Design, synthesis and antifungal activity of a novel water soluble prodrug of antifungal triazole.

Article abstract of DOI:10.1016/S0960-894X(02)00892-2

A highly potent water soluble triazole antifungal prodrug, RO0098557 (1), has been identified from its parent, the novel antifungal agent RO0094815 (2). The prodrug includes a triazolium salt linked to an aminocarboxyl moiety, which undergoes enzymatic activation followed by spontaneous chemical degradation to release 2. Prodrug 1 showed high chemical stability and water solubility and exhibited strong antifungal activity against systemic candidiasis and aspergillosis as well as pulmonary aspergillosis in rats.

Full text of DOI:10.1016/S0960-894X(02)00892-2

Bioorganic & Medicinal Chemistry Letters 13 (2003) 191–196
Design, Synthesis and Antifungal Activity of a Novel Water
Soluble Prodrug of Antifungal Triazole
Jun Ohwada,a Masao Tsukazaki,a Tadakatsu Hayase,a Nobuhiro Oikawa,a
Yoshiaki Isshiki,a Hiroshi Fukuda,a Eisaku Mizuguchi,a Masahiro Sakaitani,a
Yasuhiko Shiratori,a Toshikazu Yamazaki,b Shigeyasu Ichihara,c
Isao Umedaa,* and Nobuo Shimmaa
aDepartment of Chemistry, Nippon Roche Research Center, 200 Kajiwara, Kamakura, Kanagawa 247-8530, Japan
bDepartment of Mycology, Nippon Roche Research Center, 200 Kajiwara, Kamakura, Kanagawa 247-8530, Japan
cDepartment of Preclinical Science, Nippon Roche Research Center, 200 Kajiwara, Kamakura, Kanagawa 247-8530, Japan
Received 21 August 2002; accepted 17 October 2002
Abstract—A highly potent water soluble triazole antifungal prodrug, RO0098557 (1), has been identified from its parent, the novel
antifungal agent RO0094815 (2). The prodrug includes a triazolium salt linked to an aminocarboxyl moiety, which undergoes
enzymatic activation followed by spontaneous chemical degradation to release 2. Prodrug 1 showed high chemical stability and
water solubility and exhibited strong antifungal activity against systemic candidiasis and aspergillosis as well as pulmonary asper-
gillosis in rats.
# 2002 Elsevier Science Ltd. All rights reserved.
Introduction
formulation and the existing prodrug approaches were
not satisfactory in stability and/or water solubility, we
conducted a study on a new prodrug of 2, which should
have sufficient water solubility, stability in aqueous
solution and quantitative bioconversion.
There is a high medical need for an injectable agent with
a broad spectrum for the treatment of severe deep
mycoses of hospitalized patients. Only fluconazole
(FCZ) and amphotericin B are currently available for
parenteral use, but they have limitations in antifungal
spectra and safety, respectively.1 Most of the azoles
under development have broader spectrum but cannot
be administered parenterally due to high lipophilicity.2
There have been some efforts to overcome this problem
by using a prodrug approach.3 For example, we recently
reported a novel prodrug of azole antifungals and
Takeda Chemical Industries, Ltd. has identified a water
soluble quaternary triazolium prodrug, TAK-457.
Compound 2 obviously has two functional groups,
namely a tertiary alcohol and a triazole group, with
potential to be linked to a pro-moiety. However, the
high steric hindrance around the hydroxyl group may
limit the modification. This prompted us to design a
triazolium salt type prodrug.5 The general concept of
prodrugs is depicted in Scheme 1. The prodrug 3 con-
tains a [N-(3-acetoxypropyl)-N-methylamino]carboxy-
methyl group, and an ester group of which can be
rapidly hydrolyzed by a nonspecific enzyme, serum
esterase, to generate an alcohol intermediate, and then
undergoes rapid and spontaneous intra-molecular cycli-
zation to release 2, a cyclic carbamate and an aldehyde.
Further optimization of the pro-moiety of 3 by its con-
formational restriction to adjust the conversion rate to 2
led to the identification of 1, a new injectable prodrug of
highly potent antifungal azole 2. In this paper, we
describe the design, synthesis and biological profile of
RO0098557 (1) (Fig. 1).
Previously, we identified RO0094815 (2) which has a
broad antifungal spectrum covering Aspergillus spp.,
FCZ-resistant Candida spp. and has a good safety pro-
file including low drug–drug interaction.4 Since the
water solubility of 2 was, however, too low for parenteral
*Corresponding author. Tel.: +81-467-47-2244; fax: +81-467-45-
6824; e-mail: umedaiso@chugai-pharm.co.jp
0960-894X/03/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(02)00892-2

