Release of nitrite from the antitubercular nitroimidazole drug PA-824 and analogues upon one-electron reduction in protic, non-aqueous solvent

Article abstract of DOI:10.1039/b915877d

The one-electron reduction chemistry of the antituberculosis drug PA-824, together with a series of closely related compounds, has been investigated in irradiated anaerobic propan-2-ol solution. The protic solvent, of low dielectric constant, was chosen to mimic the environment of a water-restricting active site of a model protein, which is capable of reducing the compounds. Radiolytic reduction of the compounds containing electron donating substituents in the 2-position of the imidazole ring released nitrite, with compounds that are highly active against Mycobacterium tuberculosis exhibiting high yields of nitrite. The release of cytotoxic reactive nitrogen species through a one-electron pathway, by as yet unidentified proteins, may play a role in the activity of this class of compounds against TB. The described radiolytic quantification of nitrite release may have utility as a preliminary screening test for nitroaromatic candidate drugs against the disease.

Full text of DOI:10.1039/b915877d

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Release of nitrite from the antitubercular nitroimidazole drug PA-824 and
analogues upon one-electron reduction in protic, non-aqueous solvent†
Andrej Maroz,a Sujata S. Shinde,a,b Scott G. Franzblau,c Zhenkun Ma,d William A. Denny,b Brian D. Palmerb
and Robert F. Anderson*a,b
Received 4th August 2009, Accepted 29th September 2009
First published as an Advance Article on the web 4th November 2009
DOI: 10.1039/b915877d
The one-electron reduction chemistry of the antituberculosis drug PA-824, together with a series of
closely related compounds, has been investigated in irradiated anaerobic propan-2-ol solution. The
protic solvent, of low dielectric constant, was chosen to mimic the environment of a water-restricting
active site of a model protein, which is capable of reducing the compounds. Radiolytic reduction of the
compounds containing electron donating substituents in the 2-position of the imidazole ring released
nitrite, with compounds that are highly active against Mycobacterium tuberculosis exhibiting high yields
of nitrite. The release of cytotoxic reactive nitrogen species through a one-electron pathway, by as yet
unidentified proteins, may play a role in the activity of this class of compounds against TB. The
described radiolytic quantification of nitrite release may have utility as a preliminary screening test for
nitroaromatic candidate drugs against the disease.
Introduction
Mycobacterium tuberculosis (Mtb), the causative agent of tuber-
culosis (TB), kills approximately two million people a year and,
in its latent form, infects one third of humanity.1 The latent
form is thought to sustain the pandemic as non replicating Mtb
are much more resistant to chemotherapeutics.2 Host cells can
impose immunity through the bacteriostasis effects of endogenous
reactive nitrogen species (RNS),3 reactive oxygen species (ROS)
and nutrient depravation, all of which can drive Mtb into its non
replicating state, a state also reached in hypoxia. As most front-
line drugs are used to kill replicating Mtb, the non replicating
state is resistant. This has necessitated the use of multidrug
combinations over extended treatment times and non-compliance
in such treatment regimes has given rise to the emergence of drug-
resistant tuberculosis.4 Hence there is an urgent need to develop
new drugs to fight the disease.
The bicyclic nitroimidazoles 1 (PA-824) and 9 (OPC-67683),
Scheme 1, are new candidate drugs for treating TB,5,6 and are
undergoing clinical trials. The drugs are active against both
replicating and non-replicating forms of the disease and appear
to act in an atypical manner compared to related nitroimidazoles.
The 5-nitroimidazole metronidazole (13), for example, has modest
activity against the disease only in hypoxia7 and this activity is
Scheme 1
thought to arise from reduction of the nitro group to bioactive
intermediates. Nitroimidazoles and other nitroarenes, generally
undergo sequential reduction to initially form the nitro radical
aDepartment of Chemistry, The University of Auckland, Private Bag
anion, which under oxic conditions undergoes redox cycling to
form ROS. Further reduction of the radical anion in hypoxia, or
direct two-electron (hydride ion) reduction, leads to the formation
of nitroso and on to hydroxylamine cytotoxins.
