Structure-activity relationships for pyrido-, imidazo-, pyrazolo-, pyrazino-, and pyrrolophenazinecarboxamides as topoisomerase-targeted anticancer agents

Article abstract of DOI:10.1021/jm010330+

Heterocyclic phenazinecarboxamides were prepared by condensation of aminoheterocycles and 2-halo-3-nitrobenzoic acids, followed by reductive ring closure and amidation. They showed similar inhibition of paired cell lines that underexpressed topo II or overexpressed P-glycoprotein, indicating a non topo II mechanism of cytotoxicity and indifference to P-glycoprotein mediated multidrug resistance. Compounds with a fused five-membered heterocyclic ring were generally less potent than the pyrido[4,3-a]phenazines. A 4-methoxypyrido[4,3-a]phenazine (IC₅₀s 2.5-26 nM) gave modest (ca. 5 day) growth delays in H69/P xenografts with oral dosing.

Full text of DOI:10.1021/jm010330+

740
J . Med. Chem. 2002, 45, 740-743
Brief Articles
Str u ctu r e-Activity Rela tion sh ip s for P yr id o-, Im id a zo-, P yr a zolo-, P yr a zin o-,
a n d P yr r olop h en a zin eca r boxa m id es a s Top oisom er a se-Ta r geted An tica n cer
Agen ts
Swarna A. Gamage,† J ulie A. Spicer,† Gordon W. Rewcastle,† J ohn Milton,# Sukhjit Sohal,# Wendy Dangerfield,#
Prakash Mistry,# Nigel Vicker,# Peter A. Charlton,# and William A. Denny*,†
Auckland Cancer Society Research Centre, Faculty of Medical and Health Sciences, The University of Auckland,
Private Bag 92019, Auckland, New Zealand, and Xenova Ltd., 957 Buckingham Avenue, Slough, Berkshire SL1 4NL, U.K.
Received J uly 17, 2001
Heterocyclic phenazinecarboxamides were prepared by condensation of aminoheterocycles and
2-halo-3-nitrobenzoic acids, followed by reductive ring closure and amidation. They showed
similar inhibition of paired cell lines that underexpressed topo II or overexpressed P-
glycoprotein, indicating a non topo II mechanism of cytotoxicity and indifference to P-
glycoprotein mediated multidrug resistance. Compounds with a fused five-membered hetero-
cyclic ring were generally less potent than the pyrido[4,3-a]phenazines. A 4-methoxypyrido[4,3-
a]phenazine (IC50s 2.5-26 nM) gave modest (ca. 5 day) growth delays in H69/P xenografts
with oral dosing.
Tri- and tetracyclic DNA intercalating agents bearing
carboxamide-linked cationic side chains are well-known
as antiproliferative and anticancer agents, mediated
through their interaction with topoisomerase enzymes.1-3
Structure-activity studies show that for good activity
the carboxamide side chain must be positioned off the
short axis of the chromophore peri to the chromophore
nitrogen. For acridine-1 (2) and phenazine-2 (4) carboxa-
mides, small lipophilic substituents placed peri to the
same chromophore nitrogen, on the other ring (e.g., in
7, 8), result in increased DNA binding. Recent crystal-
lographic studies of acridine-4-carboxamide/oligonucle-
otide complexes suggest the reason for this may be
hydrophobic van der Waal interactions between these
substituents and the cytosine at the CpG step.4
a relatively short clinical half-life (0.28 h).7 In a search
for more potent dual topoisomerase inhibitors, we
considered phenazine analogues (e.g., 4), which show
broadly similar patterns of biological activity.2 Early
studies showed that the 8,9-benzophenazine analogue
(9) was a potent cytotoxin in L1210 leukemia cultures
(IC50 33 nM),2 and a drug development program based
on 9 as a lead resulted in the methoxy derivative (11;
XR 11576) that is in preclinical development.8 In this
paper we report the synthesis and structure-activity
relationships for related heterocyclic derivatives of 9
(compounds 12-28 of Table 1).
