Anti-Inflammatory Effect of Charantadiol A, Isolated from Wild Bitter Melon Leaf, on Heat-Inactivated Porphyromonas gingivalis-Stimulated THP-1 Monocytes and a Periodontitis Mouse Model

Porphyromonas gingivalis has been identified as one of the major periodontal pathogens. Activity-directed fractionation and purification processes were employed to identify bioactive compounds from bitter melon leaf. Ethanolic extract of bitter melon leaf was separated into five subfractions by open column chromatography. Subfraction-5-3 significantly inhibited P. gingivalis-induced interleukin (IL)-8 and IL-6 productions in human monocytic THP-1 cells and then was subjected to separation and purification by using different chromatographic methods. Consequently, 5β,19-epoxycucurbita-6,23(E),25(26)-triene-3β,19(R)-diol (charantadiol A) was identified and isolated from the subfraction-5-3. Charantadiol A effectively reduced P. gingivalis-induced IL-6 and IL-8 productions and triggered receptors expressed on myeloid cells (TREM)-1 mRNA level of THP-1 cells. In a separate study, charantadiol A significantly suppressed P. gingivalis-stimulated IL-6 and tumor necrosis factor-α mRNA levels in gingival tissues of mice, confirming the inhibitory effect against P. gingivalis-induced periodontal inflammation. Thus, charantadiol A is a potential anti-inflammatory agent for modulating P. gingivalis-induced inflammation.


Introduction
Periodontal diseases are complex, multifactorial diseases characterized by chronic inflammation of periodontal tissues, including gingival inflammation and alveolar bone resorption, and eventually tooth loss. Periodontitis begins as acute inflammation of the gingival tissue driven by polymicrobial infections and aggressive host immune and inflammatory responses via production of pro-inflammatory cytokines [1].
Porphyromonas gingivalis, a Gram-negative anaerobic bacterium, has been considered as a major oral pathogen in the development of chronic periodontitis [2]. P. gingivalis expresses several known virulence factors, such as lipopolysaccharide (LPS), fimbriae, proteases, and outer membrane vesicles [1]. Exposure to P. gingivalis causes innate responses through toll-like receptor (TLR)-2 and TLR-4 on the host cell surface and can trigger the production and release of pro-inflammatory mediators, such as interleukin (IL)-8 and IL-6, IL-1β, and tumor necrosis factor (TNF)-α. These pro-inflammatory cytokines play a significant role in the development of periodontitis [3]. IL-8 is produced primarily by gingival fibroblasts, gingival epithelial cells and endothelial cells. It is detectable in diseased periodontal tissues and has been associated with subclinical inflammation of the initial lesion [4]. Recently, IL-8 has been considered to be a potential therapeutic target for periodontitis [5]. IL-6 and IL-1β regulate inflammatory cell migration and osteoclastogenesis [4]. Since excessive secretion of pro-inflammatory mediators has been highly related to periodontitis pathogenesis, developing a strategy-based approach to suppress P. gingivalis-induced inflammatory responses may be a promising strategy for the alleviation of chronic periodontal disease.
Natural products from the herbal remedy, medicinal plants, functional foods, and their constituent have been considered to be effective in the prevention and treatment of periodontal diseases [6][7][8]. However, these studies did not purify specific compounds that have a meaningful anti-inflammatory effect on periodontitis from the crude extracts.

