Nicotinamide increases biosynthesis of ceramides as well as other stratum corneum lipids to improve...

British Journal of Dermatology 2000; 143: 524±531.

Nicotinamide increases biosynthesis of ceramides as wellas other stratum corneum lipids to improve the epidermalpermeability barrier

O.TANNO, Y.OTA, N.KITAMURA, T.KATSUBE AND S.INOUE

Basic Research Laboratory, Kanebo Ltd, 5-3-28 Kotobuki-cho, Odawara-shi, Kanagawa 250-0002, Japan

Accepted for publication 3 April 2000

Summary Background Stratum corneum lipids, particularly ceramides, are important components of the

epidermal permeability barrier that are decreased in atopic dermatitis and aged skin.Objectives We investigated the effects of nicotinamide, one of the B vitamins, on biosynthesis of

sphingolipids, including ceramides and other stratum corneum lipids, in cultured normal human

keratinocytes, and on the epidermal permeability barrier in vivo.Methods The rate of sphingolipid biosynthesis was measured by the incorporation of [14C]-serine

into sphingolipids.

Results When the cells were incubated with 1±30 mmol L21 nicotinamide for 6 days, the rate ofceramide biosynthesis was increased dose-dependently by 4´1±5´5-fold on the sixth day compared

with control. Nicotinamide also increased the synthesis of glucosylceramide (7´4-fold) and

sphingomyelin (3´1-fold) in the same concentration range effective for ceramide synthesis.Furthermore, the activity of serine palmitoyltransferase (SPT), the rate-limiting enzyme in

sphingolipid synthesis, was increased in nicotinamide-treated cells. Nicotinamide increased the

levels of human LCB1 and LCB2 mRNA, both of which encode subunits of SPT. This suggested thatthe increase in SPT activity was due to an increase in SPT mRNA. Nicotinamide increased not only

ceramide synthesis but also free fatty acid (2´3-fold) and cholesterol synthesis (1´5-fold). Topical

application of nicotinamide increased ceramide and free fatty acid levels in the stratum corneum,and decreased transepidermal water loss in dry skin.

Conclusions Nicotinamide improved the permeability barrier by stimulating de novo synthesis of

ceramides, with upregulation of SPT and other intercellular lipids.

Key words: ceramide synthesis, epidermal permeability barrier, nicotinamide, serine palmitoyl-

transferase, stratum corneum lipids

Nicotinamide is one of the B vitamins, deficiency of

which results in a clinical syndrome called pellagra

that is characterized by dermatitis, diarrhoea anddementia.1 Nicotinamide is necessary for the pro-

duction of nicotinamide adenine dinucleotide (NAD)

and NAD phosphate (NADP), which act as coenzymesin over 40 biochemical reactions. Although the

mechanism responsible for pellagra is obscure, defi-

ciency of NAD or NADP caused by nicotinamidedeficit is thought to result in pellagra. The three

symptoms appear frequently in pellagra, and it is rare

to find a proven case without dermatitis. However, therole of nicotinamide in the epidermis and why

nicotinamide deficit results in dermatitis have not

been determined.

Stratum corneum lipids, rich in sphingolipids, freefatty acids and cholesterol, are required for the epi-

dermal permeability barrier.2 In particular, sphingo-

lipids such as ceramides play a crucial part in epidermalbarrier homeostasis.3,4 The level of ceramides in the

stratum corneum is decreased in atopic dermatitis5 and

aged skin.6 Moreover, a recent study showed thatwinter xerosis is related to compositional changes in

stratum corneum lipids and marked reductions in

ceramide levels.7

Although nicotinamide has been applied in the

treatment of cutaneous lesions of other forms of

dermatitis because the symptoms are similar to those

524 q 2000 British Association of Dermatologists

Correspondence: Shintaro Inoue.

E-mail: [email protected]

NICOTINAMIDE INCREASES CERAMIDE BIOSYNTHESIS 525

q 2000 British Association of Dermatologists, British Journal of Dermatology, 143, 524±531

of a deficiency in this B vitamin, the relationship

between nicotinamide and stratum corneum lipids hasnot been well characterized. In this study, to investigate

the role of nicotinamide in the epidermis, especially

the relationship between nicotinamide and stratumcorneum lipids, we examined the effects of nicotin-

amide on biosynthesis of sphingolipids in cultured

human keratinocytes and its mechanism. We alsoexamined the effects of topical application of nicotin-

amide on the dry skin of volunteers.

