Anti-Inflammatory Effects of Licania macrocarpa Cuatrec Methanol Extract Target Src- and TAK1-Mediated Pathways

In this study, we investigated the anti-inflammatory effects of Licania macrocarpa Cuatrec methanol extract (Lm-ME) in vitro and in vivo and found pharmacological target proteins of Lm-ME in TLR4-mediated inflammatory signaling. This extract reduced NO production and mRNA expression of inflammatory cytokines such as iNOS, COX-2, IL-6, and IL-1β. In the NF-κB- and AP-1-mediated luciferase reporter gene assay, transcription factor activities decreased under cotransfection with MyD88 or TRIF. Phosphorylated protein levels of Src, PI3K, IKKα/β, and IκBα as well as p50 and p65 in the NF-κB signal pathway were downregulated, and phosphorylation of TAK1, MEK1/2, MKK4/7, and MKK3/6 as well as ERK, JNK, and p38 was decreased in the AP-1 signal pathway. Through overexpression of HA-Src and HA-TAK1, respectively, Lm-ME inhibited autophosphorylation of overexpressed proteins and thereby activated fewer downstream signaling molecules. Lm-ME also attenuated stomach ulcers in an HCl/EtOH-induced acute gastritis model mice, and COX-2 mRNA expression and phosphorylated TAK1 levels in gastric tissues were diminished. The flavonoids kaempferol and quercetin were identified in the HPLC analysis of Lm-ME; both are actively anti-inflammatory. Therefore, these results suggest that Lm-ME can be used for anti-inflammatory remedy by targeting Src and TAK1.


Introduction
Inflammation is the first defense against many infectious pathogens or injury, and it plays an important role in common causes of atherosclerotic cardiovascular disease, cancer, and chronic obstructive lung disease [1][2][3]. In particular, toll-like receptors (TLRs) are key factors of the innate immune system that also recognize pathogen-associated molecular patterns (PAMPs), which are molecular motifs shared by different types of microbes [4,5]. TLRs are expressed on dendritic cells, macrophages, natural killer cells, T cells, B cells, epithelial cells, endothelial cells, and fibroblasts [6]. Four particular adapter molecules of TLRs are known to be involved in signaling: MyD88, TIRAP, TRIF, and TRAM [7].
TLR4 recognizes microbial components, including lipopolysaccharides (LPS) and saturated fatty acids (SFA) [8], and can induce signaling through MyD88 and TRIF [9]. In the case of TLR4 activation, there are two different ways to activate inflammation: a MyD88-dependent pathway and a MyD88-independent pathway. In the case of the MyD88dependent pathway, MyD88 recruits IRAK4 to activate and degrade IRAK1. IRAK1 is associated with TRAF6, and the complex can activate TAK1, which activates the IκB kinase (IKK) and mitogen-activated protein kinase (MAPK). is IKK-related complex can activate NF-κB, and the MAPKrelated protein activates the AP-1 pathway [8,10]. In the MyD88-independent pathway, however, TRIF takes part in late-phase NF-κB and MAPK signaling, because TRIF recruits TRAF6 and receptor-interacting protein 1 (RIP1) to activate TAK1, resulting in NF-κB activation. Additional TLR4 downstream signaling molecules PI3K and AKT also activate the NF-κB pathway [11][12][13].
NF-κB and AP-1 transcription factors can induce inflammatory cytokines such as TNF-α, IL-1β, IL-6, iNOS, and COX-2 [14,15]. ese cytokines can induce pathological pain and promote inflammation [16]. Nitric oxide (NO), in particular, is produced by inducible nitric oxide synthase (iNOS) and plays roles in blood pressure regulation, inflammation, infection, and the onset of progression of malignant disease [17]. Poor regulation of these inflammation symptoms can lead to many diseases, including cardiovascular disease, cancer, rheumatoid arthritis, allergy, asthma, and Alzheimer's and chronic kidney disease [18]. Accordingly, many researchers have focused on inflammatory signaling targets and also identified candidate substances that prevent inflammatory cytokine production and inflammation-related protein or gene expression.
Licania macrocarpa Cuatrec belongs to the genus Licania (chrysobalanaceae), which consists of more than 200 species distributed mainly in the Americas, especially from Panama to Peru [19]. Licania species have been used as medicinal plants in South America to treat inflammation, diabetes, stomach ailments, diarrhea, and dysentery [20]. Despite traditional use, the medicinal mechanisms of the genus Licania have not been well characterized at the cellular and molecular levels. erefore, we focused on the anti-inflammatory effects of a methanol extract of L. macrocarpa Cuatrec (Lm-ME) both in vitro, by using macrophage-like RAW264.7 cells, and in vivo, by using an HCl/EtOH-induced acute gastritis mouse model. To check for anti-inflammatory effects, we conducted an NO assay in macrophage-like RAW264.7 cells and peritoneal macrophages and also examined inflammation-related mRNA expression of cytokines such as IL-6, IL-1β, iNOS, and COX-2. To find target proteins that are affected by Lm-ME in the TLR4 pathway, we utilized a luciferase reporter gene assay, western blotting, and an overexpression strategy. To identify the active components of Lm-ME, we carried out HPLC analysis using flavonoids related to anti-inflammatory effects.

