Superior effect of MP-AzeFlu than azelastine or fluticasone propionate alone on reducing inflammatory markers

Background MP-AzeFlu, intranasal formulation of azelastine hydrochloride (AZE) and fluticasone propionate (FP), is superior to AZE or FP alone for treatment of allergic rhinitis (AR). However, the precise anti-inflammatory mechanism of action of MP-AzeFlu has not been characterized. Objective To investigate the anti-inflammatory effects of MP-AzeFlu compared with AZE or FP alone in an established in vitro model of eosinophilic inflammation. Methods Nasal mucosal epithelial cells and peripheral blood eosinophils were obtained from human volunteers. Epithelial cells were stimulated with 10% fetal bovine serum (FBS) in the presence of MP-AzeFlu, AZE, or FP (1:102 to 1:105 dilution). Concentrations of interleukin (IL)-6, IL-8, and granulocyte–macrophage colony-stimulating factor (GM-CSF) were measured by ELISA. Eosinophils were incubated in 10% human epithelial cell–conditioned medium (HECM) and survival assessed by trypan blue dye exclusion. Results are expressed as mean ± SEM percentage secretion/survival compared with FBS/HECM (respectively). Results FP and MP-AzeFlu (all dilutions) and AZE (1:102) significantly reduced IL-6 secretion and eosinophil survival compared with positive controls. At 1:102 dilution, IL-6 secretion was significantly lower with MP-AzeFlu (38.3 ± 4.2%, compared with FBS = 100%) than with AZE (76.1 ± 4.9%) or FP (53.0 ± 4.9%). At 1:102 dilution, eosinophil survival was significantly lower with MP-AzeFlu at day 3 (17.5 ± 3.0%) and day 4 (2.4 ± 1.4%, compared with HECM = 100%) than with AZE (day 3: 75.2 ± 7.2%; day 4: 44.0 ± 9.7%) or FP (day 3: 38.5 ± 3.5%; day 4: 14.6 ± 4.0%). Conclusion Greater reductions in cytokine secretion and eosinophil survival observed with MP-AzeFlu in vitro may underlie MP-AzeFlu’s superior clinical efficacy vs. AZE or FP alone observed in AR patients. Electronic supplementary material The online version of this article (10.1186/s13223-018-0311-4) contains supplementary material, which is available to authorized users.


Background
Allergic rhinitis (AR) is one of the most common chronic diseases, impairing quality of life and causing billions of dollars of lost productivity annually [1]. AR is characterized by upper airway inflammation, sneezing, and nasal congestion, drainage, and itching [1]. Inflammatory mediators and cells in AR include elevated levels of proinflammatory cytokines and eosinophil infiltration [2][3][4][5].
Current guidelines recommend intranasal corticosteroids for treatment of AR and, in some cases, the use of oral or intranasal antihistamines [1,6,7]. Unfortunately, many patients do not achieve full control of their symptoms and are not satisfied with their treatment [8]. Combination therapy may  14:86 be considered for patients with inadequate response to monotherapy [1] or when a prompt response to initial therapy is desired [6]. In particular, combined intranasal azelastine hydrochloride (AZE) and intranasal fluticasone propionate (FP) in a single intranasal formulation (MP-AzeFlu) is recommended as more effective than monotherapy [8] and as a firstline option for moderate-to-severe AR [7,8].
In randomized studies, MP-AzeFlu was more effective among patients with seasonal AR and perennial AR than AZE or FP alone, [9][10][11][12][13] and was more effective among patients with non-AR than FP alone [14]. In addition, studies in real-world settings in Europe found significant improvements in AR symptoms with MP-AzeFlu therapy, [15,16] and a 1-year study of MP-AzeFlu vs. FP alone provided support for the longterm efficacy and safety of MP-AzeFlu in persistent rhinitis [17]. A retrospective US claim database study of AR patients with comorbid asthma has shown, that the AR and asthma related therapy costs were lower when the patients have been treated with MP-AzeFlu than with a free combination of intranasal steroid and intranasal antihistamine [18].
Mechanistic studies have examined the effect of FP (alone and in combination with the antihistamine loratadine) on the expression of inflammatory mediators, including proinflammatory cytokines and eosinophils [19][20][21][22][23][24]. Typically, FP downregulated cytokine expression and reduced eosinophil survival in these studies, although findings are mixed. In addition, AZEalone or in combination with other agents-was found to suppress inflammatory markers in several in vitro studies [25][26][27]. However, the mechanism of action of combined AZE and FP, specifically MP-AzeFlu's effects on inflammatory mediators, has not been characterized. In particular, it is unknown whether there may be an enhanced anti-inflammatory effect of the two drugs in combination, compared with the individual components, which might underlie the superior clinical efficacy [9][10][11][12][13][14] of the combination compared with monotherapy.
To study the role of inflammatory mediators in upper airway diseases and the mechanism of action of antiinflammatory drugs in these diseases, an in vitro model was developed utilizing cultured primary isolated nasal mucosal epithelial cell cultures and peripheral blood eosinophils [28][29][30][31]. This in vitro model has been used previously to compare the anti-inflammatory effects of a number of drugs including corticosteroids, chromones, anti-leukotrienes, and second generation antihistamines, [24,[28][29][30][31][32][33][34] demonstrating that it is a good model to study the mechanisms of action of these classes of drugs. In a previous study using this in vitro model, the combination of the corticosteroid mometasone furoate and the antihistamine desloratadine reduced interleukin (IL)-6 and (sICAM)-1 secretion and inhibited eosinophil survival induced by epithelial secretions compared with either agent alone [29].
The objective of the current study was to investigate the anti-inflammatory effects of MP-AzeFlu compared with AZE or FP alone in an established in vitro model of eosinophilic inflammation.

