Hizikia fusiformis fractions successfully improve atopic dermatitis indices in anti-CD3-stimulated splenocytes and 2,4-dinitrochlorobenzene-treated BALB/c mice

$Page 1: Hizikia fusiformis fractions successfully improve atopic dermatitis indices in anti-CD3-stimulated splenocytes and 2,4-dinitrochlorobenzene-treated BALB/c mice$
Hizikia fusiformis fractions successfully improve atopicdermatitis indices in anti-CD3-stimulated splenocytesand 2,4-dinitrochlorobenzene-treated BALB/c miceKyu Ho Leea*, Hee Jung Kima*, Hae Bok Kima, Seung Tae Kima, Young Ri Choia, Da Woom Seoa,Jung Min Yua, Su Kil Janga, Sang Moo Kimb, Do-Ik Leec and Seong Soo Jooa

aDepartment of Marine Molecular Biotechnology, College of Life Science, bDepartment of Marine Food Science and Technology,Gangneung-Wonju National University, Gangwon and cDepartment of Immunology, College of Pharmacy, Chung-Ang University, Seoul, Korea

Keywords2,4-dinitrochlorobenzene; atopic dermatitis;butanoic acid; helper T cell; Hizikia fusiformis

CorrespondenceSeong Soo Joo, Department of MarineMolecular Biotechnology, College of LifeScience, Gangneung-Wonju NationalUniversity, 120 Gangneung Daehangno,Gangneung, Gangwon 210-702, Korea.E-mail: [email protected]

Received March 6, 2013Accepted October 10, 2013

doi: 10.1111/jphp.12179

*Equally contributed to this paper.

Abstract

Objectives In the present study, we aimed to examine whether fractions from anedible sea weed, Hizikia fusiformis, had immunomodulatory effects, particularlyan anti-atopic effect, by attenuating the expression of T cell-dependent cytokinesusing in-vitro and in-vivo animal atopic dermatitis-like models.Methods The anti-atopic activities were examined in in vitro, and a 2,4-dinitrochlorobenzene (DNCB)-induced atopic dermatitis-like mouse model usingquantitative real-time polymerase chain reaction, electrophoretic-mobility shiftand histophathological analysis.Key findings Our results showed that the final fraction (F2′) of H. fusiformiscontained a higher amount of butanoic acid which was not found in the otherfractions, and effectively inhibited T cell activation by inhibiting dephosphory-lation of nuclear factor of activated T cells in electrophoretic-mobility shift assay.As a consequence, helper T cell-dependent cytokines, such as interleukin-2, -4 andinterferon-γ, were significantly inhibited while activated with an anti-CD3 anti-body. We also showed that skin challenged with DNCB successfully recoveredwhen treated with 2.5 mg/kg, comparable to that by 0.25% prednicarbate. Theseresults indicate that F2′ may contribute to inhibit T cell activation by eliminatingTh cell-dependent cytokines.Conclusions Taken together, we concluded that F2′ containing butanoic acid maybe a new functional anti-atopic candidate, which probably acts through nuclearfactor of activated T cell inactivation mechanisms.

Introduction

Hizikia fusiformis (Harvey) Okamura (syn. Sargassumfusiforme (Harvey) Setchell) is one of the most commonbrown seaweeds growing wild on rocky coastlines aroundKorea, Japan and China. H. fusiformis has traditionally beenwidely consumed in Asia as food, and much attention hasbeen paid because of its high nutritional, pharmaceuticaland industrial value. Several studies have reported that anextract of H. fusiformis shows various pharmacologicalactivity (e.g. anti-tumour, immunomodulatory, anti-inflammatory and antioxidant).[1–4] In particular, the anti-inflammatory activity of H. fusiformis extracts suggestsapplications for several acute and chronic disorders includ-ing autoimmune or autoinflammatory disorders, neurode-generative diseases and cancer.[5] Atopic dermatitis (AD)

characterized by immunoglobulin E (IgE) antibodiesassociated with helper T cells that produce type 2 (Th2)cytokines is a chronic inflammatory skin disease associatedwith cutaneous hyper-reactivity to environmental triggersthat are the result of an inappropriate immune responsetriggering inflammation.[6] The initiation phase of AD isdriven by cytokines derived from activated, allergen-specificTh2-type cells (e.g. interleukin (IL)-4, -5, -6, -10, 13),whereas the predominance of Th1 cytokines (e.g. IL-2 andinterferon (IFN)-γ) is responsible for the chronic nature ofthe AD lesions and determines disease severity.[7] Thus,inactivating T cells could be a target for treating or amelio-rating irritable AD symptoms. Topical immunosuppressantssuch as corticosteroids, cyclosporine A (CsA), tacrolimus

bs_bs_banner

And PharmacologyJournal of Pharmacy

Research Paper

© 2013 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, 66, pp. 466–476466

mailto:[email protected]

and pimecrolimus have been introduced for treatingAD.[8] However, many concerns have been reported whenused these agents in long-term therapy because of theirlimited effectiveness or adverse effects. Recently, we havescreened >30 seaweeds for anti-allergic activity includingH. fusiformis. Among them, we selected H. fusiformisas it was likely to suppress T cell activation by inhibitingTh1/Th2-driven cytokine expression. In this study, weanalysed the components contained in a H. fusiformisextract and further examined the immunosuppressiveactivity of the extract in both in-vitro and in-vivo 2,4-dinitrochlorobenzene (DNCB)-treated atopic dermatitis-like animal models which were proven as a novel model forhuman AD compared with NC/Nga models.[9]

