Discovery of highly potent novel antifungal azoles by structure-based rational design

Article abstract of DOI:10.1016/j.bmcl.2009.07.144

On the basis of the active site of lanosterol 14α-demethylase from Candida albicans (CACYP51), a series of new azoles were designed and synthesized. All the new azoles show excellent in vitro activity against most of the tested pathogenic fungi, which represent a class of promising leads for the development of novel antifungal agents. The MIC₈₀ value of compounds 8c, 8i and 8n against C. albicans is 0.001 μg/mL, indicating that these compounds are more potent than fluconazole, itraconazole and voriconazole. Flexible molecular docking was used to analyze the structure-activity relationships (SARs) of the compounds. The designed compounds interact with CACYP51 through hydrophobic, van der Waals and hydrogen-bonding interactions.

Full text of DOI:10.1016/j.bmcl.2009.07.144

Bioorganic & Medicinal Chemistry Letters 19 (2009) 5965–5969
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Discovery of highly potent novel antifungal azoles by structure-based
rational design
a
Wenya Wang a, Chunquan Sheng a, , Xiaoying Che , Haitao Ji b,c, Yongbing Cao a, Zhenyuan Miao a,
*
Jianzhong Yao a, Wannian Zhang a,
*
a School of Pharmacy, Second Military Medical University, 325 Guohe Road, Shanghai 200433, People’s Republic of China
b Department of Chemistry, Northwestern University, Evanston, IL 60208-3113, USA
c Department of Biochemistry, Molecular Biology, and Cell Biology, and Center for Drug Discovery and Chemical Biology, Northwestern University, Evanston, IL 60208-3113, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
On the basis of the active site of lanosterol 14a-demethylase from Candida albicans (CACYP51), a series of
new azoles were designed and synthesized. All the new azoles show excellent in vitro activity against
most of the tested pathogenic fungi, which represent a class of promising leads for the development of
novel antifungal agents. The MIC80 value of compounds 8c, 8i and 8n against C. albicans is 0.001 lg/
mL, indicating that these compounds are more potent than fluconazole, itraconazole and voriconazole.
Flexible molecular docking was used to analyze the structure–activity relationships (SARs) of the com-
pounds. The designed compounds interact with CACYP51 through hydrophobic, van der Waals and hydro-
gen-bonding interactions.
Received 20 May 2009
Revised 14 July 2009
Accepted 30 July 2009
Available online 3 August 2009
Keywords:
Azole
Rational design
Molecular docking
Antifungal activity
Ó 2009 Published by Elsevier Ltd.
The incidence of systemic fungal infections has been increasing
dramatically due to an increase in the number of immunocompro-
mised hosts, such as patients undergoing anticancer chemotherapy
or organ transplants and patients with AIDS.1 In contrast to the
large number of antibacterial antibiotics, there are very few anti-
fungal agents that can be used for life-threatening fungal infec-
tions. These drugs are amphotericin B,2 5-fluorocytosine, azoles
(such as fluconazole and itraconazole),3 and echinocandins (such
as caspofungin and micafungin).4 Because of their high therapeutic
index, azoles are first-line drugs for the treatment of invasive fun-
gal infections. Unfortunately, the broad use of azoles has led to
development of severe resistance,5,6 which significantly reduced
the efficacy of them. This situation has led to an ongoing search
for new azoles. Several novel triazole antifungal agents (Fig. 1),
such as voriconazole,7 posaconazole,8 ravuconazole9 and albaco-
nazole,10 are marketed or currently in the late stages of clinical
trials.
cells.12 Because eukaryotic CYP51s are membrane associated pro-
teins, solving their crystal structures remains a challenge. In our
continual interest in rational antifungal drug design, three-dimen-
sional (3D) models of CYP51 from Candida albicans (CACYP51) and
Aspergillus fumigatus (AFCYP51) have been constructed through
homology modeling.13,14 The mechanism of azoles binding with
CACYP51 has been investigated by flexible molecular docking13,15
and site-directed mutagenesis.16 On the basis of the results from
molecular modeling, highly potent new azoles15 and novel non-
azole CACYP51 inhibitors17 have been designed and synthesized,
which demonstrates that the utilization of structural information
of fungal CYP51s can accelerate the discovery of novel antifungal
agents. In this Letter, structure-based rational drug design was suc-
cessfully used to the discovery a series of new azoles with excellent
in vitro antifungal activity and broad antifungal spectrum.
