Sulfoquinovosylmonoacylglycerols regulating intestinal inflammation in co-culture system from the brown alga Turbinaria ornata

General experimental procedures

Optical rotations were measured on a Jasco P-1010 polarimeter (Jasco Co., Ltd, Tokyo, Japan) with a 5 cm-length cell. IR spectrum was recorded on a Jasco FT/IR 4100 spectrometer. Nuclear magnetic resonance (NMR) spectra were measured on a Varian VNMRS 500 MHz spectrometer (Varian Technologies, Santa Clara, CA, USA) in CD₃OD. High resolution electrospray ionization mass spectra were obtained on an SCIEX X500R instrument and tandem mass spectrometry (MS/MS) data was measured on an SCIEX QTRAP 3200 instrument (Applied Biosystems/Sciex, Foster City, CA, USA). High-performance liquid chromatography (HPLC) experiments were performed on an Agilent 1200 system (Agilent, Waldbronn, Germany) consisting of a binary pump (G1312B), autosampler (G1329B), column oven (G1316B), and 1260 infinity evaporative light scattering detector (ELSD). The LPS (Escherichia coli 0127:B8), dimethyl sulfoxide, PMA (phorbol 12-myristate 12-acetate), and monoclonal anti-β-actin antibody were purchased from Sigma-Aldrich (St. Louis, MO, USA). MEM medium, RPMI 1640 medium, fetal bovine serum (FBS), penicillin / streptomycin, and phosphate-buffered saline (PBS) were purchased from Gibco Life Technologies (Grand Island, NY, USA). The DAF-DA assay was purchased from Sigma-Aldrich. Tin protoporphyrin IX (SnPP) was purchased from Porphyrin Products (Logan, UT, USA). Primary antibodies, including inducible nitric oxide synthase (iNOS), cyclooxygenase 2 (COX-2), p65, Lamin B, phospho-p65, IκBα, phospho-ERK, ERK, phospho-p38, p38, phospho-JNK, JNK, HO-1, and Nrf2 were purchased from Cell Signaling Technology, Inc. (Beverly, MA, USA). The secondary antibody was purchased from Jackson ImmunoResearch Laboratories, Inc. (West Grove, PA, USA).

Isolation of sulfoquinovosylmonoglycerols from the brown alga Turbinaria ornata

A specimen of the brown alga T. ornata was collected in Fulhadhoo, Maldives, in August 2017. The freeze-dried specimen (1.5 kg) was extracted with 90% aq. EtOH (2 L) twice at room temperature. The extract was concentrated in vacuo to give a semi-solid, which was partitioned into four fractions by solvent extraction: hexane, chloroform, ethyl acetate, and butanol. Among them, the butanol fraction (600 mg), obtained from the collaborator (Prof. You-Jin Jeon at Jeju University), was used. For the purpose of removing any lipids, the fraction was first chromatographed on Sephadex LH-20 eluting 100% MeOH to yield four fractions: M1 (150 mg), M2 (250 mg), M3 (155 mg), and M4 (50 mg). Fractions M1 and M4 were evaluated as inactive effect on NO inhibition, but M2 and M3 showed good activity. Partial fraction of M2 (60 mg) showing the highest NO inhibition effect and interesting ¹H NMR signal was separated again by HPLC using a column Phenomenex C8 (250 × 4.6 mm, 5 μm), gradient solvent system as [A: H₂O with 0.1% formic acid, B: acetonitrile, from 30% B to 50% B for 10 min, then 50% B to 100% B for 10 min], a flow rate of 1.0 mL min⁻¹, and an ELSD detector to yield compounds 1 (2.1 mg), 2 (1.2 mg), and 3 (10.4 mg) (Fig. 1).

Fig. 1

Structure of compounds 1–3 isolated from Turbinaria ornata.

Compound 1: a colorless oil; [α]_D²⁵ = + 32.8 (c 0.1, MeOH); ¹H and ¹³C NMR see Table 1; high resolution electrospray ionization mass spectroscopy (HRESIMS) m/z 597.2946 [M-H]⁻ (calcd. m/z 597.2950 [M-H] ⁻ for C₂₇H₅₀O₁₂S).