Effects of Sub-Fractions from Leaf Extract of Wild Bitter Melon and Charantadiol A on P. gingivalis-Induced Cytokines in THP-Cells
Previous studies indicated that P. gingivalis can elicit high levels of IL-6 and IL-8 production in a variety of cell types comprising human oral epithelial cells, periodontal ligament cells and monocytes [13][14][15]. We previously demonstrated that sub-fractions, the fraction 5 (Fra. 5) and Fra. 5-2, isolated from crude WBM leaf extract inhibited P. gingivalis-stimulated IL-8 production by THP-1 cells [13]. In the present study, the Fra. 5-3 ( Figure 1) was fractionated and evaluated to determine the fractions that contained effective substances. Then, co-culture model of heat-inactivated P. gingivalis and THP-1 monocytes was used to evaluate the suppress effects on P. gingivalis-induced inflammatory responses by the components of Fra. 5-3. used to evaluate the suppress effects on P. gingivalis-induced inflammatory responses by the components of Fra. 5-3.  (2) was also found as impurity in the NMR spectra.
To determine whether Fra. 5-3 would affect cell viability, THP-1 cells were incubated firstly in culture medium supplemented with various concentrations of tested samples. No adverse effect on cell proliferation was observed when the concentration of Fra. 5-3 was below 10 μg/mL by the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay (data not shown). Figure 2 shows that the productions of IL-8 and IL-6 were significantly elevated in response to P. gingivalis stimulation. However, culture medium supplied with different concentrations of Fra. 5-3 significantly reduced respective cytokine production by as much as 85% (IL-6) and 81% (IL-8) (Figure 2). To determine whether Fra. 5-3 would affect cell viability, THP-1 cells were incubated firstly in culture medium supplemented with various concentrations of tested samples. No adverse effect on cell proliferation was observed when the concentration of Fra. 5-3 was below 10 µg/mL by the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay (data not shown). Figure 2 shows that the productions of IL-8 and IL-6 were significantly elevated in response to P. gingivalis stimulation. However, culture medium supplied with different concentrations of Fra. 5-3 significantly reduced respective cytokine production by as much as 85% (IL-6) and 81% (IL-8) ( Figure 2).
Charantadiol A, a cucurbitane-type triterpenoid, is also a biological active component in bitter melon fruit and shows hypoglycemic effect in streptozotocin-induced diabetic rats [16]. Cucurbitane-type triterpenoids exert anti-inflammatory activity by inhibiting nitric oxide production in lipopolysaccharides (LPS)-stimulated RAW 264.7 macrophage cells [17]. In this study, we have demonstrated that charantadiol A is as potent as luteolin, a well-known antioxidant and anti-inflammatory flavonoid, in suppressing P. gingivalis-induced inflammatory responses in vitro ( Figure 3).   Effects of sub-fraction (Fra. 5-3) isolated from wild bitter melon leaf extract on P. gingivalis-induced pro-inflammatory cytokine productions in human monocytic THP-1 cells. Cells were incubated with 0.1% (v/v) DMSO (as a vehicle control), or co-cultured with P. gingivalis (M.O.I. = 10) and different concentrations of sub-fraction 5-3 (2.5, 5 or 10 μg/mL) for 24 h. The cell-free culture supernatants were subsequently collected and analyzed for the content of IL-8 and IL-6. Experiments were performed three times in triplicate. Each value represents the mean ± SD. Values with different letters are significantly different at p < 0.05.
Charantadiol A, a cucurbitane-type triterpenoid, is also a biological active component in bitter melon fruit and shows hypoglycemic effect in streptozotocin-induced diabetic rats [16]. Cucurbitane-type triterpenoids exert anti-inflammatory activity by inhibiting nitric oxide production in lipopolysaccharides (LPS)-stimulated RAW 264.7 macrophage cells [17]. In this study, we have demonstrated that charantadiol A is as potent as luteolin, a well-known antioxidant and anti-inflammatory flavonoid, in suppressing P. gingivalis-induced inflammatory responses in vitro ( Figure 3).  Charantadiol A, a cucurbitane-type triterpenoid, is also a biological active component in bitter melon fruit and shows hypoglycemic effect in streptozotocin-induced diabetic rats [16]. Cucurbitane-type triterpenoids exert anti-inflammatory activity by inhibiting nitric oxide production in lipopolysaccharides (LPS)-stimulated RAW 264.7 macrophage cells [17]. In this study, we have demonstrated that charantadiol A is as potent as luteolin, a well-known antioxidant and anti-inflammatory flavonoid, in suppressing P. gingivalisinduced inflammatory responses in vitro ( Figure 3).

