Anti-Inflammatory Effects of Artemisia Leaf Extract in Mice with Contact Dermatitis In Vitro and In Vivo

The leaves of Artemisia argyi Lev. et Vant. and A. princeps Pamp. are well known medicinal herbs used to treat patients in China, Japan, and Korea with skin problems such as eczema and itching, as well as abdominal pain and dysmenorrhoea. We investigated the anti-inflammatory effects of Artemisia leaf extract (ALE) using CD mice and Raw 264.7 cells. The effects of ALE on histopathological changes and cytokine production in ear tissues were assessed in mice with CD induced by 1-fluoro-2,4-dinitrobenzene (DNFB). Moreover, the anti-inflammatory effects on production levels of prostaglandin E2 (PGE2) and nitric oxide (NO) and expression levels of cyclooxygenase 2 (COX-2) and inducible nitric oxide synthase (iNOS) were investigated in Raw 264.7 cells. Topical application of ALE effectively prevented ear swelling induced by repeated DNFB application. ALE prevented epidermal hyperplasia and infiltration of immune cells and lowered the production of interferon- (IFN-) gamma (γ), tumour necrosis factor- (TNF-) alpha (α), and interleukin- (IL-) 6 in inflamed tissues. In addition, ALE inhibited expression of COX-2 and iNOS and production of NO and PGE2 in Raw 264.7 cells. These results indicate that Artemisia leaf can be used as a therapeutic agent for inflammatory skin diseases and that its anti-inflammatory effects are closely related to the inhibition of inflammatory mediator release from macrophages and inflammatory cytokine production in inflamed tissues.


Introduction
Contact dermatitis (CD) in the framework of occupational diseases remains prevalent among workers worldwide, impacting their quality of life and workability. The employees most affected by CD are hairdressers, healthcare workers, and metal workers [1] because they are continuously exposed to harmful environments when working. As a result, employees with CD tend to use anti-inflammatory and immunomodulatory agents such as corticosteroids repeatedly [2]. Corticosteroids are effective and powerful agents for CD, but their doses should be restricted because of their adverse side effects. Herbal medicines have recently emerged in the framework of complementary and alternative medicines (CAM) for corticosteroids because they have relatively lower cost and safety [3].
Herbs belonging to the Artemisia genus are widely used as medicine worldwide. The leaves of Artemisia argyi Lev.
et Vant. and A. princeps Pamp. are frequently used as traditional or folk medicines for patients with abdominal pain, dysmenorrhoea, uterine haemorrhage, and inflammation in China, Japan, and Korea [4]. A. argyi, Chinese mugwort, is herbaceous perennial plant known in Japanese as gaiyou and in Chinese as aiye. A. princeps, Japanese mugwort, is a perennial and very vigorous plant known as yomogi in Japanese. Recently, the leaves of A. argyi and A. princeps and their components have been shown to have antitumour [5][6][7][8], antifungal [9], anticoagulant [10], antidiabetic [11], and antiinflammatory [12,13] activity.
Based on these findings, we examined the effects of Artemisia leaf extract (ALE) on inflamed tissues in mice with CD and anti-inflammatory activities in Raw 264.7 cells. Specifically, the effects of ALE on histopathological changes including ear swelling, epidermal hyperplasia, immune cell infiltration, and cytokine production such as interferon-(IFN-) gamma ( ), tumour necrosis factor-(TNF-) alpha ( ), 2 Mediators of Inflammation interleukin-(IL-) 6, and IL-10 in ear tissues were assessed in mice with CD induced by topical application of 1-fluoro-2,4dinitrobenzene (DNFB). The anti-inflammatory effects on production levels of prostaglandin E 2 (PGE 2 ) and nitric oxide (NO) and the expression levels of cyclooxygenase 2 (COX-2) and inducible nitric oxide synthase (iNOS) were investigated in Raw 264.7 cells.