Materials and Methods

Preparation of extract and fractionation

Fresh H. fusiformis was collected from the east coast ofSouth Korea in October 2012. A voucher specimen(KWNU80942) was authenticated by Prof. Joo (Biopharma-ceutical Lab, College of Life Science, Gangneung NationalUniversity, Korea), and deposited at the herbarium of thecollege of Natural Science, Kangwon National University,Korea. Epiphytes, salt and sand were completely removedwith tap water. The samples were then sanitized with 70%ethanol, rinsed with deionized water and freeze-dried.Finely ground H. fusiformis (100 g) was steeped in 1 l of80% aqueous ethanol for 24 h repeatedly for 3 days at roomtemperature. The ethanol extracts were combined, filteredthrough filter paper (Whatmann International Ltd, Maid-stone, UK) and evaporated. Fractionation was sequentiallyperformed using n-hexane, ethyl acetate (EA), chloroform,n-butanol (BuOH) and water, and each fraction was evapo-rated and dried completely. The promising anti-allergicactivity of the EA extract prompted us to isolate and analyseits active components. Activated silica gel (230–400 mesh)was packed on a glass column (300 mm × 30 mm) usingchloroform, and the crude EA extract was loaded on top ofthe silica gel. The column was eluted stepwise with 1 l ofchloroform : methanol (75 : 25, v/v), 1 l of chloroform : EA(75 : 25 to 0 : 100, v/v), 1 l of EA : acetone (75 : 25 to0 : 100, v/v) and 1 l of acetone : methanol (75 : 25 to 0 : 100,v/v). Over 80 fractions (50 ml each) were collected and con-centrated with a rotary evaporator. An aliquot of each con-centrated fraction was loaded on activated silica gel thin-layer chromatography (TLC) plates (20 cm × 20 cm). Theplates were developed using chloroform : methanol : water(75 : 25 : 10) solvent. The spots were located by exposingthe plate to iodine fumes. Fractions with the same numberof spots and similar Rf values were pooled. The pooled frac-tions were numbered (F1–F4). As fraction one (F1)obtained from first step column chromatography showed

high anti-allergic activity, it was selected for furtherfractionation. Thus, bioactive fraction F1 was furtherfractionated using silica gel (230–400 mesh) column(300 mm × 30 mm). The column was eluted stepwise with1 l of chloroform : methanol (75 : 25 to 0 : 100, v/v) and 1 lof methanol : EA (75 : 25 to 0 : 100, v/v). Over 65 fractions(30 ml each) were collected and concentrated with a rotaryevaporator. An aliquot each concentrated fraction wasloaded on an activated silica gel TLC plate. Fractions withthe same number of spots and similar Rf values were pooledand numbered (F1′–F5′) (Fig. 1). These five fractions weretested for anti-AD activity.

Gas chromatography-massspectrometry (GC-MS)

All samples were dissolved in ethanol and identified with ananalytical Agilent Technologies 5975C GC/MS instrumentequipped with a CTC CombiPAL autosampler system.Chromatographic separation was carried out using heliumcarrier gas on an HP-5 column (250 μm × 0.25 μm × 30 m,Agilent Technologies, Santa Clara, CA, USA). A 10 μlaliquot of the sample was injected into a split injector thatwas operated in split mode using a 5 : 1 split ratio with split

Dried Hizikia fusiformis powder (100 g)

Extraction with 80% ethanol

Extraction with solvents

Ethanol extract (25 g)

Residuen-Hexane extract(2.4 g)

Ethyl acetate extract (11.5 g) Residue

F1

F1' F2' F3' F4' F5'

F2 F3 F4

Figure 1 Schematic representation of the extraction of biologicallyactive fractions from ethyl acetate extract by column chromatography.The yield of hexane and ethyl acetate extract is represented as g/100 gof dry powder.

Kyu Ho Lee et al. H. fusiformis improves atopic dermatitis

© 2013 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, 66, pp. 466–476 467

flow and column flow of 5 ml/min and 1 ml/min, respec-tively. The injector temperature was held at 250°C and thetransfer line was 250°C. The GC oven was held at 50°C for3 min, ramped at 10°C/min to 280°C and held for 3 min.The ion source temperature was 250°C with an electronimpact ionization energy of −70 V. Data were collected from35 m/z to 250 m/z using a detector voltage of 1059 V follow-ing a 4 min delay. The components were identified by com-paring their relative retention times and mass spectra withWiley7N library data of the GC-MS system.

Animals

Male BALB/c mice (7 weeks old) were purchased fromSamtaco (Osan, Kyunggi-do, South Korea), and wereadapted to laboratory conditions (temperature: 20 ± 2°C,relative humidity: 50%, light/dark cycle: 12 h) for 1 week.The animal experiments were approved by the Gangneung-Wonju National University Animal Care and Use Commit-tee (approval No. GWNU-2012-26), and all procedureswere conducted in accordance with the Guide for Careand Use of Laboratory Animals published by the NationalInstitutes of Health.

Induction of AD-like skin lesions inDNCB-treated mice

DNCB (Sigma-Aldrich, St. Louis, MO, USA) dissolved inacetone was used to induce dermatitis in the BALB/c mice.Briefly, outer ears were sensitized with 20 μl of 1% DNCBdaily for 7 days. After the first challenge, 20 μl of 0.5%DNCB was repeatedly applied to the outer ears for anadditional 5 weeks at 2-day intervals. At the same time(second challenge), untreated control (acetone : oliveoil = 4 : 1 mixture), Dermatop ointment 0.25% (predni-carbate, a therapeutic control, Sanofi-Aventis, Seoul, Korea)or EA extract (1.25 and 2.5 mg/kg) dissolved in 20 μlDMSO were evenly applied daily to the entire ear skin areaof the mice (n = 3 for each group) for 2 consecutive days.No substances were applied to the skin surface on the lastday of the experiment. The mice were then sacrificed, andthe skin and spleens were collected from the same animalfor further analysis.

