Polycyclic aromatic compounds as anticancer agents: Synthesis and biological evaluation of dibenzofluorene derivatives

Article abstract of DOI:10.1016/S0968-0896(00)00213-3

Highly regioselective electrophilic substitution of dibenzofluorene was achieved and the nitro derivative was transformed to a variety of new anticancer agents. Copyright (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0968-0896(00)00213-3

Bioorganic & Medicinal Chemistry 8 (2000) 2693±2699
Polycyclic Aromatic Compounds as Anticancer Agents:
Synthesis and Biological Evaluation of Dibenzo¯uorene Derivatives
Frederick F. Becker,* Chhanda Mukhopadhyay, Linda Hackfeld, Indrani Banik
*
and Bimal K. Banik
The University of Texas, M. D. Anderson Cancer Center, Department of Molecular Pathology Box-89,
1
515 Holcombe Blvd., Houston, TX, 77030, USA
Received 10 January 2000; accepted 11 July 2000
AbstractÐHighly regioselective electrophilic substitution of dibenzo¯uorene was achieved and the nitro derivative was transformed
to a variety of new anticancer agents. # 2000 Elsevier Science Ltd. All rights reserved.
Introduction
Results and Discussion
The synthesis of polycyclic aromatic ring systems by
various methodologies has been published extensively.1
The carcinogenic properties of such compounds have
been explained by advancing a number of di€erent
Recently, we have developed a novel oxidation method
for the conversion of benzylic methylenes to benzylic
ketones in polycycylic systems by sodium bismuthate.
Thus, pentacyclic dibenz¯uorene 1 was oxidized to the
dibenz¯uorenone 2 in good yield (Scheme 1).
7
2
mechanisms. The use of polycyclic aromatic com-
pounds and their derivatives as anticancer agents has
also been explored, but less is known regarding this
We have also shown a facile reduction of the polycyclic
aromatic nitro compounds to polycyclic aromatic
amines (for example, 3 to 4) by samarium metal in the
3
activity. In the main, the antitumor activity of these
compounds is proposed to depend on intercalation with
4
8
or covalent binding to DNA. However, many other
presence of catalytic amounts of iodine (Scheme 2).
sites of interaction such as the cell membrane have been
identi®ed. From a prior limited study, we determined
that suitably substituted chrysene derivatives act on the
cancer cell through interactions with the membrane.5
Using these two methods, we examined the synthesis
and the antitumor activities of structurally complex,
angular dibenzo¯uorene [a,g] polycyclic systems with a
very reactive bridged methylene group. We hypothesized
that a bridged unit in the polycyclic aromatic system
could play an important role in this function as it could
form cation, anion, and radical intermediates. This
communication describes the electrophilic substitution
reaction of the 13H dibenzo¯uorene (1) for the ®rst
time. The structure±activity relationships of several new
diamides and diamines (13, 14, 15 and 16) are also
reported here.
Although using potentially carcinogenic/mutagenic
compounds to derive antitumor agents might appear a
questionable approach, a large body of information
supports this concept. Many reports have demonstrated
that alteration of the structure of PAH can mitigate
their deleterious e€ects, emphasizing their interaction
with speci®c cell organelles to evoke speci®c cytotoxic
2
a
reactions. As a result, many of the antitumor agents
that are in current clinical use are derived from com-
pounds such as carbazoles, anthracenes, and related
We prepared pentacyclic dibenzo[a,g]¯uorene (1) in 20%
9
6
structures.
yield by following the method reported by Harvey. The
hydrocarbon 1 can be prepared via alkylation of the
enamine 5a with the bromide 6 and cyclodehydration±
aromatization of the ketone 5b (Scheme 3).
Functionalization of benzene and naphthalene deriva-
a,10
*Corresponding author: Tel.: +1-713-792-7749; fax: +1-713-792-5940;
e-mail: banik@mdanderson.org
tives by electrophilic reaction1
is routine organic
0968-0896/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(00)00213-3

2
694
F. F. Becker et al. / Bioorg. Med. Chem. 8 (2000) 2693±2699
Scheme 1.
quartet present in 1 was absent from the nitro com-
pound 7. The spectrum of 7 showed a new singlet at d
.39 and a new doublet at d 8.44 (J=8.75 Hz). We
eliminated positions C , C , C and C for the nitro
8
1
4
7
10
group because of the singlet at d 8.44. The positions C5,
Scheme 2.
C , C and C were eliminated based on the homo-
6
8
9
nuclear decoupling and COSY experiment. That region
d 7.5±7.6) of the hydrocarbon 1 remained una€ected in
7 clearly ruled out positions C and C for the nitro
(
chemistry. The orientation of the electrophile in such
monocyclic or bicyclic derivatives is predictable. A
similar substitution reaction in a polycyclic aromatic
system is extremely dicult, and for polycyclic non-
alternate hydrocarbon it is poorly predictable. In fact,
little is known about the electrophilic substitution reac-
2
3
group in 7. We eliminated position C because of the
12
13
down®eld doublet at d 8.44. The C NMR spectrum of
7 showed the presence of nine quaternary carbons. The
signal at d 123.58 due to the C11 carbon (veri®ed by
HETCOR study) was not present in the spectrum of 7.
A new peak at d 145.24 appeared because of the nitro
group. Thus, we assigned C11 as the site of the nitro
group in 7 based on extensive NMR study.
11
tion in polycyclic aromatic system.
