Lead optimization of a pyrrole-based dihydroorotate dehydrogenase inhibitor series for the treatment of malaria

Article abstract of DOI:10.1021/acs.jmedchem.0c00311

Malaria puts at risk nearly half the world's population and causes high mortality in sub-Saharan Africa, while drug resistance threatens current therapies. The pyrimidine biosynthetic enzyme dihydroorotate dehydrogenase (DHODH) is a validated target for malaria treatment based on our finding that triazolopyrimidine DSM265 (1) showed efficacy in clinical studies. Herein, we describe optimization of a pyrrole-based series identified using a target-based DHODH screen. Compounds with nanomolar potency versus Plasmodium DHODH and Plasmodium parasites were identified with good pharmacological properties. X-ray studies showed that the pyrroles bind an alternative enzyme conformation from 1 leading to improved species selectivity versus mammalian enzymes and equivalent activity on Plasmodium falciparum and Plasmodium vivax DHODH. The best lead DSM502 (37) showed in vivo efficacy at similar levels of blood exposure to 1, although metabolic stability was reduced. Overall, the pyrrole-based DHODH inhibitors provide an attractive alternative scaffold for the development of new antimalarial compounds.

Full text of DOI:10.1021/acs.jmedchem.0c00311

Subscriber access provided by Karolinska Institutet, University Library
Article
Lead optimization of a pyrrole-based dihydroorotate
dehydrogenase inhibitor series for the treatment of Malaria
Sreekanth Kokkonda, Xiaoyi Deng, Karen L. White, Farah El Mazouni, John White, David M.
Shackleford, Kasiram Katneni, Francis C.K. Chiu, Helena Barker, Jenna McLaren, Elly Crighton,
Gong Chen, Iñigo Angulo-Barturen, Maria Belen Jimenez-Diaz, Santiago Ferrer, Leticia Huertas
Valentin, Maria Santos Martínez-Martínez, Maria Jose Lafuente-Monasterio, Chittimalla R. Kumar,
Shatrughan P. Shahi, Sergio Wittlin, David Waterson, Jeremy N. Burrows, Dave Matthews, Diana
R Tomchick, Pradipsinh K Rathod, Michael J Palmer, Susan A. Charman, and Margaret A. Phillips
J. Med. Chem., Just Accepted Manuscript • Publication Date (Web): 06 Apr 2020
Downloaded from pubs.acs.org on April 6, 2020
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 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 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.
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 97
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 97
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Lead optimization of a pyrrole-based dihydroorotate dehydrogenase inhibitor series for the
treatment of Malaria
1
2
3
2
1
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Sreekanth Kokkonda , Xiaoyi Deng , Karen L. White , Farah El Mazouni , John White , David
3
3
3
3
3
M. Shackleford , Kasiram Katneni , Francis C.K. Chiu , Helena Barker , Jenna McLaren , Elly
3
3
4#
4#
4
Crighton , Gong Chen , Inigo Angulo-Barturen , Maria Belen Jimenez-Diaz , Santiago Ferrer ,
4
4
4
Leticia Huertas Valentin , Maria Santos Martinez-Martinez , Maria Jose Lafuente-Monasterio ,
5
5
6,7
8
Chittimalla R. Kumar , Shatrughan P. Shahi , Sergio Wittlin , David Waterson , Jeremy N.
8
8
9
1
8
Burrows , Dave Matthews , Diana Tomchick , Pradipsinh K. Rathod , Michael J. Palmer , Susan
3
2*
A. Charman and Margaret A. Phillips
1Departments of Chemistry and Global Health, University of Washington, Seattle, WA 98195;
2
9
Departments of Biochemistry and Biophysics, University of Texas Southwestern Medical Center
3
at Dallas, 5323 Harry Hines Blvd, Dallas, Texas 75390-9135; Centre for Drug Candidate
Optimisation, Monash Institute of Pharmaceutical Sciences, Monash University, Parkville, VIC
4
3
052, Australia, GSK, Tres Cantos Medicines Development Campus, Severo Ochoa, Madrid,
5
6
2
8760 Spain, Syngene International Ltd, Bangalore, India, 560 099; Swiss Tropical and Public
7
Health Institute, Socinstrasse 57, 4002 Basel, Switzerland, University of Basel, 4002 Basel,
8
Switzerland, Medicines for Malaria Venture, 1215 Geneva, Switzerland.
*
Author to whom all correspondence should be addressed. margaret.phillips@utsouthwestern.edu
ACS Paragon Plus Environment

Page 3 of 97
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
#Current Address: TAD, Biscay Science and Technology Park, Astrondo Bidea, BIC Bizkaia Bd
6
12, Derio 48160, Bizkaia, Basque Country, Spain.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 97
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Abstract
Malaria puts at risk nearly half the world’s population and causes high mortality in sub-Saharan
Africa, while drug resistance threatens current therapies. The pyrimidine biosynthetic enzyme
dihydroorotate dehydrogenase (DHODH) is a validated target for malaria treatment based on our
finding that triazolopyrimidine DSM265 (1) showed efficacy in clinical studies. Herein we
describe optimization of a pyrrole-based series identified by a target-based DHODH screen.
Compounds with nanomolar potency versus Plasmodium DHODH and Plasmodium parasites were
identified with good pharmacologic properties. X-ray studies showed the pyrroles bind an
alternative enzyme conformation from 1 leading to improved species selectivity versus
mammalian enzymes and equivalent activity on P. falciparum and P. vivax DHODH. The best
lead DSM502 (37) showed in vivo efficacy at similar levels of blood exposure to 1, although
metabolic stability was reduced. Overall the pyrrole-based DHODH inhibitors provide an
attractive alternative scaffold for development of new antimalarial compounds.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
ACS Paragon Plus Environment

