Investigation of new acyloxy derivatives of cholic acid and their esters as drug absorption modifiers

Article abstract of DOI:10.1016/j.steroids.2011.04.014

Skin penetration enhancers are used in the formulation of transdermal delivery systems for drugs that are otherwise not sufficiently skin-permeable. Intestinal absorption promoters/enhancers are used as excipients in oral formulations of poorly oral-bioav

Full text of DOI:10.1016/j.steroids.2011.04.014

Steroids 76 (2011) 1082–1097
Contents lists available at ScienceDirect
Steroids
journal homepage: www.elsevier.com/locate/steroids
Investigation of new acyloxy derivatives of cholic acid and their esters as drug
absorption modifiers
a
b
b
c
b
d
ˇ
Lech Mrózek , Lenka Dvorˇáková , Zuzana Mandelová , Lucie Rárová , Anna Rezácˇová , Lukásˇ Placˇek ,
Radka Opatrˇilováb, Jirˇí Dohnalb, Oldrˇich Paletae, Vladimír Krále, Pavel Drasˇare, Josef Jampílekb,∗
a BorsodChem MCHZ, s.r.o., Chemicka 1/2039, 709 03 Ostrava-Marianske Hory, Czech Republic
b Faculty of Pharmacy, University of Veterinary and Pharmaceutical Sciences Brno, Palackeho 1/3, 612 42 Brno, Czech Republic
c Centre of the Region Hana for Biotechnological and Agricultural Research, Department of Growth Regulators, Faculty of Science, Palacky University, Slechtitelu 11, 783 71 Olomouc,
Czech Republic
d Pragolab s.r.o., Nad Krocinkou 55/285, 190 00 Prague 9, Czech Republic
e Institute of Chemical Technology, Technicka 5, 166 28 Prague 6, Czech Republic
a r t i c l e i n f o
a b s t r a c t
Article history:
Skin penetration enhancers are used in the formulation of transdermal delivery systems for drugs that
are otherwise not sufficiently skin-permeable. Intestinal absorption promoters/enhancers are used as
excipients in oral formulations of poorly oral-bioavailable drugs. Series of fourteen acyloxy derivatives of
5␤-cholic acid as potential drug absorption modifiers was generated by multistep synthesis. The synthesis
of all newly prepared compounds is presented here. Structure confirmation of all generated compounds
was accomplished by 1H NMR, 13C NMR, IR and MS spectroscopy methods. All the prepared compounds
were analyzed using RP-TLC, and their lipophilicity (RM) was determined. The hydrophobicity (log P)
and solubility (log S) of the studied compounds were also calculated using two commercially available
programs. All the target compounds were tested for their in vitro transdermal penetration activity and
as potential intestinal absorption enhancers. The anti-proliferative activity of all the final compounds
was also assessed against the human cancer cell lines: T-lymphoblastic leukemia cell line and the breast
adenocarcinoma cell line. Their cytotoxicity was also evaluated against the normal human skin fibro-
blast cells. Two compounds showed anti-proliferative effect on cancer cells without affecting the growth
of normal cells, which should be promising in potential development of new drugs. Most of the tar-
get compounds showed minimal anti-proliferative activity (IC50 > 37 ␮M), indicating they would have
low cytotoxicity when administered as chemical absorption modifiers. The relationships between the
lipophilicity and the chemical structure of the studied compounds as well as the relationships between
their chemical structure and enhancement effects are discussed in this article.
Received 25 January 2011
Received in revised form 12 April 2011
Accepted 21 April 2011
Available online 5 May 2011
Keywords:
Cholic acid derivatives
Lipophilicity
Solubility
Transdermal penetration enhancers
Intestinal absorption promoters
In vitro cytotoxicity
© 2011 Elsevier Inc. All rights reserved.
1. Introduction
intestinal absorption after oral administration or less commonly
transdermal absorption after local application [1]. One of possibili-
Development in the field of pharmaceutical administration has
resulted in the discovery of highly sophisticated drug delivery
systems that allow for the maintenance of a constant drug level
in an organism. Contrary to these revolution biopharmaceuti-
cal/galenical results, it is estimated that about 40% of marketed
active pharmaceutical ingredients and 60% of active new chemical
entities identified in R&D screening programs employed by many
pharmaceutical companies are poorly water soluble. These poorly
soluble “modern” drugs are distinguished by incomplete absorp-
tion and low, erratic bioavailability as a result of either mostly
ties to modify/optimize drug unfavourable physico-chemical prop-
erties is to use absorption modifiers, e.g., bile acid derivatives [2–5].
Cholic acid is one of the most important human bile acids. Bile
acid derivatives/analogues are an important class of compounds
with a range of pharmacological activities. Bile acids could be
easily modified by derivatisation of the functional groups on the
steroid nucleus. Nontoxic bile acid/salt derivatives (as amphiphilic
compounds) are used widely in drug formulations as excipients
(intestinal absorption enhancers, promoters) and can influence
gastrointestinal solubility, absorption and chemical/enzymatic sta-
bility of drugs [6–12]. Cholic acid derivatives were studied also
as transdermal penetration enhancers [13–21]. The reason for
their activity may be their specific features in solvation and self-
assembly [22–28].
∗
Corresponding author. Tel.: +420 724139484.
E-mail address: josef.jampilek@gmail.com (J. Jampílek).
0039-128X/$ – see front matter © 2011 Elsevier Inc. All rights reserved.
doi:10.1016/j.steroids.2011.04.014

L. Mrózek et al. / Steroids 76 (2011) 1082–1097
1083
OH
OH
OH
C H COOBn
C H COOBn
C H COOH
2 4
2
4
2
4
a
b
HO
OH
ROCO
OH
ROCO
OH
H
H
H
1
2,30
3,31
c
OH
ROCO
ROCO
C H COOBn
C H COOBn
C H COOH
2 4
2
4
2
4
d
e
AcO
OH
AcO
OCOR
HO
OCOR
H
H
H
4
5,32
6,33
ROCO
ROCO
C H COOBn
C H COOH
2 4
2
4
2,3,5-8 R= C H
d
b
3
7
1
30-35 R= C
H
15 31
ROCO
ROCO
OCOR
OCOR
H
H
7,34
8,35
Scheme 1. Synthesis of target mono-, bis-, tris(butanoyl) 3, 6, 8 and mono-, bis-, tris(hexadecanoyl) 31, 33, 35 substituted compounds: (a) C3H7COCl or C15H31COCl, DMAP,
CH2Cl2; (b) HCOONH4, Pd/C, MeOH; (c) Ac2O, KHCO3, toluene; (d) C3H7COCl or C15H31COCl, BTEAC, CaH2, toluene; (e) (i) NaOH, i-PrOH, (ii) AcOH, CH2Cl2.
Transdermal therapeutic systems are an excellent alternative
to conventional pharmaceutical administration forms. However,
the application of transdermal drug delivery faces the problem of
insufficient or no penetration of active pharmaceutical substances
through the skin, as the outermost layer of skin, namely the stratum
corneum (SC), forms a strong barrier for most of exogenous sub-
stances including drugs. The barrier function of the SC is attributed
to its multilayered wall-like structure, in which terminally differ-
entiated keratin-rich epidermal cells (corneocytes) are embedded
in an intercellular lipid-rich matrix [4,21].
Transdermal penetration enhancers are special pharmaceutical
excipients that interact with skin components, to increase pen-
etration of drugs to blood circulation after topical application.
Numerous compounds of different chemical structures were evalu-
ated as penetration enhancers and several possible mechanisms of
action of enhancers have been hypothesized, but exact mechanisms
have not been elucidated [21]. In spite of the extensive research in
this field, chemical penetration enhancers have not reached their
full potential in transdermal or topical systems so far. Transdermal
chemical penetration enhancers are compounds which can parti-
tion into and interact with the SC constituents when incorporated
into a transdermal formulation, thereby reducing the resistance of
the skin to drug diffusion [21].
di- and tri-O-acylation of cholic acid were chosen for clarify-
ing the correlations between the physico-chemical properties of
the absorption/penetration enhancers of bile acid-type inducing
the enhancement effect. Various acyl chain lengths as well as
different degree and position of substitution of the compounds dis-
cussed in this paper impart specific solvation and surface features
influencing structural modifications of the biological membranes.
Primary in vitro screening of transdermal penetration activity
of all the final synthesized compounds was performed using
a Franz cell [31] and intestinal absorption enhancement activ-
ity was evaluated using PAMPA (parallel artificial membrane
permeability assays) experiments [32,33]. All the discussed com-
pounds were evaluated for their anti-proliferative activity against
the T-lymphoblastic leukemia cell line and the breast adeno-
carcinoma cell line as well as also for their cytotoxicity against
normal human skin fibroblast cells. The relationships between the
lipophilicity/solubility and the chemical structure of the studied
compounds as well as the relationships between their chemical
structure and activity/enhancement effects (SAR) are discussed in
this article.
2. Result and discussion
The traditional lipophilicity parameter, log P, is a well-known
physico-chemical descriptor widely used in QSAR analysis. In some
experimental studies of penetration enhancement, the lipophilicity
(non-polarity) of enhancers was measured and the correspond-
ing relationship between enhancer lipophilicity and penetration
enhancement potency was investigated [21,29,30]. Therefore we
have examined both the experimental lipophilicity RM (RP-TLC)
data and calculated lipophilicity log P of all compounds in this arti-
cle. The solubility (polarity) log S of the discussed compounds was
calculated as well.
2.1. Chemistry
3␣-Mono(acyloxy) derivatives 3, 13, 31 were prepared by
the reaction of benzyl 3␣,7␣,12␣-trihydroxy-5␤-cholate (benzyl
cholate, 1) [34] and 4-dimethylaminopyridine (DMAP) with acyl
chlorides under mild conditions followed by hydrogenolysis of
protecting groups, see Schemes 1 and 2. 3␣-Mono(decanoyloxy)
derivative 13 was also prepared by modification of a synthetic
pathway described by Bonar-Law et al., using 3␣,7␣,12␣-
trihydroxy-5␤-cholic acid 14, in which the 12␣-hydroxyl was
protected with a trifluoroacetyl group [35], see Scheme 2. Methyl
3␣-decanoyloxy-7␣,12␣-dihydroxy-5␤-cholate (26) was obtained
by re-esterification of benzyl 3␣-decanoyloxy-7␣,12␣-dihydroxy-
5␤-cholate (9) by methanol, see Scheme 2.
The multistep synthesis of a series of fourteen acyloxy deriva-
tives of 5␤-cholic acid with C4, C10 and C16 linear acyl chains
as novel transdermal chemical penetration enhancers and/or as
intestinal drug absorption modifiers is described herein. Mono-,

1084
L. Mrózek et al. / Steroids 76 (2011) 1082–1097
OH
OH
C2H4COOMe
C2H4COOMe
C9H19OCO
C9H19OCO
OH
OCOC9H19
H
H
26
27
f
f
OH
R3OCO
C2H4COOBn
+
C2H4COOBn
R1
R2
C9H19
R3
H
a
1
10
H
11 C9H19 C9H19
H
C9H19
OCOR2
C9H19OCO
R1OCO
OH
12 C9H19
H
H
H
9
10-12
b
OH
OH
C2H4COOH
C2H4COOtBu
e
C9H19OCO
C9H19OCO
OH
OH
H
H
13
17
d
OH
CF3OCO
CF3OCO
C2H4COOH
c
C2H4COOtBu
C2H4COOtBu
a
HO
OH
HO
OH
C9H19OCO
OH
H
H
H
14
15
16
Scheme 2. Synthesis of target mono(decanoyl) substituted derivative 13 and mono- and 3␣,7␣-bis(decanoyl) substituted derivatives 26, 27: (a) C9H19COCl, DMAP, CH2Cl2;.
(b) HCOONH4, Pd/C, MeOH; (c) (CF3CO)2O, t-BuOH, THF; (d) MeOH, THF, NH3(aq); (e) CF3COOH, AcONa, CH2Cl2; (f) MeONa, MeOH, THF, AcOH.
The preparation of 7␣,12␣-bis(acyloxy) derivatives 6, 22, 33
included the protection of 3␣-hydroxy group of benzyl ester 1
by the acetyl group, according to Schwarz et al. [36], resulting
in compound 4 and subsequent acylation of both free 7␣,12␣-
dihydroxy moieties. Long-chain acyls were derived from fatty acids.
Finally, the acetyl group was selectively cleaved by solvolysis
of the protecting group. Methyl 3␣,7␣,12␣-tris(decanoyloxy)-
5␤-cholate (29) was obtained by re-esterification of benzyl
3␣,7␣,12␣-tris(decanoyloxy)-5␤-cholate (24) by methanol, see
Schemes 1 and 3.
2.2. Lipophilicity and solubility of the prepared compounds
in the presence of
a base. Methyl 7␣,12␣-bis(decanoyloxy)-
3␣-hydroxy-5␤-cholate (28) was prepared by mild hydrolysis
and re-esterification of benzyl 3␣,7␣,12␣-tris(decanoyloxy)-5␤-
cholate (24) by methanol, see Scheme 3. This method including the
protection of 3␣-hydroxy group in ester 1 by a benzyloxycarbonyl
moiety was developed for the selective preparation of 7␣,12␣-
bis(decanoyloxy)-3␣-hydroxy-5␤-cholic acid (22), see Scheme 3.
The preparation of 7␣,12␣-bis(acyloxy) substituted derivatives
was based on regioselective alcoholysis described by Wess et al.
[37].
3␣,7␣-Bis(decanoyloxy) derivatives 19, 27 were prepared by
modification of the procedures used for preparation of 3␣-
mono(decanoyloxy) derivatives 13 and 26, i.e. the protection
of 12˛-hydroxyl with trifluoroacetyl group (for compound 19)
or re-esterification of benzyl ester 12 (for compound 27), see
Schemes 2 and 3.
Different lipophilicity descriptors such as log kw, log P, log D,
etc. are used for structure–activity relations description and pre-
diction. However, the algorithms used in their calculation do not
take into account configuration specificity and sometimes even
regiospecificity of steroidal derivatives. Hence, having in mind the
importance of permeability and solubility (polarity) for biological
activity [38], this study compares calculated log P and log S values
with the related experimental parameter, the RM values of the final
derivatives, as a measure of their lipophilicity. The RM values were
determined by RP-TLC. Solubility was estimated using the software
that is applied by many industrial companies, ACD/Solubility DB.
This program calculates aqueous solubility values at any pH under
the standard conditions (and zero ionic strength). The accuracy of
calculations for simple structures is usually better than 0.2–0.5 log-
arithmic units. So, it is not derived from log P and takes into account
not only the pH (solubility as a function of pH) but compares the
fragmental estimations with experimental material from ca 6000
compounds databased. Nevertheless, it is important to note that all
log P and log S values calculated using various programs should be
3␣,7␣,12␣-Tris(acyloxy) derivatives 8, 25, 35 were pre-
pared by the reaction of benzyl cholate 1 with acyl chlorides
in the presence of benzyltriethylammonium chloride (BTEAC)
as a phase transfer catalyst with subsequent hydrogenolysis