We planned to link a 4-carbon side chain with a
heterocyclic base at the end to the aromatic ring
through nitrogen. Therefore, our task was to prepare
amino dibenzo¯uorene 8 for the subsequent derivatiza-
tion. Towards this goal, we reacted the ketone 2 with
nitric acid in acetic acid under di€erent conditions but
failed to produce the desired nitro derivative. However,
the hydrocarbon 1 produced a single nitro compound 7
The site of electrophilic attack was further veri®ed by
computer-assisted charge density calculation of the Whe-
13
lands intermediates. The relative energies of the Wheland
intermediates of each of the carbons were calculated as
C1 (6.947), C2 (21.278), C3 (13.68), C4 (11.008), C5
(14.403), C (16.919), C (0.0), C (10.558), C (8.591),
6
7
8
9
ꢀ
with nitric acid±acetic acid at 0±5 C in 80% yield. The
C10 (0.894), C11 (2.325), C12 (19.638), and C13 (7.358).
These values indicated that preferential attack can take
place at positions 7, 10, and 11. On the basis of NMR
data we eliminated positions 7 and 10.
location of the nitro group in the aromatic system was
determined by NMR spectra. The NMR spectrum
(400 MHz) of the known hydrocarbon 1 was taken.
Based on the homonuclear decoupling and COSY
NMR studies, all of the protons in 1 were assigned as
follows: (400 MHz, CDCl ) d 8.81 (1H, d, J=8.46 Hz,
In order to study the electrophilic substitution reaction of
the hydrocarbon 1, we carried out acetylation by acetic
anhydride in the presence of anhydrous aluminum chlo-
ride using dichloromethane as solvent. The crude NMR
showed the presence of one acetyl group. The location
of the acetyl group in 9 was tentatively assigned based
on the location of the nitro group in the aromatic sys-
tem (Scheme 1). Regioselective electrophilic substitution
3
H ), 8.50 (1H, d, J=8.68 Hz, H ), 8.02 (1H, d,
6
7
J=8.22 Hz, H ), 7.89±7.96 (3H, m, H , H , H ), 7.70
1
4
5
10
and 7.79 (2H, ABq, J=8.23 Hz, H , H ), 7.60±7.65
1
1
12
(
assignments were supported by the data reported by
1H, m, H ), 7.45±7.54 (3H, m, H , H , H ). These
8 2 3 9
1
2
Jones et al. The NMR spectrum revealed that the AB
Scheme 3.

F. F. Becker et al. / Bioorg. Med. Chem. 8 (2000) 2693±2699
2695
of 1 is interesting and the use of 9 will be described in
subsequent publications.
transformation. Therefore, in order to produce the dia-
mino compounds, reduction of 13 was carried out by this
reagent under a variety of conditions. However, neither of
the desired diamino compounds 16a or 16b could be
isolated from the complex reaction mixtures by this
method. After many experiments, diborane was found
to be the reagent of choice for this reaction and it
resulted in a good yield of the amino compound 16.
Reduction of the nitro compound 7 to the amino com-
pound 8 was carried out by following our own method-
8
ology. We demonstrated that samarium±iodine is
highly e€ective for the transformation of nitro to the
amino compound. Our next task was to prepare the side
chains 12 and to couple them to the amine 8. The acid
1
2 was prepared by re¯uxing succinic anhydride (10)
These derivatives were tested in our Center's Core
Analytical Laboratory against three tumor lines of ani-
mal origin and six of human origin, all of which have
been used in the NCI panel for the testing of che-
motherapeutic agents. As an approach to an operational
de®nition of in vitro, antitumor cytotoxicity (activity), we
chose to compare the e€ects of our compounds with those
of cisplatin.
with piperidine (11a) and N-methylpiperazine (11b). The
amine 8 was then condensed with the side chains 12 by
mixed anhydride method. Many other condensing
agents, such as DCC14 and HOBt, failed to give the
amides 13 (Scheme 4).
The desired diamides 13a and 13b were isolated by column
chromatography. The benzylic methylene group in 13
was oxidized by molecular oxygen15 to get the ketones
The antitumor activity of these newly synthesized diben-
zo¯uorene derivatives 13, 14, 15 and 16 were tested and a
comparison with respect to cisplatin shown in Table 1.
The results of these tests are of interest both in terms of
their antitumor e€ects and in the signi®cant alterations
in activity that occurred with minor structural modi®ca-
tions. Compound 13b was more active than cisplatin
against four of the ®ve tumors against which cisplatin
14a and 14b. The oxidation of 13 that has a bulky side
chain is noteworthy since it might be expected to inter-
fere with the reaction. The ketone 14 was subsequently
reduced to the alcohol 15.
Reduction of the amide functionality to the amino group
by lithium aluminum hydride is a standard chemical
Scheme 4.
a
Table 1. IC50 (mM) of compounds 13 to 15 MTT assay (72 h continuous exposure)
Cell line
Cisplatin
13a
13b
14a
14b
15a
15b
16a
16b
B16
BRO
7.33
5.66
Ð
128.63
103.44
33.22
53.50
50.83
88.06
95.64
33.44
4.31
3.88
3.45
3.88
4.31
4.09
4.31
3.88
3.88
121.93
120.64
74.15
51.24
78.26
97.94
81.72
31.35
61.40
12.35
12.77
7.33
29.70
31.64
12.05
17.44
19.16
18.94
19.80
13.99
10.08
11.05
12.51
4.17
4.04
4.28
Ð
1.61
3.64
±
HL-60
L1210
MCF-7
OVCAR 3
P388/0
PC 3
Ð
2.93
6.05
Ð
±
15.99
Ð
Ð
1.66
15.99
25.75
13.82
2.72
27.22
21.78
12.30
8.76
4.38
6.46
13.55
4.75
3.09
4.51
4.04
4.04
4.36
1.84
4.36
3.44
3.21
HT-29
>100
aAll data were provided as IC50 values (mM) and assays were conducted by 72 h continuous exposure using the MTT method. The ®nal concentra-
tion of solvent was <0.625% which was not toxic to the cells. All dilutions were made in RPMI 1640 with 10% FBS. The cytotoxity data is based
on at least three separate experiments with deviations within 0.2 mM.

