Optimization of N-benzyl-5-nitrofuran-2-carboxamide as an antitubercular agent

Article abstract of DOI:10.1016/j.bmcl.2018.12.053

The optimization campaign for a nitrofuran antitubercular hit (N-benzyl-5-nitrofuran-2-carboxamide; JSF-3449) led to the design, synthesis, and biological profiling of a family of analogs. These compounds exhibited potent in vitro antitubercular activity (MIC = 0.019–0.20 μM) against the Mycobacterium tuberculosis H37Rv strain and low in vitro cytotoxicity (CC₅₀ = 40–>120 μM) towards Vero cells. Significant improvements in mouse liver microsomal stability and mouse pharmacokinetic profile were realized by introduction of an α α-dimethylbenzyl moiety. Among these compounds, JSF-4088 is highlighted due to its in vitro antitubercular potency (MIC = 0.019 μM) and Vero cell cytotoxicity (CC₅₀ > 120 μM). The findings suggest a rationale for the continued evolution of this promising series of antitubercular small molecules.

Full text of DOI:10.1016/j.bmcl.2018.12.053

Bioorganic&MedicinalChemistryLettersxxx(xxxx)xxx–xxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Optimization of N-benzyl-5-nitrofuran-2-carboxamide as an antitubercular
agent
Ricardo Gallardo-Maciasa, Pradeep Kumarb, Mark Jaskowskia, Todd Richmannb, Riju Shresthab,
Riccardo Russob, Eric Singletonb, Matthew D. Zimmermanc, Hsin Pin Hoc, Véronique Dartoisc,
⁎
Nancy Connellb, David Allandb, Joel S. Freundlicha,b,
a Department of Pharmacology, Physiology, and Neuroscience, Rutgers University – New Jersey Medical School, Newark, NJ, USA
b Division of Infectious Disease, Department of Medicine and the Ruy V. Lourenço Center for the Study of Emerging and Re-emerging Pathogens, Rutgers University – New
Jersey Medical School, Newark, NJ, USA
c Public Health Research Institute, Rutgers University – New Jersey Medical School, Newark, NJ, USA
A R T I C L E I N F O
A B S T R A C T
Keywords:
The optimization campaign for a nitrofuran antitubercular hit (N-benzyl-5-nitrofuran-2-carboxamide; JSF-3449)
led to the design, synthesis, and biological profiling of a family of analogs. These compounds exhibited potent in
vitro antitubercular activity (MIC = 0.019–0.20 μM) against the Mycobacterium tuberculosis H37Rv strain and low
in vitro cytotoxicity (CC50 = 40– > 120 μM) towards Vero cells. Significant improvements in mouse liver mi-
crosomal stability and mouse pharmacokinetic profile were realized by introduction of an α, α-dimethylbenzyl
moiety. Among these compounds, JSF-4088 is highlighted due to its in vitro antitubercular potency
(MIC = 0.019 μM) and Vero cell cytotoxicity (CC50 > 120 μM). The findings suggest a rationale for the con-
tinued evolution of this promising series of antitubercular small molecules.
Mycobacterium tuberculosis
Nitrofuran
Benzamide
α, α-Dimethyl
Tuberculosis (TB), caused by Mycobacterium tuberculosis, is a deadly
airborne infectious disease that led to 1.7 million deaths in 2016.1 It is
believed that one-third of the world’s population is infected with TB
and ca. 10 million new cases have been reported annually. Front-line
produces nitric oxide (NO•),8 a reactive nitrogen species (RNS) that has
been demonstrated to have cidal consequences for the bacterium.3 Even
though pretomanid and delamanid are novel antitubercular agents with
efficacy against both drug-sensitive and drug-resistant TB strains, these
investigational drugs suffer from points of weakness. For example,
pretomanid and delamanid exhibit a relatively high frequency of re-
sistance (e.g., 10−5 – 10−6) coupled with dramatic losses (≥500X) of
potency versus such mutants.9,10
TB chemotherapy, developed > 50 years ago, includes
a 2-month
treatment with isoniazid (INH), rifampicin (RIF), pyrazinamide (PZA),
and ethambutol (EMB) followed by 4 months of INH and RIF. However,
the emergence of multi-drug resistant and extensively-drug resistant TB
strains has rendered TB a very challenging infectious disease to treat.2
Therefore, new chemotherapeutic agents, characterized by a novel
mechanism of action, improved efficacy, and potential for significantly
shortening the current length of treatment, are required.
