Fragment-Based Design of Novel Quinazolinon Derivatives as Human Acrosin Inhibitors

Article abstract of DOI:10.1111/cbdd.12106

Human acrosin is a promising target for the male contraceptives. On the basis of the active site of human acrosin, a series of novel quinazolinon compounds were designed by a fragment docking and growing strategy. In vitro anti-acrosin assay revealed that all the compounds showed potent human acrosin inhibitory activities. In particular, compounds 5c and 5g are more active than the known inhibitors. Molecular docking studies revealed that the quinazolinon inhibitors interacted with human acrosin mainly through hydrogen bonding and hydrophobic interactions. The binding mode was also consistent with the structure-activity relationships. The quinazolinon derivatives in this study can serve as new lead structure for the development of novel male contraceptives. On the basis of the active site of human acrosin, a series of novel quinazolinon derivatives were designed and synthesized by a fragment docking and growing strategy. Compound 5g showed potent human acrosin inhibitory activity and was more active than the known inhibitors.

Full text of DOI:10.1111/cbdd.12106

Chem Biol Drug Des 2013; 81: 437–441
Research Letter
Fragment-Based Design of Novel Quinazolinon
Derivatives as Human Acrosin Inhibitors
a potential target for the development of novel male
Weiwei Ning, Ju Zhu, Canhui Zheng, Xuefei Liu,
Yunlong Song, Youjun Zhou, Xiaomeng Zhang,
Ling Zhang, Chunquan Sheng* and Jiaguo Lv*
contraceptive agents. Currently, known human acrosin
inhibitors include bdellins (3), N-alpha-tosyl-L-lysyl-chlorom-
ethylketone (TLCK) (4), suramin (5), p-aminobenzamidine
(pAB) (6), isoxazolecarbaldehydes (7), and 4-guani-
dinobenzoates (8) (Figure 1). However, most of them have
low affinity, poor selectivity, and high toxicity. In this sce-
nario, rational inhibitor design on the basis of the three-
dimensional (3D) structure of human acrosin is an effective
way to find novel lead structures.
Department of medicinal Chemistry, School of Pharmacy,
Second Military Medical University, 325 Guohe Road,
Shanghai, 200433, China
*Corresponding authors: Chunquan Sheng,
shengcq@hotmail.com; Jiaguo Lv, ljg20060508@yahoo.
com.cn
Human acrosin is a promising target for the male con-
traceptives. On the basis of the active site of human
acrosin, a series of novel quinazolinon compounds
were designed by a fragment docking and growing
strategy. In vitro anti-acrosin assay revealed that all
the compounds showed potent human acrosin inhibi-
tory activities. In particular, compounds 5c and 5g are
more active than the known inhibitors. Molecular dock-
ing studies revealed that the quinazolinon inhibitors
interacted with human acrosin mainly through hydro-
gen bonding and hydrophobic interactions. The bind-
ing mode was also consistent with the structure–
activity relationships. The quinazolinon derivatives in
this study can serve as new lead structure for the
development of novel male contraceptives.
In our previous studies, 3D model of human b-acrosin has
been constructed by homology modeling (9) using the
crystal structures of both ram and boar acrosin (10) as the
templates. The active site of human acrosin and the bind-
ing mode of known inhibitors were investigated by the
multiple copy simultaneous search (11) and molecular
docking (12). Moreover, novel human acrosin inhibitors
have been identified by virtual screening (13). In recent
years, fragment-based drug design (FBDD) has been
emerged as a promising approach for lead discovery and
optimization (14,15). Herein, we describe fragment-based
rational design and synthesis of new quinazoline ketone
derivatives as potent human acrosin inhibitors.
Key words: binding mode, human acrosin, lead structure,
male contraceptives, quinazolinon inhibitor
The first step of structure-based drug design is to analyze
the properties of the active site. As shown in Figure 2A,
the active site of human acrosin can be divided into three
parts: P1, P2, and G pocket. The P1 pocket is a polar
and deep site lined with residues Thr216, Cys217,
Gln218, Trp243, and Gly244, which are important for
inhibitor binding. The entrance of the P2 pocket is larger
than the P1 pocket, and important residues in this area
include Arg199, Tyr196, Gly244, Val255, and Val245. The
G pocket is a narrow channel that is surrounded by
Glu120, His69, Trp243, and Ala68. In particular, Trp243 is
located near the center of the P1, P2, and G pocket and
can be used as the starting point for drug design. Thus,
the following task is to find a suitable chemical scaffold
that can be located in the center of the active site and
interact with Trp243. The scaffold library was obtained
from Ji’s de novo design method fragment hopping (16).
The fragment-like molecules were docked into the active
site of human acrosin using AUTODOCK (17). The fragments
were ranked by the binding energy, and the top 10 scored
fragments (see Appendix S1 in Supporting information)
were subjected for visual inspection of their binding mode.
Abbreviations: AIBN, azodiisobutyronitrile; FBDD, fragment-
based drug design; MCSS, multiple copy simultaneous
search; NBS, N-bromosuccinimide.
Received 15 October 2012, revised 2 January 2013 and
accepted for publication 10 January 2013
Fertility control is an important and global issue because
of overpopulation and unintended pregnancy. In contrast
to various hormonal-based oral contraceptives for women,
the discovery and development of effective, safe, and
reversible male contraceptive remain a challenge. Human
acrosin is a serine protease that is located in epididymis
and plays an important role during the process of fertiliza-
tion (1). The functions of human acrosin during fertilization
include acrosomal exocytosis, the receptor for zona pellu-
cida proteins, and facilitating the penetration of spermato-
zoa into the egg (2). Because of its special location and
high cell specificity, human acrosin has been regarded as
ª 2013 John Wiley & Sons A/S. doi: 10.1111/cbdd.12106
437