Cell culture

Mouse spleens were aseptically isolated from BALB/c miceafter sacrifice, and single primary splenocytes were preparedby mechanical dissociation in cold phosphate bufferedsaline at pH 7.2. Erythrocytes were depleted using a redblood cell lysis buffer (eBioscience, San Diego, CA, USA)containing ammonium chloride, which lyses red blood cellswith minimal effect on lymphocytes. Splenocytes were cul-tured in RPMI1640 complete medium with 10% fetal

bovine serum (Hyclone, Logan, UT, USA). The mediumalso contained an anti-CD3 monoclonal antibody (mAb)(eBioscience) and varying concentrations of lyophilizedfractions. Splenocytes were cultured at 5 × 106 cells/ml in96-well microtiter plates or at 1 × 107 cells/ml in 24-wellplates and maintained at 37°C in a humidified atmospherecontaining 5% CO2. Cell-free supernatants were collectedafter 24 h of culture and stored at −70°C until analysis.

Cellular cytotoxicity (lactate dehydrogenase(LDH)) assay

The cytotoxicity induced by the F1 and F2′ fractions of the(EA) extract was quantified by measuring LDH release.LDH content was determined using a commercial non-radioactive LDH assay kit, CytoTox 96 (Promega, Madison,WI, USA), which is based on a coupled enzymatic reactionthat results in the conversion of a tetrazolium salt into a redformazan product. The increase in the amount of formazanproduced in the culture supernatant directly correlates withthe increase in the number of lysed cells. The formazan wasquantified spectrophotometrically by measuring its absorb-ance at 490 nm (Spectra Max 340, Molecular Devices, Sun-nyvale, CA, USA). Cytotoxicity in experimental samples wasdetermined as %LDH release compared with that in cellstreated with 1% Triton X-100.

Quantitative real-time polymerase chainreaction (PCR) assay

Total RNA extracts from mouse splenocytes stimulated withanti-CD3 mAb (in vitro) or DNCB (in vivo) were preparedusing the Trizol method (Invitrogen, Carlsbad, CA, USA).cDNA was synthesized from RNA by reverse transcriptionof 1 μg of total RNA using the Improm-II reverse transcrip-tion system (Promega) and oligo dT primers in a totalvolume of 20 μl. PCR amplification was performed usingthe primers described in Table 1 (Bioneer, Deajeon, Korea).Quantitative real-time PCR reactions were run on a Rotor-Gene 6000 (Corbett Research, Sydney, Australia) usingSYBR Green PCR Master Mix (Qiagen, Valencia, CA, USA)in 20 μl reaction mixtures. Each real-time PCR master mixcontained 10 μl 2 × enzyme mastermix, 7.0 μl RNase freewater, 1 μl of each primer (10 pM each) and 1 μl dilutedtemplate. The PCR was performed with an initial pre-incubation step for 10 min at 95°C, followed by 45 cycles of95°C for 15 s, annealing at 52°C for 15 s and extension at72°C for 10 s. Melting curve analysis was used to confirmformation of the expected PCR product, and products fromall assays were additionally tested with 1.2% agarose gelelectrophoresis to confirm the correct lengths. An inter-runcalibrator was used, and a standard curve was created foreach gene to obtain PCR efficiencies. Relative sampleexpression levels were calculated using Rotor-Gene 6000

Kyu Ho Lee et al.H. fusiformis improves atopic dermatitis


Series Software 1.7, and were expressed relative to β-actinand corrected for between-run variability. Data for theexperimental samples were expressed as a percentage of theinternal control gene. Butanoic acid (BA) (Cat# B103500,Sigma-Aldrich) was used under the same condition insplenocytes for a comparative analysis.

Electrophoretic-mobility shift Assay(EMSA)

Nuclear extracts for EMSA were prepared from BALB/cmouse splenocytes using Nuclear Extraction Reagents(Pierce, Rockford, IL, USA) according to the manufactu-rer’s instructions. EMSA was performed with thePanomics EMSA kit (Panomics, Freemont, CA, USA). Inbrief, nuclear extracts containing equal amounts of pro-teins for each sample were incubated with poly (dI-dC)(1 μg/μl) for 5 min, followed by adding binding buffer(20 mM HEPES pH 7.9, 1 mM DTT, 0.1 mM EDTA,50 mM KCl, 5% glycerol and 200 μg/ml BSA) andbiotinylated oligo (10 ng/μl). A fivefold excess of unlabeledoligo was added before adding the biotinylated probe tocontrol for specificity of binding for selected samples.Binding reaction mixtures were incubated for 30 min atroom temperature. Protein-DNA complexes were separatedon 6% non-denaturing polyacrylamide gel in 0.5 × Tris-borate/EDTA buffer (0.1 M Tris, 0.09 M boric acid con-taining 1 mM EDTA) at 4°C. After electrophoresis, the gelswere transferred to Biodyne B nylon membranes (ThermoScientific, Rockford, IL, USA). The transferred oligos wereimmobilized by UV cross-linking for 10 min. The mem-branes were blocked using blocking buffer to detect thebound oligos (Panomics EMSA Gel-Shift Kit) followed byadding streptavidin-HRP. Then, the membranes were dis-tributed in a working substrate solution according to themanufacturer’s instructions for 5 min at room tempera-ture. Finally, the membranes were exposed usingHyperfilm ECL, and the bands were detected withGelquant software (MiniBIS Pro, Jerusalem, Israel).

