Introduction

Herbs have been extensively used for medicinal and edible purposes, and their consumption has increased significantly over the decades (Hussain et al., 2016). They are easy to consume with tea or food and have health benefits after a long-term intake (Sharifi-Rad et al., 2015). The development of functional foods using medicinal herbs can create enormous economic value in a short time. Medicinal herbs are also stable and have few side effects (Hussain et al., 2016). However, the bioavailability of some medicinal herbs such as ginseng, depends on the intestinal microbiome. This finding has application in bioconversion technology which involves the use of heat, enzyme treatment, and fermentation (Lee and Paik, 2017).

Inula britannica (IB) is a traditional oriental herbaceous plant of East Asia used for the treatment of bronchitis, digestive disorders, and inflammation. Various studies have shown that IB extracts have antibacterial, antitumor, antihistamine, antioxidant, and anti-inflammatory activities (Bai et al., 2005; Han et al., 2001; Khan et al., 2010; Lee et al., 2013). The flowers of IB have been used in the manufacture of medicinal products, beverages, and cottage cheese (Lee et al., 2016a). Furthermore, the antibacterial effects of IB have been reported on methicillin-resistant Staphylococcus aureus and Helicobacter pylori strains (Lee et al., 2016b). IB contains various biologically active substance such as sesquiterpene lactones (1-O-acetylbritannilactone (ABL), 1,6-O,O-diacetylbritannilactone, and eroglide), taraxasteryl acetate, patuletin, axillarin, nepitrin, and quercetin (Khan et al., 2010). ABL has been reported to have multi-functional properties including anti-inflammatory, antibacterial, antihepatitis, and antidiabetic properties (Zhengfu et al., 2015). Sesquiterpene lactone ergolide, which is known as the main compound in IB, has anti-inflammatory, apoptotic, and cytotoxic effects (Khan et al., 2010). Ergolide has been shown to suppress inducible nitric oxide synthase (iNOS) and cyclooxygenase-2 (COX-2), in addition to activating nuclear factor-kappa B (NF-κB) signaling (Han et al., 2001).

Probiotics are defined as useful microorganisms which can alter the human intestinal microbiota when ingested in a suitable amount (Lee et al., 2015). The metabolism of lactic acid bacteria (LAB) interacts with that of human nutrition in the intestine through β-glucosidase production. Plant metabolites are degraded by fermentation with probiotics or β-glucosidase treatment, leading to the enhancement of bioavailability and the reduction of bitterness through the production of aromatic compounds (Son et al., 2017). Foods that have been fermented using LAB include table olives, soybeans, ginseng, and wine. It has been reported that LAB eliminates bitterness and increase the production of isoflavone aglycone and aromatic compounds (Michlmayr and Kneifel, 2014). IB extract fermented with Lactobacillus plantarum KCCM 11613P has been shown to have an anti-microbial effect against S. aureus strains, and this antimicrobial effect was attributed to epicatechin (Bae et al., 2019).

Weissella cibaria is known as the main LAB involved in kimchi fermentation, and it has attracted increased attention as probiotic candidates (Lee et al., 2018). W. cibaria D30 was isolated from Korean kimchi, and its probiotic properties, antagonistic activity, and antioxidant effect have been demonstrated (Yu et al., 2019). In particular, W. cibaria D30 has been reported to have high β-glucosidase activity. In this study, IB was fermented with W. cibaria D30 (FIB), and its bioavailability, effect on cell viability, total phenolics, total flavonoids, and total sugar contents, and the production of ABL and eroglide were investigated. In addition, the viability and anti-inflammatory effects of FIB in RAW 264.7 cells were assessed.

Materials and methods

Preparation of IB and FIB extracts

IB was purchased from Kyungdong Market (Seoul, Korea). IB powder was diluted in 1000 mL of distilled water (1/10, w/v) and extracted by heating for 15 min at 121 °C. The extract was filtered through Whatman No. 2 paper (GE Healthcare, Chicago, IL, USA) using vacuum filtration. After lyophilization, the powder was stored at 4 °C.

