Effects of shinbuto and ninjinto on prostaglandin E2 production in lipopolysaccharide-treated human gingival fibroblasts

Previously, we revealed that several kampo medicines used for patients with excess and/or medium patterns (kakkonto (TJ-1), shosaikoto (TJ-9), hangeshashinto (TJ-14), and orento (TJ-120)) reduced prostaglandin (PG)E2 levels using LPS-treated human gingival fibroblasts (HGFs). Recently, we examined other kampo medicines used for patients with the deficiency pattern [bakumondoto (TJ-29), shinbuto (TJ-30), ninjinto (TJ-32), and hochuekkito (TJ-41)] and the herbs comprising shinbuto and ninjinto using the same experimental model. Shinbuto and ninjinto concentration-dependently reduced LPS-induced PGE2 production by HGFs, whereas hochuekkito weakly reduced and bakumondoto did not reduce PGE2 production. Shinbuto and ninjinto did not alter cyclooxygenase (COX) activity or the expression of molecules involved in the arachidonic acid cascade. Therefore, we next examined which herbs compromising shinbuto and ninjinto reduce LPS-induced PGE2 production. Among these herbs, shokyo (Zingiberis Rhizoma) and kankyo (Zingiberis Processum Rhizoma) strongly and concentration-dependently decreased LPS-induced PGE2 production. However, both shokyo and kankyo increased the expression of cytosolic phospholipase (cPL)A2 but did not affect annexin1 or COX-2 expression. These results suggest that shokyo and kankyo suppress cPLA2 activity. We demonstrated that kampo medicines suppress inflammatory responses in patients with the deficiency pattern, and in those with excess or medium patterns. Moreover, kampo medicines that contain shokyo or kankyo are considered to be effective for the treatment of inflammatory diseases.


INTRODUCTION
Periodontal disease is an inflammatory disease of the gingiva that destroys periodontal tissues. In severe cases, alveolar bone is absorbed. In inflammatory responses and tissue degradation, prostaglandin E 2 (PGE 2 ), interleukin (IL)-6, and IL-8 play important roles. As PGE 2 has several functions in vasodilation, the enhancement of vascular permeability and pain, and osteoclastogenesis induction, PGE 2 participates in inflammatory responses and alveolar bone resorption in periodontal disease (Noguchi & Ishikawa, 2007).
However, these kampo medicines are used for patients with the excess pattern or medium pattern. Kampo medicine used for those with the deficiency pattern remains to be elucidated. In the present study, we therefore examined the anti-inflammatory effects of the kampo medicines for patients with the deficiency pattern [bakumondoto , shinbuto (TJ-30), ninjinto (TJ-32), and hochuekkito (TJ-41)], which are used for the treatment of inflammatory diseases. Furthermore, we examined the effects on PGE 2 production using herbs comprising the kampo medicines that reduce PGE 2 production.

Reagents
Kampo medicines (bakumondoto, shinbuto, ninjinto, and hochuekkito) were purchased from Tsumura & Co. (Tokyo, Japan). Powders of 8 herbs (bukuryo, bushi, kankyo, kanzo, ninjin, shakuyaku, shokyo, and sojutsu) were provided by Tsumura & Co. The ingredients in shinbuto and ninjinto formulas are shown in Tables 1 and 2. Powders of kampo medicines or herbs were suspended in Dulbecco's modified Eagle's medium (D-MEM; Sigma, St. Louis, MO, USA) containing 10% heat-inactivated fetal calf serum, 100 units/ml penicillin, and 100 mg/ml streptomycin (culture medium), and were rotated at 4 • C overnight. Then, the suspensions were centrifuged and the supernatants were filtrated through a 0.45 µm-pore membrane. Lipopolysaccharide (LPS) from Porphyromonas gingivalis 381 was provided by Professor Nobuhiro Hanada (School of Dental Medicine, Tsurumi University, Japan). Arachidonic acid was purchased from Cayman Chemical (Ann Arbor, MI). Other reagents were purchased from Nacalai tesque (Kyoto, Japan).  were removed by aspiration and the cells were treated with a 100-µl mixture of WST-8 with culture medium for 2 h at 37 • C in CO 2 incubator. Optical density was measured (measured wavelength at 450 nm and reference wavelength at 655 nm) using an iMark microplate reader (Bio-Rad, Hercules, CA, USA), and the mean background value was subtracted from each value. Data is represented as means ± S.D. (n = 4).