Page 5 of 97
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Introduction
Infectious diseases cause the majority of deaths in low income countries, primarily the
1
result of respiratory infections, diarrheal diseases, HIV, malaria and tuberculosis (TB). Malaria
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
alone is responsible for 0.4 million deaths per year, mostly amongst young children in sub-Saharan
2
Africa. Most deaths are caused by Plasmodium falciparum malaria, however nearly 90 countries
are endemic with malaria, and P. vivax, while less lethal, has the greater global reach.3-5 While a
vaccine for the prevention of malaria has recently been recommended for large scale pilot
implementation in Africa, its efficacy is sub-optimal resulting in <50% protection. 6, 7 Therefore,
both prevention and treatment programs continue to rely on chemotherapy. While a number of
effective compounds have been employed, drug resistance has led to decreased efficacy in most
8
cases (e.g chloroquine) , a challenge that is shared with other infectious diseases like HIV and
TB.9 The current standard of care treatments for malaria, artemisinin combination therapies
(
ACTs)10, 11, are safe, effective and credited with declining cases of malaria worldwide. However,
parasites showing delayed clearance times to artemisinin derivatives have been identified in
southeast Asia, and combined with increasing resistance to partner compounds, these issues
underlie treatment failures in patients in this region.8, 10, 12
Concerns that ACTs may fail on a global scale are driving drug discovery efforts to find
new therapies for the treatment of malaria.4, 13 A number of new chemical entities have reached
clinical development, including three that are currently in Phase IIb human studies (the peroxide
OZ43914,
15
combined with ferroquine, KAF15616,
17
combined with lumefantrine and
spiroindolone KAE60918, 19 as a single agent for severe malaria). Combination therapy is
considered key to reduce the potential for resistance to develop towards any new treatment,
increasing the pressure on the malaria drug portfolio to identify a larger number of clinical
4
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 97
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
candidates. While most of the current clinical candidates were identified by asexual blood stage
screens, target based approaches have also been successful.4 Notably, dihydroorotate
dehydrogenase (DHODH) was validated as a new target by our group through the identification of
a clinical development candidate, the triazolopyrimidine DSM265 (1) (Fig. 1), targeting
DHODH.20-25 A second triazolopyrimidine (DSM421; 2-(1,1-difluoroethyl)-5-methyl-N-(6-
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(trifluoromethyl)pyridin-3-yl)-[1,2,4]triazolo[1,5-a]pyrimidin-7-amine)
reached
preclinical
development but was not progressed due to unexpected off-target toxicity 26, 27 and a thiophene
2
8
analog 2 (Fig. 1) developed by Genzyme reached late lead development but was not advanced
beyond this stage.
Fig. 1. Structures of selected DHODH inhibitors.
DHODH is a mitochondrial enzyme that is localized to the inner mitochondrial membrane
where it catalyzes a key step in the de novo synthesis of pyrimidines, the flavin-dependent
conversion of dihydroorotate to orotic acid.29, 30 Ubiquinone serves as the final oxidant in the
reaction. Because Plasmodium parasites lack salvage enzymes, the de novo pathway is essential to
5
ACS Paragon Plus Environment

Page 7 of 97
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
the formation of pyrimidines for DNA and RNA synthesis. As a consequence, inhibitors of P.
falciparum DHODH (PfDHODH) block parasite replication at the early trophozoite stage prior to
2
2
the burst of replication required to form the schizont. A key advantage of the target is that it is
required for both blood and liver stage development, and thus DHODH inhibitors meet the product
profile for either malaria treatment or chemoprevention.22, 31 1 has been clinically validated for the
treatment of P. falciparum infections during Phase I and IIa studies to test safety and efficacy in
volunteers and patients (Phase I24, 25 and Phase IIa ) and in human challenge studies to assess its
potential for chemoprevention.32, 33 1 showed considerable strengths in the clinic, including good
safety and a long human half-life (~100 hours) providing a single dose (400 mg dose) cure of P.
falciparum malaria in human studies in patients in Peru.23, 25 Additionally a single dose (400 mg)
given 1 day before human volunteers were challenged with the infectious mosquito stage of P.
falciparum (sporozoites) prevented infection, a result that likely supports once weekly dosing for
chemoprevention. 32, 33 Two potential liabilities have been the identification of resistance mutations
from both in vitro selections and in patients experiencing recrudescence 23, 34 and the finding that
in comparison to P. falciparum, 1 showed reduced activity on P. vivax in both ex vivo studies and
in a human phase IIa clinical trial.23, 26
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
23
To position additional DHODH inhibitors from a chemical series distinct from 1, we
undertook lead optimization of a pyrrole-based series that we identified in our original target-based
high throughput screen35 but did not publish. Our identified hit (DSM43 (3); Fig. 1) was
3
6
subsequently published by Genzyme , but was not advanced into lead optimization. Herein we
describe a structure-guided lead optimization program around the pyrrole scaffold leading to the
identification of potent PfDHODH inhibitors with in vitro antimalarial activity and with good in
6
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 97
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
vivo pharmacokinetic (PK) properties supporting in vivo efficacy in the SCID mouse model of P.
falciparum malaria.
Results.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Chemistry and structure activity relationships (SAR) of the pyrrole series on DHODH and P.
falciparum 3D7
The initial hit compound 3 was used as a starting point for a hit-to-lead optimization
program to identify PfDHODH inhibitors that were differentiated from the current clinical
candidate 1 (Fig. 1). Notably, our goals were to identify compounds that improved on the known
liabilities of 1, including development of compounds with equivalent activity on P. falciparum and
P. vivax parasites. We also sought to identify a series that maintained selectivity against human
DHODH but also showed better selectivity versus the common toxicology species (e.g. rat, mouse
and dog). While 1 is not a significant inhibitor of human DHODH, inhibition of the mouse and rat
2
2
enzyme complicated its preclinical development. Finally, we also had the objective of identifying
compounds with improved solubility to simplify formulation development. The initial hit 3 was a
sub-micromolar inhibitor of both PfDHODH and P. falciparum 3D7 (Pf3D7) parasites in culture,
it showed complete selectivity against human DHODH, and was more potent on P. vivax DHODH
(PvDHODH) than PfDHODH (Table 1). These data suggested that the pyrrole series might meet
our objective of improved P. vivax activity.
We systematically explored the SAR of the series by modifying each of the moieties
attached to the pyrrole ring, including the aryl group attached at the 4 position, the two methyl
groups attached at positions 3 and 5, and the ester group at position 2 (Table 1). The reported
pyrrole derivatives 3-74 were synthesized as shown in schemes 1-6 and supplementary schemes
7
ACS Paragon Plus Environment

