The synthesis of 2-nitroaryl-1,2,3,4-tetrahydroisoquinolines, nitro-substituted 5,6-dihydrobenzimidazo[2,1-a]isoquinoline N-oxides and related heterocycles as potential bioreducible substrates for the enzymes NAD(P)H: Quinone oxidoreductase 1 and E. coli

Article abstract of DOI:10.1016/j.bmcl.2011.10.044

A series of 2-nitroaryl-1,2,3,4-tetrahydroisoquinolines 10 and nitro-substituted 5,6-dihydrobenzimidazo[2,1-a]isoquinoline N-oxides 11 have been synthesised and evaluated as potential bioreducible substrates for the enzymes NAD(P)H: quinone oxidoreductase

Full text of DOI:10.1016/j.bmcl.2011.10.044

Bioorganic & Medicinal Chemistry Letters 21 (2011) 7447–7450
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
The synthesis of 2-nitroaryl-1,2,3,4-tetrahydroisoquinolines, nitro-substituted
5,6-dihydrobenzimidazo[2,1-a]isoquinoline N-oxides and related heterocycles
as potential bioreducible substrates for the enzymes NAD(P)H: quinone
oxidoreductase 1 and E. coli nitroreductase
Philip J. Burke a, Lai Chun Wong b, Terence C. Jenkins a, Richard J. Knox a, Stephen P. Stanforth b,
⇑
a
ˆ
Morvus Technology Ltd, Ty Myddfai, Llanarthne, Carmarthen SA32 8HZ, UK
b School of Life Sciences, Northumbria University, Newcastle Upon Tyne NE1 8ST, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of 2-nitroaryl-1,2,3,4-tetrahydroisoquinolines 10 and nitro-substituted 5,6-dihydrobenzimi-
dazo[2,1-a]isoquinoline N-oxides 11 have been synthesised and evaluated as potential bioreducible sub-
strates for the enzymes NAD(P)H: quinone oxidoreductase 1 (NQO1) and Escherichia coli nitroreductase
(NR). Also prepared and evaluated were 2-(3,5-dinitropyridin-2-yl)-1,2,3,4-tetrahydroisoquinoline 12
and 5,6-dihydro-10-nitropyrido[300,200:40,50]imidazo[20,10-a]isoquinoline 12-oxide 13. Both compounds
10b and 13 were reduced faster by human NQO1 than by CB-1954 [5-(aziridin-1-yl)-2,4-
dinitrobenzamide].
Received 24 June 2011
Revised 30 September 2011
Accepted 3 October 2011
Available online 18 October 2011
Keywords:
NAD(P)H: quinone oxidoreductase 1 (NQO1)
Escherichia coli nitroreductase
CB-1954
Ó 2011 Elsevier Ltd. All rights reserved.
1,2,3,4-Tetrahydroisoquinolines
Benzimidazo[2,1-a]isoquinoline N-oxides
Many nitroaromatic pro-drugs1 have been assessed as potential
substrates for the enzyme NAD(P)H: quinone oxidoreductase 1
(NQO1), also known as DT-diaphorase, because elevated levels of
NQO1 are often associated with tumour tissues.2 Escherichia coli
nitroreductase (NR) has also generated interest as an activator of
nitroaromatic pro-drugs through use in either anti-body directed
enzyme pro-drug therapy (ADEPT) or virus/gene-directed enzyme
pro-drug therapy (VDEPT/GDEPT).3 An important landmark in the
development of bioreducible nitroaromatic pro-drugs was the
observation that CB-1954 (Fig. 1) exhibited cytotoxicity against
rat Walker 256 carcinomas.4 Unfortunately, human cell lines,
including those expressing high levels of NQO1, were unrespon-
sive. However, CB-1954 was reduced more efficiently by E. coli
NR and this property has stimulated interest in using ADEPT or
VDEPT/GDEPT as potential methods for activating CB-1954 in
tumours.
NO2
O
_
9
6
4
1
5
10
N
6
7
3
2
8
7
11
4
NH2
N
X
N
1
8
X
O2N
5
N
+
9
N
2
N
O2N
+
Me
O
_
10
3
11
4
Cryptolepine 1
2
X = CH
X = CH
CB-1954
5 X = N
3 X = N
Figure 1. CB-1954, Cryptolepine and potential anti-cancer targets.
⇑
Corresponding author.
E-mail address: steven.stanforth@northumbria.ac.uk (S.P. Stanforth).
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.10.044