The chemical synthesis of compounds 8a–p and 9a–c is out-
lined in Schemes 1 and 2, respectively. As a key intermediate of
our designed compounds, the oxirane compound 4 was synthe-
sized by our reported procedure.15 The N-methyl-substituted
phenoxybutan-1-amine side chains 7a–p were synthesized via
two steps. Excess 1,4-dibromobutane were treated with various
substituted phenols to give 1-(4-bromobutoxy)-substituted ben-
zenes 6a–p. Compounds 6a–p were subsequently reacted with
methylamine in EtOH at room temperature to afford side chains
7a–p. The target compounds 8a–p were obtained by treating epox-
ide 4 with side chains 7a–p in the presence of triethylamine and
EtOH at 80 °C with moderate to high yields. Compounds 9a, 9b
Azole antifungals act by competitive inhibition of the lanosterol
14a
-demethylase (CYP51),11 which is the key enzyme in sterol bio-
synthesis of fungi. Selective inhibition of CYP51 would cause
depletion of ergosterol and accumulation of lanosterol and other
14-methyl sterols resulting in the growth inhibition of fungal
* Corresponding authors. Tel./fax: +86 21 81871233 (C.S.); +86 21 81870243
(W.Z.).
E-mail addresses: shengcq@hotmail.com (C. Sheng), zhangwnk@hotmail.com
(W. Zhang).
0960-894X/$ - see front matter Ó 2009 Published by Elsevier Ltd.
doi:10.1016/j.bmcl.2009.07.144

5966
W. Wang et al. / Bioorg. Med. Chem. Lett. 19 (2009) 5965–5969
O
N
N
1
OH 3
N
O
CH2
Cl
O
N
N
N
N
N
N
N
F
O
N CH2
2
N
N
N
F
Cl
Itraconazole
Fluconazole
O
N
N
OH
CH3
O
CH2
O
N
N
OH
N
N
N
N
N
CH2
N
N
N
F
N
F
F
F
F
Voriconazole
Posaconazole
CH3
O
N
CH3
OH
OH
N
N
N
N
N
F
N
CN
N
N
S
F
Cl
F
F
Albaconazole
Ravuconazole
Figure 1. Chemical structures of azole antifungal agents.
O
O
N
Cl
N
O
F
N
N
a
c
b
.
N
F
N
CH3SO3H
F
F
F
F
F
F
4
1
2
3
R
R
d
e
Br (CH2)4 Br
O
(CH2)4 Br
O(CH2)4 NHCH3
5
6a-p
7a-p
OH CH3
R
8a R = 4-CH3
8b R = 2-CH3
8c R = 4-Cl
8d R = 2-Cl
8i R = 4-Br
8j R = 3-Br
8k R = 2-Br
N
f
N
N
N
F
(CH2)4O
8l R = 4-OCH2Ph
8e R = 4-C(CH3)3
8f R = 3-CH3
8g R = 4-F
8m R = 4-OCF3
8n R = 4-NO2
8o R = 3-NO2
8p R = 2-NO2
F
8h R = 4-OCH3
8a-p
Scheme 1. Synthesis of the target compounds 8a–p. Reagents and conditions: (a) ClCH2COCl, AlCl3, CH2Cl2, 40 °C, 3 h, 50%; (b) triazole, K2CO3, CH2Cl2, rt, 24 h, 70.0%; (c)
(CH3)3SOI, NaOH, toluene, 60 °C, 3 h, 62.3%; (d) phenol, K2CO3, DMF, 70 °C, 2 h, 91.8–92.3%; (e) methylamine alcohol solution, rt, 12 h, 97–98%; (f) 4, Et3N, EtOH, reflux, 9 h,
51.2–60.3%.
and 9c were synthesized by the reduction of the nitro groups of
compounds 8n, 8o and 8p in the presence of Raney Ni and hydra-
zine hydrate (Scheme 2). All the target compounds were obtained
as racemates.
