Novel non-peptidic small molecule inhibitors of secreted aspartic protease 2 (SAP2) for the treatment of resistant fungal infections

Article abstract of DOI:10.1039/c8cc07810f

Targeting secreted aspartic protease 2 (SAP2), a kind of virulence factor, represents a new strategy for antifungal drug discovery. In this report, the first-generation of small molecule SAP2 inhibitors was rationally designed and optimized using a structure-based approach. In particular, inhibitor 23h was highly potent and selective and showed good antifungal potency for the treatment of resistant Candida albicans infections.

Full text of DOI:10.1039/c8cc07810f

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Novel non-peptidic small molecule inhibitors of
secreted aspartic protease 2 (SAP2) for the
treatment of resistant fungal infections†
Cite this: Chem. Commun., 2018,
54, 13535
Received 1st October 2018,
Accepted 8th November 2018
Guoqiang Dong,‡a Yang Liu,‡b Ying Wu,a Jie Tu,a Shuqiang Chen,a Na Liua and
a
Chunquan Sheng
*
DOI: 10.1039/c8cc07810f
rsc.li/chemcomm
Targeting secreted aspartic protease 2 (SAP2), a kind of virulence secretion of SAP was observed in drug-resistant Candida spp.
factor, represents a new strategy for antifungal drug discovery. In isolates than in the susceptible isolates.9 Thus, the SAPs offer
this report, the first-generation of small molecule SAP2 inhibitors potential targets for the discovery of novel antifungal agents.12
was rationally designed and optimized using a structure-based Moreover, targeting SAPs may provide a novel intervention
approach. In particular, inhibitor 23h was highly potent and selective strategy to overcome drug resistance. Among the SAPs, SAP2 is
and showed good antifungal potency for the treatment of resistant one of the most expressed enzymes implicated in host invasion
Candida albicans infections.
and attracted interest in early stages of drug discovery.9
Several natural products showed weak or moderate inhibi-
Invasive fungal infections (IFIs) represent a serious health risk tory activity against the extracts of SAPs.13 For SAP2, only
with high incidence and mortality in immunocompromised peptide (e.g. pepstatin A, 1, Fig. 1A)14 and peptidomimetic15,16
individuals.1 Despite the availability of several classes of anti-
fungal agents (e.g. polyenes, triazoles and echinocandins) for IFI
treatment, the mortality is still on the rise due to limited clinical
efficacy, drug-related toxicity, and severe drug resistance.2 Thus,
there is an urgent need to validate new antifungal drug targets as
well as developing a new generation of safe and effective anti-
fungal agents.3
In recent years, targeting virulence factors has represented a
new therapeutic approach in antimicrobial therapy.4 Most
of the success in this field was focused on the discovery of
new drugs targeting bacterial virulence.5–7 In contrast, effective
small molecules targeting fungal virulence factors are very
rare.8 Secreted aspartic proteases (SAPs) of Candida albicans
are a family of important virulence factors for localized and
systemic infections.9 In Candida albicans, ten distinct SAP
genes (SAP1-10) were expressed in vitro and in vivo.10 The SAPs
are essential for the fungal nutrition process and contribute to
the fungal pathogenicity due to their critical participation in
several stages of the infective process including adhesion,
invasion, tissue damage and so on.9,11 Moreover, a higher
a Department of Medicinal Chemistry, School of Pharmacy, Second Military Medical
University, 325 Guohe Road, Shanghai 200433, People’s Republic of China.
E-mail: shengcq@smmu.edu.cn
b Department of Pharmacy, No. 401 Hospital of Chinese People’s Liberation Army,
Fig. 1 Discovery of small molecule SAP2 inhibitors by virtual screening.
(A) Protocol for virtual screening of small-molecule SAP2 inhibitors;
(B) chemical structure of peptidic SAP2 inhibitors; (C) chemical structures
and inhibitory activity of three novel SAP2 inhibitors.
Qingdao 266071, People’s Republic of China
† Electronic supplementary information (ESI) available: Experimental and meth-
ods. See DOI: 10.1039/c8cc07810f
‡ G. Dong and Y. Liu contributed equally.
