Synthesis and characterization of new liver targeting 5-fluorouracil-cholic acid conjugates

Article abstract of DOI:10.1002/ardp.200900075

The objective of this work was to develop a liver-specific antihepato carcinoma agent. A series of 5-fluorouracil / cholic acid conjugates (5-FU-cholic acid conjugates) were prepared and tested for their chemical characteristics and bio-distribution properties. The in-vitro stability trial showed 5-FU-cholic acid conjugates could be completely hydrolyzed by heating at 70°C in an acidic solution, pH = 1, for 5 min. The fast and complete hydrolysis of these compounds could be compatible with a fast separation and analysis method to shorten the analysis time. The decomposition speeds of the 5-FU-cholic acid conjugates in different organs of mice at several time points after oral administration were evaluated by measuring the concentrations of regenerated 5-FU in organ tissue. The results were compared with those of the controls, which was a group of mice orally taking 5-FU. The concentrations of 5-FU in mice liver tissue were remarkably increased after oral administration of the prodrugs, and were much larger than if only orally administered 5-FU. The results suggested the feasibility to improve therapeutic efficiency of liver targeting treatments by using cholic acid as the vector of drugs.

Full text of DOI:10.1002/ardp.200900075

Arch. Pharm. Chem. Life Sci. 2009, 342, 513–520
S. Qian et al.
513
Full Paper
Synthesis and Characterization of New Liver Targeting
5-Fluorouracil-Cholic Acid Conjugates
Shan Qian, Jian-Bo Wu, Xiao-Chun Wu, Jie Li, and Yong Wu
Key Laboratory of Drug Targeting, West China School of Pharmacy, Sichuan University, Chengdu, Sichuan,
P.R. China
The objective of this work was to develop a liver-specific antihepato carcinoma agent. A series of
5-fluorouracil / cholic acid conjugates (5-FU-cholic acid conjugates) were prepared and tested for
their chemical characteristics and bio-distribution properties. The in-vitro stability trial showed
5-FU-cholic acid conjugates could be completely hydrolyzed by heating at 708C in an acidic solu-
tion, pH = 1, for 5 min. The fast and complete hydrolysis of these compounds could be compat-
ible with a fast separation and analysis method to shorten the analysis time. The decomposition
speeds of the 5-FU-cholic acid conjugates in different organs of mice at several time points after
oral administration were evaluated by measuring the concentrations of regenerated 5-FU in
organ tissue. The results were compared with those of the controls, which was a group of mice
orally taking 5-FU. The concentrations of 5-FU in mice liver tissue were remarkably increased
after oral administration of the prodrugs, and were much larger than if only orally adminis-
tered 5-FU. The results suggested the feasibility to improve therapeutic efficiency of liver target-
ing treatments by using cholic acid as the vector of drugs.
Keywords: Cholic acid / 5-Fluorouracil / Liver / Prodrugs / Targeting /
Received: April 7, 2009; accepted: April 29, 2009
DOI 10.1002/ardp.200900075
desirable target organs in the body due to its specific
mammalian hepatocyte receptor [2].
Introduction
Cholic acid 1 is a specific hepatic targeting delivery sys-
tem for oral administration. Every day, the molecule cir-
culates 6 to 10 times in the human body, with only 2-5%
runs off [3]. The existence of an active and specific Na+-
dependent cholic acid transportation system in the sinus-
oidal membrane of hepatocytes suggests cholic acid as a
putative shuttle delivering specific drugs to liver cell
[4, 5]. Based on this highly specific hepatic targeting char-
acteristic, cholic acid is possibly a good candidate for a
drug carrier with great potential.
5-Fluorouracil (5-FU) 2 is a classic anti-tumor drug
[6, 7], which plays a very important role in common
malignant tumor therapeutic practice, especially for
liver carcinomas [8, 9]. However, despite of many advan-
tages, chemotherapy using 5-FU is limited due to its short
half-life, wide distribution, low selectivity, and various
side effects [10–12]. For many years, researchers have
been seeking for derivatives of 5-FU with good therapeu-
tic efficacy and low toxicity [13, 14]. It was reported that
Chemotherapy using cytotoxic anticancer agents has
been restricted in clinical application due to the lack of
selectivity and specificity towards the target sites, thus
causing serious toxic side effects. If the distribution of
chemotherapeutic agents could be directed to specific
sites, their therapeutic effects could be greatly enhanced
while the toxic side effects would be minimized. Cur-
rently, one strategy to improve drugs' selectivity is to cou-
ple the therapeutics with ligands that can recognize spe-
cific receptors. That could increase the interactions
between malignant cells and a therapeutic moiety, and,
at the same time, decrease toxic effects on normal cells
[1]. In this respect, the liver should be one of the most
Correspondence: Yong Wu, Key Laboratory of Drug Targeting, West
China School of Pharmacy, Sichuan University, No. 17, Block 3, Southern
Renmin Road, Chengdu, Sichuan 610041, P. R. China.