Table 1

The NMR spectral data for compound 1 (500 MHz for ¹H, CD₃OD)

Cell culture

The human epithelial cell line (Caco-2) and human monocytic cell line (THP-1) were obtained from the American Type Culture Collection (ATCC, Manassas, VA, USA). The Caco-2 cells were cultured in Eagle’s minimum essential medium (EMEM; Gibco BRL, Grand Island, NY, USA) containing 20% (v/v) FBS and 1% (v/v) antibiotics (100 U mL⁻¹ penicillin and 100 μg mL⁻¹ streptomycin). THP-1 cells were cultured in RPMI1640 (Gibco BRL) containing 10% (v/v) FBS, 1% (v/v) antibiotics (100 U mL⁻¹ penicillin and 100 μg mL⁻¹ streptomycin) and 0.05 mM 2-mercaptoethanol. The cells were incubated in a humidified atmosphere of 95% air-5% CO₂. After that, THP-1 cells were treated with PMA for macrophage differentiation as below.

Differentiation of THP-1 monocyte to macrophage

THP-1 cells were differentiated in medium containing 50 nM of PMA for three days followed by incubated in fresh medium without PMA for additional two days (Daigneault et al. 2010).

In vitro co-culture model of intestine

For establishment of co-culture system that mimics human intestine, Caco-2 cells were seeded at 3.75 × 10⁵ cells per well on transwell inserts (0.4 μm pore size; Corning CoStar Corp., Cambridge, MA, USA) and maintained in growth media EMEM for 14–20 days in an incubator at 37°C with a 5% CO₂ atmosphere. Culture medium was changed every 3 days until the cells were fully differentiated (TEER value > 1,200 Ω·cm²). THP-1 cells were independently seeded at 8.5 × 10⁶ cells into the bottom plate of 6-well transwell plate in growth media RPMI1640. Then, the insert that is polarized with Caco-2 monolayer was added. The upper chamber reflects the apical side and the lower chamber the basolateral side of the intestinal epithelium (Kleiveland 2015). For evaluation of anti-inflammatory activity of the test sample, 1 μg mL⁻¹ of LPS was added on the basolateral compartment of plate and then the test sample was added to the apical compartment of plate at the concentration indicated. After 24 h incubation, the contents of nitrite and inflammatory mediators produced within culture supernatant in the basolateral side were determined.

Measurement of TEER value

TEER value was measured once a week to test monolayer integrity using a Millicell-ERS (Millipore, Burlington, MA, USA). The TEER was calculated as following:

Cell viability assay

Cell viability was measured in THP-1 cells using CCK-8 assay (Cell Counting Kit-8, #CK04; Dojindo Laboratory, Kumamoto, Japan), according to the manufacturer’s instructions. Briefly, THP-1 cells were seeded at 1 × 10⁴ cells per well in 96-well plates and incubated for 24 h. Then, the cells were treated with test sample of various concentrations for 24 h. After incubation, 10 μL of CCK-8 solution was added to each well and incubated for 3 h at 37°C. The absorbance was measured using a microplate reader (Bio-Tek Instruments Inc., Winooski, VT, USA) at 450 nm. The experiments were performed in triplicate.

Measurement of NO and prostaglandin E₂ production

THP-1 cells were seeded at 1 × 10⁴ cells per well in 96-well plates and incubated for 24 h. The cells were treated with test sample for 1 h, followed by treated with LPS (10 μg mL⁻¹) and interferon-gamma (IFNγ; 10 ng mL⁻¹) for 24 h. The content of nitrite produced in culture medium was determined using DAF-DA assay. Briefly, cells were incubated with DAF-DA (3 μmol mL⁻¹) at 37°C for 20 min. Then washed the cells and replaced with fresh medium incubated for an additional 20 min. The fluorescence was measured at 495 nm of excitation and 515 nm of emission wavelengths with a microplate reader (Bio-Tek Instruments Inc.). The concentration of prostaglandin E₂ (PGE₂) was quantified using an enzyme-linked immunosorbent assay (ELISA) kit (R&D System, Minneapolis, MN, USA) according to the manufacturer’s instructions. The culture medium as applied in ELISA assay without dilution.