Materials and methods

Cell culture

Neonatal human foreskin keratinocytes (Kurabo,Osaka, Japan) were seeded on to 60-mm or 90-mm

plastic dishes coated with collagen and were grown to

subconfluency in MCDB153 (0´1 mmol L21 Ca21

Wako Pure Chemical Industries, Kyoto, Japan) supple-

mented with insulin (5 mg L21; Wako), hydrocortisone

(180 mg L21; Merck, Rahway, NJ, U.S.A.), 2-amino-ethanol (6´1 mg L21; Nacalai Tesque, Inc., Kyoto,

Japan), O-phosphorylethanolamine (14´1 mg L21;

Tokyo Kasei, Tokyo, Japan), epidermal growth factor(100 ng L21; Sigma, St Louis, MO, U.S.A.) and bovine

pituitary extract (BPE; 0´4% v/v; Kurabo). On the 4th

day after seeding on to 60-mm dishes (1 � 105 cells perdish), the medium was changed to MCDB153 containing

the same supplements except BPE, and 1±30 mmol L21

nicotinamide (Kanto Chemical Co. Inc., Tokyo, Japan)was added. Cells were maintained at 37 8C (5% CO2)

until harvesting for lipid and enzyme analysis.

For Northern blotting analysis, cells were seeded onto 90-mm dishes (4 � 105 cells per dish) and cultured

for 2, 4 or 6 days with 10 mmol L21 nicotinamide.

Cells were then harvested for total RNA extraction.

Ceramide synthesis

Ceramides were determined after fractionation with an

Iatrobead column (Iatron Laboratories Inc., Tokyo,

Japan). Cells were cultured in 60-mm dishes withnicotinamide or nicotinic acid for 6 days and incubated

for a further 2 days in medium containing [14C]-serine

(0´5 mCi per dish; American Radiolabeled Chemicals,St Louis, MO, U.S.A.). At the end of the incubation period,

cells were harvested by trypsinization and cell numbers

were counted. After centrifugation, the cells wereresuspended in 0´8 mL of HEPES buffer (50 mmol L21

2-[4-(2-hydroxyethyl)-1-piperazyl] ethanesulphonic acid

(HEPES), 10 mmol L21 glucose, 3 mmol L21 KCl,

100 mmol L21 NaCl, 0´75 mmol L21 Na2HPO4´2H2O,pH 7´4). Lipids were extracted using Bligh-Dyer sol-

vent8 and dried with a centrifugal concentrator (Tomy

Seiko Co., Tokyo, Japan). Extracted lipids were dissolvedin 1 mL of benzene and applied to an Iatrobead

column, with 0´1 mL bed volume of Iatrobeads (6RS-

8060) in a 1-mL glass syringe. After washing with1 mL of benzene/ethyl acetate (4 : 1), the ceramide

fraction was eluted with 1 mL of ethyl acetate. The

radioactivity was counted with a liquid scintillationspectrometer (Aloka, Tokyo, Japan).

Free fatty acid, cholesterol and sphingolipid synthesis

Free fatty acids, cholesterol and sphingolipids including

ceramides were determined as follows. After incubationfor 6 h with [14C]-acetic acid (for free fatty acids and

cholesterol; 5 mCi per dish; Nycomed Amersham plc,

Bucks., U.K.) or [14C]-serine (for sphingolipids; 0´5 mCiper dish), lipids were extracted with Bligh-Dyer solvent

and separated on 20 � 10-mm silica gel thin-layer

chromatography (TLC) plates (Merck). The extractedlipids were separated by one-dimensional TLC, utilizing

hexane/diethyl ether/acetic acid (70 : 30 : 1) to 10 cm

for free fatty acids and cholesterol. On another plate,the extracted lipids were separated utilizing chloro-

form/methanol/water (40 : 10 : 1) to 4 cm as the first

development solvent and chloroform/methanol/aceticacid (192 : 7:1) to 10 cm as the second solvent for

ceramide and glucosylceramide, and utilizing chloro-

form/methanol/acetic acid/water (25 : 15 : 4 : 2) to10 cm for sphingomyelin. Radioactivity was quantified

as photostimulated luminescence (PSL) for each spot

using a BAS-1500 Bioimage analyser (Fuji Film, Tokyo,Japan). Data are shown as PSL per cell.

