Cis-unsaturated analogues of 3,8,13,18,23-pentaazapentacosane (BE-4-4-4-4): Synthesis add growth inhibitory effects on human prostate cancer cell lines

Article abstract of DOI:10.1021/jm000310s

From the results of our previous physicochemical studies of polyamine - Nucleic acid interactions, we concluded that polyamine analogues in cisoidal conformation are capable of wrapping around the major groove of the double helix, of displacing natural po

Full text of DOI:10.1021/jm000310s

404
J . Med. Chem. 2001, 44, 404-417
Cis-Un sa tu r a ted An a logu es of 3,8,13,18,23-P en ta a za p en ta cosa n e (BE-4-4-4-4):
Syn th esis a n d Gr ow th In h ibitor y Effects on Hu m a n P r osta te Ca n cer Cell Lin es
Venodhar K. Reddy, Aparajita Sarkar, Aldonia Valasinas, Laurence J . Marton, Hirak S. Basu,* and
Benjamin Frydman*
SLIL Biomedical Corp., 535 Science Drive, Suite C, Madison, Wisconsin 53711
Received J uly 21, 2000
From the results of our previous physicochemical studies of polyamine-nucleic acid interactions,
we concluded that polyamine analogues in cisoidal conformation are capable of wrapping around
the major groove of the double helix, of displacing natural polyamines from their nucleic acid
binding sites, and of inhibiting cell division. On the basis of this hypothesis, nine unsaturated
pentamines, formally derived from the cytotoxic pentamine 3,8,13,18,23-pentaazapentacosane
(BE-4-4-4-4), were prepared in an attempt to increase antineoplastic activity. Cis-double bonds
were introduced in all possible sites in the saturated pentaazapentacosane structure of BE-
4-4-4-4 to yield two pentacosenes, four pentacosadienes, two pentacosatrienes, and one
pentacosatetraene. Cis-double bonds should also provide good targets for mixed-function
oxidases that might eliminate the accumulation of unsaturated pentamines in serum, thereby
reducing systemic toxicity in animals. We determined the ability of these new pentamines to
inhibit growth in four cultured human prostate cancer cell lines (LnCap, DU145, PC-3, and
DuPro) using a MTT assay. LnCap and DU145 cells were very sensitive, PC-3 cells were
relatively resistant, and DuPro cells were intermediate in sensitivity to most of these synthetic
pentamines. In all cell lines, pentamines that had unsaturation(s) at the end of the chain showed
the highest cell growth inhibitory effects. The cellular uptake, effects on cellular polyamine
levels, and cytotoxicity of these pentamines on one representative prostate cancer cell line
(DuPro) were further examined with a colony-forming efficiency (CFE) assay. The pentamines
with unsaturation(s) at the end of the chain were once again the most cytotoxic among both
the saturated (BE-4-4-4-4) and unsaturated analogues. Appreciable amounts of all pentamines
entered DuPro cells and depleted cellular polyamine pools by day 6 of treatment. For most
pentamines, however, cell growth inhibitory and cytotoxic effects could not be directly correlated
either with their cellular uptake or with their ability to deplete cellular polyamine pools. The
position of the double bonds in the aliphatic backbone seems to be the most important
determinant of cytotoxicity. For some pentamines, however, depletion of cellular polyamines
may add to their efficacy.
In tr od u ction
experimental evidence suggests that DNA has consider-
able flexibility.6 The structural changes that take place
in DNA as well as in model oligo- and polynucleotides
upon interaction with polyamines have been extensively
studied by us and by others using both experimental
methods as well as theoretical models.7,8 It has been
conclusively shown that polyamines, including the pen-
tamine 3,8,13,18,23-pentaazapentacosane (BE-4-4-4-4),
interact with anionic DNA, induce changes in DNA
secondary structure, and thus alter the flexibility of the
DNA chain.7 It is also known that spermine and
spermidine bind to t-RNA, are needed to stabilize the
Γ-shaped structure of the t-RNAs, and are required for
the fidelity of transcription.1 RN,ωN-Bisethyl- and -bis-
methylspermine and its higher and lower homologues
were found to bind regioslectively to t-RNA as a
consequence of different hydrogen-bonding modes that
are formed between both types of molecules.9 In an
attempt to apply these extensive studies on polyamine-
nucleic acid interactions to drug discovery, pentamines
were designed where changes were made in the lengths
The polyamines putrescine, spermidine, and spermine
are organic cations found in all mammalian cells and
are required for cell proliferation.1 A number of poly-
amine analogues have shown promise as anticancer
agents.2,3 They inhibit cell growth and even kill cancer
cells both in tissue culture as well as in experimental
animal models. Most successful among these analogues
have been the RN,ωN-bisethyl derivatives of spermine
and its higher and lower homologues.4,5 Of the many
theories advanced to explain the biological effects of the
polyamines, the hypothesis centering on their binding
to nucleic acids is the most compelling.1 Both spermine
and spermidine are strong bases that are protonated
at physiological pH.2,3 These cationic polyamines can
bind to the negatively charged nucleic acids either by
electrostatic interactions or by hydrogen bonding. DNA-
binding compounds have always been of interest as
anticancer agents. Although the model of the DNA
double helix has created an image of a rigid structure,
* To whom correspondence should be addressed. Phone: 608-231-
3875, ext. 12 or 23. Fax: 608-231-3892. E-mail: hsbasu@slil.com,
bjfrydman@slil.com.
+
of the carbon chains separating the central NH2
groups, in the distribution of the positive charges along
10.1021/jm000310s CCC: $20.00 © 2001 American Chemical Society
Published on Web 12/16/2000

Cytotoxic Polyamine Analogues
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 3 405
Sch em e 1
Sch em e 2
the molecule’s backbone, and in the degree of rotational
freedom of the polyamine backbone.5,7-13 One of our
most successful designs was BE-4-4-4-4, a higher ho-
mologue of bisethylspermine. BE-4-4-4-4 has a relatively
higher affinity for DNA than has spermine, it is more
efficient than is spermine in inducing structural changes
in DNA, and unlike spermine, it inhibits nucleosome
formation in cell free systems.12 It also enters cells and
produce alterations both in chromatin structure as well
as in gene expression in cellular systems.14,15 It is
cytotoxic to many human tumor cell lines both in culture
and in nude mouse xenografts.12,13 It is presently under
consideration for clinical trials as an antineoplastic
agent.
In our search for polyamine analogues with enhanced
cytotoxic effects and decreased systemic toxicity, we
decided to refine the structure of BE-4-4-4-4 by intro-
ducing unsaturation(s) in the pentacosane backbone.
Our previous molecular dynamic studies on spermine-
DNA interactions had suggested that in a minimum
energy conformation, spermine binds to DNA in a
cisoidal conformation that wraps around the major
groove of the double helix.16 Cis-double bonds in the
alkane chain should also provide good targets for mixed-
function oxidases that could oxidize the unsaturation-
(s) to diols,17 thus facilitating the elimination of the
potentially toxic pentamines from animals. Here we
report the synthesis of mono-, di-, tri-, and tetraunsat-
urated derivatives of BE-4-4-4-4 (see Table 1 for struc-
tures). The resulting nine pentamines were evaluated
in tissue culture against human prostate cancer cells.
As the prostate gland is one of the major sites of
polyamine biosynthesis in mammals and since prostate
cancers are generally sensitive to inhibitors of polyamine
biosynthesis,18 four human prostate cancer cell lines
(LnCap, DU-145, DuPro, and PC-3) were selected for
this study. We studied the effects of the nine pentamines
on the growth of the four cell lines with a MTT assay.
We selected DuPro as a representative cell line and
further studied the effects of the pentamines on growth,
intracellular polyamine levels, and cell survival, using
a colony-forming efficiency (CFE) assay. Our results
suggest that pentamines with cis-double bonds in the
terminal segments of the pentacosane chain are not only
more cytotoxic than are those pentamines with cis-
double bonds in the central segments but are also more
cytotoxic than is the saturated parent pentamine BE-
4-4-4-4.
Ch em istr y
The synthesis of the cis-unsaturated pentamines
required the prior availability of the diamide 4 (Scheme
1). Alkylation of N-ethylmesityleneamide (1) with 4-bro-
mobutyronitrile gave the nitrile 2. The latter was then
reduced to the mesitylenediamine 3 and converted to
the diamide 4. Alkylation of 1 with N-pthaloyl-cis-4-
chloro-2-butenylamine gave phthaloylamide 5. The
phthaloyl protecting group was cleaved with hydrazine,
and the resulting amine was protected by acylation with
mesitylenesulfonyl (Mts) chloride to give 7. Monoalky-
lation of 7 with cis-1,4-dichloro-2-butene gave the
expected diene 9 (Scheme 2). Alkylation of 7 with 1,4-
diiodobutane gave the monounsaturated diamide 8.
Monoalkylation of 4 with 4-chlorobutylnitrile gave the
diamide 10, and on reduction of 10 over Raney Ni,
amine 11 was obtained. Amidation of 11 gave triamide
12 (Scheme 3). Alkylation of 4 with cis-1,4-dichloro-2-
butene gave diamide 13.
Using the abovementioned intermediates it was pos-
sible to build the backbones of several unsaturated
pentamines. Thus, the condensation of 12 and 9 gave
pentamide 14 that after deprotection from its mesityl-
enesulfonyl residues gave pentamine 15 that was iso-
lated as its pentahydrochloride (Scheme 3). Condensa-
tion of 12 and 8 gave pentamide 16; after cleavage of
the amide residues, pentamine 17 was isolated as its
pentahydrochloride. Condensation of 12 and 13 gave
pentamide 18; on deprotection pentamine 19 was ob-
tained (Scheme 4). Good reaction yields were obtained
for all the reaction sequences.
A second approach to the unsaturated pentamines
was based on the dialkylation of mesitylenesulfonamide.

