Structure-Based Rational Design of Novel Inhibitors Against Fructose-1,6-Bisphosphate Aldolase from Candida albicans

Article abstract of DOI:10.1021/acs.jcim.6b00763

Class II fructose-1,6-bisphosphate aldolases (FBA-II) are attractive new targets for the discovery of drugs to combat invasive fungal infection, because they are absent in animals and higher plants. Although several FBA-II inhibitors have been reported, none of these inhibitors exhibit antifungal effect so far. In this study, several novel inhibitors of FBA-II from C. albicans (Ca-FBA-II) with potent antifungal effects were rationally designed by jointly using a specific protocols of molecular docking-based virtual screening, accurate binding-conformation evaluation strategy, synthesis and enzymatic assays. The enzymatic assays reveal that the compounds 3c, 3e-g, 3j and 3k exhibit high inhibitory activity against Ca-FBA-II (IC₅₀ < 10 μM), and the most potential inhibitor is 3g, with IC₅₀ value of 2.7 μM. Importantly, the compounds 3f, 3g, and 3l possess not only high inhibitions against Ca-FBA-II, but also moderate antifungal activities against C. glabrata (MIC₈₀ = 4-64 μg/mL). The compounds 3g, 3l, and 3k in combination with fluconazole (8 μg/mL) displayed significantly synergistic antifungal activities (MIC₈₀ < 0.0625 μg/mL) against resistant Candida strains, which are resistant to azoles drugs. The probable binding modes between 3g and the active site of Ca-FBA-II have been proposed by using the DOX (docking, ONIOM, and XO) strategy. To our knowledge, no FBA-II inhibitors with antifungal activities against wild type and resistant strains from Candida were reported previously. The positive results suggest that the strategy adopted in this study are a promising method for the discovery of novel drugs against azole-resistant fungal pathogens in the future.

Full text of DOI:10.1021/acs.jcim.6b00763

Subscriber access provided by UB + Fachbibliothek Chemie | (FU-Bibliothekssystem)
Article
Structure-based Rational Design of Novel Inhibitors against
Fructose-1,6-Bisphosphate Aldolase from Candida albicans
Xinya Han, Xiuyun Zhu, Zongqin Hong, Lin Wei, Yanliang Ren, Fen Wan,
Shuaihua Zhu, Hao Peng, li guo, Li Rao, Lingling Feng, and Jian Wan
J. Chem. Inf. Model., Just Accepted Manuscript • Publication Date (Web): 05 May 2017
Downloaded from http://pubs.acs.org on May 5, 2017
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Chemical Information and Modeling is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 49
Journal of Chemical Information and Modeling
1
2
3
4
5
6
7
8
9
Structureꢀbased Rational Design of Novel Inhibitors
against Fructoseꢀ1,6ꢀBisphosphate Aldolase from
Candida albicans
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Xinya Han,‡,a Xiuyun Zhu,‡,a Zongqin Hong, ‡,a Lin Wei,a Yanliang Ren, a,* Fen Wan,a Shuaihua
Zhu, a Hao Peng,a Li Guo,b Li Rao,a Lingling Feng,a Jian Wan,a,*
a Key Laboratory of Pesticide & Chemical Biology (CCNU), Ministry of Education, Department
of Chemistry, Central China Normal University, Wuhan 430079, China
b Hubei Environmental Monitoring Central Station, Wuhan 430072, Hubei, China
KEYWORDS Class II fructoseꢀ1,6ꢀbisphosphate aldolases (FBAꢀII), CaꢀFBAꢀII inhibitors, C.
albicans, rational drug design, novel antifungal pathogens
ACS Paragon Plus Environment
1

Journal of Chemical Information and Modeling
Page 2 of 49
1
2
3
4
5
6
7
8
9
ABSTRACT Class II fructoseꢀ1,6ꢀbisphosphate aldolases (FBAꢀII) are attractive new targets for
the discovery of drugs to combat invasive fungal infection, because they are absent in animals
and higher plants. Although several FBAꢀII inhibitors have been reported, none of these
inhibitors exhibit antifungal effect so far. In this study, several novel inhibitors of FBAꢀII from
C. albicans (CaꢀFBAꢀII) with potent antifungal effects were rationally designed by jointly using
a specific protocols of molecular dockingꢀbased virtual screening, accurate bindingꢀconformation
evaluation strategy, synthesis and enzymatic assays. The enzymatic assays reveal that the
compounds 3c, 3e–g, 3j and 3k exhibit high inhibitory activity against CaꢀFBAꢀII (IC50 < 10
ꢁM), and the most potential inhibitor is 3g, with IC50 value of 2.7 ꢁM. Importantly, the
compounds 3f, 3g and 3l possess not only high inhibitions against CaꢀFBAꢀII, but also moderate
antifungal activities against C. glabrata (MIC80= 4–64 ꢀg/mL). The compounds 3g, 3l and 3k in
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
combination with fluconazole (8 ꢀg/mL) displayed significantly synergistic antifungal activities
(MIC80 < 0.0625 ꢀg/mL) against resistant Candida strains, which are resistant to azoles drugs.
The probable binding modes between 3g and the active site of CaꢀFBAꢀII have been proposed by
using the DOX (Docking, ONIOM and XO) strategy. To our knowledge, no FBAꢀII inhibitors
with antifungal activities against wild type and resistant strains from Candida were reported
previously. The positive results suggest that the strategy adopted in this study are a promising
method for the discovery of novel drugs against azoleꢀresistant fungal pathogens in the future.
ACS Paragon Plus Environment
2

Page 3 of 49
Journal of Chemical Information and Modeling
1
2
3
4
INTRODUCTION
5
6
7
8
Fungal infections are a serious public health problem.1 Candida is the primary systemic fungus
affecting humans. This fungus is usually existed in the oral and gastrointestinal tracts in many
individuals as a commensal pathogen but can also cause severely oral and vaginal infections
when normal fungusꢀhost were disturbed. For severely immunocompromised patients, Candida
can led to the lifeꢀthreatening systemic infections.2ꢀ4
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Limited antifungal drugs can be used against Candida infections. The drugs5ꢀ8 in routine
clinical use included, echinocandins, polyenes, azoles and flucytosine (5ꢀFC). Polyenes (E. g.
amphotericin B) are regard as bind ergosterol of the fungal plasma membrane; The azoles
targeted CYP51 (sterol 14 alphaꢀdemethylase) can inhibited ergosterol biosynthesis; The
echinocandins mainly inhibit glucansynthesis.8 5ꢀFC was first synthesized in 1957. Monotherapy
with 5ꢀFC is limited as for the frequent development of resistance. However, 5ꢀFC can be used in
combination with amphotericin B to treat severe systemic mycoses.5, 9ꢀ11 Azoles are the most
wellꢀknown antifungal drugs worldwide. However, due to the extensive use of these drugs in
humans, the resistances of pathogenic microorganisms to current drug are a significant threat to
public health.12, 13 Azole derivatives targeting CYP51 are heme iron coordinating compounds
with the potential to also interact with human CYPs; via nonspecific drugꢀmetal binding, which
can lead to various azole drug side effects.14 Therefore, there is an urgent need to quickly
discover new selective targets.
Fructoseꢀ1,6ꢀbisphosphate aldolase (FBA, EC4.1.2.13) catalyzes the reversible aldol
condensation of dihydroxyacetonephosphate (DHAP) and glyceraldehyde 3ꢀphosphate (GAP) to
yield fructoseꢀ1,6ꢀbisphosphate (FBP),15ꢀ17 and can be characterized two distinct classes as
reported by Therisod (Figure 1).1 Class I FBA (FBAꢀI), which is existed in higher organisms and
ACS Paragon Plus Environment
3

Journal of Chemical Information and Modeling
Page 4 of 49
1
2
3
4
5
6
7
8
9
some prokaryotes, forms a Schiffꢀbase intermediate between a lysine residue of the active site
and the keto substrate. In contrast, class II FBA (FBAꢀII) used the divalent metal ion to polarize
the keto carbonyl group on the substrate, and to stabilize the enediolate intermediate. Since FBAꢀ
II is existed in pathogenic microbes (E. g. fungus, bacteria and parasites) and absented in animals,
this enzyme is considered a particularly attractive new target.16, 18ꢀ20 The importance of FBAꢀII
has been illustrated by knockout studies, including M. tuberculosis,19 S. galbus,20 B. subtilis,21 P.
aeruginosa22, 23 and S. pneumonia.24 Additionally, Brown et al.25 have examined the effects of
depleting this enzyme in C. albicans to check the validity of FBAꢀII as the target of the
antifungal drug. Their results showed that when FBAꢀII is depleted to below 5% of the wildꢀtype
levels, the growth of C. albicans will be blocked. Therefore, the necessity of FBAꢀII and the
absence of human homologue have made FBAꢀII most potent as a drug target. Additionally, the
sequence conservation between fungal aldolases is significantly high (>80%), which suggests
that developing broadꢀspectrum antifungal drug that can selectively inhibit aldolase in fungus
might be possible.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
To our knowledge, several analogues1, 16, 26, 27 of FBP, DHAP and phosphoꢀglycolohydroxamic
acid (PGH) have been reported to have high inhibitory activities against FBAꢀII, and some of
these inhibitors also highly inhibit the FBAꢀII from C. albicans (CaꢀFBAꢀII). However, no
inhibition against the growth of cultivated pathogens has been observed by using these
compounds. It may be that the phosphate groups of these compounds are very polar, and they are
thus hard to cross biological membranes by simple diffusion. Therefore, the idea of preparing
inhibitors of FBAꢀII as potential new drug has received limited attention. Herein, we report
positive progress in this field. With a structureꢀbased drug design strategy, a series of novel Caꢀ
FBAꢀII inhibitors with potent antifungal effects against wild type and resistant strains from
ACS Paragon Plus Environment
4