Ning et al.
N
O
O
O
ClH2C
N
Cl
O
O
CHO
S
OHC
H2N
N
H
O2N
CH3
Cl
TLCK
(IC50 = 142.6 mM)
ISO-2
ISO-1
(IC50 = 42 mM)
(IC50 = 39 mM)
H
H2N
N
H3C
N
O
N
O
S
NH
HCl
O
N
H
NH
HN
N
O
O
H3C
NHCCH3
O
5 g
(IC50 = 80 uM)
4-guanidinobenzoates
(IC50 = 8 mM)
Figure 1: Chemical structures of human acrosin inhibitors and compound 5g.
Table 1: Chemical structure, human acrosin inhibitory activity (IC50, mM), and binding energy (Kcal/mol) of the quinazolinon derivatives
O
R
NH
N
Compounds
R
H
IC50
5.3
Binding energy
3
À7.80
5a
0.98
À9.83
H3C
H
N
CH3
5b
0.84
À9.61
H
N
H3CO
5c
5d
0.25
0.78
À11.31
À10.7
H
H
N
N
O
S
N
S
H
N
HN
O
5e
5f
1.3
À8.26
À9.33
O
N
N
H
N
H
N
S
O
0.82
O
S
H
H
N
N
O
N
H3C
O
438
Chem Biol Drug Des 2013; 81: 437–441

Novel Quinazolinon Derivatives as Human Acrosin Inhibitors
Table 1: continued
Compounds
5g
R
IC50
0.080
Binding energy
À16.24
H3C
O
S
N
N
HN
HN
O
H3C
5h
5i
1.8
2.3
À10.4
À9.98
NH
Cl
N
N
N
H
N
N
Cl
5j
2.4
À8.37
N
H3C
TLCK
142.6
À8.26
TLCK, N-alpha-tosyl-L-lysyl-chloromethylketone.
As a result, quinazolinon was finally selected as the core
structure to interact with human acrosin. Molecular dock-
ing results revealed that the quinazolinon scaffold 3 was
positioned near Trp243 and formed two hydrogen bonds
with Trp243 and Gln218, respectively (Figure 2B). To vali-
date the reliability of fragment docking, the acrosin inhibi-
method of Kennedy et al. (18). N-alpha-tosyl-L-lysyl-
chloromethylketone was used as a standard reference
drug. The IC50 values and AUTODOCK docking energies are
listed in Table 1. In general, all the compounds are active
against human acrosin with their IC50 values in the range
of 80 lM to 2.4 mM. In comparison with the activity of
known human acrosin inhibitors (Figure 1), significant
improvement of the inhibitory activity was observed for
the quinazolinon derivatives reported here. Particularly,
compound 5g showed the best inhibitory activity
(IC50 = 80 lM), which was far more potent than the refer-
ence drug TLCK. AutoDock was used to investigate the
binding mode of the synthesized compounds. Figure 2B
depicts the interaction between representative compound
5g and the active site of human acrosin. After the attach-
ment of the amine side chain to the scaffold, the position
and key interactions of the quinazolinon core are retained.
The imine nitrogen atom and amide NH of the quinazoli-
non ring formed two hydrogen bonds with Trp243 and
Gln218, respectively. The amine side chain of compound
5g was positioned into the G and P2 pocket. The NH lin-
ker also formed hydrogen bonding interaction with Glu120
in the G pocket. The sulfonamide group does not form
direct interaction with human acrosin. The terminal dimeth-
ylpyridine was located in the P2 pocket and formed van
der Waals and hydrophobic interaction. In addition, the
nitrogen atom of the pyridine ring also formed a hydrogen
bond with Arg199, which is an important residue to
achieve specificity (10).
tory activity of the quinazolinon scaffold
3
was
determined. To our delight, fragment 3 showed moderate
inhibitory activity with an IC50 value of 5.3 mM. Then, frag-
ment-based growing strategy was used to improve the
binding affinity of the quinazolinon core. In consideration
of synthetic accessibility, various amine-containing side
chains, such as substituted anilines, aliphatic amines, and
heterocycle-sulfonamide-phenyl amines, were designed to
be lined to the quinazolinon framework to form additional
hydrophobic and hydrogen bonding interactions with the
P2 and G pocket. As a result, compounds 5a–j (Table 1)
were subjected to chemical synthesis.
Scheme 1 outlines the synthetic route of the target com-
pounds. Starting from 2-amino-5-methylbenzoic acid (1),
it was condensed with propionic anhydride to afford ben-
zoxazine-4-one compound 2. After dehydration with form-
amide, intermediate
2 was converted to quinazoline
derivative 3. The methyl group of compound 3 was bro-
minated by N-bromosuccinimide (NBS) and azodiisobuty-
ronitrile to give bromomethyl derivative 4. Target
compounds 5a–j were obtained by reaction 4 with vari-
ous amines in the presence of triethylamine (TEA) and
DMF.
The binding mode of compound 5g is supported by
the SAR results of the synthesized compounds. Substi-
tuted aniline derivatives 5a and 5b are more potent
The in vitro human acrosin inhibitory activities of the syn-
thesized compounds were determined according to the
Chem Biol Drug Des 2013; 81: 437–441
439