Histopathological analysis

Ears were fixed with 10% neutral buffered formalin(sodium phosphate, monobasic (4 g), sodium phosphate,

dibasic (6.5 g), formaldehyde, 37% (100 ml) and distilledwater (900 ml)) for 5 days. The tissues were then embeddedin paraffin, cut into sections (5 μm) and stained with aH&E solution. All tissue samples were examined andimaged in a blinded fashion. Images were captured using aNikon Eclips Ti-S inverted microscope (Nikon, Tokyo,Japan) at a magnification of ×200.

Statistical analysis

The Kruskal–Wallis one-way analysis of variance with aDunnett’s post-hoc test was performed using SPSS software(v. 13) (Chicago, IL, USA) to determine if the differencesbetween groups were significant. A P < 0.05 was consideredsignificant.

Results

GC-MS analysis

We found that EA extracts contained various componentsbetween retention times of 4.154 and 21.706 min. Themajor constituents of the EA extracts from H. fusiformiswere ethyl propanoate (60.2%) and 1-ethyl butane (20.5%),which were all volatiles. Of the remaining compounds,a lesser amount of BA (0.35%) that has been reported tohave immune-modulatory and anti-inflammatory activitywas observed (Table 2). As BA was considered a majorcomponent associated with anti-inflammatory activity,[10]

we further analysed the Fl and F1′–F5′ fractions to deter-mine whether they contained BA (Figure 2a and 2b). BAwas confirmed by analysing the results with commercial BA(Sigma-Aldrich) (Figure 2c). The BA area (%) was high inF1 (0.25), F1′ (0.08) and F2′ (0.14), whereas it was notdetectable in other fractions under the same injectionconditions (50 mg/ml).

Splenocytes cytotoxicity

The cytotoxicity induced by the EA extract and the F1 andF2′ fractions of splenocytes was quantified by measuringLDH release at varying ranges of concentration (0.1–100 μg/ml). Incubating mouse splenocytes with each of the

Table 1 Primer sequences for real-time reverse transcription polymerase chain reaction

Gene Primer Amino acid sequences Product size (bp) Accession No.

IL-2 5′ Primer 5′- AGCTCTACAGCGGAAGCACA 236 NM_0083663′ Primer 5′- GTCAAATCCAGAACATGCCG

IFN-γ 5′ Primer 5′- TGAAAATCCTGCAGAGCCAG 193 NM_0083373′ Primer 5′- TGGACCTGTGGGTTGTTGAC

IL-4 5′ Primer 5′- ATATCCACGGATGCGACAAA 252 M258923′ Primer 5′- AAGCCCGAAAGAGTCTCTGC

β-actin 5′ Primer 5′- TACAGCTTCACCACCACAGC 187 NM_0073933′ Primer 5′- AAGGAAGGCTGGAAAAGAGC



study samples did not result in cell cytotoxicity except at thehighest dose. Figure 3 shows that the F1 and F2′ fractionsand the EA extract did not significantly increase LDHrelease for 0.1–10 μg/ml exposures for up to 24 h but a

higher concentration (100 μg/ml) induced a significantincrease in LDH release. As the greatest cytotoxic effect wasobserved at 100 μg/ml, no higher concentrations weretested.

Table 2 Comparison of the gas chromatography/mass spectroscopy (GC-MS) library of the EA extracta

PeakRetentiontime (min) library

Quality(%) %Area

1 4.154 1-Ethoxy butane 91 20.52 4.319 Ethyl propanoate 90 60.23 5.128 butanoic acid 53 0.44 16.566 2,7,10-Trimethylspiro[acridane-9,9′(10′H)-anthracene]-10′-one 64 5.15 19.122 N-methyl-4-(P-hydroxybenzyl)-1,2,3,4-tetrahydroisoquinoline 43 8.36 19.663 3-methyl-cyclooctane 52 0.67 19.901 2,6-Diisopropylnaphthalene(DIPN) 50 0.28 20.093 (S)-4-methyl-1-cyclopentene-carboxaldehyde 52 0.2

3-methyl-1,4-heptadiene 525-methyl-3-heptyne 49

9 20.17 5,5,8-trimethyl-4a,5,6,7,8,8a-hexahydro-N-(2′,2′-dimethyl-4′-phenyl-1′,3′-dioxanyl)pyrrolo[3,4-a]naphthalene

64 2.6

10 20.264 cis-Carvone oxide 23 0.711 21.644 (R,R)-1-((Z)-hex-1’-enyl)-2-ethenyl

cyclopropane62 0.9

12 21.706 1,3-cyclooctadiene 53 1.3

aThe components were identified based on a comparison of their relative retention times and mass spectra with the Wiley7N library data from theGC-MS system.

(a) (b) (c)

Abundance1 2

140000

130000

120000

110000

100000

90000

80000

70000

60000

50000

40000

30000

20000

10000

Time

Time

5.00 6.00

5.00 6.00

Time 5.00 6.00 Time 5.00 6.00 Time 5.00 6.00

7.00 8.00 9.00 10.00 11.00 12.00 13.00 14.00 15.00 16.00 17.00 18.00 19.00 20.00 21.00 22.00 23.00 24.00 25.00 26.00

3

Butanoic acid

Butanoic acid

OH

O

46

78

10

11

12

9

5

20000

15000

10000

5000

0

5000

10000

10000

20000

30000

10000

20000

30000

F1 F1‘ F2‘

Figure 2 Gas chromatography-mass spectroscopy (GC-MS) chromatogram of ethyl acetate extract (a), F1, F1′ and F2′ fractions (b) and butanoicacid (c). The peaks were analysed based on the GC-MS library as listed in Table 2. The presence of butanoic acid in each fraction was identified bythe same retention time (5 min) and a single butanoic acid peak was referenced as shown in (c).