Page 5 of 49
Journal of Chemical Information and Modeling
1
2
3
4
5
6
7
8
9
Candida have been successfully designed and synthesized in the present study. The probable
bindingꢀmodes between the representative hit compounds and CaꢀFBAꢀII were analyzed by
jointly using the DOX (Docking, ONIOM and XO) strategy,28 molecular dynamics (MD)
simulations and enzymatic assays.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
RESULT AND DISCUSSION
Building the 3D conformation of CaꢀFBAꢀII. Recent reviews29 have suggested that a
sequence identity that are higher than 30% is indicative of similar 3D structures. Our
ClustalW230 analysis showed that the FBAꢀII from E. coli (EcꢀFBAꢀII)31 shared the highest
identity (56.3%, as shown in Figure S1) with CaꢀFBAꢀII (GenBank Number: 224471817),
permitting a straightforward sequence alignment and guaranteeing the quality of the homology
model. Therefore, using EcꢀFBAꢀII (PDB ID: 1B57) as template, the plausible homology model
of CaꢀFBAꢀII was performed with the SWISSMODEL server.32 To reduce steric clashes, a
staged minimization was performed using SYBYL X1.3.33 The quality of the homology model of
CaꢀFBAꢀII was assessed with PROCHECK,34 ProQ,35 ProSAꢀweb,36, 37 and QMEAN.38 The
assessments revealed that the LGscore in ProQ was 6.606 (scores ≥ 2.5 indicate a very good
model), the QMEAN score was 0.533 (reliable model lies between 0 and 1), and the Zꢀscore
from ProSAꢀWeb was −9.57 (Figure S2). The quality of the 3D model structure of CaꢀFBAꢀII
were also confirmed by Ramachandran plot analyses of PROCHECK34 (Figure S3), the results
indicated that 88.5 % of the residues were distributed in the favored regions, 10.5 % were in the
additionally allowed regions, 0.5 % in the generously allowed regions, while only 0.5 % of the
residues were in the disallowed regions. These assessment results indicated that the backbones of
the modeled 3D structure of CaꢀFBAꢀII were reasonable.
ACS Paragon Plus Environment
5

Journal of Chemical Information and Modeling
Page 6 of 49
1
2
3
4
5
6
7
8
9
Identifying the binding modes of CaꢀFBAꢀII and FBP. Although EcꢀFBAꢀII shares the
highest identity (56.3%) with CaꢀFBAꢀII, no corresponding substrate is complexed in EcꢀFBAꢀII.
However, the active site of CaꢀFBAꢀII is also highly identical to that of the FBAꢀII from
Mycobacterium tuberculosis (MtꢀFBAꢀII). Furthermore, the Xꢀray crystallographic structure of
MtꢀFBAꢀII complexed with its substrate, an acyclic keto form of FBP, has been resolved (PDB:
3ELF).19 To generate a rational active site region, the resultant model structure of CaꢀFBAꢀII and
the crystal structure of MtꢀFBAꢀII were superimposed each other, and the substrate (FBP) was
then merged into the corresponding pocket of CaꢀFBAꢀII. All atoms located within 6.5 Å from
any atom of FBP were selected as part of the active site of CaꢀFBAꢀII, the corresponding
residues were considered to be involved in the active site if one or more of their atoms were
selected.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
As reported in previous studies,19, 31, 39 using a divalent metal ion, FBAꢀII could polarize the
keto carbonyl group of the substrate and stabilize the enediolate intermediate formed during
catalysis. Thus, coordination with a divalent metal ion is a predominant factor in ligands binding,
which has indeed been confirmed by the crystallographic structure of FBAꢀII complexed with its
ligands.16, 19 Therefore, the Zn (II) ion is the most important pharmacophore for design of novel
inhibitors.
Dockingꢀbased virtual screening and identification of the probe molecule. Using the active
cavities of the 3D model structure of CaꢀFBAꢀII generated above, a set of virtual screening
strategies (illustrated in Figure S4) were employed to obtain inhibitors targeting CaꢀFBAꢀII. For
the first step, a 2D ligandꢀbased search was performed to preselect the compounds from the
Specs database based on Lipinski’s rule of five,40 which is usually used as a guideline to identify
drug candidates that are more probably yield cell permeable compounds, and also allow oral
ACS Paragon Plus Environment
6

Page 7 of 49
Journal of Chemical Information and Modeling
1
2
3
4
5
6
7
8
9
administration. After the 2D ligandꢀbased search, about 110, 000 compounds were screened for
the subsequent dockingꢀbased virtual screening. All preselected 2D compounds were
transformed into 3D conformations using the CONCORD module of SYBYL X1.3.33 As
discussed above, the Zn (II) metal ion in the active site is essential to the catalysis of CaꢀFBAꢀII.
Thus, the design of inhibitors centers on a synthetically accessible scaffold that positions a Zn (II)
ion binding group. FlexXꢀPharm enables pharmacophoreꢀtype constraints to guide ligand
docking.41 Thus, we define spatial and pharmacophore constraints in the docking screening
process based on CaꢀFBAꢀII out of Specs database as mentioned before. The spatial constraint
defined by a distance from the Zn (II) ions of 3.0 Å or less was used in the FlexꢀPharm docking.
Under this constraint, only those compounds docked into the cavity of CaꢀFBAꢀII that met this
restrictive condition were immediately selected for further judgment using a scoring function.
After the FlexXꢀPharm procedures were applied to CaꢀFBAꢀII, approximately 30,000
compounds were selected.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
SurflexꢀDock42 usually dock ligands into the binding site of a protein by using an empirical
scoring function and a patented search engine. This program was particularly successful in
eliminating false positive results and retaining a large amounts of active compounds.42, 43
Therefore, SurflexꢀDock experiments were performed on CaꢀFBAꢀII, and approximately 10, 000
compounds were selected. Subsequently, those compounds with similar backbones were
collected into one category. Finally, the hit compounds (~500) were selected out by jointly
considering the docking score and docked conformations of the compounds. The virtual
screening strategy adopted in this study was also documented in our previous studies.44ꢀ46 Only
compounds with a high docking score that interacted with Zn (II) ion in CaꢀFBAꢀII were selected.
ACS Paragon Plus Environment
7

Journal of Chemical Information and Modeling
Page 8 of 49
1
2
3
4
5
6
Finally, thirty representative potential compounds were purchased from Specs Corporation, and
the corresponding inhibitory activities against CaꢀFBAꢀII were examined (Table S1).
7
8
9
As observed in Table S1, compounds 1, FBA10 and FBA26 exhibited high inhibitory ratios (>
70%) against CaꢀFBAꢀII, therefore, the IC50 values of these three compounds were determined.
Compound 1 exhibited a lower IC50 value (17 ꢀM) against CaꢀFBAꢀII than the other two
compounds. Therefore, compound 1 was selected as a probe molecule for further development
and optimization to obtain more potent inhibitors of CaꢀFBAꢀII.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Identifying the binding mode of CaꢀFBAꢀII and compound 1. To discover more potent
inhibitors, accurate insight into the binding mode of the probe molecule 1 was essential. To
obtain the startingꢀpoint bindingꢀconformation of 1 and CaꢀFBAꢀII, a DOX strategy, which has
been documented previously,28 was applied to predict the most reasonable bindingꢀconformation
of CaꢀFBAꢀII and 1. Firstly, compound 1 was docked into CaꢀFBAꢀII using the SurflexꢀDock
module in SybylꢀX1.3 software to generate as much bindingꢀconformation of 1 with CaꢀFBAꢀII
as possible. After the docking procedure, approximately 200 poses with different bindingꢀ
conformations were obtained. Then, poses with similar conformations (RMSD < 0.35 Å) were
aggregated into one category, and we selected one representative pose from each category for
further bindingꢀenergy calculations using threeꢀlayer ONIOM (ONIOM3) method in Gaussian 09
software package.47 In the ONIOM3 calculations, compound 1 was set as the high layer with a
sphere model, which was calculated at the B3LYP/6ꢀ31G(d) theoretical level; the important
residues within 10 Å depth from any given atom of 1 were considered to be the medium layer
(M), which was calculated using a semiꢀempirical method (PM6). The low layer (L) consisted of
the entire enzyme system, and was calculated by using the AMBER force field. To consider the
flexibility of 1 and the active site of CaꢀFBAꢀII, the top 10 poses with the lowest bindingꢀ
ACS Paragon Plus Environment
8

