Curcumin longa extract-loaded nanoemulsion improves the survival of endotoxemic mice by inhibiting nitric oxide-dependent HMGB1 release

Preparation of the C. longa extract-loaded nanoemulsion (CLEN)

The extract of C. longa was prepared by using 50% ethanol for 24 h. The concentrated extract contained 10.43 mg/mL curcumin. The oil phase was prepared by dissolving 173.92 g C. longa extract with 86.96 g MCT (medium chain triglyceride) oil containing 26.09 g soy lecithin. The aqueous phase was prepared by mixing 17.39 g Tween 80, 173.9 mL C. longa extract, and 869.57 mL distilled water. A coarse emulsion was prepared by magnetic stirring at room temperature for 2 h. Nanoemulsions were prepared by further homogenizing the coarse emulsion. For this, the emulsion was first subjected to high speed homogenization at 5,000 rpm for 10 min, then to ultrasonication with a Vibra Cell (VCX-750, Sonics & Materials, Inc., Sandy Hook, CT, USA) for 15 min, and finally to high pressure homogenization under 10,000 psi for three cycles. The CLEN powder was prepared by spray drying 1,000 mL CLEN and 113.06 g dextrin.

Characterization of CLEN

The mean droplet size and polydispersity index (PDI) of CLEN were determined by dynamic light scattering using a particle size analyzer (ELSZ-1000; Otsuka Electronics Co., Ltd., Osaka, Japan). Thus, 1 mL CLEN was added to a polystyrene latex cell, and the mean droplet size and PDI were measured at 25 °C with a detector angle of 90° and wavelength of 633 nm. The curcumin content in CLEN was measured by UV-Vis spectrophotometry at 425 nm.

Cell culture

Murine macrophage RAW264.7 cells were obtained from the Korean Cell Line Bank (Seoul, Korea). The cells were maintained in a humidified atmosphere of 5% CO₂ at 37 °C in Dulbecco’s modified Eagle’s medium containing 10% heat-inactivated fetal calf serum, 100 U/ml penicillin, and 100 µg/ml streptomycin.

MTT assay

RAW264.7 cells seeded into a 24-well plate were incubated with various concentrations of CLEN for 24 h, after which the MTT solution was added to the culture medium. After incubation for 4 h, the medium was removed, the crystal formazan was dissolved, and the absorbance at 570 nm was determined.

Measurement of released HMGB1

The HMGB1 released into the culture medium was measured as described previously (Hwang et al., 2015). Briefly, RAW264.7 cells grown to sub-confluency were maintained in serum-free medium for 24 h and then incubated with or without LPS in the presence or absence of CLEN for 24 h. Aliquots of conditioned culture media from equal numbers of cells were precipitated with 80% cold acetone, after which the resulting pellets were resuspended in SDS-PAGE sample buffer. The levels of HMGB1 released into the culture medium were analyzed by immunoblot.

Western blot analysis

The protein expression was measured by immunoblot analysis as described previously (Hwang et al., 2015). Briefly, RAW264.7 cells were incubated with the indicated reagents for the indicated time periods and then lysed in PRO-PREP Protein Extraction Solution (iNtRON Biotechnology, Seoul, Korea). Aliquots of whole-cell lysates or conditioned media were subjected to Western blot analysis with specific antibodies.

Measurement of NO

NO formation was determined by spectrometry that measured the levels of nitrite (an oxidized form of NO) that had accumulated in the culture medium (Seo, Nishinaka & Yabe-Nishimura, 2000). After incubation for the indicated time period, 100 µl aliquot of conditioned culture medium was reacted with 100 µl of Griess reagent.

Animal model of endotoxemia and the survival test

The Institutional Animal Care and Use Committee of Konkuk University approved all of the present animal studies (approval number: KU15140). Endotoxemic animal models were generated by intraperitoneally injecting male or female 7-week-old BALB/c mice (20–25 g) with bacterial endotoxin (10 mg/kg E. coli LPS 0111:B4) as described previously (Hwang et al., 2015). Briefly, mice were randomly assigned to the following four groups: vehicle (DMSO), 10 mg/kg LPS, 10 mg/kg LPS plus 50 mg/kg CLEN, 10 mg/kg LPS plus 50 mg/kg CLEN plus 2.5 mg/kg sodium nitroprusside (SNP), 10 mg/kg LPS plus 2.5 mg/kg SNP, 50 mg/kg CLEN, and/or 2.5 mg/kg SNP.

