Additional interaction of allophenylnorstatine-containing tripeptidomimetics with malarial aspartic protease plasmepsin II

Article abstract of DOI:10.1016/j.bmcl.2007.03.052

Based on a highly potent allophenylnorstatine-containing inhibitor, KNI-10006, against the plasmepsins of Plasmodium falciparum, we synthesized a series of tripeptide-type compounds with various N-terminal moieties and evaluated their inhibitory activitie

Full text of DOI:10.1016/j.bmcl.2007.03.052

Bioorganic & Medicinal Chemistry Letters 17 (2007) 3048–3052
Additional interaction of allophenylnorstatine-containing
tripeptidomimetics with malarial aspartic protease plasmepsin II
Koushi Hidaka,a Tooru Kimura,a Yumi Tsuchiya,a Mami Kamiya,a Adam J. Ruben,b
Ernesto Freire,b Yoshio Hayashia and Yoshiaki Kisoa,*
aDepartment of Medicinal Chemistry, Center for Frontier Research in Medicinal Science, 21st Century COE Program,
Kyoto Pharmaceutical University, Yamashina-ku, Kyoto 607-8412, Japan
bDepartment of Biology and Johns Hopkins Malaria Research Institute, Johns Hopkins University, Baltimore, MD 21218, USA
Received 27 February 2007; revised 14 March 2007; accepted 16 March 2007
Available online 21 March 2007
Abstract—Based on a highly potent allophenylnorstatine-containing inhibitor, KNI-10006, against the plasmepsins of Plasmodium
falciparum, we synthesized a series of tripeptide-type compounds with various N-terminal moieties and evaluated their inhibitory
activities against plasmepsin II. Certain phenylacetyl derivatives exhibited extremely high affinity with Ki values of less than
0.1 nM suggesting successful hydrophobic interactions.
Ó 2007 Elsevier Ltd. All rights reserved.
The most lethal malaria parasite, Plasmodium falcipa-
rum, increasingly develops resistance to available drugs
such as chloroquine or sulfadoxine-pyrimethamine.1
Drug resistance is one of the reasons for mortality due
to malaria, reaching up to 3 million people every year.
Malaria parasites invade the human host, who had been
bitten by the Anopheles mosquito carrier, through hema-
tocyte multiplication and proliferate during the erythro-
cyte cycle. In the cycle, hemoglobin is transported to
acidic food vacuoles and then cleaved by proteases into
small peptides to obtain amino acids as nutrients. This
hemoglobin degradation is essential for the parasites’
propagation.2 In the case of P. falciparum, four aspartic
proteases, named plasmepsins (Plms), participate in the
digestion.3 Recent studies suggested the importance of
inhibiting all four Plms to suppress parasite multiplica-
tion.4 Plm II has become a promising focus for new
anti-malarial drug design since the time its crystallo-
graphic data first became available.5 Thus far, several
groups have reported Plm inhibitors with potent inhibi-
tory activities.6 Our design of Plm inhibitors is based on
the hydroxymethylcarbonyl (HMC) isostere as an ideal
transition-state mimic established in renin and HIV
protease inhibitor studies.7 We have reported small
peptidic compounds incorporating an allophenylnorsta-
tine [Apns; (2S,3S)-3-amino-2-hydroxy-4-phenylbutyric
acid]-containing scaffold, that exhibited potent Plm II
inhibitory and anti-malarial activity.8 Among these ana-
logues, dipeptidomimetic KNI-10006 exhibited highly
potent inhibitory activity against Plm II (Fig. 1). KNI-
10006 also exhibited high potency against Plm I, IV,
and histo-aspartic protease.9
Structure–activity relationship (SAR) studies0at the po-
sition of KNI-10006 revealed a significant P2 contribu-
tion of the (1S,2R)-aminoindanol moiety to the tight
binding.10 The inhibitory activity of KNI-10006, how-
ever, was attenuated in P. falciparum-infected erythro-
cyte cultures. Our attention was directed to modifying
other positions in KNI-10006. Modifications of the
dipeptidic structure failed to obtain compounds with
more potent inhibitory activity against Plm II (data
not shown). In an attempt to diversify the development
of Apns-containing Plm inhibitors, we herein report
SAR on the N-terminal moiety of tripeptide-type com-
pounds containing Apns.
