Synthesis of a novel nitroimidazole-spermidine derivative as a tumor-targeted hypoxia-selective cytotoxin

Article abstract of DOI:10.1016/j.bmcl.2003.12.100

A four-step synthesis of (R,S)-N⁴-[3-(2-nitro-1-imidazolyl)-2- hydroxypropyl]-spermidine trihydrochloride (4) is described and the utilization of the polyamine active transport system for the uptake of this compound in cells is demonstrated. Thus, V79 cells pretreated with an inhibitor of spermidine biosynthesis, α-difluoromethylornithine (DFMO), are ca. 2-fold more sensitive to 4 under hypoxic conditions, compared to untreated cells. Similarly, radiosensitization of hypoxic V79 cells by 4 is improved in DFMO-pretreated cells.

Full text of DOI:10.1016/j.bmcl.2003.12.100

Bioorganic & Medicinal Chemistry Letters 14 (2004) 1519–1522
Synthesis of a novel nitroimidazole-spermidine derivative as a
tumor-targeted hypoxia-selective cytotoxin
Maria V. Papadopoulou,* Howard S. Rosenzweig and William D. Bloomer
Evanston Northwestern Healthcare, Department of Radiation Medicine, Evanston, IL 60201, USA
Received 15 July 2003; accepted 31 December 2003
Abstract—A four-step synthesis of (R,S)-N4-[3-(2-nitro-1-imidazolyl)-2-hydroxypropyl]-spermidine trihydrochloride (4) is described
and the utilization of the polyamine active transport system for the uptake of this compound in cells is demonstrated. Thus, V79
cells pretreated with an inhibitor of spermidine biosynthesis, a-difluoromethylornithine (DFMO), are ca. 2-fold more sensitive to 4
under hypoxic conditions, compared to untreated cells. Similarly, radiosensitization of hypoxic V79 cells by 4 is improved in
DFMO-pretreated cells.
# 2004 Elsevier Ltd. All rights reserved.
Polyamines such as N1-(3-aminopropyl)-1,4-butanedi-
amine (spermidine), 1,4-butanediamine (putrescine), and
N1,N4-bis(3-aminopropyl)-1,4-butanediamine (spermine)
are a requirement for the optimal growth and replica-
tion of various cell types and are present in higher con-
centrations in rapidly proliferating cells.1À4 Thus,
polyamine analogues that vary slightly in structure from
natural polyamines might compete with naturally
occurring polyamines for critical cellular binding sites.
In addition, a wide range of mammalian and bacterial
cells have an active, energy-dependent and carrier-
mediated polyamine transport system which seems to be
regulated by intracellular polyamine concentrations.5À8
Porter et al. also demonstrated that the N1 and N8 pri-
mary amines of spermidine are the most critical deter-
minants in transport specificity, leaving the central N4
secondary amine for derivatization.5 Utilization of the
polyamine transport system can afford selective accu-
mulation of polyamine analogues in neoplastic tissues
and therefore, presents a very attractive anticancer
chemotherapeutic strategy.9À15 We tried to exploit this
methodology in the development of novel bioreductive
drugs as cytotoxins and radiosensitizers of hypoxic cells,
because hypoxia in tumors is a limiting factor for suc-
cess in cancer radio/chemotherapy16À19 and because
such an approach could increase effectiveness due to
selective uptake and DNA-binding in tumor cells. In
this paper we report the synthesis of N4-[3-(2-nitro-1-
imidazolyl)-2-hydroxypropyl]spermidine trihydrochloride
(4) and its evaluation as a hypoxia-selective cytotoxin
and radiosensitizer in V79 cells. The 2-nitroimidazole
group was selected on the basis of its one-electron
reduction potential, known to be in the ‘correct’ range
of À330 to À450 mV, which allows for its exclusive
reduction under hypoxic conditions and has been
previously incorporated in a variety of bioreductive
drugs.20
The synthesis of 4 is outlined in Scheme 1. Protection of
the terminal amino groups of N4-benzyl-spermidine (1,
Aldrich) with di-tert-butyldicarbonate, followed by
debenzylation through hydrogenation in dry MeOH
with 5% Pd–C catalyst, at 55 psi (rt, overnight) led to
N1,N8-diBoc-spermidine, (2), in 98% yield. Protected
spermidine (2) was coupled with 1-(oxiranylmethyl)-2-
nitroimidazole in dry EtOH, under reflux for 12 h under
an inert atmosphere, to afford (R,S)-N4-[3-(2-nitro-1-
imidazolyl)-2-hydroxypropyl]-N1,N8-diBoc-spermidine,21
(3), in 65% yield. Compound (3) was separated as an oil
by preparative alumina TLC with 60:40 hexane:ethyl
acetate eluent (Rf=0.54). Finally, compound (3) was
deprotected and converted to its trihydrochloride salt
(4), in 97% yield, by dissolving it in dry ether and
treating with a HCl/dioxane solution.22 1-(Oxira-
nylmethyl)-2-nitroimidazole was synthesized by alkyl-
ation of 2-nitroimidazole with (R,S)-1-bromo-2,3-
* Corresponding author. Tel.: +1-847-570-2262; fax: +1-847-570-
1878; e-mail: m-papadopoulou@northwestern.edu
0960-894X/$ - see front matter # 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2003.12.100

