Synthesis and comparative spectroscopic analysis of two chenodeoxycholic acid (CDCA) derivatives with closely related 7α-ester moieties

Article abstract of DOI:10.1016/j.tetlet.2008.11.052

3α-Hydroxyl-7α-(4-pentenoyloxy)-5β-cholanoic acid (5) has been synthesized in four step reactions starting from CDCA. The serendipitous synthesis of methyl 3α-(ethoxycarbonyloxy)-7α-(allyloxycarbonyloxy)-5β-cholanoate (7) has led us to compare the spectroscopic difference of the 7α-(allyloxycarbonyloxy) group versus the 7α-(4-pentenoyloxy) group. The molecular structures of these compounds were confirmed by X-ray crystallography. Methyl 3α-(ethoxycarbonyloxy)-7α-(allyloxycarbonyloxy)-5β-cholanoate was obtained by a method, which may prove useful in the synthesis of ¹⁴C-labeled derivatives for metabolic studies.

Full text of DOI:10.1016/j.tetlet.2008.11.052

Tetrahedron Letters 50 (2009) 503–505
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Synthesis and comparative spectroscopic analysis of two chenodeoxycholic
acid (CDCA) derivatives with closely related 7a-ester moieties
Xinyan Bai a, Charles Barnes b, Jerry Ray Dias a,
*
a Department of Chemistry, University of Missouri-Kansas City, 5100 Rockhill Road, Kasas City, MO 64110, United States
b Department of Chemistry, University of Missouri—Columbia, Columbia, MO 65211, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
3
a
-Hydroxyl-7
starting from CDCA. The serendipitous synthesis of methyl 3
yloxy)-5b-cholanoate (7) has led us to compare the spectroscopic difference of the 7
oxy) group versus the 7 -(4-pentenoyloxy) group. The molecular structures of these compounds were
confirmed by X-ray crystallography. Methyl 3
a
-(4-pentenoyloxy)-5b-cholanoic acid (5) has been synthesized in four step reactions
-(ethoxycarbonyloxy)-7 -(allyloxycarbon-
-(allyloxycarbonyl-
Received 8 September 2008
Revised 13 November 2008
Accepted 14 November 2008
Available online 20 November 2008
a
a
a
a
a-(ethoxycarbonyloxy)-7a-(allyloxycarbonyloxy)-5b-cho-
lanoate was obtained by a method, which may prove useful in the synthesis of 14C-labeled derivatives
for metabolic studies.
Ó 2008 Elsevier Ltd. All rights reserved.
1. Introduction
drug delivery systems based on host/guest interactions. We have
serendipitously synthesized a CDCA derivative (7) with ambient
In recent years, there has been increased interest in developing
prodrugs to target various transporters. Some peptides and oral
drugs have been attached to bile acids to give prodrugs to improve
their bioavailability.1 Acyclovir which is an antiviral drug with a
poor intestinal permeability has been linked to acidic group of C-
24 of CDCA (chenodeoxycholic acid) to give a prodrug to target
hASBT (the human apical sodium-dependent bile acid transporter).
The oral administration of this prodrug showed that the bioavail-
ability of acyclovir ester was enhanced two times compared to that
of acyclovir alone.2,3 The use of peptide drugs is limited by their
poor oral absorption and enzymatic hydrolysis in GI tract. Drug
substrates conjugated to a compound which can be transported
efficiently by intestinal carrier system is one of the practical strat-
egies to overcome the drawback of peptide drugs. Swaan et al.4 at-
tached a peptide moiety to the 24-carboxylic acid group of cholic
acid using automatic peptide synthesizer. This resulted in in-
creased intestinal absorption and metabolic stability. Based on
the concept that hASBT is the major transporter of bile acids,
designing bile acid conjugates represents a strategy for drug deliv-
ery. Though 3H-labeled taurocholic acid has been used to study the
transportation mechanism of hASBT,5,6 the mechanism of interac-
tion of hASBT with bile acids is still not fully understood. In our
group, work has been focused on the selective etherization and
CO2 that offers a pathway to make 14C-labeled bile acid derivatives,
which can be used to explore the transport mechanism of hASBT
through labeling with 14CO2. In addition, for most bile acid based
prodrugs, the drug substrate is usually covalently coupled to C24
or C3-position, with little effort directed toward attaching drugs
to C7-position.7 The selective functionalization of 7
a-OH group
to give the novel derivative 5 could inspire the development of pro-
drugs with drug conjugated at C7 and further study their effective-
ness and bioavailability.
