Journal of Dermatological Science 68 (2012) 56–62

Orally administered sphingomyelin in bovine milk is incorporated into skinsphingolipids and is involved in the water-holding capacity of hairless mice

Yuko Haruta-Ono a, Shuichi Setoguchi b, Hiroshi M. Ueno a, Satoshi Higurashi a, Noriko Ueda a,Ken Kato a,*, Tadao Saito c, Kazuhisa Matsunaga b, Jiro Takata b

a Milk Science Research Institute, Megmilk Snow Brand Co., Ltd, 1-1-2 Minamidai, Kawagoe, Saitama 350-1165, Japanb Laboratory of Drug Design and Drug Delivery, Faculty of Pharmaceutical Sciences, Fukuoka University, 8-19-1 Nanakuma, Jonan-ku, Fukuoka 814-0180, Japanc Laboratory of Animal Products Chemistry, Graduate School of Agricultural Science, Tohoku University, 1-1 Tsutsumidori-Amamiyamachi, Aoba-ku, Sendai, Miyagi 981-8555, Japan

A R T I C L E I N F O

Article history:

Received 6 April 2012

Received in revised form 15 June 2012

Accepted 14 July 2012

Keywords:

Sphingomyelin

Glucosylceramide

Ceramide

Bovine milk

Skin

Distribution

A B S T R A C T

Background: We previously reported that dietary sphingomyelin (SM) concentrate from bovine milk

improves epidermal functions. SM is a known precursor of ceramide (Cer) in the stratum corneum (SC).

Neither the uptake nor distribution of orally administered SM nor its effects on epidermal functions have

been demonstrated.

Objective: We evaluated the effects of dietary SM on epidermal functions, and the distribution and fate of

its radiolabeled metabolites in mice orally administered [4,5-3H-sphinganyl] sphingomyelin (3H-SM).

Methods: Bovine milk SM (98% purity) was administered orally to 13-week-old hairless mice at 142 mg/

kg per day for eight weeks. Their SC hydration, transepidermal water loss (TEWL), and SC Cer content

were measured. 3H-SM was then administered orally to 10-week-old hairless mice. Its distribution and

metabolites in the skin were evaluated with whole-body autoradiography, liquid scintillation counting,

and thin-layer chromatography.

Results: SC hydration in the SM-administered mice was higher than that in control mice, whereas their

TEWL and Cer contents did not differ. Radioactivity was distributed extensively in the bodies of the

experimental mice and decreased gradually with time. In contrast, the radioactivity in the SC remained

constant after its administration, and radiolabeled SM and Cer were detected in the skin. This suggests

that dietary SM is transferred to the skin and then converted to Cer in the SC.

Conclusions: Orally administered SM is incorporated into skin SM and converted to SC Cer, which is

involved in the water-holding capacity of the SC.

� 2012 Japanese Society for Investigative Dermatology. Published by Elsevier Ireland Ltd. All rights

reserved.

Contents lists available at SciVerse ScienceDirect

Journal of Dermatological Science

jou r nal h o mep ag e: w ww .e lsev ier . co m / jds

1. Introduction

The epidermis has a considerable water-holding capacity and apermeability barrier function, which protect the organism fromdehydration, infection by pathogens, and environmental stresses,such as chemicals and physical damage. The stratum corneum (SC),the outermost layer of the epidermis, is composed of corneocytesand an intercellular lipid matrix. Ceramide (Cer), which comprisesapproximately 50% of these intercellular lipids, plays an importantrole in retaining epidermal water and in the epidermal permeabil-ity barrier function, in combination with cholesterol and free fattyacids [1–3]. The Cer in the SC consists of a heterogeneous family ofat least seven molecular groups (Cer1–7), which vary in their long-chain sphingoid base structures, their chain lengths, and the

* Corresponding author. Tel.: +81 49 242 8165; fax: +81 49 242 8157.

E-mail address: k-kato@meg-snow.com (K. Kato).

0923-1811/$36.00 � 2012 Japanese Society for Investigative Dermatology. Published b

http://dx.doi.org/10.1016/j.jdermsci.2012.07.006

a-hydroxylation of their constituent amide-linked fatty acids [4–6]. SC Cer1–7 are generated from glucosylceramide (GlcCer),whereas Cer2 and Cer5 are also generated from sphingomyelin(SM). Thus, SM and GlcCer are important precursors of SC Cer [7,8],and are synthesized by keratinocytes and stored in epidermallamellar bodies. At the transition from the stratum granulosum tothe SC, the lamellar bodies fuse with the plasma membrane of theuppermost granular cells and extrude their contents into theintercellular spaces of the SC. Hydrolytic enzymes, such as b-glucocerebrosidase and acid sphingomyelinase, then convert thesecreted GlcCer and SM, respectively, into SC Cer [9,10].