The major metabolites of 9 (by Mycobacterium bovis) and
1 (by Mtb), determined under aerobic and hypoxic conditions
respectively, are their des-nitro compounds.5,8 Analysis of inter-
mediates produced, by both chemical and radiolytic methods,
indicate that reduction of the imidazole ring occurs in preference
92019, Auckland, 1142, New Zealand. E-mail: r.anderson@auckland.ac.nz;
Fax: +64 9 3737422; Tel: +64 9 3737599 ext. 88315
bAuckland Cancer Society Research Centre, The University of Auckland,
Private Bag 92019, Auckland, 1142, New Zealand
cInstitute for Tuberculosis Research, College of Pharmacy, University of
Illinois at Chicago, 833 South Wood Street, Chicago, Illinois, 60612, USA
dGlobal Alliance for TB Drug Development, 40 Wall Street, New York,
10005, USA
† Electronic supplementary information (ESI) available: NMR spectra for
compound IV. See DOI: 10.1039/b915877d
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 413–418 | 413
©

View Article Online
to reduction of the nitro group.8,9 The F420 deazaflavin-dependent
protein Rv3547 has been shown to metabolise 1 in hypoxia10 to
form products, which include the des-nitro compound and the
release of RNS.8 It is proposed that a hydride transfer from the
reduced cofactor to the C3-position of the imidazole ring takes
place as the initial step.8 These authors propose Nef-type chemistry
for the release of nitric oxide (NO), a reaction pathway requiring
highly acidic conditions which is unknown for nitroheterocycles
in solution.11 The protonation of intermediates formed in the
active sites of proteins often occurs by proton bridging from
neighbouring amino acids and hence a ‘biological equivalent’ of
Nef-type chemistry can be invoked. In this regard, the active site of
such proteins can be described as a protic environment, even in the
absence of solvent water or with there being only a few molecules
of water present in the active site of the protein. The purified
F420 deazaflavin-dependent nitroreductase metabolises 1 in vitro
to yield RNS, as detected by the Griess reagent, and these are
suggested to be monomeric hyponitrous acid (HNO) and nitrous
acid (HNO2)8 Nitrous acid is unstable in solution and readily
breaks down to form NO, and the nitrite anion is metabolised in
cells by a number of processes, also leading to the formation of
NO.12
Fig.
1
Time resolved spectra observed following pulse radiolysis
(ca. 5 Gy in 200 ns) of 10 (0.5 mM) in deaerated propan-2-ol–water
(95 : 5% volume) measured 100 ms (᭹), 8 ms (᭺) after the electron pulse.
Insets: time dependent change of transient absorption at 450 nm.
In this study we examine a series of nitroimidazoles related
to PA-824 for possible release of RNS in a protic solvent (95%
propan-2-ol, 5% water) as a model for a water-restricted active site
of proteins which carry out one-electron reduction of substrates,
by using the a-hydroxyprop-2-yl radical ((CH3)2C∑OH) as the
reductant. This solvent has previously been used to examine
the conversion from outer sphere electron transfer in water to
inner sphere electron transfer on changing the solvent, in the
one-electron reduction of 4-nitrobenzonitrile.13 Our findings are
compared with the cytotoxicity of the compounds against Mtb
under both hypoxic and aerobic conditions to give some insight
into an alternative/additional one-electron reduction pathway of
how these drugs might be reduced to yield RNS.
Results and discussion
Pulse radiolysis studies
Time-resolved spectral changes following the pulse radiolysis of
10 in deaerated 95% propan-2-ol–5% water solutions containing
10 (0.5 mM), are presented in Fig. 1. The initial, low-intensity,
absorption spectrum of the a-hydroxyprop-2-yl radical (l <
360 nm) gave rise to the spectrum built-up over 100 ms, which
subsequently decayed to the absorption spectrum measured 8 ms
after the pulse. Both these spectra, measured at 100 ms (corrected
for its partial decay during formation) and at 8 ms after the pulse,
exhibit bleaching of the parent compound, 10, in the 330–370 nm
region. The rate of build-up in absorption of the band at 450 nm
increased with the concentration of 10, Fig. 2A, giving a calculated
rate constant of 9.1 0.3 ¥ 107 M-1 s-1. The solvated electron,
produced on the ionization of the solvent and observed at 600 nm,
decayed quickly on reaction with 10 (0.5 mM to 2.0 mM) on the
microsecond timescale, with a rate constant of 5 ¥ 109 M-1 s-1
(data not shown) but did not give rise to an intermediate which
absorbs in the region l > 270 nm. The transient spectrum (Fig. 1)
measured at 100 ms after the pulse has some similarities to that
previously measured in water, such as bleaching of the ground-
Fig. 2 Kinetic data obtained on the pulse radiolysis of 10 in deaerated
propan-2-ol–water (95 : 5% vol): (A) Dependence of the rate of formation
of the intermediate absorbing at 450 nm on the concentration of 10: (B)
Dependence of the reciprocal of the first half-life of the decay of radicals
absorbing at 450 nm on the initial radical concentration formed with
increasing radiation dose.
