oral supplementation of lithospermum erythrorhizon prevents the development of atopic dermatitis...

Copyright © 2009 John Wiley & Sons, Ltd. Phytother. Res. 23, 1250–1256 (2009)DOI: 10.1002/ptr

1250 J. M. KIM ET AL.

Copyright © 2009 John Wiley & Sons, Ltd.

PHYTOTHERAPY RESEARCHPhytother. Res. 23, 1250–1256 (2009)Published online 9 March 2009 in Wiley InterScience(www.interscience.wiley.com) DOI: 10.1002/ptr.2573

Oral Supplementation of Lithospermumerythrorhizon Prevents the Development ofAtopic Dermatitis with Reducing CeramideDegradation in the Epidermis of NC/Nga Mice

JungMin Kim1, YoungRan Kim1, DaeBang Seo2, SungHan Kim3, SangJun Lee2 and Yunhi Cho1*1Department of Medical Nutrition, Graduate School of East-West Medical Science, Kyung Hee University, Yongin-si, Gyeonggi-do,Korea2Amorepacific Corporation R & D Center, Yongin-si, Gyeonggi-do, Korea3Nutrex Co., Ltd, Seoul, Korea

Lithospermum erythrorhizon Sieb. et Zucc. (LE) is widely used in the treatment of abnormal skin conditions,but its systemic efficacy, especially in atopic dermatitis (AD), is not clear. To examine the systemic efficacy ofLE on the clinical manifestation of AD-like skin lesions, NC/Nga mice, a murine model of AD, were fed acontrol diet (group CA: atopic control) or a diet with a 70% ethanol extract from 5% LE (group LE) for 10weeks. In group LE, the clinical manifestation of AD-like skin lesions was prevented as the level of serum IgE,epidermal hyperproliferation, and the number and duration of scratching episodes, which were greater ingroup CA, were significantly reduced to a similar level of the normal control group of BALB/c mice (groupC). In addition, the level of ceramides, the major lipid maintaining the epidermal barrier, in the epidermis ofgroup LE was increased, and was inversely associated with a decreased protein level of ceramidase, an enzymeof ceramide degradation. However, the mRNA and the protein levels of serine palmitoyl transferase (enzymefor de novo ceramide synthesis) in groups C, CA and LE did not differ. It was demonstrated that oralsupplementation with LE extract prevented the development of atopic dermatitis with reducing ceramidedegradation coupled with a low expression of ceramidase protein. Copyright © 2009 John Wiley & Sons, Ltd.

Keywords: Lithospermum erythrorhizon; atopic dermatitis; ceramides; ceramidase; NC/Nga mice.

Received 10 October 2007Revised 13 March 2008

Accepted 20 March 2008

* Correspondence to: Dr Y. Cho, Department of Medical Nutrition, Gradu-ate School of East-West Medical Science, Kyung Hee University, #1Seocheon-dong, Giheung-gu, Yongin-si, Gyeonggi-do, 446-701, Korea.E-mail: [email protected]/grant sponsor: Ministry of Health and Welfare of the Republicof Korea, Health 21 R & D project; contract/grant number: 0405-FS00-0501-0014.

INTRODUCTION

Atopic dermatitis (AD) is a chronic inflammatory skindisease characterized by pruritic and eczematous skinlesions (Leung, 2000). Although genetic, pharmacolo-gical and psychological factors play an important role,an aberrant immune response and impaired epidermalbarrier function are actively involved in the pathogenesisof AD (Leung, 2000). Immunological studies have docu-mented an elevated immunoglobulin E (IgE) in ADfor which the interleukin cytokines (IL)-4, IL-5 and IL-13, produced by T-helper (Th) 2 cells, are responsible(Van der Heijden et al., 1991; Van Reijsen et al., 1992).In addition, most patients with AD have impaired in-tegrity of the epidermal barrier (Werner, 1986), whichis composed of intercellular lipid lamellae with ceramides,cholesterol and fatty acids in the stratum corneum, theoutermost layer of the epidermis (Elias and Menon,1991). An impaired barrier function leads to the clinicalmanifestation of dryness and scaliness with epidermalhyperproliferation in the epidermis of AD (Imokawaet al., 1991). Moreover, recent cutaneous reports have

demonstrated that ceramides, the major lipid maintain-ing the epidermal barrier, are decreased significantly,and the expression or activity of ceramide metabolizingenzymes are altered in the epidermis of AD (Murataet al., 1996; Aioi et al., 2001).

Therapy of AD includes the topical application ofsteroids, the avoidance of skin irritating factors andthe systemic administration of antihistamines and anti-microbials, but evidence of the side effects of usingsteroids and certain antihistamines is accumulatingsteadily (Arellano et al., 2007; Furue et al., 2004). Inthis regard, treatment alternatives using plant anddietary sources are of particular interest (Koo and Arain,1998). Lithospermum erythrorhizon Sieb. et Zucc. (LE),a plant species of the Boraginaceae family native toeast Asia, has been reported to be effective in the clini-cal treatment of abnormal skin conditions such as burnsand inflammation (Bown, 1995). Also topical applica-tion of the LE extract significantly accelerated thehealing of wounds in the skin of diabetic mice (Fujitaet al., 2003), but most of these reported activities in LEextract have been based on in vitro studies, and onlylimited information is available on the systemic efficacyof LE, especially in AD.

