Potent Inhibitors of Plasmodial Serine Hydroxymethyltransferase (SHMT) Featuring a Spirocyclic Scaffold

Article abstract of DOI:10.1002/cmdc.201800053

With the discovery that serine hydroxymethyltransferase (SHMT) is a druggable target for antimalarials, the aim of this study was to design novel inhibitors of this key enzyme in the folate biosynthesis cycle. Herein, 19 novel spirocyclic ligands based on

Full text of DOI:10.1002/cmdc.201800053

DOI: 10.1002/cmdc.201800053
Full Papers
Potent Inhibitors of Plasmodial Serine
Hydroxymethyltransferase (SHMT) Featuring a Spirocyclic
Scaffold
Geoffrey Schwertz,[a] Matthias C. Witschel,[b] Matthias Rottmann,[c, d]
Ubolsree Leartsakulpanich,[e] Penchit Chitnumsub,[e] Aritsara Jaruwat,[e]
Watcharee Amornwatcharapong,[f] Wanwipa Ittarat,[e] Anja Schꢀfer,[c, d] Raphael A. Aponte,[b]
Nils Trapp,[a] Pimchai Chaiyen,[f, g] and FranÅois Diederich*[a]
With the discovery that serine hydroxymethyltransferase
(SHMT) is a druggable target for antimalarials, the aim of this
study was to design novel inhibitors of this key enzyme in the
folate biosynthesis cycle. Herein, 19 novel spirocyclic ligands
based on either 2-indolinone or dihydroindene scaffolds and
featuring a pyrazolopyran core are reported. Strong target af-
finities for Plasmodium falciparum (Pf) SHMT (14–76 nm) and
cellular potencies in the low nanomolar range (165–334 nm)
were measured together with interesting selectivity against
human cytosolic SHMT1 (hSHMT1). Four co-crystal structures
with Plasmodium vivax (Pv) SHMT solved at 2.2–2.4 ꢁ resolution
revealed the key role of the vinylogous cyanamide for anchor-
ing ligands within the active site. The spirocyclic motif in the
molecules enforces the pyrazolopyran core to adopt a substan-
tially more curved conformation than that of previous non-spi-
rocyclic analogues. Finally, solvation of the spirocyclic lactam
ring of the receptor-bound ligands is discussed.
Introduction
The malaria landscape has changed considerably during the
last decade, with a 22% decrease in the number of cases and
a 50% decrease in the number of deaths attributed to malaria
from 2000 to 2015, by virtue of increased funding and a scale-
up of life-saving tools.[1] Nevertheless, malaria continues to
have a devastating global impact, with 216 million estimated
cases in 2016 and 445000 deaths.[2,3] Considerable efforts have
been invested toward the development of a preventive vac-
cine;[2,4] however, RTS,S/AS01 vaccine is the only candidate
that has led to encouraging results in late-stage phase III clini-
cal trials so far.[5] Malaria eradication endeavors are seriously
jeopardized by the widespread resistance of Plasmodium para-
sites to most antimalarials and, in particular, the emergence of
resistance against the first-line artemisinin-based combination
therapy.[6–8] Spurred by these concerning indicators, the scien-
tific community is actively seeking alternative treatments. In
analogy to artemisinin (1), potent inhibitors containing a phar-
macophoric peroxide bond have been investigated, such as ar-
temisone (2);[9] E209 (3);[10] artefenomel (4);[11] as well as artero-
lane (5)[12] and its regioisomeric analogues, including aminopi-
peridine 6[13] (for structures, see Section S1, Figure S1 in the
Supporting Information). Promising efficacy was also achieved
with non-endoperoxide scaffolds, including NITD609 (7),[14,15]
KAF156 (8),[16–20] DSM265 (9),[21,22] and aminoxadiazole-based li-
gands (Supporting Information, Section S1, Figure S2).[23]
[a] G. Schwertz, Dr. N. Trapp, Prof. Dr. F. Diederich
Laboratorium fꢀr Organische Chemie, ETH Zꢀrich, Vladimir-Prelog-Weg 3,
8093 Zꢀrich (Switzerland)
E-mail: diederich@org.chem.ethz.ch
[b] Dr. M. C. Witschel, Dr. R. A. Aponte
BASF SE, Carl-Bosch-Strasse 38, 67056 Ludwigshafen (Germany)
[c] Dr. M. Rottmann, A. Schꢁfer
Swiss Tropical and Public Health Institute (SwissTPH), Socinstrasse 57, 4051
Basel (Switzerland)
[d] Dr. M. Rottmann, A. Schꢁfer
Universitꢁt Basel, Petersplatz 1, 4003 Basel (Switzerland)
[e] Dr. U. Leartsakulpanich, Dr. P. Chitnumsub, A. Jaruwat, W. Ittarat
National Center for Genetic Engineering and Biotechnology, 113 Thailand
Science Park, Phahonyothin Road, Pathumthani 12120 (Thailand)
Surprisingly, little attention has been paid to the inhibition
of serine hydroxymethyltransferase (SHMT) for developing anti-
malarials. SHMT is involved in the folate biosynthesis cycle and
is a pyridoxal 5’-phosphate (PLP)-dependent enzyme that cata-
lyzes the reversible interconversion of serine and glycine with
tetrahydrofolate (H4F) that leads to the formation of the one-
carbon carrier 5,10-methylenetetrahydrofolate (5,10-CH2-
H4F).[24–30] To date, only a limited number of active substances
have been tested against Plasmodium falciparum (Pf) and Plas-
modium vivax (Pv) SHMT.[27,31] Small libraries of known antifo-
[f] W. Amornwatcharapong, Prof. Dr. P. Chaiyen
Department of Biochemistry and Center for Excellence in Protein and
Enzyme Technology, Faculty of Science, Mahidol University, 272 Rama VI
Road, Bangkok 10400 (Thailand)
[g] Prof. Dr. P. Chaiyen
Department of Biomolecular Science and Engineering, School of Biomolecu-
lar Science & Engineering, Vidyasirimedhi Institute of Science and Technolo-
gy (VISTEC), Wangchan Valley, Rayong 21210 (Thailand)
Supporting information and the ORCID identification number(s) for the
author(s) of this article can be found under:
https://doi.org/10.1002/cmdc.201800053.
ChemMedChem 2018, 13, 1 – 14
1
ꢂ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
lates, such as methotrexate (10), pyrimethamine (11), and the
triazine-based ligand WR99210 (12), together with related ana-
logues 13–17 (Supporting Information, Section S1, Figure S3)
have been screened. However, these ligands only displayed
target IC50 (half-maximal inhibitory concentration) affinities in
the high micromolar range against PfSHMT; these values are
series were highly selective against human SHMT1 and at-
tempted to rationalize these results by structural analysis. The
binding mode of these novel active substances was elucidated
by X-ray co-crystal structures with PvSHMT and resembles to
the one of their non-spiro analogues. Nonetheless, subtle dif-
ferences are pinpointed and analyzed.
close to the KM value of H4F (KM PfSHMT=190Æ20 mm).[25]
H4F
Amino acid analogues, such as d-serine, d-threonine, and thio-
semicarbazide, were also investigated, but no significant inhibi- Results and Discussion
tion was observed either.[27] We recently demonstrated that
Ligand design
pyrazolopyran-based inhibitors, such as carboxylate (Æ)-18
(Figure 1), could inhibit PfSHMT in the lower nanomolar IC50
Molecular modeling with MOLOC,[35] by using previously ob-
tained co-crystal structures with pyrazolopyran-based inhibi-
tors,[32–34] was used to guide the structural modifications of the
ligands. As shown in the schematic representation of the bind-
ing mode of carboxylate (+)-18 at the active site of PvSHMT
(PDB ID: 5GVN, 2.3 ꢁ resolution) (Figure 1), two sub-pockets
are distinguishable: the pyrazolopyran core binds into the
pterin binding pocket, which normally hosts the pterin core of
H4F (PDB ID: 4OYT, 2.4 ꢁ),[30] whereas the bicyclic scaffold occu-
pies the para-aminobenzoate (pABA) channel. The ligands re-
ported herein feature a spirocyclic motif on the 4-position of
the pyran ring to address a small lipophilic pocket within the
pterin binding site (see below). The choice of fragments occu-
pying the pABA channel was guided by the results of our pre-
vious structure–activity relationship (SAR) studies.[33,34]
2-Indolinone series
To diversify the portfolio of SHMT inhibitors, it was envisioned
to install a 2-indolinone scaffold onto the pyrazolopyran core.
A significant number of active substances feature a 2-indoli-
none scaffold;[36] a prominent example is antimalarial 7[14] dis-
cussed in the Introduction. This scaffold has also been success-
fully incorporated into drugs for cancer treatment;[37] in tyro-
sine kinase inhibitors;[38–40] and is found in natural products,[41]
such as in the welwitindolinone alkaloids.[42] In addition, this
motif has been used to conformationally lock biologically
active molecules in the form of spirocycles.[43]
Figure 1. Schematic depiction of the binding mode of pyrazolopyran-based
ligand (+)-18, as observed in a co-crystal structure analysis (Protein Data
Bank (PDB) ID: 5GVN, 2.3 ꢁ).[33] Intermolecular hydrogen bonding is indicated
by red dashed lines. pABA: para-aminobenzoate.
range and were highly effective (half-maximal effective concen-
tration (EC50)) in cell-based assays against the PfNF54
strain.[32–34] Several co-crystal structures with PvSHMT showed
that these highly preorganized ligands nicely occupy the H4F
binding pocket of the enzyme. Finally, we validated SHMT as a
druggable target for antimalarials by achieving significant in
vivo efficacy in a P. falciparum severe combined immunodefi-
ciency (SCID) mouse model.[33]
Four 2-indolinone derivatives were designed, with terminal
ortho-tolyl ((Æ)-19), 3-pyridyl ((Æ)-20), 4-pyridyl ((Æ)-21), and
aryl sulfonamide ((Æ)-22) rings (Table 1). MOLOC modeling of
the SHMT-bound 3-pyridyl analogue, (+)-20, suggested that
two favorable hydrogen-bonding interactions could be formed
through the 2-indolinone motif of (+)-20 with the side chain
of Asn354 and the backbone C=O of Lys355 (Supporting Infor-
mation, Section S4, Figure S10). All other key interactions of
the core with the surrounding amino acids were retained.