Page 9 of 49
Journal of Chemical Information and Modeling
1
2
energies were selected for further geometric optimization using XO methods.48 Finally, the
bindingꢀconformation with the lowest bindingꢀenergy as optimized with XO was selected, as
illustrated in Figure 2A.
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
As seen from Figure 2A, the hydroxyl group of compound 1 appears to be very important for
the binding of 1 to CaꢀFBAꢀII. The hydroxyl group could coordinate with the Zn (II) ion. In fact,
the hydroxyl group is easily ionized in solution when it is coordinated with the metal. Thus, the
hydroxyl group on the benzene ring of 1 was set in the ionized state in the present investigation.
The hydrogen atom on the hydroxyl group is likely to be transferred to the surrounding Asp109,
and the oxygen atom of the phenolic hydroxyl group could also form two hydrogen bonds with
the COOH moiety of Asp109 and the NH2 moiety of Asn287. Additionally, one of the carbonyl
groups of 1 appears to form a hydrogen bond with the NH moiety of Glu182, and the other
carbonyl group is likely to form a hydrogen bond with Asn36. In the middle region of 1, one
hydrogen bond between Asp289 and the NH moiety of 1 could be observed. The nitro group of 1
is likely to form two hydrogen bonds with Ser268 and Thr290. To further consider the flexibility
of the whole protein (CaꢀFBAꢀII), we took the complex (CaꢀFBAꢀII and 1) optimized by the XO
method48 as the starting point conformation, and a molecular dynamics (MD) simulation was
further performed using the Amber 12 program.47 Plots of the RMSD and simulation time are
illustrated in Figure 3. The black line represents the RMSD of the backbone Ca atoms of the
complex (CaꢀFBAꢀII and 1) between simulated trajectories and initial structure, whereas the pink
line represents the corresponding RMSD of compound 1. In Figure 3, the black line nearly
achieves a dynamic convergence at approximately 16 ns, whereas the pink line rapidly arrives at
a dynamic convergence with an RMSD value of 1 Å. The conformation of the complex (Caꢀ
FBAꢀII and 1) with the lowest kinetic energy was obtained from the simulation trajectory (16ꢀ20
ACS Paragon Plus Environment
9

Journal of Chemical Information and Modeling
Page 10 of 49
1
2
3
4
5
6
7
8
9
ns). The binding mode of CaꢀFBAꢀII to 1 optimized according to the MD simulations in a
protein environment is illustrated in Figure 2B. Compared with Figure 2A, Figure 2B illustrates
that the Oꢀ moiety of 1 remains in coordination with the Zn (II) ion, this suggest that the Oꢀ
moiety of 1 is essential to the binding of 1 to CaꢀFBAꢀII. Although the binding mode optimized
with XO shows no obvious coordination between the carbonyl group and the Zn (II) ion, the MD
result reveals that one of the carbonyl groups of 1 could also coordinate with the Zn (II) ion.
Additionally, the N atom of 1 appears to have a slight coordination with the Zn (II) ion. Our MD
result indicates that the coordination distance between the N atom and the Zn (II) ion is greater
than 3 Å (3.4 Å), the coordination bond is thus not marked in Figure 2B. These results suggest
that the Zn (II) ion is important for the binding of 1. Previously, Pombeiro et al.49, 50 reported the
crystal structure of compound 1 analogs and Cu (II)/Ni (II) ion. Our model binding mode of
compound 1 with a Zn (II) ion is similar to those of the crystal structures of compound 1 analogs
and Cu (II)/Ni (II) ion. Similar to Figure 2A, the hydrogen bonds between Asp289 and the NH
moiety of 1 are retained. Compared with Figure 2A, the hydrogen bond between the nitro group
of 1 and the OH moiety of Ser268 is retained, but the Thr290 is far from the nitro group of 1.
Indeed, the active site of FBAꢀII is highly conserved. According to the crystal structure of Ecꢀ
FBAꢀII,31 the phosphate group of EcꢀFBAꢀII could form a hydrogen bonds with Ser267, Lys325,
Thr289 and Gly227. In our present model, the nitro group of 1 is located in the phosphate subsite
of CaꢀFBAꢀII. The nitro could form one hydrogen bond with Ser268, which corresponds to
Ser267 in EcꢀFBAꢀII. Therefore, the nitro group could perhaps play the role of a phosphate
group in certain cases. After the MD simulation, one of the carbonyl groups of compound 1 was
rotated approximately 180 degrees compared with those groups optimized by XO (Figure 2A).
This result indicates that one of the carbonyl groups of 1 is rotatable and cannot easily form a
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
10

Page 11 of 49
Journal of Chemical Information and Modeling
1
2
3
4
5
6
7
8
9
stable hydrogenꢀbond with surrounding residues. Using the complex conformation of CaꢀFBAꢀII
and 1 optimized by MD (Figure 2B), the corresponding binding free energy was calculated by
XO methods to be ꢀ6.3 kcal/mol (corresponding to Kd = 24 ꢁM). This theoretical result is in the
same order of magnitude as the experimental IC50 value (17 ꢁM).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Rational design, optimization and synthesis of hit compounds. To obtain more active hit
compounds against CaꢀFBAꢀII, a structure optimization and design route of compound 1 was
next performed in terms of the binding mode of 1 to CaꢀFBAꢀII that was described above, as
illustrated in Figure 4. Starting from compound 1, the simple substituent groups, such as 1a ‒ s,
were first evaluated, and the results are summarized in Table 1. The position of the nitro group
on the benzene ring has a substantial impact on the CaꢀFBAꢀII activity. Compound 1 with the
nitro substituent at the R2 position of the benzene ring exhibits high CaꢀFBAꢀII inhibition (96 %),
while compound 1a with a nitro group at the R3 position exhibits no inhibitory activity against
CaꢀFBAꢀII (14 %). Therefore, the nitro group in the R2 position is favorable for the inhibition of
CaꢀFBAꢀII. According to the bindingꢀmode of 1 to CaꢀFBAꢀII (Figure 2B), the compound with
the nitro group in the R2 position is favorable to forming a hydrogen bond with the surrounding
residues (Ser268) in CaꢀFBAꢀII. As R2 and R3 were fixed to hydrogen, the variants of R1 position
(1b‒f) show that the compound (1f) with hydroxyl group in R1 position exhibit higher CaꢀFBAꢀ
II inhibitory activity (IC50= 38 ꢁM) than the other variants. This result further confirms that, in
accordance with the binding modes in Figure 2B, the coordination between the hydroxyl group
and the Zn (II) ion is important to the binding of the hit compounds. As R1 and R3 were fixed to
hydrogen, variants of R2 (1g‒n) were subsequently explored. The compound (1n) with a nitro
group in the R2 position exhibited higher inhibitory activity against CaꢀFBAꢀII (IC50= 20 ꢁM)
than the others R2 variants. This finding further confirms that the nitro group in the R3 position
ACS Paragon Plus Environment
11

Journal of Chemical Information and Modeling
Page 12 of 49
1
2
3
4
5
6
7
8
9
could form a hydrogen bond with Ser268, and the nitro group could possibly play the role of a
phosphate group in certain cases. However, according to the binding mode of compound 1 in
Figure 2B, the hydroxyl group and sulfonic acid group, which are similar to the nitro group,
could indeed form hydrogen bonds with Ser268. Nevertheless, the solvation effect of a hydroxyl
group and sulfonic acid group should be larger than that of a nitro group. For this reason, the
energies of solvation of 1l, 1m and 1n were next calculated at the wb97xd/6ꢀ31G*ꢀsmd theory
level using the Gaussian 09 software package, which are of 17.9 kcal/mol, 23.5 kcal/mol and
14.5 kcal/mol, respectively. The larger solvation effect counteracts the binding of the compound
to CaꢀFBAꢀII. Thus, compound 1n exhibits a higher inhibitory activity against CaꢀFBAꢀII than
1l or 1m. Regarding the structureꢀactivity relationship (SAR) analysis mentioned above, the
hydroxyl group in the R1 position and the nitro group in the R3 position are favorable for Caꢀ
FBAꢀII inhibition by the hit compounds.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Regarding the binding mode in Figure 2B, one of the methyl groups of 1 is surrounded by the
OH moiety of Thr38 and the NH2 moieties of Gln293 and Arg332. Therefore, we hypothesized
that, when the methyl in 1 is replaced by the amino group in 1s, a hydrogen bond between the
amino and surrounding residues could be formed, which might result in an increase in the
corresponding inhibitory activities of the hit compounds. However, as listed in Table 1, the
inhibitory activity of compound 1s (IC50 = 20 ꢁM) is close to that of compound 1 (IC50 =17 ꢁM).
Taken together, one can observe that it is hard to remarkably increase the CaꢀFBAꢀII inhibitory
activities of the hit compounds by varying the substituents groups of compound 1. Additionally,
some compound 1 analogues likely exist in tautomeric form in solvents.49 Therefore, it was
necessary for us to modify the framework of 1. As illustrated in Figure 2B, one of the acetyl
groups in compound 1 is located at the entrance of the active site of CaꢀFBAꢀII, and no
ACS Paragon Plus Environment
12