This journal is ©The Royal Society of Chemistry 2018
Chem. Commun., 2018, 54, 13535--13538 | 13535

View Article Online
Communication
ChemComm
SAP2 inhibitors were reported. However, it is very difficult to SAP2 by mimicking the interaction of 2 (see Fig. S2 in the ESI†
develop peptidic or peptidomimetic compounds into clinical for schematic representation of the interactions). The analysis
drugs due to significant challenges in achieving suitable of the binding mode of these SAP2 inhibitors revealed that they
pharmacokinetic characteristics. To date, only one report has share three common structural features: (1) a heterocyclic
demonstrated antifungal activity of peptidomimetic SAP2 scaffold located at the central part of the active site; (2) three
inhibitors.15 Thus, it is highly desirable to discover non-peptidic side chains attached on the scaffold to form hydrophobic and
(or non-peptidomimetic) small molecule inhibitors. Such p–p interactions with the active site; (3) polar functional groups
inhibitors are expected to possess both good SAP2 inhibitory in the side chain as hydrogen bond donors or acceptors. Among
activity and good in vivo antifungal potency. The crystal struc- the six pockets in the SAP2 active site, S1, S2, S10 and S20 are
ture of SAP2 in complex with the peptidic inhibitor A70450 important for small molecule binding.
(2, Fig. 1B) has been solved,17 offering the opportunity of
Due to the importance of selectivity in the discovery of
structure-based discovery of small molecule inhibitors. However, antifungal agents, the three SAP2 inhibitors were assayed for
therapeutically targeting SAP-like proteases by a small molecule is inhibitory activity of beta-secretase 1 (BACE1, a kind of human
a highly challenging task.18 The difficulty lies in how to effectively aspartic protease). The results indicated that compounds 3, 4
mimic the interactions of large peptidic inhibitors and achieve a and 5 were inactive at the concentration of 100 mM (inhibition
balance of binding affinity and drug-like properties. Herein, we rate o10%). Furthermore, we tested their in vitro antifungal
report structure-based discovery of the first generation of small activity against C. albicans using the standard NCCLS protocols.22
molecule SAP2 inhibitors by combining docking-based virtual Interestingly, all of them were inactive in the antifungal assay
screening and structure-based drug design.
with minimal inhibitory concentration (MIC) values larger than
Following our established virtual screening protocols 64 mg mLꢀ1. The results were well consistent with the action
(Fig. 1A and Table S1 in the ESI†),19 35 compounds (Fig. S1 mode of virulence inhibitors.
in the ESI†) were selected from about 287 000 compounds in
Inspired by these results, inhibitor 5 was selected for
the Specs 3D database by docking, scoring, visual inspection structure–activity relationship (SAR) and structural optimiza-
and Pan-assay interference compound (PAINS, http://zinc15. tion studies (Fig. 3). First, it was subjected for chemical
docking.org/patterns/home) filtration.20 The evaluation of the similarity search in the SPECS database. Seventeen analogues
C. albicans SAP2 inhibitory activity was carried out via a spectro- were selected and purchased for biological evaluation (B1–B17,
photometric assay.21 Initial screening was performed at the Fig. S3 in the ESI†). Among them, compound 6 showed the best
concentration of 100 mM and 13 out of 35 compounds showed an activity (IC50 = 10.18 mM), which was slightly better than the
inhibitory rate larger than 40% (Fig. 1B and Table S2 in the ESI†). initial hit 5 (Table S3 in the ESI†). Second, various substituents
The IC50 values were further determined for 7 compounds with were introduced onto A-ring and B-ring (compounds 12a–t,
the inhibitory rate larger than 50%. The results indicated that Table S4 in the ESI†) and the importance of the A, B, C-ring
all of them had an IC50 value lower than 100 mM. Among them,
three compounds, namely 3 (IC50 = 77.18 mM), 4 (IC50
=
36.37 mM) and 5 (IC50 = 13.78 mM), showed the best SAP2
inhibitory activity (Fig. 1C).
In order to get the binding mode of these novel SAP2
inhibitors, accurate molecular docking was performed. As
shown in Fig. 2, the small molecule inhibitors bound with
Fig. 2 Proposed binding poses of inhibitors 3 (A), 4 (B) and 5 (C) in
the active site of C. albicans SAP2. The comparison of binding modes of
small-molecule inhibitors (carbon atoms in purple) 3 (D), 4 (E) and 5
(F) with the peptide inhibitor compound 2 (carbon atoms in green) is
shown on the surface of SAP2. The figures are generated using PyMol
(http://www.pymol.org/).
Fig. 3 Structure–activity relationship and optimization process of SAP2
inhibitor 5.