KNI-2727c,d was the first compound tested as a tripepti-
dic Plm II inhibitor containing Apns.8 Previous studies
on dipeptidic Plm II inhibitors stimulated us to examine
Keywords: Malaria; Aspartic protease; Plasmepsin inhibitor; Allopheny-
lnorstatine; Peptidomimetic.
KNI-227,7c possessing dimethyl groups in the P1 resi-
0
*
Corresponding author. Tel.: +81 75 595 4635; fax: +81 75 591
9900; e-mail: kiso@mb.kyoto-phu.ac.jp
due, specifically, Dmt [(R)-5,5-dimethyl-1,3-thiazoli-
0960-894X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.03.052

K. Hidaka et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3048–3052
3049
N
HO
O
O
O
O
O
O
H
N
O
N
N
N
O
N
N
N
H
H
H
H
OH
O
OH
R
S
S
R
S
KNI-10006 Ki = 0.5 nM
KNI-272 : R = H, Ki = 1000 nM
KNI-227 : R = CH3, Ki = 36 nM
N
HO
O
O
O
H
N
O
N
H
N
N
H
O
OH
S
S
1 (KNI-10033) Ki = 3 nM
Figure 1. Structures and Plm II inhibition constants of KNI-compounds.
dine-4-carboxylic acid] (Fig. 1). KNI-227 exhibited Plm
II inhibitory activity more potently than KNI-272 and
had improved Plm II selectivity (ꢀ50-fold) over
human cathepsin D. Replacements of the P2 (R)-methyl-
thioalanine (Mta) with hydrophobic or hydrophilic res-
idues were not effective in maintaining inhibitory
activity (data not shown).0 Introduction of (1S,2R)-1-
amino-2-indanol at the P2 position of KNI-227 gave
compound 1 (KNI-10033) exhibiting potent inhibitory
activity with a Ki value of 3 nM (Fig. 1). Although the
activity in P. falciparum-infected erythrocyte cultures
was slightly more desirable (EC50 = 3.1 lM) than that
of KNI-10006 (EC50 = 6.8 lM), the enzyme inhibitory
activity of 1 was less potent than that of KNI-10006.
Therefore, our attention was shifted to modify the
N-acyl group of Mta.
HO
O
O
O
a,b,c
Boc
Boc
N
OH
N
N
N
H
H
OH
S
S
2
3
b,d,b
HO
O
S
O
O
H2N
HCl•
N
H
N
N
H
OH
S
4
Synthetic routes of the N-terminal modified tripeptidic
compounds are shown in Scheme 1. Intermediate 4
was synthesized from Boc-Dmt-OH (2) as previously re-
ported10 with subsequent peptide couplings with ben-
zotriazol-1-yloxy-tris(dimethylamino)phosphonium
hexafluorophosphate (BOP) or N-ethyl-N0-[3-(dimeth-
ylamino)propyl]carbodiimide (EDC) plus 1-hydroxy-
benzotriazole (HOBt) method and deprotections of
Boc group using HCl-dioxane. In most cases, com-
pounds 5 were synthesized by coupling with correspond-
ing acids using the BOP reagent. In the case of
compound 5d, cyclohexyl succinimidylcarbonate was
reacted at rt. For amino-substituted phenylacetyl deriv-
atives (5h–j), Boc-amino-substituted phenylacetic acids
were used and finally deprotected. The pure products
were obtained after purification by preparative HPLC
and identified by MALDI-TOF MS. Enzyme inhibitory
activities against Plm II and HIV-1 protease were evalu-
ated as previously described.8,10
e
f
e,b
1, 5a–c,
5e–g, 5k–p
5d
5h–j
HO
O
H
O
O
R
N
N
H
N
N
H
O
OH
S
S
Scheme 1. Reagents: (a) (1S,2R)-1-amino-2-indanol, BOP, Et3N
DMF; (b) 4 N HCl/dioxane, anisole; (c) Boc-Apns-OH, EDCÆHCl,
HOBt, Et3N, DMF; (d) Boc-Mta-OH, EDCÆHCl, HOBt, Et3N, DMF;
(e) RCOOH, BOP, Et3N, DMF; (f) cyclohexyl succinimidylcarbonate,
Et3N, DMF. Mta = (R)-methylthioalanine.
inhibitory activity against Plm II (Table 1). We assumed
that an intramolecular hydrogen bond between the oxy-
gen of the ether moiety and NH group of Mta was
formed. This conformational constraint would be
preferred for binding with HIV-1 protease but would
interrupt the bindings within the pockets of Plm II.