7448
P. J. Burke et al. / Bioorg. Med. Chem. Lett. 21 (2011) 7447–7450
Table 1
The alkaloid, Cryptolepine 1 (Fig. 1), has been shown to possess
a range of biological activities including anti-cancer activity.5
A
Synthesis of compounds 10 and 11
series of nitro-analogues of Cryptolepine 1 have recently been pre-
pared as potential bioreducible anti-cancer agents and these com-
pounds have been shown to be human NQO1 substrates and also
have displayed cytotoxic activity.6 However, the mechanism of ac-
tion of these Cryptolepine nitro-derivatives is as yet unclear be-
cause it has been demonstrated that these compounds are
localised within the cell cytoplasm and are unable to penetrate into
the nucleus.
In view of our previous work describing the synthesis of 5,6-
dihydrobenzimidazo[2,1-a]isoquinoline N-oxides 2 and the 5,6-
dihydropyrido[20,30:5,4]imidazo[2,1-a]isoquinoline 12-oxide 3,7 as
well as our work on other bioreducible nitroaromatics,8 we envis-
aged that nitro-containing derivatives of these compounds might
also have the potential to act as bioreducible anti-cancer agents
(Fig. 1). These heterocycles could be prepared by cyclodehydration
of the corresponding 2-nitroaryl-1,2,3,4-tetrahydroisoquinoline
derivatives 4 and 5, respectively.9 Thus, derivatives of heterocycles
4 and 5 might provide useful scaffolds for the future development
of bioreducible pro-drugs. A series of nitro-derivatives of heterocy-
cles 2–5 have therefore been prepared and evaluated as substrates
for human NQO1 and E. coli NR in order to assess their bioreduc-
ibility profiles.
The reaction of 1,2,3,4-tetrahydroisoquinoline 6, 5-nitro-1,2,3,4-
tetrahydroisoquinoline710 or7-nitro-1,2,3,4-tetrahydroisoquinoline
811 with the appropriate dinitrochloro- or dinitrofluoro-aromatic
compounds 9 under basic conditions afforded the corresponding
2-dinitroaryl-1,2,3,4-tetrahydroisoquinoline derivatives 10a–10o
(14–95%)(Scheme1, Table1).Arangeofcompounds10wereselected
for conversion into their corresponding N-oxide derivatives 11 (25–
99%) by heating in propionic acid at reflux.7,9 Heterocycles 12 (97%
yield) and 13 (quantitative yield) were prepared similarly. Included
in the potential bioreducible substrates were some trifluoromethy-
lated derivatives because of the general interest of this
functionality in medicinal chemistry.12 Experimental details are
given in the Supplementary data.
The compounds prepared in Table 1 were evaluated as human
NQO1 and E. coli NR substrates and both their specific activities
(i.e. rate of reduction of the substrate by the appropriate enzyme)
and their relative activities compared to CB-1954 are presented
in Table 2. The compounds’ specific activities were determined
by monitoring their rate of disappearance as a function of time
R1
R1
R5
R5
R4
R3
R4
R3
N
N
R2
R2
N
+
O2N
O
_
10
11
10
10
11
Yield
(%)
R1
R2
R3
R4
R5
Methoda Yield
(%)
a
b
c
d
e
f
g
h
i
j
k
l
m
n
o
H
H
H
H
H
H
H
NO2
NO2
NO2
NO2
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
NO2
NO2
NO2
NO2
NO2
CO2H
CONH2
NO2
NO2
NO2
NO2
H
H
H
H
H
CF3
CO2H
A
A
B
A
A
B
C
A
A
D
A
A
A
D
A
95
60
70
75
68
75
27
90
49
60
75
45
14
51
68
99
93
b
CONH2
H
89
CONH2
74
b
H
H
H
H
NO2
NO2
H
CF3
CO2H
H
b
82
61
b
H
CONH2
71
NO2 NO2
NO2 NO2
NO2 NO2
NO2 NO2
H
H
H
H
52
b
CF3
CO2H
H
b
c
CONH2
a
Method A: EtOH/H2O, NaHCO3, reflux; Method B: (i) DMSO, K2CO3, ca 60 °C, (ii)
concd HCl; Method C: DMSO, K2CO3, ca 60 °C; Method D: (i) EtOH/H2O, NaHCO3,
reflux, (ii) concd HCl.
b
Not prepared.
c
Compound 11o could not be characterised satisfactorily due to its insolubility.
by HPLC from which their initial rate of reaction was derived.
In some cases overlapping substrate and product peaks were ob-
served by HPLC, and in these cases, an estimation of the specific
activity was obtained by measuring the decrease in NADH
concentration with the assumption that 2 equiv of NADH are re-
quired for reduction of the nitro-group to the corresponding
hydroxylamine group. Thus, the rate of decrease of NADH
R1
R1
R1
R5
R5
R5
X
R4
R3
R4
R3
(i)
(ii)
R4
R3
N
N
R2
R2
NH
R2
N
+
NO2
O2N
O
_
9
X = Cl, F
11
10
6
7
R
R
1 = R2 = H
1 = NO2, R2 = H
8 R1 = H, R2 = NO2
Cl
N
(iii)
(ii)
N
N
N
N
NH
N
+
NO2
NO2
O2N
NO2
NO2
O
_
6
12
13
Scheme 1. Synthesis of nitro-substituted 5,6-dihydrobenzimidazole N-oxides and related derivatives. Reagents and conditions: (i) see Table 1; (ii) propionic acid, reflux; (iii)
EtOH/H2O, NaHCO3, 50–60 °C.