E-mail: tgxx903@163.com
Fax: +86 28 855-03666
i 2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

514
S. Qian et al.
Arch. Pharm. Chem. Life Sci. 2009, 342, 513–520
Compounds like N-alkyl-5-fluorouacil derivatives, for
example, have a strong N–C bond which prevents the
derivatives from being easily hydrolyzed in vivo and thus
prevents the compounds showing any antitumor activ-
ity. If an oxygen atom was introduced into the a-position
to the alkyl group, the N–C bond of these derivatives
would become more labile to hydrolytic condition and
then showed moderate antitumour activity [18]. Based on
this knowledge and in order to retain the 5-FU's antitu-
mor activity after coupling it with cholic acid, we intro-
duced an oxymethyl group at the N-1 position of 5-FU as a
spacer.
Similarly, to obtain the liver targeting property, the
choric acid molecule must satisfy the following specific
structural features [19]: there must be an electronegative
group (such as carboxyl) in its side-chain, and there must
be at least one hydroxyl group on the ring. Therefore, we
selected the most highly active site, hydroxyl-3 on the ste-
roid ring, as connection point with the spacer and 5-FU
[20]. In addition, because of the steric hindrance of the
steroid ring, we extended the hydroxyl-3 and trans-
formed it with ethylene glycol into a more active primary
alcohol. As a saturated carbon chain, the introduced
ethylene could spin freely, and can provide enough space
for an interval. The interval is necessary to ensure that
the ligand is connected to the original molecule (5-FU)
not to impede interaction between the transfer molecu-
lar (cholic acid) and the transferred molecular (5-FU), so
as to ensure the liver-targeting property of the drug.
We chose the ester bond to connect the modified 5-FU
with cholic acid, as this bond can ensure the successful
release of the original drug (5-FU) from the prodrug after
its arrival at the target site on the liver.
Figure 1. Structure of compounds 1, 2, and 3.
liposomes-encapsulated 5-FU could be selectively accu-
mulated in the liver [15].
In this study, a series of spacers with different lengths
were used to link these two molecule parts, cholic acid
and 5-FU, together, so as to produce liver-targeting pro-
drugs with good hydrophobicity, high selectivity, and
low toxicity (Fig. 1). We synthesized a series of 5-FU-cholic
acid conjugates 3a–e, and measured their chemical
stability in vitro and their bio-distribution in vivo. We
hypothesize that the 5-FU-cholic acid conjugates will be
targeted to and interact with hepatoma cells selectively
and specifically.
As shown in Scheme 1, carboxylated 5-FU 7a–e were
synthesized from 5-FU 2. First, we performed the N-
hydroxylmethylation of 5-FU 2 with formaldehyde and
condensation with diacid monobenzyl esters to get 6a–e.
Second, we catalyzed hydrogenolysis of 6a–e to get car-
boxylated 5-FU 7a–e [21]. 3-Hydroxyethyl cholic acid ben-
zyl ester 5 was synthesized from cholic acid 1 [20]. After
that, 8a–e were synthesized via acyl chloride from car-
boxylated 5-FU (7a–e) with a spacer chain [20]. Finally,
8a–e were deproteceted through hydrogenolysis to get
the sticky oils 3a–e. The structures of 5-FU-cholic acid
conjugates 3a–e were verified by elemental analyses, IR,
1H-NMR, and ESI-MS spectra. IR spectrum showed an
absorption peak of ester at 1720 cm– 1 (COOR), which sug-
gested the formation of the ester, and another absorption
peak of the carboxyl acid at 1650 cm– 1 (COOH), which
indicated the reservation of carboxyl group in the cholic
Results and discussion
Chemistry
Recent research of 5-FU derivatives show that the weaker
the chemical bonds of 5-FU derivatives, the stronger the
activity and toxicity they expose [16, 17]. We tried to
design a proper chemical bond between 5-FU and cholic
acid, which will be stable during in-vivo transportation,
but should also break and release the original drug effi-
ciently at the target site.
As a classic antitumor drug, 5-FU has been modified to
yield a large number of derivatives. Some relationships
between its chemical structure and its therapeutic activ-
ity have been identified. According to these investiga-
tions, the bond strength between the nitrogen atom in
the uracil ring of 5-FU and the substituent on the nitro-
gen atom, are important factors which could affect the
antitumor activity and the toxicity of the compounds.
1
acid molecule. H-NMR showed the existence of oxy-
methyl, the ester bond with different lengths, cholic
i 2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

Arch. Pharm. Chem. Life Sci. 2009, 342, 513–520
New Liver Targeting 5-Fluorouracil-Cholic Acid Conjugates
515
Reagents and conditions: a) PhCH2Br, DBU, 558C, 24 h; b) i: methylsulfonyl chloride, 108C, 2 h, ii: ethylene glycol, 708C, 2 h; c) i: HCHO, 608C, 50 min, ii: DCC, DMAP,
PhCH2OOC(CH2)nCOOH (n = 2, 3, 4, 6, 8), r.t., 24 h; d) H2, Pd / C, r.t., overnight; e) i: oxalyl chloride, 458, 4 h, ii: triethylamine, r.t., overnight; f) H2, Pd / C, r.t., overnight.