Preparation of ALE.
Artemisia leaf was purchased from Gwangmyungdang (Ulsan, Korea). The Artemisia leaf consisted of a mixture of A. argyi and A. princeps and was authenticated by Professor Jung-Hoon Kim, one of the authors of this study. Twenty grams of Artemisia leaf was immersed in 500 mL of methyl alcohol and sonicated for 15 min, after which they were extracted for 24 h. Following extraction, the supernatant was transferred and the Artemisia leaf was again extracted with 500 mL of methanol for 24 h. The two extracts were then combined and filtered through Whatman no. 20 filter paper, after which they were condensed using a rotary evaporator (EYELA, Tokyo, Japan). The evaporated extract was subsequently dried using a lyophilizer (Labconco, Kansas City, MO, USA), which yielded 1.04 g of freeze-dried powder (yield, 5.21%). Specimens of crude material and Artemisia leaf methanol extract (ALE, Voucher no. MH2013-040) were deposited in the herbarium located in the School of Korean Medicine, Pusan National University.

Animals.
Six-week-old male Balb/c mice were obtained from Samtako (Incheon, Korea). All mice used in this experiment were housed in the cages under specific conditions, including a 12 h light/dark cycle and specific pathogen-free conditions. In addition, mice were provided with free access to standard rodent feed and water. We conducted all animal experiments according to institutional guidelines and all experimental procedures were approved by our animal care committee (PNU-2012-0140).

CD Induction and Experimental
Schedule. CD was induced using our standard method as previously described [14]. Briefly, 0.1% DNFB (50 L) in vehicle composed of acetone and olive oil (4 : 1, AOO) was applied onto the shaved back of mice for three successive days (sensitization). Mice were then treated by application of 0.2% DNFB (30 L) in vehicle onto the backside of their ears every two days. For topical treatment with drugs, dexamethasone (DEX) and ALE were dissolved in ethanol, filtered using a syringe filter (0.45 m), and finally diluted in vehicle (AOO : ethanol, 4 : 1). ALE (30 or 300 g/ear) was topically applied onto the backside of ears for 7 days. A total of 36 mice were used in this study (NOR group and DEX group, 6 mice; CTL group and ALE group, 8 mice). The experimental procedures are summarized in Figure 1.

Effects on Ear Thickness and Weight.
Mice were sacrificed with CO 2 , after which ear pieces (5 mm in diameter) obtained via dermal punch were weighed using a microbalance and the thicknesses of both ears were measured with digimatic calipers (Mitutoyo, Kanagawa, Japan) at the same time.

Tissue Preparation and Staining.
Obtained tissues were fixed in 4% formalin for 24 h and then dehydrated using ethyl alcohol. Next, all tissues were soaked in xylene and finally embedded in paraffin. Ear tissues (4 m) were subsequently resected, after which sections were stained with haematoxylin-eosin (H/E) and observed using a light microscope (50x).

Evaluation of Hyperplasia in the Epidermis and Infiltration of Immune Cells.
To evaluate hyperplasia in the epidermis and infiltration of immune cells, five nonoverlapping fields per slide were randomly selected and captured with a light microscope. The height from the basal lamina to the top of the stratum granulosum was quantified to evaluate the epidermal thickness. Five lengths were used to calculate the mean epidermal thicknesses of each tissue slide. Total immune cell numbers were quantified by counting immune cells in the same size counting grid.

Measurement of Cytokine Production.
Cytokine levels in ear tissues were evaluated using a mouse inflammation cytometric bead array (CBA) kit (BD Biosciences, San Jose, CA, USA). Briefly, to obtain tissue lysates, resected inflamed tissues were lysed using protein extraction solution (Intron Bio, Daejeon, Korea) and a homogenizer (Next Advance, NY, USA). Next, 50 g of lysates was used to evaluate the levels of TNF-, IFN-, IL-6, and IL-10.

Determination of Nitric Oxide (NO) Production.
Cells were seeded in 96-well plates at a density of 1 × 10 4 cells/well and then incubated overnight. Next, cells were treated with the indicated concentrations of ALE for 4 h, after which they were stimulated with 1 g/mL of lipopolysaccharide (LPS) for 20 h. Following stimulation, 100 L of supernatants was mixed with 100 L of Griess Reagent (2% sulphanilamide in 10% H 3 PO 4 and 0.2% of N-(1-naphthyl)ethylenediamine in distilled water) and then incubated at room temperature for 10 min. The absorbance at 540 nm was subsequently measured using a spectrophotometer (TECAN, Männedorf, Switzerland). The production levels of NO were determined using a NaNO 2 serial dilution standard curve.

