A comparison of chloroambucil- and xylene-containing poly amines leads to improved ligands for accessing the polyamine transport system

Article abstract of DOI:10.1021/jm070794t

Several disubstituted arylene- and chloroambucil-polyamine conjugates were synthesized and evaluated for their ability to target cells via their polyamine transport system (PAT). As compared to the monosubstituted analogues, the disubstituted arylene systems were superior PAT targeting agents. Using a Chinese hamster ovary (CHO) cell line (PAT active) and its CHO-MG mutant (PAT inactive), the series was screened for their PAT targeting ability. The data were expressed as a CHOMG/CHO IC₅₀ ratio. Indeed, the disubstituted systems gave high IC₅₀ ratios (e.g., ratio > 2000), which indicated high selectivity for the PAT. The chloroambucil adducts were less toxic than the corresponding arylmethyl compounds. In this regard, having the proper recognition element (i.e., homospermidine) and cytotoxic "cargo" were deemed paramount for successful drug delivery via the PAT.

Full text of DOI:10.1021/jm070794t

J. Med. Chem. 2008, 51, 1393–1401
1393
A Comparison of Chloroambucil- and Xylene-Containing Polyamines Leads to Improved
Ligands for Accessing the Polyamine Transport System
Navneet Kaur,† Jean-Guy Delcros,‡ Jon Imran,† Annette Khaled,§ Mounir Chehtane,§ Nuska Tschammer,§
Bénédicte Martin,‡ and Otto Phanstiel IV*,†,#,§
Department of Chemistry, P.O. Box 162366, UniVersity of Central Florida, Orlando, Florida 32816-2366, Biomolecular Science Center,
UniVersity of Central Florida, 12722 Research Parkway, Orlando, Florida 32825, Groupe Cycle Cellulaire, CNRS UMR 6061 Génétique et
DéVeloppement, IFR 97 Génomique Fonctionnelle et Santé, Faculté de Médecine, UniVersité Rennes 1, 2 AV. du Pr Leon Bernard, CS 34317,
F-35043 Rennes Cédex, France, and Department of Medical Education, College of Medicine, UniVersity of Central Florida, 12201 Research
Parkway, Orlando, Florida 32816-2201
ReceiVed July 3, 2007
Several disubstituted arylene- and chloroambucil-polyamine conjugates were synthesized and evaluated for
their ability to target cells via their polyamine transport system (PAT). As compared to the monosubstituted
analogues, the disubstituted arylene systems were superior PAT targeting agents. Using a Chinese hamster
ovary (CHO) cell line (PAT active) and its CHO-MG mutant (PAT inactive), the series was screened for
their PAT targeting ability. The data were expressed as a CHOMG/CHO IC50 ratio. Indeed, the disubstituted
systems gave high IC50 ratios (e.g., ratio > 2000), which indicated high selectivity for the PAT. The
chloroambucil adducts were less toxic than the corresponding arylmethyl compounds. In this regard, having
the proper recognition element (i.e., homospermidine) and cytotoxic “cargo” were deemed paramount for
successful drug delivery via the PAT.
Introduction
Prior work in our laboratories and others has described the
structural requirements associated with the delivery of polyamine
conjugates via the PAT in mouse leukemia cells (L1210) and
CHO cells.1,4–13 In short, the number of methylene “spacer”
units, the size of the N1 substituent, and the degree of N1
substitution all influence PAT-mediated delivery.6,8–10 The linear
triamine motifs were identified as excellent vector systems for
the PAT.6,8 In particular, the homospermidine conjugate 1a (e.g.,
a 4,4-triamine, Figure 1) had 150-fold higher cytotoxicity in
CHO cells than in the mutant cell line, CHO-MGa, which was
PAT-deficient.6,8,14,15 In addition, a direct correlation was found
between the cytotoxicity and the ability of the polyamine
conjugate to use the PAT for cellular entry. This strategy was
further illustrated, when 1a was shown to be 10–30 times more
selective in killing B16 melanoma cells over Mel A (normal
melanocyte) cells.10 Therefore, the 4,4-triamine motif (homo-
spermidine) was found to impart excellent PAT selectivity in
this earlier anthracenylmethyl-polyamine series. We speculated
that if homospermidine was an optimal motif for cellular entry,
then architectures containing two or more of these recognition
elements may have enhanced PAT selectivity and biological
properties. Having that in mind, we synthesized the disubstituted
compounds 4-6 and the tris-homospermidine adduct, 7.
Beyond the multimeric forms 4–7, we also investigated the
effect of appending a different “cytotoxic cargo” to the
homospermidine motif. The chlorambucil (CA) architecture was
an attractive target for several reasons. First, chlorambucil is a
chemotherapeutic, which is given in the treatment of chronic
lymphocytic leukemia, low-grade non-Hodgkin’s lymphoma,
Hodgkin’s lymphoma, and ovarian cancer.16 Second, it is taken
up by various tumor cells by simple diffusion and exerts its
cytotoxic effects by interaction with DNA.16 Third, Cohen et
al. have reported good results in targeting tumor cells with the
chlorambucil-spermidine conjugate 8.17 Indeed, 8 showed
10000-fold more activity than chlorambucil at forming inter-
strand cross-links with naked DNA.17,18 The increased toxicity
of the conjugate may be due to the enhanced DNA binding and/
Polyamines are essential growth factors for cells. All cells
have methods of manufacturing polyamines from amino acid
sources. In addition, cells can import polyamines from outside
the cell via a process referred to as the polyamine transporter
(PATa).1 While much is known about polyamine transport in
bacteria,2 yeast,2 and leishmania,3 the mammalian polyamine
transport system is a measureable, yet poorly described, import
process. It is an important cancer target because many cancer
cells are unable to produce enough polyamines to sustain their
growth rate and rely upon polyamine import to grow.1,4–10
Therefore, an opportunity exists to selectively target rapidly
dividing cancer cells via their heavy reliance on polyamine
import. Drug-polyamine conjugates, which covalently attach a
polyamine to a known cytotoxic agent, have been shown to have
enhanced cytotoxicity to cancer cells (e.g., mouse melanoma,
B-16 cells) over their normal cell counterparts (Mel-A) in
vitro.1,4–10
In this regard, numerous N-alkyl-polyamines have been
synthesized with promising anticancer activity.11 The molecular
recognition events involved in polyamine transport have long
been known to be sensitive to the distance separating the
nitrogen centers as well as to the number of nitrogens present
in the molecule.1,4–11 However, recent systematic studies of N1-
(anthracenylmethyl)polyamines suggested that there are limits
to the structural tolerance accommodated by the PAT.4–8
* To whom correspondence should be addressed. Tel: 407-823-5410.
Fax: 407-823-2252. E-mail: ophansti@mail.ucf.edu.
†
Department of Chemistry, University of Central Florida.
Department of Medical Education, University of Central Florida.
Groupe Cycle Cellulaire, Université Rennes 1.
Biomedical Science Center, University of Central Florida.
#
‡
§
a
PAT, polyamine transporter; CHO, Chinese hamster ovary cells; CHO-
MG, Chinese hamster ovary cells polyamine transport deficient mutant;
L1210, mouse leukemia cells; HBBS, Hanks balanced salt solution; PBS,
Phosphate buffered saline.
10.1021/jm070794t CCC: $40.75
2008 American Chemical Society
Published on Web 02/19/2008

1394 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 5
Kaur et al.
Figure 1. Structures of compounds 1–13.
Scheme 1a
This report investigates both the role of the polyamine
recognition element and the nature of the appended cytotoxic
agent. We found that the disubstituted (polyamine-arylene-
polyamine) motifs are excellent PAT targeting agents and have
the highest reported CHOMG/CHO IC50 ratios (>2000).
Moreover, the chloroambucil adducts were much less toxic than
the corresponding anthracenylmethyl compounds, (e.g. 1a). In
this regard, having the proper recognition element (homosper-
midine) and cytotoxic cargo (e.g., xylene) were deemed
paramount for successful drug delivery via the PAT.
Results and Discussion
Synthesis. New conjugates 4–7 were synthesized to probe
the effect of having two or more PAT recognition elements (i.e.,
homospermidine) within the same molecule. Compound 10 and
its controls 11–13 were also synthesized to probe whether
homospermidine could enhance the PAT-mediated delivery of
the chlorambucil scaffold.
For the synthesis of compounds 4–6, the diBoc-protected
amine 14 was synthesized as reported earlier.19 As shown in
Scheme 1, reductive amination of the respective dialdehydes
15, 16, and 17 with amine 14 gave compounds 18, 19, and 20,
respectively. Boc removal with 4 N HCl provided the target
conjugates, 4–6.
Compound 7 was synthesized from the commercially avail-
able benzene-1,3,5-tricarboxylic acid trimethyl ester 21 (Scheme
2), which was reduced to alcohol 22 using LiAlH4 (78% yield).