L. Mrózek et al. / Steroids 76 (2011) 1082–1097
1085
CF OC
O
OH
C2H4COOtBu
C2H4COOH
3
a
b
15
C9H19OCO
OCOC9H19
C9H19OCO
OCOC9H19
H
H
18
19
OH
C9H19OCO
C2H4COOBn
C2H4COOBn
c
a
1
BnOOCO
OH
BnOOCO
OCOC9H19
H
H
20
21
d
C9H19OCO
C9H19OCO
C2H4COOBn
e
C2H4COOH
a
e
4
AcO
OCOC9H19
HO
OCOC9H19
H
H
23
22
C9H19OCO
C9H19OCO
C2H4COOBn
d
C2H4COOH
a
1
C9H19OCO
OCOC9H19
C9H19OCO
OCOC9H19
H
H
24
25
f
f
C9H19OCO
C9H19OCO
C2H4COOMe
C2H4COOMe
HO
OCOC9H19
C9H19OCO
OCOC9H19
H
H
28
29
Scheme 3. Synthesis of target bis-, tris(decanoyl) substituted derivatives 19, 22, 25 and 7␣,12␣-bis-, tris(decanoyl) substituted derivatives 28, 29: (a) C9H19COCl, BTEAC,
CaH2, toluene; (b) i) MeOH, THF, NH3(aq), ii) CF3COOH, AcONa, CH2Cl2; (c) BnOCOCl, pyridine; (d) HCOONH4, Pd/C, MeOH; (e) (i) NaOH, i-PrOH, (ii) AcOH, CH2Cl2; (f) MeONa,
MeOH, THF, AcOH.
understood as approximate. The results are shown in Table 1 and
illustrated in Fig. 1.
corresponds to the lipophilicity within the series of the discussed
compounds. For most of the discussed compounds a logical depen-
dence can be found: polarity/solubility (log S data) decreases with
a lipophilicity increase, see Fig. 1.
Interesting anomalies were observed for 3␣,7␣- and 7␣,12␣-
bis(acyloxy) derivatives 19/22 and 27/28: 7␣,12␣-bis(acyloxy)
substituted compounds unexpectedly expressed much lower
lipophilicity and much higher polarity than 3␣,7␣-bis(acyloxy)
substituted derivatives. They could be probably explained by dif-
ferent solvation and other types of non-covalent interactions
(especially van der Waals interactions, hydrogen bonds, dipole-
dipole interactions) of hydroxyl moieties in the C(3) and C(12)
positions of the steroidal skeleton. This fact is extremely important
As expected, compound 35 possessed the highest lipophilicity.
Contrary to all expectations and calculated log P and log S data,
compound 6 showed the lowest lipophilicity. Generally, it can be
stated that mono(acyloxy) substituted derivatives expressed the
lowest lipophilicity, except for compound 6, and tris(acyloxy) sub-
stituted compounds showed the highest lipophilicity within the
individual series. The calculated log P data and the determined
RM values correspond to the expected trend in lipophilic-
ity, increasing within the series of the evaluated compounds
(butanoyl < decanoyl < decanoyl-Me ester < hexadecanoyl deriva-
tives). This dependence is approximately linear. The RM data

1086
L. Mrózek et al. / Steroids 76 (2011) 1082–1097
Table 1
Comparison of determined RM with calculated lipophilicity (log P) and solubility (log S) values.
Compound
R1
R2
R3
R4
RM
log P (ACD/log P DB)
log S (ACD/log S DB)
3
6
8
C3H7
H
C3H7
C9H19
C9H19
H
C9H19
C9H19
C9H19
H
C9H19
C15H31
H
H
H
H
H
H
H
H
H
H
CH3
CH3
CH3
CH3
H
−0.51
−0.53
−0.34
−0.20
0.08
4.87
6.85
8.84
−2.96
−3.42
−4.71
−4.21
−6.42
−6.42
−8.55
−7.05
−9.18
−9.18
−11.22
−5.76
−8.70
−10.74
C3H7
C3H7
H
C9H19
C9H19
C9H19
H
C9H19
C9H19
C9H19
H
C15H31
C15H31
C3H7
C3H7
H
13
19
22
25
26
27
28
29
31
33
35
7.92
H
12.97
12.95
18.01
8.52
13.56
13.56
18.61
10.98
19.08
27.18
C9H19
C9H19
H
−0.12
0.44
−0.04
0.29
0.10
0.76
0.15
H
C9H19
C9H19
H
C15H31
C15H31
H
H
0.52
1.67
C15H31
Cholic acid
−
−
−
−
−0.55
2.88
−1.39
28.0
22.0
16.0
10.0
4.0
2.3. In vitro screening of transdermal penetration-enhancing
activity
R = 0.9592
The penetration enhancement activity of the prepared com-
pounds was evaluated using theophylline as a model penetrant and
propylene glycol/water 1:1 (v/v) as a donor vehicle. Theophylline
was used as a model drug of medium polarity (log P −0.06; log D8
−0.05) [39,40], as it has been extensively studied in transdermal
penetration experiments [41,42]. Most of the studies involved the
use of propylene glycol (PG) or its mixture with water or ethanol as
a donor vehicle. Previous studies have indicated that PG by itself (or
a PG/water co-solvent system) does not interfere with membranes,
but rather exhibits a synergistic effect in combination with other
penetration enhancers [43–45]. Porcine ear skin was selected for
initial evaluation of enhancement activity of prepared compounds
as this tissue is a suitable in vitro model of human skin [46,47].
Porcine skin has been shown to be histologically and biochemically
similar to human skin, therefore full-thickness pig ear skin has been
used in numerous percutaneous absorption studies [48]. Neverthe-
less, for testing of hydrophobic penetrants, dermatomed skin has
been recommended [49]. The skin permeation experiments were
performed using static Franz diffusion cells [31].
-2.0
-8.0
R = 0.8555
1.2
-14.0
-0.6
0.0
0.6
RM
1.8
Fig. 1. Comparison of log P and log S values (lipo/hydrophilicity properties) com-
puted using two programs with the determined RM values.
for the absolutely different enhancer activity of these bis(acyloxy)
substituted derivatives, see below. Some differences between the
determined lipophilicity and predicted lipophilicity/solubility data
were also observed for compounds 26, 31 and 29. All these unex-
pected differences influence penetration activity.
Generally, it could be concluded that the prediction power of
using either experimental RM or calculated log S or log P for extrap-
olation of transport modifications may be a good tool for searching
potential transdermal penetration modifiers.
As it could be expected, the presented transdermal penetra-
tion enhancement results of the target compounds and the data
obtained using PAMPA method (see below) are different (see
Table 2, where the enhancement ratios (ERs) are presented). The
principal difference is probably due to different properties of
the transport “membrane”; the PAMPA diaphragm is an artificial
model of cell lipid bilayer “simulating” intestine wall, whereas the
Table 2
ERs of the prepared target compounds and in vitro anti-proliferative activity/cytotoxicity of tested compounds on normal and cancer cell lines. IC50 (␮mol/L) assessed by
Calcein AM assay of surviving cells. ERs data for skin and anti-proliferative activity of the compounds are expressed as mean SD (n = 3 experiments), ERs data for PAMPA
are expressed as mean SD (n = 4 experiments). Means followed by different letters are significantly different at P = 0.05.
Compound
ERs
Cell lines IC50 (␮mol/L)
Skin
PAMPA
CEM
MCF7
BJ
3
6
8
1.67 0.10fg
1.57 0.11ef
1.67 0.08fg
2.13 0.13i
1.30 0.13cd
1.24 0.14bc
2.09 0.14i
1.83 0.12gh
1.08 0.12ab
0.92 0.14a
1.45 0.12de
2.01 0.11hi
1.96 0.11hi
2.01 0.11hi
1.08 0.04g
1.18 0.04h
1.16 0.04h
0.35 0.04a
0.60 0.05c
1.03 0.05g
0.73 0.05d
0.73 0.05d
0.95 0.04f
1.21 0.04h
0.93 0.05f
0.52 0.04b
0.86 0.04e
0.80 0.04e
>37
>37
>37
>37
>37
>37
>37
>37
>37
>37
33.0 0.8c
27.3 1.9b
>37
13
19
22
25
26
27
28
29
31
33
35
34.6 0.9b
>37
16.4 4.3a
>37
21.8 0.5a
>37
14.8 2.1
>37
>37
>37
>37
>37
>37
>37
>37
>37
>37
>37
>37
>37
>37
>37
>37
25.7 3.8b
>37
>37
30.4 2.5b
>37
>37

L. Mrózek et al. / Steroids 76 (2011) 1082–1097
1087
Fig. 2. ERs through porcine ear skin and PAMPA of the prepared target compounds. Control experiments used theophylline in the donor vehicle without any enhancer. ERs
data for skin were calculated from 3 experiments, ERs data for PAMPA were calculated from 4 experiments.
ear porcine skin is a much more complex naturally constructed
barrier.
P = 0.001, the ER values of compounds 3, 6, 8, 19, 22 and 29 at P = 0.01
and the ER value of compound 26 at P = 0.05.
The effect of target cholic acid derivatives on penetration
of theophylline through the porcine skin is presented in Fig. 2
(hatched columns). Control experiments were run with only theo-
phylline in the donor vehicle in the absence of any enhancer. The
used system PG/water ensured entire solubility of theophylline,
which is crucial for evaluation of penetration of theophylline
through membranes. This solvent system also provides sta-
ble emulsion or microsuspension of the investigated enhancer
[21,29,30,39–45]. A tested enhancer cannot be completely dis-
solved, similarly as in topical formulations, because by penetration
of enhancer to the skin the dynamic balance (dissolved/undissolved
part) is continuously changing. The investigated enhancer should
be understood as a pharmaceutical excipient influencing only drug
transport across the skin barrier [3,4].
The highest ERs were obtained with 13 (mono(decanoyl)) and 25
(tris(decanoyl)) derivatives with ERs of 2.13 and 2.09, respectively.
According to the above presented data (Table 2, Fig. 2 – hatched
columns), it can be concluded that all the discussed compounds
showed only moderate penetration activity. Although acylation of
cholic acid did not lead to potentially better enhancers of transder-
mal absorption, some interesting structure–activity relationships
can be observed.
Generally, it can be postulated that the highest activity within
the individual series was shown by mono- and tris(acyloxy) sub-
stituted derivatives, whereas bis(acyloxy) substituted derivatives,
especially 7␣,12␣-bis(acyloxy) substituted compounds, possessed
the lowest enhancing activity. However, comparable ER values
were obtained within the series of the least lipophilic butanoyl
derivatives (3, 6, 8) and the most lipophilic hexadodecanoyl deriva-
tives (31, 33, 35), whereas hexadodecanoyl series expressed higher
penetration-enhancing activity compared with butanoyl series.
Great differences can be observed in both decanoyl series,
whereas methyl ester series 26–29 showed to be less active
than acids 13, 19, 22, 25. From the results (Table 2, Fig. 2 –
hatched columns) it is evident that penetration–enhancement
activity is strongly dependent on balanced lipo/hydrophilic prop-
erties (lipophilicity/solubility–polarity) of the individual discussed
enhancers due to their optimal interaction with skin components.
The dependences of transdermal penetration–enhancement activ-
ity (ER) on the lipophilicity (RM values) and solubility (log S values)
of the studied compounds are shown in Fig. 3.
When compounds with the lowest penetration–enhancement
activity are eliminated, dependences illustrated in Fig. 3 can be
presented. The eliminated compounds are mostly 3␣,7␣- and
7␣,12␣-bis(acyloxy) derivatives and generally compounds with
great differences between determined and predicted lipophilic-
ity and/or solubility, i.e. compounds with strong inter- and
intramolecular interactions. The parabolic dependence of activity
on lipophilicity can be observed in Fig. 3a, and almost a mirror
image can be seen in Fig. 3b, where the dependence of activity
on solubility is demonstrated. It can be concluded that the higher
The results of one-way analysis of the variance (ANOVA) test
complemented by the Bonferroni’s multicomparison test are pre-
sented in Table 2 where differences were considered significant
at P = 0.05. Considerable differences between determined ER were
found. If comparison was made related to ER in the skin test of the
most active compound 13 (2.13), the ER values of compounds 27
and 28 were significantly different from it at the probability level
2.2
2.2
13
25
13
25
31
35
35
31
2.0
1.8
1.6
1.4
1.2
1.0
0.8
2.0
33
33
26
26
1.8
3
8
3
8
1.6
6
6
29
29
1.4
19
19
22
22
1.2
27
27
1.0
28
28
0.8
-0.6
-0.1
0.4
0.9
1.4
1.9
-12.0 -11.0 -10.0
-9.0
-8.0
-7.0
-6.0
-5.0
-4.0
-3.0
-2.0
RM
log S
Fig. 3. The dependences of transdermal penetration–enhancement activity ER on the lipophilicity RM (a) and solubility log S (b) of the studied target compounds.

1088
L. Mrózek et al. / Steroids 76 (2011) 1082–1097
1.3
1.1
0.9
0.7
0.5
0.3
1.3
6
6
8
8
28
28
1.1
3
3
22
22
29
29
27
27
25
0.9
33
33
35
35
26
19
25
26
0.7
0.5
0.3
19
31
31
13
-0.1
13
-0.6
0.4
0.9
1.4
1.9
-12.0 -11.0 -10.0
-9.0
-8.0
-7.0
-6.0
-5.0
-4.0
-3.0
-2.0
RM
log S
Fig. 4. The dependences of intestinal absorption-enhancement activity ER on the lipophilicity RM (a) and solubility log S (b) of the studied target compounds.
transdermal penetration–enhancement activity is connected with
higher solubility (Fig. 3b) and an optimum range of lipophilicity
(Fig. 3a). When lipophilicity decreases dramatically with higher
solubility, a decrease in enhancement activity can also be observed
(butanoyl derivatives 3, 6, 8). This confirms the fact that alkyl chains
C10–C16 are preferred, because transdermal chemical penetration
enhancers are compounds that intercalate between ceramides in
the SC, disrupt ceramide–ceramide bonds and, by doing this, form
a “channel” in the SC [21].
According to the above presented data (Table 2, Fig. 2 – dotted
columns), it can be concluded that all the discussed compounds
showed only moderate enhancement activity; the resting non-
mentioned compounds are decelerators. Although it can be
concluded that prepared acylcholic acid derivatives showed
only low intestinal absorption enhancement, some interesting
structure–activity relationships can be observed.
The results of one-way ANOVA test complemented by the Bon-
ferroni’s multicomparison test are presented in Table 2 where
differences were considered significant at P = 0.01. Considerable
differences between determined ER were found. If comparison was
made related to ER of the less active compound in PAMPA test, i.e.
28, the ER values of compounds 13, 19, 25, 26, 27, 29, 31, 33 and 35
significantly differed from it at the probability level P = 0.001 and
the ER values of compounds 3 and 22 at P = 0.01.
Significant changes even with small differences in the chemi-
cal structure can be observed in PAMPA experiments. Generally, it
can be stated that trends in enhancement efficacy are absolutely
opposite compared to transdermal penetration–enhancement
activity ER. While transdermal enhancement activity is the highest
for mono- a tris(acyloxy) substituted derivatives and the low-
est for 7␣,12␣-bis(acyloxy) substituted compounds, in PAMPA
experiments namely 7␣,12␣-bis(acyloxy) substituted compounds
showed the highest activities within individual series, see Table 2,
Fig. 2 – dotted columns. This fact can be probably connected with
the above discussed specific non-covalent interactions of hydroxyl
moieties, i.e. specific lipo/hydrophilic interactions and properties,
see Section 2.2.
The compounds shown in Fig. 4 can be divided into intestinal
absorption promoters/enhancers (accelerators) and compounds
with the lowest enhancer activity, i.e. the highest “decelerator
activity” (decelerators). Practically the whole series of decanoyl
derivatives 13, 19, 25 and some other mono(acyloxy) substituted
compounds, generally with relatively low lipophilicity can be clas-
sified as decelerators. Fig. 4a illustrates the linear dependence
of activity on lipophilicity; intestinal penetration–enhancement
activity dramatically decreases with lipophilicity increase (dashed
line). Fig. 4b shows linear dependence of activity on solubility
(dashed line), and it is almost a mirror image of the depen-
dence demonstrated in Fig. 4a. Absolutely different dependences
can be observed in the series of “decelerators”. Fig. 4a shows
decelerator activity decrease (intestinal enhancement activity
increase) with lipophilicity (RM) increase (dashed and dotted line),
whereas Fig. 4b illustrates decelerator activity increase (decrease
of enhancement activity) with solubility (log S) increase (dashed
and dotted line). Based on these facts, it can be concluded that
in the case of accelerants/enhancers (not decelerators) intestinal
penetration–enhancement activity is also connected with higher
2.4. In vitro screening of intestinal absorption-enhancing activity
(PAMPA experiments)
PAMPA have become a very useful and quite cheap tool for
predicting in vivo drug permeability and are well-suited as a rank-
ing tool for the assessment of compounds with passive transport
mechanisms. An absorption study of binary mixtures or final for-
mulations is also possible on PAMPA plates.
The effect of target cholic acid derivatives on the penetra-
tion of theophylline through the artificial polyvinylidene fluoride
(PVDF) PAMPA membrane is presented in Fig. 2, dotted columns.
Control experiments were run with only theophylline in the
donor vehicle in the absence of any enhancer. The used solvents
ensure solubility of theophylline as a model penetrating com-
pound, which is important for facilitation of penetration through a
membrane. This system also supports the stability of the micro-
suspension of the tested enhancer. The same as for the skin,
the ratio between theophylline penetration with and without an
enhancer was determined for the selected concentration based on
the previous experience [29,30]. Only if some of evaluated potential
enhancers expressed significant enhancer activity, the influence
of the enhancer concentration on enhancement effect would be
investigated.
Penetration in Franz cells is different from that in PAMPA tests,
because the real skin is used as a barrier in Franz cells. Another
parameter that may influence the measurements is solubility (sol-
vation) supramolecular superassembly properties of cholic acid
derivatives (donors), but explanation of this parameter is not the
aim of this study. The difference between both experiments can
also be seen in the geometrical arrangement of the experiment:
in the PAMPA method a donor is below the diaphragm, whereas
in the Franz cell the donor is above it. The greatest differences
were noticeable in the results of mono(decanoyl) derivative 13 and
mono(hexadecanoyl) derivative 31.
The highest ERs were obtained for compounds 28
(7␣,12␣-bis(decanoyl)-Me ester),
(tris(butanoyl)), (mono(butanoyl)) and 22 (7␣,12␣-
bis(decanoyl)): 1.21, 1.18 1.16, 1.08 and 1.03, respectively.
6
(7␣,12␣-bis(butanoyl)),
8
3