Materials
AZE and FP were provided by MEDA Pharma (Bad Homburg, Germany). Other materials were purchased from commercial sources (see Additional file 1).

Study population
Nasal mucosa specimens were obtained from 12 patients (nine men, three women), ranging in age from 34 to 73 years (mean ± standard deviation, 58.2 ± 3.5 years), who underwent nasal corrective surgery for septal dysmorphy, turbinate hypertrophy, or both. The diagnosis of septal dysmorphy and turbinate hypertrophy was based on the clinical history and nasal endoscopic exploration. Skin-prick test was positive (allergen sensitization) in three patients (25.0%). None of the patients in this study had clinical AR, chronic rhinosinusitis (CRS), nasal polyps (NP), and/or asthma. Patients were excluded from this study if they were receiving topical or systemic glucocorticoids or antihistamine treatment 4 weeks prior to the surgery or had an upper or lower airway infection 2 weeks prior to the surgery. All patients gave informed consent to participate in the study at the time of surgery. Tissues used in this study were obtained from the Biobank BTIRCE-R100311-016 at Institut d'Investigacions Biomèdiques August Pi i Sunyer (IDIBAPS). Scientific and Ethics Committee of Hospital Clínic de Barcelona gave the ethical clearance for this process.
Normodense eosinophils were obtained from nine volunteers (seven women, two men), ranging in age from 39 to 74 years (mean ± standard deviation, 59.1 ± 13.6 years) with > 3% peripheral blood eosinophils (mean ± standard deviation, 7.7 ± 1.4%). Patients were excluded if they received topical or systemic glucocorticoid or antihistamine treatment 4 weeks prior to blood extraction or if they had an upper or lower airway infection two weeks prior to blood extraction. Skin-prick test was positive (allergen sensitization) in four patients (44%). Two of the patients had AR (22%), four patients had CRS with NP (44%), and three patients had CRS with NP and asthma (33%). All patients gave informed consent to participate in the study prior to the venipuncture. Scientific and Ethics Committee of Hospital Clínic de Barcelona gave the ethical clearance for this process.
To study the effect of MP-AzeFlu on cytokine production, Ham's HD was switched to RPMI (1 mL) in the presence or absence of MP-AzeFlu (dilution 1:10 2 to 1:10 5 , as described in Additional file 1) or equivalent dilutions of AZE (from 2.39 × 10 −5 M to 10 −8 M) or FP (from 7.29 × 10 −6 M to 10 −9 M) for 1 h before the addition of 10% FBS. After 24 h, the supernatant was harvested from cultures, centrifuged at 400 g for 10 min at room temperature, sterilized through 0.22 µm filters, and stored at -80 °C until used. Because both AZE and FP were diluted in dimethyl sulfoxide (DMSO) when preparing the MP-AzeFlu formulation, we investigated the effect of DMSO at the highest final concentration present in the culture medium on epithelial cell viability and cytokine secretion.

Epithelial cell viability
Cell viability after treatment was analyzed by incubation of cells with the tetrazolium salt XTT (Cell Proliferation Kit II) for 3 h, following the manufacturer's instructions. Absorbance was measured in duplicate at 490 nM.

Enzyme-linked immunoassays of cytokines and sICAM-1
Concentrations of granulocyte-macrophage colonystimulating factor (GM-CSF), IL-6, IL-8, and sICAM-1 were measured in HECM using commercial enzymelinked immunosorbent assay (ELISA) kits. Cytokines selected for analysis were those found at detectable levels in previous studies in which this model was used. The assay detection ranges were 15.6-1000 pg/mL for GM-CSF, 9.38-600 pg/mL for IL-6, and 31.2-2000 pg/ mL for both IL-8 and sICAM-1. To verify that the substances used in the different experiments (AZE, FBS, FP) did not affect the ELISA results, wells containing either culture media alone or media with the highest drug concentration used in the different protocols were compared (N = 3). None of the substances showed any intrinsic effect on the ELISA final values. In order to avoid variability in cytokine concentration caused by differences in the number of cells present in each culture well, cytokine production was normalized by optical density value obtained by the cell proliferation assay.