Further oxidation of alcohol 22 using PCC resulted in trialde-
hyde 23 (51%). Reductive amination of trialdehyde 23 with
amine 14 gave the secondary amine 24. However, the isolation
of 24 was problematic due to the coelution of 14. Therefore,
a Reagents: (a) 25% MeOH/CH2Cl2. (b) 50% MeOH/CH2Cl2, NaBH4.
(c) 4 N HCl/EtOH.
or facilitated uptake via the polyamine transport system (PAT).
In compound 8, chloroambucil is conjugated to spermidine at
the N4 position of spermidine (Figure 1). However, we recently
demonstrated that the N5-alkylated analogue 9 resulted in a
complete loss of PAT targeting/selectivity as compared to the
N1-substituted parent 1a.10 We speculated that the bioactivity
and delivery profile of chlorambucil could be enhanced by
introducing this DNA cross-linking motif at the N1-position of
homospermidine. This insight led us to synthesize compounds
10–13 to evaluate this strategy.

Ligands for Accessing the Polyamine Transport System
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 5 1395
Scheme 2a
a Reagents: (a) LiAlH4/THF. (b) PCC/CH2Cl2. (c) 25% MeOH/CH2Cl2; 14. (d) 50% MeOH/CH2Cl2/NaBH4. (e) 4 N HCl/EtOH.
Scheme 3a
a Reagents: (a) 25% MeOH/CH2Cl2. (b) 50% MeOH/CH2Cl2, NaBH4. (c) 4 N HCl/EtOH.
Scheme 4a
a Reagents: (a) 25% MeOH/CH2Cl2. (b) 50% MeOH/CH2Cl2, NaBH4. (c) 4 N HCl/EtOH.
Table 1. L1210: Ki and Cytotoxicity Studies
the mixture of 14 and 24 was reacted with BOC-ON to
L1210 IC50
L1210 +
DFMO IC50 (µM)
L1210 + SPD
IC50 (µM)
selectively protect the primary amine 14, which facilitated the
purification of compound 24. Once isolated, the Boc groups of
24 were removed using 4 N HCl to obtain compound 7, albeit
in low overall yield.
As shown in Scheme 3, compounds 10–12 were synthesized
by reductive amination of aldehyde 25 with their respective di-
Boc amines 14, 26, and 27 to give the secondary amines 28,
29, and 30, respectively. Removal of the Boc group by 4 N
HCl resulted in the compounds 10-12, respectively. Similarly,
reductive amination of 25 with putrescine analogue 31 gave
the secondary amine 32, which in turn gave compound 13
(Scheme 4).
Biological Evaluation. Once synthesized, the conjugates (4–7
and 10–13) were screened for cytotoxicity in L1210, CHO, and
CHO-MG cells. L1210 cells were selected to enable compari-
sons with the published IC50 and Ki values for a variety of related
polyamine substrates.4–8 Chinese hamster ovary (CHO) cells
were chosen along with a mutant cell line (CHO-MG) to
compd
Ki (µM)
(µM)
1a
1b
2
3
4
5
6
7
10
11
12
13
CA
1.8 ( 0.1
32.2 ( 4.3
3.8 ( 0.5
4.5 ( 0.8
0.39 ( 0.05 0.78 ( 0.07
0.17 ( 0.02 0.25 ( 0.08
0.52 ( 0.11 0.16 ( 0.01
0.49 ( 0.02 122.3 ( 8.1
5.97 ( 0.98 35.4 ( 1.6
12.4 ( 1.2
2.66 ( 0.17 6.53 ( 0.42
75.0 ( 4.2
>>150
0.30 ( 0.04
6.30 ( 0.26
0.50 ( 0.03
36.3 ( 8.4
0.09 ( 0.01
9.78 ( 0.42
0.43 ( 0.02
ND
0.22 ( 0.08
ND
0.36 ( 0.03
49.3 ( 10.0
ND
ND
ND
0.83 ( 0.09
6.96 ( 0.73a
ND
ND
5.52 ( 0.09
6.42 ( 0.80
6.89 ( 0.42
102.8 ( 10.9
31.1 ( 2.0
8.13 ( 0.47
7.83 ( 0.35
2.59 ( 0.13
ND
8.14 ( 0.46
2.18 ( 0.15
14.1 ( 0.5
ND
ND
a A 100 µM concentration of SPD was used.
comment on selective transport via the PAT.6–8,12 The results
are shown in Tables 1 and 2.
L1210 Ki and IC50 Studies. The Ki values in Table 1 were
determined for [14C]spermidine uptake and reflect the affinity

1396 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 5
Kaur et al.
Table 2. L1210 Cytotoxicity of 1a in µM in the Absence or Presence
of 7
In general, the disubstituted derivatives showed higher cyto-
toxicity in L1210 cells than either the trisubstituted analogue 7
or their monosubstituted derivatives, except in the case of 1a.
IC50 in the
IC50 in the presence of different
absence of 7
concentrations of 7
The chlorambucil derivatives 10–13 were less toxic than 1a
and gave relatively higher L1210 IC50 values (Table 1). For
example, although both the homospermidine analogue 10 (IC50
) 35.4 µM) and 1a contained the same polyamine “message”,
10 was significantly less toxic than the corresponding anthryl
analogue 1a (IC50 ) 0.30 µM). Compounds 11 and 12 were
designed to probe how the orientation of the spermidine motif
affected bioactivity. While the Ki value of 12 was less than 11,
the observed IC50 values were similar (8.1 and 6.5 µM,
respectively). The most toxic analogue of the CA series was
derivative 13, which contained an appended butanediamine motif
and was later shown to not use the polyamine transport system
for cellular entry.
1a
1a + 7(5 µM)
1.27 ( 0.06
1a + 7(10 µM)
1.06 ( 0.09
1a + 7(20 µM)
1.15 ( 0.08
0.31 ( 0.02
of the polyamine derivative for the polyamine transport system
on the cell surface. The IC50 values listed in Table 1 represent
the concentration of the polyamine conjugate required to reduce
the relative cell growth by 50%. With both parameters, one can
determine whether high affinity for the transporter (e.g., low Ki
value) translated into high cytotoxicity (e.g., low IC50 value).
A priori, one may have expected that conjugates with high
PAT affinity would be transported efficiently into the cell by
this transporter. However, prior results10 showed that the Ki
values were of minimal value in predicting the cytotoxicity and
transport of these systems. Indeed, tetraamine PAT substrates
with very low Ki values (<<1 µM) have been shown to actually
inhibit their own import.7 In the series of anthracene-triamine
conjugates studied, substrates with moderate Ki values (∼2 µM)
were efficiently transported into cells via PAT.6 In summary,
the Ki values provided insight as to how alterations of polyamine
structure influence the affinity of the polyamine conjugate for
the PAT system but were often unreliable predictors of PAT
selectivity.
Using competition experiments with radiolabeled spermidine,
we were able to rank the binding affinity of the compounds for
the PAT. As shown in Table 1, the monosubstituted adduct 1a
had a Ki value of 1.8 µM, whereas the Ki value was lowered to
0.39 µM for the disubstituted moiety 4. A similar trend was
observed in case of the naphthyl (5) and benzyl (6) derivatives.
Inspection of compounds 1a, 2, and 3 shows the trend that as
the size of the cargo decreases, the Ki value increases (anthra-
cenyl > napthyl > benzyl: 1.8 < 3.8 < 4.5), indicating a lower
affinity for PAT. Surprisingly, the same trend was not observed
in the case of the disubstituted series (4–6). Naphthyl derivative
5 showed a lower Ki value than the anthracenyl (4) and benzyl
(6) derivatives. Even the trisubstituted benzyl derivative 7 gave
a lower Ki value than the monosubstituted derivatives, suggest-
ing that the di- and trisubstituted analogues 4-6 and 7 all had
significantly lower Ki values (than the triamines 1a, 2, and 3,
respectively) and were better PAT binding agents.
In the case of the chlorambucil derivatives 10–13, high Ki
values were observed, suggesting that these compounds have
lower binding affinity for PAT.
An interesting trend was apparent in the L1210 IC50 values
of the compounds 1a, 2, and 3 (Table 1). The respective IC50
value increased as the size of the cargo is decreased, for
example, anthracenyl > naphthyl > benzyl: 0.30 < 0.50 < 36.3
µM. Surprisingly, the opposite trend was observed in compounds
4–6, that is, anthracenyl > naphthyl > benzyl: 0.78 > 0.25 >
0.16 µM.
On the other hand, when each monosubstituted derivative (1a,
2, and 3) was compared with its corresponding disubstituted
derivative (4–6), no general trend was observed. The IC50
increases 2.6 times (0.30 vs 0.78 µM) in the case of the
anthracene derivatives (1a vs 4). However, in the case of the
naphthyl derivatives (2 vs 5), the IC50 was reduced in half (0.50
vs 0.25 µM). A significant difference was observed in the case
of benzyl derivatives (3 vs 6: 36.3 vs 0.16 µM) and was
previously rationalized by partial metabolism of 3 to free
homospermidine.8 In contrast, a significantly higher IC50 value
was observed in the case of trisubstituted benzyl derivative 7.