As an alternative to these nitroimidazoles, nitrofuran containing
compounds have also demonstrated antitubercular activity. For in-
stance, nitrofuranylamides, nitrofuranylpiperazines, and nitrofur-
anylisoxazolines have demonstrated in vitro and in vivo efficacy
(Fig. 2),11–16 but to the best of our knowledge have yet to transition to
clinical studies. Inspired by these compounds and our efforts with dis-
tantly related triazine17 and pyrimidine18 series, we decided to further
probe the structure-activity relationship (SAR) of nitrofuranylamides as
antitubercular agents. We commenced with the hit compound JSF-
3449 from our internal screening efforts,19 and pursued its evolution to
an α, α-dimethyl benzylamide JSF-4088 (Fig. 2). Herein, we discuss
how this optimization proceeded, outlining the hurdles that were
overcome and those that remain in the pursuit of a candidate for in vivo
Pretomanid,3 delamanid,4 and metronidazole,5 feature
a ni-
troimidazole moiety, and are of clinical relevance as antitubercular
agents (Fig. 1). Delamanid was approved in Europe in 2014, while both
it and pretomanid are in clinical trials in the United States. In a com-
pleted clinical study, metronidazole exhibited some benefit with respect
to shortening drug therapy, but neurotoxic adverse events were deemed
a critical limitation.6 It is proposed that the nitro moiety, in both pre-
tomanid and delamanid, undergoes intrabacterial drug metabolism by a
deazaflavin-dependent nitroreductase, Ddn.7 This bioactivation
⁎ Corresponding author at: Department of Pharmacology, Physiology, and Neuroscience, Rutgers University – New Jersey Medical School, Newark, NJ, USA.
E-mail address: freundjs@rutgers.edu (J.S. Freundlich).
https://doi.org/10.1016/j.bmcl.2018.12.053
Received 27 October 2018; Received in revised form 20 December 2018; Accepted 22 December 2018
0960-894X/©2018ElsevierLtd.Allrightsreserved.
Pleasecitethisarticleas:Gallardo-Macias,R.,Bioorganic&MedicinalChemistryLetters,https://doi.org/10.1016/j.bmcl.2018.12.053

R. Gallardo-Macias et al.
Bioorganic&MedicinalChemistryLettersxxx(xxxx)xxx–xxx
Figure 1. The chemical structures of delamanid, pretomanid, and metronidazole.
Figure 2. A) Nitrofuranylbenzylamides as antitubercular agents and B) highlight of our introduction of α, α-dimethyl substitution of the benzylic carbon to afford
JSF-4088.
efficacy studies.
Table 1
MIC and CC50 values of select nitrofuranylamides in μM.
JSF-3449, which appears to have first been noted as displaying
20
antitubercular activity in 1967,
exhibited in our hands a modest
minimal inhibitory concentration (MIC; minimum concentration of
compound that achieves 90% growth inhibition of the in vitro cultured
bacteria) of 0.39 μM versus the H37Rv laboratory strain of M. tubercu-
losis. A more potent (lower value) MIC was set as an optimization goal.
The hit’s Vero cell cytotoxicity CC50 (minimum compound concentra-
tion that inhibits growth of this model mammalian cell line by 50%)
was 13 μM. The SI value (CC50/MIC) of 33, acceptable for a hit com-
pound (≥10), was also targeted for improvement by not only a de-
Compound
R
M. tuberculosis MIC (μM)
Vero Cell CC50 (μM)
JSF-3449
Benzyl
0.39
3.1
13
1
2
3
4
5
6
7
8
9
Phenyl
6.7
12
crease in MIC but also by an increase in CC50
.
