Vitamin D Supplementation Reduces Induction of Epithelial-Mesenchymal Transition in Allergen Sensitized and Challenged Mice

Asthma is a chronic disease of the lung associated with airway hyperresponsiveness (AHR), airway obstruction and airway remodeling. Airway remodeling involves differentiation of airway epithelial cells into myofibroblasts via epithelial-mesenchymal transition (EMT) to intensify the degree of subepithelial fibrosis. EMT involves loss in E-cadherin with an increase in mesenchymal markers, including vimentin and N-cadherin. There is growing evidence that vitamin D has immunomodulatory and anti-inflammatory properties. However, the underlying molecular mechanisms of these effects are still unclear. In this study, we examined the contribution of vitamin D on the AHR, airway inflammation and expression of EMT markers in the airways of mice sensitized and challenged with a combination of clinically relevant allergens, house dust mite, ragweed, and Alternaria (HRA). Female Balb/c mice were fed with vitamin D-sufficient (2000 IU/kg) or vitamin D-supplemented (10,000 IU/kg) diet followed by sensitization with HRA. The density of inflammatory cells in the bronchoalveolar lavage fluid (BALF), lung histology, and expression of EMT markers by immunofluorescence were examined. Vitamin D-supplementation decreased AHR, airway inflammation in the BALF and the features of airway remodeling compared to vitamin D-sufficiency in HRA-sensitized and -challenged mice. This was accompanied with increased expression of E-cadherin and decreased vimentin and N-cadherin expression in the airways. These results indicate that vitamin D may be a beneficial adjunct in the treatment regime in allergic asthma.


Introduction
Chronic inflammation in the airways causes structural and functional changes in the lungs that may result in asthma. Physiological changes in asthma include narrowing of the airways following exposure to allergens or bronchoconstrictors, leading to airway hyperresponsiveness (AHR). Structural changes feature the characteristics of airway remodeling including epithelial cell shedding, goblet cell hyperplasia/metaplasia, subepithelial fibrosis, smooth muscle cell hyperplasia, edema, and angiogenesis [1]. Current therapies, such as corticosteroids, leukotriene antagonists, and long-acting β2 agonists, may be effective in dampening inflammation, but are ineffective in preventing or reversing airway remodeling [2], [3]. Additionally, therapies such as corticosteroids may induce apoptosis in epithelial cells, thereby contributing to epithelial shedding [4]. Thus, additional consideration must be taken to understand the cause of airway remodeling in order to acquire therapies that target molecules involved in structural alterations, including subepithelial fibrosis and epithelial thickening. The process of airway remodeling involves the release of inflammatory mediators from immune cells in the airway, including transforming growth factor (TGF)-β, tumor necrosis factor (TNF)-α, interleukin (IL)-4, and IL-13 with the disruption of the epithelium and subepithelial fibrosis [5].
Initiation of subepithelial fibrosis in the airway epithelial cell can occur through activation of pathways resulting into epithelial mesenchymal transition (EMT). Activation of EMT signaling causes the cells to differentiate into myofibroblasts, enabling invasion and migration within the epithelial layer. Increased myofibroblasts in the sub-mucosa secrete collagen and extracellular matrix, thereby contributing to the subepithelial fibrosis in airway remodeling [6].
A number of studies have examined the relationship between vitamin D and asthma with a growing body of evidence suggesting that vitamin D deficiency is linked with the pathogenesis of asthma [7]. Treatment of asthma though oral supplementation of vitamin D has been not well studied and has given mixed results in terms of patient outcomes, as recently reviewed by Yawn and colleagues [8]. Further understanding of how vitamin D functions are necessary to provide therapeutic guidelines for the use of vitamin D to control the pathogenesis of asthma.
Mouse models of allergic airway inflammation and AHR are generally used to determine the cellular and molecular mechanisms. Most often, ovalbumin (OVA) is used as an allergen. However, sometimes it is difficult to extrapolate the findings to human unless the studies are performed with clinically-relevant allergens [9]. Such allergens have been used sparingly in animal models of allergic airway inflammation. House dust mite [10], ragweed [11], and Alternaria alternata extracts [12] have been used individually in mouse models, but never in combination with one another. This study utilizes the combination of these allergens in a mouse model of asthma while evaluating the induction of EMT protein markers in the lung.

