3,3′-Disubstituted 5,5′-Bi(1,2,4-triazine) Derivatives with Potent in Vitro and in Vivo Antimalarial Activity

Article abstract of DOI:10.1021/acs.jmedchem.8b01799

A series of 3,3′-disubstituted 5,5′-bi(1,2,4-triazine) derivatives was synthesized and screened against the erythrocytic stage of Plasmodium falciparum 3D7 line. The most potent dimer, 6k, with an IC₅₀ (50% inhibitory concentration) of 0.008 μM, had high in vitro potency against P. falciparum lines resistant to chloroquine (W2, IC₅₀ = 0.0047 ± 0.0011 μM) and artemisinin (MRA1240, IC₅₀ = 0.0086 ± 0.0010 μM). Excellent ex vivo potency of 6k was shown against clinical field isolates of both P. falciparum (IC₅₀ = 0.022-0.034 μM) and Plasmodium vivax (IC₅₀ = 0.0093-0.031 μM) from the blood of outpatients with uncomplicated malaria. Despite 6k being cleared relatively rapidly in mice, it suppressed parasitemia in the Peters 4-day test, with a mean ED₅₀ value (50% effective dose) of 1.47 mg kg^-1 day^-1 following oral administration. The disubstituted triazine dimer 6k represents a new class of orally available antimalarial compounds of considerable interest for further development.

Full text of DOI:10.1021/acs.jmedchem.8b01799

Subscriber access provided by Iowa State University | Library
Article
3,3’-Disubstituted 5,5’-Bi(1,2,4-triazine) derivatives
with Potent in vitro and in vivo Antimalarial Activity
Lian Xue, Da-Hua Shi, Jitendra R. Harjani, Fei Huang, Julia Beveridge, Tamir Dingjan, Kung Ban,
Sarah Diab, Sandra Duffy, Leonardo Lucantoni, Sabine Fletcher, Francis C. K. Chiu, Scott Alan
Blundell, Katherine Ellis, Stuart A. Ralph, Grennady Wirjanata, Silvia Teguh, Rintis Noviyanti, Marina
Chavchich, Darren John Creek, Ric Price, Jutta Marfurt, Susan A. Charman, Matthew Cuellar, Jessica
Strasser, Jayme Dahlin, Michael A. Walters, Mike Edstein, Vicky M. Avery, and Jonathan B. Baell
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.8b01799 • Publication Date (Web): 04 Feb 2019
Downloaded from http://pubs.acs.org on February 6, 2019
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 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
Edstein, Mike; Australian Defence Force Malaria and Infectious Disease
Institute
Avery, Vicky; Griffith University, Eskitis Institute
Baell, Jonathan; Monash Institute of Pharmaceutical Sciences, Medicinal
Chemistry
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

Journal of Medicinal Chemistry
Page 2 of 47
1
2
3
4
5
3,3’-Disubstituted 5,5’-Bi(1,2,4-triazine) derivatives
6
7
8
9
with Potent in vitro and in vivo Antimalarial Activity
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
Lian Xue,†∆ Da-Hua Shi,‡ ∆ Jitendra R. Harjani,‡∆ Fei Huang,† Julia G.
Beveridge,‡ Tamir Dingjan,‡ Kung Ban,‡ Sarah Diab, ‡ Sandra Duffy, Leonardo
Lucantoni, Sabine Fletcher, Francis C.K. Chiu,§ Scott Blundell,§ Katherine
Ellis,# Stuart A. Ralph,€ Grennady Wirjanata,¥ Silvia Teguh,‡ Rintis Noviyanti,&
£
#
¥,
¥
µ
Marina Chavchich, Darren Creek, Ric N. Price, Jutta Marfurt, Susan A.
Charman,§ Matthew E. Cuellar,∫ Jessica M. Strasser,∫ Jayme L. Dahlin,∞ Michael
A. Walters, ∫ Michael D. Edstein,£ Vicky M. Avery and Jonathan B. Baell†,‡,*
†School of Pharmaceutical Sciences, Nanjing Tech University, No. 30 South Puzhu Road,
Nanjing 211816, People’s Republic of China.
‡Medicinal Chemistry, Monash Institute of Pharmaceutical Sciences, Monash University
(Parkville Campus), 381 Royal Parade, Parkville, Victoria 3052, Australia
Discovery Biology, Griffith Institute for Drug Discovery, Brisbane Innovation Park, Griffith
University, Nathan, Queensland 4111, Australia
#Drug Delivery Disposition and Dynamics, Monash Institute of Pharmaceutical Sciences,
Monash University (Parkville Campus), 381 Royal Parade, Parkville, Victoria 3052,
Australia
€Bio21 Institute, The University of Melbourne, Parkville, Victoria 3052, Australia
&Eijkman Institute for Molecular Biology, Jalan Diponegoro 69, Jakarta 10430, Indonesia
1
ACS Paragon Plus Environment

Page 3 of 47
Journal of Medicinal Chemistry
1
2
3
4
¥Global and Tropical Health Division, Menzies School of Health Research, Royal Darwin
5
6
7
8
9
Hospital Campus, Rocklands Drive, Casuarina, Northern Territory 0810, Australia
µCentre for Tropical Medicine and Global Health, Nuffield Department of Clinical Medicine,
University of Oxford, Oxford OX3 7LJ, UK
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 Department of Drug Evaluation, Australian Defence Force Malaria and Infectious
Disease Institute, Brisbane, Queensland 4052, Australia
§Centre for Drug Candidate Optimisation, Monash University (Parkville Campus), 381 Royal
Parade, Parkville, Victoria 3052, Australia
∫ Institute for Therapeutics Discovery and Development, 717 Delaware Street SE, University
of Minnesota, Minneapolis, USA
∞Department of Pathology, Brigham and Women’s Hospital, 75 Francis Street, Boston,
Massachusetts 02115, USA
*Correspondence to:
Jonathan B. Baell
Medicinal Chemistry, Monash Institute of Pharmaceutical Sciences,
Monash University,
Parkville, VIC 3052, Australia
Phone: 61 3 9903 9044
Email: jonathan.baell@monash.edu
KEYWORDS
Malaria, Plasmodium, antimalarial, 1,2,4-triazine
2
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 47
1
2
3
4
Abstract
5
6
7
8
9
A series of 3,3’-disubstituted 5,5’-bi(1,2,4-triazine) derivatives was synthesized and screened
against the erythrocytic stage of Plasmodium falciparum 3D7 line. The most potent dimer, 6k,
with an IC50 (50% inhibitory concentration) of 0.008 µM, had high in vitro potency against P.
falciparum lines resistant to chloroquine (W2, IC50 = 0.0047 ± 0.0011 µM) and artemisinin
(MRA1240, IC50 = 0.0086 ± 0.0010 µM). Excellent ex vivo potency of 6k was shown against
clinical field isolates of both P. falciparum (IC50 = 0.022-0.034 µM) and P. vivax (IC50 =
0.0093-0.031 µM) from the blood of outpatients with uncomplicated malaria. Despite 6k being
cleared relatively rapidly in mice, it supressed parasitemia in the Peters 4-day test, with a mean
ED50s value (50% effective dose) of 1.47 mg kg-1 day-1 following oral administration. The
disubstituted triazine dimer 6k represents a new class of orally available antimalarial
compounds of considerable interest for further development.
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
3
ACS Paragon Plus Environment

