Quercetin inhibits LPS-induced macrophage migration by suppressing the iNOS/FAK/paxillin pathway and modulating the cytoskeleton

ABSTRACT The natural flavonoid quercetin has antioxidant, anti-inflammatory, and anticancer effects. We investigated the effect of quercetin on lipopolysaccharide (LPS)-induced macrophage migration. Quercetin significantly attenuated LPS-induced inducible nitric oxide synthase (iNOS)-derived nitric oxide (NO) production in RAW264.7 cells without affecting their viability. Additionally, quercetin altered the cell size and induced an elongated morphology and enlarged the vacuoles and concentrated nuclei. Quercetin significantly disrupted the F-actin cytoskeleton structure. Furthermore, quercetin strongly inhibited LPS-induced macrophage adhesion and migration in a dose-dependent manner. Moreover, quercetin inhibited the LPS-induced expression of p-FAK, p-paxillin, FAK, and paxillin as well as the cytoskeletal adapter proteins vinculin and Tensin-2. Therefore, quercetin suppresses LPS-induced migration by inhibiting NO production, disrupting the F-actin cytoskeleton, and suppressing the FAK–paxillin pathway. Quercetin may thus have potential as a therapeutic agent for chronic inflammatory diseases.


Introduction
Macrophages play an essential role in innate immunity. Macrophages are activated by pathogen-associated molecular patterns (PAMPs) on pathogens or cell debris [1]. Activated macrophages secrete various inflammatory mediators, including nitric oxide (NO) and cytokines, which results in recruitment and activation of immune cells and communication with adaptive immunity by antigen presentation [2]. The recruitment of inflammatory cells to the infection site is vital for host defense. Disruption of this process can result in development of diseases, such as cancer, chronic inflammatory conditions, sepsis, atherosclerosis, and autoimmune disorders [3]. Therefore, regulators of macrophage mobility have therapeutic potential for chronic inflammatory diseases and cancer.
Focal adhesion kinase (FAK) is an intracellular tyrosine kinase with roles in cell motility, cell migration, and cell survival [4]. Indeed, FAK is required for macrophage motility. FAK expression is low in blood monocytes but increases in differentiated macrophages [5,6]. Loss of FAK expression reduces the motility of macrophages [7]. FAK activation is mediated by lipopolysaccharides (LPSs) and dependent on iNOS [3]. The phosphorylation of FAK by the steroid receptor coactivator family of protein tyrosine kinase (Src) at several tyrosine residues, including Tyr 925, affects formation of a complex with paxillin and vinculin and activates the downstream pathways (e.g., MAPK cascades) that control cellular morphology and motility [8][9][10][11][12]. Paxillin is a cytoskeletal protein that co-localizes with FAK at focal adhesion contacts [13]. Paxillin mediates cell movement by recruiting cytoskeletal elements and signaling molecules involved in cell attachment, spreading, and migration [14]. The FAK/paxillin interaction is modulated by the cytoskeletal adapter protein vinculin via the extracellular signal-regulated kinase (ERK) pathway [15]. Inhibition of FAK/paxillin in macrophages could prevent the development of inflammatory diseases and cancer [16,17].
Quercetin (Que; 3, 5, 7, 3, 3[3-pentahydroxyflavone]), which is a flavonoid present in fruits and Chinese herbs, has antioxidant [18] and anti-inflammatory effects [19,20]. Studies of the anti-inflammatory activity of quercetin have focused on inhibition of the NF-κB pathway, generation of proinflammatory cytokines, and release of granular components [21,22]. However, the effects of quercetin on LPS-mediated macrophage motility are unclear. Thus, in the current study we investigated the effect of quercetin on NO production and migration by LPS-activated macrophages. The results showed that quercetin inhibited LPS-induced macrophage migration by inhibiting NO production, modulating the F-actin cytoskeleton structure, and inhibiting the FAK-paxillin cascade.

Effect of quercetin on macrophage viability
Only the highest concentration of quercetin alone (39 μg/mL) reduced macrophage viability for 24h treatment; when administered together with LPS (1 μg/ mL) for 24h, LPS treatment did not exert a toxic effect on macrophages. Moreover, Quercetin at 4.9-39 μg/mL did not affect the viability of LPS-stimulated macrophages. Therefore, 4.9-39 μg/mL quercetin was used in subsequent experiments ( Figure 1).