Recently, several studies have reported that dietary constituentsplay a beneficial role in epidermal functions [11–14]. In our previousstudy, we also demonstrated that the oral administration of an SMconcentrate prepared from bovine milk to hairless mice increasedthe hydration of their SC and the SC Cer content and reduced theirtransepidermal water loss (TEWL) [15,16]. Moreover, the dailyintake of SM concentrate improves the water-holding capacity of the

y Elsevier Ireland Ltd. All rights reserved.

Table 2Compositions of experimental diets.

Componenta Content (%)

Control SM-supplemented SM-concentrate

supplemented

Casein 20.0 20.0 17.6

Corn oil 5.0 5.0 –

Olive oil 5.0 5.0 –

Safflower oil – – 4.2

SM – 0.1 –

SM concentrate – – 9.8

DL-Methionine 0.3 0.3 0.3

Mineral mixture 3.5 3.5 3.5

Vitamin mixture 1.0 1.0 1.0

Cellulose 5.0 5.0 5.0

Cornstarch 15.0 15.0 15.0

Sucrose 45.2 45.1 43.6

SM – 0.1 0.7

Linoleic acid 3.1 3.1 3.1

Linolenic acid 0.1 0.1 0.1

a The experimental diets were modified from AIN-76 (American Institute of

Nutrition, 1977), supplemented with olive oil or safflower oil.

Y. Haruta-Ono et al. / Journal of Dermatological Science 68 (2012) 56–62 57

human epidermis [17]. These results indicate that the dietary SMconcentrate improves the water-holding capacity and permeabilitybarrier function of the epidermis. However, the SM concentratecontains other components, so it is unclear whether these effects onepidermal functions are attributable to SM, or whether orallyadministered SM is distributed to the skin and converted to Cer.

In this study, we examined the effects of dietary SM onepidermal functions and its distribution and the distribution of itsmetabolites in the skin of hairless mice orally administered 3H-radiolabeled SM.

2. Materials and methods

2.1. Chemicals

SM from bovine milk (>98% purity) was purchased from NofCorporation (Tokyo, Japan). Radiolabeled [4,5-3H-sphinganyl]sphingomyelin (3H-SM; Fig. 1) was synthesized from SM byBlychem Ltd (Billingham, UK). The radiolabeled compound had aspecific activity of 17 Ci/mmol and a radiochemical purity of >99%according to thin-layer chromatography (TLC). The SM concentrate(Milk Ceramide MC-5; Megmilk Snow Brand Co., Ltd, Tokyo, Japan)was prepared from bovine milk, according to a previous report[15]. The SM concentrate contained 6.9% SM and other compo-nents, including proteins, phospholipids, triglycerides, and carbo-hydrates (Table 1). Cer standards were obtained from Larodan FineChemicals (Malmo, Sweden), and contained non-OH Cer and a-OHCer from bovine brain. Other high-grade reagents were purchasedfrom Wako Pure Chemical Industries (Osaka, Japan).

2.2. Animals

Male hairless HR-1 mice, aged 10–14 weeks and weighing 22–32 g, were purchased from Japan SLC (Hamamatsu, Japan) [16,17].The mice were housed in individual plastic cages in a temperature-and humidity-controlled room (23 8C and 50 � 5% relative humidi-ty), on a 12 h light/dark cycle. The mice were given free access to foodand distilled water. All procedures for animal care and use compliedwith the regulations established by the NIH Guide for the Care andUse of Laboratory Animals and the Experimental Animal Care and UseCommittee of Fukuoka University.

2.3. Effects of orally administered SM on epidermal functions

Twelve-week-old male HR-1 mice were used in this study. Aftera one-week adaptation period, the mice were separated into three

Table 1Composition of SM concentrate.