Profiles of Th1/Th2 cytokine mRNAs inmouse splenocytes

Splenocytes were activated by an anti-CD3 mAb to deter-mine if the F1 and F2′ fractions and the EA extract consist-ently inhibited T cell receptor-mediated Th1/Th2 cytokineexpression at the gene level. IL-2, IFN-γ and IL-4 mRNAexpression was examined in the presence of each sample atvarious concentrations (1–25 μg/ml for EA extract or10 μg/ml for the F1 and F2′ fractions). CsA was included forcomparison, and mRNA levels for the selected Th1/Th2cytokines were assessed using quantitative real-time PCR. ABuOH extract from H. fusiformis was also included for acomparative analysis. As shown in Figure 4a–c, T cells werehighly activated by the anti-CD3 mAb. However, IL-2,

IFN-γ and IL-4 mRNA expression levels were maximallyinhibited after treatment with 25 μg/ml of the EA extract.The first round fractions (F1–F4) of the EA extract revealedthat the F1 fraction was more potent for inhibiting Th1/Th2-dependent cytokines than that of the other three frac-tions (F2–F4) (Figure 5a–c). Interestingly, the second roundfractions (F1′–F5′) of the F1 fraction inhibited IL-2, IFN-γand IL-4 mRNA expression. As shown in Figure 6a–c, theF1′ and F2′ fractions, which contained BA, significantlyinhibited those cytokines but the F3′–F5′ fractions were noteffective. Thus, these results clearly demonstrate that theoverall anti-AD effect was from the BA contained inH. fusiformis. BA alone was compared as shown inFigure 6d–f, and effectively inhibited IL-2, IFN-γ and IL-4expression as expected.

Ctrl H2O2 H2O2

H2O2H2O2

0.1 1 10 1000

20

40

60

80

100

EA extract (μg/mL)

F1 (μg/mL) F2' (μg/mL)

BuOH extract (μg/mL)

Cyt

oto

xici

ty (

%)

vs C

trl

Ctrl 0.1 1 10 1000

20

40

60

80

100

Ctrl 0.1 1 10 1000

20

40

60

80

***

***

***

***

***

***

***

**Cyt

oto

xici

ty (

%)

vs C

trl

Cyt

oto

xici

ty (

%)

vs C

trl

Cyt

oto

xici

ty (

%)

vs C

trl

Ctrl 0.1 1 10 1000

20

40

60

80

100

(a) (b)

(c) (d)

Figure 3 Cytotoxicity in mouse splenocytes. Splenocytes were prepared from BALB/c mice, seeded in 96-well microplates and cultured withvarying concentrations of each fraction for 24 h. Concentration-dependent cytotoxicity, measured as percent lactate dehydrogenase (LDH)released into the culture supernatant, was compared with non-treated control cells (Ctrl). H2O2 (1 mM) was used as the positive control. Resultsare expressed as means ± standard deviations from three separate experiments. *P < 0.05, *** P < 0.001 versus Ctrl. Ctrl, control.



Ctrl - CsA 1 10 25 1 10 25 (μg/mL)

25 (μg/mL)

25 (μg/mL)

0.0

0.5

1.0

1.5

EA BuOHanti-CD3

***

******

*** ***

***

Rel

etiv

e co

nce

ntr

atio

nn

orm

aliz

ed t

o β

-act

in(I

L-2

mR

NA

)

Rel

etiv

e co

nce

ntr

atio

nn

orm

aliz

ed t

o β

-act

in(I

L-4

mR

NA

)

Rel

etiv

e co

nce

ntr

atio

nn

orm

aliz

ed t

o β

-act

in(I

FNγ

mR

NA

)

Ctrl - CsA 1 10 25 1 100.0

0.5

1.0

1.5

EA BuOH

anti-CD3

***

***

******

***

*** ***

Ctrl - CsA 1 10 25 1 100.0

0.5

1.0

1.5

EA BuOH

anti-CD3

***

****** ***

***** ***

(a)

(b)

Figure 4 Th1/Th2 cytokine mRNA expression in splenocytes. Cellswere cultured for 24 h in the presence with ethyl acetate (EA) or theBuOH extract at varying concentrations of 1–25 μg/ml and the cellswere stimulated with 1 μg/ml anti-CD3 monoclonal antibody (mAb).cyclosporine A (CsA) was used as a positive control. Isolated mRNAswere analysed by real-time reverse transcription polymerase chain reac-tion for interleukin (IL)-2 (a), interferon (IFN)-γ (b) and IL-4 (c). Resultswere internally confirmed by the comparative cycle count (Ct, cyclenumber threshold) against β-actin as the standard gene. An anti-CD3mAb group was used as a calibrator for a relative comparison foreach group. Results are mean ± standard deviation (n = 3). **P < 0.01,*** P < 0.001 versus anti-CD3-treated group. Ctrl, control.