Serine palmitoyltransferase activity

Serine palmitoyltransferase (SPT) activity was deter-

mined by a modification of a method describedpreviously.9 Keratinocytes were inoculated on to

90-mm plastic tissue culture dishes at 4 � 105 cells/

plate and cultured in 10 mL of medium under 5% CO2

at 37 8C for 4 days. The culture medium was switched

to one containing 10 mmol L21 nicotinamide followed

by further culture for 6 days. Keratinocytes wereharvested as follows. The cells were rinsed twice with

cold phosphate-buffered saline, and then scraped from

the dish. Recovered cells were suspended in HEPESbuffer [50 mmol L21 HEPES, pH 7´4, containing

10 mmol L21 ethylenediamine tetraacetic acid (EDTA),

526 O.TANNO et al.


5 mmol L21 dithiothreitol (DTT) and 0´25 mol L21

sucrose], and sonicated three times on an ice bath for5 s each at 80% power with a Handy Sonic UR-20P

(Tomy Seiko Co.). Differential centrifugation (4 8C) was

performed first at 800 g (15 min) then at 6000 g(15 min), followed by microsomal sedimentation at

100,000 g (60 min). The microsomal pellet was

resuspended in storage buffer (pH 7´4) containing50 mmol L21 HEPES, 5 mmol L21 DTT and 20%

glycerol (v/v), and stored at 2 808C until use. The

protein was determined using a BCA assay kit(Pierce Chemical Co., Rockford, IL, U.S.A.) using

bovine serum albumin as the standard. The

reaction mixture contained 50 mmol L21 pyridoxal-phosphate, 0´2 mmol L21 palmitoyl-coenzyme A

(CoA) and 0´5 mmol L21 serine, 20 mCi mL21

(1´3 � 1023 mmol L21) [3H]-l-serine (NycomedAmersham plc) in 0´2 mL of HEPES buffer

(50 mmol L21 HEPES, pH 8´3, containing 5 mmol L21

DTT and 3 mmol L21 EDTA). The assay mixture waspreincubated for 10 min at 37 8C, and the assay was

initiated by addition of 100 mg of microsomal proteinfollowed by incubation at 37 8C for 10 min. The

reaction was terminated by addition of 0´2 mL of

0´5 mol L21 NH4OH and immediate cooling on ice. Thereaction product, 3-ketodihydrosphingosine (3KDS), was

isolated as described previously,10 and the radioactivity

was counted with a liquid scintillation spectrometer(Aloka). Enzyme-specific activity is expressed as pmol of

3KDS formed min21 mg21 of protein.

Human LCB1 and LCB2 cDNA isolation

Human SPT is encoded by human LCB1 and LCB2

genes.11 Human LCB1 and LCB2 cDNAs were gener-ated by reverse transcription±polymerase chain reac-

tion (RT±PCR). The primers were designed according to

the sequences hLCB1s, hLCB1as, hLCB2s and hLCB2as,as described previously.11 Total RNA from cultured

keratinocytes was used for RT±PCR. PCR was per-

formed for 30 cycles of denaturation at 94 8C for1 min, annealing at 60 8C for 2 min and elongation at

72 8C for 2 min. The isolated cDNAs were ligated into

pGEM-T (Promega, Madison, WI, U.S.A.). Digoxigenin(DIG)-labelled anti-sense/sense RNA probes were pre-

pared from isolated cDNA using a DIG RNA labelling kit

(Boehringer Mannheim, Mannheim, Germany).

Northern blot analysis

Total RNA was isolated using an RNaid Kit (Bio 101, CA,

U.S.A). RNA samples were separated by electrophoresis

on 0´8% formaldehyde/agarose gels and transferred on

to nylon membranes. DIG-labelled LCB1 and LCB2 anti-sense RNA probes were prepared as described above.

Hybridization was performed according to the protocol

supplied with the DIG Nucleic Acid Detection Kit(Boehringer Mannheim).

Chemiluminescent detection was performed as

described previously.12 The membranes were washedat 95 8C for 3 min in 0´1% sodium dodecyl sulphate

(SDS), cooled to room temperature, and rehybridized

with a DIG-labelled glyceraldehyde-3-phosphatedehydrogenase anti-sense RNA probe (Clontech, Palo

Alto, CA, U.S.A.).

