Induction of apoptosis by aryl-substituted diamines: Role of aromatic group substituents and distance between nitrogens

Article abstract of DOI:10.1016/S0960-894X(02)00156-7

A series of aromatic substituted diamines was synthesized and characterized for their cytotoxic profiles against human breast and prostate tumor cell lines. Following a structure function analysis of the effects of changes of the benzyl substituents and t

Full text of DOI:10.1016/S0960-894X(02)00156-7

Bioorganic & Medicinal Chemistry Letters 12 (2002) 1263–1267
Induction of Apoptosis by Aryl-Substituted Diamines: Role of
Aromatic Group Substituents and Distance Between Nitrogens
Mark R. Burns,* Solveig LaTurner, Josh Ziemer, Maralee McVean, Bruce Devens,
C. Lance Carlson, Gerard F. Graminski, Scott M. Vanderwerf, Reitha S. Weeks and
Jay Carreon
MediQuest Therapeutics, Inc., 4010 Stone Way North, Seattle, WA 98103, USA
Received 18 December 2001; accepted 26 February 2002
Abstract—A series of aromatic substituted diamines was synthesized and characterized for their cytotoxic profiles against human
breast and prostate tumor cell lines. Following a structure function analysis of the effects of changes of the benzyl substituents and
the distance between amino groups the most potent analogues were analyzed biologically and were shown to induce apoptosis.
These compounds do not induce the enzyme SSAT or deplete intracellular polyamine levels, mechanisms demonstrated by other
cytotoxic polyamine analogues. # 2002 Elsevier Science Ltd. All rights reserved.
Molecules that bind to nucleic acids have a wide variety
of potent and unique biological activities.1,2 Inhibition
of transcription factor binding to the TATA box of a
DNA sequence was shown by the zinc(II) complexes of
a group of benzylated macrocyclic tetraamines includ-
ing 1.3 Subsequent work by the same group has shown
potent inhibition of HIV-1 TAR RNA–Tat peptide
interactions by related compounds.4 Recent reports
have described the cytotoxic activities of a group of
compounds related structurally by the presence of two
lipophilic aromatic substituents separated by a variable
length, flexible linker containing two cationic ammo-
nium centers. Examples of the aromatic substituents
include the acridines,5,6 bis(9-methyl-phenazines),7
anthracyclines,8 indeno[1,2-b]quinolines9 and bis (naph-
thalimides).10,11 Two of these molecules, DMP 840 212
and LU 79553 311 have been reported to be undergoing
clinical studies. Biochemical studies of the mechanism of
action of 2 has implicated the involvement of major groove
binding together with topoisomerase II poisoning.13
Additionally, the antitumor activity of several un-sub-
stituted benzyl diamines such as dibenzylputrescine 4
has been previously reported (Fig. 1).14,15 Due to our
interest in the discovery of polyamine analogues with
unique mechanisms of antitumor activities, we have
synthesized a series diaminoalkanes substituted with
various aromatic groups. This group of compounds was
evaluated by their cytotoxic activities against human
breast (MDA-MB-231) and prostate (PC-3) tumor cell
lines. The more active analogues were then evaluated in
detailed biological assays including the role played by
*Corresponding author. Tel.: +1-206-675-8880; fax: +1-206-675-
8881; e-mail: markburns@mqti.com
Figure 1. Aromatic-substituted diamines.
0960-894X/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(02)00156-7

1264
M. R. Burns et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1263–1267
polyamine metabolism in their cytotoxic mechanism.
The induction of apoptosis was demonstrated for sev-
eral of the more potent analogues by staining with
Annexin V/propidium iodide followed by flow cyto-
metry analysis.
included 1,3-diaminopropane (for 34 and 40), 1,5-dia-
minopentane (for 35 and 41), 2,5-diaminohexane17 (for
37), m-xylylenediamine (for 38) and p-xylylenediamine
(for 39).