Page 13 of 49
Journal of Chemical Information and Modeling
1
2
3
4
5
6
7
8
9
remarkable hydrogen bond between this acetyl group and the surrounding residues can be
observed. Therefore, we next questioned whether developing compounds with only one acetyl
group would affect the CaꢀFBAꢀII activity. Compounds 2 and 2a were thus synthesized and
evaluated. Due to the synthesis difficulty, the hydroxyl group could not be introduced into the
benzene ring of compounds 2 and 2a. As listed in Table 2, compound 2 exhibited no inhibitory
activity against CaꢀFBAꢀII. To analyze the reason why 2 could not inhibit CaꢀFBAꢀII, 2 was
docked into CaꢀFBAꢀII by SurflexꢀDock, and the probable bindingꢀmode of 2 shown in
illustrated in Figure 5A, in which the Arg332 in the active site of CaꢀFBAꢀII is nearby one of the
acetyl group of compound 1. Thus, we supposed that if the methyl group in 2 were replaced by
the benzene group in 2a, a π–π interaction between the benzene ring of 2a and Arg332 in Caꢀ
FBAꢀII could be formed. The possible binding mode of 2a and CaꢀFBAꢀII was also analyzed by
SurflexꢀDock software, as illustrated in Figure5B. As expected, compound 2a exhibits a higher
inhibitory activity against CaꢀFBAꢀII (IC50=26 ꢁM) than 2, as listed in Table 2.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
It should be noticed from Figure5B that the only coordination interactions are between the
hydrazine group of 2a and Zn (II), with a coordination distance of 2.8 Å, which is remarkably
shorter than that of 1 (Figure 2B). This finding suggests that the coordination between the
hydrazine group and Zn (II) plays a major role in influencing the CaꢀFBAꢀII inhibition of the hit
compounds. To increase the interaction between the hydrazine of the hit compound and Zn (II),
we attempted to design and synthesize phenylhydrazone derivatives, such as 3 and 3a–l, as listed
in Table 3, together with their corresponding inhibitions rates of enzymatic. For those
compounds with high inhibition rates (> 70%), IC50 values were also determined, and antifungal
assays were conducted. The enzymatic assay results demonstrate that the CaꢀFBAꢀII inhibitory
activities of most of these compounds are higher, with IC50 value of 2.7–20 ꢁM. In particular,
ACS Paragon Plus Environment
13

Journal of Chemical Information and Modeling
Page 14 of 49
1
2
3
4
5
6
7
8
9
compound 3g exhibited the highest inhibition against CaꢀFBAꢀII (IC50 = 2.7 ꢁM). The possible
binding modes of the representative compound 3g and CaꢀFBAꢀII were also analyzed using the
DOX strategy, as illustrated in Figure 6.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
As shown in Figure 6, on the left of compound 3g, the nitro group could form hydrogen bonds
with the surrounding Ser268 and Thr290 residues in CaꢀFBAꢀII, which is similar to the binding
modes of compounds 1, 2 and 2a. As we designed above, the coordination distance (2.2 Å)
between the hydrazone group of 3g and Zn (II) is shorter than that (2.8 Å) of 2a. As previously
documented51, the hydrazone group easily forms a coordination bond with metals, therefore,
compounds with hydrazone groups usually exhibit high biological activity. This is most likely
the major reason why most of the phenylhydrazone derivatives (3, 3a–l) possess highly
inhibitory activity against CaꢀFBAꢀII. On the right side of 3g, the π–π interaction between the
benzene group and Arg332 can be observed. In addition, hydrogen bonds between the nitro
group and Arg332 can also be found. Furthermore, the nitro group of benzene on the right side of
compound 3g (R1 position in Table 3) was replaced by other substituents to evaluate the
corresponding CaꢀFBAꢀII inhibitory activities. As seen in Table 3, when the 4’ position of
benzene was substituted by CH3, CF3 or H, the corresponding compounds exhibited weak
inhibitory activities against CaꢀFBAꢀII, as in 3, 3a and 3b. However, the introduction of benzene,
halogen, CN or NO2 at the 4’ position of benzene led to a high inhibition of CaꢀFBAꢀII by the hit
compounds, such as 3c–g, with IC50 values of 2.7 – 12.6 ꢁM. In particular, compound 3k also
showed a high level of CaꢀFBAꢀII inhibition (IC50 = 6.0 ꢁM). Therefore, one could conclude that
the compounds with conjugated and hydrogenꢀbond acceptor group at the R1 position of
phenylhydrazone derivatives, such as 3c–g and 3k, have potentially higher CaꢀFBAꢀII inhibitory
activities than the other compounds. This suggests that for phenylhydrazone derivatives, the π–π
ACS Paragon Plus Environment
14

Page 15 of 49
Journal of Chemical Information and Modeling
1
2
3
4
5
6
7
8
9
interaction between the conjugate group and the Arg332 is very important for their CaꢀFBAꢀII
inhibitory activities. Notably, compound 3g not only could form π–π interaction between
nitrobenzene and Arg332, but also could form a hydrogen bond between a nitro group and
Arg332, resulting in the highest inhibitory activity against CaꢀFBAꢀII for compound 3g.
To examine the interaction of 3g and CaꢀFBAꢀII, we have mutated the R332, T290 and S62 to
Ala (Table 4), respectively. As listed in Table 4, R332A and T290A substitutions result in the
60ꢀfold (101 ꢁM) and 130ꢀfold (230.4 ꢁM) increase of Ki values compared with wild CaꢀFBAꢀII
(1.7 ꢁM). These experimental results indicate that R332 and T290 are significantly important to
the binding of compound 3g and CaꢀFBAꢀII. In comparison, the Ki value of S62A (31.5 ꢁM)
exhibit ~15ꢀfold increase than that of wild CaꢀFBAꢀII. This suggest that the interaction between
S62 and compound 3g is exist. However, the effect of R332 and T290 on the inhibitory activity
of compound 3g is smaller than S62, the possible explain is that the hydrogenꢀbond distances of
compound 3g and R332 and T290 is 2.2 Å, which is shorter than that (2.5 Å) of compound 3g
and S62, as can be noticed From Figure 6. These experiment are consistent with our theoretical
results. Furthermore, we also mutated the residue D289, which is far from the active site of Caꢀ
FBAꢀII and hasn’t remarkable interactions with compound 3g, to Ala. As expect, compound 3g
exhibit similar D289A inhibitory activities (3.1 ꢁM) to wild CaꢀFBAꢀII (1.7 3.1 ꢁM). Thus, we
further believe that our mutant results are reliable.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
To examine the effect of the substituent group in the R2 position of the third type of hit
compounds on CaꢀFBAꢀII inhibition, compounds 3i and 3j were also synthesized. As listed in
Table 3, the CaꢀFBAꢀII inhibitory rate of compound 3i (61 %) is similar to that of 3 (43 %),
indicating that the methyl group in the R2 position slightly affects the CaꢀFBAꢀII activity.
ACS Paragon Plus Environment
15

Journal of Chemical Information and Modeling
Page 16 of 49
1
2
3
4
5
6
7
8
9
In a comparison, most of the phenylhydrazone derivatives (3, 3a–m) exhibit greater CaꢀFBAꢀII
inhibition than the other hit compounds (1, 1a–s, 2, 2a). As analyzed above, we inferred that the
key reason is the increase in the coordination interaction between the hydrazone group in
phenylhydrazone derivatives and the Zn (II) ions in CaꢀFBAꢀII. To further confirm this
speculation, we selected compounds 3f and 3g as probe molecules and attempted to design and
synthesize benzoylhydrazone derivatives, such as 4 and 4a, as listed in Table 5. We anticipated
that the introduction of a carbonyl group near the hydrazine group, such as in 4 or 4a, might be
unfavorable for the coordination of the hydrazone group and the Zn (II) ions in CaꢀFBAꢀII. As
expected, compounds 4 (21 %) and 4a (59 %) exhibited poor inhibitory activities against Caꢀ
FBAꢀII compared with compounds 3f (100 %) and 3g (100 %). These results confirm to some
extend that the hydrazone group is very important to the CaꢀFBAꢀII inhibitory activities of the
hit compounds.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Antifungal evaluation. To examine the antifungal activities of the hit compounds, those
phenylhydrazone derivatives with high CaꢀFBAꢀII inhibitions rates (3c–h, 3j–l) were further
assayed using a standard procedure for the inhibition of the growth of cultivated C. albicans
(SC5314), C. glabrata (537) and C. tropicalis. The MIC80 value of the control drug (fluconazole
(FCZ)) was 0.5 ꢀg/mL, which is in accordance with previous values.52 Our results showed that
none of our tested compounds exhibited a remarkable inhibitory activity (MIC80 > 64 ꢀg/mL)
against C. albicans and C. tropicalis. Compound 3f exhibited potential inhibitions (MIC80 = 4
ꢀg/mL) of C. glabrata (537). Compounds 3g (MIC80 = 16 ꢀg/mL) and 3l (MIC80 = 64 ꢀg/mL)
displayed moderate inhibition of C. glabrata (537). The dose response curves of compounds 3f
and 3g in the inhibition of the growth of cultivated C. glabrata (537) are illustrated in Figure S5.
To our knowledge, several FBAꢀII inhibitors have been reported previously,16, 18, 26, 27 however,
ACS Paragon Plus Environment
16