state absorbance in the 320–380 nm, however, the absorption band
in the 400–500 nm region, relative to the bleaching region, is much
more intense.
The transient species measured at 100 ms was observed to
decay by a mixed-order kinetic process and was analysed (at
450 nm) as previously described14 to ascertain both the first-order
rate constant of 55 70 s-1 and second-order rate constants of
5.6 0.2 ¥ 108 M-1 s-1; Fig. 2B. The first-order rate constant is
possibly that for the elimination of nitrite (as nitrous acid) from
II, Scheme 2, in competition (under pulse radiolysis conditions)
with a second-order decay process. The elimination of nitrous acid
in this way would lead to the transient formation of aryl radical,
414 | Org. Biomol. Chem., 2010, 8, 413–418
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
compound 12 (2-OCH3) also releases a high concentration of
nitrite upon reduction. The maximum G-value corresponds to
the ca. 75% G-value of total reducing equivalent in irradiated
propan-2-ol. The yield of nitrite was found to be unchanged
when irradiations were carried out in propan-2-ol–water mixtures
(95 : 5% vol).
Steady-state radiolysis, UV-vis spectra and mass analysis
of product
The reduction of PA-824 (1) upon its radiolysis in deaerated
propan-2-ol was monitored using UV-vis spectrophotometry in
a closed system containing an integral spectrophotometer cell.
Isosbestic points were maintained up until the last four spectra,
Fig. 4, enabling the G-value for the loss of 1 to be determined.
This was done by converting the absorption at 319 nm into
concentration and using the initial slope of a plot of the loss
of concentration against radiation dose (inset of Fig. 4). The
obtained G-value of 0.15 mmol J-1 corresponds to ca. 75% of
the estimated concentration of reducing equivalents produced in
irradiated propan-2-ol (Experimental section). The more soluble
methyl ether analogue of PA-824, 10, was used to obtain sufficient
product for chemical analysis. This compound was similarly
irradiated in deaerated propan-2-ol (5.2 mg in 25 mL) and the
decrease in its absorption band centred at 340 nm was followed
with increasing radiation dose.
Scheme 2
III, which could abstract H-atom from the solvent to give the
observed product, IV (see below).
Steady-state radiolysis, production of nitrite
All compounds were subjected to steady-state radiolysis in deaer-
ated propan-2-ol solution and the irradiated solutions analysed
for nitrite using the Griess reagent.
Dose response plots were constructed for increasing radiation
doses, Fig. 3, from which the yield of nitrite produced, (G-value
in mmol J-1), was determined for each compound. A wide range
in G-values, from 0 up to 0.15 mmol J-1, were determined for
analogues of PA-824 with E(1)≤ -500 mV. This is most likely, in
electronic terms, related to the structural necessity of possessing
an electron-donating substituent in the 2-position, as the model
Fig. 4 Changes in UV-vis absorption spectra, observed during the
g-irradiation (absorbed doses 0–500 Gy) of N2-saturated propan-2-ol
solutions of 1 (50 mM). Inset : Plot of absorbance of 1 monitored at
lmax 319 nm as a function of absorbed dose.