In our previous studies, the active site of CACYP51 has been well
characterized by multiple copy simultaneous search (MCSS).17
Moreover, the binding mode of azole antifungal agents with CA-
CYP51 can provide useful information for the rational design of

W. Wang et al. / Bioorg. Med. Chem. Lett. 19 (2009) 5965–5969
5967
OH CH3
OH CH3
R
R
a
N
N
N
N
N
F
(CH2)4O
N
N
N
F
(CH2)4O
F
F
9a R = 4-NH2
9b R = 3-NH2
9c R = 2-NH2
8n R = 4-NO2
8o R = 3-NO2
8p R = 2-NO2
Scheme 2. Synthesis of the target compounds 9a–c. Reagents and conditions: (a) Raney Ni, NH2NH2, EtOH, rt, 5 h, 94.5–96.4%.
new antifungal azoles. As discussed earlier,17 the active site of CA-
CYP51 can be divided into four pockets. The S1 pocket represents
the hydrophilic hydrogen bond binding site, which is defined by
Gln309, His310, and Ser312. Our docking studies indicate that
the C2 hydroxyl group of azole antifungal agents interacts with
the S1 pocket, possibly through the hydrogen-bonding interaction
with His310 mediated by crystal waters.13,15 The S2 pocket is
above the heme ring representing the core hydrophobic area. The
triazole ring of azole antifungal agents binds to the S2 pocket
through the formation of a coordination bond with the iron atom
of the heme group. The S3 pocket represents a narrow and hydro-
phobic cleft (facing BC loop). The difluorophenyl or dichlorophenyl
group of the azole antifungal agents is located in the S3 pocket and
forms hydrophobic interaction with Ala114, Phe126, Leu139,
Ile304 and Met306. Because the triazole ring, diflurophenyl group
and C2 hydroxyl group are regarded as essential pharmacophoric
elements of azole antifungal agents,13 they were retained in the
present study. The difference between azole antifungal agents
mainly lies in the side chains attached to C3 (Fig. 1), and most of
the recent efforts aim to optimize them.15,18–20 The C3 side chains
of the antifungal azoles are found to be located in the S4 pocket,
which represents a hydrophobic hydrogen bond binding site (fac-
ing FG loop). The fundamental purpose of the present drug design
is to find a new type of side chain, which can be well accommo-
dated in the S4 pocket. On the basis of the properties of the S4
pocket, a satisfied side chain should meet the following criteria.
(1) Considering the hydrophobic nature of the S4 pocket, a hydro-
phobic side chain is necessary; (2) The side chain should have suit-
able length. Although
a long side chain can form stronger
hydrophobic and van der Waals interactions with CACYP51, it could
also result in poor water solubility and low bioavailability. There-
fore, a balance of affinity and drug-like properties should be
achieved; (3) A hydrogen bond donor or acceptor should be intro-
duced into the side chain, because an additional hydrogen-bonding
interaction will further improve the affinity and specificity of the
inhibitors. Ser378 is a well validated residue for the hydrogen-
bonding interactions.15,17
On the basis of the above hypothesis, compounds 8a–p and 9a–
c were designed. The binding mode of the compounds was vali-
dated by flexible molecular docking (Affinity module within
InsightII 2000 software package21). Figure 2 shows the docking
conformation of compounds 8i and 8n in the active site of CA-
CYP51. The C3 side chain of the compounds was oriented into
the S4 pocket with an extended conformation. The N-methyl group
formed hydrophobic and van der Waals interactions with Phe228.
