The influence of the aziridinyl substituent of benzimidazoles and benzimidazolequinones on toxicity towards normal and Fanconi anaemia cells

Article abstract of DOI:10.1016/j.ejmech.2010.01.026

Aziridinyl substituted benzimidazolequinones are more toxic than methoxy analogues towards normal human fibroblast cells (GM00637). The aziridinyl substituent is required for hypersensitive killing of Fanconi anaemia (FA) cells (PD20i) deficient in FANCD2. Despite lacking quinone functionality, 4,7-dimethoxy-N-[(aziridin-2-yl)methyl]benzimidazole also induces hypersensitivity from FA cells, similar to their response towards mitomycin C. Expression of FANCD2 (in PD20:RV) corrects FA cell hypersensitivity supporting cellular response via the FANC pathway.

Full text of DOI:10.1016/j.ejmech.2010.01.026

European Journal of Medicinal Chemistry 45 (2010) 1873–1879
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
The influence of the aziridinyl substituent of benzimidazoles and
benzimidazolequinones on toxicity towards normal and Fanconi anaemia cells
,
Karen Faheya, Liz O’Donovan a, Miriam Carr a, Michael P. Carty b, Fawaz Aldabbagh a
*
a School of Chemistry, National University of Ireland, Galway, Ireland
b Biochemistry, School of Natural Sciences, National University of Ireland, Galway, Ireland
a r t i c l e i n f o
a b s t r a c t
Article history:
Aziridinyl substituted benzimidazolequinones are more toxic than methoxy analogues towards normal
human fibroblast cells (GM00637). The aziridinyl substituent is required for hypersensitive killing of
Fanconi anaemia (FA) cells (PD20i) deficient in FANCD2. Despite lacking quinone functionality, 4,7-dime-
thoxy-N-[(aziridin-2-yl)methyl]benzimidazole also induces hypersensitivity from FA cells, similar to their
response towards mitomycin C. Expression of FANCD2 (in PD20:RV) corrects FA cell hypersensitivity
supporting cellular response via the FANC pathway.
Received 10 October 2009
Received in revised form
9 January 2010
Accepted 12 January 2010
Available online 20 January 2010
Ó 2010 Elsevier Masson SAS. All rights reserved.
Keywords:
Aziridine
Bioreductive
Cytotoxicity
Mitomycin C
Quinones
1. Introduction
the nanomolar range (10ꢀ9 M) with greater potency under hypoxic
conditions (low pO2) [14].
Mitomycin C (MMC, Fig. 1) is the archtypical bioreductive drug,
incorporating both a quinone for reductive activation and fused
aziridine for subsequent alkylation of DNA [1]. MMC-induced cross-
links are then believed to be primarily responsible for cell death
[1–4]; initial monoalkylations at the C-1 position by DNA can be
followed by cross-link formation through the C-10 position upon loss
of the carbamate. In recent years, a significant number of alternatives
to MMC have been prepared and investigated for cytotoxicity and/or
anti-tumour activity [5–7]. These include the benzimidazolequi-
nones with [1,2-a] alicyclic ring fused [8–15] and non-fused struc-
tures (benzimidazole-4,7-diones) [16–21]. Examples include the
pyrrolo[1,2-a]benzimidazolequinones (e.g. PBI), which have been
comprehensively studied by Skibo and co-workers [8–11]. The PBIs
possess an aziridinyl substituent directly attached to the quinone
moiety and were found to show considerable in vitro cytotoxicity.
Aldabbagh and co-workers have reported other [1,2-a] alicyclic ring
fused benzimidazolequinones, including those with an additional
fused cyclopropane ring (e.g. 1) [12,13,15]. Benzimidazolequinones 1
and those without a cyclopropane ring (e.g. 2), were shown to
possess cytotoxicity towards normal human skin fibroblast cells in
Fanconi anaemia (FA) is a rare human genetic disease charac-
terised by an increased incidence of cancer in early adulthood
[22–25]. At a cellular level, it is diagnosed by hypersensitivity of FA
cells to crosslinking agents, in particular MMC. FA cells are mutant
in one of a number of genes encoding proteins in the FANC multi-
protein complex, which plays a key role in repairing MMC-induced
DNA damage. In a recent communication [21], we reported that FA
cells deficient in FANCD2 were hypersensitive to N-[(aziridin-2-
yl)methyl]benzimidazolequinone 9 (Fig. 2), comparable to their
response to MMC. The expression of FANCD2 decreased sensitivity
providing evidence for the induction of DNA damage, and a role for
the FANC pathway in partial correction of cell sensitivity.
We now clarify the requirement for the aziridinyl and quinone
functionalities in toxicity of agents towards normal cells, and
hypersensitive killing of FA cells. As part of this study we have
prepared methoxy and aziridinyl substituted five to seven-
membered [1,2-a] alicyclic ring fused benzimidazolequinones 3–8
(Fig. 1), and assessed the influence of the methoxy and aziridinyl
substituents on toxicity, as well as the effect of ring size. For
quinone produgs one- and two electron enzyme bioreductive
activation via the quinone functionality is an ideal prerequisite for
toxicity [1,5–7]. We now show that the quinone moiety is not
necessary to induce hypersensitivity from FA cells by reporting that
4,7-dimethoxy-N-[(1-tritylaziridin-2S-yl)methyl]-1H-
* Corresponding author. Tel.: þ353 91 493120; fax: þ353 91 495700
E-mail address: fawaz.aldabbagh@nuigalway.ie (F. Aldabbagh).
0223-5234/$ – see front matter Ó 2010 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2010.01.026

1874
K. Fahey et al. / European Journal of Medicinal Chemistry 45 (2010) 1873–1879
10
The synthesis of 4,7-dimethoxy-N-[(aziridin-2-yl)methyl]benzi-
O
O
O
O
OCONH2
midazole 10 was previously described, and represented the first time
that N-diazole nucleophilies had substituted at the methyl carbon of
2-methylaziridines [21]. The conversion of the 4,7-dimethoxy
substituents of 10 into the quinone 9 via the conventional method of
hydrobromic acid induced demethylation followed by room
temperature oxidation with ferric chloride of the in situ formed
hydroquinone resulted in opening of the aziridine, but this method
was successful in converting N-ethyl-4,7-dimethoxybenzimidazole
26 into quinone 27 (Scheme 3). A carefully selected acidic NBS
procedure proved successful in yielding quinone 9 [21,27].