Cytotoxicity experiments were also performed in murine
leukemia cells (L1210, Table 1) with and without the presence
of difluoromethylornithine (DFMO), a known inhibitor of
ornithine decarboxylase (ODC), the enzyme responsible for
polyamine biosynthesis. Typically, when the polyamine bio-
synthetic pathway is blocked, cells respond by increasing their
import of extracellular polyamines via PAT. Therefore, if the
conjugates are PAT-selective, one should see a lowering of their
respective IC50 values in the presence of DFMO, a molecule
that facilitates the conjugate’s import. Indeed, as shown in Table
1, the IC50 values are all lower in the presence of DFMO, with
the exception of 1b and 6. These observations are consistent
with previous studies, wherein we observed an approximate
halving of the IC50 value when the experiment is conducted in
the presence of DFMO. This indicated that the PAT-selective
drug was more potent in the presence of the ODC inhibitor,
presumably due to its increased cell uptake.
Spermidine rescue experiments were also performed to test
whether these drugs were using the PAT for cellular entry.
Because spermidine (SPD, 200 µM) is a natural PAT substrate,
it should outcompete the PAT-selective drug for cellular entry
and “rescue the cell” from the cytotoxic drug. Indeed, exogenous
spermidine is an antagonist for drug uptake and provided a
significant “import-inhibition” effect with lower amounts of drug
entering the cells after 24 h of incubation. This was observed
as a general shifting of the cytotoxicity curve (i.e., a plot of %
relative viability vs [drug] to the right and a corresponding
increased IC50 value of the drug in the presence of spermidine).
In general, PAT-selective drugs like 1a resulted in a significant
increase in IC50 value in the presence of SPD. Indeed, in the
presence of SPD, the disubstituted xylene analogues 4–6 gave
IC50 values 7, 26, and 43 times higher than the SPD-free assays
(Table 1), respectively. This result indicated that they were
excellent PAT ligands. This spermidine protection effect has
been observed previously4,5,12 and can be rationalized here by
spermidine’s competitive access to cells via the PAT.
In contrast, virtually no SPD protection was observed with
1b and the chlorambucil derivatives 10–13, suggesting that these
drugs did not use the PAT for cellular entry. In this regard, the
IC50 values remained virtually unchanged even in the presence
of excess SPD (200 µM).
CHO and CHO-MG Studies. CHO cells were chosen along
with a mutant cell line (CHO-MG) to comment on how the
synthetic conjugates gain access to cells.6–8,12 The CHO-MG
cell line was polyamine transport-deficient and was isolated after
selection for growth resistance to methylglyoxalbis(guanylhy-
drazone), MGBG, (CH3C[dN-NHC(dNH)NH2]CH[dN-

Ligands for Accessing the Polyamine Transport System
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 5 1397
Table 3. CHO Cytotoxicity Studies
NHC(dNH)NH2]) using a single-step selection after mutagen-
esis with ethylmethanesulfonate.14,15
compd CHOMG IC50 (µM) CHO IC50(µM) CHOMG/CHO IC50 ratio
For the purposes of this study, the CHO-MG cell line
represents cells with no PAT activity and provided a model for
alternative modes of entry or action, which are independent of
PAT. These alternative modes of entry include passive diffusion
or utilization of another transporter. The alternative modes of
action may also include interactions on the outer surface of the
plasma membrane or other membrane receptor interactions.
1a
1b
2
3
4
5
6
7
10
11
12
13
CA
66.7 ( 4.1
7.6 ( 0.4
>100
0.45 ( 0.10
7.7 ( 0.5
0.6 ( 0.2
>1000
148
1
>164
NAa
>2222
>833
677
NAa
1
1
1
0.6
1.32
>1000
>100
>100
50.2 ( 3.8
>500
>>100
>100
>100
17.1 ( 1.5
111.2 ( 4.6
0.045 ( 0.003
0.12 ( 0.03
0.074 ( 0.010
>500
>>100
In contrast, the parent CHO cell line represents a cell type
with high PAT activity14,15 and should be very vulnerable to
PAT targeting drugs. Therefore, highly selective PAT ligands
should give high (CHO-MG/CHO) IC50 ratios.
>100
>100
28.5 ( 1.0
84.5 ( 11.9
As reported earlier,10 dramatic differences in CHO and
CHO-MG cytotoxicity (Table 2) were observed with 1a and
2 (e.g., CHOMG/CHO IC50 ratio of 1a: 148). The benzyl
analogue 3 was shown to be partially metabolized into free
homospermidine. As such, 3 had an IC50 value >1000 µM
in both CHO and CHO-MG cell lines.8
a NA, not applicable.
Table 4. IL-3-Dependent FL5.12A B Cells: Apoptosis via Annexin-V
Staining (% Apoptosis)a
%
IL-3 (ng/mL)
UNTR
AG
1a
1b
4
The disubstituted compounds (anthryl 4, naphthyl 5, and
benzyl 6) gave the highest CHO-MG/CHO IC50 ratios ever
published and were >2222, >833, and 677, respectively. This
unprecedented PAT selectivity confirmed that these disubstituted
derivatives have extremely high selectivity toward PAT-active
CHO cells. Moreover, there was a direct correlation between
the derivative’s PAT selectivity and the size of the arylene
spacer group, wherein the larger anthracenyl ring system in 4
had the highest PAT selectivity.
2
0.1
0
3.38
3.71
44.71
5.17
3.72
79.69
67.71
80.41
>95
>95
>95
>95
5.12
6.99
16.39
a UNTR (untreated); AG (aminoguanidine). Representative data for the
apoptotic effect of compounds 1a, 1b, and 4 on FL5.12A cells; experiments
were done in triplicate with a margin error of 2%.
at neutral pH to produce aziridines.21 To investigate this issue,
an NMR experiment was conducted in a phosphate buffer at
pD 7.4 in D2O. After 48 h of incubation at 37 °C, the
chlorambucil conjugate 10 was 89% degraded by the solvent,
which explains the low cytotoxicity of the chlorambucil
derivatives. The parent chlorambucil (CA) had a CHO IC50
111.2 µM and a CHO-MG/CHO ratio of 1.32, which also
suggests non-PAT-mediated import.
Unfortunately, the selectivity of the trisubstituted compound
7 was not determined due to its low cytotoxicity in CHO and
CHO-MG cells (e.g., CHO IC50 >500 µM). However, it may
have a high affinity for binding to PAT, especially in light of
its low L1210 Ki value (which was similar to 6) and the expected
halving of the L1210 IC50 value upon DFMO treatment (Table
1). These observations coupled with the fact that spermidine
was unable to rescue L1210 cells from 7 suggests that it may
be binding to PAT but not using PAT for its cellular entry. It
is reported in the literature that multisubstituted polyamines are
PAT inhibitors.20 Furthermore, to see the PAT inhibitor effect
of 7, another experiment was conducted. L1210 cells were dosed
with 1a in the presence of 7 with the concept that if 7 acts as
an inhibitor of PAT, it would rescue the cells from 1a. Indeed,
significant rescue was observed in the presence of 7 with a 4-fold
increase in IC50 values [IC50 1a, 0.31 ( 0.02 µM; IC50 1a + 7
(5 µM), 1.27 ( 0.06 µM]. The L1210 cytotoxicity curves of
compound 1a were determined with different concentrations of
7. As shown in Table 2, significant rescue was observed at all
concentrations of 7 tested (5, 10, and 20 µM). These are
statistically significant results. Both the PAT-binding affinity
of 7 (low Ki value) and its ability to protect the cells from 1a
make 7 a relatively nontoxic inhibitor of the PAT. Indeed, a
nontoxic PAT inhibitor may find use in future biochemical
studies of polyamine transport.
The chlorambucil derivatives 10–12 were nontoxic to both
CHO and CHO-MG cells (IC50 > 100 µM), and their respective
CHO-MG/CHO IC50 ratios were not determined. However, the
fact that spermidine was unable to rescue L1210 cells from these
drugs suggest that they do not use the PAT for cellular entry.