18 JSF-3449 was profiled
Phenethyl
3.1
versus a panel of ESKAPE pathogens and was selective for M. tubercu-
losis (Supplemental Information, Table S1). Due to the potency of JSF-
3449, we further assayed JSF-3449 for in vitro mouse liver microsome
(MLM) stability and kinetic solubility (S) in pH 7.4 PBS as well as
mouse pharmacokinetic (PK) profile. JSF-3449 displayed an MLM t1/
2 = 28.5 min, and S > 500 µM. Ideally, the t1/2 would be greater than
or equal to 60 min, but the solubility was more than satisfactory
(> 100 μM).18 Oral dosing (po) of JSF-3449 in CD-1 female mice, at a
level of 25 mg/kg in 5% DMA/60% PEG300/35% D5W, showed
average (n = 2 mice) values of the maximum plasma concentration
(Cmax) of 285 ng/mL (1160 nM) and AUC0-5h of 547 h × ng/mL (Sup-
plemental Information, Figure S1). Thus, the PK profile reinforced the
needs for improving MIC and t1/2, as a correlate of oral exposure in the
mouse. With this in mind, we initiated the synthesis of a novel set of
analogs of JSF-3449 (Table 1).
Phenylpropyl
2-ClBenzyl
3-ClBenzyl
4-ClBenzyl
4-ClPhenyl
4-ClPhenethyl
4-ClPhenylpropyl
0.78
6.2
11
11
0.78
0.098
0.78
0.39
0.39
5.5
2.8
1.5
5.4
19
little with a linker length of 0 – 3 carbons, the trend for antitubercular
efficacy followed: 1 > 3 > > 0 and 2 (Table 1).
Next, we examined the effect of substituting the phenyl ortho-,
meta-, or para-position in JSF-3449 with a chlorine. The corresponding
analogs, prepared according to the route for JSF-3449 (Scheme 1),
demonstrated divergent activity from the parent compound. 4-sub-
stituted compound 6 had a 4-fold enhancement in antibacterial activity
(MIC = 0.098 μM), while 3-chloro compound 5 was twofold less potent
and 2-substituted compound 4 lost 16x. Chloro substitution at any of
the three positions of JSF-3449 did lead to a reduction in the Vero cell
CC50, although the SI was only slightly decreased when translating from
JSF-3449 (SI = 33) to 6 (SI = 29). A return to the examination of the
effect of tether length, while maintaining a 4-Cl phenyl moiety, de-
monstrated that the 0- (compound 7), 2- (compound 8), and 3-carbon
spacers (compound 9) afforded less potent compounds than the benzyl
analog 6. Analogs 9 and 8 did, however, offer reductions in the Vero
cell cytotoxicity as compared to compound 6.
As previous reports indicated that antitubercular activity is often
associated with the 5-nitrofuran moiety,11–14 we kept this moiety con-
stant in our studies. We initially examined the linker length between the
amide nitrogen and phenyl ring. Aniline 1, phenethylamine 2, and
phenylpropylamine 3 were synthesized via a coupling reaction between
the appropriate acid chloride and amine components, similar to that
used to initially prepare JSF-3449 (Scheme 1). It should be noted that 1
and 2 have been investigated previously by the Lee group in their
studies of nitrofuranylamides and exhibited similar antitubercular ac-
tivities as reported herein.13 While the Vero cell cytotoxicity changed
2

R. Gallardo-Macias et al.
Bioorganic&MedicinalChemistryLettersxxx(xxxx)xxx–xxx
Scheme 1. Reagents and conditions: a) Et3N, DCM; b) NaH, THF, 0 °C to rt.
Table 2
MIC and CC50 values of α-substituted nitrofuranylbenzylamides.
Compound
R1
R2
M. tuberculosis MIC (μM)
Vero Cell CC50 (μM)
JSF-3449
H
H
0.39
6.2
0.78
6.2
6.2
12
13
10
11
12
13
14
15
16
CH3
CH3
CH2
H
24
CH3
CH2
H
> 180
46
a
a
Isopropyl
Cyclopropyl
tert-butyl
O
22
H
22
H
25
10
O
25
3.0
a
Denotes a fused cyclopropyl ring to the benzylic carbon.