The radiation was stopped at the point where this absorption
band was completely lost (ca. 7.3 kGy) and the irradiated
sample was then subjected to analysis. Thin layer chromatography
indicated clean formation of a single product, as detected under
UV light, which was more polar than the starting 10. This product
was purified by column chromatography on silica, to give the
desnitro compound IV (R = OCH3) as a yellow oil, which
accounted for approximately three-quarters of the mass of the
starting material. The rest of the material was presumably either
broken down into non-aromatic compound(s) by the high dose
of radiation, or could not be isolated by the chromatographic
procedure. High resolution mass spectrometry of the isolated
product gave a molecular weight of 212.116, consistent with a
Fig. 3 Representative plots of the yield of nitrite upon irradiation of
compounds (50 mM) in deaerated solutions containing (i) propan-2-ol; 1,
᭢; 3, ꢀ; 6, ꢀ; 7, ᭜; 10, ꢀ; 11, ᭹; 13, ꢁ; 14, ᭺ and (ii) propan-2-ol–water
(95 : 5% volume), 1, ꢀ.
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 413–418 | 415
©

View Article Online
Table 1 Yield of nitrite formed upon radiolytic reduction of compounds in deaerated propan-2-ol; formation of radical adduct; and comparison with
the Minimum Inhibitory Concentration/mM, determined under aerobic (MABA, MIC) or anaerobic (LORA, MAC) conditions
E(1) pH 7/
mV
+13 ion m/zc
MABAa,b
aerobic
LORAa,b
anaerobic
-
Compound
G(NO2 )/mmol J-1
g mol-1
MICaaerobic
MACaanaerobic
1, PA-824
-534 7e
-534 6b
-476 6b
-488 6b
-570 6b,g
-568 6b,h
0.12 0.03
0.05 0.01
Yes
Yes
No
No
Yes
Yes
No
No
Yes
Yes
No
Yes
Yes
1.1
1.1
> 128
> 128
126
> 128
> 128
> 128
0.03f
105f
4.4
15
92
0.80d
1.2d
12d
25d
2
3
0
0
> 100i
> 100i
25i
50–100i
25i
4
89
5
0.04 0.01
0.15 0.02
0
> 128
> 128
> 128
120
250i
130d
6
0.80d
7
-500
-338
6
6
8
0
9, OPC-67683
-526 6f
-527 6e
-557 8f
-561 7f
-516 7f
0.14 0.03
0.13 0.02
0
0.14 0.02
0.025 0.005
2.54f
10
> 128f
—
11
—
12
121f
> 128f
~ 64f
13, metronidazole
ꢀ 64f
> 300j
62.5j
a MIC, minimum inhibitory concentrations towards Mycobacterium tuberculosis, MABA, microplate-based assay; LORA, low-oxygen-recovery assay;
MIC, minimum aerobic inhibitory concentration; MAC, minimum anaerobicidal concentration. b Data from reference 18; MIC for the lowest
concentration effecting an inhibition of >90%. c Mass spectrometry evidence for the formation of des-nitro-propan-2-ol adduct in anoxia. d Data
from reference 8; MIC for the concentration that will inhibit 99% of growth of inoculums. e Data from reference 9. f This work. g Data for 6-OH analogue
(without side-chain). h Data for 6-OCH3 analogue (without side-chain). i Data from reference 30. j Data from reference 16.
formula of C10H16N2O3. This corresponds to loss of the nitro
group from 10, and the addition of one propan-2-ol molecule.
The incorporation of solvent was confirmed by the 13C NMR
spectrum, which contained resonances for 10 carbon atoms, three
more than the starting material. In the 1H and 13C NMR spectra,
the chemical shifts of the oxazine ring proton and carbon atoms
were little shifted from those of the starting material, suggesting
that this ring remained unaltered. In the 1H NMR spectrum of IV
(R = OCH3), a one-proton singlet at d 6.25 ppm indicated that
a single hydrogen atom remained on the imidazole ring, with its
upfield value compared with that in the starting material being
consistent with loss of the nitro group. Two new, three-proton
singlets at d 1.44 and 1.41 ppm result from the methyl groups of the
incorporated propan-2-ol. A new D2O-exchangeable one-proton
singlet at d 4.95 ppm results from the tertiary alcohol group,
indicating that the propan-2-ol has been linked via a carbon atom,
rather than through oxygen. Complete chemical shift assignments
for the molecule were made following COSY, HMBC and HSQC
NMR experiments, and are consistent with the proposed structure.