Component Content (%)

Moisture 1.2

Protein 22.9

Fat 57.2

Ash 6.7

Carbohydrate 12.0

SM 6.9

Fig. 1. Structure of 3H-SM. Radioactive labeling was at the C4 and C5 positions of the

sphingoid moiety.

experimental groups: the control group (n = 10), the SM-supple-mented group (n = 7), and the SM-concentrate-supplementedgroup (as the positive control; n = 10). Each mouse had a similarmean body weight. All mice were given free access to one of thethree experimental diets and distilled water for eight weeks. Theexperimental diets were modified from the AIN-76 diet (Table 2).The amounts of linoleic acid (3.1%) and linolenic acid (0.1%) wereadjusted to ensure the same concentrations in the three diets withthe addition of safflower oil and/or olive oil, because they areknown to affect the permeability barrier function of the epidermis[18,19]. In the SM-concentrate-supplemented group (positivecontrol), the lipids in the diet were partly replaced by the SMconcentrate. The SM contents in the SM-supplemented and SM-concentrate-supplemented diets were 0.1% (w/v) and 0.7% (w/v),respectively. The mouse body weights were recorded once a weekand their food intake was monitored daily. The water-holdingcapacity and permeability barrier function of the epidermis wereevaluated by assessing the hydration of the SC and TEWL using aCorneometer CM825 and Tewameter TM300 (Courage andKhazaka Electronics, Cologne, Germany), respectively, at fourand eight weeks. After the eight-week feeding period, the micewere fasted for 18 h and sacrificed with diethyl ether. The wholeskin was stripped from each mouse with scissors, and the SC sheetwas isolated by incubating the skin in 5% (w/v) trypsin solution inphosphate-buffered saline (pH 7.3) for 1 h at 37 8C. The SC sheetswere dried overnight in a vacuum oven and weighed, and thenstored at �80 8C until further analysis. The Cer content of the SCwas quantified by high-performance liquid chromatography(HPLC), as described in Section 2.8.

2.4. Whole-body autoradiography

3H-SM (18.5 MBq) was diluted in 0.2 mL of 15% (w/v) SMconcentrate and administered orally to 10-week-old male HR-1mice that had been fasted for 18 h. The mice were housed singly inmetabolic cages (Metabolica, Sugiyama-gen Co., Tokyo, Japan)after the administration of 3H-SM. The mice were sacrificed 24 hafter the administration of 3H-SM, rapidly frozen at �80 8C, andembedded in carboxymethylcellulose gel. Serial sections (30 mmthick) through the sagittal plane of each mouse were made withthe tape-sectioning method using a Cryo Polycut cryostat (Reich-ert-Jung, Nussloch, Germany) at �20 8C. The sections on theadhesive tape (Yu-Ki Ban, Nitto Medical Co., Ltd, Osaka, Japan)

Y. Haruta-Ono et al. / Journal of Dermatological Science 68 (2012) 56–6258

were desiccated and placed on a TR2040 imaging plate (FujifilmCo., Ltd, Tokyo, Japan) for 10 days. Whole-body macroautoradiographs were processed photographically with an FLA7000image analyzer (Fujifilm Co., Ltd).

2.5. Quantification of radioactivity in tissues

3H-SM (3.7 MBq/head) and an SM concentrate mixture wereadministered orally to 14-week old male HR-1 mice that had beenfasted for 18 h. The mice were housed in individual wire-bottomed cages, with free access to food and water, after theadministration of the compounds. The mice were then anesthe-tized at 24, 72, or 144 h and sacrificed by the collection of bloodfrom the heart. Their organs or tissues were collected andweighed. The dorsal skin was stripped and the subcutaneoustissues were removed on ice. The skin samples were incubated in0.5% (w/v) trypsin for 2 h at 37 8C and then the SC was separatedfrom the dorsal skin. Some organs, tissues, and blood wereprepared with a sample oxidizer (model 307; Perkin-Elmer, MA,USA), and their radioactivity (tritium) was counted with a liquidscintillation counter (Packard TR2300; Perkin-Elmer). The per-centages of radioactivity in the organs or tissues relative tothe amount of radioactivity ingested were determined(means � SD, n = 3).

Fig. 2. Effects of dietary SM on the skin of hairless mice. Mice were fed with one of three e

SM-concentrate-supplemented diet (n = 10). The SC hydration (A) and TEWL (B) were m

and eight weeks. After the eight-week feeding period, the skin of the mice was stripped

*Significantly different from the control group (P < 0.05).

2.6. Skin metabolites after administration of 3H-SM

We examined the radiolabeled metabolites in the whole skin.3H-SM (18.5 MBq) in 0.2 mL of 15% (w/v) SM concentrate wasadministered orally to 14-week-old male HR-1 mice and the micewere housed in metabolic cages. The mice were anesthetized 24 hafter the administration of the concentrate and sacrificed by thecollection of blood from the heart. The dorsal skin was stripped andthe subcutaneous tissues were removed on ice. The skin samplewas minced with scissors and homogenized in an equal weight ofwater in a Micro Smash MS-100 (Tomy Seiko Co., Ltd, Tokyo,Japan). The lipid samples extracted from the homogenates weredissolved in 0.2 mL of chloroform:methanol (1:1, v/v) and stored at�80 8C until TLC analysis.