406 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 3
Reddy et al.
Sch em e 3
Sch em e 4
Thus, by alkylation of the amide with iodoalkyl diamide
8 in the presence of a phase-transfer reagent it was
possible to obtain the pentamide 20 (Scheme 5). Depro-
tection gave the cis-diene 21. Dialkylation of mesityl-
enesulfonamide with 9 gave pentamide 22; on depro-
tection the completely unsaturated pentamine 23 was
obtained. Dialkylation of mesitylenesulfonamide with
13 gave pentamide 24 that on deprotection gave pen-
tamine 25. Monoalkylation of mesitylenesulfonamide
with chloroalkyl diamide 13 gave triamide 26; by further
alkylation of 26 with iodoalkyl diamide 8 it was possible
to obtain pentamide 27. Deprotection gave pentamine
28 (Scheme 6). When mesitylenesulfonamide was first
alkylated with chloroalkyl diamide 9 to give 29, and the
latter then alkylated with 8, it was possible to obtain
pentamide 30. Deprotection of 30 gave pentamine 31.
By condensation of triamide 29 and diamide 13, it was
possible to obtain pentamide 32. Deprotection of 32 gave
pentamine 33. These synthetic procedures allowed the
obtention of all the possible cis-unsaturated isomers of
the pentamines in preparative amounts when needed.
SLIL identification numbers and chemical structures
arranged in increasing order of unsaturation are shown
in Table 1. The cell lines displayed different sensitivities
toward the pentamines. LnCaP and DU145 cells were
the most sensitive with ID50 values for all pentamines
of less than 1 µM. In contrast, DuPro and PC-3 cells
were resistant to many of the pentamines with ID50
values for some of them above the highest drug concen-
tration used (31.25 µM). Analysis of the data shown in
Table 1 indicates that those pentamines that had ID50
values less than 1 µM in all four cell lines were either
completely saturated (BE-4-4-4-4) or carried at least one
unsaturated site in the terminal segment of the hydro-
carbon chain (17 and 21). Pentamines with double bonds
in the mid-section of the polyamine chain showed
markedly less growth inhibitory activity, particularly
in DuPro and PC-3 cell lines. PC-3 cells are generally
more resistant to the pentamines than are many other
human tumor cell lines, perhaps due to a diminished
capacity of the polyamine transport system in PC-3
cells.19
We selected the DuPro cell line that exhibits an
intermediate level of sensitivity to the pentamines
(Table 1) as a representative cell line for detailed study.
All nine pentamines plus BE-4-4-4-4 at 1 and 10 µM
were assayed for their effects on changes in cell number
and in cellular polyamine levels over time. The effects
of the pentamines on DuPro cell growth along with the
Biologica l Resu lts a n d Discu ssion
The concentration of a pentamine required to inhibit
50% cell growth (ID50) after 6 days of treatment was
determined by a MTT assay. The ID50 values for each
pentamine in four human prostate cancer cell lines
(LnCap, DU145, DuPro, and PC-3) together with their

Cytotoxic Polyamine Analogues
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 3 407
Sch em e 5
Sch em e 6
cellular polyamine levels are shown in Figures 1-4. At
1 µM, BE-4-4-4-4 inhibited growth by 50% on day 6 of
treatment (Figure 1A). No other pentamine showed any
observable growth inhibitory effect at 1 µM (Figures 1A
and 2A). While all pentamines were detectable in cell
lysates, 31 and 33 were present at higher concentrations
than were the other pentamines (Figures 1B,C and
2B,C). At 1 µM, no pentamine was able to completely
deplete any of the natural polyamines. In most cases,
cellular polyamine levels were within 50% of those in
control cells and for 25, 23, 33, and 28, the polyamine
levels were actually higher than those of the controls

408 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 3
Reddy et al.
Ta ble 1. Chemical Structures and Effects of the Pentamines on Human Prostate Tumor Cell Growth as Determined by MTT Assay
on day 6 of treatment (Figure 2C). At 10 µM, however,
a marked decrease in cell growth was observed not only
for BE-4-4-4-4 but also for 17 and 21, particularly on
day 6 of treatment (Figure 3). Interestingly, 19 and 15,
which showed good growth inhibition as assessed with
the MTT assay (Table 1), showed little effect, even at
10 µM, on day 6 after treatment. Cell growth was near
that of control after an initial delay. This underscores
the necessity of confirming the MTT data with assays
that follow a time course before drawing definitive
conclusions about growth inhibitory activities. At 10 µM,
appreciable amounts of all pentamines were detected
in cell lysates (Figures 3B,C and 4B,C). While none of
the pentamines was able to completely deplete the
cellular polyamine pools, BE-4-4-4-4, 17, 19, 15, and 21
were more efficient in decreasing cellular polyamines
than were the other pentamines (compare Figure 3 with
Figure 4). Similar to the treatment with 1 µM, treat-
ment with 10 µM 25, 23, 31, and 33 resulted in an
increase in the cellular polyamine levels, particularly
on day 6 of treatment (Figure 4C).
The abilities of various pentamines to deplete cellular
polyamine pools correlate poorly with their growth
inhibitory effects. For example, at 10 µM, 19 and 15
failed to arrest cell growth but depleted cellular
polyamine levels to the same extent as did 17 and 21
(Figure 3), two very efficient growth inhibitors. On the
other hand, 10 µM 33 inhibited cell growth to the same
extent as did 10 µM BE-4-4-4-4 (compare Figure 3 with
Figure 4), but 33 had little effect on putrescine and
spermidine levels and actually increased the cellular
spermine level by day 6 of the treatment, while BE-4-
4-4-4 markedly decreased all natural polyamine pools.
Although there are some variations in the intracellular
pentamine levels, no correlation could be observed
between intracellular pentamine levels and growth
inhibitory activities. For example, on day 6 of treatment
with 10 µM pentamine, the intracellular level of the
weak growth inhibitor 28 was approximately 3-5-fold
higher than were those of the strong growth inhibitors
17 and 21 (compare Figures 3 and 4). Therefore, we had
to conclude that the growth inhibitory activities of the
pentamines are in general related neither to their
intracellular levels nor to their abilities to deplete
cellular polyamine pools. As was the case with other
strong DNA-binding polyamine analogues,20 their cy-
totoxicity may relate to their DNA-binding abilities
(Basu et al., unpublished results). If the intracellular
pentamine concentration is sufficient to displace cellular
polyamines from their DNA binding site(s), the pen-
tamine can act as a growth inhibitor even in the
presence of cellular polyamines. The depletion of cellular
polyamines, however, may help enhance the growth
inhibitory activities of a few of the pentamines.
The cytotoxic effects of pentamines that exhibit
significant growth inhibition was further studied using
a CFE assay. The CFE data for most of the active
pentamines on day 6 of treatment are shown in Figure
5; 28 was used as a control to confirm that compounds
that do not exhibit growth inhibition do not exhibit
cytotoxicity. Only two pentamines, 17 and 21, killed
more cells than did BE-4-4-4-4. Both pentamines have

Cytotoxic Polyamine Analogues
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 3 409
F igu r e 1. (A) Effects of 1 µM BE-4-4-4-4, 17, 19, 15, and 21 on the growth of DuPro cells. Symbol for each pentamine is shown
in the inset. Each data point is an average of at least three separate experiments run in duplicate. Error bars where not shown
are smaller than the symbol size. (B) Effects of 1 µM BE-4-4-4-4, 17, 19, 15, and 21 on the polyamine levels of DuPro cells on day
4 of treatment. Symbol for each polyamine is shown in the inset. Each data point is an average of two separate experiments. (C)
Effects of 1 µM BE-4-4-4-4, 17, 19, 15, and 21 on the polyamine levels of DuPro cells on day 6 of treatment. Symbol for each
polyamine is shown in the inset. Each data point is an average of two separate experiments.
their unsaturated site(s) at the terminal segment of the
hydrocarbon chain(s). Introduction of an unsaturated
site in the middle of the chain markedly reduced the
pentamine’s cytotoxicity, as compared to that of BE-4-
4-4-4, even though some of these pentamines may
induce initial growth delay.
The rationale behind the strong cytotoxic activities
of 21 and 17 is difficult to identify with precision, since
the mechanism by which polyamines induce cell death
is still controversial.1 As mentioned before (see Intro-
duction), we favor explanations that are based on the
binding of the polyamine analogues to nucleic acids. It
has been shown that DNA bending is necessary for
nucleosome formation during cell division.21 Therefore,
interference with DNA bending should inhibit chroma-
tin condensation and cell division. We suggested that a
flexible polyamine backbone is necessary for efficient
polyamine-DNA interactions.7 The flexibility of the
polyamine chain will facilitate its interaction with DNA
and help in displacing the natural polyamines from their
DNA binding site(s). Pentamines with unsaturated
site(s) at the end segments of the carbon chain (as in
17 and 21) could therefore be more efficient binders than
pentamines with unsaturated sites at the center of the
chain. Once bound, the partially restricted conformation
of the active pentamines might affect DNA function
more than their saturated counterpart by introducing
unwanted steric effects in its spatial arrangement. We
are, however, aware that subtler biochemical effects
such as induction of enzymes of the cellular apoptotic

