A class of 5-nitro-2-furancarboxylamides with potent trypanocidal activity against Trypanosoma brucei in vitro

Article abstract of DOI:10.1021/jm301215e

Recently, the World Health Organization approved the nifurtimox- eflornithine combination therapy for the treatment of human African trypanosomiasis, renewing interest in nitroheterocycle therapies for this and associated diseases. In this study, we have synthesized a series of novel 5-nitro-2-furancarboxylamides that show potent trypanocidal activity, ～1000-fold more potent than nifurtimox against in vitro Trypanosoma brucei with very low cytotoxicity against human HeLa cells. More importantly, the most potent analogue showed very limited cross-resistance to nifurtimox-resistant cells and vice versa. This implies that our novel, relatively easy to synthesize and therefore cheap, 5-nitro-2-furancarboxylamides are targeting a different, but still essential, biochemical process to those targeted by nifurtimox or its metabolites in the parasites. The significant increase in potency (smaller dose probably required) has the potential for greatly reducing unwanted side effects and also reducing the likelihood of drug resistance. Collectively, these findings have important implications for the future therapeutic treatment of African sleeping sickness.

Full text of DOI:10.1021/jm301215e

Subscriber access provided by DUKE UNIV
Article
A class of 5-nitro-2-furancarboxylamides with potent
trypanocidal activity against Trypanosoma brucei in vitro
Linna Zhou, Gavin Stewart, Emeline Rideau, Nicholas J. Westwood, and TERRY SMITH
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/jm301215e • Publication Date (Web): 03 Jan 2013
Downloaded from http://pubs.acs.org on January 5, 2013
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155 Sixteenth Street
N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society. However, no
copyright claim is made to original U.S. Government works, or works produced by employees of any
Commonwealth realm Crown government in the course of their duties.

Subscriber access provided by DUKE UNIV
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155 Sixteenth Street
N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society. However, no
copyright claim is made to original U.S. Government works, or works produced by employees of any
Commonwealth realm Crown government in the course of their duties.

Page 1 of 36
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 36
1
1
2
3
4
A class of 5ꢀnitroꢀ2ꢀfurancarboxylamides with potent trypanocidal
5
6
7
activity against Trypanosoma brucei in vitro
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
Linna Zhou,1,2 Gavin Stewart,1,2 Emeline Rideau,1,2 Nicholas J. Westwood1,2* & Terry K. Smith1,2,3
1School of Chemistry and EaStCHEM, 2Biomedical Science Research Complex, 3School of Biology,
The North Haugh, The University of St. Andrews, Fife, Scotland, U.K., KY16 9ST.
*
*To whom correspondence should be addressed:
Terry K. Smith
Tel: (0)1334ꢀ463412
Fax: (0)1334ꢀ462593
Email: tks1@stꢀandrews.ac.uk
Nicholas J. Westwood
Tel: (0)1334ꢀ463816
Fax: (0)1334ꢀ462593
Email: njw3@stꢀandrews.ac.uk
Running title: Potent nitroꢀfurancarboxylamide analogues
ACS Paragon Plus Environment

Page 3 of 36
Journal of Medicinal Chemistry
2
1
2
3
4
ABSTRACT:
5
6
7
8
Recently the World Health Organization approved the nifurtimoxꢀeflornithine combination therapy for
the treatment of human African trypanosomiasis, renewing interest in nitroꢀheterocycles therapies for
this and associated diseases. In this study we have synthesised a series of novel 5ꢀnitroꢀ2ꢀ
furancarboxylamides that show potent trypanocidal activity, ~1000ꢀfold more potent than nifurtimox
against in vitro Trypanosoma brucei with very low cytotoxicity against human HeLa cells. More
importantly, the most potent analog showed very limited crossꢀresistance to nifurtimoxꢀresistant cells
and viceꢀversa. This implies that our novel relatively ease to synthesise and therefore cheap, 5ꢀnitroꢀ2ꢀ
furancarboxylamides are targeting a different, but still essential, biochemical process to those targeted
by nifurtimox or its metabolites in the parasites. The significant increase in potency (smaller dose
probably required) has the potential for greatly reducing unwanted sideꢀeffects and also reducing the
likelihood of drugꢀresistance. Collectively these findings have important implications for the future
therapeutic treatment of African sleeping sickness.
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 4 of 36
3
1
2
3
4
INTRODUCTION
5
Infectious diseases caused by parasitic protozoa affect ca. 15% of the global population and more than
65% of the population in the Third and developing world, yet current drug therapies for protozoan
infections are woefully inadequate. As protozoan infections take their toll predominantly in the
developing world, market forces are insufficient to promote the development of novel antiꢀprotozoan
drugs. In 2000, only ca. 0.1% of global investment in health research was spent on drug discovery for
tropical diseases.
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
One such neglected parasitic disease is Human African Trypanosomiasis (HAT) or African sleeping
sickness, which is caused by the protozoan parasite Trypanosoma brucei. The World Health
Organization (WHO) estimates that HAT constitutes a serious health risk to 60 million people in subꢀ
Saharan Africa, The WHO also estimates that there are <30 000 new cases per year in subꢀSaharan
Africa and at present an annual death toll of ~8,000.1
The related disease in cattle, cattle trypanosomiasis or Nagana, also represents a major health concern
due to its devastating economic (estimated by the WHO to cause an annual economic loss of ∼US$ 4
billion.), social and nutritional impact on African families. As such, the total burden of trypanosomiasis
translates into 1,598,000 DisabilityꢀAdjusted Life Years (DALY). This is on a par with big killers such
as tuberculosis and malaria.2, 3
Treatment of HAT is solely dependent upon a repertoire of four drugs: suramin (1), pentamidine (2),
melarsoprol (3) and eflornithine (4) (Figure 1). These therapies are often toxic, difficult to administer
and increasingly have an acquired drug resistance, highlighting the urgent need for new, more effective
drug therapies. Developed before the 1950s, suramin (1) and melarsoprol (3) are used for chemotherapy
of the early stage of the disease (T. b. rhodesiense), as is pentamidine (2) (T. b. gambiense). The
arsenical melarsoprol (3) is extremely toxic, with death in ~ 6% of cases and treatment failure rates of as
high as 30% in certain areas. Treatment of the second stage of the disease, where the parasites cross the
ACS Paragon Plus Environment