We introduced N-terminal moieties that were known to
be effective in HIV-1 protease inhibition in previous
studies.7e In spite of their potent HIV-1 PR inhibitions,
similar to 1, compounds 5a and 5b exhibited less potent

3050
K. Hidaka et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3048–3052
Table 1. Inhibitory activity of N-terminal modified compounds
Table 2. Inhibitory activity of phenylacetyl derivatives against Plm II
and HIV-1 PR
HO
O
O
O
HO
R1
O
O
O
R
H
N
N
N
R2
R3
N
H
H
OH
N
N
N
S
H
H
S
O
OH
S
S
Compound
R
Plm II HIV-1 PR
Ki(nM) % inhibition
at 50 nM
Compound
R1
R2
R3
Plm II HIV-1 PR %
Ki (nM) inhibition at
50 nM
5c (KNI-10232) CH3
H
H
H
0.2
18
83
87
88
86
76
68
86
84
88
81
88
91
85
O
H
N
1 (KNI-10033)
3
99
5e (KNI-10332)
5f (KNI-10313)
H
H
CH3
H
N
CH3
H
5
4
O
5g (KNI-10372) NH2
H
5h (KNI-10368)
5i (KNI-10333)
H
H
NH2
H
H
12
NH2 60.1
H
N
O
H
5a (KNI-10061)
5b (KNI-10062)
15
9
98
99
H
5j (KNI-10341) OH
H
16
4
N
5k (KNI-10342)
5l (KNI-10343)
H
H
OH
H
H
O
OH
H
60.1
9
O
5m (KNI-10314) OCH3
H
H
N
5n (KNI-10316)
5o (KNI-10315)
5p (KNI-10395)
H
H
H
OCH3
H
H
17
4
O
OCH3
H
O
H
60.1
H
N
5c (KNI-10232)
5d (KNI-10233)
0.2
83
56
O
2006.08, Chemical Computing Group Inc., Montreal,
Canada). Several energy minimization processes with
an MMFF94x force field were additionally performed
including a water soak around the inhibitor. In the min-
imization steps, hydrogen bond interactions of HMC
with the two catalytic Asps were kept similar to that ob-
served in HIV protease complexes (PDB ID, 1HPX, and
1KZK), that is, the hydroxyl group of HMC interacted
with the carboxylate anion of one of the Asp, and the
carbonyl interacted with the carboxylic acid proton of
the other Asp.11 An energetically favored conformation
of 5i was obtained in Plm II, with Asp34 being proton-
ated (Fig. 2A). Most of the hydrogen bond interactions
were kept similar to a previous report10 with differences
near the phenylacetyl-Mta moiety. The N-terminal phe-
nyl moiety was surrounded by Phe 11, Gln12, Ile13, and
Met15, and the CH2 moiety had hydrophobic contact
with the side-chain of Met15 and was a little distant
from the side chain of Thr 114. We observed hydrogen
bond interactions of the p-amino group with two Gln
H
N
O
5
O
Therefore, we introduced two other moieties in com-
pounds 5c and 5d that would not form the intramolecu-
lar hydrogen bond interaction present in compounds 1,
5a, and 5b. We identified compound 5c (KNI-10232)
that possessed more potent Plm inhibitory activity
(Ki = 0.2 nM) than KNI-10006.
We focused on the phenylacetyl structure of 5c and
changed its o-methyl substituent. We introduced methyl,
amino, hydroxyl, and methoxy substituents at each
position on the phenyl ring (Table 2). These compounds
possessed moderate HIV-1 protease inhibition due to a
lack of the intramolecular hydrogen bond interaction
at the N-terminal moieties. As a result, highly potent
activities were shown only in the cases of o-methyl
(5c), p-amino (5i), and p-hydroxyl (5l) substitutions.
The latter two compounds exhibited extremely potent
activities (Ki 6 0.1 nM) suggesting possible hydrogen
bond interactions at the p-position to direct the orienta-
tion of the phenyl ring. Plm II favored none of the
m-substitutions, thereby implying steric hindrance at
the m-position. Interestingly, the removal of all substitu-
tions maintained highly potent activity (5p). Although
differences in their affinities are not clear, we believe that
successful hydrophobic interactions near the phenyl ring
play a key role.
