Synthesis and biological characterization of novel charge-deficient spermine analogues

Article abstract of DOI:10.1021/jm100439p

Biogenic polyamines, spermidine and spermine, are positively charged at physiological pH. They are present in all cells and essential for their growth and viability. Here we synthesized three novel derivatives of the isosteric charge-deficient spermine analogue 1,12-diamino-3,6,9-triazadodecane (SpmTrien, 5a) that are N¹-Ac-SpmTrien (5c), N¹²-Ac-SpmTrien (5b), and N¹,N¹²-diethyl-1,12-diamino-3,6,9-triazadodecane (N¹,N¹²-Et₂-SpmTrien, 5d). 5a and 5d readily accumulated in DU145 cells at the same concentration range as natural polyamines and moderately competed for the uptake with putrescine (1) but not with spermine (4a) or spermidine (2). 5a efficiently down-regulated ornithine decarboxylase and decreased polyamine levels, while 5d proved to be inefficient, compared with N¹,N¹¹-diethylnorspermine (6). None of the tested analogues were substrates for human recombinant spermine oxidase, but those having free aminoterminus, including 1,8-diamino-3,6-diazaoctane (Trien, 3a), were acetylated by mouse recombinant spermidine/spermine N ¹-acetyltransferase. 5a was acetylated to 5c and 5b, and the latter was further metabolized by acetylpolyamine oxidase to 3a, a drug used to treat Wilson's disease. Thus, 5a is a bioactive precursor of 3a with enhanced bioavailability.

