Synthesis of fluorescent C24-ceramide: Evidence for acyl chain length dependent differences in penetration of exogenous NBD-ceramides into human skin

Article abstract of DOI:10.1016/j.bmcl.2009.10.047

Topical skin lipid supplementation may provide opportunities for controlling ceramide (Cer) deficiency in skin diseases such as atopic dermatitis or psoriasis. Here we describe the synthesis of a long-chain 7-nitrobenzo[c][1,2,5]oxadiazol-4-yl (NBD)-labeled Cer and its different penetration through human skin compared to widely used short-chain fluorescent Cer tools.

Full text of DOI:10.1016/j.bmcl.2009.10.047

Bioorganic & Medicinal Chemistry Letters 19 (2009) 6975–6977
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis of fluorescent C24-ceramide: Evidence for acyl chain length
dependent differences in penetration of exogenous NBD-ceramides
into human skin
a
b
a
c
a,
ˇ
*
´
ˇ
ˇ
ˇ
Jakub Novotny , Katerina Pospechová , Alexandr Hrabálek , Robert Cáp , Katerina Vávrová
a Centre for New Antivirals and Antineoplastics, Department of Inorganic and Organic Chemistry, Charles University in Prague, Faculty of Pharmacy, Hradec Králové, Czech Republic
b Department of Biological and Medicinal Sciences, Charles University in Prague, Faculty of Pharmacy, Hradec Králové, Czech Republic
c Clinics of Surgery, University Hospital Hradec Králové, Czech Republic
a r t i c l e i n f o
a b s t r a c t
Article history:
Topical skin lipid supplementation may provide opportunities for controlling ceramide (Cer) deficiency in
skin diseases such as atopic dermatitis or psoriasis. Here we describe the synthesis of a long-chain 7-
nitrobenzo[c][1,2,5]oxadiazol-4-yl (NBD)-labeled Cer and its different penetration through human skin
compared to widely used short-chain fluorescent Cer tools.
Received 28 August 2009
Revised 9 October 2009
Accepted 10 October 2009
Available online 15 October 2009
Ó 2009 Elsevier Ltd. All rights reserved.
Keywords:
Skin
Ceramide
Stratum corneum
Fluorescent lipids
Depletion in total ceramide (Cer) content with alteration in the
Cer pattern is a common sign for skin disorders with diminished
skin barrier function, including atopic dermatitis1 and psoriasis.2
The formulations containing physiological lipids and, in particular,
Cer supplementation can improve permeability barrier homeosta-
sis.3–5 In addition, several pseudoCer and Cer analogues were able
to recover perturbed skin barrier,6–8 for example, a simple
L-serine
derivative 14S24 showed excellent activity.9,10
The mechanisms by which exogenous Cer correct the barrier
abnormalities have been under debate for some time. In contrast
to petrolatum, which remained restricted to the stratum corneum
(SC),5 short-chain fluorescent 7-nitrobenzo[c][1,2,5]oxadiazol-4-yl
(NBD)-C6-Cer penetrated into the nucleated epidermal cells within
two hours,5 was transported to the distal Golgi apparatus and then
reprocessed.11 Such fast epidermal uptake and metabolism has
been generally accepted as the mechanism by which all topical
Cer repair the perturbed skin barrier (for recent reviews, see Refs.
12–14).
However, no such evidence exists for natural long-chain Cer,
which have an average acyl chain length of 24 carbons, as no such
long-chain fluorescent tools have been commercially available.
Although short-chain Cer have been widely used as soluble and
easier-to-handle Cer mimics, many substantial differences
Scheme 1. Synthesis of NBD-labeled Cer and pseudoCer 14S24. Reagents: (i) 48%
HBr, toluene; (ii) CrO3, AcOH, H2O, acetone; (iii) DHP, PPTS, CHCl3; (iv) 3, CH2Br2,
Mg, CH3MgCl, Li2CuCl4, THF; (v) SOCl2, MeOH; (vi) HN3 in benzene, PPh3, DIAD, THF;
(vii) NBD-Cl, NaHCO3, MeOH; (viii) (1) N-hydroxysuccinimide, DCC, 4-DMAP, CHCl3;
* Corresponding author. Tel.: +420 495 067 497; fax: +420 495 067 166.
E-mail address: katerina.vavrova@faf.cuni.cz (K. Vávrová).
(2) sphingosine, CHCl3; (ix) L-Ser-O-C14H29, WSC, DMAP, CHCl3.