Effect of quercetin on NO production
NO production was detected by Griess Reagent by measuring nitrite concentration in the culture medium. We observed that LPS (1 μg/mL) strongly stimulates NO production in macrophages after incubation for 24h without affecting the cell viability. Moreover, pretreatment of macrophages with quercetin (9.5-39 µg/ mL) for 24h reduced LPS-induced NO release in a concentration-dependent manner (Figure 2(a-b). Cell viability was not affected by quercetin for 24h treatment ( Figure 1(b), suggesting that the quercetinmediated inhibition of LPS-induced NO production was not due to cytotoxicity. Moreover, quercetin at 9.5-39 µg/mL ameliorated the LPS-induced increase in the iNOS mRNA level after 5h treatment.

Quercetin influences macrophage morphology after LPS stimulation
Quercetin influenced the morphology of LPS-treated RAW264.7 macrophages. LPS and quercetin treatment caused aberrations in cell morphology in a dose-dependent manner ( Figure 3). After 24 h, the control cells were confluent and round, whereas the LPS-treated  cells were attached, of increased size, and formed long, slim pseudopodia-like protrusions (Figure 3(b)). Addition of quercetin enhanced the severity of these morphological changes ( Figure 3(c-f)). The percentages of pseudopodia cell were significantly increased after LPS treatment, However, quercetin enhanced the percentages of pseudopodia cell in a dose dependent manner compared with LPS treatment (Figure 3(g)).
The cytoskeleton plays a vital role in cell movement and morphology. We reported previously that quercetin disrupts F-actin in L929 cells [23]; therefore, we investigated the effect of quercetin on cytoskeleton  structure. After incubation for 2 h, quercetin (37 μg/ mL) altered the actin cytoskeleton (Figure 4(a-d)). The F-actin structure exhibited marked changes after treatment for 48 h (Figure 4(e-h)); in contrast, actin filaments were evident at the periphery of the control cells.
Regarding ultra-microstructural changes, untreated macrophages were small and round, with a smooth cell membrane and small vacuoles ( Figure 5(a)). LPS treatment for 24 h resulted in a significant increase in the size of the macrophages and the formation of cell-surface protrusions ( Figure 5(b), blue arrow). Macrophages formed larger phagocytic vacuoles than the untreated macrophages ( Figure 5 (b), yellow arrow) and had a significantly increased number of Golgi apparati. After quercetin treatment, the cell size and the volume of the phagocytic bulbs increased compared to those of LPS-treated cells ( Figure 5(c), yellow arrow), and the nuclear membranes were folded and concentrated with fractured chromosomes ( Figure 5(c), white arrow). Therefore, LPS significantly increased the cell size; however, quercetin did not result in a further increase in cell size ( Figure 5(d)).

Quercetin inhibits macrophage adhesion after LPS stimulation
We investigated the effect of LPS and quercetin on cell adhesion ( Figure 6(a-b)). Pretreatment of macrophages with LPS and quercetin for 24 h significantly inhibited their adhesion in a dose-dependent manner (4.9-39 μg/ mL), compared with LPS treatment. Moreover, quercetin alone treatment inhibited macrophage adhesion in a dosedependent manner, compared to the control macrophages ( Figure 6(c-d)).

Quercetin modulates inos and fak-paxillin protein levels
Because quercetin significantly inhibited NO production and iNOS expression in LPS-treated macrophages ( Figure 2(a-b)), we determined the iNOS protein level in LPS-treated macrophages. Quercetin at 4.9-39 μg/ mL decreased the iNOS protein level in a dose-dependent manner (Figure 8(b)).
Quercetin inhibited the migration of macrophages, and the FAK-paxillin pathway is associated with cell motility [24][25][26]. Thus, we investigated whether quercetin inhibited cell migration by suppressing the production of proteins involved in the FAK-paxillin pathway (Figure 8(a,c-f)). Untreated macrophages express FAK and low level of paxillin protein. LPS treatment increased the p-FAK Tyr925 , FAK, p-paxillin- Tyr118 , and paxillin protein levels compared with the untreated control macrophages. Addition of quercetin significantly reduced the magnitude of the LPS-induced increase in the p-FAK Tyr25 , FAK, p-paxillin Tyr18 , and paxillin protein levels. Moreover, LPS increased the vinculin and tesin-2 protein levels, which were regulated by quercetin (Figure 8(g-h)). Therefore, the effect of quercetin on the migration of LPS-induced macrophages is mediated by its modulation of the FAK-paxillin pathway.

Quercetin regulates the FAK and paxillin protein levels
Immunofluorescence staining showed that LPS increased FAK and paxillin expression, whereas quercetin treatment decreased their expression at the cell periphery, Moreover, LPS increased the expression of F-actin structure in macrophages cells, whereas, quercetin strongly decreased the F-actin intensity compared with LPS treatment (Figure 9).

