L-serine and glycine based ceramide analogues as transdermal permeation enhancers: Polar head size and hydrogen bonding

Article abstract of DOI:10.1016/S0960-894X(03)00409-8

Novel transdermal permeation enhancers related to stratum corneum ceramides were investigated. The synthesis of a series of simple compounds based on two selected amino acids, L-serine and glycine, and their enhancement activities are reported. Glycine derivative 3 enhanced the permeation of theophylline through human skin in vitro 12.5±0.5 times. The relationships between properties of the polar head of the compounds and their activity are discussed.

Full text of DOI:10.1016/S0960-894X(03)00409-8

Bioorganic& Medicinal Chemistry Letters 13 (2003) 2351–2353
L-Serine and Glycine Based Ceramide Analogues as Transdermal
Permeation Enhancers: Polar Head Size and Hydrogen Bonding
Katerina Vavrova,a,* Alexandr Hrabalek,a Pavel Dolezal,b
Tomas Holasa and Jarmila Zbytovskab
a
´
Department of Inorganic and Organic Chemistry, Faculty of Pharmacy, Charles University, Heyrovskeho 1203,
´
´
500 05 Hradec Kralove, Czech Republic
Department of Pharmaceutical Technology, Faculty of Pharmacy, Charles University, Heyrovskeho 1203,
´ ´
500 05 Hradec Kralove, Czech Republic
b
´
Received 10 March 2003; revised 28 March 2003; accepted 18 April 2003
Abstract—Novel transdermal permeation enhancers related to stratum corneum ceramides were investigated. The synthesis of a
series of simple compounds based on two selected amino acids, l-serine and glycine, and their enhancement activities are reported.
Glycine derivative 3 enhanced the permeation of theophylline through human skin in vitro 12.5Æ0.5 times. The relationships
between properties of the polar head of the compounds and their activity are discussed.
# 2003 Elsevier Science Ltd. All rights reserved.
The advantages of transdermal drug delivery are well
documented and include the avoidance of first-pass
metabolism by the liver, reduced fluctuations in plasma
concentration, reduction of side effects, extended dura-
tion of activity, and convenient termination of drug
administration.1 The major limitation of the transder-
mal drug delivery is the outermost layer of the skin, the
stratum corneum. The effective barrier function of the
stratum corneum has been attributed to a specific lipid
lamellae in the intercellular spaces.2 The lamellae are
composed mainly of ceramides, free fatty acids, choles-
terol, cholesteryl esters, and cholesteryl sulphate. Cer-
amides (Fig. 1) are considered to be the key molecules in
the lipid lamellar organization and the resistance to
environmental changes, chemical and physical stress.3
The ceramides comprise about 50% of the stratum cor-
neum lipids and are structurally heterogenous.4 The
common feature is a relatively small polar head and two
long, straight, saturated hydrophobicchains. The polar
head contains two to four hydroxyl groups and the
amide group. This enables ceramide to form strong
hydrogen bonding network5 necessary for the stability
of a lamellar structure.6
The use of permeation enhancers is one of the approaches
to facilitate drug diffusion through the skin. Enhancers
can interact with the stratum corneum components and
reversibly reduce skin barrier properties. Due to the
structure of the stratum corneum—continuity of the lipid
barrier—interaction with stratum corneum intercellular
lipids is of crucial importance for the effectiveness of per-
meation enhancer action. Because of their structure, many
permeation enhancers are capable of inserting themselves
between the hydrophobictails of the ecramide bilayer,
thus disturbing their packing, increasing their fluidity and
subsequently leading to easier diffusion of penetrants.7,8
There is a hypothesis that a certain similarity should
exist between an enhancer molecule and ceramides.9
However it is not clear where are the limits of the simi-
larity. In other words what features of the ceramide
molecule are necessary for maintaining the barrier
function and, on the other hand, what change in the
ceramide molecule will lead to a structure with permea-
tion enhancing properties.
