Synthesis of malarial plasmepsin inhibitors and prediction of binding modes by molecular dynamics simulations

6090

J. Med. Chem. 2005, 48, 6090-6106

Synthesis of Malarial Plasmepsin Inhibitors and Prediction of Binding Modes

by Molecular Dynamics Simulations

Karolina Ersmark,^†Martin Nervall,^‡Elizabeth Hamelink,^§Linda K. Janka,^#Jose C. Clemente,^#Ben M. Dunn,^#

Michael J. Blackman, Bertil Samuelsson,^§Johan A° qvist,^‡and Anders Hallberg*^,†

Department of Medicinal Chemistry, Uppsala University, BMC, Box 574, 751 23 Uppsala, Sweden, Department of Cell and

Molecular Biology, Uppsala University, BMC, Box 596, 751 24 Uppsala, Sweden, Division of Parasitology, National Institute

for Medical Research, The Ridgeway, Mill Hill, London NW/ 1AA, UK, Medivir AB, Lunastigen 7, 141 44 Huddinge, Sweden,

and Department of Biochemistry and Molecular Biology, University of Florida College of Medicine, P.O. Box 100245

Gainesville, Florida 32610-0245

Received May 17, 2005

A series of inhibitors of the malarial aspartic proteases Plm I and II have been synthesized

with ^L-mannitol as precursor. These inhibitors are characterized by either a diacylhydrazine

or a five-membered oxadiazole ring replacing backbone amide functionalities. Molecular

dynamics simulations were applied in the design process. The computationally predicted Plm

II K_ivalues were generally in excellent agreement with the biological results. The diacyl-

hydrazine was found to be superior over the oxadiazole as an amide bond replacement in the

Plm I and II inhibitors studied. An extensive flexibility of the S2′ pocket was captured by the

simulations predicting the binding mode of the unsymmetrical inhibitors. Plm I and II inhibitors

with single digit nanomolar K_ivalues devoid of inhibitory activity toward human Cat D were

identified. One compound, lacking amide bonds, was found to be Plm IV selective and very

potent, with a K_ivalue of 35 nM.

Introduction

inhibitors of Plm I and II have been identified.^7,12,19-34

However, achieving selectivity versus the highly homo-

logous human aspartic protease cathepsin D (Cat D),

and activity in parasite infected red blood cells have

emerged as major challanges.^7,35

Malaria, one of the most serious infectious diseases

in the world, is caused by protozoan parasites of the

genus Plasmodium. Of the four human species, Plas-

modium falciparum is by far the most lethal and is

responsible for the majority of the estimated 0.7-2.7

million malaria related deaths occurring annually.¹

Currently, there is a rapid spread of parasite drug

resistance, mainly among the P. falciparum strains,

resulting in an urgent need for new effective drugs with

new mechanisms of action.²The publication of the P.

falciparum genome has revealed a number of new

targets for drug development.³Among these are the

hemoglobin degrading aspartic proteases plasmepsin I,

II and IV (Plm I, II and IV) and the closely related

histoaspartic protease (HAP).⁴It has been shown that

hemoglobin catabolism, which takes place in an acidic

food vacuole, is essential for parasite survival both in

culture and in animal models.^5-7Besides aspartic

proteases three cysteine proteases (falcipain-1, -2, and

-3) and one metallo protease (falcilysin) have also been

identified to digest hemoglobin in the food vacuole.^8-11

Until now Plm I and II have been most extensively

studied of the four aspartic proteases, but there is a

growing interest in the recently characterized Plm IV

and HAP.^12-15Plm I and II demonstrate the highest

sequence homology and are both involved in the initial

cleavage of the native hemoglobin.^16-18Several potent

Directed by binding affinity calculations using the

linear interaction energy (LIE) method, we previously

reported the design of a series of C₂-symmetric inhibi-

tors with high affinities to both Plm I and II (K_i) 0.5-

37 nM and 6-181 nM, respectively).²⁶Notably, all of

the inhibitors demonstrated a unique selectivity versus

the human Cat D (K_i> 2000 nM).²⁶The characteristic

features of these inhibitors were (a) a 1,2-dihydroxy-

ethylene unit mimicking the tetrahedral intermediate

in the enzymatic cleavage and (b) two extended P1/P1′

side chains designed to interact with the unobstructed

S1-S3 cleft and the flexible S1′ region. Despite high

affinities in plasmepsin assays, the inhibitors exerted

moderate inhibition of parasite growth in red blood

cells.²⁶

Application of the concept of bioisosteric replacement

is an established strategy to circumvent undesirable

ADME profiles and has been extensively used as an

approach to optimize lead compounds in drug discov-

ery.^36,37In conversions of peptides to peptide mimetics

the amide bonds are often replaced.^36-38The suitability

of the replacing groups as bioisosteres depend on the

role of the amide in the molecular recognition, i.e., if

the amide group acts as a hydrogen bond donor and/or

acceptor or if it solely acts as a spacer when the molecule

interacts with its receptor.

* To whom correspondence should be addressed. E-mail:

Anders.Hallberg@orgfarm.uu.se. Phone: +46-18471 42 84, Fax: +46-

18 471 44 74.

Both experimental crystal structures^21,33,39and mo-

lecular dynamics (MD) simulations^26,27indicate that the

Plm II binding cleft is quite flexible and can accom-

modate ligands with P1/P1′ and P2/P2′ substituents of

^†Department of Medicinal Chemistry, Uppsala University.

^‡Department of Cell and Molecular Biology, Uppsala University.

National Institute for Medical Research.

^§Medivir AB.

^#University of Florida College of Medicine.

Published on Web 08/23/2005

Malarial Plasmepsin Inhibitors

Journal of Medicinal Chemistry, 2005, Vol. 48, No. 19 6091

Chart 1. Generic Structures of Plm II Substrate (A)

and Two Inhibitors (B and C) Encompassing

TS-Mimicking Scaffolds Derived from L-Mannitol

varying size. Such a situation generally makes auto-

mated docking procedures less reliable if the receptor

protein is treated as rigid, which is often the case.

Hence, automated docking of our earlier series of

inhibitors using the X-ray structure of the Plm II-

pepstatin complex was not successful. On the other

hand, we found that superposition of the inhibitor

scaffolds combined with subsequent relaxation by MD

simulations produced reasonable structures as judged

both by their energetics and by comparison to indepen-

dent crystallographic results.²⁷Thus, MD simulations

in combination with the linear interaction energy (LIE)

method resulted in theoretical binding affinities in

excellent agreement with experimental measurements

when the C₂-symmetric inhibitors were investigated.^26,27

Herein, we report the impact of replacing one or both

of the two amide bonds of the previously reported C₂-

symmetric inhibitors,²⁶by either (a) a diacylhydrazine

element with ability to participate in an extensive

hydrogen bond network or (b) a compact 1,3,4-oxadiazole

possessing only hydrogen bond accepting capacity. The

characteristic features of the central core structure of

the unsymmetrical inhibitors are illustrated in Chart

1. The design that delivered potent and selective Plm

I/Plm II inhibitors with no inhibition of the human Cat

D used a bisvinyl bromide as a starting structure and

was partly guided by MD simulations with Plm II as

target enzyme. Furthermore, we report computational

predictions of the binding conformations for these

unsymmetrical inhibitors.

Results

Chemistry. The synthetic routes to the symmetrical

(3 and 4) and to the unsymmetrical (5-13 and 16)

diacylhydrazine inhibitors as well as to the 1,3,4-

oxadiazole inhibitors (symmetrical 17 and 18 and

unsymmetrical 19-24) are outlined in Scheme 1. The

vinyl bromide fragments were intended to serve as

suitable handles for further transformations. The pre-

cursor bislactones 26 and 27 were prepared as described

previously.²⁶To produce the symmetrical diacylhydra-

zines, 3 and 4, an excess of hydrazide was used to open

the bislactone rings of 27. Subsequent dehydration of

the corresponding diacylhydrazine using Burgess re-

agent⁴⁰delivered the symmetrical 1,3,4-oxadiazoles 17

and 18. To avoid reaction between Burgess reagent and

the hydroxyl groups, protection with chlorotrimethyl-

silane was first performed and the cyclodehydration was

thereafter accomplished with Burgess reagent. Depro-

tection using potassium fluoride gave the 1,3,4-oxadia-

zoles (17 and 18). The unsymmetrical compounds were

easily available from the monolactone intermediates 28

and 29 (Scheme 1). Mono opening of the bislactones 26

or 27 with (1S,2R)-1-amino-2-indanol could be achieved

essentially as previously reported for a benzyloxy ana-

logue utilizing 2-hydroxypyridine as catalyst to deliver

the monolactones 28 and 29.⁴¹Ring opening of the

monolactone (28 or 29) using an excess of hydrazide

gave the unsymmetrical diacylhydrazines 5-13 and 16.

The unsymmetrical 1,3,4-oxadiazoles 19-24 were then

obtained by a similar procedure as used for the prepara-

tion of the symmetrical 1,3,4-oxadiazoles. Thus, the

alcohols were first protected with chlorotrimethylsilane

and subsequent cyclodehydration with Burgess reagent

and deprotection delivered the unsymmetrical 1,3,4-

oxadiazoles 19-24 in moderate to good yields over three

steps (20-70%).

The 1,2,4-triazole 25 was generated in two steps from

the monolactone intermediate 29 (Scheme 2). Nucleo-

philic ring opening of the monolactone with hydrazine

hydrate furnished the corresponding hydrazide 30.

Condensation of the hydrazide 30 with acetamidine, to

the amidrazone (not isolated), followed by thermal

cyclization provided eventually the 1,2,4-triazole 25.

For the preparation of the unsymmetrical diacylhy-

drazine analogues 14 and 15 with extended P1/P1’ arms,

microwave-assisted Suzuki couplings were conducted

(Scheme 3).^42,43The (E)-bromo diacylhydrazine precur-

sor 9 was reacted with an excess of the appropriate

organoboronic acid, a catalytic amount of Pd₂(dba)₃/

[(t-Bu)₃PH]BF₄,⁴⁴and Ba(OH)₂as a base. Microwave

irradiation at 110 °C for 15 min furnished the (E)-

methylenedioxyphenyl compound 14. With the less

reactive electron-poor acetylphenylboronic acid as re-

actant the reaction time had to be prolonged to 40 min

to obtain the (E)-acetylphenyl compound 15.

Although a small optimization of the reaction protocol

was made, 14 and 15 were both isolated in low yields

6092 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 19

Ersmark et al.

Scheme 1^a

^a(a) R₂CONHNH₂, CH₂Cl₂, reflux; (b) i. TMSCl, NEt₃, CH₂Cl₂ii. Burgess reagent, THF, reflux iii. KF‚2H₂O; (c) (1S,2R)-1-amino-2-

indanol, 2-hydroxypyridine, CH₂Cl₂(d) R₂CONHNH₂, CH₂Cl₂, reflux; (e) i. TMSCl, NEt₃, CH₂Cl₂ii. Burgess reagent, THF, reflux iii.

KF‚2H₂O.

Scheme 2^a

Scheme 3^a

^a(a) NH₂NH₂‚H₂O, EtOH (b) CH₃CNHNH₂, EtOH, reflux.

(23% and 22%, respectively). Initial attempts with Pd-

(PPh₃)₄as catalyst and Na₂CO₃as a base were unsuc-

cessful in producing full consumption of the starting

material and significant mono- and di-dehalogenation

occurred as deduced from analytical RP-LC-MS. The

change of base to Ba(OH)₂× 8H₂O increased the

reaction rate and produced less dehalogenation. How-

ever, the chromatographic purification was found very

difficult, partly due to triphenylphosphine oxide con-

taminations. This problem was circumvented by the use

of a different catalyst system, (Pd₂(dba)₃/[(t-Bu)₃PH]-

BF₄), exhibiting a comparable catalytic activity.

^a(a) RB(OH)₂, Pd₂(dba)₃/[(t-Bu)₃PH]BF₄, Ba(OH)₂‚8H₂O, DME,

H₂O, EtOH, microwave irradiation.

symmetrical diamides 1 and 2, both previously pre-

pared,²⁶were included as reference compounds. Overall,

the symmetrical inhibitors (3 and 4 and 17 and 18) were

less potent than the unsymmetrical (5-16 and 19-25).

Among the symmetrical compounds only the diacylhy-

drazine with a 4-tert-butylphenyl 4 demonstrated any

affinity in the Plm I and II assays. A comparison

between the two amide bond replacing groups, the

extended diacylhydrazine and the compact 1,3,4-oxa-

diazole, revealed that the diacylhydrazine is far superior

over the oxadiazole. The 1,3,4-oxadiazoles exhibited K_i

values, at the best in the micromolar range, and only

one of these compounds, 24, was active in the Plm I

Enzyme Inhibition. The inhibitory activities and

the selectivity of the compounds 3-25 were determined

in Plm I, Plm II, and Cat D enzyme assays, and the

results are summarized as K_ivalues in Table 1. The two

Malarial Plasmepsin Inhibitors

Journal of Medicinal Chemistry, 2005, Vol. 48, No. 19 6093

Table 1. Plasmepsin Inhibitory Activity of the Prepared Compounds Encompassing Amide Bond Replacements

6094 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 19

Table 1 (Continued)

Ersmark et al.

Malarial Plasmepsin Inhibitors

Journal of Medicinal Chemistry, 2005, Vol. 48, No. 19 6095

Table 1 (Continued)

6096 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 19

Table 1 (Continued)

Ersmark et al.