Assignment of the site of attachment of the propan-2-ol as being
at the 5-position of the imidazole ring was made following NOE
experiments performed in a ROESY experiment. The resonances
assigned to the methyl groups and alcohol of the propan-2-ol
side chain showed correlations to those of the oxazine methylene
group attached to the imidazole ring, indicating that they are close
together in space. In contrast, the singlet at d 6.25 ppm resulting
from the remaining imidazole ring proton showed a correlation
only to the methyl groups of the adjacent propan-2-ol side chain.
Similar irradiation of PA-824 (1) gave an analogous product
IV (R = OCH2C6H4-4-OCF3) in good yield, which was fully
characterised by proton and carbon NMR analyses. Irradiation of
OPC-67683 (9) under these conditions was slightly less clean, but
appeared to give an analogous compound as the major product,
as evidenced by the appearance of a characteristic singlet at d
6.23 ppm in the 1H NMR spectrum of the crude product, resulting
from the imidazole ring proton. In addition, all irradiated samples
were analysed by low resolution mass spectrometry for loss of
each parent compound and formation of their major radiolytic
products. Loss of the nitrite from each compound and formation
of their desnitro-propan-2-ol adduct was scored as a change in
mass of +13 g mol-1, Table 1.
Conclusions
Nitrite is released as an RNS following one-electron reduction of
the nitroimidazole compounds examined in this study in a protic
solvent of low dielectric constant. Such conditions may well mimic
the environment of the active site of some enzymes, which carry out
one-electron reduction of substrates, when access of the solvent is
restricted by, for example, conformational changes in the protein
structure upon substrate binding. Compound 10, PA–824 (1) and
OPC-67683 (9) yield ca. 75% of stoichiometric amounts of nitrite
following their one-electron reduction in anaerobic propan-2-ol.
The G-values for the loss of compounds 1 and 9 equal the nitrite
yield of ca. 0.15 mmol J-1 (Table 1). As we directly observe the
reaction of the a-hydroxyprop-2-yl radical with the compounds,
we can assume that the release of nitrite is associated with this
reaction and that the scavenging of the solvated electron, to form
the radical anion, does not release nitrite quantitatively. It is
possible that the radical anion reforms the parent compounds,
at ca. 50% yield, through a disproportionation reaction, with
this reaction producing a product and nitrite, Scheme 2. This
would lead to the loss of parent compound equalling the yield
of nitrite release at ca. 75% of the total yield of reducing radicals
produced upon irradiation of the solvent (0.15 mmol J-1 compared
to 0.20 mmol J-1). The production of nitrite is accompanied, under
steady-state radiolysis conditions, by the formation of an adduct
between their des-nitro forms and propan-2-ol. This is consistent
with the a-hydroxyprop-2-yl radical adding to the 3-position of the
imidazole ring, followed by the elimination of nitrous acid to form
an aryl-type radical, III, which abstracts a H-atom from the solvent
to form the product IV, Scheme 2. This mechanism of reduction
is substantially different from that for nitrobenzenoid compounds
in the same solvent, as exemplified by 4-nitrobenzonitrile. In this
416 | Org. Biomol. Chem., 2010, 8, 413–418
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
case the a-hydroxyprop-2-yl radical adds to the nitro group and
quickly eliminates forming acetone and the nitro radical anion,13
which we have determined does not lead to formation of nitrite.
High yields of nitrite are seen for the analogue of 1 with-
out the lipophilic sidechain, compound 10, the 2-methoxy-4-
nitroimidazole model compound 12, and compound 6, which
has the electron donating NH moiety in the 2-position of the
4-nitroimidazole ring. Other analogues of 1, which also possess
electron donating moieties in the 2-position, 2 and 5, also produced
nitrite upon reduction, but at much lower concentrations.
Column chromatography was carried out on silica gel, (Merck
230–400 mesh). PA-824 (1),17 compounds 2–8,18 OPC-67683 (9)19
compounds 10 and 12,9 1-methyl-4-nitro-1H-imidazole (12)20 were
synthesised as described. Metronidazole (13) was purchased from
Sigma.
