Marine Microorganism-Derived Macrolactins Inhibit Inflammatory Mediator Effects in LPS-Induced Macrophage and Microglial Cells by Regulating BACH1 and HO-1/Nrf2 Signals through Inhibition of TLR4 Activation

Recently, many natural products with unique structure and promising pharmacological potential have been reported from marine-derived microorganisms. The macrolactin A (MA), 15-epi-dihydromacrolactin F (DMF) and macrolactin F (MF) were obtained from the culture broth extract of a marine sediment derived microorganism Bacillus sp. HC001. In this study, MA, DMF and MF inhibited the production and expression of proinflammatory mediators of inducible nitric oxide synthase (iNOS) and cyclooxygenase–2 (COX-2) in LPS-stimulated RAW264.7 and BV2 cells. Also, MA, DMF and MF exert anti-inflammatory effects through the expression of heme oxygenase (HO) -1, a stress-inducing enzyme that converts heme to carbon monoxide (CO), iron and biliberdine. Toll-like receptor 4 (TLR4) expressed by lipopolysaccharide (LPS) was inhibited by increased expression of HO-1 transcription factor Nrf2 and down regulation of BTB Domain And CNC Homolog 1 (BACH1), inhibited phosphorylation of Mitogen-activated protein kinase kinase kinase 7 (MAP3K7, TAK1) and nuclear factor kappaB (NF-κB). These results show that MA, DMF and MF effectively inhibited TLR4 by regulating BACH1 and HO-1/Nrf2 signals in LPS-stimulated RAW264.7 and BV2 cells, which suggests the possibility of use as an anti-inflammatory agent.


Introduction
Inflammation is the initial bodys defense response to tissue damage and foreign pathogens, which protects and regenerates tissue in response to damage and infection, while chronic inflammation results in loss of the immune system, resulting in tissue damage and the development of various diseases [1,2]. In the innate immune response, macrophages and microglia serve as the bodys first line of defense against pathogen invasion and play an important role in promoting cellular protection and repair [3,4]. The activation of macrophages and microglia plays an important role in host defense responses to infectious agents by releasing inflammatory cytokine tumor necrosis factor-alpha (TNF-α), interleukin and pro-inflammatory mediators nitric oxide (NO) [5,6]. Toll-like receptor 4 (TLR4) is a transmembrane protein of pattern recognition receptors for Lipopolysaccharides (LPS) from Gram-positive bacteria that   5.58 ppm, d, J = 11.5 Hz) to H-24 (δ H 1.25 ppm, d, J = 6.2 Hz) suggested a long chain type of compound structure. Furthermore, the proton chemical shift of H-23 (δ H 5.00 ppm) was downfield shifted compared to the three other oxygenated methines (H-7, H-13, H-15), suggesting the presence of a cyclic ester linkage between C-1 and C-23 in compound 2. The geometries of double bonds at C-2, C-4, C-8, C-10 and C-18 were identified as Z, E, E, Z, E on the basis of their vicinal 1 H-1 H coupling constants 11.5, 15.2, 15.2, 11.1 and 15.2 Hz in 1 H NMR spectrum, respectively. The absolute configuration of C-7, C-13 and C-15 were determined by modified Mosher's method. The compound 2 was derivatized with (R)-(-)-and (S)-(+)-α-methoxy-α-(trifluoromethyl)-phenylacetyl chloride (MTPA-Cl) in anhydrous pyridine-d 5 to afford the (S)-and (R)-MTPA esters (Figures S12-S14). The ∆δ S-R values of tri-MTPA derivatives of DMF indicated 7S, 13S and 15S configurations. Based on the comparison of chemical shifts and coupling constants of 2 with the reported macrolactins, the configuration of C-23 of 2 was speculated to be R. Previously, a 15-epimer of compound 2 was reported as the semisynthetic derivatives of macrolactin F prepared for the structure elucidation of macrolactin B [36]. The absolute configuration of semisynthetic dihydromacrolactin F was revealed to be 7S, 13S, 15R, 23R. However, the compound 2 was analyzed to be opposite configuration at C-15 and it was named as 15-epi-dihydromacrolactin F (Figure 1a-c). Compounds 1 and 3 were identified to be macrolactins A and C, respectively, based on the comparison of spectroscopic data including MS spectra as well as 1 H and 13 C NMR spectra [27].