Scheme 1. The synthetic route of 3a–e.
acid, and 5-FU. Elemental analyses and ESI-MS spectra
confirmed the structures of these conjugates.
pH and temperature stability in vitro
5-FU-cholic acid conjugates with hepatoma cells target-
ing property in this study consisted of 5-FU, 5-FU-cholic
acid, or metabolism products of 5-FU in vivo after oral
administration. In order to achieve antitumor effects,
the 5-FU moiety must be released from those forms by
hydrolysis in organs and plasma. Therefore, we tested the
stability of 5-FU-cholic acid conjugate in the solutions
under different pH and temperature in vitro, to establish
the basis for further studies in vivo.
Figure 2. Hydrolysis of 5-FU-cholic acid conjugate 3c at different
pH values and temperatures in vitro.
It was shown in Fig. 2 that conjugate 3c could be com- different time points in liver were significantly higher
pletely hydrolyzed in the solution of pH = 1 at 708C than those of the controls, which orally took an equiva-
within 5 min. The fast and complete hydrolysis of these lent dose of 5-FU. After oral administration of 5-FU-cholic
compounds could be compatible with a fast separation acid conjugate 3c, the concentrations of 5-FU in liver was
and analysis method to shorten the analysis time.
maintained at a high level from 5 min to 6 h, and is still
be detectable 24 h after administration. It also showed
that shortly after the entrance of 5-FU-cholic acid conju-
gate 3c into the body, 5-FU starts to gather in the liver
Distribution and pharmacokinetic analysis in vivo
5-FU levels in the liver
As shown in the Fig. 3, after oral administration of 5-FU- and is maintained there at a relatively high level for a
cholic acid conjugate 3c, the concentrations of 5-FU at long time. However, after oral administration of 5-FU, its
i 2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

516
S. Qian et al.
Arch. Pharm. Chem. Life Sci. 2009, 342, 513–520
Figure 5. 5-FU concentration-time profiles in mice's heart tissue
following oral administration of 5-FU and 5-FU-cholic acid conju-
gate 3c.
Figure 3. 5-FU concentration-time profiles in mice's liver tissue
following oral administration of 5-FU and 5-FU-cholic acid conju-
gate 3c.
Figure 6. 5-FU concentration-time profiles in mice's lung tissue
following oral administration of 5-FU and 5-FU-cholic acid conju-
gate 3c.
Figure 4. 5-FU concentration-time profiles in mice's plasma fol-
lowing oral administration of 5-FU and 5-FU-cholic acid conju-
gate 3c.
these tissues the concentrations of regenerated 5-FU were
relatively low from 5 to 30 min after oral administration.
However, we were still troubled by the question, why
the concentrations of 5-FU in plasma and organs reached
a peak within a few minutes after oral administration of
5-FU-cholic acid conjugate 3c. It may be caused by the
ester bond, which connects 5-FU with cholic acid, and
turns out to be not stable enough. Further research for
more stable connecting bonds is still in progress.
concentration in the liver quickly decreases within
5 min and stayed at a very low level in the 24 h-testing
period.
5-FU levels in plasma, heart, and lung
As shown in Figs. 4–6, the concentration of 5-FU
decreased significantly after 30 min, this shows the weak
point of 5-FU for the clinical application. Shortly after
the oral administration of 5-FU, for instance, the blood
concentration of 5-FU reaches a peak, that results in the
high frequency of administration in 5-FU's clinical appli-
cation. Moreover, the metabolic processing of 5-FU
mainly occurres in the liver, but its metabolites have no
antitumor activity. In this sense, after the absorption in
the intestine and detoxification in the liver, 5-FU is trans-
ferred into activeless forms. Additionally, the individual
differences of the gastrointestinal system and liver
capacity between patients can also cause differences in 5-
FU's absorption and utilization. But in the case of 5-FU-
cholic acid conjugate 3c, Figs. 4, 5, and 6 showed that in
Pharmacokinetic analysis of the bio-distribution of 5-FU-
cholic acid conjugate 3c and 5-FU
Table 1 shows the pharmacokinetic parameters of 5-FU-
cholic acid conjugate 3c and 5-FU in plasma, heart, liver,
and lung tissues. In case of oral administration of 5-FU-
cholic acid conjugate 3c, the area under the regenerated
5-FU's concentration-time curves in the liver from time
zero to 24 h (AUC) was larger than that of orally taken 5-
FU itself. Actually, at all sampling points, we measured
that conjugate 3c gave higher 5-FU concentrations in the
liver than in plasma, which was quite different from
i 2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

Arch. Pharm. Chem. Life Sci. 2009, 342, 513–520
New Liver Targeting 5-Fluorouracil-Cholic Acid Conjugates
517
Table 1. Pharmacokinetic parameters of 5-FU-cholic acid conjugate 3c and 5-FU in plasma, heart, liver and lung tissues.