410 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 3
Reddy et al.
F igu r e 2. (A) Effects of 1 µM 25, 23, 31, 33, and 28 on the growth of DuPro cells. Symbol for each pentamine is shown in the
inset. Each data point is an average of at least three separate experiments run in duplicate. Error bars where not shown are
smaller than the symbol size. (B) Effects of 1 µM 25, 23, 31, 33, and 28 on the polyamine levels of DuPro cells on day 4 of
treatment. Symbol for each polyamine is shown in the inset. Each data point is an average of two separate experiments. (C)
Effects of 1 µM 25, 23, 31, 33, and 28 on the polyamine levels of DuPro cells on day 6 of treatment. Symbol for each polyamine
is shown in the inset. Each data point is an average of two separate experiments.
pathways22 or inhibition of the synthesis of cell cycle-
related proteins23 may also contribute to the cytotoxic
effects of the unsaturated pentamines. Very recent
evidence in yeast cells has shown that several of these
subtle effects, such as the expression of certain cell cycle-
related proteins, can be regulated by cellular polyamine-
induced changes in chromatin structure.24
one a monounsaturated and one a diunsaturated pen-
tamine, where the double bonds are placed at the
terminal butane segments of the pentaazapentacosane
chain, were nearly as growth inhibitory as the fully
saturated analogue (BE-4-4-4-4). Among all the pen-
tamines, they were the most efficient inhibitors of cell
growth at 10 µM when assayed by cell counting in the
DuPro cell line and were about 1 order of magnitude
more cytotoxic than was BE-4-4-4-4 as determined by
the CFE assay. It can be expected that the introduction
of cis-double bonds in the polyamine backbone would
favor their in vivo catabolism by mixed-function oxi-
dases and facilitate their elimination by conjugation
with metabolic transporters. Only after completing the
in vivo assays will it become clear if the new unsatur-
Con clu sion s
The synthesis of nine new unsaturated pentamines
formally derived from the strong cytotoxic agent BE-4-
4-4-4 was reported. The unsaturated pentamines carry
cis-double bonds at all possible positions of the pen-
taazapentacosane chain and are the mono-, di-, tri-, and
tetraunsaturated derivatives. When assayed against
four human prostate cancer cell lines, two pentamines,

Cytotoxic Polyamine Analogues
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 3 411
F igu r e 3. (A) Effects of 10 µM BE-4-4-4-4, 17, 19, 15, and 21 on the growth of DuPro cells. Symbol for each pentamine is shown
in the inset. Each data point is an average of at least three separate experiments run in duplicate. Error bars where not shown
are smaller than the symbol size. (B) Effects of 10 µM BE-4-4-4-4, 17, 19, 15, and 21 on the polyamine levels of DuPro cells on
day 4 of treatment. Symbol for each polyamine is shown in the inset. Each data point is an average of two separate experiments.
(C) Effects of 10 µM BE-4-4-4-4, 17, 19, 15, and 21 on the polyamine levels of DuPro cells on day 6 of treatment. Symbol for each
polyamine is shown in the inset. Each data point is an average of two separate experiments.
fluorescence spectrometer (340-nm excitation, 515-nm emis-
sion) for detection were used.
ated pentamines are indeed a major step toward the
development of effective cancer chemotherapeutic agents.
N-Eth yl-N-(3-cya n obu tyl)m esitylen esu lfon a m id e (2).
NaH (80%, 1.08 g, 36 nmol) was added under nitrogen to a
solution of amide 111b (6.8 g, 30 mmol) in DMF (50 mL). The
mixture was stirred for 1 h and a solution of 4-bromobuty-
ronitrile (4.88 g, 33 mmol) in DMF (10 mL) was slowly added.
The mixture was stirred overnight at 75 °C, the solvent was
distilled off, the residue taken up in chloroform, washed with
a saturated solution of ammonium chloride, dried (Na2SO4),
and evaporated. The residue was purified by flash chroma-
tography on silica gel (hexane-ethyl acetate 3:1); 8.0 g (90%)
of 2 was obtained as a colorless oil: 1H NMR (CHCl3) δ 1.05
(t, 3H), 1.90 (m, 2H), 2.30 (bm, 5H), 2.60 (s, 6H), 3.20 (q, 2H),
3.35 (t, 2H), 6.95 (s, 2H); 13C NMR (CDCl3) δ 12.50, 20.61,
22.43, 23.60, 31.05, 36.12, 40.39, 43.78, 118.62, 131.79, 132.67,
139.71, 142.41; MS-ESI (m/z) calcd 294.41 (M+), found 294.40.
Exp er im en ta l Section
Ch em istr y. NMR spectra were obtained using a Bruker
AM-250 spectrometer. Reactions were monitored using TLC
on silica gel plates (0.25 mm thick). Flash chromatography was
performed on columns packed with EM Science silica gel, 230-
400 mesh ASTM. Melting points were determined on an
Electrothermal IA 9100 digital melting point apparatus. Mass
spectra (ESI) were obtained on a PE Sciex API 365 electro-
spray triple quadrupole spectrometer; mass spectra (FAB)
were obtained on an Autospec (VG) spectrometer. HPLC
analyses of the dansyl derivatives of the polyamine pentamines
were routinely performed to check the purity of the samples.
A Vydac C-18 (300-µm pore) column for separations and a

412 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 3
Reddy et al.
F igu r e 4. (A) Effects of 10 µM 25, 23, 31, 33, and 28 on the growth of DuPro cells. Symbol for each pentamine is shown in the
inset. Each data point is an average of at least three separate experiments run in duplicate. Error bars where not shown are
smaller than the symbol size. (B) Effects of 10 µM 25, 23, 31, 33, and 28 on the polyamine levels of DuPro cells on day 4 of
treatment. Symbol for each polyamine is shown in the inset. Each data point is an average of two separate experiments. (C)
Effects of 10 µM 25, 23, 31, 33, and 28 on the polyamine levels of DuPro cells on day 6 of treatment. Symbol for each polyamine
is shown in the inset. Each data point is an average of two separate experiments.
1N,4N-Bis(m esitylen esu lfon yl)-4N-eth yl-1,4-d ia m in obu -
ta n e (4). Nitrile 2 (7.8 g, 27 mmol) was dissolved in a mixture
of ethanol (150 mL) and concentrated HCl (1.5 mL), PtO2 was
added (700 mg), and the mixture was hydrogenated at 50 psi
overnight. The catalyst was filtered off, the solvent was
evaporated and the amine 3 (7.8 g, 98%) was used in the next
step without further purification: 1H NMR (CDCl3, free base)
δ 1.00 (t, 3H), 1.55 (m, 4H), 2.25 (s, 3H), 2.80 (t, 2H), 3.20 (m,
4H), 6.95 (s, 2H); 13C NMR (CDCl3) δ 12.54, 20.69, 22.53, 24.72,
27.65, 39.92, 40.29, 44.59, 131.71, 133.21, 139.82, 142.09; MS-
FAB (m/z) calcd 299.94 (M+ + 1), found 299.45.
Mesitylenesulfonyl chloride (8.8 g, 40.5 mmol) in dioxane
(30 mL) was added dropwise to a stirred mixture of 3 (7.8 g,
27 mmol) dissolved in dioxane (60 mL) and 50% KOH (30 mL)
at 5 °C. The reaction mixture was allowed to reach 20 °C and
then kept overnight. An excess of water was added and the
mixture was extracted with chloroform, dried (Na2SO4) and
evaporated. The oily residue was crystallized from ethyl
acetate-hexane; 10.9 g (82%) of 4 was obtained: mp 71.5-72
1
°C; H NMR (CDCl3) δ 1.00 (t, 3H), 1.10-1.50 (m, 4H), 2.30
(s, 6H), 2.55, 2.60 (s, 12H), 2.85 (q, 2H), 3.15 (m, 4H), 4.70 (t,
1H), 6.95, 7.00 (s, 4H); 13C NMR (CDCl3) δ 12.70, 20.92, 21.04,
22.73, 22.92, 24.58, 26.68, 40.04, 42.02, 44.42, 131.91, 133.31,
133.64, 138.99, 140.05, 142.15, 142.35; MS-FAB (m/z) calcd
480.68 (M+), found 480.69. Anal. (C24H36N2O4S2) C, H, N.
N-Eth yl-N-(4-ph th alim ido-2-bu ten yl)m esitylen esu lfon a-
m id e (5). Amide 1 (3.1 g, 13.65 mmol) was dissolved in
anhydrous DMF (30 mL), NaH (85%, 0.42 g) was slowly added,
and the mixture was stirred at room temperature for 1 h. N-(4-
Chloro-2-butenyl)phthalimide25 (3.06 g, 13 mmol) dissolved in
20 mL of DMF was added and the mixture stirred at 80 °C for
18 h. It was then cooled to room temperature, quenched with
water (10 mL), and the solution was evaporated to dryness in
vacuo. The aqueous layer was extracted with chloroform (3 ×
25 mL), the organic layers were washed with brine (35 mL),
dried (MgSO4), and the solvent was evaporated leaving behind

