Inhibitors of plasmepsin II - Potential antimalarial agents

Article abstract of DOI:10.2533/000942904777677524

Malaria is a very serious infectious disease affecting over two billion people worldwide. Currently available antimalarial drugs are losing effectiveness due to the emergence and the spread of resistant parasite strains. In order to regain control over the disease, new treatments are urgently needed. Drug discovery efforts in this direction are most likely to be successful if they target a novel mechanism of action. Such approaches will lead to antimalarial medicines that are functionally and structurally different from the existing drugs and therefore will have the potential to overcome existing resistances. Our own efforts are focused on the aspartic protease plasmepsin II (PMII) which is a promising new drug target for future antimalarial therapies. We have found structurally simple, moderately active, non-peptide inhibitors of plasmepsin II which offer ample opportunity for further optimization efforts.

Full text of DOI:10.2533/000942904777677524

PHARMACEUTICAL INDUSTRY IN SWITZERLAND
634
CHIMIA 2004, 58, No. 9
Chimia 58 (2004) 634–639
© Schweizerische Chemische Gesellschaft
ISSN 0009–4293
Inhibitors of Plasmepsin II –
Potential Antimalarial Agents
Christoph Bossa*, Sylvia Richard-Bildsteina, Rocco Furnarib, Jean-Marc Bourgeoisb, Olivier
Corminboeufa, Corinna Grisostomia, Lionel Coppexa, Luke Harrisa, Lars Pradea, Solange Meyera,
Christoph Binkerta, Walter Fischlia, Reto Brunc, and Thomas Wellera
Abstract: Malaria is a very serious infectious disease affecting over two billion people worldwide. Currently avail-
able antimalarial drugs are losing effectiveness due to the emergence and the spread of resistant parasite strains.
In order to regain control over the disease, new treatments are urgently needed. Drug discovery efforts in this di-
rection are most likely to be successful if they target a novel mechanism of action. Such approaches will lead to
antimalarial medicines that are functionally and structurally different from the existing drugs and therefore will have
the potential to overcome existing resistances. Our own efforts are focused on the aspartic protease plasmepsin
II (PMII) which is a promising new drug target for future antimalarial therapies. We have found structurally simple,
moderately active, non-peptide inhibitors of plasmepsin II which offer ample opportunity for further optimization
efforts.
Keywords: Aspartic proteases · Enzyme inhibitors · Malaria · Medicinal chemistry · Plasmepsin II (PMII)
Introduction
About 90% of these cases can be allocated
to Sub-Saharan Africa. The remaining cas-
parasites, requires high doses and there-
fore often leads to severe side effects.
Artemisinin and its derivatives, a sec-
ond group of drugs used against resist-
ant strains, are ‘fast-acting’ drugs and
should be administered in combination
with a ‘long-acting’ drug in order to re-
duce recrudescence and to prevent or
slow the development of resistances.
These facts clearly indicate the urgent
Malaria remains one of the world’s most se- es are mostly confined to South and South-
rious infectious diseases, besides tubercu- east Asia and parts of South America. In
losis and AIDS. Malaria is the most wide- these regions, malaria, besides torturing the
spread parasitic disease in humans, with infected individuals, also severely affects
about 40% of the world’s population (~2 the functionality of societies and has a
billion people) at risk of infection, mainly strong negative impact on the economic sit-
in tropical and subtropical areas. Of the four uation and certainly contributes to econom-
species of Plasmodium responsible for ic underdevelopment [1].
malaria in man, Plasmodium falciparum
During recent years, several factors had need for the discovery of new antimalarial
medicines. Future drugs most importantly
causes the highest mortality. Every year a negative impact on the malaria problem:
about 400 million clinical cases occur, re- 1) Increased mobility of the population, must be cheap, effective and safe to use es-
sulting in more than 1 million deaths.