Animals and Diets
Female and male BALB/c mice were purchased from Harlan Laboratories (Indianapolis, IN). Mice were maintained in a specific pathogen-free environment at the Animal Resource Facility of Creighton University and monitored daily. The Institutional Animal Care and Use Committee of Creighton University approved the research protocol of this study. Food and water were provided ad libitum. Vitamin D supplemented feed and normal mouse feed containing sufficient vitamin D levels were given to breeding male and female mice. Female offspring were weaned on the same diets consisting of normal vitamin D levels of 2,000 IU/kg and vitamin Dsupplemented diet levels of 10,000 IU/kg (Harlan Laboratories, Madison, WI). No animals during the experimental procedure became ill or died prior to the experimental endpoint but were monitored before and during the experiment. Mice were monitored for distress, through the lack of movement, difficulty in breathing, and rolling on the back. If the event of any distress, the animal would be taken out of the cage and examined. If the distress were found to be due to a respiratory problem, albuterol would be administered by aerosol. In this study, none of the animals were found to be under distress at any time.

Induction of Allergic Airway Disease
Mice 3-4 weeks in age were divided into sensitized and non-sensitized groups. HRA sensitized and challenged model of allergic airway inflammation was established by intranasal installation of 200 allergic units of HDM, 50 μg of Alternaria alternata extract, and 100 μg of ragweed allergen extract (Jubilant HollisterStier Allergy, Spokane, WA). Animals received aerosol challenges of mixed antigen containing 5% HDM, 0.1% Alternaria, and 0.5% ragweed (HRA) 3 days a week for 3 weeks from days 14-30 ( Fig 1A). On day 31, AHR to methacholine was measured by whole body plethysmography (Buxco FinePointe Data Sciences International St. Paul, MN.). On day 32, mice anesthetized using sodium pentobarbital were tracheostomized and placed in a plethysmograph single chamber (Buxco FinePointe Series Resistance and Compliance (RC) Buxco Electronics, Troy, NY). Specific lung specific resistance (R L ) was measured with increasing doses of nebulized methacholine to measure AHR as previously described [13].

Bronchoalveolar Lavage Fluid and Cytokine Measurement
Immediately after euthanization of mice using 100 mg/kg dose of sodium pentobarbital i.p., samples of blood and bronchoalveolar lavage fluid (BALF) were collected from the mice. Blood was collected from the left ventricle of mice, allowed to coagulate, serum separated, and stored at -80°C for future analysis. Mouse lungs were then gently lavaged with 1 ml of PBS, centrifuged at 400 × g for 10 minutes, and the supernatants stored at −80°C. Total cell counts were performed using the Countess Automated Cell Counter (Invitrogen, Grand Island, NY). Cytospin slides were obtained using a Shandon Cytospin 4 centrifuge machine (Thermo Electron, Waltham, MA). The slides were dried and the cells were fixed and stained using Diff-Quik staining reagent (StatLab Medical Products, Lewisville, TX) according to the manufacturer's instructions. Randomly, a minimum of 300 cells were examined per cytospin slide and absolute cell numbers for each cell type were calculated according to the total leukocyte counts in BALF and the percentage of each individual cell type on the slide. ELISA measured the IgE levels in both serum and BALF according to the manufacturer's instructions using mouse IgE ELISA Detection Ready-Set-Go kit (eBioscience, San Diego, CA).