Subsequently, we sought to add different alkyl substituents to the
benzylic carbon to reduce metabolic oxidation and, thus, improve MLM
stability and mouse PK profile while potentially reducing cytotoxicity
due to metabolite formation.21 Returning to JSF-3449 as the model
compound, different groups were introduced at the benzylic carbon:
methyl (compound 10), α, α-dimethyl (compound 11), a fused cyclo-
propyl (compound 12), isopropyl (compound 13), cyclopropyl (com-
pound 14), tert-butyl (compound 15), and a carbonyl (compound 16).
The syntheses of these analogs were achieved following the routes in
Scheme 1. All of the analogs, except α, α-dimethyl 11, lost significant
aromatic ring was replaced with a subset of heterocycles. For example,
benzothiophene (compound 17), benzofuran (compound 18), and 2-
naphthalene (compound 19) analogs were prepared following synthetic
procedures depicted in Scheme 2, featuring the reductive dimethyla-
tion22 of commercially available nitriles to afford the heterocyclic α, α-
dimethylmethylamine for coupling with the 5-nitrofuroyl chloride.
Replacement of phenyl with 2-naphthyl (19; MIC = 0.0078 μM), 2-
benzothiophene (17; MIC = 0.078 μM), or 2-benzofuran (18;
MIC = 0.20 μM) all afforded improvements in whole-cell activity
(Table 4). While 17 and 18 displayed an increase in their Vero cell
cytotoxicity compared to 11, 19 did not exhibit significant cytotoxicity
(CC50 > 150 μM). The mouse PK profiles for 19 and 17 showed very
limited exposure in the plasma (Supplemental Information Figures S3
and S4, respectively). Their relative decreased metabolic stabilities
(MLM t1/2 ≤ 2.13 min) and S values may be responsible for this out-
come.
whole-cell activity (MIC = 6.2
– 25 µM) compared to JSF-3449
(Table 2). 11 displayed a twofold increase in MIC compared to JSF-
3449, but demonstrated a drastically improved CC50 of > 180 μM
(SI > 230).
Compound 11 was profiled for in vitro MLM stability, aqueous so-
lubility, and mouse PK (Table 3). In comparison to JSF-3449, 11 de-
monstrated enhanced t1/2 (meeting the goal criteria) and mouse PK
profile (Supplemental Information Figure S2) as judged by Cmax and
AUC0-5h, although its S was depressed but still acceptable at 388 μM.
The α, α-dimethyl moiety was then held constant while the pendant
We returned to the α, α-dimethyl benzylamide series and focused on
the exploration of different substituents at the 4-position leveraging the
synthetic route in Scheme 2. Analogs with an electron-donating group
on the phenyl ring such as 4-t-butyl (compound 23), 4-Me2N
3

R. Gallardo-Macias et al.
Bioorganic&MedicinalChemistryLettersxxx(xxxx)xxx–xxx
Table 3
MLM, solubility, and in vivo mouse PK (single 25 mg/kg dose po) profiles of 11 and JSF-3449.
a
a
Compound
MLM t1/2 (min)
Clint (μL/min/mg protein)
Kinetic Solubility in pH 7.4 PBS (μM)
Cmax in ng/mL (nM)
AUC0-5h (h × ng/mL)
JSF-3449
11
28.5
74.5
24.3
9.30
> 500
388
285 (1160)
443 (1620)
547
717
a
Average of two experiments.
Scheme 2. Reagents and conditions: a) CeCl3, CH3Li, THF, −78 °C to rt; b) Et3N, DCM.
Table 4
MIC and CC50 values of α, α-dimethylsubstituted nitrofuranylheteroarylamides.
Compound
R
M. tuberculosis MIC (μM)
Vero Cell
MLM t1/2 (min)
Clint (μL/min/mg protein)
Kinetic Solubility in pH 7.4 PBS (μM)
CC50 (μM)
11
17
18
19
Phenyl
0.78
> 180
38
74.5
1.87
n.d.
9.30
371
n.d.
325
388
5.61
n.d.