Measurement of Prostaglandin E2 (PGE 2 ) Production.
The production of PGE 2 was measured by enzyme-linked immunosorbent assay (ELISA) using PGE 2 ELISA assay kit (Enzo Life Science, Farmingdale, NY, USA). Briefly, cells were treated as previously described. The culture supernatants were then collected, after which the optical densities were measured using a spectrophotometer (TECAN, Männedorf, Switzerland) at a wavelength of 405 nm.

ALE Prevented Ear Swelling Induced by Repeated DNFB
Application. At the end of experiment, the thicknesses and weights of both ears were evaluated. In CTL group, repeated treatment with DNFB elevated the levels of ear thickness and weight more than three times relative to the normal group. 30 and 300 g/ear of ALE treatment effectively inhibited enlargement of ear thickness (Figure 2(a)) and 300 g/ear of ALE inhibited ear weight gain significantly (Figure 2(b)). DEX was more effective than ALE (Figure 2). a large pustule (filled arrow) and vesicles were also observed ( Figure 3(a)). Treatment with ALE effectively inhibited epidermal hyperplasia and reduced immune cell infiltration compared to the CTL mice (Figure 3(b)). DEX treatment was most effective among all experimental groups (Figure 3).

ALE Lowered the Production Levels of TNF-, IFN-,
and IL-6 in Inflamed Tissues. We also checked the effects of ALE on cytokine productions in inflamed tissues. In our results, elevated levels of TNF-, IFN-, and IL-6 production were observed in CTL mice (Figure 4). These increases in inflammatory cytokines were significantly prevented by topical application of ALE. Treatment with 300 g/ear of ALE significantly lowered the production levels of TNF-, IFN-, and IL-6, respectively. The IL-10 production levels were not affected by induction of CD or treatment with ALE, and only the DEX treated group showed lower production of IL-10 than the normal and control groups (Figure 4).

ALE Inhibited iNOS Expression and NO Production in
Raw 264.7 Cells. The inhibitory effects of ALE on iNOS expression induced by LPS were investigated. Expression of iNOS was elevated in the LPS treated CTL group compared to the normal group. Treatment with more than 25 g/mL of ALE lowered iNOS expression in a concentration dependent manner (Figure 5(a)). In addition, treatment with LPS elevated NO production level more than nine times relative to neither LPS nor ALE treated group. This increase in No production was effectively inhibited by ALE in a dose dependent fashion (Figure 5(b)).

ALE Inhibited COX-2 Expression and PGE 2 Production
in Raw 264.7 Cells. The LPS stimulated Raw 264.7 cells showed marked increases in COX-2, representative COX enzyme, and treatment with ALE inhibited COX-2 expression (Figure 6(a)). In addition, the production of PGE 2 , one of the major final metabolites, was markedly induced by LPS stimulation. Treatment with 50 mg/mL of ALE inhibited PGE 2 production significantly (Figure 6(b)).