Discussion
Quercetin suppresses the LPS-induced inflammatory response via various mechanisms [18,22,27,28]; e.g., by inhibiting the NF-κB, STAT-1, and Syk/Src/IRAK-1 pathways. However, little is known about the effect of quercetin on cell motility. Macrophages play an important role in innate immunity, and their recruitment to sites of infection or injury is important for controlling infection and tissue repair. However, uncontrolled or abnormal accumulation of monocytes can lead to development of cancer or chronic inflammation. Therefore, we assessed the effect of quercetin on LPSinduced macrophages, focusing on their motility.
In accordance with our previous report [23], quercetin alone at 39 μg/mL reduced the viability of macrophages for 24h treatment by WST-8 assay. However, quercetin inhibited LPS-induced NO release by macrophages without affecting their viability. Moreover, quercetin reduced the LPS-induced iNOS production at the protein and mRNA levels. Quercetin exerts strong effects on LPS induced iNOS mRNA levels compared with NO production in the culture medium. This may due to the sensitivity of detection methods. Griess reagent detected the Nitrite level in the culture medium, which is one of two primary stable and nonvolatile breakdown products of nitric oxide. However, RT-PCR detected the iNOS mRNA expression in whole cells. This may lead to different pattern of quercetin effect. Sustained production of NO can lead to inflammation and tumorigenesis. NO release stimulates migration of several types of cells, such as primary aortic smooth muscle cells [29] and epithelial cells [30]. Moreover, LPS induces macrophage migration through the Src-FAK cascade, which is highly dependent on iNOS [3,12]. Therefore, inhibition of NO release could be a mechanism by which quercetin inhibits the migration of LPS-stimulated macrophages.
We next investigated whether the inhibitory effects of quercetin on LPS-induced expression of iNOS is associated with the decreased cell migration and FAKpaxillin expression. Cell migration is essential for angiogenesis and wound healing. It involves formation from the plasma membrane of filopodia or lamellipodia, which are supported by actin filaments and associated proteins [14]. Quercetin altered macrophage morphology by inducing some cells to form irregular shapes (e.g., diamond) or produce slim pseudopodialike protrusions. The morphology of intracellular structures, such as vacuoles and nuclei, was also influenced by quercetin. Moreover, quercetin disrupted the structure of F-actin and reduced FAK and paxillin protein levels at focal adhesion contacts at the leading edges of  macrophages. Therefore, quercetin impairs the formation of focal adhesions and actin bundles by suppressing the production of FAK and paxillin. This is consistent with a previous report that blocking the expression or function of FAK and paxillin reduced cell mobility [31]. Moreover, we previously reported that quercetin attenuates the phagocytosis of Candida albicans by macrophages by disrupting the F-actin cytoskeleton structure [23]. Moreover, quercetin can directly bind to actin [32,33]. Therefore, quercetin likely inhibits cell migration by altering the cytoskeleton.
We next evaluated the effect of LPS and quercetin on adhesion of macrophages. Quercetin inhibited macrophage adhesion in a dose-dependent manner, whereas LPS had no effect. This may be because the larger size of the LPS-treated cells resulted in fewer cells being able to attach to the same surface area compared to untreated cells. Moreover, LPS treatment increased cell motility; moving cells are less tightly attached to the substrate. These results confirm that quercetin disrupted the cytoskeleton and decreased cell motility.
The FAK pathway is reportedly important in LPS-activated macrophages [11]. We observed that untreated macrophages express FAK, the results of this study were consistent with the other study that expression of FAK is increased in differentiated macrophages [6]. Moreover, quercetin inhibited the LPS-induced expression of p-FAK, FAK, p-paxillin, and paxillin, as well as the vinculin and tensin-2 protein levels, in a concentration-independent manner. Therefore, quercetin influences LPS-induced cell migration by modulating the FAK/paxillin pathway. These results are in agreement with our in vitro finding that, in the presence of quercetin, LPS induces macrophage migration.
In conclusion, quercetin inhibits LPS-induced NO production, disrupts the F-actin cytoskeleton, and inhibits macrophage migration. Because abnormal accumulation of macrophages can contribute to disease progression (e.g., cancer and chronic inflammatory diseases), our data suggest that iNOS/FAK are targets of quercetin in the presence of LPS-induced inflammation. Furthermore, quercetin could be used to control infection and chronic inflammation, as well as to maintain tissue homeostasis.
Cell culture RAW 264.7 murine macrophages were purchased from the Type Culture Collection of the Chinese Academy of Sciences, Shanghai, China. The cells were cultured in DMEM supplemented with 10% FBS at 37°C in a 5% CO 2 atmosphere.