H2N
Me
N
OMe
N
N
N
N
R
1
NH
N
Me
( )n
O
O
MMC
PBI
1
R = OCONH2, NH2,
n = 1, 2
NHCOCH3, NHCONH2
O
O
R
N
N
3. Results and discussion
N
N
( )n
O
O
3.1. Toxicity towards normal cells
2
n = 1, 3: R = OMe 4: R = NCH2CH2
n = 2, 5: R = OMe 6: R = NCH2CH2
n = 3, 7: R = OMe 8: R = NCH2CH2
There are no literature comparisons of the cytotoxicity of azir-
idinyl with methoxy substituted benzimidazolequinones, including
pyrrolo[1,2-a]benzimidazolequinones. We thus set about comparing
the toxicity of aziridinyl substituted [1,2-a] alicyclic ring fused ben-
zimidazolequinones 4, 6 and 8 with respective synthetic precursors
3, 5 and 7 towards the normal human skin fibroblast cell line
(GM00637) (Fig. 1 and Table 1). Increasing the size of the alicyclic
ring to six or seven membered was found to decrease potency, with
five-membered ring fused benzimidazolequinones 3 and 4 more
toxic than methoxy and aziridinyl substituted six or seven-
membered analogues. Moreover, aziridinyl substituted compounds
are more toxic towards this normal cell line than methoxy
substituted analogues with the same sized fused alicyclic ring.
Further, the aziridinyl and methoxy substituted benzimidazolequi-
nones 3–8 are less cytotoxic towards the same normal cell line than
[1,2-a] alicyclic ring fused benzimidazolequinones that lack substit-
uents on the quinone moiety (e.g. 1 and 2) [14]. Benzimidazolequi-
nones 3, 4 and 8 are more cytotoxic than MMC.
N-[(Aziridin-2-yl)methyl]benzimidazolequinone 9 is more than
eight times more cytotoxic towards normal cells than its synthetic
precursor 10 that lacks the quinone moiety. The enhancement of
toxicity towards normal cells by the quinone moiety is supported
by evaluating the toxicity of N-ethyl-4,7-dimethoxybenzimidazole
26 and its quinone adduct 27 towards the GM00637 cell line.
Compounds 26 and 27 lack the aziridine functionality, and benz-
imidazole 26, which is also lacking in the oxidizing quinone
functionality is non-toxic. However, N-ethyl-benzimidazolequi-
none 27 shows comparable toxicity to the aziridinyl containing
benzimidazolequinones.
Fig. 1. Mitomycin C and related benzimidazolequinones.
benzimidazole 10 (the synthetic precursor to 9, Fig. 2) is highly
selective towards FA cell killing at nanomolar (10ꢀ9 M)
concentrations.
2. Chemistry
The [1,2-a] alicyclic ring-fused benzimidazolequinones 3–8 were
prepared using a synthetic strategy previously described in the
literature for the pyrrolo [10] and azepino [26] benzimidazolequi-
nones. Nitration of methoxy substituted benzimidazoles 11–13 with
nitric acid gave almost equal amounts of two nitro isomers 14–19,
which were separable by column chromatography (Scheme 1). This
is due to the equal susceptibility of both adjacent positions to the
(directing) methoxy substituent to electrophilic nitration. The
nitration ceased if the temperature was lowered, with the addition of
concentrated sulfuric acid at ꢀ18 ꢁC giving both nitroisomers along
with significant quantities of the dinitrated compound. Moreover, if
the nitration time was increased or the traditional concentrated
nitric/sulfuric acid mixture was used at room temperature, the
benzimidazoles 20–22 were the sole reaction products in yields of
74–83%.
Catalytic hydrogenation of nitro compounds 14, 16 and 18 gave
amines 23–25 quantitatively, which were isolated and partially
characterized, but not purified (Scheme 2). Fremy’s oxidation gave
the benzimidazolequinones 3, 5 and 7 in about 80% yield, which
were treated with aziridine to displace the methoxy substituent to
give our targets 4, 6 and 8 in 60–88% yield. Benzimidazolequinones
3–8 were all found to be stable at room temperature and could be
stored at 0 ꢁC indefinitely without any decomposition, unlike the six-
membered analogue of PBI (Fig. 1., R]OAc or PBI-A), termed TPBI-A
[8]. The latter was reported to be unstable at room temperature,
which was given as a reason for its diminished cytotoxicity.
3.2. Toxicity towards Fanconi anaemia (FA) cells
We recently showed that nanomolar (10ꢀ9 M) concentrations of
benzimidazolequinone 9 are cytotoxic towards PD20i cells, a human
FA cell line lacking FANCD2 protein [21]. This is similar to the known
hypersensitivity of these cells to MMC [22–25]. Since compound 9
has activating quinone and reactive aziridinyl functionality, it was
interesting to determine the response of FA cells to benzimidazo-
lequinones having either the aziridinyl group attached directly at
the quinone moiety, as in 4, or replacing it with a methoxy group as
in 3. 6-Aziridinylbenzimidazolequinone 4 is found to be cytotoxic
towards PD20i cells in the 10ꢀ9 M range (Fig. 3), comparable to the
hypersensitivity of PD20i cells to MMC (Fig. 4) and N-[(aziridin-2-
yl)methyl]benzimidazolequinone 9 [21]. In contrast 6-methox-
ybenzimidazolequinone 3 is found to show negligible toxicity
towards PD20i cells at 10ꢀ9 M concentrations, implicating the
aziridinyl group in the hypersensitivity of FA cells lacking FANCD2 to
these compounds. PD20:RV is an isogenic FA cell line expressing
wild-type FANCD2 protein from an inserted transgene, which
O
O
OMe
N
N
N
N
N
N
Tr
Tr
OMe
9
10
Fig. 2. N-[(1-Tritylaziridin-2S-yl)methyl]-1H-benzimidazole(-4,7-dione).

K. Fahey et al. / European Journal of Medicinal Chemistry 45 (2010) 1873–1879
1875
NO2
MeO
MeO
MeO
O2N
N
N
N
N
N
N
HNO3
+
0 °C, 7 min
( )
( )
( )
n
n
n
14: n = 1, 43%
16: n = 2, 31%
18: n = 3, 37%
11: n = 1
12: n = 2
13: n = 3
15: n = 1, 41%
17: n = 2, 31%
19: n = 3, 35%
NO2
N
MeO
O2N
HNO3, H2SO4
rt, 24 h
N
( )
n
20: n = 1, 76%
21: n = 2, 83%
22: n = 3, 74%
Scheme 1. Nitration of alicyclic [1,2-a] ring-fused benzimidazoles.
partially corrects the sensitivity of these cells to MMC. We treated
PD20:RV cells with 3 and 4 (Fig. 3), and found that the toxicity of 4 is
significantly reduced, implicating the FANC complex in cellular
response to 4.
benzimidazolequinones. The quinone functionality increases
toxicity towards this normal cell line for N-(aziridin-2-yl)methyl
substituted compounds, with 10 significantly less toxic than 9.