Prior to fermentation, IB extract was adjusted to pH 6.5 using 5 M NaOH. W. cibaria D30 with probiotic potential was used for fermentation of IB (Yu et al., 2019). W. cibaria D30 was propagated in lactobacilli MRS broth (Becton, Dickinson and Company, Spark, MD, USA) for 24 h at 37 °C. The activated cultures were used for IB fermentation. IB extract (500 mL; 1/10, w/v) was inoculated with 3 mL of the inoculums (to 6 log CFU/mL), and cultures were incubated at 37 °C for 72 h in a shaking incubator at 50 rpm. The culture was then centrifuged at 10,000×g, 4 °C for 10 min. The supernatant was filtered through Whatman No. 2 paper using vacuum filtration, and lyophilized. The lyophilized extracts were stored at 4 °C and used for further study.

Cell viability of W. cibaria D30 during fermentation

The FIB was sampled at 0, 3, 6, 9, 12, 18, 24, 48, and 72 h during fermentation. The cell viability of W. cibaria D30 was determined by dilution of the cells, plating on MRS agar medium, and incubation at 37 °C for 24 h. Cell viability was presented as viable cell counts (log CFU/mL).

Total phenolic, flavonoid, and sugar contents

Total phenolic contents (TPC) were determined by the Folin-Ciocalteu assay (Singleton et al., 1999). Ninety microliters of 50% (v/v) Folin-Ciocalteu’s reagent (St. Louis, MO, USA), 90 μL of each extract, and 1.8 mL of 2% (w/v) Na2CO3 solution were mixed. After 30 min, the production of molybdenum oxide was measured at 752 nm. The contents of phenolic compounds were expressed as mg gallic acid equivalents (GAE)/g solid.

Total flavonoid contents (TFC) were determined by the aluminum chloride assay (Um et al., 2018). One hundred microliters of each extract, 20 μL of 5% NaNO2, and 800 μL of 60% ethanol were mixed and incubated for 6 min. The incubated mixture was added to 20 μL of 10% AlCl3 and 60 μL of 4% NaOH. After 30 min, the absorbance of the mixture was measured at 405 nm. The flavonoid content was expressed as mg quercetin equivalents (QE)/g of solid.

Total sugar contents (TSC) were determined by anthrone-sulfuric acid assay (Bae et al., 2019). Each extract (50 μL) was mixed with anthrone reagent and incubated in ice for 10 min. The mixtures were then boiled at 100 °C for 20 min and subsequently cooled at 25 °C. Next, 800 μL of water was added and absorbance measured at 620 nm. The reducing sugars content was expressed as mg sucrose equivalents (SE)/g of solid.

High-performance liquid chromatography (HPLC) of 1-O-acetylbritannilactone and ergolide

The lyophilized samples were dissolved in water and filtered through 0.20 μm membrane filters in preparation for HPLC. A liquid chromatograph (Waters 600 HPLC system, Waters Co., Milford, MA, USA) was used to analyze ABL (Chengdu Biopurify Phytochemicals Ltd., Chengdu, China) and ergolide (Chengdu Biopurify Phytochemicals Ltd.). The column used for the analysis was an Eclipse XDB-C18 (4.6 mm × 150 mm, 5 μm, Agilent Technologies Inc., Santa Clara, CA, USA). The mobile phase was composed of water with 0.1% ortho-phosphoric acid as eluent A and acetonitrile with 0.1% ortho-phosphoric acid as eluent B, and the elution gradient was as follows: 0–15 min: 90–48% A, 10–52% B; 15–20 min: 48% A, 52% B; 20–40 min: 48–20% A, 52–80% B; 40–50 min: 20–90% A, 80–10% B, at a flow rate of 0.5 mL/min. The detector was a Waters 2487 dual wavelength detector. ABL and ergolide were detected at 210 nm.

