Targeting fungal virulence factor by small molecules: Structure-based discovery of novel secreted aspartic protease 2 (SAP2) inhibitors

Article abstract of DOI:10.1016/j.ejmech.2020.112515

Secreted aspartic protease 2 (SAP2), a kind of virulence factor, is an emerging new antifungal target. Using docking-based virtual screening and structure-based inhibitor design, a series of novel SAP2 inhibitors were successfully identified. Among them, indolone derivative 24a showed potent SAP2 inhibitory activity (IC₅₀ = 0.92 μM). It blocked fungi biofilm and hypha formation by down-regulating the expression of genes SAP2, ECE1, ALS3 and EFG1. As a virulence factor inhibitor, compound 24a was inactive in vitro and showed potent in vivo efficacy in a murine model of invasive candidiasis. It represents a promising lead compound for the discovery of novel antifungal agents.

Full text of DOI:10.1016/j.ejmech.2020.112515

Journal Pre-proof
Targeting Fungal Virulence Factor by Small Molecules: Structure-based Discovery of
Novel Secreted Aspartic Protease 2 (SAP2) Inhibitors
Chenglan Li, Yang Liu, Shancao Wu, Guiyan Han, Jie Tu, Guoqiang Dong, Na Liu,
Chunquan Sheng
PII:
S0223-5234(20)30487-6
DOI:
https://doi.org/10.1016/j.ejmech.2020.112515
EJMECH 112515
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 22 March 2020
Revised Date: 19 May 2020
Accepted Date: 29 May 2020
Please cite this article as: C. Li, Y. Liu, S. Wu, G. Han, J. Tu, G. Dong, N. Liu, C. Sheng, Targeting
Fungal Virulence Factor by Small Molecules: Structure-based Discovery of Novel Secreted Aspartic
Protease 2 (SAP2) Inhibitors, European Journal of Medicinal Chemistry, https://doi.org/10.1016/
j.ejmech.2020.112515.
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition
of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of
record. This version will undergo additional copyediting, typesetting and review before it is published
in its final form, but we are providing this version to give early visibility of the article. Please note that,
during the production process, errors may be discovered which could affect the content, and all legal
disclaimers that apply to the journal pertain.
© 2020 Elsevier Masson SAS. All rights reserved.

Graphical Abstract

Targeting Fungal Virulence Factor by Small Molecules:
Structure-based Discovery of Novel Secreted Aspartic
Protease 2 (SAP2) Inhibitors
Chenglan Li a, b, 1, Yang Liu c, 1, Shancao Wu b, 1, Guiyan Han b, Jie Tu b, Guoqiang
Dong b, *, Na Liu b, *, Chunquan Sheng a, b, *
a. School of Pharmacy, Ningxia Medical University, 1160 Shengli Street, Yinchuan
750004, China
b. Department of Medicinal Chemistry, School of Pharmacy, Second Military Medical
University, 325 Guohe Road, Shanghai 200433, China
c. Department of Pharmacy, No. 971 Hospital of PLA, Qingdao 266071, China
* Corresponding authors
E-mail address: shengcq@smmu.edu.cn (C. Sheng), liuna66@aliyun.com (N. Liu),
dgq-81@163.com (G. Dong)
1 These authors contributed equally to this work

Abstract
Secreted aspartic protease 2 (SAP2), a kind of virulence factor, is an emerging
new antifungal target. Using docking-based virtual screening and structure-based
inhibitor design, a series of novel SAP2 inhibitors were successfully identified.
Among them, indolone derivative 24a showed potent SAP2 inhibitory activity (IC50 =
0.92 μM). It blocked fungi biofilm and hypha formation by down-regulating the
expression of genes SAP2, ECE1, ALS3 and EFG1. As a virulence factor inhibitor,
compound 24a was inactive in vitro and showed potent in vivo efficacy in a murine
model of invasive candidiasis. It represents a promising lead compound for the
discovery of novel antifungal agents.
KEY WORDS: Secreted aspartic protease 2, antifungal, virulence factor.

1. Introduction
Invasive Candida infections increase dramatically in recent years, representing a
serious disease with high morbidity and mortality [1]. Antifungal agents including
azoles, polyenes and chinocandins are used as frontline treatments of invasive
candidiasis. However, pathogenic fungi can be only controlled ephemerally due to the
rapid development of resistance [2-4]. Thus, discovery of novel antifungal agents
acting on new targets is urgently needed. Secreted aspartic proteases (SAPs) were
proven to be a crucial virulent factor during fungal infections [5, 6] . There are 10
SAP genes (SAP1-10) in the C. albicans genome [5, 7]. SAP genes in Candida
parapsilosis (C. parapsilosis), Candida tropicalis (C. tropicalis) and Candida
dubliniensis (C. dubliniensis) are orthologous genes [8-11]. SAP1-3 are secreted
mainly in yeast form of C. albicans, while SAP4-6 are secreted in hypha form [12,
13]. Among them, the structure and function of SAP2 has been extensively studied,
which was supposed to be a potential target for the development of novel antifungal
agents.
Fungi biofilms (particularly Candida) are often highly assembled on the surface
of medical devices implanted in individual patients and epithelial cells, which cause
the development of drug resistance [14-17]. SAPs were regarded as important factors
for biofilm formation [13, 18]. It is reported that SAPs production enhanced the
exposure of C. albicans biofilms to sub-inhibitory concentrations of fluconazole (FLC)
[13, 18]. However, the effects of SAP2 inhibitors on the biofilm formation still remain
unknown.
Peptide and peptidominetic SAP2 inhibitors have been reported [19]. However,
most of them lack antifungal potency possibly due to unfavorable physicochemical
and pharmacokinetic properties. Thus, the identification of small molecule SAP2
inhibitors is highly desirable for target validation and antifungal lead discovery. In our
previous studies, the first class of SAP2 inhibitors were identified by virtual screening
of the SPECS database [19]. Inspired by this proof-of-concept study, there is an urgent

