Flavones Isolated from Scutellariae radix Suppress Propionibacterium Acnes-Induced Cytokine Production In Vitro and In Vivo

Scutellariae radix, the root of Scutellaria baicalensis, has long been applied in traditional formulations and modern herbal medications. Propionibacterium acnes (P. acnes) in follicles can trigger inflammation and lead to the symptom of inflammatory acnes vulgaris. This study was aimed at evaluating the effect of Scutellariae radix extract and purified components isolated from it on inflammation induced by P. acnes in vitro and in vivo. The results showed the ethyl acetate (EA) soluble fraction from the partition of crude ethanolic extract from Scutellariae radix inhibited P. acnes-induced interleukin IL-8 and IL-1β production in human monocytic THP-1 cells. Seven flavones were isolated from the EA fraction by repeated chromatographies, and identified as 5,7-dihydroxy-6-methoxyflavone (FL1, oroxylin), 5,7-dihydroxy-8-methoxyflavone (FL2, wogonin), 5-hydroxy-7,8-dimethoxyflavone (FL3, 7-O-methylwogonin), 5,6′-dihydroxy-6,7,8,2′-tetramethoxy flavone (FL4, skullcapflavone II), 5,7,4′-trihydroxy-8-methoxyflavone (FL5), 5,2′,6′-trihydroxy-7,8-dimethoxyflavone (FL6, viscidulin II), and 5,7,2′,5′-tetrahydroxy-8,6′-dimethoxyflavone (FL7, ganhuangenin). They all significantly suppressed P. acnes-induced IL-8 and IL-1β production in THP-1 cells, and FL2 exerted the strongest effect with half maximal inhibition (IC50) values of 8.7 and 4.9 μM, respectively. Concomitant intradermal injection of each of the seven flavones (20 μg) with P. acnes effectively attenuated P. acnes-induced ear swelling, and decreased the production of IL-6 and tumor necrosis factor-α in ear homogenates. Our results suggested that all the seven flavones can be potential therapeutic agents against P. acnes-induced skin inflammation.


Introduction
Acne vulgaris is one of the most common skin diseases. The pathogenesis is complex and incompletely understood, but inflammation is believed to be a key component [1]. Propionibacterium acnes (P. acnes), a Gram-positive anaerobic bacterium species, may play a major role in the initiation of the inflammatory reaction by stimulating the secretion of interleukin (IL)-18, tumor necrosis [2,3]. IL-8 is the major inflammatory mediator and a strong chemotactic factor for neutrophils, basophils, and T cells. IL-8 has been implicated in mounting an inflammatory response in acne lesions [4]. In addition, the high levels of IL-1β were observed in human acne lesion, in mouse skin lesion induced by P. acnes, and in P. acnes-exposed human monocytes [5].
Scutellariae radix, the root of Scutellaria baicalensis, has been used as traditional Chinese medicine to treat allergic and inflammatory diseases in Japan and China. It is also often used to treat cardiovascular diseases, respiratory, and gastrointestinal infections [6]. Flavonoids, including baicalin, baicalein, wogonin, and oroxylin-A, have been identified in Scutellariae radix [7]. The extracts and the isolated compounds of Scutellariae radix are pharmacologically-active and show great potential in the treatment of inflammation, cancers, and virus-related diseases [8].
However, the bioactive components of Scutellariae radix have not yet been completely investigated. This study is aimed at exploring the suppressive effects of seven flavones isolated from Scutellaria radix on P. acnes-induced inflammation in vitro and in vivo, and their anti-acne potential.

Effects of Scutellariae Radix Extracts on P. acnes-Induced IL-1β and IL-8 Production in THP-Cells
Ethylacetate (EA) fraction was cytotoxic to THP-1 cells when the concentrations applied were higher than 20 μg/mL ( Figure 1A). So, the doses of 2.5, 5, and 10 μg/mL were used for the subsequent in vitro experiments. Treatment of THP-1 cells with heat-killed P. acnes evoked the production of IL-1β and IL-8. EA fraction of Scutellariae radix significantly decreased P. acnes-induced IL-1β and IL-8 production in a dose-dependent manner ( Figure 1B,C).
The butanol fraction also inhibited IL-1β and IL-8 production by P. acnes-stimulated THP-1 cells, but less effectively than the EA fraction did. The butanol fraction had half maximal inhibition (IC50) values of 80.5 and 135.5 μg/mL for the inhibition of IL-1β and IL-8 secretion, respectively, while the EA fraction had the respective IC50 values of 6.6 and 5.6 μg/mL. Therefore, EA fraction was subjected to chromatography on silica gel and further purification by semi-preparative HPLC. Seven known flavones (FL1-7) were found.