192
J. Ohwada et al. / Bioorg. Med. Chem. Lett. 13 (2003) 191–196
Figure 1. Structure of RO0098557 and RO0094815.
Scheme 1.
Scheme 2. (a) ClCOOCH2Cl, diisopropylethylamine, CH2Cl2, rt (quant); (b) 2, CH3CN, 80 ꢀC (60%); (c) (1) ClCOOCH2Cl, Et3N, CH2Cl2, rt;
(2) Ac2O, pyridine, rt (30%, two steps); (d) (1) NaI, CH3CN, 50 ꢀC ; (2) 2, CH3CN, 50 ꢀC (88%, two steps); Synthesis of 3: (1) N-3-hydroxypropyl-
N-methylamine, ClCOOCH2Cl, Et3N, CH2Cl2, rt; (2) AcCl, Et3N, CH2Cl2, rt (20%, two steps); (3) 2, NaI, CH3CN, 50 ꢀC (82%); Synthesis of 10:
(1) l-prolinol, ClCOOCH2Cl, Et3N, CH2Cl2, rt; (2) Ac2O, pyridine, rt (<10%, 2 steps); (3) 2, NaI, CH3CN, 50 ꢀC (92%); Synthesis of 11:
(1) 2-hydroxymethyl-N-methylaniline, ClCOOCH2Cl, diisopropylethylamine, CH2Cl2, rt; (2) Ac2O, diisopropylethylamine, rt (20%, two steps); (3) 2,
cat. NaI, CH3CN, reflux (63%).

J. Ohwada et al. / Bioorg. Med. Chem. Lett. 13 (2003) 191–196
193
Chemistry
four steps: (i) N-acylation of 2-methylaminoethanol,
(ii) O-acetylation, (iii) iodination of 8 with sodium
iodide in acetonitrile, and (iv) coupling with 2 in aceto-
nitrile at 50 ꢀC. The derivatives 3, 10 and 11 were simi-
larly obtained from corresponding aminoalcohols as
described in Scheme 2.
We synthesized a series of new triazolium derivatives of
2. Compounds 3, 6, 9, 10 and 11 were first prepared as
outlined in Scheme 2 in order to analyze their stability
and ability to release 2. Next, aromatic analogues 18,
19, 20, 21 and 1 were synthesized for optimization of 11
to increase its water solubility and conversion rate.
Compounds in the second series had sarcosine esters6
to make them water soluble, and they were also
designed to generate acetaldehyde7 instead of formal-
dehyde for a better safety profile. The synthetic pro-
cedures for the second series of the derivatives are
outlined in Scheme 3.
Compound 1 was prepared from 2-chloronicotinic acid
in seven steps. 12 was treated with oxalyl chloride fol-
lowed by potassium tert-butoxide in THF to give tert-
butyl ester 13. Substitution of the chlorine atom in 13
with methylamine, followed by reduction of the ester
group of 14 by lithium aluminium hydride, gave
aminoalcohol 15 as a colourless crystal. N-Acylation of
15 with 1-chloroethyl chloroformate followed by
N-Boc-sarcosine gave sarcosine ester 16. It was then
coupled with 2 in the presence of NaI in acetonitrile
followed by anion exchange chromatography with
DOWEX-1 to give 17 as an amorphous powder.
Compound 5 was prepared from N-methylaniline by
N-acylation with chloromethyl chloroformate. The
coupling of 2 with 5 was carried out in acetonitrile to
give a triazolium salt 6. Compound 9 was synthesized in
Scheme 3. (a) (1) oxalyl chloride, DMF, 0 ꢀC; (2) KOtBu, THF, À5 ꢀC (97%, two steps); (b) CH3NH2, MeOH, rt (90%); (c) LiAlH4, THF, 0 ꢀC
(80%); (d) (1) ClCOOCH(CH3)Cl, diisopropylethylamine, CH2Cl2, 0 ꢀC; (2) Boc-Sarcosine, WSCI, DMAP, CH2Cl2, 0 ꢀC (84%, two steps); (e) (1) 2,
NaI, CH3CN, 50 ꢀC; (2) DOWEX-1 ClÀ form, aqueous MeOH, rt (65%, two steps); (f) (1) HCl, EtOAc, rt; (2) lyophilization (69%, two steps);
Synthesis of 18: (1) (i) (4,5-difluoro-2-methylaminophenyl)methanol, ClCOOCH(CH3)Cl, diisopropylethylamine, CH2Cl2, 0 ꢀC; (ii) Boc-Sarcosine,
WSCI, DMAP, CH2Cl2, 0 ꢀC (quant, two steps); (2) 2, cat. NaI, CH3CN, 80 ꢀC; (50%,); (3) HCl, EtOAc, rt (90%); Synthesis of 19: (1) (i) 2-fluoro-6-
methylaminophenyl)methanol, ClCOOCH(CH3)Cl, diisopropylethylamine, CH2Cl2, 0 ꢀC; (ii) Boc-Sarcosine, WSCI, DMAP, CH2Cl2, 0 ꢀC (74%,
two steps); (2) 2, cat. NaI, CH3CN, reflux; (3) HCl, EtOAc, rt (29%, two steps); Synthesis of 20: (1) (i) (5-fluoro-2-methylaminophenyl)methanol,
ClCOOCH(CH3)Cl, diisopropylethylamine, CH2Cl2, 0 ꢀC; (ii) Boc-Sarcosine, WSCI, DMAP, CH2Cl2, 0 ꢀC (91%, two steps); (2) 2, cat. NaI,
CH3CN, 70 ꢀC (72%); (3) HCl, EtOAc, rt (88%); Synthesis of 21: (1) (i) (4-chloro-2-methylaminophenyl)methanol, ClCOOCH(CH3)Cl, diisopro-
pylethylamine, CH2Cl2, 0 ꢀC; (ii) Boc-Sarcosine, WSCI, DMAP, CH2Cl2, 0 ꢀC (71%, two steps); (2) 2, CH3CN, 65 ꢀC; (3) HCl, EtOAc, rt (65%, two
steps).

