Preliminary evaluation of the cytotoxicity of a series of tris-2-aminoethylamine (Tren) based hexadentate heterocyclic donor agents

Article abstract of DOI:10.1016/j.bmc.2005.07.007

Tachpyridine is a cytotoxic metal chelator with potential anti-tumor activity. The synthesis and evaluation of a set of derivatives of the related hexadentate heterocyclic donor agents tris-2-aminoethylamine (tren) and tris[N-(2-pyridylmethylene)-2-aminoe

Full text of DOI:10.1016/j.bmc.2005.07.007

Bioorganic & Medicinal Chemistry 13 (2005) 5961–5967
Preliminary evaluation of the cytotoxicity of a series
of tris-2-aminoethylamine (Tren) based hexadentate
heterocyclic donor agents
a
Suzy V. Torti, Rong Ma, Vincent J. Venditto, Frank M. Torti,
a
d
b
c
Roy P. Planalp and Martin W. Brechbiel
d,
*
a
Department of Biochemistry, The Comprehensive Cancer Center, Wake Forest University School of Medicine,
Winston-Salem, NC, USA
b
Department of Cancer Biology, The Comprehensive Cancer Center, Wake Forest University School of Medicine,
Winston-Salem, NC, USA
c
Department of Chemistry, University of New Hampshire, Durham, NH, USA
Radiation Oncology Branch, National Cancer Institute, National Institutes of Health, 10 Center Drive,
Building 10, Room B3B69, Bethesda, MD 20892-1002, USA
d
Received 3 March 2005; accepted 8 July 2005
Available online 22 August 2005
Abstract—Tachpyridine is a cytotoxic metal chelator with potential anti-tumor activity. The synthesis and evaluation of a set of
derivatives of the related hexadentate heterocyclic donor agents tris-2-aminoethylamine (tren) and tris[N-(2-pyridylmethylene)-2-
aminoethyl]amine (trenpyr) was performed to compare their cytotoxic activity to tachpyridine in HeLa tumor cells. Methyl groups
were added to the pyridyl ring of trenpyr, and the effects of alkyl group substitution on cell survival were assessed. Profound
cytotoxicity was observed and IC50 data were obtained in ascending order from those compounds substituted with a methyl
group at the 3-, 4-, or 5-position and lastly by the 6-methyl derivative. These results suggest that analogous derivatives with
substitution at the 3-position of the pyridyl ring deserve further exploration.
Published by Elsevier Ltd.
1. Introduction
pyr) have been studied for their efficacy in this arena
8
–10
(
Fig. 1).
Iron is a critically important metal for cell function,
survival, and replication. It is a cofactor in hemoglo-
bin and myoglobin respiratory proteins, hydroxylating
enzymes, lipoxygenases, and cyclooxygenases, and
ribonucleotide reductase, the enzyme catalyzing the
rate-limiting step in DNA synthesis. Tumor cells have
been shown to have an increased uptake and traffick-
ing of iron with concomittant membrane expression of
Our group has previously synthesized tachpyr and sever-
0–12
1
al analogs of this compound.
Tachpyr is a hexaden-
tate metal chelator with three secondary amines and
three pyridyl groups assembled in a pre-organized
framework base on the cis,cis-1,3,5,-triaminocyclohex-
ane (tach). We have previously shown that tachpyr is
exceptionally cytotoxic through the binding of cellular
iron and/or zinc, but not other metals, and that disrup-
tion of the ability to bind metal ions by alkylation of the
secondary amines effectively deactivates these com-
1
,2
transferrin receptors. Thus, depletion of intracellular
iron has been hypothesized as a viable target for can-
3
cer therapy. Iron chelation agents including deferriox-
4
,5
8
amine (DFO),
PIH) and related compounds,
pyridylmethyl)-cis,cis-1,3,5,-triaminocyclohexane (tach-
pyridoxal isonicotinoyl hydrazone
pounds. Tachpyr also has been reported to induce a
9,10
6
,7
0
00
(
and N,N ,N -tris(2-
p53-independent apoptotic mode of death.
is a lipophilic molecule, which impacts on its ability to
Tachpyr
1
3
cross cell membranes.