Effects of the Seven Flavones Isolated from EA Fraction of Scutellariae Radix on P. acnes-Induced IL-8 and IL-1β Production in Human Monocytic THP-1 Cells
Chemical names, common names, and chemical structures of the seven known flavones isolated from EA fraction of Scutellariae radix are shown in Figure 2.   Chemical names, common names, and chemical structures of the seven known flavones isolated from EA fraction of Scutellariae radix are shown in Figure 2.

Effects of the Seven Flavones Isolated from EA Fraction of Scutellariae Radix on P. acnes-Induced IL-8 and IL-1β Production in Human Monocytic THP-1 Cells
Chemical names, common names, and chemical structures of the seven known flavones isolated from EA fraction of Scutellariae radix are shown in Figure 2.   The concentrations without apparent cytotoxicity toward THP-1 cells, assayed by MTT, were used for the subsequent experiments (Table 1). Seven flavones, at various concentrations, significantly reduced IL-8 and IL-1β levels ( Table 1). The potency of the flavones was expressed as IC50 value. IC50 values of FL4 for IL-8, and FL1 and FL6 for IL1β could not be precisely determined because the degree of inhibition provided by the highest test concentration was less than 50%. The rank order of potency of these flavones for IL-8 inhibition was FL2 > FL5 > FL1 > (FL4) FL6 > FL3 > FL7; whereas for IL-1β inhibition was FL2 > FL4 > FL5 > (FL1) > FL3 > (FL6) FL7. Therefore, FL2 had the most potent inhibitory effect on P. acnes-induced IL-1β and IL-8 production in vitro.  The concentrations without apparent cytotoxicity toward THP-1 cells, assayed by MTT, were used for the subsequent experiments (Table 1). Seven flavones, at various concentrations, significantly reduced IL-8 and IL-1β levels ( Table 1). The potency of the flavones was expressed as IC 50 value. IC 50 values of FL4 for IL-8, and FL1 and FL6 for IL1β could not be precisely determined because the degree of inhibition provided by the highest test concentration was less than 50%. The rank order of potency of these flavones for IL- Therefore, FL2 had the most potent inhibitory effect on P. acnes-induced IL-1β and IL-8 production in vitro.

Effects of the Seven Flavones Isolated from Scutellariae Radix on P. acnes-Induced Ear Edema and Cytokine Production in Mouse Ear Homogenates
Mouse ear edema was induced by intradermal injection of P. acnes to mouse ears. Concomitant injection of each of the seven flavones with P. acnes afforded suppression of P. acnes-induced edema as measured by ear thickness, and all the seven flavones had almost equal effects ( Figure 3A). All flavones also significantly inhibited IL-6 ( Figure 3B) and TNF-α ( Figure 3C) production in P. acnes-treated mouse ears. For the suppression of IL-1β, except FL1 and FL2, all the other five flavones had significant effect ( Figure 3D). Our data indicated that these flavones had protective effects against P. acnes-induced skin inflammation.

Discussion
The pharmacological effects of the water and organic solvent extracts of Scutellariae radix have been extensively studied [8]. There have been over 40 flavonoids isolated from Scutellariae radix, with flavones and their glycosides being the most abundant [9]. Seven flavones isolated in this study existed in the forms of aglycones. FL1 (oroxylin A) and FL2 (wogonin) are the major flavonoids identified in Scutellariae radix [10]. FL3 to 7 are minor components [8], considerably less attention has been paid to these flavones.
The anti-inflammatory effects of FL1, 2, and 3 have been reported. Oroxylin A (FL1) inhibited LPS-induced iNOS and COX-2 gene expression by blocking NF-κB activation in vitro [11]. It also suppressed LPS-induced angiogenesis by down-regulation of toll-like receptor TLR-4 and the activity of mitogen-activated protein kinases (MAPK) [12].