The synthesis started with a selective bromination[44] of 7-
(trifluoromethyl)isatin 23 to yield 24 (Scheme 1), which was
converted quantitatively into acetal 25 prior to cross-coupling
with arylboronic acids of interest to give the bicyclic building
blocks 26a–d. Interestingly, masking of the free ketocarbonyl
group of isatin 24 was required for the successful preparation
of the biaryl fragments, as already reported by Wang and co-
workers.[45] Isatin derivatives 27a–d were prepared through
cleavage of the acetal moiety with concentrated HCl, and sub-
Following these original studies, we sought to identify a
backup lead for the initial pyrazolopyran-based ligands and
report herein two series of molecules containing a spirocyclic
motif and featuring a pyrazolopyran core. Ligands incorporat-
ing a 2-indolinone scaffold exhibited activities in the two-digit
micromolar range, which indicated that polar fragments in the
vicinity of Asn354 and Arg371 were not tolerated, and that the
lactam ring of these ligands was presumably not properly sol-
vated if bound to the receptor. In contrast, several spiro-dihy-
droindene derivatives displayed IC50 and EC50 values in the
lower nanomolar range. We found that two ligands of this
&
ChemMedChem 2018, 13, 1 – 14
www.chemmedchem.org
2
ꢂ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
Table 1. Biological activities of ligands (Æ)-19–22 and (Æ)-29–31 containing a 2-indolinone scaffold.
Compound
EC50 PfNF54 [mm]
IC50 PfSHMT [mm][a]
IC50 AtSHMT [mm]
R
(Æ)-29
(Æ)-30
>34
>250
>250
12% inhibition at 100 mm
22.7
14.3
(Æ)-19
(Æ)-20
(Æ)-21
(Æ)-22
(Æ)-31
22.1
22.8
11.6
16.7
1.5
5.2Æ0.4
1.2
>250
48% inhibition at 100 mm
>250
34.2
n.d.
9.6
n.d.[b]
>250
[a] Data are the mean ÆSD of n=3 experiments performed in duplicate. [b] n.d.: not determined.
sequent reaction with malononitrile and 3-methyl-1H-pyrazol-
5(4H)-one (28) led to the one-pot synthesis of pyrazolopyran
analogues (Æ)-19–22 (Scheme 1).[46,47] Additionally, indoline-
2,3-dione and brominated 24 were also engaged in multicom-
ponent reactions to yield ligands (Æ)-29 and (Æ)-30 (Table 1).
All ligands were tested against PfSHMT and the PfNF54
strain, and on Arabidopsis thaliana (At) SHMT[33] to refine the
SAR. The core alone ((Æ)-29) did not exhibit any efficacy in any
assays, whereas its functionalization at both meta positions led
to minor improvements (Table 1). With the exception of ortho-
tolyl derivative (Æ)-19, which could decrease the enzymatic ac-
tivities in the low micromolar range (IC50 PfSHMT=5.2 mm; IC50
AtSHMT=1.2 mm), all other ligands ((Æ)-20–22 and (Æ)-30)
were inactive in the target assays. Cell-based efficacies were
also unsatisfactory, with measured EC50 values in the two-digit
micromolar range. N-Methylation of (Æ)-21 resulted in (Æ)-31
with sevenfold enhanced potency in the cell-based assay
(EC50 =1.5 mm), which might be the result of better cell perme-
ation. However, target affinities still remained in the high mi-
cromolar range (Table 1).
The poor activity values observed with this series reflect that
a polar fragment at this position of the molecule, such as the
lactam ring in the 2-indolinone analogues, is not suitable for
binding to SHMT. Nonetheless, these disappointing results
could be rationalized by retrospective modeling of the newly
obtained co-crystal structures with two spiro-dihydroindene
derivatives (see below).
Spiro-dihydroindene series
Subsequently to the low affinities observed with the series of
molecules described above, we set out to prepare a molecule
with a gem-dimethyl moiety (in (Æ)-32), which resembles the
isopropyl fragment in the previous series,[33] instead of the car-
bonyl group (Scheme 2). It is worth noting that the gem-di-
methyl group is widely found in natural products and is used
ChemMedChem 2018, 13, 1 – 14
www.chemmedchem.org
3
ꢂ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
has a bright-orange color, which is characteristic of push–pull
p-electron conjugation, and its structure has been confirmed
by X-ray crystallography (Supporting Information, Sec-
tion S5.3.1, Figure S21).
Unfortunately, all attempts to convert 33 with 28 into target
molecule (Æ)-32 remained unsuccessful (Scheme 2 and Sup-
porting Information, Section S2.3, Table S1). Several reaction
conditions were investigated; however, changing the base, sol-
vent, and temperature had no effect on the outcome of the re-
action. In fact, starting material 33 remained unreacted, regard-
less of the reaction conditions, and installation of different N-
protecting groups did not affect its reactivity. Work on this
series was therefore halted.
Alternatively, dihydroindene 39 was prepared in two steps
from dihydroindenone 40 via 41 (Scheme 3A). Unlike indoline
33, dinitrile 39 reacted with 28 to yield ligand (Æ)-42, the
structure of which was confirmed by X-ray crystallography
(Supporting Information, Section S5.3.2, Figure S22). Notably,
steric hindrance caused by the gem-dimethyl moiety precluded
the use of a multicomponent reaction for core formation and
required the isolation of dinitrile 39 toward the synthesis of
(Æ)-42. Encouraged by this outcome, several derivatives of (Æ)-
42 were designed and synthesized, starting from 4-pyridyl ana-
logue (Æ)-43 (Scheme 3B). This terminal 4-pyridyl fragment
was chosen, in agreement with the results of our previous SAR
studies.[33] Brominated 44 was efficiently bis-methylated alpha
to the keto group to give 45, and subsequent cross-coupling
with 4-pyridylboronic acid (46) yielded 47 (Scheme 3B). Con-
formationally locked analogue (Æ)-43 was then prepared via
48.
As we reported previously,[33] incorporation of a meta sub-
stituent on the phenyl ring departing from the stereogenic
center is crucial to achieve high potency paired with favorable
physicochemical properties. Therefore, fluorine was introduced
on the 4-position of the fused phenyl ring and 4-fluorodihy-
droindenone scaffold 49 was synthesized in six steps from
benzylic alcohol 50 (Scheme 4). The latter was first converted
quantitatively into sulfonate 51 and further reacted with dieth-
yl malonate in the presence of sodium hydride to yield 52.
Saponification gave diacid 53 in high yield, after which mono-
decarboxylation occurred upon heating 53 in boiling 1,4-diox-
Scheme 1. Representative synthesis of 2-indolinone analogues (Æ)-19–22
(for Ar residues, see Table 1): a) Br2, AcOH, 258C, 23 h, 93%; b) ethylene
glycol, p-toluenesulfonic acid (pTsOH; cat.), toluene, reflux, 23 h, 99%; c) ArÀ
B(OH)2, [Pd(PPh3)2Cl2] (3.0 mol%), NaHCO3, dimethoxyethane (DME)/H2O 1:1,
reflux, 3 h, 41–92%; d) conc. HCl, MeOH, reflux, 2 h, 58–98%; e) malononi-
trile, 28, Et3N (cat.), EtOH, reflux, 2 h, 51–87%.
by medicinal chemists to modulate a broad range of proper-
ties, such as target affinity, metabolic stability, and the confor-
mation of small molecules.[48] The synthesis of the dinitrile pre-
cursor 33, starting with an intramolecular aldol-type condensa-
tion of diacid 34 to give indole 35 (Scheme 2), has been re-
ported in the literature.[49] Selective deprotection of the enol
by employing sodium sulfite led to dihydroindol-3-one 36. The
latter was treated with sodium hydride and iodomethane;
however, deprotected gem-dimethyl-indol-3-one 37 was ob-
tained directly in 62% yield, instead of the intended product
38; this was likely to be due to contamination with traces of
water based on the harsh reaction conditions. Subsequently, a
Knoevenagel condensation with malononitrile was performed
on 37 in boiling pyridine to obtain dinitrile 33. This molecule
Scheme 2. Attempt to prepare indoline analogue (Æ)-32: a) NaOAc, Ac2O, reflux, 5 h, 64%; b) Na2SO3 in H2O, EtOH, reflux, 2 h, 88%; c) NaH, MeI, toluene,
reflux, 24 h, 62%; d) malononitrile, pyridine, reflux, 62 h, 57%.
&
ChemMedChem 2018, 13, 1 – 14
www.chemmedchem.org
4
ꢂ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
Scheme 3. Synthesis of spiro-dihydroindene analogues (Æ)-42 (A) and (Æ)-43 (B): a) 1) NaH, DMF, 238C, 1 h; 2) MeI, DMF, 238C, 1.5 h, 67%; b) malononitrile,
pyridine, reflux, 64–82 h, 47–89%; c) 28, piperidine, EtOH/1,4-dioxane 1:1, microwave irradiation (mw), 658C, 21–40 h, 50–63%; d) NaH, MeI, THF, 238C, 16 h,
95%; e) 46, [Pd(PPh3)2Cl2] (5.0 mol%), K2CO3, 1,4-dioxane/H2O 5:1, 1008C, 16 h, 82%.