Cytotoxic Polyamine Analogues
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 3 413
F igu r e 5. Effects of BE-4-4-4-4 (O), 17 (9), 19 (2), 15 (1), 21 ([), and 28 (0) on the survival of DuPro cells on day 6 of treatment.
Each data point and corresponding error bars are, respectively, an average and the standard deviation of six independent
observations.
a viscous residue that solidified upon titration with hexane to
give 5. It was pure enough to be used in the next step without
purification: 4.75 g (82%); 1H NMR (CDCl3) δ 1.16 (t, J ) 7.11
Hz, 3H), 2.29 (s, 3H), 2.63 (s, 6H), 3.29 (q, J ) 7.11 Hz, 2H),
4.06 (d, J ) 5.24 Hz, 2H), 4.26 (d, J ) 5.72 Hz, 2H), 5.59 (m,
2H), 6.95 (s, 2H), 7.71 (m, 2H), 7.83 (m, 2H); 13C NMR (CDCl3)
δ 13.06, 20.89, 22.72, 34.35, 40.68, 42.01, 123.27, 126.69,
129.47, 131.90, 134.00, 140.24; MS-FAB (m/z) calcd 426.53
(M+), found 426.54.
N -E t h yl-N -(4-a m in o-2-b u t e n yl)m e sit yle n e su lfon a -
m id e (6). Amide 5 (20 g, 47 mmol) was dissolved in methanol,
hydrazine monohydrate (50 mL, 98.5 mmol) was added and
the solution stirred at 55 °C for 24 h. Initially it was a
homogeneous solution; after several hours a white solid
precipitated. The mixture was cooled to 20 °C, 300 mL of
concentrated HCl was slowly added (exothermic reaction),
stirring was continued for additional 12 h, the solution was
evaporated, and the resulting solid was dissolved in 200 mL
of water and extracted with chloroform (3 × 150 mL). The
combined organic layers were pooled, dried (MgSO4),and
evaporated to dryness. The residue was purified by flash
chromatography using hexanes-ethyl acetate (7:3) as eluant;
9.0 g (65%) of 6 was obtained: 1H NMR (CDCl3) δ 1.0 (t, J )
7.1 Hz, 3H), 2.28 (s, 3H), 2.56 (s, 6H), 2.62 (br, 2H), 3.12 (q, J
) 7.1 Hz, 3H), 3.73 (br, 2H), 3.94 (d, J ) 6.0 Hz, 2H), 5.80 (m,
2H), 6.92 (s, 2H); 13C NMR (CDCl3) δ 12.97, 20.93, 22.74, 36.43,
40.94, 42.08, 124.29, 131.89, 132.00, 132.62, 140.21, 142.67;
MS-FAB (m/z) calcd 480.68 (M+), found 296.44. It was used
without further purification in the next step.
3N,8N-Bis(m esitylen esu lfon yl)-12-ch lor o-3,8-d ia za -(5Z,-
10Z)-d od eca d ien e (9). Amide 7 (4.8 g, 10 mmol) was dis-
solved in anhydrous DMF (40 mL); NaH (0.37 g) was added
in portions while the mixture was stirred at room temperature
for 2 h. cis-1,4-Dichloro-2-butene (7.5 g, 60 mmol) in 10 mL
DMF was then added, and stirring was continued at 50 °C for
18 h. The mixture was cooled to 20 °C, quenched with 10 mL
water, the solvents evaporated, and the contents were parti-
tioned between water (50 mL) and chloroform (50 mL). The
aqueous layer was further extracted with chloroform (3 × 50
mL), the pooled organic layers dried (MgSO4), evaporated to
dryness, and 9 was purified by flash chromatography using
hexane-ethyl acetate (8.5:1.5) as eluant: 5.5 g (97%); mp 106-
1
108 °C; H NMR (CDCl3) δ 1.03 (t, 3H), 2.30 (s, 6H), 2.57 (s,
12H), 3.17 (q, 2H), 3.71 (m, 4H), 3.81 (d, 2H), 3.95 (d, 2H),
5.50 (m, 3H), 5.74 (m, 1H), 6.93 (s, 2H), 6.95 (s, 2H); 13CNMR
(CDCl3) δ 12.91, 22.70, 22.74, 38.20, 40.45, 41.60, 42.11,42.33,
128.17, 128.95, 129.34, 129.40, 131.94, 132.08, 140.23, 140.34,
142.91. Anal. (C28H39ClN2O4S2) C, H, N.
3N,8N-Bis(m esit ylen esu lfon yl)-12-iod o-3,8-d ia za -(5Z)-
d od ecen e (8) was prepared (70%) from 7 and 1,4-diiodobutane
as described above for 9. The product was purified by column
chromatography using hexanes-ethyl acetate (4:1) as eluant:
1H NMR(CDCl3) δ 1.04 (t, 3H), 1.63 (m, 4H), 2.30 (s, 6H), 2.58
(s, 12H), 3.04 (t, 2H), 3.16 (m, 4H), 3.78 (d, 4H), 5.55 (m, 2H),
6.94 (s, 2H), 6.95 (s, 2H); 13C NMR (CDCl3) δ 5.69, 12.92, 20.95,
22.72, 22.78, 28.25, 30.36, 40.47, 41.59, 42.11, 44.71, 128.34,
129.00, 131.94, 132.06, 132.60, 132.89, 140.15, 140.21, 142.50,
142.71; MS-FAB (m/z) calcd 661.67 (M+ +1), found 661.68.
3N,8N-Bis(m esitylen esu lfon yl)-3,8-diazalau r on itr ile (10)
was prepared (94%) from 4 and 4-bromobutyronitrile as
described for 2: 1H NMR (CDCl3) δ 0.97 (t, 3H), 1.40 (m, 4H),
1.85 (pent, 2H), 2.27 (t, 2H), 2.30 (s, 6H), 2.57 (s, 6H), 2.58 (s,
6H), 3.13 (m, 6H), 3.28 (t, 2H), 6.94 (s, 2H), 6.96 (s, 2H); 13C
NMR (CDCl3) δ 12.55, 14.54, 20.84, 22.64, 22.73, 23.65, 24.43,
24.57, 39.88,44.31, 44.54, 45.58, 118.69, 131.84, 132.05, 132.73,
133.36, 139.94, 142.20, 142.71; MS-FAB (m/z) calcd 548.77 (M+
+ 1), found 548.78.
1N,4N-Bis(m esitylen esu lfon yl)-4N-eth yl-1,4-d ia m in o-2-
bu ten e (7) was obtained from 6 as described for 4 in 96%
yield. It was purified by flash chromatography using hexane-
ethyl acetate (4:1.5) as eluant: mp 98-99 °C; 1H NMR (CDCl3)
δ 0.93 (t, J ) 5.85 Hz, 3H), 2.23 (s, 3H), 2.24 (s, 3H), 2.50 (s,
6H), 2.56 (s, 6H), 3.06 (q, J ) 7.15 Hz, 2H), 3.48 (t, J ) 5.99
Hz, 2H), 3.68 (d, J ) 5.72 Hz, 2H), 4.58 (t, J ) 6.24 Hz, 1H),
5.44 (m, 2H), 6.87 (s, 2H), 6.89 (s, 2H); 13C NMR (CDCl3) δ
12.80, 20.89, 22.64, 22.89, 39.01, 40.59, 41.41, 128.14, 128.46,
131.91, 139.08, 140.19, 142.26, 142.54; MS-FAB (m/z) calcd
480.68 (M+).
3N,8N-Bis(m esitylen esu lfon yl)-3,8-diazalau r ylam in e (11)
was prepared (93%) from 10 as described above for 3: 1H NMR