Cell
Culture. RAW264.7 cells were cultured in RPMI 1640 with 10% heat-inactivated FBS and antibiotics (penicillin and streptomycin). HEK293T cells were cultured in DMEM with 5% heat-inactivated FBS and antibiotics (penicillin and streptomycin). All cells were incubated at 37°C under 5% CO 2 .

Lm-ME Treatment and Administration.
A stock of Lm-ME (25.93 g) was dissolved in 100% DMSO (1.10 g/ml) to make 100 mg/ml of Lm-ME for in vitro experiments. Lm-ME was further diluted to 50-200 μg/ml for treatment with medium. For in vivo experiments, the stock of Lm-ME was suspended in 0.5% CMC at a concentration of 100 mg/kg and 200 mg/kg.

Cell Viability Test.
e cytotoxic effects of Lm-ME in RAW264.7, HEK293T, and peritoneal macrophage cells were determined by conventional MTT assay as previously described [21].

High-Performance Liquid Chromatography (HPLC).
To identify the active components of Lm-ME, high-performance liquid chromatography (HPLC) was conducted as reported previously [22]. Standard compounds quercetin, kaempferol, and luteolin were used.

Luciferase Reporter Gene Assay and Transfection.
For the luciferase assay, HEK293T cells (1.25 × 10 5 cells/well) were plated in 24-well plates for 18 h. Luciferase genes (NF-κB or AP-1), signaling construct (MyD88 or TRIF) and the β-galactosidase gene (as a control) were mixed together in Opti-MEM media for 15 min, and the transfection reagent, PEI, was diluted with Opti-MEM for 15 min. After stabilization of the genes and PEI, both were mixed together for 20 min. e mixture was then added to HEK293T cells for 24 h. For transfection of Src and TAK1, HEK293T cells (2.5 × 10 5 cells/ml) were plated in 6 wells for 18 h. e HA-Src or HA-TAK1 genes were diluted in Opti-MEM, and PEI was also diluted in Opti-MEM, each for 15 min. After stabilization of the genes and PEI, they were mixed together for 20 min. e mixture was added to HEK293T cells for 24 h.

Preparation of Peritoneal Macrophages.
Peritoneal macrophages were collected from peritoneal exudates of ICR mice that were injected intraperitoneally with 4% thioglycollate broth (1 ml, Difco Laboratories, Detroit, MI, USA). After 4 days, the mice were anesthetized with isoflurane and sacrificed. Five ml of PBS was injected into the belly and massaged. After enough massage, PBS was collected into a 50 ml tube 4 times. Cells were pelleted by centrifugation (3,000 rpm, 5 min, 4°C). After discarding the supernatant, the pelleted cells were resuspended and washed with 10 ml of RBC buffer and centrifuged. After lysis of the red blood cells, the remaining cells were resuspended with RPMI media and plated into 96 wells.

