Design, synthesis, and evaluation of β-galactosylceramide mimics promoting β-glucocerebrosidase activity in keratinocytes

Article abstract of DOI:10.1016/S0960-894X(03)00006-4

We have established an efficient synthesis of mimics of β-galactosylceramide (β-GalCer) increasing a β-glucocerebrosidase (β-GlcCer'ase) activity that associates with the skin barrier function. Among the synthetic β-GalCer analogues (6a-6e) described herein, compound 6e exhibited a potent effect on the activation of β-GlcCer'ase function in vitro and reduced the transepidermal water loss (TEWL) level in a UVB-induced barrier disrupted mice model. These findings indicated that compound 6e could be useful for cosmetics and medicines to improve skin barrier function.

Full text of DOI:10.1016/S0960-894X(03)00006-4

Bioorganic & Medicinal Chemistry Letters 13 (2003) 813–815
Design, Synthesis, and Evaluation of ꢀ-Galactosylceramide
Mimics Promoting ꢀ-Glucocerebrosidase Activity in Keratinocytes
Kyoko Fukunaga,a,b,* Masahiro Yoshida,b,y Fumio Nakajima,b,c Rie Uematsu,b
Mariko Hara,a Shintaro Inoue,a Hirosato Kondoc,{ and Shin-Ichiro Nishimurad
aBasic Research Laboratory, Kanebo Ltd., 3-28, 5-chome Kotobuki-cho, Odawara-shi, Kanagawa-ken 250-0002, Japan
bSapporo Laboratory for Glycocluster Project, Japan Bioindustry Association, Hokkaido University, Sapporo 060-0810, Japan
cDepartment of Chemistry, Nippon Organon Ltd., 5-90, 1-chome Tomobuchi-cho, Miyakojima-ku, Osaka, Japan
dLaboratory of Bio-Macromolecular Chemistry, Division of Biological Sciences, Graduate School of Science,
Hokkaido University, Sapporo 060-0810, Japan
Received 1 November 2002; accepted 27 December 2002
Abstract—We have established an efficient synthesis of mimics of b-galactosylceramide (b-GalCer) increasing a b-glucocere-
brosidase (b-GlcCer’ase) activity that associates with the skin barrier function. Among the synthetic b-GalCer analogues (6a–6e)
described herein, compound 6e exhibited a potent effect on the activation of b-GlcCer’ase function in vitro and reduced the trans-
epidermal water loss (TEWL) level in a UVB-induced barrier disrupted mice model. These findings indicated that compound 6e
could be useful for cosmetics and medicines to improve skin barrier function.
# 2003 Elsevier Science Ltd. All rights reserved.
b-Galactosylceramide (b-GalCer) is one of the simplest
glycosphingolipids having a lipophilic sphingosine and a
hydrophilic galactose moiety attached via an ether link-
age to sphingosine. b-GalCer is found abundantly in the
myelin sheath of the central and peripheral nervous
system,1,2 and acts as cell surface receptors for bacterial
toxins and viruses.3,4
diseases, such as dry skin, atopic dermatitis and psoriasis,
is only produced by b-GlcCer’ase not but by sphingo-
myelinase.9 Actually, it has been demonstrated that
b-GalCer increased not only the in vitro b-GlcCer’ase
activity but also the in vivo Cer level of stratum cor-
neum to improve the resistance against the above skin
diseases.5,10
Recently, we have reported that b-GalCer promoted the
activity of b-glucocerebrosidase (b-GlcCer’ase) in epi-
dermis and resulted in the increase of the ceramide (Cer)
levels.5 This enzyme plays a critical role for the forma-
tion and maintenance of the epidermal permeability
barrier, by which b-glucosylceramide (b-GlcCer) local-
ized in epidermis is hydrolyzed to Cers that is essential
for the formation of lamellae structure in the stratum
corneum.6,7 Moreover, we found that acylceramide8
(Cer1), an important lipid in preventing various skin
Cer is a natural product and Cer mimics have been
applied for cosmetics. However, it is very difficult to
utilize these compounds widely. Because the general
synthesis of biologically active ceramides involves two
important aspects; one is the establishement of the
chirality and the other the construction of the trans
double bond, so a practical synthesis for active ceramides
has been still studied.11ꢀ15 Therefore, our attention has
been focused on the galactose–attached Cer mimics for
increasing the b-GlcCer’ase activity. In this study, we
designed the structural assembly using the galactose and
the Cer mimic lead from l-serine, in view of the struc-
tural diversity, for example we have applied easily the
various fatty amines or fatty acids in the synthetic
strategy, as shown in Figure 1.
*Corresponding author. Tel.:+81-465-34-6149; fax:+81-465-34-3037;
e-mail: fukunaga@oda.cos.kanebo.co.jp
yCurrent address: Medicinal Chemistry Central Research Laboratory,
Sanwa Kagaku Kenkyusho Co., Ltd., 363 Shiosaki, Hokusei-cho,
Inabe-gun, Mie, 511-0406, Japan.
Here, we describe the synthesis of b-GalCer analogues
6a–6e as the general method. As shown in Scheme 1,
b-d-galactose penta acetate 1 was derived into the
{Current address: Manufacturing Technology R&D Laboratories,
Shionogi & Co., Ltd., 1-3 Kuise Terajima 2-chome, Amagasaki,
Hyogo 660-0813, Japan.
0960-894X/03/$ - see front matter # 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0960-894X(03)00006-4