We then examined the distribution of the radiolabeledcompounds in each skin layer of the mice after 3H-SM administra-tion. 3H-SM (18.5 MBq) was diluted in 0.2 mL of 15% (w/v) SMconcentrate and administered orally to 10-week-old male HR-1mice that had been fasted for 18 h. The mice were anesthetized72 h after the administration of the concentrate and sacrificed bythe collection of blood from the heart. The dorsal skin was strippedand the subcutaneous tissues were removed on ice. The skinsamples were incubated in 0.5% (w/v) trypsin for 1 h at 37 8C, andthe SC was then separated from the dorsal skin. The SC, the

xperimental diets: the control diet (n = 10), the SM-supplemented diet (n = 7), or the

easured using a Corneometer CM825 and a Tewameter TM300, respectively, at four

and SC Cer was quantified by HPLC (C). Data points are means, and bars show SD.

Y. Haruta-Ono et al. / Journal of Dermatological Science 68 (2012) 56–62 59

remaining epidermis and the dermis (epidermis + dermis), and thesubcutaneous tissue were minced separately with scissors andhomogenized in an equal weight of water in a Micro Smash MS-100. The total lipid fraction and Cer fraction were extracted fromeach as described in Section 2.7.

2.7. Lipid extraction and isolation of Cer

The total lipid fraction was extracted with a modification of theRose Gottlieb method [20]. To analyze the Cer samples, theextracted lipids were dissolved in 2 mL of hexane and fractionatedon a Mega Bond Elut silica column (Varian, Palo Alto, CA, USA)before HPLC and TLC analyses. Most of the cholesteryl sulfates,triacylglycerols, and free fatty acids were eluted with 6 mL ofhexane:diethyl ether (1:1, v/v). Cer was eluted with 6 mL ofchloroform:methanol (2:1, v/v) and then with 6 mL of chloro-form:methanol (1:2, v/v) with phospholipids. These Cer-contain-ing fractions were mixed and dried under nitrogen gas. The Cerfractions were dissolved in 2 mL of chloroform:methanol (1:1, v/v)and stored at �80 8C until HPLC and TLC analyses.

2.8. Determination of lipids

Cer contents were analyzed with the Waters 2690 Alliance HPLCsystem (Waters, Milford, MA, USA) equipped with an evaporativelight-scattering PL–ELS 2100 detector (Polymer Laboratories,Amherst, MA, USA) and a Pegasil Silica 60 column(250 mm � 4.6 mm i.d., 5 mm; Sensyu Kagaku, Tokyo, Japan). Theanalysis was performed according to Gildenast and Lasch [21]. The

Fig. 3. Whole-body autoradiographs of a hairless mouse 24 h after the administration of

SM (22.7 mCi/kg body weight). All sections are from the same animal and represent later

to high uptake of radioactivity.

radiolabeled lipid and Cer samples were separated by TLC (Silica Gel60 A, 20 cm � 10 cm plates; Merck, Darmstadt, Germany). The lipidsamples were separated twice with chloroform:methanol:water(12:6:1, v/v) to 3 cm, and then with hexane:diethyl ether:acetic acid(16:4:3, v/v) to the top of the plates. The Cer samples were developedtwice with chloroform:methanol:acetic acid (190:9:1, v/v) [22].After the radioactive metabolites were detected with an FLA7000image analyzer, the TLC plates was visualized by spraying them with10% (w/v) CuSO4 and 8% (w/v) H3PO4 in 5% (v/v) methanol andheating them to 180 8C. Cer was identified by comparing itsretardation factor (Rf) values with those of the reference standardsdescribed in Section 2.1. The Rf values of non-OH Cer and a-OH Cerwere 0.40 and 0.14, respectively.

2.9. Statistical analysis

All data are given as means � SD. They were determined withone-way ANOVA and Tukey–Kramer’s post hoc test, using StatViewver. 5 software (SAS Institute Inc., Cary, NC, USA). Differences wereconsidered significant at P < 0.05.