While investigations continue into a more detailed
understanding of the mode of action of tachpyr and
*
Corresponding author. Fax: +1 301 402 923; e-mail: martinwb@
mail.nih.gov
0968-0896/$ - see front matter. Published by Elsevier Ltd.
doi:10.1016/j.bmc.2005.07.007

5962
S. V. Torti et al. / Bioorg. Med. Chem. 13 (2005) 5961–5967
2. Results and discussion
N
N
HO
N
The tris-methylated tren compounds were prepared via a
modification of a previously reported route, reduction of
an ethyl carbamate, wherein substitution with benzyl
H
N
NH
N
N
NH
16
O
NH
carbamate was successfully employed (Fig. 2). The ini-
tially planned synthesis of the tris-methylated tren com-
pounds involved methylation of the tris-tosylated tren,
which would then be deprotected to yield the desired
product analogous to the preparation of tris-methylated
tach. However, difficulties were encountered in com-
pletely deprotecting the isolated intermediate methylat-
ed tosylated with formation of a tosyl salt observed by
NMR that was present prior to and retained even after
ion-exchange chromatography. Other protecting groups
were then investigated, including benzylchloroformate,
which was found to conveniently serve as protecting
OH
PIH
Tachpyr
OH
H N
(CH2)
N
(CH2)
2
NH
2
5
O
O
(CH2)
5
N
OH
O
H
N
H C
N
(H C)
(H C)
O
3
2
5
2
2
OH
O
group and be amenable to LiAlH reduction to generate
4
Desferrioxamine (DFO)
methyl groups. Alkylation of this triamine with 2-chlo-
romethylpyridine was accomplished by a modification
of a previously reported route that had been used to
generate a series of analogous N-alkylated tachpyr
Figure 1. Chelating agents that have been evaluated as iron depletion
cancer therapeutics agent.
1
2
derivatives.
1
0,12,14
its respective derivatives,
the nature of the che-
lating agent itself also continues to be an active area
of study. While tachpyr may benefit from the tach
framework conferring considerable pre-organization
of the six donor amines, there is also potentially some
energy cost in metal complex formation as this
chelator has to alter conformation from all equatorial
to an all axial conformation in which the metal ion
assumes a pseudo-adamantyl position within the com-
The trenpyr (1) family of compounds (Fig. 3) was
prepared as previously described for the synthesis of
1
7
trenpyr and tren(C-Me)pyr. This is a modification
of previously reported syntheses of these chelators that
had been prepared for complexation with Fe and Cu as
1
8
superoxide dismutase mimics. In brief, the requisite
pyridine carboxaldehydes were first prepared as
1
0,15
plex structure.
One may hypothesize that selection
of the three aminomethylenepyridyl metal-binding
arms assembled in a geometry that would eliminate
or lower this complex formation energy barrier might
also provide a more biologically active agent for
depleting intracellular iron. Previous results addressing
the location of alkyl groups have demonstrated
NH2
N
3
1) CBZ-Cl
2) LiAlH4
1
0
significant impact on both complex formation, and
1
H
N
5
stability,
cytotoxicity.
and that this in part correlates with
8
N
CH3
N
3
Cl
Thus, to further investigate this possibility, we have
hypothesized that ligands analogous to tachpyr based
on the well-known triamine framework provided by
tris-2-aminoethylamine (tren) should exhibit a similar
biological activity pattern. We would also predict that
while the tren-based compounds would have an
analogous cytotoxicity pattern, they might also be
less active as a whole due to having to counter entro-
py costs in complex formation. Thus, we would
validate some of the physical constraints of complex
formation as they impact biological activity. To this
end, tren, tris[N-methyl-2-(aminoethyl)]amine (NMe-
tren), tris[N-methyl-N-2-pyridylmethylene-2-(aminoeth-
yl)]amine (NMe-trenpyr), tris[N-(2-pyridylmethylene)-
CH3
3
N
N
N
Figure 2. Synthesis and structures of tren based ligands studied.
2
-aminoethyl]amine (trenpyr), tris[N-(2-pyridyl(2-ethyl-
ene))-2-2-aminoethyl]amine (tren(C-Me)pyr), and four
trenpyr analogs with methyl groups at the 3-, 4-, 5-,
and 6-position were prepared and evaluated versus
tachpyr for cytotoxic activity in HeLa cells, a cancer
cell line, as well as in normal epithelial cells.
Figure 3. Structures of trenpyr based ligands studied.

S. V. Torti et al. / Bioorg. Med. Chem. 13 (2005) 5961–5967
5963
previously reported by lithiation of the appropriate
methyl-substituted 2-bromo-pyridine, that was
intracellular iron resulting in a more toxic agent. How-
ever, the results indicate that tachpyr possesses a signif-
icant advantage originating from the more restricted
geometry of this ligand. This advantage probably pro-
vides a real influence on complex formation in regard
to overall lowered entropy, outweighing the energy cost
quenched with dimethylformamide and isolated by
fractional distillation. The aldehydes were reacted with
tren to form their respective tris-imines; the reaction
was driven to completion by Dean–Stark distillation.
The tris-imines were then directly reduced with excess
borohydride to provide, after an extractive workup,
the targeted set of chelating agents.
1
9
of conformational change from equatorial to axial.