Preparation and Staining of Lung Tissue
Lung lobes were fixed in 4% formalin and paraffin embedded. Lung lobes were removed, fixed in 4% formalin and embedded in paraffin in an automatic tissue processer. Lung tissues were sectioned to 4 μm thickness, rehydrated, and stained with hematoxylin and eosin (H&E) following manufacturer's instructions (Fisher HealthCare PROTOCOL, Waltham, MA). Mucus secretion was identified by periodic acid-Schiff (PAS) reaction (Sigma-Aldrich, St. Louis, MO). Trichrome staining (Thermo Scientific Richard-Allan Scientific, Waltham, MA) was used to identify collagen deposition. Morphometric analysis of goblet cell was performed by counting cells in PAS-stained sections and expressed as the mean number of goblet cells per 100 μm of basal lamina measured by ImageJ image analysis software (National Institutes of Health) [14]. Fibrosis was also quantified by image analysis using ImageJ software. Briefly, the peribronchial fibrosis was traced, converted to grayscale, threshold was adjusted, and the perimeter of the airway was measured. Subepithelial fibrosis is thus expressed as area of fibrosis (μm 2 ) per micrometer of basement membrane [15].  Animals received aerosol challenges of 5% HDM, 0.1% Alternaria, and 0.5% ragweed 3 days a week for 3 weeks from days 14-30. On day 31, airway hyperresponsiveness (AHR) to methacholine was measured by whole body plethysmography (WBP). On day 32, specific lung resistance (R L ) was measured by tracheostomy; (B) Graph showing the results of enhanced pause (Penh), a measurement of AHR by WBP. The results are expressed as a percentage of baseline Penh value (baseline is defined as 1) and presented as mean ± SEM of 7 mice per group. Vitamin D-sufficient PBS-sensitized and challenged mice vs Vitamin D-sufficient HRA-sensitized and challenged mice (*p <0.05, ***p <0.001); Vitamin D sufficient HRA-sensitized and challenged mice vs. Vitamin D-supplemented HRA-sensitized and challenged mice (## p <0.01); (C) Five randomly selected animals from each group were subjected to a methacholine dose response using invasive tracheostomy and R L was recorded. The results are expressed as a percentage of baseline R L value (baseline is defined as 1) and presented as mean ± SEM of 5 mice per group. Vitamin D-sufficient PBS control mice vs Vitamin D-sufficient HRA-sensitized and -challenged mice (*p <0.05, **p <0.01, ***p < 0.001); Vitamin D-sufficient HRA-sensitized and -challenged mice vs Vitamin D-supplemented HRA-sensitized and -challenged mice (#p <0.05). Laboratories, Burlingame, CA, 1/500 dilution] and nuclei were counterstained with DAPI. Fluorescent analysis was performed using an Olympus inverted fluorescence microscope (Olympus BX51) and mean fluorescent intensity was measured using ImageJ software (http:// rsbweb.nih.gov/ij/). Intensity measurements were performed by analyzing the region of interest specifically to the airway and setting a threshold according to guidelines previously set [16]. The mean fluorescent intensity from the airway was analyzed using the equation corrected total fluorescence (CTF) = integrated density-(area of region of interest x mean fluorescence of background readings) [17]. Negative controls of tissue sections were run without incubation with the primary antibody.

Statistical Analysis
Values of all measurements are reported as mean ± SEM. Multiple group comparison was performed using one-way analysis of variance with a Tukey post-hoc test. GraphPad Prism v6.0 was used to analyze data with a p value of <0.05 considered significant.

Effect of Vitamin D on AHR and Inflammatory Cells in HRA-sensitized and Challenged Mice
Following the protocol shown in Fig 1A, HRA-sensitized and challenged mice demonstrated AHR to methacholine as determined by noninvasive whole-body plethysmography (WBP) ( Fig 1B) and invasively ( Fig 1C) by measuring specific airway resistance via tracheostomy. Vitamin D-sufficient HRA-sensitized and challenged mice demonstrated significantly elevated AHR to methacholine compared to PBS control mice on a vitamin D-sufficient diet as demonstrated via WBP ( Fig 1B) and R L (Fig 1C) for doses 25-100 mg/ml and 12.5-100 mg/ml respectively. HRA-sensitized mice fed with a vitamin D-supplemented diet exhibited significant reduction in AHR compared to vitamin D-sufficient HRA-sensitized mice at 25 mg/ml ( Fig  1B) and 100 mg/ml of methacholine ( Fig 1C). There was no significant difference in AHR among PBS control vitamin D-sufficient, and PBS control vitamin D-supplemented groups. Aerosolized administration of 100 mg/ml methacholine at day 31 had the following Penh values: 6.223 ± 0.6265 in vitamin D-sufficient HRA-sensitized and challenged mice; 3.172 ± 0.6265 in vitamin D-sufficient PBS-sensitized and challenged mice; 4.75 ± 0.5951 in vitamin D-supplemented HRA-sensitized and challenged mice; 5.481 ± 0.4100 in vitamin Dsupplemented PBS-sensitized and challenged mice (n = 7 in each group) (Fig 1B).  (Fig 1C).
Vitamin D-sufficient HRA-sensitized and challenged mice demonstrated significantly increased total number of cells, macrophages, lymphocytes, and eosinophils compared to vitamin D-sufficient PBS-control mice as well as vitamin D-supplemented PBS and HRA-sensitized and challenged mice (Fig 2). No significant difference was observed in the number of neutrophils between groups.