13536 | Chem. Commun., 2018, 54, 13535--13538
This journal is ©The Royal Society of Chemistry 2018

View Article Online
ChemComm
Communication
in the SAP2 inhibitory activity was also investigated. SAP2
inhibitory assay revealed that only the 4-methyl derivative 12b
showed a slightly improved activity (IC50 = 9.14 mM). Removal of
the A-ring or B-ring (Fig. S4 in the ESI†) and replacement of the
A-ring with hydroxyl alkyl groups (compounds 18a–f, Table S5
in the ESI†) resulted in an obvious decrease of the activity. In
contrast, removal of the C-ring led to total loss of the activity
(Fig. S4 in the ESI†), indicating that the C-ring played key roles
in inhibitor binding. Moreover, removal or esterification of the
terminal carboxyl group also resulted in inactive analogues
(Fig. S4 in the ESI†), highlighting the importance of the
carboxyl group. For the position of the carboxyl side chain on
the C-ring phenyl group, meta-substituted derivatives were
Fig. 5 Kaplan–Meier survival curves of C. elegans with candidiasis treated
with compounds 5 (A), 12a (A), 12b (B), 23h (B) and fluconazole (log-
rank test: P o 0.05 for comparison between compounds 12a, 12b, 23h,
fluconazole and the blank control).
more potent than the corresponding para-position analogues model for evaluating the antifungal potency.24 Moreover, this
because of stronger hydrogen bonding interaction with SAP2 method can be used for concurrent evaluation for both anti-
(Fig. 4 and Fig. S5 in the ESI†). Finally, in order to improve the fungal activity and host toxicity.24,25 As shown in Fig. 5A, the
activity of the inhibitors, an additional carboxyl side chain was initial hit 5 was inactive at the concentration of 16 mg mLꢀ1 in
introduced in consideration of forming new hydrogen bonding the nematode assay possibly because of its poor water solubility.
interactions with the S3 pocket. Thus, a series of di-carboxyl Interestingly, compounds 12a, 12b and 23h showed excellent
derivatives were synthesized and assayed (Table S6 in the ESI†). antifungal potency. The survival rate of them (16 mg mLꢀ1
)
Consistent with molecular docking, all the di-carboxyl deriva- was above 60% at the end of the test, which was comparable
tives showed increased SAP2 inhibitory activity. In particular, to that of the fluconazole (FLC) group (32 mg mLꢀ1). The results
compound 23h showed the best activity with an IC50 value of confirmed that the small-molecule SAP2 inhibitors were effective
0.86 mM. The additional carboxyl group of 23h formed an in the treatment of C. albicans infection with low host toxicity.
additional hydrogen bond with Tyr225 (Fig. 4). Moreover, the
2,4-di-carboxyl substitution was proven to be better than the factor inhibitors has been rather limited. Only one report
corresponding 3,4-di-carboxyl substitution. showed that peptidomimetic SAP2 inhibitors were effective in
To date, the investigation for antifungal activity of virulence
Compound 23h was assayed for inhibitory activity against four a model of vaginal candidiasis.15 In antibacterials, virulence
human aspartic proteases (i.e. BACE1, cathepsin B, renin and factor inhibitors showed therapeutic advantages in stand-alone
pepsin).23 It had an IC50 value of larger than 100 mM towards the therapy7 or combination therapy.5 Thus, the in vivo antifungal
four enzymes, confirming its good selectivity. Human umbilical potency of compound 23h was further explored in the models
vein endothelial cells (HUVEC) were used to evaluate the growth of fluconazole-sensitive and fluconazole-resistant C. albicans
inhibition of compound 23h towards normal cells. To our delight, infections. ICR mice (n = 10) infected with C. albicans were
compound 23h showed an IC50 value larger than 100 mM, received i.p. injections with either a drug vehicle or compound
indicating its low host toxicity. Furthermore, the in vitro metabolic 23h (2 mg kgꢀ1 or 5 mg kgꢀ1 body weight) in 24 h intervals for
stability of compound 23h was evaluated using the liver micro- 7 days. The vehicle-treated mice died within 7 d, whereas
some assay. The results indicated that compound 23h had good animals that had received treatment with compound 23h
metabolic stability with the T1/2 value of 227.6 minutes.
displayed increased time to death and 10% to 20% survival
The SAP2 gene does not affect the in vitro growth of C. albicans. rate (vehicle vs. compound 23h, P o 0.05; Fig. 6A and B).