Page 17 of 49
Journal of Chemical Information and Modeling
1
2
3
4
5
6
7
8
9
none of these inhibitors have demonstrated inhibitory activities against microorganisms to date.
This is the first report of FBAꢀII inhibitors with in vitro antifungal activities. The discrepancy
between the inhibitions of CaꢀFBAꢀII and the antifungal assays can be partly explained by the
fact that the present hit compounds cannot easily cross biological membranes.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
As is known, the currently available azole antifungal agents can disrupt the cell membrane,
therefore, we hypothesized that the combination of azole drugs and our hit compounds may
facilitate the crossing of biological membranes by the hit compounds, which in turn could
enhance their antifungal activities. Especially, due to the extensive use of azoles drugs in humans,
the resistance of pathogenic microorganisms to current drug is a significant threat to public
health.12, 13 Therefore, combined with FCZ (8 ꢀg/mL), the potencies of our synthesized
phenylhydrazone analogues against the resistant Candida strains (100 and 103), which are
resistant to known azoles drugs (MIC80>1024 ꢀg/mL), were tested. As listed in Table 6, the
MIC80 values of our synthesized phenylhydrazone analogues alone against are greater than 64
ꢀg/mL, and the MIC80 of FCZ alone is greater than 1024 ꢀg/mL. However, compounds 3d, 3f,
3g, 3k and 3l in combination with FCZ (8 ꢀg/mL) resulted in remarkably greater inhibition than
that of either compound alone. Synergistic interactions were observed for these five hit
compounds. The combination index (CI) is regard as the gold standard for defining the
synergism of drugꢀdrug interactions.53ꢀ55 CI values=1 represent an additive effect; CI values < 1
and > 1 indicate synergistic and antagonistic interactions, respectively. Our calculated CI value
(Table 5) of compounds 3c–g, 3k and 3l, in combination with the FCZ, were well below 0.02, at
0.004 to 0.02, pointing toward a pronounced synergistic activity10. Over all, the combination
therapy experiments have provided us the first proof that our novel CaꢀFBAꢀII inhibitors, such as
3d, 3f, 3g, 3k and 3l, can combine with the known azole drugs to treat resistant fungi infections,
ACS Paragon Plus Environment
17

Journal of Chemical Information and Modeling
Page 18 of 49
1
2
3
4
5
6
which are resistant to known azole drugs. These compounds are worth developing as new
antifungal therapies.
7
8
9
CONCLUSIONS
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
FBAꢀII has always been regarded as a potent and particularly attractive new target for the
discovery of drugs to combat invasive bacterial and fungal infections because of its occurrence in
pathogenic microbes (bacteria, alga and fungi) and its absence in animals. Several inhibitors of
FBAꢀII have been reported, but none of these inhibitors have antifungal effects. By jointly using
a specific protocols of dockingꢀbased virtual screening, accurate bindingꢀconformation
evaluation strategy, a series of novel inhibitors of CaꢀFBAꢀII were rationally designed, optimized
and synthesized, and their inhibitory activities were examined by enzymatic and antifungal assay.
Following a further study of SAR analysis, the inhibitors (3c, 3e–g, 3j) and 3k with the high
inhibitory activity against CaꢀFBAꢀII (IC50 < 10 ꢁM) were designed. In particular, the compound
3g was the most potent (IC50 = 2.7 ꢁM). Furthermore, compounds 1 and 3g were selected as
representative molecules for further analysis, and the probable binding modes between these two
hit compounds and the active site of CaꢀFBAꢀII were proposed by using a combination of the
DOX strategy, MD simulations and enzymatic assays. The SAR analysis of the designed
compounds can be consistently explained according to our proposed bindingꢀmodes of 1 and 3g
to CaꢀFBAꢀII. These results revealed that the hydrazine group is a favorable scaffold for Caꢀ
FBAꢀII inhibitors. It should be noticed that the compounds 3f, 3g and 3l possessed not only high
CaꢀFBAꢀII inhibitory activities, but also moderate antifungal activities against C. glabrata
(MIC80 = 4–64 ꢀg/mL). Moreover, compounds 3g, 3l and 3k in combination with FCZ (8 ꢀg/mL)
displayed significantly synergistic antifungal activities (MIC80 < 0.0625 ꢀg/mL) against resistant
Candida strains, that are resistant to known azoles drugs (MIC80>1024 ꢀg/mL). To our
ACS Paragon Plus Environment
18

Page 19 of 49
Journal of Chemical Information and Modeling
1
2
3
4
5
6
7
8
9
knowledge, this is the first report of FBAꢀII inhibitors with antifungal activities against wild type
and resistant strains of Candida. This positive result confirmed that it is practical to use FBAꢀII
as a potential target for the discovery of drugs fighting against the invasive bacterial and fungal
infections. The present enzymatic and antifungal assay results suggest that the rational design of
CaꢀFBAꢀII inhibitors adopted in this study will be a promising road to the discovery of novel
drugs against azoleꢀresistant fungal pathogens of humans in the future.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
EXPERIMENTAL SECTIONS
Sequence alignment and homology modeling. The amino acid sequence of FBAꢀII from C.
albicans (accession number in NCBI: 224471817) was aligned with that of EcꢀFBA (PDB ID:
1B57)31 using the program ClustalW2. Based on the sequence alignment, the structure of Caꢀ
FBAꢀII was built by using the SWISSMODEL server32, employing the crystallographic structure
of 1B57 (GI: 6980738) as the protein template. A dimeric structure of CaꢀFBAꢀII was
constructed via the combination of the two subunit. The Zn (II) ion were copy from the template
structure. The dimeric conformation of CaꢀFBAꢀII were then optimized by SYBYLꢀX1.3 with
AMBER force field.56, 57
Molecular dynamics (MD) simulations of CaꢀFBAꢀII complex with compound 1. To
confirm the binding mode of probe molecule 1, MD simulations were performed by using the
PMEMD module of AMBER 12 package,47 based on the optimized CaꢀFBAꢀII complex with
compound 1. The partial atomic charges of compound 1 were calculated by restricted
electrostatic potential (RESP)58 fitting, which is implemented in the Antechamber module. The
interactions between Zn (II) ion and protein was treated strictly by van der waals and coulomb
forces.59 The protein parameters were used by AMBER ff10 force field, whereas the general
AMBER force field (GAFF) for the parameters of compound 1. The Na (I) ions were used to
ACS Paragon Plus Environment
19

Journal of Chemical Information and Modeling
Page 20 of 49
1
2
3
4
5
6
7
8
9
neutralize the system, and then an octahedral box of TIP3P water molecules, which extended 10
Å from any given atom of CaꢀFBAꢀII, were selected for solvation.60 The MD was simulated with
constant pressure and temperature (1 bar at 300 K) at periodic boundary conditions. The
temperature was maintained at 300 K using the weakꢀcoupling algorithm. The SHAKE
algorithm61 was used to constrain the bonds with hydrogen atoms. the longꢀrange electrostatic
interactions were calculated by Particle Mash Ewald.62 The nonꢀbond cutoff was set as 10 Å.
Equilibration was also monitored by combining the stability of the temperature, energies and
densities of the system, as well as the RMSD values of the backbone atoms.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
XO calculations. The relative stability of multiple possible proteinꢀinhibitor bindingꢀmodes
were obtained by XO calculation. XO method is a hybrid theoretical chemistry tool based on
ONIOM and divide&conquer method designed to provide accurate quantum chemical
description for large systems, which has been documented in previous publications.63, 64 Our
present computation system included the inhibitor molecule and the drug binding pocket
involving the residues within 10 Å from any atom of the inhibitors. The inhibitor and
surrounding residues were treated with ωB97xD/6ꢀ311++G**//6ꢀ31G* level of theory while the
rest of the system was considered using PM6 theory. The high level region was divided to 3
fragments for the XO calculation. In the free energy calculations, the solvation effect was
considered employing SMD method. The entropy was obtained by frequency analysis of the
inhibitor.
PAINS Screening. The hit compounds with high enzymatic inhibition against CaꢀFBAꢀII(IC50
< 15 ꢁM) were screened by the Pan Assay Interference Compounds (PAINS) remover tool
(Sybyl X1.365 and FAFDrug 3.0 server66) to remove the possible PAINS compounds. The results
indicate that all of our active compounds in this study have passed the PAINS filter.
ACS Paragon Plus Environment
20