Cell culture

RAW 264.7 cells were purchased from the American Type Culture Collection (Rockville, MD, USA), and the cells were cultured in DMEM (HyClone, Logan, UT, USA) supplemented with 10% fetal bovine serum (FBS, Hyclone) as well as 1% penicillin and streptomycin (Hyclone) at 37 °C in a 5% CO2 humidified incubator.

Measurement of cell viability of RAW 264.7 cells

Cell viability of RAW 264.7 cells was measured by using the MTT assay. The cells were seeded at 5 × 105 cells/well in 96 well plates. Seeded RAW 264.7 cells were pretreated with IB or FIB. After 2 h, the cells were incubated with lipopolysaccharide (LPS) (Sigma, St. Louis, MO, USA) at 37 °C in 5% CO2 for 24 h. After incubation at 37 °C for 2 h, MTT solution was removed and formazan crystals were dissolved in DMSO to obtain a colored solution. The absorbances of the samples were measured at 570 nm using a microplate reader and cell viability was calculated using the following equation. Negative control was defined treatment without LPS and sample.

$${\text{Cell viability}}\left( \% \right) = \left( {\frac{\text{Absorbance of sample}}{\text{Absorbance of negative control}}} \right) \times 100$$

Evaluation of anti-inflammatory effect

Measurement of nitric oxide (NO) production

NO concentration in the cultured medium was determined by using the Griess reaction assay (Yang et al., 2019). RAW 264.7 cells were seeded at 5 × 105 cells/well in 96-well plates. Seeded cells were pretreated with FIB at various concentrations. After 2 h, the cells were incubated with LPS at 37 °C in 5% CO2 for 24 h. The cultured medium (100 μL) was mixed with a 100 μL of Griess reagent and incubated for 15 min at room temperature. Absorbance was measured at 540 nm using a microplate reader. NO production was extrapolated from the standard curve generated from different concentrations of sodium nitrate.

Measurement of cytokine expression using quantitative real-time PCR

RAW 264.7 cells were seeded at 1 × 106 cells/well in 6-well culture plates and pre-incubated for 24 h. The cells were treated and incubated with various concentrations of FIB for 2 h. And then the wells were treated and incubated with or without LPS (1 μg/mL) for 24 h.

Total RNA was isolated using RNeasy mini kit according to the manufacturer (Qiagen, Hilden, Germany). The total RNA (2 μg) was converted to cDNA using RevertAid™ First Strand cDNA Synthesis Kit (Thermo Fisher Scientific, Waltham, MA, USA). The expression levels of induced cytokines associated with anti-inflammatory effects were determined by using the SYBR Green PCR Master mix with semi-quantitative real-time PCR (PikoReal 96, Scientific Pierce, Waltham, MA, USA). The primers used were β-actin (forward primer 5′-GTGGGCCGCCCTAGGCACCAG-3′ and reverse primer 5′-GGAGGAAGAGGATGCGGCAGT-3′), TNF-α (forward primer 5′-TTGACCTCAGCGCTGAGTTG-3′ and reverse primer 5′-CCTGTAGCCCACGTCGTAGC-3′), IL-1β (forward primer 5′-CAGGATGAGGACATGAGCACC-3′ and reverse primer 5′-CTCTGCAGACTCAAACTCCAC-3′), and IL-6 (forward primer 5′-GTACTCCAGAAGACCAGAGG-3′ and reverse primer 5′-TGCTGGTGACAACCACGGCC-3′). PCR was performed as follows: 95 °C for 2 min for polymerase activation, followed by 40 cycles of 95 °C for 5 s for denaturation and 60 °C for 15 s for annealing and extension. The results were analyzed by using the delta–delta Ct method. The melting curve was used to analyze the measurement of reaction specificity.

Western blot analysis

Western blot was carried out by the modified method of Kim et al. (2018). In NF-κB phosphorylation, RAW 264.7 cells were seeded at 5 × 105 cells/well in 6-well culture plates and pre-incubated for 24 h. Next, the cells were treated and incubated with various concentrations of FIB for 2 h. The wells were treated with or without LPS (1 μg/mL) for 24 h.