Malaria kills one person every 12 seconds.
climate changes, and novel agricultural pecially in children and in pregnant women.
practices have allowed the disease to In order to minimize problems with exist-
spread into formerly unaffected regions. ing resistances it is advisable to focus drug
2) Disease control in affected areas has discovery efforts towards compounds with
been made difficult by political conflicts. new mechanisms of action. To avoid or re-
3) The appearance of mosquito vectors re- duce the chance of side effects it is favor-
sistant to pesticides, as well as parasite able to target parasite-specific enzymes or
strains resistant to standard drugs, has parasite-specific pathways [2]. Different
drastically reduced the effectiveness of new targets are under investigation, e.g. the
most antimalarial drugs, especially falcipains or the plasmepsins I (PMI), II
chloroquine. The reduced effectiveness (PMII), and IV (PMIV) [3].
*Correspondence: Dr. C. Bossa
Tel.: +41 61 487 45 61
Fax: + 41 61 487 45 00
E-Mail: christoph.boss@actelion.com
aActelion Pharmaceuticals Ltd
Drug Discovery Chemistry & Biochemistry
Gewerbestrasse 16
of this drug has been a major setback in
During the blood stage of its life cycle
the efforts to control or eradicate malar- [4], the P. falciparum parasite reproduces
ia, since chloroquine is cheap and it is asexually and requires enormous amounts
safe to use in pregnant women. Resis- of nutrients to proliferate. Because of the
tance to other antimalarial agents such limited capacity to synthesize amino acids
as sulfadoxine-pyrimethamine, meflo- de novo, the parasites cover their needs by
quine, quinine or amodiaquine has also catabolism of human hemoglobin present
become a serious issue. Halofantrine, in red blood cells. Hemoglobin degradation
one of the few remaining therapeutic occurs within the food vacuole of the para-
compounds against resistant strains of site and is mediated by a series of parasite
CH–4123 Allschwil/BL
bEcole d’ingénieurs et d’architectes de Fribourg
Département de chimie
Bd de Pérolles 80
CH–1705 Fribourg
cSwiss Tropical Institute,
Socinstrasse 57
CH–4002 Basel

PHARMACEUTICAL INDUSTRY IN SWITZERLAND
635
CHIMIA 2004, 58, No. 9
specific proteolytic enzymes, e.g. aspartic change its conformation.
where it is urgently needed:
proteases called plasmepsins, cysteine pro-
The design of protease inhibitors gener- – High efficacy against P. falciparum, in-
teases called falcipains, or the metallopro- ally relies on a mechanism-based pharma-
cluding resistant strains
tease falcilysin (Fig. 1). As hemoglobin cophore as the central binding unit. Usual- – No side effects which would limit its
degradation is a parasite-specific catabolic ly a secondary alcohol is the structural ele-
pathway, essential for the survival of the ment of choice to inhibit aspartic proteases.
use, particularly in pregnant women or
children
Plasmodia, the enzymes involved in this This element mimics the tetrahedral transi- – Activity against the other species of
process fulfill the requirements as promis- tion state during peptide bond cleavage by
plasmodium (e.g. ovale; vivax)
ing drug targets against malaria. Plas- aspartic proteases and has been successful- – Oral bioavailability
mepsins appear to be well-suited antimalar- ly used to develop potent inhibitors based – Single dose treatment would be an asset
ia targets, since their activity is necessary on statines or hydroxyethylamine-units [8]. – Formulation must be stable in tropical
for the survival of P. falciparum parasites. However, most of these inhibitors are de-
climates
The parasites are unable to proliferate in rived from peptides and are therefore often – Cost per treatment: max. 2 USD/1.65
human red blood cells in vitro in the pres- characterized by metabolic instability, poor
ence of inhibitors of aspartic proteases [6]. pharmacokinetic properties and the inabili-
The attractiveness of plasmepsins as anti- ty to penetrate cells. Other structural ele-
Euro
malarial drug targets is enhanced by the fact ments that have also been successfully Plasmepsin II Inhibitors Based
that only eight members belonging to the applied in aspartic protease inhibitors so far on Tertiary Amines
aspartic protease family are known in man are either secondary amines in 4-aryl-
which considerably reduces the risk of side piperidine derived compounds [9] or
In order to find a new drug which is
effects. This is in contrast to other families tertiary amines in fully substituted able to fulfill the criteria listed above, it ap-
of proteases, e.g. cysteine- or metallo-pro- 4-aminopiperidine based inhibitors pears highly recommendable to look for
teases, where dozens to hundreds of family [10c][10d]. Potent, non-peptide, low mo- new molecules from structurally different
members were identified in the human lecular weight inhibitors of the aspartic pro- classes as compared to currently used anti-
genome. PMII is an endopeptidase working tease PMII have been identified, as demon- malarial therapeutics. Therefore, and due
best at low pH. It initiates the hemoglobin strated by reports in the literature [10]. No to the general interest in aspartic proteases
degradation in the food vacuole of P. falci- reports of pre-clinical or clinical investiga- as drug targets, a high-throughput FRET
parum by cleaving the β-chain of hemoglo- tions of such compounds have been detect- assay was developed for PMII [11] and a
bin between Phe33 and Leu34. The catalyt- ed so far.