In vivo studies: effects of nicotinamide on stratum

corneum lipids and barrier function

For the in vivo studies, informed consent was obtained

from volunteers with dry skin (12 men aged 26±51years). Nicotinamide (2%) in 0´1% polyoxyethylene

(20) sorbitan monolaurate (w/v Wako) solution was

applied twice daily to one shin of volunteers and 0´1%polyoxyethylene (20) sorbitan monolaurate (vehicle)

alone to the contralateral shin for 4 weeks. Trans-

epidermal water loss (TEWL) was measured usingthe TEWAMETER TM210 (Courage and Khazaka,

Cologne, Germany) after 4 weeks of application. Data

were expressed as mg cm22 h21 (mean ^ SEM). After4 weeks of application, the stratum corneum was

stripped using cyanoacrylate resin. After lipids were

extracted, ceramides, cholesterol and free fatty acidswere divided using an amino column. Ceramides and

cholesterol were applied to TLC plates and quantified as

described above. Free fatty acids were quantified by theACS/ACOD method using free fatty acid measuring

reagent (NEFA C-test Wako). The protein was extracted

from the stratum corneum by boiling in 0´05% SDSand 0´01% 2-mercaptoethanol for 1 h. Protein was

quantified as described above.

Statistical analysis

The results were analysed by Student's t-test and

Dunnett's multiple comparison procedure.13 P , 0´05

indicated statistical significance.

Results

Nicotinamide increases ceramide synthesis

We first examined the effects of nicotinamide or nico-

tinic acid on ceramide synthesis in human keratinocytes



cultured for 2 days. When nicotinamide was added to

the culture at concentrations of 1±30 mmol L21,ceramide synthesis was increased in a dose-dependent

manner [4´1±5´5-fold compared with the control (no

addition); P , 0´01; Fig. 1]. Although nicotinic acidincreased ceramide synthesis, the effect of nicotinamide

was more potent than that of nicotinic acid. The cell

numbers per plate were almost the same at all doses ofnicotinamide and nicotinic acid examined (data not

shown).

We next examined the time course of the increase inceramide synthesis by nicotinamide. In this experiment,

we examined ceramide synthesis in culture for 6 h

every 2 days after addition of 10 mmol L21 nicotin-amide. Ceramide synthesis in the nicotinamide-treated

cultures was almost the same as that in the control

cultures (no addition) until the fourth day. However,ceramide synthesis in the nicotinamide-treated cultures

increased significantly after the fourth day (P , 0´05),

reaching a maximal level on the sixth day (P , 0´05).The increase in ceramide synthesis declined after the

sixth day despite application of nicotinamide every 2days (Fig. 2).

Nicotinamide increases glucosylceramide andsphingomyelin synthesis

To determine why nicotinamide increased cera-

mide synthesis, we attempted to determine whether

nicotinamide increases the synthesis of other sphin-golipids. When the cells were incubated with

10 mmol L21 nicotinamide for 6 days, levels of

glucosylceramide and sphingomyelin synthesis werealso increased, by 7´4-fold and 3´1-fold, respectively

(Fig. 3). The degree of increase in sphingomyelin

synthesis was the same as that in ceramide synthesis.

Nicotinamide increases serine palmitoyltransferase activity

To determine the mechanism responsible for the

increase in sphingolipid synthesis, the effect of nicotin-

amide on the activity of SPT, the rate-limiting enzymein the de novo synthesis pathway of sphingolipids, was

examined. Cells were incubated with 10 mmol L21

nicotinamide for 6 days, then the microsomal fractionwas isolated by centrifugation. The SPT activity of this

fraction was increased by about 1´2-fold (P , 0´01) as

compared with that of the control (Fig. 4A).

Induction of LCB gene expression by nicotinamide

As the SPT activity was increased by nicotinamide(Fig. 4A), total RNA extracted from keratinocytes

treated with nicotinamide was subjected to Northern

blot analysis to investigate whether nicotinamideupregulates the expression of LCB1 and LCB2

mRNA. Whereas the LCB2 probe hybridized with two

Figure 1. Nicotinamide increases ceramide synthesis dose-depen-

dently. Nicotinamide (1±30 mmol L21) or nicotinic acid (1 mmol L21)

was added to the culture for 6 days. Ceramide synthesis was measured

by incorporation of [14C]-serine into ceramide for 2 days (see Materialsand methods). Mean ^ SEM; n � 3; **P , 0´01, Dunnett's multiple

comparison; dpm, disintegrations per min.

