Montelukast, a Cysteinyl Leukotriene Receptor 1 Antagonist, Induces M2 Macrophage Polarization and Inhibits Murine Aortic Aneurysm Formation

Background The pathogenesis of abdominal aortic aneurysm (AAA) is characterized by atherosclerosis with chronic inflammation in the aortic wall. Montelukast is a selective cys-LT 1 receptor antagonist that can suppress atherosclerotic diseases. We evaluated the in vitro properties of montelukast and its in vivo activities in an angiotensin II–infused apolipoprotein E–deficient (apoE−/−) AAA mouse model. Methods The mouse monocyte/macrophage cell line J774A.1 was used in vitro. M1 macrophages were treated with montelukast, and gene expressions of inflammatory cytokines were measured. Macrophages were cultured with montelukast, then gene expressions of arginase-1 and IL (interleukin)-10 were assessed by quantitative polymerase chain reaction, arginase-1 was measured by fluorescence-activated cell sorting, and IL-10 concentration was analyzed by enzyme-linked immunosorbent assay. In vivo, one group (Mont, n=7) received oral montelukast (10 mg/kg/day) for 28 days, and the other group (Saline, n=7) was given normal Saline as a control for the same period. Aortic diameters, activities of matrix metalloproteinases (MMPs), cytokine concentrations, and the number of M2 macrophages were analyzed. Results Relative to control, montelukast significantly suppressed gene expressions of MMP-2, MMP-9, and IL-1β, induced gene expressions of arginase-1 and IL-10, enhanced the expression of the arginase-1 cell surface protein, and increased the protein concentration of IL-10. In vivo, montelukast significantly decreased aortic expansion (Saline vs Mont; 2.44 ± 0.15 mm vs 1.59 ± 0.20 mm, P<.01), reduced MMP-2 activity (Saline vs Mont; 1240 μM vs 755 μM, P<.05), and induced infiltration of M2 macrophages (Saline vs Mont; 7.51 % vs 14.7 %, P<.05). Conclusion Montelukast induces M2 macrophage polarization and prevents AAA formation in apoE−/− mice.


Introduction
Abdominal aortic aneurysm (AAA) is a localized dilatation of the aorta and is a life-threatening vascular disorder [1]. The usual treatment for AAA is open surgical repair or endovascular repair to prevent rupture of the aorta. Although these techniques are effective in patients with a large AAA, few therapeutic strategies are available for surgically unsuitable patients who have a small AAA or who are otherwise at high risk or cannot undergo open or endovascular surgery because of anatomical limitations. Therefore, nonsurgical or less invasive therapeutic procedures for the treatment of AAA are urgently required. The purpose of our study was to verify whether or not montelukast, a cys-LT 1 receptor antagonist, would be effective as a minimally invasive treatment, and how macrophages were involved in its mechanism of action.
The pathogenesis of AAA is characterized by the destruction of the extracellular matrix, with chronic inflammation in the aortic wall. This inflammation is induced by the infiltration and activation of inflammatory cells such as macrophages, which release inflammatory cytokines and matrix metalloproteinases (MMPs) that cause degradation of the extracellular matrix [2].
Macrophages play an important role in regulating the inflammatory process in AAA. The two main macrophage phenotypes are M1 and M2 [3]. M1 macrophages are release 2 BioMed Research International proinflammatory cytokines and strongly promote aneurysm formation. In contrast, M2 macrophages are considered to be anti-inflammatory and are highly protective against elastin degradation and aneurysm progression. Therefore, the M1/M2 macrophage ratio plays a critical role in aneurysm development and is potentially a significant therapeutic target [4,5].
Leukotrienes (LTs), which are derived from the 5-lipoxygenase (5-LO) pathway of arachidonic acid metabolism, are potent lipid mediators with proinflammatory biological actions [6]. Among them, cysteinyl-leukotrienes (cys-LTs), which include LTC 4 , LTD 4 , and LTE 4 , increase vascular permeability and smooth muscle cell contraction, leading to bronchoconstriction and vasoconstriction in asthmatic patients [7]. Recent experimental studies reported that LTD 4 induced MMP production through a cys-LT 1 receptor in the aortic wall [8] and that montelukast, a selective cys-LT 1 receptor antagonist with anti-inflammatory effects that has been widely used for the treatment of inflammatory diseases such as asthma and allergic rhinitis, inhibited the development of atherosclerotic diseases [9,10]. Recently, Gennaro et al. showed that montelukast prevented experimental AAA [11]. However, little information is available about the role of montelukast in AAA formation. Therefore, we evaluated the in vitro properties of montelukast and its in vivo activities in an angiotensin II (Ang II)-induced apolipoprotein E-deficient (apoE −/− ) AAA mouse model.