Parameter
5-Fu-cholic acid conjugate 3c
5-Fu
liver
plasma
heart
liver
lung
plasma
heart
lung
V1/Fa) (L/kg)
24.126
0.241
26.848
0.268
73.13
4.08
0.25
0.17
1.565
0.213
90.051
5.199
0.25
18.023
0.18
58.744
0.587
31.119
0.44
0.083
1.577
70.54
51.963
0.52
25.48
17.333
0.083
0.04
37.095
0.371
50.892
5.061
0.083
0.137
33.9
116.392
1.164
15.896
21.541
0.083
CL/Fb) (L/h/kg)
AUCc) (mg N h/L) 76.55
82.478
14.152
0.083
0.049
29.23
t1/2d) z(h)
6.356
0.083
0.109
44.55
T
max* (h)
Kee) (1/h)
0.133
55.93
0.032
30.95
Cmax+ (mg/L)
25.66
18.5
a)V1/F: apparent volume of distribution.
b)
CL/F: clearance.
c)
AUC: area under (the concentration time) curve.
t1/2: biological half life.
d)
* Tmax: time of maximum concentration.
e)
Ke: rate constant.
+ Cmax: maximum concentration.
orally taken 5-FU. The table also showed that conjugate for Doctoral Programms of Colleges and Universities (Chinese
3c stayed at a higher concentration and had a longer cir- Ministry of Education; No. 20050610085).
culation time in blood, when compared with 5-FU. The
AUC and t1/2 value of conjugate 3c in plasma were The authors have declared no conflict of interest.
76.55 mg.h/L and 6.356 h, respectively, values much
larger than those of 5-FU (31.119 mg.h/L and 0.44 h,
resp.). The longer circulation time of conjugate 3c in the
blood stream may be attributable to its larger molecular
Experimental
size, resulting in a lower glomerular filtration rate in the
kidney. From this point of view, 5-FU-cholic acid conju-
Chemistry
TLC was performed using precoated silica gel GF254 (0.2 mm), and
gate 3c, with a longer retention time and a higher drug
concentration in the liver, was better than 5-FU for liver-
targeting therapy.
column chromatography was performed using silica gel (100–
200 mesh). Melting point was measured on an YRT-3 melting
point apparatus (Shantou Keyi instrument & Equipment Co. Ltd,
Shantou, China). IR spectra were obtained on a Perkin-Elmer 983
(Perkin Elmer, Norwalk, CT, USA). Elemental analyses were per-
formed by Atlantic Microlab, Atlanta, GA, USA. NMR spectra
were taken on a Varian INOVA400 (Varian, Palo Alto, CA, USA)
and on a Bruker AC-E200 (Bruker Bioscience, USA). Chemical
shifts are expressed in d (ppm) with tetramethylsilane (TMS)
functioned as the internal reference and coupling constants (J)
are expressed in Hz. Mass spectra were recorded on an Agilent
1946B ESI-MS instrument (Agilent, Palo Alto, CA, USA). The syn-
thesis route of the 5-FU-cholic acid conjugates 3a–e, starting
from a commercially available 5-FU and cholic acid, is outlined
in Scheme 1. As a representative example, the general synthetic
procedure of 3c is provided below. The synthetic procedure for
compounds 3a, 3b, 3d, and 3e is similar to that of compound 3c.
Conclusion
This research showed that after oral administration of 5-
FU-cholic acid conjugate 3c, concentrations of 5-FU in the
liver at different time points were significantly higher
than concentrations of orally administered 5-FU (giving
the same dose). This verified that the 5-FU-cholic acid con-
jugates are potential liver-targeting drugs. In the conju-
gates, 5-FU is a typical antitumor drug, while the cholic
acid is a good liver-targeting drug carrier. By coupling
these two parts together, we get a series of compounds
with high hepa-selectivity and long halflife in vivo. The
result of this study may present a way to improve the
therapeutic efficiency in cancer chemotherapy by using
cholic acid as a vector of anti-cancer drugs.
Synthesis of benzyl 3a,7a,12a-trihydroxyl-5b-cholestane-
24-carboxylic ester 4
The solution of choric acid 1 (10.0 g, 24.5 mmol), benzyl bro-
mine (4.9 g, 31.4 mmol), and 1,8-diazabicyclo[5.4.0]undec-7-ene
(DBU) (4.7 g, 31.4 mmol) in absolute DMF (30 mL) was stirred at
558C for 24 h. After distillation under reduced pressure to
remove DMF, the residue of the reaction mixture was redis-
solved with 10% hydrochloric acid, and then poured into
100 mL of ice water. The solution was extracted with ethyl ace-
This work was supported by the National Natural Science Foun-
dation of China (No. 30672537) and a Special Research Fund
i 2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

518
S. Qian et al.
Arch. Pharm. Chem. Life Sci. 2009, 342, 513–520
tate (36150 mL). The collected ethyl acetate fraction was
washed with saturated sodium bicarbonate aqua. The ethyl ace-
tate solution was dried with anhydrous Na2SO4 and concen-
trated on a rotary evaporator to afford the viscous oil 4 (8.8 g,
Synthesis of 4-((5-fluoro-2,4-dioxo-3,4-dihydropyrimidin-
1(2H)-yl)methoxy)-4-oxo hexanoic acid 7c
To the solution of benzyl ester 6c (0.6 g, 1.7 mmol) in 15 mL of
methanol was added 10% Pd / C (60 mg). The reaction mixture
was stirred at room temperature overnight to hydrogenolyze
the benzyl ester. The undisolved substances were removed by fil-
tration and the filtrate was concentrated and then purified on a
silica column using 60% acetone in petroleum ether as eluent.