need to design structurally diverse SAP2 inhibitors, which can be used as chemical
probes or lead compounds to explore biological and pharmacological functions of
SAP2 and discover novel antifungal agents.
To design new classes of SAP2 inhibitors, herein structure-based virtual
screening was performed using a larger compound library. After structure-based hit
optimization, a series of new SAP2 inhibitors were discovered. Among them,
indolone inhibitor 24a showed potent SAP2 inhibitory activity and could significantly
reduce the kidney fungal burdens in a murine model of invasive candidiasis.
Importantly, the biofilm inhibitory activity of SAP2 inhibitors was confirmed for the
first time. Compound 24a significantly inhibited biofilm formation and filamentation
of azole-resistant C. albicans and C. tropicalis.
2. Results and Discussion
2.1 Structure-based Virtual Screening of the ChemDiv Database
The virtual screening procedure is similar to that of our previous protocol (Fig.
1A) [19]. A total of about 713,000 compounds in ChemDiv database were docked and
ranked by GOLD, which was proven to be an accurate method to dock SAP2
inhibitors. The fitness score of 85 was used as a cutoff value for the selection of the
initial hits. As a result, top-ranked 7851 molecules were obtained and subjected to
docking validation and visual inspection. The selection of the candidate molecules
was based the following criteria: (1) effective mimicking of the interaction of peptide
inhibitor A70450; (2) formation of hydrogen bonding interactions with residues
involved in A704500-SAP2 interaction; (3) compounds with too many rotatable bonds
were excluded for further evaluation; (4) the binding mode of the compounds can be
reproduced by at least two docking methods. After filtration, 582 candidate
compounds were subjected to similarity analysis and 100 structurally diverse
candidates were designated for purchase. Finally, only 50 compounds turned out to be
commercially available for in vitro assaying (Fig. S1 in Supporting Information).

Fig. 1 Discovery of new SAP2 inhibitors by virtual screening. (A) Protocol for virtual
screening of small-molecule SAP2 inhibitors; (B) SAP2 inhibitory activity of
compounds selected from virtual screening at the concentration of 100 μM. (C)
Chemical structures and inhibitory activity of four novel SAP2 inhibitors.
2.2 SAP2 Inhibitory Activity Assay and Molecular Docking
Using a spectrophotometric assay [20-23], initial screening of C. albicans SAP2
inhibitory activity was performed at the concentration of 100 μM. Peptide inhibitor
pepstatin A (PepA) was used as the positive control. The results revealed that 8 out of
50 compounds showed an inhibitory rate larger than 40% with a hit rate of 16% (Fig.
1B and Table S1 in Supporting Information). The IC50 values were further determined
for 4 compounds with inhibitory rate larger than 50% and their IC50 values are in the

range of 5.04 μM to 92.3μM (Fig. 1C). Among them, compound J29 showed the best
SAP2 inhibitory activity (IC50 = 5.04 μM). Then, the binding mode of inhibitor J29
was investigated by molecule docking. As shown in Fig. 2, it bound with SAP2 by
mimicking the interaction of A704500. The pyrazolone scaffold was located at the
central site of S1, S2, S1’ and S2’ pocket. The C1, C3 and C4 side chain of pyrazolone
interacted with the S1’, S1 and S2 pocket, respectively. The C3-phenyl group formed
aromatic π-π interaction with Tyr84 and hydrophobic interactions with Ile30, Ile119
and Ile123. The C4 phenyl group formed hydrophobic interactions with Ile 305 and
Ile223. Notably, the terminal sulfonic acid group of the C4-side chain formed two
hydrogen bonds with Tyr225.
Fig. 2 Binding modes of pyrazolone SAP2 inhibitor J29. (A) Proposed binding pose
of inhibitor J29 in the active site of SAP2. The inhibitors are shown in the stick
format and hydrogen bonds are shown in dotted red lines. (B) The comparison of
binding modes of small-molecule inhibitor J29 (carbon atoms in purple) with peptide
inhibitor A-70450 (carbon atoms in green) are shown on the surface of SAP2. The
ligand binding pockets of SAP2 are labeled on the surface of the active site. The
figures are generated using PyMol (http://www.pymol.org).
2.3 Structure-Activity Relationship (SAR) of Pyrazolone SAP2 Inhibitor J29
Compound J29 was reported as a Shp-2 inhibitors for the treatment of cancer and
the benzenesulfonic acid group is necessary for the activity [24]. To validate the
importance of sulfonic acid in hydrogen bonding interaction with SAP2, compounds
without C-ring acid group were designed and synthesized (6a-6e). Moreover, the

diazanyl group was unfavorable for the drug-like properties, which was removed in
the following SAR studies (Fig. 3). In consistence with the binding mode, these
compounds were almost inactive at the concentration of 100 μM (Table 1). Then, a
series of compounds with C-ring acid bioisosteres were synthesized and assayed
(5a-5y). The SAP2 inhibitory activity was retained for most of the target compounds,
further highlighting the importance of the acid group. For the carboxylic derivatives,
compounds with di-carboxylic side chains were generally more active than the
mono-substituted analogues. Moreover, the 2,4-disubstitution was better than the
3,4-disubstitution. Compounds 5m (IC50 = 10.05 μM) and 5o (IC50 = 10.75 μM) are
best compounds in this series. Moreover, A-ring substitutions also had different
effects on the SAP2 inhibitory activity. Among the di-substituted derivatives, the
4-chloro substitution or no substitution were generally more favorable for the activity.

