Target Elucidation by Cocrystal Structures of NADH-Ubiquinone Oxidoreductase of Plasmodium falciparum (PfNDH2) with Small Molecule To Eliminate Drug-Resistant Malaria

Article abstract of DOI:10.1021/acs.jmedchem.6b01733

Drug-resistant malarial strains have been continuously emerging recently, which posts a great challenge for the global health. Therefore, new antimalarial drugs with novel targeting mechanisms are urgently needed for fighting drug-resistant malaria. NADH-ubiquinone oxidoreductase of Plasmodium falciparum (PfNDH2) represents a viable target for antimalarial drug development. However, the absence of structural information on PfNDH2 limited rational drug design and further development. Herein, we report high resolution crystal structures of the PfNDH2 protein for the first time in Apo-, NADH-, and RYL-552 (a new inhibitor)-bound states. The PfNDH2 inhibitor exhibits excellent potency against both drug-resistant strains in vitro and parasite-infected mice in vivo via a potential allosteric mechanism. Furthermore, it was found that the inhibitor can be used in combination with dihydroartemisinin (DHA) synergistically. These findings not only are important for malarial PfNDH2 protein-based drug development but could also have broad implications for other NDH2-containing pathogenic microorganisms such as Mycobacterium tuberculosis.

Full text of DOI:10.1021/acs.jmedchem.6b01733

Subscriber access provided by University of Newcastle, Australia
Article
Target Elucidation by Co-crystal Structures of NADH-
Ubiquinone Oxidoreductase of Plasmodium Falciparum (PfNDH2)
with Small Molecule to Eliminate Drug-Resistant Malaria
Yiqing Yang, You Yu, Xiaolu Li, Jing Li, Yue Wu, Jie Yu, Jingpeng
Ge, Zhenghui Huang, Lubin Jiang, Yu Rao, and Maojun Yang
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.6b01733 • Publication Date (Web): 14 Feb 2017
Downloaded from http://pubs.acs.org on February 15, 2017
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 40
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Target Elucidation by Coꢀcrystal Structures of
NADHꢀUbiquinone Oxidoreductase of Plasmodium
Falciparum (PfNDH2) with Small Molecule to
Eliminate DrugꢀResistant Malaria
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Yiqing Yang1,2,†, You Yu2,†, Xiaolu Li4,†, Jing Li2, Yue Wu1, Jie Yu2, Jingpeng Ge2, Zhenghui
Huang3, Lubin Jiang3,*, Yu Rao1,*, Maojun Yang2,*
1MOE Key Laboratory of Protein Sciences, School of Pharmaceutical Sciences, Tsinghua
University, Beijing 100084, China.
2MOE Key Laboratory of Protein Sciences, TsinghuaꢀPeking Center for Life Sciences, School of
Life Sciences, Tsinghua University, Beijing 100084, China.
3Institut Pasteur of Shanghai, CAS Key Laboratory of Molecular Virology and Immunology,
Chinese Academy of Sciences, Shanghai 200031, China.
4Department of Biochemistry and Molecular Biology, State Key Laboratory of Medical
Molecular Biology, Institute of Basic Medical Sciences, Chinese Academy of Medical Sciences,
Peking Union Medical College, Beijing 100005, China.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 40
1
2
3
4
5
6
7
8
9
ABSTRACT: Drugꢀresistant malarial strains have been continuously emerging recently, which
posts a great challenge for the global health. Therefore, new antiꢀmalarial drugs with novel
targeting mechanisms are urgently needed for fighting drugꢀresistant malaria. NADHꢀubiquinone
oxidoreductase of Plasmodium falciparum (PfNDH2) represents a viable target for antiꢀmalarial
drug development. However, the absence of structural information of PfNDH2 limited rational
drug design and further development. Herein, we report high resolution crystal structures of the
PfNDH2 protein for the first time in Apoꢀ, NADHꢀ and RYLꢀ552 (a new inhibitor)ꢀbound states.
The PfNDH2 inhibitor exhibits excellent potency against both drugꢀresistant strains in vitro and
parasiteꢀinfected mice in vivo via a potential allosteric mechanism. Furthermore, it was found
that the inhibitor can be used in combination with dihydroartemisinin (DHA) synergistically.
These findings are not only important for malarial PfNDH2 proteinꢀbased drug development, but
could also have broad implications for other NDH2ꢀcontaining pathogenic microorganisms such
as Mycobacterium tuberculosis.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
INTRODUCTION
Nobel Prize in Physiology or Medicine of 2015 was awarded to Youyou Tu for her contributions
in discovering artemisinin, which had been broadly used to treat malaria. Today malaria still
remains a major challenge to global health, which causes more than 212 million new cases and
over 429,000 deaths annually1. Although antiꢀmalarial drugs have successfully mitigated the
epidemics in the past few decades, drug resistance has been emerging rapidly2,3. In fact, clinical
evidence has reported the drug resistance for all existing classes of antiꢀmalarial medicines
including chloroquine4 and artemisinin5. For instance, chloroquineꢀresistant P. falciparum has
spread to most malariaꢀmanifested areas up to date. Even for the most effective antiꢀmalarial
ACS Paragon Plus Environment

Page 3 of 40
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
drug artemisinin, its resistance was first reported in 2008, which has become a major issue in
parts of Southeast Asia3,6ꢀ8. To tackle these important problems, there is an urgent requirement
for new antiꢀmalarial drugs.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The electron transport chain (ETC), which comprises oxidative phosphorylation ComplexꢀI to IV
located at the inner membrane of mitochondria, serves as the core for oxidation to produce ATP.
Atovaquone, which has been used as competitive inhibitor for more than 20 years in clinics,
targets the cytochrome bc1 complex (ComplexꢀIII) of the ETC to prevent the proton pumping
and causes a loss of mitochondrial membrane potential leading to organelle dysfunction9,10
.
However, the utilization of atovaquone was limited severely after the emergence of its drug
resistance. The bc1 complex activity requires the ubiquinol, which is produced by NADH
dehydrogenase (NDH, and also named ComplexꢀI)11 through the transfer of two electrons from
NADH to ubiquinone. This ubiquinolꢀproducing reaction is the rateꢀlimiting step in respiration
and is central to energy metabolism. Singleꢀsubunit typeꢀII NADH dehydrogenase (NDH2),
without the proton pumping activity, acts as an alternative to the multiꢀsubunit respiratory
ComplexꢀI12,13. NDH2 exists in many pathogenic microorganisms, such as P. falciparum,
Toxoplasma gondi and Mycobacterium tuberculosis14ꢀ16, and is essential for the viability and
longꢀterm survival of M. tuberculosis17 and P. falciparum14,18,19. The inhibition of their NDH2
could efficiently disrupt their respiration chain, so it could represent a viable target for
combatting P. falciparum, the main malaria causing parasite20,21
.
Although previous literatures reported compounds like 1 (CKꢀ2ꢀ68) (Fig. 1A) that had been
shown to target PfNDH2 with a high potency18,19, the absence of highꢀresolution structure of
PfNDH2 and inhibitorꢀbound PfNDH2 complex presents difficulties for molecular mechanism
elucidation and structureꢀbased rational drug design. Given the significance of structural
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 40
1
2
3
4
5
6
7
8
9
information of NDH2 protein for both fundamental molecular mechanism understanding and
antiꢀmalarial drug discovery, herein, we reported the high resolution crystal structures of
PfNDH2 in Apoꢀ, NADHꢀ and 2 (RYLꢀ552)ꢀbound states at 2.16, 2.70 and 2.05 Å, respectively
(Table S1). These structures revealed a possible allosteric mechanism of compound 2 with two
binding sites in PfNDH2. Allosteric ligands usually offer distinct advantages over competitive
inhibitors22,23 and have been explored increasingly in various families of proteins in recent years,
which paves a way to develop new pharmaceutical agents and increases our understanding of the
fundamental biological processes as well24ꢀ27. The potential of PfNDH2 inhibitors were evaluated
further through a series of in vitro and in vivo studies, which reflected the promising perspective
for antiꢀmalarial drug development based on PfNDH2 and its inhibitors.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
RESULTS AND DISCUSSION
Overall structure of PfNDH2. The structures of Apoꢀ, NADHꢀ and the inhibitorꢀbound
PfNDH2 complexes are solved for the first time. It was found that PfNDH2 forms a stable homoꢀ
dimer in solution and crystal structures (Fig. 1B and Fig. S1) with each monomer forms a
globular structure comprising four domains (Fig. S2A). Residues of the Cꢀterminal domain (CTD)
that support selfꢀdimerization in situ and membrane attachment28,29 are well conserved in
different Plasmodium species (Fig. 1CꢀD). Between the two canonical Rossmann fold domains
(domain A and domain B), which bind the substrates FAD and NADH (Fig. S2B), respectively,
PfNDH2 contains a domain C in the middle of the sequence (Fig. 1B and Fig. S2A). Dali
searches30 with domain C only returned entries with the highest Zꢀscore of 5.2, suggesting that no
known structure was identified to share significant homology with this domain. Evaluation of the
most similar structure from this search, the EFꢀhand domain in calcium binding protein 40
ACS Paragon Plus Environment