814
K. Fukunaga et al. / Bioorg. Med. Chem. Lett. 13 (2003) 813–815
Figure 1.
Scheme 1. Reagents and condition: (i) PhSH, BF3OEt2, CHCl3 (85%); (ii) NaOMe, MeOH (quant); (iii) BzCl, Pyr. (quant); (iv) R1NH2, WSC,
HOBt, DMF (quant); (v) NIS, TfOH, CHCl3 (75%); (vi) Pd(OH)2, H2, dioxane–MeOH (quant); (vii) R2COOH, WSC, HOBt, DMF (quant); (viii)
NaOMe, MeOH (quant).
thioglycoside donor having better reactivity as the leav-
ing group for the accessible glycosylation reaction, fol-
lowed by the change of the protecting groups from
acetate to benzoyl groups for the prevention of the
ortho-ester formation to afford compound 2. Glycosyl-
ation16,17 of 2 and 3 in the presence of N-iodosuccini-
mide (NIS) and trifluoromethansulfonic acid (TfOH)
gave the key intermediate, b-glycoside 4 in 75% yield.
The glycosylation step to compound 4 proceeded ste-
reoselectively based on the neighboring effect of the C-2
benzoyl group of 2 to afford only b-anomer 4. Removal
of the benzyloxycarbonyl group of compound 4 with
Pd(OH)2, followed by the introduction of stearic acid
gave the precursor 5 in good yields. Finally, debenzoyla-
tion of 5 gave quantitatively the desired b-GalCer ana-
logues 6a–e.
The in vitro activity of the most active compound 6e
was examined using UVB- induced barrier disrupted
mice. As shown in Table 2, the topical application of
compound 6e significantly reduced the transepidermal
water loss (TEWL)19 compared to that GalCer. This
might be due to the difference of bioavailability including
skin permeability or biological half life.
These findings suggest that the l-serine unit modified by
alkyl amine and acylalkyl groups with a moderate chain
length can mimic the ceramide moiety of natural GalCer
to maintain similar biological activities. Compound 6e
could be useful for improving skin barrier in cosmetic
and medicinal fields. Optimization of compound 6e is in
progress.
Table 1. Effects of b-GalCer analogues 6a–e on b-GlcCer’ase activity
in cultured normal human keratinocytes
The b-GlcCer’ase activity in cultured human keratino-
cytes was determined according to our previous paper.5
First of all, to examine the structure–activity relation-
ship (SAR) regarding the length of alkyl chain on
fatty acids (R2), the effects of compounds 6a–e on the
b-GlcCer’ase activities in cultured keratinocytes were
investigated as shown in Table 1.
Compd
b-GlcCer’ase activity (n=5)
Control
GalCer
6a
6b
6c
1.00ꢁ0.06
1.65ꢁ0.17**
1.24ꢁ0.07
1.37ꢁ0.12
0.87ꢁ0.03
1.20ꢁ0.10
1.42ꢁ0.11*
6d
6e
Among compounds 6a–c, compound 6c exhibited
weaker activity than that of natural b-GalCer. It might
be due to their shorter alkyl chain of R2 moiety. Next,
we investigated the SAR regarding the alkyl chain
lengths of alkyl amines (R1). As shown in Table 1,
compound 6e (R1=C6H13) had stronger activity than
those of compounds 6a and 6d (R1=C10H21, C14H29).
Optimization of the length of R1 is under investigation.
Normal human keratinocytes ( NHKs ) were cultured with GalCer or
compounds 6a–e (5 mM) or vehicle for 4 days. After extracting enzyme
from the cells, the b-GlcCer’ase activity was measured as nmol
4-methylumbeliferone synthesized/min/mg of protein. Activity was
expressed as ratio to control. Each value represents the means of five
determinations. (*P<0.05, **P<0.001; respectively, vs control) The
Dunnett’s test was used for comparisons between control and experi-
mental groups.18