Discussion
Artemisia species have traditionally been utilized for amelioration of diseases including malaria, hepatitis, parasites, and various cancers worldwide. The anti-inflammatory effects of A. princeps in antigen-stimulated T cells and regulatory T cells have been reported by Chang et al. [15]. Additionally, the major active components of Artemisia species are sesquiterpenoids and flavonoids such as eupatilin and jaceosidin, which are extracted from most herbs in Artemisia species, and these components have been reported to have anti-inflammatory effects in mice [16]. Based on the antiinflammatory and immunomodulatory effects of Artemisia species, we investigated whether ALE can reduce inflammatory reactions in an animal model of CD to ameliorate its symptoms.
The skin of CD patients tends to thicken because of chronic and repeated inflammatory reactions, which are closely related to epidermal and dermal hyperplasia [17]. In our experiment, repeated application of DNFB induced ear swelling, which was effectively inhibited by topical application of ALE (Figure 2). Given that the degree of ear swelling is recognized as an index of inflammatory reaction, these results imply that ALE can exert anti-inflammatory action in animal models of CD.
In our animal model of CD, repeated application of DNFB induced hyperplasia in the epidermis and massive infiltration of immune cells to the epidermis and into the connective tissues. In addition, spongiotic changes, vesicles, and pustules were seen in the control group (Figure 3(a)(B)). Topical application of ALE effectively inhibited hyperplasia in the epidermis and infiltration of immune cells. In addition, the area of spongiotic changes and vesicles was diminished and pustules were rarely seen in the ALE and DEX group ( Figure 3). These findings indicate that ALE can act as an anti-inflammatory agent in CD to reduce immune cell infiltration, resulting in inhibition of spongiotic changes, as well as pustule and vesicle formation.
TNF-and IFN-, hallmarks of Th1 skewing reaction, can stimulate keratinocytes, which subsequently proliferate, resulting in epidermal hyperplasia [18]. These two cytokines are also closely related to infiltration of immune cells into the inflamed tissues. In the inflammatory cascade of CD, activated keratinocytes can release TNF-and IL-6, which are growth promoting cytokines, resulting in accelerated immune cell infiltration and prolonged lifespan of immune cells [18]. IL-6 has also been reported to accelerate proliferation and migration of keratinocytes, leading to skin disorders, including psoriasis [19]. In this study, topical treatment with ALE effectively reduced production levels of TNF-, IFN-, and IL-6 in ear tissues ( Figure 4). When combined with previous results, these findings imply that ALE can inhibit hyperplasia in the epidermis and infiltration of immune cells via regulation of proinflammatory cytokines, finally leading to inhibition of enlargement of skin thickness.
Histopathological analyses revealed that many types of inflammatory cells, including macrophages, infiltrate into the dermis and epidermis in CD. Macrophages also play an important role by releasing inflammatory mediators such as NO, prostaglandins, and cytokines during initiation of CD. For these reasons, we also investigated anti-inflammatory effects of ALE in vitro using a macrophage cell line, Raw 264.7. Treatment with up to 100 g/mL of ALE had no effect on the viability of Raw 264.7 cells (data not shown). ALE significantly inhibited iNOS expression and NO production in a dose dependent manner ( Figure 5). In addition, ALE inhibited COX-2 expression and PGE 2 production induced by LPS stimulation (Figure 6). These findings indicate that ALE inhibits inflammatory response in macrophages via regulation of NO and PGE 2 production. As shown in Figures  5 and 6, ALE strongly inhibited iNOS expression and NO production, while it had little effect on PGE 2 production or COX-2 expression. This difference in efficacy may indicate that ALE can affect other subtypes of prostaglandin. It is well known that intracellular signaling pathways such as Erk, JNK, and p38 and transcription factors such as CREB and NF-B play a central role in activation and inflammatory mediator production in macrophages [20]. We checked the effects of ALE on expression levels of signaling pathways and the NF-B pathway. Treatment with ALE prevented phosphorylation of p38 but had no effect on the NF-B pathway (data not shown). Considering these results, the anti-inflammatory effects of ALE may be closely related to the p38 signaling pathway, and the NF-B pathway may not participate in the anti-inflammatory mechanisms of ALE.
Kim et al. reported immune-stimulatory effects of hot water extracts of the leaves of A. princeps [21]. Specifically, they found that extracts of A. princeps elevated production levels of NO and TNF-, which is inconsistent with our results. These differences may be attributed to different extraction methods. Many previous studies of A. princeps have shown anti-inflammatory or antiallergic effects, as well as antioxidative effects. In addition, components isolated from A. princeps such as eupatilin and jaceosidin have been reported to have inhibitory effects against IgE-induced hypersensitivity and carrageenan-induced inflammation, respectively [16,22]. However, Kim et al. did not describe why their findings were inconsistent with those of previous results. In addition, water extracts of plant materials should also be checked for microbial contamination before use because they are more easily contaminated than methanol or ethanol extracts.
Overall, the results of this study suggest that the antiinflammatory effects of ALE in CD mice are closely related to regulation of the activation of immune cells, especially macrophages and cytokine production in inflamed tissues.

Conclusion
In this study, ALE inhibited the release of NO and PGE 2 in macrophages. In addition, ALE effectively lowered the production levels of TNF-, IFN-, and IL-6 in inflamed tissues. These anti-inflammatory actions prevented epidermal hyperplasia and immune cell infiltration. Finally, ALE