Page 5 of 36
Journal of Medicinal Chemistry
4
1
2
3
4
5
6
bloodꢀbrainꢀbarrier and invade the central nervous system, is limited to melarsoprol (3) and eflornithine
(4), the ornithine decarboxylase inhibitor (difluoromethylornithine).4, 5
7
8
9
Nifurtimox (5) (Figure 1), is often used to treat Chagas' disease, caused by Trypanosoma cruzi, and has
been used as a monotherapy for melarsoprolꢀrefractory HAT on compassionate grounds, despite its low
efficacy and severe toxicity. The recent introduction of a nifurtimoxꢀeflornithine combination therapy
(NECT) by WHO in the Model Lists of Essential Medicines has been seen as highly advantageous in
terms of cost, logistics, and human resources in areas of poverty. Patients given NECT, consisting of
oral nifurtimox over 10 days with eflornithine (4) infusions for 7 days, were found to fair just as well as
those given the eflornithine monotherapy, with cure rates of around 97%.2
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 use of NECT has renewed general interest in the use of nitroheterocyclic compounds to treat a wide
6ꢀ8
range of infectious disease, including tuberculosis and hepatitis C.
This increased awareness of the
potential of nitroheterocyclic compounds has led to the reinvestigation of fexinidazole (6) (Figure 1),
which is presently in clinical trials against both earlyꢀ and lateꢀstage African sleeping sickness.6ꢀ8
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 36
5
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. Structures of clinic drugs 1ꢀ5 for sleeping sickness and clinic phase I drug Fexinidazole (6).
Alarmingly, it has also been shown that Trypanosoma brucei nifurtimoxꢀresistant cells show crossꢀ
resistance to other nitroꢀcontaining drugs, including fexinidazole, that are currently in clinical trials.9, 10
.
Several different types of studies on nitrohererocyclics have shed some light on this observed drugꢀ
resistance. Genomeꢀwide RNAi screens of nifurtimox and benznidazole (another trypanocidal
nitroheterocyclic drug) resistant Trypanosoma brucei, have identified that a decrease in nitroreductase
activty is linked to nifurtimoxꢀresistance.9, 10 This was confirmed by showing that genetic deletion of
one allele of the T. brucei’s nitroreductase, led to nifurtimoxꢀresistance. Likewise the overexpression of
this nitroreductase, but not the alternative proposed prostaglandin F2α synthase and cytochrome P450
reductase, in Trypanosoma brucei resulted in increased susceptibility to nifurtimox.11 Collectively, it is
clear that this Type I T. brucei nitroreductase, a NAD(P)H dependent, flavin binding protein is involved
in the reductive activation of nifurtimox. The nitroreductases mediate a series of twoꢀelectron
reductions of the nitro group to a nitroso intermediate, then to a hydroxylamine and eventually to the
ACS Paragon Plus Environment

Page 7 of 36
Journal of Medicinal Chemistry
6
1
2
3
4
5
6
7
8
corresponding amine. However, the true function of the T. brucei mitochondrial nitroreductase is
unknown, but its essentiality may explain why only a single allele deletion/mutation is associated with
nifurtimoxꢀresistance. 10, 11
9
As well as the development of nifurtimoxꢀresistance and its crossꢀresistance to other nitroꢀcontaining
drugs, sideꢀeffects are also a major problem for nitroꢀcontaining drugs. For example, the adverse offꢀ
target sideꢀeffects with nifurtimox lead to treatment cessation in over 30% of patients with Chagas’
disease.12 The most common sideꢀeffects are anorexia, loss of weight, psychic alterations, excitability or
sleepiness, digestive manifestations such as nausea or vomiting and occasionally intestinal colic and
diarrhoea.13 Various human enzymes have been identified to be capable of 5ꢀnitrofuran reduction in
vitro, in cells or tissues.14ꢀ16 Recently we have elucidated in collaboration with the Patton laboratory a
possible role of aldehyde dehydrogenase 2 (ALDH2) in the toxicity caused by 5ꢀnitrofuran containing
drugs such as nifurtimox.17
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
Whilst considerable effort has gone into the synthesis of nitroꢀaromatic containing compounds over the
years18ꢀ23, there remains several unexplored areas of chemical space. For example, in the context of
nifurtimox most of the efforts have been focused on changing either the furan ring to other heterocycles
(such as imidazole) or changing the substituent in the hydrazone motif,18ꢀ23 with limited increases in
activity compared to nifurtimox. Recently, two new types of typanocidal compounds,
nitrobenzylphosphoramide mustards (for example see compound 7, Figure 2)24 and aziridinyl
nitrobenzamides have been developed to target specifically T. brucei type I nitroreductase, the best
having an EC50 ~ 1 ꢁM against both T. brucei and Trypanosoma cruzi.25, 26 Several 5ꢀnitroꢀ2ꢀ
furancarboxylamide containing compounds have been reported to have potent antituberculosis and
antimicrobial activity,27ꢀ30 however, very few 5ꢀnitroꢀ2ꢀfurancarboxylamides are reported to have
typanocidal activity, primarily low activity against T. cruzi.31, 32 A nifurtimox analog (8) (Figure 2)
showed trace activity against T. cruzi in a mouse model.33 The best 5ꢀnitroꢀ2ꢀfurancarboxylamides
identified thus far, 9 and 10, have EC50 values of 42 nM and 262 nM respectively against T. cruzi
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 36
7
1
2
3
4
5
6
7
8
amastigotes (Figure 2) but have not been tested against T. brucei.34 Several 5ꢀnitroꢀ2ꢀ
furancarbohydrazides possess activity against T. brucei with the best, compound 11 having an EC50 of
0.13 ꢁM, but low selectivity against human cells. 35
9
In this study, we report that our recently discovered nitrofuran NFN1 (12a)17 also shows promising
activity against T. brucei in vitro. We have therefore synthesised a series of 5ꢀnitroꢀ2ꢀ
furancarboxylamide analogs of 12a. A lack of crossꢀresistance with nifurtimox and significantly
increased (two to three orders of magnitude) trypanocidal activity has important implications for the
future therapeutic use of these nitrofuran containing compounds in the treatment of HAT, instead of, or
in combination with nifurtimox.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Page 9 of 36
Journal of Medicinal Chemistry
8
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
Figure 2. Structures of some reported bioactive nitroaromatic compounds 7ꢀ12a.
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 10 of 36
9
1
2
3
4
Results and Discussion
5
6
7
Chemistry
8
9
5ꢀNitrofurans 12aꢀk were prepared by acylation of the corresponding 2ꢀaminoꢀthiophenes 13aꢀk with 5ꢀ
nitroꢀfurancarboxylic acid chloride 14a (Scheme 1, Table 1 for substituents). The required 2ꢀaminoꢀ
thiophenes 13aꢀh were prepared using the Gewald multicomponent reaction with various combinations
of the 2ꢀcyanoacetyl esters or amides (15aꢀ15g) with aldehydes (16a or 16b) and elemental sulphur (see
Table S1 in ESI for details).36, 37 13h was Oꢀsilylꢀprotected before coupling to 14a. This reaction gave
12h directly as a result of in situ removal of the silylꢀprotecting group. Thiophene 13i was prepared by
hydrolysis of 13a and 13j was prepared by catalytic amination of 2ꢀiodothiophene using CuI and Lꢀ
proline.38 13k and 13l (Table 2) were commercially available. Reaction of 13a (R2 = Et, R3 = CO2Et)
with 14b gave control compound 18a. An analogous reaction of 14b with aniline gave 18b (Table 2).
Furan ester 19 was commercially available.
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
OH
X
R1
a
N
N
O
O
15
15d X = NH; R1 = Etb
17
15a X = O; R1 = Eta 15e X = NH; R1 = nBub
15b X = O; R1 = Mea 15f X = NH; R1 = Ha
15c X = O; R1 = nBua 15g X = NH; R1 = CH2CH2OHb
O
16a R2 = CH2CH3
16b R2 = (CH2)3CH3
b
S8 +
R2
H
R4
O
O
R2
R4
R2
S
O
S
COR5
c
H2N
+
HN
R3
13a-kc
R3
14a R4 = NO2, R5 = Cl
14b R4 = H, R5 = Cl
19 R4 = NO2, R5 = OCH3
c
12a-k R4 = NO2
18a R2 = Et, R3 = CO2Et, R4 = H
R4
O
O
d
NHR6
22a-
qd
ACS Paragon Plus Environment