2. Results and discussion
The synthesis of compound 5 is shown in Scheme 1. The carbox-
ylic acid on CDCA (1) was methyl esterified to give ester 2.8 Then,
the hydroxyl group on 3
tected by reacting with ethyl chloroformate in dried pyridine at
À20 °C to form 7 -monohydroxyl ester 3, using the previous pub-
lished conditions.9 Ester 4 was produced by reacting 7
-hydroxyl
a-position of ester 2 was selectively pro-
a
a
group of 3 with 4-pentenoic acid, 2,6-dichlorobenzoyl chloride,
and DMAP via Yamaguchi reaction.10 Careful saponification of ester
4 gave the desired 3a-hydroxy-7a-(4-pentenoyloxy)-5b-cholanoic
acid (5), which is an important intermediate for further macrolact-
onization and ring-closing metathesis by using Grubbs’ catalyst.11
1H NMR and 13C NMR assignments of the important functional
groups of 5 were identified by applying 1H, decoupled 13C, 2D H–H
COSY and coupling 13C NMR techniques (cf. with Supplementary
data).
esterification of sterically hindered 7a-hydroxy group of CDCA to
make some novel CDCA derivatives as moieties for eventual con-
struction of macro-cyclomers, which represent another venue for
The synthesis and proposed formation of ester 7 are described
in Scheme 2. The route for synthesizing methyl 3
a-(ethoxycarbon-
* Corresponding author. Tel.: +1 816 276 2284; fax: +1 816 235 5502.
E-mail address: diasj@umkc.edu (J.R. Dias).
yloxy)-7 -(allyloxy)-5b-cholanoate (8) was designed by silylation
a
0040-4039/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2008.11.052

504
X. Bai et al. / Tetrahedron Letters 50 (2009) 503–505
The serendipitous synthesis of compound 7 suggests that the
OH
steric hindrance of alkoxide ion on 6a impedes its direct reaction
with allyl bromide compared to the more acidic CO2. The alkoxide
ion thus reacts with CO2, which is more acidic and smaller to
form more extended, less sterically hindered carboxylate anion
6b which can rapidly attack allyl bromide via a typical SN2 mech-
anism to afford compound 7. We have tried to synthesize 8 by
rigorous exclusion of atmospheric CO2. When reaction was car-
ried out either under N2 gas or in sealed glassware with degassed
reaction mixture, we failed to get even a little 8. This indicates
O
a
O
O
OH
OH
95%
b
OH
OH
91%
2
1
O
O
O
c
O
OH
O
O
O
O
94%
O
O
a
-OÀ attached
O
that it may not be possible to obtain 8 by using 7
on steroid skeleton to attack allyl bromide via SN2 mechanism
due to steric hindrance of 7 -position, where 7
-OÀ is located
O
3
4
a
a
in the concave face of steroid skeleton. Since TBAF is very hygro-
scopic, it has been difficult to remove water molecules completely
from it. The yield of compound 7 is only 30%. Large amount of
product 3 was recovered due to the presence of even trace
amount of water in reaction system. Synthesis of 7 has to be pro-
ceeded with THF freshly distilled from sodium and potassium al-
loy and anhydrous TBAF, which is dried by following procedure in
the literature.14 The use of TBAT and cesium fluoride which are
available in anhydrous form failed to deprotect the silyl group
on 6, probably, due to the steric bulk of TBAT and the low solu-
bility of CsF in THF, Although TBAT and CsF are very useful as a
good source to provide dry FÀ.15–18
OH
O
O
O
OH
5
Scheme 1. Synthesis of 5. Reagents and conditions: (a) AcCl, CH3OH, at 0 °C, 1 h; (b)
ClCOOEt, py at À20 °C, 1 h; (c) 2,6-dichlorobenzoyl chloride, 4-pentenoic acid, Et3N,
DMAP, rt, 24 h; (d) K2CO3, CH3OH, THF, reflux, 12 h; (e) 3 M HCl.
