Design, Synthesis, and Biological Evaluation of 2-Nitroimidazopyrazin-one/-es with Antitubercular and Antiparasitic Activity

Article abstract of DOI:10.1021/acs.jmedchem.8b01578

Tuberculosis and parasitic diseases, such as giardiasis, amebiasis, leishmaniasis, and trypanosomiasis, all urgently require improved treatment options. Recently, it has been shown that antitubercular bicyclic nitroimidazoles such as pretomanid and delamanid have potential as repurposed therapeutics for the treatment of visceral leishmaniasis. Here, we show that pretomanid also possesses potent activity against Giardia lamblia and Entamoeba histolytica, thus expanding the therapeutic potential of nitroimidazooxazines. Synthetic analogues with a novel nitroimidazopyrazin-one/-e bicyclic nitroimidazole chemotype were designed and synthesized, and structure-activity relationships were generated. Selected derivatives had potent antiparasitic and antitubercular activity while maintaining drug-like properties such as low cytotoxicity, good metabolic stability in liver microsomes and high apparent permeability across Caco-2 cells. The kinetic solubility of the new bicyclic derivatives varied and was found to be a key parameter for future optimization. Taken together, these results suggest that promising subclasses of bicyclic nitroimidazoles containing different core architectures have potential for further development.

Full text of DOI:10.1021/acs.jmedchem.8b01578

Subscriber access provided by University of Leicester
Article
Design, synthesis and biological evaluation of 2-nitroimidazopyrazin-
one/-es with antitubercular and antiparasitic activity
Angie M Jarrad, Chee Wei Ang, Anjan Debnath, Hye Jee Hahn, Kyra Woods, Lendl Tan,
Melissa L Sykes, Amy J. Jones, Ruby Pelingon, Mark Stuart Butler, Vicky M. Avery,
Nicholas P. West, Tomislav Karoli, Mark Arnold Thomas Blaskovich, and Matthew A. Cooper
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.8b01578 • Publication Date (Web): 23 Nov 2018
Downloaded from http://pubs.acs.org on November 24, 2018
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 90
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Manuscript
Design, synthesis and biological evaluation of 2-nitroimidazopyrazin-
one/-es with antitubercular and antiparasitic activity
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
,§
1,§
2
2
3
3
Angie M. Jarrad , Chee Wei Ang , Anjan Debnath , Hye Jee Hahn , Kyra Woods , Lendl Tan ,
4
4
1
1
4
3
Melissa L. Sykes , Amy J. Jones , Ruby Pelingon , Mark S. Butler , Vicky M. Avery , Nicholas P. West ,
Tomislav Karoli1, #, Mark A.T. Blaskovich and Matthew A. Cooper
1,*
1,*
1
Institute for Molecular Bioscience, The University of Queensland, St Lucia, Queensland, 4072,
Australia
2
Center for Discovery and Innovation in Parasitic Diseases, Skaggs School of Pharmacy and
Pharmaceutical Sciences, University of California, San Diego, La Jolla, CA 92093 USA
3
School of Chemistry and Molecular Bioscience, The University of Queensland, St Lucia, Queensland,
4072, Australia
4
Discovery Biology, Griffith Institute for Drug Discovery, Griffith University, Nathan, Queensland,
4
§
111, Australia
Authors contributed equally; *joint corresponding authors
Abstract
Tuberculosis and parasitic diseases, such as giardiasis, amebiasis, leishmaniasis and trypanosomiasis, all
urgently require improved treatment options. Recently, it has been shown that anti-tubercular bicyclic
nitroimidazoles such as pretomanid and delamanid have potential as repurposed therapeutics for the
treatment of visceral leishmaniasis. Here we show that pretomanid also possesses potent activity against
Giardia lamblia and Entamoeba histolytica, thus expanding the therapeutic potential of nitroimidazo-
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 90
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
oxazines. Synthetic analogs with the novel nitroimidazopyrazin-one/-e bicyclic nitroimidazole
chemotype were designed, synthesized and structure activity relationships generated. Selected
derivatives had potent antiparasitic and antitubercular activity whilst maintaining drug-like properties
such as low cytotoxicity, good metabolic stability in liver microsomes and high apparent permeability
across Caco-2 cells. The kinetic solubility of the new bicyclic derivatives varied, and was found to be a
key parameter for future optimization. Taken together, these results suggest promising subclasses of
bicyclic nitroimidazoles containing different core architectures have potential for further development.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Introduction
Infectious diseases are an enormous global health burden. The nitroimidazole class of antibiotics,
exemplified by metronidazole (1, Figure 1), has a long history of use to treat bacterial and parasitic
1
infections. The mode of action of nitroimidazoles involves partial reduction of the nitro group and
subsequent decomposition of the compound to give toxic radical species that cause DNA and protein
1
damage. The nitro group is activated by different mechanisms, with multiple enzymes involved in
electron transfer reactions in different target organisms, which accounts for the remarkably broad spectra
of action of the class. More recently, there have been issues with clinical efficacy that are accompanied,
in some cases, by development of resistance, which has prompted re-examination of this old, but still
widely used drug class.
New clinical applications of “old” nitroimidazoles and the development of novel nitroimidazoles with a
bicyclic core scaffold architecture have significant potential to address the emergent unmet medical need
imparted by resistant bacteria and parasites. For example, secnidazole (2, Figure 1) was recently
approved in the USA for the treatment of bacterial vaginosis, despite being available earlier as a generic
in many jurisdictions.2,3 A pediatric formulation of benznidazole 3 (Figure 1), a 2-nitroimidazole used
for treatment of Chagas disease caused by the parasite Trypanosoma cruzi, gained FDA approval in 2017
ACS Paragon Plus Environment

Page 3 of 90
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
4
for use in children. The Drugs for Neglected Diseases initiative (DNDi) is currently investigating
fexinidazole (4, Figure 1) as a candidate for oral treatment of human African trypanosomiasis (HAT), a
parasitic infection caused by Trypanosoma brucei spp. that is a public health threat to ~70 million people
5
in Africa. Furthermore, bicyclic nitroimidazoles, such as delamanid (5, Figure 1) and pretomanid (6,
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
6
PA-824), are promising new antimicrobials being developed for the treatment of tuberculosis (TB), the
number one cause of death from infectious diseases and the ninth leading cause of death worldwide.7
The nitroimidazo-oxazole 5 gained conditional approval in the European Union in 2014 for the treatment
of drug resistant TB. This agent was derived from CGI 17341 (7)8,9 and overcame the mutagenic liability
of 7 (Figure 1).10,11 Meanwhile, 6 was developed concurrently with 5, and is currently in Phase III trials.
The PK profile of 6 is superior to 5 and this permits once daily dosing, although 6 is less potent.12
Figure 1. Monocyclic nitroimidazoles 1-4 and the bicyclic variants delamanid (5), pretomanid (6), CGI 17341 (7)
and (R)-PA-824 (8).
Investigations on how 5 and 6 inhibit M. tuberculosis under aerobic and anaerobic growth conditions
revealed an interesting dual mode of action. Transcriptional profiling of M. tuberculosis treated with 6
under aerobic growth conditions, gave a response consistent with both the inhibition of cell wall mycolic
1
3
acid biosynthesis and also respiratory poisoning. Additionally, it was shown that deazaflavin dependent
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 90
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
nitroreductase (Ddn) catalyzed reduction of 6 to the des-nitro 9 metabolite and that this process generated
14
nitric oxide. Nitric oxide could be detected in Mycobacterium bovis cells treated with 6 under both
aerobic and anaerobic growth conditions and the rate of NO release in M. bovis cells correlated with the
anaerobic activities for a series of analogues, supporting the premise that this is the mode of action of
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
14
this class under anaerobic growth conditions in M. tuberculosis. In comparison, 1 is only active against
non-replicating M. tuberculosis under anaerobic growth conditions (<0.06% oxygen).15
Nitroimidazoxazine 6 was also investigated for activity against other kinetoplastid organisms in multiple
developmental life stages including L. donovani (promastigote and amastigote), T. brucei brucei
1
6
(
procyclic and bloodstream) and T. cruzi (epimastigote and amastigote), with the findings prompting
further mode of action studies in L. donovani. Compound 8, the R- enantiomer of 6, was shown to be
1
6
more effective than the S- enantiomer in an in vivo model of visceral leishmaniasis. Recently, an
NAD(P)H oxidase was identified as the activating nitroreductase (NTR2) for the nitroimidazo-oxazole/-
oxazine.17 However, the monocyclic sulfone metabolite of 4, known to be activated by a type I
nitroreductase, 18 was only marginally activated by NTR2. These results support the hypothesis that 6
is activated by an alternative mechanism of action in L. donovani under these culture conditions and
illustrates the biological complexity of the mode of action of various nitroimidazoles.
17
In the course of development of 5 and 6 as treatments for tuberculosis, over 1000 derivatives were
prepared.19 The structure activity relationships disclosed to date mostly include compounds with
modifications to the aryl side chain, with a smaller number of variants that alter the bicyclic core structure
(Figure 2A). The nitro group and stereochemistry of the side chain have been shown to be critical for
activity as both 8 and S-des-nitro-PA-824 (9) derivatives were inactive in in vitro cultures at 50 μM.16
An analog with the nitro group at the 3ʹ position was shown to be inactive, clarifying that the 2ʹ position
2
0
of the nitro group on the 6 bicyclic scaffold is important for activity. Replacement of the benzylic
oxygen with nitrogen led to an amino linked series (10a-b) with favorable in vitro activity and solubility
ACS Paragon Plus Environment