The monobenzylated analogues 5, 9, and 24 were pro-
duced by similar methods via use of N1-Boc-1,4-diamino-
butane18 as the starting amine. The anilino derivatives
The synthesis of symmetrically disubstituted derivatives
utilized the straightforward one-pot route shown in
Scheme 1. Following formation of the diimine inter-
mediate by stirring the diamine with excess aromatic
aldehyde, removal of drying agent and solvent gave a
product that was reduced by treatment with excess
NaBH4 in MeOH. Standard workup and purification
over silica gel gave the analogues in pure form.16 Other
diamines that were used in place of 1,4-diaminobutane
Table 2. Cytotoxicity assay results for polycyclic aromatic com-
pounds 3, 4 and 23–33
Analogue
Ar-
MDA-MB-231
IC50 (mM)a
PC-3
IC50 (mM)a
3
4
LU 79553
H-
0.06
192
0.07
114
23
9.1Æ2.4 (5)
10Æ3.0 (4)
24b
25
Mono-(1-naphthyl)-
>300
211
2.5
2.1
26
27
28
9.4
77
8.4
51
Scheme 1. Synthesis of arylated diamines. Reagents and conditions:
(a) ArCHO (2.2 equiv), Et3N (2.1 equiv), MgSO4, CH2Cl2; (b) NaBH4,
MeOH.
Table 1. Cytotoxicity assay results for substituted aromatic com-
pounds (3, 4, and 5–22)
2.8Æ1.0 (4)
3.2Æ1.5 (4)
29
30
31
32
1.7 (2)
1.92Æ0.09 (3)
6.6Æ4.9 (3)
3.5 (2)
1.6 (2)
0.77Æ0.07 (3)
1.85Æ0.11 (3)
2.7 (2)
Analogue
R Group
structure
MDA-MB-231
IC50 (mM)a
PC-3
IC50 (mM)a
3
4
5b
6
7
8
9c
10
11
12
13
14
15
16
17
18
19
20
21
22
LU 79553
H-
(Mono)-H-
2-NO2-
3-NO2-
0.06
192
>300
62
0.07
114
211
23
9
143
50
6
56
58
10
22
15
98
>300
56
20
>300
202
580
18
4-NO2-
(Mono)-4-NO2-
4-CH3O-
3,4-Methylenedioxy-
3,4,5-Trimethoxy-
3-Cyano-
205
201
21
70
58
42
84
19
146
>300
197
65
>300
>300
577
4-Cyano-
3-Azido-
2-NH2-
3-NH2-
4-N(CH3)2-
3,4-Dichloro-
2-PyridinylCH2-
3-PyridinylCH2-
4-PyridinylCH2-
33
0.62Æ0.17 (4)
0.60Æ0.18 (3)
aValues are from single experiments unless otherwise noted by the
number of repeats in parentheses. If n=2 average is given, if n>2 then
ÆSD is given.
aValues are from single experiments unless otherwise noted.
bMono-aryl 5 had the structure PhCH2NHCH2CH2CH2CH2NH2.
cMono-aryl
CH2NH2.
9 had the structure 4-NO2–ArCH2NHCH2CH2CH2
bMono-aryl 24 had the structure 1-napthylCH2NHCH2CH2-
CH2CH2NH2.

M. R. Burns et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1263–1267
1265
16 and 17 were produced by selective Zn/HCl reduction
of the nitrobenzyl analogues 6 and 7. The 3-azidobenzyl
analogue 15 was produced from 17 using standard con-
ditions (NaNO2, 6 N HCl, NaOAc, NaN3).19 Analogues
As shown in Table 1, the substituted benzylated di-
amines were evaluated by their ability to inhibit the
growth of MDA-MB-231 (breast) and PC-3 (prostate)
human cancer cell lines.21 In comparison to the un-sub-
stituted benzyl analogue 4, electron-withdrawing groups
in the meta-position such as the 3-NO2- analogue 7 and
the 3-cyano derivative 13 had 10- and 4-fold higher
activities, respectively. para-Substitution with the same
groups gave much lower activities (compare analogues 8
and 14). On the other hand, para-substitution with the
electron-donating group CH3O- gave an analogue 10
possessing 9-fold higher activity. Evaluation of the
polycyclic aromatic substituted diamines shown in
Table 2 showed that even greater increases in activities
were possible.
