RETRACTED ARTICLE: Synthesis and bioevaluation of aryl-guanidino polyamine conjugates targeting the polyamine transporter

Article abstract of DOI:10.1016/j.bmcl.2010.09.069

Aryl-guanidino polyamine conjugates were prepared to evaluate their recognition for polyamine transporter (PAT) via a-difluoromethylornithine (DFMO) and spermidine (SPD)-treated B16 cell lines. The potent synergistic effects of DFMO on guanidino polyamine conjugates indicated that the presence of DFMO strongly facilitates the transport of conjugates into cells via PAT on cell membrane. The apoptotic mechanisms of triamine conjugates 10 and 1 (with and without guanidine groups) revealed that they induced apoptosis of Hela cells through the mitochondrial pathway associated with lysosomes, while DFMO strongly synergizes the function of 10 without changing the apoptotic route. Taken together, guanidino polyamine conjugates can target PAT for transport as normal polyamine ones, and the presence of guanidine in the polyamine vectors does not seem to alter the cellular targets of the conjugates, which may depend mainly on the pharmacophore.

Full text of DOI:10.1016/j.bmcl.2010.09.069

Bioorganic & Medicinal Chemistry Letters 20 (2010) 6421–6425
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis and bioevaluation of aryl-guanidino polyamine conjugates
targeting the polyamine transporter
Jianhong Wang a, Zhiyong Chen a, Songqiang Xie b, , Jin Zhao a, Chaojie Wang a,
⇑
⇑
a Key Laboratory of Natural Medicine and Immuno-Engineering, Henan University, Kaifeng 475004, China
b Institute of Chemistry and Biology, Henan University, Kaifeng 475001, China
a r t i c l e i n f o
a b s t r a c t
Article history:
Aryl-guanidino polyamine conjugates were prepared to evaluate their recognition for polyamine trans-
porter (PAT) via a-difluoromethylornithine (DFMO) and spermidine (SPD)-treated B16 cell lines. The
potent synergistic effects of DFMO on guanidino polyamine conjugates indicated that the presence of
DFMO strongly facilitates the transport of conjugates into cells via PAT on cell membrane. The apoptotic
mechanisms of triamine conjugates 10 and 1 (with and without guanidine groups) revealed that they
induced apoptosis of Hela cells through the mitochondrial pathway associated with lysosomes, while
DFMO strongly synergizes the function of 10 without changing the apoptotic route. Taken together, gua-
nidino polyamine conjugates can target PAT for transport as normal polyamine ones, and the presence of
guanidine in the polyamine vectors does not seem to alter the cellular targets of the conjugates, which
may depend mainly on the pharmacophore.
Received 10 August 2010
Revised 5 September 2010
Accepted 14 September 2010
Available online 17 September 2010
Keywords:
Polyamine conjugate
Guanidine
Polyamine transporter
Apoptosis
Ó 2010 Elsevier Ltd. All rights reserved.
The biogenic polyamines, spermidine, spermine and putrecine,
are small organic molecules essential for regulation of cell growth,
proliferation and differentiation. It is well accepted that tumor
cells uptake these polyamines via the polyamine transporter
(PAT) in greater amounts than do healthy cells, from which gener-
ates the idea that some toxic agents coupled with polyamines
might have improved anti-cancer effects.1,2 Some attempts have
been made to use the polyamine moiety to ferry cytotoxic agents
that might otherwise be poorly transported into tumor cells. Re-
cently, it was found that some synthetic polyamines with the back-
bone modifications are better recognized by PAT than their native
counterparts.3,4 The polyamine-drug conjugates are, therefore,
promising chemical systems for the development of more efficient
anti-tumor agents.5–9
To date, several characteristic properties of polyamine transport
have been determined based on series of biochemical assays.2 In
one aspect, the PAT affinity of polyamine motifs is modulated by
their minor structural changes. With anthracene as a cargo, the
homospermidine conjugate (1 in Fig. 1) exhibited excellent PAT
recognition and intriguing biological properties.10 With 1 as a lead
compound, further research revealed that the N-ethyl substituted
homospermidine in conjugate 2 may serve as a less toxic poly-
amine carrier than the parent homospermidine.11 Recently we
replaced the terminal amino group of diamines12 or triamines13
with hydrazine groups, and found that the resultant conjugates
such as 3 could harness PAT to enter the cells. These findings
prompted us to look for more modified polyamines targeting PAT.