^and, not determined. ^bK_iPlm IV, >5000 nM. ^cData from ref 26. ^dData from ref 27. ^eK_iPlm IV, 478 nM. ^fK_iPlm IV, 35 nM. Inhibition

on P. falciparum infected red blood cells; 25% inhibition at 4 µM.

assay, while several of the diacylhydrazines demon-

strated K_ivalues in the low nanomolar range in both

Plm I and II assays. The 1,2,4-triazole compound 25 was

employed as a reference compound encompassing in

contrast to 19, a five-membered heterocycle with a

hydrogen bond donating NH. However, incorporation of

this heterocycle rendered a compound inactive in all

enzyme assays assessed, as was the case with 19.

The importance of the size of the P2′ side chain is

demonstrated in Table 1. Thus, replacing a methyl (5)

by a phenyl (6) and a 4-tert-butylphenyl group (7)

generated consistently more potent unsymmetrical dia-

cylhydrazine Plm I and II inhibitors. Among the un-

symmetrical oxadiazoles with these three P2′ side

chains, only the phenyl derivative 20 exhibited any

inhibitory activity for Plm II (cf. 19, 20, and 21).

Introducing an extended spacer arm and flexibility (8-

10 and 22-24) further improved the inhibitory potency.

The effect was most pronounced with the diacylhydra-

zines (8-10), but the inhibitors with a 1,3,4-oxadiazole

were still weak (22-24). A two-carbon extension (9)

between the diacylhydrazine and a phenyl seemed

optimal and gave rise to potencies three times higher

in the Plm I assay and almost seven times higher in

the Plm II assay (K_i) 9 nM and 7 nM, respectively),

compared to the parent compound 2 (K_i) 27 nM and

47 nM, respectively). Saturated basic nitrogen ring

systems at the same distance to the diacylhydrazine (the

morpholine 12 and the piperazine 13) resulted in a

substantial loss in affinity.

except the oxadiazole 24, were more active in the Plm

II than in the Plm I assay.

Three of the compounds (1, 4, and 18) were examined

in a Plm IV assay, and the K_ivalues are reported as

footnotes to Table 1. The allyloxy inhibitor 1 was found

inactive, but the diacylhydrazine 4 and the 1,3,4-

oxadiazole 18 exhibited a much higher potency in the

Plm IV assay than what was observed in the Plm I and

II assays (Plm IV K_i) 478 nM and 35 nM, respectively).

Notably, the Plm IV selective symmetrical nonpeptidic

inhibitor 18 demonstrated a 25% inhibition of parasite

growth in infected erythrocytes at 4 µM.

In general, the inhibitors did not exhibit Cat D

binding, although the symmetrical diacylhydrazine 4

and the two phenethyl diazylhydrazine inhibitors, 14

and 15, demonstrated a weak binding to the mam-

malian enzyme (K_i) 3000 nM, 800 nM, and 1200 nM,

respectively).

Molecular Dynamics Simulations and Free En-

ergy Calculations. The binding of seven of the syn-

thesized compounds and the previously prepared di-

amide 2 were analyzed by computer simulation meth-

ods. Binding free energies of 2, 5, 9, 14-16, 19 and 23

to Plm II were estimated with the linear interaction

energy (LIE) method⁴⁵and complementary docking

studies were carried out with Autodock.⁴⁶

As previously reported with similar compounds,²⁷

rigid receptor automated docking with Autodock is

problematic due to the flexible nature of the Plm II

binding site. To solve the problem of getting reasonable

starting structures for molecular dynamics (MD) simu-

lations, the ligands were superimposed onto the scaffold

of compound 1 as modeled in an earlier study.²⁷The

template 1 is C₂-symmetric whereas ligands 5, 9, 14-

16, 19 and 23 are only equivalent with regard to the

P1/P1′ positions. Two LIE calculations for each ligand

were thus required to determine the favored binding

mode, one with the diacylhydrazine or oxadiazole moiety

on the prime side and one with these linkers on the

nonprime side. A summary of the calculated data is

Surprisingly, an elongation of the P1/P1′ side chains

did not render the same positive impact as previously

observed in a series of related C₂-symmetric inhibitors

(cf. 9 with 14-15).²⁶Hence, the extended methylene-

dioxyphenyl 14 demonstrated almost the same potency

as the corresponding bis(vinyl bromide) 9, and the

acetylphenyl extension in 15 gave a significantly lower

activity. A shorter P1 and P1′ side chain, as in the

allyloxy derivative 16, provided a less potent inhibitor.

All P1/P1′ bis(vinyl bromide)-containing compounds,

Malarial Plasmepsin Inhibitors

Journal of Medicinal Chemistry, 2005, Vol. 48, No. 19 6097

Table 2. Experimental and Calculated Energetics of Inhibition of Plm II

ligand-surrounding interactions (kcal/mol)^c

vdw

l-s

el

l-s

vdw

l-s

el

l-s

compound^a

∆G_{bind, exp}(kcal/mol)^b

∆G_{bind, calc}(kcal/mol)

V

p

w

1^d

-9.6

-10.1

-8.9

-11.2

-10.9

-9.4

-9.3

-7.3

-7.6

-9.1 ( 0.4

-9.4 ( 1.0

-7.7 ( 0.7

-8.4 ( 0.8

-8.2 ( 0.9

-10.2 ( 1.1

-9.0 ( 1.0

-10.6 ( 1.3

-10.1 ( 0.8

-10.9 ( 1.0

-8.6 ( 0.4

-9.7 ( 0.7

-4.8 ( 0.6

-6.9 ( 0.5

-7.2 ( 0.5

-8.4 ( 0.2

-85.9 ( 0.1

-102.6 ( 2.6

-90.6 ( 1.3

-90.4 ( 0.6

-105.0 ( 2.0

-108.2 ( 0.7

-128.6 ( 0.7

-128.5 ( 1.1

-123.3 ( 0.6

-127.5 ( 0.5

-91.8 ( 0.1

-94.8 ( 0.7

-85.9 ( 1.4

-87.4 ( 0.0

-101.3 ( 0.5

-103.3 ( 0.1

-67.2 ( 0.1

-74.0 ( 0.4

-72.0 ( 0.2

-74.2 ( 1.1

-68.8 ( 0.3

-73.3 ( 1.9

-70.9 ( 0.7

-75.7 ( 1.2

-83.2 ( 0.5

-83.4 ( 0.9

-70.6 ( 0.4

-72.3 ( 0.8

-67.2 ( 0.6

-72.7 ( 1.1

-70.7 ( 1.0

-73.4 ( 0.3

-40.6 ( 0.1

-49.6 ( 0.9

-46.4 ( 0.2

-56.9 ( 0.7

-71.1 ( 0.5

-66.4 ( 1.6

-47.1 ( 0.1

-47.6 ( 0.4

-55.7 ( 0.3

-64.5 ( 0.9

-74.7 ( 0.7

-72.9 ( 1.0

-70.5 ( 0.9

-75.1 ( 1.8

-84.0 ( 0.9

-69.2 ( 0.9

-73.8 ( 0.2

-74.0 ( 0.1

2

5_conf.1

5_conf.2

9_conf.1

9_conf.2

14_conf.1

14_conf.2

15_conf.1

15_conf.2

16_conf.1

16_conf.2

19_conf.1

19_conf.2

23_conf.1

23_conf.2

^aconf.1 and conf.2 indicate the different orientations of the inhibitor in the active site. In conformation 1 the diacylhydrazine or

oxadiazole linker is on the nonprime side while in conformation 2 the diacylhydrazine or oxadiazole linker is on the prime side. ^bThe

0

experimental binding free energy calculated from experimentally determined K_is using ∆G

) RT ln K_i. ^cThe calculated average

electrostatic V_eland nonpolar V_vdWenergies for ligand-surrounding (l-s) interactions. The s^buⁱⁿb^d,es^xc^pripts p and w denote simulations of the

ligand in complex with the protein and free in water, respectively. ^dData from ref 27.

shown together with experimental data in Tables 1 and

2. It can be seen that both the absolute and relative

binding free energies are rather well predicted by the

simulations, and, in particular, one finds that the

ranking with respect to the (minimal) P2′ motif is

correct, by comparing 2, 5 and 19. Furthermore, the

favorable effect of the bromine substituent at the P1/

P1′ position (2 vs 1 and 9 vs 16) is also reproduced by

the calculations.

Examination of how the electrostatic and hydrophobic

contributions to the absolute binding affinity vary

among the ligands reveals that the nonpolar (hydro-

phobic) term makes the dominating contribution for all

the analyzed ligands. Furthermore, when comparing the

two different binding directions we find that the most

favored one, for all studied ligands, is with the diacyl-

hydrazine/oxadiazole oriented to the prime side. The

data in Table 2 also reveal that in all cases, except for

15, the polar contribution is the main determinant of

the favored binding direction. This is in accordance with

the general notion that while hydrophobic binding

provides most of the overall affinity, the polar interac-

tions such as H-bonds confer specificity in enzyme-

inihibitor complexes.^27,47,48

In the MD simulations of the ligand complexes the

central scaffold (i.e. the six central atoms originating

from mannitol) of the ligands adopts the same confor-

mation as observed earlier.²⁶Hence, the hydroxyl group

adjacent to P1′ makes the crucial transition state mimic

interaction with the two active aspartic acids Asp34 and

Asp214. The second central hydroxyl makes a very

strong anionic H-bond interaction with Asp214 and also

interacts with an active site water as found earlier.²⁶

The P1 group of the simulated ligands resides in the

S1 pocket of Plm II. The larger P1 substituents in 14

and 15 stretch according to the simulation through the

entire S1-S3 site, exposing the outermost part of the

P1 group to the solvent. The P1′ substituent reaches into

the S1′ site of the protein and causes a slight expansion

of S1′ by movements of the Phe294 side chain.

Figure 1. Surface representation of PlmII in complex with

14: (A) Snapshot from MD simulations with the S2′ subsite

closed as in the starting structure and (B) after the S2′ cavity

has opened. The snug fit of the P2′ phenyl group in the pocket

can be seen, as well as the accompanying conformational

change of the diacylhydrazine fragment that is necessary for

the stable H-bond network seen in Figure 2.

the S2′ pocket to fully accommodate the bulky phenyl

moiety (Figure 1). The phenyl ring of 14 enters the

expanded pocket after a few hundred picoseconds and

stays there for the rest of the simulation (>2 ns). The

observed flexibility in S2′ is found to originate from

movements of Met75. The backbone of the protein is not

displaced significantly but the methionine side chain is

rotated toward the interior of the enzyme, thereby

opening S2′ enough for the phenyl group to enter. It is

notable that the conformational change of Met75 seen

in the MD simulations is supported by experimental

During simulation of 14 with the diacylhydrazine

linker on the prime side we observed an expansion of

6098 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 19

Ersmark et al.

tion required for these hydrogen bonds, and for 5 the

terminal interaction with the hydroxyl of Tyr192 is

completely lost.

The opposite binding direction was investigated as

well for 5, 9, and 14-16, i.e., with the diacylhydrazine

linker on the nonprime side. The MD simulations show

that the phenyl fragment cannot be fully accommodated

in the S2 pocket and remains solvent exposed during

these simulations where its conformation fluctuates

substantially. In this binding mode the diacylhydrazine

mainly interacts with H-bond donors/acceptors from the

protein backbone, namely the Ser79 and Ser218 amide

nitrogens and the carbonyl of Gly216, but these inter-

actions are less stable due to the mobility of the phenyl

group.

In compounds 19 and 23 the diacylhydrazine is

replaced by the oxadiazole ring. Both the ring oxygen

and the ring nitrogens may act as hydrogen bond

acceptors, while no donor function is present in contrast

to the diacylhydrazines. Compounds 19 and 23 did not

show the same conformational stability during MD as

the other compounds examined. In the orientation with

the oxadiazole on the prime side, one of the two

nitrogens in the ring makes an H-bond to the backbone

nitrogen of Val78. The other nitrogen is exposed to

solvent, and the oxygen is forced into an unfavorable

orientation with respect to the Gly36 carbonyl group,

which destabilizes the oxadiazole compared to the

diacylhydrazine substituent. Furthermore the oxadia-

zole linker to the phenyl group is shorter in 23, making

it harder for the phenyl group to reach into the S2′

pocket. When the oxadiazole instead is located on the

nonprime side, the heterocyclic nitrogens form close

interactions with the amide hydrogen of Ser79, forcing

the ring oxygen to make unfavorable interactions with

the carbonyl oxygen of Gly216 in a manner similar to

the other conformation.

With the diacylhydrazine or heterocyclic fragment of

the investigated ligands located to the prime side, the

indanol group at the other end of the ligands is located

in the S2 subsite. There it makes hydrophobic contact

with the residues that form the S2 pocket, Val78,

Thr217, Ala219, Ile290 and Ile292, as found earlier.²⁷

The hydroxyl group of the indanol can either hydrogen

bond to the Ser218 side chain or make contact with the

solution. In the mirrored conformation, the indanol is

accommodated in S2′, but does not stretch all the way

to Met75 mentioned above. Instead it stays close to the

flap and makes hydrophobic contact only to the flap

residues Tyr77 and Val78, thereby leaving a large

portion of the aliphatic indan ring solvent exposed. The

hydroxyl group then fluctuates between interactions

with the amide oxygen of Asn76 and an internal H-bond

to the adjacent carbonyl in the ligand, as observed

earlier.²⁷

Figure 2. Comparison of H-bond network between Plm II and

three ligands: (A) pepstatin A, (B) 9, and (C) 23. Plm II is

shown in green and the ligands in yellow. Dashed yellow lines

depict H-bonds, and it can be noted that the four H-bonds

made with 9 matches the pattern of pepstatin A, while

compound 23 loses several H-bonds and also has an unfavor-

able interaction between the carbonyl of Gly36 and the

oxadiazole oxygen.