Despite the somewhat disappointing result obtained
from direct comparison of the tachpyr to trenpyr, the re-
sults obtained from examining the effects of methyl sub-
stitution, either on N–Me or on the pyridyl ring of
trenpyr were quite interesting. From the data presented
in Figures 4, and 5 and Table 1, one can clearly observe
a ranking of activity wherein the tren(3-Me)pyr exhibits
slightly greater activity than tachpyr, and tren(4-Me)pyr
and tren(C-Me)pyr are roughly equivalent to tachpyr.
The tren(5-Me)pyr appeared to be somewhat lower in
activity. Most importantly, all of these compounds were
found to be more active that the parent trenpyr with the
exceptions being the tren(6-Me)pyr and NMe-trenpyr.
Clearly, alkyl substitution of the pyridine ring as well
as on the carbon framework of trenpyr has a profound
effect on cytotoxicity in the HeLa cell line. The tren(3-
Me)pyr, tren(4-Me)pyr, and tachpyr all clearly achieve
complete cell kill at the low concentration of ꢀ7 lM
with a very steep response curve.
Comparing the cytotoxicity of tachpyr to trenpyr direct-
ly by an MTT assay indicated that tachpyr possessed
significantly greater activity at a lower dose of the
reagent to effectively achieve 100% kill of the HeLa cells
(
Figs. 4 and 5). This result was somewhat of a
disappointment as the more flexible chelator trenpyr
was proposed to have faster complexation kinetics with
1
1
40
20
100
8
0
0
0
0
0
6
4
2
We also assessed the sensitivity of normal, non-cancer,
epithelial cells to these compounds. As seen in Table 1,
alkyl substitution had parallel effects on cytotoxic effica-
cy in these cells as in the malignant HeLa cells: for
example, methylation at the 3-position enhanced cyto-
toxicity, whereas methylation at the 6-position dramati-
cally reduced toxicity. Importantly, however, in all cases
normal cells were less sensitive to the cytotoxic effect of
the chelators than the cancer cells, with IC50 increases
from 2 to 4-fold in normal cells relative to HeLa cells
(Table 1). Such preferential activity on tumor cells is
clearly an advantageous feature in the use of such chela-
tors as anti-tumor drugs.
0
5
10
15
20
25
30
35
Concentration (µM)
Figure 4. Effect of test compounds on cell viability. Hela cells were
treated for 72 h with varying concentrations of each test compound
and viability assessed using an MTT assay. Shown are means and
standard errors of three independent experiments, each performed
using six replicate cultures for each point. Tachpyr (d); trenpyr (s);
tren (.); tren(N-Me)pyr (j); tren(N-Me) (,).
1
40
20
The IC50 for the 3-methyl-substituted compound was
slightly superior to tachpyr, whereas the IC50 for the
1
4-methyl- and C-methyl-substituted compounds were
1
00
comparable to tachpyr (Table 1). While the IC50 for
the tren(5-Me)pyr was slightly higher than that of tach-
pyr, and the concentration required for complete cell kill
8
6
4
2
0
0
0
0
Table 1.
a
Compounds
Average IC50
0
HeLa
HMEC
0
20
40
60
80
100
120
140
b
Tachpyr
Trenpyr
Tren-C-pyr
5.3
11.5
30.8
13.5
14.5
15.4
17.2
263.0
Concentration (µM)
13.5
5.3
Figure 5. Effect of test compounds on cell viability. Hela cells were
treated for 72 h with varying concentrations of each test compound
and viability assessed using an MTT assay. Shown are means and
standard errors of three independent experiments, each performed
using six replicate cultures for each point. Tachpyr (d); trenpyr (s);
tren(C-Me)pyr (.); tren(3-Me)pyr (j); tren(4-Me)pyr (h); tren(5-
Me)pyr (Æ); tren(6-Me)pyr (,).
Tren(3-Me)pyr
Tren(4-Me)pyr
Tren(5-Me)pyr
Tren(6-Me)pyr
3.8
5.5
10.1
55.9
a
Average of 2–4 independent experiments.
Average of 12 independent experiments.
b

5964
S. V. Torti et al. / Bioorg. Med. Chem. 13 (2005) 5961–5967
(
ꢀ15 lM) was more than twice that of tachpyr, tren(5-
The source of the increased activity of tren(3-Me)pyr
versus trenpyr may be a result of a steric interaction.
The hypothesized mode of biological cytotoxicity of
tachpyr and its derivatives has been reported to be more
than just simple sequestration of intracellular iron.
These related trenpyr ligands could be expected to exert
an analogous reactivity. Complex formation has been
reported to involve a reduction–oxidation process
wherein the initial complexed Fe(III) is reduced to Fe(II)
while the ligand undergoes an oxidative dehydrogena-
tion process that can repeatedly cycle to ultimately form
Me)pyr was far more active than trenpyr itself. The final
derivative on the group, tren(6-Me)pyr, appears to have
drastically compromised activity, and this result clearly
sets this substitution position off as being a negative site
for modification versus all of the other ring positions.