Page 5 of 47
Journal of Medicinal Chemistry
1
2
3
4
■INTRODUCTION
5
6
7
8
Malaria is caused by protozoan parasites of the genus Plasmodium that infect red blood
cells (RBCs).1 If left untreated, malaria can progress to severe disease and death. The
introduction of an effective vaccine against malaria remains elusive and chemotherapy thus
remains central to malaria control and treatment. Quinoline compounds, such as chloroquine
and mefloquine, antifolates and artemisinin derivatives have been critical in the fight against
malaria, but parasites have shown development of resistance to these agents.2
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
Over 40% of the world’s population is at risk of malaria, with children and pregnant
women particularly vulnerable in resource-limited communities.3 Each year malaria causes
more than 400,000 deaths, mostly due to infection with Plasmodium falciparum in sub-Saharan
Africa.4 Chemotherapy remains of central importance for the control and ultimate elimination
of malaria, with artemisinin combination therapies (ACTs) being the first-line antimalarial
treatment in most endemic countries.4 However, P. falciparum resistance to artemisinin is now
documented across most of the greater Mekong region, with associated drug resistance
emerging to the longer acting partner drugs. The 2016 WHO report emphasizes the importance
of continuing investment in research and development to combat malaria and to meet the
milestones outlined in Global Technical Strategy (GTS) goals for malaria control and
elimination.5 There is, therefore, an urgent need to develop new medicines able to clear
parasites from the blood, prevent recrudescence, and reduce the duration of treatment.
As parasite resistance to conventional antimalarial agents emerge and spread, so must
the search for new chemotypes that are capable of killing the parasite through new mechanisms.
In this context, the development of OZ439, DSM265, KAE609, KAF156 and MMV390048
are notable (Figure 1).6 OZ439 (Artefenomel), a relatively newly discovered trioxolane being
developed in combination with ferroquine, is being investigated for reduced-frequency dose
regimens. This combination is currently in a phase IIb clinical trial, projected to be completed
4
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 47
1
2
3
4
5
6
7
8
9
by the end of 2018.7 DSM265 has been proven to be potentially useful in treating and protecting
against malaria in a single dose. Dihydroorotate dehydrogenase (DHODH) – an enzyme in the
malaria parasite that is essential for its survival - is selectively targeted by DSM265 without
impacting the orthologous human enzyme. Currently, DSM265 is progressing through phase
IIa clinical trials.8 KAE609 (Cipargamin), an inhibitor of PfATP4, is being developed for a
single exposure radical cure as it is considered to be very potent, fast acting and retained for
more than 8 days in plasma. Cipargamin has completed the first phase IIa clinical trials.9
KAF156 induces rapid killing of the parasite and mechanistic studies have identified decreased
susceptibility associated with mutations in three P. falciparum genes: CARL (cyclic amine
resistance locus), UDP-galactose and acetyl-CoA transporters.10 KAF156 has completed phase
IIa clinical trials in patients with single species P. falciparum or P. vivax infections.
MMV390048 is a PfPI4K inhibitor that has been developed for a single exposure radical cure
with the potential of providing chemoprophylaxis.11 Although these developments are
promising, the high attrition rate encountered during clinical development and the inevitability
of resistance demand continued efforts to discover and develop new antimalarial 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
nBu
F3C
N
nBu
N
NH2
N
O
OH
F5S
N
Cl
NH
F
NH
O
O
O
Cl
Cl
Cl
O
N
N
N
N
Cl
F
H
N
H
O
F
N
SO2Me
OZ439
DSM265
KAE609
MMV390048
KAF156
Figure 1. Antimalarial drugs in the pipeline.
Previously, we have reported preliminary observations of related 3,3’-disubstituted
5,5’-bi(1,2,4-triazine) analogues with activity against P. falciparum.12 We now describe our
structure-activity relationship (SAR) investigation of this novel chemotype with the most
potent compound of this series, 6k, exhibiting an excellent activity against chloroquine and
5
ACS Paragon Plus Environment

Page 7 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
artemisinin resistant P. falciparum lines and being efficacious orally at low doses in the Peters
4-day P. berghei rodent model. We further provide details of ex vivo studies on clinical isolates
from patients with malaria in Papua, Indonesia.
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
Chemistry. The synthesis of all dimeric 3,3’-disubstituted 5,5’-bi(1,2,4-triazine)s 6a-
6o, 7a-7l, 8a-8m , 9a was achieved by cyanide-mediated dimerization of equimolar mixtures
of the appropriately substituted monomers, as adapted from related work reported by us and
others (Scheme 4).12, 13,14 The use of two different monomers results in a product mixture
comprising heterodimer and the two possible homodimers, separation of which yields 3
products that may be tested.
Scheme 1. Synthesis of 3-(alkylthio)-1,2,4-triazinesa
NH2
.
HI/HBr
N
(a)
(b)
NH2
N
NH2
N
NH2
S
N
RS
N
RS
H
1
3
2
(R = various alkyl)
(R = various alkyl)
aReagents and conditions: (a) RBr/NaI or RI, EtOH, 81 °C; (b) aq. Glyoxal/aq. NaHCO3, 0 °C then rt.
Monomers utilized included those with thioether (3, Schemes 1 and 2) or ether (5,
Scheme 3) groups. The synthesis of 3-(alkylthio)-1,2,4-triazine monomers 3a-3c was achieved
after S-alkylating thiosemicarbazide with alkyl halides, followed by cyclocondensation with
aqueous Glyoxal (Scheme 1). An alternative approach to synthesize 3-(alkylthio)-1,2,4-
triazines 3d-3e involved a nucleophilic substitution reaction that was accomplished using 3-
(methylsulfonyl)-1,2,4-triazine 4 and mercaptans (Scheme 2).
Scheme 2. Synthesis of 3-(alkylthio)-1,2,4-triazines
6
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 47
1
2
3
4
(a)
N
N
N
5
6
N
RS
N
MeO2S
N
7
8
3
(R = various alkyl)
4
9
aReagents and conditions: (a) RSH/Na2CO3, ACN, rt.
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 synthesis of 3-(alkoxy)-1,2,4-triazine monomers 5 was achieved using the
nucleophilic substitution reaction involving alcohols and 3-(methylsulfonyl)-1,2,4-triazine 4
(Scheme 3). This reaction often proved efficient when magnesium alkoxides were employed
instead of sodium alkoxides in polar solvents such as DMF.
Scheme 3. Synthesis of 3-(alkoxy)-1,2,4-triazines
(a)
N
N
N
N
RO
N
MeO2S
N
5
(R = various alkyl)
4
aReagents and conditions: (a) NaH/ROH or MeMgI/ROH, DMF, rt.
Scheme 4. Synthesis of 3,3'-disubstituted 5,5'-bi(1,2,4-triazine)s starting with 3-substituted 1,2,4-
triazines
N
N
N
N
N
N
N
N
N
N
MeS
Me
MeS
RS
+
RO
RS
Me
N
N
N
N
N
(a)
+
3
5
+
N
N
N
N
N
SCH2CH2NMe2
S-iPr
OR'
OR'
SR'
N
N
N
N
N
N
N
N
N
N
8a-m
6n
6o
6a-o
7a-l
9a
(R = R')
aReagents and conditions: (a) aq. KCN, EtOH or Dioxane.
7
ACS Paragon Plus Environment