414 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 3
Reddy et al.
(CDCl3) δ 1.00 (t, 3H), 1.40 (m, 10H), 2.29 (s, 6H), 2.57 (b, 14H),
3.13 (m, 8H), 6.93 (s, 4H); 13C NMR (CDCl3) δ 12.72, 20.90,
22.72, 22.78, 24.67, 24.80, 30.80, 40.02, 41.61, 44.56, 45.10,
45.38, 131.87, 140.04, 142.21, 142.28; MS-FAB (m/z) calcd
552.80 (M+ + 1), found 552.81.
1,6,11-Tr is(m e sit yle n e su lfon yl)-1,6,11-t r ia za t r id e -
ca n e (12) was prepared (97%) from 11 as described for 4: 1H
NMR (CDCl3) δ 0.96 (t, 3H), 1.38 (m, 8H), 2.29 (s, 9H), 2.55
(s, 6H), 2.56 (s, 6H), 2.59 (s, 6H), 2.80 (m, 2H), 3.10 (m, 8H),
4.67 (t, 1H), 6.93 (s, 6H); 13C NMR (CDCl3) δ 12.56, 20.87,
22.70, 22.74, 22.84, 24.40, 26.45, 24.67, 26.62, 39.87, 41.88,
44.45, 45.02, 45.09, 131.86, 131.90, 133.12, 133.12, 133.32,
133.68, 138.91, 139.97, 142.02, 142.02, 142.21, 142.38; MS-
FAB (m/z) calcd 734.04 (M+), found 734.05.
49.32, 49.69, 129.11, 129.39; MS-ESI (m/z) calcd 356.61 (M+
+ 1), found 356.60.
3N,8N,13N,18N,23N-P en ta k is(m esitylen esu lfon yl)-3,8,13,-
18,23-p en ta za -(15Z)-p en ta cosen e (18) was prepared (52%)
from 12 and 13 as described for 14: 1H NMR (CDCl3) δ 0.95
(m, 6H), 1.32 (m, 12H), 2.29 (s, 15H), 2.55 (s, 30H), 3.06 (m,
16H), 3.70 (m, 4H), 5.47 (m, 2H), 6.92 (s, 10H); 13C NMR
(CDCl3) δ 12.67, 20.90, 22.71, 22.76, 24.43, 24.68, 39.97, 42.08,
44.48, 44.90, 45.61, 128.28, 128.45, 131.87, 131.93, 132.01,
139.96, 140.00, 140.12, 142.21, 142.28, 142.58; MS-FAB (m/z)
calcd 1266.80 (M+), found 1266.81.
3,8,13,18-P en taaza-(15Z)-pen tacosen e pen tah ydr och lo-
r id e (19) was prepared (96%) from 18 as described above for
15: 1H NMR (D2O) δ 1.28 (t, 6H), 1.78 (m, 12H), 3.09 (m, 16H),
3.84 (m, 4H), 5.96 (m, 2H); 13C NMR (D2O) δ 13.31, 25.61,
25.73, 45.70, 46.79, 49.05, 49.36, 49.65, 129.19; MS-ESI (m/z)
calcd 356.61 (M+ + 1), found 356.60.
3N,8N-Bis(m esitylen esu lfon yl)-12-ch lor o-3,8-diaza-(10Z)-
d od ecen e (13) was prepared (99%) from 4 and 1,4-dichloro-
1
2-butene as described for 9: H NMR (CDCl3) δ 1.01 (t, 3H),
3N,8N,13N,18N,23N-P en ta k is(m esitylen esu lfon yl)-3,8,13,-
23-p en ta za -(5Z,20Z)-p en ta cosa d ien e (20). A mixture of
potassium hydroxide (0.25 g), potassium carbonate (0.25 g) and
tetrabutylammonium bromide (0.05 g) was suspended in 15
mL benzene, mesitylenesulfonamide (0.2 g, 1 mmol) was added
to the suspension and the mixture was heated at 50 °C. Iodide
14 (2.0 g, 3 mmol) in 10 mL benzene was added, the mixture
heated under reflux for 18 h, then cooled to 20 °C, the inorganic
solids filtered off, washed with benzene (2 × 20 mL), and the
combined benzene layers washed several times with water
until the washings were neutral. The benzene was dried
(MgSO4), evaporated to dryness and the residue purified by
column chromatography using hexanes-ethyl acetate (7.5:2.5)
as eluant; 0.95 g (25%) of 20 was obtained: 1H NMR (CDCl3)
δ 1.00 (t, 6H), 1.28 (m, 8H), 2.29 (s, 15H), 2.53 (s), 2.55 (s),
2.57 (s), 3.03 (m, 8H), 3.12 (q, 4H), 3.70 (m, 8H), 5.47 (m, 4H),
6.93 (s, 10H); 13C NMR (CDCl3) δ 12.78, 20.85, 22.63, 22.69,
24.32, 24.58, 40.41, 41.43, 42.00, 44.76, 45.43, 128.08, 128.83,
131.88,131.95, 132.77, 132.85, 133.23, 139.90. 140.00, 140.08,
142.22, 142.43,142.53; MS-FAB (m/z) calcd 1264.78 (M+), found
1264.79.
1.38 (m, 4H), 2.29 (s, 3H), 2.30 (s, 3H), 2.57 (s, 6H), 2.61 (s,
6H), 3.11 (m, 4H), 3.16 (q, 2H), 3.81 (d, 2H), 3.98 (d, 2H), 5.51
(m, 1H), 5.77 (m, 1H), 6.93 (s, 2H), 6.95 (s, 2H); 13C NMR
(CDCl3) δ 12.76, 20.91, 22.71, 22.76, 24.74, 38.12, 40.08, 41.85,
44.59, 129.14, 129.25, 131.88, 132.02, 140.19, 142.21, 142.63;
MS-FAB (m/z) calcd 569.22 (M+ ), found 569.23.
3N,8N,13N,18N,23N-P en ta k is(m esitylen esu lfon yl)-3,8,13,-
18,23-p en ta a za -(15Z,20Z)-p en ta cosa d ien e (14) was pre-
pared (78%) from 12 and 9 following the procedure described
above for 9. It was purified by column chromatography using
hexane-ethyl acetate (7:3) as eluant: 1H NMR (CDCl3) δ 0.97
(t, 3H), 0.99 (t, 3H), 1.29 (m, 8H), 2.29 (s, 15H), 2.54, 2.55,
2.59 (s, 3H), 3.06 (m, 12H), 3.65 (m, 8H), 5.48 (m, 4H), 6.92 (s,
10H); 13C NMR (CDCl3) δ 12.70, 12.83 20.88, 20.91, 22.65,
22.68, 22.72, 22.74, 24.48, 24.72, 40.04, 40.47, 41.53, 42.07,
42.22, 42. 34, 44.54, 44.96, 127.94, 128.27, 129.20, 131.92,
132.05, 139.96, 140.00, 140.12, 140.16, 140.27, 142.19, 142.25,
142.47, 142.58, 142.87; MS-FAB (m/z) calcd 1264.78 (M+),
found 1264.77.
3,8,13,18-Tetr a a za -(15Z,20Z)-p en ta cosa d ien e P en ta h y-
d r och lor id e (15). Pentamide 14 (0.93 g, 0.735 mmol) was
dissolved in 20 mL anhydrous methylene chloride; phenol (3.5
g,36.8 mmol) was added, followed by 30% HBr in glacial acetic
acid (17.6 mL). The mixture was stirred overnight at 25 °C.
Water (10 mL) was added, the aqueous layer was separated,
the organic layer extracted with 5 mL of water, and the
combined aqueous layers were washed with methylene chlo-
ride (6 × 15 mL). The aqueous layer was evaporated under
vacuum to a solid residue that was dissolved in 1 mL 1 N
sodium hydroxide followed by 1 mL 50% potassium hydroxide.
The solution was extracted with chloroform (10 × 5 mL), the
combined organic layers were dried (MgSO4), evaporated, and
the residue dissolved in dry ethyl ether. Dry hydrogen chloride
was passed through the solution kept at 0 °C; pentahydro-
chloride 15 (0.33 g, 84%) precipitated, it was filtered and
washed with ether: 1H NMR (D2O) δ 1.29 (t, 3H), 1.31 (t, 3H),
1.79 (m, 8H), 3.12 (m, 12H), 3.87 (m, 8H), 5.98 (m, 4H); 13C
NMR (D2O) δ 13.36, 13.46, 25.66, 25.77, 45.44, 45.74, 46.24,
46.41, 46.84, 49.09, 49.41, 49.70, 129.02, 12916, 129.47, 129.66;
MS-ESI (m/z) calcd 354.59 (M+ + 1), found 354.60. Anal.
(C20H48Cl5N5) C, H, N.
3N,8N,13N,18N,23N-P en ta k is(m esitylen esu lfon yl)-3,8,13,-
18,23-p en ta a za -(20Z)-p en ta cosen e (16) was prepared (51%)
from 12 and 8 as described above for 14: 1H NMR (CDCl3) δ
0.97 (t, 3H), 0.99 (t, 3H), 1.29 (m, 12H), 2.29 (s, 15H), 2.55 (s),
2.56 (s), 2.57 (s), 3.10 (m, 16H), 3.70 (m, 4H), 5.47 (m, 2H),
6.93 (s, 10H); 13C NMR (CDCl3) δ 12.69, 12.83, 20.91, 22.69,
22.71, 22.76, 24.43, 24.70, 40.48, 41.11, 41.48, 44.50, 44.91,
128.13, 128.90, 131.88, 131.94, 132.01, 133.29, 139.95, 140.00,
140.15, 142.22, 142.29, 142.60; MS-FAB (m/z) calcd 1266.80
(M+), found 1266.79.
3,8,13,18,23-Tet r a a za -(5Z,20Z)-p en t a cosa d ien e p en -
ta h yd r och lor id e (21) was prepared (57%) from 20: 1H NMR
(D2O) δ 1.31 (t, 6H), 1.78 (m, 8.H), 3.15 (m, 12H), 3.83 (m,
8H), 5.96 (m, 4H); 13C NMR (CDCl3) δ 13.43, 25.64, 25.76,
45.39, 46.19, 46.77, 49.35, 49.72, 129.11, 129.41; MS-ESI (m/
z) calcd 354.59 (M+ + 1), found 354.60. Anal. (C20H48Cl5N5) C,
H, N.
3N,8N,13N,18N,23N-P en ta k is(m esitylen esu lfon yl)-3,8,13,-
18,23-p en ta a za -(5Z,10Z,15Z,20Z)-p en ta cosa tetr a en e (22)
was prepared (24%) from 9 and mesitylenesulfonamide as
1
described for 20: mp 58 °C; H NMR (CDCl3) δ 0.99 (t, 6H),
2.29 (s, 15H), 2.53 (s), 2.55 (s), 3.12 (q, 4H), 3.63 (m, 16H),
5.49 (m, 8H), 6.93 (s, 10H); 13C NMR (CDCl3) δ 12.85, 20.89,
20.92, 22.66, 40.47, 41.53, 42.19, 128.00, 128.47, 128.58,
129.11, 131.92, 132.05, 140.17, 140.30, 142.46, 142.87; MS-
FAB (m/z) calcd 1260.75 (M+), found 1260.76.
3,8,13,18,23-P en t a za -(5Z,10Z,15Z,20Z)-p en t a cosa t et -
r a en e p en ta h yd r och lor id e (23) was prepared (81%) from
22 as described above for 15: 1H NMR (D2O) δ 1.31 (t, 6H),
3.15 (q, 4H), 3.84 (m, 4H), 3.90 (m, 12H), 5.98 (m, 8H); 13C
NMR (D2O) δ 13.42, 45.41, 46.22, 46.44, 129.07, 129.37, 129.42,
129.58; MS-MALDI (m/z) calcd 350.56 (M+ + 1), found 350.55.
Anal. (C20H44Cl5N4) C, H, N.
3N,8N,13N,18N,23N-P en ta k is(m esitylen esu lfon yl)-3,8,13,-
18,23-p en ta a za -(10Z,15Z)-p en ta cosa d ien e (24) was pre-
pared (25%) from 13 as described above for 20: mp 62 °C; 1H
NMR (CDCl3) δ 0.95 (quint, 6H), 1.33 (m,8H), 2.29 (s, 15H),
2.54 (s), 2.55 (s), 3.07 (m, 12H), 3.65 (m, 8H), 5.48 (m,4H), 6.92
(s, 10H); 13C NMR (CDCl3) δ 12.69, 20.90,22.69, 22.73, 24.70,
40.03, 42.13, 42.30, 44.53, 45.59, 128.11, 128.79, 131.87,
132.00, 140.02, 140.14, 140.28, 142.17, 142.58, 142.85; MS-
FAB (m/z) calcd 1264.78 (M+), found 1264.79.
3,8,13,18,23-P en ta za -(20Z)-p en ta cosen e p en ta h yd r o-
ch lor id e (17) was prepared (79%) from 16 as described above
for 15: 1H NMR (D2O) δ 1.29 (t, 3H), 1.30 (t, 3H), 1.78 (m,
12H), 3.12 (m, 16H), 3.83 (m, 4H), 5.96 (m, 2H); 13C NMR (D2O)
δ 13.31, 13.42, 25.62, 25.75, 45.38, 45.71, 46.18, 46.76, 49.07,
3,8,13,18,23-P en ta a za -(10Z,15Z)-p en ta cosa d ien e p en -
ta h yd r och lor id e (25) was prepared from 24 (87%) as de-
1
scribed above for 15: H NMR (D2O) δ 1.28 (t, 6H), 1.79 (m,
8H), 3.10 (m, 12H), 3.87 (m, 8H), 5.98 (m, 4H); 13C NMR (D2O)