Page 5 of 90
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
properties.21,22 Modification of the imidazole ring to a pyrazole (11a) or triazole (11b) was detrimental
to M. tuberculosis activity, as was replacement of the oxygen heteroatom in the oxazine ring with
sulfoxide (11c), sulfone (11d), amino (11e) or methylene (11f) groups, although a sulfur (11g)
heteroatom in the ring was tolerated.23
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
More recent efforts have explored the activity of bicyclic nitroimidazoles against the kinetoplastids
Leishmania and trypanosomes. DNDi is currently investigating additional nitro-imidazooxazines for
24
development against visceral leishmaniasis, including DNDI-8219 (12) and DNDI-0690 (13), 6- and
7
-substituted imidazooxazines, as two promising backup candidates.25,26 Interestingly,
nitroimidazothiazine oxides 14a-d were found to display favorable activity against T. b. brucei, an animal
1
9
infective strain that is commonly used as a model of HAT. Moreover, a “thio-delamanid” derivative
27
(
15) proved to be efficacious against T. cruzi, although it was inactive against Leishmania. These
studies demonstrate that the selectivity profile toward different parasites can be altered by structural
modifications of the bicyclic scaffold and that different subclasses of the bicyclic nitroimidazoles expand
the potential therapeutic scope of this antimicrobial class.
We reasoned that bicyclic nitroimidazole derivatives could have potential against an even wider range of
28
organisms than Leishmania and trypanosomes. Previously we found that 4(5)-nitroimidazoles (16) had
potent activity against Giardia lamblia, Entamoeba histolytica, Trichomonas vaginalis and Clostridium
difficile (Figure 2B) and therefore it was hypothesized that 6 might also have activity against these
organisms. Anaerobic protozoa such as G. lamblia and E. histolytica and the anaerobic bacteria C.
difficile occupy the gut under reduced oxygen tension and cause diarrheal infections. These organisms
spread by the fecal oral route through stable cyst forms for the protozoa or through spores for C. difficile.
Metronidazole 1 can be used therapeutically for infections caused by these organisms, but alternative
treatment options are desirable. Furthermore, cyclizing the 4(5)-nitroimidazoles 16 from the 1′ imidazole
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 90
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
ring position to the carboxamide nitrogen could produce new bicyclic nitroimidazopyrazin-one/-e
scaffolds. These could potentially extend the spectra of biological activity to include M. tuberculosis and
result in altered structure activity relationships against a panel of parasitic organisms. Previous studies
have shown that compounds containing the imidazopyrazinone scaffold have a range of different
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
29
biological activities, including agonism of the GABAA receptor , antagonism of the
ionotropic transmembrane receptor,30,31 modulation of ion channels to control arrhythmia32 and
3
3
inhibition of M. tuberculosis glutamine synthetase . However, no studies have investigated the
antimicrobial activity of a nitrated derivative of this class.
We now report the findings from an evaluation of 6 against a wide range of organisms, and the subsequent
design, synthesis and biological evaluation of novel bicyclic nitroimidazoles, nitroimidazopyrazinones
(17) and nitroimidazopyrazines (18) (Figure 2B), derived from the 4(5)-nitroimidazole scaffold 16
2
8
previously reported. Given the potential for activity against many different organisms, compounds were
screened against a wide range of microorganisms including: G. lamblia, E. histolytica, T. b. brucei, L.
donovani, M. tuberculosis, C. difficile, Cryptoccoccus neoformans, Candida albicans and representative
ESKAPE bacteria: Escherichia coli, Staphylococcus aureus, Klebsiella pneumoniae, Pseudomonas
aeruginosa and Actinetobacter baumannii. Derivatives were counter-screened for cytotoxicity against
mammalian liver and kidney cell lines. Preferred derivatives with favorable antimicrobial or antiparasitic
activity were then evaluated for therapeutic properties including metabolic stability, plasma protein
binding and intestinal permeability. Most of the compounds were also assessed for aqueous kinetic
solubility. This report presents the first description of structure activity relationships and evaluation of
nitroimidazopyrazin-ones/es for antimicrobial and antiparasitic activity with therapeutic potential.
ACS Paragon Plus Environment

Page 7 of 90
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 2. A) Structural variation of bicyclic nitroimidazoles described in the literature and B) the novel
nitroimidazopyrazinones and nitroimidazopyrazinones, developed by cyclizing the 4(5)-nitroimidazole
framework, described here.
Chemistry
1
2
3
Design and Synthesis of Bicyclic Nitroimidazoles with Variation at R , R = R = H
2
8
In previous work a set of amide-substituted monocyclic nitroimidazoles were prepared and it was
hypothesized for this study that linking the amide and imidazole nitrogen could form a bicyclic
nitroimidazopyrazinone system, similar to the pretomanid 6 core (Figure 2B). A library of
1
nitroimidazopyrazinones 17a-t were prepared with different functional groups at R designed to explore
a range of physicochemical properties (Scheme 1). Initial biological results indicated that benzylic R1
groups were favorable for antimicrobial and antiparasitic activity. The library was therefore tailored to
1
contain analogs with a number of different benzyl substituted R groups (17b-j), including derivative 17c
with a 4-OCF -substituted benzyl group similar to pretomanid, and analogs with substitutions at the 3′
3
1
and 2′ positions. Compound 17k was prepared with R = CH -(2-pyridyl) to introduce a hydrogen bond
2
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 90
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
acceptor and to impart more polarity and improve aqueous solubility. A methyl group was introduced at
the benzylic position (17l) to increase the lipophilicity, bulk and potentially restrict the conformation of
the side chain. Two phenethylene derivatives (17m-n) were prepared to investigate the effect of spacing
1
the aromatic ring further from the bicyclic core. Derivatives with polar side chain R groups were also
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
synthesized, including ethyl acetate 17o, ethyl hydroxyl 17p, morpholine derivatives 17q-r and the
1
carboxamide group (17s). Compound 17t with R = cyclohexyl group was prepared to investigate the
effect of a bulky aliphatic group at this position.
Initially, nitroimidazopyrazinones 17a-f, 17k-l, 17n and 17t were prepared from 4(5)-nitroimidazole
2
8
carboxamides 16a-f , 16k-l, 16n and 16t by cyclizing the 1′ imidazole to the 2′ free amide nitrogen
(route 1, Scheme 1). This was achieved by alkylating 4(5)-nitroimidazoles with bromoacetaldehyde
34
diethyl acetal under basic conditions (K CO ) which strongly favored the 4-nitro regioisomer product.
2
3
Secondly, microwave heating (µW 120°C) of 19a-f, 19k-l, 19n and 19t under acidic conditions afforded
the bicyclic products 17a-f, 17k-l, 17n and 17t. This synthesis was based on the preparation of des-nitro-
35
imidazopyrazinone 20a, as previously described by Prévot and Leumann; however, microwave heating
at higher temperatures and for shorter periods than conventional reflux facilitated analog generation for
both steps. Secondly, inclusion of a cosolvent (aq. 2M HCl :1,4-dioxane 1:1) was necessary to solubilize
the secondary amide starting material and achieve conversion to the desired products 17b-f, 17k-l, 17n
and 17t. With this approach, products often precipitated from the reaction mixture and could be isolated
in high purity by filtration and washes (H O) alone, especially if the intermediate had been purified by
2
1
chromatography. Alternatively, the products were purified by recrystallization. In the H NMR, the
2
3
imidazopyrazinone R and R proton groups were typically two doublets (J = ~5.9 Hz) each integrating
for 1 proton at ~ δ7.4 ppm and ~ δ7.6 ppm, respectively. Furthermore, both NMR (2D HMBC)
experiments and an X-ray crystal structure of 17a confirmed the 2-position of the nitro group (Supporting
Information, Figure S1).
ACS Paragon Plus Environment