3. Results

3.1. Effects of dietary SM on SC hydration, TEWL, and Cer content of

the SC

The final body weights and amounts of food taken did not differsignificantly among the experimental groups (data not shown).The daily intake of SM was 5 mg (142 mg/kg body weight) in the

3H-SM. Autoradiography was performed at 24 h after the oral administration of 3H-

al sagittal (top figure) to mid-sagittal views (bottom figure). Black areas correspond

Y. Haruta-Ono et al. / Journal of Dermatological Science 68 (2012) 56–6260

SM-supplemented group and 32 mg (929 mg/kg body weight;13 g/kg body weight of SM concentrate) in the SM-concentrate-supplemented group. We used the SM concentrate as the positivecontrol because the dietary SM concentrate is known to improvethe epidermal functions of hairless mice and humans [15–17].

Fig. 2A shows the time-dependent changes in SC hydration inthe mice of the SM-supplemented, SM-concentrate-supplemented,and control groups. The SC hydration of the control and SMconcentrate-supplemented groups decreased during the eight-week feeding period, and the SC hydration of the SM concentrate-supplemented group was significantly higher than that of thecontrol group at four and eight weeks (P < 0.05). After the eight-week feeding period, the SC hydration of the SM-supplementedgroup was similar to the initial value, and was also significantlyhigher than that of the control group at eight weeks. In contrast,the TEWL of the SM-supplemented group was not significantlydifferent from that of the control group in all experimental periods(Fig. 2B). The Cer content was significantly higher (P < 0.05) in themice fed the SM concentrate-supplemented diet than that of themice fed the control or SM-supplemented diet at eight weeks(Fig. 2C). These results indicate that dietary SM improved thewater-holding capacity in the murine epidermis.

3.2. Whole-body autoradiography

To clarify how dietary SM contributes to the maintenance of SChydration, we investigated the uptake and distribution of dietary3H-SM in hairless mice. Whole-body autoradiographs of the mice24 h after the administration of 3H-SM are shown in Fig. 3. Allsections are from the same animal and represent lateral sagittal(top figure) to mid-sagittal views (bottom figure). The black areas

Fig. 4. Skin metabolites after the administration of 3H-SM. The lipids of the murine skin w

separated with TLC (A). The chromatograph was developed, and the radioactivity was det

extract from the whole skin. SC Cer of the murine skin was extracted 72 h after the admi

radioactivity was detected with a Fuji FLA7000 phosphorimager (right panel), the chroma

(left panel). Lane 1, Cer standard; lane 2, SM standard; lane 3, administered 3H-SM; lane

Cer fraction from the subcutaneous tissue.

correspond to high uptake of radioactivity. Radioactivity wasapparent in the stomach, vertebra, liver, kidney, heart, lung,intestine, brown adipose tissue, and fecal matter. These resultsshow that radioactivity was extensively distributed in the mousebody after the oral administration of 3H-SM.

3.3. Tissue distribution of radioactivity

Whole-body autoradiography showed high levels of radioac-tivity in the liver and brown adipose tissue. The radioactivityconcentrations were measured in those tissues and in the skin at24, 72, and 144 h after the oral administration of 3H-SM (Table 3).The liver showed the highest level of radioactivity 24 h after theadministration of 3H-SM, followed by the epidermis + dermis,brown adipose tissue, blood, subcutaneous tissue, and SC. Theradioactivity in all tissues and organs, except the subcutaneoustissue and SC, declined gradually with time up to 144 h. Theradioactivity in the liver decreased dramatically from 0.706% to0.077%. The radioactivity in the SC was maintained at about0.010%, in contrast that in the epidermis + dermis, whichdecreased from 0.109% to 0.043%. These results indicate that partof radiolabeled dietary compound was distributed to the skin andwas retained in the SC for at least 144 h after its administration.

3.4. Skin metabolites

Fig. 4A shows the TLC pattern of the lipids extracted from thewhole skin 24 h after the administration of 3H-SM. The autoradio-graph had only a single band (lane 2) corresponding to 3H-SM (lane1). This result indicates that the radiolabeled compoundsoriginating from the orally administered 3H-SM were present as

ere extracted 24 h after the administration of 3H-SM (16.1 mCi/kg body weight) and

ected with a Fuji FLA7000 phosphorimager. Lane 1, administered 3H-SM; lane 2, lipid

nistration of 3H-SM (3.6 mCi/kg body weight) and separated with TLC (B). After the

tograph was visualized by spraying it with 10% CuSO4 and 8% H3PO4 in 5% methanol

4, Cer fraction from the SC; lane 5, Cer fraction from the epidermis + dermis; lane 6,

Table 3Time course of biodistribution of 3H-SM in mice.