Various polyamine analogues containing terminal guanidino
groups have been designed to possess biochemical attributes of
antitumor, antitrypanosomal, and antidiarrheal activities well
amplified with respect to their parent polyamines.14 We thus ex-
tended our screen to this more basic group, guanidine.15 During
our work, it was reported that the transition from amine to guani-
dine might be one promising strategy for better antiproliferative
activities, and the same compound 4 was synthesized.16 However,
no guanidino triamine conjugates are reported, and little is known
about if the guanidino triamine and diamine compounds could
utilize PAT for transport. Thus, a series of four guanidino putrecine
and homospermidine conjugates were synthesized and their PAT
selectivity was evaluated.
There are many synthetic protocols to introduce the guanidyl
group into the polyamine moieties by the reaction of the corre-
sponding amino group with a suitable guanidyl reagents such as
the readily available N,N0-diBoc protected thiourea or S-meth-
ylisothiourea.17 As shown in Scheme 1, S-methylisothiouronium
sulfate 5 was first Boc protected in the presence of NaHCO3 under
quick stirring. Heating the yielded product with putrecine at 50 °C
for 1 h generated masked guanidino diamine intermediate 6
through amine addition followed by elimination of volatile methyl
mercaptan. The nucleophilic substitution of N-(4-bromobu-
tyl)phthalimide by the primary amine 6, and the subsequent
N-Boc protection yielded compound 7. Since compound 7 is
⇑
Corresponding authors. Tel.: +86 13938637212 (S.X.); tel.: +86 13619810550;
fax: +86 378 2864665 (C.W.).
E-mail addresses: kfxsq@yahoo.com.cn (S. Xie), wcjsxq@henu.edu.cn (C. Wang).
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.09.069

6422
J. Wang et al. / Bioorg. Med. Chem. Lett. 20 (2010) 6421–6425
Figure 1. The structures of compounds 1–4.
H
N
H
vi
NHBoc
N
NH
2
Ar
N
H
Ar
N
H
2HCl
BocN
HN
11, 12, 13
14, 15, 4
v
NBoc
S
i, ii
H2SO4
NH
2
HN
NH
2
BocHN
N
H
2
iii
5
6
O
NBoc
Boc
NBoc
iv
Boc
N
N
N
BocHN
N
H
NH
2
BocHN
N
H
8
O
7
v
H
N
H
H
H
N
vi
Ar
Ar
N
N
NHBoc
NH
2
N
N
H
Boc
BocN
3HCl
HN
9
10
Ar =
14
15
4, 10
Scheme 1. The synthesis of target compounds. Reagents and conditions: (i) (Boc)2O, CH2Cl2/H2O; (ii) H2N(CH2)4NH2, THF/H2O; (iii) (a) N-(4-bromobutyl)phthalimide, CH3CN,
K2CO3; (b) (Boc)2O; (iv) NH2NH2, H2O; (v) (a) ArCHO, CH3OH/CHCl3; (b) NaBH4, CH3OH/CH2Cl2; (vi) 4NHCl, 1,4-dioxane.
unstable at room temperature, it was treated immediately with ex-
cess hydrazine hydrate at room temperature to liberate the key
guanylated triamine intermediate 8.6
separated by flash column chromatography, and immediately
deprotected using 4 N HCl in dioxane to give target compounds
(10, 14, 15 and 4) in high yields as hydrochloride salts.