1.50
2-benzothiophene
2-benzofuran
2-naphthyl
0.078
0.20
40
0.0078
> 150
2.13
n.d. = not determined.
Table 6
Table 5
MIC and CC50 values of nitrofuranyl(4-substituted)benzylamides in μM.
MIC and CC50 values of α, α-dimethyl substituted nitrofuranyl(4-substituted)
benzylamides.
Compound
R
M. tuberculosis MIC
Vero Cell CC50
(μM)
(μM)
Compound
R
M. tuberculosis MIC (μM)
Vero Cell CC50 (μM)
28
29
30
31
4-chlorophenyl
0.12
0.062
1.2
4.4
8.4
18
4-chlorophenoxy
11
20
21
22
23
24
25
26
27
H
0.78
0.15
0.078
0.20
0.39
0.20
0.78
> 10
0.78
> 180
81
4-morpholinomethyl
4-chlorophenoxymethyl
4-Cl
0.039
4.0
4-OCH3
4-Me2N
4-t-Bu
4-OCF3
4-F
82
79
> 150
70
withdrawing moieties, only the 4-Cl analog (20; MIC = 0.15 µM) de-
monstrated an improvement in MIC over 11. 4-F (compound 25) and 4-
CF3 (compound 27) analogs were equipotent to 11, and the 4-CN
(compound 26) derivative failed to demonstrate significant activity
(MIC > 10 μM). All but the 4-t-butyl (23) and 4-CF3 (27) analogs ex-
hibited a significant decrease in Vero cell cytotoxicity over 11, although
the SI values for 4-OCH3 (21; SI = 1100) and 4-Cl (20; SI = 540) were
42
4-CN
4-CF3
83
> 150
(compound 22), and 4-OCF3 (compound 24) were more active than 11,
but the 4-OCH3 (21; MIC = 0.078 µM) analog was the most potent
derivative (Table 5). In contrast, amongst compounds with electron-
4

R. Gallardo-Macias et al.
Bioorganic&MedicinalChemistryLettersxxx(xxxx)xxx–xxx
Scheme 3. Reagents and conditions: a) Cs2CO3, CuI, N,N-Dimethylglycine hydrochloride, dioxane, reflux, 24 h; b) Red-Al, THF, 0 °C to rt; c) CeCl3, CH3Li, THF,
−78 °C to rt; d) Et3N, DCM.
Scheme 4. Reagents and conditions: a) K2CO3, acetone, reflux, 48 h; b) Red-Al, THF, 0 °C to rt; c) CeCl3, CH3Li, THF, −78 °C to rt; d) Et3N, DCM.
Table 7
MIC and CC50 values of α,α-disubstituted nitrofuranyl(4-substituted)benzyla-
mides in μM.
Table 9
MIC and CC50 values of α, α-dimethylsubstituted nitrofuranylbenzylamides in
μM.
Compound
R
M. tuberculosis MIC
Vero Cell CC50
(μM)
Compound
R
M. tuberculosis
Vero Cell
CC50
(μM)
MIC (μM)
(μM)
11
32
33
34
35
H
0.78
25
> 180
> 140
67
4-chlorophenyl
4-chlorophenoxy
4-morpholinomethyl
4-chlorophenoxymethyl
35
4-chlorophenoxymethyl
0.12
120
> 120
6.7
0.39
> 10
0.12
36
4-methoxyphenoxymethyl
0.078
0.39
130
37
4-trifluoromethoxyphenoxymethyl
4-trifluoromethylphenoxymethyl
2-Chlorophenoxymethyl
120
38
0.39
7.0
39
0.31
30
40 (JSF-4088)
2-methoxyphenoxymethyl
0.019
0.098
0.31
> 120
13
41
42
2-trifluoromethoxyphenoxymethyl
2-trifluoromethylphenoxymethyl
Table 8
14
In vivo mouse PK (single 25 mg/kg dose po) profile of 11 and 35.
a
a
Compound
Cmax in ng/mL (nM)
AUC0-5h (h × ng/mL)
11
35
443 (1620)
9.2 (22)
717
34
In contrast, 30, the morpholino analog, had modest antitubercular
activity (MIC = 1.2 µM). We hypothesized that addition of a mor-
pholine, as a solubilizing group, would help improve potency; how-
ever, this was not the case. All of these analogs displayed an increase
in Vero cell cytotoxicity compared to 11, although the SI values for
31 (SI = 100) and 29 (SI = 130) were sufficient.
a
Average of two experiments.
more than sufficient. Unfortunately, the mouse PK profile of 21 de-
monstrated low exposure in the plasma as compared to 11 (Supple-
mental Information, Figure S5).