ic unit within the active site of the enzyme
is formed by two aspartic acid residues
commercial library of 50000 compounds
was screened for inhibition of PMII activ-
ity. These efforts resulted in compounds
(Asp34 and Asp214) [7]. The active site Features of New Antimalarial Drugs which inhibited PMII at low µM concen-
consists of a lipophilic pocket which is cov-
ered by a flexible loop region called the
trations (depicted in Fig. 2).
Finding new antimalarial drugs still re-
Structural analysis of the compounds
‘flap’. In order to allow the substrate to ac- mains a challenging and difficult task. The depicted in Fig. 2 revealed the following
cess the active site, the ‘flap’ has to open. It future ideal medicine must combine sever- features: i) an ethylene-diamine core unit,
provides flexibility for the enzyme to al features in order to be useful and helpful ii) a tertiary amine functionality, iii) two
rather lipophilic substituents connected to
the 2nd end of the ethylenediamine unit,
and iv) all inhibitors contain a 4-pentyl-
benzoyl substituent. A graphical summary
is given in Fig. 3.
According to previous reports, it ap-
pears that an interaction of the inhibitor
with the two aspartic acid residues of the
active site should take place to provide effi-
cient inhibition of the enzyme. The tertiary
amine is the only group that can serve this
function and therefore was considered
to be mandatory. Consequently, the obvious
OXIDATION
questions were:
– What is tolerated as R3 and R4? Do the
two groups have to be the same or can
they be different?
– What is the ideal spacer?
– What is tolerated as A? (aryl? het-
eroaryl? cycloalkyl?; substitution pat-
tern?)
– Is it possible to replace the 4-pentyl-
benzoyl-unit?
On the basis of lead compound 4 we de-
cided to first tackle the questions with re-
spect to the A- and B-groups. Compounds
were prepared according to the general
pathway depicted in Scheme 1 and summa-
rized in Table 1. N,N-Dibutyl-ethylene-di-
amine (5) was alkylated in a reductive ami-
Fig. 1. Proposed hemoglobin degradation pathway within the food vacuole of P. falciparum [5]

PHARMACEUTICAL INDUSTRY IN SWITZERLAND
636
CHIMIA 2004, 58, No. 9
boxylic acid functionalities either behaved
neutral (12d,g) or led to a reduction in ac-
tivity (12h). Introduction of very lipophilic
substituents such as CF3-groups (12l,m) or
the butyl group (12o) also resulted in a
moderate to strong loss of activity on PMII.
With respect to inhibition of the human as-
partic proteases cathepsin D (CathD), in-
volved in intracellular digestion of proteins
[15] and cathepsin E (CathE), involved in
the generation of biologically active pep-
tides [16], the compounds showed an ac-
ceptable selectivity towards PMII. The se-
lectivity towards PMII is expected to in-
crease with increasing potency for the
inhibition of PMII.
In order to learn more about the influ-
ence of the conformationally restricted
amide bond at the second nitrogen atom of
the core unit, the ‘diamine’ compounds
13a–l (Table 4) were prepared. The synthe-
sis started from compound 5 (Scheme 1),
Fig. 2. Lead structures found by FRET high-throughput screening against plasmepsin II
nation procedure with an appropriate ben- able boronic acids or precursor 11b with which was reacted either with 4-bromo-
zaldehyde derivative 6 to give the interme- aryl- or heteroaryl bromides in a Suzuki re- benzaldehyde or with 4-formylphenyl-
diate secondary amine 7 [12] which subse- action [13] (Scheme 2). The experimental boronic acid under reductive amination
quently was acylated with substituted ben- set-up was optimized for parallel chemistry conditions in order to obtain analogues of
zoylchlorides 8 to give a first set of final on a 0.1 mmol scale. The Pd catalysts SK- compound 7. Coupling with 4-pentyl-ben-
compounds 9a–p [7d].