The impact of methylation of the secondary amines is
also a negative indication and while the biological activ-
ity reported for analogous tachpyr compounds indicated
N-alkylation essentially abrogated the cytotoxicity of
tachpyr, significant activity was noted for NMe-trenpyr
1
0,19
(
Fig. 4). This result does in part substantiate our
the tri-imine analog of tachpyr as its Fe(II) complex.
hypothesis that a more flexible geometry may be able
to obviate some obstacle to complex formation and
thereby retain some measure of biological activity, un-
Thus, the mode of action of these ligands in part in-
cludes formation of the tris-imine and thus a potential
change in the fundamental structure of the complex.
In the case of either tachpyr (6-coordinate), or trenpyr
8
like the N-alkylated tachpyr compounds.
(7-coordinate) this progression to the respective tris-im-
The explanation of the varying degrees of activity prob-
ably lies within the structure of the Fe complex that is
formed, the relative impediments to that formation,
and the mode of action thereafter that appears to be
an effective component of the cytotoxicity of tachpyr
and thus these analogous ligands. If one examines the
known structures of the analogous tachpyr complexes
and thus potentially those of trenpyr, these generally ap-
pear to be octahedral in geometry with the three pyridyl
donors arranged in a spiral twisted arrangement such
ine involves a ꢀflatteningꢁ of the spiral nature of the pyr-
idyl rings and a rotation away from the octahedral or
mono-capped anti-trigonal prism, respectively, to a
1
0,20
more pure trigonal prism type of structures.
A care-
ful examination of the immediate environment about the
methylene protons that undergo elimination in this
mechanism reveals that a potential interaction between
these protons and an alkyl substituent at the 3-position
on the pyridine ring exists. This steric interaction might
be partially relieved through the oxidative dehydrogena-
tion elimination reactions. The resulting sp2 centers that
form then direct the remaining proton away from the al-
kyl substituents at the 3-position on the pyridine ring.
Thus, relief of this interaction potentially provides a
driving force for the elimination reaction and thus an
enhanced degree of cytotoxicity and lowered IC50
values.
1
0,15
that a pair of enantiomeric complexes are formed.
The methyl groups added to trenpyr to form tren(3-Me)-
pyr and tren(4-Me)pyr do not appear to provide any sig-
nificant interactions or impediments in achieving this
conformation about the metal; however, the addition
of the methyl groups clearly provides a significant
enhancement on the biological activity of these com-
pounds. The cytotoxicity of these two compounds ex-
ceeds that of the parent trenpyr and is comparable to
tachpyr itself (Fig. 5 and Table 1). The tren(5-Me)pyr,
while somewhat less active than tachpyr, is also more ac-
tive than trenpyr. The enhanced activity of some of these
compounds may relate in part to their ability to assume
the geometry required for complex formation with ease
or additional driving forces; however, this enhancement
is not easily explained by structural aspects of the metal
complex alone for all of the studied compounds.
While these arguments provide some measure of expla-
nation for the extremes in biological activity for this
group of compounds, tren(3-Me)pyr versus tren(6-Me)-
pyr, likely explanations for the increased biological
activity of the tren(4-Me)pyr, tren(5-Me)pyr, and
tren(C-Me)pyr are unresolved. Differences in lipophilic-
ity and the ability to cross cell membranes may be influ-
enced by the addition of alkyl groups. Electronic
properties of the Fe complexes may also be altered in
some capacity that enhanced the elimination reaction
and the redox chemistry associated with the Fe(III)/
Fe(II) chemistry within the cellular environment. Thus,
it seems clear that other modes of action are in play in
these tren derivatives or at least are amplified in some
capacity by the use of a less pre-organized triamine plat-
form compared to tach.
An explanation for the lack of activity for the tren(6-
Me)pyr seems readily available from the previous litera-
ture reports. An attempt to form the iron Fe complex
for the evaluation of superoxide dismutase mimic activ-
ity failed and was attributed to steric hindrance from the
methyl groups in the 6-position inhibiting complex for-
1
8
mation. Complementary to this hypothesis were the
crystal structures of the Cu(II) complexes of both tach-
In conclusion, while trenpyr itself demonstrated a some-
what decreased activity as a cytotoxic intracellular iron
depletion/sequestration agent versus the analogous tach-
pyr, profound effects were evident from the addition of
methyl groups on the pyridine rings. The 3- and 4-posi-
tions clearly appear to greatly increase biological
activity to be comparable to tachpyr, along with the
5-position to a lesser extent. While tachpyr is possibly
of more fundamental interest, the triamine base
employed therein is considerably less available than tren.