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.10.047

´
J. Novotny et al. / Bioorg. Med. Chem. Lett. 19 (2009) 6975–6977
6976
between naturally occurring long-chain Cer and their short-chain
analogues have been described recently regarding their role in
both cell signaling15,16 and skin permeability.17
Thus, we hypothesized that the Cer penetration may be chain
length dependent and that the previously observed fast (within
2 h) uptake by nucleated epidermal layers5 may be specific only
for the short-chain Cer. In order to test this hypothesis, we aimed
to synthesize a series of fluorescent Cer analogues with different
acyl chain length, namely 6, 12 and 24 carbons, and to evaluate
their penetration into both disrupted and intact human skin. Fur-
tetrachlorocuprate20 afforded acid 4 in 58% yield. Its hydroxyl
was deprotected and the carboxylic group was esterified to yield
methyl ester 5 in one step. The formal one-pot conversion of hy-
droxyl to primary amino group was achieved by combination of
Mitsunobu reaction with Staudinger reduction of the azido
group.21 The resulting hydrochloride of 24-aminotetracosanoic
acid 6 was alkalized and labeled with NBD-Cl to yield 7c. The com-
mercially available
x-amino hexanoic and dodecanoic acids were
labeled in the same manner to yield 7a–b. Sphingosine was syn-
thesized by alkynylation of protected L-serinal as described previ-
thermore, we aimed at comparing the penetration of NBD-C24
-
ously.17 Then, it was N-acylated by succinimidyl esters of the
NBD acids. NBD-labeled pseudoCer 14S24 was prepared by N-acyl-
Cer with 14S24, that is, a non-physiological pseudoCer, which
has different polar head structure but the same chain length.
For the preparation of the fluorescent long-chain NBD-C24-Cer,
24-aminotetracosanoic acid was synthesized first (Scheme 1).
Mono-bromation of dodecan-1,2-diol was achieved under micro-
wave activation in a much shorter time compared to previous
method.18 The bromo alcohol 1 was oxidized to give acid 2.19 Hy-
droxyl group of 1 was protected as THP ether 3, and converted to a
Grignard reagent. Coupling of the chloromagnesium salt of 2 with
the Grignard reagent in the presence of freshly prepared lithium
ation of tetradecyl L
-serine10 with 7c using carbodiimide coupling.
The fluorescent lipids were applied on viable human skin, intact
or acetone-treated, maintained in Dulbecco’s Modified Eagle’s Med-
ium at air/liquid interface. The first penetration experiment was
realized under the same conditions as described previously for the
short-chain Cer,5 that is, 10
lM solution of the lipids in propylene
glycol/ethanol 7:3 (v/v) were applied on the skin for 2 h. The skin
sections, counterstained with Hoechst dye, were then examined
under a fluorescent microscope (for experimental details, see
Figure 1. Penetration of exogenous fluorescent Cer with different chain length and pseudoCer 14S24 into human skin. Cer (green fluorescence) were applied on the ex vivo
viable human skin in an equimolar mixture with cholesterol and lignoceric acid in propylene glycol/ethanol 7:3 vehicle at a total lipid concentration 1% for 12 h. Histological
sections are shown using brightfield microscopy with H&E staining (panel a), fluorescence microscopy with nuclei of cells (blue) stained with the Hoechst 33258 dye (b–f).
(a) and (b) controls, (c) NBD-C6-Cer, (d) NBD-C12-Cer, (e) NBD-C24-Cer, (f) NBD-14S24. Only acetone-treated skin is shown, as no significant differences in Cer penetration
were found between intact and acetone-treated skin despite significantly higher transepidermal water loss in the latter. Scale bar 50 lm.

´
6977
J. Novotny et al. / Bioorg. Med. Chem. Lett. 19 (2009) 6975–6977
Supplementary data). However, neither long nor short-chain NBD-
Cer penetrated into nucleated epidermal layers (Supplementary
data, Fig. S1). The reason for this discrepancy in NBD-C6-Cer penetra-
tion between Man et al.5 and our study may be the different barrier
properties of mice and human skin. It has been repeatedly shown
that some compounds penetrate in an almost similar manner, others
differ in at least one logarithmic order, the human skin being the less
permeable.22 Human skin has, for example, thicker stratum cor-
neum23 and responds differently to various substances and treat-
ments.24 Moreover, other studies using NBD-C6-Cer also gave
contradictory results suggesting that this effect may be vehicle-
and concentration-dependent. For example, no uptake of NBD-C6-
Cer into hairless mice epidermis from petrolatum was found.25 In
contrast, when using a complex lipid vehicle for 2 h25 and 1.7% solu-
tion in dimethyl sulfoxide, which is actually a potent permeation en-
hancer, for 4 h,26 respectively, the fluorescence was visible in the
viable epidermis.