Determination of serum HMGB1 levels and tissue iNOS expression

The HMGB1 levels in the serum of the septic mice described above were determined by collecting the blood 18 h after LPS injection and subjecting an aliquot of serum to immunoblot analysis of HMGB1. To measure iNOS expression, the tissues from the septic mice were ground under liquid nitrogen and lysed in PRO-PREP Protein Extraction Solution (iNtRON Biotechnology). Aliquots of the tissue lysates were then subjected to Western blot analysis to determine the iNOS expression levels.

Real-time polymerase chain reaction (PCR)

The levels of each mRNA were analyzed by real-time PCR. Total RNA was extracted using TRIzol reagent (Invitrogen, Carlsbad, CA, USA) and reverse-transcribed into cDNA by a TOPscript RT DryMIX kit (Enzynomics, Seoul, Korea). Equal amounts of cDNA were amplified in a 10 µl reaction solution containing 1  × SYBR PCR master mix (Takara Bio Inc., Otsu, Japan) and 10 µM primers. Real-time PCR was carried out by a Rotor Gene RG-3000 (Corbett Life Science, Sydney, Australia) according to the PCR conditions as follows: initial denaturation for 5 min at 95 °C, followed by 35 cycles of 20 s at 95 °C, 50 s at 58 °C, and 40 s at 72 °C. The primers were tumor necrosis factor (TNF)- α, 5′-TCCCAAATGGCCTCCCTCTCA-3′ and 5′-GTGGTTTGCTACGACGTGG-3′; interleukin (IL)-1 β, 5′-AAACCTTTGACCTGGGCTGTCC-3′ and 5′-TGCCTGCCTGAAGCTCTT GTT-3′; monocyte chemotactic protein (MCP)-1, 5′-TGCAGTTAACGCCCCACTCAC-3′ and 5′-CAGCTTCTTTGGGACACCTGC-3′; GAPDH, 5′-CATGGCCTTCCGTGTTCCTA-3′ and 5′-CCTGCTTCACCACCTTCTTGAT-3′.

Statistical analysis

The data are presented as mean ± standard error. Differences were assessed for significance by a one-way ANOVA followed by the Tukey–Kramer test (p < 0.05).

Characterization of CLEN and the CLEN powder

We analyzed the particle size and size distribution of the nanoemulsion and the nanoemulsion powder because these crucial variables indicate the stability, solubility, release rate, and bioavailability of nanoparticles for encapsulation (Tamjidi et al., 2013). As shown in Fig. 1, the size distribution of both CLEN and the CLEN powder showed a single spectrum. This indicates that the particle sizes of CLEN and the CLEN powder were uniformly distributed, although the distribution of the CLEN powder was broader than that of CLEN. The mean droplet size of CLEN and the CLEN powder were 323.6 nm and 492.2 nm, respectively.

Effect of CLEN on LPS-stimulated release of HMGB1 by RAW264.7 cells

To determine the optimal dosage, RAW264.7 cells were treated with 0, 0.1, 1.0, 5.0, 10, or 20 µg/ml of CLEN for 24 h. Although CLEN did not exhibited any cytotoxic effects in the concentrations below 5 µg/ml, the cell viability was significantly reduced in cells treated with 10 or 20 µg/ml CLEN (Fig. 2A). Thus, we chose the submaximal dosage of 5 µg/ml CLEN for most experiments on RAW264.7 cells.

We then assessed the anti-inflammatory activity of CLEN by culturing RAW264.7 cells with LPS in the presence and absence of CLEN and then examining their release of HMGB1. LPS increased the release of HMGB1 into the extracellular milieu and this effect was significantly suppressed by treatment with CLEN (Fig. 2B).