Fig. 3 Hit optimization strategy of the pyrazolone SAP2 inhibitor.
In order to explain the SARs, 2,4-disubstituted derivative 5m and
3,4-disubstituted derivative 5t were subjected for molecular docking. As shown in Fig.
4, their general binding conformation in the active site of SAP2 was similar to that of
inhibitor J29. Their C1, C3, C4-rings interacted with the S1’, S1 and S2 pockets,
respectively. The 2,4- and 3,4-dicarboxylic side chains had different orientations in
the S2 pocket. Compound 5m formed four hydrogen bonds with SAP2. The C2 ether
oxygen atom and carboxyl group formed two hydrogen bonds with Thr221 and
Thr222, respectively. Another two hydrogen bonds were observed between the C4
carboxyl group and Val300 and Asn301. In contrast, compound 5t only formed three
hydrogen bonds with Ile223, Val300 and Asn301. Due to weaker hydrogen bonding

interaction, compound showed decreased SAP2 inhibitory activity.
Table 1. SAP2 inhibitory activity of the pyrazolone inhibitors
Compd
5a
R1
H
R2
Inhibition (100 μM)
IC50 (μM)
62.15
55.48
75.15
nd
2-OCH2COOH
2-OCH2COOH
2-OCH2COOH
2-OCH2COOH
2-OCH2COOH
2-OCH2COOH
4-OCH2COOH
4-OCH2COOH
4-OCH2COOH
4-OCH2COOH
4-OCH2COOH
4-OCH2COOH
2,4-diOCH2COOH
2,4-diOCH2COOH
2,4-diOCH2COOH
68%
70%
60%
46%
63%
64%
94%
90%
96%
71%
93%
95%
99%
99%
99%
3-Cl
4-Cl
4-F
5b
5c
5d
5e
4-Me
4-OMe
H
72.31
70.65
18.17
26.36
12.96
39.61
21.56
20.44
10.05
11.54
10.75
5f
5g
3-Cl
4-Cl
4-F
5h
5i
5j
4-Me
4-OMe
H
5k
5l
5m
5n
5o
3-Cl
4-Cl

4-F
2,4-diOCH2COOH
2,4-diOCH2COOH
2,4-diOCH2COOH
93%
99%
94%
86%
86%
82%
70%
83%
85%
88%
12%
10%
7%
17.94
13.53
15.67
32.72
18.46
22.72
46.86
33.74
22.34
21.47
nd
5p
5q
5r
4-Me
4-OMe
2,4-diF 2,4-diOCH2COOH
5s
H
3-Cl
4-Cl
4-F
4-Me
4-OMe
H
3,4-diOCH2COOH
3,4-diOCH2COOH
3,4-diOCH2COOH
3,4-diOCH2COOH
3,4-diOCH2COOH
3,4-diOCH2COOH
4-OH
5t
5u
5v
5w
5x
5y
6a
H
4-Br
nd
6b
6c
H
4-Et
nd
H
4-OEt
5%
nd
6d
6e
H
4-N(Me)2
11%
-
nd
-
-
0.028
PepA
nd = not determined

Fig. 4 Binding modes of pyrazolone SAP2 inhibitors 5m and 5t. (A, B) Proposed
binding pose of inhibitor 5m and 5t in the active site of SAP2. The inhibitors are
shown in the stick format and hydrogen bonds are shown in dotted red lines. (C, D)
The comparison of binding modes of small-molecule inhibitor 5m and 5t (carbon
atoms in purple) with peptide inhibitor A-70450 (carbon atoms in green) are shown on
the surface of SAP2. The ligand binding pockets of SAP2 are labeled on the surface
of the active site. The figures are generated using PyMol (http://www.pymol.org).
2.4 Scaffold Hopping of Pyrazolone SAP2 Inhibitor 5m
For further investigating the importance of the pyrazolone scaffold and
improving the SAP2 inhibition activity, pyrazolidine-3,5-dione (13),
imidazolidine-2,4-dione (17) and indolin-2-one (24a) were used to replace pyrazolone.
Interestingly, SAP2 inhibitory activates revealed that indolin-2-one derivative 24a
showed the best inhibitory activity (IC50 = 0.92 μM, Table 2), which was more potent
than pyrazolone inhibitor 5m. Then, the binding mode of inhibitor 24a and 24c were
investigated by molecule docking (Fig. 5 and S2 in Supporting Information). As

shown in Fig. 5, the 2,4-dicarboxyl groups formed four hydrogen bonds with Ser36,
As24a31, Glu193 and Arg195, respectively. The phenyl group of benzyl side chain
formed aromatic π-π interaction with Tyr225. The binding mode of compound 24c
with SAP2 was similar to that of 24a.
Table 2. SAP2 inhibitory activity of the derivatives of 5m
Compd
Inhibition（100 μM）
IC50（μM）
13
94%
91%
17.43
17
23.77
24a
100%
100%
0.92
PepA
0.022
Fig. 5 Binding mode of indolone SAP2 inhibitor 24a.
2.5 Chemistry
The synthetic routes of the pyrazolone derivatives are outlined in Schemes 1.
The starting material ethyl 3-oxo-3-phenylpropanoate (1) was condensed with various
substituted phenylhydrazine hydrochlorides in the presence of AcOH to afford

pyrazolone intermediates 2. Then, esters 4 were obtained by reacting 2 with various
aldehydes containing mono- or di-ester side chain. Finally, after hydrolysis of the
ester groups using LiOH, target compounds 5a-5y were obtained. By a similar route,
1,3-diphenyl-1H-pyrazol-5(4H)-one (2) reacted with various aldehydes in the
presence of piperidine and absolute alcohol, and then after hydrolysis of the ester
groups to afford target compounds 6a-6e (Scheme 2).
Scheme 1
Reagents and conditions: (a) substituted phenylhydrazine hydrochloride, acetic acid,
Et3N, reflux, 3h, 25-57 %; (b) piperidine, absolute alcohol, 60 , 6 h, 24-85%; (c)
LiOH, THF: MeOH: H2O =3: 2: 1, rt, 6 h, 53-90%.
Scheme 2
R2
a
6a R2 = 4-OH
6b
N
O
N
O
N
N
R2 = 4-Br
6c R2 = 4-Et
6d
R2 = 4-OEt
6e R2 = 4-N(Me)2
2
Reagents and conditions: (a) various aldehyde, piperidine, absolute alcohol, 60℃, 6
h, 65-85 %.
The synthetic route of the pyrazolidinedione derivatives is shown in Scheme 3.