CC01-A and SK-CC02-A (commercially zaldehyde according to a literature protocol
Data collected in Table 1 show, that in available from Fluka or Solvias AG) were [7d] provided the ‘diamine’ Suzuki precur-
position R1 several structurally different chosen due to their superior stability, their sors, which were converted to the final
units might be tolerated, but a certain size broad scope/applicability and their excel- compounds 13a–l according to the proce-
of the substituent R1 seemed to be manda- lent solubility in dioxane which allowed for dures described in Scheme 2. Results sum-
tory. The rather small CF3 group resulted in easy addition to small vials via syringe/sep- marized in Table 4 show that the ‘diamine’
the lowest activity in all series. The most ac- tum standard-techniques in parallel chem- compounds 13a–l were generally less ac-
tive compound 9i contained the benzyloxy istry [14]. All final reaction products were tive than their direct analogues from Table
group. This unit usually is prone to meta- purified by preparative reversed phase 3, indicating the importance of the rotation-
bolic and chemical lability. We therefore HPLC and analyzed by LC-MS and occa- ally more restricted carboxamide group.
decided to focus on compound 9e, still sionally by NMR. Yields of final com- With respect to substituents on the second
showing reasonable inhibitory activity, in- pounds after purification were in the range aromatic portion the structure–activity pat-
corporating R1 into a biaryl unit often seen of 40–80%. IC50 values against PMII and tern seemed to be quite similar as compared
in drug candidates or marketed drugs. Fur- selected other aspartic proteases are given to the results of Table 3. The ‘diamine’-
thermore the biaryl unit offered a wide va- in Table 3. Compounds 12p–r, containing a based compounds showed a decreased se-
riety of chemical methods for the prepara- heteroaryl ring in position R1, exhibited lectivity towards PMII over CathD and
tion of diverse derivatives or isosteres. With PMII inhibitory activity in the same range CathE, compared to the amine–amide ana-
respect to group B it became obvious that as compound 9e with the unsubstituted logues (e.g. compare 12b to 13a).
the 4-pentyl-benzoyl unit was clearly supe- phenyl ring in the respective position. Sub-
Finally, preliminary investigations of
rior to the other substituents, e.g. 4-butyl- stitution with moderately polar, oxygen the spacer length between the two nitrogen
benzoyl, 4-propyl-benzoyl or 4-butoxy- containing groups on the second aryl ring atoms of the core unit were undertaken and
benzoyl, investigated in position B. Short- R1, resulted in increased activity towards the results are summarized in Table 5. Com-
ening the alkyl-chain or introducing PMII (see compounds 12b,c,e,f). Car- pounds 14a–j were again prepared accord-
heteroatoms led to substantial losses of in-
hibitory activity towards PMII.
Having set A–R1 as biphenyl and B–R2
as 4-pentyl-benzoyl respectively, we inves-
tigated the influence of the substituents R3
and R4. The small set of compounds pre-
pared is shown in Table 2. Comparison of
compounds 10a–d with 9e, the initial lead
structure, indicates the superiority of
lipophilic, sterically not crowded sub-
stituents in positions R3 and R4. The com-
pounds of Table 2 were prepared in an anal-
ogous sequence as described in Scheme 1.