The butyl group mimics the alkyl chain of the natural substrate
and interacted with the surrounding hydrophobic residues lined
with Val509 and Val510. Hydrogen-bonding interaction between
the oxygen atom of the side chain and Ser378 was observed. The
substituted terminal phenyl group interacted with surrounding
hydrophobic residues such as Leu376, Phe380, Leu461, Val404
and Met508. Moreover, the nitro group of compound 8n formed
Figure 2. The docking conformation of compound 8i (A) and 8n (B) in the active site of CACYP51. Important residues interacting with the compounds are shown and
hydrogen bonds are displayed through red dotted lines. The images were generated using InsightII 2000 software package.

5968
W. Wang et al. / Bioorg. Med. Chem. Lett. 19 (2009) 5965–5969
additional two hydrogen bonds with Arg381 (Fig. 2b). Compared to
fluconazole, the designed compounds bound to CACYP51 with a
lower value of interaction energy (Table 1).
In vitro antifungal activity of the synthesized compounds is
listed in Table 2. In general, all the target compounds show excel-
lent activity against most the tested fungal pathogens. The target
compounds reveal the highest activity against C. albicans and C.
tropicalis. Most of the compounds are more potent than fluconaz-
ole, itraconazole and voriconazole with their minimal inhibitory
The in vitro antifungal activity assay indicates that most of the
4-substituted and 3-substituted derivatives show higher anti-
fungal activity than 2-substituted derivatives. The docking model
shows that the active site of CACYP51 at position 2 of the bound
compound is not large enough to accommodate an additional
group. However, position 3 and 4 of the phenol group contains
an extra small pocket defined by Leu461, Val404, Met508 and
Leu403. Among the 4-substituted derivatives, their SARs are not
so obvious. In general, nitro group, halogen and small alkyl substi-
tuted compounds (e.g., compounds 8a, 8c, 8h, 8i and 8n) are favor-
able for the antifungal activity. Compound 8n has excellent
antifungal activity because the 4-nitro group can form two hydro-
gen bonds with Arg381 (Fig. 2). The hydrogen bonds will be broken
if the 4-nitro group of compound 8n is moved to the position 3 and
2 (compounds 8o and 8p), or reduced to the amino group (com-
pound 9a), which results in the decrease of the antifungal activity
by 16–64-fold. On the other hand, compounds 8c and 8i cannot
form hydrogen-bonding interaction with Arg381, but they have
the same activity as compound 8n. From the docking results, com-
pounds 8c and 8i have similar interaction energies with CACYP51
(Table 1) as compound 8n, because the halogen substitutions (Cl
or Br) at position 4 can form additional hydrophobic interaction
with surrounding Val404, Met508 and Leu403. If this position is
substituted by a less hydrophobic fluorine (compound 8g), the
antifungal activity is decreased to some extent.
concentrations (MIC80) values in the range of 0.0156–0.001 lg/
mL. Cryptococcus neoformans has a worldwide distribution and is
the most common cause of life-threatening fungal infections. All
the target compounds show higher inhibitory activity against C.
neoformans than fluconazole and itraconazole with their MIC80 val-
ues in the range of 0.25–0.004 lg/mL. Especially, the inhibitory
activity of compounds 8i and 8k is fourfold higher than that of
voriconazole. Fluconazole is not effective against A. fumigatus,
while some of our compounds show moderate activity. For exam-
ple, the MIC80 values of compounds 8c and 8i against A. fumigatus
are 4 lg/mL. However, they are less active than itraconazole and
voriconazole. For the dermatophytes (i.e., Trichophyton rubrum
and Microsporum gypseum), most of the compounds show higher
activity than fluconazole and itraconazole with their MIC80 values
in the range of 0.0625–0.0156 lg/mL. Compounds 8c, 8i, and 8n
exhibited strong in vitro antifungal activity with broad antifungal
spectrum, which were worthy of further evaluation.