Due to the potency of these analogs and having in mind to further
improve Vero cell cytotoxicity, we synthesized their corresponding
α, α-dimethyl analogs as depicted in Schemes 2, 3 and 4, featuring
reduction22 of the intermediate nitrile with CeCl3/CH3Li. Key inter-
mediates in Scheme 4 were prepared by coupling the corresponding
meta- or para-substituted phenol to 4-(bromomethyl) benzonitrile,
using K2CO3 in refluxing acetone. All four analogs succeeded with
respect to the goal of decreasing Vero cell cytotoxicity. 4-chlor-
obiphenyl (compound 32) and 4-morpholinomethyl (compound 34)
showed significantly reduced antitubercular activity (MIC > 10 μM)
towards M. tuberculosis (Table 7). In contrast, the 4-chlorophenoxy
(compound 33) and 4-chlorophenoxymethyl (compound 35) analogs
suffered more modest losses in antitubercular activity, displaying
The addition of larger substituents to the 4-position of the ben-
zylamide ring was then pursued (Table 6). Returning for the moment
to the methylene instead of α, α-dimethyl linker, we explored the
following groups: 4-chlorophenyl (compound 28), 4-chlorophenoxy
(compound 29), 4-morpholinomethyl (compound 30), and 4-chlor-
ophenoxymethyl (compound 31). The analogs were synthesized ac-
cording to the routes in Schemes 1, 3, and 4. The route in Scheme 3
depended on a copper(I)-catalyzed diaryl ether formation,23 fol-
lowed by reduction of the nitrile with Red-Al, and ended with amide
formation via reaction with 5-nitrofuroyl chloride. The aromatic
substituents afforded the more efficacious compounds: 31
(MIC = 0.039 µM), 29 (MIC = 0.062 µM), and 28 (MIC = 0.12 µM).
5

R. Gallardo-Macias et al.
Bioorganic&MedicinalChemistryLettersxxx(xxxx)xxx–xxx
Table 10
MLM, solubility, and in vivo mouse PK (single 25 mg/kg dose po) profiles of 11, 36, and JSF-4088.
a
a
Compound
MLM t1/2 (min)
Clint (μL/min/mg protein)
Kinetic Solubility in pH 7.4 PBS (μM)
Cmax in ng/mL (nM)
AUC0-5h (h × ng/mL)
11
74.5
45.0
11.6
9.30
15.4
59.8
388
443 (1620)
BLQb
717
11
36
< 0.06
0.182
JSF-4088
85 (2 1 0)
210
a
b
Average of two experiments.
Below limit of quantification of 5.0 ng/mL.
MIC values of 0.39 and 0.12 µM, respectively. In comparison to 11,
both 33 and 35 showed improved antitubercular activity. 35 de-
monstrated the best antitubercular activity with an excellent se-
lectivity towards Vero cells (SI = 1000). 35 was, thus, profiled for
mouse PK (Table 8). In comparison to 11, 35 showed much lower
exposure as judged by Cmax and AUC0-5h (Supplemental Information
Figure S6).
Acknowledgements
The authors gratefully acknowledge support from NIH grants
R33AI11167, R21AI111647, and U19AI109713.
Appendix A. Supplementary data
Finally, further optimization around 35, due to its increased in vitro
potency over that of 11, was proposed by adding different moieties such
as OCH3, OCF3, and CF3 in either the meta- or para-position of the
terminal phenoxy ring. These analogs were synthesized following the
route depicted in Scheme 4. As seen in Table 9, the 4-OCH3 (compound
36) analog showed improved antitubercular activity (MIC = 0.078 μM)
compared to 35, while maintaining low cytotoxicity (CC50 > 120 μM).