Ch em istr y
5-Amino-1-methoxyisoquinoline (33) was prepared
from 1-hydroxyisoquinoline (29) by conversion to the
1-chloro compound (30) with POCl3, followed by nitra-
tion in c.H2SO4/f.HNO3/KNO3 to give 1-chloro-5-ni-
troisoquinoline (31).9,10 Reaction of this with Na in dry
MeOH gave 1-methoxy-5-nitroisoquinoline (32),11 which
was hydrogenated over Pd/C to give the required 33 in
87% overall yield (Scheme 1A). The synthesis of 8-ami-
noisoquinoline (37) proved a particular problem. The
direct nitration of isoquinoline produces almost exclu-
sively the 5-nitroisoquinoline,12 as does nitration of
isoquinoline-N-oxide in sulfuric acid (64% 5-nitroiso-
quinoline-N-oxide and 7% 8-nitroquinoline-N-oxide).12
Reaction of the latter with POCl3 gave 8-nitroisoquino-
line (65%).13 A more direct route has been reported14
by reduction of 5-chloro-8-nitroisoquinoline with palla-
dized CaCO3 and ammonium acetate. However, an
attempted reduction of 5-chloro-8-nitroisoquinoline us-
ing the specified conditions gave only 5-chloro-8-ami-
noisoquinoline, in 70-80% yield. Removal of the 5-chlo-
ro by further hydrogenation gave small quantities of
The acridine-4-carboxamide DACA (XR-5000; 2) is a
dual inhibitor of both topo I and II enzymes5 and is
unaffected by cellular drug resistance due both to
enhanced efflux mechanisms and topoisomerase isozyme
switching (replacement of topo IIâ by topo IIR).6 How-
ever, 2 undergoes rapid oxidative metabolism and has
* To whom correspondence should be addressed. Ph: +64 9 373 7599
ext 6144. Fax: +64 9 373 7502. E-mail: b.denny@auckland.ac.nz.
† Auckland Cancer Society Research Centre.
#
Xenova Ltd.
10.1021/jm010330+ CCC: $22.00 © 2002 American Chemical Society
Published on Web 01/08/2002

Brief Articles
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 3 741
Sch em e 1a
Ta ble 1. Growth Inhibitory Properties of Tetracyclic
Phenazines in the Cell Line Panel
a
(i) POCl3; (ii) c.H2SO4/f.HNO3/KNO3; (iii) Na/MeOH; (iv) Pd/
C/H2/MeOH/Et3N; (v) NaNO2/Kl/H2O; (vi) c.H2SO4/KNO3.
Sch em e 2a
IC50
IC50
IC50
(nM)
(nM)a
ratiosb
J LA/ J LDg/
LX4/
P388c
LLd
J Lce J LC J LC H69Ph H69Pi
f
no. Fm
X
4
9
10
1100
37
9.4
10
1850
23
2500
93
0.8
0.8
1.0
0.9
A
A
H
OMe
67
35
23
294
123
1210
116
88
1.9
1.4
1.2
2.1
3.2
>4
1.3
0.64
1.2
11 A1 OMe
12
13
14
15
16
17
18
B
B
B
C
D
D
D
H
89
317
0.7
0.9
OMe
OH
H
H
OMe
OH
23
2700
16
5.6
14
11
6.4
2.4
79
28
17
0.8
3.0
0.7
1.0
2.8
0.7
21
31
133
24
24
1.05
1.2
1.2
19 D1 OMe
20 D2 OMe
21
22
23
24
25
26
27
28
E
F
G
G
H
H
I
H
H
H
Me
H
42
85
98
8.4
6.8
59
300
220
15
20
54
66
5.9
4.1
25
41
74
165
194
44
21
98
0.8
1.0
1.1
1.0
4.2
0.8
0.8
0.7
0.9
124
234
524
109
107
130
183
183
1.2
1.5
0.95
0.92
4.5
1.2
5.2
1.7
a
(i) 2-Bromo-3-nitrobenzoic acid/Cu/Cul/N-ethylmorpholine/iso-
propanol/90 °C; (ii) 5 M NaOH/NaBH4/reflux; (iii) CDI/DMF/N,N-
dimethylethylenediamine.