Indeed, 13, which was the only toxic member of the chloram-
bucil series, had a CHO-MG/CHO ratio of 0.6, confirming that
it was not PAT-selective (Table 3). The low toxicity associated
with the chlorambucil derivatives may be due to the instability
of the N-(2-chloroethyl)amino group at physiological pH. It is
reported in the literature that N-(2-chloroethyl)amines cyclize
)
IL-3-Dependent Pro-B Cell Line Studies. Having estab-
lished that the disubstituted compound 4 was extremely PAT-
selective, we evaluated it along with 1a and putrescine analogue
1b in a cytokine-dependent B cell line. It is known that
lymphomas frequently result from molecular abnormalities in
the pathways that regulate the proliferation and apoptosis of
lymphocytes. The function of polyamine transport in this process
is poorly understood. Because polyamines are essential com-
pounds for B cell growth and B cells are known to express
PATs,22 we examined the effect of compounds 1a, 1b, and 4
on an IL-3-dependent pro-B cell line, FL5.12A.23 FL5.12A B
cells were cultured for 48 h in three different concentrations of
IL-3: 2 ng/mL, which induces cell cycle progression and inhibits
apoptosis; 0.1 ng/mL, which causes cell cycle arrest but prevents
apoptosis; and 0 ng/mL (absence of IL-3), which causes cell
cycle arrest and induces apoptosis. During the last 24 h,
compound 1a, 1b, or 4 (10 µM) was added in the presence of
aminoguanidine (1 mM). Apoptosis was assessed by measuring
the binding of annexin-V-FITC to phosphatidylserine exposed
on the plasma membrane of apoptotic cells. The initial results
are shown in Table 4 and demonstrate that compound 1b
nonspecifically induced apoptosis. Even though compounds 1a
and 4 were PAT-selective, they have different abilities to induce
apoptosis. Compound 1a induced apoptosis in parallel with
decreasing amounts of IL-3, while compound 4 did not induce
apoptosis.
To examine specific polyamine transport, we performed a
spermidine rescue experiment and observed that the apoptotic-
inducing abilities of compounds 1a and 1b were reduced in the
presence of spermidine. The fact that this effect was evident
with compound 1b, which was lethal at this concentration, might

1398 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 5
Kaur et al.
Table 5. IL-3-Dependent FL5.12A B Cells: Spermidine Competition
Assay via Annexin-V Staining (% Apoptosis)a
%
IL-3 (ng/mL) UNTR SPD
1a
1a + SPD 1b 1b + SPD
2
0.1
0
5.33
4.42
5.52
6.07
67.71
80.41
37.30
33.87
46.20
>95
>95
>95
47.28
35.59
35.32
55.89 58.16 >95
a UNTR (untreated); SPD (spermidine). Representative data for the
apoptotic effect of compounds 1a and 1b on FL5.12A cells in spermidine
competition assay; experiments were done in duplicate with a margin error
of 5%.
Figure 2. Structured of antizyme inhibitors.
system in L1210 murine leukemia, two CHO cell lines (via
CHO-MG/CHO IC50 ratios), and a cytokine-dependent pro-B
cell line. Both Ki and IC50 values were determined in L1210
cells for comparison to previous published polyamine analogues.
The disubstituted compounds 4-6 represent the next generation
and a major advance in terms of PAT-selective agents. These
materials have similar cytotoxicities to the original tri-
amine series (1a, 2, and 3) in L1210 cells but are significantly
more selective in terms of entering cells with an active
polyamine transport system (e.g., CHO cells and FL5.12A cells).
This is represented by their dramatic increase in potency in the
CHO wild-type cell line over their monosubstituted counterparts
(1a, 2, and 3).
Table 6. IL-3-Dependent FL5.12A B Cells: DMSO Shock Treatment
Assay via Annexin-V Staining (% Apoptosis)a
%
IL-3 (ng/mL) UNTR DMSO
4
4 + DMSO 4 + DMSO + SPD
2
0.1
0
8.59
4.94
61.85
7.54 5.56
6.29 6.07
1.82 8.64
44.71
56.71
67.53
12.89
10.20
16.69
a UNTR (untreated); DMSO (dimethyl sulfoxide); SPD (spermidine).
Representative data for the apoptotic effect of compound 4 on FL5.12A
after DMSO shock; experiments were done in duplicate with an error margin
of 6%.
During the preparation of this manuscript, Petros et al.29
described the synthesis and antizyme-inducing effects of
compounds 33 and 34 (Figure 2). Antizyme is an inhibitor of
ODC, and its induction leads to reduced polyamine transport
and may lead to apoptosis. Because 33 and 34 have similar
structures to compounds 4-6, it is possible that these new
disubstituted materials are also antizyme-inducing agents.
Indeed, high levels of antizyme are consistent with apoptosis,
that is, programmed cell death, which has been observed with
these compounds (e.g., 1a). Future efforts will focus on pursuing
this antizyme induction hypothesis.
be due to the presence of a B cell polyamine transport system,
which may also recognize 1,4-diaminobutane (putrescine).
Different results in B and CHO cells were observed because
their vesicular trafficking patterns are different.24a Other PATs
have been identified in Escherichia coli, which are selective
for both putrescine24b and spermidine.24c PAT-selective com-
pound 1a also showed rescue in the presence of spermidine,
even in cells with different cell cycling behavior (initiated via
the corresponding IL-3 dosage). Together, the results from
Tables 4 and 5 suggest that targeting PAT in B cells for the
delivery of toxic compounds is feasible.
The highly PAT-selective compound 4 did not induce
significant apoptosis in FL5.12A cells (Table 4). A provocative
model for polyamine transport was recently published by Poulin
et al.,25,26 who described initial transport through a plasma
membrane transporter followed by sequestration of polyamine
conjugates into polyamine sequestering vesicles. One possibility
for the low toxicity of compound 4 is that it remains trapped in
vesicles and does not reach its cellular target to initiate the
apoptotic response.
To examine this hypothesis, we exposed FL5.12A cells to a brief
DMSO shock, to release vesicular contents27a,b,c and presumably
release compound 4 from the vesicle compartments. After 4 h, we
observed rapid cell death only in cells that received compound 4
and underwent DMSO shock (Table 6). Note that DMSO shock
alone was antiapoptotic, likely due to its antioxidant effects.27a,b,c
Additionally, competition with spermidine reversed the toxic effects
of compound 4-treated cells that were shocked with DMSO (Table
6). These results indicate that B cells can be targeted through PAT
and that the PAT-selective compounds likely reside in vesicles and
must escape from these vesicles to deliver their toxic cargo to cells.
Indeed, rapid vesicle sequestration has been observed with com-
pound 1a8 and other polyamine derivatives in other mammalian
cell types.25,26 As mentioned previously, B cells and CHO cells
have different cell types with different vesicular trafficking, which
explains the different results. For example, late endosome traf-
ficking does occur in CHO28 but does not occur until antigen
activation in B cells.24a
Experimental Section
Materials. Silica gel (32–63 µm) and chemical reagents were
purchased from commercial sources and used without further
purification. All solvents were distilled prior to use. All reactions
were carried out under an N2 atmosphere. 1H and 13C spectra were
recorded at 300 or 75 MHz, respectively. TLC solvent systems were
listed as volume percents, and NH4OH referred to concentrated
aqueous NH4OH. All tested compounds provided satisfactory
elemental analyses. For compounds 4 and 6, the HRMS data is
reported for the double charged species (z ) 2) and represent half
the theoretical MW (m/z).
Biological Studies. Murine leukemia cells (L1210), CHO, and
CHO-MG cells were grown in RPMI medium supplemented with
10% fetal calf serum, glutamine (2 mM), penicillin (100 U/mL),
and streptomycin (50 µg/mL). Note that the media must contain
L-proline (2 µg/mL) for proper growth of the CHO-MG cells. Cells
were grown at 37 °C under a humidified 5% CO2 atmosphere.
Aminoguanidine (2 mM) was added to the culture medium to
prevent oxidation of the drugs by the enzyme (bovine serum amine
oxidase) present in calf serum. Trypan blue staining was used to
determine cell viability before the initiation of a cytotoxicity
experiment. L1210 cells in early to mid log phase were used.
IC50 Determinations. Cell growth was assayed in sterile 96 well
microtiter plates (Becton-Dickinson, Oxnard, CA). L1210 cells were
seeded at 6 × 104 cells/mL of medium (100 µL/well). CHO and
CHO-MG cells were plated at 1 × 104 cells/mL. Drug solutions
(10 µL per well) of appropriate concentration were added at the
time of seeding for L1210 cells and after an overnight incubation
for the other cells. After exposure to the drug for 48 h, cell growth
was determined by measuring formazan formation from 3-(4,5-
dimethylthiazol-2-yl)2,5-diphenyltetrazolium using a Titertek Mul-
tiskan MCC/340 microplate reader for absorbance (540 nm)
measurements.30
Conclusions
Di- and trisubstituted polyamine derivatives were synthesized
and evaluated for their ability to target the polyamine transport

Ligands for Accessing the Polyamine Transport System
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 5 1399
Ki Procedure. The ability of the conjugates to interact with the
polyamine transport system was determined by measuring competi-
tion between the conjugates and the radiolabeled spermidine for
uptake in L1210 cells. This procedure was used to obtain the data
listed in Table 1. Initially, the Km value of spermidine transport
was determined as previously described.31
8H), 3.04 (t, 4H), 1.85–1.73 (m, 16H). 13C NMR (D2O): δ 131.3,
129.3, 128.6, 128.0, 123.8, 48.2, 47.3, 47.2, 47.1, 39.0, 24.2, 23.2,
23.1. HRMS (FAB) calcd for C28H50N6 ·6HCl (M + H - 6HCl)+,
471.4170; found, 471.4155.