Page 9 of 90
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
A second approach to prepare bicyclic nitroimidazoles 17g-j, 17m, 17o and 17q-s was developed as
shown in route 2, Scheme 1. This method was more convergent as it eliminated the need to prepare 4(5)-
nitroimidazole intermediates of each derivative. Route 2 involved alkylation of 17a with alkyl halides
under basic conditions (carbonate base) to form 17g-j, 17m, 17o and 17q-s. This approach also produced
the minor O-alkylated regioisomer in some cases. This was considered advantageous because the
alternative pyrazine ring was anticipated to have different biological activity and properties. In all cases,
use of K CO or Cs CO in DMF resulted in > 10:1 ratio of the pyrazinone to pyrazine derivatives.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
3
2
3
3
6,37
Reaction with Ag CO in toluene and heating as described in the literature,
was briefly explored and
2
3
found to increase the ratio of the minor imidazopyrazine isomer, with pyrazinone:pyrazine ratios of 2:1
for 17g:18g and 17i:18i. Unfortunately, the Ag CO /toluene method resulted in unreacted starting
2
3
material and a greater number of side products compared to K CO or Cs CO in DMF. This complicated
2
3
2
3
the separation of 18g to the required >95% purity for biological assay. Therefore, isomers 17m and 18m
were prepared using K CO in DMF.
2
3
Nitroimidazopyrazines were readily distinguished from nitroimidazopyrazinones as they were more
lipophilic, and thus were more strongly retained on C18 silica and more weakly retained on normal phase
1
silica. Furthermore, in the H NMR the resonance due to the OCH of 18i and 18m was 0.4–0.6 ppm
2
further downfield than the corresponding resonance due to the NCH group in the pyrazinone scaffold
2
1
3
(
eg 18i/17i δ 5.64 vs 5.21 ppm). In the C NMR spectrum, the corresponding OCH resonance was also
2
shifted 17–19 ppm downfield for 18i and 18m compared to the corresponding NCH resonance (eg
2
1
8i/17i δ 62.7 vs 45.7 ppm).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 90
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 1. Synthesis of imidazopyrazin-ones/-es 17a-t, 18i and 18m by route 1) condensation and dehydration
via nitroimidazole carboxamide intermediates or route 2) alkylation of 17a. i) Bromoacetaldehyde diethyl acetal,
K CO , µW 180 °C, 69%–quant. yield, ii) 2M aq. HCl (10 vol), 1,4-dioxane (10 vol), µW 120 ºC, 42–87%, (17a:
2
3
5
% aq. HCl, reflux, 66%); iii) alkyl/benzyl halide, K CO /Cs CO , DMF, rt – µW 120 ºC or Ag CO /toluene, 80
2 3 2 3 2 3
ºC as detailed in the experimental, 3-84% iv) K CO , MeOH, rt, 45%. Compounds 18g-h and 18q were detected
2
3
but not isolated in pure form.
Des-nitro Derivatives
Des-nitro derivatives 20b-c were prepared as negative control compounds, essentially as described for
the nitrated series above (Scheme 2). It was hypothesized that these compounds should be biologically
inactive if the parent compounds had a mechanism of action involving reduction of the nitro group.
ACS Paragon Plus Environment

Page 11 of 90
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 2. Synthesis of des-nitro imidazopyrazinones 20b-c. i) (COCl) , cat. DMF, DCM, 0 °C  rt, then conc.
2
2
8
NH OH, 73% ii) SOCl , reflux, then benzyl amine, 69%; iii) bromoacetaldehyde diethyl acetal, K CO , µW 120–
4
2
2
3
180 °C, 81–92%; iv) For 20a: 5% aq. HCl, reflux, 50%; For 20b: 5% aq. HCl, 80 °C, 39%; v) 4-fluorobenzylamine,
K CO , DMF, rt, 19%.
2
3
2
3
Design and Synthesis of Nitroimidazopyrazin-ones/es with Modifications at R , R
2
3
We then explored different variations at the R and R positions, established a shortened route to
synthesize the imidazopyrazinone core and further investigated the activity of the imidazopyrazine
2
3
scaffold (Scheme 3). Both core scaffolds were prepared with three different combinations of R and R
groups in order to assess how functionalization at these sites could affect activity against different
1
organisms. Since benzyl groups at R were favorable for biological activity, derivatives 24a-e containing
2
3
1
R = Ph and R = H, but including polar substituents (CH CH OAc or CH CH OH) at R to counteract
2
2
2
2
2
the increase in compound lipophilicity, were prepared. The effect of a methyl group at R = CH , as a
3
2
small, non-polar substituent (derivatives 25a-m) was also assessed. For the series with R = CH , both
3
1
1
polar and non-polar substituents at R , including R = CH CH OAc, CH CH OH and several preferred
2
2
2
2
2
3
benzyl derivatives were investigated. Lastly, matched pair analogs with R = R = CH (26a-i) with the
3
1
preferred benzyl substituents at R were prepared.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 90
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
The synthesis of the imidazopyrazinone 24a (R = Ph) was initially performed as in Scheme 3 in a similar
manner as described earlier. Intermediate 16a was alkylated with 2-bromoacetophenone at room
temperature with K CO as a base catalyst to yield 27a, which was then purified by silica
2
3
chromatography. The condensation-dehydration cyclization of 27a was achieved with 2M aq. HCl and
,4-dioxane under microwave irradiation to form the phenyl substituted imidazopyrazinone core 24a. To
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
2
3
synthesize the imidazopyrazinone cores with R = CH (25a) and R = R = CH (26a), a more concise
3
3
synthesis was designed that involved a two-step, one-pot tandem alkylation-dehydration reaction
Scheme 3). For 25a and 26a, this procedure worked well. The conditions were mild as both steps of the
reaction (alkylation and condensation-dehydration) proceeded at room temperature and the products were
(
obtained in high purity after filtration and washes (H O and MeOH) of the precipitate (67% yield over
2
2
two steps). The one pot, two-step methodology was also tested for 24a (R = phenyl), although
conversion and yields were reduced (40% isolated yield with product of 57% purity: Abs% UV254 nm
LCMS) due to impurities that formed in both steps because of the different reactivity of the alkylation
and condensation-dehydration steps. In addition, product 24a was slightly soluble in the methanol used
in the wash step to remove more non-polar impurities. An intermediate work-up procedure, or
optimization of the initial alkylation conditions could be explored to improve this method for 24a.
Alkylation of imidazopyrazinones 24a, 25a and 26a under basic conditions provided both the N- and O-
alkylated regioisomers that were readily separated by chromatography. Alkylation conditions that used
heating in the presence of Cs CO generally favored the O-alkylated imidazopyrazine product. The
2
3
method employed here was sufficient to provide both possible isomers for biological evaluation, but
further optimization of the reaction conditions could be explored in the future to alter the ratio of N- to
O- alkylated product.38 Again, the acetate groups were removed using K CO /MeOH to give the
2
3
nitroimidazopyrazin-one/-es 24d-e and 25d-e.
ACS Paragon Plus Environment