Assessment of IgE and Vitamin D Levels in Vitamin D-sufficient and Vitamin D-supplemented Mice
The levels of IgE in serum and in the BALF of HRA-sensitized and challenged mice were significantly increased in vitamin D-sufficient group compared to vitamin D-sufficient PBS-sensitized and challenged as well as vitamin D-supplement PBS and HRA-sensitized and challenged mice. Therefore, mice on a supplemented vitamin D diet had decreased levels of IgE in both serum ( Fig 3A) and in the BALF (Fig 3B). The serum 25(OH)D levels in vitamin D-supplement mice were significantly higher than in the vitamin D-sufficient group mice (Fig 3C).

Effect of Vitamin D Status on Differences in the Degree of Airway Remodeling After HRA Sensitization and Challenge
After tracheostomy animals were sacrificed; lungs were harvested and sectioned, followed by staining with H&E, PAS, and trichrome stains to examine histological hallmarks of asthmatic Effect of vitamin D on total and differential leukocytes in BALF of PBS and HRA-sensitized mice. Total cells in the BALF were counted using Countess Automated Cell Counter and differential analysis was performed using standard morphological criteria. A minimum of 300 cells were examined per cytospin slide and absolute cell numbers for each cell type were calculated according to the total leukocyte counts in BALF and the percentage of each individual cell type on the slide. Data are shown as mean ± SEM for seven animals in each group (*p < 0.05; **p < 0.01; ***p < 0.001).
doi:10.1371/journal.pone.0149180.g002 Vitamin D Reduces EMT in Asthma airways. Airways of vitamin D-sufficient PBS mice displayed normal parenchyma (Fig 4A), with no mucus staining (Fig 4B), and little collagen staining around the respiratory epithelium (Fig 4C). Vitamin D-sufficient HRA mice exhibited signs of mild airway remodeling, showing airway epithelial cell hypertrophy (Fig 4D), mild mucus staining (Fig 4E), and collagen deposition ( Fig 4F). However, in vitamin D-supplemented mice, features of airway remodeling were  absent in both PBS and HRA groups (Fig 4G and 4J). Also, there was no positive staining to mucus (Fig 4H and 4K) and collagen (Fig 4I and 4L), as determined by periodic acid Schiff (PAS) (Fig 5A) and trichrome stains (Fig 5B), respectively.

Effect of Vitamin D on the Expression of E-cadherin, Vimentin, and Ncadherin in the Lungs of HRA-Sensitized and Challenged Mice
Expression of E-cadherin was observed in the lung tissue of PBS control mice in both vitamin D-sufficient and supplemented groups (Fig 6A and 6C). The epithelium in the HRA-sensitized and challenged mice demonstrated significant decreased expression of E-cadherin ( Fig 6B) as determined using ImageJ software. E-cadherin expression was protected by the vitamin D-supplemented diet, as observed in the vitamin D-supplemented HRA-sensitized and challenged mice (Figs 6D and 5E). The expression of vimentin in the lung tissue was marginally expressed in the peribronchial area in the vitamin D-sufficient PBS mice (Fig 7A and 7C). However, in the vitamin D-sufficient HRA-sensitized and challenged mice, vimentin expression was prominently increased in expression and found to be within the epithelial layer of the airway (Fig 7B). Vimentin expression was found to be significantly increased in vitamin D-supplemented HRA-sensitized and challenged mice (Fig 7D) compared to vitamin D-supplemented PBS-sensitized and challenged mice (Fig 7C). However, the expression of vimentin in the airway epithelium of vitamin D-sufficient HRA mice was significantly greater than in the vitamin D-supplemented HRA-sensitized and challenged mice (Fig 7E).
N-cadherin expression was minimal in the lamina propria in both groups of PBS mice ( Fig  8A and 8C). N-cadherin was expressed in the blood vessels of the airway (Fig 8A), consistent with the reports of N-cadherin being expressed normally in vascular smooth muscle cells [18]. N-cadherin was expressed in vitamin D-sufficient HRA-sensitized and challenged mice ( Fig  8B). This expression was decreased by vitamin D supplementation, as determined by vitamin D-supplemented HRA-sensitized and challenged mice (Fig 8D). However, vitamin D supplementation was not enough to prevent N-cadherin expression by HRA, as evidenced by comparing with the vitamin D-supplemented PBS-sensitized and challenged mice (Fig 8E).
These results confirm our earlier studies in bronchial epithelial cells that in an inflammatory environment, vitamin D increases the expression of E-cadherin, and decreases the expression of vimentin and N-cadherin [19].