Figure 2. Time course of ceramide synthesis after addition of

nicotinamide. Cells were cultured for 0, 2, 4, 6, 8 or 10 days

following addition (X) of 10 mmol L21 nicotinamide or without

addition (W). Ceramide synthesis is expressed as the ratio to that onday 0 (2´02 � 103 ^ 70´2 PSL per plate). Mean ^ SEM; n � 3;

*P , 0´05, Student's t-test.

528 O.TANNO et al.


transcripts (8´0 kb and 2´3 kb), the LBC1 probedetected only a single transcript (3´0 kb). When cells

were cultured with 10 mmol L21 nicotinamide for 2 or

6 days, the levels of LCB1 and LCB2 transcripts werealmost the same as those of controls. In contrast, 4-day

treatment significantly increased the levels of both

LCB1 and LCB2 transcripts compared with the controls(1´8-fold; Fig. 4b).

Nicotinamide increases not only ceramide synthesisbut also free fatty acid and cholesterol synthesis

The effects of nicotinamide on synthesis of other stratum

corneum lipids were examined. When the cells were

incubated with nicotinamide, levels of free fatty acids

and cholesterol synthesis were also increased, by 2´3-fold and 1´5-fold, respectively (Table 1).

Effects of nicotinamide on stratum corneum lipid levels

and barrier function

Topical application of nicotinamide resulted in an

increase in stratum corneum free fatty acid andceramide levels expressed in mg mg21 protein, by

67% (P , 0´1) and 34% (P , 0´05), respectively

(Fig. 5A). After application for 4 weeks, TEWL of theskin treated with nicotinamide was 27% lower than

that of skin treated with vehicle (Fig. 5B).

Discussion

Nicotinamide increased biosynthesis of ceramides aswell as other stratum corneum intercellular lipids (free

fatty acids and cholesterol). Topical application of

nicotinamide also increased the levels of ceramides,free fatty acids and cholesterol in the stratum corneum.

Because the effect of nicotinamide on the synthesis of

ceramide was more potent than that of nicotinic acid,the former may act like NAD. We examined the effects

of some derivatives of nicotinamide and found that

6-aminonicotinamide increased the synthesis of cera-mide (data not shown). Because 6-aminonicotinamide

is metabolized to a biologically inactive analogue of

Figure 4. Nicotinamide increases activity and mRNA of serine pal-mitoyltransferase (SPT). (a) Cells were cultured with 10 mmol L21

nicotinamide for 6 days. SPT activity was measured in the

microsomal fraction (see Materials and methods). Mean ^ SEM;

n � 3; **P , 0´01. (b) Northern blot analysis of LCB1 and LCB2transcripts in nicotinamide-treated cells. Cells were cultured with

10 mmol L21 nicotinamide for 2, 4 or 6 days. Equal amounts of total

RNA (5 mg per lane) were used for Northern blot analyses. The filters

were first hybridized with either the LCB1 or LCB2 probe, thenstripped and rehybridized with the glyceraldehyde-3-phosphate

dehydrogenase (G3PDH) probe.

Figure 3. Nicotinamide increases glucosylceramide and sphingo-

myelin synthesis. Cells were cultured with 10 mmol L21 nicotinamide

for 6 days. Glucosylceramide, sphingomyelin and ceramide syntheses

are expressed as ratios to the respective control (no addition ofnicotinamide). Control values were 3´23 � 1024 ^ 0´27 � 1024

PSL per cell (ceramide), 1´81 � 1024 ^ 0´16 � 1024 PSL per cell

(glucosylceramide) and 1´65 � 1024 ^ 0´16 � 1024 PSL per cell(sphingomyelin). Mean ^ SEM; n � 4; **P , 0´01, Student's t-test.