For MAPKs phosphorylation, RAW 264.7 cells were seeded at 1 × 106 cells/well in 6-well culture plates and pre-incubated for 24 h. Next, the cells were treated and incubated with various concentrations of FIB for 2 h. The wells were treated with or without LPS (1 μg/mL) for 2 h.

After washing the cells with ice-cold PBS, cellular proteins were obtained in RIPA buffer supplemented with protease and phosphatase inhibitors. Cell lysates were kept on an ice-bath for 2 min and sonicated at 10 Amps (pulse on, 5 s; pulse off, 5 s) thrice. After a further 20 min, cell lysates were centrifuged at 14,000×g, 4 °C for 20 min. The total protein concentration of harvested supernatants was measured by using DC ™ protein assay kit (Bio-Rad, Hercules, CA, USA). An aliquot (30 μg) of each lysate was separated with 8–12% SDS-PAGE and subsequently transferred onto PVDF membrane. The immunoblots were blocked with 5% skim milk in tris buffered saline with Tween 20 (TBS-T) buffer for 1 h at 25 °C. The immunoblots were incubated with appropriately diluted specific primary antibodies for 16 h at 4 °C. After brief washing with TBS-T, the immunoblots were incubated with secondary antibodies for 2 h at room temperature. Each blot was detected on x-ray blue film by using enhanced chemiluminescence (Bio-Rad). The intensity of blots was determined by using Image J software (National Institutes of Health, Bethesda, MD, USA).

Statistical analysis

Statistical analysis was performed using IBM SPSS statistics version 18 software (IBM, New York, NY, USA). Data are shown as means ± standard deviations of three independent experiments. The mean values were compared using a one-way analysis of variance. Significant treatment means were separated by Tukey’s test; p < 0.05 was considered significant.

Results and discussion

Effect of fermentation of IB with W. cibaria D30

The cell viability and pH changes during fermentation of IB with W. cibaria D30 are described in Fig. 1A. The viable cell counts and pH changed from 6.05 to 9.33 log CFU/mL and 6.5 to 4.98 during fermentation, respectively. The viable cell counts increased until 9 h, and then decreased until 72 h. The decrease in the number of viable cells observed after 9 h was possibly due to the consumption of nutrients and production of antimicrobial metabolites such as organic acids. It has been reported that W. cibaria D30 produces organic acids including lactic acid and butyric acid as metabolites during fermentation (Yu et al., 2019).

Fig. 1
figure 1

Fermentaiton of IB with W. cibaria D30 and HPLC analysis. (A) Cell growth and pH of IB fermented with W. cibaria D30. ●, Total bacterial count in FIB; ■, pH value of FIB. (B) HPLC data of O-acetylbritannilactone (ABL) and ergolide contents of IB; FIB; and standard compound. (C) Concentration of ABL and ergolide. Peak identification: 1, ABL; 2, ergolide. ***p < 0.001, compared to IB

Full size image

The phytochemical properties of IB and FIB are shown in Table 1. The TPC of IB and FIB was increased from 0.047 to 0.149 mg GAE/g (p < 0.001), whereas the TFC was decreased from 0.178 to 0.113 mg QE/g (p < 0.001) by fermentation. No significant difference was detected in the TSC of IB and FIB. IB contains steroids, terpenoids (sesquiterpene lactones, diterpenes, and triterpenoids), and phenolics (Khan et al., 2010). Fermentation of phenolic compounds of IB by LAB strains can lead to the production of the aglycone forms which are more bioactive (Đorđević et al., 2010; Lee and Paik, 2017). These results may lead to increase of TPC and decrease of TFC. IB can also be influenced by β-glucosidase or organic acids produced by W. cibaria D30 during fermentation.

