J Ishikawa et al.

Ceramide Profile in Atopic Dermatitis

skeleton and delayed spreading of keratinocyte lines derived from EBS patients. J Cell
Sci 108(Pt 11):346371
Oji V, Hautier JM, Ahvazi B et al. (2006) Bathing
suit ichthyosis is caused by transglutaminase1 deficiency: evidence for a temperaturesensitive phenotype. Hum Mol Genet 15:
308397

Persikov AV, Pillitteri RJ, Amin P et al. (2004)

Stability related bias in residues replacing
glycines within the collagen triple helix (GlyXaa-Yaa) in inherited connective tissue disorders. Hum Mutat 24:3307
Xu K, Nowak I, Kirchner M et al. (2008)
Recombinant collagen studies link the severe
conformational changes induced by osteo-

genesis imperfecta mutations to the disruption of a set of interchain salt bridges. J Biol
Chem 283:3433744
Woodley DT, Hou Y, Martin S et al. (2008)
Characterization of molecular mechanisms
underlying mutations in dystrophic epidermolysis bullosa using site-directed mutagenesis. J Biol Chem 283:1783845

Changes in the Ceramide Profile of Atopic

Dermatitis Patients
Journal of Investigative Dermatology (2010) 130, 25112514; doi:10.1038/jid.2010.161; published online 24 June 2010

TO THE EDITOR
We wish to report the characteristic
differences that have been identified
between the ceramide (CER) profiles
of atopic dermatitis (AD) patients and
those of healthy individuals.
AD is characterized by impaired
stratum corneum (SC) functions, which
can be indicated by either an increase
in transepidermal water loss (TEWL) or
a decrease in water-holding function
(capacitance). On the other hand, a
close relationship between the impaired SC functions and CERs has been
reported (Elias, 1983; Melnik et al.,
1988; Yamamoto et al., 1991; Motta
et al., 1993, Paige et al., 1994; Di
Nardo et al., 1998; Matsumoto et al.,
1999; Macheleidt et al., 2002; Cho
et al., 2004). More recent research has
attemptedbut failedto uncover the
diagnostic potential of CER profiling
for skin pathologies, including AD
(Farwanah et al., 2005).
Human SC CERs can be divided into
11 groups according to their fatty acid
and sphingoid structures (Masukawa
et al., 2008), as shown in Figure 1a:
CER[NDS] contains non-OH fatty
acids [N] and dihydrosphingosines
[DS]; CER[NS] contains [N] and sphingosines [S]; CER[NH] contains [N] and
6-hydroxy sphingosines [H]; CER[NP]
contains [N] and phytosphingosines
[P]; CER[ADS] contains a-OH fatty
acids [A] and [DS]; CER[AS] contains
[A] and [S]; CER[AH] contains [A] and
[H]; CER[AP] contains [A] and [P];

CER[EOS] contains ester-linked fatty

acids, and o-OH fatty acids [EO] and
[S]; CER[EOH] contains [EO] and [H];
and CER[EOP] contains [EO] and [P].
These classes have been further subdivided into species by chain length.
To date, we have identified 350 species
in human SC, using normal-phase
liquid chromatography coupled with
electrospray ionizationmass spectrometry (Masukawa et al., 2008, 2009;
it is worth noting that the former was
used to separate CERs into classes
based on the polarities). Using the same
procedure, we obtained an exhaustive
CER profile of the skin of AD patients
and healthy individuals to find a characteristic difference between their profiles.
The study was conducted according
to the Declaration of Helsinki Principles. The protocol was approved by the
ethics committees of the Kao Corporation and Dokkyo Medical University.
Informed consent to participate in this
study was obtained from the patients.
Eight subjects with mild levels of AD
(four men and four women; age
range, 1637; mean, 28) and seven
healthy individuals with no history of
skin disorder (four men and three
women; age range, 2537; mean, 31)
were enrolled in this study. First, we
evaluated TEWL and capacitance, respectively, using a Tewameter TM
210 and a Corneometer CM 825
(Courage Khazaka Electronic GmbH,
Cologne, Germany) in the affected