L. Mrózek et al. / Steroids 76 (2011) 1082–1097
1089
solubility (Fig. 4b) and relatively low lipophilicity (Fig. 4a). 7␣,12␣-
Bis(decanoyl) derivative 28 showed the highest activity again,
probably due to specific strong inter- and intramolecular interac-
tions.
As all the butanoyl derivatives (3, 6, 8) expressed the balanced
transdermal and intestinal penetration–enhancement activity,
contrary to the rest of the discussed compounds, it can be suggested
that these butanoyl derivatives as amphiphilic compounds show
high hydrotropic effect and can reduce the resistance of membranes
to drug diffusion [50,51].
found. If comparison was made related to IC50 values of the less
anti-proliferatively active (insignificantly toxic) compound against
CEM cells, i.e. 8 (33.0 ␮mol/L) the IC50 values of compounds 13
and 22 were significantly different from it at the probability level
P = 0.01 and the IC50 value of compound 31 at P = 0.05. Similarly, in
the cytotoxicity test against MCF7 if related to the IC50 value of the
less anti-proliferatively active (insignificantly toxic) compound 13
(34.6 ␮mol/L), the IC50 value of compound 22 significantly differed
from it at the probability level P = 0.001.
The highest activity of 28 and the total loss of the intestinal
penetration–enhancement activity of the majority of decanoyl and
all hexadecanoyl derivatives compared to transdermal absorption
can be explained also by a “cut-off” effect [52,53]. This “cut-off”
effect is observed for n-alkyl substituted series of amphiphilic com-
pounds, when the dependence of biological activity upon the alkyl
chain length shows quasi-parabolic or bilinear dependence. The
total inversion of activity is connected with the structural diversity
of skin and PAMPA membrane. The hydrophobic parts of surfac-
tants interact with lipidic parts of biological membranes. However,
the water solubility of surfactants with longer alkyl chains is lim-
ited, and too large values of surfactant partition coefficient do
not enable the penetration of such molecules through hydrophilic
(aqueous) regions of biological membranes. Consequently, the final
concentration of long-chain surfactants in the membrane will be
lower than that of surfactants with shorter alkyl chains, which
results in the loss of activity. It is suggested that the lateral expan-
sion of the phospholipid bilayer of biological membranes caused by
the intercalation of long-chain amphiphilic molecules between the
phospholipid molecules and the mismatch between their hydro-
carbon chains lengths results in the creation of free volume in
the bilayer hydrophobic region [53]. According to the free volume
theory the extent of membrane disturbance due to surfactant incor-
poration depends on the size of free volume created under its alkyl
chain which can be then filled up with chains of neighbouring lipids
as well as on the partition coefficient of the surfactants [52,54].
Therefore the most effective disturbance of the membrane and thus
the highest inhibitory effect will be exhibited by surfactants with
middle alkyl chain length ensuring not only sufficiently high free
volume under an alkyl chain but also high concentration of the sur-
factant in the membrane due to the suitable value of the surfactant
partition coefficient [55].
3. Conclusions
In this work a series of fourteen final acyloxy derivatives
of 5␤-cholic acid with C4, C10 and C16 linear acyl chains were
prepared as novel potential transdermal penetration enhancers
and intestinal drug absorption modifiers. Experimental (relative)
lipophilicity RM, calculated from RP-TLC measurements was com-
pared with predicted log P and log S values. The determined RM
values as well as the calculated log P and log S values were com-
pared with the membrane permeability influence studied by the
PAMPA method and transdermal absorption in vitro. The compar-
ison of the values influencing theophylline transport through the
artificial and natural membranes (RM, log P, log S) with experimen-
tal ERs of the cholic acid ester derivatives confirmed the expected
correlation. The ability of the final compounds to enhance the
penetration of theophylline through porcine skin was examined
using a Franz cell, and the intestinal drug absorption enhanc-
ing effect was evaluated by means of PAMPA experiments. The
highest transdermal penetration–enhancement activity in this
study was exhibited by compounds 13 (3␣-mono(decanoyl))
and 25 (3␣,7␣,12␣-tris(decanoyl)), while the highest intestinal
absorption-enhancement activity was exhibited by compounds
28 (7␣,12␣-bis(decanoyl)-Me ester) and 6 (7␣,12␣-bis(butanoyl)).
All the compounds were additionally evaluated for their anti-
proliferative/cytotoxic activity against two human cancer cell lines
and against normal human skin fibroblast cells. Two compounds,
13 and 31, showed anti-proliferative effect on cancer cells with-
out affecting the growth of normal cells, which suggests that
these compounds would have low cytotoxic side-effects when
administered as enhancers/excipients. Ten other target compounds
exhibited limited cytotoxicity, so they could be used as absorption
modifiers. Although all the discussed compounds expressed only
moderate intestinal absorption/transdermal penetration enhance-
ment effects, nevertheless the obtained data provided important
parameters for correlations between solubility/lipophilicity and
enhancement activity, and noteworthy structure–activity rela-
tionships were found for subsequent structure optimization and
rationalization of the design of novel potential cholic acid-type
enhancers.
2.5. In vitro anti-proliferative/cytotoxicity assays
To evaluate the cytotoxic properties of tested compounds cells
of different histopathological origin were used: a T-lymphoblastic
leukemia cell line (CEM), a breast adenocarcinoma cell line (MCF7),
and, as controls, normal human skin fibroblast cells (BJ). Treat-
ment with compounds 8 (tris(butanoyl)), 13 (mono(decanoyl)), 22
(7␣,12␣-bis(decanoyl)), and 31 (mono(hexadecanoyl)) resulted in
dose-dependent inhibition of cancer cells viability. Compound 22
showed the highest cytotoxicity but compounds 8, 13 and 31 had
cytotoxic effect on cancer cells without affecting the growth of nor-
mal BJ cells, which should be promising in potential development
of new antiproliferative drugs.
Other discussed final compounds demonstrated poor anti-
proliferative effect (or insignificant cytotoxicity effect) against all
the cell lines, with IC50 values greater than 37 ␮mol/L, see Table 2.
These results suggest that the poor anti-proliferative activity of the
target compounds will lead to limited cytotoxicity if they are used
in vivo as absorption modifiers.
4. Experimental
4.1. Chemistry
All reagents were purchased from Sigma–Aldrich (Schnell-
dorf, Germany) and Merck (Darmstadt, Germany). Kieselgel 60,
0.063–0.200 mm (Merck, Darmstadt, Germany) was used for
column chromatography. Thin layer chromatography (TLC) experi-
ments were performed on aluminium foil-backed silica gel 40 F254
plates (Merck, Darmstadt, Germany). Detection was performed by
spraying with a solution of 20 g of Ce(SO4)2 in 200 ml 10% H2SO4
and subsequent heating. The melting points were determined
on a Boetius apparatus (Nagema, Germany) and are uncorrected.
Infrared (IR) spectra were recorded on a Smart MIRacleTM ATR
ZnSe for NicoletTM 6700 FT-IR Spectrometer (Thermo Scientific,
The results of one-way ANOVA test complemented by the Bon-
ferroni’s multicomparison test are presented in Table 2 where
differences were considered significant at P = 0.05. Considerable
differences between determined anti-proliferative IC50 values were

1090
L. Mrózek et al. / Steroids 76 (2011) 1082–1097
USA). The spectra were obtained by accumulation of 256 scans
with 2 cm−1 resolution in the 4000–600 cm−1 region. Parame-
ter “zero-filling” was 0. Happ–Gensel apodisation function was
used. Elemental analyses were performed, using a Vario EL III Uni-
versal CHNOS Elemental Analyzer (Elementar Analysensysteme,
Germany). All 1H and 13C NMR spectra were recorded on a Bruker
Avance-250 FT-NMR spectrometer (250 MHz for 1H and 62.9 MHz
for 13C, Bruker Comp., Karlsruhe, Germany). Chemicals shifts are
reported in ppm (ı) using internal Si(CH3)4 as the reference, with
diffuse, easily exchangeable signals being omitted. Mass spectra
were measured using a LTQ Orbitrap Hybrid Mass Spectrometer
(Thermo Electron Corporation, USA) with direct injection into an
APCI source (400 ◦C) in the negative mode.
Benzyl 3␣-butanoyloxy-7␣,12␣-dihydroxy-5␤-cholate (2):
Butanoyl chloride (1.50 ml, 14.33 mmol) dissolved in CH2Cl2
(20 ml) was slowly dropwise added to the solution of benzyl
cholate (1, 6.03 g, 12.09 mmol) and DMAP (1.79 g, 14.65 mmol) in
CH2Cl2 (40 ml) at 0 ◦C (ice bath) under argon. Then the mixture was
stirred and warmed to a room temperature. After 2 h it was poured
into aqueous AcOH (100 ml, 6%), and the aqueous layer was washed
with CH2Cl2. Combined organic layers were washed with saturated
aqueous NaHCO3 and water and evaporated under reduced pres-
sure. The crude product was purified by flash chromatography on
silica gel (CH2Cl2/MeOH/AcOH 300:5:1 to 300:30:2). This provided
a colourless oil. Yield: 2.79 g (40%). 1H NMR (250 MHz, CDCl3),
ı: 0.67 (s, 3H, CH3), 0.90 (s, 3H, CH3), 0.93 (m, 3H, CH3), 0.97 (d,
3H, CH3, J = 6.87 Hz), 1.05–2.50 (m, 30H), 3.85 (m, 1H, C(7)H), 3.97
(m, 1H, C(12)H), 4.59 (m, 1H, C(3)H), 5.11 (2H, PhCH2), 7.30–7.40
(m, 5H, Ph). 13C NMR (62.5 MHz, CDCl3), ı: 12.69, 13.81, 17.48,
18.69, 22.69, 23.29, 26.88, 26.91, 27.56, 28.54, 31.02, 31.45, 34.55,
34.84, 35.05, 35.24, 35.42, 36.81, 39.72, 41.39, 42.22, 46.71, 47.41,
66.26, 68.41, 73.05, 74.14, 128.32, 128.37, 128.68, 136.28, 173.48
(COOR), 174.14 (COOR). HR-MS: for C35H51O6 [M−H]− calculated:
567.3764 m/z; found: 567.3772 m/z.
(62.5 MHz, CDCl3), ı: 12.63, 17.45, 21.55, 22.61, 23.28, 26.73, 26.78,
27.54, 28.42, 30.98, 31.43, 34.56, 34.81, 35.02, 35.25, 35.30, 39.64,
41.32, 41.61, 42.06, 46.66, 47.36, 66.22, 68.39, 73.03, 74.44, 128.27,
128.33, 128.64, 136.23, 170.85 (COOR), 174.13 (COOR). HR-MS: for
C33H47O6 [M−H]− calculated: 539.3451 m/z; found: 539.3463 m/z.
Benzyl 3␣-acetoxy-7␣,12␣-bis(butanoyloxy)-5␤-cholate (5):
Ester 4 (2.50 g, 4.63 mmol), BTEAC (0.41 g, 1.80 mmol), CaH2 (0.82 g,
19.48 mmol) and butanoyl chloride (2 ml, 19.11 mmol) dissolved
in toluene (40 ml) were refluxed under argon for 1 h. Then the
mixture was poured into aqueous AcOH (70 ml, 4%). The aqueous
layer was washed with toluene (3 × 50 ml), and combined organic
extracts were washed with saturated aqueous NaHCO3 (200 ml)
and brine and evaporated under reduced pressure. The crude prod-
uct was purified by flash chromatography on silica gel (petroleum
ether/acetone 640:240). This provided a colourless oil. Yield: 2.96 g
(94%). 1H NMR (250 MHz, CDCl3), ı: 0.70 (s, 3H, CH3), 0.82 (d,
3H, CH3, J = 6.14 Hz), 0.92 (s, 3H, CH3), 0.98 (m, 3H, CH3), 1.03 (t,
3H, CH3, J = 7.42 Hz), 1.07–2.50 (m, 32H), 2.01 (s, 3H, CH3CO), 4.56
(m, 1H, C(3)H), 4.94 (m, 1H, C(7)H), 5.10 (m, 1H, C(12)H), 5.10 (2H,
PhCH2), 7.30–7.40 (m, 5H, Ph). 13C NMR (62.5 MHz, CDCl3), ı: 12.20,
13.87, 14.05, 17.62, 18.51, 18.88, 21.46, 22.53, 23.06, 25.39, 26.82,
27.31, 28.74, 30.94, 31.32, 31.47, 34.43, 34.74, 34.79, 34.90, 36.98,
37.03, 38.03, 40.94, 43.34, 45.16, 47.55, 66.24, 70.61, 73.96, 75.17,
128.31, 128.36, 128.64, 136.16, 170.58 (COOR), 172.73 (COOR),
172.81 (COOR), 173.90 (COOR). HR-MS: for C41H59O8 [M−H]− cal-
culated: 679.4288 m/z; found: 679.4295 m/z.
7␣,12␣-Bis(butanoyloxy)-3␣-hydroxy-5␤-cholic acid (6): The
mixture of ester 5 (2.81 g, 4.13 mmol), i-PrOH (37 ml) and aque-
ous NaOH (9 ml; 4%) was stirred at 80 ◦C for 30 min. It was poured
into diluted aqueous AcOH (100 ml; 2.9%). The aqueous layer was
washed with CH2Cl2 (3 × 50 ml) and combined organic extracts
were washed with water (100 ml) and evaporated under reduced
pressure. The crude product was purified by flash chromatogra-
phy on silica gel (hexane/Et2O/AcOH 250:250:10 to 60:300:8). This
provided a white crystalline compound. Yield: 1.90 g (84%). Mp.
133–136 ◦C. 1H NMR (250 MHz, CDCl3), ı: 0.73 (s, 3H, CH3), 0.83 (d,
3H, CH3, J = 6.21 Hz), 0.91 (s, 3H, CH3), 0.99 (m, 3H, CH3), 1.02 (t, 3H,
CH3, J = 7.47 Hz), 1.05–2.50 (m, 33H), 3.49 (m, 1H, C(3)H), 4.94 (m,
1H, C(7)H), 5.11 (m, 1H, C(12)H), 5.60 (s, 1H). 13C NMR (62.5 MHz,
CDCl3), ı: 12.25, 13.85, 14.07, 17.62, 18.57, 18.79, 22.61, 23.06,
25.49, 27.32, 28.84, 30.57, 30.74, 31.04, 31.60, 34.42, 34.84, 35.00,
36.95, 37.03, 38.05, 38.97, 41.13, 43.39, 45.20, 47.52, 70.71, 71.79,
75.19, 173.03 (COOR), 173.10 (COOR), 179.40 (COOH). IR (cm−1):
ꢀ(OH) 3438, ꢀ(CH) 2932, 2870, ꢀ(C O) 1724, ı(CH) 1448, ꢀ(CO)
1253, 1181, 1072, 1039. Anal. Calc. for C32H52O7 (548.75): 70.04%
C, 9.55% H; found: 69.50% C, 9.97% H. HR-MS: for C32H51O6 [M−H]−
calculated: 547.3635 m/z; found: 547.3646 m/z.
3␣-Butanoyloxy-7␣,12␣-dihydroxy-5␤-cholic acid (3): Benzyl
ester 2 (2.79 g, 4.91 mmol) and HCOONH4 (1.05 g, 16.7 mmol) were
dissolved in MeOH (70 ml), and 10% Pd/C (0.29 g) was added. The
mixture was stirred under argon for 1.5 h. Then the second portion
of 10% Pd/C (0.29 g) was added to the mixture and it was stirred for
next 2 h. Then the suspension was filtered through celite and evap-
orated under reduced pressure. The crude product was purified by
flash chromatography on silica gel (CH2Cl2/AcOEt/AcOH 400:100:4
to 250:250:4). This provided a white crystalline compound. Yield:
1.37 g (58%). Mp. 127–130 ◦C. 1H NMR (250 MHz, CDCl3), ı: 0.70
(s, 3H, CH3), 0.91 (s, 3H, CH3), 0.93 (m, 3H, CH3), 1.00 (d, 3H, CH3,
J = 5.84 Hz), 1.05–2.50 (m, 30H), 3.87 (m, 1H, C(7)H), 4.00 (m, 1H,
C
(12)H), 4.59 (m, 1H, C(3)H). 13C NMR (62.5 MHz, CDCl3), ı: 12.68,
13.81, 17.45, 18.69, 22.65, 23.31, 26.81, 26.88, 27.62, 28.43, 30.86,
31.11, 34.58, 34.86, 35.05, 35.37, 35.40, 36.81, 39.64, 41.36, 42.15,
46.70, 47.28, 68.53, 73.21, 74.19, 173.53 (COOR), 179.32 (COOH). IR
(cm−1): ꢀ(OH) 3331, ꢀ(CH) 2936, 2863, ꢀ(C O) 1719, ı(CH) 1465,
ꢀ(CO) 1250, 1185, 1075, 1020. Anal. Calc. for C28H46O6 (478.66):
70.26% C, 9.69% H; found: 70.02% C, 10.16% H. HR-MS: for C28H45O6
[M−H]− calculated: 477.3216 m/z; found: 477.3227 m/z.
Benzyl 3␣,7␣,12␣-tris(butanoyloxy)-5␤-cholate (7): Benzyl
cholate 1 (4.00 g, 8.02 mmol), BTEAC (0.71 g, 3.12 mmol) CaH2
(2.02 g, 47.98 mmol) and butanoyl chloride (5.00 ml, 47.77 mmol)
dissolved in toluene (100 ml) were refluxed under argon for 3 h.
Then the mixture was poured into aqueous AcOH (100 ml, 6%). The
aqueous layer was washed with toluene (3 × 50 ml) and combined
organic extracts were washed with saturated aqueous NaHCO3
(200 ml) and brine and evaporated under reduced pressure. The
crude product was purified by flash chromatography on silica gel
(CH2Cl2/MeOH/AcOH 300:5:1). This provided a colourless oil. Yield:
4.4 g (77%). 1H NMR (250 MHz, CDCl3), ı: 0.70 (s, 3H, CH3), 0.82 (d,
3H, CH3, J = 6.13 Hz), 0.92 (s, 3H, CH3), 0.94 (m, 3H, CH3), 0.98 (m,
3H, CH3), 1.03 (m, 3H, CH3), 1.10–2.50 (m, 36H), 4.58 (m, 1H, C(3)H),
4.93 (m, 1H, C(7)H), 5.09 (m, 1H, C(12)H), 5.10 (2H, PhCH2), 7.30–7.38
(m, 5H, Ph). 13C NMR (62.5 MHz, CDCl3), ı: 12.23, 13.71, 13.90,
14.10, 17.66, 18.53, 18.70, 18.93, 22.54, 23.09, 25.37, 26.86, 27.34,
28.75, 30.98, 31.35, 31.49, 34.46, 34.78, 34.82, 34.97, 36.74, 37.05,
37.09, 38.06, 40.94, 43.36, 45.19, 47.60, 66.28, 70.69, 73.63, 75.24,
128.35, 128.40, 128.68, 136.19, 172.72 (COOR), 172.82 (COOR),
Benzyl 3␣-acetoxy-7␣,12␣-dihydroxy-5␤-cholate (4): The mix-
ture of ester 1 (10.04 g, 20.14 mmol) and KHCO3 (8.81 g, 88 mmol)
in toluene (80 ml) and Ac2O (6.6 ml; 69.8 mmol) was refluxed under
argon for 20 min. Then the mixture was poured on ice, the aqueous
layer was separated and washed with toluene (50 ml). Combined
organic extracts were washed with brine and evaporated under
reduced pressure. The crude product was purified by crystalliza-
tion from toluene, and a white crystalline compound was obtained.
Yield: 4.94 g, (45%). 1H NMR (250 MHz, CDCl3), ı: 0.67 (s, 3H, CH3),
0.90 (s, 3H, CH3), 0.97 (d, 3H, CH3, J = 5.97 Hz), 1.00–2.50 (m, 26H),
1.99 (s, 3H, CH3CO), 3.84 (m, 1H, C(7)H), 3.97 (m, 1H, C(12)H), 4.57
(m, 1H, C(3)H), 5.11 (2H, PhCH2), 7.30–7.40 (m, 5H, Ph). 13C NMR