Effect of Charantadiol A on TREM-1 mRNA Expression Level in P. gingivalis-Stimulated THP-1 Cells
The triggering receptor expressed on myeloid cells-1 (TREM-1) is a cell surface receptor of the immunoglobulin superfamily expressed on polymorphonuclear leukocytes, Molecules 2021, 26, 5651 5 of 10 monocytes, macrophages, dendritic cells, vascular smooth muscle cells, and is upregulated in the presence of inflammation to amplify pro-inflammatory cytokine production [18,19]. Co-triggering of TREM-1 and TLR4 results in a synergistic increase in TLR4-mediated pro-inflammatory cytokine and chemokine secretion [20]. Exposure of P. gingivalis induces significantly higher expression of TREM-1 mRNA and upregulates the expression of the TREM-1/DAP12 pathway in monocytes [21,22]. Willi and co-workers reported that periodontitis patients have higher TREM-1 gingival expression than healthy controls [23]. Doxycycline is used as an adjunct treatment in clinical periodontal therapy and has been shown to reduce P. gingivalis-induced IL-8 secretion by inhibiting TREM-1 expression and release [24]. Consistent with the previous findings [21,22,24], the TREM-1 mRNA level was significantly elevated in response to P. gingivalis (Figure 4). Treatments of charantadiol A significantly inhibited bacterially induced TREM-1 mRNA expression (Figure 4), and this effect may partly account for its anti-inflammatory property. Our present results show for the first time that charantadiol A downregulated P. gingivalis-induced TREM-1 expression. However, additional studies are needed to further support for the possible mechanisms underlying inhibitory effect of charantadiol A on pro-inflammatory cytokine production.

THP-1 Cells
The triggering receptor expressed on myeloid cells-1 (TREM-1) is a cell surface re ceptor of the immunoglobulin superfamily expressed on polymorphonuclear leukocytes monocytes, macrophages, dendritic cells, vascular smooth muscle cells, and is upregu lated in the presence of inflammation to amplify pro-inflammatory cytokine production [18,19]. Co-triggering of TREM-1 and TLR4 results in a synergistic increase in TLR4-me diated pro-inflammatory cytokine and chemokine secretion [20]. Exposure of P. gingivali induces significantly higher expression of TREM-1 mRNA and upregulates the expression of the TREM-1/DAP12 pathway in monocytes [21,22]. Willi and co-workers reported tha periodontitis patients have higher TREM-1 gingival expression than healthy controls [23] Doxycycline is used as an adjunct treatment in clinical periodontal therapy and has been shown to reduce P. gingivalis-induced IL-8 secretion by inhibiting TREM-1 expression and release [24]. Consistent with the previous findings [21,22,24], the TREM-1 mRNA leve was significantly elevated in response to P. gingivalis (Figure 4). Treatments of charantadio A significantly inhibited bacterially induced TREM-1 mRNA expression (Figure 4), and this effect may partly account for its anti-inflammatory property. Our present results show fo the first time that charantadiol A downregulated P. gingivalis-induced TREM-1 expres sion. However, additional studies are needed to further support for the possible mecha nisms underlying inhibitory effect of charantadiol A on pro-inflammatory cytokine pro duction.

Effect of Charantadiol A on IL-6 and TNFα mRNA expression in P. gingivalis-Stimulated Gingival Tissue of Mice
The pro-inflammatory cytokines, IL-1, IL-6, and TNF-α, appear to have central role in periodontal tissue destruction [25]. IL-6 plays a crucial role mainly in the initiation and acute phase of periodontitis [26]. Additionally, IL-6 plays a role in the transition between acute and chronic inflammation, it enhances T-cell proliferation and accelerates of bone resorption by increasing osteoclast formation [27]. IL-6 is highly expressed in inflamed periodontal tissue and gingival crevicular fluid, which has been shown to be related to the severity of periodontitis [26,28]. TNF-α possesses a wide range of immune-regulatory functions to stimulate the production of chemokines or cyclooxygenase products, which