X-ray structures of Plm II complexes with ligands

having large P2′ groups (1LEE, 1LF2).³³

The ability of MD simulation to elicit the opening of

the S2′ pocket prompted us to also examine this

conformation in combination with the other compounds.

The other diacylhydrazine containing compounds 9, 15

and 16, also showed highly increased stability in the

P2′ region when the pocket was enlarged. The adopted

conformation of the P2′ group allows for strong H-bond

interactions between the diacylhydrazine spacer and

Gly36, Asn76, Val78 and Tyr192, as seen in Figure 2.

The predicted binding conformation also allows the

phenyl group of the ligand to reside in a predominantly

hydrophobic environment in the S2′ pocket formed by

Met75, Tyr77, Leu131 and Ile133.

Complementary docking of ligand 16 was performed

with Autodock to investigate if the binding mode found

here could be reproduced when the Met75 side chain

was rotated manually in the Plm II structure. The

rotation was made by shifting the torsion Câ-Cγ-Sδ-

Cꢀ 180°, thus causing a substantial movement of the

Cꢀ atom resulting in an expansion of the S2′ pocket.

When docking was done with the 1SME¹⁹structure with

the side chain of Met75 rotated, we still found no

Compound 5 is similar to 9 but lacks the terminal

benzyl group in P2′, and the effect of this is that the

hydrogen bonds between the diacylhydrazine and the

protein are somewhat weakened. The phenyl moiety of

9 bound in S2′ thus appears to stabilize the conforma-

Malarial Plasmepsin Inhibitors

Journal of Medicinal Chemistry, 2005, Vol. 48, No. 19 6099

reasonable solutions. However, when the structure

1LF2³³was used, where the S2′ pocket is in the open

conformation, several good solutions were found. Clus-

tering the 150 independent solutions gave a small

cluster of 2 and a larger of 22 with low docking energy

score. The cluster with only 2 members had the diacyl-

hydrazine in the nonprime position, and the larger

cluster had the diacylhydrazine on the prime side. In

the other structure denoted 1LF3,³⁹the side chain of

Met75 has the same conformation as in 1SME, i.e., S2′

is closed, but the active site of PlmII is generally more

open than in 1SME. Docking was also performed in

1LF3 without rotating the Met75 side chain. The

outcome was clustered as earlier, resulting in 2 clusters

of 29 and 18 solutions, respectively. Both clusters had

the diacylhydrazine on the prime side differing es-

sentially only in the position of the terminal phenyl

group. When the Met75 side chain was rotated we found

3 clusters containing 36, 3 and 17 solutions, respectively.

The cluster of 36 solutions had the diacylhydrazine

spacer on the prime side, and so did the cluster of 17.

The larger cluster of the two had only few solutions in

which the phenyl was docked in the open S2′, but in

the latter a majority had the phenyl enclosed by the S2′.

The small cluster of three solutions had the opposite

conformation, i.e., the diacylhydrazine was found in the

nonprime position. These calculations thus further

support the conformations found by MD, but also

illustrate how automated rigid-receptor docking can be

sensitive to protein conformation, especially when bind-

ing pockets are partially collapsed.

the obtained binding directions of the two new amide

replacements provided more detailed predictions re-

garding structure-activity relationships for these types

of compounds.

For all of the unsymmetrical inhibitors, the most

favorable binding direction predicted by the MD simula-

tions was with the diacylhydrazine/oxadiazole spacer on

the prime side rather than on the nonprime side, as

depicted in Chart 1. In practically all cases polar

interactions were more favorable for this pose than the

reverse one. For compounds with larger P2′ substitu-

ents, the capability of S2′ to open up a large hydrophobic

pocket appears essential for accommodating this binding

mode, as illustrated in Figure 1. The flexibility in S2′

mainly arises from different conformations of the Met75

side chain, and the ability of MD simulations to capture

the opening of S2′ is a pertinent example of induced fit

between ligand and receptor. Furthermore, it is quite

remarkable how well the protein can adopt to this large

phenethyl P2′ group by essentially only one conforma-

tional change of a single side chain. The predicted

protein conformational change is also supported by

experimental data on the complex of Plm II with the

inhibitor rs370, where a bulky 4-aminophenyl P2′

substituent occupies the same space as the phenyl

group.³³Both the calculated binding affinities and

complementary docking predicted the same favored

binding mode. The diacylhydrazine element is able to

establish a rigid hydrogen bond network involving the

backbone of Gly36, Asn76, Val78 and the Tyr192 side

chain in the active site of Plm II, which mimics the

network formed by pepstatin A in complex with Plm II

(Figure 2). It is reasonable to expect that such an

H-bond pattern also occur with the natural peptide

substrate. The heterocyclic oxadiazole in compound 23

functions as a shorter linker to the phenyl group in the

P2/P2′ position than the diacylhydrazine linker. Even

with an extra carbon atom, as in the phenylpropyl chain

of 24, the phenyl ring is not likely to be as stable as in

the diacylhydrazine case since the oxadiazole is found

to be unable to make the rigid H-bond pattern observed

with the diacylhydrazine. Even if 19 and 23 are less

potent inhibitors the calculated binding affinities for 19

and 23 still favor the binding mode with the oxadiazole

on the prime side, consistent with the diacylhydrazine

compounds. The affinity difference between the oxadia-

zole and diacylhydrazine inhibitors can best be under-

stood by comparing 5 and 19 where the only difference

is the amide bond replacement. Analysis of the differ-

ence in nonbonded interactions in 5 and 19 (Figure 3)

clearly shows that the largest difference involves an

unfavorable interaction of the oxadiazole oxygen of 19

with the carbonyl oxygen of Gly36 while the diacyl-

hydrazine in 5 forms a hydrogen bond with the same

Gly36 carbonyl.

Discussion

We have previously identified highly potent inhibitors

of Plm I and II comprising a C₂-symmetric 1,2-dihy-

droxyethylene scaffold.^26,27With respect to the direction

of the P2 and P2′ side chains in these inhibitors (1 and

2),²⁷as deduced from molecular modeling a diacylhy-

drazine element (3-16) or a 1,3,4-oxadiazole (17-24)

should constitute proper amide bond replacements. The

effect of replacing the amino indanols, previously em-

ployed both as P2 and P2′ side chains, were simulta-

neously examined. As demonstrated in Table 1, several

potent inhibitors of both Plm I and II with K_ivalues in

the low nanomolar range were identified (9-11 and 14).

The symmetrical compounds (3 and 4, and 17 and 18),

in which both of the amide bonds in the backbone as

well as the indanol had been replaced, were generally

significantly less active in the Plm I and II assays as

compared to the unsymmetrical variants.

With the introduced asymmetry in the P2/P2′ posi-

tions, predictions regarding possible binding modes of

the new substituents are important to further ligand

development. Automated docking, as a tool for deter-

mining binding modes, is less reliable when a high

degree of torsional freedom is present in the ligands,

especially in combination with a flexible receptor as Plm

II. As previously, we attempted to overcome this prob-

lem by using MD to examine the binding of the new

ligands to Plm II.^26,27The average structures obtained

from the simulations did not offer any differences

regarding the central parts of the ligands (the two

central hydroxyls and the P1/P1′ groups) from the

earlier studies on symmetrical inhibitors.^26,27However,

The somewhat overpredicted affinity for 15 (K_i^calc

)

12 vs K^e_i^xpt) 145 nM) was found to originate from a

hydrophobic collapse of the ligand in the simulated

unbound state. This phenomenon was examined in an

additional simulation of 15 employing ligand restraints

in water to prevent hydrophobic collapse, and the

corresponding results (K^c_i^alc) 50 nM) show better

agreement with experimental result. The only term

affected in the water simulation turns out to be the

6100 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 19

Ersmark et al.

Conclusion

Potent Plm I and II inhibitors derived from L-

mannitol and characterized by either a diacylhydrazine

or a five-membered oxadiazole ring replacing the prime

side amide have been designed and synthesized. MD

simulations were applied in the design process. The

computationally predicted Plm II K_ivalues were gener-

ally in excellent agreement with the biological results

and the predicted bioactive conformations were sup-

ported by crystal structures of other inhibitors. The

diacylhydrazine was found to be superior over the

oxadiazole as amide bond replacement in Plm I and II

inhibitors. According to the computationally predicted

conformation this was primarily attributed to the dia-

cylhydrazine being able to form an extensive hydrogen

bond network, similar to pepstatin A, to the active site.

Moreover, an extensive flexibility of the S2′ pocket was

captured by the simulations, predicting the binding

mode of the unsymmetrical inhibitors to be with the

amide bond replacement orientated to the prime side,

as illustrated in Chart 1.

All inhibitors prepared display a unique selectivity

over the human Cat D. One of the symmetrical com-

pounds with both of the two amide bonds replaced by

oxadiazole rings was found to be strongly active toward

Plm IV. This Plm IV selective compound (18) was also

effective in culture inhibiting infected erythrocytes and

could possibly serve as a lead for further optimization.

Actually, this inhibitor, although with a relatively high

molecular weight (Mw 760), was totally devoid of amide

bonds and represents, to the best of our knowledge, the

so far most potent nonpeptidic inhibitor of Plm IV

disclosed (K_i) 35 nM).

Figure 3. Plot of the difference in average nonbonded

interactions between compound 5 and 19 to each Plm II

residue (a negative value corresponds to stronger interactions

of ligand 5 with the specific amino acid). The interactions are

averaged over 1 ns and separated into van der Waals and

electrostatic terms. The negative electrostatic value for Gly36

clearly shows the more favorable interaction (H-bond) of 5,

which in the case of 19 is replaced by an oxygen-oxygen

repulsion.

nonpolar one, while the electrostatic contribution is

unaffected by the restraints. When similar restraints

were applied to other ligands, the effect on the energet-

ics was insignificant. We therefore suspect that the

observed hydrophobic collapse may be related to the

recently reported finding that several current force

fields, including the one used here, has slightly too

favorable nonpolar interactions.^49,50This overstabilizes

the “folded” state of peptide- or polymer-like molecules

and, in reality, the tendency to start packing hydropho-

bic groups should probably occur somewhat later on the

molecular size-scale.

Within the unsymmetrical diacylhydrazine series of

bis(vinyl bromides) (5-13), inhibitors with higher af-

finity to both Plm I and II compared to the parent

compound 2 were encountered (e.g. 9, 10, and 11).

Unfortunately, an extension of the P1 and P1′ side chain

did not result in the same increase in activity as

previously reported for the C₂-symmetrical diamides.

Experimental Section

Chemistry. General Information. All microwave reac-

tions were conducted in heavy-walled glass Smith process vials

sealed with aluminum crimp caps fitted with a silicon septum.

The microwave heating was performed in a Smith Synthesizer

single-mode microwave cavity producing continuous irradia-

tion at 2450 MHz (Biotage AB, Uppsala, Sweden). Reaction

mixtures were stirred with a magnetic stirring bar during the

irradiation. The temperature, pressure and irradiation power

were monitored during the course of the reaction. After

completed irradiation, the reaction tube was cooled with high-

pressure air until the temperature had fallen below 39 °C. The

hydrazides used for the synthesis of the compounds 3-8 and

11 were purchased from Sigma-Aldrich. ¹H and ¹³C NMR

spectra were recorded on a JEOL JNM-EX 270 spectrometer

at 270.2 and 67.8 MHz, respectively or on a JEOL JNM-EX400

spectrometer at 399.8 and 100.5 MHz, respectively. HMBC and

HSQC spectra were recorded on a Varian UNITY INOVA

spectrometer (¹H at 499.9 MHz). Chemical shifts are reported

as δ values (ppm) indirectly referenced to TMS via the solvent

residual signal. Optical rotations were obtained on a Perkin-

Elmer 241 polarimeter. Specific rotations ([R]_D) are reported

in deg/dm, and the concentration (c) is given in g/100 mL in

the specified solvent. Elemental analyses were performed by

Analytische Laboratorien, Lindlar, Germany. Flash column

chromatography was performed on Merck silica gel 60, 0.04-

0.063 mm. Thin-layer chromatography was performed using

aluminum sheets precoated with silica gel 60 F₂₅₄(0.2 mm; E.

Merck, eluent 0-10% MeOH in CHCl₃) and visualized with

UV light and ninhydrin. Analytical RP-LC-MS was performed

on a Gilson HPLC system with a Chromolith Performance RP-

18e 100 × 4.6 mm (Merck KGaA) column, with a Finnigan

AQA quadropole mass spectrometer, at a flow rate of 4 mL/

min, using different gradients of CH₃CN in 0.05% aqueous

Notably, in the Plm IV assay the opposite relative

activity between the oxadiazole and the diacylhydrazine

series was observed, based on the activities of two

inhibitors (4 and 18). The 1,3,4-oxadiazole 18 demon-

strated a significantly higher inhibitory activity than

the diazylhydrazine 4. Whether this behavior is due to

the different nature of the flap loop in the two proteins,

with the more flexible Gly in Plm IV instead of the Val

at position 78 of Plm II, is currently being investigated

by computational methods. The fact that, the oxadiazole

18 that is Plm IV selective and displayed inhibition of

parasite growth in cell culture suggests that other

plasmepsins than Plm I/Plm II might be essential for

the parasite replication, as has previously been pro-

posed.^13,51,52In fact Plm IV is the ortholog of the only

food vacuole plasmepsin found in the three remaining

human Plasmodium species, P. vivax, P. malaria, and

P. ovale, making it a potential target for multi species

antimalarials.^13,52In this context, it is important to

emphasize that recent studies of P. falciparum plasme-

psin knock-out clones implies the significance of inhibit-

ing several of the food vacuole plasmepsins in order to

efficiently kill the parasite.^53,54

Malarial Plasmepsin Inhibitors

Journal of Medicinal Chemistry, 2005, Vol. 48, No. 19 6101

formic acid. Preparative RP-LC-MS was performed on Gilson

HPLC system with a Zorbax SB-C8, 5µm 21.2 × 150 mm

(Agilent Technologies) column, with a Finnigan AQA quadro-

pole mass spectrometer, at a flow rate of 15 mL/min.

as a white solid: [R]²³_D+39.0° (c ) 0.96, MeOH). Anal. (C₂₈H₃₁-

Br₂N₃O₈) C, H, N.