Pure compound 7c (0.7 g, 87%) was obtained as white solid. M.p.:
1
72.4%). H-NMR (CDCl3) d: 0.66 (s, 3H), 0.89 (s, 3H), 0.95 (d, J =
6.0 Hz, 3H), 1.0 (s, 3H), 3.46 (m, 1H), 3.85 (d, 1H), 3.98 (t, J = 2.4 Hz,
1H), 5.10 (s, 2H), 7.36 (m, 5H); IR (KBr): 3421, 2932, 2825, 1732,
1087 cm– 1; MS m/z: 499 [M + H+]. Anal. calcd. for C31H46O5: C,
74.66; H, 9.30; O, 16.04. Found: C, 74.57; H, 9.34; O, 16.12.
124–1268C; IR (KBr): 3400, 1723, 1555, 1382 cm– 1 1H-NMR
;
(CDCl3) d: 1.45–1.67 (m, 4H), 2.25 (t, J = 6.2 Hz, 2H), 2.38 (t, J =
6.1 Hz, 2H), 5.64 (d, J = 6.5 Hz, 2H), 7.62 (d, J = 5.2 Hz, 1H), 8.45 (br,
1H); MS m/z: 289 [M + H+]. Anal. calcd. for C11H13FN2O6: C, 45.84; H,
4.55; F, 6.59; N, 9.72; O, 33.31. Found: C, 45.79; H, 4.53; F, 6.57; N,
9.71; O, 33.38.
Synthesis of benzyl 7a,12a-dihydroxyl-3a-
hydroxylethoxyl-5b-cholestane-24-carboxylic ester 5
Methylsulfonyl chloride (1.7 g, 15.0 mol) was added dropwise
into a solution of benzyl ester 4 (5.0 g, 10.0 mol) in absolute pyr-
idine (40 mL) at 108C. After stirring at room temperature for 2 h,
the reaction mixture was poured into 100 mL of 10% sulphuric
acid solution, followed by extraction with ethyl acetate
(36150 mL). The collected ethyl acetate fractions were dried
with anhydrous Na2SO4 and concentrated on a rotary evaporator
to give methylsulfonyl chloride derivative. The solution of this
newly produced methylsulfonyl chloride derivative was stirred
in ethylene glycol (8 mL) at 708C for 2 h. After cooling to room
temperature, the reaction mixture was poured into 100 mL of
10% sulphuric acid solution, followed by extraction with ethyl
acetate (36150 mL). The collected ethyl acetate fractions were
washed with dilute hydrochloric acid, saturated sodium bicar-
bonate aqua. and water sequentially, and then dried with anhy-
drous Na2SO4 and concentrated on a rotary evaporator to afford
the crude product. Pure 5 (1.9 g, 35%) was obtained as oil after
purifying on a silica chromatography column using acetone /
petroleum ether (1 : 8, v / v) as eluent. 1H-NMR (CDCl3) d: 0.66 (s,
3H), 0.89 (s, 3H), 0.95 (d, J = 6.0 Hz, 3H),1.0 (s, 3H), 3.45(m, 2H),
3.68 (m, 2H), 3.57 (m, 1H), 3.83 (m, 1H), 3.99 (t, J = 2.4 Hz, 1H),
5.12 (s, 2H), 7.36 (m, 5H). IR (KBr): 3423, 2923, 2853, 1754,
1092 cm– 1; MS m/z: 543 [M + H+]. Anal. calcd. for C33H50O6: C,
73.03; H, 9.29; O, 17.69. Found: C, 73.07; H, 9.34; O, 17.74.