L. Mrózek et al. / Steroids 76 (2011) 1082–1097
1091
173.19 (COOR), 173.95 (COOR). HR-MS: for C43H65O8 [M−H]− cal-
culated: 707.4601 m/z; found: 707.4610 m/z.
temperature for 1.5 h. The temperature was allowed to rise and
maintained at −30 to −40 ◦C during 2 h. Then it was cooled down to
−57 ◦C, and tert-BuOH (100 ml, 1.046 mol) was added slowly. After
16 h the mixture was poured on ice at the laboratory temperature.
The aqueous layer was washed with Et2O (250 ml). Et2O layer was
washed with ice-cooled aqueous NaOH (150 ml; 2 M) and satu-
rated aqueous NaHCO3 (150 ml) and evaporated under reduced
pressure. The residue was mixed with MeOH (150 ml) and THF
(150 ml) at 0 ◦C; the saturated aqueous solution of NaHCO3 (75 ml)
was added, and mixture was stirred for 1 h. The second portion of
the saturated aqueous solution of NaHCO3 (20 ml) was added, and
the mixture was stirred at room temperature for 3 h and poured
into the mixture of Et2O (500 ml) and ice-cooled water (300 ml).
After shaking, the Et2O layer was washed with water (300 ml) and
phosphate buffer (300 ml; 1 M) and evaporated under reduced
pressure. The crude product was purified by crystallization form
hexane, and a white crystalline compound was obtained. Yield:
9.42 g (68%). Mp. 123–125 ◦C. 1H NMR (250 MHz, CDCl3), ı: 0.79 (s,
3H, CH3), 0.81 (d, 3H, CH3, J = 6.47 Hz), 0.90 (s, 3H, CH3), 0.80–2.3
(m, 26H), 1.44 (s, 9H, (CH3)3C–OCO), 3.43 (m, 1H, C(3)H), 3.88
(m, 1H, C(7)H), 5.31 (m, 1H, C(12)H). 13C NMR (62.5 MHz, CDCl3),
ı: 12.32, 17.62, 22.64, 22.77, 22.90, 25.51, 27.29, 27.77, 28.24,
30.71, 30.85, 32.35, 34.60, 34.69, 35.15, 39.33, 39.85, 41.49, 43.49,
45.24, 47.65, 67.91, 71.92, 80.17 (C(CH3)3), 81.01, 114.79 (q, CF3CO,
1JCF = 286.20 Hz), 157.06 (q, CF3CO, 2JCF = 42.14 Hz), 173.53 (COOR).
HR-MS: for C30H46F3O6 [M−H]− calculated: 559.3325 m/z; found:
559.3337 m/z.
3␣,7␣,12␣-Tris(butanoyloxy)-5␤-cholic acid (8): Benzyl ester 7
(4.4 g, 6.21 mmol) and HCOONH4 (1.60 g, 25.37 mmol) were dis-
solved in MeOH (70 ml), and 10% Pd/C (0.29 g) was added. The
mixture was stirred under argon for 1.5 h. Then the second portion
of 10% Pd/C (0.29 g) was added to the mixture, and it was stirred for
next 2 h. After that the suspension was filtered through celite and
evaporated under reduced pressure. The crude product was puri-
fied by flash chromatography on silica gel (CH2Cl2/MeOH/AcOH
300:5:1 to 300:15:1.5). This provided a colourless oil. Yield: 3.0 g
(78%). 1H NMR (250 MHz, CDCl3), ı: 0.74 (s, 3H, CH3), 0.84 (d, 3H,
CH3, J = 6.20 Hz), 0.92 (s, 3H, CH3), 0.94 (m, 3H, CH3), 0.98 (t, 3H, CH3,
J = 7.46 Hz), 1.04 (t, 3H, CH3, J = 7.29 Hz), 1.10–2.50 (m, 37H), 4.59
(m, 1H, C(3)H), 4.95 (m, 1H, C(7)H), 5.12 (m, 1H, C(12)H), 9.50 (s, 1H,
COOH). 13C NMR (62.5 MHz, CDCl3), ı: 12.23, 13.68, 13.87, 14.07,
17.63, 18.52, 18.68, 18.91, 22.51, 23.06, 25.35, 26.83, 27.32, 28.73,
30.72, 31.03, 31.46, 34.44, 34.75, 34.81, 34.94, 36.72, 37.03, 37.08,
38.04, 40.90, 43.34, 45.19, 47.53, 70.72, 73.65, 75.25, 172.79 (COOR),
172.87 (COOR), 173.27 (COOR), 179.85 (COOH). IR (cm−1): ꢀ(CH)
2961, 2873, ꢀ(C O) 1726, ı(CH) 1456, ꢀ(CO) 1252, 1180, 1089. Anal.
Calc. for C36H58O8 (616.84): 69.87% C, 9.45% H; found: 68.05% C,
9.48% H. HR-MS: for C36H57O6 [M−H]− calculated: 617.4053 m/z;
found: 617.4049 m/z.
Benzyl decanoyloxy-hydroxy-cholates 9–12: Decanoyl chlo-
ride (3.75 ml, 18.07 mmol) dissolved in CH2Cl2 (40 ml) was slowly
dropwise added to the solution of benzyl cholate (1, 10.03 g,
20.12 mmol), and DMAP (3.08 g; 25.21 mmol) in CH2Cl2 (60 ml)
at 0 ◦C (ice bath) under argon. Then the mixture was stirred and
warmed to a room temperature. After 2 h it was poured into aque-
ous AcOH (100 ml, 6%), and the aqueous layer was washed with
CH2Cl2. Combined organic layers were washed with saturated
aqueous NaHCO3 and water and evaporated under reduced pres-
sure. The crude product was purified by flash chromatography on
silica gel (toluene/AcOEt/AcOH 100:5:0.5 to 0:100:1). This provided
four products:
Benzyl 3␣-decanoyloxy-7␣,12␣-dihydroxy-5␤-cholate (9): A
colourless oil. Yield: 2.40 g (37%). 1H NMR (250 MHz, CDCl3), ı:
0.66 (s, 3H, CH3), 0.88 (m, 3H, CH3), 0.90 (s, 3H, CH3), 0.97 (d,
3H, CH3, J = 5.88 Hz), 1.00–2.5 (m, 42H), 3.84 (m, 1H, C(7)H), 3.97
(m, 1H, C(12)H), 4.58 (m, 1H, C(3)H), 5.11 (2H, PhCH2), 7.3–7.4 (m,
5H, Ph). 13C NMR (62.5 MHz, CDCl3), ı: 12.60, 14.20, 17.43, 22.58,
22.77, 23.28, 25.19, 26.73, 26.87, 27.56, 28.40, 29.30, 29.37, 29.40,
29.53, 30.96, 31.39, 31.96, 34.62, 34.81, 34.94, 35.04, 35.28, 35.32,
39.58, 41.35, 42.09, 46.66, 47.34, 66.21, 68.42, 73.11, 74.14, 128.26,
128.31, 128.62, 136.22, 173.59 (COOR), 174.14 (COOR). HR-MS: for
C41H63O6 [M−H]− calculated: 651.4703 m/z; found: 651.4714 m/z.
Benzyl 7␣-decanoyloxy-3␣,12␣-dihydroxy-5␤-cholate (10): A
colourless oil. Yield: 1.51 g (11%). 1H NMR (250 MHz, CDCl3), ı:
0.64 (s, 3H, CH3), 0.88 (m, 3H, CH3), 0.90 (s, 3H, CH3), 0.96 (d, 3H,
CH3, J = 5.93 Hz), 1.0–2.5 (m, 42H), 3.47 (m, 1H, C(3)H), 3.97 (m, 1H,
tert-Butyl 3␣-decanoyloxy-7␣-hydroxy-12␣-trifluoroacetoxy-
5␤-cholate (16): The solution of ester 15 (1.99 g, 3.55 mmol), DMAP
(0.13 g, 1.06 mmol) and TEA (1.1 ml, 7.9 mmol) in CH2Cl2 (55 ml)
was mixed with decanoyl chloride (1.6 ml, 7.7 mmol), and the mix-
ture was stirred for 50 min at room temperature. Then it was
evaporated under reduced pressure. The crude product was puri-
fied by flash chromatography on silica gel (toluene/AcOEt/AcOH
100:5:0.5). This provided a colourless oil. Yield: 1.43 g (56%).1H
NMR (250 MHz, CDCl3), ı: 0.75–0.94 (m, 12H, 4 × CH3), 0.95–2.38
(m, 42H), 1.44 (s, 9H, (CH3)3C–OCO), 3.88 (m, 1H, C(7)H), 4.55
(m, 1H, C(3)H), 5.32 (m, 1H, C(12)H). 13C NMR (62.5 MHz, CDCl3),
ı: 12.32, 14.19, 17.62, 22.59, 22.78, 22.90, 25.17, 25.49, 26.66,
27.30, 27.75, 28.23, 29.38 (t, JCF = 9.1 Hz), 30.86, 31.98, 32.36,
34.56, 34.60, 34.62, 34.85, 35.36, 39.29, 41.24, 43.50, 45.25, 47.66,
1
67.84, 73.77, 80.16, 80.95, 114.80 (q, CF3CO, JCF = 286 Hz), 157.02
2
(q, CF3CO, JCF = 42.14 Hz), 173.50 (COOR), 173.72 (COOR). HR-
MS: for C40H64F3O7 [M−H]− calculated: 713.4682 m/z; found:
713.4691 m/z.
tert-Butyl 3␣-decanoyloxy-7␣,12␣-dihydroxy-5␤-cholate (17):
The solution of ester 16 (1.43 g, 2.0 mmol) in THF (40 ml) and MeOH
(40 ml) was mixed with concentrated NH3(aq) (30.5 ml) and the
mixture was stirred at room temperature for 1 h, poured into phos-
phate buffer (200 ml, pH 7) and washed with Et2O (2 × 160 ml).
Combined Et2O extracts were washed with brine and evaporated
under reduced pressure. The crude product was purified by flash
chromatography on silica gel (toluene/AcOEt/AcOH 100:5:0.5). This
provided a yellowish oil. Yield: 1.33 g (90%). 1H NMR (250 MHz,
CDCl3), ı: 0.68 (s, 3H, CH3), 0.86 (m, 3H, CH3), 0.89 (s, 3H, CH3),
0.96 (d, 3H, CH3, J = 6.24 Hz), 1.00–2.00 (m, 37H), 1.43 (s, 9H,
(CH3)3C–OCO), 2.03–2.40 (m, 7H), 3.84 (m, 1H, C(7)H), 3.98 (m, 1H,
C
(12)H), 4.89 (m, 1H, C(7)H), 5.10 (2H, PhCH2), 7.30–7.40 (m, 5H,
Ph). 13C NMR (62.5 MHz, CDCl3), ı: 12.57, 14.18, 17.37, 27.33, 28.20,
28.65, 29.29, 29.37, 29.47, 29.57, 30.57, 30.94, 31.43, 31.58, 31.95,
34.39, 34.99, 35.10, 35.22, 38.26, 39.06, 41.22, 42.16, 46.60, 47.29,
66.18, 70.73, 71.78, 72.72, 128.25, 128.30, 128.60, 136.17, 173.44
(COOR), 173.96 (COOR). HR-MS: for C41H63O6 [M−H]− calculated:
651.4703 m/z; found: 651.4716 m/z.
C
(12)H), 4.56 (m, 1H, C(3)H). 13C NMR (62.5 MHz, CDCl3), ı: 12.62,
Benzyl 3␣,7␣-bis(decanoyloxy)-12␣-hydroxy-5␤-cholate (11)
yield about 21%, was isolated together with benzyl 3␣,12␣-
bis(decanoyloxy)-7␣-hydroxy-5␤-cholate (12), yield about 12%,
and was not further purified (about 12%).
tert-Butyl 3␣,7␣-dihydroxy-12␣-trifluoroacetoxy-5␤-cholate
(15): The solution of 3␣,7␣,12␣-trihydroxy-5␤-cholic acid (14,
10.08 g, 24.67 mmol) in THF (250 ml) cooled to −45 ◦C under argon
was mixed with (TFAc)2O (62 ml, 446 mmol) and stirred at this
14.21, 17.48, 22.59, 22.78, 23.27, 25.19, 26.76, 26.84, 27.58, 28.23,
28.36, 29.30, 29.38, 29.40, 29.54, 31.04, 31.97, 32.65, 34.59, 34.82,
34.94, 35.02, 35.23, 35.33, 39.60, 41.33, 42.11, 46.65, 47.41, 68.46,
73.16, 74.16, 80.13, 173.73 (COOR), 173.88 (COOR). HR-MS: for
C
38H65O6 [M−H]− calculated: 617.4859 m/z; found: 617.4870 m/z.
3␣-Decanoyloxy-7␣,12␣-dihydroxy-5␤-cholic
acid
(13):
Method A: Benzyl ester 9 (2.33 g, 3.57 mmol) and HCOONH4
(1.05 g, 16.7 mmol) were dissolved in MeOH (50 ml), and 10% Pd/C