Abbreviations: AD, atopic dermatitis; CER, ceramide; SC, stratum corneum; TEWL, transepidermal water
loss

and unaffected sites on the forearm

skin of AD patients and on the forearm
skin of healthy individuals. After confirming that TEWL and capacitance
were significantly higher and lower,
respectively, in the affected sites of the
AD patients as compared with those of
the unaffected sites of the AD patients
and healthy individuals (Figure 1b and
c), we sampled SC specimens from
these skin sites by performing tape
stripping 10 times. CERs were in turn
extracted from these specimens and
analyzed with an Agilent 1100
Series LC/MSD single-quadrupole system equipped with an electrospray
ionization source, ChemStation software, a 1,100-well-plate autosampler
(Agilent Technologies, Palo Alto, CA),
and an Inertsil SIL 100A-3, 1.5 mm
i.d.  150 mm column (GL Science,
Tokyo, Japan). These procedures are
detailed in our previous reports
(Masukawa et al., 2008, 2009).
The levels of total CER as well as
those of CER[NH], CER[NP], CER[EOS],
CER[EOH], and CER[EOP] classes were
found to be significantly lower in the
affected sites of AD patients as compared with the normal skin sites of
healthy individuals (Figure 1d and e).
These findings are consistent with
previous findings (Imokawa et al.,
1991; Yamamoto et al., 1991; Di Nardo
et al., 1998; Bleck et al., 1999; Macheleidt et al., 2002). In addition, this study
revealed that the expression levels of
CER[AS] were significantly higher in the
affected sites of AD patients as compared with the normal sites of healthy
individuals.
www.jidonline.org 2511

J Ishikawa et al.
Ceramide Profile in Atopic Dermatitis

Non-hydroxy fatty acid -Hydroxy fatty acid Esterified -hydroxy fatty acid
[EO]
[N]
[A]
OH

Fatty acid

Sphingoid

OH

Dihydrosphingosine
[DS] H2N

OH

OH

CER[NDS]

CER[ADS]

Unidentified

CER[NS]

CER[AS]

CER[EOS]

CER[NH]

CER[AH]

CER[EOH]

CER[NP]

CER[AP]

CER[EOP]

OH
OH

Sphingosine
H2N
[S]

OH

OH

6-Hydroxy sphingosine
H2N

[H]

OH
OH

OH

Phytosphingosine
H2N

OH

15

20

ed
ct

tro

fe
Af

na
U

CER[NS]
0.6

MeanSD

0.5
ng per g protein

B
A
B

0.4
A

0.3
0.2
0.1

AP

EO
S
EO
H
EO
P

0
AH

on

ct
ffe

fe
Af

MeanSD

S
N
S
N
H
N
P
AD
S
AS

ed

ed
ct

tro
on
C

Un
af
fe
cte
d

Ceramide classes
10
9
8
7
6
5
4
3 A
2
1
0
D

ng per g protein

Af
fe
ct
ed

10

10

20

30

tro

25

40

on

30

MeanSD

45
40
35
30
25
20
15
10
5
0

50

Total ceramides

MeanSD

ed

MeanSD

Capacitance

AU
35

ct

TEWL
g m2 h
60

ffe

OH

na

OH

ng per g protein

[P]

A
A

A
B A

32

36

A
A
A

34

A
A

B
A

38

40 42 44 46 48
Total carbon atoms

50

52

54

Figure 1. Stratum Corneum (SC) functions and ceramide profiles in AD patients and healthy subjects. (a) Structure and nomenclature of ceramides
from the human SC. The mean values of (b) transepidermal water loss (TEWL), (c) capacitance, and the mean levels of (d) total ceramides, of (e) each
ceramide class, and of (f) each species in CER[NS] in the affected and unaffected sites of AD patients (n 8) and in the normal skin sites of healthy
individuals (n 7) are indicated by black, gray, and white bars, respectively. The uppercase letters (A, B) denote significant difference by Bonferronis
post hoc multiple comparison test (Po0.05 and Po0.01, respectively).