2
696
F. F. Becker et al. / Bioorg. Med. Chem. 8 (2000) 2693±2699
between 10 and 20 mM against three, and above 20 mM
against three. In a similar manner 15b had an IC50 less
than 5 mM against two lines, less than 10 mM against
three, and above 10 mM against four.
Experimental
General methods
Scheme 5.
All reactions described in this paper were carried out
under a well-ventilated hood. CH Cl and THF were
2
2
was used. This places it as one of the most e€ective
compounds we have yet derived from a polycyclic
hydrocarbon. The addition of a ketone group at posi-
tion 13 created the ¯uorenone compound 14b and
resulted in a decrease of activity against all but the ani-
mal leukemic lines L1210 and P388. The presence of a
piperidine heterocyclic group at the terminus of the
alkyl chain appears to render these compounds rela-
tively inactive when compared with those that terminate
with the heterocyclic N-methyl piperazine group. Com-
pounds 13a versus 13b and 14a versus 14b are striking
examples of this drastic alteration in activity and appear
to represent an o€/on phenomenon. While addition of a
hydroxyl group at position 13 in compound 15b pro-
duced either a reduction or no signi®cant alteration in
activity when compared with its parent compound 13b,
this addition in 15a signi®cantly increased activity when
compared with 13a.
dried and freshly distilled in the usual way before use.
IR spectra were recorded on a Perkin±Elmer instrument
and UV spectra were recorded on a Perkin±Elmer
instrument in nm. NMR spectra were recorded on Bruker
200 MHz and 300 MHz spectrometers. Chemical shifts
were reported as d values in parts per million down®eld
from tetramethyl silane as the internal standard in
CDCl . Mass spectra were obtained on a Micromass
3
VG platform with a single quadrapole and ®tted with an
electrospray source. Elemental analyses were performed
by Schwarzkopf Microanalytical Laboratory, Inc., New
York. Melting points were taken in open capillary tube
and are not corrected. Column chromatography was
carried out with Aldrich silica gel (230 mesh). TLC was
run with precoated silica gel plate. Na SO was used as
2
4
the drying agent after all the extractions.
11-Nitro-13H-dibenzo¯uorene (7). To an ice-cold solu-
tion of the hydrocarbon 1 (900 mg, 3.38 mmol) in THF
(30 mL) was added glacial acetic acid (50 mL). Nitric acid
(90%, 14 mL) was added dropwise to the stirred solution
5
Previous work in our laboratory with chrysene as the
PAH nucleus demonstrated that reduction of the dia-
mides signi®cantly increased antitumor activity, espe-
cially for the relatively inactive compounds. Reduction
of 13a and 13b resulted in their diamine derivatives 16a
and 16b, respectively. An impressive increase was seen
when 13b was compared to 16b against B16 and
OVCAR; while in all other instances, 16b were equally
active.
ꢀ
of the hydrocarbon 1 at 0±5 C. After the addition was
ꢀ
complete, stirring was continued for 2 h (0 C±RT). The
reaction mixture was poured into crushed ice, ®ltered,
and washed with water until free from acid. The crude
product was crystallized from CH Cl ±hexanes to
2
2
a€ord 7 (80% yield).
1
1-Amino-13H-dibenzo¯uorene (8). Nitro compound 7
Even more striking was the enhancement of activity
when the diamine 16a was compared with its diamide
analogue 13a. While IC50 of the latter was greater than
(0.80 mmol) was reduced by samarium metal (3 mmol)
in the presence of iodine as described in our previous
publication to a€ord the amine 8 as a dark semi-solid
7
3
0 mM against every cell line, 16a demonstrated IC50 of
mass; yield 90%; UV: lmax 231.96 (log e=2.26), lmax
264.26 (log e=2.76), lmax 366.09 (log e=4.33); IR
(neat) 3422, 3346, 3050, 2954, 2917, 2849, 1699, 1621,
less than 5 mM against seven out of seven tumor lines.
The cellular basis for the remarkable di€erences of
activity of the two series of compounds in all cell lines
has not been identi®ed.
�
1 1
1587, 1515, 1456 cm ; H NMR (200 MHz) d 8.84±8.82
(1H, d), 8.47±8.44 (1H, d), 8.04±7.80 (3H, m), 7.69±7.43
(
7H, m), 7.16 (1H, s), 4.18 (2H, s); 13C NMR (400 MHz)
1
1
1
43.60, 141.55, 140.80, 139.05, 131.10, 130.45, 130.14,
28.89, 128.64, 127.55, 126.45, 126.28, 124.62, 124.28,
24.07, 123.76, 123.30, 121.89, 120.95, 107.78, 36.48;
Conclusion
+
In conclusion, we have demonstrated for the ®rst time
electrophilic substitution reaction of 13H-dibenzo[a,g]-
Mass: 282.04 (M+H) .
¯
uorene (1). The prominent antitumor activity of some
11-Acetyl-13H-dibenzo¯uorene (9). To an ice-cold solu-
tion of the hydrocarbon 1 (900 mg, 3.38 mmol) in dry
CH Cl (50 mL) was added acetic anhydride (460 mg,
1
6
of the new compounds 13, 14, 15 and 16 opens up an
approach to research at the chemical:biologic interface.