Single-crystal structures of 5 and 7 were determined by X-ray
diffraction as shown in Figures 1 and 2.19 The crystal of 5 is ortho-
O
O
O
O
O
rhombic (P 21 21 21 space group), and 7a-OH has been replaced by
OH
Si
i
O
O
4-pentenoyloxy group, therefore, there is no hydrogen bond in-
O
O
O
95%
volved in 7
work of
structure.20 The 3
a
5
-position, which makes the hydrogen-bonding net-
less complex than that of CDCA single-crystal
O
3
6
a
-hydroxyl and 24-carboxylic acid groups pro-
vide two hydrogen-bonding sites, and each site forms two hydro-
gen bonds with other two molecules. Each molecule is
5
O
O
O
interconnected with four molecules via two different hydrogen
bonds (O1?O4: 2.73 Å, O5?O1: 2.61 Å) in Figure 1a. O4 on carbox-
ylic acid group is hydrogen bond acceptor, and O5–H of carboxylic
O
O
O
O
iii
O
O
O
O
O
6a
O
6b
O
acid is hydrogen bond donor, but O1 on 3a-OH group acts as both
v
CH2
Br
hydrogen bond donor and acceptor. The distance of hydrogen bond
between H10 and O4 is 1.94 Å, and between H50 and O1 is 1.85 Å as
shown in Figure 1a. The crystal structure of compound 7 is mono-
clinic (C2) and tightly compacted and arrayed by extensive inter-
molecular van der Waals interactions as shown in Figure 2. The
distances of H4ÁÁÁH19A, O6ÁÁÁH6A, O1ÁÁÁH19, and H31CÁÁÁO7 are 2.36 Å,
2.36 Å, 2.54 Å, 2.64 Å, respectively, as shown in Figure 2a. These
intermolecular hydrophobic interactions are a significant driving
force to make compound 7 form a single crystal (Fig. 2) even
i
CH2
Br
O
O
O
O
O
O
O
O
O
O
O
O
O
O
7
8
though the 3a-OH, 7a-OH, and 24-carboxylic acid functional
Scheme 2. Synthesis and proposed formation of 7. Reagents and conditions: (i)
TMCS, imidazole, THF, rt, 12 h; (ii) TBAF, THF, rt, 10 h; (iii) CO2 in the air; (iv) allyl
bromide, rt, 6 h.
groups are replaced by three different ones. There are distinctive
differences in the ethenyl moieties belonging to the 4-pentenoyl
(5) and allyloxycarbonyl (7) groups. In the 1H NMR and 13C NMR
spectra, both the methylene and the methine protons and carbons
of the latter are more deshielded due to the influence of the extra
oxygen. In the X-ray crystal structures, the carbonyl moiety in the
former is directed toward, and in the latter away from the steroid
skeleton. Also, the carbonyl moiety tends to be anti to the 7b-
hydrogen in the former and syn in the latter.
In summary, the structures of 5 and 7 have been fully character-
ized by 1H NMR, 13C NMR, HRFAB, and X-ray crystallography. Fur-
ther efforts will be focused on improving the yield of 7 and on
applying this method to other compounds with hydroxyl func-
tional groups. A novel method has evolved for attaching allyloxy-
of 7a-OH on 7a-monohydroxyl ester 3 with TMCS and imidazole in
THF,12 followed by dried TBAF desilylation of 6 to presumably pro-
duce an alkoxide intermediate, which undergoes SN2 with allyl
bromide to afford allyl ether.13 The results of mass spectroscopy
further informed us that the molecular weight of this compound
(562.3492) is 44 mass units higher than that of anticipated product
(8) with MW (518.36). The 44 mass units difference is consistent
with CO2 insert between alkoxide ion and allyl group. X-ray crys-
tallography proved that CO2 is really involved in this reaction.