Page 9 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
SAR for P. falciparum. Testing in vitro against the asexual blood stages of P.
falciparum utilized our optimized and miniaturized assay as previously reported.15 We have
not shown in Table 1 data for all possible dimeric reaction products arising from Scheme 4, but
rather just the more informative results. Relative to the symmetrical thiomethyl dimer 7a (IC50
= 0.026 M), extension of one of the S-methyl groups to propyl (6a, IC50 = 0.022 ± 0.004 M),
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
isopropyl (6b, IC50 = 0.026 M), isobutyl (6c, IC50 = 0.028 M), or cyclopentyl (6e, IC50 =
0.015 M), led to similarly potent compounds. Further branching to the tert-butyl (6d, IC50 =
0.053 ± 0.015 M) or expansion to cyclohexyl (6f, IC50 = 0.041 ± 0.026 M) led to a marginal
drop in bioactivity. The suggestion that steric bulk might become limiting was starkly shown
by the relative loss of activity in the tert-butylphenyl analogue 6g (IC50 = 0.910 ± 0.004 M)
and the 2-pyridyl analogue 6h (IC50 = 1.565 ± 0.361 M). However, it is the tert-butyl group
in the former and the endocyclic nitrogen atom in the latter that are likely dominating negative
influences, rather than an aryl ring per se, as impressive activity was recouped in thioanisole
derivative 6m (IC50 = 0.086 ± 0.019 M). Returning to alkyl extensions, incorporation of a
heteroatom was less favorable to varying degrees (for example 6i with IC50 = 0.951 ± 0.025
M, 6j with IC50 = 0.12 M, 6l with IC50 = 0.11 M) except for the N,N-dimethylaminoethyl
derivative 6k, which exhibited an impressive IC50 of 0.0080 M. In contrast to the limited
tolerance to the nature of the 3-substituent, substitution at the 6-position led to marked
abrogation of activity, even with a group as innocuous as methyl (6n with an IC50 2.4 M
compared with 6b with an IC50 of 0.026 M and 6o with IC50 of 1.4 M compared with 6k
with an IC50 of 0.0080 M).
A natural consequence of the synthetic route deployed is provision of homodimers, and
a selection of the most interesting of these are shown in Table 1. Here, homodimers with
relatively smaller S-alkyl groups are generally quite potent, and 7a-7f display IC50 values that
8
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 47
1
2
3
4
5
6
7
8
9
range from 0.017 M for the S-propyl homodimer 7c to 0.12 M for the S-isopropyl dimer 7e.
Interestingly, negative steric hindrance effects were markedly increased relative to
heterodimeric counterparts, and S-cyclopentyl homodimer 7h (IC50 = 0.42 M), S-cyclohexyl
homodimer 7i (IC50 = 1.470 ± 0.170 M) and especially S-tert-butyl homodimer 7g (IC50 =
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
3.360 ± 2.376 M) were markedly less active. Given this, the inactivity of 7j (63% inhibition
at 40 µM) was not surprising. In particular, the marked loss in activity observed with the
modifications of 6e resulting in 7h suggested a nonsymmetric site of action. Homodimers 7k
and 7l were intriguing and while the former showed only a slight activity loss (IC50 = 0.41 M)
relative to its heterodimeric counterpart 6l, the latter compound’s activity was much weaker
(IC50 = 1.3 M) than its heterodimeric counterpart 6j (IC50 = 0.12 M).
Table 1. SAR of 3,3’-Disubstituted 5,5’-bi(1,2,4-triazine)s against in vitro P. falciparum
proliferation
Cpd
R
R’
IC50 (M) a
Unsymmetrical 3-alkylthio-3’-alkylthio-5,5'-bi(1,2,4-triazine)
6a
6b
6c
6d
6e
6f
Me
Me
Me
Me
Me
Me
Me
Me
Me
Pr
iPr
0.022 ± 0.004
0.026
iBu
0.028
tBu
0.053± 0.015
0.015
cyclopentyl
cyclohexyl
para-tBu-phenylene
2-pyridyl
-(CH2)2OH
0.041± 0.026
0.91± 0.004
1.565 ± 0.361
0.951 ± 0.025
6g
6h
6i
9
ACS Paragon Plus Environment

Page 11 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
6j
6kb
6l
Me
Me
-(CH2)3OH
-(CH2)2NMe2
-(CH2)2OMe
-C6H4-para-OMe
iPr
0.12
0.0080c,d
0.11
Me
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
6m
6n
6o
Me
0.086 ± 0.019
2.4
Me, 6-Me
Me, 6-Me
-(CH2)2NMe2
1.4
Symmetrical 3-alkylthio-3’-alkylthio-5,5’-bi(1,2,4-triazine)
7ab
7bb
7cb
7db
7e
R= R’= Me
R= R’= Et
0.026
0.022
R= R’= nPr
0.017
R= R’= nBu
0.096
R= R’= iPr
0.12
7f
R= R’= iBu
0.050
7g
7h
7i
R= R’= tBu
3.36 ± 2.376
0.42
R= R’= cyclopentyl
R= R’= cyclohexyl
R= R’= para-t-Bu-phenylene
R= R’= -(CH2)2-OMe
R= R’= -(CH2)3OH
1.47 ± 0.170
63% at 40 µM
0.41
7j
7k
7l
1.3
Unsymmetrical 3-alkylthio-3’-alkoxy-5,5'-bi(1,2,4-triazine)
8ab
8b
8c
Me
Me
Me
Me
Me
Bn
0.15
0.039 ± 0.016
0.76
-(CH2)2OH
-(CH2)2-OMe
8d
0.395 ± 0.025
10
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 47
1
2
3
4
5
6
7
8
9
8e
8f
Me
-CH2CF3
-(CH2)2SiMe3
-(CH2)2-4-Me-thiazole
-CH(Me)CH2OMe
-(CH2)2NMe2
Cyclopropyl
0.091 ± 0.023
Me
0.033 ± 0.001
0.069 ± 0.024
1.1
8g
8h
8ib
8j
Me
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
Me
Me
0.080
-(CH2)2NMe2
0.013 ± 0.001
0.20
8k
8l
iPr
iPr
iPr
Me
-(CH2)2OMe
0.20
8m
-CH(Me)CH2OMe
0.61
Symmetrical 3-alkoxy-3’-alkyloxy-5,5’-bi(1,2,4-triazine)
R= R’= Me 4.1
9a
aValues represent outcomes from one or two experiments (for two experiments, the values
represent the mean, with ‘±’ sign representing the standard deviation,) against P. falciparum
3D7 strain, erythrocyte stage (chloroquine control, IC50 = 0.004 µM). bThese compounds are
c
reported in ref. 12. Cytotoxicity was measured for 6k, displaying an IC50 against HEK 293
cells of 21 M, hence a selectivity index of >2600 (puromycin control, IC50 = 0.41 µM).
dTested as the hydrochloride salt.
Our recent success14 in optimizing a working method to obtain good yields of 3-alkoxy-
1,2,4-triazines 5 meant that it was possible for us to synthesize a number of 3-alkoxy-3’-
alkylthio heterodimers according to Scheme 4. Our interest in these compounds was influenced
by our prior hypothesis that the microsomal instability of 7a could be mediated by the presence
of two thioether groups.14 As the preliminary investigations illustrated, replacement of one
thiomethyl group in 7a with a methoxy group resulted in a substantial loss of activity but the
resulting heterodimer 8a was nevertheless still relatively potent, exhibiting an IC50 value of
11
ACS Paragon Plus Environment