That the agents described herein are producing certain
of their in vitro e€ects against human and animal cancer
lines through a speci®c cytotoxicity is exempli®ed by the
2
2
4.5 mmol) followed by anhydrous aluminum chloride
(800 mg, 6.0 mmol), and the mixture was stirred at room
temperature overnight. The reaction mixture was
decomposed with dilute hydrochloric acid and was
extracted with CH Cl , washed with brine, dried, and
®
nding, against nine tumor lines, that 14b had an IC50
less than 5 mM against two, less than 10 mM against one,
2
2

F. F. Becker et al. / Bioorg. Med. Chem. 8 (2000) 2693±2699
2697
evaporated. The crude product was crystallized from
ꢀ
e=4.35), lmax 358.28 (log e=4.016); IR (neat). 3262,
� 1
CH Cl ±hexanes to a€ord 9 (90% yield), mp 141 C,
2
2920, 1643, 1538, 1442, 1292, 1259, 1143, 803, 740 cm
1
.
2
�
1
1
IR(CH Cl ) 1667 cm ; H NMR (200 MHz) d 9.08±
H NMR (200 MHz) d 9.15 (1H, s, ArNHCO-), 8.85
(J=8.40 Hz, 1H, d, H ), 8.45 (J=8.70 Hz, 1H, d, H ),
2
2
8.76 (2H, m, Ar), 8.45 (1H, d, J=8.78 Hz, Ar), 8.10
7
6
(
J=7.83 Hz, Ar), 7.74±7.42 (4H, m, Ar), 4.16 (2H, s,
1H, s, H ), 8.02 (1H, d, J=7.82 Hz, Ar), 7.91 (2H, d,
8.30 (1H, s, H ), 8.12 (J=8.32Hz, 1H, d, H ), 8.05
12 10
(J=8.03 Hz, 1H, d, H ), 7.91 (J=8.00 Hz, 2H, d, Ar),
1
1
2
benzylic CH ), 2.80 (3H, s, COCH ).
2
7.74±7.34 (4H, m, Ar), 4.20 (2H, s, benzylic CH ), 3.84±
2
3
3
.66 (2H, brt, -CONCH ), 3.62±3.42 (2H, brt, -CON-
2
0
0
0
N-[2 -(13 H-Dibenzo[a,g]-¯uorenyl)]-4-(1 -piperidinyl)-
butane-1,4-dicarboxiamide (13a). To an ice cold solu-
tion of the acid 12a (1.065 g, 5.76 mmol) in dry CH Cl2
CH ), 3.05±2.75 (4H, m,-COCH CH CO-), 2.52±2.18
2 2 2
(7H, m, -(CH ) NCH , with a singlet at d 2.30 for -
2
2
3
13
NCH3); C NMR (300 MHz) d 171.692, 170.628,
142.137, 140.963, 139.871, 134.221, 131.718, 131.569,
130.334, 129.762, 128.563, 127.604, 126.633, 126.388,
126.192, 125.248, 125.117, 124.179, 124.055, 122.261,
121.314, 117.702, 54.8863, 54.6109, 45.9947, 45.2696,
41.9344, 36.5679, 32.8116, 29.5007; Mass (ES ): 464,
257, 191, 146, 145. Anal. calcd for C H N O : C, 77.70;
H, 6.30; N, 9.10%. Found: C, 77.46; H, 6.01; N, 8.46%.
2
(
50 mL) was added dry triethylamine (801 mL, 5.76 mmol)
followed by freshly distilled isobutyl chloroformate
664 mL, 5.12 mmol). Stirring was continued in cold
conditions for 5 min. Next this mixed anhydride was
added dropwise to an ice cold suspension of the 11-
amino 13H-dibenzo¯uorene (8) (719 mg, 2.56 mmol) in
dry CH Cl (50 mL), and stirring was continued over-
(
+
30
29
3
2
2
2
night, after which TLC in ethyl acetate:hexanes (20%)
showed the absence of the starting amine. The CH Cl
layer was washed successively with HCl (5%), brine,
0
0
0
0
N-[2 -(13 H-Dibenzo[a,g]-¯uorene-13 -one]-4-(1 -piperidi-
nyl)-butane-1,4-dicarboxiamide (14a). To a solution of
2
2
ꢀ
NaHCO (5%), and brine and then dried. Removal of
3
13a (665 mg, 1.48 mmol) in dry THF (40 mL) at � 78 C
the solvent under vacuo a€orded the crude product,
which was puri®ed by column chromatography over
silica gel and elution with ethyl acetate to furnish 13a
under argon was added a solution of n-BuLi (1.8 mL,
2.5 M, 4.44 mmol) in cyclohexane. The deep-yellow
colored solution was stirred for 1 h, 30 min at this tem-
ꢀ
(
800 mg, 70%); mp 186±188 C; UV (CH Cl ) l
perature, and then dry O was bubbled through the
2
2
2
max
2
59.41 (log e=4.67), lmax 358.33 (log e=4.43); IR
solution for 1 h. The temperature was allowed to rise
to room temperature while O2 continued to bubble
through the solution for an additional 3 h. The reaction
was quenched by the addition of water (20 mL) and
CH Cl (30 mL), and stirring was continued for 30 min.