1.1
4.7
0.8
0.9
Me
159
dimethylethylenediamine to give the required carboxa-
mides of Table 1. Preparation of the chiral side chain
amines for 19 and 20 has been reported.8
J
doxorubicin
22
9.6 4.39 12.7
27 137
a
IC50; concentration of drug (nM) to reduce cell number to 50%
of control cultures (see text). Number is the average of at least
two independent determinations. Coefficient of variation 12-18%.
Resu lts a n d Discu ssion
b
Ratios of IC50s in the cell lines shown. c Murine P388 leukemia.
The compounds were evaluated in a panel of cultured
human and murine tumor cells (Table 1). The murine
P388 leukemia5 and murine Lewis lung carcinoma15
lines provide comparison with previous compounds. The
three human leukemia (J urkat) lines16 provide insights
into the mechanism of cytotoxicity. The J LA and J LD
lines, developed for resistance to amsacrine and doxo-
rubicin, respectively, are 123-fold and 110-fold more
resistant than the wild-type J LC, respectively, to the
topo II inhibitor amsacrine because of a reduced level
of topo II enzyme.16 Table 1 shows the IC50 ratios (J LA/
J LC and J LD/J LC) for the new compounds; ratios of ca.
<2-fold suggest a non topo II mediated mechanism of
action. The H69 parental human small cell lung carci-
noma cell line (H69/P) and the derived drug resistant
line H69/LX4, which overexpresses P-glycoprotein,8
were used to evaluate the abilities of the compounds to
overcome this type of multidrug resistance.
d
Murine Lewis lung carcinoma. e J LC: wild-type human J urkat
leukemia. f J LA (amsacrine-resistant J urkat)/J LC. J LD (doxoru-
g
h
bicin-resistant J urkat)/J LC. H69P: parental human small cell
lung carcinoma cell line. LX4: H69 line overexpressing P-
glycoprotein.
i
8-aminoisoquinoline, together with ring-reduced prod-
ucts and poor overall recovery of material. We finally
prepared 8-aminoisoquinoline (37) by diazotization/
iodination of 5-aminoisoquinoline (34), followed by
nitration of the resulting 5-iodoisoquinoline (35) with
H2SO4/KNO3 to give 5-iodo-8-nitroisoquinoline (36).
Reduction of this with Pd/C/H2 in MeOH gave the
desired 37 quantitatively (Scheme 1B).
Reaction of these aromatic amines and 2-iodo- or
2-bromo-3-nitrobenzoic acids under J ourdan-Ullmann
conditions2 gave 2-arylamino-3-nitrobenzoic acids
(Schemes 2 and 3) in good yields (except with very
electron deficient amines such as 37). Reductive cycliza-
tion with NaBH4/NaOMe then gave the tetracyclic
phenazine acids, which were condensed with N,N-
The 8,9-benzophenazine 9 is significantly more potent
than the phenazinecarboxamide 4 (30-60-fold across
the cell line panel; Table 1), possibly due to increased

742 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 3
Brief Articles
Sch em e 3a
F igu r e 1. Growth delays induced in H69/P xenografts in CD1
mice by compounds 11 ([) and 17 (1), using oral dosing on a
q7d×3 schedule. Data are expressed as mean ( SEM from
groups of eight mice.
pyrazino[2,3-a]phenazine 22, combining the aza atoms
of both 12 and 21, was broadly intermediate in potency.