The targeted guanidino diamine/triamine derivatives were pre-
pared using the known straightforward route.16 The one-pot
reductive amination of 8/6 to 9/11–13 was achieved in two steps
by condensation of the primary amine with one equivalent of aryl
aldehydes, and subsequently reduced by treatment with excess
NaBH4 in CH2Cl2/MeOH. Since the purification of 9/11–13 was cru-
cial for the purity of target compounds, these intermediates were
The in vitro cytotoxicities of the synthesized compounds were
evaluated against mouse leukemia L1210, melanoma B16, and
HeLa cells by MTT assays.11 As shown in Table 1, the guanidine
and hydrazine-modified polyamine conjugates displayed much
less in vitro potency than the reference compound 1. It is notewor-
thy that the guanidino diamine conjugate 4 exhibited higher
potency than its triamine counterparts 10, while parent triamine

J. Wang et al. / Bioorg. Med. Chem. Lett. 20 (2010) 6421–6425
6423
Table 1
Cytotoxicity results for target compounds in Hela, L1210 and B16 cells
Compound No.
IC50 (lM)
Hela
L1210
B16
B16 + SPD
B16 + DFMO
(B16)/(B16 + DFMO)
10
14
15
4
3
1
P50
34.36
>50
14.37
25.22
1.09
26.95
18.08
ꢀ50
49.12
36.71
>50
>50
12.67
1.40
>50(76.47)
>50(57.23)
>50
30.43
23.58
11.21
12.31
10.84
5.77
6.85
0.97
4.38
2.98
>5
>8
1.85
1.44
9.58
16.92
1.04
2.46
IC50 values were from three-independent experiments, and the standard deviations were less than 10%.
IC50 values were given only if they were less than 50 M, which was the maximum concentration tested. The data in parentheses (above 50
l
lM) were extrapolated ones.
conjugates are generally more toxic than diamine ones in the
similar series.4,6 In the only example of Ohara et al., a similar
phenomenon was observed by the comparison of the more active
guanidino diamine pyrene conjugate with the corresponding
guanidino tetraamine one.16 It is thus speculated that the diamine
conjugates benefit more than their tri- and tetraamine counter-
parts by the introduction of guanidine.
The growth inhibitory effect was determined in B16 cells coin-
cubated with excess spermidine (SPD) or a-difluoromethylorni-
thine (DFMO) in order to investigate whether these conjugates
could enter cells via PAT. DFMO, which depletes intracellular poly-
amine level by inhibition of ornithine decarboxylase (ODC), could
increase uptake of exogenous polyamines. Therefore, cells treated
with DFMO should be more susceptible to polyamine derivatives
and should provide lower IC50 values. As expected, exposure of
B16 cells to these conjugates in the presence of DFMO significantly
enhanced their growth inhibitory activity, especially for conjugates
with polyaromatic groups as cargoes (Table 1). In addition,
although DFMO exerts synergistic effects on all these terminal gua-
nidino or hydrazino substituted derivatives, the guanidino poly-
amine derivatives seem to be more sensitive to DFMO than their
parent and hydrazino polyamine conjugates.
Another assessment of the conjugates’ ability to target the PAT
was conducted by SPD protection experiments. The existence of
additional SPD in the system generates a competition with the
polyamine conjugates for the PAT protein on the cell membrane,
and therefore reduces the uptake of these conjugates which in turn
will reduce the cell death. The added SPD led to an obvious de-
crease in cell death (higher IC50) for hydrazino and guanidino poly-
amine conjugates (except 4) as reference compound 1.
Earlier studies with 1 revealed a direct correlation between
conjugate uptake and cytotoxicity.18 DFMO treatment increased
the import of 1 and resulted in a more toxic compound (lower
IC50 value). In a similar fashion, DFMO should facilitate the entry
of these novel conjugates and give lower IC50 values in the pres-
ence of DFMO. We thus measured the intracellular fluorescence
intensity of compound 10 and 1 in Hela cells by high content
with PAT may inhibit their entrance into the targeted cells as pre-
viously reported tetraamine conjugates,19 which may be offset by
DFMO.