Discussion
Structural changes in the airway epithelium lead to abnormal lung function and AHR. Physical alterations in the asthmatic epithelium include loss of E-cadherin [20] and induction of myofibroblast markers, N-cadherin and vimentin [21]. Myofibroblasts are a part of the wound healing response [22] and have been implicated in the invasion of cancer cells into the epithelium Vitamin D Reduces EMT in Asthma compartment of the colon [23]. Damage to the airway epithelium from allergens or other environmental factors can lead to the release of inflammatory mediators that promote migration of the epithelial cell to allow for wound repair. Myofibroblasts can be derived from epithelial cells through induction of EMT. EMT in wound healing is a normal biological procedure. However, in the context of an allergic response, this process becomes detrimental to the airway allowing transient properties to become permanent.
Corticosteroid treatment has been found to be ineffective against TGF-β mediated differentiation of myofibroblasts [24], [25]. For that reason alternates to anti-inflammatory treatments are required. Vitamin D has emerged as an effective anti-inflammatory agent and has been found to be effective in asthma, specifically in glucocorticoid-resistant asthma [26]. However, we still do not fully understand the underlying cellular and molecular mechanisms of vitamin D effects in the pathogenesis and/or treatment of asthma. In vitro studies have observed antiinflammatory effects of vitamin D in airway smooth muscle cells [27], fibroblasts [28], and epithelial cells [29]. In vivo studies also suggest improvement in lung function and improved structural integrity in the lung with vitamin D administration [30].
In this study, we found that vitamin D decreased the amount of IgE in both serum and in the BALF. This is consistent with the reports of high concentrations of serum total IgE in asthmatic patients with vitamin D deficiency [31]. While vitamin D was found to have no effect on total IgE in a vitamin D-deficient OVA-mouse model [32], supplementation of vitamin D in the diet of vitamin D-deficient mice has been reported to decrease eosinophil and neutrophil numbers in the BALF [33]. In other studies, administration of 1,25(OH) 2 D 3 suppressed allergen-specific IgE levels, eosinophilic airway inflammation, AHR, along with reduced IL-5 and IL-13 levels in BALF and reduced production of Th2 cytokines by T-cells [34]. Overall, most of the findings, including ours in this study, support the beneficial effect of vitamin D in attenuating allergic airway inflammation and asthma.
The immunomodulatory properties of vitamin D have been associated with several inflammatory diseases [35]. In asthma, vitamin D deficiency has been linked with worse outcomes [36][37][38]. The use of vitamin D supplementation as a form of therapy in patients with asthma has not been without controversy. A randomized double-blind study analyzing children on glucocorticoid steroid with or without 500 IU/day of vitamin D 3 found significant improvement in the absolute and predicted forced expiratory volume in 1 second (FEV 1 ) as well as decreased exacerbations [39]. Another study in asthmatic subjects using 0.25 μg calcitriol capsules twice daily demonstrated significant improvement in the absolute and percent predicted in FEV 1 [40]. Vitamin D supplementation was also found to improve FEV 1 in 24 weeks in patients on glucocorticoid steroids after receiving an initial bolus of 100,000 IU followed by 50,000 IU weekly [26]. Conversely, another recent study also evaluating patients on glucocorticoid steroids with or without an initial dose of 100,000 IU of vitamin D 3 followed by a daily dose of 4,000 IU found no difference in asthma control or quality of life. Results of this trial also indicated that the total inhaled corticosteroid dose was lower in the vitamin D 3 treatment group [41]. Vitamin D insufficiency in patients with chronic kidney disease can be combated by a weekly dose of 10,000 IU oral cholecalciferol supplement, raising 25-hydroxyvitamin D levels twofold [42]. The guidelines of the Institute of Medicine classify risk of vitamin D deficiency when serum levels of 25(OH)D are below 20 ng/ml (50 nM) [43]. Many clinical laboratories continue reporting a value of 20 ng/ml as inadequate and that 21-29 ng/ml (52.5-72.5 nM) as insufficient levels [44]. However, in patients at risk for vitamin D deficiency, the Institute of Medicine recommended a daily intake of 800 IU of vitamin D3 to raise the blood level of 25(OH)D. The Endocrine Society Task Force has proposed that 1,500-2,000 IU of vitamin D daily is required to raise blood 25(OH)D levels to a sufficient range [45]. In this study, we administered 10,000 IU vitamin D3 to mice to achieve the serum levels of 25(OH)D that is generally found in human with 3,000-4,000 IU vitamin D3 supplementation without any toxicity. The difference in the magnitude of increased serum 25(OH)D levels in mice and human could be related to differential rate of vitamin D metabolism. Nonetheless, the findings support the role of supplemental vitamin D in the attenuation of hallmark features of allergic asthma.
We report findings that vitamin D supplementation is associated with lower AHR to methacholine with decreased intensity of inflammation, decreased bronchoconstrictory response to methacholine, and decreased serum and BAL fluid IgE levels compared to vitamin D-sufficiency in HRA-sensitized and challenged mice. Increased inflammation in the lung was also displayed in the cells recovered from the BAL fluid with increased macrophages, lymphocytes, and eosinophils in the vitamin D-sufficient compared to vitamin D supplemented HRA-sensitized and challenged mice. These results are consistent with previously published results where vitamin D supplementation, while not fully reversing the effects of allergic airway inflammation, did reduce AHR, airway remodeling and BALF cytokines levels [30].