1092
L. Mrózek et al. / Steroids 76 (2011) 1082–1097
1H, C(3)H), 4.90 (m, 1H, C(7)H). 13C NMR (62.5 MHz, CDCl3), ı: 12.67,
14.23, 17.46, 22.68, 22.81, 23.16, 25.21, 25.25, 26.84, 27.39, 28.27,
28.72, 29.31, 29.42, 29.47, 29.55, 29.57, 29.67, 30.76, 31.13, 31.49,
32.00, 32.03, 34.50, 34.91, 35.11, 35.14, 38.29, 41.08, 42.28, 46.69,
47.40, 70.71, 72.91, 73.91, 173.35 (COOR), 173.58 (COOR), 179.61
(COOH). IR (cm−1): ꢀ(OH) 3529, ꢀ(CH) 2919, 2852, ꢀ(C O) 1732,
1705, ı(CH) 1378, ꢀ(CO) 1169. Anal. Calc. for C44H76O7 (717.07):
73.70% C, 10.68% H; found 73.54% C, 11.27% H. HR-MS: for C44H75O7
[M−H]− calculated: 715.5513 m/z; found: 715.5474 m/z.
(0.27 g) was added. The mixture was stirred under argon for 1 h.
Then the second portion of 10% Pd/C (0.2 g) was added to the
mixture, and it was stirred for next 2 h. Then the suspension was
filtered through celite and evaporated under reduced pressure.
The crude product was purified by flash chromatography on silica
gel (CH2Cl2/MeOH/AcOH 30:1.5:0.3). This provided a colourless
oil. Yield: 1.43 g (71%).
Method B: tert-Butyl ester 17 (1.30 g, 2.20 mmol) was dissolved
in CH2Cl2 (38 ml), the solution was cooled to 0 ◦C and mixed
with TFAcOH (7.5 ml) under argon. The reaction mixture was
warmed to the room temperature and after 2 h poured into aque-
ous AcONa (100 ml; 8%). The aqueous mixture was washed with
CH2Cl2 (2 × 100 ml) and evaporated under reduced pressure. The
crude product was purified by flash chromatography on silica
gel (CH2Cl2/Et2O/AcOH 100:8:0.5 to 100:64:0.4). This provided a
colourless oil. Yield: 0.60 g (53%). 1H NMR (250 MHz, CDCl3), ı:
0.70 (s, 3H, CH3), 0.87 (m, 3H, CH3), 0.90 (s, 3H, CH3), 0.99 (d,
3H, CH3, J = 5.76 Hz), 1.03–2.50 (m, 46H), 3.86 (m, 1H, C(7)H), 3.99
(m, 1H, C(12)H), 4.58 (m, 1H, C(3)H). 13C NMR (62.5 MHz, CDCl3), ı:
12.68, 14.24, 17.47, 22.65, 22.81, 23.32, 25.23, 26.82, 26.88, 27.63,
28.45, 29.34, 29.41, 29.44, 29.58, 30.91, 31.16, 32.01, 34.59, 34.87,
34.98, 35.07, 35.37, 35.41, 39.66, 41.37, 42.17, 46.70, 47.26, 68.52,
73.19, 74.18, 173.71 (COOR), 179.15 (COOH). IR (cm−1): ꢀ(OH) 3431,
ꢀ(CH) 2922, 2855, ꢀ(C O) 1706, ı(CH) 1464, ꢀ(CO) 1247, 1186,
1072, 1023. Anal. Calc. for C34H58O6 (562.82): 72.56% C, 10.39% H;
found 71.84% C, 10.84% H. HR-MS: for C34H57O6 [M−H]− calculated:
561.4155 m/z; found: 561.4168 m/z.
Benzyl 3␣-benzyloxycarboxy-7␣,12␣-dihydroxy-5␤-cholate
(20). The solution of benzyl cholate 1 (5.03 g, 10.1 mmol) in dry
pyridine (75 ml) was cooled to 0 ◦C and mixed under argon with
BnOCOCl (4.37 g, 25.6 mmol). Then the mixture was stirred for 4 h,
and another portion of BnOCOCl (4.38 g, 25.7 mmol) was added.
After 20 h the solvents were evaporated under reduced pressure.
The residue was dissolved in toluene (50 ml) and washed with
aqueous AcOH (60 ml, 17%). The aqueous layer was washed with
toluene (2 × 50 ml). Combined organic extracts were washed
with brine and evaporated under reduced pressure. The crude
product was purified by flash chromatography on silica gel
(CH2Cl2/MeOH/AcOH 300:5:1). This provided a colourless oil.
Yield: 2.97 g (46%). 1H NMR (250 MHz, CDCl3), ı: 0.65 (s, 3H,
CH3), 0.89 (s, 3H, CH3), 0.96 (d, 3H, CH3, J = 5.95 Hz), 1.00–2.50 (m,
27H), 3.83 (m, 1H, C(7)H), 3.95 (m, 1H, C(12)H), 4.45 (m, 1H, C(3)H),
5.10 (2H, PhCH2), 5.12 (s, 2H, PhCH2), 7.30–7.40 (m, 10H, 2 × Ph).
13C NMR (62.5 MHz, CDCl3), ı: 12.64, 17.43, 22.54, 23.26, 26.67,
26.70, 27.52, 28.41, 30.96, 31.41, 34.43, 34.76, 34.89, 35.16, 35.22,
39.65, 41.29, 42.00, 46.64, 47.30, 66.23, 68.30, 69.27, 72.94, 78.59,
128.29, 128.32, 128.34, 128.46, 128.64, 135.63, 136.23, 154.81
(OCOO), 174.16 (COOR). HR-MS: for C39H51O7 [M−H]− calculated:
631.3713 m/z; found: 631.3726 m/z.
tert-Butyl
3␣,7␣-bis(decanoyloxy)-12␣-trifluoroacetoxy-
5␤-cholate (18): tert-Butyl ester 15 (4 g, 7.13 mmol), BTEAC
(0.61 g, 2.7 mmol) CaH2 (1.2 g, 28.5 mmol) and decanoyl chloride
(5.6 g, 29.4 mmol) dissolved in toluene (100 ml) were refluxed
under argon for 2 h. Then the mixture was poured into aqueous
AcOH (250 ml, 2%). The aqueous layer was washed with toluene
(3 × 50 ml) and combined organic extracts were washed with
saturated aqueous NaHCO3 (200 ml) and brine and evaporated
under reduced pressure. The crude product was purified by flash
chromatography on silica gel (CH2Cl2/Et2O/AcOH 100:2:0.4). This
provided a colourless oil. Yield: 2.29 g (37%). 1H NMR (250 MHz,
CDCl3), ı: 0.76 (s, 3H, CH3), 0.80 (d, 3H, CH3, J = 6.39 Hz), 0.86 (m,
6H, 2 × CH3), 0.91 (s, 3H, CH3), 0.95–2.50 (m, 56H), 1.41 (s, 9H,
(CH3)3C–OCO), 4.54 (m, 1H, C(3)H), 4.92 (m, 1H, C(7)H), 5.31 (m, 1H,
Benzyl
3␣-benzyloxycarboxy-7␣,12␣-bis(decanoyloxy)-
5␤-cholate (21): Ester 20 (0.20 g, 0.32 mmol), BTEAC (0.03 g,
0.13 mmol) CaH2 (0.05 g, 1.19 mmol) and decanoyl chloride
(0.25 g, 1.31 mmol) dissolved in toluene (5 ml) were refluxed
for 2 h under argon. Then the mixture was poured into aqueous
AcOH (20 ml, 2.5%). The aqueous layer was washed with CH2Cl2
(3 × 20 ml), and combined organic extracts were washed with
brine and evaporated under reduced pressure. The residue (0.26 g,
87%) was used in the subsequent step without further purification.
1H NMR (250 MHz, CDCl3), ı: 0.69 (s, 3H, CH3), 0.80 (d, 3H, CH3,
J = 6.13 Hz), 0.83–0.90 (m, 6H, 2 × CH3), 0.92 (s, 3H, CH3), 0.95–2.45
(m, 57H), 4.46 (m, 1H, C(3)H), 4.92 (m, 1H, C(7)H), 5.09 (m, 1H,
C(12)H), 5.10 (2H, PhCH2), 5.13 (2H, PhCH2), 7.30–7.40 (m, 10H,
2 × Ph). 13C NMR (62.5 MHz, CDCl3), ı: 12.29, 14.22, 17.66, 22.56,
22.80, 23.10, 25.27, 25.45, 25.57, 26.89, 27.32, 28.85, 29.35, 29.39,
29.45, 29.47, 29.58, 29.60, 29.65, 30.97, 31.39, 31.51, 32.01, 32.02,
34.47, 34.71, 34.80, 35.04, 35.16, 38.04, 41.05, 43.41, 45.23, 47.63,
66.26, 69.51, 70.43, 75.03, 78.28, 128.33, 128.38, 128.45, 128.63,
128.67, 128.71, 135.46, 136.20, 154.75 (OCOO), 173.14 (COOR),
173.22 (COOR), 173.86 (COOR). HR-MS: for C59H87O9 [M−H]−
calculated: 939.6428 m/z; found: 939.6439 m/z.
Benzyl 3␣-acetoxy-7␣,12␣-bis(decanoyloxy)-5␤-cholate (23):
Ester 4 (3.00 g, 5.55 mmol), BTEAC (0.53 g, 2.33 mmol) CaH2 (0.97 g,
23.0 mmol) and decanoyl chloride (4.6 ml; 22.0 mmol) dissolved
in toluene (50 ml) were refluxed for 2 h under argon. Then the
mixture was poured into aqueous AcOH (50 ml, 6%). The aqueous
layer was washed with toluene (3 × 50 ml) and combined organic
extracts were washed with saturated aqueous NaHCO3 (2 × 50 ml)
and water (100 ml) and evaporated under reduced pressure. The
crude product was purified by flash chromatography on silica gel
(toluene/AcOEt 100:2 to 100:10). This provided a colourless oil.
Yield: 3.4 g (72%). 1H NMR (250 MHz, CDCl3), ı: 0.70 (s, 3H, CH3),
0.81 (d, 3H, CH3, J = 6.08 Hz), 0.88 (m, 6H, 2 × CH3), 0.92 (s, 3H, CH3),
0.95–2.45 (m, 57 H), 2.01 (s, 3H, CH3CO), 4.57 (m, 1H, C(3)H), 4.93
(m, 1H, C(7)H), 5.1 (m, 1H, C(12)H), 5.10 (2H, PhCH2), 7.32–7.37
C
(12)H). 13C NMR (62.5 MHz, CDCl3), ı: 12.17, 14.17, 17.66, 21.95,
22.48, 22.77, 22.90, 25.09, 25.19, 25.28, 26.67, 27.15, 28.20, 28.63,
29.23, 29.40, 29.52, 29.61, 30.86, 31.37, 31.97, 32.44, 34.37, 34.61,
34.75, 34.81, 34.95, 35.23, 37.94, 40.87, 43.20, 45.19, 47.76, 70.21,
1
73.39, 80.15, 80.74, 114.82 (q, CF3CO, JCF = 286.41 Hz), 156.79
2
(q, CF3CO, JCF = 41.93 Hz), 172.86 (COOR), 173.36 (COOR), 173.52
(COOR). HR-MS: for C50H82F3O8 [M−H]− calculated: 867.6040 m/z;
found: 867.6052 m/z.
3␣,7␣-Bis(decanoyloxy)-12␣-hydroxy-5␤-cholic acid (19): The
solution of ester 18 (2.28 g, 2.62 mmol) in THF (55 ml) and MeOH
(55 ml) was mixed with concentrated NH3(aq) (38 ml, 23%), and
the mixture was stirred at room temperature for 1 h, poured into
phosphate buffer (500 ml; 1 M; pH 7) and washed with AcOEt
(3 × 250 ml). Combined AcOEt extracts were evaporated under
reduced pressure and then dissolved in CH2Cl2 (35 ml). TFAcOH
(10 ml) was added at 12 ◦C and the mixture was stirred at room tem-
perature for 2 h. Then it was poured into aqueous AcONa (200 ml;
10%). The aqueous layer was washed with CH2Cl2 (3 × 100 ml).
Combined organic extracts were washed with brine (100 ml) and
evaporated under reduced pressure. The crude product was puri-
fied by flash chromatography on silica gel (CH2Cl2/Et2O/AcOH
800:32:3.2). This provided a white crystalline compound. Yield:
0.98 g (52%). Mp. 129–131 ◦C. 1H NMR (250 MHz, CDCl3), ı: 0.68
(s, 3H, CH3), 0.87 (m, 6H, 2 × CH3), 0.91 (s, 3H, CH3), 0.98 (d, 3H,
CH3, J = 6.04 Hz), 1.00–2.50 (m, 58 H), 4.00 (m, 1H, C(12)H), 4.58 (m,