Effect of Charantadiol A on IL-6 and TNFα mRNA expression in P. gingivalis-Stimulated Gingival Tissue of Mice
The pro-inflammatory cytokines, IL-1, IL-6, and TNF-α, appear to have central roles in periodontal tissue destruction [25]. IL-6 plays a crucial role mainly in the initiation and acute phase of periodontitis [26]. Additionally, IL-6 plays a role in the transition between acute and chronic inflammation, it enhances T-cell proliferation and accelerates of bone resorption by increasing osteoclast formation [27]. IL-6 is highly expressed in inflamed periodontal tissue and gingival crevicular fluid, which has been shown to be related to the severity of periodontitis [26,28]. TNF-α possesses a wide range of immune-regulatory functions to stimulate the production of chemokines or cyclooxygenase products, which consequently amplifies the degree of inflammation [27]. TNF-α has shown to participate in the initiation of periodontitis by injuring the oral mucosa barrier. Moreover, a high level of circulating TNF-α derived from periodontal tissue may contribute to systemic inflammation-associated diseases [26]. In this study, we showed that charantadiol A can affect immune responses in P. gingivalis-stimulated mouse gingival tissue. As shown in Figure 5, P. gingivalis-induced IL-6 and TNF-α mRNA expressions were attenuated by respective co-injection of charantadiol A (5 µg) or luteolin (50 µg).
consequently amplifies the degree of inflammation [27]. TNF-α has shown to participate in the initiation of periodontitis by injuring the oral mucosa barrier. Moreover, a high level of circulating TNF-α derived from periodontal tissue may contribute to systemic inflammation-associated diseases [26]. In this study, we showed that charantadiol A can affect immune responses in P. gingivalis-stimulated mouse gingival tissue. As shown in Figure 5, P. gingivalis-induced IL-6 and TNF-α mRNA expressions were attenuated by respective co-injection of charantadiol A (5 μg) or luteolin (50 μg). The effectiveness of a conventional mechanical treatment against gingivitis is clear. However, topical adjunctive therapy with antimicrobials or anti-inflammatory agents has been applied for periodontal treatment [29]. Most natural products have been applied topically (as mouthwash, toothpaste, chewing gum etc.). Evidences show a beneficial effect of anti-inflammatory agents against gingivitis, either as a single treatment modality or as an adjunctive therapy [30]. Hence, it is worthy to investigate natural products which possess the beneficial effect on gingival inflammation. We previously described methods for isolating and purifying of kuguacin R and TCD and demonstrated their anti-inflammatory action in vitro and in vivo [13]. We showed that pro-inflammatory cytokine (IL-6 and IL-8) expression was induced by P. gingivalis infection but decreased by treatment with kuguacin R or TCD [13]. The activation of mitogen-activated protein kinase (MAPK), a signaling pathway for pro-inflammatory cytokines in periodontitis [31], was modulated by kuguacin R or TCD [13]. However, the yield of charantadiol A is lower than that of kuguacin R and TCD, making it difficult to acquire enough of an amount of charantadiol A to explore more details of its mechanism. Therefore, we are not able to make further analysis as we did on kuguacin R and TCD in our previous research. Nevertheless, the shortages of inexpensive, pure kuguacin R, TCD and charantadiol A are still limiting the exploration of their potentially beneficial applications to human health. Certainly, future investigations on the toxicological and pharmaceutical evaluation of these cucurbitane triterpenoids are expected.