Table 1. Nicotinamide increases the synthesis of stratum corneumintercellular lipids in cultured keratinocytes

Rate of synthesis (PSL 1024 cells in 6 h)

Control Nicotinamide

Free fatty acids 109 ^ 7´73 248 �̂ 12´0***Cholesterol 346 �^ 28´7 507 �̂ 14´1**

Nicotinamide (10 mmol L21) was added to keratinocyte cultures. Free

fatty acid and cholesterol synthesis was measured as incorporation of

[14C]-acetic acid for 6 h (see Materials and methods). Mean ^ SEM;n � 3; **P , 0´01; ***P , 0´001, Student's t-test.



NAD, the result suggests that nicotinamide, as well as

6-aminonicotinamide, increase ceramide synthesisby different mechanisms from those associated with

NAD.

We also demonstrated that nicotinamide increasedboth the specific activity of SPT per mg microsomal

protein and the SPT mRNA level, associated with aconcomitant increase in ceramide synthesis. These

results indicated that the increase in sphingolipid

synthesis with nicotinamide may be due to an increasein SPT activity. First, in the biosynthesis pathway of

ceramide, palmitoyl-CoA is conjugated with serine by

STP, and produces 3KDS. Next, 3KDS is reduced todihydrosphingosine and, finally, ceramide is produced

via N-acyldihydrosphingosine. The reduction step,

3KDS to dihydrosphingosine, is catalysed by NADP-dependent reductase. However, STP does not depend on

NAD/NADP. The increases in activity of SPT and SPT

mRNA level enhanced by nicotinamide also demon-strate that the effect of nicotinamide is not exclusively

due to cofactor-limited enzyme activity.

Nicotinamide increased not only ceramide synthesisbut also free fatty acid and cholesterol synthesis. These

results cannot be explained by the increase in SPT

activity. There may be other mechanisms involved in

the increase of free fatty acid and cholesterol synthesisby nicotinamide. This is now under investigation in our

laboratory.

Both increases in ceramide synthesis and in themRNA levels of LCB1 and LCB2 were delayed by 4 days

after addition of nicotinamide to the cultures (Figs 2

and 4). When the nicotinamide was applied to thecultured cells from the fourth day, the peak increase in

ceramide synthesis shifted to the 10th day (data not

shown). These findings suggest that several steps arerequired to activate ceramide synthesis following

nicotinamide treatment. A recent study showed that

ultraviolet (UV) B-induced increases in mRNA levels ofLCB1 and LCB2 in cultured human keratinocytes were

delayed by 48 h after UVB irradiation.14 Tumour

necrosis factor-a and interleukin-1a were also foundto induce delayed increases in hepatic ceramide

synthesis, SPT activity and SPT mRNA levels 24 h

after treatment in HepG2 cells.15 On the basis of thesefindings, it was suggested that the UVB-induced

increases in sphingolipid production may be regulatedby increased epidermal cytokine levels. The nicotina-

mide-induced increase in sphingolipid production may

also be regulated in an autocrine or paracrine mannerby epidermal cytokines. It is not clear why the

nicotinamide-induced increase in biosynthesis of cer-

amides was decreased with prolongation of the culturetime. As the biosynthesis of ceramide was significantly

increased by nicotinamide, ceramides and other lipids

rapidly accumulated in the cells with prolongation ofthe culture time (data not shown). Feedback inhibition

or repression by these lipids accumulating in the cells

might occur and decrease the nicotinamide-inducedincrease in biosynthesis of ceramides.

A recent study showed that vitamin C also increases

the ceramide and glucosylceramide contents in recon-stituted epidermis.16 Although the contents of gluco-

sylceramide and of ceramides 6 and 7 were markedly

increased in vitamin C-supplemented medium, levels ofother ceramides were not significantly altered. On the

other hand, nicotinamide increased the total ceramide

content, but not those of specific ceramides (data notshown). Thus, the mechanism responsible for the

increase in ceramide synthesis by nicotinamide appears

to be different from that induced by vitamin C.Some studies have shown that transient increases in

ceramide levels are associated with the induction of

apoptosis.17,18 These transient increases in ceramidelevels were associated with activation of sphingomyelin

hydrolysis.19,20 In the present study, cell numbers in

Figure 5. Topical application of nicotinamide increases the level of

ceramide and other intercellular lipids in the stratum corneum and

results in improvement of the deficient permeability barrier function.