Page 13 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
0.15 M against P. falciparum 3D7. Furthermore, functional groups with increased
lipophilicity in 8b and 8e-8g such as benzyloxy (8b, IC50 = 0.039 ± 0.016 M), trifluoroethoxy
(8e, IC50 = 0.091 ± 0.023 M), and trimethylsilylethoxy (8f, IC50 = 0.033 ± 0.001 M) led to
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
levels of potency approaching that of 7a (IC50 0.026 M). We were intrigued by 8f, since
silicon has unique properties in terms of large covalent radius, enhanced lipophilicity and
ability to alter metabolic pathways, and has been put to good use in a myriad of medicinal
chemistry applications.16, 17
In parallel with the SAR just discussed, higher polarity sidechains such as those in 8c
(IC50 = 0.76 M) and 8d (IC50 = 0.395 ± 0.025 M) led to significant activity loss and this was
exacerbated by branching, with 8h showing an IC50 of 1.1 M. In agreement is the observation
that additional heteroatoms in the sidechain was not necessarily unfavorable; the N,N-
dimethylaminoethoxy sidechain of 8i led to potent activity, with an IC50 of 0.080 M. The
related heterodimer 8j was even more potent (IC50 = 0.013 ± 0.001 M), where the basic
sidechain was installed in the thioether linkage and the ether alkyl group was cyclopropyl.
Analogues 8k-8m can be compared with the S-isopropyl parent 6b. Here, it is observed that
the methoxy counterpart 8k (IC50 = 0.20 M) and methoxyethoxy counterpart 8l (IC50 = 0.20
M) are less tolerated than the methylthio of 6b (IC50 = 0.026 M) and that again, branching
of the -carbon is detrimental, with 8m returning an IC50 of 0.61 M.
ALARM NMR. It is plausible that a thioether group attached to an electron-deficient
position in the 1,2,4-triazine harbors some non-specific thiol reactivity. Although not
precluding development, it is preferable for drug candidates to not exhibit this trait. ALARM
NMR is an industry-developed heteronuclear multiple quantum coherence (HMQC) counter
screen to detect non-specific protein thiol reactivity.18, 19 As shown in Figure 2, 6k did not
perturb the La antigen conformation, suggesting it is unlikely to react indiscriminately with
protein cysteine thiol residues.
12
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 47
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
DMSO
6k
Figure 2. Compound 6k does not perturb the La antigen conformation by ALARM NMR
counter-screen for nonspecific compound thiol reactivity. Shown are 1H-13C HMQC spectra
of key 13C-labeled methyl groups of the La antigen after incubation with either DMSO or 6k,
which was incubated with the La antigen probe in the presence (blue spectra) and absence (red
spectra) of excess DTT. Data are normalized to DMSO vehicle control. ALARM NMR-
positive compounds perturb the La antigen, as evidenced by pronounced significant peak shifts
and/or peak signal attenuations in the absence of DTT.19
Physicochemical data and microsomal stability studies. A selection of compounds
was subjected to preliminary evaluation of drug-likeness. As shown in Table 2, the molecular
weight of each compound was well below 500 g/mol, topological polar surface area ranged
from 77 to 118 Å2, calculated distribution coefficients (clogD) at pH 7.4 ranged from -0.6 to
3.6, and kinetic aqueous solubility in some cases was higher than 100 g mL-1. These properties
are likely to predispose these triazine dimers favorably towards oral bioavailability, albeit the
13
ACS Paragon Plus Environment

Page 15 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
solubility range was wide with two compounds (7d and 7f) showing poor aqueous solubility.
The better solubility for compounds such as 6j, 7l, 8i and 9a can be attributed to a marked
influence of the sidechain polarity. It is worth pointing out that some compounds such as 6k
and 8i, with a tertiary amino group (–NMe2) attached at the end of the side chain, will be mostly
ionized according to pKa calculations (8.2 and 8.4, respectively), aiding solubility. The fact
that the solubility of 8i is so much better than that of 6k testifies to the greater polarity of an
ether group relative to a thioether group.
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
Table 2. Key physicochemical parameters and in vitro metabolic stability of selected
3,3’-disubstituted 5,5’-bi(1,2,4-triazine)s
Solubility
In vitro CLint
Partition
Degradation
half-life
Microsome-
b
-1
(µL / min / mg
(µg mL )
Coefficients
c
PSAa
predicted EH
[h/m]
Cpd
protein)c
[h/m]
cLogD
(min)c [h/m]
cLogP
pH 2.0 pH 6.5
7.4
6.3–
12.5
50–
100
6.3–
12.5
6.3–
12.5
3.1–
6.3
6.3–
12.5
50–
100
6.3–
12.5
3.1–
6.3
6e
6j
77
98
82
77
2.3
0.5
1.0
1.2
2.3
4 / na
16 / 5
421 / na
111 / 350
15 / 138
146 / 655
0.94 / >0.95
0.82 / 0.88
0.38 / 0.75
0.85 / 0.93
0.5
0.1
1.2
6k
7a
113 / 13
12 / 3
3.1–
6.3
7c
7d
7f
77
77
77
77
2.8
3.6
3.5
3.5
2.8
3.6
3.5
3.5
7 /na
19 / 2
46 / 2
32 / 3
252 / na
92 / 1031
38 / 1066
54 / 516
0.91 / >0.95
0.79 / 0.96
0.60 / 0.96
0.68 / 0.92
˂ 1.6
1.6–
3.1
˂ 1.6
1.6–
3.1
7h
NA
NA
14
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 47
1
2
3
4
5
6
7
8
9
12.5–
12.5–
25
7k
7l
96
118
91
0.6
-0.3
0.4
0.6
-0.3
-0.6
-0.4
68 / 14
57 / 18
26 / 124
31 / 97
20 / 56
˂ 7 / 9
0.50 / 0.73
0.55 / 0.68
0.44 / 0.55
˂ 0.2 / 0.162
25
60–
100
50–
100
50–
100
25–50
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
8i
25–50
87 / 31
9a
96
-0.4
˃ 100
˃ 247 / 189
aCalculated using Chem Axon JChem software.
bKinetic solubility determined by nephelometry.
cDegradation half-life and in vitro intrinsic clearance determined in human (h) or mouse (m)
liver microsomes and predicted hepatic extraction ratio calculated therefrom.
Ensuring that compounds are not highly metabolized is important for good oral
bioavailability and to achieve a reasonable in vivo half-life (T1/2). The metabolic stability of
compounds shown in Table 2 was evaluated in human and mouse liver microsomes (referred
o
to as HLM and MLM, respectively) at 37 C. The metabolic reaction was initiated by the
addition of the NADPH-regenerating buffer system. Substrate depletion was used to calculate
the first order degradation T1/2, in vitro intrinsic clearance value (CLint) and the in vivo hepatic
extraction ratio (EH), which provides guidance on the structures of 3,3’-disubstituted dimers
that are most vulnerable to hepatic metabolism.
From the data in Table 2 it is clear that compounds 6e and 7c, both of which are very
potent, were rapidly metabolised in both HLM and MLM. Stability was somewhat improved
with an increase in the alkyl chain in going from compound 7c to 7d (HLM T1/2 = 19 min),
where the S-alkyl group increases in length from three to four carbon atoms, although the
corresponding IC50 values also increased from 0.017 to 0.096 µM. The branching of the side
15
ACS Paragon Plus Environment