Materials and Methods
This study was performed in accordance with the Guide for the Care and Use of Laboratory Animals (National Institutes of Health, Publication 85-23, National Academy Press, Washington, DC, USA, revised in 2011) and was approved by the Institute for Laboratory Animal Research at the Nagoya University Graduate School of Medicine (Protocol No. 30003).
. . Cell Culture and Treatment. Murine macrophages of the J774A.1 cell line were purchased from the National Institutes of Biomedical Innovation, Health, and Nutrition (Ibaraki, Osaka, Japan) and cultured in Dulbecco's modified essential medium (DMEM) (Gibco, Thermo Fisher Scientific, Waltham, MA, USA) supplemented with 10% fetal bovine serum (FBS; Gibco, Thermo Fisher Scientific). Macrophages were seeded on 24-well tissue culture plates (Iwaki; AGC, Inc., Tokyo, Japan) at a concentration of 4×10 5 cells per well and incubated for 4 days at 37 ∘ C in a humidified atmosphere with 5% CO 2 . To increase the release of MMPs and inflammatory cytokines from inflammatory M1 macrophages, the cells were stimulated by TNF-(20 ng/ml) at 37 ∘ C in 5% CO 2 for 24 hours. After that, the medium was replaced with 1 ml DMEM with 10% FBS containing 2 M (n=6) or 20 M (n=6) montelukast (Tokyo Kasei Kogyo Co., Ltd., Tokyo, Japan) for 12 hours [12]. We used the 20 M dose because a previous report demonstrated that 20 M montelukast reduced oxidized low-density lipoprotein-induced monocyte adhesion to human umbilical vein endothelial cells [13].
. . Animal Preparation and Drug Administration. AAA model mice were generated as previously described [14,15]. Briefly, apoE −/− mice on the C57BL/6 background were purchased from the Jackson Laboratory (Sacramento, CA, USA) and maintained on a regular diet under standard conditions (24 ∘ C, 63% humidity). Male apoE −/− mice (6 to 8 months old) were infused with 1000 ng/kg/min Ang II (Calbiochem, Darmstadt, Germany) for 28 days through subcutaneous osmotic minipumps (model 2004; DURECT, Cupertino, CA, USA) that were implanted under anesthesia with isoflurane. Several studies reported that the efficacy of montelukast against asthma was dose dependent when administered orally [16][17][18], so we presumed that its effectiveness in this setting could be concentration dependent. Therefore, one group (Mont, n=7) received oral montelukast (10 mg/kg/day) through a gastric tube once a day for 28 days, and the other group (Saline, n=7) was administered oral normal Saline for the same period (Supplementary Figure  1). We examined the maximum AAA diameters of mice by ultrasonography (10 MHz; GE Healthcare, Chicago, IL, USA) before pump implantation and every week thereafter. Mice were sacrificed at 28 days after administration, and the portion of the aorta between the descending aorta and the iliac bifurcation was harvested. The maximum AAA diameter from the thoracoabdominal aorta to the suprarenal abdominal aorta was measured, as previously described [14,   15]. AAA was defined as a 50% increase relative to the external aortic diameter before pump implantation [19]. Images of aortas were captured with a DP70 digital camera (Olympus, Tokyo, Japan) connected to stereomicroscope, and 2-mmlong segments of isolated aorta were cut at the infradiaphragm for histologic examination. The residual aorta was minced and homogenized for analysis of MMP enzymatic activity and protein expression.
. . Elastica Van Gieson Staining. For the determination of elastin degradation, we performed Elastica van Gieson staining as previously described [14]. Briefly, harvested aortas were embedded in OCT compound (Tissue-Tek; Sakura Finetek USA, Torrance, CA, USA), and we prepared frozen slices (8-m thickness) using a microtome cryostat (Leica microsystems). The frozen cross-sections were stained with Weigert's resorcin-fuchsin (Muto Pure Chemicals, Tokyo, Japan) according to the manufacturer's protocol, and stained samples were photographed with an FSX-100 microscope (Olympus, Tokyo, Japan). The number of intact elastin fibrils on the aortic wall was analyzed using ImageJ software (National Institutes of Health, Bethesda, MD, USA). The images were used to quantify the percent area of elastin and the percent area of medial components between the elastic lamellae relative to the total medial tissue area [20].
. . Statistical Analysis. Statistical significance was calculated and compared between 2 groups using the 2 test or unpaired t-test as appropriate, and between 3 groups using one-way ANOVA and subsequent Tukey-Kramer multiple comparison. The SPSS software package (SPSS Inc., Chicago, IL, USA) was used in all cases. Data are shown as the mean ± standard error of the mean. P values less than 0.05 were considered statistically significant. . . Aortic Diameter Reduction. During this experiment, no mice died or experienced complications. The maximal aortic diameters in the Saline and montelukast groups, respectively, were 0.91 ± 0.04 mm and 0.99 ± 0.06 mm before administration (P=.27), 1.48 ± 0.12 mm and 1.16 ± 0.04 mm at day 7 (P<.05), 1.70 ± 0.14 mm and 1.30 ± 0.14 mm at day 14 (P=.08), 2.02 ± 0.14 mm and 1.36 ± 0.13 mm at day 21 (P<.01), and 2.10 ± 0.15 mm and 1.45 ± 0.15 mm at day 28 (P<.05) (Figures 3(a), 3(b), and 3(c)). At day 28, aortic diameters under microscopy were significantly expanded in the Saline group compared with the montelukast group (Saline vs Mont; 2.44 ± 0.15 mm vs 1.59 ± 0.20 mm, P<.01, Figure 3(d)). The incidence rate of AAAs in the Saline group was significantly higher than that in the montelukast group (Saline vs Mont; 100 % vs 14.3 %, P<.01, Figure 3(e)). . . Suppression of MMP-Activity. The activity of endogenous MMP-2 was significantly lower in the montelukast group than in the Saline group (endogenous MMP-2: Saline vs Mont; 1234 ± 199 M vs 734 ± 71 M, P<.05, Figure 5).