Previous work revealed that 1 could produce cytotoxicity via
apoptosis,18 we thus detected how 10, with the mere structural dif-
ference of guanidine group in homospermidine moiety from 1, kill
cells. Lysosomes are highly dynamic intracellular organelles that
involve with the biosynthetic, endocytic, and autophagic path-
ways. They control the recycling of the majority of cellular macro-
molecules and organelles through more than 50 acid hydrolases. It
has been demonstrated that lysosomal membrane permeabiliza-
tion (LMP) triggers either necrotic or apoptotic cell death path-
ways. Tumor cells may be preferentially sensitive to agents that
trigger the lysosomal apoptosis pathway, suggesting that a reason-
able therapeutic window could be achieved for the LMP-inducing
agents. Therefore, LMP-inducing agents have great potential to be
developed as novel cancer therapy.20
The induction of LMP by 10 was analyzed using the lysosomo-
tropic weak base acridine orange (AO), which prefers to accumu-
late in normal lysosomes. AO is a metachromatic fluorophore
turning red in lysosomes after excitation with blue light. LMP re-
sults in the relocation of AO from lysosomes to cytosol, and the
fluorescence changes from red to green. Thus, red fluorescence
assaying (AO uptake method) was used here to evaluate pro-
nounced lysosomal rupture. As shown in Figure 3A, compounds
10 (10 lM) and DFMO (200 lM) induced moderately decrease in
the amount of cells with normal AO red fluorescence, whereas, this
effect was significant enhanced after 10 and DFMO co-treatment,
indicating significant LMP. To determine the consequence of LMP
which will lead to apoptosis or necrosis, we utilized the fluores-
cence staining using Annexin V-FITC and propidium iodide (PI).
PI is a DNA-binding fluorochrome which intercalates in the dou-
ble-helix. If the cell membrane is normal, PI could not permeate
and nucleus staining. Whereas, the membrane of necrotic cells is
not integrated, PI could permeate and nucleus staining. When cells
are in the state of apoptosis, the phosphatidylserine (PS) will evag-
inate on the cell surface and Annexin V-FITC will bind with PS.18 As
shown in Figure 3B, the cell numbers of Annexin V-FITC staining
significantly increased after 10 and DFMO co-treatment, although
DFMO had no effect and 10 had weak effect. Simultaneously, PI
staining was not obvious even at 10 and DFMO co-treatment, how-
ever, the PI nucleus staining of compound 1 was significant
(Fig. 3C). Those results indicated that the antiproliferative effect
of 10 and DFMO co-treatment was mainly due to cellular apopto-
sis, however, compound 1 could trigger apoptosis and necrosis.
Apoptosis is mainly controlled by two major pathways, namely
the mitochondrial pathway and membrane death receptor path-
way.21 In mitochondrial pathway, mitochondria have a crucial po-
sition in apoptosis control. The loss of mitochondria membrane
potential (MMP) induces cytochrome c release from the mitochon-
dria to the cytoplasm, which leads to the activation of caspase-9
and downstream cleavage of caspase-3. The membrane death
screening (HCS). As expected, the decreased IC50 value (6.43 lM)
of 10 against Hela cells in the presence of DFMO corroborated
the existence of a strong synergistic effect. In Figure 2, both com-
pounds 10 and 1 displayed upregulated cellular uptake in a time-
and dose-dependent manner. However, it seems that the effect of
DFMO on the cellular intake of 10 is more sensitive than that of
1, which is in accordance with the change of IC50 values.
Indeed, the PAT of cells can tolerate the terminal modifications
of the homospermidine motif from amine to guanidine. However,
the data also revealed that the elevated basicity from amine to gua-
nidine groups leads to attenuated potency of the homospermidine
conjugate. It is rationally speculated that the high basicity of guan-
idinium group (pKa 13.5) in polyamine conjugates may cause their
stronger interaction with PAT due to a higher preponderance of
protonated species at physiological pH. This strong interaction

6424
J. Wang et al. / Bioorg. Med. Chem. Lett. 20 (2010) 6421–6425
Figure 2. The intracellular fluorescence intensity of 10 and 1 with absent or present DFMO on Hela cells. (A) 15 min; (B) 30 min; (C) 60 min. Each point represents the
means S.D. from four independent experiments.