194
J. Ohwada et al. / Bioorg. Med. Chem. Lett. 13 (2003) 191–196
corresponding primary alcohol intermediates. However,
there was a considerable difference in the cyclization rate
of releasing the active substance 2. The intermediates
generated from 3, 9 and 10 were slowly converted to 2 (T1/
2: >10h, >10h and >3.8 h, respectively). In contrast,
the intermediate generated from 11 rapidly converted to
2 with a half life of 13 min. Compound 6 was resistant
to direct hydrolysis by esterase in plasma, indicating the
carbamate linkage to be stable in plasma. Thus, the high
conversion rate of 11 into 2 can be explained by facile
intra-molecular cyclization of the intermediate alcohol
as a result of conformational restriction of the pro-
moiety.
Table 1. Solubility in water and conversion rate of the prodrugs in
rat plasma
Compd Solubility in water
(mg/mL)
Conversion rate in rat plasma (T1/2)
Prodrug
Intermediate
6
9
3
10
11
<1
<1
<1
<1
1
29 h
<2 mina
<2 mina
<2 mina
<2 mina
–
>10 h
>10 h
3.8 h
13 min
aHydrolysis of the acetate group.
Finally, removal of Boc group of 17 by HCl in ethyl
acetate and lyophilization gave 1 as a white powdery
solid.8 The derivatives 18, 19, 20 and 21 were obtained
from corresponding aminoalcohols in a similar manner
as described above (Scheme 3).
Calculated preferred conformations of the intermediates
for 10 and 11 are shown in Figure 2. The faster conver-
sion rate of 11 as compared with that of 10 is explained
by the fact that the carbamate group of the intermediate
from 11 locates perpendicular to the phenyl ring due to
steric repulsion between the N-methyl group and the
ortho substituent, the hydroxymethyl group in the
intermediate, which forces the nucleophile, OH, close to
the carbonyl carbon. On the other hand, the carbamate
group in 10 is almost coplanar with the proline ring,
which is not favorable for intramolecular cyclization
between the carbonyl group and the primary hydroxy
group. We surmised that prodrugs should be even more
rapidly cleaved in human plasma in order to not affect
the PK profile of 2 and have higher water solubility for
parenteral formulation. Thus, we synthesized additional
analogues of 11 having both electron withdrawing
groups and a solubilizing moiety, sarcosine ester, on the
All of the prodrugs prepared were sufficiently stable for
biological evaluation.
Results and Discussion
We initially compared the conversion rate of the deriva-
tives 3, 6, 9, 10 and 11 in rat plasma and their solubility in
water. As shown in Table 1, conformational restriction
had a definitive effect on the conversion rate of the
intermediates in rat plasma. Compounds 3, 9, 10 and 11
were rapidly hydrolyzed by esterase in rat plasma to give
Figure 2. Conformational analysis of two intermediates by MOPAC (RO0094815 was replaced by 1-methyl-1,2,4-triazole in this calculation).