N-(4-Amino-butyl)-N′-(4-{[4-(4-amino-butylamino)butylamino]-
methyl}benzyl)butane-1,4-diamine, Hydrochloride Salt, 6. White
1
The ability of conjugates to compete for [14C]spermidine uptake
was determined in L1210 cells by a 10 min uptake assay in the
presence of increasing concentrations of competitor, using 1 µM
[14C]spermidine as substrate. Ki values for inhibition of spermidine
uptake were determined using the Cheng-Prusoff equation32 from
the IC50 value derived by iterative curve fitting of the sigmoidal
equation describing the velocity of spermidine uptake in the presence
of the respective competitor.33 L1210 cells were grown and maintained
according to established procedures34 and were washed twice in HBSS
prior to the transport assay.
solid; yield, 93%. H NMR (D2O): δ 7.59 (m, 4H), 4.30 (s, 4H),
3.17 (t, 4H), 3.11 (t, 8H), 3.05 (t, 4H), 1.78 (m, 16H). 13C NMR
(D2O): δ 132.4, 130.9, 64.3, 51.1, 47.4, 47.3, 47.1, 39.3, 24.5, 23.4,
23.3. HRMS (FAB) calcd for C24H48N6 ·6HCl [(M + 2H - 6HCl)/
2]+, 211.2043; found, 211.2043.
N-(4-Amino-butyl)-N-(3,5-bis-{[4-(4-amino-butylamino)butylamino]-
methyl}benzyl)butane-1,4-diamine, Hydrochloride Salt, 7. White
soild; yield, 97%. 1H NMR (300 MHz, D2O): δ 7.64 (s, 3H,
aromatic), 4.32 (s, 6H, CH2), 3.18 (t, 6H, CH2), 3.07 (m, 18H, CH2),
1.77 (m, 24H, CH2). 13C NMR (D2O): δ 135.6, 135.1, 53.2, 49.8,
49.7, 41.7, 26.9, 25.8, 25.7. HRMS (FAB) m/z calcd for
C33H78N9Cl9 (M + H - 9HCl)+, 592.5749; found, 592.5749.
N-(4-Amino-butyl)-N′-{4-[bis-(2-chloro-ethyl)amino]benzyl}-
butane-1,4-diamine, Hydrochloride Salt, 10. White solid; yield,
B Cell Culture. FL5.12A cells were a kind gift from James
McCubrey, East Carolina University (Greenville, SC). Cells were
cultured in RPMI1640, supplemented with 10% heat-inactivated
fetal calf serum (FCS), antibiotics (penicillin, 100 units/mL;
streptomycin sulfate, 100 µg/mL), and IL-3 (2, 0.1, or 0 ng/mL) at
37 °C, in an atmosphere of 95% air and 5% CO2 under humidified
conditions. Aminoguanidine (1 mM) was added as an inhibitor of
amine oxidase derived from FCS and had no effect on the various
parameters of the cell measured in this study.
Annexin-V Staining. Apoptosis was assessed using the Annexin-
V-FITC Apoptosis Detection Kit I (BD Biosciences). FL5.12A cells
were cultured in 25 cm2 flasks and preincubated for 24 h with
different concentrations of IL-3 in 24 well plates (5 × 106 cell/
well) and then treated with 10 µM concentrations of compounds
1a, 1b, and 4 along with a 10 µM concentration of aminoguanidine.
After 24 h of incubation, cells were washed twice with ice-cold
PBS and then stained with Annexin-V using the Annexin-V-FITC
Apoptosis Detection Kit I following the manufacturer’s protocol
(BD Biosciences). FITC fluorescence was detected using a BD
FACSCalibur flow cytometer. Ten thousand events were acquired
in each sample. The percents listed in Tables 3-5 are % annexin
V-stained cells out of the total population. This directly cor-
responded to % of cells in apoptosis.
Spermidine Competition Assay in B Cells. FL5.12A cells were
cultured in 25 cm2 flasks and then incubated in 24 well plates (5 ×
106 cell/well) with different concentrations of IL-3 for 24 h; the cells
were treated with the indicated concentration of compounds 1a, 1b,
and 4 overnight in the presence or absence of SPD (100 µM), and
cell apoptosis was assessed by annexin-V staining as described before.
DMSO Shock Assay in B Cells. FL5.12A cells were cultured
in 25 cm2 flasks and then incubated in 24 well plates (5 × 106
cell/well) with different concentrations of IL-3 for 24 h; cells were
then treated with compound 4 overnight in the presence or absence
of SPD (100 µM) and then were washed and incubated in 10%
DMSO for 2 min, washed twice with Hank’s media, and incubated
for 2 h at 37 °C. Apoptosis was assessed with annexin-V-FITC
staining as previously described.
General Procedure for Removal of BOC Group. The respec-
tive solution of the BOC-protected compound dissolved in absolute
ethanol (13 mL) was stirred at 0 °C for 10 min. A 4 N HCl solution
(22 mL) was added dropwise to the reaction mixture and stirred at
0 °C for 20 min and then at room temperature overnight. The
solution was concentrated in vacuo to give the respective amine
HCl salt as a solid.
1
96%. H NMR (300 MHz, D2O): δ 7.41 (d, 2H, aromatic), 7.05
(d, 2H, aromatic), 4.16 (s, 2H, CH2), 3.87 (t, 4H, CH2), 3.72 (t,
4H, CH2), 3.06 (m, 8H, CH2), 1.75 (m, 8H, CH2). 13C NMR (D2O):
δ 131.7, 114.7, 53.7, 50.7, 47.1, 47.1, 46.1, 42.1, 40.4, 39.0, 24.2,
23.1, 23.0. HRMS (FAB) m/z calcd for C19H37N4Cl5 (M + H -
3HCl)+, 389.2233; found, 389.2235.
N-(3-Amino-propyl)-N′-{4-[bis-(2-chloro-ethyl)amino]benzyl}-
butane-1,4-diamine, Hydrochloride Salt, 11. White solid; yield,
1
93%. H NMR (300 MHz, D2O): δ 7.47 (d, 2H, aromatic), 7.17
(d, 2H, aromatic), 4.20 (s, 2H, CH2), 3.92 (t, 4H, CH2), 3.70 (t,
4H, CH2), 3.19–3.05 (m, 8H, CH2), 2.09 (q, 2H, CH2), 1.78 (m,
4H, CH2). 13C NMR (D2O): δ 143.2, 132.2, 131.9, 124.5, 116.8,
55.1, 50.6, 47.3, 46.3, 44.8, 39.8, 36.8, 24.0, 23.1, 23.0. HRMS
(FAB) m/z calcd for C18H35N4Cl5 (M + H - 3HCl)+, 375.2080;
found, 375.2069.
N1-(3-{4-[Bis-(2-chloro-ethyl)amino]benzylamino}propyl)bu-
tane-1,4-diamine, Hydrochloride Salt, 12. White solid; yield, 91%.
1H NMR (300 MHz, D2O): δ 7.44 (d, 2H, aromatic), 7.11 (d, 2H,
aromatic), 4.21 (s, 2H, CH2), 3.89 (t, 4H, CH2), 3.71 (t, 4H, CH2),
3.19–3.07 (m, 6H, CH2), 3.03 (t, 2H, CH2), 2.12 (q, 2H, CH2),
1.76 (m, 4H, CH2). 13C NMR (D2O): δ 144.6, 131.8, 122.4, 115.5,
54.2, 50.9, 47.3, 44.7, 43.9, 40.2, 39.0, 24.2, 23.0, 22.9. HRMS
(FAB) m/z calcd for C18H35N4Cl5 (M + H - 3HCl)+, 375.2077;
found, 375.2066.
N1-{4-[Bis-(2-chloro-ethyl)amino]benzyl}butane-1,4-di-
amine, Hydrochloride Salt, 13. White solid; yield, 96%. 1H NMR
(300 MHz, D2O): δ 7.50 (d, 2H, aromatic), 7.21 (d, 2H, aromatic),
4.22 (s, 2H, CH2), 3.93 (t, 4H, CH2), 3.71 (t, 4H, CH2), 3.11 (t,
2H, CH2), 3.04 (t, 2H, CH2), 1.77 (m, 4H, CH2). 13C NMR (D2O):
δ 142.9, 131.9, 124.8, 117.0, 55.2, 50.6, 46.4, 39.7, 39.0, 24.3,
23.0. HRMS (FAB) m/z calcd for C15H27N3Cl4 (M + H - 2HCl)+,
318.1501; found, 318.1495.