1520
M. V. Papadopoulou et al. / Bioorg. Med. Chem. Lett. 14 (2004) 1519–1522
Scheme 1. Reagents and conditions: (a) (t-Boc)2O, ET3N, THF, rt (85%); (b) Pt/C, H2, MeOH (98%); (c) 1-(oxiranylmethyl)-2-nitroimidazole,
EtOH, reflux (65%); (d) HCl gas in dioxane, THF (97%).
epoxypropane, followed by base-catalyzed ring closure,
generally according to the method of Beaman et al.23
cells not pretreated with DFMO, did not hinder the
uptake of 4, resulting thus in unaltered CT50 values
(Table 1). On the other hand, 4 mM spermidine given in
parallel with 4 in DFMO-pretreated cells, caused a
slight but statistically significant increase in the aerobic
toxicity of 4, which was further increased by using 0.5
mM spermidine. To the contrary, the hypoxic toxicity
of 4 was slightly decreased by 4 mM spermidine
(Table 1). Therefore, the potentially antagonistic effect
of spermidine is not clear at this moment and should be
further clarified. However, one can speculate that per-
haps different polyamine-binding proteins on the sur-
face of V79 cells are responsible for the transport of
compound 4 and spermidine. For instance, in the L1210
murine leukemia and A549 human lung carcinoma cells
many polyamine-binding proteins have been found.28
The cytotoxicity of 4 was assessed in suspended V79
Chinese hamster lung fibroblasts under aerobic or
hypoxic exposures, by using the clonogenic assay.24
Cytotoxicity of 4 was also determined after aerobic
pretreatment of the same cells (while in exponential
growth as monolayers) for 48 h with dl-a-difluoro-
methylornithine (DFMO), an inhibitor of ornithine
decarboxylase (ODC), the first and rate limiting enzyme
involved in the polyamine biosynthetic pathway.25,26
The radiosensitizing ability of 4 was also tested under
hypoxic or aerobic conditions in suspended V79 cells,
previously pretreated or not with 100 mM DFMO for
48 h as monolayers.
The presence of an active polyamine transport system in
V79 cells was tested by measuring the uptake of radio-
labeled 14C-spermidine (0.1–1.0 mCi, specific activity of
ca. 100 mCi/mmol) before or after polyamine depletion
by exposing the cells, while in monolayers, to 100 mM
DFMO for 48 h and then measuring radioactivity by
scintillation counting. Thus, the uptake of 14C-spermi-
dine was increased by 2.1 fold in DFMO-pretreated
cells (Fig. 1).
Compound 4, at 0.6 and 0.8 mM, was able to increase
the responsiveness of hypoxic DFMO-pretreated V79
cells to ionizing radiation, by a factor of 1.5 and 2.0,
respectively (Fig. 2). The corresponding factor in
hypoxic cells not pretreated with DFMO was 1.0 and
Cells were relatively insensitive to compound 4, since
cytotoxicity was obtained at mM concentrations (Table
1) whereas the nitroimidazole-based bioreductive com-
pound NLCQ-1, {4-[3-(2-nitro-1-imidazolyl)propyl-
amino]-7-chloroquinoline hydrochloride} which enters
cells through passive diffusion and targets DNA
through weak intercalation, demonstrates CT50H values
at mM concentrations in various cell-lines.27 However, 4
was still a more potent but less selective hypoxic cyto-
toxin than misonidazole [1-(2-nitro-1-imidazolyl)-2-
hydroxy-3-methoxypropane], a bioreductive compound
which also enters cells through passive diffusion but
does not target DNA. Thus, misonidazole demonstrates
a CT50H of 7.5Æ0.5 mM and a CT50A of 120.6Æ7.3
mM, in V79 cells. The cytotoxicity of 4 was statistically
increased in DMFO-pretreated cells (1 mM, 48 h), to a
similar degree under both aerobic or hypoxic condi-
tions, consistent with an increased uptake of 4 through
the polyamine transport system. Similar increases of the
IC50 values have been observed before in DFMO-trea-
ted L1210 cells for several polyamine conjugates.15
However, 4 mM of spermidine co-administered with 4 in
Figure 1. 14C-Spermidine uptake by V79 cells. Suspensions of V79
cells (105 cells/mL, 1 mL) in duplicate were incubated at 37 ꢀC for 30
min with various 14C-spermidine concentrations (0.1–1.0 mCi, the bulk
of it being cold spermidine). A second set of cells, pretreated as
monolayers with DFMO (100 mM, 48 h) were similarly treated with
14C-spermidine. Following incubation, the cells were prepared for
scintillation.