The aziridinyl substituent, but not the quinone moiety was
found necessary to induce hypersensitive killing of the Fanconi
anaemia (FA) cell line (PD20i) deficient in FANCD2. 4,7-Dimethoxy-
N-[(aziridin-2-yl)methyl]benzimidazole 10 is found to be most
selective for killing PD20i at 10ꢀ9 M range. Aziridinyl substituted
compounds 4, 9 and 10 have only one position for DNA alkylation
(at the aziridine), and should not form cross-links in DNA, unlike
MMC which can form cross-linked adducts with DNA [1–4]. Thus it
can be inferred that cross-linking is not necessary for the hyper-
sensitive toxicity of these compounds towards FA cells. The
hypersensitivity of FA cells towards the aziridinyl containing
benzimidazoles and benzimidazolequinones and MMC is corrected
with expression of FANCD2 (in PD20:RV), supporting the conclu-
sion that the FANC pathway plays a key role in the cellular response
to these compounds.
Having confirmed the requirement for the aziridinyl group in
the hypersensitive response of FA cells, we now decided to inves-
tigate the role of the quinone, and thus bioreductive activation. FA
cells were thus treated with 4,7-dimethoxybenzimidazole 10
(Fig. 4). PD20i cells were found to be hypersensitive to 10, and the
response was similar to the effect of MMC on these cells, which was
found to be marginally greater than benzimidazolequinone 9 [21].
Furthermore, expression of FANCD2 in PD20:RV cells was found to
almost completely correct cellular sensitivity towards benzimid-
azole 10 resulting in selectivity towards the killing of FA cells
deficient in FANCD2.
4. Conclusion
Aziridinyl substituted [1,2-a] alicyclic ring fused benzimidazo-
lequinones are more cytotoxic than methoxy analogues towards
the normal human fibroblast cell line (GM00637). Five-membered
ring fused compounds are more toxic towards this normal cell line
than six and seven-membered ring fused analogues. N-Ethyl-4,7-
dimethoxybenzimidazole 26 is non-toxic, but the quinone
analogue 27 has comparable toxicity to the aziridinyl containing
5. Experimental
5.1. Materials
The synthesis of alicyclic ring-fused benzimidazoles 11–13 was
carried out starting from commercially available materials (Sigma–
NO2
NH2
Fremy's salt,
MeO
MeO
N
N
N
N
KH2PO4
H2, Pd-C, EtOH
40 psi, rt, 8 h
rt, 30 min
( )
( )
n
n
14, 16, 18
23-25 (100%)
O
O
H
N
MeO
N
N
N
N
O
N
O
( )
MeOH, rt, 2 h
( )
n
n
4: n = 1, 88%
6: n = 2, 77%
8: n = 3, 60%
3: n = 1, 77%
5: n = 2, 83%
7: n = 3, 78%
Scheme 2. Elaboration to give target alicyclic [1,2-a] ring-fused benzimidazolequinones.

1876
K. Fahey et al. / European Journal of Medicinal Chemistry 45 (2010) 1873–1879
100
OMe
OMe
O
N
N
N
(i)
(ii)
N
80
R
R
O
(i) 10: R = CHCH2NTr
(ii) 26: R = CH3
9: R = CHCH2NTr
27: R = CH3
60
Scheme 3. Synthesis of non-ring fused benzimidazolequinones (i) R]CHCH2NTr: NBS
(1.1 equiv), H2SO4 (trace), THF (aq), 0 ꢁC, 40 min, 56%; (ii) R]CH3: (a) HBr (aq), reflux,
3 h, (b) FeCl3 (aq), rt, 18 h, 53% (See Fig. 2 for stereochemistry).
40
20
0
Aldrich) using previously described procedures [10,26]. The
synthesis of compounds 3, 4, 11 and 14 was reported by Zhou and
Skibo, although 13C NMR spectra were not included [10]. We have
previously reported the synthesis and characterization of
compounds 8, 18 and 25, [26] and 9, and 10 [21]. Methods and
spectroscopic characterization of new compounds is given below.
0
2
4
6
Concentration (x 10-3 µ M)
8
10
Fig. 3. The effect of the aziridinyl substituent on Fanconi anaemia cells deficient in
FANCD2 (PD20i, closed symbols), and expressing FANCD2 protein (PD20:RV, open
symbols), as determined using the MTT assay following treatment with 6-methoxy-
2,3-dihydro-1H-pyrrolo[1,2-a]benzimidazole-5,8-dione 3 (C,B) and 6-(N-aziridinyl)-
2,3-dihydro-1H-pyrrolo[1,2-a]benzimidazole-5,8-dione 4 (:,
data point is the mean of at least three independent experiments.
5.2. Instrumentation
Instrument details and chromatography specifications are given
in the Supplementary Information of a recent communication [21].
J values are reported in Hz. NMR assignments were supported by
HETCOR 1H–13C NMR 2D spectra for compounds, and by DEPT for
compounds 14 and 20. Assignments were supported by NOE
difference spectra in which pre-saturation of the 9-H signal of 17
resulted in enhancement of the signal of 1-CH2.
EPSRC National Mass Spectrometry Service (University of Swan-
sea) carried out low-resolution CI on the Micromass Quattro II triple
quadrupole instrument and high-resolution mass spectrometry on
the Finnegan MAT 900 XLT by manual peak matching by using
Electrospray (ESI) for compounds 15 and 19. High resolution mass
spectra for all other compounds were carried out by University
College London, Mass Spectroscopy Facility using a Thermo Finnigan
MAT900xp operating in chemical ionization (CI, CIþ) mode with
methane as the carrier gas, and accurate masses were calculated
against reference ‘‘heptacosa’’. The precision of all accurate mass
measurements is better than 4 ppm.