This study investigated the systemic efficacy of theLE extract on the clinical manifestation of AD-like skinlesions in NC/Nga mice, a recently recognized murinemodel of AD (Aioi et al., 2001; Matsuda et al., 1997),and delineated a possible mechanism for this efficacybased on an altered level of ceramides. In addition to

AD PREVENTION BY LITHOSPERMUM ERYTHRORHIZON 1251


determining the serum level of IgE, scratching behaviorand epidermal hyperproliferation, the epidermal levelof ceramides, as well as the mRNA and protein expres-sions of ceramide metabolizing enzymes, were evalu-ated in the epidermis of NC/Nga mice fed LE extractfor 10 weeks.

MATERIALS AND METHODS

Preparation of LE extract. The dried root of LE wascollected from Fushun, Liaoning province in China(October 2005). A voucher specimen (05-10-009) wasdeposited at the herbarium of Nutrex Co., Ltd (Seoul,Korea). LE was prepared as a dried powder from thediluted ethanol extract, and provided by Nutrex Co.,Ltd. Specifically, LE (100 g) was cut into small piecesand extracted with 500 mL of 70% ethanol at 85–90 °Cfor 12 h with stirring. The extraction procedure wasrepeated twice. The combined solvent was filtered througha Whatman no. 2 paper, and lyophilized to yield 34.9 gof LE extract. The standardized major bioactive com-ponent, lithospermic acid, was identified at a concentra-tion of 2.0 mg/g in the LE extract (documents wereprovided by Amorepacific Corporation R & D Center(Gyeonggi-do, Korea)).

Mice and diets. Six week old male BALB/c mice werepurchased from SLC Japan (Shizuoka, Japan). Six weekold male NC/Nga mice were also purchased from SLCJapan (Shizuoka, Japan) and assigned into two groups.One group was fed a diet supplemented with the driedpowder of a 70% ethanol extract from 50 g/kg LE(group LE). The other group, which served as the atopycontrol, was fed a control diet (group CA). The com-positions of the experimental diets are shown in Table 1,and mice were fed each of the respective diets for 10weeks to evaluate the systemic efficacy of LE extracton the clinical manifestation of AD-like skin lesions.Group C, the normal control group which consisted ofBALB/c mice, was fed a control diet for 10 weeks.

During the entire 10 week feeding period, all micewere maintained under conventional laboratory con-ditions without air filtration to raise atopic dermatitisspontaneously as described previously (Aioi et al., 2001).The mice were housed under conditions of controlledtemperature (22–24 °C), humidity (55–60%) and light(lights on from 07:00 to 19:00). Food intake and bodyweights of all groups were monitored weekly over the10 week feeding period, in which there were no signifi-cant differences in any of the groups. Animal care andhandling conformed to the accepted NIH guidelines(NIH publication No. 80-23, 1978). At the end of week10, all mice were killed by cervical dislocation, andepidermal strips were removed after overnight incuba-tion of whole skin in an ice cold 1:1 mixture of dispaseII (2.4 U/mL, Roche, Indianapolis, IN) and RPMIsupplemented with 10% fetal bovine serum at 4 °C(Maciejewski et al., 2006).

Measurement of serum IgE. Blood was collected fromthe retro-orbital plexus prior to killing and was allowedto clot for 30 min at room temperature. The serum wasprepared by centrifugation (at 800 × g for 10 min at4 °C) and stored at −20 °C until use. The serum level

Table 1. Composition of experimental diets (g/kg)

Groupa

Ingredient C CA LE

Caseinb 230 230 230Corn starch 372 372 354.54L-Cystine 3 3 3Corn oil 100 100 100Cellulose powder 50 50 50Vitamin mixturec 10 10 10Salt mixtured 35 35 35Sucrose 200 200 200Extract of Lithospermum 0 0 17.46erythrorhizon (LE)