Page 5 of 40
Journal of Medicinal Chemistry
1
2
(CBP40) (PDB ID: 1IJ6) showed that none of the residues responsible for the Ca2+ꢀbinding occur
3
4
5
6
in the extended PfNDH2 domain (Fig. S2C) and the function of domain C is undisclosed so far.
7
8
9
The binding mode of PfNDH2 with compound 2. To understand the mechanism underlying the
inhibitory function, we tried to coꢀcrystallize PfNDH2 with its inhibitors. The previously
reported compound 1 displays poor water solubility because of the strong intermolecular πꢀπ
stacking interactions18, which always results in the precipitation of compound 1 from the coꢀ
crystallization solution. To solve this challenge, new inhibitors with both excellent potency and
improved aqueous solubility should be obtained as tool compounds. The quinolinꢀ4ꢀone
compounds could be synthesized under three different reaction conditions with related starting
materials and intermediates which were reported previously (Scheme 1, d, j and i)18,19. Some
derivatives of 1ꢀ(2ꢀhydroxyphenyl)ethanone such as compound 42 could react with aromatic
aldehydes via aldol condensation to generate intermediates of α, βꢀunsaturated ketones that were
used further as starting materials to prepare chromenꢀ4ꢀone by iodine catalyzed intramolecular
cyclization (Scheme 1, g and h). To improve the aqueous solubility and meet the needs for our
coꢀcrystallization study, we modified the molecules with hydrophilic functional groups including
–F, ꢀC(=O)OH, ꢀC(=O)NHNH2, and ꢀNHC(=O)CF3, etc. (Scheme 1, Table S3), which led to the
identification of some highly effective small molecules including compound 2 and 3 (RYLꢀ552S)
(Fig. 1A). It was found that only compound 2 could be coꢀcrystallized with PfNDH2 successfully
after many attempts, which might be due to a better combination of the aqueous solubility
(CLogP=5.8) and inhibitory potency (IC50=3.73 nM) of compound 2 (Fig. 1A). The results may
indicate the 5ꢀF in compound 2 can contribute more than the 7ꢀCl in compound 1 in terms of
binding affinity. Unexpectedly, compound 2 inserts into two different deep pockets in each
PfNDH2 molecule (Fig. 2A). Two compound 2 molecules insert into the pocket between
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 40
1
2
PfNDH2 homoꢀdimer interface (Fig. 2B) which exists in situ28,29. The 4ꢀoxoꢀ5ꢀfluoro element of
the quinolone scaffold grips the backbone carbonyl of G87 and side chain amino group of K523
with the help of two bridge water molecules. And the carbonyl oxygen atom also forms another
hydrogen bond with the side chain of N92. The bisaryl part linked by methylene is clamped by
the hydrophobic surface produced from V91, I170, L174 and I532. The terminal
trifluoromethoxy forms two hydrogen bonds with Y74 and K533. More interestingly, instead of
the predicted Q binding sites (catalytic domains), the other two molecules are deeply buried in
the pocket formed by the helixꢀ17 and 18 at the Cꢀterminal separately (Fig. 2C). It is found that
the 4ꢀoxoꢀ5ꢀfluoro element of quinolone has two hydrogen bonds with the side chain of K501.
The nitrogen atom from quinolone ring forms a hydrogen bond network with one water molecule,
which further bridges with terminal carboxylic acid group of E218 and guanidine group of R529.
And two different πꢀπ stacking interactions around the quinolone ring (one likes T shape with
edge to face style between 6ꢀH of quinolone and W500; the other is face to face between
quinolone and Y475) contributes more binding affinity. We designed single and double
mutations of the residues that could interfere with the binding of compound 2 in one or the other
binding pocket (N92R, D171R, Y475R, S497R, N92R/Y475R or N92R/S497R; Fig. 2D and Fig
S3). Each of these engineered single mutations significantly reduced the ability of compound 2
to interfere with PfNDH2 activity, as reflected in the increased IC50 value by 3ꢀ to 10ꢀfold.
Despite the strong influences of the mutations inhibition, these single mutations in themselves
showed only small effects on the native catalytic ability of the enzyme (Fig. 2E). Although the
N92R/S497R double mutant lost about 50% of the enzyme activity, the compound 2 greatly
reduce the inhibition activity by 34ꢀfold, while the N92R/Y475R mutant protein lost 95% of the
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Page 7 of 40
Journal of Medicinal Chemistry
1
2
3
4
enzyme activity. These findings are consistent with the single mutant results which indicates the
5
6
Y475R has a greater effect than S497R (IC50, 31 vs 17 nM).
7
8
9
Further ligand-based SAR study verified the cocrystal structure results. The ligandꢀbased
SAR (structureꢀactivity relationship) studies are consistent well with the coꢀcrystal structure
results of compound 2 (Fig. 3AꢀB and Table S3). For example: 1) 5ꢀF of compound 2 can form a
hydrogen bond with K501 (Fig. 3C). In contrast, if the F atom is installed at other sites including
6, 7 or 8 position, it can not help the carbonyl group of quinolone accommodating the H2O
bridge (Fig. 3A). In addition, the 6ꢀF substitution may decrease the πꢀπ stacking interaction with
W500 (Fig. 3A) and eject the compound out of the pocket. These structural observations are also
consistent well with our experimental results that the overall inhibitory order for R1 is 5ꢀF>7ꢀ
F>8ꢀF>6ꢀF (Fig. 3B); 2) the NꢀH of the quinolone is essential for the inhibitory activity, as it
serves as a hydrogen bond donor (Fig. 3A). However, the flavone compounds would lost this
critical interaction, which results in less effective inhibition (Fig. 3B, ꢀNHꢀ is better than ꢀOꢀ for
X); 3) The R2 group is oriented towards the hydrophobic clamp composed of V91, I170, L174
and I532 (Fig. 2B). As a result, the polar and hydrophilic functional groups like ꢀC(=O)OMe, ꢀ
C(=O)OH, ꢀC(=O)NHNH2 (Fig. 3B) are not compatible for binding interactions. Because the
linker Z is located at the gap of the same hydrophobic pocket as well, it is not surprising that
methylene is better than ꢀOꢀ (Fig. 3B). 4) 4ꢀOCF3 or 4ꢀSCF3 is the best substituent group for R3
(Fig. 3B), which can be explained by the hydrogen bonding formations between them and
residues Y74 and K533 via the fluoride atoms (Fig. 3A).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The proposed allosteric inhibitory mechanism by compound 2. In the enzyme kinetic studies
of PfNDH2 with different compound 2 concentrations (Fig. S4), the effective Vmax decreases
with inhibition but the Km remains constant, which indicates that compound 2 is a nonꢀ
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 40
1
2
3
4
5
6
7
8
9
competitive inhibitor. Then we compared the structure of compound 2 bound PfNDH2 with the
Apoꢀ as well as NADHꢀbound forms and searched for conformational changes that could explain
the inhibitory effect of compound 2 on enzyme function (Fig. 4AꢀC). In these structural
comparisons, the Rossmann fold involved in FADꢀbinding (amino acids 41ꢀ152) superposed
with root mean square deviation values (RMSDs) of less than 0.18 Å. In the NADHꢀbound form
of PfNDH2, the main chain extending from helixꢀ7 was shifted to a more closed configuration,
toward the NADH molecule, than in the Apoꢀform (Fig. 4A); However, in the compound 2
bound form, the main chain of the NADH binding region was moved away by about 1.1 Å from
the position of the NADH molecule in the NADHꢀbound form (Fig. 4B). At the dimer interface,
binding of the compound 2 pushed the main chain of the each monomer away from the interface
by up to 1.0 Å (Fig. 4C). Our observations that 1) most mutations at the compound 2 binding
sites impaired compound 2 inhibition but produced only a small effect on the catalytic ability of
PfNDH2, 2) compound 2 behaved as a nonꢀcompetitive inhibitor and 3) binding of compound 2
to distant regions of the protein produced structural shifts at the PfNDH2 catalytic site, together
suggested a potential allosteric effect of compound 2 on NADH binding and PfNDH2 activity.
To test this hypothesis, we performed Surface Plasmon Resonance (SPR) assays for the NADH
binding affinity of PfNDH2 in the presence or absence of compound 2. Results from these assays
and calculated dissociation constants (KDs) confirmed that compound 2 decreased the binding
affinity of the NADH for PfNDH2 by about 7 fold in these experiments (Fig. 4D and Fig. S5).
We also found that compound 2 was hard to come off from PfNDH2 in the SPR studies (Fig. S6)
which could explain the high inhibitory effect of compound 2 to PfNDH2 in terms of the binding
affinity. In the PfNDH2 enzyme, FAD binds deeply within the first of two Rossmann domains of
NDH2 monomer, where the isoalloxazine ring of this cofactor lies in parallel closely from the
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Page 9 of 40
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
nicotinamide ring of NADH in the second Rossmann domain. This close proximity is thought to
support the transfer of electrons from NADH to FAD, forming FADH2 which interacts and
converts ubiquinone to ubiquinol. The effect of compound 2’s binding is to transmit a shift of
helix 7 in the opposite direction, forcing NADH to dissociate more readily from PfNDH2 (Fig.
4E). We suggest that these and possibly other allosteric effects of compound 2 impair electron
transfer from NADH to FAD, compromising subsequent electron transfer through FADH2ꢀ
ubiquinone complex and blocking the reduction of ubiquinone to ubiquinol, which might affect
the dynamic of the homoꢀdimerization of the proteins finally.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Compound 2 and 3 kill drug-resistant strains of P. falciparum. To further evaluate the antiꢀ
malarial potential of PfNDH2 inhibitors, we tested compound 2 and 3 against ten P. falciparum
strains that are from different malariaꢀendemic areas worldwide and carry various phenotypes of
resistance to major antimalarial drugs in clinical use31. Doseꢀresponse data from these
experiments yielded characteristic sigmoid response curves and a narrow range of EC50 values (2
to 12 nM; Fig. 5A and Fig. S7) which, in this low nM range, are desirable for drug discovery and
are characteristic of firstꢀline antimalarials such as artemisinin derivatives and the
spiroindolones32. There was no correlation of compound 2 or 3 EC50 with resistance to any
particular antiꢀmalarial drug, reinforcing the potential of these two compounds. We also tested
whether parasites were completely killed or some could persist through an exposure to high
concentration levels of drug. After 72 hr exposure to 10 × EC50 concentrations of compound 2 or
3, no recrudescence of 3D7, Dd2 or 803 parasites was observed in cultures maintained for 28
days post treatment (Fig. 5B).
Transcripts of the PfNDH2 gene were detected at highest level in late asexual stages (Fig. 5C)
and suggested experiments to determine the stageꢀspecific inhibitory activity of compound 2 or 3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 40
1
2
3
4
5
6
7
8
9
against P. falciparum. We first examined morphological changes of drugꢀtreated parasites by
microscopy. Treatments initiated in the ring stage did not cause obvious alterations of parasite
morphology until 36 h post invasion (Fig. 5D). Further experiments was conducted, in which
compound 2 or 3 was applied to synchronized early ring stages and then removed from culture at
2 h intervals after the parasites entered mature stages of development (34 h trophozoite and
older). It was found that these compounds reduced parasitemia in the subsequent cycle by ≥ 50%
only if drug pressure was maintained for more than 40 h post invasion (Fig. 5E). Much mature
stage parasites may therefore be particularly susceptible to the effects of compound 2 and 3
during the intraerythrocytic life cycle of P. falciparum.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The anti-malarial potential of PfNDH2 inhibitors. MTT cell proliferation assays showed no
reduced viability of HeLa cells at compound 2 and 3 concentrations of up to 2 ꢁM (Fig. 6A),
suggesting a good therapeutic window for these compounds. To evaluate the compound 2 and 3
against malaria infection in vivo, we took advantage of the high sequence conservation of NDH2
proteins among different Plasmodium species (Fig. S8) and observed the outcomes after
treatment of P. yoelii infection in mice (Fig. 6BꢀE). Twentyꢀfour hours after inoculation of P.
yoelii parasites (1×106), we treated the groups of three mice with single intraperitoneal doses of
10 mg/kg, 30 mg/kg or 90 mg/kg of compound 2 or 3. Pyrimethamine and dihydroartemisinin
served as two antiꢀmalarial drugs for control experiments. Like pyrimethamine and
dihydroartemisinin, compound 2 eliminated P. yoelii infection at all tested dosages, whereas
compound 3 failed to cure infected mice at the dose of 10 mg/kg (Fig. 6B). Mice given no
treatment (DMSO only) did not survive after 9 days of infection.
To test whether the new PfNDH2 inhibitors were also able to cure mice with high parasitemia,
we administered four daily subcutaneous doses of compound 2 (10 mg/kg/d) or 3 (30 mg/kg/d)
ACS Paragon Plus Environment