tration of exogenous NBD-Cer across SC. Although short-chain Cer
are useful experimental tools in many sphingolipid studies, they
cannot be used as general Cer mimics, particularly when considering
Cer effects in lipid membranes, including SC lipid lamellae. Thus, the
mechanism ofaction ofCer inskinbarrierrepair, whichwasbasedon
a behavior of a short-chain Cer, should be reconsidered.
Acknowledgements
This work was supported by the Centre for New Antivirals and
Antineoplastics (1M0508) and the Ministry of Education of the
Czech Republic (MSM0021620822).
Supplementary data
Supplementary data (experimental procedures and Fig. S1)
associated with this article can be found, in the online version, at
doi:10.1016/j.bmcl.2009.10.047.
In order to reproduce the above mentioned in vivo results5 and
to be able to distinguish between Cer with various acyl chain
length, the tissue was exposed to approximately three orders-of-
magnitude higher concentration of the lipids for 12 h (Fig. 1, see
Supplementary data for details).
References and notes
1. Imokawa, G.; Abe, A.; Jin, K.; Higaki, Y.; Kawashima, M.; Hidano, A. J. Invest.
Dermatol. 1991, 96, 523.
2. Motta, S.; Monti, M.; Sesana, S.; Mellesi, L.; Ghidoni, R.; Caputo, R. Arch.
Dermatol. 1994, 130, 452.
3. Chamlin, S. L.; Frieden, I. J.; Fowler, A.; Williams, M.; Kao, J.; Sheu, M.; Elias, P.
M. Arch. Dermatol. 2001, 137, 1110.
4. Chamlin, S. L.; Kao, J.; Frieden, I. J.; Sheu, M. Y.; Fowler, A. J.; Fluhr, J. W.;
Williams, M. L.; Elias, P. M. J. Am. Acad. Dermatol. 2002, 47, 198.
5. Man, M. Q.; Feingold, K. R.; Elias, P. M. Arch. Dermatol. 1993, 129, 728.
6. Matsuki, H.; Kiyokane, K.; Matsuki, T.; Sato, S.; Imokawa, G. Exog. Dermatol.
2004, 3, 293.
7. Imokawa, G.; Yada, Y.; Higuchi, K.; Okuda, M.; Ohashi, Y.; Kawamata, A. J. Clin.
Invest. 1994, 94, 89.
8. Takagi, Y.; Nakagawa, H.; Higuchi, K.; Imokawa, G. Dermatology 2005, 211, 128.
9. Vavrova, K.; Hrabalek, A.; Mac-Mary, S.; Humbert, P.; Muret, P. Br. J. Dermatol.
2007, 157, 704.
10. Vavrova, K.; Zbytovska, J.; Palat, K.; Holas, T.; Klimentova, J.; Hrabalek, A.;
Dolezal, P. Eur. J. Pharm. Sci. 2004, 21, 581.
11. Mao-Qiang, M.; Brown, B. E.; Wu-Pong, S.; Feingold, K. R.; Elias, P. M. Arch.
Dermatol. 1995, 131, 809.
12. Elias, P. M. J. Invest. Dermatol. 2005, 125, 183.
13. Feingold, K. R. J. Lipid Res. 2007, 48, 2531.
14. Loden, M. J. Eur. Acad. Dermatol. Venereol. 2005, 19, 672.
15. Goni, F. M.; Alonso, A. Biochim. Biophys. Acta 2006, 1758, 1902.
16. van Blitterswijk, W. J.; van der Luit, A. H.; Veldman, R. J.; Verheij, M.; Borst, J.
Biochem. J. 2003, 369, 199.
17. Novotny, J.; Janusova, B.; Novotny, M.; Hrabalek, A.; Vavrova, K. Skin Pharmacol.
Physiol. 2009, 22, 22.
18. Chong, J. M.; Heuft, M. A.; Rabbat, P. J. Org. Chem. 2000, 65, 5837.
19. Klimentova, J.; Kosak, P.; Vavrova, K.; Holas, T.; Hrabalek, A. Bioorg. Med. Chem.
2006, 14, 7681.