Starting from di-tert-butylhydrazine-1,2-dicarboxylate, intermediate 10 was
synthesized via electrophilic substitution, deprotection and acylation. By a similar
procedure in Scheme 1, intermediate 10 reacted with disubstituted benzaldehyde in
the presence of piperidine and absolute alcohol to afford target compound 13. Using a
similar protocol, imidazolidine-2,4-dione derivative 17 was synthesized (Scheme 4).
Scheme 3
Reagents and conditions: (a) NaH, DMF, 0 , 4 h; (b) TFA, DCM, rt, 1 h; (c)
malonyl dichloride, triethylamine, DCM, 0 , 0.5 h; (d) piperidine, DCM, rt,
overnight; (e) LiOH·H2O, THF: MeOH: H2O = 3: 2: 1, rt, 0.5 h.
Scheme 4
Reagents and conditions: (a) NH4OAc, AcOH, 115 , reflux, 6 h; (b) NaH, DMF,
0 , 4 h; (c) LiOH·H2O, THF: MeOH: H2O = 3: 2: 1, rt, 0.5 h.
Indolin-2-one derivatives 24a-24h were synthesized according to the procedures
outlined in Scheme 5. Substituted phenylboronic acids were reacted with

6-bromoindolin-2-one to give compound 20. Following similar synthetic steps,
compound 23 reacted with disubstituted benzaldehyde to afford target compounds
24a-24h.
Scheme 5
CHO
O
OH
H
N
OEt
B
EtO
O
O
OH
O
H
O
O
Br
N
19
H
11
N
O
a
b
O
O
O
EtO
OEt
O
20
18
21
R
R
N
N
24a R = Me
24b
24c R = Cl
O
Br
O
R
R = F
d
22
24d R = OMe
c
24e
R = SO2Me
O
24f R = CF3
24g R = Ac
24h
O
O
O
O
O
HO
EtO
R = CN
OH
OEt
O
O
24a - 24h
23
Reagents and conditions: (a) Na2CO3, PdP(ph3)4, PhMe, 100 , N2, reflux, 5 h; (b)
piperidine, absolute alcohol, 78 , reflux, 3 h; (c) NaH, DMF, 0 , 4 h; (d) LiOH·H2O,
THF: MeOH: H2O= 3: 2: 1, rt, 1 h.
2.6 Biological Evaluation and SAR of Inhibitor 24a
As a promising lead compound for SAP2, a series of novel derivatives of
inhibitor 24a were designed and synthesized (Scheme 5). Their SAP2 inhibitory
activities are summarized in Table 3 and S2 in Supporting Information. Various
substituents on benzyl benzene ring were investigated. The SAP2 inhibitory activity
was retained for most of the target compounds. Among them, methyl, chloro,
methylsulfonyl, trifluoromethyl, and cyano substituents resulted in better activity than
fluoride, methoxyl and carboxyl substituents, which inhibited SAP2 100% at
concentration of 100 μM. The in vitro antifungal activities of selected target
compounds (24a, 24c, 24e, 24f) were measured. The results showed that indolone
SAP2 inhibitors could not inhibit the growth of C. albicans and had no in vitro

synergistic effect with FLC (Tables S3 and S4 in Supporting Information).
Table 3. SAP2 inhibitory activity of the indolone inhibitors
Compd
24a
R
4-Me
4-F
Inhibition (100 μM)
100%
58%
24b
4-Cl
100%
29%
24c
4-OMe
4-SO2Me
4-CF3
4-Ac
4-CN
-
24d
100%
100%
90%
24e
24f
24g
100%
100%
24h
PepA
2.7 Compound 24a Inhibited the Formation of Azole-resistant C. albicans and C.
tropicalis Biofilms
Due to formation of biofilms, current antifungal treatments are accompanied by
therapeutic failure and drug-tolerance [17]. Fungi biofilms are consist of dense layers
protecting fungi from antifungal drugs [15, 25, 26]. SAPs are known as virulence
factors associated with biofilm and hyphal formation. However, whether SAP2
inhibitors block biofilm and hyphal formation is still unknown. Therefore, selected
SAP2 inhibitors were assayed for inhibitory activity against biofilm formation against
clinical FLC-resistant C. albicans and C. tropicalis isolates using FLC as the control

group. The results revealed that compounds 24a, 24c, 24e, 24f and 24h
dose-dependently inhibited biofilm formation. Among them, inhibitors 24c and 24h
inhibited C. albicans biofilm formation larger than 50% at high concentration (64
μg/mL), while compounds 24a, 24c and 24f achieved more than 50% biofilm
formation in C. tropicalis cells. In contrast, the inhibition of biofilm formation was
lower than 50% when treated with FLC at 64 μg/mL (Fig. 6A, 6B). Furthermore, we
investigated the synergistic effect compound 24a in combination with and FLC at
different concentrations. The result showed that biofilm formation was effectively
inhibited at low concentration (0.5 μg/mL, P < 0.05) and nearly 90% inhibition could
be achieved at 32 μg/mL (Fig. 6C). Thus, SAP2 inhibitor could serve as a potential
sensitizer of FLC, providing a new strategy to treat biofilm-related resistant
infections.