Scheme 4. Preparation of building block 49 and ligand (Æ)-57: a) Methylsulfonyl chloride, Et3N, THF, 08C, 30 min, quant.; b) NaH, diethyl malonate, THF, reflux,
16 h, 61%; c) 50% NaOH in H2O, EtOH, reflux, 2.5 h, 94%; d) 1,4-dioxane, reflux, 22 h, quant.; e) 1) (COCl)2, CH2Cl2, 238C, 4 h; 2) AlCl3, CH2Cl2, reflux, 3.5 h, 65%;
f) NaH, MeI, THF, 238C, 16 h, 87%; g) 46, [Pd(PPh3)2Cl2] (5.0 mol%), K2CO3, 1,4-dioxane/H2O 5:1, 1008C, 16 h, 76%; h) malononitrile, pyridine, reflux, 64 h, 77%;
i) 28, piperidine, EtOH/1,4-dioxane 1:1, mw, 65 8C, 40 h, 70%.
ane to quantitatively form propionic acid 54. Intermediate 54
was first treated with oxalyl chloride, followed by aluminum
chloride to prepare the 2,3-dihydroindenone scaffold 55
through a Friedel–Crafts-type acylation.[50] Finally, compound
49 was synthesized by bis-methylation of 55. Building block 49
was subsequently subjected to Suzuki–Miyaura cross-coupling
conditions[51] with 46, leading to biaryl 56, which was convert-
ed into ligand (Æ)-57 via dinitrile 58 (Scheme 4). Enantiomer
separation by HPLC with a chiral stationary phase (CSP) al-
lowed us to obtain enantiomerically pure (+)-57 and (À)-57
(Supporting Information, Section S2.5).
isolate enantiopure (+)-63 and (À)-63, which were deprotect-
ed to obtain (+)-59 and (À)-59, respectively (Supporting Infor-
mation, Sections S2.4 and 2.5).
The spiro-dihydroindene ligands presented herein were
tested against At- and PfSHMT, and in cell-based assays against
the PfNF54 strain (Table 2). The core shows good sub-micromo-
lar affinity for PfSHMT (IC50 (Æ)-42=0.201 mm), which is in con-
trast to the inactive 2-indolinone analogues (Table 1). Interest-
ingly, a considerable affinity gain was measured for (Æ)-43,
which featured a terminal 4-pyridyl moiety (IC50 (Æ)-43=
0.030 mm); this highlights the importance of filling the pABA
channel to inhibit PfSHMT activity efficiently. Introducing a fluo-
ride into ligand (Æ)-57 did not substantially affect the target
affinity, whereas replacing the pyridyl ring with a carboxypiper-
idyl moiety in compounds (Æ)-59 and (Æ)-63 also resulted in
activity in the two-digit nanomolar range (Table 2). These re-
sults were nicely underpinned by the inhibitory constants
against AtSHMT, although the affinity values were generally
one order of magnitude lower (Table 2).
Additionally, we set out to prepare carboxylate (Æ)-59 by
performing a Buchwald–Hartwig amination,[52] involving fluo-
roindene 49 and benzyl (piperidin-4-yl)acetate (60), which
gave intermediate 61 (Scheme 5). The latter was further react-
ed with malononitrile in boiling pyridine to yield dinitrile 62.
Subsequently, spirocyclic ligand (Æ)-63 was prepared in a mod-
erate yield and palladium-catalyzed hydrogenation gave the
target molecule (Æ)-59. Chiral HPLC separation allowed us to
ChemMedChem 2018, 13, 1 – 14
www.chemmedchem.org
5
ꢂ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
Scheme 5. Preparation of carboxylate (Æ)-59: a) 60, [Pd2(dba)3] (2.0 mol%; dba: dibenzylidenacetone), 2-dicyclohexylphosphino-2’,4’,6’-triisopropylbiphenyl (X-
Phos), Cs2CO3, 1,4-dioxane, 1108C, 16 h, quant.; b) malononitrile, pyridine, reflux, 96 h, 79%; c) 28, piperidine, EtOH/1,4-dioxane 1:1, mw, 65 8C, 90 h, 35%;
d) H2, [Pd/C], MeOH, 238C, 4 h, 97%.
Table 2. Biological activities of spiro-dihydroindene ligands.
Compound
EC50 PfNF54 [mm]
IC50 PfSHMT[mm][a]
IC50 AtSHMT[mm][a]
IC50 hSHMT1[mm][a]
SF[b]
R
(Æ)-42
(Æ)-43
25.5
0.201Æ0.004
0.030Æ0.001
2.86Æ0.280
8.86Æ0.68
44.1
1.6
2.358
0.173Æ0.009
0.047Æ0.002
(Æ)-57
(+)-57
(À)-57
0.619
0.334
10.213
0.039Æ0.001
0.014Æ0.002
0.339Æ0.006
0.195Æ0.004
0.281Æ0.016
2.27Æ0.156
0.047Æ0.001
0.027Æ0.001
0.125Æ0.003
1.2
1.9
0.4
(Æ)-59
(+)-59
(À)-59
2.131
0.861
>21.5
0.061Æ0.001
0.043Æ0.002
2.51Æ0.03
0.112Æ0.006
0.163Æ0.007
4.69Æ0.200
0.057Æ0.001
0.019Æ0.001
0.440Æ0.010
0.9
0.4
0.2
(Æ)-63
(+)-63
(À)-63
0.252
0.165
2.487
0.076Æ0.002
0.060Æ0.002
>25.0
0.648Æ0.038
0.494Æ0.043
4.79Æ0.861
0.029Æ0.001
0.029Æ0.001
0.248Æ0.005
0.4
0.5
<0.1
(Æ)-64
>36.0
4.36Æ0.72
47% inhibition at 100 mm
>250
57.3
[a] Data are the mean ÆSD of n=3 experiments performed in duplicate. [b] Selectivity factor: (IC50 hSHMT1)/(IC50 PfSHMT).
The cell-based SAR is in good agreement with the target af-
finities (Table 2). More than 10-fold potency improvement was
obtained upon occupying the pABA channel with a 4-pyridyl
fragment (EC50 (Æ)-42=25.5 mm>EC50 (Æ)-43=2.358 mm). The
addition of a fluoride to the meta position of the fused phenyl
ring had a significant impact, as exemplified by the sub-micro-
molar potency measured for (Æ)-57 (EC50 =0.619 mm). A better
EC50 value was measured for benzyl ester (Æ)-63 (EC50 =
0.252 mm), presumably as a result of improved cellular permea-
tion due to its significant lipophilic character, because the cor-
responding carboxylate (Æ)-59 is 10-fold less potent (EC50 =
2.131 mm). It is noteworthy that the gem-dimethyl moiety on
the dihydroindene scaffold is essential for activity because its
removal led to a considerable loss of affinity (ligand (Æ)-64:
IC50 PfSHMT=4.36 mm; EC50 >36 mm; Table 2).
&
ChemMedChem 2018, 13, 1 – 14
www.chemmedchem.org
6
ꢂ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
Chiral recognition is effective for this class of molecules
(Table 2). The (+)-enantiomer is much more active on the
target than the (À)-enantiomer in all head-to-head comparison
pairs (e.g., (À)-57: 339 nm>(+)-57: 14 nm and (À)-59:
2510 nm>(+)-59: 43 nm). The strong preference for the
(+)-enantiomer was also observed in the cellular assay, in
which ligand (+)-63 displayed the best potency of the series
(EC50 =165 nm). Finally, no adverse cytotoxicity was measured
for enantiopure (+)-57, (+)-59, and (+)-63 on a rat myoblast
cell line (L6) with IC50 values ranging from 5.53 to >100 mm
(Supporting Information, Section S3, Table S2).
Crystal structure determination of PvSHMT in complex with
(+)-57, (+)-59, (+)-65, and (+)-66
Ternary complexes of PvSHMT, PLP-Gly, and either (+)-57 or
(+)-59 were crystallized by using the microbatch method. Ad-
ditionally, two previously reported pyrazolopyran-based li-
gands ((+)-65 or (Æ)-66, see Figure 2) were also co-crystallized
Figure 2. Molecular structures and selected biological data of ligands (+)-65
and (Æ)-66.[33]
with PvSHMT.[33] These two complexes with non-spirocyclic an-
alogues provided useful comparisons to elaborate the differen-
ces in binding upon spirocycle formation, as further discussed
below. The co-crystals diffracted to 2.4, 2.3, 2.2, and 2.2 ꢁ reso-
lution, respectively, and belonged to the C2 space group. The
structures were solved by molecular replacement by using the
coordinates of a chain A protomer of PvSHMT (PDB ID: 4OYT)
as the template.[30] Importantly, both active sites of PvSHMT
homodimer were populated with a ligand in each complex
and the conformations of the ligands occupying the binding
sites were very similar (Supporting Information, Section S5.1,
Figures S13 and S14).
Figure 3. Co-crystal structure of PvSHMT with PLP-Gly and (+)-59 (PDB ID:
5YG4, 2.3 ꢁ). A) Overall binding mode, and B) water network near Glu56. The
water molecules are represented as red spheres. The mesh spans the
volume of the binding pocket. Distances are given in ꢁ. Color code: Cprotein
gray, C(+)-59 gold, CPLP-Gly salmon, F light cyan, N blue, O red, P orange, S
yellow.
mide with the side chain of Thr357 and the backbone C=O of
Leu124 and Gly128 (Figure 3A and Supporting Information,
Section S5.1, Figure S15A). In addition, the nitrile moiety of the
ligands is engaged in favorable parallel stacking with the side
chain of Asn356 (d(CꢀN···Csp2 )=3.8–4.1 ꢁ).[53] The sp3-oxygen
atom of the pyran ring is at a short distance from Thr183, es-
tablishing a favorable hydrogen bond, but it is also in proximi-
ty to the backbone C=O of Gly128 (d(O···O=C)=3.4 ꢁ), which
presumably results in a slightly repulsive interaction.