Page 9 of 97
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
-2. We utilized various protocols for the preparation of the pyrrole motif with different
substitutions. Knorr condensation of α-amino β-keto ester with aryl substituted 2,4-pentanedione
gave the aryl substituted dimethyl pyrroles (Scheme 1). Mono methyl substituted pyrroles were
prepared by Friedel-Crafts acylation on the corresponding pyrrole followed by step wise reduction
of the carbonyl group resulting in pyrrole ester intermediates (Scheme 2). 3-methoxymethyl and
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
-hydroxymethyl substituted pyrroles were obtained via silver-catalyzed isocyanide-alkyne
3
7
cycloaddition (Scheme 3). The 5-CF substituted pyrrole ester precursor 108 was prepared from
3
3
8
the pentafluorophenyl ester 106, and subsequent iodination and trifluoromethylation afforded the
precursor 108 leading to derivative 54 (Scheme 4). Friedel-Crafts arylation with some of the
substituted aryl acid chlorides on the pyrrole proved difficult and gave poor yields. These phenyl
substitutions were therefore prepared in a similar manner via the α,β alkynyl ketone as described
3
9
in Scheme 3 starting from aryl aldehydes (Scheme 5). Similar reaction conditions were used to
synthesize other pyrrole scaffolds as shown in Scheme 6 and supplementary schemes from the
appropriate starting materials. Finally, hydrolysis of pyrrole esters obtained in the various
protocols/methods followed by coupling with a variety of amines yielded pyrrole amides. Our
initial analysis of the resultant compounds included determination of their inhibitory activity
versus PfDHODH and PvDHODH, their selectivity versus human DHODH, and their potency
against Pf3D7 parasites in whole cell assays (Tables 1-2, 4-7). Compounds that showed good
potency and selectivity in these assays were then profiled further.
Optimization of the Aryl group. To first further validate the series before beginning a full
scale hit-to-lead program, a handful of additional ester analogs (4-11) were synthesized that
explored different benzyl groups at the 4 position of the pyrrole with varied para, ortho and meta
substituents (Scheme 1 and Tables 1 and 2). The data showed that para-substituted benzyl analogs
8
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 97
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
yielded the most potent activity on both Plasmodium DHODHs and on Pf3D7. Substitution of the
benzyl moiety with a para-CF benzyl (4) led to a 4-fold increase in potency against PfDHODH
3
and Pf3D7 whereas replacement with para-CF -pyridin-3-yl (9) led to a 15-25-fold improvement
3
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
in enzyme and parasite activity over 3, and also to better activity than para-OMe (6) or para-iPr
(
8) (Table 1). Meta-substituted benzyl rings were less active (5, 7), as was the para-CF -pyridin-
3
2
-yl analog (10). Methylation of the pyrrole nitrogen led to complete loss of activity (11). Both
and 9 retained good activity on PvDHODH while showing complete selectivity versus the human
4
enzyme. We next tested a panel of mammalian DHODH enzymes from common species used in
toxicologic studies (mouse, rat and dog), and again found that both 4 and 9 retained complete
selectivity, while neither compound inhibited any of these mammalian enzymes up to the top
concentration tested (100 µM) (Table 2). This contrasts with 1, which showed low to mid µM
activity on the mouse, rat and dog enzymes while not significantly inhibiting hDHODH. Thus,
these data confirmed that the pyrrole series could deliver both improved P. vivax DHODH activity
and reduced mammalian DHODH inhibitory activity, meeting a key objective for a backup series
for 1. Having met these objectives, we committed to a full scale hit-to-lead medicinal chemistry
program on this series.
Scheme 1
O
O
O
O
Me
Me
i
Me
Me
Ar
75
77a-h
Me
COOEt
Me
Ar
Ar
Me
(iv)
iii
+
R
Me
N
N
H
O
O
H
O
NOH
O
ii
3-10
EtO
Me
EtO
12-25
O
Me
(v)
Me
76
Ar
78
COOEt
Me
N
Me
11
Reagents and conditions: (i) ArCH2Br, Acetone, K2CO3, 60 oC, 12h (ii) NaNO2 , AcOH, 0 oC- rt, 3h, (iii) Zn, AcOH, 80
oC, 6-8 h (iv) a) NaOH, EtOH, 80 oC, 2-4h b) EDC,TEA, DMF, RT, 4-6h (v) 4 -> 11, MeI, KOtBu, 18crown6, benzene, 6h
9
ACS Paragon Plus Environment

Page 11 of 97
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 1:
Me
Ar
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
3
COOEt
2
Me 5
N
H
Cmpd
Cmpd ID
Ar
IC50 (µM)
EC50 (µM)
Pf3D7 cells
0.0060 ± 0.0019
(9)
a
PfDHODH
.030 ±0.014
12)
PvDHODH
0.072±0.028
(11)
hDHODH
>100 (4)
0
1
2
3
4
DSM265
NA
NA
(
Genz-667348 *
0.022
0.042
>30
>100 (2)
>100
0.007
0
.27 ±0.092
0.045
(0.04 – 0.050)
0.014
DSM43
OEt
0.23, 0.25
0.061, 0.071
(
3)
.10±0.033
3)
0
DSM483
CF3
OEt
(
(0.013-0.016)
0.83
(0.74-0.92)
0.021
(0.018-0.025)
0.53
(0.28-0.78)
5
6
DSM484
DSM485
>100
>100
0
.055
0
.52
0.59
(0.47-0.75)
OMe
(0.036-0.084)
(0.46-0.60)
OMe
2.3
(2.0-2.7)
0.47±0.065
(3)
7
DSM486
>100
10 (6.7 – 15)
OMe
Me
0
.57
(0.46-0.71)
.018
(0.016-
.020)
.3
0.040
(0.03-0.054)
0.0094
(0.0077-
0.011)
0.47
(0.2-1.1)
8
9
DSM487
DSM491
>100
>100
Me
0
N
0.0092
(0.0076-0.011)
CF3
0
N
0
0.06
(0.05-0.072)
0.10
(0.090-0.11)
10
DSM520
CF3
CF3
>100
nd
(0.25-0.37)
1
1
DSM490**
>100
16 (11-21)
nd
Pf, P. falciparum; Pv, P. vivax; h, human. For each experiment, triplicate data were collected at each inhibitor
concentration. A 3-fold dilution series was used to determine IC50’s and either a 3-fold or 4-fold dilution series
was used to determine the EC50’s. Values in parenthesis represent the 95% confidence interval. For key active
compounds, multiple independent experiments were performed and for these compounds data represent mean ±
standard deviation with the number of independent experiments in parenthesis. If only 2 biological replicates were
2
8
collected than both values are shown. * data taken from . **Compound has an N-methyl as shown in Scheme
1
.
1
0
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 97
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table 2. Inhibitory activity on mammalian DHODHs
Cmpd
hDHODH
mDHODH
0.11, 0.14
2.1±0.62 (6)
>100 (1)
>100 (1)
>100 (1)
>100 (2)
>100 (1)
nd
rDHODH
0.017±0.0046 (3)
4.4±2.7 (8)
>100 (1)
dDHODH
0.20, 0.27
14 ± 6.7 (5)
>100 (1)
>100 (1)
>100 (1)
>100 (2)
>100 (1)
nd
Teriflunomide* 0.30±0.052 (6)
1
4
6
9
>50 (10)
>100 (1)
>100 (1)
>100 (1)
>100 (2)
>100 (3)
>100 (3)
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
>100 (1)
>100 (1)
3
7
5
0
>100 (2)
4
6
>100 (1)
nd
h, human; m, mouse; r, rat; d, dog. Number of independent experiments in parenthesis. nd, not
determined. *Human DHODH inhibitor teriflunomide ((Z)-2-cyano-3-hydroxy-N-(4-
(trifluoromethyl)phenyl)but-2-enamide) is an approved drug for the treatment of rheumatoid
4
0
arthritis and multiple sclerosis, data similar to previous reports .
X-ray structure of 4 bound to PfDHODH. To allow incorporation of a structure-based
approach to optimize for good binding to PfDHODH, we solved the X-ray structure of the enzyme
bound to 4 (Fig. 1) to a resolution of 1.78 A with an Rfree and Rwork of 0.194 and 0.151 respectively
(Fig. 2 and Table 3). 4 bound adjacent to the flavin mononucleotide (FMN) cofactor in a binding
pocket that overlaps the 1-binding pocket in the region between R265 and H185 (Fig. 2A), but
which differs significantly in the position of the respective substituted phenyl rings (Fig. 2B). The
o
different binding modes arise from the differing conformation of F188, which is rotated 180 to
the “up” position in the 4 structure compared to the “down” position observed when bound to 1.
2
2
The 4 binding configuration is similar to that observed for the thiophene-based inhibitor 2 28
(Fig. 1 and 2C). The pyrrole nitrogen of 4 forms a hydrogen bond with H185 (replacing the
interaction with the bridging NH in 1), while the carbonyl forms an interaction with R265
replacing the interaction with the pyridine nitrogen in 1). The ethyl ester linkage of 4 is bound in
(
1
1
ACS Paragon Plus Environment