Ning et al.
than the purine derivative 5i and piperidine derivative
5j because they form stronger hydrophobic interaction
with the G pocket. However, all of them have short
side chain and cannot interact with the P2 pocket.
Therefore, they are less active than the compounds
with long side chain (e.g., compounds 5c–g), which
have lower interaction energies with human acrosin
(Table 1). In comparison with compound 5g, the inhibi-
tory activity of demethyl derivative 5e is decreased
because it forms lower hydrophobic interaction with
the P2 pocket. Similarly, the replacement of terminal
phenyl group of compound 5c with less hydrophobic
thiazole (compound 5d) and methylisoxazole (com-
pound 5f) leads to the decrease in the inhibitory activ-
ity. Compound 5g is the most active compound
because its terminal dimethylpyrimidine group can form
both strong hydrophobic interaction and additional
hydrogen bonding interaction with the P2 pocket.
Moreover, several acrosin inhibitors, such as TLCK and
isoxazole aldehydes, could bind covalently with acro-
sin. Because covalent inhibitors are being used more
frequently in the pharmaceutical industry, it might be
applied to further optimize compound 5g by incorpo-
rating a proper substituent to covalently interact with
key residues of acrosin (e.g., Asp189, Ser221).
A
Tyr196
p2
Glu120
Va1245
Gly244
G
Asp123
His69
Trp243
p1
Gln218
Ser221
Thr216
Cys217
B
Arg199
Glu120
In summary, computational fragment-based approaches
were used to design a series of quinazolinon derivatives
as novel human acrosin inhibitors. The quinazolinon deriv-
ative showed improved inhibitory activity over the known
human acrosin inhibitors. The most active compound 5a
has good drug-like properties, which represents a good
starting point for lead optimization. Furthermore, the frag-
ment docking and fragment growing strategy used in this
study should have general implications for structure-based
rational drug design.
G
p2
Trp243
Gln218
p1
Acknowledgment
This work was supported by the Shanghai population and
Family Planning Commission (Grant Nos. 2007JG-02) and
National Natural Science Foundation of China (Grant No.
81222044).
Figure 2: (A) The active site of human acrosin; (B) the binding
mode of inhibitor 5g in the active site of human acrosin and
hydrogen bonds are displayed as dotted lines.
O
O
H3C
NH
H3C
H3C
COOH
NH2
a
O
b
N
N
1
3
2
O
N
O
N
BrH2C
NH
R
NH
c
d
5a- j
4
Scheme 1: Reagents and conditions: (a) Propionic anhydride, reflux, 5 h, 69.3%; (b) HCONH2, 150 °C, 7 h, 75.5%; (c) N-
bromosuccinimide (NBS), azodiisobutyronitrile (AIBN), CHCl3, reflux, 4 h, 67.8%; (d) substituted amines, Et3N, DMF, 10 h, 6.9–98.7%.
440
Chem Biol Drug Des 2013; 81: 437–441