(
neat) 3264, 2934, 1648, 1538, 1509, 1442, 1259, 804,
�
1
1
741 cm
ArNHCO-), 8.86 (J=8.40 Hz, 1H, d, H ), 8.47 (J=
;
H NMR (200 MHz) d 9.39 (1H, s,
7
8
1
9
.70 Hz, 1H, d, H ), 8.37 (1H, s, H ), 8.16 (J=8.31 Hz,
6 12
H, d, H ), 8.05 (J=8.03 Hz, 1H, d, H ), 7.91 (J=
1
.29 Hz, 2H, d, Ar), 7.66±7.45 (4H, m, Ar), 4.24 (2H, s,
2
2
The organic layer was collected, washed with brine, and
dried. On evaporation of the solvent under vacuo, the
crude product was obtained, which was crystallized
from CH Cl /hexanes to yield 14a (452 mg, 66%); mp
0
1
benzylic CH ), 3.78±3.58 (2H, brt, -CONCH ), 3.56±
2
2
3.38 (2H, brt,-CONCH ), 3.05±2.75 (4H, m, -COCH
CH CO-), 1.82±1.48 (6H, brs, -NCH (CH ) CH -);
2
2
2
2
13
ꢀ
C
NMR (400 MHz): 170.929, 169.439, 141.249, 139.962,
226±228 C; UV (CH Cl ), l
229.60 (log e=4.22),
2
2
2 3
2
2
2
max
lmax 285.05 (log e=4.28), lmax 331.39 (log e=4.05); IR
(neat) 3244, 2938, 1698, 1644, 1518, 1445, 1265, 815,
1
1
1
4
2
38.970, 133.079, 130.982, 130.609, 129.398, 128.808,
27.584, 126.607, 125.620, 125.392, 125.169, 124.231,
24.089, 123.156, 123.076, 121.385, 120.342, 116.504,
5.524, 42.215, 35.613, 32.088, 28.705, 25.325, 24.618,
�
1 1
801, 735, 703 cm ; H NMR (200 MHz) d 9.43 (1H, s,
ArNHCO-), 8.94 (J=8.44 Hz, 1H, d, Ar), 8.52±8.48
(1H, m, Ar), 8.24±8.00 (3H, m, Ar), 7.90 (J=8.45 Hz,
1H, d, Ar), 7.70 (J=8.16 Hz, 1H, d, Ar), 7.65±7.48 (3H,
m, Ar), 7.39 (J=8.06 Hz, 1H, d, Ar), 3.82±3.64 (2H, brt,
-CONCH ), 3.62±3.38 (2H, brt, -CONCH ), 2.87 (4H,
+
3.442; Mass (ES ): 449, 279, 205, 167, 149, 145. Anal.
calcd for C H N O : C, 80.30; H, 6.30; N, 6.20%.
3
Found: C, 80.36; H, 6.02; N, 6.27%.
0
28
2
2
2
2
s,-COCH CH CO-), 1.80±1.50 (6H, brs, -NCH2
2
2
0
0
0
13
N-[2 -(13 H-Dibenzo[a,g]-¯uorenyl)]-4-(4 N-methyl-piper-
azinyl)-butane-1,4-dicarboxiamide (13b). To an ice-cold
solution of the acid 12b (1.188 g, 5.94 mmol) in dry
CH Cl (100 mL) was added dry triethylamine (826 mL,
(CH ) CH -); C NMR (300 MHz, CDCl ) d 195.748,
2
3
2
3
171.875, 170.471, 147.007, 137.940, 135.158, 134.817,
133.708, 132.037, 130.713, 130.103, 129.348, 129.263,
128.138, 127.830, 127.487, 126.996, 126.236, 124.990,
124.336, 123.285, 120.887, 114.342, 46.562, 43.236,
2
2
5
.94 mmol) followed by freshly distilled isobutyl
+
chloroformate (686 mL, 5.29 mmol) and stirring was
continued in cold conditions for 5 min. Next, this mixed
anhydride was added dropwise to an ice cold suspension
of the 11-amino 13H-dibenzo¯uorene (8) (742 mg,
26.331, 25.625, 24.419, 22.662, 14.125; Mass (ES ): 463,
186, 152, 147, 145. Anal. calcd for C H N O : C,
30
26
2
3
77.90; H, 5.70; N, 6.10%. Found: C, 77.69; H, 5.64; N,
5.99%.
2
.64 mmol) in dry CH Cl (60 mL), and stirring was
2 2
0
0
0
0
continued overnight; after that, the TLC in ethyl acet-
ate:hexanes (20%) showed the absence of the starting
amine. The CH Cl layer was washed successively with
N-[2 -(13 H-Dibenzo[a,g]-¯uorene-13 -one]-4-(4 N-methyl-
piperazinyl)-butane-1,4-dicarboxiamide (14b). To a solu-
tion of 13b (500 mg, 1.08 mmol), in dry THF (50 mL) at
2
2
ꢀ
NaHCO (5%) and brine and then dried. Removal of
3
� 78 C under argon was added a solution of n-BuLi
the solvent under vacuo a€orded the crude product,
which was puri®ed by column chromatography over
silica gel and elution with MeOH to furnish 13b (800 mg,
(1.30 mL, 2.5 M, 3.24 mmol) in cyclohexane. The deep-
yellow colored solution was stirred for 1 h, 30 min at
this temperature, and then dry O was bubbled through
2
ꢀ
6
5%); mp 200±202 C; UV (CH Cl ) l
259.42 (log
max
the solution for 1 h. The temperature was allowed to rise
2
2

2
698
F. F. Becker et al. / Bioorg. Med. Chem. 8 (2000) 2693±2699
to room temperature while O2 continued to bubble
through the solution for an additional 3 h. The reaction
was quenched by the addition of water (20 mL) and
CH Cl (50 mL), and stirring was continued for 30 min.