Page 21 of 49
Journal of Chemical Information and Modeling
1
2
3
4
General Synthetic methods.
5
6
7
8
The hit compounds 1a–s in this study were prepared as previous documents,67 which were
from a diazonium solution and βꢀdiketone, with sodium acetate in acetonitrile at a temperature
below 5 ºC (Scheme S1). The diazonium solution was prepared using a substituted aniline with
NaNO2 and HBF4 in acetonitrile at a temperature below 5 ºC. The NaNO2 was added in portions
because the reaction was vigorous and highly exothermic. The hit compounds 2 and 2a were
prepared from a diazonium solution and ethyl acetoacetate with potassium acetate and 36% HCl
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
º
in an aqueous potassium hydroxide solution at a temperature below 5 C (Scheme S2), as
documented previously.68 The diazonium solution was prepared using a substituted aniline with
º
NaNO2 and HBF4 in acetonitrile at a temperature below 5 C. Compounds 3, 3a–m were
prepared69 by phenylhydrazine and benzaldehyde in absolute ethanol (Scheme S3). According to
the previous report,70 The hit compounds 4 and 4a were prepared by benzoyl hydrazine and
benzaldehyde in absolute ethanol, as illustrated in Scheme S4.
Enzyme inhibition activities and SiteꢀDirected Mutagenesis. The half maximal inhibitory
concentration (IC50) of some hit compounds with high inhibitions have been examined at the Caꢀ
FBAꢀII recombinant protein level by linear regression analysis in Origin 7.5 program package to
evaluate the inhibitory activities of these compounds. The gene sequence of CaꢀFBAꢀII was
obtained from NCBI (XM_717597.1).71 CaꢀFBAꢀII was expressed in E. coli BL21(DE3) cells, it
was purified as previous document.72 Commercial preparations from Sigma of glycerol 3ꢀ
phosphate dehydrogenase (GPDH) from rabbit muscle and triosephosphate isomerase (TIM)
from rabbit muscle were used. The inhibitory activity of hit compounds against CaꢀFBAꢀII were
determined by using an NADH linked enzymatic assay.73, 74 The activity of cleavage reaction
was examined in Trisꢀpotassiumꢀacetate buffer (0.1 M, pH 7.4, with 0.2 M potassiumꢀacetate for
ACS Paragon Plus Environment
21

Journal of Chemical Information and Modeling
Page 22 of 49
1
2
3
4
5
6
7
8
9
FBAꢀII), NADH (0.6 mM), glycerol 3ꢀphosphate dehydrogenase (GPDH, 0.3 U, Sigma),
triosephosphate isomerase (TIM, 1 U, Sigma), aldolase (0.4 ꢁg/mL), and various concentrations
of FBP (50‒1000 ꢁM) in a cuvette to give the final volume of 0.4 mL. The reaction was carried
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
º
out for 5 min at 30 C. The decrease of NADH absorbance at 340 nm was monitored by
spectrophotometer (SpectraMax M5, Molecular Devices). The kinetic data were analyzed by
Nonlinear regression by using the Hill relation yields a maximal specific activity (Vmax) of 70
U/mg, a kcat of 46 sꢀ1 and a Km of 61 ꢁM, which is similar to the literature value (Km= 85 ꢁM).16
To determine the IC50 values, the initial rate data taken from a saturating substrate
concentration, a fixed effector concentration, and systematically varied concentrations of test
V =V −(V −V )/((I /I)n +1)
compounds were fit to a Hill equation,
75, where V, V0, and V∞
0
0
∞
0.5
are the velocity, maximum velocity (at I = 0), and the limiting velocity (at I saturating),
respectively; n is the Hill coefficient associated with inhibitor. All kinetic parameters of Caꢀ
FBAꢀII were determined by plotting the velocity vs substrate concentrations according to Hill
equation in origin 7.0 software. The inhibition assay was run under a concentration of
compounds 3c and 3g, as illustrated in Figure 7.
Mutations were performed by introducing specific base changes into a doubleꢀstranded DNA
plastid. All siteꢀdirected mutagenesis procedures were accomplished as our previous
publication44. Additionally, the parental methylated and hemimethylated DNA were digested by
Nde I and BamH I restriction enzyme (Fermentas, China). Expression and purification of mutant
CaꢀFBAꢀII were similar to those of wildꢀtype enzyme. By using SDSꢀPAGE, the identical
motilities were exhibited for all purified enzymes (wild and mutant) to ensure the purities more
than 95%.
ACS Paragon Plus Environment
22

Page 23 of 49
Journal of Chemical Information and Modeling
1
2
3
4
5
6
7
8
9
Antifungal Assays. Antifungal activities of some of the hit compounds, FCZ was measured by
means of the minimum inhibitory concentration (MIC) of the C. albicans strain (SC5314), C.
glabrata strain (537) and C. tropicalis in 96ꢀwell microꢀtest plates, as given in document M27ꢀ
A376 and M38ꢀA277 from the National Committee for Clinical Laboratory Standards (NCCLS).
Additionally, the MIC80 values were defined as being the minimum concentrations of the drugs
that inhibited the growth by more than 80% in comparision with that of the drugꢀfree wells. The
C. albicans strain (SC5314), C. glabrata strain (537) and C. tropicalis were obtained from the
Second Military Medical University. The test compounds were dissolved in DMSO, and serially
diluted in the range of 0.125 to 64.0 ꢁg/mL for the checkerꢀboard microꢀdilution assay. DMSO
was used as a growth control, FCZ was used as controls against the tested fungi. The microtiter
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
º
plates were incubated at 35 C for 48 h for the Candida species. The optical density was
monitored at 630 nm and the background absorptions were subtracted from the value of each
well.
Assays were performed on resistant Candida strains (100 and 103) in term of broth
microdilution checkerboard method. The final combination concentrations ranged from 0.125 to
64 ꢁg/mL or 0.016 to 8 ꢁg/mL for test compounds and 8 ꢁg/mL for FCZ. The final
concentrations of test compounds are from 0.125 to 64 ꢁg/mL, and FCZ is ranged from 2 to 1024
ꢁg/mL. The combination index (CI) was calculated according to the following equation:
ꢀꢁꢂ
CI = ꢀꢁꢂ
+
ꢃꢆ， where MICꢇꢈ and MICꢉꢈ are the MICs of the combination and alone of
ꢀꢁꢂ
ꢅꢆ
ꢃꢄ
ꢀꢁꢂ
ꢅꢄ
test compounds, and MICꢇꢊ and MICꢉꢊ are the MICs of the combination and alone of FCZ,
respectively.
ACS Paragon Plus Environment
23

Journal of Chemical Information and Modeling
Page 24 of 49
1
2
3
4
Supporting Information. Supporting Information is Available: Experimental procedures,
5
6
analytical data, enzyme assays, and biological evaluation.
7
8
9
AUTHOR INFORMATION
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Corresponding Author
*To whom correspondence should be addressed. Phone: +86ꢀ27ꢀ67862022. Fax: +86ꢀ27ꢀ
67862022. Eꢀmails: renyl@mail.ccnu.edu.cn; jianwan@mail.ccnu.edu.cn
Author Contributions
The manuscript was written through contributions of all authors. All authors have given approval
to the final version of the manuscript. ‡These authors contributed equally. (Match statement to
author names with a symbol)
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENT
This work was supported by the Natural Science Foundation of China (Nos. 21572077,
21472061, 21373094, 21202056 and 21272089), the Program for the PCSIRT (No. IRT0953),
and the selfꢀdetermined research funds of CCNU from the colleges’ basic research and operation
of MOE (Nos. CCNU16A02041 and CCNU14A05006).
ABBREVIATIONS
FBAꢀI, class I fructoseꢀ1,6ꢀbisphosphate aldolase; FBAꢀII, class II fructoseꢀ1,6ꢀbisphosphate
aldolase; CaꢀFBAꢀII, class II fructoseꢀ1,6ꢀbisphosphate aldolase from C. albicans; EcꢀFBAꢀII,
class II fructoseꢀ1,6ꢀbisphosphate aldolase from E. coli; class I fructoseꢀ1,6ꢀbisphosphate
ACS Paragon Plus Environment
24