General Procedure for Reductive Amination in the Synthesis
of Disubstituted Derivatives. To a stirred solution of 14 (2.5 mmol)
in 25% MeOH/CH2Cl2 (20 mL) was added a solution of respective
dicarbaldehyde (1 mmol) in 25% MeOH/CH2Cl2 (15 mL) under
N2. The mixture was stirred at room temperature overnight until
the imine formation was complete [as monitored by the disappear-
ance of the 1H NMR (CDCl3) signal of aldehyde]. The solvent was
removed in vacuo, the solid residue was dissolved in 50% MeOH/
CH2Cl2 (40 mL), and the solution was cooled to 0 °C. NaBH4 (6
equiv) was added in small portions to the solution, and the mixture
was stirred at rt overnight. The solvent was removed in vacuo, and
the solid residue was dissolved in CH2Cl2 (50 mL) and washed
with 10% aqueous Na2CO3 solution (3 × 30 mL). The CH2Cl2 layer
was separated, dried over anhydrous Na2SO4, filtered, and removed
in vacuo to give an oily residue. The oil was purified by flash
column chromatography to yield the product.
N-(4-Amino-butyl)-N′-(10-{[4-(4-amino-butylamino)butylamino]-
methyl}anthracen-9-ylmethyl)butane-1,4-diamine, Hydrochlo-
ride Salt, 4. Yellow solid; yield, 95%. 1H NMR (D2O): δ 8.19 (d,
4H), 7.79 (d, 4H), 4.83 (s, 4H), 3.30 (t, 4H), 3.11 (m, 12H), 1.79
(m, 16H). 13C NMR (D2O): δ 129.7, 127.7, 124.7, 124.0, 47.7,
47.2, 47.1, 42.8, 39.0, 24.2, 23.2, 23.1. HRMS (FAB) calcd for
C32H52N6 ·6HCl [(M + 2H - 6HCl)/2]+, 261.2199; found, 261.2199.
N-(4-Amino-butyl)-N′-(4-{[4-(4-amino-butylamino)butylamino]-
methyl}naphthalen-1-ylmethyl)butane-1,4-diamine, Hydrochlo-
ride Salt, 5. Yellow solid; yield, 94%. 1H NMR (D2O): δ 8.20 (m,
2H), 7.78 (m, 2H), 7.72 (s, 2H), 4.83 (s, 4H), 3.28 (t, 4H), 3.11 (t,
(4-tert-Butoxycarbonylamino-butyl)-(4-{[10-({4-[tert-butoxycar-
bonyl-(4-tert-butoxycarbonylamino-butyl)amino]butylamino}meth-
yl)anthracen-9-ylmethyl]amino}butyl)carbamic Acid tert-Bu-

1400 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 5
Kaur et al.
tyl Ester, 18. Pale yellow viscous oil (92%), Rf ) 0.3 (5% MeOH/
dissolved in THF (45 mL) and stirred for 20 min at 0 °C. A solution
of 2-(tert-butoxycarbonyloyimino)-2-phenylacetonitrile (BOC-ON,
0.23 g, 0.93 mmol) was added dropwise with constant stirring. After
the addition was complete, the reaction was stirred for 2 h at 0 °C
under a N2 atmosphere. Upon completion, the solution was
concentrated in vacuo, and the residue was redissolved in CH2Cl2
and washed with a saturated aqueous Na2CO3. The organic layer
was separated, dried over anhydrous Na2SO4, filtered, and concen-
trated. Flash column chromatography of the residue gave pure 24
as a colorless oil (79 mg). Yield, 6%; Rf ) 0.35 (1% NH4OH/
1
0.5% NH4OH/CH2Cl2). H NMR (CDCl3): δ 8.36 (d, 4H), 7.50
(d, 4H), 4.82 (br m, 2H), 4.68 (s, 4H), 3.13 (m, 12H), 2.87 (t, 4H),
1.65–1.30 (m, 52H). 13C NMR (CDCl3): δ 155.9, 155.4, 131.9,
129.9, 125.6, 124.8, 79.1, 78.9, 53.5, 50.3, 46.9, 46.7, 46.0, 40.2,
28.6, 28.5, 27.5. HRMS (FAB) m/z calcd for C52H84N6O8 (M +
H)+, 921.6423; found, 921.6414.
(4-tert-Butoxycarbonylamino-butyl)-(4-{[4-({4-[tert-butoxycar-
bonyl-(4-tert-butoxycarbonylamino-butyl)amino]butylamino}-
methyl)naphthalen-1-ylmethyl]amino}butyl)carbamic Acid tert-
Butyl Ester, 19. Pale yellow viscous oil (60%), Rf ) 0.3 (7%
MeOH/0.5% NH4OH/CH2Cl2). 1H NMR (CDCl3): δ 8.12 (m, 2H),
7.50 (m, 2H), 7.38 (s, 2H), 4.87 (br m, 2H), 4.19 (s, 4H), 3.10 (m,
12H), 2.73 (t, 4H), 1.65–1.35 (m, 52H). 13C NMR (CDCl3): δ 155.9,
155.4, 135.3, 132.0, 125.7, 125.4, 124.2, 79.1, 51.7, 49.7, 46.9,
46.7, 40.2, 28.5, 28.5, 27.4, 26.7, 26.1, 25.8. HRMS (FAB) m/z
calcd for C48H82N6O8 (M + H)+, 871.6267; found, 871.6211.
(4-tert-Butoxycarbonylamino-butyl)-{4-[4-({4-[tert-butoxycar-
bonyl-(4-tert-butoxycarbonylamino-butyl)amino]butylamino}-
methyl)benzylamino]butyl}carbamic Acid tert-Butyl Ester, 20.
Pale yellow viscous oil (54%), Rf ) 0.38 (6% MeOH/0.5% NH4OH/
CH2Cl2). 1H NMR (CDCl3): δ 7.17 (m, 4H), 4.83 (br m, 2H), 3.68
(s, 4H), 3.06 (m, 12H), 2.55 (m, 4H), 1.60–1.21 (m, 52H). 13C
NMR (CDCl3): δ 155.9, 155.4, 138.8, 128.1, 126.8, 79.1, 78.8,
64.3, 53.5, 49.0, 46.9, 46.7, 40.2, 28.5, 28.5, 27.5, 27.3, 26.6, 26.0,
25.6. HRMS (FAB) m/z calcd for C44H80N6O8 (M + H)+, 821.6110;
found, 821.6083.
1,3,5-Tris(hydroxymethyl)benzene, 22. Trimethyl-1,3,5-ben-
zenetricarboxylate 21 (2 g, 7.9 mmol, Acros Chemicals) in dry THF
(30 mL) was added through a pressure-equalized addition funnel
into a 250 mL flask containing LiAlH4 (0.90 g, 23.6 mmol) in dry
THF (65 mL) at 0 °C under a N2 atmosphere. The mixture was
allowed to warm to room temperature and stirred for 4 h. The
reaction was quenched by the slow addition of a 1:1 mixture of
Celite and KHSO4. The suspension was filtered, and the Celite was
washed with MeOH (100 mL). The solvent was removed under
reduced pressure, and triol 22 was obtained in 78% yield (1.05 g).
1H NMR of the product matched that of the authentic material. 1H
NMR (300 MHz, DMSO): δ 7.18 (s, 3H), 5.17 (br s, 3H), 4.50 (s,
6H).
1
6.5% CH3OH/CH2Cl2). H NMR (300 MHz, CDCl3): δ 7.14 (s,
3H, aromatic), 4.69 (s, 3H, NH), 3.75 (s, 6H, CH2), 3.15 (m, 18H,
CH2), 2.65 (t, 6H, CH2), 1.65- 1.15 (m, 78H, CH2, CH3). 13C NMR
(CDCl3): δ 156.0, 155.6, 140.5, 126.7, 79.3, 79.2, 54.1, 49.5, 47.1,
46.9, 40.4, 28.7, 28.6, 27.6, 27.5, 26.8, 26.1, 25.8. HRMS (FAB)
m/z calcd for C63H117O12N9 (M + H)+, 1192.8894; found, 1192.9008.
General Procedure for Reductive Amination in the Synthesis
of Chlorambucil Derivatives. To a stirred solution of the respective
Boc-protected amine (1.5 mmol) in 25% MeOH/CH2Cl2 (20 mL)
was added a solution of commercially available aldehyde 25 (1
mmol) in 25% MeOH/CH2Cl2 (15 mL) under N2. The mixture was
stirred at room temperature overnight until the imine formation was
complete [as monitored by the disappearance of the 1H NMR
(CDCl3) aldehyde signal]. The solvent was removed in vacuo, the
solid residue was dissolved in 50% MeOH/CH2Cl2 (40 mL), and
the solution was cooled to 0 °C. NaBH4 (3 equiv) was added in
small portions to the solution, and the mixture was stirred at rt
overnight. The solvent was removed in vacuo, and the solid residue
was dissolved in CH2Cl2 (50 mL) and washed with 10% aqueous
Na2CO3 solution (3 × 30 mL). The CH2Cl2 layer was separated,
dried over anhydrous Na2SO4, filtered, and removed in vacuo to
give an oily residue. The oil was purified by flash column
chromatography to yield the product.
(4-{4-[Bis-(2-chloro-ethyl)amino]benzylamino}butyl)-(4-tert-
butoxycarbonyl-amino-butyl)carbamic Acid tert-Butyl Ester, 28.