Page 13 of 97
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
a hydrophobic pocket near the FMN cofactor formed by V532, I272, I263, E182 and G181,
overlapping the binding site of the C2 -CF CH of 1. The benzyl-CF of 4 forms interactions that
2
3
3
are completely distinct from the aniline-SF of 1, as noted above. The 4 benzyl-binding pocket sits
5
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
between the N-terminal  helices 1 and 2. These helices are composed of mostly hydrophobic
amino acids that are thought to pack up against the mitochondrial membrane. The benzyl-CF of
3
4 instead makes close contacts (<5 Å) with L187, L191, C175, F171, Y168, F188 and M536.
Fig. 2. X-ray structures of PfDHODH bound to 4 and 60. A. 4 bound to PfDHODH showing amino acid
residues within 4Å of the inhibitor. Dashed lines indicate H-bond interactions. Protein is displayed in pink
and 4 is displayed in tan as sticks within shaded spheres. B. Comparison of the binding mode of 4
2
2
(
colored as in A) with 1 (purple) (pdb 4RX0). C. Comparison of the binding mode of 4 (colored as in A)
with 2 (light purple) (pdb 3O8A).28 D. Overlay of the binding modes of 4 (colored as in A) and 60
(green) showing residues within 4Å of the inhibitors. The coordinates for the structure of PfDHODH
bound to 4 (pdb 6VTY) and 60 (pdb 6VTN) have been submitted to the Protein Data Bank.
1
2
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 97
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table 3. PfDHODH co-inhibitor X-ray diffraction data collection and refinement statistics
Data collection
Crystal
PfDHODH∆384-413- 4
PfDHODH∆384-413- 60
P64
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Space group
P2 2 2
1 1
1
Wavelength (Å)
Resolution (Å)
0.97924
0.97926
42.03 – 1.78 (1.81-1.78)
161,464 (8,016)
9.8 (9.0)
42.38 – 2.25 (2.29-2.25)
26,640 (1,344)
11.3 (10.2)
100.0 (100.0)
5.1 (100)
Unique reflections
Multiplicity
Data completeness (%)
Rmerge (%)a
100.0 (100.0)
8.3 (100)
Rpim (%)b
3.0 (65.6)
1.9 (71.9)
I/σ(I)
26.2 (1.1)
37.9 (1.0)
Wilson B-value (Å2)
Refinement
19.0
29.4
Resolution range (Å)
No. of reflections Rwork/Rfree
Data completeness (%)
No. of atoms/water
Rwork/Rfree(%)
42.02 – 1.78 (1.82-1.78)
148,953/2,002 (4,308/57)
92.2 (38.3)
42.38 – 2.25 (2.33-2.25)
21,071/1,482 (453/31)
79.2 (18.9)
12,086/897
2,958/72
15.14/19.39 (23.85/30.16)
0.020
20.24/21.87 (23.72/39.70)
0.004
R.m.s.d. bond length (Å)
R.m.s.d. bond angle (°)
Mean B (protein/ligand, Å2)
Ramachandran plot (%)
1.497
0.609
28.9/20.4
45.4/23.3
(
favored/additional)c
97.96/2.04
98.12/1.82
aRmerge = 100   |I — I |/  I , where the outer sum (h) is over the unique reflections and the
h
i
h,i
h
h
i
h,i
inner sum (i) is over the set of independent observations of each unique reflection.
b
1/2
Rpim = 100   [1/(n - 1)] |I — I |/  I , where n is the number of observations of
h
i
h
h,i
h
h
i
h,i
h
reflections h.
cAs defined by the validation suite MolProbity 41
Replacement of ethyl ester with amides. The ester linkage in compounds 3-11 is
metabolically unstable and thus was replaced to advance the series towards in vivo activity. Based
1
3
ACS Paragon Plus Environment

Page 15 of 97
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
on the small size of the ethyl ester pocket (Fig. 2), as well as by analogy to previously identified
2
1
moieties that bound well to this site in either the triazolopyrimidine (e.g. 1) or in the thiophene
2
8
series (e.g. 3) , a library of small cyclic and linear chain amides (12-25) were synthesized to
replace the ester as described in Scheme 1 and Table 4. These modifications were made in the
context of 4-CF -benzyl, 4-CF -3-pyridinyl and 4-CF -2-pyridinyl groups at C4 of the pyrrole ring.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
3
3
The acid 12 was inactive, but several amide replacements were well tolerated. Within a series,
PfDHODH inhibitory activity was best for the smaller substituents with ethyl amines 13 and 16
showing better activity than propyl amines 14 and 17 and with activity dropping off significantly
for butyl amine 15. Cyclopropyl amine 18 showed the best potency of the tested derivatives with
cyclobutyl amine 19 a close second. Azetidines 20 and 24 were tolerated with a 5-fold reduction
in activity compared to the cyclopropyl amine 18, but dimethyl 22 or diethyl 21 amines showed
poor to no activity. Activity was best for 4-CF -3-pyridinyl, with the order of preference 4-CF -
3
3
3-pyridinyl 16-19 > 4-CF -2-pyridinyl 23-25> 4-CF -benzyl 13-14. Activity against PvDHODH
3
3
and against Pf3D7 parasites paralleled the activity observed for PfDHODH, with the best
compounds 16 and 18 showing sub-50 nM activity on all three (Table 4). Complete species
selectivity was also maintained in the amide series for all tested compounds, with no observed
inhibition of mammalian enzymes (Table 4).
Table 4: Replacement of the C2 pyrrole ester with amides
Ar
Me
R
Me
N
H
O
Cmpd
Cmpd ID
Ar
R
IC50 (µM)
EC50 (µM)
a
PfDHODH
PvDHODH
hDHODH
Pf3D7 cells
1
2
3
DSM489
DSM496
CF3
CF3
OH
>100
>100
0.031
nd
>10
N
1
0.14
>100
0.051
H
1
4
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 97
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(
(
0.12-0.17)
.3
0.27-0.34)
(0.023-0.041)
0.059
(0.05-0.069)
(0.014-0.11)
0.11
(0.059-0.22)
0
CF3
CF3
N
H
>100
>100
>100
>100
>100
>100
nd
14
DSM495
DSM494
DSM497
DSM498
DSM499
DSM500
DSM508
DSM509
DSM517
DSM519
DSM521
DSM522
0.71
N
H
15
16
17
9.9±1.7 (3)
1.8, 0.54
(0.57-0.90)
N
N
N
N
N
N
N
0.054
(0.049-0.06)
0.072
(0.066-0.078) (0.034-0.044)
0.050 ±0.013
3)
0.036
(0.03-0.043)
0.038
0.033
(0.018-0.059)
0.047
(0.014-0.16)
0.036±0.0046
(3)
N
H
CF3
CF3
CF3
CF3
CF3
CF3
CF3
CF3
CF3
CF3
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
N
H
H
0.022
(0.019-0.026)
0.036
(0.028-0.045)
18
N
(
H
0.066
19
20
21
N
0.072, 0.042
0.14±0.035
(0.057-0.077)
0.13 ± 0.023
(3)
N
nd
nd
nd
nd
(
3)
N
13 (11 – 15)
>30
nd
>10
>10
22
N
nd
N
N
N
0.44
(0.38-0.49)
1.8
0.34
(0.31-0.38)
1.0
H
23
24
25
nd
N
0.6
(0.52-0.7)
0.22
(0.18-0.27)
N
>100
>100
(
1.5-2.0)
(0.99-1.1)
H
0.41
0.26-0.65)
N
0.47, 0.68
(
See Table 1 footnote. nd, not determined.
Replacement and modification of the pyrrole ring methyl substituents. The importance
of the methyl substituents on the pyrrole ring was next assessed by either removal or by
replacement with alcohols, methoxy ether, or CF and these variations were made within the
3
context of a wider range of amides compared to those in Table 4. Pyrrole derivatives (26-55) were
synthesized by Friedel-Crafts acetylation of the corresponding pyrrole with aryl acid chloride in
the presence of Lewis acid (AlCl ) or via the alkynyl ketone (Schemes 2-4, Supporting Information
3
Scheme S1 and Table 5). Removal of both pyrrole methyl groups 26 led to a 50-fold drop in
potency against the Plasmodium enzymes and Pf3D7 when compared to the matched cyclopropyl
compound 18 containing both methyls (Tables 4 and 5). Methyl at C5 was detrimental to activity,
1
5
ACS Paragon Plus Environment