M. V. Papadopoulou et al. / Bioorg. Med. Chem. Lett. 14 (2004) 1519–1522
Table 1. Aerobic and hypoxic toxicity of compound 4 in V79 cells
1521
a
b
CT50A
CT50H
Ratioc
ÀDFMO
2.60Æ0.24
1.28Æ0.20
2.0
2.2
2.3
1.3
+DFMO (1 mM)d
1.63Æ0.18 (P<0.001)
2.64Æ0.23 (P>0.05)
1.15Æ0.06 (P<0.005)
0.63Æ0.02 (P<0.001)
0.74Æ0.07 (P<0.005)
1.15Æ0.09 (P>0.05)
0.88Æ0.07 (P<0.05)
ND
ÀDFMO+spermidine (4 mM)e
+DFMO (0.1 mM)+spermidine (4 mM)
+DFMO (1 mM)+ spermidine (0.5 mM)
a,bCT50 values determined under aerobic (A) or hypoxic (H) conditions in suspended cells and represent the drug concentration (mM) required to
reduce clonogenicity to 50% of corresponding controls. Values are means ÆSEM (4 determinations).
c
Ratio of CT50 values in air and N2 (CT50A/CT50H).
Cells were exposed to DFMO under aerobic conditions for 48 h, while in monolayers.
d
e
Spermidine was administered to the cells in parallel and under the same conditions with 4. P values were determined by using the Student’s t-test
in comparisons with corresponding controls.
References and notes
1. Horn, Y.; Beal, S. L.; Walach, N.; Lubich, W.; Spigel, L.;
Marton, L. Cancer Res. 1982, 42, 3248.
2. Pegg, A. E.; McCann, P. P. Am. J. Physiol. (Cell Physiol.)
1982, 12, C212.
3. Tabor, C. W.; Tabor, H. Annu. Rev. Biochem. 1984, 53,
749.
4. Casero, R. A., Jr.; Woster, P. M. J. Med. Chem. 2001, 44, 1.
5. Porter, C. W.; Bergeron, R. J.; Stolowich, N. J. Cancer
Res. 1982, 42, 4072 1982.
6. Seiler, N.; Delcros, J. G.; Moulinoux, J. P. Int. J. Bio-
chem. Cell Biol. 1996, 28, 843.
7. Phanstiel, O., IV; Price, H. L.; Wang, L.; Juusola, J.;
Kline, M.; Shah, S. M. J. Org. Chem. 2000, 65, 5590.
8. Wang, L.; Price, H. L.; Juusola, J.; Kline, M.; Phanstiel,
O., IV J. Med. Chem. 2001, 44, 3682.
9. Edwards, M. L.; Snyder, R. D.; Stemerick, D. M. J. Med.
Chem. 1991, 34, 2414.
10. Stark, P. A.; Thrall, B. D.; Meadows, G. G.; Abdel-
Monem, M. M. J. Med. Chem. 1992, 35, 4264.
11. Kramer, D. L.; Miller, J. T.; Bergeron, R. J.; Khomutov,
R.; Khomutov, A.; Porter, C. W. J. Cell. Physiol. 1993,
155, 399.
Figure 2. Radiosensitization of V79 cells by compound 4. Cells were
exposed or not to DFMO (1 mM, 48 h) as monolayers. Then, cells
12. Zhuo, J.-C.; Cai, J.; Soloway, A. H.; Barth, R. F.; Adams,
D. M.; Ji, W.; Tjarks, W. J. Med. Chem. 1999, 42, 1282.
13. Cullis, P. M.; Green, R. E.; Merson-Davies, L.; Travis, N.
Chem. Biol. 1999, 6, 717.
were exposed to 0.6 or 0.8 mM of 4 under aerobic or hypoxic condi-
tions (37 ꢀC, 4 h) in suspension, before irradiation at 25 ꢀC. Cells were
then processed for clonogenicity.
14. Carlisle, D. L.; Devereux, W. L.; Hacker, A.; Woster,
P. M.; Casero, R. A., Jr. Clin. Cancer Res. 2002, 8, 2684.
15. Wang, C.; Delcros, J.-G.; Biggerstaff, J.; Phanstiel, O., IV