Thus, tren, being a commercially available polyamine
1
5
pyr and analogous tach(6-Me)pyr derivative. The
Cu(II) complex of tachpyr itself was found to be a near
regular six coordinate octahedron with minimal distor-
tion while the complex with tach(6-Me)pyr was five
coordinate with one of the pyridyl rings rotated away
from metal eliminating the sixth nitrogen donor. Evalu-
ation of the in vitro stability of a radio–copper complex
of this latter ligand indicated poor complex stability
thereby providing a clear explanation for the lack of
activity from tren(6-Me)pyr.

S. V. Torti et al. / Bioorg. Med. Chem. 13 (2005) 5961–5967
5965
starting material, might serve as a more accessible entry
point for investigating novel intracellular iron depletion/
sequestration agents. The results reported here clearly
eliminate substitution at the 6-position of the pyridyl ring
in future compounds. Lastly, the results from the 3-Me
derivative appear to indicate that this position is well suit-
ed for further investigation to expand understanding
what limitations on substitution at this position might
exist. The methyl group may well prove to be the limit
in size and any larger group might provide a negative
influence on biological activity. However, to address this
hypothesis fully will require development of a reasonable
entry into 3-alkyl-pyridine-2 carboxaldehydes. All of
these variables are the topic of ongoing studies and will
be reported in due course.
separated and the aqueous layer was extracted with
CHCl3 (2 · 100 mL). The organic layers were com-
bined, dried over MgSO , and filtered. The filtrate
4
was evaporated in vacuo to give the tris-carbamate
(89.95 g, 82%).
1
H NMR (DMSO) d 2.56 (tbd, 6H), 3.21 (tbd, 6H), 5.02
(s, 6H), 5.37 (tbd, 1H), 7.24–7.27 (m, 15H); C NMR
1
3
(DMSO) d 38.93, 53.83, 66.64, 127.95, 128.40, 136.65,
156.93. Mass Spect. (FAB) 549 (M +1); Anal. Calcd
+
for C H N O : C, 65.68; H, 6.61; N, 10.21. Found:
3
0
36
4
6
C, 65.70; H, 6.61; N, 10.15.
4.2. Tris-[N-methyl-2-(aminoethyl)]amine trihydrochlo-
ride (NMe-trenÅ3HCl) (2)
Tren-cbz (89.95 g, 0.164 mol) in THF (250 mL) was
added drop wise to a suspension of LiAlH (30.03 g,
3. Experimental
4
0.791 mol) in THF (700 mL). After complete addition,
the solution was refluxed for 18 h. The reaction mix-
3
.1. Materials and methods
ture was then quenched (carefully with H O (50 mL)
2
All chemicals and solvents were purchased from Fluka,
Sigma, or Aldrich and were used as received. Trenpyr,
and 17.80 M aq KOH (50 mL) water. After filtering,
the precipitate was rinsed with H O, the filtrate was
2
tren(C-Me)pyr,
4
2
3-methyl-pyridine-2-carboxaldehyde,
-methyl-pyridine-2-carboxaldehyde, 5-methyl-pyridine-
-carboxaldehyde, and 6-methoxy-pyridine-2-carboxalde-
adjusted to pH 8.0 with HCl (aq), and extracted with
CH Cl (2 · 200 mL). The organic layers were com-
2
2
bined, dried over MgSO , and filtered. The filtrate
4
1
2,17
hyde were prepared as previously reported.
was concentrated in vacuo to yield an oily brown res-
idue. This oil was then purified by vacuum distillation
collecting the product at 130 ꢀC, 5.5 mmHg. The pure
oil was then converted to the Tris–HCl salt by bub-
bling HCl (g) through a solution in dioxane. The
product was obtained as a precipitated yellow powder,
21.45 g (44%).
1
13
H and C NMR were obtained using a Varian Gemini
00 instrument and chemical shifts are reported in parts
3
per million on the d scale relative to TMS, TSP, or sol-
vent. Proton chemical shifts are annotated as follows:
parts per million [multiplicity, integral, coupling con-
stant (Hz)). Chemical ionization mass spectra (CI-MS)
were obtained on a Finnegan 3000 instrument. Fast
atom bombardment mass spectra (FAB-MS) were taken
on a Extrel 400. Elemental analyses were obtained from
Atlantic Microlabs (Norcross, GA, USA).
1
H NMR (D O) d 2.38 (s, 9H), 2.77 (t, J = 6.90, 6H),
2
1
3
2.95 (t, J = 6.90, 6H); C NMR (D O) d 35.13, 46.74,
2
+
50.90; Mass Spect. (FAB) 189 (M +1); Anal. Calcd
for C H N (H O)(HCl)2.5: C, 36.34; H, 9.66; N,
9
24
4
2
18.83. Found: C, 36.77; H, 9.69; N, 18.51.
Analytical HPLC was performed using a Beckman sys-
tem with Model 114M pumps controlled by System
Gold software and a Model 165 dual wavelength detec-
tor set at 254 and 280 nm. Chromatography was per-
formed using an Altex C-18 reverse phase column
4.3. Tris[N-methyl-N-2-pyridylmethylene-2-(aminoeth-
yl)]amine (NMe-trenpyr) (5)
A suspension of NMe-trenÆ3HCl (0.50 g, 1.69 mmol)
(
of 0–100% B/25 min (solvent A = 0.05 M Et N/HOAc,
pH 5.5, solvent B = MeOH) at 1.0 mL/min.