Compounds 23-28 contained a fused five-membered
ring to explore placing a small lipophilic substituent
(methyl in compounds 26 and 28) in a slightly different
spatial position compared to the OMe of the benzo- and
pyridophenazines. The pyrazolo[3,4-a]phenazines 23
and 24 have no substituent in this position, but differ
by an N-methyl group on the chromophore concave
surface. The NMe analogue 24 is considerably more
potent than both the NH compound 23 and the parent
benzophenazine 9 in the P388, LL, and J urkat cell lines
but not in the H69 lines, while the situation is reversed
in the 3H-pyrazolo[4,3-a]phenazines 25 and 26 (despite
the N-methyl group of 26 being in a similar spatial
region to the OMe of the benzo- and pyridophenazines).
The imidazo[4,5-a]phenazine 27 and the methylpyrrolo-
[3,2-a]phenazine 28 were considerably less active than
the parent benzophenazine 9.
Compound 17 was evaluated in vivo in H69/P xeno-
grafts in CD1 mice, using oral dosing on a q7d×3
schedule, and compared with the lead compound 11 (XR
11576).8 Both were tested at the highest dose that gave
no obvious adverse effects. Although 17 appeared more
toxic than 11 in this model (10 versus 75 mg/kg), it was
less effective at its MTD, providing a modest (ca. 5 day)
delay in growth of the tumor to 5× initial volume,
compared with an approximate 11 day growth delay for
11 after oral dosing (Figure 1).
a
(i) 2-Iodo-3-nitrobenzoic acid/Cu/Cul/N-ethylmorpholine/iso-
propanol/70 °C; (ii) CH2N2/Et2O/20 °C; (iii) 50% EtOH/2 M NaOH/
NaBH4/70-75 °C; (iv) CDI/DMF/50 °C/N,N-dimethylethylenedi-
amine.
DNA binding;2 recent crystal structure studies of related
acridinecarboxamides/oligonucleotide complexes show
that the extra aromatic ring could contribute additional
hydrophobic binding.4 The 4-OMe group in 10 results
in a further 2-3-fold increase in potency, and the
development of side chain SAR for this chromophore led
to 11 (XR11576), which has been selected for preclinical
development.8
The pyrido[2,3-a]phenazine compounds 12 and 13
were on average 3-4-fold less potent than the ben-
zophenazine analogues 9 and 10. In both series, sub-
stitution with OMe resulted in increased potency, and
both series had low ratios in the J LC/J LA and H69/LX4
cell line pairs, suggesting respectively nontopo II medi-
ated mechanism of action and little sensitivity to
P-glycoprotein-mediated efflux. The OH analogue 14
was considerably less potent than the OMe analogue.
In this isomer series the aza atom can potentially form
metal chelates, but it is not known if this is a factor in
its biological effects. The pyrido[3,4-a]phenazine isomer
15 was more potent than 9 in the P388, LL, and J LC
lines (although less active in the H69 lines), but the
OMe analogue could not be prepared.
Con clu sion s
The tetracyclic pyridophenazine-11-carboxamides, and
especially the pyrido[4,3-a]phenazine isomers, are po-
tent cytotoxins with an interesting profile of in vitro and
in vivo biological activity, with much higher potency
than the related phenazinecarboxamides,2 while retain-
ing desirable properties such as equivalent activity in
topo II deficient and P-glycoprotein overexpressing cell
lines.
The most interesting were the pyrido[4,3-a]phena-
zines 16-20. Both 16 and 17 were, on average, more
potent than the parent compounds 9 and 10. The
unsubstituted 16 showed ratios of about 3 in the J urkat
cell line pairs, still suggestive of a non-topo II action,
but both had values of unity or less in the H69/LX4 pair,
suggesting no sensitivity to P-glycoprotein mediated
resistance. The OH analogue 18 was again less potent
than the OMe. In contrast to the benzophenazines,8 the
chiral side chain analogues 19 and 20 did not show any
increase in potency over 17. The pyrido[3,2-a]phenazine
isomer 21 was of comparable potency to 9, while the
Exp er im en ta l Section
Ch em istr y. Analyses were carried out in the Microchemical
Laboratory, University of Otago, Dunedin, NZ. Melting points
were recorded on an Electrothermal melting point apparatus.