2-(6-Methoxy-6,7-dihydro-5H-imidazo[2,1-b][1,3]oxazin-3-yl)-2-
propanol (IV)
A deaerated solution of the oxazine (10) (5.2 mg, 0.025 mmol)
in propan-2-ol (25 mL) was irradiated to an absorbed dose of
7.2 kGy. The solution was evaporated to dryness in vacuo and the
residue was chromatographed on silica. Ethyl acetate eluted fore
fractions, then ethyl acetate–methanol (9 : 1) eluted the product
(IV, R = OCH3) as a yellow oil (3.9 mg, 73%). 1H NMR (DMSO-
d6) d 6.25 (s, 1H, H-2), 4.95 (s, 1H, OH), 4.48 (dt, J = 13.3, 2.2 Hz,
1H, H-5), 4.40 (dt, J = 11.8, 2.7 Hz, 1H, H-7), 4.21 (br d, J =
11.8 Hz, 1H, H-7), 4.08 (dd, J = 13.3, 3.5 Hz, 1H, H-5), 3.87
(m, 1H, H-6), 3.33 (s, 3H, OCH3), 1.44, 1.41 (2 s, 6H, CH3). 13C
NMR d 148.6 (C-8a), 132.7 (C-3), 117.4 (C-2), 68.5 (C-6), 66.3 (C-
OH), 65.9 (C-7), 55.5 (OCH3), 45.1 (C-5), 29.7 (CH3), 29.4 (CH3).
HREIMS found: M = 212.1160. C10H16N2O3 requires 212.1161.
Activity against Mtb does qualitatively correlate with the pro-
duction of nitrite, for compounds possessing the protein-targeted
lipophilic side-chain, Table 1. (The reason for the disparity in
the reported activities of 6 is unknown). However, the fact that
some of the compounds are active under hypoxia, even when
no nitrite is produced, means that other mechanisms operate as
well in hypoxia. It has been reported that hydride-transfer from
the F420 cofactor leads to RNS formation8 and this two-electron
pathway could be the major route to produce cytotoxic RNS under
hypoxia. It is of interest that metronidazole, a 5-nitroimidazole,
can release a low yield of nitrite upon one-electron reductions
under the conditions of this present study. Conventionally, the
anaerobic activity of metronidazole against protozoa and bacteria
is thought to arise from the formation, and subsequent reactions,
multi-electron reduced intermediates, such as its hydroxylamine
derivative.15 Recently a 5-nitroimidazole with aerobic activity has
been reported. This same study also found that some of PA-
824-related compounds tested retained anaerobic sensitivity in
mutants lacking the F420-dependent nitroreductase activity.16 This
observation implies that other pathways are also operable in
activating these compounds to cytotoxins which are active against
Mtb. The procedures described in this paper may have utility as
a preliminary screening test for potential nitroaromatic candidate
drugs, which can be activated by a one-electron pathway, in the
fight against TB.
Methods
Pulse radiolysis experiments were carried out at room temperature
(22
1
◦C) using the University of Auckland’s 4 MeV linear
accelerator, to deliver typical absorbed doses of 2.5–5 Gy in 200 ns.
The optical detection system and method of dosimetry have been
described previously.21 Steady-state radiolysis experiments were
performed using a 60Co g-source delivering a dose rate of 7 Gy
(J kg-1) min-1. Samples for irradiation were evacuated and purged
with O2-free N2 gas in glass tubes for three cycles. Nitrite was
determined using the Griess reagent.22 Reagents, sulfanilic acid
(0.1%), 0.1 mL, HCl (0.2 M), 2 mL, N-1-naphthalenediamine
dihydrochloride (0.2%), 0.1 mL and water, (1 mL) were added to
test samples (1 mL). Colour was allowed to develop for 15 min
in the dark and then read at 545 nm (e = 28340 M-1 cm-1) for
comparison with standard solutions of sodium nitrite treated in
the same way.