. . Elastic Lamellae in Medial
However, there was no significant difference between the two groups in the activity of endogenous MMP-9.

Discussion
Leukotrienes, which are derived from arachidonic acid by 5-LO in conjunction with 5-LO-activating protein (FLAP), are potent leukocyte chemoattractants and mediators of inflammation. LTA 4 , which is produced from arachidonic acid, is hydrolyzed by LTA 4 hydrolase into LTB 4 or conjugated by LTC 4 synthase (LTC 4 S) to form cys-LTs, including LTC 4 , LTE 4 , and LTD 4 [9]. Animal and human studies revealed that cys-LTs were involved in the pathogenesis of atherosclerosis and that the 5-LO pathway played an important role in the formation of AAA. Increased expression of cys-LTs was detected in human atherosclerotic lesions [22], and mice with 5-LO gene deletion demonstrated a significant reduction of atherosclerosis [23]. In the walls of human AAA, mRNA levels of 5-LO, FLAP, and LTC 4 S were significantly increased, and incubation of aortic wall tissues with arachidonic acid produced a significant amount of cys-LTs [24]. These findings show that cys-LTs induce the release of MMPs, including MMP-2 and MMP-9, as well as inflammatory cytokines such as ILs and TNF-, from macrophages through cys-LT receptor 1.
Montelukast, a selective cys-LT receptor 1 antagonist, has been widely used to treat asthma and allergic rhinitis. Its mechanisms of action are considered to include stimulating neutrophil infiltration [25], balancing oxidantantioxidant status, and decreasing plasma C-reactive protein levels. Some studies showed that montelukast could inhibit the inflammatory response related to atherosclerosis [9,26]. A previous animal study demonstrated that montelukast inhibited the development of atherosclerosis in association with decreased expression of MMP-2 and MMP-9 [27], and another experimental study indicated that montelukast produced anti-atherogenic effects related to MCP-1 downregulation [28]. Recently, Gennaro et al. demonstrated that montelukast reduced experimental AAA by inhibiting the release of MMP-9 and macrophage inflammatory protein-1 [11]. However, they did not investigate the relationship between montelukast and anti-inflammatory effects, especially those mediated by M2 macrophages. In contrast, we revealed that M2 macrophage polarization by montelukast participated in the prevention of AAA formation.
We demonstrated that montelukast suppressed the gene expressions of inflammatory cytokines in stimulated macrophages induced by TNF-in vitro. The gene expressions of MMP-2, MMP-9, and IL-1 were significantly lower in macrophages cocultured with montelukast than in nontreated macrophages. IL-1 is considered to be an important mediator of inflammation and is believed to be critical in experimental AAA formation [29].
While montelukast suppressed the expression of certain genes in our study, it also promoted gene expressions of IL-10 and arginase-1, enhanced the protein expression of arginase-1, and increased the protein concentration of IL-10 in macrophages incubated with montelukast. Our data thus indicate for the first time that montelukast induces M2 macrophages, which play a key role in suppressing AAA formation. Montelukast not only potently inhibits cys-LT but also strongly induces M2 macrophage polarization, and consequently it is reasonable to suppose that administration of montelukast could inhibit the development and expansion of AAA. Moreover, we found that montelukast significantly decreased the infiltration of M1 inflammatory macrophages and increased the infiltration of M2 anti-inflammatory