(2R,3R,4R,5R)-2,5-Bis[(2E)-3-bromo-2-propenyloxy]-6-

[N’-(4-tert-butylbenzoyl)hydrazino]-3,4-dihydroxy-N1-

[(1S,2R)-2-hydroxyindan-1-yl]-6-oxo-hexanamide (7). Com-

pound 7 was prepared according to general procedure I using

monolactone 29 (39 mg, 0.070 mmol) and 4-tert-butylbenzoic

hydrazide (69 mg, 0.36 mmol) dissolved in CH₂Cl₂(1.5 mL).

Purification by flash chromatography (0-10% MeOH in CHCl₃)

and RP-LC-MS (30 min gradient of 30-70% CH₃CN in 0.05%

aqueous formic acid) gave 7 (36 mg, 69%) as a white solid:

[R]²³_D+9.2° (c ) 1.14, CHCl₃). Anal. (C₃₂H₃₉Br₂N₃O₈) C, H, N.

(2R,3R,4R,5R)-2,5-Bis[(2E)-3-bromo-2-propenyloxy]-

3,4-dihydroxy-N1-[(1S,2R)-2-hydroxyindan-1-yl]-6-oxo-6-

[N’-(2-phenylethanoyl)hydrazino]-hexanamide (8). Com-

pound 8 was prepared according to general procedure I using

monolactone 29 (95 mg, 0.17 mmol) and phenylacetic hy-

drazide (255 mg, 1.7 mmol) dissolved in CH₂Cl₂(2 mL).

Purification by flash chromatography (0-6% MeOH in CHCl₃)

gave 8 (92 mg, 76%) as a white solid: [R]²³_D+30.1° (c ) 1.25,

MeOH); Anal. (C₂₉H₃₃Br₂N₃O₈) C, H, N.

(2R)-1N-(2-Hydroxyindan-1-yl)-2-(2-propenyloxy)-2-

[(2R,3R,4R)-4-(2-propenyloxy)-3-hydroxy-5-oxo-tetrahy-

drofuran-2-yl]-ethanamide (28). The bislactone 26 (78 mg,

0.31 mmol) and 2-hydroxypyridine (29 mg, 0.30 mmol) were

dissolved in dry CH₂Cl₂(5 mL) and stirred at room tempera-

ture for 30 min. The solution was cooled to 0 °C, and (1S,2R)-

1-amino-2-indanol (46 mg, 0,31 mmol) was added rapidly. The

reaction mixture was allowed to attain room temperature and

stirred overnight for 17 h. The reaction was monitored by TLC

and analytical RP-LC-MS. The solvent was removed under

reduced pressure, and subsequent purification by flash chro-

matography (CHCl₃) gave 28 (52 mg, 42%) as a white solid:

[R]²⁰-31.0° (c ) 1.00, CH₃CN). Anal. (C₂₁H₂₅NO₇) C, H, N.

D

(2R)-2-[(2E)-3-Bromo-2-propenyloxy]-2-{(2R,3R,4R)-4-

[(2E)-3-bromo-2-propenyloxy]-3-hydroxy-5-oxo-tetrahy-

drofuran-2-yl}-1N-(2-hydroxyindan-1-yl)-ethanamide (29).

The bislactone 27 (400 mg, 0.97 mmol) and 2-hydroxypyridine

(92 mg, 0.97 mmol) were dissolved in dry CH₂Cl₂(20 mL) and

stirred at room temperature for 30 min. The solution was

cooled to 0 °C, and (1S,2R)-1-amino-2-indanol (144 mg, 0,97

mmol) was added rapidly. The reaction mixture was allowed

to attain room temperature and stirred overnight for 15 h. The

reaction was monitored by TLC and analytical RP-LC-MS. The

solvent was removed under reduced pressure, and subsequent

purification by flash chromatography (CHCl₃) gave 29 (238 mg,

44%) as a white solid: [R]²³_D-17.0° (c ) 1.06, CH₃CN). Anal.

(C₂₁H₂₃Br₂NO₇) C, H, N.

(2R,3R,4R,5R)-2,5-Bis[(2E)-3-bromo-2-propenyloxy]-

3,4-dihydroxy-N1-[(1S,2R)-2-hydroxyindan-1-yl]-6-oxo-6-

[N’-(3-phenylpropanoyl)hydrazino]-hexanamide (9). Com-

pound 9 was prepared according to general procedure I using

monolactone 29 (100 mg, 0.18 mmol) and 3-phenyl-propionic

acid hydrazide (prepared according to literature procedure,⁵⁵

246 mg, 1.5 mmol) dissolved in CH₂Cl₂(3 mL). Purification

by flash chromatography (2-5% MeOH in CHCl₃) gave 9 (96

mg, 74%) as a white solid: [R]²³+33.3° (c ) 0.87, MeOH).

D

Anal. (C₃₀H₃₅Br₂N₃O₈) C, H, N.

(2R,3R,4R,5R)-2,5-Bis[(2E)-3-bromo-2-propenyloxy]-

3,4-dihydroxy-N1-[(1S,2R)-2-hydroxyindan-1-yl]-6-oxo-6-

[N′-(4-phenylbutanoyl)hydrazino]-hexanamide (10). Com-

pound 10 was prepared according to general procedure I using

monolactone 29 (99 mg, 0.18 mmol) and 4-phenylbutyric acid

hydrazide (prepared according to literature procedure,⁵⁵301

mg, 1.7 mmol) dissolved in CH₂Cl₂(2 mL). Purification by flash

chromatography (0-5% MeOH in CHCl₃) gave 10 (98 mg, 74%)

as a white solid: [R]²⁰_D+30.3° (c ) 1.00, MeOH). Anal. (C₃₁H₃₇-

Br₂N₃O₈) C, H, N.

(2R,3R,4R,5R)-2,5-Bis[(2E)-3-bromo-2-propenyloxy]-

1,6-(N’-ethanoylhydrazino)-3,4-dihydroxyhexan-1,6-di-

on (3). Bislactone 27 (100 mg, 0.24 mmol) and acetic hydrazide

(130 mg, 1.8 mmol) were dissolved in dry CH₂Cl₂(5 mL), and

the reaction mixture was stirred at reflux for 56 h. The reaction

was monitored by TLC and analytical RP-LC-MS. The solvent

was removed under reduced pressure, and subsequent puri-

fication by flash chromatography (3-10% MeOH in CHCl₃)

gave 3 (41 mg, 30%) as a white solid: [R]²³_D+36.2° (c ) 0.82,

MeOH). Anal. (C₁₆H₂₄Br₂N₄O₈) C, H, N.

(2R,3R,4R,5R)-2,5-Bis[(2E)-3-bromo-2-propenyloxy]-

3,4-dihydroxy-N1-[(1S,2R)-2-hydroxyindan-1-yl]-6-[N’-

(3-1H-indol-3-yl-propanoyl)hydrazino]-6-oxo-hexana-

mide (11). Compound 11 was prepared according to general

procedure I using monolactone 29 (65 mg, 0.12 mmol) and

3-indolepropionoic acid hydrazide (237 mg, 1.2 mmol) dissolved

in CH₂Cl₂(1 mL). Purification by flash chromatography (0-

10% MeOH in CHCl₃) gave 11 (45 mg, 51%) as a white solid:

(2R,3R,4R,5R)-2,5-Bis[(2E)-3-bromo-2-propenyloxy]-

1,6-Bis[N’-(4-tert-butylbenzoyl)hydrazino]-3,4-di-

hydroxyhexan-1,6-dion (4). Bislactone 27 (50 mg, 0.12

mmol) and 4-tert-butylbenzoic hydrazide (165 mg, 0.86 mmol)

were dissolved in dry CH₂Cl₂(3 mL), and the reaction mixture

was stirred at reflux for 64 h. The reaction was monitored by

TLC and RP-LC-MS. The solvent was removed under reduced

pressure, and subsequent purification by flash chromatogra-

phy (0-6% MeOH in CHCl₃) gave 4 (67 mg, 70%) as a white

[R]²⁰+31.5° (c ) 0.99, MeOH). Anal. (C₃₂H₃₆Br₂N₄O₈) C, H,

D

N.

solid: [R]²³+35.2° (c ) 0.49, MeOH). Anal. (C₃₄H₄₄Br₂N₄O₈)

D

(2R,3R,4R,5R)-2,5-Bis[(2E)-3-bromo-2-propenyloxy]-

3,4-dihydroxy-N1-[(1S,2R)-2-hydroxyindan-1-yl]-6-[N’-(3-

morpholin-4-yl-propanoyl)hydrazino]-6-oxo-hexan-

amide (12). Compound 12 was prepared according to general

procedure I using monolactone 29 (30 mg, 0.053 mmol) and

3-(morpholin-4-yl)propionic acid hydrazide (prepared according

to literature procedure,⁵⁶93 mg, 0.54 mmol) dissolved in CH₂-

Cl₂(0.8 mL). Purification by RP-LC-MS (30 min gradient of

5-50% CH₃CN in 0.05% aqueous formic acid), filtration

through basic aluminum oxide 90 (0.063-0.2 mm; E. Merck)

and flash chromatography (3-10% MeOH in CHCl₃) gave 12

(18 mg, 45%) as a white solid: [R]¹⁸_D+28.3° (c ) 0.91, MeOH).

Anal. (C₂₈H₃₈Br₂N₄O₉) C, H, N.

(2R,3R,4R,5R)-2,5-Bis[(2E)-3-bromo-2-propenyloxy]-

3,4-dihydroxy-N1-[(1S,2R)-2-hydroxyindan-1-yl]-6-oxo-6-

{N’-[3-(4-methylpiperazin-1-yl)-propanoyl]hydrazine}-

hexanamide dihydrochloride (13). Compound 13 was

prepared according to general procedure I using monolactone

29 (35 mg, 0.062 mmol) and 3-(4-methylpiperazin-1-yl)propi-

onic acid hydrazide (prepared according to literature proce-

dure,⁵⁷100 mg, 0.54 mmol) dissolved in CH₂Cl₂(0.8 mL).

Purification by RP-LC-MS (30 min gradient of 0-55% CH₃-

CN in 0.05% aqueous formic acid) gave 13 (11 mg, 22%) as

C, H, N.

General Procedure I. Synthesis of the Unsymmetrical

Diacylhydrazines 5-13 and 16. A mixture of the mono-

lactone and hydrazide was refluxed in dry CH₂Cl₂(17-47 h).

The reaction was monitored by TLC and analytical RP-LC-

MS. Purification as indicated gave 5-13 and 16.

(2R,3R,4R,5R)-2,5-Bis[(2E)-3-bromo-2-propenyloxy]-6-

(N’-ethanoylhydrazino)-3,4-dihydroxy-N1-[(1S,2R)-2-hy-

droxyindan-1-yl]-6-oxo-hexanamide (5). Compound 5 was

prepared according to general procedure I using monolactone

29 (85 mg, 0.15 mmol) and acetic hydrazide (115 mg, 1.6 mmol)

dissolved in CH₂Cl₂(2 mL). Purification by flash chromatog-

raphy (0-10% MeOH in CHCl₃) gave 5 (61 mg, 64%) as a white

solid: [R]²³+35.8° (c ) 0.34, MeOH). Anal. (C₂₃H₂₉Br₂N₃O₈)

D

C, H, N.

(2R,3R,4R,5R)-6-(N’-Benzoylhydrazino)-2,5-Bis[(2E)-3-

bromo-2-propenyloxy]-3,4-dihydroxy-N1-[(1S,2R)-2-hy-

droxyindan-1-yl]-6-oxo-hexanamide (6). Compound 6 was

prepared according to general procedure I using monolactone

29 (85 mg, 0.15 mmol) and benzoic hydrazide (209 mg, 1.5

mmol) dissolved in CH₂Cl₂(2 mL). Purification by flash

chromatography (2-8% MeOH in CHCl₃) gave 6 (74 mg, 71%)

6102 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 19

Ersmark et al.

the formate salt. A small portion was transformed to the

MeOH (5 mL) and stirred at room temperature for 18 h. The

reaction was monitored by analytical RP-LC-MS. The solvent

was removed under reduced pressure, and the residue was

partitioned between CHCl₃(6 mL) and H₂O (6 mL). The

aqueous layer was extracted with CHCl₃(2 × 6 mL), and the

combined organic layers were dried (Na₂SO₄), and concen-

trated. Purification by RP-LC-MS (30 min gradient of 90-

100% CH₃CN in 0.05% aqueous formic acid) gave 18 (23 mg,

dihydrochloride salt and analyzed: [R]²⁰+19.9° (c ) 0.47,

D

MeOH). Anal. (C₂₉H₄₃Br₂Cl₂N₅O₈) C, H, N.