a Group C, BALB/c mice fed control diet; Group CA and LE, NC/Nga mice fed control diet (group CA) or LE diet (group LE).b Casein (nitrogen × 6.25), 870 g/kg.c Vitamin mix composition, AIN – 93 vitamin mix 310025 (DyetsInc, Bethlehem, PA, USA): niacin 3 g/kg, calcium pantothenate1.6 g/kg, pyrridoxine HCl 0.06 g/kg, thiamine HCl 0.6 g/kg, ribo-flavin 0.6 g/kg, folic acid 0.2 g/kg, biotin 0.2 g/kg, vitamin E ac-etate (500 IU/g) 15 g/kg, vitamin B12 (0.1%) 2.5 g/kg, vitamin Apalmitate (500 000 IU/g) 0.8 g/kg, vitamin D3 (400 000 IU/g) 0.25 g/kg, vitamin K1/dextrose mix (10 mg/g) 7.5 g/kg and sucrose967.23 g/kg.d Salt mix composition : AIN – 93 G salt mix 210025 (Dyets Inc,Bethlehem, PA, USA): calcium carbonate 357 g/kg, potassiumphosphate (monobasic) 196 g/kg, potassium citrate H2O 70.78 g/kg, sodium chloride 74 g/kg, potassium sulfate 46.6 g/kg, mag-nesium oxide 24 g/kg, ferric citrate U.S.P 6.06 g/kg, zinc carbon-ate 1.65 g/kg, manganous carbonate 0.63 g/kg, cupric carbonate0.3 g/kg, potassium iodate 0.01 g/kg, sodium selenate 0.01025 g/kg, ammonium paramolybdate 4H2O 0.00795 g/kg, sodiummetasilicate 9H2O 1.45 g/kg, chromium potassium sulfate 12H2O0.275 g/kg, lithium chloride 0.0714 g/kg, boric acid 0.0815 g/kg,sodium fluoride 0.0635 g/kg, nickel carbonate 0.0318 g/kg, am-monium vanadate 0.066 g/kg, and finely powdered sucrose221.026 g/kg.

of immunoglobulin E (IgE) was determined using acommercially available enzyme-linked immunosorbentassay (ELISA) kit (Biosource International, Comarillo,CA, USA) by following the manufacturer’s instructions(Engvall and Perlman, 1971).

Evaluation of epidermal proliferation. Evaluation of theepidermal proliferation was performed biochemicallyby a [3H]thymidine incorporation assay (Halprin et al.,1979). 4 mm2 epidermal biopsies were placed in 1×DMEM (Dulbecco’s modified Eagle medium) in which0.1 µCi [3H]thymidine had been dissolved. After incu-bation for 3 h at 37 °C, the epidermis was frozen rapidlyin liquid N2 to stop DNA synthesis. Once the epidermisthawed, it was placed in 0.5 M NaOH, and heated at95 °C for 30 min to dissolve the tissue and release DNA.Aliquots were placed on cellulose filters (Millipore,Bedford, MA) previously treated with 1% trichloro-acetic acid. Filters with absorbed DNA were dried and[3H]thymidine incorporated into DNA was measuredby scintillation counting. Aliquots of the solution, afterbeing heated at 95 °C, were dissolved in ethidium bro-mide (EtBr) buffer (10 µg/mL in 0.1 M Tris-HCl, 0.1 M

NaCl, pH 7.4). The quantitative fluorescence of theDNA-EtBr reaction was used to measure endogenousDNA levels in the samples using a fluorometer (Modulus



Fluorometer, Turner Biosystems, Sunnyvale, CA). Thespecific [3H]thymidine incorporation was calculated asthe amount of cpm[3H]thymidine/µg DNA.

Evaluation of scratching behavior. Scratching behaviorwas observed on week 10 as described previously(Yamaguchi et al., 2001). Briefly, the mice were placedindividually into a clear plastic cage for 60 min ofhabituation. Their behavior was then recorded usingan unmanned digital video camera (Sony TRV, Japan)for 30 min. The videotapes were reviewed and thenumber of scratching episodes was counted. A series ofscratching movements made only with the hind paw(Yamaguchi et al., 2001) was counted as one scratchingepisode. The duration of each scratching episode wasassessed macroscopically using the following scoringprocedure (0, no scratching; 2, scratching for <1.5 s; 4,scratching for ≥1.5 s) (Norikazu et al., 2003). The totalscore of the scratching duration was defined as the sumof all the individual scores accumulated during 30 minof video taping.

Ceramide quantitation. Epidermal strips were homo-genized with a polytron as reported previously (Kimet al., 2006). A portion of the skin homogenate was usedto measure the protein concentration by a modifiedLowry method (Lowry et al., 1951) and the rest wasextracted with chloroform (CHCl3): methanol (MeOH)(2:1, v/v) to obtain total lipids (Folch et al., 1957). Theextracted lipids were fractionated into ceramides by highperformance thin layer chromatography (HPTLC) on0.20 mm silica gel 60-coated plates (Merck, Darmstadt,Germany) as described previously (Uchida et al., 2000).Specifically, the samples applied to the plates with aLinomat 5 autosampler (CAMAG, Muttenz, Switzer-land) were first developed to 1.0 cm and then to 3.5 cmin CHCl3: MeOH: acetone (76:20:4, v/v/v). They werethen developed to 7.5 cm in CHCl3: acetone: MeOH(80:10:10, v/v/v) and finally developed to the top inCHCl3: ethylacetate: ether: MeOH (76:20:6:2, v/v/v/v).Each stage of development was carried out by an auto-matic multiple development (AMD) apparatus (CAMAG,Muttenz, Switzerland) after the plates were completelyair-dried. The fraction containing ceramides that hadco-migrated with the respective standards were scannedat 420 nm by a TLC III scanner (CAMAG, Muttenz,Switzerland). The level of ceramides was quantitatedin each sample by calibration curves with the variousconcentrations of external standards of ceramides, andexpressed as µg ceramides/µg protein.