Figure 3. Apoptosis was detected in Hela cells after 10 (10 lM) or 1 (5 lM) treatment with absent or present DFMO (200 lM) for 48 h using high content screening (HCS). (A)
Lysosomal membrane stability was determined by assessing AO red/green fluorescence. (B) Apoptosis was detected by Annexin V-FITC and Hoechst33342 double staining. (C)
Necrosis was detected by PI and Hoechst33342 double staining (magnification, 20ꢁ).
receptor pathway is characterized by the binding between cell
death ligands and cell death receptors, and the subsequent activa-
tion of caspase-8 and -3.22 To reveal the precise molecular mecha-
nism of 10 and/or DFMO-induced apoptosis in Hela cells, we
observed the effect of 10 and/or DFMO on caspase-3, caspase-8,
caspase-9, MMP and cytochrome c. As shown in Figure 4A, 10 or
DFMO had no obvious effect on caspase-3, caspase-8 and cas-
pase-9, however, 10 and DFMO co-administration induced cas-
pase-9 and caspase-3 activation but did not affect caspase-8,
indicating that Hela cells apoptosis which induced by 10 and
DFMO co-administration via mitochondrial pathway, this apopto-
tic effect was similar to compound 1. Further, we found that DFMO
and 10 co-treatment also induced MMP loss followed by the
release of cytochrome c with the corresponding decrease in mito-
chondria (Fig. 4B).
Generally speaking, the SPD/DFMO experiments corroborated
that the guanidino polyamine conjugates could target PAT for drug
delivery as parent polyamine conjugates. However, the less active
guanidino triamine conjugate compared to corresponding diamine
derivatives, which is different from previous reports, indicated that
the presence of guanidine in conjugates may inhibit their own
transport via PAT. In this case, the co-administration of DFMO is
critical for better cancer therapy.2 The apoptotic mechanism re-
vealed that both the normal and guanidino polyamine conjugates

J. Wang et al. / Bioorg. Med. Chem. Lett. 20 (2010) 6421–6425
6425
Figure 4. The change of MMP and Cyto C, caspase-3, caspase-8 and caspase-9 were detected using Rh123 staining and immunofluorescence assay by HCS, respectively. Each
**
*
point represents the means S.D. from four independent experiments. Compared with control p <0.05, p <0.01.
4. Wang, C.; Delcros, J.-G.; Biggerstaff, J.; Phanstiel, O. J. Med. Chem. 2003, 46,
trigger apoptosis through the lysosomes-associated mitochondrial
pathway, supported by the upregulation of caspases 3 and 9. DFMO
strongly synergized the functions of 10, but did not alter the apop-
totic pathway, which corroborated that DFMO only facilitated the
transport of 10 into cells, and could not change the cellular targets
of the conjugates for cytotoxicity.
It is also intriguing that the guanidino putrecene-anthracene
conjugate 4 displayed strong DFMO effect and reversed SPD effect.
Because the natural polyamines in body often reduce the antitu-
mor efficacy of polyamine-based drugs, the elevated potency of
this drug in the presence of SPD and DFMO favors its further study
as one lead compound.
2663.
5. Phanstiel, O.; Kaur, N.; Delcros, J.-G. Amino Acids 2007, 33, 305.
6. Tian, Z. Y.; Xie, S. Q.; Du, Y. W.; Ma, Y. F.; Zhao, J.; Gao, W. Y.; Wang, C. J. Eur. J.
Med. Chem. 2009, 44, 393.
7. Tian, Z. Y.; Xie, S. Q.; Mei, Z. H.; Zhao, J.; Gao, W. Y.; Wang, C. J. Org. Biomol.
Chem. 2009, 7, 4651.