Experimental
Synthesis
All reagents used were of analytical grade. Sodium formate,
sodium hydroxide, perchloric acid and phosphate buffers were
obtained from Merck and potassium thiocyanate from Riedel-de
Haen. All other reagents were obtained from Aldrich Chemical
Company. All solutions were prepared in water purified by the
Millipore “Milli-Q” system. Solution pH values were adjusted
using the phosphate salts (5 mM) and either NaOH or HClO4
when necessary. Analyses were carried out in the Microchemical
Laboratory, University of Otago, Dunedin, NZ. Melting points
were determined on an Electrothermal 2300 Melting Point ap-
paratus. NMR spectra were obtained on a Bruker Avance 400
spectrometer. Spectra are referenced to Me4Si. Chemical shifts
and coupling constants were recorded in units of ppm and Hz,
respectively. High resolution mass spectra were determined on a
VG-70SE mass spectrometer using an ionizing potential of 70 eV
at a nominal resolution of 1000. Atmospheric pressure chemical
ionisation mass spectra (APCI-MS) were determined for methanol
elutions on a ThermoFinnigan Surveyor MSQ spectrometer.
Solutions in organic solvents were dried with anhydrous Na2SO4.
One-electron reduction potentials, E(1)
The one-electron reduction potentials of compounds 1, 9, 11–
13, E(A/A∑-), vs. NHE, were determined at pH 7.0 in aqueous
solution (5 mM phosphate buffer) by establishing redox equilibria
between three mixtures of the one-electron reduced compounds
∑
and the reference compounds tetraquat (E(TeQ2+/TeQ+ ) = -635
∑
7 mV)23 or triquat (E(TQ2+/TQ+ ) = -548
7 mV).24 The
radiolytically-produced e- species was used as the primary
aq
reductant, with the oxidizing primary radicals (∑OH and H-atom)
being scavenged by added 2-methylpropan-2-ol (0.1 M) to form an
inert radical. The DE values were calculated from the equilibrium
constants, Ke (established within 50 ms), using the Nernst equation.
Ke
TeQ+⋅ /TQ+⋅ + A
TeQ2+ /TQ2+ + A⋅−
ꢀꢀꢀꢀꢁ
ꢂꢀꢀꢀꢀ
The value for 13 (metronidazole) was found to be lower than that
in the literature.25 All these new data, together with data previous
published for the analogues of PA-824, are presented in Table 1.
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 413–418 | 417
©

View Article Online
Radiolysis of propan-2-ol
6 C. K. Stover, P. Warrener, D. R. VanDevanter, D. R. Sherman, T. M.
Arain, M. H. Langhorne, S. W. Anderson, J. A. Towell, Y. Yuan, D. N.
McMurray, B. N. Kreiswirth, C. E. Barry and W. R. Baker, Nature,
2000, 405, 962–966.
7 L. G. Wayne and H. A. Sramek, Antimicrob. Agents Chemother., 1994,
37, 2054–2058.
8 R. Singh, U. Manjunatha, H. I. M. Boshoff, Y. H. Ha, P.
Niyomrattanakit, R. Ledwidge, C. S. Dowd, I. Y. Lee, P. Kim, L. Zhang,
S. Kang, T. H. Keller, J. Jiricek and C. E. Barry 3rd, Science, 2008, 322,
1392–1395.
9 R. F. Anderson, S. S. Shinde, A. Maroz, M. Boyd, B. D. Palmer and
W. A. Denny, Org. Biomol. Chem., 2008, 6, 1973–1980.
10 U. H. Manjunatha, H. Boshoff, C. S. Dowd, L. Zhang, T. J. Albert,
J. E. Norton, L. Daniels, T. Dick, S. S. Pang and C. E. Barry, III, Proc.
Natl. Acad. Sci. U. S. A., 2006, 103, 431–436.
High energy electrons and g-rays ionize alcohols to form initially
an electron and radical cation pair. A small proportion of these
primary radicals escape recombination with the electron becoming
solvated, ca. 0.10
0.02 mmol J-1,26–28 and the radical cation
undergoing ion molecule reactions to form alkoxyl and alkyl
radicals.29 The alkoxyl radicals are oxidizing in nature and because
of the relatively lower bond strength of the a-C–H bond of propan-
2-ol, a proportion of the radical cation yield gives rise to the a-
hydroxyprop-2-yl radical giving an overall yield (G-value) of one-
electron reductants in irradiated propan-2-ol of ca. 0.2 mmol J-1.
11 R. Ballini and M. Petrini, Tetrahedron, 2004, 60, 1017–1047.
12 J. O. Lundberg, E. Weitzberg and M. T. Gladwin, Nat. Rev. Drug
Discovery, 2008, 7, 156–167.