Ctrl - CsA F1 F2 F3 F4 (10 μg/mL)0.0

0.5

1.0

1.5

anti-CD3

******

**

***


0.5

1.0

1.5

anti-CD3

***

******


0.5

1.0

1.5

anti-CD3

***

*****

Rel

etiv

e co

nce

ntr

atio

nn

orm

aliz

ed t

o β

-act

in(I

L-2

mR

NA

)

Rel

etiv

e co

nce

ntr

atio

nn

orm

aliz

ed t

o β

-act

in(I

L-4

mR

NA

)

Rel

etiv

e co

nce

ntr

atio

nn

orm

aliz

ed t

o β

-act

in(I

FNγ

mR

NA

)

(a)

(b)

(c)

Figure 5 Quantitative mRNA analysis of Th1/Th2 cytokines insplenocytes. Cells were cultured for 24 h in the presence of fractionsF1–F5 at a fixed concentration of 10 μg/ml, and cells were stimulatedwith 1 μg/ml anti-CD3 monoclonal antibody (mAb). cyclosporine A(CsA) (0.2 μg/ml) was used as the positive control. Isolated mRNAswere analysed by real-time reverse transcription polymerase chainreaction for interleukin (IL)-2 (a), interferon (IFN)-γ (b) and IL-4 (c).Results were internally confirmed by the comparative cycle count(Ct, cycle number threshold) against β-actin as the standard gene.An anti-CD3 mAb group was used as a calibrator for a relativecomparison for each group. Results are mean ± standard deviation(n = 3). **P < 0.01, *** P < 0.001 versus anti-CD3-treated group.Ctrl, control.



Ctrl - CsA F1' F2' F3' F4' F5' (10 μg/mL)0.0

0.5

1.0

1.5

anti-CD3

***

***

***

***** ***


0.5

1.0

1.5

anti-CD3

***

******

** **


0.5

1.0

1.5

anti-CD3

***

*****

*

Rel

etiv

e co

nce

ntr

atio

nn

orm

aliz

ed t

o β

-act

in(I

L-2

mR

NA

)

Rel

etiv

e co

nce

ntr

atio

nn

orm

aliz

ed t

o β

-act

in(I

L-4

mR

NA

)

Rel

etiv

e co

nce

ntr

atio

nn

orm

aliz

ed t

o β

-act

in(I

FNγ

mR

NA

)

Rel

etiv

e co

nce

ntr

atio

nn

orm

aliz

ed t

o β

-act

in(I

FNγ

mR

NA

)(a)

Rel

etiv

e co

nce

ntr

atio

nn

orm

aliz

ed t

o β

-act

in(I

L-2

mR

NA

)

(d)

(b)

(c)

Rel

etiv

e co

nce

ntr

atio

nn

orm

aliz

ed t

o β

-act

in(I

L-4

mR

NA

)

(f)(e)

Ctrl - CsA 1 2 (mM)

2 (mM)2 (mM)

0.0

0.5

1.0

1.5

anti-CD3

BA

***

******

Ctrl - CsA 10.0

0.5

1.0

1.5

anti-CD3

BA

****** ***

Ctrl - CsA 10.0

0.5

1.0

1.5

anti-CD3

BA

***

***

***

Figure 6 Levels of Th1/Th2 cytokines in splenocytes when treated with the F1′–F5′ fractions and butanoic acid. Cells were cultured for 24 h in thepresence of fractions F1′–F5′ at a fixed concentration of 10 μg/ml or butanoic acid at 1–2 mM. Cells were stimulated with 1 μg/ml anti-CD3 mono-clonal antibody (mAb), and cyclosporine A (CsA) (0.2 μg/ml) was used as a positive control. Isolated mRNAs were analysed by real-time reverse tran-scription polymerase chain reaction for interleukin (IL)-2 (a), interferon (IFN)-γ (b) and IL-4 (c). IL-2 (d), IFN-γ (e) and IL-4 (f) were also quantified for abutanoic acid (BA) reference check. Results were internally confirmed by the comparative cycle count (Ct, cycle number threshold) against β-actin asa standard gene. An anti-CD3 mAb group was used as a calibrator for a relative comparison for each group. Results are mean ± standard deviation(n = 3). *P < 0.05, **P < 0.01, *** P < 0.001 versus anti-CD3-treated group. Ctrl, control.



Histological findings and Th1/Th2-mediatedcytokines from splenocytes inDNCB-treated mice

Tissues were stained with H&E after formalin fixation tomeasure skin thickness and cell infiltration into the dermis.As shown in Figure 7a, skin thickness increased in DNCB-treated mice compared with those of normal mice (Ctrl),but decreased in the positive control (0.25% prednicar-bate, DNCB + PC), and EA extract (1.25 and 2.5 mg/kg,DNCB + EA). Higher concentrations of the EA extract(2.5 mg/kg) were expected to have more potential to recoverAD-like skin compared with that of the 0.25% predni-carbate ointment therapeutic control. These findings wereconfirmed in the Th1 and Th2 cytokine mRNA analysisfrom splenocytes in DNCB-treated mice. As shown inFigure 7b and 7c, the EA extracts significantly inhibited IL-4and IL-5 mRNA expression, suggesting that may play animportant role suppressing T cell activation in the spleen,which is the largest secondary immune organ in the body,and is responsible for initiating immune reactions to blood-borne antigens. In addition, a comparison of the majororgans (liver, kidney, thymus and spleen) and body weightdemonstrated that application of study samples were notharmful (data not shown). Moreover, total amount ofextract was too small (1.25 and 2.5 mg/kg = approximately0.031 and 0.062 mg/mouse), and no critical toxicity inorgans was expected.

Inactivation of nuclear factor of activated Tcell (NFAT) in anti-CD3-treated splenocytes

As a complex T cell antigen receptor signalling cascadeleads to T cell activation and cytokine secretion via NFATdephosphorylation, we confirmed whether the most effec-tive final fraction (F2′) was correlated with inhibiting NFATactivation using the EMSA supershift assay. Figure 8suggests that the F2′ fraction successfully inhibited depho-sphorylation of cytosolic NFAT (NFATc), and therebyregulated the transcriptional programmes leading to T cellactivation when NFATc is translocated into the nucleus,which allows it to bind to specific DNA sequences.[11]

Interestingly, these results suggest that the F2′ fraction mayhave strong potential to inhibit T cell activation as observedwith CsA.