Pale yellow viscous oil (91%); Rf ) 0.5 (6% MeOH/0.5% NH4OH/
CH2Cl2). 1H NMR (CDCl3): δ 7.19 (d, 2H), 6.63 (d, 2H), 5.05
(brs, 1H), 3.69 (m, 6H), 3.61 (m, 4H), 3.12 (m, 6H), 2.62 (t, 2H),
1.55–1.30 (m, 26H). 13C NMR (CDCl3): δ 155.8, 155.3, 144.8,
129.5, 129.1, 111.8, 79.0, 78.8, 53.5, 53.1, 48.9, 46.9, 40.5, 40.2,
28.5, 28.4, 27.4, 27.2, 26.4, 26.1, 25.2. HRMS (FAB) m/z calcd
for C29H50N4O4Cl2 (M + H)+, 589.3282; found, 589.3277.
(4-{4-[Bis-(2-chloro-ethyl)amino]benzylamino}butyl)-(3-tert bu-
toxycarbonylamino-propyl)carbamic Acid tert-Butyl Eester, 29.
Pale yellow viscous oil (90%); Rf ) 0.3 (5% MeOH/0.4% NH4OH/
CH2Cl2). 1H NMR (CDCl3): δ 7.18 (d, 2H), 6.63 (d, 2H), 3.74–3.35
(m, 10H), 3.24–3.04 (m, 6H), 2.61 (t, 2H), 1.64 (m, 2H), 1.52–1.41
(m, 22H). 13C NMR (CDCl3): δ 155.9, 144.9, 129.6, 128.7, 111.8,
79.4, 78.8, 53.5, 53.1, 50.0, 48.8, 46.9, 44.3, 43.7, 40.5, 37.4, 28.4,
28.2, 27.1, 26.4, 26.2. HRMS (FAB) m/z calcd for C28H48N4O4Cl2
(M + H)+, 575.3130; found, 575.3092.
1,3,5-Triformyl Benzene, 23. 1,3,5-Tris(hydroxymethyl)benzene
22 (1.05 g, 6.25 mmol) was suspended in CH2Cl2 (25 mL), and
solid pyridinium chlorochromate (PCC, 5.98 g, 27.74 mmol) was
added. After 30 min of stirring, the reaction mixture was diluted
with acetone (10 mL) and was allowed to stir for 3 h. The
precipitated chromium salts were filtered off and washed with
CH2Cl2. The organic phase was washed with a saturated solution
of aqueous Na2CO3 three times, then separated and dried over
anhydrous Na2SO4, filtered, and concentrated to provide a solid.
Column chromatography (100% CH2Cl2) afforded 23 as white
1
crystals (0.51 g, 51%). H NMR (300 MHz, CDCl3): δ 10.21 (s,
3H, CHO), 8.66 (s, 3H, aromatic).
(3-{4-[Bis-(2-chloro-ethyl)amino]benzylamino}propyl)-(4-tert-
butoxycarbonyl-amino-butyl)carbamic Acid tert-Butyl Ester, 30.
Pale yellow viscous oil (93%); Rf ) 0.3 (5% MeOH/0.4% NH4OH/
CH2Cl2). 1H NMR (CDCl3): δ 7.18 (d, 2H), 6.61 (d, 2H), 4.88
(brs, 1H), 3.73–3.58 (m, 10H), 3.24–3.06 (m, 6H), 2.59 (t, 2H),
1.68 (q, 2H), 1.53–1.41 (m, 22H). 13C NMR (CDCl3): δ 155.8,
155.4, 144.8, 129.6, 129.0, 111.8, 79.1, 78.8, 53.5, 53.2, 46.5, 46.2,
45.1, 44.4, 40.5, 40.2, 28.5, 28.4, 27.4, 25.9, 25.5.
(4-{4-[Bis-(2-chloro-ethyl)amino]benzylamino}butyl)carbamic Acid
tert-Butyl Ester, 32. Pale yellow viscous oil (69%); Rf ) 0.3 (5%
MeOH/0.3% NH4OH/CH2Cl2). 1H NMR (CDCl3): δ 7.17 (d, 2H),
6.61 (d, 2H), 5.17 (br s, 1H), 3.72–3.54 (m, 10H), 3.07 (m, 2H),
2.61 (t, 2H), 1.52–1.35 (m, 13H). 13C NMR (CDCl3): δ 155.8,
144.8, 129.5, 129.0, 111.8, 78.7, 53.4, 53.1, 48.7, 40.5, 28.5, 27.9,
27.3. HRMS (FAB) m/z calcd for C20H33N3O2Cl2 (M + H)+,
418.2026; found, 418.2006.
{4-[3,5-Bis-({4-[tert-butoxycarbonyl-(4-tert-butoxycarbonyl-
amino-butyl)amino]butylamino}methyl)benzylamino]butyl}-(4-
tert-butoxycarbonylamino-butyl)carbamic Acid tert-Butyl Ester,
24. 1,3,5-Triformyl benzene 23 (0.180 g, 1.11 mmol) was dissolved
in 25% MeOH/CH2Cl2 (10 mL). A solution of Boc-protected
homospermidine 14 (1.44 g, 4.011 mmol) in 25% MeOH/CH2Cl2
(10 mL) was added via an addition funnel. The reaction mixture
was stirred overnight under a N2 atmosphere. Loss of the starting
material was monitored via 1H NMR spectroscopy and the
disappearance of the aldehyde proton at 10.21 ppm. Upon conver-
sion of the starting material, the solvent was removed in vacuo
and the crude material was redissolved in a solution of 50% MeOH/
CH2Cl2. To this new solution was added NaBH4 (0.45 g, 11.9 mmol)
at 0 °C. The solution was stirred overnight under a N2 atmosphere.
The solvent was removed in vacuo, and flash column chromatog-
raphy (1% NH4OH/5% CH3OH/CH2Cl2) provided a mixture of the
coeluting BOC-protected homospermidine 14 and the desired
product 24 (1.08 g). To aid in the chromatographic separation of
24 and 14, another reaction was carried out. The mixture was
Acknowledgment. We thank Dr. David Powell at the Uni-
versity of Florida mass spectrometry facility for his kind assistance
in acquiring the mass spectra of these compounds. O.P. acknowl-

Ligands for Accessing the Polyamine Transport System
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 5 1401
(18) Holley, J. L.; Mather, A.; Wheelhouse, R. T.; Cullis, P. M.; Hartley,
J. A.; Bingham, J. P.; Cohen, G. M. Targeting of tumor cells and
DNA by a chlorambucil-spermidine conjugate. Cancer Res. 1992, 52,
4190–4195.
(19) Gardner, R. A.; Belting, M.; Svensson, K.; Phanstiel, O., IV. Synthesis
and transfection efficiencies of new lipophilic polyamines. J. Med.
Chem. 2007, 50, 308–318.
(20) Covassin, L.; Desjardins, M.; Charest-Gaudreaults, R.; Audette, M.;
Bonneau, M.-J.; Poulin, R. Synthesis of spermidine and norspermidine
dimers as high affinity polyamine transport inhibitors. Bioorg. Med.
Chem. Lett. 1999, 9, 1709–1714.
edges the Broad Medical Research Program of the Eli and Edythe
L. Broad Foundation for their partial support of this research and
thanks Dr. W. F. Flintoff (University of Western Ontario) for his
gracious donation of the CHO-MG cell line.14
Supporting Information Available: Elemental analyses and 1H
and 13C spectra for compounds 4–7, 10–13, 18–20, 24, 28, 29, and
32. This material is available free of charge via the Internet at http://
pubs.acs.org.
(21) (a) Golding, B. T.; Kebbell, M. J. Chemistry of nitrogen mustard
[2-chloro-N-(2-chloroethyl)-N-methylethanamine] studied by nuclear
magnetic resonance spectroscopy. J. Chem. Soc. Perkin Trans. 2, 1987,
705–713. (b) Ringdahl, B.; Resul, B.; Ehlert, F. J.; Jenden, D. J.;
Dahlbom, R. The conversion of 2-chloroalkylamine analogs of
oxotremorine to aziridinium ions and their interactions with muscarinic
receptors in the guinea pig ileum. Mol. Pharmacol. 1984, 26, 170–
179.
(22) Algate, P. A.; Steelman, L. S.; Mayo, M. W.; Miyajima, A.; McCubrey,
J. A. Regulation of the interleukin-3 (IL-3) receptor by IL-3 in the
fetal liver-derived FL5.12 cell line. Blood 1994, 83, 2459–2468.
(23) DeBenedette, M.; Olson, J. W.; Snow, E. C. Expression of polyamine
transporter activity during B lymphocyte cell cycle progression.
J. Immunol. 1993, 150, 4218–4224.
(24) (a) Bertram, E. M.; Hawley, R. G.; Watts, T. H. Overexpression of
rab7 enhances the kinetics of antigen processing and presentation with
MHC class II molecules in B cells. Int. Immunol. 2002, 14, 309–318.