Organ or tissue Radioactivity distribution (%)

24 h 72 h 144 h

Blood 0.049 � 0.020a 0.019 � 0.005 0.012 � 0.002

Liver 0.706 � 0.139 0.167 � 0.047 0.077 � 0.015

Brown adipose tissue 0.068 � 0.021 0.040 � 0.008 0.026 � 0.006

Subcutaneous tissue 0.027 � 0.011 0.032 � 0.003 0.036 � 0.016

Epidermis + dermis 0.109 � 0.035 0.142 � 0.030 0.043 � 0.011

SC 0.012 � 0.004 0.009 � 0.001 0.012 � 0.006

a Values are the ratio of the radioactivity in organs or tissues per the injected dose

(means � SD, n = 3).

Y. Haruta-Ono et al. / Journal of Dermatological Science 68 (2012) 56–62 61

SM in the skin. No Cer was detected because its concentration waslow. The radioactivity of incorporated SM relative to the amount ofradioactivity ingested is 0.117%. To examine whether theadministered 3H-SM was metabolized to Cer in the skin, the Cerin the SC, epidermis + dermis, and subcutaneous tissues wasextracted and analyzed with TLC 72 h after the 3H-SM wasadministered (Fig. 4B). Radiolabeled compounds were observedaround the Rf values of non-OH Cer and a-OH Cer (lane 4) andbetween them. The radiolabeled compounds in the epidermis + -dermis and in the subcutaneous tissue were detected around the Rfof non-OH Cer, and were observed at the same position as thesample administered to the mice (lane 3, lane 5, and lane 6,respectively). These results demonstrate that the radiolabeledcompounds were incorporated into SM, which was ultimatelyconverted to Cer in the SC.

4. Discussion

We have previously reported that the oral administration of anSM concentrate from bovine milk improved the water-holdingcapacity and permeability barrier function of the murine andhuman epidermis [15–17]. The SM concentrate also increased theSC Cer content of hairless mice [15,16]. However, it was unclearwhether and how orally administered SM improves theseepidermal functions. The present study was conducted toinvestigate the effects of dietary SM from bovine milk on theseepidermal functions and to clarify the distribution of orallyadministered 3H-SM and its metabolites in murine skin.

A dietary SM concentrate, used as the positive control, at a doseof 0.7% SM (i.e., 32 mg/day, 929 mg/kg body weight) improved SChydration and increased the Cer content of the SC (Fig. 2A). At eightweeks, the SC hydration in mice fed a diet supplemented with 0.1%SM (i.e., 5 mg/day. 142 mg/kg body weight) was significantlyhigher than that of mice fed the control diet (Fig. 2A), even thoughthe dietary SM content of the SM-supplemented group was lowerthan that of the SM-concentrate-supplemented group. This resultclearly shows for the first time that dietary SM improves theepidermal water-holding capacity of hairless mice.

To confirm how dietary SM contributes to the retention ofepidermal water, we investigated the distribution and fate of orallyadministered 3H-SM. Based on a whole-body autoradiographexamined 24 h after ingestion, the radioactivity was widespread inthe mouse body, and was particularly strong in the liver and brownadipose tissue (Fig. 3). The liver is an important organ in thebiosynthesis of sphingolipids, as well as other lipids [23]. Thefunction of the brown adipose tissue is to catabolize fatty acids andgenerate heat in response to food intake or cold exposure [24]. Ithas been suggested that dietary SM may affect lipid metabolism inthe liver and brown adipose tissue.

The observed radioactivity was the highest in the liver at 24 h,and then decreased markedly to 144 h. This indicates that theradiolabeled compounds were transported from the liver to otherorgans and tissues, including the skin. The radioactivity in the SC

relative to that in the blood was 0.24 at 24 h after theadministration of 3H-SM, but had increased to 1.0 at 144 h. Theradioactivity in the SC remained steady for 144 h, whereas that inthe epidermis + dermis had decreased to one half the value at 24 h.These findings indicate that the radiolabeled compounds originat-ing from the dietary SM were exported via the blood to the skin,and then accumulated in the SC.