(2R,3R,4R,5R)-2,5-Bis(2-propenyloxy)-3,4-dihydroxy-

N1-[(1S,2R)-2-hydroxyindan-1-yl]-6-oxo-6-[N’-(3-phenyl-

propanoyl)hydrazino]-hexanamide (16). Compound 16

was prepared according to general procedure I using mono-

lactone 28 (51 mg, 0.13 mmol) and 3-phenyl-propionic acid

hydrazide (prepared according to literature procedure,⁵⁵210

mg, 1.3 mmol) dissolved in CH₂Cl₂(1.5 mL). Purification by

flash chromatography (0-4% MeOH in CHCl₃) gave 16 (53

58%) as a white solid: [R]²³+81.1° (c ) 0.88, CHCl₃). Anal.

D

(C₃₄H₄₀Br₂N₄O₆) C, H, N.

General Procedure II. Synthesis of the Unsymmetri-

cal 1,3,4-Oxadiazoles 19-24. A mixture of the diacylhydra-

zine, chlorotrimethylsilane, and triethylamine in dry CH₂Cl₂

was stirred at room temperature (16-28h). The reaction was

monitored by TLC and analytical RP-LC-MS. Purification as

indicated gave the TMS-protected diacylhydrazine.

In the subsequent cyclization TMS-protected diacylhydra-

zine and Burgess reagent in dry THF were stirred at reflux

(4-6 h). The reaction was monitored with analytical RP-LC-

MS. The solvent was removed under reduced pressure, and

the crude product was (if nothing else is indicated) deprotected

without further purification. The crude product and KF‚2H₂O

were dissolved in MeOH and stirred at room temperature (14-

22h). The reaction was monitored with RP-LC-MS. Purification

as indicated gave 19-24.

mg, 74%) as a white solid: [R]²¹+45.9° (c ) 1.02, MeOH).

D

Anal. (C₃₀H₃₇N₃O₈) C, H, N.

(1R,2R,3R,4R)-1,4-Bis[(2E)-3-bromo-2-propenyloxy]-

1,4-bis(5-methyl-[1,3,4]oxadiazol-2-yl)-butane-2,3-diol (17).

To a stirred solution of diacyl hydrazine 3 (75 mg, 0.13 mmol)

in dry CH₂Cl₂/THF 1:1 (6 mL) were added chlorotrimethyl-

silane (510 µL, 4.0 mmol) and triethylamine (650 µL, 4.7

mmol). Since the reaction was slow, additional amounts of

chlorotrimethylsilane (510 µL, 4.0 mmol after 2 h) and tri-

ethylamine (650 µL, 4.7 mmol after 2 h) were added. The

reaction mixture was stirred at room temperature for 19 h,

and the reaction was monitored by TLC and analytical RP-

LC-MS. The solvent was removed under reduced pressure, and

subsequent purification by flash chromatography (0-0.5%

MeOH in CHCl₃) gave the TMS-protected diacyl hydrazine (51

(2R,3R,4R,5R)-2,5-Bis[(2E)-3-bromo-2-propenyloxy]-

3,4-dihydroxy-N1-[(1S,2R)-2-hydroxyindan-1-yl]-5-(5-meth-

yl-[1,3,4]oxadiazol-2-yl)-pentanamide (19). Protection of

the diacylhydrazine 5 (49 mg, 0.077 mmol) was accomplished

according to general procedure II using chlorotrimethylsilane

(490 µL, 3.9 mmol) and triethylamine (560 µL, 4.0 mmol)

dissolved in dry CH₂Cl₂(3 mL). Purification by flash chroma-

tography (CHCl₃) gave the TMS-protected diacyl hydrazine (62

mg, 56%) as a white solid: [R]²³-89.4° (c ) 1.04, CHCl₃).

D

Anal. (C₂₂H₄₀Br₂N₄O₈Si₂) C, H, N.

TMS-protected diacyl hydrazine (51 mg, 0.072 mmol) and

Burgess reagent (104 mg, 0.44 mmol) were dissolved in dry

THF (0.5 mL), and the reaction mixture was stirred at reflux

for 3 h. The reaction was monitored by analytical RP-LC-MS.

The solvent was removed under reduced pressure, and the

residue was partitioned between CHCl₃(10 mL) and H₂O (10

mL). The aqueous layer was extracted with CHCl₃(2 × 10 mL),

and the combined organic layers were washed with H₂O (30

mL), dried (Na₂SO₄) and concentrated. The crude product and

KF‚2H₂O (400 mg, 4.2 mmol) were dissolved in MeOH (3 mL),

and stirred at room temperature for 18 h. The reaction was

monitored by analytical RP-LC-MS. The solvent was removed

under reduced pressure and the residue was partitioned

between CHCl₃(30 mL) and brine (30 mL). The aqueous layer

was extracted with CHCl₃(2 × 30 mL) and the combined

organic layers were washed with H₂O (30 mL), dried (Na₂SO₄),

and concentrated. Purification by flash chromatography (0-

4% MeOH in CHCl₃) gave 17 (10 mg, 27%) as a white solid:

mg, 95%) as a white solid: [R]²⁴+35.7° (c ) 0.90, CHCl₃).

D

Anal. (C₃₂H₅₃Br₂N₃O₈Si₃) C, H, N.

The subsequent cyclization was carried out essentially

according to general procedure II using the TMS-protected

diacyl hydrazine (77 mg, 0.090 mmol) and Burgess reagent

(86 mg, 0.36 mmol) dissolved in dry THF (0.5 mL). Since the

reaction was slow, an additional amount of Burgess reagent

(30 mg, 0.13 mmol after 3 h) was added. In the deprotection

of the crude product KF‚2H₂O (500 mg, 5.3 mmol) and MeOH

(3 mL) was used. The solvent was removed under reduced

pressure, and the residue was partitioned between CHCl₃(10

mL) and brine (10 mL). The aqueous layer was extracted with

CHCl₃(3 × 10 mL), and the combined organic layers were

dried (Na₂SO₄) and concentrated. Purification by RP-LC-MS

(30 min gradient of 20-65% CH₃CN in 0.05% aqueous formic

[R]²³+56.6° (c )1.03, CHCl₃). Anal. (C₁₆H₂₀Br₂N₄O₆) C, H,

D

N.

(1R,2R,3R,4R)-1,4-Bis[(2E)-3-bromo-2-propenyloxy]-

1,4-bis[5-(4-tert-butylphenyl)-[1,3,4]oxadiazol-2-yl]-butane-

2,3-diol (18). To a stirred solution of diacyl hydrazine 4 (49

mg, 0.061 mmol) in dry CH₂Cl₂(3 mL) were added chlorotri-

methylsilane (150 µL, 1.2 mmol) and triethylamine (190 µL,

1.4 mmol). Since the reaction was slow, additional amounts

of chlorotrimethylsilane (150 µL, 1.2 mmol and 200 µL, 1.6

mmol after 4 and 22 h, respectively) and triethylamine (250

µL, 1.8 mmol and 250, 1.8 mmol after 4 and 22 h, respectively)

were added. The reaction mixture was stirred at room tem-

perature for 48 h. The reaction was monitored by TLC and

analytical RP-LC-MS. The solvent was removed under reduced

pressure, and subsequent purification by flash chromatogra-

phy (pentane/CHCl₃/MeOH, 60:40:1) gave the TMS-protected

diacyl hydrazine (48 mg, 83%) as a white solid: [R]²³_D-34.1°

(c ) 1.00, CHCl₃). Anal. (C₄₀H₆₀Br₂N₄O₈Si₂) C, H, N.

TMS-protected diacyl hydrazine (49 mg, 0.052 mmol) and

Burgess reagent (188 mg, 0.79 mmol) were dissolved in dry

THF (3 mL). Since the reaction was slow, an additional amount

of Burgess reagent (70 mg, 0.29 mmol and 187 mg, 0.78 mmol

after 2 and 3 h, respectively) was added. The reaction mixture

was stirred at reflux for 4 h. The reaction was monitored by

analytical RP-LC-MS. The solvent was removed under reduced

pressure, and the residue was dissolved in CHCl₃, filtered

through a pad of silica (3 cm) and concentrated. The crude

product and KF‚2H₂O (700 mg, 7.4 mmol) was dissolved in

acid) gave 19 (12 mg, 21%) as a white solid: [R]²³+44.4° (c

D

)1.15, MeOH). Anal. (C₂₃H₂₇Br₂N₃O₇) C, H, N.

(2R,3R,4R,5R)-2,5-Bis[(2E)-3-bromo-2-propenyloxy]-

3,4-dihydroxy-N1-[(1S,2R)-2-hydroxyindan-1-yl]-5-(5-phen-

yl-[1,3,4]oxadiazol-2-yl)-pentanamide (20). Protection of

the diacylhydrazine 6 (100 mg, 0.14 mmol) was accomplished

according to general procedure II using chlorotrimethylsilane

(910 µL, 7.2 mmol) and triethylamine (1.0 mL, 7.5 mmol)

dissolved in dry CH₂Cl₂(5 mL). Purification by flash chroma-

tography (0-0.5% MeOH in CHCl₃) gave the TMS-protected

diacyl hydrazine (113 mg, 88%) as a white solid: [R]²⁴_D+27.5°

(c ) 0.97, CHCl₃). Anal. (C₃₇H₅₅Br₂N₃O₈Si₃) C, H, N.

The subsequent cyclization was carried out according to

general procedure II using the TMS-protected diacyl hydrazine

(109 mg, 0.12 mmol) and Burgess reagent (86 mg, 0.36 mmol)

dissolved in dry THF (0.8 mL). In the deprotection of the crude

product KF‚2H₂O (650 mg, 6.9 mmol) and MeOH (5 mL) was

used. The solvent was removed under reduced pressure, and

the residue was partitioned between CHCl₃(20 mL) and H₂O

(20 mL). The aqueous layer was extracted with CHCl₃(3 ×

20 mL), and the combined organic layers were dried (Na₂SO₄)

and concentrated. Purification by flash chromatography (1-

2% MeOH in CHCl₃) gave 20 (64 mg, 79%) as a white solid:

[R]²⁰+48.8° (c ) 0.91, MeOH). Anal. (C₂₈H₂₉Br₂N₃O₇) C, H,

D

N.

Malarial Plasmepsin Inhibitors

Journal of Medicinal Chemistry, 2005, Vol. 48, No. 19 6103

2% MeOH in CHCl₃) gave 23 (25 mg, 47%) as a white solid:

(2R,3R,4R,5R)-2,5-Bis[(2E)-3-bromo-2-propenyloxy]-5-

[5-(4-tert-butylphenyl)-[1,3,4]oxadiazol-2-yl]-3,4-dihydroxy-

N1-[(1S,2R)-2-hydroxyindan-1-yl]-pentanamide (21). Pro-

tection of the diacyl hydrazine 7 (66 mg, 0.087 mmol) was

accomplished according to general procedure II using chlorot-

rimethylsilane (1.1 mL, 8.7 mmol) and triethylamine (1.2 mL,

8.6 mmol) dissolved in dry CH₂Cl₂(6 mL). Purification by flash

chromatography (CHCl₃) gave the TMS-protected diacyl hy-

drazine (78 mg, 92%) as a white solid: [R]²⁴_D+26.7° (c )1.04,

CHCl₃). Anal. (C₄₁H₆₃Br₂N₃O₈Si₃) C, H, N.

The subsequent cyclization was carried out essentially

according to general procedure II using the TMS-protected

diacyl hydrazine (78 mg, 0.080 mmol) and Burgess reagent

(57 mg, 0.24 mmol) dissolved in dry THF (0.7 mL). Since the

reaction was slow, an additional amount of Burgess reagent

(20 mg, 0.084 mmol after 4 h) was added. The solvent was

removed under reduced pressure, and the residue was parti-

tioned between CHCl₃(10 mL) and H₂O (10 mL). The aqueous

layer was extracted with CHCl₃(3 × 10 mL), and the combined

organic layers were dried (Na₂SO₄) and concentrated. In the

deprotection of the crude product KF‚2H₂O (450 mg, 4.8 mmol)

and MeOH (4 mL) was used. The solvent was removed under

reduced pressure, and the residue was partitioned between

CHCl₃(10 mL) and H₂O (10 mL). The aqueous layer was

extracted with CHCl₃(3 × 10 mL), and the combined organic

layers were dried (Na₂SO₄) and concentrated. Purification by

flash chromatography (0-3% MeOH in CHCl₃) gave 21 (42

mg, 71%) as a white solid: [R]²²_D+50.5° (c )1.22, CHCl₃). Anal.

(C₃₂H₃₇Br₂N₃O₇) C, H, N.

(2R,3R,4R,5R)-5-(5-Benzyl-[1,3,4]oxadiazol-2-yl)-2,5-Bis-

[(2E)-3-bromo-2-propenyloxy]-3,4-dihydroxy-N1-[(1S,2R)-

2-hydroxyindan-1-yl]-pentanamide (22). Protection of the

diacyl hydrazine 8 (57 mg, 0.080 mmol) was accomplished

according to general procedure II using chlorotrimethylsilane

(510 µL, 4.0 mmol) and triethylamine (580 µL, 4.2 mmol)

dissolved in dry CH₂Cl₂(3 mL). Purification by flash chroma-

tography (0-0.5% MeOH in CHCl₃) gave the TMS-protected

diacyl hydrazine (63 mg, 85%) as a white solid: [R]¹⁹_D+36.5°

(c ) 0.99, CHCl₃). Anal. (C₃₈H₅₇Br₂N₃O₈Si₃) C, H, N.