Effect of MA, DMF and MF on Cell Viability of RAW264.7 and BV2 Cells
First, MTT analysis was performed to evaluate the effect of MA, DMF and MF on cytotoxicity in RAW264.7 and BV2 cells. After treatment with 5~40 µM MA, DMF and MF for 2 h and then treatment with LPS 1 µg/mL for 24 h, the cell viability of both RAW264.7 and BV2 cells did not affect compared to the control group ( Figure 1d). Therefore, subsequent experiments were conducted at concentrations ranging from 5~40 µM.

Effect of MA, DMF and MF on NO Production and Pro-Inflammatory Cytokines in LPS-Stimulated RAW264.7 and BV2 Cells
Next, pro-inflammatory cytokines are essential mediators for regulating host responses to inflammation. Therefore, we investigated the effects of MA, DMF and MF on NO and PGE 2 production and pro-inflammatory cytokine expression in LPS-stimulated RAW264.7 and BV2 cells.   MA, DMF and MF inhibited NO and PGE2 production in RAW264.7 and BV2 cells. Therefore, the effects of MA, DMF and MF on the expression of iNOS and COX-2 catalyzing the production of NO and PGE2 were evaluated by western blot analysis. As a result it was found that iNOS and COX-2, which were increased in RAW264.7 and BV2 by LPS, were inhibited in a concentration-dependent manner by MA (Figure 3a   Previous studies have shown that HO-1 expression mediates anti-inflammatory effects, and the nuclear translocation of activated Nrf2 plays an important role in the major HO-1 induction [37]. Therefore, we identified whether the anti-inflammatory activity of MA, DMF and MF could be associated with HO-1 induction. As a result, MA, DMF and MF increased HO-1 expression in a concentration-dependent manner, and 40 µm expressed HO-1 similar to cobalt protoporphyrin (CoPP), an HO-1 inducer (Figure 4a). In addition, MA, DMF and MF induced Nrf2 nuclear translocation in RAW264.7 and BV2 cells, while Nrf2 levels in nuclear fractions were increased while the cytosolic fraction was degradated ( Figure 4b).

MA, DMF and MF Inhibit TLR4 Expression by LPS and Regulate BACH1 and HO-1/Nrf2 Signals.
Activation of TLR4 signal transduction by LPS raises pro-inflammatory cytokines. Therefore, we investigated the effect of MA, DMF and MF on TLR4 protein expression by LPS. In addition, the effect of the interaction of Nrf2 and BACH1, a regulator of HO-1, on the expression of TLR4 protein were investigated. First of all, MA, DMF and MF concentration-dependently inhibited TLR4 protein expression induced by LPS (Figure 5a). In addition, BACH1 and TLR4 were down-regulated and Nrf2 was up-regulated according to the expression of HO-1 increased by MA, DMF and MF. These results were confirmed more clearly when the co-treated of HO-1 inducer CoPP and MA, DMF and MF ( Figure 5b).

Effect of MA, DMF and MF on LPS-Induced Phosphorylation of TAK1, MAPK and NF-κB
TAK1 is an upstream signaling molecule of NF-κB that regulates inflammatory genes and pro-inflammatory cytokines, and MAPK is known to be involved in the regulation of pro-inflammatory mediator expression in LPS-treated RAW 264.7 cells [38]. Therefore, we investigated the effect of MA, DMF and MF on the phosphorylation of TAK1, MAPK and NF-κB in LPS-stimulated RAW264.7 and BV2 cells through western blot analysis. MA, DMF and MF inhibited the expression of phosphorylated TAK1 (Figure 6a), ERK, JNK and p38 (Figure 6b), and also inhibited the expression of NF-κB p65 and phosphorylated IκBα (Figure 7a,b).  . Cells were pre-incubated with or without MA, DMF and MF for 2 h, following treatment with LPS for 30 min. Relative protein expressions of cytosolic and nuclear p65, IkBα and p-IkBα to those in control were subjected to western blot analysis. * p < 0.05 compared with LPS-treated group.