Page 13 of 90
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
3
Scheme 3. Synthesis of nitroimidazopyrazinon-one/-es 24a-e, 25a-m and 26a-i with variations at R and R . i) 2-
Bromoacetophenone, K CO , DMF, rt, 60%; ii) 2M HCl, 1,4-dioxane, µW 120 ºC, 84%; iii) 25a & 26a: α-halo
2
3
ketone, K CO , DMF, rt; iv) 25a & 26a: 2M HCl, 1,4-dioxane, rt, 40–67% over two steps; v) K CO or Cs CO ,
2
3
2
3
2
3
DMF, rt – µW 80 ºC, 41–90% yield inclusive of both isomers; vi) K CO , MeOH, rt, 10–83% yield. Compound
2
3
25k was detected but not isolated.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 90
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Results and Discussion
Antiparasitic and Antimicrobial Activity of Pretomanid
Both pretomanid 6 and its enantiomer 8 were found to be ~2-fold more active than metronidazole 1
against G. lamblia (Table 1). Both enantiomers also had activity against E. histolytica, with 8 active at
comparable levels to 1, and 6 ~2-fold less active than 1. Compound 6 had moderate activity against C.
difficile (MIC = 4 µg/mL), which was 8-fold less potent than metronidazole 1 (MIC = 0.5 µg/mL) (Table
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
). These results expand the spectrum of action of 6, which was previously described to inhibit M.
tuberculosis, Leishmania spp. and Trypanosoma spp.16,26 The enantiomers 6/8 have been reported to have
1
6
more significant differences in activities against L. donovani, trypanosomes and M. tuberculosis. For
example, 8 was ~5-fold more active than 6 against L. donovani in both the promastigote and amastigote
1
6
(intracellular macrophage) assays. While 6 had weak activity against T. cruzi and T. b. brucei in the
parasite life stages relevant to mammalian infection, 8 had slightly enhanced activity, similar to the
16
39
results against L. donovani. In contrast, 8 was inactive against M. tuberculosis. Compound 8 has
previously been shown to bind to the M. tuberculosis Ddn enzyme involved in reductive activation of 6,
3
9
but it could not be turned over by the enzyme. These differences in selectivity of the enantiomers 6 and
8
likely indicate differences in the mode of action, respective targets, or uptake of compound by the
respective organisms. Nonetheless, the results suggest that 6 or newer derivatives may also be repurposed
toward enteric parasites. More potent compounds could also potentially be identified in the future by
screening a library of pretomanid analogs from the TB alliance/DNDi collections, given that more than
000 analogs have been prepared.19
1
ACS Paragon Plus Environment

Page 15 of 90
1
Journal of Medicinal Chemistry
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table 1. Activity of pretomanid (6) against enteric pathogens.
Antiparasitic
Antibacterial
Compound
IC50 (µM) (pIC50 ± SE)
MIC (µg/mL)
G. lamblia
E. histolytica
4.3 (5.4 ± 0.02)
9.3 (5.0 ± 0.03)
5.3 (5.3 ± 0.08)
C. difficile
7.2 (5.1 ± 0.02)
3.0 (5.5 ± 0.02)
3.2 (5.5 ± 0.08)
0.5
4
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Metronidazole (1)
Pretomanid (6)
(R)-PA-824 (8)
N.D
1
2
3
SAR of Bicyclic Nitroimidazoles with Variation at R , R = R = H
Given the broad activity of pretomanid 6, bicyclic nitroimidazoles 17a-t, 18i and 18m were tested against
a wide panel of organisms. These organisms included M. tuberculosis grown under aerobic (normoxic)
and non-replicating (hypoxic) conditions, G. lamblia and E. histolytica, L. donovani (intracellular
amastigote assay), T. b. brucei and C. difficile. To gain a clear understanding of the spectra of action,
compounds were also screened against representative ESKAPE bacterial pathogens S. aureus (ATCC
4
3300), E. coli (ATCC 25922), K. pneumoniae (ATCC 700603), A. baumannii (ATCC 19606) and P.
aeruginosa (ATCC 27853) and fungal pathogens, C. albicans (ATCC 90028) and C. neoformans (H99
type strain, ATCC 208821) at the Community for Open Antimicrobial Drug Discovery (CO-ADD).40
The cytotoxicity against mammalian liver HepG2 and kidney HEK293 cell lines was also evaluated.
Gratifyingly, several compounds showed potent activity against M. tuberculosis, G. lamblia and T. b.
brucei with the SAR discussed in detail below (Table 2). The N- and O-alkylated regioisomers had
varying profiles, with the nitroimidazopyrazinones 17a-t consistently lacking activity against L.
donovani, E. histolytica or C. difficile. However, nitroimidazopyrazine 18m was distinguished by its
activity against E. histolytica (IC = 7.8 µM, Supporting Information, Table S5) and also G. lamblia
5
0
(IC50 = 5.2 µM), despite no activity against M. tuberculosis compared to the imidazopyrazinone
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 90
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
derivative 17m. The screen against ESKAPE and fungal pathogens revealed that these compounds were
selective for M. tuberculosis, G. lamblia and T. b. brucei as they were not active against the other
organisms (MIC >32 µg/mL) (Supporting Information, Table S5). Importantly, the majority of the
compounds were not cytotoxic at the highest concentration tested (CC >100 µM) against mammalian
5
0
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
liver HepG2 and kidney HEK293 cell lines (CC > 100 µM) (Supporting Information, Table S5).
5
0
1
1
However, compounds 17b (R = CH Ph) and 17t (R = cyclohexyl) were moderately cytotoxic against
2
mammalian liver and kidney cell lines (CC = 98–123 µM) compared to compounds 17a and 17c-s
5
0
(CC50 >100 µM). Nitroimidazopyrazine derivatives 18i and 18m were also not cytotoxic against
mammalian cell lines (CC >100 µM). This activity profile suggests a specific mechanism of action in
5
0
M. tuberculosis, G. lamblia and T. b. brucei rather than a general non-specific toxic effect.
1
Nitroimidazopyrazinones 17b-i with a monosubstituted benzyl group at R displayed potent activity
against M. tuberculosis (MICnormoxia = 0.06 –1 µg/mL) at levels up to 2-fold better than pretomanid 6
(MICnormoxia = 0.25–0.5 µg/mL). There was a slight preference for substitution at the 3′ over the 2′ and 4′
positions of the phenyl group (cf. 17f-g with 17i and 17c-d). However, there was otherwise little
1
electronic effect; both electron donating (17g, R = CH (3-CH -Ph)) and withdrawing substituents (17h,
2
3
1
R = CH (3-CF -Ph)) had equal activity (MIC = 0.06 µg/mL). Poor solubility of 17j in the
normoxia
2
3
compound stock solutions was suspected to have contributed to the unexpected weak activity of this
disubstituted-fluorine derivative against M. tuberculosis and the other organisms. Replacement of the
phenyl group (17b) with a pyridyl moiety (17k) was unfavorable and resulted in a loss of activity against
M. tuberculosis (17k MICnormoxia >32 µg/mL). A methyl substituent at the benzylic position (17l) was
well tolerated (MICnormoxia = 0.5 µg/mL), but extension of the aromatic linkage with an ethyl bridge in
the phenethyl derivatives (17m-n) resulted in a 16–32-fold loss of activity against M. tuberculosis, with
bacteriostatic activity of MICnormoxia = 4-8 µg/mL compared to respective benzyl derivatives 17c-d with
1
MICnormoxia = 0.125-0.5 µg/mL. None of the derivatives with polar groups at R (17o-s) had activity
ACS Paragon Plus Environment