L. Mrózek et al. / Steroids 76 (2011) 1082–1097
1093
(m, 5H, Ph). 13C NMR (62.5 MHz, CDCl3), ı: 12.25, 14.19, 17.65,
21.48, 22.61, 22.78, 23.08, 25.21, 25.47, 25.61, 26.98, 27.31, 28.79,
29.43, 29.45, 29.47, 29.53, 29.59, 29.65, 30.96, 31.36, 31.50, 32.00,
34.48, 34.78, 34.80, 35.04, 35.14, 35.19, 38.05, 41.04, 43.39, 45.19,
47.64, 66.25, 70.58, 74.09, 75.14, 128.32, 128.38, 128.66, 136.19,
170.55 (COOR), 172.91 (COOR), 173.01 (COOR), 173.85 (COOR).
HR-MS: for C53H83O8 [M−H]− calculated: 847.6166 m/z; found:
847.6175 m/z.
7␣,12␣-Bis(decanoyloxy)-3␣-hydroxy-5␤-cholic acid (22).
Method A: Benzyl ester 21 (0.10 g, 0.106 mmol) and HCOONH4
(0.06 g, 0.95 mmol) were dissolved in MeOH (5 ml), and 10% Pd/C
(0.01 g) was added. The mixture was stirred under argon for 1 h.
Then the suspension was filtered through celite and evaporated
under reduced pressure. The crude product was purified by flash
chromatography on silica gel (CH2Cl2/MeOH/AcOH 300:15:3). This
provided a colourless oil. Yield: 70 mg (92%).
3␣,7␣,12␣-Tris(decanoyloxy)-5␤-cholic acid (25): Benzyl ester
24 (4.04 g, 4.2 mmol) and HCOONH4 (1.09 g, 17.3 mmol) were dis-
solved in MeOH (40 ml), and 10% Pd/C (0.39 g) was added. The
mixture was stirred under argon for 3 h. Then the suspension was
filtered through celite and evaporated under reduced pressure. The
crude product was purified by flash chromatography on silica gel
(CH2Cl2/MeOH/AcOH 300:5:1). This provided a colourless oil. Yield:
3.62 g (99%). 1H NMR (250 MHz, CDCl3), ı: 0.73 (s, 3H, CH3), 0.83
(d, 3H, CH3, J = 6.00 Hz), 0.88 (m, 9H, 3 × CH3), 0.92 (s, 3H, CH3),
0.95–2.50 (m, 73H), 4.58 (m, 1H, C(3)H), 4.94 (m, 1H, C(7)H), 5.11
(m, 1H, C(12)H). 13C NMR (62.5 MHz, CDCl3), ı: 12.29, 14.20, 17.66,
22.61, 22.81, 23.10, 25.23, 25.48, 25.62, 27.02, 27.35, 28.80, 29.32,
29.42, 29.47, 29.48, 29.56, 29.59, 29.65, 29.73, 30.74, 31.08, 31.49,
32.00, 32.05, 34.49, 34.81, 34.84, 35.10, 35.16, 35.24, 38.07, 41.03,
43.41, 45.22, 47.63, 70.66, 73.77, 75.19, 172.95 (COOR), 173.07
(COOR), 173.45 (COOR), 179.75 (COOH). IR (cm−1): ꢀ(CH) 2923,
2854, ꢀ(C O) 1729, ı(CH) 1466, ꢀ(CO) 1247, 1174, 1103, 1008. Anal.
Calc. for C54H94O8 (871.61): 74.44% C, 10.87% H; found: 73.71% C,
11.83% H. HR-MS: for C54H93O8 [M−H]− calculated: 869.6870 m/z;
found: 869.6876 m/z.
Methyl 3␣-decanoyloxy-7␣,12␣-dihydroxy-5␤-cholate (26):
To benzyl ester 9 (2.51 g, 3.84 mmol) dissolved in THF (6 ml) and
MeOH (34 ml), MeONa (0.11 g, 2.04 mmol) was added stepwise,
and the solution was stirred at room temperature for 45 min.
Then it was poured into aqueous AcOH (100 ml, 0.5%). The aque-
ous layer was washed with CH2Cl2 (3 × 50 ml), and combined
organic extracts were washed with brine and evaporated under
reduced pressure. The crude product was purified by crystallisa-
tion from hexane. This provided a white crystalline compound.
Yield: 0.70 g (32%). Mp. 94–97 ◦C. 1H NMR (250 MHz, CDCl3), ı:
0.69 (s, 3H, CH3), 0.87 (m, 3H, CH3), 0.90 (s, 3H, CH3), 0.98 (d,
3H, CH3, J = 6.13 Hz), 1.00–2.50 (m, 43H), 3.66 (s, 3H, COOCH3),
3.85 (m, 1H, C(7)H), 3.97 (m, 1H, C(12)H), 4.57 (m, 1H, C(3)H).
13C NMR (62.5 MHz, CDCl3), ı: 12.68, 14.24, 17.48, 22.66, 22.80,
23.29, 25.21, 26.84, 26.87, 27.57, 28.50, 29.31, 29.40, 29.42, 29.56,
31.01, 31.18, 31.99, 34.56, 34.83, 34.96, 35.04, 35.30, 35.37, 39.67,
41.35, 42.19, 46.69, 47.36, 51.64, 68.41, 73.07, 74.12, 173.67
(COOR), 174.84 (COOR). IR (cm−1): ꢀ(OH) 3598, 3534, ꢀ(CH)
2928, 2863, ꢀ(C O) 1738, ı(CH) 1366, ꢀ(CO) 1167. Anal. Calc. for
Method B: The mixture of ester 23 (1.66 g, 1.92 mmol), i-PrOH
(17 ml) and aqueous NaOH (4.2 ml, 4%) was stirred at 80 ◦C for
30 min. It was poured into diluted aqueous AcOH (80 ml, 2.5%). The
aqueous layer was washed with CH2Cl2 (3 × 50 ml), and combined
organic extracts were washed with water (100 ml) and evaporated
under reduced pressure. The crude product was purified by flash
chromatography on silica gel (toluene/AcOEt/AcOH 100:10:1 to
100:15:1). This provided a colourless oil. Yield: 0.72 g (51%).
Method C: The mixture of methyl ester 28 (3.47 g, 4.75 mmol),
THF (27 ml), i-PrOH (26 ml) and aqueous NaOH (26 ml, 0.8%,
5.25 mmol NaOH) was stirred for 20 h under room temperature.
Then it was poured into diluted aqueous AcOH (250 ml, 1%). The
aqueous layer was washed with CH2Cl2 (3 × 100 ml), and combined
organic extracts were washed with water (100 ml) and evaporated
under reduced pressure. The crude product was purified by flash
chromatography on silica gel (CH2Cl2/MeOH/AcOH 300:5:1). This
provided a colourless oil. Yield: 2.55 g (75%). 1H NMR (250 MHz,
CDCl3), ı: 0.73 (s, 3H, CH3), 0.83 (d, 3H, CH3, J = 6.27 Hz), 0.88 (m, 6H,
2 × CH3), 0.91 (s, 3H, CH3), 0.95–2.45 (m, 58H), 3.48 (m, 1H, C(3)H),
4.93 (m, 1H, C(7)H), 5.11 (m, 1H, C(12)H). 13C NMR (62.5 MHz, CDCl3),
ı: 12.31, 14.21, 17.66, 22.66, 22.80, 23.11, 25.27, 25.48, 25.56,
27.35, 28.88, 29.40, 29.44, 29.48, 29.53, 29.64, 29.70, 30.75, 31.09,
31.62, 32.01, 32.03, 34.45, 34.85, 35.08, 35.21, 38.09, 39.11, 41.20,
43.44, 45.24, 47.63, 70.69, 71.86, 75.17, 173.26 (COOR), 179.50
(COOH). IR (cm−1): ꢀ(OH) 3429, ꢀ(CH) 2923, 2854, ꢀ(C O) 1727,
ı(CH) 1466, ꢀ(CO) 1247, 1176, 1073, 1015. Anal. Calc. for C44H76O7
(717.07): 73.70% C, 10.68% H; found: 72.81% C, 11.27% H. HR-MS: for
C
35H60O6 (576.84): 72.87% C, 10.48% H; found: 72.65% C, 10.95%
H. HR-MS: for C35H59O6 [M−H]− calculated: 576.4077 m/z; found:
576.4081 m/z.
Methyl 3␣,7␣-bis(decanoyloxy)-12␣-hydroxy-5␤-cholate (27):
The mixture of benzyl esters 11 and 12 (1.33 g, 1.65 mmol)
was mixed with THF (5 ml), MeOH (20 ml) and MeONa (0.09 g,
1.67 mmol) and stirred at room temperature for 30 min. Then
it was poured into aqueous AcOH (250 ml, 0.4%). The aqueous
layer was washed with CH2Cl2 (3 × 100 ml), and combined organic
extracts were washed with brine and evaporated under reduced
pressure. The crude product was purified by flash chromatogra-
phy on silica gel (petroleum ether/acetone/AcOH 800:100:10). This
yielded a crystalline material 0.95 g (79%) that was a mixture of
methyl 3␣,7␣-bis(decanoyloxy)-12␣-hydroxy-5␤-cholate (27, 90%
of mixture) and methyl 3␣,12␣-bis(decanoyloxy)-7␣-hydroxy-
5␤-cholate (10% of mixture). Crystallisation from toluene/hexane
mixture provided 0.6 g (50%) of 27 as a white crystalline compound.
Mp. 100–102 ◦C. 1H NMR (250 MHz, CDCl3), ı: 0.68 (s, 3H, CH3), 0.87
(m, 6H, 2 × CH3), 0.92 (s, 3H, CH3), 0.97 (d, 3H, CH3, J = 6.12 Hz),
1.00–2.50 (m, 57H), 3.65 (s, 3H, COOCH3), 3.99 (m, 1H, C(12)H), 4.58
(m, 1H, C(3)H), 4.90 (m, 1H, C(7)H). 13C NMR (62.5 MHz, CDCl3), ı:
12.67, 14.23, 17.51, 22.68, 22.81, 23.16, 25.22, 25.26, 26.85, 27.40,
28.29, 28.72, 29.31, 29.43, 29.55, 29.57, 29.67, 31.00, 31.22, 31.50,
32.01, 32.03, 34.50, 34.91, 35.11, 35.17, 38.29, 41.09, 42.29, 46.70,
47.43, 51.64, 70.69, 72.87, 73.88, 173.31 (COOR), 173.53 (COOR),
174.70 (COOR). IR (cm−1): ꢀ(OH) 3529, ꢀ(CH) 2921, 2851, ꢀ(C O)
1735, 1705, ꢀ(CO) 1167. Anal. Calc. for C45H78O7 (731.10): 73.93% C,
C
44H75O7 [M−H]− calculated: 715.5513 m/z; found: 715.5528 m/z.
Benzyl 3␣,7␣,12␣-tris(decanoyloxy)-5␤-cholate (24): Benzyl
cholate 1 (8.17 g, 16.4 mmol), BTEAC (1.44 g, 6.3 mmol) CaH2
(4.55 g, 108 mmol) and decanoyl chloride (20 ml, 96.4 mmol) dis-
solved in toluene (200 ml) were refluxed for 3 h under argon.
Then the mixture was poured into aqueous AcOH (200 ml, 6%).
The aqueous layer was washed with toluene (3 × 100 ml), and
combined organic extracts were washed with saturated aqueous
NaHCO3 (200 ml) and brine and evaporated under reduced pres-
sure. The crude product was purified by flash chromatography on
silica gel (toluene/Et2O/TEA 180:18:2). This provided a yellowish
oil. Yield: 15.6 g (99%). 1H NMR (250 MHz, CDCl3), ı: 0.63 (s, 3H,
CH3), 0.81 (d, 3H, CH3, J = 6.10 Hz), 0.87 (m, 9H, 3 × CH3), 0.91 (s,
3H, CH3), 1.00–2.50 (m, 73 H), 4.58 (m, 1H, C(3)H), 4.93 (m, 1H,
C(7)H), 5.10 (m, 1H, C(12)H), 5.1 (2H, PhCH2), 7.30–7.37 (m, 5H, Ph).
13C NMR (62.5 MHz, CDCl3), ı: 12.23, 14.18, 17.63, 22.57, 22.76,
22.78, 23.06, 24.82, 25.19, 25.44, 25.58, 26.99, 27.30, 28.77, 29.17,
29.28, 29.34, 29.39, 29.43, 29.45, 29.51, 29.56, 29.60, 29.67, 30.94,
31.34, 31.46, 31.97, 32.02, 34.04, 34.45, 34.76, 34.80, 35.11, 35.18,
38.03, 41.00, 43.37, 45.17, 47.63, 66.24, 70.62, 73.73, 75.16, 128.30,
128.36, 128.63, 136.17, 172.88 (COOR), 173.01 (COOR), 173.37
(COOR), 173.86 (COOR). HR-MS: for C61H99O8 [M−H]− calculated:
959.7418 m/z; found: 959.7429 m/z.