Dietary SM is hydrolyzed to sphingosine and fatty acids. Most ofthe absorbed sphingosine is rapidly metabolized to palmitic acid[25,26], but part of it is reincorporated into more complexsphingolipids and transported to organs and tissues [26,27]. In thisstudy, radiolabeled SM was seen in the skin 24 h after theadministration of 3H-SM, and radiolabeled Cer appeared in the SC72 h after its administration (Fig. 4A and B). It is possible that thedegradation products of the dietary SM and/or their metaboliteswere transported from various organs and tissues to the skin,where they were incorporated into sphingolipids. SM is present inlarge quantities both in the precursor of SC Cer and in the outerleaflet of the plasma membrane of epidermal cells. Epidermal SM isclassified into three subfractions: sphingomyelin 1 (SM1), SM2,and SM3. SM1 and SM3 are the precursors of Cer2 and Cer5,respectively [7]. The radiolabeled compounds observed around thepositions of non-OH Cer and a-OH Cer in the TLC analysis werepresumed to be Cer1, Cer2, or Cer5 (Fig. 4B). SM2 is a plasmamembrane component in epidermal cells [7]. It is likely that thereconstructed SM is predominantly produced in the skin (Fig. 4A).In contrast, radiolabeled GlcCer, another precursor from whichCer1–7 are generated, was not detected in the skin (Fig. 4A).However, Cer species were observed between the positions of non-OH Cer and a-OH Cer on TLC, which were presumed to be Cer3 and/or Cer4, and were thought to be generated from GlcCer (Fig. 4B).These observations suggest that the reconstructed sphingolipidsthat originate from dietary SM are converted to SC Cer.

As shown Fig. 2C, the Cer content of the SC did not increase inmice orally administered SM. The radiolabeled compoundsoriginating from dietary SM were distributed in the skin in theform of SC Cer (Fig. 4B). However it is unknown whether Ceroriginated from dietary SM improves epidermal hydration in thisstudy. It seems likely that orally administered SM influences otherfactors involved in the water-holding capacity of the epidermis,such as the composition of the intracellular lamellar lipids, theformation of the cornified envelope, and the production of naturalmoisturizing factors. These factors may be related to theimprovement of epidermal hydration correlated to Cer. Therefore,dietary SM could improve the water-holding capacity of theepidermis in mice and humans, even at low doses.

Previous studies have reported that dietary SM from bovinemilk influences lipid metabolism, the inhibition of cholesterolabsorption, the apoptosis of colon cancer cells, and the maturationof intestinal cells in infants [28–30]. SM from bovine milk has alsobeen used as an important nutritional supplement. This study hasdemonstrated for the first time that dietary SM from bovine milk isbeneficial in maintaining healthy skin.

In conclusion, the oral administration of SM from bovine milkimproved the water-holding capacity of the murine epidermis, andthat radiolabeled compounds originating from dietary SM weredistributed in the skin in the form of reconstructed SM and SC Cer.Our results suggest that dietary SM is involved in maintaining skinhydration by supplying a source of SM and Cer to the skin.

References

[1] Schurer NY, Elias PM. The biochemistry and function of stratum corneumlipids. Processing of epidermal glucosylceramide is required for optimalmammalian cutaneous permeability function. Adv Lipid Res 1991;24:27–56.

[2] Elias PM, Menon GK. Structural and lipid biochemical correlates of the epi-dermal permeability barrier. Adv Lipid Res 1991;33:301–13.

Y. Haruta-Ono et al. / Journal of Dermatological Science 68 (2012) 56–6262

[3] Brod J. Characterization and physiological role of epidermal lipids. Int JDermatol 1991;30:84–90.

[4] Wertz PW, Downing DT. Ceramides of pig epidermis: structure determination.J Lipid Res 1983;24:759–65.

[5] Wertz PW, Miethke MC, Long SA, Straus JS, Downing DT. The composition ofthe ceramides from human stratum corneum and from comedones. J InvestDermatol 1985;84:410–2.

[6] Uchida Y, Iwamori M, Nagai Y. Distinct differences in lipid compositionbetween epidermis and dermis from footpad and dorsal skin of guinea pigs.Jpn J Exp Med 1988;58:153–61.

[7] Uchida Y, Hara M, Nishio H, Sidransky E, Inoue S, Otsuka F, et al. Epidermalsphingomyelins are precursors for selected stratum corneum ceramides. JLipid Res 2000;41:2071–82.

[8] Hamanaka S, Hara M, Nishio H, Otsuka F, Suzuki A, Uchida Y. Human epidermalglucosylceramides are major precursors of stratum corneum ceramides. JInvest Dermatol 2002;119:416–23.

[9] Holleran WM, Takagi Y, Menon GK, Legler G, Feingold KR, Elias PM. Processingof epidermal glucosylceramides is required for optimal mammalian cutaneouspermeability barrier function. J Clin Invest 1993;91:1656–64.

[10] Schmuth M, Man MQ, Weber F, Gao W, Feingold KR, Fritsch P, et al. Perme-ability barrier disorder in Niemann–Pick disease: sphingomyelin–ceramideprocessing required for normal barrier homeostasis. J Invest Dermatol2000;115:459–66.