6
) for 24 h at 37 ꢁC. Each
chromatography using silica gel as absorbent with ethyl acetate
as eluent to yield the title compound (53 mg, 43%) as a pale
yellow solid, Rf 0.40 (ethyl acetate); mp 188–190 ꢁC, (mp 182 ꢁC
[10]); (found: C, 56.7; H, 4.8; N, 18.3. C11H11N3O3 requires C,
56.6; H, 4.8; N, 18.0%); dC (100 MHz; CDCl3; Me4Si) 24.0 (3-CH2),
26.2 (2-CH2), 43.4 (1-CH2), 57.8 (Me), 107.8 (CH), 112.9 (CH),
128.4, 131.0, 142.3, 147.9 (all C) and 164.8 (Ar–3a-C).
5.3.2. 6-Methoxy-7-nitro-2,3-dihydro-1H-pyrrolo[1,2-a]
benzimidazole (15)
(50 mg, 41%) as a pale yellow solid, Rf 0.31 (ethyl acetate); mp
199–201 ꢁC. nmax (neat)/cmꢀ1 1637, 1583, 1518 (NO2), 1490, 1461,
1436, 1317 (NO2), 1290, 1243, 1228, 1199, 1160, 1118, 1057 and 1017;
5.3. General method for mono-nitration of benzimidazoles
100
80
60
40
20
0
5.3.1. 6-Methoxy-5-nitro-2,3-dihydro-1H-pyrrolo[1,2-
a]benzimidazole (14)
6-Methoxy-2,3-dihydro-1H-pyrrolo[1,2-a]benzimidazole 11
(0.100 g, 0.53 mmol) and fuming nitric acid (2 mL) were stirred
at 0 ꢁC for 7 min. The solution was neutralised with saturated
sodium bicarbonate and extracted with chloroform (2 ꢂ 15 mL).
The combined organic extracts were dried (Na2SO4) and evap-
orated to dryness. The brown residue was purified by column
Table 1
Effect of compounds on the growth of normal human fibroblasts.
Compound
IC50
(
m
mol/L)a
Compound
IC50 (m
mol/L)a
GM00637
GM00637
MMC
0.90 ꢃ 0.03
0.36 ꢃ 0.09
0.29 ꢃ 0.03
>1.5
3
4
5
6
7
8
9
10
26
27
0.56 ꢃ 0.06
0.15 ꢃ 0.03
1.26 ꢃ 0.26
>10
0
2
4
6
8
10
1.06 ꢃ 0.15
Fig. 4. The response of Fanconi anaemia cells deficient in FANCD2 (PD20i, closed
symbols), and expressing FANCD2 protein (PD20:RV, open symbols), as determined
using the MTT assay following treatment with 4,7-dimethoxy-N-[(1-tritylaziridin-
(2S)-yl)methyl]-1H-benzimidazole 10 (A,>) and MMC (-,,) for 24 h at 37 ꢁC. Each
data point is the mean of at least three independent experiments.
>1.5
0.58 ꢃ 0.18
a
IC50 represents the compound concentration required for the reduction of the
mean cell viability to 50% of the control value after incubation for 24 h at 37 ꢁC, as
determined using the MTT assay.

K. Fahey et al. / European Journal of Medicinal Chemistry 45 (2010) 1873–1879
1877
dH (400 MHz; CDCl3; Me4Si) 2.73–2.79 (2H, m, 2-CH2), 3.07 (2H, t, J
7.7, 3-CH2), 3.95 (3H, s, Me), 4.12 (2H, t, J 7.1, 1-CH2), 7.24 (1H, s, 5-H)
and 7.89 (1H, s, 8-H); dC (100 MHz; CDCl3; Me4Si) 24.1 (3-CH2), 26.0
(2-CH2), 43.5 (1-CH2), 57.1 (Me), 103.4 (5-CH), 107.9 (8-CH), 125.4,
135.5, 150.3, 153.12 (all C) and 166.57 (3a-C); m/z (ESI) 234.0874
(MHþ. C11H12N3O3 requires 234.0873), 234 (37%), 204 (100), 189
(11) and 174 (14).
(400 MHz; CDCl3; Me4Si) 2.79–2.86 (2H, m, 2-CH2), 3.15 (2H, t, J
7.6, 3-CH2), 3.99 (3H, s, Me), 4.24 (2H, t, J 7.3, 1-CH2) and 8.10
(1H, s, 8-H); dC (100 MHz; CDCl3; Me4Si) 24.3 (3-CH2), 26.2 (2-
CH2), 44.1 (1-CH2), 65.3 (Me), 110.1 (8-CH), 128.6, 136.0, 138.0,
142.7, 144.5 (all C) and 169.5 (3a-C).
5.4.2. 7-Methoxy-6,8-dinitro-6,8-1,2,3,4-tetrahydropyrido[1,2-
a]benzimidazole (21)
5.3.3. 7-Methoxy-6-nitro-1,2,3,4-tetrahydropyrido[1,2-
7-Methoxy-1,2,3,4-tetrahydropyrido[1,2-a]benzimidazole
12
a]benzimidazole (16)
(50 mg, 0.25 mmol) in 50:50 mixture of concentrated sulfuric acid
and fuming nitric acid (10 mL) at 20 ꢁC for 24 h gave after column
chromatography the title compound (60 mg, 83%) as a yellow solid, Rf
0.43 (chloroform–methanol 95:5); mp 154–156 ꢁC; (found: C, 49.7;
H, 4.4; N, 19.4. C12H12N4O5 requires C, 49.3; H, 4.1; N, 19.2%); nmax
(neat)/cmꢀ1 1632, 1590, 1523 (NO2), 1478, 1440, 1369, 1328 (NO2),
1243, 1213, 1160, 1099 and 1049; dH (400 MHz; CDCl3; Me4Si) 2.06–
2.10 (2H, m, CH2), 2.18–2.20 (2H, m, CH2), 3.17 (2H, t, J 6.4, 4-CH2),
4.05 (3H, s, CH3), 4.18 (2H, t, J 6.1, 1-CH2) and 8.15 (1H, s, 9-H);
dC(100 MHz; CDCl3; Me4Si) 19.9 (CH2), 22.1 (CH2), 25.8 (4-CH2), 43.4
(1-CH2), 65.3 (Me),109.4 (9-CH),131.1,135.8,138.0,139.2,143.3 (all C)
and 160.2 (4a-C).