8. Kaur, N.; Delcros, J.-G.; Imran, J.; Khaled, A.; Chehtane, M.; Tschammer, N.;
Martin, B.; Phanstiel, O. J. Med. Chem. 2008, 51, 1393.
9. Barret, J. M.; Kruczynski, A.; Vispé, S.; Annereau, J. P.; Brel, V.; Guminski, Y.;
Delcros, J.-G.; Lansiaux, A.; Guilbaud, N.; Imbert, T.; Bailly, C. Cancer Res. 2008,
68, 9845.
10. Xie, S. Q.; Liu, G. C.; Ma, Y. F.; Cheng, P. F.; Wu, Y. L.; Wang, M. W.; Ji, B. S.; Zhao,
J.; Wang, C. J. Toxicol. In Vitro 2008, 22, 352.
11. Wang, J.; Xie, S.; Li, Y.; Guo, Y.; Ma, Y.; Zhao, J.; Phanstiel, O.; Wang, C. Bioorg.
Med. Chem. 2008, 16, 7005.
12. Zhou, J. J.; Huang, H.; Xie, S. Q.; Wang, Y. X.; Zhao, J.; Wang, C. J. Chin. Chem. Lett.
2008, 19, 99.
Acknowledgments
13. Zhou, J.J. Master thesis of Henan University, 2007.
14. (a) Pallan, P. S.; Ganesh, K. N. Biochem. Biophys. Res. Commun. 1996, 222, 416;
(b) Constantinos, M.; Garnelis, T.; Pantazaka, E.; Papaioannou, D. Tetrahedron
Lett. 2004, 45, 8815; (c) Dardonville, C.; Brun, R. J. Med. Chem. 2004, 47, 2296;
(d) Manetti, F.; Cona, A.; Angeli, L.; Mugnaini, C.; Raffi, F.; Capone, C.; Dreassi,
E.; Zizzari, A. T.; Tisi, A.; Federico, R.; Botta, Maurizio. J. Med. Chem. 2009, 52,
4774.
This work was supported by National Science Foundations of
China (20872027, 90913001) and Henan Natural Science Founda-
tion (082102270002).
15. Chen, Z. Y. Master thesis of Henan University, 2007.
16. Ohara, K.; Smietana, M.; Restouin, A.; Mollard, S.; Borg, J. P.; Collette, Y.;
Vasseur, J. J. J. Med. Chem. 2007, 50, 6465.
17. Guo, Z.-X.; Cammidge, A. N.; Horwell, D. C. Synth. Commun. 2000, 30, 2933.
18. Xie, S.-Q.; Wang, J.-H.; Ma, H.-X.; Cheng, P.-F.; Zhao, J.; Wang, C.-J. Toxicology
2009, 263, 127.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2010.09.069.
19. Wang, C.; Delcros, J. G.; Cannon, L.; Konate, F.; Carias, H.; Biggerstaff, J.;
Gardner, R. A.; Phanstiel, O. J. Med. Chem. 2003, 46, 5129.
20. (a) Boya, P.; Kroemer, G. Oncogene 2008, 27, 6434; (b) Chen, Z.; Liang, X.; Zhang,
H.; Xie, H.; Liu, J.; Xu, Y.; Zhu, W.; Wang, Y.; Wang, X.; Tan, S.; Kuang, D.; Qian,
X. J. Med. Chem. 2010, 53, 2589.
References and notes
1. Palmer, A. J.; Wallace, H. M. Amino Acids 2010, 38, 415.
2. Xie, S.-Q.; Wang, J.-H.; Zhang, Y.-H.; Wang, C.-J. Expert Opin. Drug Deliv. 2010, 7,
1049.
3. Gardner, R. A.; Delcros, J. G.; Konate, F.; Breitbeil, F., III; Martin, B.; Sigman, M.;
Phanstiel, O. J. Med. Chem. 2004, 47, 6055.
21. Goldschneider, D.; Mehlen, P. Oncogene 2010, 29, 1865.
22. Grimm, S.; Noteborn, M. Trends Mol. Med. 2010, 16, 88.