Apparently, the compound was methyl 3
a
-(ethoxycarbonyloxy)-
-(allyloxycarbonyloxy)-5b-cholanoate (7) instead of methyl
-(ethoxycarbonyloxy)-7 -(allyloxy)-5b-cholanoate (8) synthe-
carbonyloxy group to 7a-position via first deprotecting silyl
7
3
a
a
group, then introducing CO2, and finally adding allyl bromide into
reaction system. This route may find application in 14C-labeled
substrates of drugs via 14CO2. Compound 7 and its derivatives will
a
sized by following the designed route.

X. Bai et al. / Tetrahedron Letters 50 (2009) 503–505
505
Figure 1. Single-crystal X-ray structure of 5. (a) Distances (in Å) of the hydrogen bonds observed in 5. (b) Top view of hydrogen bond network formed by five molecules of 5.
Figure 2. Single-crystal X-ray structure of 7. (a) Distances (in Å) of intermolecular interaction observed in the 7. (b) Top view of unit cell and packing of 7 from b-axis.
6. Weinman, S. A.; Carruth, M. W.; Dawson, P. A. J. Biol. Chem. 1998, 273, 34691–
have a lot of potential applications in the study of the metabolic
mechanism of drugs in human’s liver.
34695.
7. Kramer, W.; Schuber, G.; Bickel, M.; Hoffmann, A.; Baringhaus, K. H., et al Falk
Symp. 1993, 68, 161–176.
8. Gao, H.; Dias, J. R. Eur. J. Org. Chem. 1998, 719–724.
9. Gao, H.; Dias, J. R. Org. Prep. Proced. Int. 1999, 31, 145–166.
10. Dhimitruka, I.; SantaLucia, J., Jr. Org. Lett. 2006, 8, 47–50.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.tetlet.2008.11.052.
11. Lee, C. W.; Choi, T.-L.; Grubbs, R. H. J. Am. Chem. Soc. 2002, 124, 3224–3225.
12. Li, Y.; Dias, J. R. Synthesis 1997, 425.
13. Marzinzik, A. L.; Felder, E. R. Molecules 1997, 2, 17–30.
14. Wiles, C.; Watts, S.; Haswell, S. J.; Villar, E. P. Lab On a Chip 2001, 100.
15. Aggarwal, V. K.; Alonso, E.; Ferrara, M.; Spey, S. E. J. Org. Chem. 2002, 67, 2335–
2344.
References and notes
16. Pilcher, A. S.; DeShong, P. J. Org. Chem. 1996, 61, 6901.
17. Poissonnet, G.; Theret-Bettiol, M.-H.; Dodd, R. H. J. Org. Chem. 1996, 61, 2273–
2282.
1. Sievanen, E. Molecules 2007, 12, 1859–1889.
2. Tolle-Sander, S.; Lentz, K. A.; Maeda, D. Y.; Coop, A.; Polli, J. E. Mol. Pharmaceut.
2004, 1, 40–48.
18. Fraser, C. L.; Anastasi, N. R.; Lamba, J. J. S. J. Org. Chem. 1997, 62, 9314–9317.
19. Cambridge Crystallographic Data Centre: Allocated the deposition numbers
CCDC 701461 and CCDC 701462.
20. Alvarez, M.; Jover, A.; Carrazana, J.; Meijide, F.; Soto, V. H.; Vázquez Tato, J.
Steroids 2007, 72, 535–544.
3. Balakrishnan, A.; Polli, J. E. Mol. Pharmaceut. 2006, 3, 223–230.
4. Swaan, P. W.; Hillgren, K. M.; Szoka, F. C., Jr.; Oie, S. Bioconj. Chem. 1997, 8, 520–
525.
5. Alrefai, W. A.; Sarwar, Z.; Tyagi, S.; Saksena, S.; Dudeja, P. K.; Gill, R. K. Am. J.
Physiol. Gastrointest. Liver Physiol. 2005, 288, G978–G985.