Cytotoxic Polyamine Analogues
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 3 415
δ 12.70, 25.00 25.13, 45.10, 45.81, 46.21, 48.44, 48.78, 128.44,
128.85; MS-ESI (m/z) calcd 354.59 (M+ + 1), found 354.60.
Anal. (C20H48Cl5N5) C, H, N.
8H), 3.66 (m, 12H), 5.48 (m, 6H), 6.93 (s, 10H); 13C NMR
(CDCl3) δ 12.51, 12.63, 20.84, 20.86, 22.63, 22.65, 22.84, 24.61,
38.92, 40.40, 41.40, 42.11, 42.18, 44.44, 45.48, 127.95, 128.07,
128.49, 128.62, 128.80, 131.76, 131.83, 131.85, 131.88, 132.01,
138.05, 139.01, 140.07, 140.13, 140.24, 142.15, 142.21, 142.87;
MS-FAB (m/z) calcd 1262.77 (M+), found 1262.76. Anal. (C65
H91N5O10S5) C, H, N.
3,8,13,18,23-P en taaza-(5Z,10Z,15Z)-pen tacosatr ien e pen -
ta h yd r och lor id e (33) was obtained (54%) from 32 following
the procedure described for 15: mp 270 °C dec; 1H NMR (D2O)
δ 1.31 (m, 6H), 1.80 (m, 4H), 3.10 (m, 8H), 3.86 (m, 12H), 5.98
(m, 6H); 13C NMR (D2O) δ 13.30, 13.42, 25.58, 25.70, 45.69,
46.21, 46.43, 46.81, 49.02, 49.37, 129.00, 129.15, 129.37,
129.59; MS-ESI (m/z) calcd 352.58 (M+ + 1), found 352.59.
Anal. (C20H46N5Cl5 ) C, H, N.
Biology. Ma t er ia ls: DuPro cells were obtained from M.
Eileen Dolan of the University of Chicago, Chicago, IL. All
other cell lines used in this study were obtained from the
American Type Culture Collection (Rockville, MD). Tissue
culture medium was obtained from Fisher Scientific (Itasca,
IL) and fetal bovine serum was obtained from Gemini Bio-
products, Inc. (Calabasas, CA). All other reagents were
analytical grade. Deionized double-distilled water was used
in all studies.
1,6,11-Tr is(m e sit yle n e su lfon yl)-1,6,11-t r ia za -(3Z)-
tr id ecen e (26). Mesitylenesulfonamide (1.47 g, 7.38 mmol)
was dissolved in 50 mL anhydrous DMF, and NaH (85%, 0.3
g) was added under nitrogen. The mixture was stirred at 20
°C, 13 (1.40 g, 2.46 mmol) dissolved in 25 mL dry DMF was
added, and the solution was heated at 65 °C during 18 h. The
mixture was then cooled, 10 mL water was added, the solvent
evaporated to dryness, and the solid residue partitioned
between 40 mL water and 40 mL chloroform. The aqueous
layer was extracted with chloroform (2 × 30 mL), the combined
organic layers were washed with water (3 × 50 mL), dried
(MgSO4), and evaporated. The residue was purified by column
chromatography using hexanes-ethyl acetate (7.5:2:5); 1.7 g
1
(97%) of 26 was obtained: mp 50 °C; H NMR (CDCl3) δ 0.94
(t, 3H), 1.30 (m, 4H), 2.29 (s), 2.30 (s), 2.55 (s, 12H), 2.65 (s,
6H), 3.11 (m, 6H), 3.52 (m, 1H), 3.65 (m, 2H), 3.71 (m, 1H),
4.82 (br, 1H), 5.47 (m, 2H), 6.93 (s, 4H), 6.96 (s, 2H); 13C
NMR (CDCl3) δ 12.50, 20.91, 22.71, 22.76, 22.83, 22.91, 24.66,
38.98, 39.85, 42.15, 42.26, 44.50, 128.06, 128.51, 131.86,
131.91, 138.18, 140.00, 140.14, 140.28, 142.17, 142.65. Anal.
(C37H53N3O6S3) C, H, N.
3N,8N,13N,18N,23N-P en ta k is(m esitylen esu lfon yl)-3,8,13,-
18,23-p en ta a za -(10Z,20Z)-p en ta cosa d ien e (27) was pre-
pared (50%) from 26 and 8 as described above for 16: 1H NMR
(CDCl3) δ 0.99 (quint, 6H), 1.33 (m, 8H), 2.29 (s, 15H), 2.55
(s), 2.57 (s), 3.10 (m, 12H), 3.70 (m, 4H), 3.77 (m, 4H), 5.42
(m, 4H), 6.93 (s, 10H); 13C NMR (CDCl3) δ 12.70, 12.71, 20.89,
22.66, 22.72, 22.78, 22.81, 24.60, 26.53, 40.39, 41.37,41.87,
42.20, 45.47, 128.26, 128.62, 131.78, 131.84, 131.86,131.92,
132.77, 138.92, 139.96, 140.09 140.17, 142.57, 142.63. Anal.
(C65H92N5O10S5) C, H, N.
3,8,13,18,23-P en ta a za -(10Z,20Z)-p en ta cosa d ien e p en -
ta h yd r och lor id e (28) was prepared (40%) from 27 following
the procedure described for 15: 1H NMR (D2O) δ 1.28 (t, 3H),
1.30 (t, 3H), 1.70-1.89 (m, 8H), 3.00-3.20 (m, 12H), 3.78-
3.80 (m, 12H), 3.78-3.90 (m, 8H), 5.90-6.10 (m, 4H); 13C NMR
(D2O) δ 13.45, 25.71, 25.76, 26.77, 45.40, 46.20, 46.78, 49.10,
49.40, 129.10, 129.40. Anal. (C20H48Cl5N5) C, H, N.
1,6,11-Tr is(m esit ylen esu lfon yl)-1,6,11-t r ia za -(3Z,8Z)-
tr id eca tr ien e (29) was prepared (94%) from 9 following the
procedure described for 16: 1H NMR (CDCl3) δ 0.97 (t, 3H),
2.30 (s, 9H), 2.53 (s, 6H), 2.58 (s,6H), 2.64 (s, 6H), 3.10 (q, 2H),
3.49 (m, 2H), 3.59 (m, 2H), 3.67 (m, 4H), 5.46 (m, 4H), 6.93 (s,
6H); 13C NMR (CDCl3) δ 12.70, 20.86, 20.89, 22.67, 22.81,
22.88, 39.01, 40.49, 41.51, 42.23, 42.52, 127.71, 127.97, 128.95,
129.18, 131.84, 131.93, 131.96, 131.99, 132.05, 139.08, 140.17,
140.21, 140.30. Anal. (C37H51N3O6S3) C, H, N.
3N,8N,13N,18N,23N-P en ta k is(m esitylen esu lfon yl)-3,8,13,-
18,23-pen taaza-(5Z,10Z,20Z)-pen tacosatr ien e (30) was pre-
pared (82%) by the condensation of 29 and 8 as described for
16: 1H NMR (CDCl3) δ 0.99 (t, 6H), 1.33 (4H), 2.29 (s, 15H),
2.55 (s), 2.57 (s), 3.07 (m, 8H), 3.70 (m, 12H), 5.46 (m, 6H),
6.92 (s, 10H); 13C NMR (CDCl3) δ 12.69, 12.80, 20.84, 22.62,
22.68, 22.73, 22.77, 24.58, 26.55, 40.44, 41.51, 41.86, 42.04,
42.24, 45.49, 128.10, 128.25, 128.52, 128.62, 128.82, 131.89,
131.95, 132.79, 138.89, 140.07, 140.14, 140.23, 141.94, 142.44,
142.53, 142.82; MS-FAB (m/z) calcd 1262.77 (M+), found
1262.78.
Tissu e cu lt u r e:12 Cells were seeded into 75-cm2 culture
flasks with 15 mL of Eagle’s minimal essential medium
supplemented with 10% fetal calf serum and nonessential
amino acids. The flasks were incubated in a humidified 95%
air/5% CO2 atmosphere. The cells were grown for at least 24
h to ensure that they were in the log phase of growth. They
were then treated with the pentamines. Cells were harvested
by treatment for 5 min with STV (saline A, 0.05% trypsin,
0.02% EDTA) at 37 °C. The flasks were rapped on a lab bench,
pipetted several times and aliquots of the cell suspension were
withdrawn and counted using a Coulter particle counter that
was standardized for each cell line using a hemacytometer.
P olya m in e a n a lysis:26 Approximately 1 × 106 cells were
taken from harvested samples and centrifuged at 1000 rpm
at 4 °C for 5 min. The cells were washed twice by resuspending
them in chilled Dulbecco’s isotonic phosphate buffer (pH 7.4)
and centrifuged at 1000 rpm at 4 °C. The supernatant was
decanted and 250 µL of 2% perchloric acid was added to the
cell pellet. The cells were then sonicated and the lysates were
kept at 4 °C for at least 1 h. After centrifugation at 8000g for
5 min, the supernatant was removed for analysis. An ap-
propriate volume of the supernatant (50-100 µL) was fluo-
rescence-labeled by derivatization with dansyl chloride fol-
lowing procedures described elsewhere.26 Each sample was
loaded onto a C-18 high-performance liquid chromatog-
raphy column and separated at the analytical laboratory of
the University of Wisconsin Comprehensive Cancer Center
(UWCCC) using a previously published procedure.26 Peaks
were detected and quantitated using a Shimadzu HPLC
fluorescence monitor that was coupled to a Spectra-Physics
peak integrator. Because polyamine levels vary with environ-
mental conditions, control cultures were sampled for each
experiment.
MTT a ssa y:5 Trypsinized cell suspensions were diluted to
seed a 80-µL suspension of 500 cells into each well of a 96-
well Corning microtiter plate. The plates were incubated
overnight at 37 °C in a humidified 95% air/5% CO2 atmos-
phere. 20 µL of appropriately diluted stock solutions of each
drug was added to the middle 8 columns of the microtiter
plates. Each drug concentration was run in quadruplicate. The
outer columns of the plates were used for buffer controls. Cells
were incubated with the drug for 6 days. 25 µL of a 5 mg/mL
solution of 3-(4,5-dimethythiazol-2-yl)-2,5-diphenyltetrazolium
bromide (MTT) was added to each well and the plates were
incubated at 37 °C for 4 h. Cells were then lysed by incubating
at 37 °C overnight with 100 µL of lysis buffer (500 mL of the
lysis buffer contained: 100 g lauryl sulfate (SDS), 250 mL of
N,N-dimethylformamide in 2 mL of glacial acetic acid, pH 4.8).
The color developed was read at room temperature at 570 nm
in a E-max precision microplate reader (Molecular Devices
3,8,13,18,23-P en taaza-(5Z,10Z,20Z)-pen tacosatr ien e pen -
ta h yd r och lor id e (31) was obtained from 30 (65%) following
the procedure described above for 15: 1H NMR (D2O) δ 1.31
(t, 6H), 1.79 (m, 4H), 3.12 (m, 8H), 3.83 (m, 12H), 5.96 (m,
6H); 13C NMR (D2O) δ 13.48, 25.69, 26.76, 41.67, 45.44, 46.24,
46.45, 46.80, 49.41, 129.00, 129.12, 129.45, 129.71; MS-ESI
(m/z) calcd 352.58 (M+ + 1), found 352.59. Anal. (C20H46Cl5N5)
C, H, N.
3N,8N,13N,18N,23N-P en ta k is(m esitylen esu lfon yl)-3,8,13,-
18,23p en ta za -(5Z,10Z,15Z)-p en ta cosa tr ien e (32) was ob-
tained (89%) by condensation of 29 and 13 following the
procedure described for 16: 1H NMR (CDCl3) δ 0.95 (m, 6H),
1.33 (m, 4H), 2.29 (s, 15H), 2.54 (s), 2.55 (s), 2.57 (s), 3.09 (m,