Western blotting analysis of SPT and ceramidase proteins.Antibodies against serine palmitoyl transferase (SPT),a ceramide generating enzyme in the de novo synthesispathway, and ceramidase, the ceramide degradativeenzyme, were both customized by Labfrontier Co.,Ltd (Suwon, Republic of Korea). Specifically, rabbitpolyclonal antibodies against SPT and ceramidase weregenerated using the following peptides as antigens:DRPFDETTYEETED which is the coding sequencefor amino acids 547–560 of SPT (Kawamoto et al., 1995),and RTEYKKIRDDDLRH-C which is the cystine addedcoding sequence for amino acids 167–180 of ceramidase(Marra et al., 1999). Rabbits were immunized with500 µg of each antigen in Freund’s complete adjuvant,followed by a second injection in Freund’s incomplete

adjuvant 14 days later (Nagasaki et al., 2001). Total serumIgG was isolated by protein A purification and anti-bodies specific for either SPT or ceramidase peptides wereaffinity-purified (Nagasaki et al., 2001). The specificityof each antibody was confirmed by ELISA analysis(Nagasaki et al., 2001) and the antibodies were used at1:50–1:200 dilutions.

Protein extract (15 µg/each) from the epidermis wereseparated electrophoretically on SDS-polyacrylamidegels (SDS-PAGE, 10%) and then blotted to nitrocellu-lose membranes (Kim et al., 2005). Proteins were detectedwith horseradish peroxidase coupled with anti-IgG directedagainst either SPT (67 KDa) or ceramidase (45 KDa)using the ECL detection system (Amersham, UK). Theband intensity was quantified by densitometry.

RT-PCR analysis of SPT and ceramidase mRNAs. TotalRNA was extracted from the epidermis using TRizol®

reagent (Gibco BRL, Grand Island, NY), followed byisopropanol precipitation. Isolated RNA was quanti-fied spectrophotometrically based on the formula 1UA260nm = 40 µg/mL RNA, and then electrophoresed on1.2% agarose gels to assess the quality of RNA.

Reverse transcription of the isolated total RNA (1 µg)was performed for 1 h at 37 °C using 0.25 units of AMVreverse transcriptase (Seoulin Bioscience Inc. Republicof Korea) in 10 mM Tris-HCl buffer (pH 8.3) contain-ing 50 mM KCl, 5 mM MgCl2, 0.5 unit RNase inhibitor,and 1 mM dNTP mixture (final volume: 20 µL). The re-action was terminated by heating at 94 °C for 5 min,followed by cooling at 4 °C for 5 min. The resultingcDNA was added to 20 µL of PCR reaction mixture[10 mM Tris-HCl buffer (pH 8.3), 50 mM KCl, 2.5 mM

MgCl2, 2 µM each of 5′ and 3′ primers, and 2.5 units ofTaq polymerase (Seoulin Bioscience Inc. Republic ofKorea)]. The primer sequences were: mouse serinepalmitoyl transferase (SPT), 5′-GACTGAGCCTTTC-TGGTGCT-3′ (forward) and 5′-AATGCCATTTGGA-GTGAAGG-3′ (reverse) (UniSTS D19271) (Kawamotoet al., 1995); mouse ceramidase, 5′-TGTGGAGCAGA-AAATCAAGC-3′ (forward) and 5′-CAACATCATG-GGGCACTTAG-3′ (reverse) (UniSTS AI662009) (Marraet al., 1999) and mouse glyceraldehyde 3′-phosphatedehydrogenase (GAPDH), 5′-CCATGGAGAAGGCT-GGGG-3′ (forward) and 5′-CAAAGTTGTCATGGA-TGACC-3′ (reverse). Amplications of SPT and GAPDHwere both initiated by 1 min of denaturation at 94 °C,followed by 30 cycles at 65 °C for 60 s and 72 °C for60 s. Amplication of ceramidase was initiated by 1 minof denaturation at 94 °C, followed by 32 cycles at 65 °Cfor 1.5 min and 72 °C for 60 s.

The PCR products (SPT: 159 bp, ceramidase: 103 bp,GAPDH: 200 bp) were analysed by 3.0% agarose gelelectrophoresis with ethidium bromide. For relativequantitation, the densitometry value of each band forSPT and ceramidase was normalized to that of the house-keeping gene, GAPDH, which was used as an internalstandard.

Statistical analysis. Data are expressed as mean ± SEM(n = 10). All data were analysed with one-way ANOVAusing SAS statistical procedures (SAS 6.03, SAS Insti-tute, Cary, NC) and the differences among the groups(C, CA and LE) were determined by Tukey’s multiplecomparison test. Differences with p < 0.05 were consideredsignificant.