Measurement of prostaglandin E 2 (PGE 2 ), interleukin (IL)-6, and IL-8
HGFs were seeded in 96-well plates (10,000 cells/well) and incubated in serum-containing medium at 37 • C overnight. Then, the cells were treated with varying concentrations of each kampo medicine (0, 0.01, 0.1, or 1 mg/ml) or each herb (0, 10, 30, or 100 µg/ml) in the absence or presence of LPS (10 ng/ml) for 24 h (200 µl per well) in triplicate or quadruplicate for each sample. After the culture supernatants were collected, viable cell numbers were measured using WST-8 as described above. The concentrations of PGE 2 , IL-6, and IL-8 in the culture supernatants were measured by enzyme-linked immunosorbent assay (ELISA) according to the manufacturer's instructions (PGE 2 ), Cayman Chemical; IL-6 and IL-8, Thermo Fisher Scientific Inc., Camarillo, MA, USA), and were adjusted by the number of viable cells. Data are represented as pg or ng per 10,000 cells (mean ± S.D.).

Measurement of cyclooxygenase (COX)-2 activity
COX-2 activity was evaluated as reported previously (Wilborn et al., 1995) with slight modification. In brief, to estimate COX-2 activity, HGFs were treated with LPS and herbs for 8 h, washed, and incubated in culture medium containing exogenous arachidonic acid (10 µM). The concentrations of PGE 2 in the supernatants were measured by ELISA. Data are represented as pg per 10,000 cells (mean ± S.D.).

Preparation of cell lysates
HGFs were cultured in 60-mm dishes and treated with combinations of LPS and herbs for the indicated times. Then, cells were washed twice with Tris-buffered saline, transferred into microcentrifuge tubes, and centrifuged at 6,000 × g for 5 min at 4 • C. Supernatants were aspirated and cells were lysed on ice in lysis buffer (50 mM Tris-HCl, pH 7.4, 1% Nonidet P-40, 0.25% sodium deoxycholate, 150 mM NaCl, 1 mM ethyleneglycol bis(2aminoethylether)tetraacetic acid (EGTA), 1 mM sodium orthovanadate, 10 mM sodium fluoride, 1/100 volume of protease inhibitor cocktail (Nacalai tesque)) for 30 min at 4 • C. Samples were next centrifuged at 12,000 × g for 15 min at 4 • C, and supernatants were collected. The protein concentration was measured using a BCA Protein Assay Reagent kit (Pierce Chemical Co., Rockford, IL, USA).

Western blotting
The samples (10 µg of protein) were fractionated in a polyacrylamide gel under reducing conditions and transferred onto a polyvinylidene difluoride (PVDF) membrane (Hybond-P; GE Healthcare, Uppsala, Sweden). The membranes were blocked with 5% ovalbumin for 1 h at room temperature and incubated with primary antibody for an additional 1 h. The membranes were further incubated with horseradish peroxidase-conjugated secondary antibodies for 1 h at room temperature. Protein bands were visualized with an ECL kit (GE Healthcare). Densitometric values of each band were calculated using ImageJ software.

Statistical analysis
Differences between groups were evaluated by the two-tailed pairwise comparison test with a pooled variance, followed by correction with the Holm method (total 10 null hypotheses; five null hypotheses without kampo vs. with kampo in the absence or presence of LPS in Fig. 1, total 10 null hypotheses; three null hypotheses without kampo vs. with kampo in the absence of LPS, three null hypotheses without kampo vs. with kampo in the presence of LPS, and four null hypotheses without LPS vs. with LPS in Fig. 2). Differences between the control group and experimental groups were evaluated by a two-tailed Dunnett's test. HGFs were treated with combinations of LPS (0 or 10 ng/ml) and kampo medicine (0, 0.5, 1, 2, 5, or 10 mg/ml) for 24 h. Then, the numbers of viable cells were measured with WST-8. Open circles, treatment without LPS; closed circles, treatment with 10 ng/ml of LPS. * P < 0.05, * * P < 0.01, * * * P < 0.001 (without vs. with kampo medicine). P values were calculated by pairwise comparisons and corrected with the Holm method (10 null hypotheses).
Full-size DOI: 10.7717/peerj.4120/ fig-1 All computations were performed with the statistical program R (R Core Team, 2017). Dunnett's test was performed using the 'glht' function in the 'multcomp' package. Values with P < 0.05 were considered significantly different.