1
were analyzed by TLC and H NMR. Selected anal-
ogues were further analyzed by 13C NMR, HPLC and
HRMS methods. All results were consistent with the
proposed structures. Several compounds have been
previously reported.20aÀe
Table 3. Cytotoxicity assay results for diamine variations of analogue
23
Analogue
Diamine
Ar=
MDA-MB-231
IC50 (mM)a
PC-3
IC50 (mM)a
The 1- and 2-substituted naphthalene derivatives 23 and
25 showed dramatic 21- and 90-fold increases in
potency when compared to 4. It is interesting to note
that the mono-substituted 1-naphthyl analogue 24
showed a precipitous loss of activity. The quinoline
analogue 26 gave an activity equivalent to 23 that, when
compared to the 4-pyridinyl analogue 22, again shows
the significance of the polycyclic structure. Various
34
23
35
36b
37
38
39
5.1
4.3
Table 4. Cytotoxicity assay results for diamine variations of
analogue 30
9.1Æ2.4 (5)
10Æ3.0 (4)
Analogue
Diamine structure
Ar Structure=
MDA-MB-231
IC50 (mM)a
PC-3
IC50 (mM)a
8.3
6.5
25
5.0
37
7.0
40
30
41
2.0
0.80
19
20
1.92Æ0.09 (3) 0.77Æ0.07 (3)
7.0Æ1.0 (3)
14Æ7.0 (3)
1.3
0.48
aValues are from single experiments unless otherwise noted by the
number of repeats in parentheses. If n=2 average is given, if n>2 then
ÆSD is given.
bStructure for 36 is (1-naphthylCH2CH2NHCH2CH2)2.
aValues are from single experiments except for 30 whereÆSD is given.
Table 5. Cytotoxic activities of 23, 30, and 33 against multiple cell types
Cell line
Tumor type
IC50 values (mM)a
2
3
30
MDA-231
PC-3
SK-OV-3
A375
SK-Mel-5
T47.D
Mes-SA
Mes-SA/Dx5
Breast
Prostate
Ovarian
Melanoma
Melanoma
Breast
9.1Æ2.4 (5)
10Æ3 (4)
8.5 (2)
5.4Æ0.54 (4)
4.6
6.5
10.1
7.2
1.92Æ0.09 (3)
0.62Æ0.17 (4)
0.60Æ0.18 (3)
0.68 (2)
0.40
0.77Æ0.07 (3)
0.74 (2)
0.59Æ0.08 (3)
—
—
—
—
0.94
0.65
0.64
0.61
Uterine
MDR Uterine
aValues are from single experiments unless otherwise noted by the number of repeats in parentheses. If n=2 average is given, if n>2 then
ÆSD is given.

1266
M. R. Burns et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1263–1267
other readily available substitutions on the aromatic
aldehyde starting material including 4-methoxy-
naphthylene 28, 4-benzyloxyphenyl 29, 4-biphenyl 30,
2-fluorenyl 31, 9-anthracene 32, and the very potent
4-diphenylamine 33 together supported the favorable
activity gained through the presence of a polycyclic
aromatic ring system.
Following the demonstration of the improvement in
activity shown by the polycyclic aromatics in comparison
to the benzyl analogues, the effect of changes in the
diamine structure was explored. As shown by analogues
34, 23, and 35 in Table 3, variation in the length of the
aliphatic chain linker had minor effect on the activities.
Compound 36, with an extra methylene between the
naphthalene group and the amine had a reduced activity
in comparison with 23. Addition of methyl groups on
the a-carbon of the diamine gave an analogue 37 with
slightly better activity when compared to 23. Substitu-
tion of either a 3- or 4-xylylenediamine (38 and 39)
resulted in analogues with activities similar to 23. These
data show that the structure of the linker between the
nitrogen atoms has a minor role in activity.
