Synthesis and structure-activity relationships of novel fungal chitin synthase inhibitors

Article abstract of DOI:10.1016/S0960-894X(00)00257-2

A novel Candida albicans chitin synthase 1 (CaChs1) inhibitor, RO-41-0986 (1) was discovered by random screening. Systematic modification led to the identification of a highly potent CaChs1 inhibitor, RO-09-3024 (2), having strong antifungal activity against Candida spp. in vitro. (C) 2000 Elsevier Science Ltd. All rights reserved.

Full text of DOI:10.1016/S0960-894X(00)00257-2

Bioorganic & Medicinal Chemistry Letters 10 (2000) 1459±1462
Synthesis and Structure±Activity Relationships of Novel Fungal
Chitin Synthase Inhibitors
Kazunao Masubuchi,a Mikio Taniguchi,a Isao Umeda,a Kazuo Hattori,a
Hitomi Suda,a Yasunori Kohchi,a Yoshiaki Isshiki,a Toshiya Sakai,a
Masami Kohchi,a Michio Shirai,a Hisafumi Okabe,b Masayuki Sudoh,b
Toshikazu Yamazakib 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
Received 28 March 2000; accepted 26 April 2000
AbstractÐA novel Candida albicans chitin synthase 1 (CaChs1) inhibitor, RO-41-0986 (1) was discovered by random screening.
Systematic modi®cation led to the identi®cation of a highly potent CaChs1 inhibitor, RO-09-3024 (2), having strong antifungal
activity against Candida spp. in vitro. # 2000 Elsevier Science Ltd. All rights reserved.
Introduction
exhibit in vivo ecacy in a murine systemic candidiasis
model. Therefore, we initiated a chemical modi®cation
study of 1 with the aim of identifying a novel chitin syn-
thase inhibitor that has potent antifungal activity against
Candida albicans, Cryprococcus neoformans, and Asper-
gillus fumigatus in vivo. In this paper, we describe the
synthesis and structure±activity relationships (SAR) of a
novel series of chitin synthase inhibitors.
The medical need for safe and e€ective systemic anti-
fungal agents with novel modes of action has been
intensi®ed due to an increasing number of deep-seated
fungal infections and resistance development.1 Chitin is
a major and essential component of the fungal cell wall,
which is widely distributed among yeast and mycelial
fungi but absent in mammalian cells.2 Thus, it is a very
attractive target for designing antifungal agents. Poly-
oxins and nikkomicins3 are known competitive inhibitors
of chitin synthase, but they have shown only limited
antifungal activity against Coccidiodes and Blastomyces
spp.
Chemistry
Since we found that 3,4-dihydro derivatives of 1
retained both the chitin synthase inhibitory activity and
in vitro antifungal activity against Candida albicans, we
selected 5-amino-3,4-dihydro-1H-quinolin-2-one 5 as
the representative core structure. The general synthetic
procedure for the synthesis of 3,4-dihydro derivatives of
1 is illustrated in Scheme 1.
A novel chitin synthase inhibitor, RO-41-0986 (1: E:Z
2:1 mixture), was discovered from the Roche compound
libraries by random screening (Fig. 1). It speci®cally
inhibits Candida albicans chitin synthase 14,5 (IC50=0.07
mM) in a non-competitive manner. Though 1 is structu-
rally similar to terbina®ne6 [IC50 (CaChs1)=11.3 mM,
IC50 (squalene epoxidase)=0.03 mM)], it did not inhibit
squalene epoxidase (IC50>100 mM) at all (Fig. 2). Fur-
thermore, 1 showed only weak in vitro antifungal activity7
(IC50=3.3 mg/mL) against Candida albicans, and was
inactive against Cryprococcus and Aspergillus spp.
Despite its strong CaChs1 inhibitory activity, 1 did not
Compound 5 was synthesized from 2,6-dinitro-benzalde-
hyde 3 by treatment with methyl (triphenyl-phosphor-
anylidene) acetate in re¯uxing benzene followed by
hydrogenation over 10% Pd(OH)2-C in 78% yield. The
N-monoalkylation of the aromatic amine of 5 was carried
out by the conventional alkylation with the appropriate
alkyl halide (e.g., 1-bromo-6,6-dimethyl-2-hepten-4-yne8)
in the presence of K2CO3 or the reductive alkylation
with the appropriate aldehyde such as 6,6-dimethyl-hepta-
2,4-diynal and NaBH3CN. Then, the second N-alkyla-
*Corresponding author. Tel.: +81-467-47-2280; fax: +81-467-45-
6824; e-mail: nobuo.shimma@roche.com
0960-894X/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(00)00257-2