A significant difference was further

detected by dividing each class into
species. As a result, we found in
AD patients that the larger species
(450 total carbons) of the CER[NS],
CER[NDS], CER[NH], CER[AS], and
CER[AH] tended to be expressed at
lower levels, whereas the smaller
species (o40 total carbons) of the

CER[NS], CER[NDS], and CER[AS]

tended to be expressed at higher levels
than the equivalent levels in healthy
individuals. A typical example of this
tendency is CER[NS] (illustrated in
Figure 1f), in which the smaller species
(o40 total carbons) were expressed at
significantly higher levels and the larger
species (450 total carbons) were

2512 Journal of Investigative Dermatology (2010), Volume 130

expressed at significantly lower levels

in the affected sites of AD patients as
compared with those of healthy individuals.
In addition, we detected a parameter
that reflects the SC barrier impairments
in the CER profile. A correlation with
the values of SC barrier functions was
detected in the expression level of CER

J Ishikawa et al.
Ceramide Profile in Atopic Dermatitis

Table 1. Correlation coefficient (a) between the SC functional parameters

(TEWL and capacitance) and ceramide level, and (b) between the
parameters and the average number of carbon atoms per class
Correlation coefficients
TEWL

Capacitance

carbon atoms of CER[NDS], CER[NS],

CER[NH], CER[AS], and CER[AH].
Taken together, our findings suggest
that the CER profile is a candidate
marker for the diagnosis and prognosis
of AD.

(a) Levels of CER

Total CER

0.675 c

0.629 b

CONFLICT OF INTEREST

CER[NDS]

0.638 b

0.615 b

The authors state no conflict of interest.

CER[NS]

0.282

CER[NH]

0.730 c

0.613 b

CER[NP]

0.740 c

0.630 b

CER[ADS]

0.385

0.232

0.417 a

CER[AS]

0.451 a

CER[AH]

0.696 c

0.670 c

CER[AP]

0.437 a

0.506 a

CER[EOS]

0.593 b

0.571 b

CER[EOH]

0.681 c

0.666 c

CER[EOP]

0.691 c

0.656 c

0.773 c

0.563 b

CER[NDS]

0.862 c

0.659 c

CER[NS]

0.852 c

0.641 c

CER[NH

0.784 c

0.629 b

CER[NP]

0.190

0.076

CER[ADS]

0.216

0.159

CER[AS]

0.637 c

0.512 a

CER[AH]

0.724 c

0.687 c

CER[AP]

0.229

C34-CER[NS]

0.360

(b) Average numbers of total carbons

CER[EOS]

Junko Ishikawa1, Hirofumi Narita2,

Naoki Kondo2, Mitsuyuki Hotta1,
Yutaka Takagi1, Yoshinori Masukawa2,
Takashi Kitahara1, Yoshinori Takema1,
Satomi Koyano3, Soji Yamazaki3 and
Atsushi Hatamochi3
1

Biological Science Laboratory, Kao

Corporation, Tochigi, Japan; 2Analytical
Science Laboratory, Kao Corporation, Tochigi,
Japan and 3Department of Dermatology,
Dokkyo University School of Medicine,
Tochigi, Japan
E-mail: ishikawa.junko1@kao.co.jp

REFERENCES

0.129

0.342

0.524 a

CER[EOH]

0.045

0.165

CER[EOP]

0.847 c

0.633 b

Abbreviations: CER, ceramide; SC, stratum corneum; TEWL, transepidermal water loss.
The lowercase letters (a, b, and c) denote significant correlations (Po0.05, Po0.01, and Po0.001,
respectively).