Binding mode of spiro-dihydroindene analogues (+)-57 and
(+)-59
From the two complexes, it can be derived that (+)-57 and
(+)-59 are (S)-configured (Figure 3 and Supporting Information,
Section S5.1, Figure S15). The overall conformation of the spiro-
cyclic system in each co-crystal structure is in good agreement
with the small-molecule X-ray crystal structure of (Æ)-42 (Sup-
porting Information, Section S5.3.2, Figure S22). The binding
mode of these ligands resembles that of previous pyrazolopyr-
an-based molecules;[32–34] however, subtle differences are dis-
tinguishable. The pyrazolopyran core forms three strong hy-
drogen-bonding interactions through the vinylogous cyana-
The small lateral pocket lined by the Cys364 loop is not
properly filled by the meta-fluoro-substituted phenyl ring of
the dihydroindene fragment of (+)-57 and (+)-59, leading to a
small void. In the complex with (+)-59, a water molecule (W1)
occupies this small cavity, solvates the CꢀN moiety, and estab-
ChemMedChem 2018, 13, 1 – 14
www.chemmedchem.org
7
ꢂ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
lishes a short orthogonal dipolar interaction with the Csp2ÀF
bond[54,55] (d(O···Csp2ÀF=3.1 ꢁ), angle (OÀCsp2ÀF)=788; Fig-
ure 3A). The lower resolution of the co-crystal structure with
(+)-57 did not allow refinement of water molecules with high
accuracy in this small pocket (Supporting Information, Sec-
tion S5.1, Figure S15A).
The pyrazole ring is oriented in such a way that the pyrazole
NH cannot interact directly with Glu56 (Figure 3). Indeed, the
two X-ray co-crystal structures revealed that the pyrazolopyran
core was not aligned with Glu56 to form a strong direct ionic
hydrogen bond with its side chain. In the structure with
(+)-57, only a weak hydrogen-bonding contact, d(N···ÀOÀC)=
3.6 ꢁ, is observed (Supporting Information, Section S5.1, Fig-
ure S15A). In the structure with (+)-59, the pyrazole N2ÀH of
the ligand and Glu56 are at a distance of 4.2 ꢁ (d(N···ÀOÀC))
and an interaction between the two centers is mediated by hy-
drogen bridging to water molecule W2 (Figure 3). The latter is
part of a small linear water network involving W4 and W5 and
undergoes four bonding interactions in a tetrahedral arrange-
ment, which fix it tightly in the crystal structure. Glu56 is fur-
ther solvated by W3, which establishes additional strong hy-
drogen-bonding interactions with the phosphate moiety of the
PLP-Gly Schiff base and His129 (Figure 3). The terminal sub-
stituents attached to the spiro-dihydroindene core nicely
occupy the pABA channel (Figure 3 and Supporting Informa-
tion, Section S5.1, Figure S15). In addition, the carboxylate
moiety at the termini of ligand (+)-59 is solvated by a water
molecule (W6), which is at short distance from a second water
molecule (W7, d(O···O)=2.9 ꢁ; Figure 3B).
Figure 4. Overlay of the PvSHMT–ligand complexes obtained with (+)-59
(PDB ID: 5YG4, 2.3 ꢁ) and (+)-65 (PDB ID: 5YFZ, 2.2 ꢁ). The water molecules
are represented as red spheres. PLP-Gly is omitted for clarity. The mesh
spans the volume of the binding pocket. Distances are given in ꢁ. Color
code: CPvSHMT–(+)-59 gray, CPvSHMT–(+)-65 green, C(+)-59 gold, C(+)-65 pink, F light
cyan, N blue, O red, S yellow.
zolopyran core than that in (+)-65 (Supporting Information,
Section S5.1, Figure S19). Indeed, the angle between the plane
spanned by the vinylogous cyanamide and that spanned by
the pyrazole ring is equal to 138 in (+)-65, whereas a substan-
tially larger angle of 238 is measured for spirocyclic analogues
(Æ)-42 and (+)-59. Consequently, the pyrazole fragment of
(+)-59 does not point toward Glu56 and deviates by 12.58
compared with the original binding mode (Figure 4).
In both complexes, the gem-dimethyl fragment on the spiro-
dihydroindene scaffold and the methyl moiety on the pyrazole
ring provide optimal filling of the binding pocket (Supporting
Information, Section S5.1, Figures S15B and S16A). The methyl
group is loosely surrounded by the lipophilic side chains of
Tyr63 and Phe266, while the gem-dimethyl moiety interacts
with both Tyr64 and the polarizable hydrogen-bonding array
between the side chain of Arg371 and the carboxylate of the
PLP-Gly Schiff base.
Finally, the positioning of one CH3 group (from the gem-di-
methyl moiety in (+)-59 and the isopropyl moiety in (+)-65) in
the back of the pocket varies markedly. Subsequently, Tyr64
swings away (shift of the phenolic O atom by 1.6 ꢁ) to com-
pensate for the small void created by the different positioning
of this methyl moiety in the PvSHMT–(+)-59 complex
(Figure 4). These two co-crystal structures underline the para-
mount importance of the vinylogous cyanamide in the binding
mode of both ligand classes to PvSHMT. Additionally, they sug-
gest that the pyrazole fragment can possibly be replaced by al-
ternative heterocyclic moieties without affecting the target af-
finity much (IC50 (+)-59: 43 nm; IC50 (+)-65: 110 nm). Thereby,
interactions of the ligand core with the side chain of Glu56, as
in the non-spirocyclic series, are not essential for strong bind-
ing of the spirocyclic scaffolds.
Comparison between (+)-59 and its non-spirocyclic
analogue (+)-65
Superposition of the PvSHMT-(+)-59 complex with the one of
its closely related non-spirocyclic analogue (+)-65 revealed an
overall good match of the different ligands, but not through-
out the entire structures (Figure 4). On one hand, the terminal
carboxylate moiety and vinylogous cyanamide overlay perfect-
ly; the latter clearly anchors the ligands within the PvSHMT
binding pocket. On the other hand, the phenyl rings departing
from the stereogenic center adopt two distinct orientations. In
(+)-65, the phenyl ring is oriented toward the lateral cavity
lined by the Cys364 loop, whereas in (+)-59, the fused phenyl
ring is rotated by 19.68 with respect to the plane of the afore-
mentioned moiety, and W1 occupies the cavity instead
(Figure 4). Additionally, the spirocyclic motif in (+)-59 enforces
a more tetrahedral arrangement at the stereogenic center,
leading to a significantly less planar conformation of the pyra-
Water network solvating PvSHMT–ligand complexes
As detailed above, the terminal carboxylate of spirocyclic ana-
logue (+)-59 is solvated by a water molecule (W6) that does
not interact with the protein (Figure 3B). Conversely, in the co-
crystal structures with (+)-65 and (+)-66, the area around the
terminal carboxylate at the opening of the pABA channel is
solvated by an acyclic water network W8 to W15 (Figure 5).
&
ChemMedChem 2018, 13, 1 – 14
www.chemmedchem.org
8
ꢂ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
hSHMT1. For instance, there is a 23-fold difference in affinity
for hSHMT1 between (+)-59 and (À)-59 (IC50 hSHMT1=0.019
and 0.440 mm, respectively), whereas the affinity discrepancy is
considerably larger (58-fold) for PfSHMT (IC50 PfSHMT=0.043
and 2.51 mm, respectively).
To rationalize these results, we superimposed the co-crystal
structure of (+)-59 with PvSHMT, with that of rabbit (r) SHMT1
(PDB ID: 1LS3)[64] and human mitochondria SHMT2 (PDB ID:
5V7I), which was recently reported by Ducker and co-workers
(Figure 6).[62] We pinpointed three major differences between
Figure 5. Superposition of the PvSHMT–ligand complexes with (+)-59 (gold;
PDB ID: 5YG4, 2.3 ꢁ), (+)-65 (pink; PDB ID: 5YFZ, 2.2 ꢁ), and (+)-66 (cyan;
PDB ID: 5YG2, 2.2 ꢁ). Water molecules are represented as spheres and col-
ored as their corresponding ligand. PLP-Gly is omitted for clarity. The dashes
represent hydrogen-bonding interactions. Color code: Cprotein–(+)-59 gold,
C
protein–(+)-65 gray, Cprotein–(+)-66 cyan, F light cyan, N blue, O red, S yellow.
Two of these water molecules, W11 and W12, encircled in red,
solvate the carboxylate and stabilize the complexes with
PvSHMT (Figure 5 and Supporting Information, Section S5.1,
Figures S17B and S18B).[56–58] In the co-crystal structure with
(+)-59, water molecules solvating residues of the protein, such
as the backbone C=O of Phe266 and Val141 (W10 and W13, re-
spectively), and the side chains of Tyr110/Ser268 together with
the backbone C=O of Lys140 (W15) are conserved; however,
no water molecules are observed at the encircled positions
(Supporting Information, Section S5.1, Figures S16B).
Figure 6. Superposition of the PvSHMT–(+)-59 complex (gray; PDB ID: 5YG4,
2.3 ꢁ) with rabbit SHMT1 (purple; from PDB ID: 1LS3, 2.7 ꢁ)[64] and human
SHMT2 (cyan; from PDB ID: 5V7I, 2.5 ꢁ).[62] Only relevant residues are shown
as stick representation. The mesh spans the volume of the binding pocket.
PLP-Gly is omitted for clarity. Distances are given in ꢁ. Color code: C(+)-59
gold, F light cyan, N blue, O red, S yellow.
Notably, Lys139 points into a small pocket at the periphery
of the pABA channel, establishes several hydrogen bonds with
water molecules and/or amino acid residues, and adopts two
slightly different orientations independently of the nature of
the water cluster (Figure 5). Thus, this residue is not available
for interacting with the ligand bound in the active site, which
possibly explains why it does not rotate toward any ligands to
interact with their carboxylate side chain, as suggested by mo-
lecular dynamics.[33]
the three structures that could possibly explain the observed
selectivity. First, the threonine residue (Thr183) in Plasmodial
SHMT, which establishes a strong hydrogen bond with the
pyran ring of (+)-59, is replaced by a serine residue (Ser175/
226) in mammalian SHMTs that cannot interact with the ligand
in the observed conformation. Second, the position of the as-
paragine side chain fluctuates notably. In the rSHMT1 structure,
Asn347 is at a sub-van der Waals distance to the ligand, which
leads to potential repulsive interactions, whereas Asn410 from
hSHMT2 is too far away to establish any attractive interaction
(Figure 6). Conversely, Asn356 from PvSHMT is at an ideal posi-
tion for interacting with (+)-59 and avoiding repulsion. Finally,
Thr411 from hSHMT2 is turned away from the ligand, which
prevents the formation of a hydrogen bond with the nitrile
moiety of the ligand. Although these structural differences
may explain the observed selectivity, it is important to empha-
size that our analysis is based on the overlay of co-crystal
structures, and a more detailed molecular dynamics simula-
tion-based approach would certainly help to further clarify
these results.