addition of water (10 mL) and the reactants were
extracted with CH Cl (50 mL), washed with brine, and
2
2
dried. On removal of the solvent under vacuo, crude
product was crystallized from CH Cl /hexanes to yield
2
2
2
2
ꢀ
The organic layer was collected, washed with brine, and
dried. On evaporation of the solvent under vacuo, the
crude product was crystallized from CH Cl ±hexanes to
140 mg (70%) of pure 15b; mp 166±168 C; UV
(CH Cl ) l 227.17 (log e=4.46), lmax 261.71 (log
2
2
max
e=4.66), lmax 366.17 (log e=4.18); IR (neat) 3267,
2803, 1624, 1539, 1508, 1445, 1276, 1290, 1258, 1183,
2
2
ꢀ
furnish 14b (350mg, 68%); mp 204±206 C; UV (CH Cl ),
2
2
�
1 1
lmax 228.88 (log e=4.29), lmax 285.58 (log e=4.30), lmax
1146, 1001, 812, 748, 686 cm ; H NMR (200 MHz) d
9.06 (1H, s, ArNHCO-), 8.66 (J=8.48 Hz, 1H, d, Ar),
8.44 (J=8.26 Hz, 1H, d, Ar), 8.26 (J=8.14 Hz, 1H, d,
Ar), 8.14 (1H, s, H ), 8.04±7.70 (3H, m, Ar), 7.66±
3
1
7
8
8
7
7
30.50 (log e=4.06); IR (neat) 3231, 2920, 2852, 1699,
645, 1540, 1519, 1459, 1276, 1262, 1004, 815, 751,
�
1 1
01 cm ; H NMR (200MHz) d 9.17 (1H, s, ArNHCO-),
.93 (J=8.57 Hz, 1H, d, Ar), 8.48±8.35 (1H, m, Ar),
.30±7.96 (3H, m, Ar), 7.86 (J=8.54 Hz, 1H, d, Ar),
.69 (J=8.33 Hz, 1H, d, Ar), 7.60±7.42 (3H, m, Ar),
.38 (J=7.82 Hz, 1H, d, Ar), 3.86±3.40 (4H, m, CON
12
7.32 (4H, m, Ar), 5.82 (1H, s, H ), 3.70±3.12 (4H,
m, -CON(CH ) -), 2.84±2.50 (4H, m,-COCH CH CO-),
13
2
2
2
2
2.35±2.06 (7H, m, (-CH ) NCH , with a singlet at 2.09
2
2
+
3
for -NCH ); Mass (ES ): 480, 464, 147, 145. Anal.
3
(
CH ) ), 2.97 (4H, s, -COCH CH CO-), 2.62±2.16 (7H,
2
calcd for C H N O : C, 75.10; H, 6.10; N, 8.80%.
3
2
2
2
30 29
3
m, (-CH ) NCH with a singlet at d 2.31 for-NCH );
Found: C, 75.59; H, 6.19; N, 8.08%.
2
2
3
3
1
3
C NMR (300 MHz) d 195.507, 171.684, 170.674,
46.779, 138.031, 135.070, 134.472, 133.633, 131.810,
30.830, 129.982, 129.267, 129.123, 128.098, 127.749,
27.446, 126.858, 126.163, 124.891, 124.270, 123.225,
20.801, 114.643, 54.920, 54.643, 46.003, 45.332, 31.590,
0
0
0
1
1
1
1
2
N-[2 -(13 H-Dibenzo[a,g]-¯uorenyl]-4-(1 -piperidinyl)-
butane-1,4-diamine (16a). To a solution of 13a (50 mg,
0.12 mmol) in THF (50 mL) were added borane±methyl
sulphide complex (67 mL, 5 M) solution in diethyl ether
+
ꢀ
2.657, 14.123; Mass (ES ): 478, 464, 147, 145. Anal.
under argon at 0 C, and the mixture was re¯uxed for
calcd for C H N O : C, 75.50; H, 5.70; N, 8.80%.
3
Found: C, 75.19; H, 5.84; N, 7.99%.
16 h. Then, HCl (15 mL, 5%) was added, and the mix-
ture was re¯uxed for another 10 h. The solution was
cooled, added to 1 M sodium hydroxide solution,
extracted with ethyl acetate, washed with brine, and
dried. On removal of the solvent, the crude diamine
solidi®ed, which was crystallized from ethyl acetate:
3
0
27
3
0
0
0
0
N-[2 -(13 H-Dibenzo[a,g]-¯uorene-13 -hydroxy]-4-(1 -
piperidinyl)-butane-1,4-dicarboxiamide (15a). To a solu-
tion of 14a (200 mg, 0.43 mmol) in absolute EtOH
ꢀ
(
20 mL) was added NaBH (47 mg, 1.29 mmol) at 0±5 C
4
hexanes (20:80) to give the desired diamine 16a (35 mg,
ꢀ
and the solution was stirred for 30 min. The stirring was
continued for an additional 4 h at room temperature.
The reaction was quenched by the careful addition of
75%); mp 166±168 C; UV (CH Cl ), l
229.42 (log
2
2
max
e=4.50), lmax 268.38 (log e=4.55), lmax 377.24 (log
e=4.25); IR (neat) 3383, 2932, 2767, 1619, 1587, 1570,
1532, 1473, 1443, 1345, 1289, 1182, 1122, 1040, 804, 766
water (10 mL). The mixture was extracted with CH Cl2
2
�
1 1
(
2Â20 mL), and the organic layer was washed with brine
and 683 cm ; H NMR (200 MHz) d 8.81 (J=8.11 Hz,
1H, d), 8.44 (J=8.70 Hz, 1H, d), 8.14±7.90 (4H, m, Ar),
7.76±7.32 (4H, m, Ar), 6.97 (1H, s, H ), 4.24 (2H, s,
and dried. On removal of the solvent under vacuo, the
crude product was crystallized from CH Cl ±hexanes to
yield 15a (150 mg, 75%); mp 220±222 C; UV (CH Cl ),
2
2
2
12
ꢀ
benzylic CH ), 3.40 (J=6.56 Hz, 2H, t, ArNHCH -),
2
2
2
lmax 228.00 (log e=4.66), lmax 261.79 (log e=4.89),
lmax 366.69 (log e=4.42); IR (neat) 3263, 2935, 2855,
2.60±2.24 [6H, brt, H C-N(CH ) -], 1.98±1.42 (10H, m,
2 2 2
methylenes); 13C NMR (300 MHz) d 144.274, 143.297,
141.144, 138.722, 130.925, 130.487, 130.205, 128.641,
127.459, 126.725, 126.276, 126.195, 124.353, 124.292,
123.758, 123.712, 123.049, 121.090, 120.871, 102.292,
58.903, 54.602, 44.259, 36.837, 27.324, 25.807, 24.818,
1
647, 1624, 1538, 1443, 1276, 1262, 1183, 1088, 808,
�
1
1
748, 684 cm
; H NMR (200 MHz) d 9.26 (1H, s,
ArNHCO-), 8.72 (J=8.62 Hz, 1H, d, Ar), 8.41
J=8.07 Hz, 1H, d, Ar), 8.20±8.12 (2H, m, Ar with a
singlet at 8.28 for H ), 8.00±7.65 (3H, m, Ar), 7.60±7.28
(
+
24.388; Mass (ES ): 421, 281, 252, 232, 211, 108.