Page 11 of 40
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
to mice with ~30% infected erythrocytes four days after inoculation. These treatments achieved
dramatic reductions of parasitemia by day 10 (Fig. 6C, D). Mice treated with 10 mg/kg/d
compound 2 were completely clear of parasites by day 17 and showed 100% survival at day 40
(Fig. 6E). However, the compound 3 treatments of 30 mg/kg/d resulted in only 67% survival of
the mice (Fig. 6E). These in vivo data from mice, in contrast to the higher activity of compound 3
than of compound 2 against P. falciparum in vitro, suggest reduced bioavailability or more rapid
elimination of compound 3 (or active metabolites) than of compound 2. We also tested potential
drug interactions of compound 2 with a number of major antiꢀmalarial drugs against P.
falciparum in in vitro doseꢀresponse assays33,34. Isobologram analysis showed strong synergistic
interaction of compound 2 with dihydroartemisinin (ΣFICs=0.56±0.07) (Fig. 6F). Conversely,
antagonism occurred between compound 2 and quinine (ΣFICs=2.33±0.48), and a potential
antagonistic/additive interaction was evident with chloroquine (ΣFICs=1.52±0.19) (Fig. 6F).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
SUMMARY AND CONCLUSIONS
In the present study, we solved the crystal structures of PfNDH2 and complexes with the small
molecule that could inhibit the enzyme activity and parasites growth both in vitro and in vivo. In
the most reported cases, the allosteric ligands exert their effects by inducing conformational
changes of a target protein that are then transduced from the allosteric binding pocket to the
orthosteric domain and/or directly to effect protein catalytic sites. In our case, the compound 2
seemed to bind to the two allosteric binding sites, which could influence the NADH binding (Fig.
4). Interestingly, both proposed allosteric binding sites are associated with the CTD of PfNDH2
that enabled the NDH2 to form a homoꢀdimer in situ and attach on the membrane28,29. Binding of
compounds to this domain can result in greater potential selectivity that could provide maximal
benefit with minimal side effects. The new compound 2 displays a strong synergistic interaction
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 40
1
2
3
4
5
6
7
8
9
with dihydroartemisinin in killing parasites (Fig. 6F), which is consistent with the previous
studies that the artemisinin might act on mitochondrial ETC to inhibit parasite growth35. It has
been demonstrated that in P. falciparum the complete Fꢀtype ATP synthase appears to be
imperfect and it could not synthesize ATP by itself36,37, in consistence with a more recent report
that indicated the glycolysis as the main ATP generating process of the bloodꢀstage parasites38.
The mitochondrial ETC mainly functions as the regeneration of ubiquinone for pyrimidine
biosynthesis by dihydroorotate dehydrogenase, which is important for the development of the
bloodꢀstage P. falciparum11. This could explain why both compound 2 and 3 mainly inhibit
trophozoite/schizont stages (Fig. 5DꢀF). It is important to note that PfNDH2 is also expressed in
gametocyte/ookinete/sporozoite stages beyond the late bloodꢀstage39, suggesting a potential
activity of PfNDH2 inhibitors against both sexual and exoꢀerythrocytic forms of parasites. The
recommended treatment for both uncomplicated and severe malaria is a combination of two or
more antiꢀmalarials with different mechanisms of action40ꢀ42. Such as the atovaquone whose
monotherapy would give rapid rise drug resistance corresponding to the bc1 mutations43, it was
used as a combination component with proguanil in the drug Malarone. Considering the
emergence of artemisininꢀresistant P. falciparum strains, compound 2 based molecules may
provide an additional choice for the clinical use of ACT against malaria due to its synergistic
activity with dihydroartemisinin.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In summary, PfNDH2 inhibitors such as compound 2 could eliminate malaria both in vitro and in
vivo with a possible allosteric mechanism. The new mechanism could be suggested to attack
pathogens that rely on NDH2 for mitochondrial respiratory chain function in drug discovery. So
it not only is important for malarial PfNDH2 proteinꢀbased drug development, but also could has
broad implications for the treatment of other NDH2ꢀcontaining pathogenic microorganisms.
ACS Paragon Plus Environment