Nicotinamide (2%) was topically applied twice daily to one shin ofvolunteers and vehicle [0´1% polyoxyethylene (20) sorbitan mono-

laurate] alone to the contralateral shin for 4 weeks. (A) Stratum

corneum was stripped by cyanoacrylate resin after 4 weeks of

application. After lipids were extracted, ceramides, cholesterol andfree fatty acids were quantified (see Materials and methods). The

level of lipids was expressed in mg mg21 protein. (B) Transepi-

dermal water loss (TEWL) was measured after 4 weeks of appli-cation. Mean ^ SEM; n � 12; 1P , 0´1; *P , 0´05; **P , 0´01,

Student's t-test.

530 O.TANNO et al.


cultures treated with nicotinamide were the same as

those in control cultures. Moreover, the amount ofsphingomyelin was not decreased and cells did not

show apoptotic morphology. Therefore, it is unlikely

that the increase in ceramide level induced by nicotin-amide leads to an apoptotic response of keratinocytes.

We have previously shown that nicotinamide accel-

erates cornified envelope formation in keratinocytesand expression of differentiated type keratin K1 over

the same concentration range as used in the present

study.21 Ceramide levels have been reported to beincreased at the final stage of terminal differentia-

tion,22,23 and to be involved in the regulation of

differentiation of human keratinocytes.24 It is unclearwhether the increase in ceramide level caused by

nicotinamide results in enhanced differentiation of

keratinocytes or vice versa, although an acceleratorof keratinocyte differentiation such as ethyl a-d-gluco-

side25 cannot always increase ceramides under our

conditions (data not shown). The relationship betweenthe nicotinamide-induced increase in ceramide levels

and the enhanced differentiation of keratinocytes isnow under investigation in our laboratory.

In an in vivo study, topical application of nicotina-

mide to the dry skin of human volunteers increasedceramide and free fatty acid levels in the stratum

corneum, and tended to increase the cholesterol level.

TEWL of the skin was decreased by the application ofnicotinamide to dry skin. The level of ceramides in the

stratum corneum was correlated directly with the TEWL

of the skin (coefficient of correlation r � 2 0´31,P , 0´01). These data indicate that an increase in

the levels of intercellular lipids, especially ceramides,

in the stratum corneum resulted in an improvement indeficient epidermal permeability barrier. The results of

the present study suggest that nicotinamide acts on

epidermal cells to improve the epidermal permeabilitybarrier by increasing synthesis of ceramide and other

stratum corneum intercellular lipids, although nico-

tinamide, one of the niacin compounds, has been usedfor the treatment of cutaneous lesions because symp-

toms were similar to those of nicotinamide deficiency.

In conclusion, we have shown that nicotinamideincreases biosynthesis of intercellular lipids such as

sphingolipids including ceramide, glucosylceramide

and sphingomyelin, free fatty acids and cholesterol incultured human keratinocytes. These increases are due

in part to the upregulation of SPT. The present study is

the first to demonstrate that nicotinamide increasesceramides and other intercellular lipids in the stratum

corneum, and suggests that nicotinamide may play a

part in the maintenance of the epidermal permeability

barrier via regulating stratum corneum intercellularlipid synthesis and/or cellular differentiation.

Acknowledgments

The authors thank Mr Masayuki Matsumoto, Ms Rie

Hikima and Mr Masakatsu Ota for their technicalassistance and support in the in vivo studies. We express

our gratitude to Dr Yoshikazu Uchida for his kind and

helpful advice.

References

1 Elvehjem CA, Madden RJ, Strong FM, Woolley DW. The isolation

and identification of the anti-black tongue factor. J Biol Chem1938; 123: 137±49.

2 Williams ML, Elias PM. The extracellular matrix of stratumcorneum: role of lipids in normal and pathological function. CRC

Crit Rev Ther Drug Carrier Syst 1987; 3: 95±122.

3 Grubauer G, Feingold KR, Harris RM, Elias PM. Lipid content and

lipid type as determinants of the epidermal permeability barrier.

J Lipid Res 1989; 30: 89±96.

4 Elias PM, Menon GK. Structural and lipid biochemical correlates

of the epidermal permeability barrier. Adv Lipid Res 1991; 24:

1±26.

5 Yamamoto A, Serizawa S, Ito M, Sato Y. Stratum corneum lipid

abnormalities in atopic dermatitis. Arch Dermatol Res 1991; 283:219±23.