Consistently, all the small-molecule SAP2 inhibitors reported Although fluconazole is used as a first-line antifungal therapy
herein also had no in vitro antifungal activity against C. albicans in clinics, it has suffered from severe drug resistance. In the
(MIC 4 64 mg mLꢀ1). Thus, a series of in vivo studies were model of mice (n = 10) infected with the fluconazole-resistant
performed to evaluate the antifungal potency. Candida-mediated clinical isolate, the fluconazole-treated mice (0.5 mg kgꢀ1) died
Caenorhabditis elegans assay was proven to be a facile in vivo within 14 d. In contrast, animals treated with both fluconazole
(0.5 mg kgꢀ1) and 23h (2 mg kgꢀ1) displayed a significantly
increased survival rate (about 50% at the endpoint, Fig. 6C and D),
indicating that the SAP2 inhibitor had synergistic effects with
fluconazole. Taken together, these data demonstrated for the
first time that the combination therapy of SAP2 inhibitors and
fluconazole is an effective strategy to combat the drug resistance.
In summary, this report describes the discovery of the first
small molecule SAP2 inhibitors as novel antifungal agents. The
results presented herein have several important implications
for discovery and development of SAP2-based anti-virulence
agents. First, our findings provide compelling evidence that
SAP2 is a tractable target for small molecule modulation.
Fig. 4 Proposed binding poses of inhibitor 23h in the active site of
C. albicans SAP2.
This journal is ©The Royal Society of Chemistry 2018
Chem. Commun., 2018, 54, 13535--13538 | 13537

View Article Online
Communication
ChemComm
Notes and references
1 G. D. Brown, D. W. Denning and S. M. Levitz, Science, 2012, 336, 647.
2 F. C. Odds, Rev. Iberoam. Micol., 2005, 22, 229; M. A. Pfaller, D. J. Diekema,
M. G. Rinaldi, R. Barnes, B. Hu, A. V. Veselov, N. Tiraboschi, E. Nagy and
D. L. Gibbs, J. Clin. Microbiol., 2005, 43, 5848.
3 D. W. Denning and M. J. Bromley, Science, 2015, 347, 1414.
4 S. W. Dickey, G. Y. C. Cheung and M. Otto, Nat. Rev. Drug Discovery,
2017, 16, 457.
5 D. T. Hung, E. A. Shakhnovich, E. Pierson and J. J. Mekalanos,
Science, 2005, 310, 670.
6 C. I. Liu, G. Y. Liu, Y. Song, F. Yin, M. E. Hensler, W. Y. Jeng,
V. Nizet, A. H. Wang and E. Oldfield, Science, 2008, 319, 1391.
7 J. Zhang, H. Liu, K. Zhu, S. Gong, S. Dramsi, Y. T. Wang, J. Li,
F. Chen, R. Zhang, L. Zhou, L. Lan, H. Jiang, O. Schneewind, C. Luo
and C. G. Yang, Proc. Natl. Acad. Sci. U. S. A., 2014, 111, 13517.
8 X. Li, Y. Hou, L. Yue, S. Liu, J. Du and S. Sun, Antimicrob. Agents
Chemother., 2015, 59, 5885.
9 N. C. Silva, J. M. Nery and A. L. Dias, Mycoses, 2014, 57, 1.
10 D. H. Navarathna, J. M. Hornby, N. Hoerrmann, A. M. Parkhurst,
G. E. Duhamel and K. W. Nickerson, J. Antimicrob. Chemother., 2005,
56, 1156.