HCl/EtOH-Induced Gastritis
Mice. Acute gastritis was induced with 300 μl of 150 mM HCl in 60% EtOH as previously reported in ICR mice [14]. We orally injected 0, 100, or 200 mg/kg of Lm-ME for the gastritis-induced groups and 40 mg/kg of ranitidine for the control group 3 times in ICR mice. Eight hour after the last oral injection, 300 μl of 150 mM HCl/60% EtOH was orally administrated for 1 h and ICR mice were anesthetized with isoflurane and sacrificed. e stomachs were excised, rinsed with PBS, and cut along the greater curvature. e area of mucosal erosive lesions was measured by Image J and counting the pixels of lesions.

Real-Time Polymerase Chain Reaction (Real-Time PCR).
e mRNA expression level of gastritis stomach sample was quantified in gastritis stomach samples. mRNA was isolated with TRIzol Reagent (Gibco) according to the manufacturer's manual. mRNA levels were quantified by real-time reverse transcription-polymerase chain reaction using SYBR premix Ex Taq (PCR Biosystems, London, UK). e realtime mRNA expression results were calculated relative to the housekeeping gene GAPDH. e primer sequences are listed in Table 2.

Statistical Analyses.
All data presented in this study are expressed as means ± standard deviation (SD) from three independent experiments. For statistical comparisons, ANOVA/Scheffe's post hoc test or the Kruskal-Wallis/ Mann-Whitney tests were employed. P values <0.05 were considered statistically significant.

Lm-ME Reduced Nitric Oxide (NO) Production.
To figure out the inflammatory effects of Lm-ME, we first checked the inhibitory effect on NO production in LPS-induced RAW264.7 cells and peritoneal macrophages (Figure 1(a)). We used L-NAME as a positive control, because L-NAME exhibits selectivity for inhibition of NOS [26], and we say that it decreased NO production ( Figure 1(b)). Importantly, Lm-ME (50-200 μg/ml) did not show any cytotoxicity in RAW264.7 cells, HEK293T cells, or peritoneal macrophages by MTT assay (Figure 1(c)), so NO decreases were not due to cell death. Next, to verify the components of Lm-ME that Table 1: Sequences of primers used in RT-PCR analysis.

Gene
Sequences Evidence-Based Complementary and Alternative Medicine 3 might be responsible for anti-inflammatory effects, we performed HPLC analysis using flavonoids, including quercetin, kaempferol, and luteolin, because they are known to have anti-inflammatory effects [27]. Lm-ME contained 2.287 mg/g of quercetin and 0.477 mg/g of kaempferol ( Figure 1(d)), implying that these flavonoids (quercetin and kaempferol) in Lm-ME suppressed LPS-induced NO production.

Anti-Inflammatory Effects of Lm-ME on Transcriptional
Level. Treatment of RAW264.7 cells with LPS can induce mRNA expression of proinflammatory molecules [28]. Accordingly, we investigated mRNA expression levels of proinflammatory genes. iNOS, COX-2, and IL-6 were strongly inhibited by 100 and 200 μg/ml Lm-ME, and IL-1β was also slightly inhibited by Lm-ME (Figure 2(a)). In addition, we checked the transcriptional inhibitory effect of Lm-ME using a luciferase reporter gene assay. We transfected MyD88 or TRIF, which are key molecules in TLR4 signals, into macrophages with NF-κB and AP-1 luciferase promoter genes [29]. Transfection of adaptor molecules (MyD88 and TRIF) in HEK293T cells mimicked the activation of RAW264.7 cells with LPS [30]. NF-κB luciferase activity was decreased by Lm-ME in a dose-dependent manner in both MyD88 and TRIF-transfected conditions ( Figure 2(b)). Moreover, AP-1 luciferase activity was decreased by Lm-ME in MyD88 and TRIF-transfected conditions ( Figure 2(c)). Based on the luciferase assay results, we next explored whether Lm-ME downregulated nuclear translocation levels of c-Jun and c-Fos. At 30 and 60 min, the Lm-ME-treated group had lower c-Jun and c-Fos protein levels in the nucleus (Figure 2(d)). e protein levels of MyD88 (by FLAG) and TRIF in luciferase assay looked to be similar between all groups (Figures 2(b)-2(e)). In conclusion, based on luciferase activity and nuclear c-Jun and c-Fos results, Lm-ME can alleviate the LPS-induced signaling pathway.