Page 11 of 36
Journal of Medicinal Chemistry
10
1
2
3
4
5
6
7
8
Scheme 1. Synthetic route to 5ꢀnitroꢀ2ꢀfurancarboxylamide analogs 12aꢀk: Reagents and conditions: (a) i) (COCl)2,
DCM/DMF, RT, 2 hrs; ii) R1NH2, Et3N, DCM, 52% for 15d; 49% (15e); 37% (15g); (b) S8, 16a or 16b, Et2NH, DMF, RT,
o/n, 66% for 13a; 59% (13b); 76% (13c); 53% (13d); 49% (13e); 46% (13f); 48% (13g); 46% (13h); for synthesis of 13iꢀj
see ESI; 13k and 13l were commercially available; (c) 13aꢀk, Et3N, DCM, RT with 14a, 68% for 12a; 62% (12b); 65%
(12c); 58% (12d); 57% (12e); 57% (12f); 54% (12g); 47% (12h); 25% (12i); 41% (12j); 51% (12k); with 14b, 56% for 18a;
(d) RNH2, Et3N, DCM, RT with 14a, 65% for 22a; 61% (22b); 72% (22c); 67% (22d); 37% (22e); 50% (22f); 45% (22g);
41% (22h); 66% (22i); 39% (22j); 51% (22k); 73% (22l); 51% (22m); 68% (22n); 67% (22o); 69% (22p); 48% (22q); with
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
a
b
14b 73% for 18b (see Table 2 for structure of 18b) 15aꢀc and 15f were commercially available; 15d, 15e and15g were
prepared from 17; c for R2 and R3 substituents see Table 1; c for R6 substituents see Tables 2 and 3.
Analogs 20aꢀc and 21 (Table 2) based on 12g but containing alternative nitroaromatic rings were
prepared by coupling the corresponding acid chlorides with 13g (see ESI for details). Similarly, reaction
of 14a with a range of amines and anilines led to the synthesis of 22aꢀ22q (Scheme 1). Demethylation
of 22o to give 22r was followed by selective Oꢀallylation to give 22s (Scheme 2), the structure of which
was confirmed by Xꢀray crystallographic analysis (data not shown). When a longer reaction time and an
excess amount of allyl bromide was used in the Oꢀallylation of 22r, the diallylated analogue 22t was
formed (Scheme 2). MonoꢀOꢀallylation of 22p gave 22u (Scheme 2). Methylation of 22q to give 22v
was achieved using MeI under basic conditions in moderate yield (Scheme 2). Conversion of 5ꢀ
nitrofuran aldehyde 23 to imines 24a or 24b enabled the synthesis of the corresponding amines 25a and
25b using standard reductive amination conditions (Scheme 2).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 36
11
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
Scheme 2. Synthetic routes to Series 2 and Series 3 nitrofurancarboxylamides: Reagents and conditions: (a) pyridine
hydrochloride, 160 oC, MW, 5 mins, 52%; (b) 22r or 22p, allyl bromide (2 or 10 equivs.), K2CO3, acetone, RT, 2 hrs for 22s
(75%), 24h for 22t (55%); 24h for 22u (72%); 22q, MeI, K2CO3, acetone, 3h, for 22v (51%); (c) aniline or 13g, DCM, RT,
89% for 24a, 63% for 24b. (d) NaBH4, DCM, RT, 79% for 25a, 87% for 25b.
Structure activity relationships
In comparison to nifurtimox, NFN1 was two orders of magnitude more potent with an EC50 of 31.3 ±
3.1 nM (Table 1, c.f. entries 1 and 2). The nitro group in NFN1 was found to be essential with no
activity being observed up to the solubility limit of 18a (entry 3). Removal, or worse, elongation of the
alkyl substituent in NFN1 led to reduced activity (c.f. entry 2 and entries 4 and 5) and whilst removal of
the ester group initially appeared to be tolerated in the R2 = H series (c.f. entries 4 and 6), it was found
that the methyl and nꢀbutyl ester analogs (12c and 2d) of NFN1 had similar activity to NFN1 (c.f. entry
2 and entries 7 and 8). Conversion of the ester groups in NFN1 and 12d to the corresponding amides in
12e and 12f led to a significant loss in activity (c.f. entries 2 and 8 with entries 9 and 10 respectively).
This observation, coupled with the decrease in activity associated with amide 12h (entry 11), may be
explained by the fact that hydrolysis of the ester groups in 12aꢀd and 12k in the parasite gives the
ACS Paragon Plus Environment