The subsequent cyclization was carried out according to

general procedure II using the TMS-protected diacyl hydrazine

(56 mg, 0.060 mmol) and Burgess reagent (58 mg, 0.24 mmol)

dissolved in dry THF (0.5 mL). In the deprotection of the crude

product KF‚2H₂O (345 mg, 3.7 mmol) and MeOH (4 mL) was

used. The solvent was removed under reduced pressure, and

the residue was partitioned between CHCl₃(10 mL) and H₂O

(10 mL). The aqueous layer was extracted with CHCl₃(3 ×

10 mL), and the combined organic layers were dried (Na₂SO₄)

and concentrated. Purification by flash chromatography (0-

3% MeOH in CHCl₃) gave 22 (19 mg, 45%) as a white solid:

[R]²⁰_D+29,4° (c ) 0.94, CHCl₃). Anal. (C₂₉H₃₁Br₂N₃O₇‚¹/₂H₂O)

C, H, N.

[R]²⁰+33.1° (c ) 0.92, MeOH). Anal. (C₃₀H₃₃Br₂N₃O₇) C, H,

D

N.

(2R,3R,4R,5R)-2,5-Bis[(2E)-3-bromo-2-propenyloxy]-

3,4-dihydroxy-N1-[(1S,2R)-2-hydroxyindan-1-yl]-5-[5-(2-

phenylpropyl)-[1,3,4]oxadiazol-2-yl]-pentanamide (24).

Protection of the diacyl hydrazine 10 (89 mg, 0.12 mmol) was

accomplished according to general procedure II using chlorot-

rimethylsilane (760 µL, 6.0 mmol) and triethylamine 870 µL,

6.2 mmol) dissolved in dry CH₂Cl₂(4.5 mL). Purification by

flash chromatography (0-0.5% MeOH in CHCl₃) gave the

TMS-protected diacyl hydrazine (82 mg, 72%) as a white

solid: [R]¹⁹+32.8° (c ) 1.02, CHCl₃). Anal. (C₄₀H₆₁Br₂N₃O₈-

D

Si₃) C, H, N.

The subsequent cyclization was carried out according to

general procedure II using the TMS-protected diacyl hydrazine

(81 mg, 0.085 mmol) and Burgess reagent (60 mg, 0.25 mmol)

dissolved in dry THF (0.6 mL). In the deprotection of the crude

product KF‚2H₂O (474 mg, 5.0 mmol) and MeOH (4 mL) was

used. The solvent was removed under reduced pressure, and

the residue was partitioned between CHCl₃(20 mL) and H₂O

(20 mL). The aqueous layer was extracted with CHCl₃(3 ×

20 mL), and the combined organic layers were dried (Na₂SO₄)

and concentrated. Purification by flash chromatography (0-

2% MeOH in CHCl₃) gave 24 (36 mg, 59%) as a white solid:

[R]¹⁹+32.9° (c )1.01, CHCl₃). Anal. (C₃₁H₃₅Br₂N₃O₇) C, H,

D

N.

(2R,3R,4R,5R)-2,5-Bis[(2E)-bromo-2-propenyloxy]-5-

hydrazinocarbonyl-3,4-dihydroxy-N1-[(1S,2R)2-hydroxy-

indan-1-yl]-pentanamide (30). The monolactone 29 (40 mg,

0.071 mmol) was suspended in EtOH (4 mL) and hydrazine

hydrate (4.4 µL, 0.072 mmol) was added. After the addition

the suspension clearified and after approximately 1 h the

product started to precipitate. The reaction mixture was

stirred at room temperature for 8 h. The reaction was

monitored by TLC and analytical RP-LC-MS. The precipitated

product was filtered off and washed with EtOH. Additional

product was provided by flash chromatography (5-10% MeOH

in CHCl₃) of the filtrate to give 30 (in a total of 36 mg, 85%)

as white crystals: [R]²³_D+31.2° (c )1.10, MeOH). Anal. (C₂₁H₂₇-

Br₂N₃O₇) C, H, N.

(2R,3R,4R,5R)-2,5-Bis[(2E)-3-bromo-2-propenyloxy]-

3,4-dihydroxy-N1-[(1S,2R)-2-hydroxyindan-1-yl]-5-(5-meth-

yl-4H-[1,2,4]triazol-3-yl)-pentanamide (25). Acetamidine

hydrochloride (13.4 mg, 0.14 mmol) was dissolved in dry EtOH

(0.5 mL), and sodium methoxide (7.6 mg, 0.14) was added. The

mixture was stirred for 30 min, and the precipitated salt was

filtered off. To the filtrate was added hydrazide 30 (56 mg,

0.094 mmol), and the resulting reaction mixture was stirred

at reflux for 22 h. The reaction was monitored by TLC and

analytical RP-LC-MS. The solvent was removed under reduced

pressure, and subsequent purification by flash chromatogra-

phy (0-5% MeOH in CHCl₃) gave 25 (19 mg, 33%) as a white

solid: [R]¹⁹+15.5° (c )1.14, CHCl₃). Anal. (C₂₃H₂₈Br₂N₄O₆)

(2R,3R,4R,5R)-2,5-Bis[(2E)-3-bromo-2-propenyloxy]-

3,4-dihydroxy-N1-[(1S,2R)-2-hydroxyindan-1-yl]-5-[5-(2-

phenylethyl)-[1,3,4]oxadiazol-2-yl]-pentanamide (23). Pro-

tection of the diacyl hydrazine 9 (68 mg, 0.094 mmol) was

accomplished according to general procedure II using chlorot-

rimethylsilane (570 µL, 4.5 mmol) and triethylamine (650 µL,

4.7 mmol) dissolved in dry CH₂Cl₂(3.5 mL). Purification by

flash chromatography (0-0.5% MeOH in CHCl₃) gave the

TMS-protected diacyl hydrazine (78 mg, 88%) as a white

D

C, H, N.

(2R,3R,4R,5R)-2,5-Bis[(2E)-3-(3,4-methylenedioxy-

phenyl)-2-propenyloxy]-3,4-dihydroxy-N1-[(1S,2R)-2-hy-

droxyindan-1-yl]-6-oxo-6-[N’-(3-phenylpropanoyl)hydrazino]-

hexanamide (14). A mixture of diacyl hydrazine 9 (40 mg,

0.055 mmol), 3,4-methylenedioxyphenylboronic acid (55 mg,

0.33 mmol), Ba(OH)₂× 8H₂O (60 mg, 0.19 mmol), Pd₂(dba)₃

(5.0 mg, 0.0055 mmol), [(t-Bu)₃PH]BF₄(6.4 mg, 0.022 mmol),

and DME/H₂O/EtOH, 12:4:3 (4 mL) was added to a heavy-

walled Smith vial. The tube was capped and the mixture was

stirred at 110 °C for 15 min in the microwave cavity.

Consumption of starting material was checked by RP-LC-MS.

The crude product was partitioned between CHCl₃(50 mL)

and 0.2 M HCl (aq) (30 mL). The organic layer was dried (Na₂-

SO₄) and concentrated. Purification by RP-LC-MS (30 min

gradient of 30-77% CH₃CN in aqueous formic acid) gave 14

(10 mg, 23%) as a white solid: [R]²⁰_D+34.7° (c ) 0.39, MeOH).

Anal. (C₄₄H₄₅N₃O₁₂) C, H, N.

solid: [R]²²+30.5° (c ) 0.92, CHCl₃). Anal. (C₃₉H₅₉Br₂N₃O₈-

D

Si₃) C, H, N.

The subsequent cyclization was carried out according to

general procedure II using the TMS-protected diacyl hydrazine

(72 mg, 0.076 mmol) and Burgess reagent (56 mg, 0.23 mmol)

dissolved in dry THF (0.5 mL). In the deprotection of the crude

product KF‚2H₂O (400 mg, 4.2 mmol) and MeOH (4 mL) was

used. The solvent was removed under reduced pressure, and

the residue was partitioned between CHCl₃(20 mL) and H₂O

(20 mL). The aqueous layer was extracted with CHCl₃(3 ×

20 mL), and the combined organic layers were dried (Na₂SO₄)

and concentrated. Purification by flash chromatography (0-

(2R,3R,4R,5R)-2,5-Bis[(2E)-3-(4-acetylphenyl)-2-pro-

penyloxy]-3,4-dihydroxy-N1-[(1S,2R)-2-hydroxyindan-1-yl]-

6104 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 19

Ersmark et al.

6-oxo-6-[N’-(3-phenylpropanoyl)hydrazino]-hexan-

amide (15). A mixture of diacyl hydrazine 9 (30 mg, 0.041

mmol), 4-acetylphenylboronic acid (41 mg, 0.25 mmol), Ba-

(OH)₂× 8H₂O (46 mg, 0.14 mmol), Pd₂(dba)₃(3.8 mg, 0.0041

mmol), [(t-Bu)₃PH]BF₄(4.8 mg, 0.017 mmol), and DME/H₂O/

EtOH, 12:4:3 (3 mL) was added to a heavy-walled Smith vial.

The tube was capped, and the mixture was stirred at 110 °C

for 40 min in the microwave cavity. Consumption of starting

material was checked by RP-LC-MS. The crude product was

partitioned between CHCl₃(50 mL) and 0.2 M HCl (aq) (30

mL). The organic layer was dried (Na₂SO₄) and concentrated.

Purification by RP-LC-MS (30 min gradient of 30-75% CH₃-

CN in aqueous formic acid) gave 15 (7.4 mg, 22%) as a white

solid: [R]²⁰_D+28.7° (c ) 0.66, MeOH). Anal. (C₄₆H₄₉N₃O₁₀) C,

H, N.

Reasonable starting structures for MD simulation are of

importance for the convergence of binding free energy calcula-

tions. The ligand must thus initially be positioned relatively

close to the native binding mode in order for the configura-

tional sampling to converge around the true minimum within

reasonable computing time. In earlier work we predicted the

binding mode of compound 1 which has the same scaffold as

the present ligands.²⁷We were thus able to superimpose the

compounds here onto the scaffold of 1 bound to Plm II from

our earlier work and use that as a starting point for simula-

tions. The Plm II X-ray structure in complex with pepstatin

A was chosen as the starting protein model to take advantage

of the fact that it is often easier to equilibrate a ligand-

receptor complex by expanding a cavity rather than by

collapsing it. Of the available Plm II X-ray models in complex

with an inhibitor, the pepstatin A model was the most

collapsed.

All MD calculations were done out with the Q package⁶²

using the force field parameters of Gromos87.⁶³Partial charges

for the oxadiazole heterocyclic ring in ligand 23 and the vinyl

bromide in 9 and 23 were calculated with the HF/6-31G* basis

set after structure optimization with AM1 in Gaussian98.⁶⁴

All other atoms were assigned standard force field partial

charges. Asp34 was protonated at the oxygen pointing toward

Asp214. Asp214 was left unprotonated and was assigned a

total charge of -1. The protonation and charge protocol was

copied from earlier work.^26,27

The protein-ligand complexes were solvated in a 20 Å

sphere of SPC water.⁶¹The sphere was centered on the central

carbon of the ligand closest to the P1′ group. Water was added

on a grid with a grid spacing of 3.1 Å and with a minimum

distance of 2.4 Å to heavy atoms in the solute. Atoms outside

the sphere were restrained to their initial positions with a

harmonic potential with a force constant of 100 kcal/(mol Å²)

and were only allowed to interact through bonded terms. All

interactions involving ligand atoms were explicitly calculated

within the sphere but for protein-protein, protein-water and

water-water interactions a cutoff of 10 Å was applied. For

long-range electrostatic interactions beyond this cutoff the local

reaction field multipole expansion method⁶⁵was used. Atoms

in the outer shell of the simulation sphere, between 16 and

20 Å from the center, were weakly restrained with a force

constant of 10 kcal/(mol Å²) to get a softer transition from

restrained to unrestrained atoms across the sphere boundary.

The systems were equilibrated by slowly raising the tem-

perature from 1 to 300 K in five steps. Loose restraints were

used to keep the four hydrogen bonds involved in the extended

â conformation between the ligands and the protein intact. The

two central hydroxyls of the ligands interacting with the

catalytic dyad, Asp34 and Asp214, were also kept in place with

restraints during the equilibration. The restrained equilibra-

tion was performed for 100 ps with gradually decreasing

restraints, increasing time step and increasing temperature.

A time step of 0.2 fs was used in the initial equilibration and

was increased to a final value of 1.0 fs. The systems were

finally subjected to unrestrained equilibration of at least 500

ps before data collection. During data collection energies were

recorded every 25 fs and energy averaging was done on more

than 1 ns of MD.

Assays and K_iDeterminations. Plm II was prepared

according to Westling et al.,⁵⁸and the expression and purifica-

tion of Plm I will be published elsewhere (manuscript in

preparation). The Plm IV assay was performed as described

elsewhere.^13,52Human liver Cat D was purchased from Sigma-

Aldrich, Sweden. The activities of Plm I, Plm II and Cat D

were measured essentially as described earlier,²¹using a total

reaction volume of 100 µL. The concentration of pro-Plm II

was 3 nM, the amount of Plm I was adjusted to give similar

catalytic activity and 50 ng/mL pro-Cat D was used. The pro-

sequence of Plm II was cleaved off by preincubation in assay

reaction buffer (100 mM sodium acetate buffer (pH 4.5), 10%

glycerol and 0.01% Tween 20) at room temperature for 40 min

and Cat D was activated by incubation in the same reaction

buffer at 37 °C for 20 min. The reaction was initiated by the

addition of 3 µM substrate (DABCYL-Glu-Arg-Nle-Phe-Leu-

Ser-Phe-Pro-EDANS, AnaSpec Inc, San Jose, CA), and hy-

drolysis was recorded as the increase in fluorescence intensity

over a 10 min time period, during which the rate increased

linearly with time. Stock solutions of inhibitors in DMSO were

serially diluted in DMSO and added directly before addition

of substrate, giving a final DMSO concentration of 1%. IC₅₀

values were obtained by assuming competitive inhibition and

fitting Langmuir isotherm (v_i/v_o) 1/(1 + [I]/IC₅₀)) to the dose

resonse data (Grafit), where v_iand v_oare the initial velocities

for the inhibited and uninhibited reaction respectively and [I]

is the inhibitor concentration.⁵⁹The K_iwas subsequently

calculated by using K_i) IC₅₀/(1 + [S]/K_m)⁶⁰and a K_mvalue

determined according to Michaelis-Menten.