Western blot

Protein isolated form THP-1 cells incubated in co-culture system was applied to western blot assay. After incubation of cells with test sample and LPS as indicated, THP-1 cells were washed three times with ice-PBS and extracted with RIPA buffer containing a protease inhibitor cocktail (Santa Cruz Biotechnology, Santa Cruz, CA, USA) for 30 min on ice. Protein lysates were centrifuged at 13,000 ×g for 30 min at 4°C. Thirty micrograms of the lysed proteins quantified by Bradford assay were separated on sodium dodecyl sulfate polyacrylamide gel electrophoresis (8–12%) at 100 V and transferred to polyvinylidene difluoride membrane. The membranes were blocked with 5% nonfat milk in PBST buffer for 1 h at room temperature. The membranes were incubated with 1:1,000 diluted primary antibodies (Cell Signaling Technology, Inc., Danvers, MA, USA) at 4°C overnight. The membranes were washed three times with PBST buffer followed by incubated with 1:10,000 diluted secondary antibodies for 1 h at room temperature. Bands were visualized using ECL solution (Thermo Scientific, Rockford, IL, USA) and calibrated by the Chemidoc Imaging System (Bio-Rad, Hercules, CA, USA). The density value of the protein bands was normalized to Lamin B (nuclear) or β-actin (total protein or cytosol).

Statistical analysis

All data were through Shapiro-Wilk normality test and Levene’s equality of variance test by SPSS ver. 17.0 program (SPSS Inc., Chicago, IL, USA). Then, the results were analyzed and converted to graph using Prism ver. 5.00 software (GraphPad Software, San Diego, CA, USA); one-way ANOVA were used for parametric data or Kruskal-Wallis for non-parametric data with Dunnett or Dunns post hoc tests, respectively. Statistical significance is indicated by p < 0.05. Data were expressed as the mean ± standard deviation of three independent experiments.

Anti-inflammatory sulfoquinovosylmonoacylglycerols from the brown alga Turbinaria ornata using bioactivity-guided separation

Four fractions (M1-M4) of the butanolic layer of T. ornata extract separated by the Sephadex LH20 were applied to NO assay. Fraction M2 showed the most potent inhibitory activity on NO production induced by LPS. Hence, fraction M2 was further separated and purified using HPLC to yield several compounds including three sulfoquinovosylmonoacylglycerols(SQMGs). Compound 1 was obtained as a colorless oil and its molecular formula was determined as C₂₇H₅₀O₁₂S by the combination of negative HRESIMS and the ¹³C NMR spectrum. The ¹H NMR spectrum of 1 showed signals distributed in the range of 3.1–4.3 ppm, and a triplet at δ_H 2.36, a multiplet at δ_H 1.61, a broad singlet at δ_H 1.31 and a triplet peak at δ_H 0.89, indicating characteristic signals for a class of glycerolipids. An α-form glycopyranose was assigned by the COSY correlations from the anomeric proton at δ_H 4.77 (J = 3.7 Hz). The coupling constant of each proton and the carbon chemical shifts at C-6″, along with the molecular formula, suggested a 6-deoxy-6-sulfo-α-D-glucopyrose moiety. The chemical shifts of the protons and carbons of the unit were consistent with the those of the known ones (Hirao et al. 2012). Further analysis of the COSY and TOCSY spectra led to assign the position of one olefinic group within the long carbon chain to be Δ7′. This was also corroborated by the heteronuclear multiple bond correlation (HMBC) correlations from H-2′, H-3′, H-5′ to C-4′ and from H-4′, H-6′, H-7′ to C-5′. The COSY correlation from the olefinic protons (H-8′) to the remaining oxymethine proton allowed the hydroxy group to be attached at C-9′. In addition, the position of the functional groups in the carbon chain was determined by MS/MS analysis of the carbon chain cleaved by hydrolysis of 1 with 0.1 N HCl (Supplementary Figs S1–S16). The geometry of the olefinic group was established as E by the large proton coupling constant (J = 15.4 Hz). The observation of carbons paired from C-3′ to C-11′ revealed that compound 1 is a mixture of two epimers consisting of 9(R) and 9(S). Finally, HMBC correlations from H-3 to C-1″ and from H-1 to the carbonyl carbon at C-1′ established the planar structure of 1. Thus, compound 1 was determined as 1-O-[(7E)-9-hydroxyoctadec-7-enoyl]-3-O-(6-sulfo-α-D-quinovopyranosyl)-sn-glycerol. Compared with 1, compounds 2 and 3 did not display the downfield shifted proton signals corresponding to olefinic groups, but consisted of saturated acyl groups. From the analysis of NMR and HRESIMS, compounds 2 and 3 were identified as 1-stearoyl-3-O-(6-sulfo-α-D-quinovopyranosyl)-sn-glycerol and 1-palmitoyl-3-O-(6-sulfo-α-D-quinovopyranosyl)-sn-glycerol, respectively (Sakamoto et al. 2000, Hanashima et al. 2001).