Page 17 of 90
1
Journal of Medicinal Chemistry
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
against M. tuberculosis (MICnormoxia >32 µg/mL). Lastly, compound 17t with an aliphatic cyclohexyl
1
group at R was also inactive (MIC
>32 µg/mL), indicating a preference for an aromatic group
normoxia
rather than non-specific lipophilic group.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
In general, compounds with activity against M. tuberculosis grown in normoxic conditions were also
active against non-replicating M. tuberculosis under hypoxic conditions, although generally 2–16-fold
less active. By comparison, isoniazid, a compound that inhibits cell wall synthesis in aerobically respiring
M. tuberculosis (MIC = 0.04 µg/mL), did not inhibit growth in the hypoxic conditions (MIC >5 µg/mL).
Metronidazole 1 was inactive (MIC >32 µg/mL) in this hypoxic-recovery assay, indicating the medium
was not sufficiently anaerobic to show activity of 1 against M. tuberculosis.
The structure activity relationships observed for G. lamblia were similar to M. tuberculosis and several
compounds (17c-d, 17f, 17l-n) were found to have improved activity (up to 4-fold) relative to
metronidazole 1 (IC = 1.6 to 3.5 µM cf. 1 IC = 7.2 µM). Interestingly, an expanded range of the
5
0
50
1
1
derivatives with lipophilic R groups had potent activity against G. lamblia. For example, 17m (R =
1
CH CH (4-OCF -Ph)) and 17n (R = CH CH (4-Me-Ph)) had similar activity to the benzyl-substituted
2
2
3
2
2
analogs 17c-d against G. lamblia (IC = 1.6–3.2 µM vs 3.4–3.5 µM, respectively). In addition, the
5
0
cyclohexyl derivative 17t was potent against G. lamblia at equivalent levels to the metronidazole 1
control (17t G. lamblia IC = 5.2 µM).
5
0
There were also similar trends in the SAR observed against T. b. brucei although with a number of
differences compared to M. tuberculosis and G. lamblia. For example, the activity of the benzyl series
1
7b-i against T. b. brucei was sensitive to the particular nature of the benzyl substitution. For example,
1
1
17c (R = CH (4-OCF -Ph)) and 17e (R = CH (4-F-Ph)) were active against T. b. brucei (IC = 1.4 and
2
3
2
50
1
1
2
.9 µM, respectively), but compounds 17b (R = CH Ph) and 17f (R = CH (3-OCF -Ph)) had reduced
2
2
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 90
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
activity at the highest concentrations tested (40 and 16 µM, respectively). Another difference observed
1
for T. b. brucei was an increased tolerance for polar substituents at R . For example, compounds with
polar ethyl acetate (17o) or morpholino groups (17q-r) displayed moderate activity against T. b. brucei
(IC = 2.9–6.5 µM), whereas against M. tuberculosis and G. lamblia these examples were essentially
5
0
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
devoid of activity. The unsubstituted imidazopyrazinone compound 17a with H at R displayed some
activity against T. b. brucei (97% inhibition at 40 µM) but was insufficiently active at 20 µM to determine
an IC value. The cyclohexyl derivative 17t also had good activity against T. b. brucei (IC = 1.4 µM),
5
0
50
similarly to G. lamblia. These examples demonstrate that it is possible to identify compounds with
selective antiparasitic activity, although all compounds with M. tuberculosis activity tended to display
some activity towards parasites. The selectivity observed for the particular organisms may be due to
differences in the compound activation or uptake of the compounds and is an area of ongoing
investigation.
1
2
3
Table 2. Activity of imidazopyrazinones and imidazopyrazinones with variation at R , R = R = H.
Compound
Molecular
weight
Antibacterial
MIC (µg/mL)
Antiparasitic
IC50 (µM) (pIC50 ± SE)d
No.
Core
R1
(g/mol)
M.
M.
G. lambliab
T. b. bruceic
tuberculosisa
tuberculosisa
normoxia
>32
hypoxia
>32
1
Metronidazole 1
171.2
359.3
180.1
270.2
354.2
284.3
288.2
354.2
7.2 (5.1 ± 0.02)
3.0 (5.5 ± 0.02)
>25 (<4.6)
>40 (<4.4)
19 (4.7 ± 2.4)
97% I @ 40 µM
73%I @ 40 µM
1.4 (5.9 ± 0.08)
>40 (<4.4)
Pretomanid 6
0.25-0.5
>32
1
7a
17b
7c
7d
7e
17f
A
A
H
>32
4-8
CH
2
Ph
0.5-1
0.5
5.0 (5.3 ± 0.05)
3.5 (5.5 ± 0.01)
3.4 (5.5 ± 0.03)
6.4 (5.2 ± 0.07)
1.7 (5.8 ± 0.03)
1
A
A
A
A
CH
CH
CH
CH (3-OCF
2
(4-OCF
3
-Ph)
-Ph)
(4-F-Ph)
-Ph)
1-4
1
2
(4-CH
3
0.125
1
1-4
1
2
2
2.9 (5.5 ± 0.41)
38% I @ 16 µM
2
3
0.125
0.5-2
ACS Paragon Plus Environment