(total, class) and also in the average

number of carbon atoms in each CER
class. Table 1a and b summarizes the
correlation coefficients and their
significance. Previously, a significant
negative correlation with TEWL value
was found in both the content ratio and
the quantity of CER[NP], and a significant positive correlation was found in
the content ratio of CER[AH] (Di Nardo
et al., 1998). In our study, a significant
correlation with TEWL value was

observed but was not limited to the

two classes. A strong correlation with
the impaired SC barrier functions was
found in the expression level of
C34-CER[NS] as well as in the average
number of carbon atoms of CER[EOP],
whereas a strong correlation with the
improved SC barrier functions was
found in the expression levels of
CER[NDS], CER[NH], CER[AH], CER
[AP], CER[EOS], CER[EOH], and CER
[EOP] and in the average number of

Bleck O, Abeck D, Ring J, Hoppe U, Vietzke JP,

Wolber R et al. (1999) Two ceramide
subfractions detectable in Cer(AS) position
by HPTLC in skin surface lipids of
non-lesional skin of atopic eczema. J Invest
Dermatol 113:894900
Cho Y, Lew BL, Seong K, Kim NI (2004)
An inverse relationship between ceramide
synthesis and clinical severity in patients
with psoriasis. J Korean Med Sci 19:
85963
Di Nardo A, Wertz P, Giannetti A, Seidenari S
(1998) Ceramide and cholesterol composition of the skin of patients with
atopic dermatitis. Acta Dermatol Venereol
78:2730
Elias PM (1983) Epidermal lipids, barrier function,
and desquamation. J Invest Dermatol
80(Suppl):44s9s
Farwanah H, Raith K, Neubert RH, Wohlrab J
(2005) Ceramide profiles of the uninvolved
skin in atopic dermatitis and psoriasis
are comparable to those of healthy skin.
Arch Dermatol Res 296:51421
Imokawa G, Abe A, Jin K, Higaki Y, Kawashima
M, Hidano A (1991) Decreased level of
ceramides in stratum corneum of atopic
dermatitis: an etiologic factor in atopic dry
skin? J Invest Dermatol 96:5236
Macheleidt O, Kaiser HW, Sandhoff K (2002)
Deficiency of epidermal protein-bound
omega-hydroxyceramides in atopic dermatitis. J Invest Dermatol 119:16673
Masukawa Y, Narita H, Sato H, Naoe A, Kondo N,
Sugai Y et al. (2009) Comprehensive
quantification of ceramide species in
human stratum corneum. J Lipid Res 50:
170819

www.jidonline.org 2513

C Busch et al.
The p53 Pathway Predicts Survival in Melanoma

Masukawa Y, Narita H, Shimizu E, Kondo N,

Sugai Y, Oba T et al. (2008) Characterization
of overall ceramide species in human stratum
corneum. J Lipid Res 49:146676
Matsumoto M, Umemoto N, Sugiura H, Uehara M
(1999) Difference in ceramide composition
between dry and normal skin in patients
with atopic dermatitis. Acta Dermatol Venereol 79:2467

Melnik B, Hollmann J, Plewig G (1988)

Decreased stratum corneum ceramides
in atopic individualsa pathobiochemical
factor in xerosis? Br J Dermatol 119:
5479

Paige DG, Morse-Fisher N, Harper JI (1994)

Quantification of stratum corneum ceramides and lipid envelope ceramides in
the hereditary ichthyoses. Br J Dermatol
131:237

Motta S, Monti M, Sesana S, Caputo R, Carelli S,

Ghidoni R (1993) Ceramide composition
of the psoriatic scale. Biochim Biophys Acta
1182:14751

Yamamoto A, Serizawa S, Ito M, Sato Y (1991)

Stratum corneum lipid abnormalities in
atopic dermatitis. Arch Dermatol Res
283:21923

Alterations in the p53 Pathway and p16INK4a Expression

Predict Overall Survival in Metastatic Melanoma Patients
Treated with Dacarbazine
Journal of Investigative Dermatology (2010) 130, 25142516; doi:10.1038/jid.2010.138; published online 27 May 2010