[11] Tsuji K, Mitsutake S, Ishikawa J, Takagi Y, Akiyama M, Shimizu H, et al. Dietaryglucosylceramide improves skin barrier function in hairless mice. J DermatolSci 2006;44:101–7.

[12] Uchiyama T, Nakano Y, Ueda O, Mori H, Nakashima M, Noda A, et al. Oral intakeof glucosylceramide improves relatively higher level of transepidermal waterloss in mice and healthy human subjects. J Health Sci 2008;54:559–66.

[13] Ideta R, Sakuta T, Nakano Y, Uchiyama T. Orally administered glucosylcer-amide improves the skin barrier function by upregulating genes associatedwith the tight junction and cornified envelope formation. Biosci BiotechnolBiochem 2011;75:1516–23.

[14] Ishikawa J, Takada S, Takagi Y, Masukawa Y, Mizutani Y, Igarashi Y. Dietaryglucosylceramide is absorbed into the lymph and increases levels of epidermalsphingolipids. J Dermatol Sci 2009;56:216–8.

[15] Haruta Y, Ueda N, Kato K, Tsuji S, Yoshioka T. Clinical test of sphingomyelinconcentrate from bovine milk for skin. Milk Sci 2009;58:135–41.

[16] Haruta Y, Kato K, Yoshioka T. Dietary phospholipid concentrate from bovinemilk improves epidermal function in hairless mice. Biosci Biotechnol Biochem2008;72:2151–7.

[17] Haruta-Ono Y, Ueno H, Ueda N, Kato K, Yoshioka T. Investigation into thedosage of dietary sphingomyelin concentrate in relation to the improvementof epidermal function in hairless mice. Anim Sci J 2012;83:178–83.

[18] Jeffrey LM, Phillip WW, Donald CS, Donald TD. Effect of essential fatty aciddeficiency on epidermal O-acylsphingolipids and transepidermal water loss inyoung pigs. Biochim Biophys Acta 1987;921:191–7.

[19] Phillip WW, Donald CS, William A, Kathi CM, Donald TD. Essential fatty acidsand epidermal integrity. Arch Dermatol 1987;123:1381–4.

[20] International Dairy Federation General Secretariat. Determination of fat con-tent (Rose Gottlieb reference method). International IDF standard 9C,Dried Milk, Dried Whey, Dried Buttermilk and Dried Butter Serum. 1987:p.1–7.

[21] Gildenast T, Lasch J. Isolation of ceramide fractions from human stratumcorneum lipid extracts by high-performance lipid chromatography. BiochimBiophys Acta 1997;1346:69–74.

[22] Imokawa G, Abe A, Jin K, Higaki Y, Kawashima M, Hidano A. Decreased level ofceramides in stratum corneum of atopic dermatitis: an etiologic factor inatopic dry skin? J Invest Dermatol 1991;96:523–6.

[23] Stein O, Oette K, Hollander G, Dabach Y, Ben-Naim M, Stein Y. Divergent fate ofunsaturated and saturated ceramides and sphingomyelins in rat liver and cellsin culture. Biochim Biophys Acta 1989;1003:175–82.

[24] Cannon B, Nedergaard J. Brown adipose tissue: function and physiologicalsignificance. Physiol Rev 2004;84:277–359.

[25] Nilsson A. Metabolism of sphingomyelin in the intestinal tract of the rat.Biochim Biophys Acta 1968;164:575–84.

[26] Schmelz EM, Crall KJ, Larocque R, Dillehay DL, Merril Jr AH. Uptake andmetabolism of sphingolipids in isolated intestinal loops of mice. J Nutr1994;124:702–12.

[27] Ueda O, Uchiyama T, Nakashima M. Distribution and metabolism of sphingo-sine in skin after oral administration to mice. Drug Metab Pharmacokinet2010;25:456–65.

[28] Motouri M, Matsuyama H, Yamamura J, Tanaka M, Aoe S, Iwanaga T, et al. Milksphingomyelin accelerates enzymatic and morphological maturation ofintestine in artificially reared rats. J Pediatr Gastroenterol Nutr 2003;36:241–7.

[29] Eckhardt ERM, Wang D, Donvan JM, Carey MC. Dietary sphingomyelinsuppresses intestinal cholesterol absorption by decreasing thermodynamicactivity of cholesterol monomers. Gastroenterology 2002;122:948–56.

[30] Dillehay DL, Webb SK, Schmelz E, Merril Jr AH. Dietary sphingomyelin inhibits1,2-dimethyl-hydrazine-induced colon cancer in CF1 mice. J Nutr1994;124:615–20.