Page 19 of 90
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Compound
Core
Molecular
weight
Antibacterial
MIC (µg/mL)
Antiparasitic
IC50 (µM) (pIC50 ± SE)d
No.
R1
(g/mol)
M.
M.
G. lambliab
T. b. bruceic
5.4 (5.3 ± 0.23)
5.3 (5.3 ± 0.18)
tuberculosisa
tuberculosisa
normoxia
hypoxia
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
7g
A
A
CH
2
(3-CH
3
-Ph)
284.3
338.2
0.06
70-90% I @
9.0 (5.0 ± 0.03)
7.1 (5.1 ± 0.03)
0
.125-8
µg/mL
1
7h
CH
2
(3-CF
3
-Ph)
0.06
70-90% I @
0
.06-16
µg/mL
8
17i
7j
7k
7l
17m
A
A
A
A
A
CH
CH
CH
CHMe(4-F-Ph)
CH CH (4-OCF
Ph)
2
(2-OCF
(2,4-F-Ph)
(2-pyridyl)
3
-Ph)
354.2
306.2
271.2
302.3
368.3
1
16-32
>32
8.2 (5.1 ± 0.03)
>50 (<4.3)
6.0 (5.2 ± 1.5)
>40 (<4.4)
1
2
>32
>32
4-6.3
>32
1
2
14 (4.8 ± 0.05)
2.1 (5.7 ± 0.03)
1.6 (5.8 ± 0.04)
9.7 (5.0 ± 0.18)
94% I @ 40 µM
44%I @ 40 µM
1
0.5
2
2
3
-
8* (70%I)
1
7n
A
CH
2
CH
2
(4-CH
3
-
298.3
4* (78%I)
>32
3.2 (5.5 ± 0.04)
>40 (<4.4)
Ph)
17o
A
A
A
CH
CH
CH
morpholine
CH CO-
morpholine
CH CONH
cyclohexyl
CH (2-OCF -Ph)
CH CH (4-OCF
Ph)
H37Rv, M. tuberculosis-normoxia primary screen at 32 µg/mL n = 3, M. tuberculosis-hypoxia primary screen at 32 µg/mL n = 1–3, MIC-
2
CH
CH
CH
2
OAc
266.2
224.2
293.3
>32
>32
>32
>32
>32
>32
65 (4.2 ± 0.03)
6.5 (5.2 ± 0.22)
87%I @ 40 µM
2.9 (5.5 ± 0.14)
1
1
7p
7q
2
2
OH
>50 (<4.3)
2
2
-
~50 (~4.3)
17r
A
2
307.3
>32
>32
>50 (<4.3)
4.1 (5.4 ± 6.8)
1
7s
7t
8i
18m
A
A
B
B
2
2
237.2
262.3
354.2
368.3
>32
>32
>32
>32
>32
>32
>32
>32
>50 (<4.3)
>40 (<4.4)
1.4 (5.9 ± 0.13)
>40 (<4.4)
N.D
1
5.2 (5.3 ± 0.03)
12 (4.9 ± 0.05)
5.2 (5.3 ± 0.03)
1
2
3
2
2
3
-
a
normoxia/hypoxia of active compounds n = 3–6. Isoniazid control M. tuberculosis-normoxia MIC = 0.04 µg/mL, M. tuberculosis-hypoxia
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 90
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
MIC >5 µg/mL; b WB; pentamidine control IC50 = 0.002 µM, diminazine aceturate IC50 = 0.062 µM, puromycin IC50 = 0.05 µM, n = 2; d
the percentage inhibition (%I) at the highest concentration tested is reported for compounds that were not sufficiently active to determine
an IC50. *Increasing concentrations of compound did not inhibit growth further.
c
Comparison of Bicyclic Antimicrobial Activity to Monocyclic 4- and 4(5)-Nitroimidazole
Carboxamide Analogs
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Given the activity of the bicyclic derivatives against M. tuberculosis, we screened a number of previously
28
reported 4-nitroimidazole carboxamides 28b-k and 4(5)-nitroimidazole carboxamides 16a, 16c, 16e
and 16f, along with intermediates 16b and 16d prepared for this study, against M. tuberculosis under
normoxic and hypoxic growth conditions. This would allow us to compare the SAR between the
monocyclic and bicyclic series. Note that the 2-nitro group of the imidazopyrazinones occupies the
equivalent position of the nitro group in the 4-nitroimidazole series (Figure 3). Interestingly, none of the
4- or 4(5)- monocyclic nitroimidazoles tested were active against M. tuberculosis grown aerobically at
3
2 µg/mL (Supporting Information, Table S6). Under hypoxic growth conditions, weak activity was
observed for 28j and 16c (60-75% inhibition at 32 µg/mL), but this also correlated with increased
cytotoxicity against the mammalian kidney cell line for these compounds (Supporting Information, Table
S6). Therefore, the rigid, bicyclic nature of 17b-f appears necessary for antitubercular activity, especially
against actively dividing M. tuberculosis cells, but is not essential for activity against other parasites. A
comparison of the activity of the 4- and 4(5)- nitroimidazole carboxamide series with that of the bicyclic
derivatives is summarized in Figure 3.
ACS Paragon Plus Environment

Page 21 of 90
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 3. 4-Nitroimidazoles 28b-k and 4(5)-nitroimidazole carboxamide 16b-f, compared to imidazopyrazinones
1
7b-f.
Evaluation of Des-nitro Derivatives for Biological Activity
Two des-nitro derivatives 20b-c were prepared as negative control compounds, to confirm that the
proposed mode of action relies on reduction of the nitro group. The des-nitro imidazopyrazinone
derivatives 20b-c were tested for activity and indeed, were found to be inactive against M. tuberculosis,
G. lamblia, T. b. brucei and E. histolytica (complete profiling results are detailed in the Supporting
Information, Table S7). This result is consistent with the inactivity of a des-nitro pretomanid analog 9
against M. tuberculosis under both aerobic and anaerobic growth (MIC >160 µM and >500 µM,
respectively),41 and supports a critical role for the nitro group in the mechanism of action of
nitroimidazopyrazinones.
2
3
SAR of Bicyclic Nitroimidazoles with Variation at R = Ph or CH , R = H or CH
3
3
2
3
Compounds 24a-e, 25a-m and 26a-i with R = Ph or CH and R = H or CH were evaluated for
3
3
antiparasitic and antimicrobial in the same manner as described for 17a-t, 18i, 18m (Table 3). Both the
2
3
R and R sites were found to influence activity against different pathogens to varying degrees.
2
3
Interestingly, the modifications R and R on the imidazopyrazinone series was found to be particularly
beneficial for activity against T. b. brucei. A number of imidazopyrazinones were active against T. b.
brucei between 0.2 and 0.9 µM (24a-b, 25b, 25f, 25h, 26b, 26d, 26f and 26h). While some of the
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 90
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
pyrazine matched pairs generally lost activity, several compounds (24c, 25m, 26e, 26g and 26i)
1
2
maintained activity in the 1.2–2.2 µM range. Furthermore, when R = CH CH OAc, compound 24b (R
2
2
2
2
=
Ph) and 25b (R = CH ) showed 7.6–10-fold improvement compared to 17o (R = H). This benefit was
3
1
1
also clear for the derivatives with benzyl substitutions at R . For example, compound 17f (R = CH (3-
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
2
OCF -Ph), R = H, R = H) showed incomplete inhibition at 16 µM, while the matched pairs 25j (R =
3
3
3
2
3
CH , R = H) and 26f (R = CH , R = CH ) inhibited T. b. brucei at 1.1 and 0.40 µM, respectively. This
3
3
3
2
3
last example also demonstrates a slight preference for methyl group substitutions at both R and R on
the pyrazinone ring, as this resulted in activity that was consistently about two-fold improved.
2
3
The methyl and phenyl groups explored at R and R did not result in any overall improvement or loss
of activity against G. lamblia, as the activity was generally in the same range. As observed for
imidazopyrazine 18m, the expanded series of imidazopyrazine derivatives also had activity against G.
2
3
lamblia. When R = Ph, R = H, the pyrazine derivatives 24c and 24e performed better than the
2
pyrazinone analogs 24b and 24d against G. lamblia. However, similar activity was observed when R =
3
2
3
CH , R = H (e.g. matched pair 25h and 25i) and R = CH , R = CH (e.g. matched pair 26d and 26e),
3
3
3
although the pyrazine was slightly worse by two-fold for a few derivatives.
2
Activity against E. histolytica was also improved for the imidazopyrazinone series when R = CH and
3
3
R = CH . Both imidazopyrazinones (e.g 26b and 26f) and imidazopyrazines (e.g 26c and 26g) were
3
active, with a slight preference for the imidazopyrazinone core. Despite this improvement, none of the
derivatives had activity comparable to metronidazole 1, with the active derivatives having only moderate
to weak activity (10–34 µM). This suggests distinct requirements for activity in E. histolytica compared
to T. b. brucei and G. lamblia. Nonetheless, it is possible that further modification of the scaffold may
lead to identification of an optimal “sweet spot” for activity against E. histolytica.
ACS Paragon Plus Environment