Page 13 of 40
Journal of Medicinal Chemistry
1
2
3
4
EXPERIMENTAL SECTION
5
6
7
Chemistry. Synthesis of compounds. All commercial chemical materials (Energy, Ouhe,
8
9
Aladdin, J&K Chemical Co. Ltd) were used without further purification. All solvents were
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
13
analytical grade. The H NMR and C NMR spectra were recorded on a Bruker AVANCE III
400 MHz spectrometer in CDCl3, CD3OD, or d6ꢀDMSO using tetramethylsilane (TMS) or
solvent peak as a standard. All 13C NMR spectra were recorded with complete proton decoupling.
Lowꢀresolution mass spectral analyses were performed with an Waters AQUITY UPLCTM/MS.
Analytical thinꢀlayer chromatography (TLC) was performed on Yantai Chemical Industry
Research Institute silica gel 60 F254 plates, and flash column chromatography was performed on
Qingdao Haiyang Chemical Co. Ltd silica gel 60 (200ꢀ300 mesh). A BUCHI Rotavapor Rꢀ3 was
used to remove solvents by evaporation. The purity of all the final tested compounds is more
than 95% confirmed by NMR and UPLC. The general procedures for synthesis of target
molecules compound 2 and 3 was described as below. Synthesis of other intermediates and
compounds was followed as procedure for compound 2 or achieved as shown in Supporting
Information.
5-Fluoro-3-methyl-2-(4-(4-(trifluoromethoxy)benzyl)phenyl)quinolin-4(1H)-one (2). To a 50
ml roundꢀbottom flask, was added 554 mg (1.8 mmol, 1.0 eq) compound 9, 370 mg (1.8 mmol,
1.0 eq) compound 6, 120 mg (0.63 mmol, 0.35 eq) PTSA.H2O and 10 ml nꢀBuOH. After reflux
at 130 ˚C for 16 h, the solvent was removed by evaporation, 100 ml EtOAc was added and the
mixture washed with 50 ml saturated NaHCO3 solution, then 100 ml saturated sodium chloride
solution. The organic later was dried over 1 g anhydrous Na2SO4 and purified by 200ꢀ300 mesh
silica gel flash column chromatography (Hexane:EtOAc= 1:1), yielding 200 mg compound 2
(I.Y. =30%). 1HꢀNMR(400MHz, d6-DMSO, ppm): 1.82(s, 3H), 4.07(s, 2H) , 6.92(dd, J = 8.0 Hz,
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 40
1
2
13
3
J = 11.8 Hz, 1H), 7.30(d, J = 8.1 Hz, 1H), 7.36ꢀ7.55(m, 8H), 11.55(s, 1H). CꢀNMR(100MHz,
d6-DMSO, ppm): 175.61, 160.55(d, JCꢀF = 258.4 Hz), 146.90, 142.46, 141.86(d, JCꢀF = 4.3 Hz),
140.60, 132.62, 131.70(d, JCꢀF = 10.1 Hz), 130.58, 129.20, 128.89, 121.22, 120.18(d, JCꢀF = 254.2
Hz), 116.18, 114.16(d, JCꢀF = 4.2 Hz), 113.04(d, JCꢀF = 8.7 Hz), 108.35(d, JCꢀF = 20.9 Hz), 40.07,
12.10. LCꢀMS: calculated for C24H18F4NO2 [M+H]+: 428.12, found 428.28.
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
5-Fluoro-3-methyl-2-(4-(4-((trifluoromethyl)thio)benzyl)phenyl)quinolin-4(1H)-one (3). To
a 50 ml roundꢀbottom flask, was added 1.0 g (3.0 mmol, 1.0 eq) compound 10, 650 mg (3.0
mmol, 1.0 eq) compound 6, 240 mg (1.26 mmol, 0.4 eq) PTSA.H2O and 10 ml nꢀBuOH. After
reflux at 130 ˚C for 16 h，most solvent was removed by evaporation, 50 ml EtOAc were added
and the mixture was washed with 50 ml saturated NaHCO3 solution, then 100 ml saturated
sodium chloride solution. The organic layer was dried over 1 g anhydrous Na2SO4 and purified
by 200ꢀ300 mesh silica gel flash column chromatography (Hexane:EtOAc= 1:1), yielding 320
1
mg compound 3 (I.Y. = 22%). HꢀNMR(400 MHz, d6-DMSO, ppm): 1.82(s, 3H), 4.11(s, 2H) ,
6.93(m，1H), 7.37ꢀ7.46(m, 8H), 7.65ꢀ7.67(m, 2H), 11.55(s,1H). 13CꢀNMR(100MHz, d6-DMSO,
ppm): 175.49, 160.51(d, JCꢀF = 257.6 Hz), 146.71, 144.94, 141.95, 141.81(d, JCꢀF = 4.1 Hz),
136.49,132.69, 131.58(d, JCꢀF = 10.7 Hz), 130.35, 129.65(q, JCꢀF = 305.7 Hz), 129.17, 128.91,
120.51, 116.10 , 114.08(d, JCꢀF = 4.00 Hz), 113.00(d, JCꢀF = 8.7 Hz), 108.24(d, JCꢀF = 20.8 Hz),
40.40, 12.02. LCꢀMS: calculated for C24H18F4NOS [M+H]+: 444.10, found 444.18.
5-Fluoro-1H-benzo[d][1,3]oxazine-2,4-dione (4). To a 100 ml roundꢀbottom flask, was added
4.7 g (30 mmol, 1.0 eq) 2ꢀaminoꢀ6ꢀfluorobenzoic acid, 5 g (36 mmol,1.2 eq) K2CO3, 40 ml
EtOAc and 9 g (30 mmol 1.0 eq) triphosgene. After stirring at room temperature for 24 h, the
ACS Paragon Plus Environment