Fig. 6 Inhibition of biofilm formation by SAP2 inhibitors. (A) The biofilm formation
of azole-resistant C. albicans 0304103 on different concentrations of compound 24a,

24c, 24e, 24f, 24h and FLC. (B) The biofilm formation of azole-resistant C. tropicalis
10186 on different concentrations of compound 24a, 24c, 24e, 24f, 24h and FLC. (C)
The biofilm formation of azole-resistant C. albicans 0304103 treated with both
compound 24a and FLC. The results were represented as the mean ± SD by three
independent experiments. Statistical significance between groups was determined by
Student’s t test. * P < 0.05, ** P < 0.01, *** P < 0.001, **** P < 0.0001.
2.8 Compound 24a Inhibited Filamentation of Azole-sensitive and Azole-resistant C.
albicans and C. tropicalis
Hypha contributes to adhesion and invasion into hosts and is essential for biofilm
formation [27]. On the basis of the biofilm inhibitory activity, compound 24a was
selected to evaluate the effects on yeast-to-hypha morphological transition of
azole-sensitive and azole-resistant C. albicans and C. tropicalis. As shown in Fig. 7,
azole-sensitive and azole-resistant C. albicans formed long hyphae, while C.
tropicalis formed short hypha. Compound 24a inhibited hyphal formation almost
completely at high concentration of 128 μg/mL. At lower concentrations (8-32
μg/mL), compound 24a could not inhibit hyphal formation obviously against both the
azole-sensitive and azole-resistant C. albicans (Fig. S3 in Supporting Information). In
contrast, compound 24a showed better hyphae inhibitory activity against
azole-resistant C. tropicalis. At the concentrations in the range of 4-16 μg/mL,
compound 24a effectively inhibited hyphal formation of azole-resistant C. tropicalis.
At the high concentration of 64 μg/mL, hyphal formation was completely inhibited
(Fig. 7 and S3 in Supporting Information). The results suggested that SAP2 inhibitor
24a had potent inhibition effect against morphological transition of C. tropicalis.

Fig. 7 Inhibition of filamentation by compound 24a. DIC microscope photographs
on filamentation of azole-resistant C. albicans 0304103, azole-sensitive C. albicans
(C. alb.) SC5314, azole-resistant C. tropicalis (C. tro.) 10186.
2.9 Effects of Indolone Compounds on the Expression of SAP2 and Biofilm Formation
Related Gene in C. albicans cells
Given the biofilm formation inhibitory effects of compound 24a, the expression
levels of SAP2 and biofilm formation related genes (ECE1, ALS3, EFG1, TEC1 and
CPH1) were determined by qRT-PCR. The expression of SAP2 was down-regulated
obviously for compounds 24a, 24c, 24f, 24g and 24h, which was consistent well with
their SAP2 inhibitory activity (Fig. 8A). Extent of cell elongation protein 1 (ECE1) is
a cytolytic peptide toxin and damages host epithelial membranes. Agglutinin-like
protein 3 (ALS3) acts on the adhesion of yeast-to-host tissue, which is important in
biofilm formation. The expressions of ECE2 and ALS3 were down-regulated,
suggesting that compound 24a might act on the polarization and adhesion C.albicans
cells. Enhanced filamentous growth protein 1 (EFG1), transcription activator TEC1

and transcription factor CPH1 are regarded as the critical proteins in signal
transduction pathways of hypha growth. EFG1 is involved in cAMP-dependent
pathway and is perceived to be a major regulator of hyphae, while CPH1 depends on a
MAPK signaling pathway [28]. After the treatment of compound 24a, EFG1 was
down-regulated and CPH1 and TEC1 were up-regulated, indicating that compound
24a might inhibit hypha growth through the cAMP-dependent pathway. The results
could also explain the fact that compound 24a only inhibited hypha formation on high
concentrations (Fig. 8B).
Fig. 8 (A) The effects of compounds 24a, 24c, 24f, 24g and 24h on the expression
levels of SAP2 gene. (B) The effects of compounds 24a on the expression of fungal
filamentation and biofilm formation related genes. Real-time RT-PCR experiment was
used to investigate gene expression changes in C. albicans 0304103 strain. The
concentration of compounds was 64 μg/mL.
2.10 In vivo Antifungal Activity
To investigate the in vivo antifungal effect of SAP2 inhibitors, an invasive C.
albicans infectious murine model was constructed by injecting C. albicans 0304103
cells into the tail veins. The blank control group was treated with saline, the positive
control group was treated with FLC at 10 mg/kg, and the treatment group was
administrated with 24a at 10 mg/kg. Each mouse was injected intraperitoneally once
daily and the administration lasted for four days. Then Kidney fungal burdens were
determined. The results showed that compound 24a significantly reduced the fungal
burden in the kidney (P < 0.001), which was comparable to FLC (P < 0.0001, Fig. 9).

Thus, SAP2 inhibitor 24a offered a new lead compound for the potential therapy of
invasive candidiasis.
Fig. 9 In vivo activites of compound 24a and FLC in the C. albicans 0304103
infectious murine model. The fungal burden of kidney (in log10 CFU/mL of kidney)
were plotted as mean ± SD. Statistical significance between groups was determined
by Student’s t test. **** P < 0.0001, *** P < 0.001, ns represents no significant
difference.
3. Conclusion
In summary, a series of novel small-molecule SAP2 inhibitors were identified by
structure-based virtual screening and rational inhibitor design. After step-by-step lead
optimizations, indolinone derivative 24a was successfully identified with potent SAP2
inhibitory activity. Further biological evaluations indicated that compound 24a
effectively blocked fungi biofilm and hypha formation by down-regulating the
expression of genes SAP2, ECE1, ALS3 and EFG1. In a murine model of invasive
candidiasis, compound 24a showed potent in vivo antifungal potency and significantly
reduced kidney fungal burdens. Targeting virulence factor is a new strategy for the
development of new generation of antifungal agents. The small-molecule SAP2