Plant Materials
WBM (a cultivar of Hualien-1) leaves were obtained from the Hualien District Agricultural Research and Extension Station, Hualien, Taiwan. The fresh aerial parts of WBM were harvested. WBM leaves were collected and then a voucher specimen (number NTNUHung-2014-09) was deposited in the Department of Human Development and Family Studies, National Taiwan Normal University, Taipei, Taiwan. The voucher speci- The effectiveness of a conventional mechanical treatment against gingivitis is clear. However, topical adjunctive therapy with antimicrobials or anti-inflammatory agents has been applied for periodontal treatment [29]. Most natural products have been applied topically (as mouthwash, toothpaste, chewing gum etc.). Evidences show a beneficial effect of anti-inflammatory agents against gingivitis, either as a single treatment modality or as an adjunctive therapy [30]. Hence, it is worthy to investigate natural products which possess the beneficial effect on gingival inflammation. We previously described methods for isolating and purifying of kuguacin R and TCD and demonstrated their anti-inflammatory action in vitro and in vivo [13]. We showed that pro-inflammatory cytokine (IL-6 and IL-8) expression was induced by P. gingivalis infection but decreased by treatment with kuguacin R or TCD [13]. The activation of mitogen-activated protein kinase (MAPK), a signaling pathway for pro-inflammatory cytokines in periodontitis [31], was modulated by kuguacin R or TCD [13]. However, the yield of charantadiol A is lower than that of kuguacin R and TCD, making it difficult to acquire enough of an amount of charantadiol A to explore more details of its mechanism. Therefore, we are not able to make further analysis as we did on kuguacin R and TCD in our previous research. Nevertheless, the shortages of inexpensive, pure kuguacin R, TCD and charantadiol A are still limiting the exploration of their potentially beneficial applications to human health. Certainly, future investigations on the toxicological and pharmaceutical evaluation of these cucurbitane triterpenoids are expected.

Plant Materials
WBM (a cultivar of Hualien-1) leaves were obtained from the Hualien District Agricultural Research and Extension Station, Hualien, Taiwan. The fresh aerial parts of WBM were harvested. WBM leaves were collected and then a voucher specimen (number NTNUHung-2014-09) was deposited in the Department of Human Development and Family Studies, National Taiwan Normal University, Taipei, Taiwan. The voucher specimen of the plant was authenticated by Dr. Po-Jung Tsai, Professor, National Taiwan Normal University, Taipei, Taiwan. After cleaning with water, the WBM leaves were air-dried and ground using a blender. Powdered WBM leaves were stored in the dark at −20 • C until used.
Nuclear magnetic resonance ( 1 H NMR and 13 C NMR) techniques were used for the structure elucidation of the compounds. NMR spectra were recorded on a Bruker spectrometer (400 MHz for 1 H NMR and 100 MHz for 13 C NMR) instrument and using CDCl 3 as solvent (Supplementary Figure S1).

Preparation of Heat-Inactivated P. gingivalis
The P. gingivalis strain BCRC14417 was obtained from the Bioresource Collection and Research Center, Hsinchu, Taiwan. Bacterial suspensions to induce periodontitis of mice were prepared by a method described elsewhere [13]. Briefly, P. gingivalis was cultured anaerobically in tryptic soy broth (TSB, Difco, Detroit, MI, USA) supplemented with 2.5% yeast extract, hemin, and menadione at 37 • C. The numbers of bacteria were determined with a spectrophotometer (at an optical density at 600 nm) based on a standard curve established by colony formation on bacterial plates. To prepare heat-inactivated P. gingivalis, bacterial suspensions in phosphate-buffered saline (PBS) were heated at 80 • C for 30 min, washed with PBS, and re-suspended in RPMI 1640 medium (Gibco, Carlsbad, CA, USA).

Stimulation of THP-1 Cells with P. gingivalis and Cytokine Measurements
Fra. 5-3 and charantadiol A were re-dissolved in dimethyl sulfoxide (DMSO; RDH Chemical Co., Spring Valley, CA, USA) to 20 mg/mL of stock solution for the sequential experiments. A well-established co-culture model of P. gingivalis and THP-1 cells was used to investigate the anti-inflammatory properties of WBM leaf extracts [13]. Briefly, THP-1 cells (2 × 10 5 cells/well) were seeded in 96-well plates with serum-free medium and were stimulated with heat-inactivated P. gingivalis at multiplicity of infection (M.O.I.) of 10 (bacteria/THP-1 cell) alone or in combination with various concentrations of tested WBM extraction samples, DMSO (0.1%) as a vehicle control, and luteolin (Sigma, as a positive control) at 37 • C with 5% CO 2 humidified atmosphere. After incubation for 24 h, the cell-free supernatants were collected, and the amount of IL-6 or IL-8 was determined using the respective enzyme immunoassay kits (Invitrogen, Carlsbad, CA, USA).