1460
K. Masubuchi et al. / Bioorg. Med. Chem. Lett. 10 (2000) 1459±1462
tion, N-acylation or N-sulfonylation was carried out by
the reductive alkylation, conventional acylation or sul-
fonylation, respectively, in a similar manner.
over 20% Pd(OH)2-C in 98% yield. Diyne 11 was
obtained by successive reductive alkylation with 6,6-
dimethyl-hepta-2,4-diynal and NaBH3CN followed by
treatment with formaldehyde solution and NaBH3CN.
Conversion of the hydroxyl group of 11 to the amine 12
was carried out in 3 steps; (i) O-mesylation, (ii) sub-
stitution of the resulting mesylate with azide group, and
(iii) reduction of the azide with Ph3P. The l-valylester
13 was synthesized by treatment with Boc-(l)-Val-
OpNP followed by deprotection with TFA. The phos-
phate derivative 14 was prepared by treatment of 11
with diethyl chloro-phosphate in 14% yield.
The hydrophilic derivatives (11±14) were synthesized
from 8 as outlined in Scheme 2.
Treatment of 8 with glyoxylic acid in the presence of
Cs2CO3 followed by methylation with Me2NCH(OMe)2
gave the hydroxylester 9 in 80% yield. The bicyclic
compound 10 was synthesized from 9 by hydrogenation
Results and Discussion
Preliminary SAR obtained from the random screening
result
Figure 1.
The preliminary SAR (CaChs1 inhibitory activity)
obtained from the random screening result indicated that
the important structural elements of 1 are the bicyclic
structure (cf. compounds 19 and 20), both N(1)-H and C-
2 carbonyl group (cf. compounds 15, 16 and 17), and the
unsaturated alkyl chain (cf. compound 18) (Fig. 3).
In addition, we found that saturation of the 3,4-double
bond of 1 did not a€ect the enzyme inhibitory activity.
Figure 2.
Modi®cation of the C-5 amino group: introduction of R2
The results of the modi®cation of the C-5 amino group
are summarized in Table 1. When an additional alkyl
group (R2) was introduced on the C-5 amino group,
their chitin synthase inhibitory activity was dramatically
improved. The N-butyl derivative 26 showed the max-
imum inhibitory activity: 200 times more active than 1.
However, the in vitro antifungal activity was retained
only when the alkyl group was methyl or ethyl. The
introduction of an acyl or sulfonyl group (R2) resulted
in lost activity.
Scheme 1. General synthetic route of RO-41-0986 derivatives. (a)
Ph3P=CHCO2CH3, benzene, re¯ux (100%); (b) H2, 10% Pd(OH)2-C,
MeOH, rt (78%); (c) RCH2Br, K2CO3, DMF, rt or RCHO,
NaBH3CN, AcOH, MeOH, rt (ꢀ62%); (d) R0CHO, NaBH3CN,
AcOH, MeOH, rt (ꢀ100%) or R0COCl, or R0SO2Cl.
Scheme 2. Synthetic route of the hydrophilic derivatives. (a) OHC-
CO2H, Cs2CO3, MeOH, re¯ux (88%); (b) Me2NCH(OMe)2, CH2Cl2, rt
(91%); (c) H2, 20% Pd(OH)2-C, MeOH, rt (98%); (d) (i) 6,6-dimethyl-
hepta-2,4-diynal, NaBH3CN, AcOH, MeOH, rt; (ii) HCHO,
NaBH3CN, AcOH, MeOH, rt (90%); (e) (i) MsCl, collidine, CH2Cl2,
0 ꢁC; (ii) NaN3, DMF, rt; (iii) Ph3P, THF, rt then H2O, 50 ꢁC (51%); (f)
Boc-(l)-Val-OpNP, DMAP, HOBT, THF, rt (49%); (g) TFA, CH2Cl2,
rt (100%); (h) (EtO)2POCl, DMAP, Pyridine, CH2Cl2, rt (14%).
Figure 3. Preliminary data obtained from the random screening.
(CaChs1:IC50=mM).