Page 17 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
chain appeared to have a favorable effect on the metabolic stability. For example, 7f with an
isobutyl group as the S-alkyl substituent was more stable in HLM (T1/2 = 46 min) when
compared with 7d that has an n-butyl group as the S-alkyl substituent (HLM T1/2 = 19 min).
However, the increase in the T1/2 was also associated with a two-fold reduction in potency.
Compounds such as 7k and 7l, with O-atoms on the alkyl chain resulted in an HLM degradation
T1/2 of nearly one hour and a marginally improved MLM T1/2 (14-18 min), albeit that these
compounds have respective IC50 values of 0.408 and 1.347 µM and represent significantly
lower potency. Compounds 6k (IC50 = 0.0080 M) and 8i (IC50 = 0.080 M) have significant
structural analogy and exhibited higher HLM/MLM T1/2 values of 113/13 and 87/31 min,
respectively. It is worth noting that both 6k and 8i have good potency against the 3D7 line,
with 6k being the most potent in the series of compounds investigated in these studies. In the
set of compounds included in this study, 9a is by far the most metabolically stable compound,
with a HLM/MLM T1/2 greater than 247/189 min but has significantly lower potency (IC50 of
4.1 µM). On the basis of these analyses, 6k was chosen for further biological investigation,
principally because of its marked potency but also because we deemed its microsomal stability
to still be workable.
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
Drug susceptibility studies in vitro: comparison of 6k with chloroquine-resistant
and artemisinin-resistant P. falciparum lines. Drug resistance is a problem in antimalarial
drug therapy. New chemotypes are likely to offer new mechanisms of action and good
antimalarial potencies across drug resistant P. falciparum lines. Compound 6k was tested in
the [3H]-hypoxanthine growth inhibition assay20 with a 48 h exposure period against the P.
falciparum D6 line that is sensitive to chloroquine and most other antimalarials, and W2, which
is resistant to chloroquine and pyrimethamine. Table 3 shows both IC50 and IC90 values for the
compounds investigated. It is clear that the potency of 6k against the D6 and W2 lines was
16
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 47
1
2
3
4
5
6
comparable, with IC50 values less than 0.01 µM against both strains, whereas the chloroquine
IC50 was 10-fold higher in the chloroquine resistant W2 strain compared to D6.
7
8
9
Comparative potency measurements were also carried out against the artemisinin
sensitive P. falciparum MRA1239 line and the artemisinin-resistant MRA1240 line.25
MRA1239 is susceptible to dihydroartemisinin and mefloquine but resistant to chloroquine,
while MRA1240 is resistant to all of the aforementioned drugs. We note that 6k yielded potent
IC50 values of 0.0109 µM and 0.0086 µM against MRA1239 and MRA1240, respectively. It is
clear from these results using the [3H]-hypoxanthine growth inhibition assay, that 6k has good
potency against both chloroquine and artemisinin-resistant strains when tested under these
continuous drug exposure conditions.
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
Table 3. In vitro antimalarial activity of 6k and chloroquine against various P.
falciparum linesa
D6
W2
MRA1239
IC50
(M)
MRA1240
IC50
IC90
IC50
(M)
IC90
Cpd
IC50 (M) IC90 (M)
IC90 (M)
(M)
(M)
(M)
0.0084 ±
0.0008
0.0129 ±
0.0004
0.0047 ± 0.0083 ± 0.0109 ±
0.0155 ±
0.0027
0.131 ±
0.002
0.0086 ± 0.0130 ±
6k
0.0011
0.150 ±
0.030
0.0012
0.316 ±
0.010
0.0015
0.0757 ±
0.0010
0.0010
0.097 ±
0.021
0.0017
0.217 ±
0.071
Chloro- 0.0160 ±
quine 0.0032
0.0196 ±
0.0060
aValues are mean of two independent experiments with each assay performed with three
technical replicates. Results expressed as mean ± SD.
Ex vivo drug susceptibility of known antimalarial drugs and 6k. Chloroquine,
piperaquine, mefloquine, artesunate and 6k were screened against the chloroquine-sensitive
FC27 line and the chloroquine-resistant K1 line. These results were compared with drug
17
ACS Paragon Plus Environment