20. Müller, S.; Schmidt, R. R. J. Prakt. Chem. 2000, 342, 779.
21. Fabiano, E.; Golding, B. T.; Sadeghi, M. M. Synthesis 1987, 190.
22. Simon, G. A.; Maibach, H. I. Skin Pharmacol. Appl. Skin Physiol. 1998, 11, 80.
23. Bronaugh, R. L.; Stewart, R. F.; Congdon, E. R. Toxicol. Appl. Pharmacol. 1982, 62,
481.
Under these conditions, fluorescence of the short-chain NBD-
C6-Cer was observed within the viable epidermis in both intact
and acetone-treated skin, which is fully in accordance with Man
et al.5 On the contrary, both lipids with 24 carbons acyl chain
length, that is, NBD-C24-Cer and NBD-14S24, stayed in several
upper SC layers. One experiment was also performed on a non-via-
ble skin to exclude the possibility of Cer metabolism and it yielded
essentially the same results. This suggests that the long-chain lip-
ids (a) display much delayed kinetics compared to NBD-C6-Cer or
(b) they may not be able to reach nucleated epidermis at all.
The reason for the observed chain length dependence in Cer
penetration may be the difference in partitioning and mobility of
the short- and long-chain Cer. The better penetration of shorter
Cer through the lipid membranes may be explained by their smal-
ler molecule and faster exchange between the lipid bilayers. For
example, it has been found that the exchange of natural Cer be-
tween lipid membranes requires days,27 while it takes less than a
minute for short chain, fluorescent Cer to exchange between lipid
vesicles.28 An easier transbilayer movement of C6-Cer compared
to C16-Cer was observed in phospholipid membranes.29 Short-
chain Cer were also found to increase skin permeability.17
Moreover, the long-chain Cer may be too hydrophobic to parti-
tion from SC into a hydrophilic viable epidermis. Indeed, free nat-
ural long-chain Cer cannot exist in solution in biological fluids or in
cytosol. Long-chain Cer stay relatively tightly bound to the mem-
brane where they are generated,30,31 while short Cer can leave
the membrane and translocate to other membranes. Another
example is an elegant study comparing the effects of short- and
long-chain Cer that showed that although exogenous NBD-C6-Cer
taken up via the plasma membrane was converted into glucosylCer
in the Golgi (the same was suggested to occur with exogenous Cer
in the skin11), long-chain Cer generated in the plasma membrane
did not reach the Golgi and thus were not glycosylated.32
24. Bond, J. R.; Barry, B. W. J. Invest. Dermatol. 1988, 90, 810.
25. Silvander, M.; Ringstad, L.; Ghadially, R.; Skold, T. Lipids Health Dis. 2006, 5, 12.
26. Yatvin, M. B.; Stowell, M. H. B. U.S. 6387,876 B1, 2002.
27. Simon, C. G., Jr.; Holloway, P. W.; Gear, A. R. Biochemistry 1999, 38, 14676.
28. Pagano, R. E.; Martin, O. C. Biochemistry 1988, 27, 4439.
29. Shabbits, J. A.; Mayer, L. D. Biochim. Biophys. Acta 2003, 1612, 98.
30. Venkataraman, K.; Futerman, A. H. Trends Cell. Biol. 2000, 10, 408.
31. Chatelut, M.; Leruth, M.; Harzer, K.; Dagan, A.; Marchesini, S.; Gatt, S.; Salvayre,
R.; Courtoy, P.; Levade, T. FEBS Lett. 1998, 426, 102.
32. Tepper, A. D.; Diks, S. H.; van Blitterswijk, W. J.; Borst, J. J. Biol. Chem. 2000, 275,
34810.
33. Kuerschner, L.; Ejsing, C. S.; Ekroos, K.; Shevchenko, A.; Anderson, K. I.; Thiele,
C. Nat. Methods 2005, 2, 39.
However, it should be kept in mind that these initial results were
obtained using NBD-labeled lipids and should be confirmed using
more physiological fluorescent Cer, for example, with polyene fluo-
rophores, or radiolabeled lipids (for an excellent comparison and
discussion on these Cer tools, see Refs. 33,34). Nevertheless, these
results clearly showed chain length dependent differences in pene-
34. van Meer, G.; Liskamp, R. M. Nat. Methods 2005, 2, 14.