Anti-inflammatory activities of the isolated compounds 1–3 were screened using THP-1 macrophage cells activated by LPS and IFNγ treatment. Prior to evaluating anti-inflammatory activities of 1–3, the cytotoxicities of the compounds were measured using CCK-8 assay. All compounds showed no cytotoxicity at the concentration range up to 10 μM (cell viability > 98% of control). To evaluate the inhibitory activity of 1–3 on NO production, THP-1 cells were treated with each compound (1, 5, and 10 μM) for 1 h and then challenged with LPS (1 μg mL⁻¹) and IFNγ (10 ng mL⁻¹). After 24 h of incubation, the content of NO that is released into culture medium was detected using DAF-DA. Among the compounds tested, 1 and 2 showed the significant inhibition on NO production induced by LPS and IFNγ, while 3 was inactive. Inhibitory activities of 1–3 on PGE₂ production were further investigated (Fig. 2). As a result, the pretreatment of cells with 1 or 2 for 24 h led to a significant decrease of LPS and IFNγ-induced PGE₂ production in a concentration-dependent manner, whereas a mild decrease by 3 was observed at the concentration of 10 μM.

Fig. 2

Cytotoxicity and the effects of compounds 1 (A), 2 (B), and 3 (C) to inhibit the production of nitric oxide (NO) and prostaglandin E₂ (PGE₂) in THP-1 cells. For cytotoxicity assay, THP-1 macrophages differentiated by PMA (phorbol-12-myristate-13-acetate) were treated with various concentration of each compound (1, 5, and 10 μM) for 24 h. For measurement of NO and PGE₂ content, THP-1 macrophages were treated with each compound (1, 5, and 10 μM) followed by co-treated with lipopolysaccharide (LPS) and interferon γ (IFNγ) for 24 h. CP, compound. Results are presented as the mean standard deviation of triplicate experiments. ***p < 0.001 compared to LPS-treated cells.

The effects of 1 and 2 on the expression of pro-inflammatory proteins and the phosphorylation of mitogen-activated protein kinase family in intestinal co-culture system

Recently, the advanced and novel co-culture model of Caco-2 and PMA-differentiated THP-1 cells that resembles human intestine has been established (Kämpfer et al. 2017). This in vitro model is featured by representing numerous hallmarks of intestinal inflammation. During developing co-culture model, TEER value is extensively used to measure the resistance of tight junctions of cell monolayer. The TEER was measured using an Ohm-meter to assess the tightness of the Caco-2 cell layer on the apical side of the insert as well as the barrier integrity throughout co-culture with THP-1 cells. In our co-culture system, TEER of Caco-2 layer subsequently increased at 4 days after plating and reached the maximum value at day 21 (Fig. 3). In this co-culture system, the effects of 1 and 2 on the expression of iNOS and COX-2 as well as the phosphorylation of mitogen-activated protein kinase (MAPK) family was measured to evaluate the action mechanism of the compounds regulating intestinal inflammation. As shown in Fig. 4, western blot analysis showed that the expression levels of iNOS and COX-2 were upregulated by co-treatment of LPS and IFNγ, whereas the pretreatment of cells with 1 or 2 (1, 5, and 10 μM) significantly attenuated this upregulation. In addition, the induced phosphorylation of p38, ERK1/2, and JNK induced by LPS and IFNγ was attenuated by pretreatment of 1 or 2. At the concentration of 1 μM, the significant decrease on phospho-ERK1/2 and phospho-JNK by 1 and 2, respectively, were observed (Fig. 5).