Table 2. Clinical manifestation of atopic dermatitis in NC/Nga mice

Group

AD related parameter C CA LE

Serum IgE (IU/mL) 18.9 ± 2.61b 42.2 ± 4.02a 17.1 ± 2.69b

Hyperproliferation (cpm [3H]thymidine/µgDNA) 59.9 ± 5.17b 161.1 ± 7.79a 50.6 ± 7.74b

Scratching Total number of scratching episodes for 30 min 8.9 ± 1.18ab 10.8 ± 0.86a 7.5 ± 0.51b

Total score of scratching duration for 30 min 55.2 ± 4.10b 84.8 ± 5.46a 54.9 ± 4.30b

Values are mean ± SEM (n = 10). Means with different letters in the same rows differ p < 0.05.Group C, BALB/c mice fed control diet; Group CA and LE, NC/Nga mice fed control diet (group CA) or LE diet (group LE).Scratching movements only by the hind paw for 30 min were counted as scratching episodes.Total score of scratching duration: the sum of the individual scores for 30 min ranging from 0 to 4 (0, no scratching; 2, scratchingfor <1.5 s; 4, scratching for ≥1.5 s).

Figure 1. Ceramide levels in the epidermis of BALB/c mice fedcontrol diet only (group C) and NC/Nga mice fed either controldiet (group CA), or LE diet (group LE) for 10 weeks. Values aremean ± SEM (n = 10). Mean with different letters differ p < 0.05.

RESULTS

Clinical manifestation of atopic dermatitis

Elevation of serum IgE and epidermal hyperprolifera-tion have been demonstrated as clinical manifestationsof AD-like skin lesions in NC/Nga mice (Aioi et al.,2001; Matsuda et al., 1997). When the serum level ofIgE and [3H] thymidine incorporation into epidermiswere determined at 10 weeks (Table 2), the evidenceof AD was found in group CA. The IgE level of groupCA was two fold higher than that of group C, and theepidermis of group CA incorporated more [3H]thymidinethan the control epidermis of group C, suggesting thatthe epidermis of group CA was hyper-proliferatingwith increased DNA synthesis. The serum level of IgEand DNA synthesis in group LE were less than in groupCA, but similar to those in group C.

In addition to the elevation of serum IgE and epider-mal hyperproliferation, the NC/Nga mice with AD hadspontaneous scratching episodes on their face, ears andthe rostal part of their back using their hind paws(Yamaguchi et al., 2001), and had a longer duration ofscratching, especially more than 1.5 s (Norikazu et al.,2003). Analysis of the scratching behavior demonstratedthat in group LE, the total number of scratching epi-sodes and the total score of the scratching duration,which were greater in group CA, were significantly re-duced to a level similar to group C (Table 2). Theseresults indicate that oral supplementation with LEextract prevented the clinical manifestation of AD-likeskin lesion in NC/Nga mice.

Ceramide quantitation

Ceramides play a crucial role in maintaining the lamel-lar integrity of the epidermal barrier (Imokawa et al.,1991). Depletion of ceramides has been suggestedto be an etiological factor for epidermal barrier dis-ruption, ultimately inducing the clinical manifestationof dryness and scaliness with epidermal hyperpro-liferation in the epidermis of AD (Imokawa et al.,1991). The quantification of ceramides revealed thatpreventing the clinical manifestation of AD-like skinlesions (Table 2) was associated with increasing thelevel of ceramides in the epidermis (Fig. 1). In the

hyperproliferative epidermis of group CA, the levelof ceramides was significantly less than that of groupC. In contrast, the level of ceramides in the epidermisof group LE was similar to that of group C as theepidermal proliferation in groups LE and C werecomparable.

Expression of SPT and ceramidase

A balance between the activity and expression of ceramidemetabolizing enzymes regulates the level of ceramidesin the epidermis (Imokawa, 2001). To further delineatethe mechanism for the altered level of ceramideswith AD induction and oral supplementation of LE,the protein and mRNA expressions of serine palmitoyltransferase (SPT), the rate-limiting enzyme in the de novosynthesis of ceramides, and ceramidase, the degradativeenzyme (Imokawa, 2001), were determined (Figs 2 and3). Protein levels of SPT in groups C, CA and LE didnot significantly differ (Fig. 2), but the protein level ofceramidase in the epidermis of group CA was greaterthan that of group C (888.9% of group C) (Fig. 2), sug-gesting that a high expression of ceramidase protein,the degradative enzyme, is the metabolic feature lead-ing to decreased levels of ceramides in the hyperpro-liferative epidermis of group CA (Fig. 1, Table 2). Afterthe LE extract was supplemented for 10 weeks in groupLE, the protein level of ceramidase expression was



Figure 2. Altered protein levels of SPT and ceramidase in theepidermis of BALB/c mice fed control diet only (group C) andNC/Nga mice fed either control diet (group CA), or LE diet (groupLE) for 10 weeks. (A) Western blotting analysis of SPT andceramidase proteins in the epidermis of mice. (B) The signalintensities from multiple experiments of A were quantified andthe integrated areas were normalized, first to the correspond-ing value of actin and then to the signal observed in the normalcontrol group (group C). All values are mean ± SEM (n = 10).Mean with different letters differ p < 0.05.