6 Imokawa G, Abe A, Jin K et al. Decreased level of ceramides in

stratum corneum of atopic dermatitis: a factor in atopic dry skin?J Invest Dermatol 1991; 96: 523±6.

7 Rawlings AV, Watkinson A, Rogers J et al. Abnormalities instratum corneum structure, lipid composition and desmosome

degradation in soap-induced winter xerosis. J Soc Cosmet Chem

1994; 45: 203±20.

8 Bligh EG, Dyer WJ. A rapid method of total lipid extraction and

purification. Can J Biochem Physiol 1959; 37: 911±17.

9 Williams RD, Wang E, Merrill AH. Enzymology of long-chain base

synthesis by liver: characterization of serine-palmitoyl transferase

in rat liver microsomes. Arch Biochem Biophys 1984; 228:282±91.

10 Holleran WM, Williams ML, Gao WN et al. Serine-palmitoyltransferase activity in cultured human keratinocytes. J Lipid Res

1990; 31: 1655±61.

11 Weiss B, Stoffel W. Human and murine serine-palmitoyl-CoA

transferase cloning, expression and characterization of the key

enzyme in sphingolipid synthesis. Eur J Biochem 1997; 249:

239±47.

12 Engler-Blum G, Meier M, Frank J, Muller G. Reduction of

background problem in nonradioactive Northern and Southernblot analyses enables higher sensitivity than 32P-based hybrid-

izations. Anal Biochem 1989; 210: 235±44.

13 Dunnett CW. A multiple comparison procedure for comparing

several means with a control. J Am Stat Assoc 1955; 50:

1096±121.

14 Farrell AM, Uchida Y, Nagiec MM et al. UVB irradiation stimulates

sphingolipid synthesis in cultured human keratinocytes. J Lipid

Res 1998; 39: 2031±8.

15 Memon RA, Holleran WM, Moser AH et al. Endotoxin and



cytokines increase hepatic sphingolipid biosynthesis and producelipoproteins enriched ceramides and sphingomyelin. Arterioscler

Thromb Vasc Biol 1998; 18: 1257±65.

16 Ponec M, Weerheim A, Kempenaar J et al. The formation ofcompetent barrier lipids in reconstructed human epidermis

requires the presence of vitamin C. J Invest Dermatol 1997;

109: 348±55.

17 Okazaki T, Bell RM, Hannun YA. Sphingomyelin turnoverinduced by vitamin D3 in HL-60 cells: role in cell differentiation.

J Biol Chem 1989; 261: 19076±80.

18 Okazaki T, Bielawska A, Bell RM. Role of ceramide as a lipidmediator of alpha, 25-dihydroxyvitamin D3-induced HL-60 cell

differentiation. J Biol Chem 1990; 265: 15823±31.

19 Kim MY, Linardic C, Obeid L, Hannun Y. Identification of

sphingomyelin turnover as an effector mechanism for the actionof tumor necrosis factor a and g-interferon. J Biol Chem 1991;

266: 484±9.

20 Verheij M, Bose R, Lin XH et al. Requirement for ceramide-initiated SAPK/JNK signaling in stress-induced apoptosis. Nature

1996; 380: 75±9.

21 Kitamura N, Tanno O, Ota Y, Yasuda R. Effects of niacinamide onthe differentiation of human keratinocyte. J Dermatol Sci 1996;

12: 202(Abstr.).

22 Gray GM, Yardley HJ. Different populations of pig epidermal cells:

isolation and lipids in neonatal mouse stratum granulosum andstratum corneum. J Lipid Res 1975; 16: 441±7.

23 Lampe MA, Williams ML, Elias PM. Human epidermal lipid:

characterization and modulation during differentiation. J Lipid

Res 1983; 24: 131±40.24 Uchida Y, Elias PM, Holleran WM. Coordinate regulation of

keratinocyte proliferation by ceramides and glucosylceramides.

J Invest Dermatol 1994; 102: 594(Abstr.).25 Kitamura N, Ota Y, Haratake A et al. Effects of ethyl a-d-glucoside

on skin barrier disruption. Skin Pharmacol 1997; 101: 153±9.

Nicotinamide increases biosynthesis of ceramides as well as other stratum corneum lipids to improve...

Documents

Transcript of Nicotinamide increases biosynthesis of ceramides as well as other stratum corneum lipids to improve...