Page 25 of 49
Journal of Chemical Information and Modeling
1
2
3
4
5
6
7
8
9
aldolase from rabbit muscle (RaꢀFBAꢀI); DHAP, dihydroxyacetonephosphate; GAP,
glyceraldehyde 3ꢀphosphate; GPDH, glycerol 3ꢀphosphate dehydrogenase; TIM, triosephosphate
isomerase; FBP, fructoseꢀ1,6ꢀbiphosphate; MD, molecular dynamic; BCE, bindingꢀconformation
evaluation; ONIOM, ourꢀown Nꢀlayered integrated molecular orbital and molecular mechanics;
ONIOM3, threeꢀlayer ONIOM; Fluconazole (FCZ); DOX, Docking, ONIOM and XO
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
REFERENCES
1.
Fonvielle, M.; Coincon, M.; Daher, R.; Desbenoit, N.; Kosieradzka, K.; Barilone, N.;
Gicquel, B.; Sygusch, J.; Jackson, M.; Therisod, M., Synthesis and biochemical evaluation of
selective inhibitors of class II fructose bisphosphate aldolases: towards new synthetic antibiotics.
Chem. Eur. J. 2008, 14, 8521ꢀ8529.
2.
Calderone, R. A.; Fonzi, W. A., Virulence factors of Candida albicans. Trends Microbiol.
2001, 9, 327ꢀ335.
3.
Calderone, R. A., Candida and Candidiasis. ASM Press, Washington, D. C. 2002.
4.
Odds, F. C., Candida and Candidosis, 2nd ed. Bailliere Tindall, London,United Kingdom.
1988.
5.
Vermes, A., Flucytosine: a review of its pharmacology, clinical indications,
pharmacokinetics, toxicity and drug interactions. J. Antimicrob. Chemother. 2000, 46, 171ꢀ179.
6. Goldman, W. E.; Butts, A.; Krysan, D. J., Antifungal Drug Discovery: Something Old and
Something New. PLoS Pathog. 2012, 8, e1002870ꢀ73.
7. Ashbee, H. R.; Barnes, R. A.; Johnson, E. M.; Richardson, M. D.; Gorton, R.; Hope, W.
W., Therapeutic drug monitoring (TDM) of antifungal agents: guidelines from the British Society
for Medical Mycology. J. Antimicrob. Chemother. 2013, 69, 1162ꢀ1176.
8.
Trends Microbiol. 2003, 11, 272ꢀ279.
9. Cui, J. H.; Ren, B.; Tong, Y. J.; Dai, H. Q.; Zhang, L. X., Synergistic combinations of
Odds, F. C.; Brown, A. J. P.; Gow, N. A. R., Antifungal agents: mechanisms of action.
antifungals and antiꢀvirulence agents to fight against Candida albicans. Virulence 2015, 6, 362ꢀ
371.
ACS Paragon Plus Environment
25

Journal of Chemical Information and Modeling
Page 26 of 49
1
2
3
4
10.
Troskie, A. M.; Rautenbach, M.; Delattin, N.; Vosloo, J. A.; Dathe, M.; Cammue, B. P.;
5
6
7
8
Thevissen, K., Synergistic activity of the tyrocidines, antimicrobial cyclodecapeptides from
Bacillus aneurinolyticus, with amphotericin B and caspofungin against Candida albicans
biofilms. Antimicrob. Agents Ch. 2014, 58, 3697ꢀ3707.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
11.
Baddley, J. W.; Pappas, P. G., Antifungal Combination Therapy. Drugs 2005, 65, 1461ꢀ
1480.
12.
Soulsby, E. J., Resistance to antimicrobials in humans and animals. BMJ 2005, 331,
1219ꢀ1220.
13.
Edmond, M. B.; Wallace, S. E.; McClish, D. K.; Pfaller, M. A.; Jones, R. N.; Wenzel, R.
P., Nosocomial bloodstream infections in United States hospitals: a threeꢀyear analysis. Clin.
Infect. Dis. 1999, 29, 239ꢀ244.
14.
Zhang, W.; Ramamoorthy, Y.; Kilicarslan, T.; Nolte, H.; Tyndale, R. F.; Sellers, E. M.,
Inhibition of cytochromes P450 by antifungal imidazole derivatives. Drug Metab. Dispos. 2002,
30, 314ꢀ318.
15.
mechanism, inhibitors and structural aspects. Prog. Biophys. Mol. Biol. 1995, 63, 301ꢀ340.
16. Daher, R.; Coincon, M.; Fonvielle, M.; Gest, P. M.; Guerin, M. E.; Jackson, M.; Sygusch,
Gefflaut, T.; Blonski, C.; Perie, J.; Willson, M., Class I aldolases: substrate specificity,
J.; Therisod, M., Rational design, synthesis, and evaluation of new selective inhibitors of
microbial class II (zinc dependent) fructose bisꢀphosphate aldolases. J. Med. Chem. 2010, 53,
7836ꢀ7842.
17.
Haake, V.; Zrenner, R.; Sonnewald, U.; Stitt, M., A moderate decrease of plastid aldolase
activity inhibits photosynthesis, alters the levels of sugars and starch, and inhibits growth of
potato plants. Plant J. 1998, 14, 147ꢀ157.
18.
Li, Z.; Liu, Z.; Cho, D. W.; Zou, J.; Gong, M.; Breece, R. M.; Galkin, A.; Li, L.; Zhao,
H.; Maestas, G. D.; Tierney, D. L.; Herzberg, O.; DunawayꢀMariano, D.; Mariano, P. S., Rational
design, synthesis and evaluation of first generation inhibitors of the Giardia lamblia fructoseꢀ
1,6ꢀbiphosphate aldolase. J. Inorg. Biochem. 2011, 105, 509ꢀ517.
19.
Pegan, S. D.; Rukseree, K.; Franzblau, S. G.; Mesecar, A. D., Structural Basis for
Catalysis of a Tetrameric Class IIa Fructose 1,6ꢀBisphosphate Aldolase from Mycobacterium
tuberculosis. J. Mol. Biol. 2009, 386, 1038ꢀ1053.
20.
Wehmeier, U., Molecular cloning, nucleotide sequence and structural analysis of the
ACS Paragon Plus Environment
26

Page 27 of 49
Journal of Chemical Information and Modeling
1
2
3
4
Streptomyces galbus DSM40480 fda gene: the S. galbus fructoseꢀ1,6ꢀbisphosphate aldolase is a
5
6
member of the class II aldolases. FEMS Microbiol. Lett. 2001, 197, 53ꢀ58.
7
8
21.
Kobayashi, K.; Ehrlich, S. D.; Albertini, A.; Amati, G.; Andersen, K. K.; Arnaud, M.;
Asai, K.; Ashikaga, S.; Aymerich, S.; Bessieres, P.; Boland, F.; Brignell, S. C.; Bron, S.; Bunai,
K.; Chapuis, J.; Christiansen, L. C.; Danchin, A.; Debarbouille, M.; Dervyn, E.; Deuerling, E.;
Devine, K.; Devine, S. K.; Dreesen, O.; Errington, J.; Fillinger, S.; Foster, S. J.; Fujita, Y.;
Galizzi, A.; Gardan, R.; Eschevins, C.; Fukushima, T.; Haga, K.; Harwood, C. R.; Hecker, M.;
Hosoya, D.; Hullo, M. F.; Kakeshita, H.; Karamata, D.; Kasahara, Y.; Kawamura, F.; Koga, K.;
Koski, P.; Kuwana, R.; Imamura, D.; Ishimaru, M.; Ishikawa, S.; Ishio, I.; Le Coq, D.; Masson,
A.; Mauel, C.; Meima, R.; Mellado, R. P.; Moir, A.; Moriya, S.; Nagakawa, E.; Nanamiya, H.;
Nakai, S.; Nygaard, P.; Ogura, M.; Ohanan, T.; O'Reilly, M.; O'Rourke, M.; Pragai, Z.; Pooley,
H. M.; Rapoport, G.; Rawlins, J. P.; Rivas, L. A.; Rivolta, C.; Sadaie, A.; Sadaie, Y.; Sarvas, M.;
Sato, T.; Saxild, H. H.; Scanlan, E.; Schumann, W.; Seegers, J. F.; Sekiguchi, J.; Sekowska, A.;
Seror, S. J.; Simon, M.; Stragier, P.; Studer, R.; Takamatsu, H.; Tanaka, T.; Takeuchi, M.;
Thomaides, H. B.; Vagner, V.; van Dijl, J. M.; Watabe, K.; Wipat, A.; Yamamoto, H.; Yamamoto,
M.; Yamamoto, Y.; Yamane, K.; Yata, K.; Yoshida, K.; Yoshikawa, H.; Zuber, U.; Ogasawara, N.,
Essential Bacillus subtilis genes. Proc Natl Acad Sci U S A 2003, 100, 4678ꢀ4683.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
22.
Jacobs, M. A.; Alwood, A.; Thaipisuttikul, I.; Spencer, D.; Haugen, E.; Ernst, S.; Will, O.;
Kaul, R.; Raymond, C.; Levy, R.; ChunꢀRong, L.; Guenthner, D.; Bovee, D.; Olson, M. V.;
Manoil, C., Comprehensive transposon mutant library of Pseudomonas aeruginosa. Proc. Natl
Acad. of Sci. USA 2003, 100, 14339ꢀ14344.
23.
Liberati, N. T.; Urbach, J. M.; Miyata, S.; Lee, D. G.; Drenkard, E.; Wu, G.; Villanueva,
J.; Wei, T.; Ausubel, F. M., An ordered, nonredundant library of Pseudomonas aeruginosa strain
PA14 transposon insertion mutants. Proc Natl Acad Sci U S A 2006, 103, 2833ꢀ2838.
24.
Song, J. H.; Ko, K. S.; Lee, J. Y.; Baek, J. Y.; Oh, W. S.; Yoon, H. S.; Jeong, J. Y.; Chun,
J., Identification of essential genes in Streptococcus pneumoniae by allelic replacement
mutagenesis. Mol. Cells 2005, 19, 365ꢀ374.
25.
Rodaki, A.; Young, T.; Brown, A. J., Effects of depleting the essential central metabolic
enzyme fructoseꢀ1,6ꢀbisphosphate aldolase on the growth and viability of Candida albicans:
implications for antifungal drug target discovery. Eukaryot. Cell 2006, 5, 1371ꢀ1377.
26.
Desvergnes, S.; CourtiolꢀLegourd, S.; Daher, R.; Dabrowski, M.; Salmon, L.; Therisod,
ACS Paragon Plus Environment
27