Figure 3. Altered mRNA levels of SPT and ceramidase in theepidermis of BALB/c mice fed control diet only (group C) andNC/Nga mice fed either control diet (group CA), or LE diet (groupLE) for 10 weeks. (A) RT-PCR analysis of mRNA for SPT andceramidase in epidermis of mice. (B) The signal intensities frommultiple experiments of A were quantified and the integratedareas were normalized, first to the corresponding value of theGAPDH internal control and then to the signal observed inthe normal control group (group C). All values are mean ± SEM(n = 10). Mean with different letters differ p < 0.05.

DISCUSSION

This study investigated the systemic efficacy of LEextract on the clinical manifestation of AD-like skinlesions in NC/Nga mice. In preliminary dose-findingexperiments with the LE extract, no systemic efficacywas found at the 1% level, while 5% and 10% of LEextract (50 g or 100 g LE in kg diet) improved the clini-cal manifestation of AD-like skin lesions in NC/Ngamice: hence, the 5% level was used in the present study.When oral supplementation with 70% ethanol extractfrom 5% LE (group LE) was parallel to the inductionof AD for 10 weeks, the serum IgE levels and thymidineincorporation into the epidermis of group LE wereless than that of group CA but were similar to thosein group C (Table 2). Furthermore, the total numberof scratching episodes and the total score of scratching

significantly less than that of group CA and similar tothat of group C, indicating that the accumulation ofceramides in the epidermis of group LE is caused by adecrease of ceramide degradation coupled with a lowexpression of ceramidase protein.

There was no alteration in the mRNA levels of SPTin groups C, CA and LE (Fig. 3) as similar to the SPTprotein levels (Fig. 2). Furthermore, the reduced levelsof ceramidase protein with LE supplementation resultedfrom changes in the mRNA levels. Again, the mRNAlevel of ceramidase expression in group LE was signifi-cantly less than that in group CA. Curiously, the mRNAlevel of ceramidase in group CA was similar to thatin group C, despite the high induction of ceramidaseprotein. Whether this indicates that the translation ofthis mRNA is induced or that the stability of this pro-tein is increased by the rise in AD requires furtherinvestigation.



duration, which were both greater in group CA, weresignificantly reduced to a similar level to group C, dem-onstrating a novel observation: oral supplementationwith LE is capable of preventing the development ofAD-like skin lesions in NC/Nga mice.

Although ceramides play a crucial role in maintainingthe lamellar integrity of the epidermal barrier (Imokawaet al., 1991), little is known about the systemic modula-tion of ceramide levels related to preventing the devel-opment of AD-like skin lesions. The data in Fig. 1 andTable 2 indicate that the level of ceramides in theepidermis is inversely correlated with the degree of clini-cal manifestation of AD in NC/Nga mice. This decreasedlevel of ceramides in group CA agrees with previousstudies in either AD patients or AD induced NC/Ngamice, which demonstrated that the ceramide level wassignificantly reduced in the epidermis of AD (Imokawaet al., 1991; Murata et al., 1996; Aioi et al., 2001). Incontrast, the level of ceramides in the epidermis of groupLE was less than that of group CA but similar to thatof group C as itchiness and scaliness with epidermalhyperproliferation were less than that of group CA,suggesting that preventing the development of AD-likeskin lesions with oral supplementation of LE extractcan be attributed to the increased level of ceramides.

The mRNA and protein expressions of serine palmitoyltransferase (SPT) in the de novo ceramide synthesispathway were not altered by either AD induction or byoral supplementation of LE (Figs 2 and 3). However,the protein level of ceramidase, the degradative enzymeof ceramide was increased in group CA, suggesting thata high protein expression of ceramidase is a metabolicfeature leading to decreased levels of ceramides in theepidermis of group CA. The protein level of ceramidasein the epidermis of group LE was significantly lessthan that of group CA and comparable to that of groupC, indicating that the accumulation of ceramides inthe epidermis of group LE is caused by a decrease ofceramide degradation coupled with a low expressionof the ceramidase protein. In AD induced NC/Nga mice,the decreased level of ceramides is reported to be par-allel to the increased activity of ceramidase or decreasedactivity of sphingomyelinase (SMase), a ceramide gen-erating enzyme from sphingomyelin (Aioi et al., 2001).In addition, unusual expression of sphingomyelin deacylase,which metabolizes sphingomyelin to sphingosyl phos-phorylcholine, could also underlie the decreased levelof ceramides in the lesional epidermis of patients withAD (Murata et al., 1996). Although modulation of theexpression or activity of ceramide metabolizing enzymessuch as sphingomyelinase with oral supplementationof 5% LE requires further elucidation, the results inTable 2 and Figs 1–3 clearly demonstrate that the

improved clinical manifestation of group LE is causedby a decrease of ceramide degradation coupled with alow expression of ceramidase protein.