Discussion
The pharmacological effects of the water and organic solvent extracts of Scutellariae radix have been extensively studied [8]. There have been over 40 flavonoids isolated from Scutellariae radix, with flavones and their glycosides being the most abundant [9]. Seven flavones isolated in this study existed in the forms of aglycones. FL1 (oroxylin A) and FL2 (wogonin) are the major flavonoids identified in Scutellariae radix [10]. FL3 to 7 are minor components [8], considerably less attention has been paid to these flavones.
To compare the in vitro inhibitory effects of seven flavones on P. acnes-induced IL-8 and IL-1β production, IC 50 values were used. We found wogonin (FL2) was the most potent (Table 1). FL1, 2, 4, and 5 with IC 50 values smaller than 20 µM for both IL-1β and IL-8 were more potent than FL3, 6, and 7 which had IC 50 values larger than 50 µM. The two hydroxyl groups at C5 and C7 in FL1, 2, 5 might contribute to their high potency. FL7 also has hydroxylated C5 and C7, but showed the lowest potency probably because it has too many substituents in B ring. FL3, 4, 6 with a methoxylated C7, possessing only one hydroxyl group at C5 were assumed to have lower potency. However, FL4 was an exception, probably due to the extra methoxyl group at C6. The number of methoxyl group may also influence the potency. FL1, 2, 5 have only one methoxyl group, while the less potent flavones, FL3, 6, 7 have two methoxyl groups. So, fewer than two methoxyl groups might be better for potency. However, FL4 has four methoxyl groups, but still had high potency. Nikaido, et al. [20] investigated the structure-activity relationship of flavones from Scutellariae radix for cAMP phosphodiesterase inhibition, and found when the flavones had more than five substituents, those with more methoxyl groups were more effective [20]. FL4 and FL7 have six, and FL6 has five substituents. FL4 has four methoxyl groups, while FL6 and 7 has two. The higher potency of FL4 than FL6 and 7 appeared to be consistent with their hypothesis [20].
To our best knowledge, this is the first study to test the suppression effects of these flavones on P. acnes-induced inflammation in vivo. Due to the complicated cell populations involved in ear inflammation, and a fixed dose used for all the flavones, the effectiveness of each flavone cannot be completely compared as was ranked in vitro. However each of the seven flavones effectively reduced P. acnes-induced ear swelling and strongly suppressed the production of TNF-α and IL-6 in mice ( Figure 3). IL-6, a pro-inflammatory and chemotactic factor [22], was measured in mice instead of IL-8 that we measured in human THP-1 cells because mice do not produce IL-8 [23].
Luteolin, a positive control in this study, is also a flavone, but without any methoxyl group. Our previous study found it inhibited P. acnes-induced pro-inflammatory cytokines expression by inactivating NF-κB through the suppression of mitogen-activated protein kinases (MAPK) phosphorylation [24]. Although the molecular mechanisms by which these seven flavones from Scutellariae radix modulate pro-inflammatory cytokine levels are not elucidated, our in vitro and in vivo results suggest each of the seven flavones has promising therapeutic potential against inflammatory acne vulgaris. Our results open a new aspect of the pharmacological role of these flavones.

Culture of P. acnes and Preparation of Heat-Killed Bacteria
The strain of P. acnes (BCRC10723, isolated from facial acne) was obtained from the Bioresource Collection and Research Center (Hsinchu, Taiwan). P. acnes was cultured in brain heart infusion (BHI) broth (Difco, Detroit, MI, USA) with 1% glucose in an anaerobic atmosphere using BBL GasPak systems (Becton Dickinson Microbiology Systems, Cockeysville, MD, USA) at 37˝C. A spectrophotometer OD600 (Chrom Tech, Apple Valley, MN, USA) was used to determine the bacterial log phase of growth. The log-phase bacterial culture was harvested, washed thrice with phosphate-buffered saline (PBS) and re-suspended in PBS for induction of mouse ear edema. For the in vitro experiments, heat-killed P. acnes was prepared. Log-phase bacteria were washed with PBS, incubated at 100˝C for 30 min, re-suspended in RPMI medium (Gibco, Carlsbad, CA, USA) and stored at 4˝C until use.