Synthesis of 5-FU-benyl cholic ester conjugate 8c
The solution of acid 5 (0.4 g, 1.5 mmol) in oxalyl chloride (10 mL)
was stirred at 458C for 4 h under the protection of nitrogen. The
reaction mixture was evaporated under reduced pressure to
remove excess oxalyl chloride to give an oil. The oily compound
was dissolved in 5 mL of dichloromethane. This solution and
pure triethylamine (0.1 g, 1.5 mmol) was added to 10 mL alcohol
7c (0.5 g, 1.0 mmol) dichloromethane solution under the protec-
tion of nitrogen. After stirring at room temperature overnight,
the reaction mixture was poured into 10 mL water. The organic
layer was separated and washed by brine, and then dried by
anhydrous Na2SO4 and concentrated on a rotary evaporator to
afford the crude product. Pure 8c (0.54 g, 72%) was obtained as
an oil after purifying on a chromatography column using petro-
leum ether / dichloromethane / acetone (30 : 30 : 1) as eluent. IR
(KBr): 3421, 2943, 2814, 1753, 1323, 1054 cm– 1; 1H-NMR (CDCl3)
d: 0.67 (s, 3H), 0.88 (s, 3H), 0.98 (s, 3H), 1.0 (d, J = 6.0 Hz, 3H), 1.5–
2.0 (m, 12H), 3.56 (t, J = 4.8 Hz, 2H), 3.86 (s, 1H), 4.00 (s, 1H), 4.21
(t, J = 4.8 Hz, 2H), 5.11 (s, 2H), 5.63 (s, 2H), 7.33 (s, 5H), 7.62 (s, 1H),
9.50 (s, 1H); MS m/z: 813 [M + H+]. Anal. calcd. for C44H61FN2O11: C,
65.01; H, 7.56; F, 2.34; N, 3.45; O, 21.65. Found: C, 65.12; H, 7.45;
F, 2.25; N, 3.54; O, 21.53.
Synthesis of benzyl (5-fluoro-2,4-dioxo-3,4-
dihydropyrimidin-1(2H)-yl) methyl hexanoate 6c
Synthesis of 5-FU-cholic acid conjugate 3c
The solution of 5-FU 2 (0.5 g, 4.0 mmol) in 37% formaldehyde
and water (5 mL) was stirred at 608C until the solid completely
disappeared and stirring at 608C was continued for another
50 min. A colorless and viscous oil was obtained. To the above
mentioned oily solution in 12 mL of acetonitrile were added 6-
(benzyloxy)-6-oxobutanoic acid (1.3 g, 5.6 mmol), DCC (1.2 g,
5.6 mmol), and DMAP (30 mg, 0.3 mmol). The reaction mixture
was stirred at 08C for 1 h, and then stirred at room temperature
for 24 h. The undisolved substances in the reaction mixture
were removed by filtration, and the filtrate was evaporated
under reduced pressure to remove the solvent. The residue was
dissolved in 50 mL acetic ether and washed with dilute hydro-
chloric acid, saturated sodium bicarbonate aqua., and water,
sequentially. Then, the acetic ether solution was concentrated
and purified on a silica column using 25% acetone in petroleum
ether as eluent. Pure compound 6c (1.1 g, 72%) was obtained as
To the solution of benzyl ester 8c (0.3 g, 0.4 mmol) in 20 mL of
methanol was added 10% Pd / C (30 mg). The reaction mixture
was stirred at room temperature overnight to hydrogenolyze
the benzyl ester. The undisolved substances were removed by fil-
tration and the filtrate was concentrated and purified on a silica
column using dichloromethane / petroleum ether / acetone
(15 : 15 : 1) as eluent. Pure compound 3c (0.2 g, 87%) was
obtained as an oil. IR (KBr): 3438, 2935, 2866, 1720 1650, 1387,
1094 cm– 1; 1H-NMR (CDCl3) d: 0.68 (s, 3H), 0.88 (s, 3H), 0.98 (s, 3H),
1.0 (d, J = 6.0 Hz, 3H), 1.5–2.0 (m, 8H), 1.5–2.0 (m, 8H), 3.56 (t, J =
4.8 Hz, 2H), 3.86 (s, 1H), 4.00 (s, 1H), 4.21 (t, J = 4.8 Hz, 2H), 5.13 (s,
2H), 7.62 (s, 1H), 9.51 (s, 1H); MS m/z: 723 [M + H+]. Anal. calcd. for
C37H55FN2O11: C, 61.48; H, 7.67; F, 2.63; N, 3.88; O, 24.35. Found:
C, 61.55; H, 7.56; F, 2.46; N, 3.75; O, 24.35.
white solid. M.p.: 118–1208C; IR (KBr): 1724, 1554, 1363 cm– 1
;
5-FU-cholic acid conjugate 3a
1H-NMR (CDCl3) d: 1.66 (s, 4H), 2.38 (d, J = 6.0 Hz, 4H), 5.10 (s, 2H),
5.63 (d, J = 6.5 Hz, 2H), 7.34 (s, 5H), 7.60 (d, J = 5.2 Hz, 1H), 8.42 (br,
1H); MS m/z: 379 [M + H+]. Anal. calcd. for C18H19FN2O6: C, 57.14; H,
5.06; F, 5.02; N, 7.40; O, 25.37. Found: C, 57.23; H, 5.12; F, 5.11; N,
7.42; O, 25.42.