Page 15 of 40
Journal of Medicinal Chemistry
1
2
3
4
mixture was poured into 100 ml saturated NaHCO3 solution and filtered to collect the solid,
5
6
compound 4 (4.9g, I.Y. = 86%).
7
8
9
2-(4,4-Dimethyl-4,5-dihydrooxazol-2-yl)-3-fluoroaniline (6). To a suspension of 4.7 g (23
mmol, 1.0 eq) compound 4 in 25 ml chlorobenzene, was added 6 ml (60 mmol, 2.5 eq) 2ꢀaminoꢀ
2ꢀmethylpropanꢀ1ꢀol and 820 mg (6 mmol, 0.25 eq) ZnCl2. After reflux at 140 ˚C for 18 h,
solvent was removed by evaporation, 100 ml DCM was added and the mixture was washed with
100 ml water. The organic layer was dried over 1.0 g anhydrous Na2SO4 and purified by 200ꢀ300
mesh siliconꢀgel flash column chromatography (Hexane: EtOAc= 10:1) to yield 1.9 g compound
6 (I.Y. = 35%). 1HꢀNMR(400 MHz, CDCl3, ppm): 7.04(m, 1H), 6.41(d, J = 8.3 Hz, 1H), 6.33(m,
1H), 6.00(b, 1H), 4.02(s, 2H), 1.35(s, 6H). LCꢀMS: calcd for C11H14N2O [M+H]+: 209.10, found
209.11.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1-(4-(Bromomethyl)phenyl)propan-1-one (7). To a solution of 2.22 g (15 mmol, 1.0 eq) 1ꢀ(pꢀ
tolyl)propanꢀ1ꢀone in 20 ml anhydrous MeCN, was added 3.0 g (17 mmol, 1.1 eq) Nꢀ
bromosuccinimide (NBS) and 246 mg 2,2'ꢀazobisꢀ(2ꢀmethylpropionitrile) (AIBN) under argon.
After reflux at 90 ˚C for 7 h，the solvent was removed by evaporation, and 100 ml DCM were
added followed by washing with 100 ml water. The organic layer was dried over 1.0 g anhydrous
Na2SO4 and purified by 200ꢀ300 mesh silica gel flash column chromatography
(Hexane:EtOAc=30:1) to get 3.1g compound 7 as colorless oil (I.Y. =85%). 1HꢀNMR(400 MHz,
CDCl3, ppm): 7.93(d, J = 8.3 Hz, 2H), 7.47(d, J = 8.3 Hz, 2H), 4.49(s, 2H), 2.98(q, J = 7.2 Hz,
2H), 1.22(t, J = 7.2 Hz, 3H). LCꢀMS: calcd for C10H12BrO [M+H]+: 226.00, found 226.11.
1-(4-(4-Ttrifluoromethoxy)benzyl)phenyl)propan-1-one (9). To a 100ml roundꢀbottom flask,
was added 3.1 g (14 mmol, 1.0 eq) compound 7, 63 mg (0.28 mmol, 0.02 eq) Pd(OAc)2, 147 mg
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 40
1
2
3
4
5
6
7
8
9
(0.56 mmol, 0.04 eq) PPh3, 6 g (28 mmol, 2.0 eq) K3PO4, 3.2 g (15.5 mmol, 1.1 eq) (4ꢀ
(trifluoromethoxy)phenyl)boronic acid and 30 ml toluene under argon. After stirring at 80 ˚C for
12 h，the solvent was removed by evaporation，100 ml water were added and the mixture was
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
extracted with 100 ml DCM. The organic layer was purified by 200ꢀ300 mesh silica gel flash
column chromatography (Hexane:EtOAc=30:1), yielding 3.0 g compound 9 as colorless oil (I.Y.
=71%).
1-(4-(4-((Trifluoromethyl)thio)benzyl)phenyl)propan-1-one (10). To a 100 ml roundꢀbottom
flask, was added 1.1g (5 mmol, 1.0 eq) compound 7, 198 mg (0.25 mmol, 0.05 eq)
Pd(dppf)2Cl2.DCM (CAS: 95464ꢀ05ꢀ4), 2 g (10 mmol, 2.0 eq) K3PO4, 1.2 g (5.5 mmol, 1.1 eq)
4,4,5,5ꢀtetramethylꢀ2ꢀ(4ꢀ((trifluoromethyl)thio) phenyl)ꢀ1,3,2ꢀdioxaborolane (CAS: 1005206ꢀ25ꢀ
6) and 30 ml toluene under argon. After stirring at 100 ˚C for 12 h，the solvent was removed by
evaporation，100 ml water were added and the mixture was extracted with 70 ml DCM. The
organic layer was purified by 200ꢀ300 mesh silica gel flash column chromatography
(Hexane:EtOAc = 30:1), yielding 1.0 g compound 10 as colorless oil (I.Y. =64%). 1HꢀNMR(400
MHz, CDCl3, ppm): 7.80(d, J = 8.2 Hz, 2H), 7.56(d, J = 8.04 Hz, 2H), 7.23(m, 4H), 4.05(s, 2H),
2.96(q, J = 7.2 Hz, 2H), 1.21(d, J = 7.2 Hz, 2H). LCꢀMS: calcd for C17H15F3OS [M+H]+: 325.08,
found 325.11.
Biology. Crystallization, data collection and structure determination. Crystals were grown at
18 ˚C by hangingꢀdrop method by mixing 1 µl protein (10 mg/ml) with 1 µL reservoir solution
(2.7 M sodium acetate trihydrate, pH 4.8 or 10.0 to 10.6). The crystals appeared overnight and
grew to full size in about 2 to 3 days. To obtain the NADHꢀbound PfNDH2 crystals, the crystals
were soaked at 18 ˚C in cryoꢀprotectant (reservoir solution with 25% glycerol) with 10 mM
ACS Paragon Plus Environment

Page 17 of 40
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
NADH for about 1ꢀ3 minutes. Inhibitors were coꢀcrystallized at a concentration of 0.5 mM with
PfNDH2. The crystals and their complexes were then transferred to liquid nitrogen for
subsequent analysis. Crystallographic data were collected at the Shanghai Synchrotron Radiation
Facility (SSRF) beamline BL17U, integrated and scaled using the HKL2000 package44. The
molecular replacement method was employed to solve the phase problem using the program
Phaser45 from the CCP4 suite46, with the crystal structure of Ndi1 from Saccharomyces cerevisae
(PDB code 4G6G; 36 % sequence identity) as the initial search model. Data collection statistics
are summarized in Table S1. The final model rebuilding was performed using COOT47 and the
protein structure was refined with PHENIX48 against the highꢀresolution native data using NCS49
and stereochemistry information as restraints. Full data collection and structure statistics are
summarized in Table S1.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Enzyme activity assays for PfNDH2. The activity of PfNDH2 was tested firstly. The
Kcat=664.3±31.7 Sꢀ1, Km=34.9±3.9 ꢁM (Fig. S4). The Km was similar to the published data50.
The enzymatic activity of wild type and mutant PfNDH2 proteins was measured
spectrophotometrically using NADH and Ubiquinoneꢀ1 (UQ1) as substrates. Standard assays
were carried out at 25 ˚C in a 1.6 ml reaction mixture containing 50 mM MOPS buffer pH 7.0,
150 mM NaCl, 1 mM EDTA, 0.01% TritonꢀXꢀ100, 50 ꢁM NADH, 30 ꢁM UQ1, 0.5 nM enzymes
and selected concentrations of inhibitors. Reactions were initiated by the enzyme addition.
Progress of the reaction was monitored continuously by following the decrease of signal from
NADH at 340 nm, in a Lambda 45 spectrophotometer (PerkinElmer Life Sciences) equipped
with a magnetic stirrer in the cuvette holder. Activities were calculated using an NADH
extinction of 6220 cmꢀ1Mꢀ1 at 340 nm.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 40
1
2
3
4
5
6
7
8
9
Parasites growth inhibition assays. For synchronized assay, asynchronous cultures of schizont
stage parasites were pretreated with 5% sorbital. Plasmodium falciparum strains as mentioned in
the text (1% parasitemia, 2% hematocrit) at the mid ring stage (~6ꢀ10 hours postꢀinvasion) were
used to test antimalarial effects in 96ꢀwell plates. Parasites were incubated in triplicate in a 96
well plate compound 2, 3, or artemisinin at 200 nM. Parasites were further diluted at equal ratio
containing 1% parasitemia, 2% hematocrit from the maxim concentration of 100 nM. Parasites
were allowed to grow for 72ꢀhour incubation period under an atmosphere of gas mixture
containing 5% CO2, 5% O2, and 90% N2 at 37 °C. After 72 hours, the media were replaced with
100 ꢁl lysis buffer (10 mM KHCO3, 150 mM NH4Cl, 0.1 mM EDTA) consisting of 5×SYBR
Green I (Invitrogen; supplied in 10,000×concentration). The plates were then incubated for 20
min in the dark. The fluorescence signal was quantitated at 485 nm excitation and 538 nm
emission by a microplate fluorometer. Stage of action studies were carried out by immediately
adding 10×EC50 inhibitor and parasites were visualized at 16, 24, 32, 40, 48, 60 h after
synchronization. Giemsa staining of fixed smears was for comparison of parasite morphology at
each time point. Parasitemia was quantified by a light microscope (100 × objective lens)
equipped with a PowerShot G6 digital camera (Leica). To test parasite maturation throughout a
single (~48 h) asexual bloodꢀstage cycle, compounds were removed every 2 h starting at life
cycle t=34 h (trophozoite stage). The growth inhibition was determined by SYBR Green I DNA
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ASSOCIATED CONTENT
Supporting Information. This material is available free of charge via the Internet at
http://pubs.acs.org. Other experimental details including: protein expression and purification; the
DNA point mutants; in vivo efficacy studies on mice; MTT assay; SPR study; Figure S1ꢀS8,
Table S1ꢀS3 and synthesis of compounds.
ACS Paragon Plus Environment