Page 13 of 36
Journal of Medicinal Chemistry
12
1
2
3
4
5
6
7
8
9
corresponding carboxylic acidꢀcontaining bioactive metabolites, something that is unlikely to occur for
the amides 12e and 12f. However, acid 12i showed much less activity with an EC50 of 959.0 ± 33.9 nM
(entry 12), which may be due to a lack of cell permeability. The primary amide 12g was found to be the
most active thiopheneꢀcontaining analog with an EC50 of 17.3 ± 2.4 nM (entry 13). Furthermore,
analogs 12aꢀk showed no significant toxicity against the human HeLa cell line even at a concentration
above 20 ꢁM. The highly selective toxicity of these analogs against the parasite suggests these
nitrofuran analogs have real potential for therapeutic applications.
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 1: SAR study of analogs 12aꢀ12k.
T. brucei
EC50
HeLa
EC50
(ꢁM)
Selectivity
Index
Entry
Compound
R2
R3
R4
(nM)
a
Nifurtimox (5)
1
ꢀ a
ꢀ a
ꢀ a
2400 ± 100
91.2±11.3
~38
2
3
NFN1 (12a)b
18a
Et
Et
H
CO2Et
CO2Et
CO2Et
CO2Et
H
NO2
H
31.3±3.1
>20000
>20
>20
>20
>20
>20
>20
>20
>20
>20
>20
>20
>640
N/Ac
>278
>160
>322
>950
>625
>136
>60
4
NO2
NO2
NO2
NO2
NO2
NO2
NO2
NO2
NO2
72.3 ± 6.1
125.3±6.0
62.6 ± 2.7
21.0±1.2
12k
5
nBu
H
12b
6
12j
7
Et
Et
Et
Et
Et
Et
CO2Me
12c
n
8
CO2 Bu
32.6±0.7
12d
9
CONHEt
CONHnBu
147.0±10.1
336.3±18.8
79.6 ± 6.7
959.0 ± 33.9
12e
10
11
12
12f
CONH(CH2)2OH
CO2H
>253
N/Ac
12h
12i
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 36
13
1
2
3
13
Et
CONH2
NO2
17.3±2.4
>20
>1155
12g
4
5
a for structure see Figure 1; b see reference17; c not applicable as essential no activity against T .brucei was observed for 12i.
6
7
8
9
Encouraged by the potent activity of 12g (Table 1), it was decided to prepare more analogs based on
this structure. Fragments 19 and 13g along with 13l were shown to be inactive against T. brucei,
suggesting that both the 5ꢀnitrofuran and thiophene fragments are required (Table 2, c.f. entry 1 and
entries 2ꢀ4). Changing the nitrofuran motif in 12g to either a nitrophenyl or nitropyrazole ring also lead
to a 100ꢀfold or more drop in activity (c.f. entry 1 and entries 5ꢀ8). Interestingly, the position of the
nitro group in the nitrophenyl ring had an influence as 20b, an analog with the nitro group in the metaꢀ
position was more activity than either 20a (orthoꢀNO2) or 20c (paraꢀNO2). Studies next focused on
replacing the thiophene motif in 12aꢀk. By incorporating a cyclohexyl ring to give 22a, a 10ꢀfold
decrease in activity was observed compared to 12g (entry 9), whilst the phenyl (22b) and benzyl (22c)
analogs were essentially equipotent with 12g (c.f. entry 1 and entries 10 and 11). Incorporation of a
nitrogen atom into the 2ꢀposition of the phenyl ring in 22c to give 22d led to a 10ꢀfold drop in activity
(entry 12). The nitrofuran group in 22b was shown to be essential (c.f. entries 10 and 13) and the amide
linkage in 22b and 12g was also shown to be important, as the amine analogs 25a and 25b showed a
dramatic decrease in activity (entries 14 and 15).
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: SAR study of Series 2 analogues
T. brucei
EC50
HeLa
EC50
(ꢁM)
Selectivity
Index
Structure
Substituent
Entry
Compound
(ꢁM)
O2N
O
O
S
HN
ꢀ
ꢀ
0.0173 ± 0.0024
67.0 ± 2.6
>20
>20
>1155
N/Aa
1
2
12g
19
O
NH2
O2N
O
O
O
ACS Paragon Plus Environment