Regulatory Mechanism of Lm-ME in NF-κB and AP-1
Pathways. Since proinflammatory cytokines (e.g., IL-6 and IL-1β) are related to the NF-κB pathway, we expected Lm-ME to attenuate signal proteins passing through the NF-κB pathway. To explore the effect of Lm-ME on NF-κB signaling, we treated cells with Lm-ME at a concentration of 200 μg/ml, because 200 μg/ml of Lm-ME showed the highest NO inhibition without cytotoxicity (Figures 1(a) and 1(b)). When we checked western blotting in a time-dependent manner, the protein levels of phosphorylation of NF-κB transcription subunits p50 and p65 were downregulated starting 5 and 15 min after treatment with 200 μg/ml Lm-ME, and phospho-IκBα clearly decreased starting after 5 min (Figure 3(a)). In LPS induction, activation of c-Src and ubiquitous Src tyrosine kinase is required for the NF-κB pathway activation in macrophages [31]. To check responses over a short time period, we conducted western blotting after 3 and 5 min in LPS-treated RAW264.7 cells to determine the amount of phosphorylation of protein tyrosine kinase (Src) and phosphoinositide 3-kinases (p85). Both were suppressed by Lm-ME starting at 3 min (Figure 3(b)), implying that Src is a target protein of Lm-ME.

Anti-Inflammatory Effects of Lm-Me by Targeting Src and TAK1
Kinases. Based on previous western blotting results, we hypothesized that Src and TAK1 could be targeted by Lm-ME. To test this hypothesis, we utilized a strategy of overexpression of Src and TAK1 and determined whether autophosphorylation of Src and TAK1 and phosphorylation of downstream molecules decreased or not. First, we overexpressed HA-Src for 24 h in HEK293T cells and then treated the cells with 200 μg/ml of Lm-ME for 24 h. c-Src is autophosphorylated on Tyr-416, a residue in the middle of the carboxyl terminus [33], so we checked the levels of phosphorylated Src and p-p85, which is phosphorylated by p-Src. Following treatment with Lm-ME, p-Src and p-p85 levels decreased (Figure 4(a)). Next, we sought to verify whether Src inhibition could block NO production. We used PP2, which was developed to inhibit Src family members [34]. RAW264.7 cells were pretreated with 10, 20, or 40 μM of PP2 for 30 min and then treated with LPS for 24 h. As measured in the cell culture supernatant, PP2 blocked NO production without cytotoxicity (Figures 4(b) and 4(c)). In the AP-1 pathway, we assumed TAK1 could be the target protein of Lm-ME (Figure 3), so we overexpressed the TAK1 gene in HEK293T cells. TAK1 is also known as an  autophosphorylated kinase within its activation loop [35]. We determined that phosphorylated TAK1 was activated and that Lm-ME could block autophosphorylation of TAK1. In addition, MKK4, a downstream molecule, was also inhibited by Lm-ME (Figure 4(d)). We further tested whether the TAK1 inhibitor 5Z-7-oxozeaenol could also inhibit NO production. As expected, 5Z-7-oxozeaenol reduced NO production without cytotoxicity (Figures 4(e) and 4(f )). Based on these findings, we conclude that Lm-ME can block phosphorylation of Src and TAK1.  Figure 3: Inhibitory effects of Lm-ME on the NF-κB and AP-1 pathways. (a) RAW264.7 cells were pretreated 200 μg/ml of Lm-ME for 30 min, followed by 1 μg/ml of LPS for the indicated times (5, 15, 30, or 60 min). Cell lysates were prepared and the phosphorylated and total forms of IKKα/β, p50, p65, and IκBα were determined by western blotting analysis. (b) Lm-ME-pretreated RAW264.7 cells were exposed to LPS for the indicated times (3 or 5 min), and cell lysates were obtained. Phosphorylated and total forms of Src and p85 were checked by western blotting analysis. (c-e) RAW 264.7 cells were pretreated with 200 μg/ml of Lm-ME and incubated with LPS for the indicated times (3, 5, 15, 30, or 60 min). Phosphorylated and total forms of ERK, JNK, p38, MEK1/2, MKK4, MKK3/6, and MKK7 or TAK1, IRAK4, and IRAK1 were determined by western blotting analysis.