5 lm particles, 4.6 · 250 mm) and a binary gradient
and Na CO (3.04 g, mmol) in DMF (30 mL) was stir-
2
3
red at 80 ꢀC as 2-chloromethylpyridine HCl (0.89 g,
5.43 mmol) in DMF (10 mL) was added. Stirring was
continued for 18 h after which the reaction mixture
was evaporated in vacuo. The residue was taken up in
CHCl (40 mL) and extracted with H O (2 · 30 mL),
3
4
. Chemistry—ligand synthesis
3
2
saturated brine solution (30 mL). The organic layer
was then dried over Na SO and filtered. The filtrate
4
(
.1. Tris[N-carbobenzyloxy-(2-aminoethyl)]amine
tren-cbz)
2
4
was evaporated to give leave the product as a red oil
0.54 g, 69%).
(
A solution of tren (29.20 g, 0.200 mol) in benzene
225 mL) and H O (100 mL) was stirred at 5 ꢀC for
1
(
1
H NMR (CDCl ) d 2.24 (s, 9H), 2.53 (br t, 6H), 2.66 (br
3
2
h. Benzylchloroformate (52.96 g, 0.312 mol) was
t, 6H), 3.65 (s, 6H), 7.13 (t, J = 6.75, 3H), 7.34 (d,
J = 7.80, 3H), 7.62 (t, J = 9.15, 3H), 8.53 (d, J = 5.10,
3H); C NMR (CDCl ) d 42.58, 52.56, 55.21, 63.83,
then added drop wise to the solution. After complete
addition, a second portion of benzylchloroformate
1
3
3
(
53.14 g, 0.313 mol) was added simultaneously with
121.42, 122.53, 135.83, 148.60, 159.01; Mass Spect.
(FAB) 462 (M +1); Anal. Calcd for C H N (H O)1.5:
C, 66.36; H, 8.66; N, 20.06. Found: C, 65.82; H, 8.44; N,
20.15.
+
an 18.61 M aqueous solution of KOH (35 mL). The
reaction mixture was stirred for 2 h at 5 ꢀC and then
for 18 h at room temperature. The layers were
2
7
39
7
2

5966
S. V. Torti et al. / Bioorg. Med. Chem. 13 (2005) 5961–5967
+
(FAB) 462 (M +1); Anal. Calcd for C H N : C,
4.4. Tris[N-(2-(3-methyl-pyridylmethylene))-2-aminoeth-
yl]amine (tren(3-Me)pyr)
2
7
39
7
70.23; H, 8.53; N, 21.24. Found: C, 70.60; H, 8.89; N,
1.16. HPLC t = 19.22 min.
2
R
Tren (1.00 g, 6.845 mmol) was reacted with 3-methyl-
pyridine-2-carboxaldehyde (2.486 g, 20.55 mmol) in ben-
zene (150 mL) at reflux and driven to completion by use
of a Dean–Stark trap for 18 h. The reaction solution was
decanted and concentrated to an oil by rotary evapora-
tion. The isolated crude imine was taken up in methanol
4.7. Tris[N-(2-(6-methyl-pyridylmethylene))-2-aminoeth-
yl]amine (tren(6-Me)pyr)
Tren (3.00 g, 6.845 mmol) was reacted with 6-methyl-
pyridine-2-carboxaldehyde (7.46 g, 20.55 mmol) in ben-
zene (150 mL) and the crude imine was reduced with
sodium borohydride (0.80 g, 21 mmol) in methanol
(125 mL) to provide the the final product as a dark oil
(2.90 g, 92%).
(
2
125 mL) and reduced with sodium borohydride (0.80 g,
1 mmol) for 18 h. The solution was taken to dryness by
rotary evaporation and the residue partitioned between
CHCl (125 mL) and 5% aqueous NaHCO (125 mL)
3
3
with vigorous mixing for 2 h. Thereafter, the mixture
was poured into a separatory funnel, and the CHCl3
solution retained. The aqueous was extracted with
CHCl (100 mL), the CHCl portion combined, dried
1
H NMR (DMSO-d ) d 7.46 (t, 1H, J = 7.8), 7.099 (d,
6
1H, J = 7.8), 6.968 (d, 1H, J = 7.5), 3.863 (s, 2H), 3.14
(br s, 1H), 2.766 (t, 2H, J = 5.1), 2.677 (t, 2H, J = 5.1),
2.485 (s, 3H); C NMR (DMSO-d ) d 158.56, 157.90,
3
3
1
3
over MgSO , filtered, and rotary evaporated to leave
4
6
the product as a dark oil, that was dried in vacuo and
stored under refrigeration (2.78 g, 88%).