inhibitors reported herein represent promising antifungal lead compounds. They are
also valuable tools to explore the druggability and biological functions of SAP2.
Further optimization of SAP2 inhibitors is ongoing.
4. Experimental Section
4.1 Chemistry
The reagents and solvents used in the experiments were commercially available.
Column chromatography was performed on 200-400 mesh silica gel. 13C-NMR and
1H-NMR spectra were recorded on Bruker AVANCE600 or AVANCE300
spectrometer (Bruker Company, Germany) with CDCl3 or DMSO-d6 as the solvents
and TMS as the internal. Chemical shift was expressed in ppm (δ). The coupling
constant is reported in hertz (Hz). The melting point was measured by a microscope
melting point apparatus using an XT4A type temperature control system. The
chemical purities were analyzed by high performance liquid chromatography (HPLC)
using a flow of 0.4 mL/min on a C18 column with CH3CN/H2O as the mobile phase.
All target compounds exhibited purities greater than 95%.
4.1.1. (Z)-2,2'-((4-((5-oxo-1,3-diphenyl-1,5-dihydro-4H-pyrazol-4-ylidene) methyl)
-1,3-phenylene)bis(oxy))diacetic acid (5m)
To a solution of intermediate 4 (0.20 g, 0.47 mmol, 1 equiv) in mixed solvent
(THF: MeOH: H2O = 3: 2: 1), LiOH·H2O (0.03 g, 0.71 mmol, 1.5 equiv) was added.
The mixture was reacted at room temperature for 0.5 h, then the solvent was removed
under reduced pressure. Water (20 mL) and 1M HCl were added to adjust the PH to
2.0 to produce precipitation. Then, the precipitation was filtered, washed with water
and recrystallized from ethanol to give compound 5m (0.14 g, yield: 64%) as a orange
1
solid. m.p.: 230-232 ºC. H-NMR (300 MHz, DMSO-d6) δ: 13.18 (brs, 2H), 9.31 (s,
1H), 8.31 (s, 1H), 7.98 (d, 2H, J = 7.75 Hz), 7.17-7.79 (m, 8H), 6.62-6.81 (m, 2H),
13
4.86 (s, 2H), 4.83(s, 2H). C-NMR (75 MHz, DMSO-d6) δ: 169.91, 165.13, 162.50,

161.11, 153.39, 143.82, 138.81, 135.79, 132.74, 131.13, 130.18, 129.44, 129.32,
129.22, 128.84, 128.49, 128.24, 125.36, 121.93, 119.37, 118.83, 116.11, 107.65,
100.27, 66.06, 65.34. ESI-MS (m/z): 473.46 [M+1].
The preparation methods of compounds 5a-5l, 5n-5y and 24a-24h are the same
as compound 5m. Compounds 6a-6e were prepared in the same way as compound 4.
4.1.2. (Z)-3-oxo-3-(4-((5-oxo-1,3-diphenyl-1,5-dihydro-4H-pyrazol-4-ylidene)methyl)
phenyl)propanoic acid (5g).
1
Orange solid, yield: 72%, m.p.: 125-127 ºC. H-NMR (300 MHz, DMSO-d6) δ:
13.25 (br s, 1H), 8.57-8.70 (m, 2H), 7.98 (d, 2H, J = 9.17 Hz), 7.43-7.77 (m, 6H),
7.06-7.41 (m, 5H), 4.83 (s, 2H). 13C-NMR (75 MHz, DMSO-d6) δ: 191.78, 170.07,
162.91, 162.36, 158.59, 153.39, 150.64, 138.71, 137.47, 132.15, 131.04, 130.26,
129.46, 129.38, 128.81, 128.72, 128.49, 128.29, 126.70, 125.46, 123.03, 119.43,
115.21, 65.31. ESI-MS (m/z): 397.00 [M-1].
4.1.3.
(Z)-3,3'-(4-((1-(3-chlorophenyl)-5-oxo-3-phenyl-1,5-dihydro-4H-pyrazol-4-
ylidene)methyl)-1,3-phenylene)bis(3-oxopropanoic acid) (5n)
1
Orange solid, yield: 73%, m.p.: 228-229 ºC. H-NMR (300 MHz, DMSO-d6) δ:
13.20 (br s, 2H), 9.27 (s, 1H), 8.33 (s, 1H), 7.89-8.17 (m, 2H), 7.43-7.80 (m, 6H),
7.29 (d, 1H, J = 7.99 Hz), 6.63-6.86 (m, 2H), 4.86 (s, 2H), 4.82 (s, 2H). 13C-NMR (75
MHz, DMSO-d6) δ: 169.88, 165.35, 162.68, 161.26, 153.96, 144.37, 140.02, 135.86,
133.76, 131.13, 130.91, 130.32, 129.46, 129.27, 128.89, 128.28, 127.92, 124.87,
121.51, 118.39, 117.30, 116.07, 107.77, 100.26, 66.07, 65.36. ESI-MS (m/z): 505.11
[M-1].
4.1.4.
(Z)-3,3'-(4-((1-(4-chlorophenyl)-5-oxo-3-phenyl-1,5-dihydro-4H-pyrazol-4-
ylidene) methyl)-1,3-phenylene) bis (3-oxopropanoic acid) (5o).
1
Orange solid, yield: 76%, m.p.: 241-243 ºC. H-NMR (300 MHz, DMSO-d6) δ:
13.17 (brs, 2H), 9.29 (s, 1H), 8.32 (s, 1H), 8.03 (d, 2H, J = 9.00 Hz), 7.41-7.82 (m,
7H), 6.63-6.80 (m, 2H), 4.86 (s, 2H), 4.82 (s, 2H). 13C-NMR (75 MHz, DMSO-d6) δ:
169.89, 165.28, 162.52, 161.22, 153.75, 144.19, 137.64, 135.81, 130.98, 130.65,