NMR spectra were obtained on a Bruker DRX-400 spectrom-
1
eter (400 MHz for H and 100 MHz for 13C) and are referenced

Brief Articles
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 3 743
to Me4Si. Column chromatography was carried out on Merck
300-400 mesh silica gel (flash) or Merck 24-40 µm silica gel
(dry flash). Petroleum ether (PE) refers to the fraction boiling
at 40-60 °C; hexanes refers to the fraction boiling at 60-65
vitro and in vivo protocols, analyses, and additional references.
This material is available free of charge via the Internet at
http://pubs.acs.org.
°C. Mass spectra were determined on
spectrometer.
a VG-75SE mass
Refer en ces
(1) Atwell, G. J .; Rewcastle, G. W.; Baguley, B. C.; Denny, W. A.
Potential antitumor agents. 50. In vivo solid tumor activity of
derivatives of N-[2-(dimethylamino)ethyl]acridine-4-carboxam-
ide. J . Med. Chem. 1987, 30, 664-669.
(2) Rewcastle, G. W.; Denny, W. A.; Baguley, B. C. Potential
antitumor agents. 51. Synthesis and antitumor activity of
phenazine-1-carboxamides. J . Med. Chem. 1987, 30, 843-851.
(3) Deady, L. W.; Kaye, A. J .; Finlay, G. J .; Baguley, B. C.; Denny,
W. A. Synthesis and antitumor properties of N-[2-(dimethylami-
no)ethyl]carboxamide derivatives of fused tetracyclic quinolines
and quinoxalines; a new class of putative topoisomerase inhibi-
tors. J . Med. Chem. 1997, 40, 2040-2046.
(4) Adams, A.; Guss, J . M.; Collyer, C. A.; Denny, W. A.; Prakash,
A.; Wakelin, L. P. G. Acridinecarboxamide topoisomerase
poisons: structural and kinetic studies of the DNA complexes
of 5-substituted 9-amino-(N-(2-dimethylamino)ethyl)acridine-4-
carboxamides. Mol. Pharmacol. 2000, 58, 649-658.
(5) Twelves, C. J .; Gardner, C.; Flavin. A.; Sludden, J .; Dennis, I.;
De Bono, J .; Beale, P.; Vasey, P.; Hutchison, C.; Macham, M.
A.; Rodriguez, A.; J udson, I.; Bleehen, N. M. Phase I and
pharmacokinetic study of DACA (XR5000): a novel inhibitor of
topoisomerase I and II. Br. J . Cancer 1999, 80, 1786-1791.
(6) Finlay, G. J .; Marshall, E. S.; Mathews, J . H. L.; Paull, K. D.;
Baguley, B. C. In vitro assessment of N-[2-(dimethylamino)ethyl]-
acridine-4-carboxamide (DACA), a new DNA-intercalating an-
titumor drug with reduced sensitivity to multidrug resistance.
Cancer Chemother. Pharmacol. 1993, 31, 401-406.
(7) Kestell, P.; Dunlop, I. C.; McCrystal, M. R.; Evans, B. D.; Paxton,
J . W.; Gamage, R. S. K. A.; Baguley, B. C. Plasma pharmaco-
kinetics of N-[2-(dimethylamino)ethyl]acridine-4-carboxamide in
a phase I trial. Cancer Chemother. Pharmacol. 1999, 44, 45-
50.