Page 17 of 97
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
leading to a further 4-fold drop in PfDHODH activity relative to the compound lacking both
methyls (27 vs 26). The best activity was observed for compounds containing methyl at C3 (37-
41, 43-45, 47, 49, 51) in the absence of the C5 methyl, with those linked to cyclopropyl 37 or
cyclobutyl amine 51 showing the highest potency on Plasmodium DHODH and Pf3D7 (Table 5).
Amides linked to carbon rings with >4 atoms showed poor to no activity (28, 29, 32, 33, 34, 42)
regardless of the position of the pyrrole methyl. Branched chain alkyl groups including dimethyl
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
4
5
0, tert-butyl 41, -CH CF 39 and azetidine 38, including rings with fluorine substitutions (45 and
2 3
7) all provided sub-100 nM activity, with 45 showing similar activity to the cyclobutyl analog
1. Fluorination of cyclobutyl amine led to reduced activity versus 51 with the difluoro analog 48
showing less activity than the monofluoro 49 analog. Substitution of the C3 methyl with an alcohol
3 or ether 52 led to a 200-fold and 20-fold reduction in potency, respectively, against PfDHODH
5
relative to 37. Within the context of Me at C3, replacement of the C5 methyl with CF 54 reduced
3
activity 80-fold, as did analogs of 27 and 37 with pyrimidinyl-CF (55 and 56) capping the pyrrole
3
4
position. A strong correlation between potency on PfDHODH and Pf3D7 was observed for this
set, and equivalent to better potency was maintained for these compounds on PvDHODH. No
inhibition of the human enzyme was observed for any of the tested compounds.
1
6
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 97
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Scheme 2
O
OH
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
X
Ar
Ar
Ar
X
X
X
(i)
(ii)
(iii)
COOEt
COOEt
Y
Y
Y
N
H
N
COOEt
N
N
COOEt
Y
H
H
H
8
2
87
92
79-81
8
8
3a,b
4a,b
88a,b
89a,b
90, 91
93a,b
94a,b
95,67
85,86
X=H, Y=H: 26
X=H, Y=Me: 27-36
X=Me, Y=H: 37-51
(v)
Ar
(iv)
Ar
X
X
Ar =
Ar =
CF3
N
R
Y
Y
N
N
COOH
H
H
O
N
9
6
7a,b
98a,b
9
X = H, Y=Me: 55
X = Me, Y =H: 56
26
9
27-51
CF3
NC
N
5
5,56,68
9
Ar =
X=Me, Y=H
N
N
68
67
Reagents and Conditions: (i) AlCl , CH Cl , 0 °C-RT, 16 h (ii) NaBH , EtOH, 1 h, RT (iii) TFA, triethylsilane,
3
2
2
4
CH Cl , 85 °C, 1 h (iv) NaOH, H O, MeOH, 80 °C, 2 h (vi) Cyclopropyl amine, HATU, TEA, CH Cl , RT, 4 h
2
2
2
2
2
Scheme 3
O
OH
O
O
O
CF3
(iv)
HO
(i)
(ii)
CF3
(iii)
N
O
O
N
CF3
N
N
H
COOEt
N
COOEt
O
H
100
101
102
103
O
O
CF3
CF3
(
vii)
CF3
(v)
CF3
(vi)
N
N
H
N
H
N
N
N
COOEt
COOH
N
H
N
N
N
H
H
H
O
O
104
105
52
53
Reagents and Conditions: (i) N-methoxy-N-methyl-6-(trifluoromethyl)nicotinamide, n-BuLi, THF, -78 °C, 2 h (ii) Ethyl isocyanoacetate, Ag2CO3, NMP, 85 °C, 2 h
iii) NaBH4, EtOH, 0 °C-RT, 1 h (iv) TFA, triethylsilane, CH2Cl2, 85 °C, 1 h (v) NaOH, EtOH:H2O, 80 °C, 2 h (vi) Cyclopropyl amine, HATU, Et3N, CH2Cl2, 4-8 h.
(vii) BBr3, CH2Cl2, 0 °C, 1h.
(
1
7
ACS Paragon Plus Environment