macrophages in an in vivo experiment. The mechanism whereby montelukast prevents AAA formation therefore seems to involve enhancing anti-inflammatory activity by inducing M2 macrophages, leading to a reduction in size of AAA. M2 macrophages are divided into three phenotypes, namely, M2a, M2b, and M2c, and M2c has the strongest suppressive effect [30]. Several studies reported that polarization of the M2c phenotype was induced by IL-10, which is an antiinflammatory cytokine [30,31]. In addition, some reports showed that treatment with montelukast increased IL-10 levels in serum and inhibited inflammation [32,33]. Further experiments are needed to determine how M2 macrophage polarization is influenced by montelukast. The in vivo experiments in this study investigated Ang II-infused AAA in apoE −/− mice. Angiotensin II infusion promotes macrophage accumulation and a vascular inflammatory response in the adventitia of apoE −/− mice, a mechanism that is similar to that in human AAA models such as those involving calcium chloride or elastase [34]. We showed that orally administered montelukast suppressed the formation and progression of AAA, decreased the degradation of the medial elastin area, regulated the expression of inflammatory proteins in the aortic wall, and inhibited MMP-2 activity. Our findings suggest that montelukast suppresses the destruction of the extracellular matrix in the aortic wall by inhibiting the infiltration of inflammatory factors and attenuating the activity of MMP, resulting in the prevention of AAA formation. Given that montelukast significantly promotes TIMP-2, whose primary role is to regulate MMP-2 enzyme activity, the drug might suppress MMP-2 by affecting TIMP-2. This mechanism differs from that of doxycycline, which is a potent antibiotic against microbial infections and which directly inactivates MMPs by combining with their active zinc site [12,35]. The finding that montelukast had no significant effect on MMP-9 activity in our in vivo experiment differs from the results of previous studies [11,27]. In our in vitro experiment, montelukast significantly suppressed MMP-9 gene expression in macrophages. Although MMP-9 is secreted by a large number of cell types, including neutrophils, fibroblasts, and endothelial cells, no studies have investigated whether montelukast influences MMP-9 secretion by each of these cell types. Thus, it is possible that MMP-9 secretion by these cells might have impacted our results.
Limitations of the current study should be mentioned. First, it is important to note that the in vivo dose of montelukast used in this study was higher than the dose generally administered for treatment of asthma. Referring to previous reports [18] and considering that montelukast has few side effects, we selected a dose of 10 mg/kg/day in mice, while in humans the standard dose is 10 mg/day in adults and 5 mg/day in children. Second, although Ang II-infused AAA in apoE −/− mice was used as a model in this study because the underlying pathophysiologic mechanism resembles that of human AAA, we must verify the efficacy of montelukast in another animal model, such as a calcium chloride-induced or elastase-perfusion mouse model.

Conclusions
Our results suggest that montelukast induces M2 macrophage polarization and prevents AAA formation in apoE −/− mice. Montelukast might have therapeutic potential in preventing the development of AAA.

Data Availability
All data generated or analyzed during this study are included in this article.

Conflicts of Interest
The authors declare that they have no conflicts of interest.