(8) Vicker, N.; Burgess, L.; Chuckowree, I. S.; Dodd, R.; Folkes, A.;
Hardick, D. J .; Hancox, T. C.; Miller, W.; Milton, J .; Sohal, S.;
Wang, S.; Wren, S. P.; Charlton, P.; Dangerfield, W.; Liddle, C.;
Mistry, P.; Stewart, A.; Denny, W. A. Novel angular ben-
zophenazines: dual topoisomerase I and topoisomerase II inhibi-
tors as potential anticancer agents. J . Med. Chem. 2002, 45,
721-739.
(9) Suto, M. J .; Turner, W. R.; Werbel, L. M. Preparation of
substituted dihydroisoquinolinones and related compounds as
radiation therapy potentiators and chemotherapeutic agents.
Eur. Pat. Appl. EP 355,750; CAN 113, 132025.
(10) Robinson, R. A. 1-(Dialkylaminoalkylamino)isoquinolines. J . Am.
Chem. Soc. 1947, 69, 1939-1942.
(11) Wozniak, M.; Baranski, A.; Nowak, K.; Poradowska, H. Regi-
oselectivity of the amination of some nitroisoquinolines by liquid
ammonia/potassium permanganate. Leibigs Ann. Chim. 1990,
653-657.
(12) Srivastava, S. K.; Chauhan, P. M. S.; Agarwal, S. K.; Bhaduri,
A. P.; Singh, S. N.; Nigar Fatma, N.; Chatterjee, R. K.; Bose, C.;
Srivastava, V. M. L. Syntheses and antifilarial profile of 5-amino
and 5,8-diaminoisoquinoline derivatives: a new class of antifi-
larial. Bioorg. Med. Chem. Lett. 1996, 6, 2623-2628.
(13) Dewar, M. J . S.; Maitlis, P. M. Electrophilic substitution. Part
XI. Nitration of some six-membered nitrogen-heterocyclic com-
pounds in sulphuric acid. J . Chem. Soc. 1957, 2521-2528.
(14) Osborn, A. R.; Schofield, K.; Short, L. N. Aminoisoquinolines,
-cinnolines, and -quinazolines. (A) The basic strengths and
ultraviolet absorption spectra. (B) Infrared spectra. J . Chem. Soc.
1956, 4191-4206.
(15) Wilkoff, L. J .; Dulmadge, E.; Chopra, D. P. Viability of cultured
Lewis lung cell populations exposed to â-retinoic acid (40753).
Proc. Soc. Exp. Biol. Med. 1980, 163, 233-236.
Ar om a tic Am in es. 5-Amino-1-methoxyisoquinoline (33)
and 8-aminoisoquinoline (37) were prepared by the routes
outlined in Scheme 1A and 1B, respectively. Experimental
details are provided in Supporting Information.
N-[2-(Dim eth ylam in o)eth yl]4-m eth oxypyr ido[2,3-a ]ph -
en a zin e-11-ca r b oxa m id e (13): E xa m p le of Gen er a l
Meth od . 4-Methoxy-8-aminoquinoline (38) (1.15 g, 6.6 mmol),
2-iodo-3-nitrobenzoic acid (3.86 g, 13.20 mmol), and Cu/CuI
(catalytic) in N-ethylmorpholine (10 mL) and 2-propanol (10
mL) were heated for 2 days at 70 °C, then cooled, boiled in
aqueous ammonia and charcoal/Celite, and filtered. The
filtrate was acidified with c.HCl, and the precipitate was
recrystallized from MeOH/H2O to give 2-(4-methoxy-8-quino-
lineamino)-3-nitrobenzoic acid (41) (1.6 g, 74%): mp (MeOH/
H2O) 228-231 °C (dec); 1H NMR [(CD3)2SO] δ 4.21 (s, 3H,
OCH3), 7.22 (d, J ) 7.9 Hz, 1 H, ArH), 7.23 (d, J ) 8.0 Hz, 1
H, ArH), 7.36 (br s, 1 H, ArH), 7.48 (t, J ) 7.8 Hz, 1 H, ArH),
7.86 (d, J ) 7.9 Hz, 1 H, ArH), 8.17 (dd, J ) 8.2, 1.5 Hz, 1 H,
H-4 or H-6), 8.25 (dd, J ) 7.7, 1.5 Hz, 1 H, H-4 or H-6), 8.95
(d, J ) 5.6 Hz, 1 H, ArH), 10.40 (brs, 1 H, NH), 13.50 (br, 1 H,
CO2H). HRMS [M+] calcd C17H13N3O5: 339.0855. Found:
339.0857. Anal. (C17H13N3O5‚2H2O) C, N: H; found, 3.7;
calculated, 4.5%.