In summary, computer modeling was successfully used to the
rational design of novel antifungal azoles. In vitro antifungal activ-
ity assay indicates that the new azoles show excellent activity
against both systemic pathogenic fungi and dermatophytes. Most
of the compounds show higher activity than fluconazole and itrac-
Table 1
Calculated interaction energies (kcal/mol) for the complexes of representative
compounds with the active site of CACYP51
onaozle with MIC80 values in the range of 0.0156–0.001
lg/mL. In
particular, the most active compounds 8c, 8i and 8n show broad
antifungal spectrum and are more potent than fluconazole, itraco-
nazole and voriconazole, which are promising leads for the devel-
opment of novel antifungal agents. Moreover, the therapeutic side
effects of azole antifungal agents are partly due to the interactions
of azoles with human CYP51. Ser378 is conserved across the fungal
CYP51 enzymes and the corresponding residue in human is Ile379.
Because the synthesized compounds can form hydrogen-bonding
interaction with Ser378, they may show better selectivity toward
fungal CYP51 than marketed azole antifungal agents. Further phar-
macological and toxicological evaluation of these highly potent
compounds is in progress.
Compounds
Evdw
Eelect
Etotal
8a
8b
8c
8i
8n
8o
9a
À62.2
À60.1
À64.9
À65.5
À61.9
À62.4
À60.4
À54.8
À3.4
À1.2
À5.5
À5.2
À9.6
À2.1
À2.3
À3.2
À65.6
À61.3
À70.4
À70.7
À71.5
À63.5
À62.7
À58.0
Fluconazole
Table 2
In vitro antifungal activity of the target compounds (MIC80
a
, l )
g mLÀ1
Compd
C. alb.
C. neo.
C. tro.
T. rub.
A. fum.
M. gyp.
Acknowledgements
8a
8b
8c
8d
8e
8f
8g
8h
8i
0.0156
0.0625
0.001
0.0156
0.0625
0.0156
0.0156
0.0156
0.001
0.0625
0.0156
0.0156
0.0156
0.001
0.0156
0.0625
0.0625
0.0156
0.0625
0.25
0.0625
0.0156
0.0156
0.0156
0.25
0.0625
0.0625
0.0625
0.004
0.0625
0.004
0.0156
0.0625
0.0156
0.0625
0.25
0.004
0.0156
0.001
0.0156
0.0625
0.0156
0.004
0.004
0.001
0.001
0.0156
0.0156
0.004
0.004
0.004
0.0625
0.0156
0.0156
0.0156
1
0.0156
0.0625
0.0625
0.0625
0.0625
0.0625
0.0156
0.0156
0.0156
0.0625
0.0625
0.0625
0.0156
0.0625
0.0625
0.0625
1
16
>64
4
>64
>64
64
16
16
4
64
0.0625
0.0625
0.0156
0.0625
0.0625
0.0625
0.0625
0.0625
0.0156
0.0156
0.0625
0.0156
0.0625
0.0156
0.0625
0.0625
0.25
This work was supported in part by the National Natural
Science Foundation of China (Grant Nos. 30930107), Shanghai
Rising-Star Program (Grant Nos. 09QA1407000) and Shanghai
Leading Academic Discipline Project (Project Nos. B906).
Supplementary data
8j
8k
8l
>64
32
>64
64
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2009.07.144.
8m
8n
8o
8p
9a
9b
9c
FLZ
ITZ
VOR
64
References and notes
>64
>64
>64
>64
>64
2
0.25
0.25
1
1
0.5
0.0156
1. Fridkin, S. K.; Jarvis, W. R. Clin. Microbiol. Rev. 1996, 9, 499.
2. Gallis, H. A.; Drew, R. H.; Pickard, W. W. Rev. Infect. Dis. 1990, 12, 308.
3. Sheehan, D. J.; Hitchcock, C. A.; Sibley, C. M. Clin. Microbiol. Rev. 1999, 12, 40.
4. Denning, D. W. J. Antimicrob. Chemother. 2002, 49, 889.
5. Casalinuovo, I. A.; Di Francesco, P.; Garaci, E. Eur. Rev. Med. Pharmacol. Sci. 2004,
8, 69.