136.71, 121.51, 119.19, 54.92, 54.16, 47.28, 24.50; Mass
Spect. (FAB) 462 (M +1); Anal. Calcd for C H N :
+
2
7
39
7
C, 70.23; H, 8.53; N, 21.24. Found: C, 70.50; H, 8.71;
N, 21.56. HPLC t = 19.03 min.
1
H NMR (DMSO-d ) d 8.284 (d, 1H, J = 5.1), 7.356 (d,
6
R
1
H, J = 7.2), 7.012 (dd, 1H, J = 7.5, 4.5), 4.56 (br s, 1H),
.924 (s, 2H), 2.901 (t, 2H, J = 5.7), 2.779 (t, 2H,
3
J = 5.7), 2.251 (s, 3H);
5. Biological methods
1
3
C NMR (DMSO-d ) d
6
1
4
53.16, 146.24, 137.43, 130.89, 121.79, 54.00, 51.37,
7.43, 18.02; Mass Spect. (FAB) 462 (M +1); Anal.
5
.1. In vitro cellular proliferation assay
+
Calcd for C H N : C, 70.23; H, 8.53; N, 21.24. Found:
39
C, 70.37; H, 8.56; N, 21.12. HPLC t = 17.89 min.
R
2
7
7
Hela cells were obtained from the American Type Cul-
ture Collection and grown in a humidified 5% CO atmo-
2
sphere at 37 ꢀC in DME medium (Gibco BRL)
supplemented with 10% fetal bovine serum and penicil-
lin/streptomycin. Human mammary epithelial cells
4.5. Tris[N-(2-(4-methyl-pyridylmethylene))-2-amino-
ethyl]amine (tren(4-Me)pyr)
(
HMEC) were obtained from Cambrex and propagated
Tren (1.00 g, 6.845 mmol) was reacted with 4-methyl-
pyridine-2-carboxaldehyde (2.486 g, 20.55 mmol) in
benzene (150 mL) and the crude imine was reduced with
sodium borohydride (0.80 g, 21 mmol) in methanol
in a humidified 5% CO atmosphere at 37 ꢀC in mamma-
2
3
ry epithelial basal medium (Cambrex). 2–5 · 10 cells
were plated in 96-well tissue culture dishes and allowed
to attach overnight before test compounds were added.
Six replicate cultures were used for each point. After
(
(
125 mL) to provide the final product as a dark oil
2.62 g, 83%).
7
3
2 h, viability was assessed using an MTT assay in which
-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium
1
H NMR (DMSO-d ) d 8.336 (d, 1H, J = 4.8), 7.139 (s,
6
bromide is added to the medium and the formation of a
reduced product is assayed by measuring the optical den-
sity at 560/650 nm after 3 h. Color formation is propor-
tional to viable cell number. In some cases, MTS
(3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-
1
J = 5.4), 2.695 (t, 2H, J = 5.7), 2.292 (s, 3H);
H), 6.932 (d, 1H, J = 7.2), 3.910 (s, 2H), 2.801 (t, 2H,
C
1
3
NMR (DMSO-d ) d 158.72, 149.03, 147.63, 123.33,
21
6
1
4
8
23.05, 54.52, 53.98, 47.14, 21.02; Mass Spect. (FAB)
+
62 (M +1); Anal. Calcd for C H N : C, 70.23; H,
.53; N, 21.24. Found: C,70.39; H, 8.53; N, 21.12.
2
7
39
7
2-(4-sulfophenyl)-2H-tetrazolium, inner salt; Promega),
a water soluble substrate, was used in place of MTT.
HPLC t = 20.12 min.
R
4.6. Tris[N-(2-(5-methyl-pyridylmethylene))-2-aminoeth-
yl]amine (tren(5-Me)pyr)
Acknowledgment
This work was supported in part by grant No. DK
57781 (S.V.T.) from the National Institutes of Health.
This research was supported in part by the Intramural
Research Program of the NIH, National Cancer Insti-
tute, Center for Cancer Research.
Tren (1.00 g, 6.845 mmol) was reacted with 5-methyl-pyr-
idine-2-carboxaldehyde (2.486 g, 20.55 mmol) in benzene
(
150 mL) and the crude imine was reduced with sodium
borohydride (0.80 g, 21 mmol) in methanol (125 mL) to
provide the final product as a dark oil (2.74 g, 87%).