P. J. Burke et al. / Bioorg. Med. Chem. Lett. 21 (2011) 7447–7450
7449
Table 2
Activity data of potential human NQO1 and E. coli NR substrates as determined by HPLC
Compound
Human NQO1
mol/min/mg)
E. coli NR
mol/min/mg)
Specific activitya
(l
Relative to CB-1954
Specific activitya
(l
Relative to CB-1954
CB-1954
10a
10b
10c
10d
10e
10f
10g
10h
10ib
10j
10k
10lc
10md
10n
10o
11a
11be
11d
11e
11h
11i
6.2 ꢀ 10ꢁ3
<0.001
0.110
0.011
0.053
0.042
<0.001
<0.001
<0.001
—
(0.038)
0.0014
—
0.076
(0.020)
<0.001
0.016
—
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
0.013
0.170
1.0
1.86
0.44
41.49
4.22
(2.82)
4.40
0.02
<0.001
0.44
—
(9.85)
1.11
—
2.39
(14.8)
1.67
0.50
1.0
0.2
22.3
2.3
(1.5)
2.4
0.01
<0.001
0.2
—
(5.3)
0.6
—
1.3
(7.4)
0.9
0.3
<0.16
17.7
1.8
8.6
6.8
<0.16
<0.16
<0.16
—
(6.1)
0.23
—
12.3
(3.2)
<0.16
2.6
—
—
—
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
2.1
(1.62)
<0.001
(1.51)
<0.001
<0.001
(1.3)
1.49
(0.9)
<0.001
(0.8)
<0.001
<0.001
(0.6)
0.8
11k
11l
12
13
27.4
54.0
29.0
a
Numbers in parenthesis determined by monitoring the decrease in NADH concentration. Other values determined by monitoring the decrease in substrate concentration.
Specific activities determined at 37 °C.
b
Compound was insoluble in 10 mM phosphate buffer, pH 7 + 10% DMSO.
The compound precipitated out of solution in buffer with 10% DMSO + 500
c
l
M NADH.
d
Compound shows poor solubility. Enzyme work carried out using 50
compound and no effect of the DMSO on the enzyme activity.
l
M compound in the buffer with 10% DMSO. Rates assume a starting concentration of 50 lM
e
Multiple reduction products observed.
concentration has been divided by 2 in order to produce an
estimation of specific activity and these values are given in
parenthesis in Table 2.
In conclusion, the trifluoromethylated derivative 10b has been
established to be a significantly better substrate for human
NQO1 and E. coli NR than CB-1954. Some other N-nitroaryl-
1,2,3,4-tetrahydroisoquinoline derivatives were also better human
NQO1 and E. coli NR substrates than CB-1954 but only marginally
so. Most of the N-oxide derivatives evaluated showed little, if
any, activity towards these two enzymes with the exception of
heterocycle 13 which was reduced over twenty times faster than
CB-1954. These studies suggest that the synthesis of other trifluo-
romethylated 1,2,3,4-tetrahydroisoquinoline derivatives would be
useful in order to obtain a more detailed structure–activity profile.
Within the series of compounds 10, the trifluoromethylated
derivatives 10b and 10m exhibited human NQO1 activities that
were higher than CB-1954. The activity of the other trifluorome-
thylated substrate, compound 10i, could not be assessed because
of the relative insolubility of this compound. Compound 10b also
exhibited the best E. coli NR activity of this series of compounds
whereas compound 10m was only comparable in activity to CB-
1954. The amide isomers, compounds 10d and 10e which bear
the closest structural resemblance to CB-1954, were better sub-
strates for human NQO1 by a factor of 8.6 and 6.8, respectively,
than CB-1954 but were only slightly better substrates for E. coli
NR. The two isomeric carboxylic acid derivatives, compounds 10j
and 10n, in which one nitro-group was present in the 1,2,3,4-tetra-
hydroisoquinoline moiety of the molecule, both showed higher
activities towards human NQO1 and E. coli NR than CB-1954 but
the rate of disappearance of NADH was used to estimate the activ-
ities for these two substrates. Compounds 10f and 10g, both bear-
ing a 2,6-dinitro substitution pattern in the N-aryl ring were
essentially devoid of activity in both the human NQO1 and E. coli
NR assays.
Acknowledgements
We thank Morvus Technology Limited and the University of
Northumbria for a studentship (to L.C.W). We also thank the EPSRC
National Mass Spectrometry Centre, Swansea, for high-resolution
mass spectra.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2011.10.044.
The relative activities of the nitro-containing 5,6-dihydrobenz-
imidazo[2,1-a]isoquinoline N-oxide derivatives 11 compared to
CB-1954 were disappointing. Only compound 11a showed human
NQO1 activity greater than CB-1954. All the other derivatives
exhibited little, if any, activity with human NQO1 and E. coli NR.
However, this is in marked contrast to the pyridine analogue, com-
pound 13, which displayed significantly better relative activities
towards both human NQO1 and E. coli NR than CB-1954 (27.4
and 29.0, respectively).
References and notes
1. Chen, Y.; Hu, L. Med. Res. Rev. 2009, 29, 29. and references cited therein.
2. (a) Cresteil, T.; Jaiswal, A. K. Biochem. Pharmocol. 1991, 42, 1021; (b) Riley, R. J.;
Workman, P. Biochem. Pharmacol. 1992, 43, 1657.
3. (a) Melton, R. G.; Knox, R. J. Enzyme-Prodrug Strategies for Cancer Therapy;
Kluwer Academic/Plenum: New York, 1999. p. 272; (b) Denny, W. A. Curr.
Pharm. Des. 1996, 2, 281; (c) Denny, W. A. Eur. J. Med. Chem. 2001, 36, 577; (d)
Denny, W. A.; Wilson, W. R. J. Pharm. Pharmacol. 1998, 50, 387; (e) Denny, W. A.