RNA Extraction of THP-1 Cells and Quantitative Real-Time Polymerase Chain Reaction (PCR)
THP-1 cells were cultured in 6-cm cell culture dishes (4 × 10 6 cells/dish) for 24 h, and then co-incubated with heat-inactivated P. gingivalis (M.O.I. = 10) with various concentrations of tested samples (charantadiol A or luteolin). Cells were harvested and washed with PBS. Total RNA of human THP-1 cell samples was extracted and isolated with the TRIzol reagent (Invitrogen), according to the manufacturer's instructions. cDNA was then synthesized from the RNA in a reaction mixture of oligo (dT) primers and reverse transcriptase (Promega, Madison, WI, USA), following the manufacturer's instructions. Primers and probes were selected for the genes: GAPDH (glyceraldehyde-3-phosphate dehydrogenase) was used as the housekeeping gene. We used the forward 5 -CCA TAG GAG AGC AAC AGA-3 and reverse 5 -GCC TCG TTC TAG TCA CAT ACA-3 primers for triggering receptor expressed on myeloid cells (TREM-1), and the forward 5 -GTG AAG GTC GGA GTC AAC G-3 and reverse 5 -TGA GGT CAA TGA AGG GGT C-3 primers for GAPDH. These primer pairs amplified, respectively, a 106 bp fragment of the TREM-1 cDNA and a 112 bp fragment of the GAPDH cDNA. Real-time PCRs were conducted in an iCycler iQ Real-Time detection system (Bio-Rad, Hercules, CA, USA) using iQ TM SYBR Green Supermix (Bio-Rad). The relative amounts of the PCR products were analyzed by iQ™5 optical system software (ver. 2.1; Bio-Rad). All expression levels were normalized using the GAPDH as an internal standard in each sample. Fold expression was defined as the fold increase relative to controls.

Effect of Charantadiol A on P. gingivalis-Induced Cytokine Expression In Vivo
We evaluated the protective effects of charantadiol A or luteolin on P. gingivalisstimulated periodontal inflammation in a mouse model by using the method described elsewhere [13]. Six-week-old male C57BL/6 mice were obtained from the National Laboratory Animal Center (Taipei, Taiwan). The mice were housed in groups of 5 per cage, under standard temperature-controlled conditions with a 12 h/12 h light-dark cycle and free access to food and water throughout the experiments. All animal experiments were conducted in accordance with the Guide for the Care and Use of Laboratory Animals and were approved by the Animal Care Committee of the National Taiwan Normal University (IACUC Permit No. 103020). Throughout the period of the study, mice were fed with sterile standard solid mice chow diet and sterile water. Periodontitis was induced by an intra-gingival injection of heat-inactivated P. gingivalis according to the methods by Tsai et al. [13]. After 1 week of adaptation, animals were randomly divided into five groups (n = 5). Heat-inactivated P. gingivalis (1 × 10 9 CFU in PBS) or PBS (as vehicle control) was injected once daily into the mandibular (lower inset) gingival tissues of mice for 3 days.
To study the effects of charantadiol A or luteolin, they were respectively administered once daily for 3 days with co-injection of heat-inactivated P. gingivalis suspensions. After 14 days of bacterial injection, mice were then sacrificed with carbon dioxide asphyxiation. The gingival tissues were excised for the extraction of total RNA. P. gingivalis-induced IL-6 and TNF-α expression were determined by reverse transcription qualitative polymerase chain reaction (RT-qPCR) as previously described [13].

Statistical Analysis
All data are presented as means ± SD. Statistical analyses were performed using the SPSS 20.0 statistical package (Chicago, IL, USA). The data were evaluated for statistical significance with the one-way ANOVA followed by Duncan's multiple range tests. A p value of <0.05 was considered statistically significant.

Conclusions
In conclusion, we have demonstrated that charantadiol A suppressed P. gingivalisstimulated TREM-1 expression, thereby reducing the levels of pro-inflammatory mediators in THP-1 cells. Furthermore, charantadiol A exerted anti-inflammatory effect in periodontitis mimicking conditions in mice. Altogether, charantadiol A is an attractive cucurbitane for periodontitis treatments, and more investigations can be expected for further support the efficacy of charantadiol A on periodontitis.