Journal of Chemical Information and Modeling
Page 28 of 49
1
2
3
4
5
6
7
8
9
M., Synthesis and evaluation of malonateꢀbased inhibitors of phosphosugarꢀmetabolizing
enzymes: class II fructoseꢀ1,6ꢀbisꢀphosphate aldolases, type I phosphomannose isomerase, and
phosphoglucose isomerase. Bioorg. Med. Chem. 2012, 20, 1511ꢀ1520.
27.
phosphate Aldolases. ACS Med. Chem. Lett. 2010, 1, 101ꢀ104.
28. Rao, L.; Chi, B.; Ren, Y.; Li, Y.; Xu, X.; Wan, J., DOX: A new computational protocol for
Daher, R.; Therisod, M., Highly Selective Inhibitors of Class II Microbial Fructose Bisꢀ
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
accurate prediction of the protein–ligand binding structures. J. Comput. Chem. 2016, 37, 336ꢀ
344.
29.
Moser, D.; Wisniewska, J. M.; Hahn, S.; Achenbach, J.; Buscató, E. l.; Klingler, F.ꢀM.;
Hofmann, B.; Steinhilber, D.; Proschak, E., DualꢀTarget Virtual Screening by Pharmacophore
Elucidation and Molecular Shape Filtering. ACS Med. Chem. Lett. 2012, 3, 155ꢀ158.
30.
Larkin, M. A.; Blackshields, G.; Brown, N. P.; Chenna, R.; McGettigan, P. A.;
McWilliam, H.; Valentin, F.; Wallace, I. M.; Wilm, A.; Lopez, R.; Thompson, J. D.; Gibson, T. J.;
Higgins, D. G., Clustal W and Clustal X version 2.0. Bioinformatics 2007, 23, 2947ꢀ2948.
31.
Hall, D. R.; Leonard, G. A.; Reed, C. D.; Watt, C. I.; Berry, A.; Hunter, W. N., The crystal
structure of Escherichia coli class II fructoseꢀ1,6ꢀbisphosphate aldolase in complex with
phosphoglycolohydroxamate reveals details of mechanism and specificity. J. Mol. Biol. 1999,
287, 383ꢀ394.
32.
homologyꢀmodeling server. Nucleic Acids Res. 2003, 31, 3381ꢀ3385.
33. Sybyl X version 1.3; Tripos Inc.: St. Louis, MO, 2011; http://www.tripos.com (accessed
Nov 12, 2011).
Schwede, T.; Kopp, J.; Guex, N.; Peitsch, M. C., SWISSꢀMODEL: an automated protein
34.
Laskowski, R. A.; MacArthur, M. W.; Moss, D. S.; Thornton, J. M., PROCHECK: a
program to check the stereochemical quality of protein structures. J. Appl. Crystallogr. 1993, 26,
283ꢀ291.
35.
measures for protein threading models. BMC Bioinformatics 2001, 2, 5ꢀ19.
36. Wiederstein, M.; Sippl, M. J., ProSAꢀweb: interactive web service for the recognition of
errors in threeꢀdimensional structures of proteins. Nucleic. Acids Res. 2007, 35, w407ꢀw410.
37. Sippl, M. J., Recognition of errors in threeꢀdimensional structures of proteins. Proteins
Struct. Funct. Bioinf. 1993, 17, 355ꢀ362.
Cristobal, S.; Zemla, A.; Fischer, D.; Rychlewski, L.; Elofsson, A., A study of quality
ACS Paragon Plus Environment
28

Page 29 of 49
Journal of Chemical Information and Modeling
1
2
3
4
38.
Benkert, P.; Schwede, T.; Tosatto, S., QMEANclust: estimation of protein model quality
5
6
7
8
by combining a composite scoring function with structural density information. BMC Struct.
Biol. 2009, 9, 35ꢀ51.
39.
de la Paz Santangelo, M.; Gest, P. M.; Guerin, M. E.; Coincon, M.; Pham, H.; Ryan, G.;
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Puckett, S. E.; Spencer, J. S.; GonzalezꢀJuarrero, M.; Daher, R.; Lenaerts, A. J.; Schnappinger,
D.; Therisod, M.; Ehrt, S.; Sygusch, J.; Jackson, M., Glycolytic and nonꢀglycolytic functions of
Mycobacterium tuberculosis fructoseꢀ1,6ꢀbisphosphate aldolase, an essential enzyme produced
by replicating and nonꢀreplicating bacilli. J. Biol. Chem. 2011, 286, 40219ꢀ40231.
40.
Lipinski, C. A.; Lombardo, F.; Dominy, B. W.; Feeney, P. J., Experimental and
computational approaches to estimate solubility and permeability in drug discovery and
development settings. Adv. Drug Del. Rev. 2012, 64, Supplement, 4ꢀ17.
41.
type constraints. J. Comput. Aided Mol. Des. 2002, 16, 129ꢀ149.
42. Jain, A., SurflexꢀDock 2.1: Robust performance from ligand energetic modeling, ring
flexibility, and knowledgeꢀbased search. J. Comput. Aided Mol. Des. 2007, 21, 281ꢀ306.
43. Jain, A. N., Surflex:ꢂ Fully Automatic Flexible Molecular Docking Using a Molecular
SimilarityꢀBased Search Engine. J. Med. Chem. 2003, 46, 499ꢀ511.
44. Li, D.; Gui, J.; Li, Y. J.; Feng, L. L.; Han, X. Y.; Sun, Y.; Sun, T. L.; Chen, Z. G.; Cao, Y.;
Hindle, S.; Rarey, M.; Buning, C.; Lengauer, T., Flexible docking under pharmacophore
Zhang, Y.; Zhou, L.; Hu, X. P.; Ren, Y. L.; Wan, J., StructureꢀBased Design and Screen of Novel
Inhibitors for Class II 3ꢀHydroxyꢀ3ꢀmethylglutaryl Coenzyme A Reductase from Streptococcus
Pneumoniae. J. Chem. Inf. Model. 2012, 52, 1833ꢀ1841.
45.
Zhang, Q. Y.; Li, D.; Wei, P.; Zhang, J.; Wan, J.; Ren, Y. L.; Chen, Z. G.; Liu, D. L.; Yu,
Z. N.; Feng, L. L., StructureꢀBased Rational Screening of Novel Hit Compounds with Structural
Diversity for Cytochrome P450 Sterol 14 alphaꢀDemethylase from Penicillium digitatum. J.
Chem. Inf. Model. 2010, 50, 317ꢀ325.
46.
Han, X.; Zhu, X.; Zhu, S.; Wei, L.; Hong, Z.; Guo, L.; Chen, H.; Chi, B.; Liu, Y.; Feng,
L.; Ren, Y.; Wan, J., A Rational Design, Synthesis, Biological Evaluation and Structure–Activity
Relationship Study of Novel Inhibitors against Cyanobacterial Fructoseꢀ1,6ꢀbisphosphate
Aldolase. J. Chem. Inf. Model. 2016, 56, 73ꢀ81.
47.
Case, D. A.; Darden, T. A.; Cheatham, T. E.; Simmerling, C. L.; Wang, J.; Duke, R. E.;
Luo, R.; Walker, R. C.; Zhang, W.; Merz, K. M.; Roberts, B.; Hayik, S.; Roitberg, A.; Seabra, G.;
ACS Paragon Plus Environment
29