K. Masubuchi et al. / Bioorg. Med. Chem. Lett. 10 (2000) 1459±1462
1461
Modi®cation of the ene-yne side chain
mixture, the preferable geometry of the double bond
was investigated with the 3,4-dihydro derivatives of 1.
The result indicates that both isomers 22 and 31 showed
strong enzyme inhibitory activity, but the E-isomer was
4 times more active than the Z-isomer. The saturation
or replacement of this double bond with a phenyl group
resulted in reduction of inhibitory activity. Surprisingly,
the replacement of the double bond with the triple bond
(2) signi®cantly increased the CaChs1 inhibitory activity
(IC50=0.14 nM) as well as in vitro antifungal activity as
compared with 22. Thus, the unsaturated moiety was
®xed as a diyne group, and the terminal tert-butyl group
was further modi®ed (Table 3).
Since the unsaturated side chain was found to be impor-
tant for chitin synthase inhibitory activity from the
screening result, the further optimization of this moiety
was carried out. The results of the side chain modi®ca-
tion are summarized in Tables 2 and 3. Since 1 is an E/Z
Table 1. Modi®cation of C-5 amino group: R2a
The replacement of one methyl of the tert-butyl group
of 2 with OH, MeO or phenyl resulted in signi®cant loss
of activity. From these data, the best side chain was
identi®ed as 6,6-dimetyl-hepta-2,4-diynyl group.
Compound
R2
CaChs1
In vitro antifungal activity
C. albicans CY1002
IC50 (mg/mL)
IC50 (nM)
21
22
23
24
25
26
27
28
H
Methyl
Ethyl
n-Propyl
iso-Propyl
n-Butyl
n-Pentyl
Benzyl
Acetyl
70.9
2.77
0.9
0.68
1.23
0.35
1.1
54
775
391
71.3
3200
>200
0.19
0.27
Table 3. Modi®cation of the side chain: R6a
>200
>200
>200
>200
>200
>200
>200
3.30
Compound
R6
Enzyme
inhibition
CaChs1
In vitro
29
30
Mesyl
antifungal activity
IC50 (mg/mL)
Candida albicans CY1002
1
Nikkomycin Z
12
IC50 (nM)
aMedium: YPD broth, Incubation: 27 ꢁC, 24 h; Inoculum size:
1±2Â104 (CFU/mL).
2
0.14
0.07
34
35
7.88
213
200
>200
Table 2. Modi®cation of the side chain: R1a
36
1031
13,700
3200
3.5
>200
12
37
Compound
R1a
Enzyme
inhibition
CaChs1
In vitro
antifungal activity
IC50 (mg/mL)
Candida albicans
CY1002
Nikkomicin Z
aMedium: YPD broth, Incubation: 27 ꢁC, 24 h; Inoculum size:
1±2Â104 (CFU/mL).
IC50 (nM)
22
31
32
33
2.77
13.2
8.95
7.22
0.19
Table 4. In vitro antifungal spectrum of the compound 2 and nik-
komycin Za
95
Compound 2
IC50 (mg/mL)
Nikkomycin Z
IC50 (mg/mL)
Species
>200
>200
C. albicans CY1003
C. albicans CY3003
C. albicans CY1124b
C. glabrata CY1014
C. tropicalis CY1038
C. krusei CY1191
C. parapsilosis CY1028
C. neoformans CY1059c
A. fumigatus CF1003
0.10
0.09
0.17
0.47
0.07
0.31
0.50
81
7.0
15
>200
73
>200
13
>200
11
2
0.14
0.07
12
>200
43
Nikkomicin Z
3200
aMedium: YPD broth for yeast, YPDA for ®lamentous fungi; Incu-
bation: 27 ꢁC, Inoculum size: 1±2Â104 (CFU/mL), 24 h.
bAzole resistant AD.
aMedium: YPD broth, Incubation: 27 ꢁC, 24 h; Inoculum size:
1±2Â104 (CFU/mL).
c5-FC resistant 422.