Page 19 of 40
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Accession Codes. The atomic coordinates and structure factors (code 5JWA for Apo, 5JWB for
NADH and 5JWC for compound 2) have been deposited in the Protein Data Bank
(http://www.pdb.org/). Authors will release the atomic coordinates and experimental data upon
article publication.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
AUTHOR INFORMATION
Corresponding Author
Maojun Yang, maojunyang@tsinghua.edu.cn, +86ꢀ010ꢀ62789400;
Yu Rao, yrao@tsinghua.edu.cn, +86ꢀ010ꢀ62782025;
Lubin Jiang, lbjiang@ips.ac.cn, +86ꢀ021ꢀ54923072.
Author Contributions
†These authors contributed equally. Y.R., M.Y., and L.J. conceived the study. X.L., Y.Y., J.L.,
J.Y. and J.G. purified the proteins, grew the crystals and performed the enzyme activity analyses.
YQ.Y. and Y.R. designed and synthesized inhibitors. Y.Y. and M.Y. collected synchrotron
diffraction data and solved the crystal structures. Z.H. and L.J. performed the in vitro and in vivo
studies. Y.R., Y.M., L.J., Y.Y and YQ.Y. analyzed the data and wrote the paper.
Funding Sources
This work was supported by grants from the Ministry of Science and Technology of China
(2016YFA0501100), the National Science Fund for Distinguished Young Scholars (31625008),
the National Natural Science Foundation of China (21302106, 81573277, 81622042,
81361120405, 21532004 and 31570733), Tsinghua University Initiative Scientific Research
Program and National Institutes of Health (RO1AI116466).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 40
1
2
3
4
Notes
5
6
7
The authors declare no competing financial interest.
8
9
ACKNOWLEDGMENTS
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
We would like to thank the staff at the SSRF BL17U beamline for their assistance in data
collection and thank Dr. Heng Wang of Peking Union Medical Sciences for helpful discussion.
ABBREVIATIONS
NADH, Reduced form of nicotinamideꢀadenine dinucleotid; PfNDH2, Ubiquinone
oxidoreductase of Plasmodium falciparum; ETC, Electron transport chain; CTD, Cꢀterminal
domain; AA, amino acids; FAD, Flavin adenine dinucleotide; RMSDs, Rootꢀmean square
deviation values; SPR, Surface Plasmon Resonance; ATP, Adenosine triphosphate; SAR,
Structureꢀactivity relationship; PYR, Pyrimethamine; DHA, Dihydroartemisinin; DMSO,
Dimethyl sulfoxide; FICs, fractional inhibitory concentrations.
REFERENCES
(1) World Malaria Report; World Health Organization: Geneva, www.who.int, (accessed
December 13, 2016).
(2) Ariey, F.; Witkowski, B.; Amaratunga, C.; Beghain, J.; Langlois, A. C.; Khim, N.; Kim, S.;
Duru, V.; Bouchier, C.; Ma, L.; Lim, P.; Leang, R.; Duong, S.; Sreng, S.; Suon, S.; Chuor, C.
M.; Bout, D. M.; Menard, S.; Rogers, W. O.; Genton, B.; Fandeur, T.; Miotto, O.; Ringwald, P.;
Le Bras, J.; Berry, A.; Barale, J. C.; Fairhurst, R. M.; BenoitꢀVical, F.; MercereauꢀPuijalon, O.;
ACS Paragon Plus Environment

Page 21 of 40
Journal of Medicinal Chemistry
1
2
3
4
5
6
Menard, D. A molecular marker of artemisininꢀresistant Plasmodium falciparum malaria. Nature
2014, 505, 50ꢀ55.
7
8
9
(3) St. Laurent, B.; Miller, B.; Burton, T. A.; Amaratunga, C.; Men, S.; Sovannaroth, S.; Fay, M.
P.; Miotto, O.; Gwadz, R. W.; Anderson, J. M.; Fairhurst, R. M. Artemisininꢀresistant
Plasmodium falciparum clinical isolates can infect diverse mosquito vectors of Southeast Asia
and Africa. Nat. Commun. 2015, 6, DOI: 10.1038/ncomms9614.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(4) Wellems, T. E.; Hayton, K.; Fairhurst, R. M. The impact of malaria parasitism: from
corpuscles to communities. J. Clin. Invest. 2009, 119, 2496ꢀ2505.
(5) Dondorp, A. M.; Fairhurst, R. M.; Slutsker, L.; MacArthur, J. R.; M.D., J. G. B.; Guerin, P.
J.; Wellems, T. E.; Ringwald, P.; Newman, R. D.; Plowe, C. V. The threat of artemisininꢀ
resistant malaria. N. Engl. J. Med. 2011, 365, 1073ꢀ1075.
(6) Pantaleo, A.; Pau, M. C.; Chien, H. D.; Turrini, F. Artemisinin resistance, some facts and
opinions. J. Infect. Dev. Countries 2015, 9, 597ꢀ599.
(7) Ashley, E. A.; Dhorda, M.; Fairhurst, R. M.; Amaratunga, C.; Lim, P.; Suon, S.; Sreng, S.;
Anderson, J. M.; Mao, S.; Sam, B.; Sopha, C.; Chuor, C. M.; Nguon, C.; Sovannaroth, S.;
Pukrittayakamee, S.; Jittamala, P.; Chotivanich, K.; Chutasmit, K.; Suchatsoonthorn, C.;
Runcharoen, R.; Hien, T. T.; ThuyꢀNhien, N. T.; Thanh, N. V.; Phu, N. H.; Htut, Y.; Han, K.ꢀT.;
Aye, K. H.; Mokuolu, O. A.; Olaosebikan, R. R.; Folaranmi, O. O.; Mayxay, M.; Khanthavong,
M.; Hongvanthong, B.; Newton, P. N.; Onyamboko, M. A.; Fanello, C. I.; Tshefu, A. K.; Mishra,
N.; Valecha, N.; Phyo, A. P.; Nosten, F.; Yi, P.; Tripura, R.; Borrmann, S.; Bashraheil, M.;
Peshu, J.; Faiz, M. A.; Ghose, A.; Hossain, M. A.; Samad, R.; Rahman, M. R.; Hasan, M. M.;
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 40
1
2
3
4
5
6
7
8
9
Islam, A.; Miotto, O.; Amato, R.; MacInnis, B.; Stalker, J.; Kwiatkowski, D. P.; Bozdech, Z.;
Jeeyapant, A.; Cheah, P. Y.; Sakulthaew, T.; Chalk, J.; Intharabut, B.; Silamut, K.; Lee, S. J.;
Vihokhern, B.; Kunasol, C.; Imwong, M.; Tarning, J.; Taylor, W. J.; Yeung, S.; Woodrow, C. J.;
Flegg, J. A.; Das, D.; Smith, J.; Venkatesan, M.; Plowe, C. V.; Stepniewska, K.; Guerin, P. J.;
Dondorp, A. M.; Day, N. P.; White, N. J. Spread of artemisinin resistance in Plasmodium
falciparum malaria. N. Engl. J. Med. 2014, 371, 411ꢀ423.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(8) Djimde, A. A.; Makanga, M.; Kuhen, K.; Hamed, K. The emerging threat of artemisinin
resistance in malaria: focus on artemetherꢀlumefantrine. Expert Rev. Anti-Infect. Ther. 2015, 13,
1031ꢀ1045.
(9) Srivastava, I. K.; Rottenberg, H.; Vaidya, A. B. Atovaquone, a broad spectrum antiparasitic
drug, collapses mitochondrial membrane potential in a malarial parasite. J. Biol. Chem. 1997,
272, 3961ꢀ3966.
(10) Birth, D.; Kao, W. C.; Hunte, C. Structural analysis of atovaquoneꢀinhibited cytochrome
bc1 complex reveals the molecular basis of antimalarial drug action. Nat. Commun. 2014, 5,
DOI: 10.1038/ncomms5029.
(11) Painter, H. J.; Morrisey, J. M.; Mather, M. W.; Vaidya, A. B. Specific role of mitochondrial
electron transport in bloodꢀstage Plasmodium falciparum. Nature 2007, 446, 88ꢀ91.
(12) Melo, A. M.; Bandeiras, T. M.; Teixeira, M. New insights into type II NAD (P) H: quinone
oxidoreductases. Microbiol. Mol. Biol. Rev. 2004, 68, 603ꢀ616.
(13) Duarte, M.; Tomás, A. M. The mitochondrial complex I of trypanosomatidsꢀan overview of
current knowledge. J. Bioenerg. Biomembr. 2014, 46, 299ꢀ311.
ACS Paragon Plus Environment

