328
V. Castro-Castillo et al. / Bioorg. Med. Chem. Lett. 23 (2013) 327–329
problem is the screening of plant extracts and secondary metabo-
lites against the parasite.
Although 1-azabenzanthrones 9, 10 and 11 have been prepared
previously by dehydrogenation of their 2,3-dihydro analogues, the
low yields obtained following the published procedure of Walker
and Kempton,9 makes such an approach impractical for this pur-
pose. Therefore, we have only used this route to obtain 9A, 11A
and 10A and have preferred the methodology used by Kunitomo
et al.10 as exemplified in Scheme 2 for the synthesis of 11, which
was subsequently methylated to afford 10.
The bromination of 9 in CH3CN afforded 7 and 8 in poor yields.
The nitro (12, 13) and amino (4, 14) derivatives were prepared as
published.6
Cultures of the chloroquine-resistant FCR-3 strain of P. falcipa-
rum were grown at 37 °C in a 5% CO2 atmosphere on RPMI 1640
medium supplemented with gentamycin 0.1 mg/mL and 10%
heat-inactivated A+ human serum, as previously described.11 The
synthetic derivatives, dissolved in DMSO, were added at final con-
The oxoisoaporphines, exemplified by structures 2, 3 and 4, con-
stitute a small family of natural products characterised by their pla-
nar 7H-dibenzo[de,h]quinolin-7-one skeleton, commonly known in
the dye and pigment industries as 1-azabenzanthrone.3–5 Two oxo-
isoaporphines isolated from Menispermum dauricum DC, bearing a
nitrogen substituent at C-6 of the B ring (daurioxoisoaporphines A
and B, 2 and 3), and lakshminine (4), isolated from Sciadotenia tox-
ifera Krukoff & A.C. Smith, have demonstrated toxicity in several cell
lines.6 Due to their planar structure, these compounds could be ex-
pected to intercalate between DNA base pairs, and thus inhibit cell
replication and possibly exhibit anticancer properties. They might
also be cytotoxic as a consequence of their azaquinonoid structure,
which could reasonably interfere with mitochondrial electron
transport. While the literature records no data on the antimalarial
activity of alkaloids of this family, a number of synthetic 1-azaben-
zanthrones have been screened and have been found to be weakly
to moderately cytotoxic.6,7 We have now tested a collection of syn-
thetic 1-azabenzanthrones (including the oxoisoaporphine lakshm-
inine—4) in vitro against a chloroquine-resistant P. falciparum
strain.
centrations ranging from 250 to 0.1 lM. All experiments were re-
peated twice with two or three replicates each. The final DMSO
concentration was never greater than 0.1%. In vitro antimalarial
activity was measured using the [3H]-hypoxanthine (MP Biomedi-
cals, USA) incorporation assay.12 Results were expressed as the
concentration resulting in 50% inhibition (IC50) that was calculated
by a nonlinear regression logistic dose response model and the
mean IC50 values and standard deviation for each compound was
calculated. Values are means of n = 4–6.
4 R3= H, R4= H, R5= MeO, R6= NH2; 2,3
5 R3= H, R4= H, R5= H, R6= H; 2,3
6 R3= Br, R4= H, R5= H, R6= H; 2,3
R4 R3
The IC50 values of the tested 1-azabenzanthrones are collated in
Table 1, together with their cytotoxicities to normal human fibro-
blasts and their selectivity indices (S.I.).13
R5
R6
7 R3= Br, R4= H, R5= MeO, R6= H; 2,3
8 R3= H, R4= Br, R5= MeO, R6= H; 2,3
9 R3= H, R4= H, R5= MeO, R6= H; 2,3
9A R3= H, R4= H, R5= MeO, R6= H; 2,3
10 R3= H, R4= H, R5= MeO, R6= MeO; 2,3
11 R3= H, R4= H, R5= MeO, R6= OH; 2,3
12 R3= H, R4= H, R5= MeO, R6= NO2; 2,3
13 R3= H, R4= NO2, R5= MeO, R6= H; 2,3
14 R3= H, R4= NH2, R5= MeO, R6= H; 2,3
3
2
N
Of the series of 1-azabenzanthrones tested, only the 5-methoxy
compound (9) and its dihydro analogue (9A) exhibited low micro-
molar antiplasmodial activities. Although the latter is an order of
magnitude less potent than the reference drug chloroquine (1),
our results suggest that 9A might serve as a lead compound for fur-
ther development. Interestingly, when our values for 9 and 9A are
compared with the toxicities to normal human cells, these com-
pounds showed more than 10-fold antiplasmodial selectivity while
the others were either nonselective or more toxic to the fibroblasts
than to the parasite. Although this series of compounds is insuffi-
cient to draw firm conclusions as to a structure–activity relation-
ship, almost all the analogues with substituents on ring B have
O
1-Azabenzanthrone 5 was prepared improving on a published
procedure (Scheme 1).8 The synthesis of 5 was carried out in good
yield and without side product formation. Aromatisation of 5A (see
below) by air oxidation over Pd/C similarly afforded 5. The bromin-
ation of 5 in CH3CN afforded 6 in poor yield.
antiplasmodial activities at concentrations 6100
lM, while the
unsubstituted 1-azabenzanthrone and its 3-bromo derivative have
IC50 P 200
lM.
O
O
N
a
b
N
N
O
O
N
OH
A (96%)
B (77%)
N
O
N
d
c
e
O
O
HO
C (99%)
OMe
OH
D (54%)
E
Br
N
N
N
g
h
f
O
O
O
5
(80%)
6
(30%)
5A (66%)
Scheme 1. Reagents and conditions: (a) NaBH4, MeOH/dioxane, reflux, 3 h; (b) 37% HCl, reflux, 1 h; (c) 0.5 M KOH/MeOH, air, reflux, 24 h; (d) Me2SO4/MeOH, reflux, 3 h; (e)
37% HCl, reflux, 4 h; (f) H2SO4/SO3, 0–5 °C, 24 h; (g) Pd/C, toluene, reflux, 24 h; (h) Br2, CH3CN, 80 °C, 24 h.
Castro-Castillo, Vicente