Isolation of peripheral blood eosinophils
Isolation of eosinophils from peripheral blood samples is described in Additional file 1.

Assessment of eosinophil survival
Eosinophils (2.5 × 10 5 cells/well) were incubated on 24-well tissue culture plates with RPMI (2 mL) in the presence or absence of MP-AzeFlu (dilution 1:10 2 to 1:10 5 ) or equivalent dilutions of AZE (from 2.39 × 10 −5 M to 10 −8 M) or FP (from 7.29 × 10 −6 M to 10 −9 M) for 1 h before the addition of epithelial cell secretions at 10%. Eosinophil survival index was assessed at 24 h (day 1), 48 h (day 2), 72 h (day 3), and 96 h (day 4) of incubation by trypan blue dye exclusion. Because dead eosinophils become lysed and, consequently, the number of cells present in the culture wells decreases, the results were calculated using the eosinophil survival index instead of the percentage of surviving cells. The eosinophil survival index was calculated as follows: number of eosinophils recovered multiplied by percentage of eosinophil viability divided by number of eosinophils delivered on day 0.

Effect of FBS on cytokine and sICAM-1 secretion
In nasal mucosal epithelial cell cultures, FBS increased the secretion of IL-6, IL-8, GM-CSF, and sICAM-1 compared with control medium (Table 1).

Comparison of MP-AzeFlu, AZE, and FP effects on FBS-induced cytokine secretion at the same drug dilutions
When comparing the effect of FP with MP-AzeFlu (from dilutions 1:10 2 to 1:10 5 ), there were no significant differences on the inhibition of GM-CSF or IL-8 secretion. However, with each drug at dilution 1:10 2 , the inhibitory effect of MP-AzeFlu on IL-6 secretion was significantly greater than that of AZE or FP, as shown by the lower levels of IL-6 secretion with MP-AzeFlu ( Fig. 1; see Table 2 for underlying findings at dilution 1:10 2 ). In addition, the effect of FP was significantly greater than the effect of AZE. At higher dilutions (1:10 3 to 1:10 5 ) of FP and MP-AzeFlu, there were no significant differences between drugs in the inhibition of IL-6 secretion.

Time course of MP-AzeFlu (dilution 1:10 2 ) effects on HECM-induced eosinophil survival from days 1 to 4
HECM at 10% from nasal mucosal epithelial cells significantly increased eosinophil survival when compared with control medium from days 1 to 4 (Fig. 2). MP-AzeFlu at 1:10 2 dilution showed a time-dependent inhibitory effect on HECM-induced eosinophil survival from days 2 to 4.

Dose response and time course of MP-AzeFlu, AZE, and FP on HECM-induced eosinophil survival at days 3 and 4
At days 3 and 4, MP-AzeFlu and FP (dilution 1:10 2 to 1:10 5 ) significantly inhibited HECM-induced eosinophil survival (Table 3). However, AZE showed an inhibitory effect only at dilution 1:10 2 . At days 3 and 4, the inhibitory effect of MP-AzeFlu at dilution 1:10 2 was significantly greater than either AZE or FP at the same dilution, as shown by the lower levels of eosinophil survival with MP-AzeFlu (Fig. 3, see Table 3 for underlying findings at dilution 1:10 2 ). In addition, the effect of FP at dilution 1:10 2 was significantly greater than the effect of AZE at the same dilution. No differences were found when comparing the inhibitory effect of FP with that of MP-AzeFlu from dilutions 1:10 3 to 1:10 5 .