Effect of Lm-ME on HCl/EtOH-Induced Acute Gastritis.
To investigate whether Lm-ME has pharmacological effects in vivo, the HCl/EtOH-induced gastritis animal model was employed under oral administration conditions. Since HCl/ EtOH treatment causes injury and ulcer of the stomach, DAMPs could be released and seem to manage gastric inflammation [36,37]. Interestingly, it has been reported that DAMP (e.g., ATP and high mobility group protein B1 (HMGB1))-mediated inflammation induces inflammation by interaction with TLRs (e.g., TLR4) and subsequent  Evidence-Based Complementary and Alternative Medicine activation of the MyD88-dependent NF-κB signaling pathway [36]. In the end, DAMPs can induce inflammatory cytokines such as IL-6, IL-1β, and TNF-α [38]. So, we used HCl/EtOH-induced gastritis model to check anti-inflammatory effect of Lm-ME. e 200 mg/kg Lm-ME group had the fewest stomach inflammatory blood lesions compared to 100 mg/kg Lm-ME and ranitidine ( Figure 5(a)). In gastritis stomach samples, the level of COX-2 mRNA decreased after treatment with 200 mg/kg Lm-ME ( Figure 5(c)). We next analyzed gastritis protein levels of TAK1 in its total and phosphorylated forms. Phosphorylated TAK1 was decreased by treatment with 200 mg/kg Lm-ME and 40 mg/kg ranitidine ( Figure 5(d)). In conclusion, Lm-ME alleviated acute gastritis symptoms by inhibition of TAK1.