Fig. 3

Co-culture system consists of Caco-2 epithelial cells and PMA (phorbol-12-myristate-13-acetate)-differentiated THP-1 macrophage cells (left), and TEER values for 21 days (right).

Fig. 4

Compounds 1 and 2 down-regulated the expression of inducible nitric oxide synthase (iNOS) and cyclooxygenase 2 (COX-2) in intestinal co-culture system. Compounds 1 or 2 (1, 5, and 10 μM) were individually treated into the apical compartment of the Caco-2/THP-1 co-culture system. After 1 h, lipopolysaccharide (LPS; 1 μg mL⁻¹) and interferon γ (IFNγ; 10 ng mL⁻¹) were co-treated to the basolateral compartment. The expression levels of iNOS and COX-2 in cells were analyzed by western blot after 24 h of LPS + IFNγ challenge. (A) The representative blots. (B) The calculated intensities after normalization to β-actin. CP, compound. Results are presented as the mean standard deviation of triplicate experiments. ***p < 0.001 compared to LPS-treated cells.

Fig. 5

Compounds 1 and 2 attenuated the phosphorylation of mitogen-activated protein kinase (MAPK) family in intestinal co-culture system. Compounds 1 or 2 (1, 5, and 10 μM) were individually treated into the apical compartment of the Caco-2/THP-1 co-culture system. After 1 h, lipopolysaccharide (LPS; 1 μg mL⁻¹) and interferon γ (IFNγ; 10 ng mL⁻¹) were co-treated to the basolateral compartment. The phosphorylation levels of MAPKs (ERK, p38, and JNK) were analyzed by western blot after 15–30 min of LPS + IFNγ challenge. (A) The representative blots. (B) The calculated intensities after normalization to β-actin. CP, compound. Results are presented as the mean standard deviation of triplicate experiments. ***p < 0.001 compared to LPS-treated cells.

The effects of 1 and 2 on the nucleus translocation of nuclear factor-κB and its inhibitor, IκB-α in intestinal co-culture system

It has been known that interaction of p65 unit of nuclear factor-κB (NF-κB) with specific gene sets promotes subsequent transcription and expression of inflammatory mediators in macrophages (Aggarwal et al. 2004, Sanjeewa et al. 2019). Based on the inhibitory activities of 1 and 2 on the expressions of pro-inflammatory proteins and MAPK family, the effects of these compounds regulating p65 localization in nucleus and cytoplasm have been observed. After treatment of cells with compounds followed by LPS and IFNγ challenge, whole cell lysates were biochemically fractionated into nuclear and cytoplasmic extracts to measure localization of p65 and phosphorylated p65. As shown in Fig. 6, the treatment of cells with 1 or 2 (1, 5, and 10 μM) markedly inhibited the phosphorylation of p65 in cytoplasm and weakened the translocation of p65 into nucleus. Significantly upregulated total levels of IκB-α protein by 1 and 2 in the cytoplasm may explain the enhanced binding of IκB-α to p65, which indicates that NF-κB access to the nucleus binds to the promoter can be prevented. The increase of IκB-α level in cytoplasm was more significant by 1 at the concentration of 10 μM. These results suggest that interfering with the p65 subunit of NF-κB and a specific set of target genes by 1 or 2 results in subsequent down-regulation of transcription and expression of inflammatory mediators (Oeckinghaus and Ghosh 2009, Saegusa et al. 2010).

Fig. 6

Compounds 1 and 2 inhibited NF-κB translocation into the nucleus in intestinal co-culture system. Compounds 1 or 2 (1, 5, and 10 μM) were individually added into the apical compartment of the Caco-2/THP-1 co-culture system. After 1 h, lipopolysaccharide (LPS; 1 μg mL⁻¹) and interferon γ (IFNγ; 10 ng mL⁻¹) were co-treated to the basolateral compartment and the cells were incubated overnight. Cell extracts were biochemically separated into nuclear and cytoplasmic fractions, and the expression of p65, phosphorylated-p65, Lamin B, and IκBα in cells was analyzed by western blot. (A) The representative blots. (B) The calculated intensities after normalization to Lamin B (nucleus) or β-actin (cytosol and whole cell). CP, compound. Results are presented as the mean standard deviation of triplicate experiments. *p < 0.05 and ***p < 0.001 compared to LPS-treated cells.