Table 1 Total phenolic, flavonoid, and sugar contents of IB and FIB
Full size table

ABL and ergolide as fermentation products of IB

ABL and sesquiterpene lactone ergolide produced by IB have been reported as cytotoxic agents to cancer cells (Song et al., 2005; Zhengfu et al., 2015). Ergolide induced apoptosis through DNA fragmentation, activation of caspase-3, cleavage of PARP, and cytochrome c release in Jurkat T cells (Song et al., 2002). Figure 1B–C shows the ergolide contents of IB and FIB, and the standard compounds identified by HPLC analysis. The regression curve of ABL was \({\text{y}} = 43,000\times {\text{x}} - 16,000 \left( {R^{2} = 0.9998} \right)\) and the retention time was 22.783 min. The regression curve of ergolide was \({\text{y}} = 32,300\times {\text{x}} + 910 \left( {R^{2} = 0.9999} \right)\) and the retention time was 27.49 min. ABL concentration significantly increased from 1.38 to 4.13 μg/mg by fermentation (p < 0.001). Fermentation significantly decreased the concentration of ergolide from 5.24 to 0.94 μg/mg (p < 0.001). The change in the concentrations of ABL and ergolide after fermentation could influence the viability and the anti-inflammatory effects of IB.

Effect of IB and FIB on cell viability in LPS-induced RAW 264.7 cells

The effect of IB and FIB (100–400 μg/mL) on the cell viability of RAW 264.7 cells was examined using the MTT assay (Table 2). IB treatment reduced the cell viability of RAW 264.7 cells in a dose-dependent manner. There was no significant difference in the viability of FIB-treated cells and control cells. Treatment with IB and FIB at 400 μg/mL resulted in cell viability of 31.73% (p < 0.001) and 94.53%, respectively. The concentration of IB should be kept below 200 μg/mL (< 80% viability) in in vitro anti-inflammatory assay; however, FIB could be used up to 400 μg/mL. Oh et al. (2012) also reported that the cytotoxicity of Oyaksungisan was lowered by fermentation with Lactobacillus spp. The increased viability of FIB could be attributed to the decreased concentration of ergolide. Further research is necessary to elucidate the mechanism behind the viability increment, as FIB contains various compounds that can interact to exert complementary effects.

Table 2 Cell viability of fermented IB with W. cibaria D30 against RAW 264.7 cells
Full size table

Inhibitory effect of FIB on NO production and proinflammatory cytokines expression in LPS-induced RAW 264.7 cells

It has been demonstrated that the anti-inflammatory effects of medicinal herbs are due to the inhibition of expression of inflammatory cytokines (Han et al., 2011; Hussain et al., 2016). Han et al. (2011) reported the anti-inflammatory effects of non-fermented and fermented Sophora flavescens. IB effectively inhibits the production of NO at relatively low concentrations compared with other herbal plants (Lee et al., 2011). However, the anti-inflammatory effect of FIB has not been previously investigated. In this study, NO production was investigated in LPS-induced RAW 264.7 cells. FIB remarkably reduced NO production in a dose-dependent manner (Fig. 2A). In particular, FIB at 400 μg/mL inhibited NO production by more than 92.4% in treated cells compared with cells treated with LPS alone. These results indicate that FIB effectively suppressed the secretion of nitrite from LPS-stimulated RAW 264.7 cells. Lee et al. (2011) showed that IB dissolved in DMSO at 5 μg/mL inhibited NO production by 45.3%. The effect of IB is also influenced by extraction solvents.

Fig. 2
figure 2

Anti-inflammatory effects of FIB in lipopolysaccharide (LPS)-induced RAW 264.7 cells. (A) NO production (μM); relative intensity of (B) TNF-α; (C) IL-1β; (D) IL-6; (E) iNOS; (F) COX-2 expression. ***p < 0.001, compared to LPS only-treated cells; different letters indicate significant differences (p < 0.05) as determined by Tukey’s test

Full size image

FIB influenced the expression of proinflammatory cytokines, TNF-α, IL-1β, IL-6, iNOS, and COX-2, produced by LPS stimulation (Fig. 2B–F). FIB effectively reduced mRNA expression of proinflammatory cytokines in a dose-dependent manner. In particular, 400 μg/mL of FIB showed a suppressive effect on the production of most proinflammatory cytokines. Of these cytokines, IL-1β, IL-6, and iNOS were inhibited at all the tested concentrations.