Affinity of spiro-dihydroindene analogues for hSHMT1
Human cytosolic SHMT1 and mitochondrial SHMT2 are of
emerging interest as potential anticancer targets.[59–63] We
therefore tested the potent spiro-dihydroindene analogues
against hSHMT1 (Table 2). Interestingly, ligands (Æ)-42 and (Æ)-
64 are highly selective for PfSHMT (selectivity factors of 44.1
and 57.3, respectively), whereas other ligands of this series
have similar affinities for both species (e.g., (+)-57 or (Æ)-59;
Table 2). The SAR is identical for both species: occupying the
pABA channel with a terminal fragment significantly enhancing
the activity ((Æ)-42 vs. (Æ)-43) and adding a fluoride on the
fused phenyl ring having no effect on target affinity. It is also
noteworthy that chiral recognition is less pronounced with
ChemMedChem 2018, 13, 1 – 14
www.chemmedchem.org
9
ꢂ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
Retrospective modeling of 2-indolinone analogue (+)-20
Conclusions
The new co-crystal structures obtained with the spirocyclic an-
alogues (+)-57 and (+)-59 gave us the opportunity of revising
our initial modeling of 2-indolinone (+)-20. By remodeling, we
observed a substantial change in the binding mode of the
latter (Figure 7 and Supporting Information, Section S4,
Figures S11 and S12) in which the lactam fragment does not
interact with Asn354 and Lys355 as initially predicted, but
rather undergoes a weak hydrogen bond with Tyr64. There-
fore, it is not properly solvated by residues of the protein and
the confined environment surrounding the N atom precludes
its solvation by a water molecule (d(N···NAsn354 and N···CTyr64)=
3.7 ꢁ; Figure 7). It should be noted that the side chain of
Asn354 solvates the neighboring backbone C=O of Lys355,
and it is therefore not available for interaction with the ligand
(Supporting Information, Section S4, Figure S11).
Alternative active substances to the original pyrazolopyran-
based lead series were investigated. Nineteen spirocyclic li-
gands featuring a pyrazolopyran core were rationally designed
and synthesized. Seven analogues based on a 2-indolinone
scaffold were essentially inactive, despite the fact that our ini-
tial modeling suggested the establishment of two favorable
polar interactions through their lactam ring with residues of
PvSHMT, in addition to those arising from the core of the mole-
cules. Conversely, sub-micromolar activity was obtained by re-
placement of the 2-indolinone scaffold by a gem-dimethyldihy-
droindene motif. An important affinity gain was achieved by
properly occupying the pABA channel with terminal fragments
attached to the core of the molecules. Although adding a fluo-
rine on the meta position of the fused phenyl ring did not
affect target affinity, it considerably improved cell-based effica-
cy. Additionally, the gem-dimethyl moiety on the dihydroin-
dene scaffold proved to be indispensable because its removal
led to a complete loss of efficacy. Chiral recognition is effective
with this class of ligands, and benzyl ester (+)-63 displayed
the best cellular potency of the series with an EC50 value of
165 nm.
Two co-crystal structures of (+)-57 and (+)-59 with PvSHMT
were resolved. The binding mode of these conformationally
constrained analogues resembled that of previous pyrazolopyr-
an-based ligands, and we showed that the vinylogous cyana-
mide acted in both series as an anchor to bind to the receptor.
Nonetheless, subtle changes were observed. In the spirocyclic
analogues (+)-57 and (+)-59, the orientation of the fused
phenyl ring was substantially different to that observed with
other non-spirocyclic pyrazolopyran-based ligands, such as car-
boxylate (+)-65. Consequently, the small lateral pocket lined
by the Cys364-loop was not properly filled by the ligands,
which allowed for the insertion of a water molecule. The short
ionic hydrogen bond between the pyrazole ring and Glu56
was lost due to the curved conformation of the core enforced
by the spirocyclic motif in (+)-57 and (+)-59, in favor of a
water-mediated interaction. This observation is important be-
cause it suggests that alternative fragments to the pyrazole
ring can potentially be introduced in future generations of
SHMT inhibitors. Overall, these new spirocyclic analogues bind
strongly to PvSHMT through an array of hydrogen-bonding
and dipolar interactions, which is nicely reflected by high
target affinities. In addition, these two complexes helped to
revise our modeling of the 2-indolinone series, and we postu-
late that an improper solvation of the lactam ring in the com-
plex is responsible for the poor affinities values.
Figure 7. Proposed binding mode of (+)-20 within the active site of PvSHMT
based on the novel co-crystal structure obtained with (+)-57 (PDB ID: 5YG3,
2.4 ꢁ), as modeled with MOLOC. PLP-Gly is omitted for clarity. Water mole-
cules from the PvSHMT–ligand complex 5YG3 were used for modeling and
are represented as red spheres. Distances are given in ꢁ. Color code: CPvSHMT
gray, CLigand lime, F light cyan, N blue, O red, S yellow.
In addition, we examined complexes from the PDB with li-
gands featuring a five-membered lactam ring to underpin our
observations from modeling (Supporting Information, Sec-
tion S5.2, Figure S20). Of the 202 co-crystal structures retrieved,
we found that 40% of the lactams were solvated by a water
molecule through their C=O moiety and 33% through their
NH moiety. Above all, the lactam ring is solvated by residues
of the protein (either NÀH···O=Cprotein or C=O···HÀO,Nprotein inter-
actions) in the majority (>70%) of the hits. In conclusion, our
retrospective modeling suggests that the improper solvation
of the lactam moiety within the binding pocket is unfavorable
for strong binding to PvSHMT, and is presumably the origin of
the poor affinities observed for this series of ligands.
Finally, we reported several spirocyclic compounds with se-
lectivity for PfSHMT over human cytosolic SHMT1, and by
virtue of a structural analysis we proposed that three amino
acid residues are important for discrimination between the
species, namely, Thr183 (replaced by a serine in hSHMT1),
Asn356, and Thr357.
&
ChemMedChem 2018, 13, 1 – 14
www.chemmedchem.org
10
ꢂ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
of protein–ligand complexes were prepared by mixing purified
PvSHMT protein (60 mL, 20–25 mgmLÀ1) with 1 mm PLP; 60 mm b-
mercaptoethanol; 90 mm glycine; and 5.7 mm (+)-57, (+)-59,
(+)-65, or (Æ)-66. The mixture was equilibrated on ice for 30 min
to allow complete complex formation. The crystallization drop was
composed of 1 mL each of a crystallization solution and the protein
complex. Protein crystals of PvSHMT were grown at 293 K in 20–
24% (w/v) polyethylene glycol 4000 (PEG4000), 0.06–0.12m NaCl,
and 0.1m Tris-HCl buffer at pH 8.5.
Experimental Section
Chemical synthesis: All synthetic protocols and analytical data for
the preparation of all ligands reported herein and their respective
intermediates are detailed in Sections S1 and S6 in the Supporting
Information.
In vitro antimalarial activity: P. falciparum drug-sensitive NF54
strain was cultivated in a variation of the medium previously de-
scribed, consisting of RPMI 1640 supplemented with 0.5% ALBU-
MAX II, 25 mm HEPES, 25 mm NaHCO3 buffer (pH 7.3), 0.36 mm hy-
poxanthine, and 100 mgmLÀ1 neomycin.[65,66] Human erythrocytes
served as host cells. Cultures were maintained in an atmosphere of
3% O2, 4% CO2, and 93% N2 in modular chambers at 378C. Com-
pounds were dissolved in DMSO (10 mgmLÀ1), diluted in hypoxan-
thine-free culture medium, and titrated in duplicate over a 64-fold
range in 96-well plates. Infected erythrocytes (1.25% final hema-
tocrit and 0.3% final parasitemia) were added to the wells. After
48 h incubation, 0.25 mCi [3H]hypoxanthine was added and plates
were incubated for an additional 24 h. Parasites were harvested
onto glass-fiber filters, and radioactivity was counted by using a
MicroBeta2 plate liquid scintillation counter (PerkinElmer). The re-
sults were recorded and expressed as a percentage of the untreat-
ed controls. EC50 values were estimated by linear interpolation.[67]
PvSHMT crystal structure data collection, structure determina-
tion, and refinement: A single crystal was flash-vitrified in liquid
nitrogen with 20% glycerol in a crystallizing agent as a cryoprotec-
tant. X-ray diffraction data were collected at 100 K at a wavelength
of 1 ꢁ by using an ADSC Quantum-315 charge-coupled device
(CCD) detector at beamline 13B1, NSRRC, Taiwan. Data were pro-
cessed by using the HKL2000 package.[68] X-ray diffraction data and
refinement statistics are listed in the Supporting Information, Sec-
tion S5.1, Table S3. The structure of PvSHMT was determined by
molecular replacement by using Phaser in the CCP4 suite[69] with a
chain A protomer of PvSHMT coordinate (PDB ID: 4OYT)[30] as the
template. Model building and structure refinement were per-
formed by using Coot[70] and Refmac5.[71] The ligand structure was
prepared by using HyperChem.[72]
Enzymatic PfSHMT assay: Assay reactions (200 mL total volume)
contained PfSHMT (ꢁ0.5 mm), l-serine (2 mm), (6S)-H4F (0.4 mm), b-
nicotinamide adenine dinucleotide phosphate (b-NADP+;
0.25 mm), and the coupling enzyme methylene tetrahydrofolate
dehydrogenase (FolD, 5 mm) in 50 mm HEPES, pH 7.0, containing
1 mm dithiothreitol (DTT) and 0.5 mm ethylenediaminetetraacetic
acid (EDTA). Inhibitors (1 mL) with various concentrations were
added to this mixture, and initial rates of the reaction were moni-
tored to measure the activity of the enzyme. The inhibitors were
dissolved in DMSO, and control assays without inhibitor, but in the
presence of 0.5% DMSO (final concentration), were also per-
formed.