Page 19 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
susceptibilities measured using Plasmodium clinical isolates obtained from patients with
symptomatic malaria presenting to a clinic in Papua province in eastern Indonesia. High levels
of multidrug-resistant P. falciparum and chloroquine-resistant P. vivax have been reported in
this location previously.26, 27 Patients with a mono-infection with either P. falciparum or P.
vivax with a peripheral parasitaemia of between 2,000 and 80,000 µL-1 were included in the
study.
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
Only blood samples that showed ˃60% parasites at the ring development stage were
used after passing through Plasmodipur® filters for white blood cell depletion. Typically,
parasites were incubated with test compound at 37oC for 35–56 h until ˃ 40% of parasites
reached the mature schizont stage for the control with no antimalarial. When this in vitro assay
was performed using the FC27 line, the IC50 for 6k was 0.0153 µM, comparable with that for
chloroquine (IC50 = 0.0139 µM) and much lower than that for piperaquine (IC50 = 0.0385 µM)
or mefloquine (IC50 = 0.0390 µM; Table 4). Against the K1 line, the IC50 for 6k was 0.0465
µM, significantly lower than that for chloroquine (IC50 = 0.1112 µM), and similar to that for
piperaquine (IC50 = 0.0508 µM), but higher than that for mefloquine (IC50 = 0.0061 µM). As
expected, artesunate exhibited good potencies with both FC27 (IC50 = 0.0029 µM) and K1
(IC50 = 0.0023 µM) lines.
Table 4. Ex vivo antimalarial activity of known antimalarial drugs and 6k against
laboratory lines and clinical field isolates of P. falciparum and P. vivax
P. falciparum lab lines
IC50 (M)
P. falciparum clinical field
isolates
P. vivax clinical field
isolates
Median
Median
FC27
Compound
K1 (CQR)
(CQs)
n (%)
(Range) IC50,
n (%) (Range)) IC50,
M
M
18
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 47
1
2
3
4
5
6
7
8
9
0.0484
(0.0193-
0.2633)
p=0.003
0.0452
0.0950 (0.0435-
11
Chloroquine
Piperaquine
Mefloquine
0.0139
0.0385
0.0390
0.1112
0.0508
0.0061
10 (100)
10 (100)
10 (100)
0.3258)a
(100)
p=0.005
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
0.0207 (0.0113-
0.0570)
11
(0.0096-
0.1096)
p=0.041
0.0440
(100)
p=0.508
0.0187 (0.0100-
0.0453)
11
(0.0191-
0.0691)
p=0.004
0.0033
(100)
p=0.202
0.0021 (0.0009-
0.0052)
11
(0.0011-
0.0089)
p=0.003
0.0158
Artesunate
0.0029
0.0153
0.0023
0.0465
10 (100)
10 (100)
(100)
p=0.005
0.0267 (0.0221-
0.0346)
11
6k
(0.0093-
0.0305)
(100)
CQS, chloroquine-sensitive laboratory strain
CQR, chloroquine-resistant laboratory strain
aComparison with 6k (Wilcoxon rank sum test); p values represent the significance of difference with
6k.
In the ex vivo drug susceptibility assays using the clinical P. vivax isolates, the median
IC50 for 6k was 0.0158 µM, which was significantly lower than that for chloroquine (IC50 =
0.0484 µM), piperaquine (IC50 = 0.0452 µM), or mefloquine (IC50 = 0.0440 µM). When tested
19
ACS Paragon Plus Environment

Page 21 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
against P. falciparum field isolates, the median IC50 for 6k was 0.0267 µM, significantly lower
than that for chloroquine (IC50 = 0.0950 µM), but similar to that for piperaquine (IC50 = 0.0207
µM) and mefloquine (IC50 = 0.0187 µM). Artesunate was the most potent of the tested
antimalarials with a median IC50 of 0.0021 µM against P. falciparum and 0.0033 µM against
P. vivax isolates.
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
Pharmacokinetics (PK) of 6k in mice following intravenous (IV) and oral
administration. Although the in vitro metabolism half-life of 6k in MLM of 13 min was
relatively short, its notably potent in vitro antimalarial activity justified further evaluation of in
vivo potency in a murine malaria model. The pharmacokinetic properties were assessed using
non-fasted, male, Swiss outbred mice (24–33 g) that had full access to food and water before
and after the sampling period. Compound 6k was administered by a bolus IV injection into the
tail vein at a dose of 3 mg kg-1 or orally by gavage at doses of 3 and 30 mg kg-1. Groups of two
mice were used for each dose. Blood samples were collected into tubes containing heparin and
a stabilization cocktail for up to 24 h either by a submandibular bleed (120 µL) or terminal
cardiac puncture (600 µL) with a maximum of two samples per mouse. Samples were
centrifuged for the collection of plasma which was stored at -20°C until analysis.
After IV administration, measurable (greater than 20 nM) quantities of 6k were
observed for only up to 30 min (Figure 3A) indicating very high in vivo clearance consistent
with the low metabolic stability observed in MLM. Following oral administration at 3 mg kg-
1, 6k was detected in only one mouse 30 min after the dose was administered (Figure 3B). At
the 30 mg kg-1 dose level, measurable quantities of 6k were observed at 60 and 120 min time
points for one mouse. To rule out poor permeability as a contributor to the low oral
bioavailability, the transport of 6k was assessed across Caco-2 cell monolayers. In both the
A-B and B-A directions, 6k exhibited well-defined flux profiles with high mass balance (85%
± 8 for A-B and 90% ± 2 for B-A). Values for Papp in the A-B direction (49 ± 5 x 10-6 cm/s)
20
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 47
1
2
3
4
5
6
7
8
9
indicate that permeability across the Caco-2 cell monolayer was high, whilst the efflux ratio
(1.1 ± 0.2) suggests that 6k was not subject to polarised transport in this test system. Reasonable
aqueous solubility and good permeability across Caco-2 cell monolayers suggests that the
limited plasma exposure of 6k following oral administration is likely due to extensive first-
pass metabolism. Nonetheless, due to its single digit nanomolar potency against P. falciparum,
6k was progressed to efficacy studies in the Peters 4-day murine model.
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
10
1
10
1
B
A
0.1
0.01
0.1
0.01
*
*
*
0
30
60
90
120 1440
0
30
60
90
120 1440
Time (min)
Time (min)
Figure 3. Plasma concentrations of 6k in male Swiss outbred mice following (A) IV
administration at 2.8 mg kg-1 (B) oral administration at 2.9 mg kg-1 (filled square) and 28 mg
kg-1 (open triangle). Data points represent individual samples with n=2 samples per time
point except where shown by * where only one measurement was available.
In vivo efficacy of 6k in the Peters 4-day test. Swiss outbred ARC (Animal Resource
Centre, Murdoch, WA) female mice that were 5–7 weeks old with a mean (± SD) body weight
of 28.1 ± 2.3 g were used in these studies. A tolerability assessment was first carried out to
establish if 6k posed any gross toxicity to mice. Three oral dose regimens were assessed: 4, 8
and 16 mg kg-1 day-1 for 4 days with mice in a group of three for each dose. Each dose was
administered at 24 h intervals. The animals were observed for physical distress 2 h after the
administration and at two time points during the day; one between 7–9 am and one between
21
ACS Paragon Plus Environment