416 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 3
Reddy et al.
K. Interaction between polyamine analogues with antiprolifera-
tive effects and t-RNA: A 15N NMR study. Cell Mol. Biol. 1994,
40, 933-944. (d) Frydman, L.; Rossomando, P. C.; Frydman, V.;
Fernandez, C. O.; Frydman, B.; Samejima, K. Interaction of
natural polyamine with t-RNA: A 15N NMR study. Proc. Natl.
Acad. Sci U.S.A. 1992, 89, 9186-9190. (e) Frydman, B.; de los
Santos, C.; Frydman, R. B. A 13C NMR study of [5,8-13C2]-
spermidine binding to t-RNA and to Escherichia coli macromol-
ecules. J . Biol. Chem. 1990, 265, 20874-20878.
Corp., Sunnyvale, CA) and data were analyzed using manu-
facturer supplied cell survival software.
CF E a ssa y:12 All of the cell lines that were used in this
assay have previously been optimized with respect to the
necessary incubation times for observable colony formation.
Both floating and attached cells were harvested and centri-
fuged at 1000 rpm for 10 min at 4 °C. The cell pellets were
resuspended and replated in quadruplicate at appropriate
dilution into 60-mm plastic Petri dishes. The Petri dishes were
prepared not more than 24 h in advance with 4 mL of
supplemented Eagle’s minimum essential medium containing
5-10% fetal bovine serum (standardized for each cell line).
Cells were incubated for the previously standardized number
of days in a 95% air/5% CO2 atmosphere. The plates were
stained with 0.125% crystal violet in methanol and counted.
Results are expressed as a surviving fraction of an appropriate
control.
(10) (a) Basu, H. S.; Schweitert, H. C. A.; Feuerstein, B. G.; Marton,
L. J . Effect Of Structural Variations Of Spermine On Its
Interaction With DNA. Biochem. J . 1990, 269, 329-334. (b)
Basu, H. S.; Shafer, R. H.; Marton, L. J . A stopped-flow H-D
exchange kinetic study of spermine-polynucleotide interaction.
Nucleic Acids Res. 1987, 15, 5873-5886. (c) Basu, H. S.;
Feuerstein, B. G.; Zarling, D. A.;. Shafer, R. H.;. Marton, L. J .
Recognition of Z-RNA and Z-DNA determinants by polyamines
in solution: experimental and theoretical studies. J . Biomol.
Struct. Dyn. 1988, 6, 299-309. (d) Delcros, J . G.; Sturkenboom,
M. C.; Basu, H. S.; Shafer, R. H.; Szollosi, J .; Feuerstein, B. G.;
Marton, L. J . Differential effects of spermine and its analogues
on the structures of polynucleotides complexed with ethidium
bromide. Biochem. J . 1993, 291, 269-74. (e) Basu, H. S.; Marton,
L. J . The Interaction Of Spermine And Pentamines With DNA.
Biochem. J . 1987, 244, 243-246. (f) Basu, H. S.; Marton, L. J .;
Pellarin, M.; Deen, D. F.; McManis, J . S.; Liu, C. Z.; Bergeron,
R. J .; Feuerstein, B. G. Design and testing of novel cytotoxic
polyamine analogues. Cancer Res. 1994, 54, 6210-6214.
(11) (a) Bergeron, R. J .; McMannis, J . S.; Liu, Ch. Z.; Feng, Y.;
Weimar, W. R.; Luchetta, G. R.; Wu, Q.; Ortiz-Ocasio, J .; Vinson,
J . R. T.; Kramer, D.; Porter, C. W. Antiproliferative properties
of polyamine analogues: A structure-activity study. J . Med.
Chem. 1994, 37, 3464-3476. (b) Bergeron, R. J .; Feng, Y.;
Weimar, W. R.; McMannis, J . S.; Dimove, H.; Porter, C.; Raisler,
B.; Phanstiel, O. A comparison of structure-activity relation-
ships between spermidine and spermine analogue antineoplas-
tics. J . Med. Chem. 1997, 40, 1475-1494.
Ack n ow led gm en t. We are grateful to Ms. Amy C.
Harms and Mr. J ames F. Brown for their expert help
in securing the mass spectra and to Ms. Kendra Tutsch
(UWCCC) for performing the HPLC analyses. These
assays were performed at the Mass Spectrometry-
Bioanalytical Facility of the Biotechnology Center of the
University of Wisconsin-Madison and the analytical
laboratory of UWCCC.
Refer en ces
(1) Cohen, S. S. A Guide to the Polyamines; Oxford University
Press: Oxford, 1998.
(2) Marton, L. J .; Pegg, A. E. Polyamines as targets for therapeutic
interactions. Annu. Rev. Pharmacol. Toxicol. 1995, 33, 55-91.
(3) Frydman, B.; Valasinas, A. Polyamine based chemotherapy of
cancer. Exp. Opin. Ther. Patents 1999, 9, 1055-1068.
(4) (a) Bergeron, R. J .; Neims, A. H.; McMannis, J . S.; Hawthorne,
T. R.; Vinson, J . R. T.; Bartell, R.; Ingeno, M. J . Synthetic
polyamine analogues as antineoplastics. J . Med. Chem. 1988,
31, 1183-1190. (b) Bernacki, R. J .; Bergeron, R. J .; Porter, C.
W. Antitumor activity of N,N′-bis(ethyl)spermine homologues
against human MALME-3 melanoma xenografts. Cancer Res.
1992, 52, 2424-2430. (c) Porter, C. W.; Bernacki, R. J .; Miller
J .; Bergeron, R. J . Antitumor activity of N1,N11-bis(ethyl)-
norspermine against human melanoma xenografts and possible
biochemical correlates of drug action. Cancer Res. 1993, 53, 581-
586. (d) Chang, B. K.; Bergeron, R. J .; Porter, C. W.; Vinson, J .
R. T.; Liang, Y.; Libby, P. R. Regulatory and antiproliferative
effects of N-alkylated polyamine analogues in human and
hamster pancreatic adenocarcinoma cell lines. Cancer Chemo-
ther. Pharmacol. 1992, 30, 183-188. (e) Basu, H. S.; Pellarin,
M.; Feuerstein, B. G.; Deen, D. F.; Marton, L. J . Effects of
polyamine analogues BE-3-7-3, 3-8-3, and BE-3-8-3 on human
brain tumor cell growth and survival. Anti-cancer Res. 1993, 13,
1525-1532. (f) Bergeron, R. J .; Miller, R.; Bussenius, J .; Mc-
Manis, J . S.; Merriman, R. L.; Smith, R. E.; Yao, H.; Weimar,
W. R. Synthesis and evaluation of hydroxylated polyamine
analogues as antiproliferatives. J . Med. Chem. 2000, 43, 224-
235.
(5) Reddy, V. K.; Valasinas, A.; Sarkar, A.; Basu, H. S.; Marton, L.
J .; Frydman, B. Conformationally restricted analogues of 1N,12N-
bisethylspermine: Synthesis and growth inhibitory effects on
human tumor cell lines. J . Med. Chem. 1998, 41, 4723-4732
and references therein.
(6) Perkins, T. T.; Smith, D. E.; Chu, S. Direct observation of tube-
like motion of a single polymer chain. Science 1994, 264, 819-
822.
(7) Feuerstein, B. G.; Williams, L. D.; Basu, H. S.; and Marton, L.
J . Implications and concepts of polyamine-nucleic acid interac-
tions. J . Cell. Biochem. 1991, 46, 37-47.
(8) Marquet, R.; Houssier C. Different binding modes of spermine
to A-T and G-C base pairs modulate the bending and stiffening
of the DNA double helix. J . Biomol. Struct. Dyn. 1988, 6, 235-
246.
(12) Basu, H. S.; Pellarin, M.; Feuerstein, B. G.; Shirahata, A.;
Samejima, K.; Deen, D. F.; Marton, L. J . Interaction of a
polyamine analogue, 1,19-bis-(ethylamino)-5,10,15-triazanona-
decane (BE-4-4-4-4), with DNA and effect on growth, survival,
and polyamine levels in seven human brain tumor cell lines.
Cancer Res. 1993, 53, 3948-3955.
(13) Dolan, M. E.; Fleig, M. J .; Feuerstein, B. G.; Basu, H. S.; Luk,
G. D.; Casero, R. A., J r.; Marton, L. J . Effect of 1,19-bis-
(ethylamino)-5,10,15-triazanonadecane on human tumor xe-
nografts. Cancer Res. 1994, 54, 4698-4702.
(14) Basu, H. S.; Smirnov, I. V.;. Peng, H. F.; Tiffany, K.; J ackson,
V. Effects of spermine and its cytotoxic analogues on nucleosome
formation on topologically stressed DNA in vitro. Eur. J .
Biochem. 1997, 243, 247-258.
(15) Basu, H. S.; Dreckschmidt, N.; Tu, L.; Chanbusarkam, L.
Polyamine Analogue Bis(ethylamino)-5,10,15-triazanonadecane
(BE-4-4-4-4) Enhances Simian Virus-40 Late Gene Expression.
Cancer Chemother. Pharmacol. 1998, 43, 336-340.
(16) (a) Feuerstein, B. G.; Pattabiraman, N.; Marton, L. J . Molecular
mechanics of the interactions of spermine with DNA: DNA
bending as a result of ligand binding. Nucleic Acids Res. 1990,
18, 1271-1282. (b) Feuerstein, B. G.; Pattabiraman, N.; Marton,
L. J . Molecular dynamics of spermine-DNA interactions: se-
quence specificity and DNA bending for a simple ligand. Nucleic
Acids Res. 1989, 17, 6883-6892. (c) Feuerstein, B. G.; Pattabi-
raman, N; Marton, L. J . Spermine-DNA interactions: a theoreti-
cal study. Proc. Natl. Acad. Sci. U.S.A. 1986, 83, 5948-5952.
(17) Bellucci, G.; Berti, G.; Chiappe, C.; Lippi, A.; Marioni, F. The
metabolism of carbamazepine in Humans: Steric Course of the
Enzymatic Hydrolysis of 10,11-Epoxide. J . Med. Chem. 1987,
30, 768-773.
(18) J effers, L.; Church, D.; Basu, H.; Marton, L.; Wilding, G. Effects
of the polyamine analogues BE-4-4-4-4, BE-3-7-3, and BE-3-3-3
on the proliferation of three prostate cancer cell lines. Cancer
Chemother. Pharmacol. 1997, 40, 172-179.
(19) Mi, Z.; Kramer, D. L.; Miller, J . T.; Bergeron, R. J .; Bernacki,
R.; Porter, C. W. Human prostatic carcinoma cell lines display
altered regulation of polyamine transport in response to polyamine
analogues and inhibitors. Prostate 1998, 34, 51-60.
(20) Delcros, J . G.; Sturkenboom, M. C.; Basu, H. S.;. Shafer, R. H.;
Szollosi, J .; Feuerstein, B. G.; Marton, L. J . Differential effects
of spermine and its analogues on the structures of polynucle-
otides complexed with ethidium bromide. Biochem. J . 1993, 291,
269-274.
(21) Basu, H. S.; Marton, L. J . Biological and therapeutic implications
of effects of polyamines on chromatin condensation. In Poly-
amines: Regulations and Molecular Interactions; Casero, R. C.
A., J r., Ed.; RG Landes Co.: Austin, TX, 1995; pp 101-128.
(9) (a) Frydman, B.; Westler W. M.; Valasinas, A.; Kramer, D. L.;
Porter, C. W. Regioslective binding of spermine 1N,12N-bis-
(methyl)spermine and 1N,12N-bsi(ethyl)spermine to t-RNAPhe as
revealed by 750 MHz 1H NMR and its possible correlation with
cancer cell growth and cell cycles. J . Braz. Chem. Soc. 1999, 10,
334-340. (b) Frydman, B.; Westler, W. M.; Samejima, K.
Spermine binds in solution to the TΨC loop of t-RNAPhe
:
Evidence from a 750 MHz 1H NMR analysis. J . Org. Chem. 1996,
61, 2588-2589. (c) Fernandez, C. O.; Frydman B,; Samejima,