Acknowledgements
Work at ETH Zꢀrich was supported by a grant from the ETH Re-
search Council (ETH-01 13-2). At ETH Zꢀrich, we thank Michael
Solar for recording the small-molecule X-ray crystal structures, as
well as Dr. Bruno Bernet for help with the compound characteri-
zations. We are grateful to Abhijit Kundu and Surajit Sadhukhan
at TCG Lifesciences for providing, free-of-charge, several 2-indoli-
none building blocks for the synthesis. Marcel Kaiser and his
technicians are acknowledged for performing the cytotoxicity
assays. Protein–ligand studies, including Plasmodium and human
enzyme inhibitions, and protein crystallography were financially
supported by grants from the Cluster Program and Management
Office, the National Science and Technology development Agency
(NSTDA), Thailand (CPMO-P-13-00835), the National Center for
Genetic Engineering and Biotechnology (Platform P-14-50241),
the Thailand Research Fund (RTA5980001), Mahidol University,
and the Vidyasirimedhi Institute of Science and Technology
(VISTEC). We gratefully acknowledge the NSTDA Characterization
and Testing Center (NCTC), and the NSTDA (Thailand) for single
crystal X-ray diffraction D8 venture, and thank Waraporn Pinyo
for helping with the D8 Venture system at NCTC. We are thankful
to the National Synchrotron Radiation Research Center (Taiwan)
for the beamline 13B1 and the operating staff. The experimental
facility and the technical services were provided by the “Synchro-
tron Radiation Protein Crystallography Facility of the National
Core Facility Program for Biotechnology, Ministry of Science and
Technology”, and the “National Synchrotron Radiation Research
Center”; a national user facility supported by the Ministry of Sci-
ence and Technology, Taiwan, ROC.
Enzymatic AtSHMT assay: Assay reactions (200 mL total volume)
contained AtSHMT (1 mg), l-serine (20 mm), (6S)-H4F (0.3 mm), b-
nicotinamide adenine dinucleotide (b-NAD+; 2 mm) and the cou-
pling enzyme FolD (20 mg) in 50 mm potassium phosphate buffer,
pH 7.4, containing 7.5 mm DTT. Inhibitors (1 mL) with various con-
centrations (final concentrations from 1–1000 nm) were added to
this mixture, and initial rates of the reaction monitored to measure
the amount of uninhibited enzyme. The inhibitors were dissolved
in 80% DMSO, and the control assays without inhibitor in the pres-
ence of 1% DMSO (final concentration) were also performed. The
accumulation of NADH was followed for 20 min at l=340 nm by
using a BioTek Synergy HTX plate reader.
Enzymatic human SHMT1 assay: Assay reactions (200 mL total
volume) contained hSHMT1 (0.1 mm), l-serine (2 mm), (6S)-H4F
(0.08 mm), NADP+ (0.25 mm), and the FolD coupling enzyme
(6 mm) in 50 mm HEPES, pH 7.0, containing 1 mm DTT and 0.5 mm
EDTA. Inhibitors (1 mL) with various concentrations were added to
this mixture, and initial rates of the reaction were monitored to
measure the activity of the enzyme. All inhibitors were dissolved in
DMSO. The control reaction was assayed as described above, but
in the presence of 0.5% DMSO instead of inhibitor.
Crystallization of recombinant PvSHMT and compounds (+)-57,
(+)-59, (+)-65, and (Æ)-66: PvSHMT was crystallized by using a mi-
crobatch method in a 60-well plate (Ø 1 mm at bottom of each
well) covered with baby oil (6 mL; Johnson; a mixture of mineral
oil, olive oil, and vitamin E, PZ Johnson, Thailand). Crystal structures
Conflict of interest
The authors declare no conflict of interest.
ChemMedChem 2018, 13, 1 – 14
www.chemmedchem.org
11
ꢂ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
[15] N. J. White, S. Pukrittayakamee, A. P. Phyo, R. Rueangweerayut, F.
Nosten, P. Jittamala, A. Jeeyapant, J. P. Jain, G. Lefꢆvre, R. Li, B. Magnus-
son, T. T. Diagana, F. J. Leong, N. Engl. J. Med. 2014, 371, 403–410.
[16] T. Wu, A. Nagle, K. Kuhen, K. Gagaring, R. Borboa, C. Francek, Z. Chen,
D. Plouffe, A. Goh, S. B. Lakshminarayana, J. Wu, H. Q. Ang, P. Zeng,
M. L. Kang, W. Tan, M. Tan, N. Ye, X. Lin, C. Caldwell, J. Ek, S. Skolnik, F.
Liu, J. Wang, J. Chang, C. Li, T. Hollenbeck, T. Tuntland, J. Isbell, C. Fis-
chli, R. Brun, M. Rottmann, V. Dartois, T. Keller, T. Diagana, E. Winzeler, R.
Glynne, D. C. Tully, A. K. Chatterjee, J. Med. Chem. 2011, 54, 5116–5130.
[17] A. Nagle, T. Wu, K. Kuhen, K. Gagaring, R. Borboa, C. Francek, Z. Chen,
D. Plouffe, X. Lin, C. Caldwell, J. Ek, S. Skolnik, F. Liu, J. Wang, J. Chang,
C. Li, B. Liu, T. Hollenbeck, T. Tuntland, J. Isbell, T. Chuan, P. B. Alper, C.
Fischli, R. Brun, S. B. Lakshminarayana, M. Rottmann, T. T. Diagana, E. A.
Winzeler, R. Glynne, D. C. Tully, A. K. Chatterjee, J. Med. Chem. 2012, 55,
4244–4273.
[18] K. L. Kuhen, A. K. Chatterjee, M. Rottmann, K. Gagaring, R. Borboa, J.
Buenviaje, Z. Chen, C. Francek, T. Wu, A. Nagle, S. W. Barnes, D. Plouffe,
M. C. S. Lee, D. A. Fidock, W. Graumans, M. van de Vegte-Bolmer, G. J.
van Gemert, G. Wirjanata, B. Sebayang, J. Marfurt, B. Russell, R. Suwanar-
usk, R. N. Price, F. Nosten, A. Tungtaeng, M. Gettayacamin, J. Sattabong-
kot, J. Taylor, J. R. Walker, D. Tully, K. P. Patra, E. L. Flannery, J. M. Vinetz,
L. Renia, R. W. Sauerwein, E. A. Winzeler, R. J. Glynne, T. T. Diagana, Anti-
microb. Agents Chemother. 2014, 58, 5060–5067.
Keywords: co-crystal structures · inhibitors · malaria · serine
hydroxymethyltransferase · spiro compounds
[1] World Health Organization, 2016 World Malaria Report, ISBN 978-92-4-
151171-1, Geneva, December 2016.
[2] M. A. Phillips, J. N. Burrows, C. Manyando, R. H. van Huijsduijnen, W. C.
Van Voorhis, T. N. C. Wells, Nat. Rev. Dis. Primers 2017, 3, 17050.
[3] World Health Organization, 2017 World Malaria Report, ISBN 978-92-4-
156552-3, Geneva, November 2017.
[4] World Health Organization, Rainbow Tables: http://www.who.int/immu-
nization/research/development/Rainbow_tables/en/, accessed January
6, 2018.
[5] The RTS, S Clinical Trials Partnership, PLoS Med. 2014, 11, e1001685.
[6] A. M. Dondorp, R. M. Fairhurst, L. Slutsker, J. R. MacArthur, J. G. Breman,
P. J. Guerrin, T. E. Wellems, P. Ringwald, R. D. Newman, C. V. Plowe, N.
Engl. J. Med. 2011, 365, 1073–1075.
[7] E. A. Ashley, M. Dhorda, R. M. Fairhurst, C. Amaratunga, P. Lim, S. Suon,
S. Sreng, J. M. Anderson, S. Mao, B. Sam, C. Sopha, C. M. Chuor, C.
Nguon, S. Sovannaroth, S. Pukrittayakamee, P. Jittamala, K. Chotivanich,
K. Chutasmit, C. Suchatsoonthorn, R. Runcharoen, T. T. Hien, N. T. Thuy-
Nhien, N. V. Thanh, N. H. Phu, Y. Htut, K.-T. Han, K. H. Aye, O. A. Mokuolu,
R. R. Olaosebikan, O. O. Folaranmi, M. Mayxay, M. Khanthavong, B. Hon-
gvanthong, P. N. Newton, M. A. Onyamboko, C. I. Fanello, A. K. Tshefu,
N. Mishra, N. Valecha, A. P. Phyo, F. Nosten, P. Yi, R. Tripura, S. Borrmann,
M. Bashraheil, J. Peshu, M. A. Faiz, A. Ghose, M. A. Hossain, R. Samad,
M. R. Rahman, M. M. Hasan, A. Islam, O. Miotto, R. Amato, B. MacInnis, J.
Stalker, D. P. Kwiatkowski, Z. Bozdech, A. Jeeyapant, P. Y. Cheah, T. Sa-
kulthaew, J. Chalk, B. Intharabut, K. Silamut, S. J. Lee, B. Vihokhern, C.
Kunasol, M. Imwong, J. Tarning, W. J. Taylor, S. Yeung, C. J. Woodrow,
J. A. Flegg, D. Das, J. Smith, M. Venkatesan, C. V. Plowe, K. Stepniewska,
P. J. Guerin, A. M. Dondorp, N. P. Day, N. J. White, N. Engl. J. Med. 2014,
371, 411–423.