Discussion
The main pattern recognition receptor TLR4 in LPS-induced inflammation is involved in the expression of pro-inflammatory genes and secretion of pro-inflammatory molecules by NF-κB and MAPK activation, which plays an important role in the release of pro-inflammatory mediators such as TNF-α, IL-1β, IL-6 and PGE2 [39,40]. In this study, MA, DMF and MF inhibited the expression and production of pro-inflammatory cytokines and pro-inflammatory mediators in LPS-stimulated RAW264.7 and BV2 cells and inhibited TLR4 protein expression by LPS-stimulated expression. LPS, a major component of E. coli cell membranes, is widely used in in vivo and in vitro inflammation models, and studies previously reported that activation of MAPK and NF-κB by LPS regulates many genes in the inflammatory response [41]. In our results, MA, DMF and MF inhibited the phosphorylation of MAPK and NF-κB in inflammatory responses induced by LPS. Therefore, it is suggested that the inhibitory effect of the expression of pro-inflammatory cytokines and pro-inflammatory mediators are shown by inhibiting the activities of MAPK and NF-κB.
Inflammation is a host response to various stimuli, including oxidative stress and infection leading to the release of large amounts of inflammatory mediators, and in this pathophysiological state HO-1 is a potent antioxidant that is activated and exerts a cellular protective effect [42,43]. Nuclear transcription factor Nrf2 is known to be one of the important antioxidant regulators involved in maintaining the redox state for defense against oxidative stress in cells, and this defense mechanism is mediated transcription and protein of ARE-and Nrf2 dependent antioxidant factors including HO-1 and Gpx1 [44,45]. BACH1 is a heme sensing protein that loses its inhibitory activity upon heme levels and heme binding in cells, and high levels of heme induce degradation of BACH1 [46]. In particular, BACH1 and Nrf2 act as inhibitors and activators of the HO-1 gene, and previously reported that, during the signaling of TLR4 by LPS, the HO-1 expression is regulated by BACH1 and activates the Nrf2 antioxidant stress pathway [37]. Therefore, we investigated the correlation between BACH1 and Nrf2/HO-1 in RAW264.7 and BV2 cells activated with TLR4 signaling by LPS stimulation. As a result, MA, DMF and MF increased HO-1 expression in a concentration-dependent manner, and inhibited TLR4 activation by LPS. In addition, the increased HO-1 expression, according to the regulation of Nrf2 was shown to decrease the expression of BACH1, and this result became more apparent when CoPP (HO-1 inducer) is processed simultaneously. Nrf2/HO-1 signals have been shown to inhibit inflammation by modulating various inflammatory signals, including MAPK, STAT3 and AKT, among them activation of MAPK is induced by TAK1, and expression of TAK1 enhances expression of Nrf2 by activating antioxidant response element (ARE) [47]. TLR4 mediated intracellular signaling pathways in these inflammatory responses are important for induction of innate immune responses [7]. In particular, TLR4-mediated signaling pathways have been shown to inhibit the transcription of cytokines and chemokines that participate in the initiation or regulation of the inflammatory response by inhibiting activation of NF-κB and MAPK [48].

Summary and Conclusion
To summarize this study, the effects of microorganism-derived chemical MA, DMF and MF on LPS-stimulated RAW264.7 and BV2 cells were investigated. In the study, MA, DMF and MF influenced LPS-induced TLR4 activity and inhibition by regulating BACH1 and HO-1/Nrf2 signals. In addition, MA, DMF and MF down-regulated pro-inflammatory cytokines and mediators by inhibiting the phosphorylation of MAPK and NF-κB p65. Thus, this study suggests that marine microbial chemicals MA, DMF and MF are potential therapeutic agents for inflammatory diseases.

Isolation of Bacterial Strain and Taxonomic Identification
The microbial strain (HC001) was isolated from the marine sediment collected from Je-ju Island in 2013. The dried sediment was stamped on SYP SW solid medium (soluble starch 10 g/L; yeast extract 4 g/L; peptone 2 g/L; agar 16 g/L; sterilized and filtered sea water 1 L) plate for several days at 27 • C, and the observed colonies were isolated. One of the isolated strains was identified as Bacillus sp. HC001 based on 16S rDNA sequence (blast top strain: Bacillus amyloliquefaciens strain Lx-11, 99.79% similarity) and the strain was stored at College of Pharmacy, Yeungnam University.