Table 4 shows the lack of effect from alteration in the
distance between the amino groups of analogue 30 had
on its cytotoxic activity. Both the three and five
methylene group containing analogues 40 and 41 gave
similar results to those shown by 30.
We next explored the effects of several of the most
active analogues on a variety of human cancer cell lines.
These results are shown in Table 5. Noteworthy data
include the retention of the effects of analogues 23 and
33 on both the wild-type (Mes-SA) and multidrug
resistant (Mes-SA/Dx5) uterine carcinoma cell types.
Results from an Annexin V/propidium iodide fluor-
escent cell-staining assay22 showed a significant popu-
lation of 30 and 33 treated A375 melanoma cells were
undergoing apoptosis following a 3-day treatment with
drug. Lesser amounts of 23 treated cells were induced to
undergo an apoptotic cell death cascade. As shown by
two representative examples in Figure 2, treatment with
30 or 33 at 3 and 5 mM concentrations, respectively,
resulted in the appearance of 61 and 77% of cells with
apoptotic characteristics.
We have explored the mechanism of action of the most
active analogues based on their structural similarity to
putrescine. We have tested the ability of 23 and 33 to
influence the levels of the natural polyamines in treated
MDA-MB-231 cells and saw no changes in the levels of
putrescine, spermidine, or spermine (data not shown).
Compound 30 induced a slight 9Æ5-fold increase in the
level of SSAT activity in SK-MEL-28 cells.23 A Co-
incubation with a large excess of putrescine (50-fold
higher than drug concentration) had no effect on the
cytotoxicity of 23 or 33 on MDA-MB-231 cells. Co-
incubation with the polyamine biosynthesis inhibitor a-
difluoromethylornithine (DFMO) had no effect on the
IC50 values of these two compounds. Based on these
observations we have concluded that these compounds
act via a mechanism(s) distinct from those determined
for other cytotoxic polyamine analogues.24
Figure 2. Annexin/propidium iodide assays were performed using
A375 melanoma cells following 3 days of drug treatment. Apoptotic
cell death was assessed by annexin V-fluorescein isothiocyanate bind-
ing and propidium iodide co-staining according the manufacturer’s
protocol (R&D Systems, Minneapolis, MN, USA). Analysis by flow
cytometry used a Becton Dickinson FACScan analyzer.
An alternative hypothesis is that these analogues may
induce their biological effects through interactions with
DNA. The presence of two ammonium centers in these
molecules may enable their targeting to the anionic
phosphodiester backbone of nucleic acids, followed by

M. R. Burns et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1263–1267
1267
the aromatic portion of the molecules increasing their
affinity through either a hydrophobic or intercalative
mechanism. Initial indications of the drug’s ability to
interact with nucleic acids were assessed by a DNA/
ethidium bromide dye displacement assay.25 Results
supported the moderate ability of these molecules to
interact with DNA. A DC50 of 8.7, 8.7 and 9.4 mM for
polydAdT were found for 23, 30, and 33, respectively.
Distamycin A gave a DC50 value of 44 mM using the
same conditions.
stirred at 25 ꢀC for 2 h. Filtration and evaporation gave a yel-
low oil as the crude diimine. This was suspended in 300 mL of
CH3OH and 5.46 g (144 mmol, 5.3 equiv) of solid NaBH4 was
carefully added portion-wise. The resulting solution was stir-
red at 25 ꢀC for 20 h when the reaction was quenched by the
careful addition of 6 N HCl until strongly acidic (ꢁ35 mL).