Page 23 of 90
1
Journal of Medicinal Chemistry
2
3
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Modifications explored at R and R were mostly detrimental for activity against M. tuberculosis.
2
3
Activity against M. tuberculosis activity was abolished when R = CH or Ph and R = H. Consistent with
3
the previous SAR study, all of the imidazopyrazines were found to be inactive (MIC >32 µg/mL).
2
Nonetheless, activity was retained for three imidazopyrazinone examples (26b, 26d and 26f) when R =
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
1
R = CH and R was a benzyl group. Compounds 26b, 26d and 26f also displayed activity against M.
3
tuberculosis under hypoxic growth conditions and against G. lamblia, E. histolytica and T. b. brucei.
These results indicate that it is possible to identify compounds with broad spectrum activity against both
M. tuberculosis and parasites.
Continuing the trend previously observed for compounds 17a-t, 18i and 18m, none of the additional
imidazopyrazin-ones/-es synthesized had appreciable activity against ESKAPE bacteria or fungal
pathogens (MIC ≥ 32 µg/mL) (Supporting Information, Table S8). Both the nitroimidazopyrazinones
and nitroimidazopyrazines series were also not cytotoxic against mammalian liver and kidney cell lines
(CC >100 µM, 24d >75 µM due to compound limitations), except for 26f that displayed moderate
5
0
cytotoxicity in both cell lines (CC = 26–80 µM) (Supporting Information, Table S8).
5
0
2
3
To summarize, the SAR of the R , R sites was determined for both the nitroimidazopyrazinone and
nitroimidazopyrazine series. Several compounds with potent activity against M. tuberculosis, G. lamblia
and T. b. brucei were identified. Further work confirming the target/mode of action in M. tuberculosis,
G. lamblia and T. b. brucei could aid in explaining the differences in activity. Additional studies to
measure the reduction potential may provide insight into the differences in the activity between the
nitroimidazopyrazinones and nitroimidazopyrazines.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 90
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
3
Table 3. SAR exploring the differences between the R and R groups for the imidazopyrazinone (core A) versus
the imidazopyrazine (core B).
Compound
Molecular
weight
Antibacterial
MIC (µg/mL)
Antiparasitic
IC50 (µM) (pIC50 ± SE)e
No.
Core
R1
R2
R3
(g/mol)
M.
M.
G. lambliab
E.
T. b
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
tuberculosisa
tuberculosisa
histolyticac
bruceid
normoxia
hypoxia
Metronidazole 1
Pretomanid 6
H
171.2
359.3
256.2
342.3
342.3
300.3
300.3
194.2
280.2
280.2
238.2
238.2
>32
>32
7.2 (5.1 ±
.02)
4.3 (5.4 ±
0.02)
>40 (<4.4)
0
0.25-0.5
>32
1
3.0 (5.5 ±
0.02)
9.3 (5.0 ±
0.03)
19 (4.7 ±
2.4)
2
4a
A
A
B
A
B
A
A
B
A
B
>32
>32
>32
>32
>32
>32
>32
>32
>32
>32
10 (5.0 ±
>25 (<4.6) 0.22 (6.7 ±
0.071)
0
.03)
2
4b
CH
2
CH
2
OAc
>32
11 (5.0 ±
0.05)
34 (4.5 ±
0.1)
0.86 (6.1 ±
0.085)
2
4c
>32
1.9 (5.7 ±
>50 (<4.3)
1.5 (5.8 ±
0.085)
Ph
H
0
.05)
2
4d
CH
2
CH
2
OH
>32
38 (4.4 ±
0.03)
>50 (<4.3)
>50 (<4.3)
>50 (<4.3)
41% I @
N.D.
2
4e
>32
5.2 (5.3 ±
1.1 (6.0 ±
0.12)
0
.05)
2
5a
H
>32
~50 (~4.3)
~50 (~4.3)
3.6 (5.4 ±
0.20)
2
5b
CH
2
CH
2
OAc
>32
0.65 (6.2 ±
0.021)
5
0 µM
2
5c
>32
21 (4.7 ±
0.06)
42% I @
50 µM
>40 (<4.4)
CH
3
H
2
5d
CH
2
CH
2
OH
>32
>50 (<4.3)
>50 (<4.3)
4.6 (5.3 ±
0
.092)
2
5e
>32
80% I @
50 µM
>50 (<4.3)
56% I @
40 µM
ACS Paragon Plus Environment

Page 25 of 90
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Compound
R1
Molecular
weight
Antibacterial
MIC (µg/mL)
Antiparasitic
IC50 (µM) (pIC50 ± SE)e
No.
Core
R2
R3
(g/mol)
M.
M.
G. lambliab
E.
T. b
tuberculosisa
tuberculosisa
histolyticac
bruceid
normoxia
hypoxia
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
5f
A
B
A
B
A
A
B
A
A
B
A
B
A
CH
2
(4-OCF
Ph)
3
-
368.3
368.3
298.3
298.3
368.3
368.3
368.3
208.2
382.3
382.3
312.3
312.3
382.3
>32
>32
5.8 (5.2 ±
0.04)
>50 (<4.3) 0.41 (5.4 ±
0.057)
25g
>32
>32
N.D.
>32
>32
>32
>32
1
>32
>32
N.D.
>32
>32
>32
>32
4
7.4 (5.1 ±
9.5 (5 ±
0.07)
>40 (<4.4)
0
.03)
2
5h
CH
2
(4-CH
Ph)
3
-
4.4 (5.4 ±
0.03)
10 (5.0 ±
1.0)
0.56 (6.3 ±
0.028)
25i
5.7 (5.2 ±
30 (4.5 ±
0.1)
N.D.
0
.03)
2
5j
CH
2
2
(3-OCF
Ph)
3
3
-
2.8 (5.6 ±
0.01)
16 (4.8 ±
0.1)
1.1 (5.9 ±
0.021)
25l
CH
(2-OCF
Ph)
-
3.3 (5.5 ±
0.01)
9.1 (5 ±
0.08)
2.1 (5.7 ±
0.29)
2
5m
5.7 (5.2 ±
>50 (<4.3)
1.2 (5.9 ±
0.13)
0
.02)
26a
H
~50 (~4.3)
>50 (<4.3)
2.6 (5.6 ±
0
.17)
2
6b
CH
2
(4-OCF
Ph)
3
-
4.2 (5.4 ±
0.04)
10 (5 ±
0.1)
0.24 (5.7 ±
0.0071)
26c
>32
2
>32
4
8.6 (5.1 ±
17 (4.8 ±
0.09)
10 (5.0 ±
0.20)
0
.04)
CH
3
CH
3
2
6d
CH
2
(4-CH
Ph)
3
-
5.2 (5.3 ±
0.02)
33 (4.5 ±
0.04)
0.25 (6.6 ±
0.0071)
26e
>32
1
>32
16-32
5.8 (5.2 ±
>50 (<4.3)
2.2 (5.6 ±
0.31)
0
.02)
2
6f
CH
2
(3-OCF
Ph)
3
-
1.7 (5.8 ±
0.01)
15 (4.8 ±
0.03)
0.40 (6.4 ±
0.0071)
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 90
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Compound
R1
Molecular
weight
Antibacterial
MIC (µg/mL)
Antiparasitic
IC50 (µM) (pIC50 ± SE)e
No.
Core
R2
R3
(g/mol)
M.
M.
G. lambliab
E.
T. b
tuberculosisa
tuberculosisa
histolyticac
bruceid
normoxia
hypoxia
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
6g
B
A
B
382.3
382.3
382.3
>32
>32
3.9 (5.4 ±
.01)
27 (4.6 ±
0.02)
1.4 (5.9 ±
0.24)
0
26h
CH
2
(2-OCF
Ph)
3
-
>32
>32
>32
>32
5.3 (5.3 ±
0.03)
18 (4.7 ±
0.02)
0.78 (6.1 ±
0.090)
2
6i
>50 (<4.3)
>50 (<4.3)
1.9 (5.7 ±
0
.27)
a
H37Rv, M.tuberculosis-normoxia primary screen at 32 µg/mL n = 3, M. tuberculosis-hypoxia primary screen at 32 µg/mL n = 1–3, MIC-
normoxia/hypoxia of active compounds n = 3–6. Isoniazid control M. tuberculosis-normoxia MIC = 0.04 µg/mL, M. tuberculosis-hypoxia
b
c
d
MIC >5 µg/mL; WB; HM1:IMSS; pentamidine control IC50 = 0.002 µM, diminazine aceturate IC50 = 0.062 µM, puromycin IC50 = 0.05
e
µM, n = 2; the percentage inhibition (%I) at the highest concentration tested is reported for compounds that were not sufficiently active to
determine an IC50
.
Microsomal Stability, Plasma Stability and Plasma Protein Binding
A set of ten compounds was selected for initial assessment of drug-like properties (microsomal stability,
plasma stability and plasma protein binding (PPB)) based on their potency and structural diversity (Table
4
). It was also envisioned that these studies would provide clarity as to whether there was a clear benefit
to either the nitroimidazopyrazinone or nitroimidazopyrazine series from the perspective of ADMET
properties. Nitroimidazopyrazinones were stable in human liver microsomes (HLM) after 2 h of
1
2
3
incubation, regardless of the structural differences in R , R and R . However, 18i, 25g and 26c from the
nitroimidazopyrazine series only showed moderate stability in HLM, independent of differences at R1,
2
3
R , and R , with 44-69% of compound remaining after 2 h (Table 4).
ACS Paragon Plus Environment