7-Methoxy-1,2,3,4-tetrahydropyrido[1,2-a]benzimidazole
12
(0.100 g, 0.49 mmol) and fuming nitric acid (2 mL) at 0 ꢁC for 7 min
gave after column chromatography the title compound (38 mg, 31%),
as a pale yellow solid, Rf 0.47 (ethyl acetate-methanol 95:5); mp
176–178 ꢁC; (found: C, 58.0; H, 5.4; N, 16.9. C12H13N3O3 requires C,
58.3; H, 5.3; N, 17.0%); nmax (neat)/cmꢀ1 1588, 1514 (NO2), 1486,
1442, 1411, 1371, 1313 (NO2), 1269, 1219, 1181, 1162 and 1079; dH
(400 MHz; CDCl3; Me4Si) 1.97–2.03 (2H, m, CH2), 2.09–2.15 (2H, m,
CH2), 3.10 (2H, t, J 6.4, 4-CH2), 3.92 (3H, s, Me), 4.06 (2H, t, J 6.2,
1-CH2), 6.92 (1H, d, J 8.8, 8(9)-H) and 7.34 (1H, d, J 8.8, 9(8)-H);
dC(100 MHz; CDCl3; Me4Si) 20.4 (CH2), 22.4 (CH2), 25.7 (4-CH2),
42.9 (1-CH2), 57.8 (Me), 107.7 (CH), 112.1 (CH), 130.7, 130.8, 136.9,
148.4 (all C) and 155.5 (4a-C).
5.4.3. 3-Methoxy-2,4-dinitro-7,8,9,10-tetrahydro-6H-azepino[1,2-
a]benzimidazole (22)
5.3.4. 7-Methoxy-8-nitro-1,2,3,4-tetrahydropyrido[1,2-
a]benzimidazole (17)
3-Methoxy-7,8,9,10-tetrahydro-6H-azepino[1,2-a]benzimidazole
13 (0.200 g, 0.93 mmol) in 50:50 mixture of concentrated sulfuric
acid and fuming nitric acid (10 mL) at 20 ꢁC for 24 h gave after
column chromatography the title compound (0.210 g, 74%) as
(38 mg, 31%) as a pale yellow solid, Rf 0.31 (ethyl acetate-
methanol 95:5); mp 183.5–184.5 ꢁC; (found: C, 57.9; H, 5.4; N, 17.2.
C12H13N3O3 requires C, 58.3; H, 5.3; N, 17.0%); nmax (neat)/cmꢀ1
1634, 1593, 1512 (NO2), 1482, 1445, 1425, 1366, 1308 (NO2), 1271,
1237, 1199, 1160, 1147, 1074 and 975; dH (400 MHz; CDCl3; Me4Si)
1.96–2.04 (2H, m, CH2), 2.06–2.15 (2H, m, CH2), 3.08 (2H, t, J 6.4,
4-CH2), 3.95 (3H, s, Me), 4.06 (2H, t, J 6.2, 1-CH2), 7.23 (1H, s, 6-H)
and 7.88 (1H, s, 9-H); dC (100 MHz; CDCl3; Me4Si) 20.3 (CH2), 22.4
(CH2), 25.6 (4-CH2), 42.9 (1-CH2), 57.1 (Me), 102.5 (6-CH), 107.4
(9-CH), 127.7, 135.0, 147.2, 150.7 (all C) and 157.2 (4a-C).
a
yellow solid, Rf 0.43 (chloroform–methanol 95:5); mp
166–168 ꢁC; (found: C, 51.1; H, 4.6; N, 18.5. C13H14N4O5 requires C,
51.0; H, 4.6; N, 18.3%); nmax (neat)/cmꢀ1 1632, 1588, 1523 (NO2),
1474, 1443, 1380, 1326 (NO2), 1287, 1268, 1253, 1198, 1162, 1089 and
1044; dH (400 MHz; CDCl3; Me4Si) 1.76–1.83 (2H, m, CH2), 1.85–1.90
(2H, m, CH2), 1.95–2.00 (2H, m, CH2), 3.14 (2H, t, J 5.7, 6-CH2), 3.99
(3H, s, CH3), 4.22 (2H, t, J 5.0, 10-CH2), 8.10 (1H, s, 1-H); dC (100MHz;
CDCl3; Me4Si) 25.2 (CH2), 28.6 (CH2), 30.7 (6-CH2), 30.9 (CH2), 46.2
(10-CH2), 65.7 (Me), 109.9 (1-CH), 132.6, 136.4, 138.7, 139.0, 143.1
(all C) and 166.0 (5a-C).
5.3.5. 3-Methoxy-2-nitro-7,8,9,10-tetrahydro-6H-azepino[1,2-
a]benzimidazole (19)
yellow solid; Rf 0.44 (ethyl acetate-methanol 95:5); mp
163–165 ꢁC; nmax (neat)/cmꢀ1 1632, 1590, 1519 (NO2), 1448, 1317
(NO2), 1264, 1210, 1168, 1087, 1066 and 1008; dH (400 MHz; CDCl3;
Me4Si) 1.79–1.87 (4H, m, CH2), 1.94–1.96 (2H, m, CH2), 3.10 (2H, t, J
5.7, 6-CH2), 3.97 (3H, s, Me), 4.15 (2H, t, J 5.2, 10-CH2), 7.26 (1H, s,
4-H) and 7.90 (1H, s, 1-H); dC (100 MHz; CDCl3; Me4Si) 25.2 (CH2),
28.5 (CH2) 30.3 (6-CH2), 30.7 (CH2), 45.2 (10-CH2), 57.1 (Me), 102.9
(4-CH), 107.4 (1-CH), 128.9, 135.7, 146.6, 150.2 (all C), 162.7 (5a-C);
m/z (ESI) 262.1187 (MHþ. C13H16N3O3 requires 262.1186), 262 (21%),
232 (100), 217 (17) and 202 (16).
5.5. General method for formation of benzimidazolequinones
5.5.1. 6-Methoxy-2,3-dihydro-1H-pyrrolo[1,2-a]benzimidazole-5,8-
dione (3)
5-Nitrobenzimidazole 14 (70 mg, 0.30 mmol) and Pd–C
(10%, 10 mg) in ethanol (60 mL) was agitated under 40 psi hydrogen
at 20 ꢁC for 8 h. The catalyst was removed by filtration and the filtrate
evaporated to dryness to give 5-amino-6-methoxy-2,3-dihydro-1H-
pyrrolo[1,2-a]benzimidazole (23) (60 mg, 100%) as a yellow residue
(not purified further); dH (400 MHz; CDCl3; Me4Si) 2.65–2.73 (2H, m,
2-CH2), 3.02 (2H, t, J 7.3, 3-CH2), 3.87 (3H, s, CH3), 4.03 (2H, t, J 7.1,
1-CH2), 4.37 (2H, bs, NH2), 6.61 (1H, d, J 8.5, Ar-H) and 6.83 (1H, d, J
8.5, Ar-H); dC (100 MHz; CDCl3; Me4Si) 23.6 (3-CH2), 26.3 (2-CH2),
42.9 (1-CH2), 57.6 (Me), 97.4 (CH), 108.8 (CH), 127.9, 128.4, 137.9, 141.4
(all C) and 159.9 (3a-C).