Page 19 of 97
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Scheme 4
CF3
N
(i)
CF3
CF3
(
ii)
N
PFP
O
N
PFP
O
OH
I
N
N
H
N
H
H
O
O
O
106
107
98b
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
CF3
CF3
(iii)
PFP
O
(
iv)
N
N
H
N
CF3
CF3
N
H
N
H
O
O
108
54
Reagents and Conditions: (i) Pentafluorophenol (PFP-OH), DCC, THF, RT, 1 h (ii) NIS, DMF, RT, 1 h (iii) Methyl 2,2-difluoro-2-(fluorosulfonyl)acetate,
CuI, NMP, MW, 120 °C, 2 h (iv) THF, RT, 30 min
Table 5: Replacement or modification of the pyrrole ring methyl groups
N
N
N
F3C
F3C
X
X
R
R
Y
Y
N
H
N
H
O
O
2
6-54
55,56
Cmp
d
Cmpd ID
X
Y
R
IC50 (µM)
EC50 (µM)
PfDHODH
PvDHODH
1.2
(0.94-1.5)
hDHODH
Pf3D7 cellsa
H
N
2.5
1.9-3.2)
26
27
28
29
DSM501
DSM503
DSM506
DSM507
DSM511
DSM512
H
H
H
H
H
H
H
>100
>100
nd
1.7, 0.50
(
H
N
2.3
Me
Me
Me
Me
Me
10 ± 0.96 (3)
>100 (2)
3.6 (3.1-4.2)
>10
(1.5-3.6)
N
nd
nd
nd
nd
N
>100
2.8
2.1-3.8)
5.7
(4.4-7.4)
>100
nd
>20
H
1.9
(1.2-3.0)
3.5
30
N
nd
(
H
31
N
nd
(2.5 – 4.9)
H
N
O
32
33
34
DSM513
DSM514
DSM515
H
H
H
Me
Me
Me
nd
nd
nd
nd
nd
nd
>10
>10
>10
H
N
NH
>100
>100
H
N
N
H
3
5
6
DSM518
DSM524
H
H
Me
Me
N
NH
>100
>100
nd
nd
nd
nd
>20
>20
H
N
3
N
1
8
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 97
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
H
N
0.020 ± 0.0054
0.014 ±
0.0046 (3)
0.21
0.014 ± 0.0055
37
DSM502
DSM525
DSM528
DSM529
DSM530
DSM531
DSM532
DSM536
DSM538
DSM539
DSM546
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
H
H
H
H
H
H
H
H
H
H
H
>100 (3)
>100
>100
>100
>100
nd
(
7)
(6)
0.082
(0.073-0.093)
0.17
38
39
40
N
(
0.15-0.2)
(0.17-0.26)
0.21
(0.16-0.27)
0.080
0.051
(0.043-0.061)
0.018
(0.015-0.021)
0.064
(0.051-0.082)
N
CF3
0.044, 0.059
H
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
H
N
0.013
(0.011-0.016)
0.051
(
0.062-0.1)
H
0.11
41
N
(
0.084-0.15)
.8
(0.029-0.090)
9
4.8
42
43
44
N
nd
(
7.4-12.8)
(3.5 – 6.5)
0.065 ± 0.029
(4)
0.0099 ±0.0020
(3)
0.098
0.041
(0.034-0.049)
0.0075±
0.0031 (3)
0.018
>100
>100 (3)
>100 (3)
nd
(0.077-0.12)
.018±0.0021
3)
0
(
F
F
0.033±0.0079
(3)
0.029 ± 0.0079
(4)
45
N
±0.0025 (3)
H
4
6
7
>100
nd
>100
N SO Me
2
0.12
(0.11-0.14)
0.2
(0.17-0.25)
0.051
(0.036-0.072)
4
N
F
>100
H
F
3.6
1.3
(0.86-2.0)
48
DSM552
DSM553
DSM559
Me
Me
Me
Me
H
H
H
N
nd
nd
nd
nd
nd
(
4.0-5.4)
F
H
N
0.32
0.17
(0.16-0.19)
49
F
(0.40 – 0.90)
.3
6.3-12)
H
9
N
50
nd
3.9 (3.3 – 6.9)
(
F3C
H
N
0.019 ±
0.027
(0.023-0.032)
5
1
2
DSM560
DSM534
DSM535
DSM526
DSM527
DSM533
H
H
0.044 ±0.19 (3)
0.45
>100 (3)
>100
nd
0.0058 (3)
H
N
0.07
O
5
0.24±0.046 (3)
1.2±0.64 (3)
(
0.34-0.6)
(0.05-0.096)
nd
H
N
4.3
OH
53
H
(
2.5-7.4)
1.6
H
N
0.77
(0.60 - 1.0)
54
55
56
Me
CF
3
nd
nd
nd
(1.4-1.9)
>100 (2)
0.88
H
N
H
Me
H
nd
>10
H
N
0.21
(0.17-0.26)
Me
>100
0.39±0.085 (3)
(
0.78-1.0)
See Table 1 footnote. nd, not determined
1
9
ACS Paragon Plus Environment

Page 21 of 97
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Modification of the 4-position benzylic substitutions. To further explore the potential for
the benzyl group to yield potent compounds we synthesized additional analogs (57 – 68) in the
context of either cyclopropyl amide or ethyl ester (Scheme 5 and Table 6). 4-CF -benzyl (57, 58),
3
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
or this ring with a single ortho fluoro (59, 60) or ortho chloro (65, 66) all showed similar activity
to 4-CF -pyridinyl 37, with the cyclopropyl amide showing better activity than the ethyl ester.
3
There was a significant drop in activity upon adding a second ortho-fluoro to the ring (61, 62),
with this substitution leading to complete loss of activity for the ethyl ester. Addition of a single
ortho O-methyl to the 4-CF benzyl ring led to ~10-fold drop in potency (63, 64). Replacement of
3
the 4-CF with CN in the context of the pyridinyl ring 67 led to a potency drop of 30-fold, whereas
3
removal of the 4-CF led to complete loss in activity (68).
3
Comparison of the X-ray structure of PfDHODH bound to cyclopropyl amide 60 versus
ethylester 4. Having identified several potent pyrrole analogs containing amide replacements for
the original ester, we solved the X-ray structure of PfDHODH bound to cyclopropyl amide
derivative 60 (Fig. 1 and Table 6). The structure of PfDHODH bound to 60 was solved to 2.3 Å
in a space group of P6 with the cell dimension of a=b=84.8, c=138.0 and one molecule of
4
PfDHODH in the asymmetric unit binding modes (Table 3 and Fig. 2D). The structures of
PfDHODH bound to 60 superimposed with the 4-bound structure with an RMSD = 0.26 over 1199
atoms. The positions of 60 and 4 in the inhibitor binding site were similar, although 60 formed an
additional H-bond between the amide NH and H185 and as a consequence 60 was shifted closer
to H185 than 4 (Fig. 2D). H-bonds between the pyrrole NH and H185 and the carbonyl oxygen
and R265 were observed for both 60 and 4. Thus interestingly for 60, H185 is involved in a
bifurcated H-bonding interaction. Differences in the binding mode between the ester and the amide
2
0
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 97
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
may explain why benzyl modifications had differential impact depending on which was present
(Table 6).
Scheme 5:
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
O
Ar
OH
O
O
(i)
(ii)
(iii)
Ar
Ar
Ar
09-113
Ar = 109) 4-CF -ph
N
COOEt
H
1
114-118
1
19-123
124-128
3
1
1
1
1
10) 4-CF -2-F-ph
11) 2,5-di-F-4-Br-ph
(iv)
3
126
129 (2,5-di-F-4-CF3-ph)
12) 4-CF -2-OMe-ph
3
13) 4-CF -2-Cl-ph
3
OH
Ar
Ar
Ar
(vii)
(v)
(vi)
H
N
COOEt
N
N
COOEt
N
H
H
H
O
130-134
57,59,61,63,65,
58,60,62,64,66
Reagents and Conditions: (i) 1-Propynylmagnesium bromide, THF, 0 °C-RT, 2 h (ii) Dess Martin, CH Cl , RT, 2 h (iii)
2
2
Ethyl isocyanoacetate, Ag CO , NMP, 80 °C, 2 h (iv) methyl 2,2-difluoro-2-(fluorosulfonyl)acetate, KI, CuI, NMP, 120
2
3
°
C, 16 h (v) NaBH , EtOH, 0 °C-RT, 1 h (vi) TFA, triethylsilane, CH Cl , 85 °C, 1 h (vii) (a) NaOH, EtOH:H O, 80 °C, 2
4 2 2 2
h (b) Amine, HATU, Et N, CH Cl , 4-8 h.
3
2
2
Table 6: Modification of the 4-position benzylic substitutions
Ar
Me
R
N
H
O
Cmpd
Cmpd ID
Ar
R
IC50 (µM)
PvDHODH
0.019
EC50 (µM)
Pf3D7 cells
a
PfDHODH
.13
0.11-0.15)
0.048 ±0.019 0.0067 ±0.017
hDHODH
>100
0
0.060
CF3
CF3
57
DSM561
DSM562
DSM556
DSM557
O
(
(0.014-0.027)
(0.05-0.07)
0.018 ±0.0038
(3)
H
>100 (3)
>100
5
5
6
8
9
0
N
(
3)
(3)
0.024
(0.018-0.03)
0.015 ± 0.0080
(3)
F
F
0.035
(0.029-0.043)
0.044 ±0.19
0.028
CF3
CF3
O
(0.024-0.031)
0.016 ±0.0056
(5)
H
N
>100 (3)
(
3)
2
1
ACS Paragon Plus Environment