The root of Lithospermum erythrorhizon Sieb. et Zucc.(Boraginaceae) (LE) has been used traditionally as anherbal medicine for the treatment of various abnormalskin conditions (Bown, 1995; Fujita et al., 2003; Kimet al., 2006). As for the bioactive components, naptha-quinones, such as shikonin and its derivatives, wereextensively characterized mostly in the organic extractof LE, which were reported to exert antiproliferative,antiinflammatory and antimicrobial activities mostly fromin vitro studies (Hayashi, 1977; Lu and Liao, 1990; Sasakiet al., 2002). However, since the 70% ethanol extractof LE was prepared by decocting at 85–90 °C for 12 hin this study, the well-characterized bioactive com-ponents such as napthaquinones as well as other organiccomponents may have been degraded, and thereforelithospermic acid, a caffeic acid oligomer (Anja et al.,2006), may have been identified as a standardized majorbioactive component at a concentration of 2 mg/g inthe LE extract. Lithospermic acid has been identifiedat a level of 4.735% in the LE extract (Toshi et al., 1997),and on the other hand, glycans such as lithosperman A,B and C have also been identified, which are homo-genous polysaccharides with neutral rhamnose, fucose,arabinose and galactose (Konno et al., 1985). Concerningthe dermatological activities of lithospermic acid andglycans, oral administration of lithospermic acid hasbeen found to inhibit prolyl hydroxylation in collagensynthesis, offering a potential means for treating fibroticdiseases such as systemic scleroderma and keloids(Shigematsu et al., 1994). The profile of neutral glycansin itchy skin has been reported to be significantlyaltered when compared with normal skin (Hirokoet al., 2007), but little is known about the effects oflithospermic acid or lithosperman A, B and C on alter-ing the cera-mide level or reducing ceramide degrada-tion in the epidermis.

Although further studies are required to investigatethe systemic modulation of LE on ceramide metaboliz-ing enzymes in depth and to delineate the mechanismfor the altered level of ceramides, this in vivo studydemonstrated a novel observation that oral supplemen-tation with 5% LE extract prevented the developmentof atopic dermatitis by reducing ceramide degradationcoupled with a low expression of ceramidase protein.

Acknowledgements

This study was supported by a grant of the Korean Health 21 R & Dproject, the Ministry of Health and Welfare of the Republic of Korea(no. 0405-FS00-0501-0014).

REFERENCES

Aioi A, Tonogaito H, Suto H et al. 2001. Impairment of skinbarrier function in NC/Nga Tnd mice as a possible modelfor atopic dermatitis. Br J Dermatol 144: 12–18.

Anja W, Steven JOM, Robert GB et al. 2006. Reassignment ofthe configuration of salvianolic acid B and establishmentof its identity with lithospermic acid B. J Nat Prod 69: 1231–1233.

Arellano FM, Wentworth CE, Arana A et al. 2007. Risk oflymphoma following exposure to calcineurin inhibitors and

topical steroids in patients with atopic dermatitis. J InvestDermatol 127: 808–816.

Bown D. 1995. Encyclopedia of Herbs and Their Uses. DorlingKindersley: London.

Elias PM, Menon GK. 1991. Structural and lipid biochemical correlatesof the epidermal permeability barrier. Adv Lipid Res 24: 1–26.

Engvall E, Perlman P. 1971. Enzyme-linked immunosorbentassay (ELISA). Quantitative assay of immunoglobulin G.Immunochemistry 8: 871– 874.



Folch J, Lee M, Sloane-Stanley GH. 1957. A simple method forthe isolation and purification of total lipids from animaltissues. J Biol Chem 226: 497–509.

Fujita N, Sakaguchi I, Kobayashi H et al. 2003. An extract ofthe root of Lithospermun erythrorhizon accelerates woundhealing in diabetic mice. Biol Pharm Bull 26: 329–335.

Furue M, Terao H, Moroi Y et al. 2004. Dosage and adverseeffects of topical tacrolimus and steroids in daily manage-ment of atopic dermatitis. J Dermatol 31: 277–283.

Halprin KM, Taylor JR, Levine V et al. 1979. A combined alkaliextraction-ethidium bromide technique for the measurementof DNA in small pieces of tissue. J Invest Dermatol 73: 359–363.

Hayashi M. 1977. Pharmacological studies of shikon and tooki (3).Effect of topical application of the ether extracts and shiunkoon inflammatory reactions. Nippon Yakurigaku Zasshi (inJapanese) 63: 205–214.

Hiroko I, Masashi A, Hiroaki N et al. 2007. N-linked neutraloligosaccharides in the stratum corneum of normal andichthyotic skin. Arch Dermatol Res 298: 403–407.

Imokawa G. 2001. Lipid abnormalities in atopic dermatitis.J Am Acad Dermatol 45: S29–S32.

Imokawa G, Abe A, Jin K et al. 1991. Decreased levels of ceramidesin stratum corneum of atopic dermatitis: an etiological factorin atopic dry skin. J Invest Dermatol 96: 523–526.