11 A. L. Santos and L. A. Braga-Silva, Mini-Rev. Med. Chem., 2013, 13, 155.
12 C. Sheng and W. Zhang, Curr. Med. Chem., 2011, 18, 733.
13 X. C. Li, M. R. Jacob, D. S. Pasco, H. N. ElSohly, A. C. Nimrod,
L. A. Walker and A. M. Clark, J. Nat. Prod., 2001, 64, 1282; Z. Zhang,
H. N. ElSohly, M. R. Jacob, D. S. Pasco, L. A. Walker and A. M. Clark,
J. Nat. Prod., 2002, 65, 979.
14 K. Stewart and C. Abad-Zapatero, Curr. Med. Chem., 2001, 8, 941.
15 A. Trabocchi, C. Mannino, F. Machetti, F. De Bernardis, S. Arancia,
R. Cauda, A. Cassone and A. Guarna, J. Med. Chem., 2010, 53, 2502.
16 T. Sato, M. Shibazaki, H. Yamaguchi, K. Abe, H. Matsumoto and
M. Shimizu, J. Antibiot., 1994, 47, 588.
17 S. M. Cutfield, E. J. Dodson, B. F. Anderson, P. C. Moody, C. J. Marshall,
P. A. Sullivan and J. F. Cutfield, Structure, 1995, 3, 1261; C. Abad-Zapatero,
R. Goldman, S. W. Muchmore, C. Hutchins, K. Stewart, J. Navaza,
C. D. Payne and T. L. Ray, Protein Sci., 1996, 5, 640.
18 M. Drag and G. S. Salvesen, Nat. Rev. Drug Discovery, 2010, 9, 690.
19 G. Dong, C. Sheng, S. Wang, Z. Miao, J. Yao and W. Zhang, J. Med.
Chem., 2010, 53, 7521.
20 J. Baell and M. A. Walters, Nature, 2014, 513, 481; C. Aldrich, C. Bertozzi,
G. I. Georg, L. Kiessling, C. Lindsley, D. Liotta, K. M. Merz, Jr.,
A. Schepartz and S. Wang, J. Med. Chem., 2017, 60, 2165.
21 C. Buchold, Y. Hemberger, C. Heindl, A. Welker, B. Degel, T. Pfeuffer,
P. Staib, S. Schneider, P. J. Rosenthal, J. Gut, J. Morschhauser,
G. Bringmann and T. Schirmeister, ChemMedChem, 2011, 6, 141.
22 Z. Jiang, N. Liu, D. Hu, G. Dong, Z. Miao, J. Yao, H. He, Y. Jiang,
W. Zhang, Y. Wang and C. Sheng, Chem. Commun., 2015, 51, 14648;
C. Sheng, W. Zhang, H. Ji, M. Zhang, Y. Song, H. Xu, J. Zhu, Z. Miao,
Q. Jiang, J. Yao, Y. Zhou and J. Lu, J. Med. Chem., 2006, 49, 2512.
23 M. S. Malamas, J. Erdei, I. Gunawan, J. Turner, Y. Hu, E. Wagner,
K. Fan, R. Chopra, A. Olland, J. Bard, S. Jacobsen, R. L. Magolda,
M. Pangalos and A. J. Robichaud, J. Med. Chem., 2010, 53, 1146.
24 J. Breger, B. B. Fuchs, G. Aperis, T. I. Moy, F. M. Ausubel and
E. Mylonakis, PLoS Pathog., 2007, 3, e18.
Fig. 6 In vivo antifungal potency of compound 23h. (A) Kaplan–Meier
survival of ICR mice infected with 0.5 ꢁ 107 CFU of C. albicans (ATCC
strain SC5314). ICR mice (n = 10) received vehicle or compound 23h
(2 mg kgꢀ1 body weight) treatment via i.p. injection at 24 h intervals for 7 d.
Statistical significance was examined with the log-rank test (vehicle vs.
compound 23h, P o 0.05). (C) Kaplan–Meier survival of ICR mice (n = 10)
infected with 0.5 ꢁ 107 CFU of C. albicans (fluconazole-resistant clinical
isolate 103). ICR mice were treated with fluconazole (0.5 mg kgꢀ1) and a
combination of fluconazole (0.5 mg kgꢀ1) and compound 23h (2 mg kgꢀ1).
Statistical significance was examined with the log-rank test (fluconazole
group vs. combination group, P o 0.005).
Second, for the first time, our study demonstrates that small
molecule-mediated inhibition of SAP2 can lead to therapeutic
effects for the treatment of IFIs. Third, the combination therapy
of the small molecule SAP2 inhibitor with fluconazole could be
an effective strategy to overcome drug resistance. Taken
together, these findings provide an important starting point
for the design of SAP2 inhibitors as an effective antivirulence
strategy to develop a new generation of antifungal agents.
This work was supported by the National Natural Science
Foundation of China (Grants 81573283 and 81725020 to C. S.)
and the Science and Technology Commission of Shanghai
Municipality (Grant 17XD1404700 to C. S.).
Conflicts of interest
25 R. Pukkila-Worley, A. Y. Peleg, E. Tampakakis and E. Mylonakis,
Eukaryotic Cell, 2009, 8, 1750.
There are no conflicts to declare.
13538 | Chem. Commun., 2018, 54, 13535--13538
This journal is ©The Royal Society of Chemistry 2018