The P. falciparum-infected erythrocyte assay was performed

as described earlier.²⁶

MD Simulations and Free Energy Calculations. The

linear interaction energy method (LIE)⁴⁵is a useful tool for

predicting absolute binding free energies of biomolecular

complexes. According to this approach the free energy of

binding can be estimated as:

vdW

el

∆G_bind) R( V_l-s

_p- V_l-s

_w) + â( V_l-s

-

p

el

V_l-_{s w}) + γ(0.1)

where V_l-sis the thermodynamical average of the potential

energy between the ligand and its surrounding environment.

The subscripts p and w denote the calculated average of the

ligand bound to the solvated receptor and free in solution,

respectively, and the superscripts vdW and el denote the

nonpolar (van der Waals) and polar (electrostatic) contribu-

tions to the potential energy average. The polar contribution

is derived from the linear response approximation and depends

on the chemical nature of the ligand. Here â ) 0.33 was used

as obtained in ref 61 for compounds with two or more hydroxyl

groups. The nonpolar part of the potential energy is scaled by

an empirically determined factor. As in ref 61 a value of 0.181

was used for R, and the constant γ was set to zero. Molecular

dynamics (MD) trajectories were used to obtain the required

averages, and the precision of these quantities was estimated

by dividing the relevant trajectory in two halves and compar-

ing their corresponding averages. The difference between them

is the estimated error, which will be small for well equilibrated

systems.

The simulations of the ligands in the unbound state were

performed in a similar manner. The ligands were solvated in

a 20 Å radius and were equilibrated for 50 ps at 300 K. The

geometrical centers of the ligands were restrained to the center

of the sphere by a harmonic potential of 50 kcal/(mol Å²)l to

keep the ligands from drifting out toward the boundary of the

sphere. Radial and polarization restraints⁶⁶were applied to

the water in all simulations to mimic bulk water at the sphere

boundary. Time averaging was done on trajectories of more

than 1 ns.

Rigid receptor docking was carried out with Autodock3.⁴⁶

The dockings were made with ligand 16 using 10 million

energy calculations per run and a population of 150 individu-

als. Being the smallest ligand in the computationaly evaluated

set it is more likely than the other to fit into a rigid binding

Malarial Plasmepsin Inhibitors

Journal of Medicinal Chemistry, 2005, Vol. 48, No. 19 6105

(14) Wyatt David, M.; Berry, C. Activity and inhibition of plasmepsin

site. The 1SME, 1LF2 and 1LF3 X-ray structures were used

as models of Plm II and were initially processed with the

standard Autodock procedure. In a final step the hydrogen

used in our simulations on Asp214 was added manually and

the partial charges of Asp214 were modified to the protonated

Asp residue in the AMBER95 force field.⁶⁷The ligand was

assigned Gasteiger charges as implemented in Autodock. The

rigid root was picked automatically and all amide bonds were

kept rigid through the conformational search. One hundred

fifty separate runs were made in each protein model, and the

results were clustered with an RMSD of 2.5 Å.

Additional dockings were made where the Cꢀ of Met75 in

1SME and 1LF3 moved by a 180° rotation around the Câ-

Cγ-Sδ-Cꢀ torsion, thereby enlarging the S2′ pocket. In the

1LF2 model of PlmII the S2′ is already in the open conforma-

tion, and hence no rotation was made there.

IV,

a new aspartic proteinase from the malaria parasite,

Plasmodium falciparum. FEBS Lett. 2002, 513, 159-162.

(15) Bjelic, S.; Aaqvist, J. Computational Prediction of Structure,

Substrate Binding Mode, Mechanism, and Rate for a Malaria

Protease with a Novel Type of Active Site. Biochemistry 2004,

43, 14521-14528.

(16) Nezami, A.; Freire, E. The Integration of Genomic and Structural

Information in the Development of High Affinity Plasmepsin

Inhibitors. Int. J. Parasitol. 2002, 32, 1669-1676.

(17) Goldberg, D. E.; Slater, A. F. G.; Beavis, R.; Chait, B.; Cerami,

A.; Henderson, G. B. Hemoglobin Degradation in the Human

Malaria Pathogen Plasmodium falciparum: A Catabolic Path-

way Initiated by a Specific Aspartic Protease. J. Exp. Med. 1991,

173, 961-969.

(18) Gluzman, I. Y.; Francis, S. E.; Oksman, A.; Smith, C. E.; Duffin,

K. L.; Goldberg, D. E. Order and Specificity of the Plasmodium

falciparum Hemoglobin Degradation Pathway. J. Clin. Invest.

1994, 93, 1602-1608.

(19) Silva, A. M.; Lee, A. Y.; Gulnik, S. V.; Majer, P.; Collins, J.; Bhat,

T. N.; Collins, P. J.; Cachau, R. E.; Luker, K. E.; Gluzman, I. Y.;

Francis, S. E.; Oksman, A.; Goldberg, D. E.; Erickson, J. W.

Structure and Inhibition of Plasmepsin II, a Hemoglobin-

Degrading Enzyme from Plasmodium falciparum. Proc. Natl.

Acad. Sci. U.S.A. 1996, 93, 10034-10039.

Acknowledgment. We thank the Swedish Founda-

tion for Strategic Research (SSF) and the Swedish

Research Council (VR) for financial support. We also

thank Biotage for providing a Smith Synthesizer.

(20) Moon, R. P.; Tyas, L.; Certa, U.; Rupp, K.; Bur, D.; Jacquet, C.;

Matile, H.; Loetscher, H.; Grueninger-Leitch, F.; Kay, J.; Dunn,

B. M.; Berry, C.; Ridley, R. G. Expression and Characterization

of Plasmepsin I from Plasmodium falciparum. Eur. J. Biochem.

1997, 244, 552-560.

(21) Haque, T. S.; Skillman, A. G.; Lee, C. E.; Habashita, H.;

Gluzman, I. Y.; Ewing, T. J. A.; Goldberg, D. E.; Kuntz, I. D.;

Ellman, J. A. Potent, Low-Molecular-Weight Non-Peptide In-

hibitors of Malarial Aspartyl Protease Plasmepsin II. J. Med.

Chem. 1999, 42, 1428-1440.

(22) No¨teberg, D.; Hamelink, E.; Hulte´n, J.; Wahlgren, M.; Vrang,

L.; Samuelsson, B.; Hallberg, A. Design and Synthesis of

Plasmepsin I and Plasmepsin II Inhibitors with Activity in

Plasmodium falciparum-Infected Cultured Human Erythrocytes.

J. Med. Chem. 2003, 46, 734-746.

(23) Dahlgren, A.; Kvarnstrom, I.; Vrang, L.; Hamelink, E.; Hallberg,

A.; Rosenquist, A.; Samuelsson, B. New Inhibitors of the Malaria

Aspartyl Proteases Plasmepsin I and II. Bioorg. Med. Chem.

2003, 11, 3423-3437.

(24) Oscarsson, K.; Oscarson, S.; Vrang, L.; Hamelink, E.; Hallberg,

A.; Samuelsson, B. New Potent C2-Symmetric Malaria Plas-

mepsin I and II Inhibitors. Bioorg. Med. Chem. 2003, 11, 1235-

1246.

(25) No¨teberg, D.; Schaal, W.; Hamelink, E.; Vrang, L.; Larhed, M.

High-speed Optimization of Inhibitors of the Malarial Proteases

Plasmepsin I and II. J. Comb. Chem. 2003, 5, 456-464.

(26) Ersmark, K.; Feierberg, I.; Bjelic, S.; Hamelink, E.; Hackett, F.;

Blackman, M. J.; Hulten, J.; Samuelsson, B.; Åqvist, J.; Hallberg,

A. Potent Inhibitors of the Plasmodium falciparum Enzymes

Plasmepsin I and II Devoid of Cathepsin D Inhibitory Activity.

J. Med. Chem. 2004, 47, 110-122.

(27) Ersmark, K.; Feierberg, I.; Bjelic, S.; Hulte´n, J.; Samuelsson,

B.; Åqvist, J.; Hallberg, A. C₂-Symmetric Inhibitors of Plasmo-

dium falciparum Plasmepsin II: Synthesis and Theoretical

Predictions. Bioorg. Med. Chem. 2003, 11, 3723-3733.

(28) Carroll, C. D.; Patel, H.; Johnson, T. O.; Guo, T.; Orlowski, M.;

He, Z.-M.; Cavallaro, C. L.; Guo, J.; Oksman, A.; Gluzman, I.

Y.; Connelly, J.; Chelsky, D.; Goldberg, D. E.; Dolle, R. E.

Identification of Potent Inhibitors of Plasmodium falciparum

Plasmepsin II from an Encoded Statine Combinatorial Library.

Bioorg. Med. Chem. Lett. 1998, 8, 2315-2320.

(29) Carroll, C. D.; Johnson, T. O.; Tao, S.; Lauri, G.; Orlowski, M.;

Gluzman, I. Y.; Goldberg, D. E.; Dolle, R. E. Evaluation of a

Structure-Based Statine Cyclic Diamino Amide Encoded Com-

binatorial Library Against Plasmepsin II and Cathepsin D.

Bioorg. Med. Chem. Lett. 1998, 8, 3203-3206.

Supporting Information Available: ¹H and ¹³C NMR

spectra of the compounds 3-25 and 28-30. This information

is available free of charge via the Internet at http://

pubs.acs.org.

References

(1) Breman, J. The Ears of the Hippopotamus: Manifestations,

Determinants, and Estimates of the Malaria Burden. Am. J.

Trop. Med. Hyg. 2001, 64 (1, 2)S, 1-11.

(2) Wongsrichanalai, C.; Pickard, A. L.; Wernsdorfer, W. H.; Mesh-

nick, S. R. Epidemiology of Drug-Resistant Malaria. Lancet

Infect. Dis. 2002, 2, 209-218.

(3) Gardner, M. J.; Hall, N.; Fung, E.; White, O.; Berriman, M.;

Hyman, R. W.; Carlton, J. M.; Pain, A.; Nelson, K. E.; Bowman,

S.; Paulsen, I. T.; James, K.; Eisen, J. A.; Rutherford, K.;

Salzberg, S. L.; Craig, A.; Kyes, S.; Chan, M.-S.; Nene, V.;

Shallom, S. J.; Suh, B.; Peterson, J.; Angiuoli, S.; Pertea, M.;

Allen, J.; Selengut, J.; Haft, D.; Mather, M. W.; Vaidya, A. B.;

Martin, D. M. A.; Fairlamb, A. H.; Fraunholz, M. J.; Roos, D.

S.; Ralph, S. A.; McFadden, G. I.; Cummings, L. M.; Subrama-

nian, G. M.; Mungall, C.; Venter, J. C.; Carucci, D. J.; Hoffman,

S. L.; Newbold, C.; Davis, R. W.; Fraser, C. M.; Barrell, B.

Genome Sequence of the Human Malaria Parasite Plasmodium

falciparum. Nature (London) 2002, 419, 498-511.

(4) Banerjee, R.; Liu, J.; Beatty, W.; Pelosof, L.; Klemba, M.;

Goldberg, D. E. Four Plasmepsins are Active in the Plasmodium

falciparum Food Vacuole, Including a Protease with an Active-

Site Histidine. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 990-995.

(5) Rosenthal, P. J. Plasmodium falciparum: Effects of Proteinase

Inhibitors on Globin Hydrolysis by Cultured Malaria Parasites.

Exp. Parasitol. 1995, 80, 272-281.

(6) Olson, J. E.; Lee, G. K.; Semenov, A.; Rosenthal, P. J. Antima-

larial Effects in Mice of Orally Administered Peptidyl Cysteine

Protease Inhibitors. Bioorg. Med. Chem. Lett. 1999, 7, 633-638.

(7) Boss, C.; Richard-Bildstein, S.; Weller, T.; Fischli, W.; Meyer,

S.; Binkert, C. Inhibitors of the Plasmodium falciparum Parasite

Aspartic Protease Plasmepsin II as Potential Antimalarial

Agents. Curr. Med. Chem. 2003, 10, 883-907.

(8) Salas, F.; Fichmann, J.; Lee, G. K.; Scott, M. D.; Rosenthal, P.

J. Functional Expression of Falcipain, a Plasmodium falciparum

Cysteine Proteinase, Supports its Role as a Malarial Hemo-

globinase. Infect. Immun. 1995, 63, 2120-2125.

(9) Shenai, B. R.; Sijwali, P. S.; Singh, A.; Rosenthal, P. J.

Characterization of Native and Recombinant Falcipain-2, a

Principal Trophozoite Cysteine Protease and Essential Hemo-

globinase of Plasmodium falciparum. J. Biol. Chem. 2000, 275,

29000-29010.

(10) Sijwali, P. S.; Shenai, B. R.; Gut, J.; Singh, A.; Rosenthal, P. J.

Expression and characterization of the Plasmodium falciparum

hemoglobinase falcipain-3. Biochem. J. 2001, 360, 481-489.