13 V. Jagannadham and S. Steenken, J. Am. Chem. Soc., 1984, 106, 6542–
6551.
14 R. F. Anderson, S. S. Shinde, M. P. Hay, S. A. Gamage and W. A.
Denny, J. Am. Chem. Soc., 2003, 125, 748–756.
15 R. M. J. Ings, J. A. McFadzean and W. E. Ormerod, Biochem.
Pharmacol., 1974, 23, 1421–1429.
(CH3)2CHO∑ +(CH3)2CHOH→(CH3)2C∑OH + (CH3)2CHOH
One electron reduction of the nitroheterocyclic compounds
dissolved in propan-2-ol then occurs by (i) direct reaction with
the solvated electron and (ii) reaction with the a-hydroxyprop-2-yl
radical.
16 P. Kim, L. Zhang, U. H. Manjunatha, R. Singh, S. Patel, J. Jirieck,
T. H. Keller, H. I. Boshoff, C. E. Barry and C. S. Dowd, J. Med. Chem.,
2009, 52, 1317–1328.
17 US Pat, 5,668,127.
18 A. M. Thompson, A. Blaser, R. F. Anderson, S. S. Shinde, S. G.
Franzblau, Z. Ma, W. A. Denny and B. D. Palmer, J. Med. Chem.,
2009, 52, 637–645.
19 H. Sasaki, Y. Haraguchi, M. Itotani, H. Kuroda, H. Hashizume,
T. Tomishige, M. Kawasaki, M. Matsumoto, M. Komatsu and H.
Tsubouchi, J. Med. Chem., 2006, 49, 7854–7860.
20 P. Benjes and R. Grimmett, Heterocycles, 1994, 37, 735–738.
21 R. F. Anderson, W. A. Denny, W. Li, J. E. Packer, M. Tercel and W. R.
Wilson, J. Phys. Chem. A, 1997, 101, 9704–9709.
22 P. Griess, Ber. Bunsenges. Physik. Chem., 1879, 12, 426–428.
23 R. F. Anderson, Ber. Bunsenges. Physik. Chem., 1976, 80, 969–972.
24 E. Stekhan and T. Kuwana, Ber. Bunsenges. Physik. Chem., 1974, 78,
253–259.
Acknowledgements
This work was supported by Grant 07/243 from the Health
Research Council of New Zealand, and by the Global Alliance
for TB Drug Development, New York.
References
1 World health organization (WHO), Tuberculosis Fact Sheet No. 104,
Geneva, Switzerland, 2007.
2 H. I. Boshoff and C. E. Barry, 3rd, Nat. Rev. Microbiol., 2005, 3, 70–80.
3 J. Chan, K. Tanaka, D. Carroll, J. Flynn and B. R. Bloom, Infect.
Immun., 1995, 63, 736–740.
4 M. Laurenzi, A. Ginsberg and M. Spielgelman, Infect. Disord.: Drug
Targets, 2007, 7, 105–119.
5 M. Matsumoto, H. Hashizume, T. Tomishige, M. Kawasaki, H.
Tsubouchi, H. Sasaki, Y. Shimokawa and M. Komatsu, PLoS Med.,
2006, 3, e466.
25 P. Wardman, J. Phys. Chem. Ref. Data, 1989, 18, 1637–1755.
26 J. C. Russell and G. R. Freeman, J. Phys. Chem., 1968, 72, 808–815.
27 M. C. Sauer, S. Arai and L. M. Dorfman, J. Chem. Phys., 1965, 42,
708–712.
28 W. V. Sherman, J. Phys. Chem., 1966, 70, 667–672.
29 F. P. Sargent and E. M. Gardy, Can. J. Chem., 1974, 52, 3545–3650.
30 P. Kim, S. Kang, H. I. Boshoff, J. Jiricek, M. Collins, R. Singh, U. H.
Manjunatha, P. Niyomrattanakit, L. Zhang, M. Goodwin, T. Dick,
T. H. Keller, C. S. Dowd and C. E. Barry, III, J. Med. Chem., 2009, 52,
1329–1344.
418 | Org. Biomol. Chem., 2010, 8, 413–418
This journal is
The Royal Society of Chemistry 2010
©