Conclusions
The findings in this study suggest that vitamin D supplementation reduces the effect of allergic airway inflammation in a mouse model with nasal aerosolization of clinically relevant allergens. This study further indicates a role for the epithelium in mediating airway remodeling in asthma. However, further research will need to be conducted in order to elucidate the effect of vitamin D on antigen presentation by dendritic cells to T cells and the immune response in this mouse model. These results also suggest that the epithelium is a therapeutic target for the prevention of airway remodeling in asthma.

Acknowledgments
As the senior corresponding author, I warrant that the manuscript, as submitted, has been reviewed by and approved by all named authors; that the corresponding author is empowered by all of the authors to act on their behalf with respect to the submission of the manuscript; that the article is original; that the article does not infringe upon any copyright or other proprietary right of any third party; that neither the text nor the data reported have been published previously (abstracts excepted); and that the article or a substantially similar article is not under consideration by another journal at this time. All authors have read the journal's authorship statement. No writing assistance was utilized in the production of this manuscript. The magnification); AW: Airway BV: Blood vessel. Right panels: Sections stained using rabbit anti-N-cadherin antibody and goat anti-rabbit Alexa Fluor 549 as secondary antibody; Left panels: Merged Alexa Fluor 549 and DAPI used to stain the nuclei; (C, D) Expression of N-cadherin in the lungs of vitamin D-supplemented PBS control mice compared to vitamin D-supplemented HRA-sensitized and challenged mice (40x magnification); (E) Corrected total fluorescence (CTF) of proteins in the airways was measured in AU using ImageJ software. The results are presented as mean ± SEM of 7 mice per group with 10 measurements per mouse, *p <0.05, ** p <0.01, ****p <0.0001. doi:10.1371/journal.pone.0149180.g008 Vitamin D Reduces EMT in Asthma content of this article is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.

Author Contributions
Conceived and designed the experiments: KDF SCH DKA. Performed the experiments: KDF SCH. Analyzed the data: KDF DKA. Contributed reagents/materials/analysis tools: DKA. Wrote the paper: KDF DKA.