J. Med. Chem. 2003, 46, 2663.
16. Hockel, M.; Schlenger, K.; Mitze, M.; Schaffer, U.;
Vaupel, P. Seminars in Radiat. Oncol. 1996, 6, 3.
17. Siemann, D. W. Int. Radiat. Oncol. Biol. Phys 1984, 10,
1585.
1.2 at 0.6 and 0.8 mM of 4, respectively. No radio-
sensitization was observed by 4 in aerobic V79 cells
regardless of pretreatment with DFMO (Fig. 2). Mis-
onidazole at 0.8 mM increased the responsiveness of V79
cells by a factor of 1.8, regardless of pretreatment with
DFMO, as was expected for compounds that do not utilize
the active polyamine transport system to enter cells.
18. Brown, J. M. Mol. Med. Today 2000, 6, 157.
19. Bussink, J.; Kaanders, J. H. A. M.; van der Kogel, A. J.
Radiother. Oncol. 2003, 67, 3.
The above results suggest that optimized nitroimida-
zole-based polyamine derivatives, obtained perhaps
through terminal amine alkylation, could be of interest
as hypoxia-selective cytotoxins and radiosensitizers for
their selective uptake by cancer cells through the poly-
amine transport system. Hypoxic potency rather than
selectivity would be more meaningful in such systems.
20. Denny, W. A. Eur. J. Med. Chem. 2001, 36, 577.
1
21. Compound 3: Clear oil; yield 65%; H NMR (500 MHz)
(CDCl3) d 7.27 (s, 1H), 7.07 (s, 1H), 4.86 (s, broad, 1H),
4.74 (s, broad, 1H), 4.65 (d, J=13.5 Hz, 1H), 4.15–4.06
(m, 2H), 3.97 (m, 1H), 3.22–3.03 (m, 4H), 2.60–2.34 (m,
6H), 1.58 (m, 2H), 1.43–1.39 (m, 22H). HRMS (VG 70-
250SE mass spectrometer): calcd for C23H43N6O7,
[M+H]+: m/z 515.3193. Found: 515.3191 (100%). Anal.
(C23H42N6O7) C, H, N.
Acknowledgements
22. Compound 4: White solid, very hygroscopic; yield 97%;
1H NMR (500 MHz) (D2O, 4.8 d) d 7.49 (s, 1H), 7.24 (s,
1H), 4.51 (t, broad, 1H), 4.37 (dd, J=11.7 Hz, 3.3 Hz,
1H), 3.38 (t, broad, 6H), 3.08 (m, 5H), 2.18 (m, 2H), 1.70–
1.90 (m, 4H). FAB (VG 70-250SE mass spectrometer):
The authors would like to thank Ms. Mira K. Rao for
her technical assistance and the Merrell Dow Research
Institute, Cincinnati, for generously providing DFMO.

1522
M. V. Papadopoulou et al. / Bioorg. Med. Chem. Lett. 14 (2004) 1519–1522
calcd for C13H29Cl2N6O3 [M-Cl]+: m/z 387; Found: 387.
Anal. (C13H29N6Cl3O3) C, H, N, Cl.
26. Marton, L. J.; Pegg, A. E. Annu. Rev. Pharmacol. Toxicol.
1995, 35, 55.
23. Beaman, A. G.; Tautz, W.; Duschinsky, R. Antimicrob.
Agents Chemother. 1986, 520.
27. Papadopoulou, M. V.; Ji, M.; Rao, M. K.; Bloomer,
W. D. Oncol. Res. 2000, 12, 185.
24. Papadopoulou, M. V.; Ji, M.; Rao, M. K.; Bloomer,
W. D. Oncol. Res. 1996, 8, 425.
28. Felschow, D. M.; Mi, A.; Stanek, J.; Frei, J.; Porter, C. W.
Biochem. J. 1997, 328, 889.
25. Metcalf, B. W.; Bey, P.; Danzin, C.; Jung, M. J.; Casara,
P.; Vevert, J. P. J. Am. Chem. Soc. 1978, 100, 2551.