Page 15 of 36
Journal of Medicinal Chemistry
14
1
2
3
4
5
6
7
8
9
N/Aa
N/Aa
N/Aa
>8
R1
R2
S
R1=Et, R2= H
R1=H, R2= Me
orthoꢀPh
32.6 ± 1.9
>20
>20
>20
>20
3
4
5
6
13g
13l
H2N
O
65.1 ± 3.2
26.1 ± 2.5
2.6 ± 0.2
NH2
20a
20b
O
Ar
O2N
S
metaꢀPh
10
HN
11
12
13
14
15
16
17
O
paraꢀPh
Ortho-Pyrazole
Cyclohexyl
Ph
>20
>20
>20
>20
>20
>20
>20
>20
N/Aa
N/Aa
>86
7
8
20c
21
NH2
>300
0.2310 ± 0.0313
0.0281 ± 0.0015
0.0273 ± 0.0018
0.2910 ± 0.0315
>20
9
22a
22b
22c
22d
18b
>710
>732
>68
10
11
12
13
18
O2N
O
O
19
20
21
22
23
24
25
26
27
Bn
Cyc
HN
CH2ꢀ2ꢀPy
No NO2, Ph
N/Aa
O2N
O
ꢀ
ꢀ
113.3 ± 14.7
5.3 ± 0.8
>20
>20
N/Aa
14
15
25a
25b
NH
28
O2N
O
29
30
31
32
33
S
HN
>4
O
NH2
a not applicable as no activity or low activity against T.brucei was observed for 18b and 25a.
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
Based on the potent activity of 22b, the effect of incorporating additional substituents into the phenyl
ring was investigated (Table 3). Substitution at the paraꢀ(R3) and metaꢀ(R4)ꢀpositions with either
electronꢀwithdrawing (entries 3 and 5ꢀ7) or electronꢀdonating groups (entries 4 and 8) proved
detrimental with the exception of the incorporation of the metaꢀCF3 group in 22e (EC50 of 22e = 4.69 ±
0.14 nM, entry 2). Interestingly, incorporation of a metaꢀCF3 into the phenyl ring of the benzyl analog
22c (Table 2) gave analog 22l which had an EC50 = 17.4 ± 3.0 nM, suggesting that the use of a phenyl
substituent (as in 22e) may be preferred to the benzyl substituent in 22l. Combining the metaꢀCF3
substituent with either an additional metaꢀmethoxyꢀ or a second metaꢀCF3 substituent led to a decrease
in activity (entries 9 and 11), whereas the metaꢀCF3 group was able to override the effect of a paraꢀ
methoxy substituent (c.f. entries 8 and 12). Incorporation of a paraꢀhydroxy in both the metaꢀCF3 and
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 36
15
1
2
3
4
5
6
7
8
9
unsubstituted series led to a significant reduction in activity (entries 13 and 14), but extending the paraꢀ
alkoxy chain from methoxy in 22o to allyloxy led to the most active compound prepared, analog 22s,
which had an EC50 against T. brucei of 2.4 ± 0.3 nM (entry 15). The positive role of the paraꢀallyloxy
substituent was also observed in the unsubstituted series (c.f. entries 8 and 16). The incorporation of an
additional allyloxyꢀsubstituent on the amide nitrogen in 22s led to a significant loss in activity (c.f.
entries 15 and 17).
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
Interestingly, the lack of significant biological activity displayed by the 2ꢀtrifluoromethylphenol 22r,
e.g. EC50 value of 2.2 ± 0.1 ꢀM versus 2.4 ± 0.3 nM for 22s or 7.8 ± 0.3 nM for 22o – might be due to
the high instability of 22r in the tested conditions, as the result of fluorine elimination generating a
quinone methide, as previously reported.39, 40 Masking the phenol as allyl or methyl ethers, e.g. like in
22s or 22o, respectively, restored the trypanocidal activity against T. brucei.
Finally, the special role played by the metaꢀCF3 substituent, was further highlighted by a more than 5ꢀ
fold drop in activity observed on its replacement by a metaꢀCH3 substituent (c.f. entries 12 and 18). As
was the case for analogs 12aꢀ12k, all the analogs shown in Tables 2 and 3 show low toxicity against the
human HeLa cell line.
Table 3: SAR study of series 3 analogues
R2
R4
O2N
O
N
O
R3
R1
T.brucei
EC50
HeLa
EC50
(ꢁM)
Selectivity
Index
Entry
Compound
R1
R2
R3
R4
(nM)
ACS Paragon Plus Environment

Page 17 of 36
Journal of Medicinal Chemistry
16
1
2
3
4
5
6
7
8
9
1
2
H
H
H
CF3
Br
H
H
H
H
28.1 ± 1.5
4.69 ± 0.14
103.8 ± 7.7
58.6 ± 2.9
126.3 ± 10.2
411.9 ± 41.6
680.0 ± 6.1
76.5 ± 5.5
26.6 ± 2.6
17.4 ± 3.0
128.0 ± 5.1
7.8 ± 0.3
>20
>20
>20
>20
>20
>20
>20
>20
>20
>20
>10
>20
>20
>20
>20
>20
>20
>20
>710
>4454
>192
>345
>158
>48
22b
22e
22f
3
H
H
H
4
H
OCH3
H
H
H
22g
22h
22i
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
5
H
CF3
H
6
H
H
Cl
H
7
H
H
COOCH3
OCH3
H
H
>29
22j
8
H
H
H
>261
>752
>1149
>78
22k
22m
22l a
22n
22o
22r
22p
22s
22u
22t
9
H
CF3
ꢀ
OCH3
ꢀ
10
11
12
13
14
15
16
17
18
ꢀ
ꢀ
H
CF3
CF3
CF3
H
H
CF3
H
H
OCH3
OH
>2565
>9
H
H
2200 ± 100
394.4 ± 21.0
2.4 ± 0.3
H
OH
H
>50
H
CF3
H
Oꢀallyl
Oꢀallyl
Oꢀallyl
OCH3
H
>8330
>258
>9
H
H
7.73 ± 0.72
2200 ± 100
38.0 ± 3.7
Oꢀallyl
H
CF3
CH3
H
H
>525
22v
a structure of 5ꢀNitroꢀNꢀ(3ꢀ(trifluoromethyl)benzyl)furanꢀ2ꢀcarboxamide (22l)
O2N
O
O
HN
CF3
CrossꢀResistance Studies
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The striking increase in potency of the 5ꢀnitroꢀ2ꢀfurancarboxylamide analogs compared to nifurtimox
against bloodstream T. brucei, suggests they have a different or possibly additional mode of action to
that of nifurtimox, i.e. in addition to the activation by the T. brucei nitroreductase. In order to
investigate this possibility it was decided to generate drug resistant cell lines to nifurtimox and the most
potent of the 5ꢀnitroꢀ2ꢀfurancarboxylamide analogs, 22s (EC50 = 2.4 ± 0.3 nM). This would allow
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 36
17
1
2
3
4
5
6
subsequent crossꢀresistance studies to be conducted. Previous lab generated nifurtimox ꢀresistant strains
have shown that deletion or mutation of one of the alleles of the nitroreductase confers resistance 9, 10
.
7
8
9
As with these previous studies nifurtimox ꢀresistant parasites were generated easily by culturing
bloodstream T. brucei in the continuous presence of nifurtimox. A stepwise increase in the
concentrations of nifurtimox, initially starting at 1.5 ꢁM, increasing to 2.5, 4, 5, 7.5, 10, 15, 20, 25, 30,
40, 45, 50 ꢁM on days 2, 5, 9, 14, 23, 39, 55, 65, 76, 90, 109, 120 respectively was used (Fig 3A). At
143 days these nifurtimox ꢀresistant parasites were cloned by serial dilution, while maintaining the 50
ꢁM nifurtimox.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Compound 22sꢀresistant parasites were initially harder to generate. Selection studies started at 1 nM of
22s, increasing to 2, 3, 5, 7.5, 10, 15, 20, 25, 30, 35 40, 50 ꢁM on days 2, 5, 9, 26, 36, 47, 58, 70, 76,
103, 130, 148 respectively (Fig. 3B). At 160 days the 22sꢀresistant parasites were cloned by serial
dilution, while maintaining a 50 nM concentration of 22s. Both of the drugꢀresistant cellꢀlines were
stable for 30 days in the absence of their respective drugs. The nifurtimoxꢀresistant parasites had a
doubling time of ~11.5 ± 0. 6 h compared to wildꢀtype ~7.8 ± 0.4 h, while 22sꢀresistant parasites had a
doubling time of ~13.3 ± 1.1 h.
ACS Paragon Plus Environment