(b) Pistocchi, R.; Kashiwagi, K.; Miyamoto, S.; Nukui, E.; Sadakata,
Y.; Kobayashi, H.; Igarashi, K. Characteristics of the operon for a
putrescine transport system that maps at 19 minutes on the Escherichia
coli chromosome. J. Biol. Chem. 1993, 268, 146–152. (c) Kashiwagi,
K.; Innami, A.; Zenda, R.; Tomitori, H.; Igarashi, K. The ATPase
activity and the functional domain of PotA, a component of the
spermidine-preferential uptake system in Escherichia coli. J. Biol.
Chem. 2002, 277, 24212–24219.
(25) Soulet, D.; Covassin, L.; Kaouass, M.; Charest-Gaudreaults, R.;
Audette, M.; Poulin, R. Role of endocytosis in the internalization of
spermidine-C2-BODIPY, a highly fluorescent probe of polyamine
transport. Biochem. J. 2002, 367, 347–357.
(26) Soulet, D.; Gagnon, B.; Rivest, S.; Audette, M.; Poulin, R. A
fluorescent probe of polyamine transport accumulates into intracellular
acidic vesicles via a two-step mechanism. J. Biol. Chem. 2004, 279,
49355–49366.
(27) (a) Geden, S. E.; Gardner, R. A.; Fabbrini, M. S.; Ohashi M.; Phanstiel,
O., IV; Teter, K. Lipopolyamine treatment increases the efficacy of
intoxication with saporin and an anti-cancer saporin conjugate. FEBS
Journal 2007, 274, 4825–4836. (b) Kawai, S.; Nishizawa, M. New
procedure for DNA transfection with polycation and dimethyl sul-
foxide. Mol. Cell. Biol. 1984, 4, 1172–1174. (c) Willis, C. L.; Ray,
D. E. Antioxidants attenuate MK-801-induced cortical neurotoxicity
in the rat. Neurotoxicology 2007, 28, 161–167.
(28) Ohashi, M.; Miwako, I.; Nakamura, K.; Yamamoto, A.; Murata, M.;
Ohnishi, S.; Nagayama, K. An arrested late endosome-lysosome
intermediate aggregate observed in a Chinese hamster ovary cell
mutant isolated by novel three-step screening. J. Cell. Sci. 1999, 112,
1125–1138.
(29) Petros, L. M.; Graminski, G. F.; Robinson, S.; Burns, M. R.; Kisiel,
N.; Gesteland, R. F.; Atkins, J. F.; Kramer, D. L.; Howard, M. T.;
Weeks, R. S. Polyamine analogs with xylene rings induce antizyme
frameshifting, reduce ODC activity and deplete cellular polyamines.
J. Biochem. 2006, 140, 657–666.
(30) Mosmann, T. Rapid colorimetric assay for cellular growth and survival:
Application to proliferation and cytotoxicity assays. J. Immunol.
Methods 1983, 65, 55–63.
(31) Clément, S.; Delcros, J. G.; Feuerstein, B. G. Spermine uptake is
necessary to induce haemoglobin synthesis in murine erythroleukemia
cells. Biochem. J. 1995, 312, 933–938.
(32) Cheng, Y. C.; Prusoff, W. H. Relationship between the inhibition
constant (Ki) and the concentration of inhibitor which causes 50%
inhibition (IC50) of an enzymatic reaction. Biochem. Pharmacol. 1973,
22, 3099–3108.
(33) Torossian, K.; Audette, M.; Poulin, R. Substrate protection against
inactivation of the mammalian polyamine transport system by 1-ethyl-
3-(3-dimethylaminopropyl)-carbodiimide. Biochem. J. 1996, 319, 21–
26.
(34) Bergeron, R. J.; Müller, R.; Bussenius, J.; McManis, 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.
References
(1) (a) Cullis, P. M.; Green, R. E.; Merson-Davies, L.; Travis, N. Probing
the mechanism of transport and compartmentalisation of polyamines
in mammalian cells. Chem. Biol. 1999, 6, 717–729, and refs therein.
(b) Seiler, N.; Dezeure, F. Polyamine transport in mammalian cells.
Int. J. Biochem. 1990, 22, 211–218. (c) Seiler, N.; Delcros, J.-G.;
Moulinoux, J. P. Polyamine transport in mammalian cells. An update.
Int. J. Biochem. Cell Biol. 1996, 28, 843–861.
(2) Igarashi, K.; Kashiwagi, K. Polyamine transport in bacteria and yeast.
Biochem. J. 1999, 344, 633–642, and refs therein.
(3) Hasne, M.-P.; Ullman, B. Identification and characterization of a
polyamine permease from the protozoan parasite Leishmania major.
J. Biol. Chem. 2005, 280, 15188–15194.
(4) Phanstiel, O., IV; Price, H. L.; Wang, L.; Juusola, J.; Kline, M.; Shah,
S. M. The effect of polyamine homologation on the transport and
cytotoxicity properties of polyamine-(DNA-intercalator) conjugates.
J. Org. Chem. 2000, 65, 5590–5599.
(5) Wang, L.; Price, H. L.; Juusola, J.; Kline, M.; Phanstiel, O., IV. The
influence of polyamine architecture on the transport and topoisomerase
II inhibitory properties of polyamine DNA-intercalator conjugates.
J. Med. Chem. 2001, 44, 3682–3691.
(6) Wang, C.; Delcros, J.-G.; Biggerstaff, J.; Phanstiel, O., IV. Synthesis
and biological evaluation of N1-(anthracen-9-ylmethyl)triamines as
molecular recognition elements for the polyamine transporter. J. Med.
Chem. 2003, 46, 2663–2671.
(7) Wang, C.; Delcros, J.-G.; Biggerstaff, J.; Phanstiel, O., IV. Molecular
requirements for targeting the polyamine transport system: synthesis
and biological evaluation of polyamine-anthracene conjugates. J. Med.
Chem. 2003, 46, 2672–2682.
(8) Wang, C.; Delcros, J.-G.; Cannon, L.; Konate, F.; Carias, H.;
Biggerstaff, J.; Gardner, R. A.; Phanstiel, O., IV. Defining the
molecular requirements for the selective delivery of polyamine-
conjugates into cells containing active polyamine transporters. J. Med.
Chem. 2003, 46, 5129–5138.
(9) Kaur, N.; Delcros, J.-G.; Martin, B.; Phanstiel, O., IV. Synthesis and
biological evaluation of dihydromotuporamine derivatives in cells
containing active polyamine transporters. J. Med. Chem. 2005, 48,
3832–3839.
(10) Gardner, R. A.; Delcros, J.-G; Konate, F.; Breitbeil, F., III; Martin,
B.; Sigman, M.; Phanstiel, O., IV. N1-substituent effects in the selective
delivery of polyamine-conjugates into cells containing active polyamine
transporters. J. Med. Chem. 2004, 47, 6055–6069.
(11) (a) Bergeron, R. J.; Feng, Y.; Weimar, W. R.; McManis, J. S.; Dimova,
H.; Porter, C.; Raisler, B.; Phanstiel, O. A comparison of structure-
activity relationships between spermidine and spermine analogue
antineoplastics. J. Med. Chem. 1997, 40, 1475–1494. (b) Kramer,
D. L.; Miller, J. T.; Bergeron, R. J.; Khomutov, R.; Khomutov, A.;
Porter, C. W. Regulation of polyamine transport by polyamines and
polyamine analogs. J. Cell. Physiol. 1993, 155, 399–407.
(12) Delcros, J.-G.; Tomasi, S.; Carrington, S.; Martin, B.; Renault, J.;
Blagbrough, I. S.; Uriac, P. Effect of spermine conjugation on the
cytotoxicity and cellular transport of acridine. J. Med. Chem. 2002,
45, 5098–5111.
(13) Covassin, L.; Desjardins, M.; Charest-Gaudreault, R.; Audette, M.;
Bonneau, M. J.; Poulin, R. Synthesis of spermidine and norspermidine
dimers as high affinity polyamine transport inhibitors. Bioorg. Med.
Chem. Lett. 1999, 9, 1709–1714.
(14) Mandel, J. L.; Flintoff, W. F. Isolation of mutant mammalian cells
altered in polyamine transport. J. Cell. Physiol. 1978, 97, 335–344.
(15) Byers, T. L.; Wechter, R.; Nuttall, M. E.; Pegg, A. E. Expression of
a human gene for polyamine transport in chinese hamster ovary cells.
Biochem. J. 1989, 263, 745–752.
(16) Bank, B. B.; Kanganis, D.; Liebes, L. F.; Silber, R. Chlorambucil
pharmacokinetics and DNA binding in chronic lymphocytic leukemia
lymphocytes. Cancer Res. 1989, 49, 554–559.
(17) Cohen, G. M.; Cullis, P. M.; Hartley, J. A.; Mather, A.; Symons,
M. C. R.; Wheelhouse, R. T. Targeting of cytotoxic agents by
polyamines: Synthesis of chlorambucil-spermidine conjugate. J. Chem.
Soc., Chem. Commun. 1992, 298–300.
JM070794T