Effects of shinbuto and ninjinto on the arachidonic acid cascade
To clarify the mechanism of how shinbuto and ninjinto reduced LPS-induced PGE 2 production more directly, we examined the effects of these two kampo medicines on the arachidonic acid cascade. First, we examined the effects of shinbuto and ninjinto on COX activity. In order to bypass PLA 2 , we added exogenous arachidonic acid to HGFs treated with LPS alone or LPS plus kampo medicine (shinbuto or ninjinto). Then, we measured the PGE 2 level produced by COX. However, shinbuto and ninjinto did not affect LPS-induced PGE 2 production (Fig. 3). Next, we examined whether shinbuto and ninjinto affected the expression of molecules in the arachidonic acid cascade. cPLA 2 , which is the most upstream enzyme in the arachidonic acid cascade, releases arachidonic acid from plasma membranes. Shinbuto slightly reduced cPLA 2 expression and ninjinto slightly increased cPLA 2 expression (Fig. 4A). COX-2 was weakly expressed in the absence of LPS, and the treatment with LPS alone increased COX-2 expression. However, shokyo did not alter but kankyo slightly increased LPSinduced COX-2 expression (Fig. 4). Annexin1 (also named lipocortin1) is produced by glucocorticoids and inhibits cPLA 2 activity (Gupta et al., 1984;Wallner et al., 1986). Shinbuto and ninjinto slightly increased annexin1 expression (Fig. 4A) in a concentrationdependent manner (Fig. 4B).
Lastly, we evaluated the effects of shinbuto and ninjinto on ERK phosphorylation. cPLA 2 is directly phosphorylated and activated by phosphorylated ERK (Lin et al., 1993;Gijón et al., 1999). Therefore, we examined whether shinbuto and ninjinto suppressed LPS-induced ERK phosphorylation. LPS treatment enhanced ERK phosphorylation at 0.5 h and its phosphorylation was attenuated. However, 1 mg/ml of shinbuto or ninjinto did not affect LPS-induced ERK phosphorylation (Fig. 5).

Effects of herbs on PGE 2 production and molecular expression in the arachidonic acid cascade
We examined whether herbs which comprising shinbuto and ninjinto affected LPS-induced PGE 2 , IL-6 and IL-8 production by HGFs. When HGFs cells were treated with 10 ng/ml of LPS, HGFs cells produced large amounts of PGE 2 . Bukuryo increased LPS-induced PGE 2 production. Shokyo, kankyo and kanzo strongly and significantly reduced LPS-induced PGE 2 production (Fig. 6A). Moreover, shokyo and kankyo decreased PGE 2 production in a concentration-dependent manner (Figs. 6D-6E). Other herbs had little or no effect on PGE 2 production. Bukuryo increased LPS-induced IL-6 and IL-8 production, and kankyo increased IL-8 production (Figs. 6B-6C). Kanzo reduced IL-6 production (Fig. 6B). We then examined whether shokyo and kankyo affected the expression of molecules in the arachidonic acid cascade. Both shokyo and kankyo increased the expression of cPLA 2 but did not affect annexin1 or COX-2 expression (Fig. 7).