12
˚
residues. In a recent report by Aqvist’s group, other
possible conformations in the opposite direction were
suggested using X-ray crystallographic data (PDB,
2ANL) of an Apns-containing compound KNI-764
and Plm IV from Plasmodium malariae (PmPM IV).13
Sequence identity between Plm II and PmPM IV is
72% including differences around binding pockets such
as Val78/Gly and Tyr192/Phe. One of the possible con-
formations of 5i in Plm II with a protonated Asp214 is
illustrated in Figure 2B. Dramatic movement of Val78
in the flap region of Plm II accommodated for the ben-
zyl group of Apns and N-terminal group, and disrupted
closing of the flap. On the other hand, some hydrogen
bond interactions were consistent with those suggested
The speculative interaction of 5i (KNI-10333) with Plm
II was confirmed by computational simulation (Fig. 2).
The inhibitor was manually docked into Plm II (PDB
ID, 1SME) by using a modeling package (MOE
˚
by Aqvist et al. The latter aminoindanol moiety could
form hydrogen bond interactions with Ser218, and not
with Tyr192 in Figure 2A. The interaction energy, which

K. Hidaka et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3048–3052
3051
Figure 2. Two speculative binding models of compound 5i [green in (A) or orange in (B)] and Plm II (blue ribbon). One of the catalytic two
aspartates was intentionally protonated [Asp34 in (A), Asp214 in (B)].
is the sum of the electrostatic and van der Waals ener-
gies calculated by MOE,14 between inhibitor and
enzyme of the latter conformation (Figure 2B,
ꢁ152 kcal/mol) was less favorable than that of the for-
mer (Figure 2A, ꢁ184 kcal/mol) because the N-terminal
group of the latter was surrounded mostly by water
molecules, and partially by flap and pocket residues
(Val78, Phe294, and Pro295).
References and notes
1. World Malaria Report 2005, WHO and UNICEF; 2005.
2. Banerjee, R.; Goldberg, D. E. In Antimalarial Chemother-
apy; Rosenthal, P. J., Ed.; Humana Press: Totowa, NJ,
2001; pp 43–63.
3. Banerjee, R.; Liu, J.; Beatty, W.; Pelosof, L.; Klemba, M.;
Goldberg, D. E. Proc. Natl. Acad. Sci. U.S.A. 2002, 99,
990.
4. Omara-Opyene, A. L.; Moura, P. A.; Sulsona, C. R.;
Bonilla, J. A.; Yowell, C. A.; Fujioka, H.; Fidock, D. A.;
Dame, J. B. J. Biol. Chem. 2004, 279, 54088.
5. Silva, A. M.; Lee, A. Y.; Gulnik, S. V.; Majer, P.; Collins,
J.; Bhat, T. N.; Collins, P. J.; Cachau, R. E.; Lurer, K. E.;
Gluzman, I. Y.; Francis, S. E.; Oksman, A.; Goldberg, D.
E.; Erickson, J. W. Proc. Natl. Acad. Sci. U.S.A. 1996, 93,
10034.
We evaluated the inhibitory activity of compound 5c
against P. falciparum. The EC50 value in erythrocyte cul-
tures was 4.7 lM. This significant difference from the Ki
value (0.2 nM) suggests a need to modify its chemical
properties such as decreasing the number of amide
bonds, or improving hydrophobic–hydrophilic balance
for membrane permeability to deliver inhibitor to the
target Plms.
6. See a review, (a) Ersmark, K.; Samuelsson, B.; Hall-
berg, A. Med. Res. Rev. 2006, 26, 626; Recent reports
´
(b) Ersmark, K.; Nervall, M.; Gutierrez-de-Teran, H.;
Hamelink, E.; Janka, L. K.; Clemente, J. C.; Dunn, B.
M.; Gogoll, A.; Samuelsson, B.; Aqvist, J.; Hallberg,
In conclusion, we synthesized a series of Apns-contain-
ing Plm inhibitors at tripeptid0ic size. Starting from
˚
0
KNI-272, replacements with P1 dimethyl groups, P2
A. Bioorg. Med. Chem. 2006, 14, 2197; (c) Corminb-
oeuf, O.; Dunet, G.; Hafsi, M.; Grimont, J.; Grisos-
tomi, C.; Meyer, S.; Binkert, C.; Bur, D.; Jones, A.;
Prade, L.; Brun, R.; Boss, C. Bioorg. Med. Chem. Lett.