Cytotoxic Polyamine Analogues
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 3 417
(22) Eiseman, J . L.; Rogers, F. A.; Guo, Y.; Kauffman, J .; Sentz, D.
L.; Klinger, M. F.; Callery, P. S.; Kyprianou, N. Tumor-targeted
apoptosis by a novel spermine analogue, 1,12-diaziridinyl-4,9-
diazadodecane, results in therapeutic efficacy and enhanced
radiosensitivity of human prostate cancer. Cancer Res. 1998, 58,
4864-4870.
(23) Fredlund, J . O.; J ohansson, M.; Baldetorp, B.; Oredsson, S. M.
Abnormal DNA synthesis in polyamine deficient cells revealed
by bromodeoxyuridine-flow cytometry technique. Cell Prolif.
1994, 27, 243-56.
(24) Pollard, K. J .; Samuels, M. L.; Crowley, K. A.; Hansen, J . C.;
Peterson, C. L. Functional interaction between GCN5 and
polyamines: a new role for core histone acetylation. EMBO J .
1999, 18, 5622-33.
(25) Wright, W. B., J r.; Press, J . B.; Chan, P. S.; Marisco, J . W.; Haug,
M. F.; Lucas, J .; Tauber, J .; Tomcufcik, A. S. Thromboxane
synthetase inhibitors and antihypertensive agents I: 1N-[(1H-
Imidazol-1-yl)alkyl]aryl amides and N-[(1H-1,2,4-triazol-1-yl)-
alkyl]aryl Amides. J . Med. Chem. 1986, 29, 523-530.
(26) Kabra, P. M.; Lee, H. K.; Lubich, W. P.; Marton, L. J . Solid-
phase extraction and determination of dansyl derivatives of
unconjugated and acetylated polyamines by reversed-phased
liquid chromatography. J . Chromatogr. Biomed. Appl. 1986, 380,
19-32.
J M000310S