IR (KBr): 3442, 2928, 2869, 1723, 1652, 1376, 1096 cm– 1; 1H-NMR
(CDCl3) d: 0.67 (s, 3H), 0.88 (s, 3H), 0.98 (s, 3H), 1.0 (d, J = 6.0 Hz,
3H), 1.5–2.0 (m, 4H), 3.56 (t, J = 4.8 Hz, 2H), 3.86 (s, 1H), 4.00 (s,
1H), 4.21 (t, J = 4.8 Hz, 2H), 5.13 (s, 2H), 7.62 (s, 1H), 9.51 (s, 1H);
MS m/z: 695 [M + H+]. Anal. calcd. for C35H51FN2O11: C, 60.50; H,
i 2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

Arch. Pharm. Chem. Life Sci. 2009, 342, 513–520
New Liver Targeting 5-Fluorouracil-Cholic Acid Conjugates
519
7.40; F, 2.73; N, 4.03; O, 25.33. Found: C, 60.52; H, 7.43; F, 2.70; N,
4.09; O, 25.29.
ples. Then the animals were dissected and each tested organ was
removed, including heart, lung, and liver. Organs were rinsed
with cold normal saline, blotted dry with a paper towel,
extracted with methanol, diluted, centrifuged, and dispensed in
plastic sample vials. As a polar compound, the retention time of
5-FU was relatively short on HPLC column. Though some endoge-
nous substances in mice's organ tissue may interfere with the
determination of 5-FU concentration, under the experimental
condition, which we adopted, with 5-BU (5-bromouracil) as inter-
nal standard [22–24], 5-FU showed a significant peak on the
spectrum with a smooth baseline. Endogenous substances from
mouse tissue did not interfere with the separation of 5-FU. This
method is simple, sensitive, with high recovery ratio and good
accuracy, which is feasible for the determination of 5-FU in this
kind of biological samples.
5-FU-cholic acid conjugate 3b
IR (KBr): 3435, 2937, 2863, 1723, 1652, 1384, 1093 cm– 1; 1H-NMR
(CDCl3) d: 0.67 (s, 3H), 0.88 (s, 3H), 0.98 (s, 3H),1.0 (d, J = 6 Hz, 3H),
1.5–2.0 (m, 8H), 1.5–2.0 (m, 6H), 3.56 (t, J = 4.8 Hz, 2H), 3.86 (s,
1H), 4.00 (s, 1H), 4.21 (t, J = 4.8 Hz, 2H), 5.13 (s, 2H), 7.62 (s, 1H),
9.51 (s, 1H); MS m/z: 709 [M + H+]. Anal. calcd. for C36H53FN2O11: C,
61.00; H, 7.54; F, 2.68; N, 3.95; O, 24.83. Found: C, 61.02; H, 7.53;
F, 2.65; N, 3.95; O, 24.81.
5-FU-cholic acid conjugate 3d
IR (KBr): 3436, 2934, 2869, 1724, 1653, 1384, 1093 cm– 1; 1H-NMR
(CDCl3) d: 0.68 (s, 3H), 0.88 (s, 3H), 0.98 (s, 3H),1.0 (d, J = 6.0 Hz,
3H), 1.5–2.0 (m, 8H), 1.5–2.0 (m, 12H), 3.56 (t, J = 4.8 Hz, 2H),
3.86 (s, 1H), 4.00 (s, 1H), 4.21 (t, J = 4.8 Hz, 2H), 5.13 (s, 2H), 7.62 (s,
1H), 9.51 (s, 1H); MS m/z: 751 [M + H+]. Anal. calcd. for
C39H59FN2O11: C, 62.38; H, 7.92; F, 2.53; N, 3.73; O, 23.44. Found:
C, 62.36; H, 7.94; F, 2.51; N, 3.71; O, 23.46.
HPLC analysis
The HPLC system consisted of an SPD-10A variable UV-VIS detec-
tor and a set of Model LC-10AT liquid chromatograph including
a manometric module as well as a dynamic mixer from Shi-
madzu (Tokyo, Japan). The mobile phase consists of pure water,
which was filtered through a 0.45 mm membrane filter before
use. A phenomenex ODS column (4.66200 mm, 5 lm; Phenom-
enex, Torrance, CA, USA) was eluted with the mobile phase at a
flow rate of 1.0 mL/min. The eluate was monitored by measuring
the absorption at 256 nm with a sensitivity of AUFS 0.01 at 258C.
The average retention time of 5-FU was approximately 7 min.
5-FU-cholic acid conjugate 3e
IR (KBr): 3436, 2934, 2868, 1724, 1653, 1389, 1090 cm– 1. 1H-NMR
(CDCl3) d: 0.67 (s, 3H), 0.88 (s, 3H), 0.98 (s, 3H),1.0 (d, J = 6.0 Hz,
3H), 1.5–2.0 (m, 8H), 1.5–2.0 (m, 16H), 3.56 (t, J = 4.8 Hz, 2H),
3.86 (s, 1H), 4.00 (s, 1H), 4.21 (t, J = 4.8 Hz, 2H), 5.13 (s, 2H), 7.62 (s,
1H), 9.51 (s, 1H); MS m/z: 779 [M + H+]. Anal. calcd. for
C41H63FN2O11: C, 63.22; H, 8.15; F, 2.44; N, 3.60; O, 22.59. Found:
C, 63.21; H, 8.13; F, 2.46; N, 3.60; O, 22.61.
References
Pharmacological evaluation
[1] M. A. Theresa, Nature 2002, 2750–2761.