Page 23 of 97
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
F
>
30
nd
nd
nd
nd
nd
>10
CF3
CF3
6
6
6
6
6
1
2
3
4
5
DSM563
DSM564
DSM566
DSM567
DSM570
DSM571
DSM584
DSM585
O
F
F
H
1.2
(1.1-1.4)
0.42
(0.26-0.67)
0.81
(0.4-1.6)
0.26
N
F
MeO
1
.1
nd
CF3
O
(0.66-1.8)
0.11
0.093-0.14)
MeO
Cl
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
0.092
(0.064-0.13)
0.059
(0.049-0.072)
0.014 ±0.0050
(3)
>100
>100
>100 (3)
nd
H
N
CF3
CF3
CF3
(
(0.18-0.39)
0.042
0.029
O
(
0.036-0.048)
0.017
0.0061 (3)
(0.025-0.033)
0.020 ± 0.00086
(5)
Cl
H
N
66
±
0.63
nd
0.20± 0.017 (3)
CN
6
6
7
8
O
N
(0.51-0.77)
H
N
>100
nd
nd
>10
N
See Table 1 footnote. nd, not determined.
Replacement or modification of the bridging methyl. Finally, the impact of replacing or
modifying the bridging methyl was explored (Schemes 6 and S2; Table 7). Oxidation to a ketone
(69 and 71) led to almost complete loss of activity, whereas replacement with oxygen (74) led to
a ~40-fold drop in potency against PfDHODH. In contrast, modification of the bridging carbon
with hydroxyl (73) caused only a 4-fold drop in potency, demonstrating that hydroxyl modification
of this position is tolerated. Additionally since 73 is racemic, it is likely that the active enantiomer
will be 2-fold more potent than what is observed for the racemic mixture. Attempts to replace the
bridging carbon with NH were made but this compound was not synthetically feasible in our hands.
SAR summary. We identified a number of compounds (37, 45, 51, 60 and 66) with sub-
3
0 nM potency on PfDHODH and Pf3D7 parasites that were of interest for potential follow up.
Overall, we observed a strong correlation between PfDHODH and activity on Pf3D7 throughout
the series and across the full dose response range of measured values (Fig. 3). Good activity on
PvDHODH was also observed, with a tendency for greater potency on PvDHODH than
PfDHODH. No inhibition of the mammalian enzymes was observed for compounds in the series.
2
2
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 97
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Fig. 3. SAR correlation between inhibitory activity on PfDHODH and Pf3D7 parasites. Data are
taken from Tables 1, 3-6. For compounds where either the PfDHODH IC or Pf3D7 EC was
5
0
50
greater than the top concentration tested in the study (e.g. > 100 M) were not included.
Scheme 6:
O
O
OH
O
O
O
(i)
NOH
O
(ii)
O
EtO
Me
+
N
N
+
Cl
CF3
CF3
135
136
137
78
O
O
CF3
CF3
H
N
N
(iii)
N
COOEt
N
H
N
H
O
138
74
o
Reagents and conditions: (i) Cs CO Acetone (ii) Zn, AcOH, reflux, 6h (iii) a) NaOH, EtOH-H O, 80 C; b) Amine,
2
3,
2
HATU, Et N, CH Cl , 4h
3
2
2
2
3
ACS Paragon Plus Environment

Page 25 of 97
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 7: Bridge modifications
Cmpd
CmpdID
DSM504
Structure
IC50 (µM)
EC50 (µM)
a
PfDHODH
21
(17-25)
PvDHODH hDHODH Pf3D7 cells
O
69
nd
nd
nd
6.5
(6.0-7.0)
F3C
Me
N
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Me
COOEt
N
H
O
7
1
3
DSM505
DSM558
>100 (2)
nd
>10, >20
F3C
Me
N
H
N
Me
N
H
O
OH
7
0.074
(0.059-
0.092)
0.23
(0.18-0.29)
>100
0.044
(0.038-
0.052)
F3C
Me
N
N
H
N
N
H
O
74
DSM537
F3C
O
Me
0.8
(0.68-0.94)
0.16
(0.10-0.25)
>100
0.29
(0.24-0.36)
H
N
Me
N
H
O
See Table 1 footnote.
In vitro ADME. A subset of compounds was evaluated to assess their physicochemical
properties and to determine their propensity to be metabolized using in vitro human and mouse
liver microsome assays (Table 8). Compounds were typically selected based on having
demonstrated good potency in the PfDHODH and Pf3D7 assays, however some compounds were
chosen to explore the SAR around this subset of ADME properties. The Log D for compounds
7
.4
containing an amide was generally in good physicochemical space (range 1.7 to 4.1). Kinetic
solubility (phosphate buffered saline pH 6.5) ranged from poor to quite good, with compounds that
contained an azetidine attached to the carbonyl bond showing the best solubility (20, 38, 45 and
4
7). The azetidine contains a tertiary nitrogen that is unable to act as a H-bond donor, in contrast
to the amide secondary nitrogen observed in most of the other non-ester compounds. Thus, the
poorer solubility of the amides may potentially result from intramolecular H-bonding within the
crystal lattice. Within the amide series, compounds containing a cyclopropyl amide (27, 37, 55
2
4
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 97
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
and 71) were more soluble than those with other amide substituents, while replacement of the
pyridinyl-CF with a pyrimidinyl-CF improved solubility (55 vs 27). Compounds with benzyl-
3
3
CF were predictably less soluble (58 vs 37), and addition of an ortho-chloro led to further
3
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
reductions (66). Variation of the number of methyl groups on the pyrrole ring (0-2), had little
impact on solubility although those with a single methyl were improved over the others (27 and
3
7 vs 18 and 26).
Metabolic stability ranged from good to poor across the series, but for many compounds
species differences were also observed between human and mouse microsomes (Table 8).
Cyclopropyl amides with either para-CF -pyridine or para-CF -pyrimidine (18, 27, 37, 55, 56, 71,
3
3
73 and 74) showed the best stability with compounds containing pyrimidine performing best (55
and 56). The difluoroazetidine 38 showed excellent stability in human liver microsomes but was
much more rapidly metabolized in mouse microsomes, and this was likewise a problem for three
potentially interesting compounds with good potency 51, 60 and 66. This issue would likely make
development of these compounds difficult as it suggests that there will be poor plasma exposure
in the mouse efficacy model. Replacement or modification of the bridging carbon did not
significantly alter stability, e.g. hydroxylation (73 vs 37), oxidation to a carbonyl (71 vs 18), or
replacement with oxygen (74 vs 18). Not surprisingly esters (3 and 9) showed poor solubility and
were very rapidly metabolized.
The estimated extent of protein binding (using an albumin-based column and
chromatographic method) ranged from ~43% (74) to 98% (66) and the extent of binding strongly
correlated with Log D across the series.
2
5
ACS Paragon Plus Environment