5.4. General method for di-nitration of benzimidazoles
5.4.1. 6-Methoxy-5,7-dinitro-2,3-dihydro-1H-pyrrolo[1,2-
a]benzimidazole (20)
6-Methoxy-2,3-dihydro-1H-pyrrolo[1,2-a]benzimidazole 11
(0.370 g, 2.0 mmol) in 50:50 mixture of concentrated sulfuric
acid and fuming nitric acid (20 mL) was stirred at 0 ꢁC for 24 h.
The solution was basified with concentrated ammonium
hydroxide to pH 12 and extracted with chloroform (3 ꢂ 50 mL).
The combined organic extracts were dried (Na2SO4) and evap-
orated to dryness to yield the title compound (0.416 g, 76%) as
A mixture of amine 23 (60 mg, 0.30 mmol) and monobasic
potassium phosphate (0.112 g, 0.82 mmol) in water (14 mL) was
added to Fremy’s salt (0.210 g, 0.75 mmol) and monobasic potas-
sium phosphate (0.280 g, 2.1 mmol) in water (42 mL). The mixture
was stirred at 20 ꢁC for 30 min and extracted with chloroform
(2 ꢂ 15 mL). The combined organic extracts were dried (Na2SO4),
evaporated to dryness and purified by column chromatography
using silica gel as absorbent with chloroform as eluent. The yellow
quinone crystals isolated were recrystallised from chloroform/
hexane to yield the title compound (50 mg, 77%), as a bright yellow
solid, Rf 0.41 (chloroform–methanol 95:5); mp 211 ꢁC (decomp)
a
yellow solid, Rf 0.37 (chloroform–methanol 95:5); mp
160–162 ꢁC; (found: C, 47.6; H, 3.8; N, 20.5. C11H10N4O5 requires
C, 47.5; H, 3.6; N, 20.1%); nmax (neat)/cmꢀ1 1581, 1533 (NO2),
1499, 1441, 1372, 1330 (NO2), 1302, 1198, 1078 and 1050; dH

1878
K. Fahey et al. / European Journal of Medicinal Chemistry 45 (2010) 1873–1879
(mp 221 ꢁC [10]); (found: C, 60.6; H, 4.6; N, 12.4. C11H10N2O3
requires C, 60.6; H, 4.6; N, 12.8%); dC (100 MHz; CDCl3; Me4Si) 22.8
(3-CH2), 26.5 (2-CH2), 45.1 (1-CH2), 56.9 (Me), 105.2 (7-CH), 130.4,
144.1, 160.6, 160.8 (all C), 174.9 (C]O) and 178.0 (C]O).
5.6.2. 7-(N-Aziridinyl)-1,2,3,4-tetrahydropyrido[1,2-
a]benzimidazole-6,9-dione (6)
Benzimidazolequinone 5 (40 mg, 0.17 mmol) and ethyleneimine
(0.48 mL, 9.0 mmol) in methanol (6.4 mL) at room temperature for
2 h gave after column chromatography the title compound (32 mg,
77%) as a red solid, Rf 0.43 (chloroform–methanol 95:5); mp
153–155 ꢁC (decomp); (found: C, 64.0; H, 5.1; N, 17.6. C13H13N3O2
requires C, 64.2; H, 5.5; N,17.3%); nmax (neat)/cmꢀ11677 (C]O), 1634
(C]O), 1577, 1521, 1476, 1244, 1139 and 1082; dH (400 MHz; CDCl3;
Me4Si) 1.91–1.97 (2H, m, CH2), 1.99–2.04 (2H, m, CH2), 2.25 (4H, s,
aziridine-CH2), 2.99 (2H, t, J 6.2, 4-CH2), 4.29 (2H, t, J 5.7, 1-CH2) and
5.79 (1H, s, 8-H); dC (100 MHz; CDCl3; Me4Si) 19.7 (CH2), 22.2 (CH2),
24.9 (4-CH2), 27.8 (aziridine-CH2), 45.4 (1-CH2), 115.7 (8-CH), 130.6,
139.9, 151.3, 157.0 (all C), 176.7 (C]O) and 178.6 (C]O); m/z (CI)
244.1084 (MHþ. C13H14N3O2 requires 244.1086), 244 (100%).
5.5.2. 7-Methoxy-1,2,3,4-tetrahydropyrido[1,2-a]benzimidazole-
6,9-dione (5)
6-Nitrobenzimidazole 16 (50 mg, 0.20 mmol) upon hydrogena-
tion gave 6-amino-7-methoxy-1,2,3,4-tetrahydropyrido[1,2-a]benz-
imidazole (24) (44 mg, 100%) as a yellow residue (not purified
further); dH (400 MHz; CDCl3; Me4Si) 1.91–1.97 (2H, m, CH2), 2.02–
2.07 (2H, m, CH2), 3.04 (2H, t, J 6.2, 4-CH2), 3.87 (3H, s, Me), 3.97 (2H, t,
J 6.1, 1-CH2), 4.43 (2H, bs, NH2), 6.58 (1H, d, J 8.5, Ar–H) and 6.85
(1H, d, J 8.5, Ar–H); dC (100 MHz; CDCl3; Me4Si) 20.5 (CH2), 22.5 (CH2),
25.2 (4-CH2), 42.3 (1-CH2), 57.5 (Me), 96.4 (CH), 108.8 (CH), 127.2,
130.4, 131.4, 141.5 (all C) and 150.2 (4a-C).