References and notes
1
H NMR (DMSO-d ) d 8.4–8.28 (m, 1H), 7.45–7.35 (m,
6
1
2
3
1
H), 7.22–7.15 (m,1H), 3.868 (s, 2H), 3.235 (br s, 1H),
.741 (q, 2H, J = 5.4), 2.647 (t, 2H, J = 5.4), 2.274 (s,
1
. Faulk, W. P.; His, B. L.; Stevens, P. J. Lancet 1980, 2,
90–392.
2. Seymour, G. J.; Walsh, M. D.; Lavin, M. F.; Struttun, G.;
3
1
3
H); C NMR (DMSO-d ) d 156.16, 149.60, 136.99,
6
31.25, 121.89, 54.37, 54.01, 47.09, 18.13; Mass Spect.
Gardiner, R. A. Urol. Res. 1987, 15, 341–344.

S. V. Torti et al. / Bioorg. Med. Chem. 13 (2005) 5961–5967
5967
3
4
5
. Le, N. T. V.; Richardson, D. R. Biochim. Biophys. Acta
2
13. Zhao, R.; Planalp, R. P.; Ma, R.; Greene, B.; Jones, B. T.;
Brechbiel, M.; Torti, F. M.; Torti, S. V. Biochem.
Pharmacol. 2004, 67, 1677–1688.
14. Ye, N.; Park, G.; Przyborowska, A. M.; Sloan, P. E.;
Clifford, T.; Bauer, C. B.; Broker, G. A.; Rogers, R. D.;
Ma, R.; Torti, S. V.; Brechbiel, M. W.; Planalp, R. P.
J. Chem. Soc., Dalton Trans. 2004, 1304–1311.
15. Park, G.; Dadachova, E.; Przyborowska, A.; Lai, S.;
Broker, G.; Rogers, R. D.; Planalp, R. P.; Brechbiel, M.
W. Polyhedron 2001, 20, 3155–3163.
16. Schmidt, H.; Lensink, C.; Xi, S. K.; Verkade, J. G. Z.
Anorg. Allg. Chem. 1989, 578, 75–80.
17. Camphausen, K.; Sproull, M.; Tanawa, S.; Sankineni, S.;
Scott, T.; M e´ nard, C.; Coleman, C. N.; Brechbiel, M. W.
Bioorg. Med. Chem. 2003, 11, 4287–4291.
18. Nagano, T.; Hirano, T.; Hirobe, M. Free Radical Res.
Commun. 1991, 12–13, 221–227.
19. Park, G.; Przyborowska, A. M.; Ye, N.; Tsoupas, N. M.;
Bauer,C.B.;Broker,G.A.;Rogers,R.D.;Brechbiel,M.W.;
Planalp, R. P. J. Chem. Soc. Dalton Trans. 2003, 318–324.
20. Morgenstern-Badarau, I.; Lambert, F.; Renault, J. P.;
Cesario, M.; Marechal, J.-D.; Maseras, F. Inorg. Chim.
Acta 2000, 297, 338–350.
002, 1602, 31–46.
. Hann, H. W.; Stahlhut, M. W.; Rubin, R.; Maddrey, W.
C. Cancer 1992, 70, 2051–2056.
. Kemp, J. D.; Smith, K. M.; Kanner, L. J.; Gomez, F.;
Thorson, J. A.; Naumann, P. W. Blood 1990, 76, 991–
1
025.
. Richardson, D. R.; Tran, E. H.; Ponka, P. Blood 1995, 86,
295–4306.
. Chaston, T. B.; Lovejoy, D. B.; Watts, R. N.; Richardson,
D. R. Clin. Cancer Res. 2003, 9, 402–414.
. Torti, S. V.; Torti, F. M.; Whitman, S. P.; Brechbiel, M.
W.; Park, G.; Planalp, R. P. Blood 1998, 92, 1384–1389.
. Abeysinghe, R.; Greene, B. T.; Haynes, R.; Willingham,
M. C.; Naryanan, V. L.; Sausville, E. A.; Planalp, R. P.;
Brechbiel, M. W.; Torti, F. M.; Torti, S. V. Carcinogenesis
6
7
8
9
4
2
0. Park, G.; Lu, F. H.; Ye, N.; Brechbiel, M. W.; Torti, S. V.;
001, 22, 1607–1614.
1
Torti, F. M.; Planalp, R. P. J. Biol. Inorg. Chem. 1998, 3,
4
49–457.
1. Bowen, T.; Planalp, R. P.; Brechbiel, M. W. Bioorg. Med.
Chem. Lett. 1996, 6, 807–810.
2. Ma, D.; Lu, F.; Overstreet, T.; Milenic, D.; Brechbiel, M.
W. Nucl. Med. Biol. 2002, 29, 91–105.
1
1
21. Mosmann, T. J. Immunol. Methods 1983, 65, 55–63.