Page 27 of 90
1
Journal of Medicinal Chemistry
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
To further understand how the metabolic stability could impact in vivo efficacy studies, metabolic
stability was evaluated in CD-1 mouse liver microsomes (MLM) because the CD-1 mouse strain is
intended to be used for future in vivo efficacy studies. As for HLM, the majority of the
nitroimidazopyrazinones 17c-f, 17i and 25f displayed excellent stabilities toward MLM. However, one
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
3
of the imidazopyrazinone analogs, 26b (R = R = CH ) showed significant degradation after 2 h, with
3
only 3.6% of intact compound remaining (compared to 92% in HLM) (Table 4). Comparison of 25f with
3
26b, which differ by the presence of a hydrogen or methyl group at R respectively, indicated that the
methyl group at R3 was responsible for the high intrinsic clearance of 26b in MLM.
Nitroimidazopyrazines, 18i, 25g and 26c were also not stable in MLM (<0.5–8% remaining), suggesting
a metabolic liability of this series.
Compound stability in both human and mouse (CD-1) plasma was assessed for compounds 17c-f, 25f-g
and 26b-c (Table 4). All of the examples had >90% of compound remaining after 2 h at 37 °C, similar
to pretomanid 6. Therefore, both the nitroimidazopyrazinones and nitroimidazopyrazines series were
stable in human and mouse plasma, despite the instability of nitroimidazopyrazines observed in liver
microsomes.
Plasma protein binding affects compound bioavailability and tissue distribution in vivo. Therefore, to
understand PPB for the nitroimidazopyrazines and nitroimidazopyrazines series, selected potent
compounds (17c, 17e-f, 25f-g and 26b) were evaluated for PPB using the ultrafiltration method (Table
4). Most of the tested compounds, except 17e (82% bound) had high PPB at >95%, which might limit
the concentration of free compounds at the site of infection. However, as many approved and clinical
drugs, including 6, tend to have high PPB, this parameter is not recommended to be optimized in early
drug design.42
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 90
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table 4. In vitro physicochemical properties for representative compounds.
Compound
Core
R1
R2
R3
Microsomal stability
% remaining at 2 h)
Plasma stability
Plasma
(
(% remaining at 2 h)
protein
binding
(%)
Human
Mouse
CD-1)
Human
Mouse (CD-
1)
(
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Pretomanid 6
CH (4-OCF
CH (4-CH
CH (4-F-Ph)
97 ± 5.0
>99
92 ± 2.7
>99
96 ± 6.9
>99
96 ± 1.9
>99
97 ± 1.1
98 ± 0.4
N.D
1
7c
17d
7e
7f
7i
18i
5f
5g
6b
6c
A
A
A
A
A
B
A
B
A
B
2
3
-Ph)
2
3
-Ph)
97 ± 1.1
99
96 ± 6.0
>99
>99
>99
1
2
98 ± 1.4
>99
>99
82 ± 0.3
99 ± 0.2
N.D
H
H
1
CH
CH
CH
CH
CH
CH
CH
2
2
2
2
2
2
2
(3-OCF
(2-OCF
(2-OCF
(4-OCF
(4-OCF
(4-OCF
(4-OCF
3
3
3
3
3
3
3
-Ph)
-Ph)
-Ph)
-Ph)
-Ph)
-Ph)
-Ph)
>99
>99
97 ± 2.2
N.D
1
>99
>99
N.D
51 ± 18
>99
<0.5
N.D
N.D
N.D
2
>99
92 ± 4.1
94 ± 4.7
97 ± 1.4
>99
99 ± 4.8
>99
98 ± 0.4
>99
CH
3
3
H
2
69 ± 4.3
92 ± 1.1
44 ± 0.98
7.9 ± 4.0
3.6 ± 0.1
<0.5
2
>99
99 ± 0.1
N.D
CH
CH
3
2
>99
Values are presented as mean of three replicates ± SD. N.D represents not determined. Microsome stability verapamil control = 2% (MLM),
% (HLM) remaining at 30 min; plasma stability eucatropine control = 28% (mouse), 21% (human) remaining at 2 h; plasma protein
9
binding sulfamethoxazole control = 68% bound.
Caco-2 Intestinal Epithelium Permeability
An in vitro Caco-2 monolayer bidirectional assay was used to evaluate intestinal epithelium permeability
43
and to predict the oral absorption of the nitroimidazopyrazinone series. Three derivatives (17c, 17f and
2
6b) were selected based on potency against M. tuberculosis, G. lamblia and T. b. brucei and to permit
direct comparison of R = R = H (17c) versus R = R = CH (26b). Pleasingly, compounds 17c, 17f and
2
3
2
3
3
26b were highly permeable (Table 5). The apparent permeability coefficient (P ) apical to basal (A to
app
-6
B) had values of >20 × 10 cm/s, similar to pretomanid 6 and propranolol, with the latter used as a
positive control for high permeability. No metabolism of the compounds by Caco-2 cells was observed.
This was evident from the high recovery of tested nitroimidazopyrazinones obtained in both directions,
ACS Paragon Plus Environment

Page 29 of 90
1
Journal of Medicinal Chemistry
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
similar to the control propranolol (A to B measured). Nitroimidazopyrazinones 17c, 17f and 26b and 6
also showed a low efflux ratio of <2 (cf. digoxin efflux ratio >360 as it is a substrate for P-gp). Therefore,
the tested bicyclic nitroimidazoles do not appear to be substrates of efflux transporters. These results
suggest that 17c, 17f and 26b have desirable therapeutic properties and may display good oral absorption
properties in vivo.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Table 5. Caco-2 permeability of selected analogs.
-6
Compound
Core
R1
R2
R3
P
app (10 cm/s)
Mean recovery (%)
Efflux ratio
A to B
29
B to A
22
A to B
81
A to B
101
N.D
N.D
88
Pretomanid 6*
0.74
N.D
N.D
>360
0.88
1.0
Fenoterol
Propranolol
Digoxin
0.27
21
N.D
N.D
8.4
92
70
<0.020
25
<72
83
1
7c
7f
26b
A
A
A
CH
2
2
(4-OCF
(3-OCF
3
3
-Ph)
-Ph)
H
H
H
H
22
97
1
CH
24
24
88
99
CH
2
(4-OCF
3
-Ph)
CH
3
CH
3
21
20
83
97
0.95
N.D represents not determined. * Comparable results with the literature.22
Kinetic Solubility
Compound solubility impacts in vivo absorption, efficacy and dosing and is thus an important parameter
to optimize in drug development. For example, the recently approved TB drug, delamanid has limited
44
solubility which requires twice daily dosing, and new bicyclic analogs with improved solubility would
be of great interest. According to the generic criteria for hits and leads in infectious diseases identified
by Japanese Global Health Innovative Technology (GHIT) Fund and its key partners, a lead should have
acceptable physicochemical properties, with solubility at least >10 μM in phosphate-buffered saline pH
7
.4 (PBS).45
ACS Paragon Plus Environment