Amine 24 (44 mg, 0.20 mmol), monobasic potassium phosphate
(0.326 g, 2.40 mmol) and Fremy’s salt (0.175 g, 0.65 mmol) in water
(46 mL) at 20 ꢁC for 30 min gave after column chromatography and
recrystallization the title compound (39 mg, 83%) as orange crystals,
Rf 0.53 (chloroform–methanol 95:5); mp 208–210 ꢁC (decomp);
(found: C, 62.1; H, 4.7; N,11.9. C12H12N2O3 requires C, 62.1; H, 5.2; N,
12.1%); nmax (neat)/cmꢀ11686 (C]O),1639 (C]O),1590,1530,1483,
1432, 1323, 1239, 1174, 1101 and 1036; dH (400 MHz; CDCl3; Me4Si)
1.91–1.99 (2H, m, CH2), 1.99–2.02 (2H, m, CH2), 2.94 (2H, t, J 6.6,
4-CH2), 3.80 (3H, s, Me), 4.26 (2H, t, J 6.0, 1-CH2) and 5.61 (1H, s,
8-H); dC (100 MHz; CDCl3; Me4Si) 19.6 (CH2), 22.1 (CH2), 24.8
(4-CH2), 45.4 (1-CH2), 56.7 (Me), 105.8 (8-CH), 130.6, 139.4, 151.7
(all C), 159.8 (Ar-4a-C), 175.1 (C]O) and 178.4 (C]O); m/z (CI)
233.0927 (MHþ. C12H13N2O3 requires 233.0926), 233 (100%).
5.7. Synthesis of 1-ethyl-1H-benzimidazole-4,7-dione (27)
1-Ethyl-4,7-dimethoxy-1H-benzimidazole 26 [28] (0.200 g,
0.97 mmol) in 48% hydrobromic acid (7 mL) was heated at reflux for
3 h. The reaction was cooled and evaporated to dryness. An aqueous
solutionofferricchloride (0.7M, 7 mL)wasaddedandstirredatroom
temperature for 18 h. Saturated sodium acetate solution (10 mL) was
added, extracted with chloroform, and dried (MgSO4). The organic
extracts were evaporated to dryness to give a brown residue, which
was purified by column chromatography using silica gel as absorbent
with ethyl acetate as eluent to give the title compound (90 mg, 53%) as
a yellow solid; Rf 0.41 (ethyl acetate); mp 141–142 ꢁC; nmax (neat)/
cmꢀ11663 (C]O), 1587, 1508, 1337, 1235, 1116, 1045; dH (400 MHz;
CDCl3; Me4Si) 1.49 (3H, t, J 7.3, CH3), 4.38 (2H, q, J 7.3, CH2), 6.60 (1H,
d(ABq), J 10.5, 5(6)-H), 6.67 (1H, d(ABq), J 10.5, 6(5)-H), 7.77 (1H, s, 2-
H); dC (100 MHz; CDCl3; Me4Si) 16.1 (CH3), 42.3 (CH2), 130.1 (C),
136.23 (CH),136.74 (CH),142.60 (2-CH),178.41 (C]O),180.96 (C]O);
m/z (CI) 177.0656 (MHþ. C9H9N2O2 requires 177.0664).
5.5.3. 3-Methoxy-7,8,9,10-tetrahydro-6H-azepino[1,2-
a]benzimidazole-1,4-dione (7)
Amine 25 [26] (40 mg, 0.17 mmol), monobasic potassium phos-
phate (0.294 g, 2.16 mmol) and Fremy’s salt (0.158 g, 0.59 mmol) in
water (43 mL) at 20 ꢁC for 30 min gave after column chromatography
and recrystallization the title compound (33 mg, 78%) as a yellow
solid, Rf 0.56 (chloroform–methanol 95:5); mp 179–181 ꢁC; (found:
C, 63.8; H, 5.3; N, 11.2. C13H14N2O3 requires C, 63.4; H, 5.7; N, 11.4%);
nmax (neat)/cmꢀ1 1690 (C]O), 1648 (C]O), 1632, 1589, 1524, 1471,
1437, 1322, 1297, 1238, 1134, 1094 and 1029; dH (400 MHz; CDCl3;
Me4Si) 1.67–1.73 (2H, m, CH2), 1.76–1.82 (2H, m, CH2), 1.86–1.91 (2H,
m, CH2), 2.97 (2H, t, J 5.5, 6-CH2), 3.80 (3H, s, Me), 4.54–4.55 (2H, m,
10-CH2) and 5.62 (1H, s, 2-H); dC (100 MHz; CDCl3; Me4Si) 25.0 (CH2),
28.2 (CH2), 29.3 (6-CH2), 30.8 (CH2), 45.9 (10-CH2), 56.8 (Me), 106.4
(2-CH),131.1,139.0,158.3,159.7 (all C),175.3 (C]O) and 179.1 (C]O).
5.8. Cell culture and cytotoxicity measurements
5.8.1. Cell culture
An SV40-transformed normal human skin fibroblast cell line
(repository number GM00637) was obtained from the National
Institute for General Medical Sciences (NIGMS) Human Genetic Cell
Repository (Coriell Institute for Medical Research, New Jersey, USA).
The PD20i and PD20:RV human fibroblast cell lines were obtained
from the Oregon Health and Science University (OHSU) Fanconi
Anaemia Cell Repository (Portland, OR, USA). The SV40-transformed
normal human skin fibroblast cell line (GM00637) was grown in
minimum essential media (MEM) Eagle-Earle BSS supplemented
with 15% uninactivated fetal bovine serum, (FBS), penicillin–strep-
5.6. General method for formation of aziridinyl substituted
benzimidazolequinones
tomycin, 2 mM
L
-glutamine, 2 ꢂ essential and non-essential amino
acids and vitamins. PD20i and PD20:RV human fibroblast cell lines
were grown in minimal essential media (alpha modification) sup-
plemented with 15% uninactivated fetal bovine serum, penicillin–
5.6.1. 6-(N-Aziridinyl)-2,3-dihydro-1H-pyrrolo[1,2-
a]benzimidazole-5,8-dione (4)
A solution of benzimidazolequinone 3 (65 mg, 0.30 mmol) and
ethyleneimine (0.83 mL, 15.5 mmol) in methanol (10.4 mL) was
stirred at room temperature for 2 h. The mixture was evaporated to
dryness and the residue purified by column chromatography using
silica gel as absorbent with chloroform as eluent. The quinone was
recrystallized from chloroform/hexane to yield the title compound
(60 mg, 88%) as a red solid, Rf 0.34 (chloroform–methanol 95:5);
mp 195–196 ꢁC (decomp), (mp 206 ꢁC [10]); (found: C, 62.9; H, 4.7;
N, 18.7). C12H11N3O2 requires C, 62.9; H, 4.8; N, 18.3%; dC(100 MHz;
CDCl3; Me4Si) 22.7 (3-CH2), 26.5 (2-CH2), 27.9 (aziridine-CH2), 45.1
(1-CH2), 114.8 (7-CH), 130.4, 144.7, 157.7, 160.4 (all C), 176.6 (C]O)
and 178.1 (C]O); m/z (CI) 230.0920 (MHþ. C12H12N3O2 requires
230.0930), 230 (100%).
streptomycin and 2 mM L-glutamine. All cells were incubated at
37 ꢁC under a humidified atmosphere containing 5% CO2.