Effect of CLEN on the NO production and iNOS expression of LPS-treated RAW264.7 cells

NO is reported to mediate the LPS-induced release of HMGB1 by RAW264.7 cells (Jiang & Pisetsky, 2006). Therefore, we examined whether CLEN treatment alters the NO production of LPS-treated RAW264.7 cells. While LPS significantly upregulated NO formation, concomitant CLEN treatment reduced it in a dose-dependent manner: the maximal inhibitory effect was observed with 5 µg/ml CLEN (Fig. 3A). Consistent with this result, CLEN treatment also suppressed the LPS-induced expression of iNOS, TNF- α, IL-1 β, and MCP-1 by RAW264.7 cells (Figs. 3B and 3C).

To confirm that CLEN inhibits the LPS-induced release of HMGB1 by suppressing NO formation in RAW264.7 cells, the cells were treated with LPS with and without CLEN in the presence or absence of the iNOS inhibitor ^NG-monomethyl-L-arginine (L-NMMA). Indeed, L-NMMA significantly potentiated the ability of CLEN to suppress the LPS-induced secretion of HMGB1 (Fig. 3D). This suggests that CLEN inhibits HMGB1 release by suppressing the NO signaling cascade in RAW264.7 cells exposed to LPS.

Role of mitogen-activated protein (MAP) kinases in CLEN-mediated inhibition of HMGB1 release

To identify the signaling cascade(s) involved in CLEN-mediated inhibition of HMGB1 release by RAW264.7 cells, we analyzed the effect of CLEN on the LPS-induced activation of three MAP kinases, namely, extracellular signal-regulated kinase (ERK), p38, and JNK. As shown in Fig. 4A, LPS activated all three MAP kinases, and their activation peaked at 15 min after stimulation. p38 and ERK activation was sustained for up to 60 min, whereas JNK activation was more transient. Concomitant treatment with CLEN significantly suppressed JNK activation, but not p38 and ERK activation (Fig. 4B). Thus, CLEN appears to inhibit HMGB1 release by suppressing the JNK-mediated signaling cascade.

To verify this, we treated LPS-stimulated RAW264.7 cells with CLEN in the presence and absence of SP600125, which is a specific inhibitor of JNK. As shown in Fig. 5, LPS induced the HMGB1 release in a parallel pattern of NO formation and iNOS expression, whereas CLEN significantly suppressed the release of HMGB1, with concomitant reduction in NO level and iNOS expression. This CLEN-mediated inhibitory activity is further potentiated in the presence of SP600125, suggesting that CLEN inhibits the LPS-induced release of HMGB1 by suppressing the JNK-mediated NO signaling cascade.

Effect of CLEN on endotoxin-induced death in a murine model of septicemia

Our findings led us to experiments in mice with endotoxemia, which is a model of systemic inflammation. These mice were used to examine whether CLEN treatment can protect the mice from death by suppressing HMGB1 release into the circulation during endotoxemia. Thus, the male or female mice were injected with LPS in the presence or absence of CLEN and/or SNP to verify gender difference. LPS alone led to the death of all mice (100%) within 72 h of the injection. However, concomitant i.p. treatment with CLEN reduced the mortality rate to 50%. This protective effect of CLEN was not observed in mice treated with SNP, indicating that CLEN-mediated blockade of NO signaling is critical in LPS-induced lethality. Interestingly, there were no gender difference in the inhibitory effects of CLEN against LPS-induced lethality (Fig. 6). Moreover, all of the CLEN-treated animals who survived continued to survive up to 15 days after LPS challenge; there were no late deaths. This suggests that CLEN treatment protected the mice from lethal endotoxemia.

Since HMGB1 exacerbates inflammatory responses during LPS-induced pathogenic endotoxemia (Wang et al., 1999), we examined the effect of CLEN on the circulating levels of HMGB1. While the circulating levels of HMGB1 rose following LPS challenge, concomitant CLEN treatment significantly reduced these levels (Fig. 7A). We also found that concomitant CLEN treatment markedly inhibited the LPS-induced iNOS expression in various tissues, including the heart, lung, liver, and kidney (Fig. 7B). These data suggest that CLEN protected the mice from endotoxin-induced death by blocking NO signaling, thereby suppressing the release of HMGB1 into the circulation.