The resulting mixture was diluted by the addition of 300 mL of
H2O and 300 mL CH2Cl2. The aqueous layer was made basic
by the addition of excess 4 N NaOH. Two additional washes
of the basic aqueous layer with equal portions of CH2Cl2 were
performed. The combined organic layers were washed with
brine, dried and evaporated to give 10.38 g yellow oil as the
crude product. Column chromatography using CHCl3/iPrOH/
concd NH4OH 88:10:2 gave 8.42 g (84%) white foam. The
analytical sample was crystallized from EtOH. 1H NMR
CHCl3, ppm from internal Si(CH3)4 d 8.22 (d, 2H), 7.96 (d,
2H), 7.84 (d, 2H), 7.55 (m, 8H), 4.32 (s, 4H), 2.84 (m, 4H),
1.71 (m, 4H), 1.42 (br s, 2H); 13C NMR (CHCl3, ppm from
residual solvent signal): d 135.9, 133.6, 131.6, 128.5, 127.5,
125.9, 125.7, 125.4, 125.2, 123.4, 51.6, 49.8, 27.9. HRMS by
MALDI-FTMS: calcd for C26H28N2Na+ 391.2150. Obsd:
391.2155. Samples were analyzed for purity by HPLC over a
C-18 column using a 40 min gradient of 5–100% CH3CN in
H2O both containing 0.1% TFA.
In conclusion, we have obtained nearly a 300-fold
improvement in the cytotoxic properties of a series of
compounds through optimization of the aromatic con-
stituent attached to a central diamine core of the mol-
ecule. Large polycyclic aromatic substitution led to the
working hypothesis that these molecules may be inter-
acting with nucleic acids via a combination of ionic and
either hydrophobic or intercalating interactions. The
observed ability of several of these compounds to
induce apoptosis provided an alternative suggestion that
interaction with apoptosis pathway components may be
playing a role in their cytotoxic activities.26
17. Moyano, N.; Frydman, J.; Buldain, G.; Ruiz, O.; Fryd-
man, R. B. J. Med. Chem. 1990, 33, 1969.
18. Krapcho, A. P.; Kuell, C. S. Synth. Commun. 1990, 20, 2559.
19. Cledera, P.; Avendano, C.; Menendez, J. C. Tetrahedron
1998, 54, 12349.
References and Notes
20. (a) Analogue 3 has been described in ref 11. (b) Analogue
4 has been described in ref 14. (c) Analogue 22 was described
by: Bergeron, R. J.; Wiegand, J.; Weimar, W. R.; Snyder, P. S.
Pharmacol. Res. 1998, 38, 367. (d) Analogue 20 was described
by: Yan, Q.; Anderegg, G. Inorg. Chim. Acta 1985, 105, 121.
(e) Analogues 21 and 25 were reported by: Colautti, A.;
Maurich, V. Boll. Chim. Farm. 1972, 111, 593.
1. Nadassy, K.; Wodak, S. J.; Janin, J. Biochemistry 1999, 38,
1999.
2. Dervan, P. B. Bioorg. Med. Chem. 2001, 9, 2215.
3. Kikuta, E.; Koike, T.; Kimura, E. J. Inorg. Biochem. 2000,
79, 253.
4. Kikuta, E.; Aoki, S.; Kimura, E. J. Am. Chem. Soc. 2001,
123, 7911.
21. Cells were plated in 96-well plates in the respective media
(MDA-MB-231 and A375 used DMEM while PC-3 used
F12K. SK-OV-3, Mes-SA and Mes-SA/Dx5 cells were cul-
tured in McCoy’s 5A media. SK-Mel-5 cells used EMEM
while T47.D cells used RPMI 1640 media. All media con-
tained 1 mM aminoguanidine, 50 U/mL penicillin, 50 mg/mL
streptomycin and 10% fetal bovine serum. A375 and T47.D
used 1 mM sodium pyruvate while A375, T47.D and MDA-
MB-231 had 2 mM l-glutamine. Additionally, T47.D cells had
0.4% insulin). After 24 h, drug was added and cells were
allowed to grow for an additional 72 h. Cell number was
determined by MTS assay used as described by manufacturer
(Promega).
5. Canellakis, E. S.; Shaw, Y. H.; Hanners, W. E.; Schwartz,
R. A. Biochim. Biophys. Acta 1976, 418, 277.
6. Canellakis, E. S.; Bellantone, R. A. Biochim. Biophys. Acta
1976, 418, 290.