Discussion

AD is an inflammatory, non-contagious and pruritic skindisorder that takes a chronic relapsing course. AD is dueto a skin hypersensitivity reaction and inflammation.AD represents an epidermal barrier dysfunction, andimmunoallergic events contribute to the particular inflam-matory microenvironment of the skin. Immune system

Ctrl

DNCB+EA (1.25 mg/kg)DNCB

DNCB+PC DNCB+EA (2.5 mg/kg)

Ctrl - PC 1.25 2.5 (mg/kg)0

1

2

3

**

******

DNCB-tretedEA extract

Ctrl - PC 1.25 2.5 (mg/kg)0

2

4

6

8

DNCB-treated

***

******

EA extract

Rel

etiv

e co

nce

ntr

atio

nn

orm

aliz

ed t

o β

-act

in(I

L-4

mR

NA

)

Rel

etiv

e co

nce

ntr

atio

nn

orm

aliz

ed t

o β

-act

in(I

L-5

mR

NA

)

(a)

(b)

(c)

Figure 7 Histological recovery of skin lesions 5 weeks after treatmentand the effect on Th2 cytokine mRNA expression in splenocytes of 2,4-dinitrochlorobenzene (DNCB)-treated mice. (a) Tissues were fixed with10% neutral buffered formalin, embedded in paraffin and stained withthe H&E (Ctrl, untreated; DNCB-treated; PC (0.25% prednicarbate,therapeutic control); EA, EA extract-treated). (b,c) Each compound wasevenly applied to the skin lesions for 5 weeks and real-time reversetranscription polymerase chain reaction analysis on the splenocytes wasperformed to assess the IL-4 and IL-5 expression. Experiments wereperformed in triplicate and the results are expressed as the mean ± S.D.Th2 cytokines were compared in triplicate and the results are expressedas the mean ± SD (n = 3). **P < 0.01, *** P < 0.001 versus DNCB-treated group. Ctrl, control.



dysregulation of the two major CD4+ T cell subsets isan important aspect of AD pathophysiology.[12] Accord-ingly, AD is characterized by increased inflammatorycell infiltration into the skin and Th2-dominated immun-ity that leads to elevated levels of serum IgE and peri-pheral eosinophilia.[13] Moreover, AD inflammation isassociated with an increased number of Th2 cells in acuteskin lesions, overexpressing Th2-type cytokines (e.g. IL-4,IL-5 and IL-13).[14] However, in chronic phase of AD, Thcell patterns are switched into a Th1-type cytokine associ-ated with IFN-γ expression, followed by infiltration ofinflammatory dendritic epidermal cells, macrophages andeosinophils that produce IL-12.[15] Much research overthe past decades has focused on AD. These studiesindicate that the basic cause of this disease is allergicinflammation accompanied by a variety of immunologicalabnormalities.[16] Thus, the most promising anti-AD drugsmay be compounds that are immune-suppressive. Thesecompounds should be able to control the imbalance ofTh1/Th2 cells, to be safe for humans and to preventrelapses when used for long-term treatments.[17]

In this study, we found that H. fusiformis containedvarious volatile components that exhibit varying antioxi-dant and anti-inflammatory activity.[18,19] In the GC-MSanalysis, BA was consistently found in the EA extract andthe F1, F1′ and F2′ fractions where Th1 and Th2-typecytokines such as IL-2, IFN-γ and IL-4 were significantlyinhibited. BA is a major metabolite in the colonic lumenarising from bacterial fermentation of dietary fibre and actsas a physiological regulator of homeostasis in colonic epi-thelial cells by balancing proliferation, differentiation andapoptosis.[20,21] Furthermore, BA enhances apoptosis of Tcells in the colonic tissue and thereby eliminates the source

of inflammation (i.e. IFN-γ).[22] Bailón et al.[23] reported thatBA suppresses IFN-γ, IL-4, IL-5 and IL-10 expression inconcanavalin A-activated T lymphocytes. These results ledus to further investigate the anti-AD properties ofH. fusiformis. Our results demonstrate that the EA extractand the F1 fraction significantly suppressed expression ofthe Th1 and Th2-type cytokines IL-2, IFN-γ and IL-4.However, only the F1′ and F2′ fractions from the F1 fractionsignificantly inhibited those cytokines, whereas F3′, F4′ andF5′ were ineffective. The GC-MS analysis showed that theEA extract and the F1, F1′ and F2′ fractions coincidentlycontained BA. Interestingly, more BA was identified in theF2′ fraction (0.14%) than that in the F1′ fraction (0.08%).This suggests that the more BA presents, leads to moreeffective anti-atopic activity. These in-vitro results werecoincidently shown in DNCB-treated AD-like mice. H&Estaining showed that the 2.5 mg/kg EA extract-treatedmice recovered as much as the positive control (0.25%prednicarbate), and that the Th2-type cytokines IL-4and IL-5 were also significantly suppressed even at lowerdoses (1.25 mg/kg). Notably, F2′ clearly inhibited NFATcdephosphorylation in the EMSA gel shift assay. Thus, webelieve that the H. fusiformis fractions appreciably blockedsynthesis of Th1/Th2 cytokines by inhibiting depho-sphorylation of NFATc, which can induce the over-expression of AD cytokine genes in the nucleus, thereby ini-tiating differentiation of Th0 cells to Th2 cells in AD skinlesions. Regardless of the finding that BA might be themajor component inhibiting NFATc dephosphorylation,more attention should be paid to the EA extract and theF1 and F2′ fractions, which significantly suppressed theTh1/Th2-type cytokines at much lower concentration(<25 μg/ml) compared with that of a previously reportedstudy that was effective at the mM level (molecular weight:88.1).[23]

Conclusions

We first determined that H. fusiformis contained BA asa major compound to inhibit the Th type-1 and -2cytokines. The consistent results of inhibiting the T cell-dependent activation between the in-vitro and in-vivostudies strongly suggests that the H. fusiformis fractioncontaining a high amount of BA may have anti-AD activ-ity by eliminating the source of inflammation (i.e. IFN-γproduction) from T cells and thereby effectively suppress-ing the processes triggering NFAT activation. In conclu-sion, our results demonstrate that the F2′ fraction fromH. fusiformis may have ideal properties as an effectiveanti-AD candidate. The inhibited dephosphorylation ofNFATc, which is a major pathway of calcineurin inhibitors,also provides more evidence that the F2′ fraction mightdirectly contribute to T cell inactivation.