*Corresponding author. Tel.: +42-49-5067497; fax: +42-49-5514330;
e-mail: vavrovak@faf.cuni.cz
Figure 1. Ceramide 2.
0960-894X/03/$ - see front matter # 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0960-894X(03)00409-8

´ ´
K. Vavrova et al. / Bioorg. Med. Chem. Lett. 13 (2003) 2351–2353
2352
The aim of this study was to synthesize a series of ceramide
analogues with different polar head size and structure and
to test their skin permeation enhancing properties in vitro.
Furthermore, we wanted to assess the structure–activity
relationships to obtain more information about the impor-
tance of the similarity of these enhancers to ceramides.
Structures of the analogues were based upon the knowledge
of ceramide biosynthesis (originating from l-serine).10 In
general, the substances consist of a polar head and two
hydrophobic chains. The polar head contains amino acids
or a hydroxy acid; and the hydrophobic chains are unbran-
ched, both saturated and unsaturated, 11–17 carbons long.
reacted with amino acid esters 1, 2 via the mixed anhy-
dride method to yield 13, 14. The final products 15, 16
were prepared by oxidizing the double bond with
KMnO4 in the presence of a crown-ether.15
The compounds were tested in vitro using the modified
Franz diffusion cell; theophylline was chosen as a model
penetrant of medium polarity. Samples of excided
human cadaver skin (300 mm thick, samples from iden-
tical areas on a thigh) were used. Donor samples (200
mL, occluded conditions) were prepared as 5% w/w
suspensions of theophylline in a given vehicle (infinite
dose) with 1% w/w of the suspended, partly dissolved
enhancer (except control samples). Azone1 (N-dodecyl-
azepan-2-one)16 was also tested under the same condi-
tions to serve as a standard. Theophylline in the
acceptor phase samples was determined by HPLC. For
details of the permeation experiments and HPLC deter-
mination, see ref 17.
The synthetic schemes are depicted in Figure. 2.11 The
first series of the analogues is the most similar to natural
ceramides. The polar head is formed by one amino acid,
serine or glycine. Primarily esters of l-serine 2 (yield 85%)
and glycine 1 (yield 91%) were prepared via reaction with
the pertinent alcohol and dry HCl under nitrogen.12
Amino acid esters then reacted with N-hydroxy-
succinimide ester (R1COO-SI) to obtain compounds 3–6.
Lauric acid and oleic acid were converted into R1COO-SI
with N-hydroxysuccinimide and dicyclohexyl carbodi-
imide in ethyl acetate in yields of 73 and 76%.
Cumulative amounts of theophylline (mg/cm2) corrected
for acceptor sample replacement were plotted against time
(h) and fluxes (mg/cm2/h) of theophylline through human
skin were calculated. All the tested compounds are
characterized by the enhancement ratio (ER, ratio of the
means of the pertinent flux values and the control ones).
Statistical analysis was performed using Student’s t-test.
N-Lauroyloxysuccinimide was also used for the pre-
paration of N-acylated amino acids 7,813 (90–93%),
which then reacted with appropriate amino acid esters
1,2 to form ceramide analogues of the second series with
a polar head formed by two amino acids 9–11.
Results and Discussion
In the third series of the analogues, maleicacid mono-
ester 12 (81%) was prepared first.14 The monoester then
After the preliminary tests (data not shown) water and
isopropyl myristate (IPM) were chosen as the repre-
sentative hydrophilicand lipophilicvehilces. In the
present study, we have chosen analogues with the same
chain length (12C) to evaluate the characteristics of the
polar head of the compounds. An oleic acid derivative
was also tested, because the presence of cis double bond
in the centre of the hydrophobic chain proved itself to
disrupt ceramide lamellar organization more than a
saturated chain.18
The ERs of the compounds from water and IPM sus-
pensions are summarized in Table 1.
Surprisingly, the most potent enhancer was compound
3, the simpliest one, having glycine as the polar head.