Page 19 of 36
Journal of Medicinal Chemistry
18
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
Figure. 3. Generation of nifurtimox and 22s resistance cellꢀlines: Schematic representation of the
generation of a (A) nifurtimoxꢀresistant and (B) 22sꢀresistant cell lines in T. brucei. Each point
55
56
57
58
59
60
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 36
19
1
2
3
4
represents the concentration at that time point when cells were checked and divided if required.
5
6
7
8
9
The sensitivities of the cloned nifurtimoxꢀresistant and 22sꢀresistant parasites to nifurtimox, 22s and the
diamidine pentamidine (2) were determined and compared to wildꢀtype cells. The nifurtimoxꢀresistant
cells did not greatly alter (~2ꢀfold increase) the sensitivity to pentamidine (2) compared to wildꢀtype
cells (Table 4), in accordance with previous lab generated nifurtimoxꢀresistant cellꢀlines.9,10 22sꢀ
Resistant parasites also showed very little alteration in the EC50 (1.1 ± 0.1 nM) of pentamidine (2),
compared to wildꢀtype cells (Table 4).
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 nifurtimoxꢀresistant cells were found to be 9ꢀ10ꢀfold less sensitive to nifurtimox than wildꢀtype
cells, with EC50s of 20.9 ± 1.7 and 2.1 ± 0.2 ꢁM, respectively (Table 4), in accordance with previous
findings.9 22sꢀResistant cells were found to be ~14ꢀfold less sensitive to 22s than wildꢀtype cells, with
EC50s of 29.3 ± 2.0 and 2.4 ± 0 .3 nM, respectively (Table 4). Even at this level of resistance 22s is still
~70 times more potent than nifurtimox in wildꢀtype T. brucei.
To investigate crossꢀresistance between nifurtimox and 22s, the EC50s of the drugꢀresistant cell lines
versus the alternative nitrofuran containing compounds were determined (Table 4). The results showed
that nifurtimox and 22s showed a low level of crossꢀresistance. Nifurtimoxꢀresistant cells showed a ~3ꢀ
fold increase in the EC50 of 22s, whilst nifurtimox also showed ~3ꢀfold increase in the EC50 against 22sꢀ
resistant cells. These findings imply that these two nitrofuran compounds, nifurtimox and 22s, are
trypanocidal as a result of targeting primarily different biochemical process within the parasites.
However, the low level of crossꢀresistance suggests there may be some minor overlap in mode of
action. This issue is discussed in more detail later. These findings have important implications for the
therapeutic use of this new generation of nitrofuran compounds as part of novel combination therapies
or more importantly potentially replacing nifurtimox in a clinical setting.
Table 4: Crossꢀresistance studies, EC50s of wildꢀtype and cloned drugꢀresistant cell lines.
ACS Paragon Plus Environment

Page 21 of 36
Journal of Medicinal Chemistry
20
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
Cell Type
WildꢀType
Nifurtimox
Resistant
Nifurtimox (ꢁM)a Compound 22s (nM) Pentamidine (nM)b
2.1 ± 0.2
2.4 ± 0.3
7.8 ± 0.4
1.0 ± 0.1
2.4 ± 0.2
20.9 ± 1.7
Compound 22s
Resistant
7.3 ± 0.5
29.3 ± 2.0
1.1 ± 0.1
a EC50 literature value of nifurtimox in wild type T. brucei 2.4 ± 0.1 ꢁM, and in nifurtimox resistant
cells 20.1 ± 0.9 ꢁM. 9
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
b EC50 literature value of pentamidine (2) in wild type T. brucei 0.95 ± 0.02 nM, and in nifurtimox
resistant cells 2.6 ± 0.1 nM. 9
Figure. 4. Studies of possible synergy between nifurtimox (5) and 22s in T. brucei.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 36
21
1
2
3
4
5
6
7
8
9
As nifurtimox and 22s apparently act through primarily different modes of action, this suggests that
their combined action against the parasite may be synergistic. As such a series of EC50s were
determined for 22s in the presence of various concentrations of nifurtimox (Figure 4). At low
concentrations of nifurtimox (<700 nM), synergy can be observed as the EC50s are below the straight
diagonal line that represents what would be expected for an additive trypanocidal effect of the two
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
compounds. Above concentrations of 700 nM nifurtimox, there is a distinct lack of synergy, if anything
competition occurs between the two inhibitors. This taken together with the observed low level of crossꢀ
resistance (Table 4) may indicate that nifurtimox at high concentrations is competing with the same
biochemical process which is normally targeted by 22s. This further highlights the possible multiꢀtarget
nature for nifurtimox that has been suggested previously.9ꢀ14 This type of information is relevant for the
potential pharmacological applications of these novel nitroꢀfurancarboxylamides as a combinational
therapy, would likely lead to the use of lower nifurtimox doses.
Table 5: EC50s of typical series 1 and series 2 analogs against T. b. rhodesiense
T. brucei
Entry
Compound
rhodesiense
EC50 (nM)
1
2
3
4
5
6
7
8
nifurtimox (5)a 1708 ± 197
NFN1 (12a)
44.2 ± 4.5
201.8 ± 18.7
23.1 ± 2.9
9.4 ± 1.8
12e
12g
22e
22h
159.0 ± 12.1
2.9 ± 0.4
22s
Melarsoprol (3)b
7.8 ± 0.9
ACS Paragon Plus Environment

