cells and eosinophils in BAL samples ( Figure 6E ). The same pattern was also noted for serum levels of HDM sIgE and sIgG2a ( Figure 6F ). Similarly, mice pretreated with VD₃ showed clear reduction for IL-4 in BAL samples as compared to V₂O₅ co-exposure-treated mice ( Figure 6G ). In contrast, significant reductions were observed for TSLP, and IL-25 in BAL samples of VD₃-pre-treated mice compared with V₂O₅ co-exposure-treated mice ( Figure 6H ). These preliminary findings suggest that VD₃ can, at least partially, alleviate vanadium co-exposure-induced cytokine release and airway inflammation.

Next, we investigated whether pre-treatment with VD₃ could inhibit V₂O₅ co-exposure-induced airway fibrosis/remodeling. Masson trichrome staining was performed in the lung tissues. Mice pretreated with VD₃ showed a significant inhibition of airway thickening and collagen deposition/fibrosis caused by V₂O₅ co-exposure ( Figures 7A, B ). The reduced airway fibrosis was further confirmed by immunofluorescence staining with α-SMA and vimentin. Expression of both α-SMA ( Figures 7C, D ) and vimentin ( Figures 7E, F ) was reduced in the airways of VD₃ pre-treated mice as compared to those treated with V₂O₅ co-exposure. Collectively, these data suggest that VD can inhibit vanadium co-exposure-induced airway fibrosis/remodeling.