1
2
(
4H, m, Ar), 5.83 (J=10.28 Hz, 1H, d, H ), 3.75±3.55
13
0
0
0
(
2H, brt, -CONCH ), 3.50±3.35 (2H, brt,-CONCH ),
2
N-[2 -(13 H-Dibenzo[a,g]-¯uorenyl)]-4-(4 N-methyl-piper-
azinyl)-butane-1,4-diamine (16b). To a solution of 13b
(100 mg, 0.22 mmol) in THF (60 mL) was added bor-
ane±methyl sulphide complex (130 mL), 5 M solution in
2
3
.00±2.62 [5H, m, -COCH CH CO- with a doublet at
2 2
2.70 (J=10.28 Hz, C -OH, which disappeared with
D O)], 1.78±1.44 [6H, brs, -NCH (CH ) CH -]; Mass
13
2
2
2 3
2
+
ꢀ
(
ES ): 465, 447, 168. Anal. calcd for C H N O : C,
3
diethyl ether under argon at 0 C, and the mixture was
3
0
28
2
7
5
7.90; H, 5.70; N, 6.10%. Found: C, 77.92; H, 5.92; N,
.63%.
re¯uxed for 16 h. Then, distilled water (15 mL) was
added and the re¯uxing was continued for an additional
1
0 h. The solution was cooled, added to 1 M sodium
0
0
0
0
N-[2 -(13 H-Dibenzo[a,g]-¯uorene-13 -hydroxy]-4-(4 N-
methyl-piperazinyl)-butane-1,4-dicarboxiamide (15b). To
a solution of 14b (200 mg, 0.42 mmol) in absolute EtOH
hydroxide solution, extracted with ethyl acetate, washed
with brine, and dried. On removal of the solvent, the
crude diamine solidi®ed, which was crystallized from
ethyl acetate:hexanes (20:80) to give the desired diamine
(
20 mL) was added NaBH (46 mg, 1.26 mmol) at 0±
4
C, and the solution was stirred under this condition
ꢀ
ꢀ
5
16b (62 mg, 66%); mp 140±142 C; UV (CH Cl ), l
2
2
max
for 30 min. The temperature was allowed to rise to room
temperature, and stirring was continued for an addi-
tional 4 h. The reaction was quenched by the careful
227.48 (log e=4.43), lmax 268.32 (log e=4.45), lmax
377.26 (log e=4.14); IR (neat) 376, 2936, 2802, 1619,
1587, 1570, 1514, 1415, 1344, 1241, 1165, 1012, 806, 741,

F. F. Becker et al. / Bioorg. Med. Chem. 8 (2000) 2693±2699
2699
�
1 1
and 684 cm ; H NMR (200 MHz) d 8.80 (J=8.35 Hz,
1
5. Becker, F. F.; Banik, B. K. Bioorg. Med. Chem. Lett. 1998,
8, 2877. For an example, see: Jorgensen, K.; Ipsen, J. H. Bio-
chem. Biophys. Acta 1991, 1062, 227.
H, d), 8.42 (J=8.70 Hz, 1H, d), 8.14±7.88 (4H, m, Ar),
7.72±7.30 (4H, m, Ar), 6.95 (1H, s, H ), 4.22 (2H, s,
benzylic CH ), 3.38 (J=6.62Hz, 2H, t, ArNHCH -),
12
6
. For some recent examples, see: (a) Cherubim, P.; Deady, L.
2
2
W.; Dorkos, M.; Quazi, N. H.; Baguley, B. C.; Denny, W. A.
Anti-cancer Drug Design 1993, 8, 429. (b) Palmer, B. D.; Lee,
H. H.; Baguley, B. C.; Denny, W. A. J. Med. Chem. 1992, 35,
2
1
1
1
1
1
4
.68±2.30 [10H, brt, (5X-NCH )], 2.30 (3H, s, NCH ),
2 3
1
3
.98±1.56 (4H, m, methylenes); C NMR (300 MHz) d
44.251, 143.231, 141.112, 138.728, 130.938, 130.482,
30.202, 128.645, 127.474, 126.789, 126.285, 126.209,
24.345, 124.322, 123.784, 123.705, 123.035, 121.045,
20.865, 102.332, 58.0515, 54.9487, 53.0379, 45.9162,
2
7
58.
. Banik, B. K.; Venkatraman, M. S.; Mukhopadhyay, C.;
Becker, F. F. Tetrahedron Lett. 1998, 39, 7243. Also see: (a)
Banik, B. K.; Ghatak, A.; Venkatraman, M. S.; Becker, F. F.