Page 23 of 36
Journal of Medicinal Chemistry
22
1
2
3
4
5
6
43
a EC50 literature value of nifurtimox in T. b. rhodesiense 1.5 ꢁM
.
7
8
b EC50 literature value of melarsoprol against T. b. rhodesiense 6 nM . 43
9
10
11
It was decided to test some of the 5ꢀnitroꢀ2ꢀfurancarboxylamides from this study against cultured T. b.
rhodesiense, one of the humanꢀinfective subspecies (Table 5). The potency of a selection of analogs
against T. b. rhodesiense were similar to those observed for T. b. brucei, suggesting that these novel
nitroꢀfurans could be considered lead candidates for the next step towards a novel treatment for HAT, as
well as the related animal disease, Nagana.
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
Conclusions
In this work, we designed and synthesised a series of nitrofuranꢀcontaining analogs that were tested for
trypanocidal and cytotoxicity activity against cultured bloodstream form T. b. brucei and human HeLa
cells respectively. The analog 22s showed an Ec50 of 2.4 ± 0.3 nM, three orders of magnitude more
potent than nifurtimox with a selectivity index >8000. SAR studies showed that all fragments of the
compounds were required for activity. In addition, both the nitrofuran and amide functional groups and
the amide NH were necessary. The thiophene ring in the starting analog 12a can be replaced by a range
of other substituents. However, in the thiophene series, an ester group was required and is likely
converted to the corresponding carboxylic acid in the parasite. Several of our analogs were also tested
against T. b. rhodesiense showing analogous activity to that in T. b. brucei. This exciting result
demonstrates that this series of analogs is worthy of further study as a potential therapy for HAT.
Importantly the crossꢀresistance studies showed that the 5ꢀnitroꢀ2ꢀfurancarboxylamide analogs in this
study have low levels of resistance to nifurtimoxꢀresistant cells and vice versa, whilst showing no
resistance to pentamidine (2). These results indicate that the trypanocidal modeꢀofꢀaction of the 5ꢀnitroꢀ
2ꢀfurancarboxylamide analogs in this study does not rely upon nitroreductase activation unlike
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 36
23
1
2
3
4
5
6
7
8
9
nifurtimox and other nitroheterocyclic compounds that are presently in clinical trials against HAT.
However, the lack of expected synergy at high concentrations of nifurtimox suggests it may be
interacting with the novel mode of action which is primarily targeted by our novel 5ꢀnitroꢀ2ꢀ
furancarboxylamide analogs.
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
Our data suggests that these new 5ꢀnitroꢀ2ꢀfurancarboxylamide analogs have a potential therapeutic
implication as a single reagent in regards to potency and selectivity. Also combination therapies can be
considered due to their low crossꢀresistance to other trypanocidal drugs.
This idea is supported by our results that show there is a synergistic relationship between nifurtimox
and 22s at concentrations of nifurtimox of <700 nM (Figure 4). Whilst there remain many more
challenges to the development of a HAT therapeutic than have not been addressed here (see Table S6
for a detailed discussion of drugꢀlikeness parameters associated with our analogs), we believe that the
novel mechanistic and high in vitro potency of the antiꢀtrypanosomal compounds presented here renders
them worthy of further study in the drug discovery context.
ACS Paragon Plus Environment