Both the 4-OCF3 (compound 37) and 4-CF3 (compound 38) analogs
were ca. 3-fold less potent than 35, and showed relatively high Vero cell
cytotoxicity. The 2-Cl (compound 39) and 2-CF3 (compound 42) ana-
logs exhibited similar sub-optimal profiles. 2-OCF3 (compound 41) was
slightly more potent versus M. tuberculosis than 35, but its CC50 was
only 13 μM. 2-OCH3 (compound 40; JSF-4088) satisfyingly demon-
strated improved potency (MIC = 0.019 μM) and Vero cell cytotoxicity
(CC50 > 120 μM) than 35.
Supplementary data to this article can be found online at https://
doi.org/10.1016/j.bmcl.2018.12.053.
References
1. World Health Organization. Global Tuberculosis Report. 2017.
2. Shah NS, Auld SC, Brust JC, et al. Transmission of Extensively Drug-Resistant
Tuberculosis in South Africa. N Engl J Med. 2017;376(3):243–253.
3. Stover CK, Warrener P, VanDevanter DR, et al. A small-molecule nitroimidazopyran
drug candidate for the treatment of tuberculosis. Nature. 2000;405:962–966.
4. Matsumoto M, Hashizume H, Tomishige T, et al. OPC-67683, a nitro-dihydro-imi-
dazooxazole derivative with promising action against tuberculosis in vitro and in
mice. PLoS Med. 2006;3:2131–2144.
5. Wayne LG, Sramek HA. Metronidazole Is bactericidal to dormant cells of
Mycobacterium tuberculosis. Antimicrob Agents Chemother. 1994;38:2054–2058.
6. Carroll MW, Jeon D, Mountz JM, et al. Efficacy and safety of metronidazole for
pulmonary multidrug-resistant tuberculosis. Antimicrob Agents Chemother.
2013;57(8):3903–3909.
7. Manjunatha UH, Boshoff H, Dowd CS, et al. Identification of a nitroimidazo-oxazine-
specific protein involved in PA-824 resistance in Mycobacterium tuberculosis. Proc Natl
Acad Sci. 2006;103:431–436.
As both 36 and JSF-4088 showed improved antitubercular activity
(in the absence of significant growth inhibition of the ESKAPE bacteria;
Supplemental Information Table S1) over 35, these two compounds
were profiled for S, MLM t1/2, and mouse PK (Table 10). 36 and JSF-
4088 did not improve upon the MLM stability or aqueous solubility of
11, with the aqueous solubility of 36 being unacceptably low
(≤0.06 μM). It was, therefore, not surprising, that plasma exposure of
36 was below the lower limit of quantification (LLOQ = 5.0 ng/mL).
JSF-4088 showed quantifiable absorption in the mice (Supplemental
Information Figure S7), but still failed to improve upon the profile of
11. In addition, it demonstrated poor accumulation in mouse lungs
(mean Clung/Cplasma = 0.06) at t = 5 h post administration of a single
25 mg/kg po dose (Supplemental Information, Table S2).
8. Singh R, Manjunatha U, Boshoff HI, et al. PA-824 kills nonreplicating Mycobacterium
tuberculosis by intracellular NO release. Science. 2008;322(5906):1392–1395.
9. Bloemberg GV, Keller PM, Stucki D, et al. Acquired Resistance to Bedaquiline and
Delamanid in Therapy for Tuberculosis. N Engl J Med. 2015;373(20):1986–1988.
10. Haver HL, Chua A, Ghode P, et al. Mutations in genes for the F420 biosynthetic
pathway and a nitroreductase enzyme are the primary resistance determinants in
spontaneous in vitro-selected PA-824-resistant mutants of Mycobacterium tubercu-
losis. Antimicrob Agents Chemother. 2015;59(9):5316–5323.
11. Fan YL, Wu JB, Ke X, Huang ZP. Design, synthesis and evaluation of oxime-func-
tionalized nitrofuranylamides as novel antitubercular agents. Bioorg Med Chem Lett.