Discussion
This report shows clear mechanistic effects that are consistent with and may underlie the superior clinical efficacy of MP-AzeFlu compared with corticosteroid or antihistamine alone. Key findings of this study include: (1) FP and MP-AzeFlu at all tested dilutions, and AZE at 1:10 2 dilution, significantly reduced secretion of IL-6 compared with FBS-induced secretion; (2) at 1:10 2 dilution of each agent, the reduction of IL-6 secretion by MP-AzeFlu was significantly greater than with AZE or FP alone; (3) FP and MP-AzeFlu, at all tested dilutions, and AZE at dilution 1:10 2 , significantly reduced eosinophil survival at days 3 and 4 compared with HECM alone; and (4) at 1:10 2 dilution of each agent, the decrease of eosinophil survival induced by MP-AzeFlu at days 3 and 4 was significantly greater than with AZE or FP alone.
In the present study, the corticosteroid FP and the formulation MP-AzeFlu reduced IL-6 and GM-CSF secretions from nasal mucosa epithelial cells relative to FBS. This is consistent with previous findings for the intranasal corticosteroids budesonide, beclomethasone dipropionate, mometasone furoate, FP, and fluticasone furoate [23,24,29,30,35]. In this study, MP-AzeFlu findings for IL-8, GM-CSF, and sICAM-1 differed little from FP alone. We also found that FP and MP-AzeFlu reduced eosinophil survival relative to HECM, consistent with previous findings for intranasal corticosteroids [24,29,30].
In the present study, the antihistamine AZE reduced IL-6 and GM-CSF secretion and eosinophil survival only at 1:10 2 dilution. These reductions are consistent with previous findings for the antihistamine desloratadine [29,31]. Our findings for AZE are also consistent with those of previous in vitro studies that found AZE decreased inflammatory markers [25][26][27]. A possible mechanism for these effects has been suggested by research showing that AZE exhibits direct activity on transient receptor potential vanilloid 1 ion channels in mouse neuronal cells [36]. We found that AZE did not significantly reduce ICAM-1 expression, in contrast to an earlier study that found such a reduction [37].
The stronger effects of MP-AzeFlu on IL-6 secretion and eosinophil survival in the current study, compared with either AZE or FP alone, mirror and possibly underlie the stronger clinical efficacy of the MP-AzeFlu combination compared with its components [9][10][11][12][13][14]. The relative magnitude of in vitro effects for MP-AzeFlu vs. AZE or FP alone (Figs. 1, 3) appear similar to the relative magnitude of clinical effects of these agents in treatment of AR. For example, the reduction in total nasal symptom

Table 2 Effect of AZE, FP, and MP-AzeFlu on cytokine and sICAM-1 secretion from nasal mucosal epithelial cells
Results are expressed as mean ± SEM. The Wilcoxon signed-rank test was used for analysis. *P < 0.05 compared with 10% FBS-induced secretion There is good reason to believe our in vitro findings may elucidate the mechanism of action of MP-AzeFlu in AR. The inflammatory mediators IL-6, GM-CSF, and IL-8, as well as eosinophils, play important roles in airway inflammatory diseases. IL-6 is a proinflammatory cytokine with pleiotropic expressions consistent with a primary role in the pathogenesis of local inflammation [39]. IL-6 mediates many biologic functions, acting as an endogenous pyrogen, stimulating the acute phase response, stimulating T lymphocytes, inducing terminal differentiation of B lymphocytes, and stimulating immunoglobulin production [39]. It is well known that CRS is associated with elevated levels of IL-6, as upregulation has been reported in peripheral blood mononuclear cells (PBMCs), [40] on sinus mucosa biopsies, [41][42][43] and in nasal secretions [44] from patients suffering from CRS. In fact, increased expression of IL-6 messenger RNA in nasal mucosa biopsies of patients suffering from persistent AR has been reported [45], pollen exposure to patients with AR significantly increased IL-6 in nasal secretions [5], and nasal secretions were increased in allergic patients after intranasal administration of IL-6 [39]. In addition, it has been reported that primary cultures of human nasal epithelial cells from patients with AR showed significant upregulation in the release of IL-6 [46,47].
GM-CSF plays a pivotal role in the maturation, chemotaxis, survival and activation of eosinophils [34,48]. GM-CSF has also been involved in the regulation of glandular secretion by inducing lactoferrin release in nasal mucosa [49]. Furthermore, it has been reported that a significant correlation exists between GM-CSF concentrations in nasal secretions and in allergenspecific immunoglobulin E antibodies to house dust mite Dermatophagoides pteronyssinus in patients with persistent AR [50].
On the other hand, it has been reported that CRS is associated with elevated levels of IL-8 in PBMCs, [51] sinus mucosa biopsies, [43,45] and nasal secretions [44]. IL-8, in addition to its potent activity on neutrophils, can cause basophil histamine release and co-induce chemotactic activity for primed eosinophils [52,53]. The activation and infiltration of eosinophils in AR and their release of proinflammatory mediators has also been described [54,55]. Furthermore, high levels of IL-8 were detected in nasal secretions of patients with AR after the nasal provocation test [4], and upregulation in the release of IL-8 has been reported in primary cultures of human nasal epithelial cells from patients with AR [46].
Finally, it has been reported that the levels of sICAM-1 and soluble vascular adhesion molecule-1 of patients

Availability of data and materials
The datasets generated during the current study are not publicly available because the data reside in a proprietary database maintained by Meda