7. Gamage, S. A.; Spicer, J. A.; Finlay, G. J.; Stewart, A. J.;
Charlton, P.; Baguley, B. C.; Denny, W. A. J. Med. Chem.
2001, 44, 1407.
8. Chaires, J. B.; Leng, F. F.; Przewloka, T.; Fokt, I.; Ling,
Y. H.; Perez-Soler, R.; Priebe, W. J. Med. Chem. 1997, 40, 261.
9. Deady, L. W.; Desneves, J.; Kaye, A. J.; Finlay, G. J.;
Baguley, B. C.; Denny, W. A. Bioorg. Med. Chem. 2000, 8, 977.
10. Kirshenbaum, M. R.; Chen, S. F.; Behrens, C. H.; Papp,
L. M.; Stafford, M. M.; Sun, J. H.; Behrens, D. L.; Fredericks,
J. R.; Polkus, S. T.; Sipple, P. Cancer Res. 1994, 54, 2199.
11. Brana, M. F.; Castellano, J. M.; Moran, M.; Perez de
Vega, M. J.; Perron, D.; Conlon, D.; Bousquet, P. F.;
Romerdahl, C. A.; Robinson, S. P. Anticancer Drug. Des.
1996, 11, 297.
12. O’Reilly, S.; Baker, S. D.; Sartorius, S.; Rowinsky, E. K.;
Finizio, M.; Lubiniecki, G. M.; Grochow, L. B.; Gray, J. E.;
Pieniaszek, H. J.; Donehower, R. C. Ann. Oncol. 1998, 9, 101.
13. Nitiss, J. L.; Zhou, J.; Rose, A.; Hsiung, Y.; Gale, K. C.;
Osheroff, N. Biochemistry 1998, 37, 3078.
14. Aizencang, G.; Harari, P.; Buldain, G.; Guerra, L.; Pickart,
M.; Hernandez, P.; Frydman, B. Cell. Mol. Biol. 1998, 44, 615.
15. Badolo, L.; Gelbcke, M.; Dubois, J.; Hanocq, M. Phar-
mazie 1998, 53, 15.
16. The synthesis of 23 is given as a representative example.
All reagents were used as received from Aldrich in their purest
and driest forms available. The mixture produced by addition
of 2.38 g (27 mmol) of 1,4-diaminobutane, 9.37 g (60 mmol, 2.2
equiv) of 1-naphthaldehyde, 8.0 mL (57.4 mmol, 2.1 equiv) of
Et3N and 6.0 g of anhyd MgSO4 in 100 mL of CH2Cl2 was
22. Moore, A.; Donahue, C. J.; Bauer, K. D.; Mather, J. P.
Methods Cell Biol. 1998, 57, 265.
23. The SSAT experiment was performed according to the
procedure used by: Porter, C. W.; Ganis, B.; Libby, P. R.;
Bergeron, R. J. Cancer Res. 1991, 51, 3715. When the positive
control DENSPM was assayed in the same manner a
6969Æ1062-fold induction of SSAT was observed.
24. Casero, R. A.; Woster, P. M. J. Med. Chem. 2001, 44, 1.
25. These experiments were performed using an adaptation of
the procedures previously described by: Jenkins, T. C. Drug-
DNA Interactions Protocols; Fox, K. R. Ed.; In Methods in
Molecular Biology; Humana: Totowa, 1997; Vol. 90, p 195.
Results described used the following conditions: 2.5 mM poly-
dAdT, 1.5 mM ethidium bromide in 3.75 mM Tris/NaCl at pH
8.0. DC50 is the concentration required to quench half the
fluorescent signal obtained by excitation at 535 nm and emis-
sion at 590 nm.
26. Stefanelli, C.; Bonavita, F.; Stanic’, I.; Pignatti, C.; Fla-
migni, F.; Guarnieri, C.; Caldarera, C. M. FEBS Lett. 1999,
451, 95.