Kawamoto S, Okubo K, Yoshii J et al. 1995. Analysis ofgene expression in mouse embryogenesis by 3′-directedcDNA sequencing. http://www.ncbi.nlm.nih.gov/entrez/viewer.fcgi?view =gb&val =D19271.

Kim J, Kim H, Jeong DH et al. 2006. Comparative effect ofgromwell (Lithospermum erythrorhizon) extract and borageoil on reversing epidermal hyperproliferation in guinea pigs.Biosci Biotechnol Biochem 70: 2086–2095.

Kim J, Um SJ, Woo J et al. 2005. Comparative effect of seedsof Rhynchosia volubilis and soybean on MG-63 humanosteoblastic cell proliferation and estrogenicity. Life Sci 78:30–40.

Konno C, Mizuno T, Hikino H. 1985. Isolation and hypoglycemicactivity of lithospermans A, B, and C, glycans of Lithospermumerythrorhizon roots. Planta Med 2: 157–158.

Koo J, Arain S. 1998. Traditional Chinese medicine for thetreatment of dermatologic disorders. Arch Dermatol 134:1388–1393.

Leung DYM. 2000. Atopic dermatitis: new insights and opportu-nities for therapeutic intervention. J Allergy Clin Immunol105: 860–876.

Lowry OH, Rosebrough NJ, Farr AL et al. 1951. Protein meas-urement with the folin phenol reagent. J Biol Chem 193:265–275.

Lu G, Liao J. 1990. Detection of the anti-cancer biological effectof naphthaquinone pigment-LIII. Zhong Xi Yi Jie He Za Zhi(in Chinese) 10: 422–425.

Maciejewski LD, Richman JG, Hakak Y et al. 2006. Langerhanscells release prostaglandin D2 in response to nicotinic acid.J Invest Dermatol 126: 2637–2646.

Marra M, Hillier L, Kucaba T et al. 1999. The Wash U-NCIMouse EST Project. http://www.ncbi.nlm.nih.gov/entrez/viewer.fcgi?view=gb&val=A1662009.

Matsuda H, Watanabe N, Geba GP et al. 1997. Development ofatopic dermatitis like skin lesion with IgE hyperproductionin NC/Nga mice. Int Immunol 9: 461–466.

Murata Y, Ogata J, Higaki Y et al. 1996. Abnormal expression ofsphingomyelin acylase in atopic dermatitis: an etiologicfactor for ceramide deficiency? J Invest Dermatol 106: 1242–1249.

Nagasaki T, Yasuda S, Imai T. 2001. Preparation of antibodyagainst agatharesinol, a norlignan, using a hapten-carrierconjugate. Phytochemistry 58: 833–840.

Norikazu T, Iwao A, Michio K. 2003. Analysis of the spontaneousscratching behavior by NC/Nga mice: a possible approachto evaluate antipruritics for subjects with atopic dermatitis.Eur J Pharmacol 471: 223–228.

Sasaki K, Abe H, Yoshizaki F. 2002. In vitro antifungal activity ofnaphthoquinone derivatives. Biol Pharm Bull 25: 669–670.

Shigematsu T, Tajima S, Nishikawa T et al. 1994. Inhibition ofcollagen hydroxylation by lithospermic acid magnesium salt,a novel compound isolated from Salviae miltiorrhizaeRadix. Biochim Biophy Acta 1200: 79–83.

Toshi O, Yoshie M, Koichiro S et al. 1997. Caffeic acid derivativesin lamiaceae and boraginaceae plants. Jpn J Food Chem 4:11–15.

Uchida Y, Hara M, Nishio H et al. 2000. Epidermal sphingo-myelins are precursors for selected stratum corneumceramides. J Lipid Res 41: 2071–2082.

Van der Heijden FL, Wierenga EA, Bos JD et al. 1991. High numberof IL-4-producing CD4+ allergen-specific T lymphocytes inatopic dermatitis lesional skin. J Invest Dermatol 97: 389–394.

Van Reijsen FC, Bruijnzeel-Koomen CA, Kalthoff FS et al. 1992.Skin derived aeroallergen specific T-cell clones of Th2phenotype in patients with atopic dermatitis. J Allergy ClinImmunol 90: 184–193.

Werner Y. 1986. The water content of the stratum corneum inpatients with atopic dermatitis. Acta Derm Venerol (Stockh)65: 102–105.

Yamaguchi T, Maekawa T, Nishikawa Y et al. 2001. Characterizationof itch-associated responses of NC mice with mite-inducedchronic dermatitis. J Dermatol Sci 25: 20–28.

http://www.ncbi.nlm.nih.gov/entrez/viewer.fcgi?view =gb&val =D19271

http://www.ncbi.nlm.nih.gov/entrez/viewer.fcgi?view=gb&val=A1662009

oral supplementation of lithospermum erythrorhizon prevents the development of atopic dermatitis...

Documents