DISCUSSION
In our previous studies, we reported the importance of HGFs in the study of periodontal disease (Kamemoto et al., 2009;Ara et al., 2010;Nakazono et al., 2010;Ara et al., 2012;Kitamura, Urano & Ara, 2014;Ara & Sogawa, 2016), because HGFs are the most prominent cells in periodontal tissue. Moreover, LPS-treated HGFs produce inflammatory chemical mediators, such as PGE 2 and inflammatory cytokines such as IL-6 and IL-8 (Sismey-Durrant & Hopps, 1991;Bartold & Haynes, 1991;Tamura et al., 1992). Moreover, HGFs continue to produce PGE 2 (Ara et al., 2008a), IL-6, and IL-8 (Ara et al., 2009) in the presence of LPS. Therefore, the large amount of chemical mediators and cytokines derived from HGFs may be contained in periodontal tissues. From these findings, we believe that examining the effects of drugs on HGFs is needed in the study of periodontal disease.
Recently, we found that shokyo suppressed LPS-induced PGE 2 production by HGFs and that shokyo may suppress PLA 2 activity (Ara & Sogawa, 2016). In the present study, we examined the effects of kankyo in comparison with shokyo. Shokyo and kankyo increased cPLA 2 expression but did not alter annexin1 expression (Fig. 7). Moreover, we revealed that shinbuto and ninjinto, which contain shokyo and kankyo respectively, did not alter PGE 2 production when arachidonic acid was added to bypass the upstream pathway (Fig. 3). These data suggest that shokyo and kankyo did not affect the downstream pathway of arachidonic acid, which includes COX-2 and PGE synthase. In addition, shinbuto and ninjinto did not affect ERK phosphorylation (Fig. 5). From our findings described above, we were unable to explain the mechanism of the reduction in PGE 2 production. As gingerols in ginger are reported to inhibit both calcium-independent PLA 2 (iPLA 2 ) and cPLA 2 activities (Nievergelt et al., 2011), shokyo and kankyo are suggested to inhibit PLA 2 as discussed in the previous study (Ara & Sogawa, 2016). Previously, we reported that cPLA 2 is the main isoform in HGFs (Ara & Sogawa, 2016) among the subtypes such as cPLA 2 , iPLA 2 , and secretory PLA 2 (sPLA 2 ) (Burke & Dennis, 2009). Therefore, shokyo and kankyo may mainly inhibit cPLA 2 activity in HGFs. We found that orento decreases LPS-induced PGE 2 production via the suppression of ERK phosphorylation (Ara et al., 2010). However, orento may also reduce LPS-induced PGE 2 production by inhibition of cPLA 2 activity because orento contains kankyo.
We demonstrated that shokyo and kankyo concentration-dependently reduced LPSinduced PGE 2 production (Fig. 6A), and that the effects of kankyo are slightly stronger than those of shokyo (Figs. 6D-6E). In previous study, 6-and 8-gingerols were found to not inhibit cPLA 2 activity, but 10-gingerol and 6-, 8-, and 10-shogaols did (Nievergelt et al., 2011). Therefore, the difference in these effects on PGE 2 production between shokyo and kankyo may be due to the amount of shogaols in these herbs.
We demonstrated that shinbuto and ninjinto slightly increased annexin1 expression (Fig. 4). However, the involvement of annexin1 in the reduction in PGE 2 production is unlikely. Shokyo and kankyo did not alter annexin1 expression (Fig. 7). All 4 herbs other than shokyo in shinbuto did not reduce PGE 2 production, but rather, bukuryo increased PGE 2 production (Fig. 6A). Similarly, kanzo in ninjinto increased annexin1 expression in HGFs, and kanzo also inhibited COX activity (Ara & Sogawa, 2016). The 2 residual herbs other than kankyo and kanzo did not reduce PGE 2 production (Fig. 6A). Therefore, the increased annexin1 expression did not contribute to decreased PGE 2 production.
At the herb level, we were unable to clarify which herbs affect cytokine production. Bukuryo in shinbuto increased LPS-induced IL-6 and IL-8 production (Figs. 6B-6C). Therefore, this effect of shinbuto on increased IL-6 production may be due to bukuryo. However, shinbuto did not alter IL-8 production even though it contains bukuryo. Moreover, although ninjinto increased LPS-induced IL-6 production, kanzo reduced IL-6 production, and the other three herbs, kankyo, sojutsu, and ninjin, did not alter IL-6 production. Similarly, although ninjinto did not alter IL-8 production, kankyo increased IL-8 production. Therefore, the effects of herbs on IL-6 and IL-8 production are considered to not be due to a single herb but to the combination of herbs.

CONCLUSION
We demonstrated that shinbuto and ninjinto reduced LPS-induced PGE 2 production by HGFs. Moreover, shokyo and kankyo, which are included in these kampo medicines respectively, concentration-dependently reduced LPS-induced PGE 2 production. However, shokyo and kankyo did not alter the expression of the molecules in the arachidonic acid cascade, suggesting that shokyo and kankyo inhibit cPLA 2 activity. Therefore, the kampo medicines that contain shokyo or kankyo may have the ability to reduce PGE 2 production. We found that the kampo medicines used for patients with the deficiency pattern also have anti-inflammatory effects in those with the excess pattern or medium pattern. We expect kampo medicines to be used for improving inflammatory diseases, such as periodontal disease and stomatitis, in patients with any pattern.