J. Ohwada et al. / Bioorg. Med. Chem. Lett. 13 (2003) 191–196
195
Table 2. Solubility in water and conversion rate of the prodrugs in
plasma
phenylring of the pro-moiety. Conversion rates and
water solubility of the new derivatives, 18, 19, 20, 21
and 1 are summarized in Table 2. At this stage, we
evaluated the conversion rate in rat, monkey and
human plasma to determine whether or not species dif-
ference exists. All compounds in Table 2 showed a more
rapid conversion rate and higher water solubility com-
pared to 11. Especially compound 1, having a pyridine
ring, quickly converted into 2 in the plasma of the three
species (T1/2: <1 min), and exhibited extremely high
solubility in water (>100 mg/mL). Furthermore, 1 was
found to stoichiometrically release 2 and the cyclized
compound 229 in human plasma as shown in Figure 3,
suggesting that 1 underwent a rapid and quantitative
intramolecular cyclization.
Compd
Solubility in water
(mg/mL)
Conversion rate in plasma
(T1/2) (min)
Prodrug
Intermediate
Rat <2
Monkey 6
Human 6
5
6
8
18
19
20
21
1
>10
>10
Rat <2
Monkey 2
Human 2
3
3
2
Rat <2
Monkey 2
Human 2
9
8
8
>10
Rat <2
Monkey 2
Human 3
4
5
4
>10
Prodrug 1 was quite stable in an aqueous buffer solution
at pH 1 to 4 (>99% of 1 remained after 7 h), suggesting
sufficient stability for parenteral use. The PK profile after
intravenous (iv) bolus and oral (po) administration of 1
to monkey at a dose of 3 mg/kg is shown in Figure 4. In
Rat <1
Monkey <1
Human <1
Not detected
Not detected
Not detected
>100
Figure 3. Conversion of 1 in human plasma.
Figure 4. Plasma level of 2 after administration of 1 to monkey at a dose of 3 mg/kg.