Effect of FIB on NF-κB activation in LPS-induced RAW 264.7 cells

NF-κB is a transcription factor that induces proinflammatory genes, such as iNOS, COX-2, TNF-α, IL-1β, and IL-6 (Kim et al., 2013). The influence of FIB on the expression of inflammatory mediators associated with the NF-κB pathway, p65 and IκBα were investigated (Fig. 3). Figure 3 shows the influence of FIB on NF-κB activated by phosphorylation upon LPS stimulation. LPS increased p-p65 and decreased IκBα. The expression level of p-p65 was markedly inhibited in the presence of FIB at a concentration of 400 μg/mL, and the inhibitory effect increased in a dose-dependent manner (Fig. 3B). p-p65 affected the levels of IκBα and p-IκBα in the cytoplasm (Fig. 3C–D). FIB significantly repressed IκBα phosphorylation in a dose–dependent manner, implying that FIB reduced IκBα degradation and NF-κB activation. These findings are consistent with the reports of other studies, indicating that NF-κB elements are present on the promoter of the iNOS, COX-2, TNF-α, IL-6, and IL-10 genes (Ahn et al., 2005; Barnes and Karin, 1997; Chen et al., 2000). Furthermore, ABL decreased NO production and iNOS expression in RAW 264.7 cells (Tavares and Seca, 2019).

Fig. 3
figure 3

Effect of FIB on protein expression level of NF-κB phosphorylation in LPS-treated RAW 264.7 cells. (A) NF-κB protein expression level was measured by western blot analysis. (B) p-p65 (normalized to p65) expressed. (C) p-IκBα (normalized to β-actin) expressed. (D) IκBα (normalized to β-actin) expressed.***p < 0.001, compared to only LPS-treated cells; different letters indicate significant differences (p < 0.05) as determined by Tukey’s test

Full size image

Effect of FIB on the phosphorylation of MAPKs in LPS-induced RAW 264.7 Cells

MAPKs pathways are the major pathways induced by LPS in RAW 264.7 cells. Western blot analysis was carried out to investigate the effect of FIB on the phosphorylation of MAPKs induced by LPS stimulation (Fig. 4). FIB reduced the levels of p-JNK, p-ERK, and p-p38 (Fig. 4B–D). The expression levels of p-p38, p-ERK, and p-JNK were dramatically reduced by FIB at 400 μg/mL, in a dose-dependent manner. These results suggest that the inhibitory effect of FIB on the phosphorylation of MAPKs was involved in the decreased production of proinflammatory mediators in RAW 264.7 cells.

Fig. 4
figure 4

Effect of FIB on protein expression level of MAPKs phosphorylation in LPS-inducd RAW 264.7 cells. (A) MAPKs protein expression level was measured by western blot analysis. (B) p-JNK (normalized to JNK) expressed. (C) p-ERK1/2 (normalized to ERK1/2) expression. (D) p-p38 (normalized to p38) expressed. **p < 0.05, compared to LPS-only treated cells; ***p < 0.001, compared to LPS-only treated cells; different letters indicate significant differences (p < 0.05) as determined by Tukey’s test

Full size image

In conclusion, the fermentation of IB with probiotic W. cibaria D30 were carried out, causing an increase in the levels of TPC and ABL but a decrease in the levels of TFC and ergolide. These results influenced the viability of RAW 264.7 cells and induced anti-inflammatory effects. FIB has higher viability than IB, which is mediated by its modified components. FIB at a higher concentration than that of IB exhibited anti-inflammatory effects at a lower toxicity by inhibiting the NF-κB and MAPKs pathways. Therefore, the fermentation of IB with W. cibaria D30 could lead to an improved anti-inflammatory medicinal herbal agent with higher viability.