Discussion
e genus Licania displays the largest number of biological activities among Chrysobalanaceae species and is used widely in Venezuela for anti-inflammatory properties [39]. In Northeastern Brazil, Licania leaves have been used to treat diabetes, stomach aches, diarrhea, and dysentery [20]. However, the underlying anti-inflammatory mechanisms in Lm-ME in LPS-induced RAW264.7 cells and an HCl/EtOHinduced acute gastritis model have not previously been reported. erefore, this study focused on the effects and molecular target proteins of Lm-ME to better illuminate anti-inflammatory molecular mechanisms.
When RAW264.7 cells and peritoneal macrophages are treated with LPS, they produce NO [40]. We observed that  Figure 5: In vivo anti-inflammatory effects of Lm-ME. ICR mice were orally injected with 0, 100, or 200 mg/kg Lm-ME or 40 mg/kg of ranitidine 3 times over 2 days. Eight hours after the last oral injection, 300 μl of 150 mM HCl/60% EtOH was orally administrated for 1 h. Stomachs were excised from the mice, and stomach lesions imaged (a). By using Image J software, stomach lesions induced by HCl/EtOH were measured (b). (c) mRNA expression levels of COX-2 from gastritis samples were measured by quantitative real-time PCR. (d) Total and phosphorylated TAK1 levels from gastritis mice were determined by western blotting analysis.
Evidence-Based Complementary and Alternative Medicine 9 NO production was reduced by Lm-ME (Figure 1(a)) without any cytotoxicity in various LPS-stimulated cell types (Figure 1(b)). ese results indicate that Lm-ME can have anti-inflammatory effects on macrophage-like RAW264.7 cells and peritoneal macrophages. Lm-ME significantly decreased mRNA levels of proinflammatory cytokines such as iNOS, COX-2, IL-6, and IL-1β in LPS-induced RAW264.7 cells (Figure 2(a)). ese proinflammatory cytokines are related to inflammatory pain and disease [16]. IL-1β cytokines and IL-6 are related to rheumatologic autoimmune diseases, and many treatment strategies have targeted cytokine production. In addition, NO release and iNOS expression can be caused by osteoarthritis and systemic lupus erythematosus (SLE), as well as various types of intestinal inflammation [41]. We found that the HCl/EtOH-induced acute gastritis model of damage-associated inflammatory disease was palliated by Lm-ME, which reduced the mRNA levels of COX-2 and protein levels of phosphorylated TAK1 (Figures 5(c) and 5(d)). e regulatory effects of Lm-ME on NO production and expression of other cytokines suggest the possibility that Lm-ME can be used as an effective alternative remedy in inflammatory-related disorders.
Interestingly, Lm-ME targeted Src tyrosine kinase by blocking autophosphorylation of Src and suppressing downstream signaling molecules PI3K, IKKα/β, IκBα, p50, and p65 (Figures 3(a) and 3(b)). Central role of Src in controlling inflammatory signaling cascades for NF-κB activation has been previously reported [46,47]. Due to the functional response of Src in inflammatory signaling, inflammation-regulatory activities of Src including cytokine production, migration of macrophages, and inflammatory mediator production have been also found in LPS-treated macrophages [48,49]. Several compounds or extracts with Src inhibitory property were also proved as anti-inflammatory reagents [14,[50][51][52][53]. In the AP-1 pathway, Lm-ME suppressed activation of only TAK1 and not IRAK1 and IRAK4 (Figure 3(e)). IRAK is membrane associated as part of a complex (IRAK-TRAF6-TAK1-TAB2-TAB3). e important step in inflammation signaling requires IRAK1 to be consecutively phosphorylated on threonine 209, ubiquitinated and degraded.
is step allows the TRAF6-TAK1-TAB2-TAB3 complex to translocate from the membrane to the cytosol, and degradation is a necessary step in activation of the TAK1-dependent pathway [54,55]. TAK1 plays roles in both NF-κB and AP-1 pathways, which phosphorylate and activate IKK and MKK6 (an upstream molecule of JNK and p38) [56]. Indeed, it was reported that TAK1 inhibitor 5Z-7-oxozeaenol can prevent picryl chloride-induced ear swelling [57]. Some anti-inflammatory plants or compounds  Figure 6: Anti-inflammatory mechanisms of Lm-ME. Lm-ME targeted Src and TAK1 kinases in inflammatory NF-κB and AP-1 pathways. Consequently, the activation of transcription factors was blocked, and various inflammatory cytokines and mediators were reduced. such as Momordica charantia and torilin were found to suppress TAK1 as a pharmacological target [12,58], implying that this enzyme plays a positive role in inflammatory responses. Lm-ME suppressed the activation of TAK1 in both LPS/TLR4-activated macrophages (Figure 3(e)) and in HCl/EtOH-induced gastritis symptom ( Figure 5(d)), so it is possible that the observed downregulation of IKKα/β resulted from both Src and TAK1 inhibition.
In conclusion, we found that Lm-ME exerted anti-inflammatory effects by targeting Src-and TAK1-dependent pathways. Lm-ME decreased NO production and transcription of proinflammatory cytokines in vitro and also reduced stomach lesions, levels of COX-2 mRNA, and phosphorylated TAK1 in an HCl/EtOH-induced gastritis model in vivo. Based on these findings, L. macrocarpa Cuatrec could be a useful herbal medicine for the prevention of Src-and TAK1-related inflammatory diseases ( Figure 6).

Data Availability
e data used to support the findings of this study are available from the corresponding author upon request.

Conflicts of Interest
e authors have no conflicts of interest to declare.