1094
L. Mrózek et al. / Steroids 76 (2011) 1082–1097
10.75% H; found: 73.73% C, 11.43% H. HR-MS: for C45H77O7 [M−H]−
calculated: 729.5669 m/z; found: 729.56468 m/z.
3␣-Hexadecanoyloxy-7␣,12␣-dihydroxy-5␤-cholic acid (31):
Benzyl ester 30 (4.57 g, 6.20 mmol) and HCOONH4 (3.07 g;
48.68 mmol) were dissolved in MeOH (150 ml), and 10% Pd/C
(0.11 g) was added. The mixture was stirred under argon for 1.5 h.
Then the second portion of 10% Pd/C (0.54 g) was added to the mix-
ture and it was stirred for next 2 h. After that the suspension was
filtered through celite and evaporated under reduced pressure. The
crude product was purified by flash chromatography on silica gel
(CH2Cl2/MeOH/AcOH 600:30:6). This provided a yellowish waxy
compound. Yield: 3.51 g (88%). 1H NMR (250 MHz, CDCl3), ı: 0.69
(s, 3H, CH3), 0.86 (m, 3H, CH3), 0.89 (s, 3H, CH3), 0.98 (d, 3H, CH3,
J = 5.73 Hz), 1.02–2.05 (m, 47H), 2.10–2.50 (m, 6H), 3.85 (m, 1H,
Methyl 7␣,12␣-bis(decanoyloxy)-3␣-hydroxy-5␤-cholate (28):
Benzyl ester 24 (6.01 g, 6.25 mmol), THF (30 ml), MeOH (60 ml)
and MeONa (0.36 g, 6.67 mmol) were mixed and stirred at room
temperature for 16 h. Then the mixture was poured into aque-
ous AcOH (250 ml, 1%). The aqueous layer was washed with
CH2Cl2 (3 × 100 ml), and combined organic extracts were washed
with brine and evaporated under reduced pressure. The crude
product was purified by flash chromatography on silica gel
(toluene/Et2O/TEA 160:40:2). This provided a colourless oil. Yield:
4.30 g (94%). 1H NMR (250 MHz, CDCl3), ı: 0.71 (s, 3H, CH3), 0.80
(d, 3H, CH3, J = 6.26 Hz), 0.87 (m, 6H, 2 × CH3), 0.89 (s, 3H, CH3),
0.95–2.50 (m, 57H), 3.46 (m, 1H, C(3)H), 3.64 (s, 3H, COOCH3), 4.91
(m, 1H, C(7)H), 5.08 (m, 1H, C(12)H). 13C NMR (62.5 MHz, CDCl3),
ı: 12.29, 14.21, 17.66, 22.67, 22.80, 23.10, 25.26, 25.47, 25.56,
27.35, 28.88, 29.39, 29.44, 29.47, 29.52, 29.54, 29.63, 29.68, 30.73,
30.96, 31.11, 31.61, 32.01, 34.44, 34.87, 35.07, 35.20, 38.06, 39.13,
41.18, 43.43, 45.21, 47.56, 51.60, 70.65, 71.82, 75.15, 173.23 (COOR),
174.57 (COOR). IR (cm−1): ꢀ(OH) 3507, ꢀ(CH) 1727, ꢀ(CO) 1247,
1171, 1075. Anal. Calc. for C45H78O7 (731.10): 73.93% C, 10.75% H;
C
(7)H), 3.99 (m, 1H, C(12)H), 4.56 (m, 1H, C(3)H), 5.30 (s, OH). 13C
NMR (62.5 MHz, CDCl3), ı: 12.62, 14.22, 17.43, 22.57, 22.80, 23.31,
24.85, 25.20, 26.70, 26.87, 27.63, 28.32, 29.22, 29.34, 29.44, 29.47,
29.61, 29.78, 29.82, 30.84, 31.17, 32.04, 34.16, 34.60, 34.85, 34.96,
35.05, 35.33, 35.42, 39.56, 41.35, 42.05, 46.66, 47.23, 68.56, 73.27,
74.20, 173.69 (COOR), 179.58 (COOH). IR (cm−1): ꢀ(OH) 3485, 3333,
ꢀ(CH) 2918, 2851, ꢀ(C O) 1723, 1709, ı(CH) 1464, ꢀ(CO) 1258,
1176, 1037. Anal. Calc. for C40H70O6 (646.98): 74.26% C, 10.91% H;
found: 73.94% C, 11.62% H. HR-MS: for C40H69O6 [M−H]− calcu-
found: 72.52% C, 11.38% H. HR-MS: for C45
lated: 729.5669 m/z; found: 729.56476 m/z.
H
77O7 [M−H]− calcu-
lated: 645.5094 m/z; found: 645.5101 m/z.
Benzyl 3␣-acetoxy-7␣,12␣-bis(hexadecanoyloxy)-5␤-cholate
Methyl 3␣,7␣,12␣-tris(decanoyloxy)-5␤-cholate (29): Benzyl
ester 24 (2.52 g, 2.62 mmol), THF (6 ml), MeOH (25 ml) and MeONa
(0.15 g, 2.78 mmol) were mixed and stirred at room temperature for
30 min. Then the mixture was poured into aqueous AcOH (250 ml,
2%). The aqueous layer was washed with CH2Cl2 (3 × 100 ml), and
combined organic extracts were washed with brine and evapo-
rated under reduced pressure. The crude product was purified by
flash chromatography on silica gel (toluene/Et2O/TEA 180:18:2).
This provided a colourless oil. Yield: 1.9 g (82%). 1H NMR (250 MHz,
CDCl3), ı: 0.66 (s, 3H, CH3), 0.75 (d, 3H, CH3, J = 6.2 Hz), 0.81 (m, 9H,
3 × CH3), 0.85 (s, 3H, CH3), 0.90–2.40 (m, 73H), 3.58 (s, 3H, COOCH3),
4.51 (m, 1H, C(3)H), 4.87 (m, 1H, C(7)H), 5.03 (m, 1H, C(12)H). 13C NMR
(62.5 MHz, CDCl3), ı: 12.31, 14.22, 17.71, 22.64, 22.81, 22.83, 23.13,
25.25, 25.51, 25.64, 27.05, 27.38, 28.83, 29.34, 29.44, 29.49, 29.52,
29.58, 29.61, 29.66, 29.74, 31.01, 31.16, 31.52, 32.02, 32.07, 34.52,
34.86, 34.90, 35.12, 35.18, 35.25, 38.09, 41.06, 43.44, 45.24, 47.64,
51.62, 70.66, 73.77, 75.21, 172.92 (COOR), 173.06 (COOR), 173.42
(COOR), 174.58 (COOR). IR (cm−1): ꢀ(CH) 2922, 2854, ꢀ(C O) 1729,
ı(CH) 1378, ꢀ(CO) 1171. Anal. Calc. for C55H96O8 (885.35): 74.61% C,
10.93% H; found: 73.45% C, 11.88% H. HR-MS: for C55H95O8 [M−H]−
calculated: 883.7027 m/z; found: 883.69904 m/z.
(32): Ester 4 (2.51 g, 4.64 mmol), BTEAC (0.42 g, 1.84 mmol),
CaH2 (0.79 g, 18.76 mmol) and hexadecanoyl chloride (5.5 ml,
18.15 mmol) dissolved in toluene (40 ml) were refluxed under
argon for 1 h. Then the mixture was poured into aqueous
AcOH (50 ml; 6%). The aqueous layer was washed with toluene
(3 × 50 ml), and combined organic extracts were washed with sat-
urated aqueous NaHCO3 (200 ml) and brine and evaporated under
reduced pressure. The crude product was purified by flash chro-
matography on silica gel (petroleum ether/acetone 800:165). This
provided a colourless oil. Yield: 3.18 g (67%). 1H NMR (250 MHz,
CDCl3), ı: 0.70 (s, 3H, CH3), 0.81 (d, 3H, CH3, J = 6.10 Hz), 0.88 (m, 6H,
2 × CH3), 0.92 (s, 3H, CH3), 0.95–2.50 (m, 80H), 2.00 (s, 3H, CH3CO),
4.57 (m, 1H, C(3)H), 4.93 (m, 1H, C(7)H), 5.10 (m, 1H, C(12)H), 5.10 (2H,
PhCH2), 7.30–7.40 (m, 5H, Ph). 13C NMR (62.5 MHz, CDCl3), ı: 12.25,
14.22, 17.65, 21.49, 22.60, 22.80, 23.09, 25.22, 25.47, 25.63, 26.98,
27.31, 28.78, 29.44, 29.48, 29.51, 29.55, 29.61, 29.67, 29.73, 29.78,
29.84, 30.95, 31.34, 31.50, 32.04, 34.48, 34.76, 34.80, 35.04, 35.16,
35.20, 38.05, 41.04, 43.38, 45.19, 47.63, 66.25, 70.58, 74.09, 75.14,
128.32, 128.37, 128.65, 136.19, 170.54 (COOR), 172.90 (COOR),
173.00 (COOR), 173.84 (COOR). HR-MS: for C65H107O8 [M−H]− cal-
culated: 1015.8044 m/z; found: 1015.8056 m/z.
Benzyl
3␣-hexadecanoyloxy-7␣,12␣-dihydroxy-5␤-cholate
7␣,12␣-Bis(hexadecanoyloxy)-3␣-hydroxy-5␤-cholic
acid
(30): Hexadecanoyl chloride (4.4 ml, 14.52 mmol) dissolved in
CH2Cl2 (20 ml) was slowly dropwise added to the solution of ben-
zyl cholate 1 (6.02 g, 12.07 mmol) and DMAP (1.80 g, 14.73 mmol)
in CH2Cl2 (40 ml) at 0 ◦C (ice bath) under argon. Then the mixture
was stirred and warmed to a room temperature. After 2 h it was
poured into aqueous AcOH (100 ml, 6%), and the aqueous layer
was washed with CH2Cl2. Combined organic layers were washed
with saturated aqueous NaHCO3 and water and evaporated
under reduced pressure. The crude product was purified by flash
chromatography on silica gel (toluene/AcOEt/AcOH 500:25:2.5 to
500:70:2.5). This provided a colourless oil. Yield: 5.20 g (58%). 1H
NMR (250 MHz, CDCl3), ı: 0.66 (s, 3H, CH3), 0.88 (m, 3H, CH3),
0.89 (s, 3H, CH3), 0.97 (d, 3H, CH3, J = 5.9 Hz), 1.00–2.50 (m, 54 H),
3.84 (m, 1H, C(7)H), 3.97 (m, 1H, C(12)H), 4.58 (m, 1H, C(3)H), 5.11
(2H, PhCH2), 7.30–7.37 (m, 5H, Ph). 13C NMR (62.5 MHz, CDCl3), ı:
12.61, 14.22, 17.44, 22.59, 22.79, 23.29, 25.20, 26.74, 26.88, 27.57,
28.40, 29.32, 29.42, 29.47, 29.60, 29.74, 29.77, 29.81, 30.97, 31.39,
32.03, 34.62, 34.82, 34.95, 35.05, 35.28, 35.33, 39.59, 41.36, 42.09,
46.66, 47.35, 66.22, 68.44, 73.13, 74.15, 128.26, 128.32, 128.63,
136.23, 173.60 (COOR), 174.15 (COOR). HR-MS: for C47H75O6
[M−H]− calculated: 735.5642 m/z; found: 735.5651 m/z.
(33): The mixture of ester 32 (2.94 g, 2.89 mmol), i-PrOH (25 ml)
and aqueous NaOH (6 ml; 4.3%) was stirred at 80 ◦C for 30 min.
Then it was poured into diluted aqueous AcOH (110 ml; 2.7%). The
aqueous layer was washed with CH2Cl2 (3 × 50 ml), and combined
organic extracts were washed with water (100 ml) and evapo-
rated under reduced pressure. The crude product was purified by
flash chromatography on silica gel (petroleum ether/Et2O/AcOH
375:125:10 to 50:150:4). This provided a colourless oil. Yield:
2.03 g (79%). 1H NMR (250 MHz, CDCl3), ı: 0.73 (s, 3H, CH3),
0.83 (d, 3H, CH3, J = 6.26 Hz), 0.88 (m, 6H, 2 × CH3), 0.91 (s, 3H,
CH3), 0.95–2.50 (m, 81H), 3.48 (m, 1H, C(3)H), 4.93 (m, 1H, C(7)H),
5.10 (m, 1H, C(12)H). 13C NMR (62.5 MHz, CDCl3), ı: 12.32, 14.23,
17.67, 22.66, 22.82, 23.11, 25.26, 25.49, 25.56, 27.33, 28.89,
29.41, 29.50, 29.54, 29.56, 29.65, 29.72, 29.76, 29.80, 29.85, 30.72,
31.09, 31.63, 32.06, 34.45, 34.82, 35.08, 35.20, 38.10, 39.13, 41.20,
43.43, 45.25, 47.64, 70.68, 71.86, 75.16, 173.24 (COOR), 179.49
(COOH). IR (cm−1): ꢀ(OH) 3435, ꢀ(CH) 2921, 2852, ꢀ(C O) 1729,
1709, ı(CH) 1466, ꢀ(CO) 1250, 1176, 1073, 1005. Anal. Calc. for
C56H100O7 (885.39): 75.97% C, 11.38% H; found: 75.31% C, 12.00%
H. HR-MS: for C56H99O6 [M−H]− calculated: 883.7391 m/z; found:
883.7403 m/z.

L. Mrózek et al. / Steroids 76 (2011) 1082–1097
1095
Benzyl 3␣,7␣,12␣-tris(hexadecanoyloxy)-5␤-cholate (34):
Benzyl cholate 1 (2.00 g, 4.01 mmol), BTEAC (0.36 g, 1.58 mmol),
CaH2 (1.01 g, 24.00 mmol) and hexadecanoyl chloride (7.5 ml,
24.70 mmol) dissolved in toluene (50 ml) were refluxed under
argon for 5 h. Then the mixture was poured into aqueous
AcOH (50 ml; 6%). The aqueous layer was washed with toluene
(3 × 50 ml), and combined organic extracts were washed with
saturated aqueous NaHCO3 (200 ml) and brine and evaporated
under reduced pressure. The crude product was purified by flash
chromatography on silica gel (toluene/AcOEt, 0–8%). This provided
a colourless oil. Yield: 2.95 g (60%). 1H NMR (250 MHz, CDCl3), ı:
0.70 (s, 3H, CH3), 0.81 (d, 3H, CH3, J = 6.08 Hz), 0.88 (m, 9H, 3 × CH3),
0.92 (s, 3H, CH3), 0.95–2.45 (m, 108H), 4.57 (m, 1H, C(3)H), 4.93
(m, 1H, C(7)H), 5.09 (m, 1H, C(12)H), 5.1 (2H, PhCH2), 7.30–7.37 (m,
5H, Ph). 13C NMR (62.5 MHz, CDCl3), ı: 12.27, 14.23, 17.68, 22.62,
22.82, 23.11, 25.25, 25.47, 25.64, 27.03, 27.34, 28.80, 29.36, 29.51,
29.58, 29.68, 29.73, 29.82, 29.85, 29.89, 30.97, 31.37, 31.51, 32.07,
34.50, 34.79, 34.85, 35.11, 35.17, 35.23, 38.08, 41.05, 43.40, 45.21,
47.67, 66.27, 70.63, 73.74, 75.18, 128.34, 128.39, 128.67, 136.21,
172.87 (COOR), 173.00 (COOR), 173.36 (COOR), 173.86 (COOR).
HR-MS: for C79H135O8 [M−H]− calculated: 1212.0235 m/z; found:
1212.0247 m/z.
3␣,7␣,12␣-Tris(hexadecanoyloxy)-5␤-cholic acid (35): Ester 34
(2.96 g, 2.44 mmol) and HCOONH4 (0.62 g, 9.83 mmol) were dis-
solved in MeOH (70 ml), and 10% Pd/C (0.30 g) was added. The
mixture was stirred under argon for 3 h. Then the suspension
was filtered through celite and evaporated under reduced pres-
sure. The crude product was purified by flash chromatography
on silica gel (CH2Cl2/MeOH/AcOH 300:5:1). This provided a white
crystalline compound. Yield: 2.56 g (93%). Mp. 53–55 ◦C. 1H NMR
(250 MHz, CDCl3), ı: 0.73 (s, 3H, CH3), 0.83 (d, 3H, CH3, J = 6.00 Hz),
0.88 (m, 9H, 3 × CH3), 0.92 (s, 3H, CH3), 1.00–2.50 (m, 109), 4.58
(m, 1H, C(3)H), 4.93 (m, 1H, C(7)H), 5.10 (m, 1H, C(12)H). 13C
NMR (62.5 MHz, CDCl3), ı: 12.31, 14.24, 17.69, 22.62, 22.83, 23.12,
25.26, 25.49, 25.65, 27.04, 27.35, 28.81, 29.36, 29.52, 29.59, 29.69,
29.74, 29.83, 29.86, 29.89, 30.73, 31.06, 31.51, 32.08, 34.51, 34.81,
34.86, 35.13, 35.18, 35.25, 38.10, 41.05, 43.41, 45.24, 47.67, 70.66,
73.77, 75.19, 172.93 (COOR), 173.05 (COOR), 173.43 (COOR), 179.61
(COOH). IR (cm−1): ꢀ(CH) 2916, 2849, ꢀ(C O) 1727, 1706, ı(CH)
1468, ꢀ(CO) 1248, 1171, 1071, 1008. Anal. Calc. for C72H130O8
(1123.80): 76.95% C, 11.66% H; found: 76.52% C, 12.35% H. HR-
MS: for C72H129O8 [M−H]− calculated: 1121.9687 m/z; found:
1121.9700 m/z.
(and zero ionic strength). The accuracy of calculations (according
to the vendor) for simple structures is usually better than 0.2–0.5
logarithmic units (for complex structures it is better than 0.5–1.0
logarithmic units). Solubility is not derived from log P and takes
into account not only the pH (solubility as a function of pH) but
compares the fragmental estimations with experimental material
from ca 6000 compounds databased.
4.3. In vitro screening of penetration enhancing activity and
sample analysis
Skin samples were obtained from porcine ear. Full thickness
dorsal skin was cut in fragments and stored at −20 ◦C until uti-
lized. Skin samples were slowly thawed (at 4 ◦C overnight and
then at ambient temperature) before each experiment. The pene-
tration enhancing activity of newly synthesized target compounds
was evaluated in vitro, using a vertical Franz diffusion cell (SES –
Analytical Systems, GmbH, Germany), with a donor surface area of
0.635 cm2 and receptor volume of 5.2 ml. The skin was mounted
between the donor and receptor compartments of the Franz diffu-
sion cell with the epidermal side up. The receptor compartment was
filled with phosphate buffered saline (pH 7.4) and was maintained
at 37 0.5 ◦C, while using circulating water bath. The receptor
compartment content was continuously stirred using a magnetic
stirring bar. The skin was kept in contact with the receptor phase
for 0.5 h prior to the experiment. The donor samples were pre-
pared by dissolving the tested enhancer (1 mg) in propylene glycol
(0.5 ml), and the solution of theophylline (5 mg) in water (0.5 ml)
was added. This mixture was shaken vigorously and then sonicated
for 10 min at 40 ◦C, then this stable system (dissolved theophylline
in enhancer emulsion) was applied to the skin surface and the donor
compartment of the cell was covered by Parafilm®. The control
samples were prepared in the same manner without enhancers.
Samples (0.5 ml) of the receptor phase were withdrawn at seven
pre-determined time intervals over 24 h (1, 2, 4, 6, 8, 12 and
24 h) and the cell was refilled with an equivalent amount of fresh
buffer solution. A minimum of three determinations was performed
using skin fragments from a minimum of two animals for each
compound. The samples were stored at −18 ◦C until analyzed by
HPLC.
Analysis of samples for theophylline content was performed
using an Agilent 1200 series HPLC system, equipped with a diode
array detection (DAD) system, a quaternary model pump and an
automatic injector (Agilent Technologies, Germany). Data acquisi-
tion was performed using ChemStation chromatography software.
A Waters Symmetry® C8 5 ␮m, 4.6 mm × 250 mm (Waters Corp.,
Milford, MA, USA) chromatographic column was used. A mix-
ture of MeCN (HPLC grade, 50.0%) and H2O (HPLC – Mili-Q
Grade, 50.0%) was used as a mobile phase. The total flow of
the column was 0.5 ml/min, injection 10 ␮l, column temperature
25 ◦C and sample temperature 10 ◦C. The detection wavelength of
280 nm was chosen. The retention time (tR) of theophylline was
5.07 0.05 min.
4.2. Lipophilicity and solubility determination/calculations
RM values were determined from the RP-18 TLC measurements.
The solution of a compound in CH2Cl2 was spotted on a TLC plate
(TLC silica gel 60 RP-18 F254s, 20 cm × 20 cm, Merck), 1.5 cm from
the edge. The volatiles were carefully evaporated, and the plate
was developed by MeOH/AcOH 100:1 solvent mixture for 2.5 h.
After drying, the plate was sprayed for the detection by solution of
Ce(SO4)2 in H2SO4 and heated. RM data were obtained from equa-
tion: RM = log(1/Rf − 1). RM values as well as log k, log kw or Kovats
retention indexes can be used as the lipophilicity index converted
to log P scale [1,56].
Log P values (i.e., the logarithm of the partition coefficient
for n-octanol/water) were predicted using ACD/Log P DB soft-
ware (ACD/Labs, ver. 12.01, Advanced Chemistry Development,
Inc., Toronto, ON, Canada, 2010). Log S values (as ACD/Labs aque-
ous log S at pH 7.4) were calculated by ACD/Log S DB software
(ver. 12.01, Advanced Chemistry Development, Inc., Toronto, ON,
Canada, 2010). The results are shown in Table 1.
The cumulative amounts of theophylline that penetrated
through the skin into the receptor compartment (␮g/cm2), were
corrected for sample removal, and plotted against time (h). An
approximately linear dependence was found (R2 ≥ 0.98), and steady
state fluxes (␮g/cm2/h) were calculated using the linear region of
the plots. ERs were calculated as ratios of theophylline flux with
and without the enhancer. The results are summarized in Table 2.
4.4. PAMPA experiments
ACD/Solubility DB (http://www.acdlabs.com/products/pc
admet/physchem/physchemsuite/) is a program that calculates
aqueous solubility values at any pH under the standard conditions
The penetration enhancing activity of newly synthesized final
compounds was evaluated in vitro, using a vertical PAMPA sys-
tem (BD GentestTMPre-Coated PAMPA Plate System, 96 wells,