0.25
1
0.25
0.125
0.0156
0.0625
0.25
1
0.125
0.0156
0.125
0.0156
0.0156
0.001
0.25
6. Hoffman, H. L.; Ernst, E. J.; Klepser, M. E. Expert Opin. Invest. Drugs
2000, 9, 593.
7. Chandrasekar, P. H.; Manavathu, E. Drugs Today (Barc) 2001, 37, 135.
8. Herbrecht, R. Int. J. Clin. Pract. 2004, 58, 612.
a
Abbreviations: C. alb., Candida albicans; C. neo., Cryptococcus neoformans; C. tro.,
Candida tropicalis; T. rub., Trichophyton rubrum; A. fum., Aspergillus fumigatus; M.
gyp., Microsporum gypseum; FLZ: Fluconazole; ITZ: Itraconazole; VOR: Voriconazole.

W. Wang et al. / Bioorg. Med. Chem. Lett. 19 (2009) 5965–5969
5969
9. Arikan, S.; Rex, J. H. Curr. Opin. Invest. Drugs 2002, 3, 555.
10. Capilla, J.; Yustes, C.; Mayayo, E.; Fernandez, B.; Ortoneda, M.; Pastor, F. J.;
Guarro, J. Antimicrob. Agents Chemother. 2003, 47, 1948.
11. Aoyama, Y.; Yoshida, Y.; Sato, R. J. Biol. Chem. 1984, 259, 1661.
12. Lamb, D. C.; Kelly, D. E.; Venkateswarlu, K.; Manning, N. J.; Bligh, H. F.; Schunck,
W. H.; Kelly, S. L. Biochemistry 1999, 38, 8733.
16. Chen, S. H.; Sheng, C. Q.; Xu, X. H.; Jiang, Y. Y.; Zhang, W. N.; He, C. Biol. Pharm.
Bull. 2007, 30, 1246.
17. Ji, H.; Zhang, W.; Zhang, M.; Kudo, M.; Aoyama, Y.; Yoshida, Y.; Sheng, C.; Song,
Y.; Yang, S.; Zhou, Y.; Lu, J.; Zhu, J. J. Med. Chem. 2003, 46, 474.
18. Aher, N. G.; Pore, V. S.; Mishra, N. N.; Kumar, A.; Shukla, P. K.; Sharma, A.; Bhat,
M. K. Bioorg. Med. Chem. Lett. 2009, 19, 759.
13. Ji, H.; Zhang, W.; Zhou, Y.; Zhang, M.; Zhu, J.; Song, Y.; Lu, J. J. Med. Chem. 2000,
43, 2493.
19. Giraud, F.; Guillon, R.; Loge, C.; Pagniez, F.; Picot, C.; Le Borgne, M.; Le Pape, P.
Bioorg. Med. Chem. Lett. 2009, 19, 301.
14. Sheng, C.; Zhang, W.; Zhang, M.; Song, Y.; Ji, H.; Zhu, J.; Yao, J.; Yu, J.; Yang, S.;
Zhou, Y.; Lu, J. J. Biomol. Struct. Dyn. 2004, 22, 91.
20. Lebouvier, N.; Pagniez, F.; Duflos, M.; Le Pape, P.; Na, Y. M.; Le Baut, G.; Le
Borgne, M. Bioorg. Med. Chem. Lett. 2007, 17, 3686.
15. Sheng, C.; Zhang, W.; Ji, H.; Zhang, M.; Song, Y.; Xu, H.; Zhu, J.; Miao, Z.; Jiang,
Q.; Yao, J.; Zhou, Y.; Lu, J. J. Med. Chem. 2006, 49, 2512.
21. InsightII 2000: Molecular Simulation Inc. 9685 Scranton Road, San Diego, CA
92121-3752, 1999.