Ctrl - CsA F2'ShiftedNF-ATcProbe

anti-CD3 - + + +

Figure 8 Electrophoretic-mobility shift assay (EMSA) for NFATc.Splenocytes were pre-treated with the F2′ fraction and cyclosporine A(CsA) for 0.5 h and then activated with anti-CD3 monoclonal antibody(mAb) for 1 h, followed by preparation of nuclear extracts for EMSA.All experiments were repeated three times; data from a representativeexperiment are shown. Ctrl, control.



Declaration

Funding

This study was supported by a grant of the KoreaHealthcare Technology R&D Project, Ministry for Health,Welfare and Family Affairs, Republic of Korea (A091121).

References

1. Shan BE et al. Immunomodulat-ing activity of seaweed extract onhuman lymphocytes in vitro. Int JImmunopharmacol 1999; 21: 59–70.

2. Yang EJ et al. Inhibitory effect of Jejuendemic seaweeds on the productionof pro-inflammatory mediators inmouse macrophage cell line RAW264.7. J Zhejiang Univ Sci B 2010; 11:315–322.

3. Siriwardhana N et al. Antioxidantactivity of Hizikia fusiformis on reac-tive oxygen species scavenging andlipid peroxidation inhibition. Food SciTechnol Int 2003; 9: 339–346.

4. Kang C-H et al. Ethyl alcohol extractof Hizikia fusiforme induces caspase-dependent apoptosis in human leuke-mia U937 cells by generation ofreactive oxygen species. Trop J PharmRes 2011; 10: 739–746.

5. Dinarello CA. Anti-inflammatoryagents: present and future. Cell 2010;140: 935–950.

6. Leung DY, Bieber T. Atopic dermati-tis. Lancet 2003; 361: 151–160.

7. Mastrandrea F. The potential roleof allergen-specific sublingual immu-notherapy in atopic dermatitis. AmJ Clin Dermatol 2004; 5: 281–294.

8. Akhavan A, Rudikoff D. Atopic der-matitis: systemic immunosuppressivetherapy. Semin Cutan Med Surg 2008;27: 151–155.

9. Lee K-S et al. A novel model forhuman atopic dermatitis: applicationof repeated DNCB patch in BALB/cmice, in comparison with NC/Ngamice. Lab Anim Res 2010; 26: 95–102.

10. Mishiro T et al. Butyric acid attenu-ates intestinal inflammation in murineDSS-induced colitis model via milk fatglobule-EGF factor 8. Lab Invest 2013;93: 834–843.

11. Zhou B et al. Regulation of the murineNFATc1 gene by NFATc2. J Biol Chem2002; 277: 10704–10711.

12. Levy ML. Atopic dermatitis: under-standing the disease and its manage-ment. Curr Med Res Opin 2007; 23:3091–3103.

13. Leung DY. Atopic dermatitis: the skinas a window into the pathogenesis ofchronic allergic disease. J Allergy ClinImmunol 1995; 96: 302–318.

14. Leung DYM et al. New insight intoatopic dermatitis. J Clin Invest 2004;113: 651–657.

15. Leung DYM. Atopic dermatitis: newinsights and opportunities for thera-peutic intervention. J Allergy ClinImmunol 2000; 105: 860–876.

16. Hussain I, Kline JN. DNA, theimmune system, and atopic disease.

J Investig Dermatol Symp Proc 2004; 9:23–28.

17. Wollenberg A et al. Proactive treat-ment of atopic dermatitis in adultswith 0.1% tacrolimus ointment.Allergy 2008; 63: 742–750.

18. Jang H-W, Ka M-H. Antioxidantactivity and characterization of vola-tile extracts of Capsicum annuum L.,and Allium spp. Flavour Fragrance J2008; 23: 178–184.

19. Podlogar JA, Verspoh EJ. Anti-inflammatory effects of ginger andsome of its components in humanbronchial epithelial (BEAS-2B) cells.Phytother Res 2002; 26: 333–336.

20. von Engelhardt W et al. Role of short-chain fatty acids in the hind gut. VetQuart 1998; 20: S52–S59.

21. Augenlicht LH et al. Short chain fattyacids and colon cancer. J Nutr 2002;132: 3804S–3808S.

22. Zimmerman MA et al. Butyrate sup-presses colonic inflammation throughHDAC1-dependent Fas upregulationand Fas-mediated apoptosis of T cells.Am J Physiol Gastrointest Liver Physiol2012; 302: 1405–1415.

23. Bailón E et al. Butyrate in vitroimmune-modulatory effects mightbe mediated through a proliferation-related induction of apoptosis.Immunobiology 2010; 215: 863–

873.



Documents

Hizikia fusiformis fractions successfully improve atopic dermatitis indices in anti-CD3-stimulated splenocytes and 2,4-dinitrochlorobenzene-treated BALB/c mice