Page 23 of 40
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(14) Biagini, G. A.; Fisher, N.; Shone, A. E.; Mubaraki, M. A.; Srivastava, A.; Hill, A.; Antoine,
T.; Warman, A. J.; Davies, J.; Pidathala, C.; Amewu, R. K.; Leung, S. C.; Sharma, R.; Gibbons,
P.; Hong, D. W.; Pacorel, B.; Lawrenson, A. S.; Charoensutthivarakul, S.; Taylor, L.; Berger, O.;
Mbekeani, A.; Stocks, P. A.; Nixon, G. L.; Chadwick, J.; Hemingway, J.; Delves, M. J.; Sinden,
R. E.; Zeeman, A. M.; Kocken, C. H.; Berry, N. G.; O'Neill, P. M.; Ward, S. A. Generation of
quinolone antimalarials targeting the Plasmodium falciparum mitochondrial respiratory chain for
the treatment and prophylaxis of malaria. Proc. Natl. Acad. Sci. U. S. A. 2012, 109, 8298ꢀ8303.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(15) Yano, T.; Rahimian, M.; Aneja, K. K.; Schechter, N. M.; Rubin, H.; Scott, C. P.
Mycobacterium tuberculosis type II NADHꢀmenaquinone oxidoreductase catalyzes electron
transfer through a twoꢀsite pingꢀpong mechanism and has two quinoneꢀbinding sites.
Biochemistry 2014, 53, 1179ꢀ1190.
(16) Verner, Z.; Skodova, I.; Polakova, S.; DurisovaꢀBenkovicova, V.; Horvath, A.; Lukes, J.
Alternative NADH dehydrogenase (NDH2): intermembraneꢀspaceꢀfacing counterpart of
mitochondrial complex I in the procyclic Trypanosoma brucei. Parasitology 2013, 140, 328ꢀ337.
(17) Sassetti, C. M.; Rubin, E. J. Genetic requirements for mycobacterial survival during
infection. Proc. Natl. Acad. Sci. U. S. A. 2003, 100, 12989ꢀ12994.
(18) Pidathala, C.; Amewu, R.; Pacorel, B.; Nixon, G. L.; Gibbons, P.; Hong, W. D.; Leung, S.
C.; Berry, N. G.; Sharma, R.; Stocks, P. A.; Srivastava, A.; Shone, A. E.; Charoensutthivarakul,
S.; Taylor, L.; Berger, O.; Mbekeani, A.; Hill, A.; Fisher, N. E.; Warman, A. J.; Biagini, G. A.;
Ward, S. A.; O'Neill, P. M. Identification, design and biological evaluation of bisaryl quinolones
targeting Plasmodium falciparum type II NADH:quinone oxidoreductase (PfNDH2). J. Med.
Chem. 2012, 55, 1831ꢀ1843.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 40
1
2
3
4
5
6
7
8
9
(19) Sharma, R.; Lawrenson, A. S.; Fisher, N. E.; Warman, A. J.; Shone, A. E.; Hill, A.;
Mbekeani, A.; Pidathala, C.; Amewu, R. K.; Leung, S.; Gibbons, P.; Hong, D. W.; Stocks, P.;
Nixon, G. L.; Chadwick, J.; Shearer, J.; Gowers, I.; Cronk, D.; Parel, S. P.; O'Neill, P. M.; Ward,
S. A.; Biagini, G. A.; Berry, N. G. Identification of novel antimalarial chemotypes via
chemoinformatic compound selection methods for a highꢀthroughput screening program against
the novel malarial target, PfNDH2: increasing hit rate via virtual screening methods. J. Med.
Chem. 2012, 55, 3144ꢀ3154.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(20) Saleh, A.; Friesen, J.; Baumeister, S.; Gross, U.; Bohne, W. Growth inhibition of
Toxoplasma gondii and Plasmodium falciparum by nanomolar concentrations of 1ꢀhydroxyꢀ2ꢀ
dodecylꢀ4 (1H) quinolone, a highꢀaffinity inhibitor of alternative (type II) NADH
dehydrogenases. Antimicrob. Agents Chemother. 2007, 51, 1217ꢀ1222.
(21) Fisher, N.; Warman, A. J.; Ward, S. A.; Biagini, G. A. Type II NADH: quinone
oxidoreductases of Plasmodium falciparum and Mycobacterium tuberculosis: kinetic and high‐
throughput assays. Methods Enzymol. 2009, 456, 303ꢀ320.
(22) Wermuth, C. G. The Practice of Medicinal Chemistry. Academic Press: New York, 2008;
Ch. 3.
(23) Conn, P. J.; Jones, C. K.; Lindsley, C. W. Subtypeꢀselective allosteric modulators of
muscarinic receptors for the treatment of CNS disorders. Trends Pharmacol. Sci. 2009, 30, 148ꢀ
155.
(24) Hardy, J. A.; Wells, J. A. Searching for new allosteric sites in enzymes. Curr. Opin. Struct.
Biol. 2004, 14, 706ꢀ715.
ACS Paragon Plus Environment