We next focused our efforts on structural
variations in the biaryl unit. Therefore,
compounds 12a–r were prepared by react-
Fig. 3. General structure of plasmepsin II inhibitors
ing precursor 11a with commercially avail-

PHARMACEUTICAL INDUSTRY IN SWITZERLAND
637
CHIMIA 2004, 58, No. 9
Table 2. Variations in R3 and R4
Table 1. Biological data of the 1st set of plasmepsin II inhibitors
ing to chemical steps outlined in Schemes 1
and 2 by using the template N,N-dibutyl-
propylene diamine. The amide substituent
was kept constant as the 4-pentyl-benzoyl
group. The picture obtained with respect to
inhibitory potency towards PMII was un-
clear. The elongation of the linker to three
carbons had a different impact on the re- the compounds were tested against P. falci-
quirements for R1 as in the case of the C2 parum K1 strain in human red blood cells
linker. The results called for further investi- (RBC) [17]. A selection of compounds is
gations of the spacer length vs. substitution depicted in Table 6. Initial activities detect-
patterns at the groups A-R1 and B-R2.
ed in the RBC assay were below 1 µM
In order to get an idea about the anti- which is promising and clearly shows the
parasitic activities of our PMII inhibitors, potential of the structural class of PMII in-
hibitors described in this account. The IC50
shifts from the isolated enzyme assay to the
RBC assay are between 1 and 12 fold. In the
diamine series (compounds 13, Table 4) the
IC50 shifts are smaller as compared to the
shifts observed in the amine–amide series
(compounds 12, Table 3).
Conclusions/Outlook
We have presented the first results from
a program aiming at the identification of
small molecule, non-peptide inhibitors of
the P. falciparum parasite aspartic protease
PMII. A promising series of structurally
simple lead compounds, accessible by
straightforward chemistry, was identified.
First optimization efforts have been per-
formed, resulting in clear perspectives for
further work: The length and the nature of
the linker between the two nitrogen atoms
relative to the substituents at the remaining
Scheme 1. Synthesis of the 1st set of plasmepsin II inhibitors
positions has to be further investigated. The

PHARMACEUTICAL INDUSTRY IN SWITZERLAND
638
CHIMIA 2004, 58, No. 9
search for more drug-like replacements of
the dibutyl-amino unit and for isosteric re-
placements of the biaryl unit is far from fin-
ished. Another feature of the present in-
hibitors asking for further optimization is
the very lipophilic pentyl side chain which
might be metabolically labile. Such com-
pounds might exhibit a poor pharmacoki-
netic profile. We will further characterize
the PMII inhibitors in secondary assays in
whole cells in order to judge their potential
for side effects. In addition we will try to
solve the X-ray structure of PMII with an
inhibitor described above in order to per-
form further work on a rational basis of
molecular design.
Acknowledgements
Scheme 2. Synthesis of the 2nd set of plasmepsin II inhibitors. Pd. cat: SK-CC01-A or SK-CC02-A
from Solvias AG; a detailed description of the experimental procedure for the Suzuki reaction is giv-
en in [14].
The authors would like to thank Gabi Vor-
burger, Pascal Rebmann, Christine Berthion,
Anne Bodlenner, Annabelle Gillig, Heiner Bam-
merlin, Walter Schmutz and Loic Dayon for ex-
pert technical assistance used for the synthesis,
purification and characterization of compounds,
Dr. Daniel Bur for helpful and stimulating dis-
cussions, and Prof. Henri Ramuz for his moti-
vating constant support.
Table 3. Investigations of the biaryl unit
Received: June 18, 2004
[1] a) J.A. Radding, Annu. Rep.Med. Chem.
1999, 34, 159; b) D.F. Wirth, Annu. Rep.
Med. Chem. 1999, 34, 349; c) R.G. Ridley,
Nature 2002, 415, 686; d) J. Wiesner, R.
Ortmann, H. Jomaa, M. Schlitzer, Angew.
Chem. 2003, 115, 5432.
[2] a) A. Bell, Curr. Opin. Anti-infect. Invest.
Drugs 2000, 2, 63; b) P.J. Rosenthal, L.H.
Miller, in ‘Antimalarial Chemotherapy’,
Ed. P.J. Rosenthal, Humana Press, New
Jersey, 2001, p. 3.
[3] a) C. Berry, Curr. Opin. Drug Disc. Dev.
2000, 3, 624. b) D.M. Wyatt, C. Berry,
FEBS Lett. 2002, 513, 159.
[4] D.S. Fullerton, in ‘Textbook of Organic
Medicinal and Pharmaceutical Chem-
istry’, Eds. J.N. Delgado, W.A. Remers,
Lippincott-Raven Publishers, Philadel-
phia – New York, 1998, p. 235.