WST-8 assay
WST-8 assays were performed as reported previously [34]. RAW267.4 macrophages were seeded into 96-well cell culture plates at 5 × 10 5 cells/mL in DMEM containing the indicated concentrations of quercetin with or without LPS (1 μg/mL) for 24 h. Next, 10 μL WST-8 per well were added, and the plate was placed on a plate shaker for 1 min to ensure optimal mixing. After incubation for 30 min, the optical density (OD) at 450 nm was determined using a microtiter plate reader (Bio-Rad, Hercules, CA). The mean and standard deviation of three replicates were determined.

Cell morphology
RAW267.4 macrophages were seeded into 96-well cell culture plates at 5 × 10 5 cells/mL in DMEM containing the indicated concentrations of quercetin with or without LPS (1 μg/mL) for 24 h. Images of three randomly selected fields were obtained using an inverted microscope. The percentage of cells with pseudopodia was calculated using Image J software as the number of cells with pseudopodia ÷ the total number of cells ×100%.

Nitric oxide assay
The NO concentrations in cell culture supernatants were determined by measuring the accumulation of nitrite (NO 2 − ) [29]. Briefly, 100 μL of Griess reagent (1% sulfanilamide, 0.1% naphthylethylene diamine dihydrochloride, and 2.5% phosphoric acid) were mixed with an equal volume of culture supernatant in 96-well flat-bottomed microplates and incubated at room temperature for 10 min. The OD at 540 nm was read using a microtiter plate reader. Nitrite concentrations were determined from a standard curve established using serial dilutions of NaNO 2 .

Adhesion assay
Adhesion assays were performed as reported previously [35]. Cells were plated in 12-well plates at 5 × 10 5 cells/ mL for 12 h and subsequently incubated with quercetin or LPS (1 μg/mL) for 24 h. Cells were harvested and reseeded in 24-well plates for 3 h. Cells were washed and fixed with 4% PFA at room temperature for 15 min. Cristal violet staining solution (Beyotime, Beijing, China) was next added, and the plates were incubated for 10 min at room temperature. Images of five randomly selected fields were obtained using an inverted microscope. Cells were enumerated using Image J software. The percentage of adhesion compared to LPS treatment or control group was calculated.

Transwell assay
Transwell assays were performed in triplicate as reported previously [35]. Cells (4 × 10 4 /well) were placed in Transwell R cell culture chambers (8 mm pore size; Corning, Lowell, MA, USA). Cell suspension was placed in the upper chamber of a Transwell R insert and incubated with the indicated concentrations of quercetin and DMSO in serum-free DMEM. The lower chamber was filled with DMEM containing 20% FBS as a chemoattractant, and the system was incubated under the normal culture condition for 24 h. The lower surface of the membrane was treated with 4% PFA and stained with crystal violet. Cells in five random fields per chamber were counted using an inverted microscope and Image J software. The percentage of migration was calculated and compared to that under LPS treatment.

Transmission electron microscopy
Transmission electron microscopy (TEM) was performed as described previously [36]. Briefly, macrophages (1 × 10 6 cells/mL) were incubated for 24 h with or without LPS and quercetin (9.5 μg/mL) in six-well plates. The cells were next harvested and fixed in 2.5% glutaraldehyde for 2 h at 4°C. Samples were subsequently rinsed with 0.1 M PBS, fixed in 1% osmium tetroxide for 1-2 h, and dehydrated sequentially in 50, 70, 80, 90, and 100% ethanol for 15 min. The specimens were embedded in Epon™ and observed by TEM (CM100, Philips, The Netherlands). Cell size was calculated using Image J software by measuring cell diameters in four randomly selected fields.

F-actin cytoskeleton staining
Immunofluorescence staining for F-actin was performed as reported previously [23]. Macrophages (1 × 10 5 /mL) were seeded in a four-well plates with/ without quercetin (0.4, 4, and 37 μg/mL) and incubated for 2 and 48 h under standard conditions. The cells were washed and fixed with 3.7% PFA for 15 min at room temperature and then washed and permeabilized with 0.1% Triton X in PBS for 5 min. The cells were washed and incubated with Alexa Fluor® 488-phalloidin for 1 h, subsequently stained with DAPI (1 μg/mL), and visualized using a BZ-8000 digital fluorescence microscope equipped with BZ-analysis software (Keyence GmbH, Neu-Isenburg, Germany).

Statistical analysis
Data are presented as means ± SD of at least three experiments. A value of p < 0.05 by Student's t-test was taken to indicate statistical significance.