Page 25 of 36
Journal of Medicinal Chemistry
24
1
2
3
Experimental
4
5
6
7
BIOLOGY
8
9
Materials
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
All materials unless stated were purchased from either Sigma/Aldrich, or Invitrogen.
Trypanocidal Studies
The trypanocidal activity over a 72 hour period was determined using the Alamar BlueTM viability, as
described previously41. Nifurtimox, pentamidine, melarsoprol and various novel nitrofuran analogues
were tested against Trypanosoma brucei brucei bloodstreamꢀform (strain 427) or drugꢀresistant strains
were cultured at 37°C in HMI9 medium supplemented with 10% fetal calf serum and 2.5 ꢁg ml–1 G418
as described previously42. T. brucei rhodesiense (strain Z310) were cultured in a similar manner, but in
the absence of G418. For the synergistic experiments, various concentrations of nifurtimox (0, 200, 400,
600, 800, 1000, 1500, 2000, 2400 nM) were used in conjunction with a serial dilution of 22s.
Experiments were conducted in replicates of four; the data was fitted using GraFit software to obtain
EC50 ± standard deviations and slope factors.
Cytotoxicity Studies
Cytotoxic affects against HeLa cells were determined in a similar manner. Briefly, the cells were
cultured in DMEM supplemented with 10% fetal calf serum and 2 mM LꢀGlutamine. Cells were plated
at initial cell concentration of 2 x 104 cells/well and incubated with the compounds for ~65 hours prior
to addition of Alamar BlueTM solution for a further 5 hours.
Generation of drugꢀresistant T. brucei cell lines.
Drugꢀresistant T. brucei cell lines were generated by subꢀculturing bloodstream trypanosomes in the
continuous presence of either nifurtimox or compound 22s. Parasites were exposed to stepwiseꢀ
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 36
25
1
2
3
4
5
6
7
8
9
increased concentrations of drug, starting at appropriate subꢀlethal concentrations, until they were
routinely growing between 20ꢀ25 times their respective original ED50s. After 150ꢀ160 days in culture,
drugꢀresistant parasites were cloned by limiting dilution, and used for further study. Cell doubling times
were determined in replicates of 4, over a 96 h period.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Page 27 of 36
Journal of Medicinal Chemistry
26
1
2
3
Chemistry
4
5
6
7
8
9
General. Chemicals and reagents were obtained from either Aldrich or AlfaꢀAesar, except nifurtimox
(5) from Bayer Argentina. All reactions involving moisture sensitive reagents were performed in oven
dried glassware under a positive pressure of argon. Dichloromethane (DCM) was obtained dry from a
solvent purification system (MBraun, SPSꢀ800). Melting points were recorded in open capillaries using
an Electrothermal 9100 melting point apparatus. Infrared spectra were recorded on a Perkin Elmer
Spectrum GX FTꢀIR spectrometer using thin films on KBr (for solids) discs or Nujol (for liquids). Low
resolution (LR) and high resolution (HR) electrospray mass spectral (ESꢀMS) analyses were acquired
by electrospray ionisation (ESI) within the School of Chemistry, University of St Andrews. Low and
high resolution ESI MS were carried out on a Micromass LCT spectrometer or at a high performance
orthogonal acceleration reflecting TOF mass spectrometer, coupled to a Waters 2975 HPLC. Nuclear
magnetic resonance (NMR) spectra were acquired either on a Bruker Avance 300 (1H, 300.1 MHz; 13C,
75.5 MHz) or on a Bruker Avance 400 (1H, 400 MHz; 13C, 100.1 MHz) spectrometer and in the
deuterated solvent stated. 13C NMR spectra were acquired using the PENDANT or DEPTQ pulse
sequences.
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 characterization of 2ꢀaminothiophenes 13aꢀk, compound 20aꢀc, 21, intermediates and other
known compounds please see supporting information (SI). The prepared analogues were analyzed by
HPLC with purity above 95% (SI).
General Procedure for the synthesis of furancarboxyl amides 12aꢀk, 18aꢀb and 22aꢀq
The furoic acid chlorides 14a and 14b were prepared in situ: thionyl chloride (1.10 eq.) was added
dropwise to a mixture of 5ꢀnitrofuranꢀ2ꢀcarboxylic acid or 2ꢀfuroic acid (1.10 eq.), triethylamine (1.50
eq.) in DCM (0.4 M) under a N2 atmosphere. The reaction mixture was stirred at room temperature for
5 hours. Then crude 14a or 14b was added to another flask containing the corresponding amine or
aniline (1.00 eq.) and triethylamine (2.00 eq.) in DCM (0.4 M). The reaction mixture was stirred at
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 36
27
1
2
3
4
5
6
room temperature for 5 hours. The solvent was then removed under reduced pressure and the crude
reaction mixture was purified by column chromatography.
7
Details of the synthesis of furancarboxyl amides 12a, 12g, 22e, 22h and 22s are given here in the
main text of the paper, the remainder (12bꢀ12f, 12hꢀ12k, 22aꢀ22d, 22f, 22g, 22iꢀ22t and 22u) are
described in SI.
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
Ethyl 5ꢀethylꢀ2ꢀ(5ꢀnitrofuranꢀ2ꢀcarboxamido)thiopheneꢀ3ꢀcarboxylate (12a): The general
procedure was followed using 2ꢀamiothiophene 13a (3.99 g, 20.0 mmol). The crude reaction mixture
was purified by column chromatography on silica gel (5:1, hexane/ethyl acetate) to afford the product as
o
an orange solid (4.60 g, 13.6 mmol, 68%). Mp: 127ꢀ128 C; IR (KBr) νmax = 3120 (s) (NH), 2958 (m)
(CꢀH), 1667 (s) (C=O), 1569 (s), 1531 (s) (NO2), 1350 (s) (NO2), 1280 (s) (CꢀO), 1266 (s) (CꢀO) cmꢀ1.
3
3
1H NMR (400 MHz, d6ꢀAcetone): δ 11.75 (br. s, 1H), 7.58 (d, J= 3.9 Hz, 1H), 7.42 (d, J= 3.9 Hz,
3
1H), 6.85 (t, 4J= 1.1 Hz, 1H), 4.28 (q, J= 7.1 Hz, 2H), 2.67 (dq, 3J= 7.5 Hz, 4J= 1.1 Hz, 2H), 1.27 (t,
3
3J= 7.1 Hz, 3H), 1.17 (t, J= 7.5 Hz, 3H). 13C NMR (100 MHz, d6ꢀAcetone): δ 165.8 (C), 153.6 (C),
153.1 (C), 147.4 (C), 145.6 (C), 139.2 (C), 120.6 (CH), 118.6 (CH), 115.0 (C), 113.8 (CH), 61.7 (CH2),
23.2 (CH2), 15.9 (CH3), 14.6 (CH3). LRMS (ES+): m/z (%) 360.87 (100) [M+Na]+; HRMS (ES+): m/z
calcd for C14H14N2O6NaS [M+Na]+: 361.0470; found 361.0469.
Nꢀ(3ꢀcarbamoylꢀ5ꢀethylthiophenꢀ2ꢀyl)ꢀ5ꢀnitrofuranꢀ2ꢀcarboxamide (12g): The general procedure
was followed using 2ꢀamiothiophene 13g (170 mg, 1.00 mmol). The crude reaction mixture was
purified by column chromatography on silica gel (1:1, hexane/ethyl acetate) to afford the product as a
o
yellow solid (167 mg, 0.54 mmol, 54%). Mp: (Dec.> 224 C); IR (KBr) νmax = 3476 (m), 3341 (m)
(NH), 3211 (m), 3143 (m), 1659 (s) (C=O), 1591 (s), 1565 (s), 1534 (s) (NO2), 1349 (s) (NO2), 739 (m)
1
3
3
cmꢀ1. H NMR (400 MHz, d6ꢀAcetone): δ 7.56 (d, J= 3.9 Hz, 1H), 7.36 (d, J= 3.9 Hz, 1H), 7.05 (t,
3
4
3
13
4J= 1.1 Hz, 1H), 6.80 (br. s, 1H), 2.66 (dq, J= 7.5 Hz, J= 1.1 Hz, 2H), 1.16 (t, J= 7.5 Hz, 3H). C
ACS Paragon Plus Environment