5-FU was supplied by Nantong Pharmaceutical Co. Ltd. (Nantong,
P. R.China); 5-Bu by Fluke Ltd. (Jiangsu, P. R. China); all other
chemicals and reagents were of analytical grade, and were
obtained commercially. All liquid reagents were distilled before
use. Female Kunming mice weighing 18–22 g were received
from Sichuan Industrial Institute of Antibiotics (P. R. China) and
were fasted for 24 h. Since the experiment could be finished
within 48 h, there was no significant change in mice's body
weights during the experiment.
[2] A. P. Davis, R. P. Bonar–Law, J. K. M. Sanders, Eds. Elsevier
Oxford, 1996, 4, 257–286.
[3] G. M. H. William, L. H. Victor, F. S. Dale, Hepatology 1991,
13, 68–72.
[4] A. F. Hofmann, Raven Press, New York, 3rd Ed, 1994, pp.
677–718.
[5] W. Kramer, G. Wess, G. Schuber, M. Bickel, F. Girbig, J. Biol.
Chem. 1992, 267, 185–198.
[6] S. Cao, Y. M. Rustum, Cancer Res. 2000, 60, 3717–3721.
5-FU-cholic acid pH and temperature stability in vitro
A solution of 5-FU-cholic acid (50 lg/mL) was incubated in buffer
with varying pH values (1, 3, 5, 8) and different temperatures
(25, 40, 55, 708C) under the protection of nitrogen. At a predeter-
mined time interval, a 20 lL portion of the solution was
removed, and the concentration of 5-FU was measured by HPLC
as described below.
[7] R. Yoshikawa, M. Kusunoki, H. Yanagi, M. Noda, et al., Can-
cer Res. 2001, 61, 1029–1037.
[8] C. Heidelberger, N. K. Chaudhuri, P. Danneberg, Nature
1957, 179, 663–666.
[9] E. O. Aboagye, A. Saleem, V. J. Cunningham, S. Osman, P.
M. Price, Cancer Res. 2001, 61, 4937–4941.
[10] C. Kuropkat, K. Griem, J. Clark, E. R. Rodriguez, et al., J.
Clin. Oncol. 1999, 22, 466–470.
5-FU-cholic acid distribution in mice tissue in vivo
According to the requirements of the National Act on the usage
of experimental animals (P. R. China), the Sichuan University
Animal Ethical Experimentation Committee, approved all pro-
cedures of our in-vivo studies. 5-FU and 5-FU-cholic acid conju-
gate 3c were intragastrically administrated into the mice. At
each preparation and sampling time point, five mice were
treated with a single dose of prodrug equivalent to 20 mg/kg of
5-FU. At the indicated time, the animals were sacrificed and
blood samples were collected from the ocular artery directly
after removing eyeball, and were treated to obtain plasma sam-
[11] D. V. Sapti, C. S. McHenry, R. T. Raines, Biochemistry 1987,
26, 8606–8613.
[12] W. Parulekar, R. W. de Marsh, R. Wong, Int. J. Radiat. Oncol.
Biol. Phys. 2004, 58, 1487–1495.
[13] R. B. Diasio, Drugs 1999, 58, 119–126.
[14] T. Spector, D. J. Porter, S. T. Nel, Drugs Future 1994, 19,
566–571.
[15] F. Fata, I. G. Ron, N. Kemeny, E. O'Reilly, et al., Cancer 1999,
86, 1129–1134.
i 2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

520
S. Qian et al.
Arch. Pharm. Chem. Life Sci. 2009, 342, 513–520
[16] T. B. Cai, X. Tang, J. Nagorski, Bioorg. Med. Chem. 2003, 11,
[20] G. Wess, W. Kramer, W. Bartmann, Tetrahedron Lett. 1992,
33, 195–198.
4971–4975.
[21] Q. L. Jiang, L. Y. Wu, Y. Wu, West Chin. J. Pharm. Sci. 2005, 1,
[17] C. Allegra, Lippincott Health Promot. Lett. 1997, 432.
12–14 (in Chinese).
[18] O. Shochiro, N. Toshio, T. Hiromi, M. Haruki, et al., Chem.
Pharm. Bull. 1987, 35, 3894–3897; S. Ozaki, Y. Watanabe, T.
Hoshiko, H. Mizuno, et al., Chem. Pharm. Bull. 1984, 32,
733–738; A. Buur, L. Trier, C. Magnusson, P. Artursson,
Int. J. Pharm. 1985, 24, 43.
[22] F. Chu, J. K. Gu, W. H. Liu, J. Chromatogr. B Biomed. Sci. Appl.
2003, 795, 377–382.
[23] W. Wattanatorn, H. L. Mcleod, J. Cassidy, J. Chromatogr. B
Biomed. Sci. Appl. 1997, 702, 233–237.
[19] Z. F. Stephan, E. C. Yurrachek, Biochem. Pharmacol. 1992, 43,
[24] L. Zufa, A. Aldaz, J. Gimdez, J. Chromatogr. B Analyt. Technol.
Biomed. Life Sci. 2004, 809, 51–58.
1969–1974.
i 2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com