5.8.2. Cytotoxicity measurement
Growth inhibition (cell viability) was determined using the MTT
colorimetric assay [29]. Cells were plated into 96-well plates at
a density of 10,000 cells per well (PD20i and PD20:RV cells were
plated in parallel in the same plate) and allowed to adhere over
a period of 48 h. Benzimidazolequinone and MMC (Sigma) solu-
tions (5
mL) were applied in DMSO (2.5% final concentration in
well), and the plates were incubated at 37 ꢁC under a humidified
atmosphere containing 5% CO2 for 24 h. Control cells were exposed
to an equivalent concentration of DMSO alone. MTT (100 mg) was

K. Fahey et al. / European Journal of Medicinal Chemistry 45 (2010) 1873–1879
1879
[4] M. Tomasz, Y. Palom, Pharmacol. Ther. 76 (1997) 73–87.
[5] L. Garuti, M. Roberti, D. Pizzirani, Mini-Rev. Med. Chem. 7 (2007) 481–489.
added and the cells were incubated for another 3 h. The superna-
tant was then removed carefully by pipetting. The resultant MTT
formazan crystals were dissolved in 100 mL of DMSO and absor-
[6] M.A. Colucci, C.J. Moody, G.D. Couch, Org. Biomol. Chem.
637–656.
6 (2008)
[7] Y. Chen, L. Hu, Med. Res. Rev. 29 (2009) 29–64.
[8] E.B. Skibo, I. Islam, M.J. Heileman, W.G. Schulz, J. Med. Chem. 37 (1994)
78–92.
[9] E.B. Skibo, I. Islam, W.G. Schulz, R. Zhou, L. Bess, R. Boruah, Synlett (1996)
297–309.
[10] R. Zhou, E.B. Skibo, J. Med. Chem. 39 (1996) 4321–4331.
[11] A. Suleman, E.B. Skibo, J. Med. Chem. 45 (2002) 1211–1220.
[12] J. O’Shaughnessy, D. Cunningham, P. Kavanagh, D. Leech, P. McArdle,
F. Aldabbagh, Synlett (2004) 2382–2384.
bance was determined on a plate reader at 550 nm with a reference
at 690 nm. Cell viability is expressed as a percentage of the DMSO-
only treated control value. Dose-response curves were analyzed by
nonlinear regression analysis and IC50 values were estimated by
using GraphPad Prism software, v. 5.02 (GraphPad Inc., San Diego,
CA, USA). The in vitro activity of the drugs towards GM00637 is
expressed as IC50 (i.e. concentration required for the reduction of
the mean cell viability to 50%).
[13] J. O’Shaughnessy, F. Aldabbagh, Synthesis (2005) 1069–1076.
[14] M. Lynch, S. Hehir, P. Kavanagh, D. Leech, J. O’Shaughnessy, M.P. Carty,
F. Aldabbagh, Chem. Eur. J. 13 (2007) 3218–3226.
[15] S. Hehir, L. O’Donovan, M.P. Carty, F. Aldabbagh, Tetrahedron 64 (2008)
4196–4203.
Acknowledgements
[16] C. Flader, J. Liu, R.F. Borch, J. Med. Chem. 43 (2000) 3157–3167.
[17] L. Garuti, M. Roberti, M. Malagoli, T. Rossi, M. Castelli, Bioorg. Med. Chem. Lett.
10 (2000) 2193–2195.
[18] L. Garuti, M. Roberti, D. Pizzirani, A. Pession, E. Leoncini, V. Cenci, S. Hrelia, IL
Farmaco 59 (2004) 663–668.
[19] K.-H. Chung, S.-Y. Hong, H.-J. You, R.-E. Park, C.-K. Ryu, Bioorg. Med. Chem. 14
(2006) 5795–5801.
[20] A. Gellis, H. Kovacic, N. Boufatah, P. Vanelle, Eur. J. Med. Chem. 43 (2008)
1858–1864.
This publication has emanated from research conducted with
financial support from Science Foundation Ireland (07/RFP/
CHEF227). The authors acknowledge the receipt of embark initiative
postgraduate scholarships for KF and LOD from the Irish Research
Council for Science, Engineering and Technology funded by the
National Development Plan. This work was also supported in part by
Health Research Board funding to MPC.
[21] L. O’Donovan, M.P. Carty, F. Aldabbagh, Chem. Commun. (2008)
5592–5594.
[22] A.D. Auerbach, S.R. Wolman, Nature 261 (1976) 494–496.
[23] M.D. Tischkowitz, S.V. Hodgson, J. Med. Genet. 40 (2003) 1–10.
[24] L.J. Niedernhofer, A.S. Lalai, J.H.J. Hoeijmakers, Cell (Cambridge, M.A. USA) 123
(2005) 1191–1198.
[25] T. Taniguchi, A.D. D’Andrea, Blood 107 (2006) 4223–4233.
[26] K. Fahey, F. Aldabbagh, Tetrahedron Lett. 49 (2008) 5235–5237.
[27] D.W. Kim, H.Y. Choi, K.-J. Lee, D.Y. Chi, Org. Lett. 3 (2001) 445–447.
[28] R.W. Middleton, H. Monney, J. Parrick, Synthesis (1984) 740–743.
[29] T. Mosmann, J. Immunol. Methods 65 (1983) 55–63.
References
[1] H.A. Seow, P.G. Penketh, R.P. Baumann, A.C. Sartorelli, in: H. Sies, L. Packer
(Eds.), Methods in Enzymology 382 (Quinones and Quinone Enzymes, Part B),
Elsevier, 2004, pp. 221–233.
[2] V.N. Iyer, W. Szybalski, Science 145 (1964) 55–58.
[3] M. Tomasz, R. Lipman, D. Chowdary, J. Pawlak, G.L. Verdine, K. Nakanishi,
Science 235 (1987) 1204–1208.