Page 25 of 40
Journal of Medicinal Chemistry
1
2
3
4
(25) Christopoulos, A. Allosteric binding sites on cellꢀsurface receptors: novel targets for drug
5
6
discovery. Nat. Rev. Drug Discovery 2002, 1, 198ꢀ210.
7
8
9
(26) Nussinov, R.; Tsai, C. J. Allostery in disease and in drug discovery. Cell 2013, 153, 293ꢀ
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
305.
(27) Smet, F. D.; Christopoulos, A.; Carmeliet, P. Allosteric targeting of receptor tyrosine
kinases. Nat. Biotech. 2014, 32, 1113ꢀ1120.
(28) Feng, Y.; Li, W.; Li, J.; Wang, J.; Ge, J.; Xu, D.; Liu, Y.; Wu, K.; Zeng, Q.; Wu, J. W.;
Tian, C.; Zhou, B.; Yang, M. Structural insight into the typeꢀII mitochondrial NADH
dehydrogenases. Nature 2012, 491, 478ꢀ482.
(29) Iwata, M.; Lee, Y.; Yamashita, T.; Yagi, T.; Iwata, S.; Cameron, A. D.; Maher, M. J. The
structure of the yeast NADH dehydrogenase (Ndi1) reveals overlapping binding sites for waterꢀ
and lipidꢀsoluble substrates. Proc. Natl. Acad. Sci. U. S. A. 2012, 109, 15247ꢀ15252.
(30) Holm, L.; Rosenstrom, P. Dali server: conservation mapping in 3D. Nucleic Acids Res.
2010, 38, W545ꢀ549.
(31) Mu, J.; Myers, R. A.; Jiang, H.; Liu, S.; Ricklefs, S.; Waisberg, M.; Chotivanich, K.;
Wilairatana, P.; Krudsood, S.; White, N. J.; Udomsangpetch, R.; Cui, L.; Ho, M.; Ou, F.; Li, H.;
Song, J.; Li, G.; Wang, X.; Seila, S.; Sokunthea, S.; Socheat, D.; Sturdevant, D. E.; Porcella, S.
F.; Fairhurst, R. M.; Wellems, T. E.; Awadalla, P.; Su, X.Z. Plasmodium falciparum genomeꢀ
wide scans for positive selection, recombination hot spots and resistance to antimalarial drugs.
Nat. Genet. 2010, 42, 268ꢀ271.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 40
1
2
3
4
5
6
7
8
9
(32) Rottmann, M.; Mcnamara, C.; Yeung, B. K. S.; Lee, M. C. S.; Zou, B.; Russell, B.; Seitz,
P.; Plouffe, D. M.; Dharia, N. V.; Tan, J. Spiroindolones, a potent compound class for the
treatment of malaria. Science 2010, 329, 1175ꢀ1180.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(33) Fidock, D. A.; Rosenthal, P. J.; Croft, S. L.; Brun, R.; Nwaka, S. Antimalarial drug
discovery: efficacy models for compound screening. Nat. Rev. Drug Discovery 2004, 3, 509ꢀ520.
(34) Hegreness, M.; Shoresh, N.; Damian, D.; Hartl, D.; Kishony, R. Accelerated evolution of
resistance in multidrug environments. Proc. Natl. Acad. Sci. U. S. A. 2008, 105, 13977ꢀ13981.
(35) Moon, A. M.; Biggs, H. M.; Rubach, M. P.; Crump, J. A.; Maro, V. P.; Saganda, W.;
Reddy, E. A. Artemisinin directly targets malarial mitochondria through its specific
mitochondrial activation. PLoS One 2014, 9, e89814.
(36) 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.; Fairlamb, A. H.; Fraunholz, M. J.; Roos, D. S.; Ralph, S. A.; McFadden, G. I.;
Cummings, L. M.; Subramanian, 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 2002, 419, 498ꢀ511.
(37) Vaidya, A. B.; Mather, M. W. A postꢀgenomic view of the mitochondrion in malaria
parasites. Curr. Top. Microbiol. Immunol. 2005, 295, 233ꢀ250.
ACS Paragon Plus Environment

Page 27 of 40
Journal of Medicinal Chemistry
1
2
3
4
(38) Boysen, K. E.; Matuschewski, K. Arrested oocyst maturation in Plasmodium parasites
5
6
lacking type II NADH:ubiquinone dehydrogenase. J. Biol. Chem. 2011, 286, 32661ꢀ32671.
7
8
9
(39) Le, R. K.; Zhou, Y.; Blair, P. L.; Grainger, M.; Moch, J. K.; Haynes, J. D.; De, L. V. P.;
Holder, A. A.; Batalov, S.; Carucci, D. J. Discovery of gene function by expression profiling of
the malaria parasite life cycle. Evaluation of inꢀhospital management for febrile illness in
Northern Tanzania before and after 2010 World Health Organization Guidelines for the
treatment of malaria. Science 2003, 301, 1503ꢀ1508.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(40) Moon, A. M.; Biggs, H. M.; Rubach, M. P.; Crump, J. A.; Maro, V. P.; Saganda, W.;
Reddy, E. A. Evaluation of inꢀhospital management for febrile illness in Northern Tanzania
before and after 2010 World Health Organization Guidelines for the treatment of malaria. PLoS
One 2014, 9, e89814.
(41) Reyburn, H. New WHO guidelines for the treatment of malaria. BMJ 2010, 340, c2637.
(42) Khan, M. A.; Smego, R. A., Jr.; Razi, S. T.; Beg, M. A. merging drugꢀꢀresistance and
guidelines for treatment of malaria. J. Coll. Physicians Surg. Pak. 2004, 14, 319ꢀ324.
(43) Srivastava, I. K.; Vaidya, A. B. A mechanism for the synergistic antimalarial action of
atovaquone and proguanil. Antimicrob. Agents Chemother. 1999, 43, 1334ꢀ1339.
(44) Otwinowski, Z.; Minor, W. Processing of Xꢀray diffraction data collected in oscillation
mode. Methods Enzymol. 1997, 276, 307ꢀ326.
(45) McCoy, A. J.; GrosseꢀKunstleve, R. W.; Adams, P. D.; Winn, M. D.; Storoni, L. C.; Read,
R. J. Phaser crystallographic software. J. Appl. Crystallogr. 2007, 40, 658ꢀ674.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 40
1
2
3
4
5
6
(46) Bailey, S. The CCP4 suite: programs for protein crystallography. Acta Crystallogr., Sect. D:
Biol. Crystallogr. 1994, 50, 760ꢀ763.
7
8
9
(47) Emsley, P.; Cowtan, K. Coot: modelꢀbuilding tools for molecular graphics. Acta
Crystallogr., Sect. D: Biol. Crystallogr. 2004, 60, 2126ꢀ2132.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(48) Adams, P. D.; GrosseꢀKunstleve, R. W.; Hung, L. W.; Ioerger, T. R.; McCoy, A. J.;
Moriarty, N. W.; Read, R. J.; Sacchettini, J. C.; Sauter, N. K.; Terwilliger, T. C. PHENIX:
building new software for automated crystallographic structure determination. Acta Crystallogr.,
Sect. D: Biol. Crystallogr. 2002, 58, 1948ꢀ1954.
(49) Terwilliger, T. Rapid automatic NCS identification using heavyꢀatom substructures. Acta
Crystallogr., Sect. D: Biol. Crystallogr. 2002, 58, 2213ꢀ2215.
(50) Biagini, G. A.; Viriyavejakul, P.; O'Neill, P. M.; Bray, P. G.; Ward, S. A. Functional
characterization and target validation of alternative complex I of Plasmodium falciparum
mitochondria. Antimicrob. Agents. Chemother. 2006, 50, 1841ꢀ1851.
ACS Paragon Plus Environment

Page 29 of 40
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. Inhibitors and overall structure of PfNDH2. (A) Three effective inhibitors of
PfNDH2. The IC50 and EC50 values of new compounds 2 and 3 are from three separate
experiments. The CLogP values are calculated by the software ChemDraw. (B) PfNDH2 forms a
homoꢀdimer. The CTDs (from 465 to 533AA) of two monomers of PfNDH2 are colored in blue
and hot pink, respectively. The domain C (from 359 to 430AA) is colored in purple. The
remaining parts of the two monomers of PfNDH2 are colored in green and cyan, respectively.
FAD is shown as stick representation. (C) Detail of interactions between the α18 helices of the
two PfNDH2 monomers. (D) The amino acids associated with dimerization are indicated by red
stars. Secondary structural elements of NDH2s are indicated above the sequences. Conserved
residues in the sequence alignment are shaded by the color scheme of the CLC (Cakeꢀloving
ACS Paragon Plus Environment