130.27, 129.45, 129.28, 129.24, 129.07, 128.85, 128.27, 127.19, 121.62, 120.66,
116.08, 107.74, 100.26, 66.08, 65.36. ESI-MS (m/z): 505.00 [M-1].
4.1.5. (Z)-3,3'-(4-((5-oxo-1,3-diphenyl-1,5-dihydro-4H-pyrazol-4-ylidene) methyl)
-1,2-phenylene) bis (3-oxopropanoic acid) (5t).
1
Orange solid, yield:85%, m.p.: 252-253 ºC. H-NMR (300 MHz, DMSO-d6) δ:
13.11 (brs, 2H), 8.65 (s, 1H), 7.91-8.14 (m, 3H), 7.40-7.80 (m, 8H), 7.00-7.29 (m, 2H),
13
4.89 (s, 2H), 4.79 (s, 2H). C-NMR (75 MHz, DMSO-d6) δ: 170.19, 170.01, 162.41,
153.41, 152.79, 150.87, 147.06, 138.70, 131.26, 131.04, 130.25, 129.45, 129.41,
129.37, 129.06, 128.54, 126.88, 125.50, 123.18, 119.53 (2C), 118.95, 118.48, 113.25,
65.38, 65.34. ESI-MS (m/z): 573.31 [M+1].
4.1.6. (Z)-4-(4-hydroxybenzylidene)-2,5-diphenyl-2,4-dihydro-3H-pyrazol-3-one (6a).
1
Orange solid, yield: 65%, m.p.: 242-243 ºC. H-NMR (300 MHz, DMSO-d6) δ:
10.94 (s, 1), 8.53-8.69 (m, 2), 8.00 (d, 2H, J = 7.92 Hz), 7.39-7.80 (m, 8H), 7.16-7.32
(m, 1H), 6.94 (d, 2H, J = 8.16 Hz). 13C-NMR (75 MHz, DMSO-d6) δ: 191.40, 163.94,
163.77, 162.50, 153.48, 151.08, 138.83, 138.28, 132.55, 131.21, 130.16, 129.42,
129.38, 129.32, 129.02, 128.90, 128.52, 125.33, 125.18, 121.66, 119.36, 116.30.
ESI-MS (m/z): 341.45 [M+1].
4.1.7. (E)-2,2'-((4-((1-(4-methylbenzyl)-2-oxo-6-(p-tolyl) indolin-3-ylidene) methyl)-
1,3-phenylene) bis (oxy)) diacetic acid (24a).
1
Orange solid, yield: 79.5%. H-NMR (600 MHz, DMSO-d6) δ: 13.11 (s, 2H),
7.84 (s, 1H), 7.75 (d, J = 8.1 Hz, 1H), 7.65 (d, J = 7.8 Hz, 1H), 7.51 (d, J = 7.4 Hz,
2H), 7.24 (t, J = 7.7 Hz, 4H), 7.21 (s, 1H), 7.19 (d, J = 7.3 Hz, 1H), 7.12 (d, J = 7.5
Hz, 2H), 6.67 (d, J = 8.4 Hz, 2H), 5.02 (s, 2H), 4.86 (s, 2H), 4.77 (s, 2H), 2.31 (s, 3H),
2.22 (s, 3H). ESI-MS (m/s): 564.40 [M+1], 562.28 [M-1]. 13C-NMR (150 MHz,
DMSO-d6) δ: 129.49, 129.17, 127.26, 126.38, 42.32, 40.04, 39.84, 39.63 (d, J = 21.0
Hz), 39.42, 39.31, 39.21, 20.61. HRMS m/z calcd for C34H25N2O7 [M-H]- 562.1871,
found 562.18923. HPLC purity 95.62 %. Retention time: 5.649 min, eluted with

CH3CN: H2O = 4: 1.
4.1.8. (E)-2,2'-((4-((1-(4-chlorobenzyl)-2-oxo-6-(p-tolyl) indolin-3-ylidene) methyl)
-1,3-phenylene) bis(oxy)) diacetic acid (24c).
1
Orange solid, yield: 62.7%. H-NMR (600 MHz, DMSO-d6) δ: 13.01 (s, 2H),
7.88 (s, 1H), 7.79 (d, J = 8.3 Hz, 1H), 7.70 (d, J = 8.1 Hz, 1H), 7.55 (d, J = 8.2 Hz,
2H), 7.41 (d, J = 8.4 Hz, 4H), 7.30-7.25 (m, 3H), 7.25-7.22 (m, 1H), 6.71 (d, J = 9.7
13
Hz, 2H), 5.10 (s, 2H), 4.88 (s, 2H), 4.80 (s, 2H), 2.34 (s, 3H). C-NMR (150 MHz,
DMSO-d6) δ: 169.75, 169.71, 161.12, 158.10, 143.14, 141.23, 137.29, 136.71, 135.88,
132.53, 131.92, 130.80, 129.48, 129.16, 128.62, 126.42, 123.86, 122.35, 119.89,
119.72, 116.18, 107.04, 106.42, 100.24, 64.99, 64.65, 41.92, 20.64.
4.1.9. (E)-2,2'-((4-((2-oxo-6-(p-tolyl)-1-(4-(trifluoromethyl) benzyl) indolin-3-ylidene)
methyl)-1,3-phenylene) bis(oxy)) diacetic acid (24f).
1
Orange solid, yield: 95.7%. H NMR (600 MHz, DMSO-d6) δ 13.05 (s, 2H),
7.87 (s, 1H), 7.79 (d, J = 8.4 Hz, 1H), 7.71 (m, J = 13.4, 8.2 Hz, 3H), 7.59 (d, J = 8.0
Hz, 2H), 7.54 (d, J = 8.2 Hz, 2H), 7.29 (s, 1H), 7.27 – 7.23 (m, 3H), 6.72 – 6.68 (m,
2H), 5.20 (s, 2H), 4.88 (s, 2H), 4.79 (s, 2H), 2.33 (s, 3H). 13C-NMR (150 MHz,
DMSO-d6) δ: 169.73, 169.69, 167.98, 161.13, 158.10, 143.13, 141.72, 141.29, 137.31,
136.67, 132.60, 130.81, 129.47, 127.94, 126.43, 125.55, 123.81, 122.39, 119.97,
119.73, 116.18, 106.98, 106.45, 100.25, 65.10, 64.82, 42.23, 20.63.
4.1.10. (E)-2,2'-((4-((1-(4-acetylbenzyl)-2-oxo-6-(p-tolyl) indolin-3-ylidene) methyl)
-1,3-phenylene) bis(oxy))diacetic acid (24g).
1
Orange solid, yield: 63.6%. H-NMR (600 MHz, DMSO-d6) δ :13.02 (s, 2H),
7.93 (d, J = 8.3 Hz, 2H), 7.88 (s, 1H), 7.79 (d, J = 8.4 Hz, 1H), 7.69 (d, J = 8.0 Hz,
1H), 7.53 (d, J = 8.1 Hz, 2H), 7.49 (d, J = 7.4 Hz, 2H), 7.27-7.22 (m, 4H), 6.69 (s,
1H), 6.64 (s, 1H), 5.17 (d, J = 10.2 Hz, 2H), 4.87 (s, 2H), 4.79 (s, 2H), 2.53 (d, J = 6.0
Hz, 3H), 2.32 (d, J = 4.8 Hz, 3H). 13C-NMR (150 MHz, DMSO-d6) δ :197.37, 169.76,
169.72, 167.95, 161.13, 158.11, 143.17, 142.16, 141.23, 137.29, 136.70, 135.98,
132.58, 130.81, 129.47, 128.63, 127.35, 126.41, 123.84, 122.36, 119.91, 119.82,