1462
K. Masubuchi et al. / Bioorg. Med. Chem. Lett. 10 (2000) 1459±1462
Table 5. In vitro and in vivo antifungal activities of the compound 2 derivatives at C-3 position
Compound
C-3 substituent
CaChs1
In vitro antifungal activity IC50 (mg/mL)
In vivoa
IC50 (nM)
C. albicans
CY1002
C. neoformans
CY1057
A. fumigatus
CF1003
T/C (%)
at 5 mg/kg
2
11
13
12
H
OH
O-(l)-valyl
NH2
0.14
3
14.9
12.3
0.76
n.t.c
0.07
0.13
1.3
20
2.1
1.2
>200
>200
16
84
82
19
35
21
50
34
39
15
15
0.23
3.4
23
14
Flucanozole
OP(ˆO)(OE)2
n.t.c
38b
aTOKA assay in systemic candidiasis in mice; infection: C. albicans CY1002 5Â105 cells/mouse; treatment: 0, 2, 4 h after infection by iv; vehicle: 5%
DMSO 5% PEG300-30% HPCD; T/C=CFU (colony forming unit) of treatment/CFU of controlÂ100.
bTreatment: 0, 2, 4 h after infection by iv (1 mg/kg).
cn.t.=not tested.
The antifungal spectra of the compound 2 and nikko-
mycin Z are shown in Table 4. 2 was active against a wide
range of Candida species, including an azole-resistant
strain, weakly active against Aspergillus fumigatus, and
inactive against Cryprococcus neoformans. It was sig-
ni®cantly better than nikkomycin Z.
in vitro antifungal activity against most of Candida spp.,
including azole-resistant spp. and exhibit in vivo ecacy
in TOKA assay in mice systemic candidiasis model. Fur-
ther modi®cation study is in progress.
References and Notes
Modi®cation of pyrid-2-one moiety
1. Tuite, M. F. TIBTECH 1996, 14, 219.
2. Ruiz-Herrera, J. Fungal Cell Wall: Structure, Synthesis, and
Assembly; CRC Press, 1992; pp 1.
3. Zhang, D.; Miller, M. J. Curr. Pharm. Des. 1999, 5, 73.
4. Sudoh, M.; Watanabe, M.; Mio, T.; Nagahashi, S.;
Yamada-Okabe, H.; Takagi, M.; Arisawa, M. Microbiology
1995, 141, 2676.
5. Mio, T.; Yabe, T.; Sudoh, M.; Satoh, Y.; Nakajima, T.;
Arisawa, M.; Yamada-Okabe, H. J. Bacteriol. 1996, 178, 2416.
6. Ryder, N. S.; Mieth, H. Curr. Top Med. Mycol. 1991, 4,
220.
7. Antifungal susceptibility assays were performed by the
broth microdilution method according to M27-A guidelines
recommended by the National Committee for Clinical
Laboratory Standards (NCCLS).
Compound 2 exhibited only weak in vivo ecacy in the
target organ kidney assay (TOKA)9 in a systemic candi-
diasis model in mice (Table 5), despite its extremely potent
CaChs1 inhibitory activity. The possible reasons for the
weak in vivo activity are rapid metabolism (major meta-
bolism: N-demethylation in mouse liver microsome/37 ꢁC,
20 min), strong serum protein binding (>99%: mouse),
and low water solubility.
Thus, we synthesized more hydrophilic analogues of 2 as
summarized in Scheme 2. The (l )-valyl ester 13 showed
the improved in vivo ecacy in the TOKA assay as
compared with 2 (Table 5), though the ecacy of 13
was still weaker than that of ¯uconazole.
8. Stutz, A. J. Med. Chem. 1984, 27, 1539.
9. Abruzzo, G. K.; Flattery, A. M.; Gill, C. J.; Kong, L.;
Smith, J. G.; Krupa, D.; Pikounis, V. B.; Kropp, H.; Bartizal,
K. Antimicrob. Agents Chemother. 1995, 39, 1077.
In summary, we have identi®ed novel and speci®c CaChs1
inhibitors, RO-09-3024 (2) and its analogues, which show