196
J. Ohwada et al. / Bioorg. Med. Chem. Lett. 13 (2003) 191–196
Table 3. In vivo efficacy of 1 against systemic candidiasis in rat
Rat systemic candidiasis; ED50 (mmol/kg) on day 14
nitrogen containing hetero-aromatic ring. We identified
the triazolium salt prodrug RO0098557 (1) that showed
potent antifungal activity against both systemic candi-
diasis and aspergillosis as well as pulmonary aspergil-
losis in rat in both iv and po administration. Because
there is few injectable antifungal azole agent with a
broad spectrum, 1 could be a promising drug for treat-
ment of systemic fungal infections both parenterally and
orally.
C. albicansa
C. tropicalisa
CY1002
CY3003
CY5042
RO0098557
ITCZ
iv
po
4.0
4.0
1.5
2.6
0.9
0.8
po
4.7
1.7
1.7
aTreatments: 0, 4, 24, and 48 h after infection (b.i.d.Â1+q.d.Â2).
References and Notes
1. (a) Maesaki, S.; Kohno, S. Saishin Igaku 1992, 47, 277. (b)
Ohno, R.; Mizoguchi, H.; Kume, H. Nikkei Medical Aug. 10,
1993, 40. (c) Saxena, S.; Khan, J. A.; Ghosh, P. C. J. Anti-
microb. Chemother. 1998, 42, 635. (d) Petit, C.; Cheron, M.;
Joly, V.; Rodrigues, J. M.; Bolard, J.; Gaboriau, F. J. Anti-
microb. Chemother. 1998, 42, 779.
Table 4. In vivo efficacy of 1 against aspergillosis in rat
Rat aspergillosis; ED50 (mmol/kg) on day 14
Systemica
Systemicb
Pulmonaryc
A. fumigatus
CF924390e
A. fumigatus
2. (a) Fung-Tomc, J. C.; White, T. C.; Minassian, B.; Huczko,
E.; Bonner, D. P. Diagn. Microbiol. Infect. Dis. 1999, 35, 163.
(b) Kitazaki, T.; Ichikawa, T.; Tasaka, A.; Hosono, H.; Mat-
sushita, Y.; Hayashi, R.; Okonogi, K.; Itoh, K. Chem. Pharm.
Bull. 2000, 48, 1935. (c) Van Cutsem, J. Mycoses 1989, 32, 14.
(d) Nakamura, H.; Yamada, A.; Ohki, H.; Maki, K.; Kitano,
Y.; Takasugi, H.; Tanaka, H. Abstracts, The 17th Symposium
on Medicinal Chemistry, 6th Annual Meeting of Division of
Medicinal Chemistry, Abstract No. 1P-09, Nov. 19–21, 1997;
The Pharmaceutical Society of Japan: Tsukuba, Japan.
3. (a) Ohwada, J.; Murasaki, C.; Yamazaki, T.; Ichihara, S.;
Umeda, I.; Shimma, N. Bioorg. Med. Chem. Lett. 12, 2775. (b)
Ichikawa, T.; Kitazaki, T.; Matsushita, Y.; Yamada, M.;
Hayashi, R.; Yamaguchi, M.; Kiyota, Y.; Okonogi, K.; Itoh,
K. Chem. Pharm. Bull. 2001, 49, 1102. (c) Ueda, Y; Matis-
kella, J. D.; Golik, J; Hudyma, T. W.; Chen, C. WO 01/52852,
2001.
4. Ohwada, J.; Tsukazaki, M.; Hayase, T.; Oikawa, N.;
Isshiki, Y.; Umeda, I.; Yamazaki, T.; Ichihara, S.; Shimma, N.
Abstracts, The 21st Symposium on Medicinal Chemistry, 10th
Annual Meeting of Division of Medicinal Chemistry, Abstract
No. 1P-06, Nov. 28–30, 2001; The Pharmaceutical Society of
Japan: Kyoto, Japan.
5. Bogardus, J. B.; Higuchi, T. J. Pharm. Sci. 1982, 71, 729.
6. Among several amino acid esters, sarcosine ester was cho-
sen from its physicochemical and safety property.
CF1003d
CF924390e
RO0098557
iv
po
12
14
6.0
9.9
6.8
8.9
ITCZ
po
8.9
7.2
2.5
aNormal host.
bImmunosuppression: cyclophophamide 100 mg/kg, ip on day À4 and
cortisone acetate 125 mg/kg, sc on day À1.
cImmunosuppression: cortisone acetate 125 mg/kg, sc on days À6, À4,
and À1 and low protein diet.
dTreatments: 0, 4, 24, 28, 48, 52, 72, 76 h after infection (b.i.d. Â4).
eTreatments: 0, 4, 24, 28, 48 and 52 h after infection (b.i.d. Â3).
both administrations, the active drug 2 was quickly
formed and slowly eliminated with T1/2 of 9.8 h (iv) and
12.8 h (po). Oral bioavailability of 2 was fairly good in
monkey (87%) after oral administration of 1. The in
vivo efficacy of 1 in various infection models in rats is
summarized in Tables 3 and 4. All the ED50 values were
measured on Day 14. In the systemic candidiasis model
in rats, 1 exhibited almost equal to or better activity
than ITCZ against Candida albicans CY1002, CY3003,
and a fluconazole resistant strain, C. tropicalis, in both
iv and oral administration. 1 was also strongly active in
the systemic and pulmonary aspergillosis model in rats
in both administrations (ED50: 6.0–14 mmol/kg).
7. Acetaldehyde is most likely safe at the clinical dose of the
prodrug.
8. Compound 1 was determined to be mono HCl salt. Anal.
.
calcd for C35H35F2N8O5S Cl.HCl: Cl, 8.98. Found: Cl, 9.27.
9. Compound 22 as well as 1 showed no toxicity at the clinical
dose in rat and monkey, respectively.
In summary, we developed a widely applicable prodrug
technique for solubilization of compounds having the