[5] a) J. Ziegler, R. Linck, D.W. Wright, Curr.
Med. Chem. 2001, 8, 171; b) D.E. Gold-
berg, A.F.G. Slater, A. Cerami, G.B. Hen-
derson, Proc. Natl. Acad. Sci. USA 1990,
87, 2931.
[6] A. Nezami, I. Luque, T. Kimura, Y. Kiso,
E. Freire, Biochemistry 2002, 41, 2273.
[7] A.M. Silva, A.Y. Lee, S.V. Gulnik, P. Ma-
jer, J. Collins, T.N. Bhat, P.J. Collins, R.E.
Cachau, K.E. Luker, I.Y. Gluzman, S.E.
Francis, A. Oksman, D.E. Goldberg, J.W.
Erickson, Proc. Natl. Acad. Sci. USA
1996, 93, 10034.
[8] a) D.J. Maly, L. Huang, J.A. Ellman,
ChemBioChem 2002, 3, 16; b) T. Mineo,
M.A. Avery, Chemtracts Org. Chem.
2001, 14, 518.
[9] A. Binggeli,V. Breu, D. Bur, W. Fischli, R.
Gueller, G. Hirth, H.-P. Märki, M. Müller,
C. Oefner, H. Stadler, E. Vieira, M. Wil-

PHARMACEUTICAL INDUSTRY IN SWITZERLAND
639
CHIMIA 2004, 58, No. 9
Table 4. Variations of the biaryl unit in the ‘diamine’ series
Table 5. Variation of the spacer length
helm, W. Wostl (F. Hoffmann-LaRoche
Ltd) WO 09709311, 1997.
Table 6. Results of the compounds tested in P. falciparum infected red
blood cells (assay according to [17])
[10] a) D.A. Carcache, S.R. Hörtner, A. Bertogg,
C. Binkert, D. Bur, H.-P. Märki, A. Dorn, F.
Diederich, ChemBioChem 2002, 11, 1137;
b) C. Boss, S. Richard-Bildstein, T. Weller,
W. Fischli, S. Meyer, C. Binkert, Curr. Med.
Chem. 2003, 10, 883; c) C. Boss, W. Fischli,
S. Meyer, S. Richard-Bildstein, T. Weller
(Actelion Pharmaceuticals Ltd) WO
00224649, 2002; d) C. Boss, W. Fischli,
S. Meyer, S. Richard-Bildstein, T. Weller
(Actelion Pharmaceuticals Ltd) WO
00238534, 2002.
[11] E.D. Matayoshi, G.T. Wang, G.A. Krafft, J.
Erickson, Science 1990, 247, 954.
[12] A.F. Abdel-Magid, K.G. Carson, B.D. Har-
ris, C.A. Maryanoff, R.D. Shah, J. Org.
Chem. 1996, 61, 3849.
[13] N. Miyaura, A. Suzuki, Chem. Rev. 1995,
95, 2457.
[14] a) A. Schnyder, F. Naud, M. Thommen, H.-
U. Blaser, sp2 2003, 32; b) SolviasApplica-
tion Note, 09/01, 2001; c)A. Schnyder, A.F.
Indolese, M. Studer, H.-U. Blaser, Angew.
Chem. Int. Ed. 2002, 41, 3668; d)A. Schny-
der, T. Aemmer,A.F. Indolese, U. Pittelkow,
M. Studer, Adv. Synth. Catal. 2002, 344.
[15] M.J. Hurley, L.B. Larsen,A.L. Kelly, P.L.H.
McSweeney, Int. Dairy Journal 2000, 10,
673.
[16] P.J. Tatnell, W.E. Lees, J. Kay, FEBS Lett.
1997, 408, 62.
[17] R.G. Ridley, H. Matile, C. Jaquet,A. Dorn,
W. Hofheinz, W. Leupin, R. Masciadri,
F.P. Theil, W.F. Richter, M.A. Girometta,
A. Guenzi, H. Urwyler, E. Gocke, J.M.
Potthast, M. Csato, A. Thomas, W. Peters,
Antimicrob. Agents Chemother. 1997, 41,
677.