Determination of the Viability of THP-1 Cells
The human monocytic THP-1 cell line (BCRC 60430) was obtained from the Bioresource Collection and Research Center (Hsinchu, Taiwan). Cells were maintained in RPMI 1640 supplemented with 10% heat-inactivated fetal bovine serum (FBS, Gibco), penicillin (100 U/mL), and streptomycin (100 µg/mL) at 37˝C in a humidified atmosphere with 5% CO 2 . To determine the cytotoxicity of tested samples, a suspension of THP-1 cells (1ˆ10 6 cells/mL) was treated with various concentrations of tested samples in 96-well culture plates for 24 h at 37˝C. Whole cell suspension was taken from each well and centrifuged for 4 min at 600 g. Supernatant was removed, and the pellet was incubated with 100 µL of MTT reagent (Sigma-Aldrich) for 3 h at 37˝C. Samples were centrifuged for 2 min at 4500 g, and supernatant was gently removed. Finally, the converted dye from the MTT reagent in cell pellets was solubilized with 500 µL of isopropanol/HCl, and then 200 µL of each sample was transferred in duplicates to a 96-well plate. The absorbance was measured using a Synergy HT multi-detection micro-plate reader (BioTek, Winooski, VT, USA) at 540 nm with 690 nm as the reference wavelength.

Measurement of Cytokine Production in Human Monocytic THP-1 Cells
THP-1 cells were seeded at 1ˆ10 6 cells/mL in 24-well plates with serum-free medium, and were treated with heat-killed P. acnes (7.5ˆ10 7 CFU/mL; multiplicity of infection (M.O.I.) = 75) alone or in combination with different concentrations of tested samples for a 24-h incubation. Cell-free supernatants were collected, and concentrations of IL-1β and IL-8 were analyzed with respective enzyme immunoassay kits (BioLegend, San Diego, CA, USA). The half maximal inhibitory concentration (IC 50 ) values were estimated using a nonlinear regression algorithm (SigmaPlot 12; SPSS Inc. Chicago, IL, USA).

P. acnes-Induced Ear Edema and Measurement of Cytokine Levels In Vivo
Eight-week-old male ICR mice were purchased from the Animal Center of College of Medicine, National Taiwan University, Taipei, Taiwan. Animal experiments were approved by the Animal Care Committee of the National Taiwan Normal University. Mice were fed with chow diet and water ad libitum. To examine the anti-inflammatory effect of flavones in vivo, an intradermal injection model was employed [24]. In the preliminary test, 10 µL of flavones (up to 20 µg/site) was injected into mice ears. No noticeable skin irritation occurred (data not shown). Hence, flavones (20 µg/site) were used for the following experiments. Mice were randomly grouped (n = 5 per group). P. acnes (6ˆ10 7 CFU per 10 µL in PBS) was injected into the left ear of ICR mice. Right ears received an equal amount (10 µL) of PBS. Ten microliters of flavones in 5% DMSO in PBS was injected into the same location of both ears right after P. acnes or PBS injection. Twelve hours after bacterial injection, the increase in ear thickness was measured using a micro-caliper (Mitutoyo, Kanagawa, Japan). The increase in ear thickness of the P. acnes-injected ear was calculated and expressed as percentage of the PBS-injected control.

Measurement of Cytokine Levels In Vivo
After thickness had been measured, the ears were excised (n = 5 per group). The ear samples were homogenized using a BioMasher III (Nippi Inc., Tokyo, Japan) in radioimmunoprecipitation (RIPA) buffer (G-Biosciences, St. Louis, MO, USA) supplemented with 1 mM phenylmethylsulfonyl fluoride (PMSF) for 1 min on ice. The homogenates were vortexed and centrifuged at 10,000 g for 10 min at 4˝C. The supernatant was reserved and stored at´80˝C for the determinations of TNF-α, IL-1β, and IL-6 levels according to the manufacturer's instruction (BioLegend).

Statistical Analysis
All data are presented as the mean˘standard deviation (SD). Statistical analyses were performed using the SPSS 19.0 statistical package. 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.