116.18, 107.07, 106.44, 100.24, 65.13, 64.84, 42.39, 26.61, 20.62.
4.1.11. (E)-2,2'-((4-((1-(4-cyanobenzyl)-2-oxo-6-(p-tolyl) indolin-3-ylidene) methyl)
-1,3-phenylene)bis(oxy))diacetic acid (24h).
1
Orange solid, yield: 76.5%. H NMR (600 MHz, DMSO-d6) δ 13.19 (s, 2H),
7.87 (s, 1H), 7.84 – 7.77 (m, 3H), 7.70 (d, J = 8.0 Hz, 1H), 7.58 – 7.51 (m, 4H), 7.29
– 7.22 (m, 4H), 6.70 (d, J = 10.4 Hz, 2H), 5.19 (s, 2H), 4.87 (s, 2H), 4.79 (s, 2H), 2.32
13
(s, 3H). C-NMR (150 MHz, DMSO-d6) δ: 170.28, 170.25, 168.48, 161.64, 158.60,
143.52, 143.14, 141.77, 137.82, 137.11, 133.10, 131.32, 129.96, 128.54, 126.92,
124.19, 122.89, 120.50, 120.20, 119.15, 116.61, 110.62, 107.44, 106.88, 100.69,
65.90, 65.43, 60.21, 42.82, 21.12. HRMS m/z calcd for C34H25N2O7 [M-H]- 573.1667,
found 573.16796.
4.2 Biological activity
4.2.1 Strains, Culture and Reagents
Strains were routinely incubated in YPD (1% yeast extract, 2% peptone and 2%
dextrose) at 30°C in a shaking incubator. C. albicans SC5314, C. albicans 0304103,
C. tropicalis 10186 strains were provided by Changzheng Hospital of Shanghai,
China. Stock solutions of all compounds were prepared in DMSO at 2 mg/mL.
4.2.2 Biofilm Formation Assay
Suspension of C. alb.0304103 (1.0 × 105 CFU/mL) or C. tro 10186 cells (1.0 ×
105 CFU/mL) was added to a 96-well plate and incubated at 37 for 90 minutes.
Then rinse with PBS medium three times. Subquently, compounds were added at
different concentrations with FLC as the positive control. Then the fungi were
cultured at 37 for 24h. Finally, a semi-quantitative determination of biofilm
formation was performed using XTT reduction method.
4.2.3 Hyphal Formation Assay
C. alb. 0304103, C.alb.SC5314 and C. tro.10186 cells suspension (1.0 × 105

CFU/mL in 1.5 mL) were added to a 12-well plate (Corning, USA) and treated by
different concentrations of compound 24a with FLC as the positive drug. Then,
microscopic observation and photographing were conducted to observe and record the
growth of hyphae in each group after incubation at 37 ° C for 3h (C.alb. 0304103,
C.alb.5314) or 7 h (C. tro. 10186).
4.2.4 In vivo Antifungal Potency
Female ICR mice (20 ~ 25 g in weight) were used as experimental animals. The
immunodeficient mouse model was established by intraperitoneal injection with 0.2
mL cyclophosphamide (100 mg/kg, saline). After 24 hours, 0.2 mL (1 × 105 CFU/mL)
of C. albicans 0304103 was injected by intravenous. Mice were divided into three
groups with 4 mice in each group. FLC (10 mg/kg) and compound 24a (10 mg/kg)
were injected intraperitoneally 24 h after inoculation. After four consecutive days of
administration, mice were sacrificed. Kidney tissue was removed and homogenized.
The total fungal load was calculated by spreading a certain dilution on the YPD agar,
and incubatione at 30 ° C.
4.2.5 SAP2 Inhibitory Activity Assay.
To asses the SAP2 inhibitory activity, FRET assay was performed. SAP2 enzyme
was expressed and purified by Wuxi Biortus Biosciences Co. Ltd. (Wuxi, China). The
FRET pair-labeled substrate (Dabcyl-Arg-Lys-Pro-Ala-Leu-Phe-Phe-Arg-Leu-Glu
(EDANS)-Arg-OH) was synthesized and purchased from GL Biochem (Shanghai) Ltd.
A 50 mM sodium citrate buffer (pH = 4.5) containing 50 mM NaCl was prepared as
assay buffer. The reaction mixture contained 185 µL buffer and 5 µL SAP2 solution (a
linear fluorescence increase of 100 units/min). The text compounds were diluted in
DMSO, and 5 µL of the dilution was added and preincubated with SAP2 for 30 min
before the addition of substrate. The final concentrations of text compounds were
between 100 µM and 0.01 µM (10 µM ~ 0.001 nM for pepstatin A). Finally, 5 µL
substrate solution (18.75 µM in DMSO) was added and the mixture was incubated at
30 °C for 30 min in a final volume of 200 μL. Fluorescence was determined by