[8] World Health Organization, Artemisinin and Artemisinin-Based Combina-
tion Therapy Resistance, Global Malaria Programme (GMP)_2016.11,
Geneva, October 2016.
[9] R. K. Haynes, B. Fugmann, J. Stetter, K. Rieckmann, H.-D. Heilmann, H.-
W. Chan, M.-K. Cheung, W.-L. Lam, H.-N. Wong, S. L. Croft, L. Vivas, L.
Rattray, L. Stewart, W. Peters, B. L. Robinson, M. D. Edstein, B. Kotecka,
D. E. Kyle, B. Beckermann, M. Gerisch, M. Radtke, G. Schmuck, W.
Steinke, U. Wollborn, K. Schmeer, A. Rçmer, Angew. Chem. Int. Ed. 2006,
45, 2082–2088; Angew. Chem. 2006, 118, 2136–2142.
[10] P. M. O’Neill, R. K. Amewu, S. A. Charman, S. Sabbani, N. F. Gnꢀdig, J.
Straimer, D. A. Fidock, E. R. Shore, N. L. Roberts, M. H.-L. Wong, W. D.
Hong, C. Pidathala, C. Riley, B. Murphy, G. Aljayyoussi, F. J. Gamo, L.
Sanz, J. Rodrigues, C. G. Cortes, E. Herreros, I. Angulo-Barturꢃn, M. B. Ji-
mꢃnez-Dꢄaz, S. F. Bazaga, M. S. Martꢄnez-Martꢄnez, B. Campo, R. Sharma,
E. Ryan, D. M. Shackleford, S. Campbell, D. A. Smith, G. Wirjanata, R.
Noviyanti, R. N. Price, J. Marfurt, M. J. Palmer, I. M. Copple, A. E. Mercer,
A. Ruecker, M. J. Delves, R. E. Sinden, P. Siegl, J. Davies, R. Rochford,
C. H. M. Kocken, A.-M. Zeeman, G. L. Nixon, G. A. Biagini, S. A. Ward, Nat.
Commun. 2017, 8, 15159.
[11] S. A. Charman, S. Arbe-Barnes, I. C. Bathurst, R. Brun, M. Campbell, W. N.
Charman, F. C. K. Chiu, J. Chollet, J. C. Craft, D. J. Creek, Y. Dong, H.
Matile, M. Maurer, J. Morizzi, T. Nguyen, P. Papastogiannidis, C. Scheurer,
D. M. Shackleford, K. Sriraghavan, L. Stingelin, Y. Tang, H. Urwyler, X.
Wang, K. L. White, S. Wittlin, L. Zhou, J. L. Vennerstrom, Proc. Natl. Acad.
Sci. USA 2011, 108, 4400–4405.
[12] J. L. Vennerstrom, S. Arbe-Barnes, R. Brun, S. A. Charman, F. C. K. Chiu, J.
Chollet, Y. Dong, A. Dorn, D. Hunziker, H. Matile, K. McIntosh, M. Padma-
nilayam, J. S. Tomas, C. Scheurer, B. Scorneaux, Y. Tang, H. Urwyler, S.
Wittlin, W. N. Charman, Nature 2004, 430, 900–904.
[13] B. R. Blank, J. Gut, P. J. Rosenthal, A. R. Renslo, J. Med. Chem. 2017, 60,
6400–6407.
[14] M. Rottmann, C. McNamara, B. K. S. Yeung, M. C. S. Lee, B. Zou, B. Rus-
sell, P. Seitz, D. M. Plouffe, N. V. Dharia, J. Tan, S. B. Cohen, K. R. Spencer,
G. E. Gonzꢅlez-Pꢅez, S. B. Lakshminarayana, A. Goh, R. Suwanarusk, T.
Jegla, E. K. Schmitt, H.-P. Beck, R. Brun, F. Nosten, L. Renia, V. Dartois,
T. H. Keller, D. A. Fidock, E. A. Winzeler, T. T. Diagana, Science 2010, 329,
1175–1180.
[19] F. J. Leong, R. Zhao, S. Zeng, B. Magnusson, T. T. Diagana, P. Pertel, Anti-
microb. Agents Chemother. 2014, 58, 6437–6443.
[20] N. J. White, T. T. Duong, C. Uthaisin, F. Nosten, A. P. Phyo, B. Hanboonku-
nupakarn, S. Pukrittayakamee, P. Jittamala, K. Chuthasmit, M. S. Cheung,
Y. Feng, R. Li, B. Magnusson, M. Sultan, D. Wieser, X. Xun, R. Zhao, T. T.
Diagana, P. Pertel, F. J. Leong, N. Engl. J. Med. 2016, 375, 1152–1160.
[21] J. M. Coteron, M. Marco, J. Esquivias, X. Deng, K. L. White, J. White, M.
Koltun, F. El Mazouni, S. Kokkonda, K. Katneni, R. Bhamidipati, D. M.
Shackleford, I. Angulo-Barturen, S. B. Ferrer, M. B. Jimꢃnez-Dꢄaz, F.-J.
Gamo, E. J. Goldsmith, W. N. Charman, I. Bathurst, D. Floyd, D. Mat-
thews, J. N. Burrows, P. K. Rathod, S. A. Charman, M. A. Phillips, J. Med.
Chem. 2011, 54, 5540–5561.
[22] M. A. Phillips, J. Lotharius, K. Marsh, J. White, A. Dayan, K. L. White, J. W.
Njoroge, F. El Mazouni, Y. Lao, S. Kokkonda, D. R. Tomchick, X. Deng, T.
Laird, S. N. Bhatia, S. March, C. L. Ng, D. A. Fidock, S. Wittlin, M. La-
fuente-Monasterio, F. J. G. Benito, L. M. S. Alonso, M. S. Martinez, M. B. Ji-
menez-Diaz, S. F. Bazaga, I. Angulo-Barturen, J. N. Haselden, J. Louttit, Y.
Cui, A. Sridhar, A.-M. Zeeman, C. Kocken, R. Sauerwein, K. Dechering,
V. M. Avery, S. Duffy, M. Delves, R. Sinden, A. Ruecker, K. S. Wickham, R.
Rochford, J. Gahagen, L. Iyer, E. Riccio, J. Mirsalis, I. Bathhurst, T. Rueckle,
X. Ding, B. Campo, D. Leroy, M. J. Rogers, P. K. Rathod, J. N. Burrows,
S. A. Charman, Sci. Transl. Med. 2015, 7, 296ra111.
[23] E. Sandoval, M. J. Lafuente-Monasterio, M. J. Almela, P. CastaÇeda,
M. B. J. Dꢄaz, M. S. Martꢄnez-Martꢄnez, J. Vidal, ꢇ. Angulo-Barturen, P. Bam-
borough, J. Burrows, N. Cammack, M. J. Chaparro, J. M. Coterꢈn, C. de
Cozar, B. Crespo, B. Dꢄaz, G. Drewes, E. Fernꢅndez, S. Ferrer-Bazaga, M. T.
Fraile, F. J. Gamo, S. Ghidelli-Disse, R. Gꢈmez, J. Haselden, S. Huss, M. L.
Leꢈn, J. de Mercado, S. J. F. Macdonald, J. I. M. Hernando, S. Prats, M.
Puente, A. Rodrꢄguez, J. C. de la Rosa, L. Rueda, C. Selenski, P. Willis,
D. M. Wilson, M. Witty, F. Calderꢈn, J. Med. Chem. 2017, 60, 6880–6896.
[24] N. Nirmalan, P. Wang, P. F. G. Sims, J. E. Hyde, Mol. Microbiol. 2002, 46,
179–190.
[25] S. Maenpuen, K. Sopitthummakhun, Y. Yuthavong, P. Chaiyen, U. Leart-
sakulpanich, Mol. Biochem. Parasitol. 2009, 168, 63–73.
[26] K. Sopitthummakhun, S. Maenpuen, Y. Yuthavong, U. Leartsakulpanich,
P. Chaiyen, FEBS J. 2009, 276, 4023–4036.
[27] K. Sopitthummakhun, C. Thongpanchang, T. Vilaivan, Y. Yuthavong, P.
Chaiyen, U. Leartsakulpanich, Malar. J. 2012, 11, 194.
[28] W. Pornthanakasem, D. Kongkasuriyachai, C. Uthaipibull, Y. Yuthavong,
U. Leartsakulpanich, Malar. J. 2012, 11, 387.
[29] P. Chitnumsub, W. Ittarat, A. Jaruwat, K. Noytanom, W. Amornwatchara-
pong, W. Pornthanakasem, P. Chaiyen, Y. Yuthavong, U. Leartsakulpa-
nich, Acta Crystallogr. Sect. D 2014, 70, 1517–1527.
[30] P. Chitnumsub, A. Jaruwat, P. Riangrungroj, W. Ittarat, K. Noytanom, W.
Oonanant, J. Vanichthanankul, P. Chuankhayan, S. Maenpuen, C.-J.
Chen, P. Chaiyen, Y. Yuthavong, U. Leartsakulpanich, Acta Crystallogr.
Sect. D 2014, 70, 3177–3186.
&
ChemMedChem 2018, 13, 1 – 14
www.chemmedchem.org
12
ꢂ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
[31] C. K. T. Pang, J. H. Hunter, R. Gujjar, R. Podutoori, J. Bowman, D. G. Mu-
deppa, P. K. Rathod, Mol. Biochem. Parasitol. 2009, 168, 74–83.
[32] M. C. Witschel, M. Rottmann, A. Schwab, U. Leartsakulpanich, P. Chit-
numsub, M. Seet, S. Tonazzi, G. Schwertz, F. Stelzer, T. Mietzner, C. McNa-
mara, F. Thater, C. Freymond, A. Jaruwat, C. Pinthong, P. Riangrungroj,
M. Oufir, M. Hamburger, P. Mꢀser, L. M. Sanz-Alonso, S. Charman, S. Wit-
tlin, Y. Yuthavong, P. Chaiyen, F. Diederich, J. Med. Chem. 2015, 58,
3117–3130.