Table 1. Enhancement ratios of the tested compounds in theophyl-
line permeation through human skin from its aqueous and IPM sus-
pensions
Compd
ER (water)ÆSD
ER (IPM)ÆSD
Control
3
5
6
9
11
1.00Æ0.15
12.50Æ0.50**,++
2.74Æ0.71
1.00Æ0.08
1.11Æ0.16
1.21Æ0.16*,+
1.49Æ0.27+
1.09Æ0.01
1.07Æ0.02
1.45Æ0.34+
0.60Æ0.36
1.26Æ0.25
2.28Æ0.64
1.26Æ0.06
16
Azone1
1.39Æ0.18*
2.50Æ0.93
Figure 2. Synthesis of ceramide analogues: (i) R1COO-SI, CHCl3, rt,
24 h (55–82%); (ii) (1) TEA, THF, À20 ꢀC, 15 min; (2) ClCOOEt,
À20 ꢀC, 15 min, (3) 1(2), rt, 2 h (33–42% for 9–11, 78 and 91% for 14
and 13, resp); (iii) (1) KMnO4, DH-18–C-6, CH2Cl2, À50 ꢀC, 3 h; (2)
Na2SO4; (3) H2SO4 (22–26%).
n=3–4.
*p<0.05 versus control; **p<0.001 versus control; +p<0.05 versus
Azone; ++p<0.001 versus Azone.

´ ´
K. Vavrova et al. / Bioorg. Med. Chem. Lett. 13 (2003) 2351–2353
2353
The compound was five times more active than Azone1
Acknowledgements
under the same conditions. The presence of the hydroxy-
methyl group in compound 5, an l-serine derivative,
decreased the permeation enhancing activity sig-
nificantly. Compound 5 has stronger ability to form
H-bonds compared to 3. Since H-bonding is essential for
the dense and rigid ceramide arrangement in the stratum
corneum intercellular spaces, this finding can be related to
the mechanism of action of this type of enhancers. The
importance of H-bonding ability is also well documented
in the process of permeation through the skin.19
This work was supported by the Charles University
(grant 256/2001/B-CH/FaF) and the Ministry of Edu-
cation of the Czech Republic (grant MSM 111600001).
References and Notes
1. Buyuktimkin, N.; Buyuktimkin, S.; Rytting, J. H. In
Transdermal and Topical Drug Delivery Systems; Ghosh, T.
K., Pfister, W. R., Eds.; Interpharm: Buffalo Grove, IL, 1997;
pp 357–476.
2. Wertz, P. W.; van den Bergh, B. Chem. Phys. Lipids 1998,
91, 85.
3. Elias, P. M. J. Invest. Dermatol. 1983, 80, 44.
4. Wertz, P. W. Adv. Drug Deliver. Rev. 1996, 18, 283.
5. Abraham, W.; Downing, D. T. Pharm. Res. 1992, 9, 1415.
6. Kim, D. H.; Oh, S. G.; Lee, Y. J.; Kim, Y. J.; Kim, H. K.;
Kang, H. H. Colloids Surf. A: Physicochem. Eng. Asp. 2002,
207, 177.
Compound 6, which has an oleyl residue instead of the
lauryl one in compound 5, shown substantially lower
activity. The conformational kink in the oleyl chain due to
the presence of the cis double bond, often reported as the
cause of high enhancing activity of oleic acid and its deri-
vatives, does not seem to contribute significantly to the
activity of the present ceramide analogues with oleyl chain.
Compounds 9 and 11, the dipeptide derivatives, exhib-
ited lower activity compared to 3 and 5. The possible
explanation is that the bigger the polar head, the lower
the ability of the compounds to incorporate themselves
between the hydrophobic chains of ceramides in the
stratum corneum. Similarly to the difference between 3
and 5, compound 9 showed higher ER than 11, which
has two additional H-bonding groups.
7. Barry, B. W. In Pharmacology of the Skin; Shroot, B.,
Schaefer, H., Eds; Karger: Basel, 1987; p 121.