(11) Eggleson, K. K.; Duffin, K. L.; Goldberg, D. E. Identification and

Characterization of Falcilysin, a Metallopeptidase Involved in

Hemoglobin Catabolism within the Malaria Parasite Plasmo-

dium falciparum. J. Biol. Chem. 1999, 274, 32411-32417.

(12) Nezami, A.; Kimura, T.; Hidaka, K.; Kiso, A.; Liu, J.; Kiso, Y.;

Goldberg, D. E.; Freire, E. High-Affinity Inhibition of a Family

of Plasmodium falciparum Proteases by a Designed Adaptive

Inhibitor. Biochemistry 2003, 42, 8459-8464.

(30) Dolle, R. E.; Guo, J.; O’Brien, L.; Jin, Y.; Piznik, M.; Bowman,

K. J.; Li, W.; Egan, W. J.; Cavallaro, C. L.; Roughton, A. L.; Zhao,

W.; Reader, J. C.; Orlowski, M.; Jacob-Samuel, B.; DiIanni

Carroll, C. A Statistical-Based Approach to Assessing the

Fidelity of Combinatorial Libraries Encoded with Electrophoric

Molecular Tags. Development and Application of Tag Decode-

Assisted Single Bead LC/MS Analysis. J. Comb. Chem. 2000, 2,

716-731.

(31) Jiang, S.; Prigge, S. T.; Wei, L.; Gao, Y.-E.; Hudson, T. H.;

Gerena, L.; Dame, J. B.; Kyle, D. E. New Class of Small

Nonpeptidyl Compounds Blocks Plasmodium falciparum Devel-

opment In Vitro by Inhibiting Plasmepsins. Antimicrob. Agents

Chemother. 2001, 45, 2577-2584.

(32) Nezami, A.; Luque, I.; Kimura, T.; Kiso, Y.; Freire, E. Identifica-

tion and Characterization of Allophenylnorstatine-Based Inhibi-

tors of Plasmepsin II, an Antimalarial Target. Biochemistry

2002, 41, 2273-2280.

(13) Dame, J. B.; Yowell, C. A.; Omara-Opyene, L.; Carlton, J. M.;

Cooper, R. A.; Li, T. Plasmepsin 4, the Food Vacuole Aspartic

Proteinase Found in all Plasmodium Spp. Infecting Man. Mol.

Biochem. Parasitol. 2003, 130, 1-12.

6106 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 19

Ersmark et al.

(33) Asojo, O. A.; Afonina, E.; Gulnik, S. V.; Yu, B.; Erickson, J. W.;

Randad, R.; Medjahed, D.; Silva, A. M. Structures of Ser205

Mutant Plasmepsin II from Plasmodium falciparum at 1.8 Å in

Complex with the Inhibitors rs367 and rs370. Acta Crystallogr.

D 2002, D58, 2001-2008.

(34) Johansson, P.-O.; Chen, Y.; Belfrage, A. K.; Blackman, M. J.;

Kvarnstroem, I.; Jansson, K.; Vrang, L.; Hamelink, E.; Hallberg,

A.; Rosenquist, A.; Samuelsson, B. Design and Synthesis of

Potent Inhibitors of the Malaria Aspartyl Proteases Plasmepsin

I and II. Use of Solid-Phase Synthesis to Explore Novel Statine

Motifs. J. Med. Chem. 2004, 47, 3353-3366.

(52) Li, T. Y.; C. A.; Beyer, B. B.; Hung, S.-H.; Westling, J.; Lam; M.

T.; Dunn, B. M.; Dame, J. B. Recombinant Expression and

Enzymatic Subsite Characterization of Plasmepsin 4 from the

Four Plasmodium Species Infecting Man. Mol. Biochem. Para-

sitol. 2004, 135, 101-109.

(53) Omara-Opyene, A. L.; Moura, P. A.; Sulsona, C. R.; Bonilla, J.

A.; Yowell, C. A.; Fujioka, H.; Fidock, D. A.; Dame, J. B. Genetic

Disruption of the Plasmodium falciparum Digestive Vacuole

Plasmepsins Demonstrates Their Functional Redundancy. J.

Biol. Chem. 2004, 279, 54088-54096.

(54) Liu, J.; Gluzman, I. Y.; Drew, M. E.; Goldberg, D. E. The Role

of Plasmodium falciparum Food Vacuole Plasmepsins. J. Biol.

Chem. 2005, 280, 1432-1437.

(35) Berry, C. Plasmepsins as Antimalarial Targets. Curr. Opin. Drug

Discovery Des. 2000, 3, 624-629.

(36) Patani, G. A.; LaVoie, E. J. Bioisosterism: A Rational Approach

in Drug Design. Chem. Rev. 1996, 96, 3147-3176.

(55) Prata, J. V.; Clemente, D.-T. S.; Prabhakar, S.; Lobo, A. M.;

Mourato, I.; Branco, P. S. Intramolecular Addition of Acyldiaz-

enecarboxylates Onto Double Bonds in the Synthesis of Hetero-

cycles. J. Chem. Soc., Perkin Trans. 1 2002, 513-528.

(56) Henichart, J. P.; Houssin, R.; Lablanche, B. Synthesis and

Structure of 1,2,4-Triazoline-5-thiones Having a Dialkylamino-

ethyl Substituent at the 3 or 4 Position. J. Heterocycl. Chem.

1977, 14, 615-619.

(57) Massarani, E.; Nardi, D.; Rossi, S.; Tajana, A.; Degen, L.

Antibacterial Nitrofuran Derivatives. 3. 5-Nitro-2-furaldehyde

Piperazinoacylhydrazones. J. Med. Chem. 1971, 14, 635-638.

(58) Westling, J.; Cipullo, P.; Hung, S. H.; Saft, H.; Dame, J. B.; Dunn,

B. M. Active Site Specificity of Plasmepsin II. Protein Sci. 1999,

8, 2001-2009.

(59) Copeland, R. A.; Ed. Enzymes: A Practical Introduction to

Structure, Mechanism, and Data Analysis; VCH: New York,

1996; 300 pp.

(60) Cheng, Y.; Prusoff, W. H. Relationship Between the Inhibition

Constant (K1) and the Concentration of Inhibitor which Causes

50 Per Cent Inhibition (I50) of an Enzymatic Reaction. Biochem.

Pharmacol. 1973, 22, 3099-3108.

(37) Chen, X.; Wang, W. The Use of Bioisosteric Groups in Lead

Optimization. Annu. Rep. Med. Chem. 2003, 38, 333-346.

(38) Giannis, A.; Kolter, T. Peptide Mimetics for Receptor Ligands:

Discovery, Development, and Medicinal Perspectives. Angew.

Chem. 1993, 105, 1303-1326.

(39) Asojo, O. A.; Gulnik, S. V.; Afonina, E.; Yu, B.; Ellman, J. A.;

Haque, T. S.; Silva, A. M. Novel Uncomplexed and Complexed

Structures of Plasmepsin II, an Aspartic Protease from Plas-

modium falciparum. J. Mol. Biol. 2003, 327, 173-181.

(40) Brain, C. T.; Paul, J. M.; Loong, Y.; Oakley, P. J. Novel Procedure

for the Synthesis of 1,3,4-Oxadiazoles from 1,2-Diacylhydrazines

Using Polymer-Supported Burgess Reagent under Microwave

Conditions. Tetrahedron Lett. 1999, 40, 3275-3278.

(41) Oscarsson, K.; Lahmann, M.; Lindberg, J.; Kangasmetsa, J.;

Unge, T.; Oscarson, S.; Hallberg, A.; Samuelsson, B. Design and

Synthesis of HIV-1 Protease Inhibitors. Novel Tetrahydrofuran

P2/P2′-Groups Interacting with Asp29/30 of the HIV-1 Protease.

Determination of Binding from X-ray Crystal Structure of

Inhibitor Protease Complex. Bioorg. Med. Chem. 2003, 11, 1107-

1115.

(42) Larhed, M.; Moberg, C.; Hallberg, A. Microwave-Accelerated

Homogeneous Catalysis in Organic Chemistry. Acc. Chem. Res.

2002, 35, 717-727.

(43) Larhed, M.; Lindeberg, G.; Hallberg, A. Rapid Microwave-

Assisted Suzuki Coupling on Solid-Phase. Tetrahedron Lett.

1996, 37, 8219-8222.

(44) Netherton, M. R.; Fu, G. C. Air-Stable Trialkylphosphonium

Salts: Simple, Practical, and Versatile Replacements for Air-

Sensitive Trialkylphosphines. Applications in Stoichiometric and

Catalytic Processes. Org. Lett. 2001, 3, 4295-4298.

(45) A° qvist, J.; Medina, C.; Samuelsson, J. E. A New Method for

Predicting Binding Affinity in Computer-Aided Drug Design.

Protein Eng. 1994, 7, 385-391.

(61) Marelius, J.; Hansson, T.; Aqvist, J. Calculation of Ligand

Binding Free Energies from Molecular Dynamics Simulations.

Int. J. Quantum Chem. 1998, 69, 77-88.

(62) Marelius, J.; Kolmodin, K.; Feierberg, I.; A° qvist, J. Q:

A

Molecular Dynamics Program for Free Energy Calculations and

Empirical Valence Bond Simulations in Biomolecular Systems.

J. Mol. Graph. Model. 1998, 16, 213-225, 261.

(63) van Gunsteren, W. F. B.; H. J. C. Groningen Molecular Simula-

tion (GROMOS) Library Manual; Nijenborgh: Groningen, The

Netherlands, 1987.

(64) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;

Robb, M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery,

J. A., Jr.; Stratmann, R. E.; Burant, J. C.; Dapprich, S.; Millam,

J. M.; Daniels, A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.;

Tomasi, J.; Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.;

Pomelli, C.; Adamo, C.; Clifford, S.; Ochterski, J.; Petersson, G.

A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.; Malick, D. K.; Rabuck,

A. D.; Raghavachari, K.; Foresman, J. B.; Cioslowski, J.; Ortiz,

J. V.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.;

Komaromi, I.; Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith, T.;

Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Gonzalez, C.;

Challacombe, M.; Gill, P. M. W.; Johnson, B. G.; Chen, W.; Wong,

M. W.; Andres, J. L.; Head-Gordon, M.; Replogle, E. S.; Pople,

J. A. Gaussian 98, revision x.x; Gaussian, Inc.: Pittsburgh, PA,

1998.

(65) Lee, F. S.; Warshel, A. A Local Reaction Field Method for Fast

Evaluation of Long-Range Electrostatic Interactions in Molec-

ular Simulations. J. Chem. Phys. 1992, 97, 3100-3107.

(66) King, G.; Warshel, A. A Surface-Contained All-Atom Solvent

Model for Effective Simulations of Polar Solutions. J. Chem.

Phys. 1989, 91, 3647-3661.

(46) Morris, G. M.; Goodsell, D. S.; Halliday, R. S.; Huey, R.; Hart,

W. E.; Belew, R. K. and Olson, A. J. Automated Docking Using

a Lamarckian Genetic Algorithm and Empirical Binding Free

Energy Function. J. Comput. Chem. 1998, 19, 1639-1662.

(47) Hulten, J.; Bonham, N. M.; Nillroth, U.; Hansson, T.; Zuccarello,

G.; Bouzide, A.; Aqvist, J.; Classon, B.; Danielson, U. H.; Karlen,

A.; Kvarnstrom, I.; Samuelsson, B.; Hallberg, A. Cyclic HIV-1

Protease Inhibitors Derived from Mannitol: Synthesis, Inhibi-

tory Potencies, and Computational Predictions of Binding Af-

finities. J. Med. Chem. 1997, 40, 885-897.

(48) Kuntz, I. D.; Chen, K.; Sharp, K. A.; Kollman, P. A. The Maximal

Affinity of Ligands. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 9997-

10002.

(49) Shirts, M. R.; Pitera, J. W.; Swope, W. C.; Pande, V. S. Extremely

Precise Free Energy Calculations of Amino Acid Side Chain

Analogues: Comparison of Common Molecular Mechanics Force

Fields for Proteins. J. Chem. Phys. 2003, 119, 5740-5761.

(50) Oostenbrink, C.; Villa, A.; Mark, A. E.; van Gunsteren, W. F. A

Biomolecular Force Field Based on the Free Enthalpy of Hydra-

tion and Solvation: The GROMOS Force-Field Parameter Sets

53A5 and 53A6. J. Comput. Chem. 2004, 25, 1656-1676.

(51) Coombs, G. H.; Goldberg, D. E.; Klemba, M.; Berry, C.; Kay, J.;

Mottram, J. C. Aspartic Proteases of Plasmodium falciparum

and other Parasitic Protozoa as Drug Targets. Trends Parasitol.

2001, 17, 532-537.

(67) Cornell, W. D.; Cieplak, P.; Bayly, C. I.; Gould, I. R.; Merz, K.

M., Jr.; Ferguson, D. M.; Spellmeyer, D. C.; Fox, T.; Caldwell,

J. W.; Kollman, P. A. A Second Generation Force Field for the

Simulation of Proteins, Nucleic Acids, and Organic Molecules.

J. Am. Chem. Soc. 1996, 118, 2309.

JM050463L

Article Doi

Synthesis of malarial plasmepsin inhibitors and prediction of binding modes by molecular dynamics simulations

DOI: 10.1021/jm050463l

Source and publish data:

Authors:

Article abstract of DOI:10.1021/jm050463l

Full text of DOI:10.1021/jm050463l

Products guided by the article

R&D Labs maybe for 866755-67-1

Relevant to this article

Hot Product