A solution of acid 41 (1.35 g, 4.00 mmol) in 5 N aqueous
NaOH (50 mL) was treated with NaBH4 (0.76 g, 20 mmol),
and the mixture was then heated under reflux for 3 h, cooled,
and acidified with c.HCl. The resulting solution was neutral-
ized with aqueous ammonia to precipitate 4-methoxypyrido-
[2,3-a]phenazine-11-carboxylic acid (42) (71%): mp (MeOH/
H2O) 281-285 °C; 1H NMR (CF3CO2D) δ 4.62 (s, 3 H, OCH3),
7.95 (d, J ) 7.7 Hz, 1 H, H-3), 8.53 (dd, J ) 8.9, 7.3 Hz, 1 H,
H-9), 8.76 (d, J ) 9.7 Hz, 1 H, H-5), 8.96 (dd, J ) 8.9, 1.2 Hz,
1 H, H-8), 8.98 (d, J ) 9.7 Hz, 1 H, H-6), 9.16 (dd, J ) 7.2, 1.1
Hz, 1 H, H-10), 9.37 (d, J ) 7.1 Hz, 1 H, H-2). Anal.
(C17H11N3O3) C, H, N.
A mixture of 42 (0.41 g, 1.33 mmol) and CDI (0.43 g, 2.66
mmol) in DMF (10 mL) was heated and stirred at 50-60 °C
for 2 h, then treated with N,N-dimethylethylenediamine (0.5
mL, excess) at room temperature for 1 h. Evaporation of
solvent and dilution with water, followed by extraction into
CH2Cl2 (3 × 50 mL), gave a crude product (0.39 g), which was
chromatographed on alumina. Elution with MeOH/CH2Cl2 (1:
99) gave 13 (58%): mp (CH2Cl2/hexane) 180-181 °C; 1H NMR
[CDCl3] δ 2.41 [s, 6 H, N(CH3)2], 2.96 [t, J ) 6.9 Hz, 2 H, CH2N-
(CH3)2], 3.91 (q, J ) 6.3 Hz, 2 H, CH2NH), 7.12 (d, J ) 5.4 Hz,
1 H, H-3), 8.03 (d, J ) 9.5 Hz, 1 H, H-5), 8.05 (dd, J ) 8.5, 7.2
Hz, 1 H, H-9), 8.39 (dd, J ) 8.6, 1.4 Hz, 1 H, H-8), 8.45 (d, J
) 9.6 Hz, 1 H, H-6), 8.97 (d, J ) 5.4 Hz, 1 H, H-2), 9.02 (dd,
J ) 7.2, 1.4 Hz, 1 H, H-10), 11.88 (br s, 1 H, CONH). Anal.
(C21H21N5O2) C, H, N.
Ack n ow led gm en t. We thank Drs. Bruce Baguley
and Graeme Finlay for some of the cell culture data.
This work was supported by Xenova Ltd. and by the
Auckland Division of the Cancer Society of New Zealand.
(16) Finlay, G. J .; Baguley, B. C.; Snow, K.; J udd, W. Multiple
patterns of resistance of human leukemia cell sublines to
amsacrine analogs. J . Natl. Cancer Inst. 1990, 82, 662-667.
Su p p or tin g In for m a tion Ava ila ble: Full synthetic de-
tails for all compounds, together with brief outlines of the in
J M010330+