Page 23 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
3–5 pm. Physical adverse events such as inability to move, reduced appetite, extreme pallor,
ruffled hair, and loss of body weight were monitored. None of the mice exhibited any of the
aforementioned signs indicating that the dose regimens of 6k used in the Peters 4-day test were
not likely to be harmful to the mice.
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
For the Peters 4-day test, mice were inoculated intraperitoneally with 20 × 106 P.
berghei-infected RBCs.21 Mice were treated with two-fold increasing doses (ranging from
0.125–16 mg kg-1 day-1 with eight increasing doses) in a group of six. The first dose was
administered 2 h after inoculation of infected RBCs (D0). The same dose was administered for
three more days after each 24 h period. Blood samples were collected on D+4 for analysis by
blood film microscopy. Two different routes of administration (oral and subcutaneous) were
evaluated. The subcutaneous route is known to provide good drug absorption and rapid action,
while the oral route bioavailability is contingent on the ability of the drug to be absorbed and
avoid first pass elimination by the liver. The mean (± SD) ED50 and ED90 values obtained for
subcutaneous administration of 6k were 0.21 ± 0.12 mg kg-1 day-1 and 0.60 ± 0.02 mg kg-1 day-
1, respectively. For oral administration, the mean ED50 and ED90 values of 6k were 1.47 ± 0.01
mg kg-1 day-1 and 3.43 ± 0.40 mg kg-1 day-1, respectively. In contrast, the ED50 (oral) values
for chloroquine and dihydroartemisinin were 1.1 mg kg-1 day-1 and 1.3 mg kg-1 day-1,
respectively, while the untreated vehicle control mice had to be euthanized after 6–7 days post
parasite inoculation. Hence, 6k is efficacious when dosed subcutaneously and maintained good
oral efficacy. The 6-fold drop in oral efficacy of 6k relative to its subcutaneous efficacy might
be expected if 6k is subject to extensive first pass metabolism, assuming that 6k is the active
species in vivo. The fact that 6k has oral activity, despite its pharmacokinetic profile as shown
in Figure 3A, led us to consider what metabolites might plausibly be produced in vivo that
might contribute to oral efficacy.
22
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 47
1
2
3
4
5
6
7
8
9
Preliminary metabolite identification. We selected five compounds (6k, 7a, 7d, 8i,
9a) that exhibited low to high rates of degradation in HLM and MLM (Table 2). For most of
the compounds, metabolic stability parameters were broadly comparable between species,
except for 6k where the rate of degradation was lower in HLM compared with MLM. A
preliminary metabolite screen was conducted for each compound by monitoring for the
accurate mass of likely metabolites. Mono- and bis-oxygenation, dehydrogenation,
demethylations/dealkylations, deaminations (to alcohols or acids), were searched for by MS
under positive ionization mode. For the purposes of this preliminary study, full structure
elucidation/confirmation using MS/MS was not carried out. While for 7a and 9a, no
interpretable metabolite signals were found, for each of 6k, 7d and 8i, a putative metabolite
corresponding to mono-oxygenation (mass of parent molecule + 16 Da) was detected following
incubation with both HLM and MLM. Plausible sites for mono-oxygenation for all three
compounds comprise the thioether sulfur atom or a ring nitrogen atom, while 6k and 8i offer
the additional prospect of tertiary amine nitrogen atom mono-oxygenation. It is interesting that
7a did not give rise to an obvious mono-oxygenation peak whereas closely related analogue 7d
did. The behavior of 7d was noteworthy in other ways. In the absence of cofactors, 7d showed
significant degradation in both HLM and MLM (approximately 41 - 51% loss of parent over
the course of incubation), suggesting a contribution of non-specific degradation to the overall
degradation rate of this compound. However, none of these metabolic transformations were
detected in the control incubations.
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
N
N
N
N
N
N
N
S
N
N
S
S
N
N+
O-
S
N
N
N
M+16 (II)
M+16 (I)
23
ACS Paragon Plus Environment

Page 25 of 47
Journal of Medicinal Chemistry
1
2
3
4
Figure 4. Proposed structures for metabolites of 6k. Dotted
5
6
lines indicate possible sites of metabolism.
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
In this preliminary study, metabolite standards were not available and LC-MS
conditions were therefore optimized for the parent compound only. Incubation conditions were
also optimized for assessing the rate of loss of each parent compound (i.e. using low substrate
and low microsomal protein concentrations). To more accurately identify and confirm the
formation of the major metabolites of 6k in HLM and MLM, incubations were conducted at a
high substrate (10 μM) and high protein (1 mg mL-1) concentration to facilitate metabolite
identification (SI). Two putative metabolites consistent with mono-oxygenation ([MH+] 326,
M+16 (I & II)) were detected. M+16 (I) was detected in HLM only, while M+16 (II) was
present in both HLM and MLM. Based on analysis of the MS/MS data, M+16 (I) is likely
derived from oxygenation at the methylsulfanyl-triazine region of the molecule, while M+16
(II) is likely an N-oxide occurring at the N,N-dimethylamine (Figure 4). Based on peak area
comparison, the putative N-oxide, M+16 (II), is the predominant metabolite in both species. In
addition, two putative metabolites corresponding to bis-oxygenation ([MH+] 342, M+32) and
demethylation ([MH+] 296, M-14) were identified by accurate mass measurement. However,
due to poor sensitivity, structure assignment was not possible for either product.
For M+16 (I), four variants are possible if oxidation of any of the ring nitrogen atoms
is considered feasible in addition to oxidation of the sulfur atom. Further work is in progress to
synthesise these possible metabolites, confirm their structure and test their in vitro ADME and
activities as well as their in vivo efficacy.
■ CONCLUSION
24
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 47
1
2
3
4
5
6
7
8
9
Our analysis of the SAR of 3,3’-disubstituted 5,5’-bi(1,2,4-triazine) derivatives has established
that 6k possesses single digit nanomolar in vitro antimalarial activity against the 3D7 line, and
also against chloroquine-, dihydroartemisinin, and mefloquine-resistant P. falciparum lines. In
ex vivo studies of P. falciparum and P. vivax clinical field isolates collected in Timika,
Indonesia, 6k proved to be more active than chloroquine and comparable with piperaquine
when tested against P. falciparum isolates, and displayed greater activity than chloroquine,
piperaquine and mefloquine, against P. vivax isolates. When evaluated in the P. berghei rodent
model, 6k proved to be efficacious in supressing the development of parasites, with an ED50
value of 0.21 mg kg-1day-1 when administered subcutaneously and 1.47 mg kg-1 day-1 when
administered orally, making it comparable to chloroquine and 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
The relatively poor systemic exposure is somewhat at odds with the oral efficacy
exhibited by 6k and may suggest that efficacy is driven by active metabolites. However, at this
stage, the possibility that 6k acts by an unusual in vivo PK/PD relationship cannot be ruled out
and further work is required to better understand these intriguing effects.
This new class of orally active antimalarial bitriazines is an attractive proposition for
further optimizing towards a preclinical candidate, and we will report on these efforts in due
course.
■ EXPERIMENTAL SECTION
Chemistry. General Experimental Methods. For all synthetic endeavours
commercially available reagents were used without further purification. Column
chromatography was performed using silica gel 60 (40–60 µm). The solvents for
chromatography were used without purification. The reactions were monitored by TLC on
25
ACS Paragon Plus Environment