Page 27 of 97
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 8: Physicochemical properties, in vitro metabolism.
b
Kinetic solubility
Compd Log DpH 7.4 range in pH 6.5
PBS (µg/mL)
CL (H/M)
(µL/min/mg
protein)
int
cPPBa (%
bound)
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
3
9
3.6
4.5
3.7
2.8
3.2
3
12.5-25
<1.6
97.4
nd
<7/8
c
c
188 /470
1.6-3.1
12.5-25
6.3-12.5
12.5-25
>50
93.6
77.1
84.5
82.8
77.2
66.7
72.6
81.5
87.5
76.2
65.9
81.9
74.9
cnd
83.5
44.5
nd
286/779
70/54
1
6
7
8
0
1
1
2
172/64
15/33
2.9
2.6
2.8
3.2
3.5
2.8
2.7
3.3
3.3
cnd
3.2
2.5
2.5
3.7
3.8
4.1
2.6
2.4
3.2
31/53
26
27
12.5-25
25-50
31/96
36/41
28
29
37
38
39
6.3-12.5
12.5-25
25-50
226/138
173/93
10/24
>100
<7/192
15/57
12.5-25
25-50
45
47
21/94
25-50
7/145
51
55
56
58
60
66
12.5-25
50-100
12.5-25
3.1-6.3
1.6-3.1
<1.6
29/204
<7/8
<7/nd
19/173
15/184
14/264
19/138
8/24
97.1
97
98.1
73.2
50.4
89.5
68
71
12.5-25
25-50
73
6.3-12.5
20/40
2
6
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 97
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
7
4
1.7
12.5-25
42.7
8/22
a
Binding estimated using a chromatographic method with an albumin-based column as
described.
21
b
Human (H) or mouse (M) liver microsomes
Measurable loss of 3 in control incubations without cofactor suggesting the contribution of
non-NADPH mediated metabolism to overall degradation rate.
cnd, could not determine; nd, not determined
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
c
Identification of 37 as a lead compound. Taking into consideration both the potency data
against Pf3D7 parasites and key parasite enzymes (PfDHODH and PvDHODH), and the in vitro
pharmacology, we selected 37 for additional biological, ADME and pharmacokinetic profiling. It
was the only compound with sufficiently strong properties across these criteria to suggest it could
meet development criteria.
Additional Parasitology Profiling of 37. As 37 showed good potency against both Pf3D7
and the Plasmodium enzymes (Table 5), we undertook additional parasitology studies to further
define its profile. Good activity was also observed for P. falciparum Dd2, which is an isolate that
is resistant to multiple clinically used drugs (EC 0.016 vs 0.014 µM on Dd2 vs 3D7, respectively)
5
0
(
Tables 5 and 9), thus demonstrating that like other DHODH inhibitors (e.g. 1), 37 is not cross
resistant with 8-aminoquinolines or DHFR inhibitors. To provide further demonstration that cell
killing occurs through DHODH inhibition we tested 37 against a P. falciparum strain that has been
transformed with yeast DHODH and is resistant to both P. falciparum DHODH and cytochrome
bc1 inhibitors. 42, 43 Inclusion of proguanil in this study further allowed us to distinguish between
bc1 and DHODH inhibitors, as resistance of the yeast DHODH strain to bc1 inhibitors is rescued
in the presence of proguanil. As expected for a DHODH inhibitor, and like 1, the yeast DHODH
2
7
ACS Paragon Plus Environment

Page 29 of 97
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
strain was resistant to 37, and this resistance was not rescued by proguanil consistent with DHODH
inhibition as the mechanism of parasite killing (Table 9).
To assess cross-resistance with 1, we tested 37 against a panel of 1 resistant cell lines that
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
we previously described22, 34 (Table 9). The cell line containing a DHODH gene amplification
(1D3) showed a similar 5-fold resistance to both 1 and 37, consistent with their common
mechanism of action. Cell lines harboring DHODH point mutations caused a 15-40-fold shift in
EC for 1, but the effects on 37 were less (range 0.8-6-fold), with the L531F mutation actually
5
0
leading to slightly improved binding of 37. The additional data set collected for 1 in this study
3
4
agreed closely with the previously published results , data are included in Table 9 for comparison.
The largest effect (6-fold) for 37 was caused by the C276F mutation, which is the mutation that
23
was observed clinically. However, this mutation caused a significantly smaller shift for 37 than
for 1. Overall, the greatly reduced effects of the 1 selected mutations on 37 efficacy are consistent
with the different binding modes of the triazolopyrimidine and pyrrole series compounds (Fig. 2).
Finally, 37 did not show any evidence of cytotoxicity when tested against both human
HepG2 cells and mouse L1210 cells (CC > 50 µM) (Table 9). Preliminary safety pharmacology
5
0
studies showed that 37 did not inhibit the hERG channel, nor a set of additional ion channels that
were tested (Table 9).
Table 9. Activity of 37 on drug resistant P. falciparum and mammalian cell lines
P.
Resistance
Profile
Selection EC 37 (µM)
Fold
EC 1 (µM) Fold-
5
0
50
falciparum
Strain
Dd2
Agent
change
change
CQ, Py, SD,
MF, CG
na
0.016±0.0070(6)
NA
0.0048,
NA
0.0046#
(parental)
2
8
ACS Paragon Plus Environment