7450
P. J. Burke et al. / Bioorg. Med. Chem. Lett. 21 (2011) 7447–7450
Curr. Pharm. Des. 2002, 8, 1349; (f) Roldán, M. D.; Pérez-Reinado, E.; Castillo, F.;
7. Hedley, K. A.; Stanforth, S. P. J. Heterocycl. Chem. 1995, 32, 529.
8. Burke, P. J.; Chun Wong, L.; Clegg, W.; Harrington, R. W.; Jenkins, T. C.; Knox, R.
J.; Meikle, I. T.; Stanforth, S. P. Tetrahedron Lett. 2010, 51, 3918.
9. Meth-Cohn, O. Adv. Heterocycl. Chem. 1996, 65, 1.
10. Zhu, Z.; Furr, J.; Buolamwini, J. K. J. Med. Chem. 2003, 46, 831.
11. Ajao, J. F.; Bird, C. W. J. Heterocycl. Chem. 1985, 22, 329.
12. Cherkupally, P.; Beier, P. Tetrahedron Lett. 2010, 51, 252. and references cited
therein.
Moreno-Vivián, C. FEMS Microbiol. Rev. 2008, 32, 474.
4. Cobb, L. M.; Connors, T. A.; Elso, L. A.; Khan, A. H.; Mitchley, B. V. C.; Ross, W. J.
C.; Whisson, M. E. Biochem. Pharmacol. 1969, 18, 1519.
5. Lavardo, J.; Reszka, A. P.; Moreria, R.; Neidle, S.; Paulo, A. Bioorg. Med. Chem.
Lett. 2010, 20, 7042.
6. Seville, S.; Phillips, R. M.; Shnyder, S. D.; Wright, C. W. Bioorg. Med. Chem. 2007,
15, 6353.