8. Marjukka Suhonen, T.; Bouwstra, J. A.; Urtti, A. J. Con-
trol. Release 1999, 59, 149.
9. Hrabalek, A.; Dolezal, P.; Palat, K. In Perspectives of Per-
cutaneous Penetration 7A: Brain, K. R.; James V. J.; Walters,
K. A., Eds; STS: Cardiff, 2000; p 70.
10. Huwiler, A.; Kolter, T.; Pfeilschifter, J.; Sandhoff, K.
Biochim. Biophys. Acta 2000, 1485, 63.
11. All compounds were fully characterized by spectral meth-
ods, and their purity checked by elemental analysis. Repre-
sentative data of the most potent derivative; compound 3: mp
81–82 ꢀC; IR (KBr) nmax 3333, 2918, 2849, 1740, 1641, 1548,
1244 cmÀ1; 1H NMR (300 MHz, CDCl3): d 5.97 (1H; bs; NH);
4.14 (2H; t; J=6.9 Hz; COO–CH2); 4.03 (2H; d; J=4.9 Hz;
CH2NH); 2.23 (2H; t; J=7.6 Hz; COCH2); 1.70–1.55 (4H; m;
2 CH2); 1.40–1.15 (34H; m; 17 CH2); 0.87 (6H; t; J=6.6 Hz; 2
CH3); 13C NMR (75 MHz, CDCl3) d 173.2; 170.2; 65.7; 41.3;
36.4; 31.9; 29.6; 29.5; 29.5; 29.4; 29.3; 29.2; 28.5; 25.8; 25.6;
22.7; 14.1. Anal. calcd for C26H51NO3: C, 73.36; H, 12,08; N,
3,29. Found: C, 72,98; H, 12,41; N, 3,49.
12. Limanov, V. E.; Svitova, I. R.; Kruchenok, T. B.; Tsvir-
ova, I. M.; Yaroslavskaya, L. A. Pharm. Chem. J. 1984, 18,
708.
13. Lapidot, Y.; Rappoport, S.; Wolman, Y. J. Lip. Res.
1967, 8, 142.
A member of the third series, compound 16 has also low
ER. This corresponds to the previously suggested rela-
tionships, as 16 has a large polar head and three
hydroxyl groups.
The compounds were not active when in the IPM sus-
pension (the reason was not estimated further).
From the previous studies it is only known that an
enhancer molecule must have a polar head and a
hydrophobic chain to ensure interaction with ceramides.
The optimal chain length is 10–12 carbons. We hypo-
thesize that a relatively small polar head of the ceramide
analogues (similarly to ceramides) with a limited ability
to act as a H-bond donor (in contrast to ceramides) is
important for the enhancing activity. Further evaluation
of this hypothesis is in progress.
14. Huber, Lutton J. Amer. Chem. Soc. 1957, 79, 3919.
15. Mukaiyama, T.; Tabusa, F.; Suzuki, K. Chem. Lett. 1983,
173.
16. Stoughton, R. B. Arch. Dermatol. 1982, 118, 474.
17. Vavrova, K.; Hrabalek, A.; Dolezal, P. Arch. Dermatol.
Res. 2002, 294, 383.
18. Oh, H.-J.; Oh, Y.-K.; Kim, C.-K. Int. J. Pharm. 2001, 212,
63.
19. Pugh, W. J.; Roberts, M. S.; Hadgraft, J. Int. J. Pharm.
1996, 138, 149.
20. Vavrova, K.; Hrabalek, A.; Dolezal, P.; Holas, T. CZ
Patent PV 2003-540, 2003.
Apart from these theoretical findings, ceramide analo-
gue 3 derived from glycine seems to be a promising
permeation enhancer,20 because it has a high enhance-
ment ratio (12.50) at low concentration together with
expected low toxicity. As there is a variety of enzymes in
the living epidermis, degradation of enhancer 3 into
non-toxicmetabolites after it reaches the stratum cor-
neum/stratum granulosum interface could be expected.