2006, 16, 6194; (d) Boss, C.; Corminboeuf, O.; Grisos-
tomi, C.; Meyer, S.; Jones, A. F.; Prade, L.; Binkert,
C.; Fischli, W.; Weller, T.; Bur, D. Chem. Med. Chem.
2006, 1, 1341; (e) Dell’Agli, M.; Parapini, S.; Galli, G.;
Vaiana, N.; Taramelli, D.; Sparatore, A.; Liu, P.;
Dunn, B. M.; Bosisio, E.; Romeo, S. J. Med. Chem.
2006, 49, 7440.
aminoindanol and modifications of the N-terminal
group led to compounds exhibiting more potent activity
than KNI-10006, that are more than 10 thousand times
more potent than KNI-272. Evaluations of their anti-
malarial activity and further modifications using hydro-
phobic N-terminal moieties to develop anti-malarial
agents are under investigation.
7. (a) Izuka, K.; Kamijo, T.; Harada, H.; Akahane, K.;
Kubota, T.; Umeyama, H.; Ishida, T.; Kiso, Y. J. Med.
Chem. 1990, 33, 2707; (b) Mimoto, T.; Imai, J.; Tanaka,
S.; Hattori, N.; Kisanuki, S.; Akaji, K.; Kiso, Y. Chem.
Pharm. Bull. 1991, 39, 3088; (c) Mimoto, T.; Imai, J.;
Kisanuki, S.; Enomoto, H.; Hattori, N.; Akaji, K.; Kiso,
Y. Chem. Pharm. Bull. 1992, 40, 2251; (d) Kiso, Y.;
Matsumoto, H.; Mizumoto, S.; Kimura, T.; Fujiwara, Y.;
Akaji, K. Biopolymers 1999, 51, 59; (e) Takaku, H.;
Nojima, S.; Mimoto, T.; Terashima, K.; Kiso, Y. PCT Int.
Appl., WO9829118.
8. Nezami, A.; Luque, I.; Kimura, T.; Kiso, Y.; Freire, E.
Biochemistry 2002, 41, 2273.
9. Nezami, A.; Kimura, T.; Hidaka, K.; Kiso, A.; Liu,
J.; Kiso, Y.; Goldberg, D. E.; Freire, E. Biochemistry
2003, 42, 8459; Highlighted in Science 2003, 301,
143.
Acknowledgments
This research was supported in part by the Frontier
Research Program, the 21st Century COE Program
of Ministry of Education, Culture, Sports, Science,
and Technology of Japan (MEXT), and grants from
MEXT. E.F. and A.J.R. also thank the Johns Hopkins
Malaria Institute for a Pilot Research Grant and a
Graduate Student Fellowship. We gratefully acknowl-
edge Ms. Y. Iteya for synthetic assistance, and Mr.
T. Hamada for the determination of HIV-1 PR inhib-
itory activity. We are grateful to Dr. J.-T. Nguyen for
consult and revision of the manuscript. We thank the
Swiss Tropical Institute for performing the erythrocyte
inhibition assays.

3052
K. Hidaka et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3048–3052
10. Kiso, A.; Hidaka, K.; Kimura, T.; Hayashi, Y.;
Nezami, A.; Freire, E.; Kiso, Y. J. Peptide Sci. 2004,
10, 641.
11. Wang, Y. X.; Freedberg, D. I.; Yamazaki, T.; Wingfield,
P. T.; Stahl, S. J.; Kaufman, J. D.; Kiso, Y.; Torchia, D.
A. Biochemistry 1996, 35, 9945.
13. Clemente, J. C.; Govindasamy, L.; Madabushi, A.; Fisher,
S. Z.; Moose, R. E.; Yowell, C. A.; Hidaka, K.; Kimura,
T.; Hayashi, Y.; Kiso, Y.; Agbandje-McKenna, M.;
Dame, J. B.; Dunn, B. M.; McKenna, R. Acta. Crystal-
logr. Sect. D 2006, 62, 246.
14. Halgren, T.A. J. Comput. Chem. 1996, 17, (Part I) p. 490,
(Part II) p. 520, (Part III) p. 553, (Part IV) p. 587, (Part V)
p. 616.
´
12. Gutierrez-de-Teran, H.; Nervall, M.; Dunn, B. M.;
˚
Clemente, J. C.; Aqvist, J. FEBS Lett. 2006, 580, 5910.