Page 27 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
Silica Gel 60F-254 plates with detection by UV light and/or KMnO4 stain (1.50 g KMnO4, 10.0
g K2CO3, and 1.25 mL 10% NaOH in 200 mL water).
7
8
9
Purity. The purity of all compounds submitted for biological testing was >95%, as
determined using methods described (vide infra).
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
NMR. 1H and 13C NMR spectra were recorded at 400.13 and 100.62 MHz respectively,
on a Bruker Avance III Nanobay spectrometer with a BACS 60 sample changer. The NMR
solvents were purchased from Cambridge Isotope Laboratories. Chemical shifts (δ, ppm) are
reported relative to the solvent peak (CDCl3): 7.26 [1H] or 77.16 [13C]; DMSO d6: 2.50 [1H] or
39.52 [13C]). Proton resonances are annotated as: chemical shift (δ), multiplicity (s, singlet; d,
doublet; m, multiplet), coupling constant (J, Hz), and the number of protons.
LCMS (protocol-A). Low resolution mass spectrometry analyses were performed with
an Agilent 6100 Series Single Quad LC/MS coupled with an Agilent 1200 Series HPLC, 1200
Series G1311A quaternary pump, 1200 series G1329A thermostated autosampler and 1200
series G1314B variable wavelength detector. The conditions for liquid chromatography were:
reverse-phase HPLC analysis fitted with a Phenomenex Luna C8(2) 5 µm (50 x 4.6 mm) 100
Å column; column temperature: 30°C; injection volume: 5 µL; solvent: 99.9% acetonitrile,
0.1% formic acid; gradient: 5–100% of solvent over 10 min; detection: 254 nm. The conditions
for mass spectrometry were: quadrupole ion source; ion mode: multimode-ES; drying gas
temp: 300°C; vaporizer temperature: 200°C; capillary voltage: 2000 V (positive), 4000 V
(negative); scan range: 100-1000 m/z; step size: 0.1 sec; acquisition time: 10 min.
LCMS (protocol-B). LRMS [M+H]+ of compounds was analysed on an Agilent
UHPLC/MS 1260/6120 system with the following technical information. Pump: 1260 Infinity
G1312B Binary pump; autosampler: 1260 Infinity G1367E 1260 HiP ALS; detector: 1290
Infinity G4212A 1290 DAD. LC conditions: reverse-phase HPLC analysis; column: Poroshell
26
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 47
1
2
3
4
5
6
7
8
9
120 EC-C18 3.0 X 50mm 2.7-Micron; column temperature: 35°C; injection volume: 1 μL; flow
rate: 1 mL min-1. Solvent A: 99.9% water, 0.1% formic acid, solvent B: 99.9% ACN, 0.1%
formic acid, gradient: 5-100% of solvent B in solvent A over 3.8 mins. Gradient takes 4 min to
get to 100% solvent B in solvent A; maintain for 3 min and a further 3 min to get back to the
original 5% solvent B in solvent A. MS conditions: ion source: Quadrupole, ion mode: API-
ES, drying gas temp: 350°C; capillary voltage (V): 3000 (positive); capillary voltage (V): 3000
(negative); scan 52 range: 100-1000; step size: 0.1 s; acquisition time: 5 min. LC/MSD
Chemstation Rev.B.04.03 coupled with Masshunter Easy Access Software managed the
running and processing of samples.
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
HRMS. High-resolution MS was performed with an Agilent 6224 TOF LC/MS coupled
to an Agilent 1290 Infinity LC. All data were acquired and reference mass corrected via a dual-
spray electrospray ionisation (ESI) source. Each scan or data point on the total ion
chromatogram (TIC) is an average of 13,700 transients, producing a spectrum every second.
Mass spectra were created by averaging the scans across each peak and subtracting the
background from the first 10 s of the TIC. Acquisition was performed using the Agilent Mass
Hunter Data Acquisition software ver. B.05.00 Build 5.0.5042.2 and analysis was performed
using Mass Hunter Qualitative Analysis ver. B.05.00 Build 5.0.519.13. Acquisition
parameters: mode, ESI; drying gas flow, 11 L min-1; nebuliser pressure, 45 psi; drying gas
temperature, 325°C; voltages: capillary, 4000 V; fragmentor, 160 V; skimmer, 65 V; octapole
RF, 750 V; scan range, 100–1500 m/z; positive ion mode internal reference ions, m/z
121.050873 and 922.009798. LC conditions: Agilent Zorbax SB-C18 Rapid Resolution HT
(2.1 × 50 mm, 1.8 mm column), 30°C; sample (5 µL) was eluted using a binary gradient
(solvent A: 0.1% aq. HCO2H; solvent B: 0.1% HCO2H in CH3CN; 5–100% B [3.5 min], 0.5
mL min-1).
27
ACS Paragon Plus Environment

Page 29 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
General synthetic procedures. Triazines of two classes were used in cyanide mediated
dimerizations, namely 3-(alkylthio)-1,2,4-triazines 3a-3e, and 3-(alkoxy)-1,2,4-triazines 5. The
synthesis of each class is fairly general or has been adopted and applied from a specifically
described procedure in the literature.
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
Synthesis of 3-(alkylthio)-1,2,4-triazines 3a-3e. The synthesis of 3-(alkylthio)-1,2,4-
triazine monomers 3a-3c was achieved through alkylation of thiosemicarbazide with alkyl
halides followed by cyclization with glyoxal. This synthetic methodology was the same as what
has been previously reported by us for the synthesis of 3-(methylthio)-1,2,4-triazine.14 An
alternative approach to synthesizing 3-(alkylthio)-1,2,4-triazines 3d-3e involved a nucleophilic
substitution reaction that was accomplished using 3-(methylsulfonyl)-1,2,4-triazine 4 and
mercaptans.14 3-(Cyclopentylthio)-1,2,4-triazine (3a), 3-((1,2,4-triazin-3-yl)thio)propan-1-ol
(3b) and 3-((2-methoxyethyl)thio)-1,2,4-triazine (3c) were synthesized via the former
procedure while 3-((4-methoxybenzyl)thio)-1,2,4-triazine (3d) and 3-(isobutylthio)-1,2,4-
triazine (3e) were synthesized via the latter procedure. The synthesis of all other 3-(alkylthio)-
1,2,4-triazines has been reported by our group in an earlier publication.14
Synthesis of 3-(alkoxy)-1,2,4-triazines 5. The synthesis of 3-(alkoxy)-1,2,4-triazine
monomers 5 was achieved using the nucleophilic substitution reaction involving magnesium
alkoxides and 3-(methylsulfonyl)-1,2,4-triazine 4 in DMF. The syntheses of Mg-alkoxides
were accomplished by reacting alcohols with MeMgI in THF solutions. The exact procedural
details used in the synthesis of 3-(alkoxy)-1,2,4-triazine 5 have been described in detail in our
earlier publication.14 The synthesis and characterisation of all 3-(alkoxy)-1,2,4-triazines 5 that
were utilised in the synthesis of 3,3’-disubstituted 5,5’-bi(1,2,4-triazine)s discussed in this
study have been reported by our group in the past.14
28
ACS Paragon Plus Environment