2018;28(18):3064–3066.
12. Hurdle JG, Lee RB, Budha NR, et al. A microbiological assessment of novel ni-
trofuranylamides as anti-tuberculosis agents. J Antimicrob Chemother.
2008;62(5):1037–1045.
13. Tangallapally RP, Yendapally R, Lee RE, et al. Synthesis and evaluation of ni-
trofuranylamides as novel antituberculosis agents. J Med Chem.
A family of novel N-benzyl-5-nitrofuran-2-carboxamides was de-
signed and synthesized to optimize the hit compound JSF-3449. While
addressing limitations in cytotoxicity and metabolic stability, the ad-
dition of an α, α-dimethyl moiety was a key modification. 11, in par-
ticular, showed improved selectivity versus Vero cells (SI > 230), en-
hanced MLM stability (t1/2 = 74.5 min), and enhanced mouse PK
profile, while only demonstrating small setbacks with regard to MIC
and S. Further SAR optimization of 11 led to JSF-4088 with sig-
nificantly improved in vitro potency (MIC = 0.019 μM) and Vero cell
cytotoxicity window (SI > 6300). However, JSF-4088 was unable to
improve upon the still modest mouse PK profile of 11. Further PK
profile optimization remains as a challenge to this series of compounds
with outstanding in vitro potency against M. tuberculosis. This challenge
may be answered by merging our use of α, α-dimethyl substitution of
the benzylic carbon in conjunction with a broader range of substitutions
of the pendant phenyl moiety as exemplified in reports of nitrofuranyl
amides by Tangallapally et al.,13,15 which despite generally being less
potent in vitro versus M. tuberculosis and more cytotoxic to Vero cell
cultures did display in vivo efficacy in an interferon-γ knockout mouse
model of infection.
2004;47(21):5276–5283.
14. Yempalla KR, Munagala G, Singh S, et al. Nitrofuranyl Methyl Piperazines as New
Anti-TB Agents: Identification, Validation, Medicinal Chemistry, and PK Studies. ACS
Med Chem Lett. 2015;6(10):1041–1046.
15. Tangallapally RP, Yendapally R, Lee RE, Lenaerts AJ, Lee RE. Synthesis and eva-
luation of cyclic secondary amine substituted phenyl and benzyl nitrofuranyl amides
as novel antituberculosis agents. J Med Chem. 2005;48(26):8261–8269.
16. Rakesh Bruhn DF, Scherman MS, et al. Pentacyclic nitrofurans with in vivo efficacy
and activity against nonreplicating Mycobacterium tuberculosis. PLoS ONE.
2014;9(2)::e87909.
17. Ekins S, Reynolds RC, Kim H, et al. Bayesian models leveraging bioactivity and cy-
totoxicity information for drug discovery. Chem Biol. 2013;20:370–378.
18. Inoyama D, Paget SD, Russo R, Novel,, et al. Pyrimidines as Antitubercular Agents.
Antimicrob Agents Chemother. 2018;62(3)..
19. Wilson R, Kumar P, Parashar V, et al. Antituberculosis thiophenes define a require-
ment for Pks13 in mycolic acid biosynthesis. Nat Chem Biol. 2013;9(8):499–506.
20. Rusan M, Petrovanu O. Leonte C. In vitro antimicrobial effect of some benzylamine
and thienylamine derivatives. Analele Stiintifice ale Universitatii AI I Cuza din Iasi,
Sectiunea 1c Chimie. 1967;13(2):153–156.
21. Talele TT. Natural-Products-Inspired Use of the gem-Dimethyl Group in Medicinal
Chemistry. J Med Chem. 2018;61(6):2166–2210.
22. Ciganek E. Tertiary carbinamines by addition of organocerium reagents to nitriles
and ketimines. J Org Chem. 1992;57:4521–4527.
23. Ma D, Cai Q. N, N-dimethyl glycine-promoted Ullmann coupling reaction of phenols
and aryl halides. Org Lett. 2003;5(21):3799–3802.
6