Synth. Commun. 2000, 30, 2701. (b) Banik, B. K.; Ghatak, A.;
Mukhopadhyay, C.; Becker, F. F. J. Chem. Res. (S) 2000,
4.2362, 40.9995, 36.8472, 29.7030, 27.2063, 24.8132;
+
Mass (ES ): 436, 281, 260, 239, 219.
1
8
08.
. Banik, B. K.; Mukhopadhyay, C; Venkatraman, M. S.;
Acknowledgements
Becker, F. F. Tetrahedron Lett. 1998, 39, 7247. Also see: (a)
Banik, B. K.; Zegrocka, O.; Banik, I.; Hackfeld, L.; Becker, F.
F. Tetrahedron Lett. 1999, 40, 6731. (b) Ghatak, A.; Becker, F.
F.; Banik, B. K. Tetrahedron Lett. 2000, 41, 3793. (c) Basu, M.
K.; Becker, F. F.; Banik, B. K. Tetrahedron Lett. 2000, 41,
We gratefully acknowledge the funding support received
for this research project from the Hubert L. and Olive
Stringer Chair for the partial support of this research.
We are thankful to NIH Cancer Center Support Grant,
5
603. (d) Banik, B. K.; Suhendra, M.; Banik, I.; Becker, F. F.
Synth. Commun. 2000, in press.
. Harvey, R. G.; Pataki, J.; Cortez, C.; DiRaddo, P.; Yang,
C. J. Org. Chem. 1991, 56, 1210.
5
-P30-CA16672-25, in particular the shared resources of
9
the Pharmacology and Analytical Center Facility. The
antitumor activity was assayed by Dr. Robert Newman
in the Pharmacology and Analytical Core Laboratory
1
6
0. (a) Ranu, B. C.; Ghosh, K.; Jana, U. J. Org. Chem. 1996,
1, 9546. (b) Heaney, H. In Comprehensive Organic Synthesis;
(
2P30CA16672-23).
Trost, B. M., Ed.; Pergamon: Oxford, 1991; Vol. 2, p 753. (c)
Olah, G. A. Friedel-Crafts and Related Reactions; Interscience:
New York, 1964.
References and Notes
. For example, see: (a) Harvey, R. G. Polycyclic Aromatic
Hydrocarbons; Wiley-VCH, 1997. (b) Clar, E. Polycyclic
Hydrocarbons; Academic: New York, 1964.
11. (a) Abu-Shqara, E.; Yang, C. X.; Harvey, R. G. J. Org.
Chem. 1992, 57, 3312. (b) Minabe, M.; Cho, B. P.; Harvey, R.
V. J. Am. Chem. Soc. 1989, 111, 3809. (c) Yoshida, M.; Hish-
ida, K.; Minabe, M.; Suzuki, K. J. Org. Chem. 1980, 45, 1783.
(d) Yoshida, M.; Minabe, M.; Suzuki, K. J. Org. Chem. 1979,
44, 3029. (e) Yoshida, M.; Nagayama, S.; Minabe, M.; Suzuki,
K. J. Org. Chem. 1979, 44, 1915.
1
2
2
. (a) Zhang, F.-J.; Cortez, C.; Harvey, R. G. J. Org. Chem.
000, 65, 3952. (b) Harvey, R. G. Polycyclic Aromatic Hydro-
carbons: Chemistry and Carcinogenicity; Cambridge University
Press, 1991; Chapter 4. (c) Di Raddo, P.; Chan, T. J. Org.
Chem. 1982, 47, 1427. (d) Lehr, R. E.; Yagi, H.; Thakker, D.
R.; Levin, W.; Conney, A. H.; Jerina, D. M. Carcinog.-
Comprv. 1978, 3, 231. (e) Thakker, D. R.; Yagi, H.; Lu, A. Y.
H.; Levin, W.; Conney, A. H.; Jerina, D. M. Proc. Natl. Acad.
Sci. USA 1976, 73, 3381.
12. Jones, D. W.; Matthews, R. S.; Bartle, K. D. E. Spectro-
chim Acta, Part A 1972, 28, 2053.
13. Dewar, M. J. S.; Dennington, II, R. D. J. Am. Chem. Soc.
1989, 111, 3804. Chem 3-D Pro version 4.0 was used to obtain
the starting geometries by doing MM2 calculations. DEWAR-
PI was used to calculate the energies of the Wheland Inter-
mediates in kcal/mol for substitution at each position.
DEWAR-PI was available from the Quantum Chemistry Pro-
gram Exchange (QCPE). Dewar, M. J. S.; Ruiz, J. M. QCPE
Bull. 1988, 8, No. 1, 50.
14. Holzapfel, C. W.; Pettit, G. R. J. Org. Chem. 1985, 50,
2323.
15. Harvey, R. G.; Abu-Shqara, E.; Yang, C. X. J. Org.
Chem. 1992, 57, 6313.
3. Smalley, R. V.; Goldstein, D.; Bullowski, D.; Hannon, C.;
Buchler, D.; Knudsen, C.; Tuttle, R. L. Invesi. New Drugs
1
4
992, 10, 107.
. For some recent references, see: (a) Sami, S. M.; Dorr, R.
T.; Alberts, D. S.; Remers, W. A. J. Med. Chem. 1993, 36, 765.
b) Motohashi, N.; Gollapudi, S. R.; Emrani, J.; Bhattiprolu,
(
K. R. Cancer Invest. 1991, 9, 305. (c) Denny, W. A.; Rew-
castle, G. W.; Baguley, B. C. J. Med. Chem. 1990, 33, 814. (d)
Palmer, B. D.; Rewcastle, G. W.; Atwell, G. J.; Baguley, B. C.;
Denny, W. A. J. Med. Chem. 1988, 31, 707.
16. The reactions were carried out under a well-ventilated
hood. Some of the compounds described here are potential
carcinogens.