TO THE EDITOR
Malignant melanoma is one of the least
chemosensitive human cancers. The
cause of drug resistance in melanoma
remains poorly understood. So far, no
cytotoxic regimen has revealed superiority compared to dacarbazine monotherapy (Eggermont and Kirkwood, 2004).
Although mutations inactivating the
TP53 gene (encoding the p53 protein)
are rare events in melanoma (Hartmann
et al., 1996), p53 may be inactivated
through other mechanisms. The murine
double-minute oncogene (MDM2) binds
to and ubiquitinates p53 (Piette et al.,
1997), and MDM2 amplification has
been found responsible for p53 inactivation in some malignancies (Momand
et al., 1998). Recently, an MDM2
promoter polymorphism (SNP309T4G)
increasing MDM2 expression through
enhanced binding of the Sp1 transcriptional activator was shown to enhance
carcinogenesis (Bond et al., 2004).
p14ARF activates p53 by inhibiting
MDM2 binding (Figure 1a); thus, loss of
p14ARF may reduce p53 function
(Efeyan and Serrano, 2007). Although
p14ARF in general is activated in
response to oncogenic stimulation, there
is evidence that p14ARF may also be
involved in response to genotoxic stress
(Christophorou et al., 2006). p14ARF is
encoded by the alternative reading

frame of the CDKN2A gene. Although

most CDKN2A mutations identified in
melanoma affect the p16INK4a transcript only (Grafstrom et al., 2005;
Knappskog et al., 2006; Goldstein
et al., 2007), mutations affecting the 50
half of exon 2 may affect both transcripts.
Here, we analyzed pretreatment
tumor biopsies from 85 patients
receiving dacarbazine for advanced
malignant melanoma (Busch et al.,
unpublished data) for potential disturbances in p53, MDM2, p14ARF, or
p16INK4a. Pretreatment biopsies from
all patients were snap frozen in liquid
nitrogen. Each patient provided written
informed consent, and the study
protocol was approved by the regional
ethical committee. The study was
conducted according to the Declaration
of Helsinki Principles. Of the 85 patients, 75 were evaluable for clinical
response according to the UICC-criteria
(Busch et al., unpublished data). RNA
was extracted with TRIzol (Invitrogen,
Carlsbad, CA), and cDNA was synthesized with reverse transcriptase (Roche
Diagnostics, Basel, Switzerland). The
tumors were screened for TP53 and
CDKN2a mutations and MDM2 promoter status by PCR amplification and
subsequent sequencing. In addition,
intragenetic deletions or amplifications
in CDKN2A or MDM2 were screened

Abbreviations: BMI-1, polycomb ring finger oncogene; MDM2, mouse double-minute homolog;
OS, overall survival

2514 Journal of Investigative Dermatology (2010), Volume 130

for using MLPA (MRC Holland,

Amsterdam, the Netherlands; Schouten
et al., 2002). mRNA expression
analysis of p16INK4a and BMI-1 was
performed in a duplex qRT-PCR system
(LightCycler 480; Roche Diagnostics)
using TaqMan probes and b-2-microglobulin as an internal reference.
Dilutions of PCR products of these
genes were included in each run to
make a standard curve for calculation
of relative concentrations.
We considered the p53 pathway
to be disturbed in tumors harboring
one of the following events: TP53 or
p14ARF mutations causing amino-acid
changes, biallelic p14ARF deletions,
MDM2 amplification, and/or homozygosity for the SNP309G haplotype.
To evaluate the impact of p16INK4a
expression levels, we compared tumors
expressing p16INK4a above or below
median of the analyzed samples.
Potential
correlations
between
molecular parameters and clinical
outcome were tested using the Fishers
exact test comparing objective responders (complete response or partial
response) to nonresponders (progressive
disease or stable disease) or stable
disease complete response/partial
response versus progressive disease.
Univariate analyses were performed
analyzing different factors impact on
time to progression and overall survival
(OS) using log-rank test. Variables
with P-value o0.10 were entered into

Copyright of Journal of Investigative Dermatology is the property of Nature Publishing Group and its content
may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express
written permission. However, users may print, download, or email articles for individual use.