1096
L. Mrózek et al. / Steroids 76 (2011) 1082–1097
http://www.bdbeurope.com). The donor samples were prepared
by dissolving the tested enhancer (16.9 ␮mol) and theophylline
(83.3 ␮mol) in ethanol (0.3 ml) and water (2.7 ml). This mixture was
shaken vigorously for 30 s and then sonicated for 10 min at 40 ◦C.
The control samples were prepared in the same manner without
enhancers. As a receptor phase carbonate buffer saline (pH 7.4)
was used. About 0.5 h before the experiment the PAMPA system
was taken out from the freezer and warm up to the room tempera-
ture. The receptor phase (200 ␮l/well) was pipetted into the upper
wells. The donor phase (stable system of dissolved theophylline in
enhancer emulsion/microsuspension) was pipetted into the lower
ones (300 ␮l/well). After 5 h 10 ␮l of the receptor phase was taken
from each well and mixed with physiological solution (990 ␮l). A
minimum of four determinations was performed. The samples were
stored at −18 ◦C until analyzed by HPLC.
4.6. Statistical analysis
All experiments were carried out in triplicate (ERs data for
skin and anti-proliferative activity) or quadruplicate (PAMPA). Data
were expressed as means SD. Differences were evaluated by one-
way ANOVA test completed by the Bonferroni’s multicomparison
test (ORIGIN PRO7). The differences were considered significant at
P = 0.05. Student’s t-test was used to detect statistical differences
(P = 0.001, P = 0.01 or P = 0.05) between the less active compound
and other compounds from the series.
Acknowledgments
This study was supported by GACR P304/11/2246, IGA VFU Brno
20/2010/FaF, grant No. ED0007/01/01 of the Centre of the Region
Hana for Biotechnological and Agricultural Research, the Ministry
of Education of the Czech Republic MSM6046137305 and 2B06024
(SUPRAFYT) and GACR P304/10/1951.
Analysis of samples for theophylline content was performed,
using a Waters Alliance 2695 XE HPLC separation module and
a Waters Photodiode Array Detector 2996 (Waters Corp., Mil-
ford, MA, USA). A Waters Symmetry® C8 5 ␮m, 4.6 mm × 250 mm
(Waters Corp., Milford, MA, USA) chromatographic column was also
used. The HPLC separation process was monitored by EmpowerTM
2 Chromatography Data Software, Waters 2009 (Waters Corp., Mil-
ford, MA, USA). The mixture of MeCN (HPLC grade, 50.0%) and H2O
(HPLC – Mili-Q Grade, 50.0%) was used as a mobile phase. The
total flow of the column was 0.5 ml/min, injection 10 ␮l, column
temperature 25 ◦C and sample temperature 10 ◦C. The detection
wavelength of 280 nm was chosen. The retention time (tR) of
theophylline was 5.07 0.05 min. ERs were calculated as ratios of
theophylline flux with and without the enhancer. The results are
summarized in Table 2.
References
[1] Kerns EH, Li D. Drug-like properties: concept structure design and methods.
San Diego, CA, USA: Elsevier; 2008.
[2] Hofmann AF, Mysels KJ. Colloid Surf B 1988;30:145.
[3] Kim KH, Singh BN. Drug delivery – oral route. In: Swarbrick J, Boylan JC, editors.
Encyclopedia of pharmaceutical technology. 2nd ed. New York, NY, USA: Marcel
Dekker; 2002. p. 899–900.
[4] Smith EW, Maibach HI. Percutaneous penetration enhancers. 2nd ed. Boca
Raton, FL, USA: CRC Press; 2006.
[5] Sharma P, Chawla HPS, Panchagnula R. Drugs Fut 1999;24:1221.
[6] Hosny EA, Khan-Ghilzaib NM, Elmazar MM. Pharm Acta Helv 1997;72:203.
[7] Reddy SM, Sinha VR, Reddy DS. Drugs Today 1999;35:537.
[8] Michael S, Thole M, Dillmann R, Fahr A, Drewe J, Fricker G. Eur J Pharm Sci
2000;10:133.
[9] Cernea S, Kidron M, Wohlgelernter J, Modi P, Raz I. Clin Ther 2004;26:2084.
[10] Sharma P, Varma MVS, Chawla H, Panchagnula PSR. Farmaco 2005;60:870.
[11] Mikov M, Fawcett JP, Kuhajda K, Kevresan S. Eur J Drug Metab Pharmacokinet
2006;31:237.
4.5. In vitro anti-proliferative/cytotoxicity assay
Dulbecco’s modified Eagle’s medium (DMEM), l-glutamine,
trypsin, penicillin and streptomycin were purchased from Sigma
(MO, USA); foetal bovine serum (FBS) and Calcein AM from Invit-
rogen (CA, USA). The cell lines used for screening, T-lymphoblastic
leukemia CEM, breast adenocarcinoma MCF7, and human foreskin
fibroblasts BJ, were obtained from the American Type Culture Col-
lection (Manassas, VA, USA). Cells were cultured in DMEM medium
(Sigma, MO, USA). All media used were supplemented with 10%
heat-inactivated foetal bovine serum, 2 mM l-glutamine, 10,000 U
penicillin and 10 mg/ml streptomycin. The cell lines were main-
tained under standard cell culture conditions at 37 ◦C and 5% CO2
in a humid environment. Cells were subcultured twice or three
times a week using the standard trypsinization procedure. Sus-
pensions with approximately 1.0 × 105 cells/ml (0.5 × 105 cells/ml
for BJ) were distributed in 96-well microtiter plates, and after 12 h
of stabilization the tested compounds were added at the desired
concentrations in Lutrol F 127. Control cultures were treated with
Lutrol F 127 alone, and the final concentration of Lutrol F 127 in the
reaction mixture never exceeded 0.5%. In most cases six serial 3-
fold dilutions of the test substances were added at time zero in 20 ␮l
aliquots to the microtiter plate wells, and the highest final concen-
tration in the wells was 37 ␮M. After incubation for 72 h, Calcein AM
solution (2 ␮M, Molecular Probes, Invitrogen, CA, USA) was added,
and the cells were incubated for a further hour. The fluorescence
from viable cells was then quantified, using a Fluoroskan Ascent flu-
orometer (Labsystems, Finland). The percentage of surviving cells in
each well was calculated from the equation IC50 = (OD drug exposed
well/mean OD control wells) × 100%. Each compound was tested in
triplicate, and the entire test was repeated at least three times. The
IC50 value, the concentration lethal for 50% of tumor cells of each
tested substance, was calculated from the obtained dose response
curves [57].
[12] Sang KK, Dong YL, Eunhye L, Yong KL, Choong YK, Hyun TM, et al. J Controlled
Release 2007;120:4.
[13] Duchateau GSMJE, Zuidema J, Merkus FWHM. Int J Pharm 1986;31:196.
[14] Xin HZ, Alain LWP. Int J Pharm 1991;69:29.
[15] Zhou XH, Po ALW. Int J Pharm 1991;69:29.
[16] Barry BW. Mol Aspects Med 1991;12:195.
[17] Carelli V, Dicolo G, Nannipieri E, Serafini MF. Int J Pharm 1993;89:81.
[18] Hadgraft J. Pharmaceutical skin penetration enhancers. New York, NY, USA:
Marcel Dekker; 1993.
[19] Lassmann-Vague V, Raccah D. Diabetes Metab 2006;32:513.
[20] Hassan N, Ahad A, Ali M, Ali J. Exp Opin Drug Deliv 2010;7:97.
[21] Jampilek J, Brychtova K. Med Res Rev 2011;31, doi:10.1002/med.20227.
[22] Araki YI, Lee S, Sugihara G, Furuichi M, Yamashita S, Ohseto F. Colloid Surf B
1996;8:81.
[23] Hishikawa Y, Sada K, Watanbe R, Miyata M, Hanabusa K. Chem Lett 1998:795.
[24] Valenta C, Nowack E, Bernkop-Schnurch A. Int J Pharm 1999;185:103.
[25] Willemen HM, Vermonden TA, Marcelis TM, Sudholter EJR. Eur J Org Chem
2001:2329.
[26] Willemen HM, Vermonden T, Marcelis ATM, Sudholter EJR. Langmuir
2002;18:7102.
[27] Dukh M, Saman D, Kroulik J, Cerny I, Pouzar V, Kral V, et al. Tetrahedron
2003;59:4069.
[28] Valkonen A, Lahtinen M, Virtanen E, Kikkonen S, Kolehmainen E. Biosens Bio-
electron 2004;20:1233.
[29] Brychtova K, Jampilek J, Opatrilova R, Raich I, Csollei J. Bioorg Med Chem
2010;18:73.
[30] Brychtova K, Opatrilova R, Raich I, Kalinowski DS, Dvorakova L, Placek L, et al.
Bioorg Med Chem 2010;18:8556.
[31] Franz TJ. J Invest Dermatol 1975;64:190.
[32] Kansy M, Senner F, Gubernator K. J Med Chem 1998;41:1007.
[33] Avdeef A, Tsinman O. Eur J Pharm Sci 2006;28:43.
[34] Anelli LP, Lattuada L, Uggeri F. Synth Commun 1998;28:109.
[35] Bonar-Law RP, Davis AP, Sanders JKM. J Chem Soc Perkin Trans 1990;1:2245.
[36] Schwarz V, Pihera P, Halaskova J. Ceskoslov Farm 1984;8:327.
[37] Wess G, Kramer W, Bartmann W, Enhsen A, Glombik H, Mullner S, et al. Tetra-
hedron Lett 1992;33:195.
[38] Dewitte RS. Mod Drug Discov 2004;7:41.
[39] Ni N, El-Sayed MM, Sanghvi T, Yalkowsky SH. J Pharm Sci 2000;89:1620.
[40] Katz M, Ben-Shlush I, Kolusheva S, Jelinek R. Pharm Res 2006;23:580.
[41] Fang JY, Tsai TH, Hung CF, Wong WW. J Pharm Pharmacol 2004;56:1493.

L. Mrózek et al. / Steroids 76 (2011) 1082–1097
1097
[42] Sloan KB, Beall HD, Taylor HE, Getz JJ, Villaneuva R, Nipper R, et al. Int J Pharm
1998;171:185.
[43] Evrard D, Touitou E, Kolusheva S, Fishov Y, Jelinek R. Pharm Res 2001;18:943.
[44] Williams AC, Barry BW. Int J Pharm 1989;56:43.
[50] Suzuki H, Sunada H. Chem Pharm Bull 1998;46:125.
[51] Da Silva RC, Spitzer M, Da Silva LHM, Loh W. Thermochim Acta 1999;328:161.
[52] Devinsky F, Kopecka-Leitmanova A, Sersen F, Balgavy P. J Pharm Pharmacol
1990;42:790.
[45] Yamane MA, Williams AC, Barry BW. J Pharm Pharmacol 1995;47:978.
[46] Jacobi U, Kaiser M, Toll R, Mangelsdorf S, Audring H, Otberg N, et al. Skin Res
Technol 2007;13:19.
[47] Herkenne C, Naik A, Kalia YN, Hadgraft J, Guy RH. Pharm Res 2006:1823–50.
[48] Meyer W, Schwarz K, Neurand KT. Curr Probl Dermatol 1978;7:39.
[49] Wilkinson SC, Maas WJM, Nielsen JB, Greaves LC, Sandt JJM, Williams FM. Int
Arch Occup Environ Health 2006;79:405.
[53] Balgavy P, Devinsky F. Adv Colloids Interface Sci 1996;66:23.
[54] Sarapuk J, Kubica K. Cell Mol Biol Lett 1998;5:261.
[55] Kralova K, Sersen F, Devinsky F, Lacko I. Tenside Surf Det 2010;47:288.
[56] Pliska V, Testa B, van de Waterbeemd H. Lipophilicity in drug action and toxi-
cology. Germany: VCH Weinheim; 1996.
[57] Krystof V, McNae IW, Walkinshaw MD, Strnad M. Cell Mol Life Sci
2005;62:1763.