Novel Quinazolinon Derivatives as Human Acrosin Inhibitors
12. Zhang J., Zheng C., Sheng C., Zhou Y., Zhu J., Lv J.
(2009) Characterization of human acrosin active site
and binding modes with its inhibitors. Chem J Chinese
U;30:2409–2414.
13. Liu X., Dong G., Zhang J., Qi J., Zheng C., Zhou Y.,
Zhu J., Sheng C., Lu J. (2011) Discovery of novel
human acrosin inhibitors by virtual screening. J Com-
put Aided Mol Des;25:977–985.
References
1. Klemm U., Muller-Esterl W., Engel W. (1991) Acrosin,
the peculiar sperm-specific serine protease. Hum
Genet;87:635–641.
2. Howes L., Jones R. (2002) Interactions between zona
pellucida glycoproteins and sperm proacrosin/acrosin
during fertilization. J Reprod Immunol;53:181–192.
3. Baskova I.P., Zavalova L.L. (2001) Proteinase inhibitors
from the medicinal leech Hirudo medicinalis. Biochem-
istry;66:703–714.
14. Loving K., Alberts I., Sherman W. (2010) Computa-
tional approaches for fragment-based and de novo
design. Curr Top Med Chem;10:14–32.
15. Rees D.C., Congreve M., Murray C.W., Carr R. (2004)
Fragment-based lead discovery. Nat Rev Drug Dis-
cov;3:660–672.
4. Pakzad R. (1989) The effect of the trypsin inhibitor ap-
rotinin (Trasylol) and TLCK on the gelatinolytic activity
of acrosin and the motility of rabbit sperm in vitro. Z
Mikrosk Anat Forsch;103:8–13.
ꢀ
16. Ji H., Stanton B.Z., Igarashi J., Li H., Martasek P.,
Roman L.J., Poulos T.L., Silverman R.B. (2008) Mini-
mal pharmacophoric elements and fragment hopping,
an approach directed at molecular diversity and iso-
zyme selectivity. Design of selective neuronal nitric
oxide synthase inhibitors. J Am Chem Soc;130:3900–
3914.
5. Jones R., Parry R., Lo Leggio L., Nickel P. (1996) Inhi-
bition of sperm-zona binding by suramin, a potential
‘lead’ compound for design of new anti-fertility agents.
Mol Hum Reprod;2:597–605.
6. Fraser L.R. (1982) p-Aminobenzamidine, an acrosin
inhibitor, inhibits mouse sperm penetration of the zona
pellucida but not the acrosome reaction. J Reprod Fer-
til;65:185–194.
7. Gupta G., Jain R.K., Maikhuri J.P., Shukla P.K., Kumar
M., Roy A.K., Patra A., Singh V., Batra S. (2005) Dis-
covery of substituted isoxazolecarbaldehydes as potent
spermicides, acrosin inhibitors and mild anti-fungal
agents. Hum Reprod;20:2301–2308.
8. Bourinbaiar A.S., Lee-Huang S. (1995) Acrosin inhibi-
tor, 4′-acetamidophenyl 4-guanidinobenzoate, an
experimental vaginal contraceptive with anti-HIV activ-
ity. Contraception;51:319–322.
17. Morris G.M., Goodsell D.S., Halliday R.S., Huey R.,
Hart W.E., Belew R.K., Olson A.J. (1998) Automated
docking using a Lamarckian genetic algorithm and an
empirical binding free energy function.
J Comb
Chem;19:1639–1662.
18. Kennedy W.P., Kaminski J.M., Van der Ven H.H.,
Jeyendran R.S., Reid D.S., Blackwell J., Bielfeld P.,
Zaneveld L.J. (1989) A simple, clinical assay to evalu-
ate the acrosin activity of human spermatozoa. J An-
drol;10:221–231.
Supporting Information
9. Lu J.G., Sheng C.Q., Zhang M., Ji H.T., Zhang W.N.,
Zhou Y.J., Zhu J., Jiang J.H. (2006) Homology model-
ing of human acrosin and its molecular docking study
with KF950. Acta Chim Sinica;64:1073–1078.
10. Tranter R., Read J.A., Jones R., Brady R.L. (2000)
Effector sites in the three-dimensional structure of
mammalian sperm beta-acrosin. Structure;8:1179–
1188.
Additional Supporting Information may be found in the
online version of this article:
Appendix S1. Top 10 ranked scaffolds for fragment-
based drug design. Experimental details, NMR and MS
data for the synthesized compounds. The binding mode of
scaffold 3 in the active site of human acrosin. Molecular
weight (MW), ligand efficiency (LE) and ligand-lipophilicity
efficiency of the target compounds.
11. Miranker A., Karplus M. (1991) Functionality maps of
binding sites: a multiple copy simultaneous search
method. Proteins;11:29–34.
Chem Biol Drug Des 2013; 81: 437–441
441