Culture of Bacterial Strain
The strain Bacillus sp. HC001 was inoculated in SYP SW medium (soluble starch 10 g/L; yeast extract 4 g/L; peptone 2 g/L; sterilized and filtered sea water 1 L) as seed culture. The seed cultures were incubated in a shaking incubator with 150 rpm at 25 • C for 3 days. The seed cultures were inoculated in SYP medium (1 L of media in 2.5 L culture flask, totally 35 flasks) and incubated in a shaking incubator with 150 rpm at 25 • C for 7 days.

Extraction and Isolation
The culture broth of Bacillus sp. HC001 was extracted twice with the same volume of ethyl acetate (EtOAc) and the combined organic layer was concentrated in vacuo to give 1.9 g of crude extract. The crude extract was subjected to vacuum liquid chromatography (VLC) using step gradient elution with 0-100% MC in MeOH to obtain fractions A-G. fraction E (195.5 mg) was further separated using reverse-phase preparative HPLC (gradient elution, a mixture of 50% acetonitrile changed to 60% acetonitrile in water for 1 h, flow rate 6 mL/min) to afford macrolactin A (1, MA, 9.7 mg, Rt = 29.

NO Production and Assays
RAW264.7 and BV2 cells (1 × 10 6 cells/mL) in 96-well plates and incubated overnight. Samples of various concentrations were treated for 2 h, followed by LPS (1 µg/mL) for 24 h. The absorbance of the final product was then measured spectrophotometrically at 540 nm using an ELISA plate reader according to the depicted method. Nitrite concentration in the measured sample was calculated from the calibration curve.

Cytokine Analysis
RAW264.7 and BV2 cells cultured in 24-well plates (6 × 10 5 cells/well) were treated with MA, DMF, MF and LPS (1 µg/mL) for 24 h. The supernatant of each well was harvested; the PGE2 content was measured using the mouse PGE2 ELISA kit (R & D Systems) and the IL-6 and TNF-α contents were determined using the mouse IL-6 ELISA kit (R & D Systems) and mouse TNF-α ELISA kit (Enzo Life Sciences) according to the manufacturers protocol. The microplate reader was measured at a wavelength of 540 nm.

Western Blot Analysis
RAW264.7 and BV2 cells were treated with 5, 10, 20 and 40 µM of MA, DMF and MF for 2 h subsequent to the addition of 1 µg/mL of LPS. The protein samples were prepared using RIPA lysis buffer (Thermo Scientific, MA, USA), which contained protease and phosphatase inhibitors, and quantified using a BCA protein estimation kit (Thermo Scientific, MA, USA). A total of 30 µg of protein was loaded on 12% SDS PAGE and transferred onto a nitrocellulose (NC) membrane (Millipore, MA, USA). Then, the membrane was blocked with 5% skimmed milk in PBST (PBS containing 0.05% Tween-20) for 2 h, and incubated with the desired primary antibody overnight at 4 • C. Several groups of samples were separated by SDS-PAGE for western blot analysis and then incubated with primary antibody overnight at 4 • C. The secondary antibody was then incubated for 2 h at room temperature. Immune blotting reagents ECL-plus (GE Healthcare Life Science, Tokyo, Japan) were then used to detect immunoreactive bands using FujiFilm LAS4000. Gray values of the bands were quantified by J-Image software and quantified by β-actin.

Cytosolic and Nuclear Protein Extraction
RAW264.7 macrophages and BV2 microglia cells were seeded at 5 × 10 6 cells/mL. The harvested cells were then lysed on ice for 20 min with radioimmunoprecipitation assay (RIPA) buffer (Thermo Fisher Scientific, MA, Waltham, MA), and the isolated cytoplasm and nuclei were removed using the NE-PER nuclear and cytoplasmic extraction reagent kit (Pierce Biotechnology, Rockford, IL, USA) according to the manufacturers instructions. Each was used for western blot analysis.

Statistical Analysis
All experiments were presented as mean ± SD from at least 3 independent experiments, all statistical analysis was performed with SPSS 12.0.1 for Windows, using one-way analysis of variance (ANOVA). The p-value was considered statistically significant when it was less than 0.05.

Conflicts of Interest:
The authors declare that there is no conflict of interests.