K. Fukunaga et al. / Bioorg. Med. Chem. Lett. 13 (2003) 813–815
815
Table 2. Effects of the GalCer analogue 6e on UVB-induced murine
epidermal barrier disruption
3. Bhat, S.; Spitalnik, S. L.; Gonzalez-Scarano, F.; Silberberg,
D. H. Proc. Natl. Acad. Sci. U.S.A. 1991, 88, 7131.
4. Abul-milh, M.; Paradis, S.; Dubreuil, J. D.; Jacques, M.
Infect. Immun. 1999, 67, 4983.
Compd
TEWL
5. Hara, M.; Uchida, Y.; Haratake, A.; Mimura, K.; Hama-
naka, S. J. Dermatol. Sci. 1998, 16, 111.
6. Holleran, W. M.; Takagi, Y.; Menon, G. K.; Jackson,
S. M.; Lee, J. M.; Feingold, K. R.; Elias, P. M. J. Lipid Res.
1994, 35, 905.
7. Takagi, Y.; Kriehuber, E.; Imokawa, G.; Elias, P. M.;
Holleran, W. M. J. Lipid Res. 1999, 40, 861.
8. Behne, M.; Uchida, Y.; Ski, T.; de Montellano, P. O.; Elias,
P. M.; Holleran, W. M. J. Invest. Dermatol. 2000, 114, 185.
9. Uchida, Y.; Hara, M.; Nishio, H.; Sidransky, E.; Inoue, S.;
Otsuka, F.; Suzuki, A.; Elias, P. M.; Holleran, W. M.;
Hamanaka, S. J. Lipid Res. 2000, 41, 2071.
Control (3 days)
GalCer
6e
Control (4 days)
GalCer
6e
5.55ꢁ0.68
5.24ꢁ1.48
2.01ꢁ0.44 **
6.05ꢁ0.43
3.68ꢁ0.28* (n=4)
2.19ꢁ0.39***
GalCer, compound 6e or vehicle (10 mL/cm2) was applied daily to
dosal skin of mice before irradiation and for 2 days after irradiation
(0.15 J/cm2). TEWL was expressed as ratio to control water loss
before application of each compound. Each value represents the
meansꢁSEM of four or five animals. (*P<0.05, **P<0.01,
***P<0.001, respectively, vs each control).
10. Haratake, A.; Uchida, Y.; Mimura, K.; Holleran, W. M.
J. Invest. Dermatol. 1994, 102, 537.
11. Schmidt, R. R.; Zimmerman, P. Tetrahedron Lett. 1986,
27, 481.
Acknowledgements
12. Kiso, M.; Nakamura, A.; Tomita, Y.; Hasegawa, A.
Carbohydr. Res. 1986, 158, 101.
13. Obayashi, M.; Schlosser, M. Chem. Lett. 1985, 1715.
14. Bourin, R. H.; Rapoport, H. J. Org. Chem. 1981, 46, 4394.
15. Julina, R.; Herrig, T.; Bernet, B.; Vasella, A. Helv. Chim.
Acta 1986, 69, 368.
The present work was supported by a grant for
‘Research and Development on Glycocluster Control-
ling Biomolecules’ from the New Energy and Industrial
Technology Development Organization (NEDO).
16. Konradsson, P.; Udodong, U. E.; Fraser-Reid, B. Tetra-
hedron Lett. 1990, 31, 4313.
17. Veeneman, G. H.; van Leeuwmen, S. H.; van Boom, J. H.
Tetrahedron Lett. 1990, 31, 1331.
References and Notes
1. Ledeen, R. W. J. Neurosci. Res. 1984, 12, 147.
2. Dasgupta, S.; Everhart, M. B.; Bhat, N. R.; Hogan, E. L.
Dev. Neurosci. 1997, 19, 152.
18. Dunnett, C. W. Biometrics 1964, 20, 482.
19. Graubauer, G.; Elias, P. M.; Feingold, K. R. J. Lipid Res.
1989, 30, 323.