[53] M. Harder, B. Kuhn, F. Diederich, ChemMedChem 2013, 8, 397–404.
[54] R. Paulini, K. Mꢉller, F. Diederich, Angew. Chem. Int. Ed. 2005, 44, 1788–
1805; Angew. Chem. 2005, 117, 1820–1839.
[55] K. Mꢉller, C. Faeh, F. Diederich, Science 2007, 317, 1881–1886.
[56] S. G. Krimmer, M. Betz, A. Heine, G. Klebe, ChemMedChem 2014, 9, 833–
846.
[57] M. Betz, T. Wulsdorf, S. G. Krimmer, G. Klebe, J. Chem. Inf. Model. 2016,
56, 223–233.
[33] G. Schwertz, M. C. Witschel, M. Rottmann, R. Bonnert, U. Leartsakulpa-
nich, P. Chitnumsub, A. Jaruwat, W. Ittarat, A. Schꢀfer, R. A. Aponte, S. A.
Charman, K. L. White, A. Kundu, S. Sadhukhan, M. Lloyd, G. M. Freiberg,
M. Srikumaran, M. Siggel, A. Zwyssig, P. Chaiyen, F. Diederich, J. Med.
Chem. 2017, 60, 4840–4860.
[34] G. Schwertz, M. S. Frei, M. C. Witschel, M. Rottmann, U. Leartsakulpanich,
P. Chitnumsub, A. Jaruwat, W. Ittarat, A. Schꢀfer, R. A. Aponte, N. Trapp,
K. Mark, P. Chaiyen, F. Diederich, Chem. Eur. J. 2017, 23, 14345–14357.
[35] P. R. Gerber, K. Mꢉller, J. Comput.-Aided Mol. Des. 1995, 9, 251–268.
[36] G. S. Singh, Z. Y. Desta, Chem. Rev. 2012, 112, 6104–6155.
[37] A. Leoni, A. Locatelli, R. Morigi, M. Rambaldi, Expert Opin. Ther. Pat.
2016, 26, 149–173.
[38] G. J. Roth, A. Heckel, F. Colbatzky, S. Handschuh, J. Kley, T. Lehmann-
Lintz, R. Lotz, U. Tontsch-Grunt, R. Walter, F. Hilberg, J. Med. Chem. 2009,
52, 4466–4480.
[39] P. C. Peng, Y. D. Su, J. Feng, J. H. Fu, J. L. Yanng, L. Xiao, J. H. Peng, Y. L.
Li, L. Zhang, B. Hu, Y. Zhou, F. Q. Li, B. B. Fu, L. G. Lou, A. S. Gong, G. H.
She, W. H. Sun, X. T. Mong, J. Med. Chem. 2010, 53, 8140–8149.
[40] G. J. Roth, R. Binder, F. Colbatzky, C. Dallinger, R. Schlenker-Herceg, F.
Hilberg, S.-L. Wollin, R. Kaiser, J. Med. Chem. 2015, 58, 1053–1063.
[41] C. V. Galliford, K. A. Scheidt, Angew. Chem. Int. Ed. 2007, 46, 8748–8758;
Angew. Chem. 2007, 119, 8902–8912.
[58] S. G. Krimmer, J. Cramer, M. Betz, V. Fridh, R. Karlsson, A. Heine, G.
Klebe, J. Med. Chem. 2016, 59, 10530–10548.
[59] M. Marani, A. Paone, A. Fiascarelli, A. Macone, M. Gargano, S. Rinaldo, G.
Giardina, V. Pontecorvi, D. Koes, L. McDermott, T. Yang, A. Paiardini, R.
Contestabile, F. Cutruzzolꢊ, Oncotarget 2015, 7, 4570–4583.
[60] D. Kim, B. P. Fiske, K. Birsoy, E. Freinkman, K. Kami, R. L. Possemato, Y.
Chudnovsky, M. E. Pacold, W. W. Chen, J. R. Cantor, L. M. Shelton, D. Y.
Gui, M. Kwon, S. H. Ramkissoon, K. L. Ligon, S. W. Kang, M. Snuderl,
M. G. Vander Heiden, D. M. Sabatini, Nature 2015, 520, 363–367.
[61] S. Bernhardt, M. Bayerlovꢅ, M. Vetter, A. Wachter, D. Mitra, V. Hanf, T.
Lantzsch, C. Uleer, S. Peschel, J. John, J. Buchmann, E. Weigert, K.-F.
Bꢉrrig, C. Thomssen, U. Korf, T. Beissbarth, S. Wiemann, E. J. Kantelhardt,
Breast Cancer Res. 2017, 19, 112.
[62] G. S. Ducker, J. M. Ghergurovich, N. Mainolfi, V. Suri, S. K. Jeong, S. H.-J.
Li, A. Friedman, M. G. Manfredi, Z. Gitai, H. Kim, J. D. Rabinowitz, Proc.
Natl. Acad. Sci. USA 2017, 114, 11404–11409.
[63] B. Wang, W. Wang, Z. Zhu, X. Zhang, F. Tang, D. Wang, X. Liu, X. Yan, H.
Zhuang, Clin. Neurol. Neurosurg. 2017, 154, 28–33.
[64] T.-F. Fu, J. N. Scarsdale, G. Kazanina, V. Schirch, H. T. Wright, J. Biol. Chem.
2003, 278, 2645–2653.
[65] W. Trager, J. B. Jensen, Science 1976, 193, 673–675.
[66] A. Dorn, R. Stoffel, H. Matile, A. Bubendorf, R. G. Ridley, Nature 1995,
374, 269–271.
[67] W. Huber, J. C. Koella, Acta Trop. 1993, 55, 257–261.
[68] “Processing of X-ray Diffraction Data Collected in Oscillation Mode”: Z.
Otwinowski, W. Minor in Methods in Enzymology, Vol. 276: Macromolecu-
lar Crystallography, Part A (Eds.: C. W. Carter, Jr., R. M. Sweet), Academic
Press, New York, 1997, pp. 307–326.
[42] J. R. Reyes, J. Xu, K. Kobayashi, V. Bhat, V. H. Rawal, Angew. Chem. Int. Ed.
2017, 56, 9962–9966; Angew. Chem. 2017, 129, 10094–10098.
[43] M. Arthuis, R. Pontikis, G. G. Chabot, J. Seguin, L. Quentin, S. Bourg, L.
Morin-Allory, J.-C. Florent, ChemMedChem 2011, 6, 1693–1705.
[44] A. Huang, A. Moretto, K. Janz, M. Lowe, P. W. Bedard, S. Tam, L. Di, V.
Clerin, N. Sushkova, B. Tchernychev, D. H. H. Tsao, J. C. Keith, Jr., G. D.
Shaw, R. G. Schaub, Q. Wang, N. Kaila, J. Med. Chem. 2010, 53, 6003–
6017.
[69] A. J. McCoy, R. W. Grosse-Kunstleve, P. D. Adams, M. D. Winn, L. C. Storo-
ni, R. J. Read, J. Appl. Crystallogr. 2007, 40, 658–674.
[45] C. H. Wang, A. R. White, S. N. Schwartz, S. Alluri, T. M. Cattabiani, L. K.
Zhang, T. M. Chan, A. V. Buevich, A. K. Ganguly, Tetrahedron 2012, 68,
9750–9762.
[46] G. Vasuki, K. Kumaravel, Tetrahedron Lett. 2008, 49, 5636–5638.
[47] S. Gogoi, C.-G. Zhao, Tetrahedron Lett. 2009, 50, 2252–2255.
[48] T. T. Talele, J. Med. Chem. 2018, 61, 2166–2210.
[70] P. Emsley, B. Lohkamp, W. G. Scott, K. Cowtan, Acta Crystallogr. Sect. D
2010, 66, 486–501.
[71] G. N. Murshudov, P. Skubꢅk, A. A. Lebedev, N. S. Pannu, R. A. Steiner,
R. A. Nicholls, M. D. Winn, F. Long, A. A. Vagin, Acta Crystallogr. Sect. D
2011, 67, 355–367.
[72] HyperChemTM Professional 7.51, Hypercube Inc., 1115 NW 4th Street,
Gainesville, FL 32601 (USA).
[49] S. Matsumoto, D. Samata, M. Akazome, K. Ogura, Tetrahedron Lett.
2009, 50, 111–114.
[50] N. O. Calloway, Chem. Rev. 1935, 17, 327–392.
Manuscript received: January 23, 2018
Revised manuscript received: February 25, 2018
[51] C. C. C. J. Seechurn, M. O. Kitching, T. J. Colacot, V. Snieckus, Angew.
Chem. Int. Ed. 2012, 51, 5062–5085; Angew. Chem. 2012, 124, 5150–
5174.
Accepted manuscript online: February 27, 2018
[52] P. Ruiz-Castillo, S. L. Buchwald, Chem. Rev. 2016, 116, 12564–12649.
Version of record online: && &&, 0000
ChemMedChem 2018, 13, 1 – 14
www.chemmedchem.org
13
ꢂ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

FULL PAPERS
G. Schwertz, M. C. Witschel, M. Rottmann,
U. Leartsakulpanich, P. Chitnumsub,
A. Jaruwat, W. Amornwatcharapong,
W. Ittarat, A. Schꢁfer, R. A. Aponte,
N. Trapp, P. Chaiyen, F. Diederich*
Curved analogues: Potent antimalarial
inhibitors of the enzyme serine hydroxy-
methyltransferase that feature a spirocy-
clic core were uncovered by structure-
based design. The binding mode of
these novel active substances was eluci-
dated by X-ray co-crystal structures,
which reveal the key contribution of the
vinylogous cyanamide fragment and a
substantial curving of the core induced
by the spirocyclic motif.
&& – &&
Potent Inhibitors of Plasmodial Serine
Hydroxymethyltransferase (SHMT)
Featuring a Spirocyclic Scaffold
&
ChemMedChem 2018, 13, 1 – 14
www.chemmedchem.org
14
ꢂ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!