Full text of DOI:10.1021/jm100439p

5738 J. Med. Chem. 2010, 53, 5738–5748
DOI: 10.1021/jm100439p
Synthesis and Biological Characterization of Novel Charge-Deficient Spermine Analogues
‡,^
†
§
‡
‡
Janne Weisell,*,†,^ Mervi T. Hyvonen, Merja R. Hakkinen, Nikolay A. Grigorenko, Marko Pietila, Anita Lampinen,
€
€
€ €
€
Sergey N. Kochetkov, Leena Alhonen, Jouko Vepsalainen, Tuomo A. Keinanen, and Alex R. Khomutov
‡
†
‡
€
†Department of Biosciences, and ‡A. I. Virtanen Institute, Biocenter Kuopio, University of Eastern Finland, Yliopistonranta 1E, 70210 Kuopio,
Finland, §BASF Schweiz AG, P.O. Box, CH 4002, Basel, Switzerland, and Engelhardt Institute of Molecular Biology, Russian Academy of
Sciences, Moscow 119991, Russia. ^ These authors contributed equally to this work.
Received April 10, 2010
Biogenic polyamines, spermidine and spermine, are positively charged at physiological pH. They are
present in all cells and essential for their growth and viability. Here we synthesized three novel deriva-
tives of the isosteric charge-deficient spermine analogue 1,12-diamino-3,6,9-triazadodecane (SpmTrien,
5a) that are N1-Ac-SpmTrien (5c), N12-Ac-SpmTrien (5b), and N1,N12-diethyl-1,12-diamino-3,6,9-
triazadodecane (N1,N12-Et2-SpmTrien, 5d). 5a and 5d readily accumulated in DU145 cells at the same
concentration range as natural polyamines and moderately competed for the uptake with putrescine (1)
but not with spermine (4a) or spermidine (2). 5a efficiently down-regulated ornithine decarboxylase and
decreased polyamine levels, while 5d proved to be inefficient, compared with N1,N11-diethylnorsper-
mine (6). None of the tested analogues were substrates for human recombinant spermine oxidase, but
those having free aminoterminus, including 1,8-diamino-3,6-diazaoctane (Trien, 3a), were acetylated by
mouse recombinant spermidine/spermine N1-acetyltransferase. 5a was acetylated to 5c and 5b, and the
latter was further metabolized by acetylpolyamine oxidase to 3a, a drug used to treat Wilson’s disease.
Thus, 5a is a bioactive precursor of 3a with enhanced bioavailability.
Introduction
4a can also be catabolized to 2 by spermine oxidase (SMO)
without prior acetylation step. The expression of the rate-
limiting enzymes of polyamine metabolism is tightly regulated
by multiple steps, including complex transcriptional, post-
transcriptional, translational, and posttranslational mechan-
isms. Spd and Spm synthases are constitutively expressed and
regulated by the abundance of their substrates, especially by
the level of decarboxylated S-adenosylmethionine.
Different polyamine structural analogues have been de-
signed for research purposes and as potential therapeutic
compounds.3 Among the most studied potential polyamine-
based cancer chemotherapeutics are the terminally bis-N-alky-
lated analogues, such as N1,N11-diethylnorspermine (DENSpm,
6). Many of them superinduce SSAT, induce polyamine efflux,
down-regulate biosynthetic enzymes, and compete for poly-
amine uptake, resulting in dramatic polyamine depletion and
cessation of cell growth or apoptosis. In addition, some of
these analogues seem to have cytostatic/cytotoxic effects that
are not directly related to polyamine depletion.4 By contrast,
C-methylated polyamine analogues, such as R-methylspermi-
dine, generally support cell growth and mimic the natural
polyamines in many of their cellular functions.5-7 Apart from
these analogue types, charge-deficient analogues have been
synthesized in order to elucidate the importance of proper charge
distribution. Among these compounds are difluorospermidines8
and aminooxy9 and oxa derivatives10 of 2 and 4a.
The polyamines spermidine (Spd,a 2) and spermine (Spm,
4a), andtheir precursorputrescine(Put,1), are organiccations
existing at millimolar concentrations in eukaryotic cells.1 At
physiological pH their amino groups are protonated,2 en-
abling them to interact with negatively charged cellular
components such as DNA, RNA, and anionic phospholipids.
Polyamines participate in the regulation of important cellular
functions, such as proliferation, differentiation, and function-
ing of N-methyl-D-aspartate receptor and inward rectifier
potassium channel.1 Disturbances in the polyamine home-
ostasis have been linked to various pathological conditions,
such as cancer, acute pancreatitis, diabetes, and decreased
immune response.
Themain regulatory enzymesofpolyamine biosynthesisare
ornithinedecarboxylase(ODC) andS-adenosyl-L-methionine
decarboxylase (AdoMetDC), while spermidine/spermine N1-
acetyltransferase (SSAT) is the key enzyme of acetylpolya-
mineoxidase(APAO)-mediatedcatabolism/interconversion.1
*To whom correspondence should be addressed. Phone: þ358-
403552237. Fax: þ358-17 163259. E-mail: Janne.Weisell@uef.fi.
a Abbreviations: AdoMetDC, S-adenosyl-L-methionine decarboxy-
lase; APAO, acetylpolyamine oxidase; N1,N12-Et2-SpmTrien, N1,N12
-
diethyl-1,12-diamino-3,6,9-triazadodecane; DFMO, R-difluoromethy-
lornithine (2,5-diamino-2-(difluoromethyl)pentanoic acid); DENSpm,
N1,N11-diethylnorspermine (N1,N11-Et2-norSpm); N1-Ac-Spd, N1-acet-
ylspermidine; N1-Ac-Spm, N1-acetylspermine; NMD, nonsense-mediated
mRNA decay; ODC, ornithine decarboxylase; Put, putrescine (1,4-
diaminobutane); Spd, spermidine (1,8-diamino-4-azaoctane); Spm,
spermine (1,12-diamino-4,9-diazadodecane); SMO, spermine oxidase;
SpmTrien, 1,12-diamino-3,6,9-triazadodecane; SSAT, spermidine/sper-
mine N1-acetyltransferase; Trien, triethylenetetraamine (1,8-diamino-
3,6-diazaoctane).
Triethylenetetraamine (1,8-diamino-3,6-diazaoctane, Trien,
3a) is a charge-deficient analogue of 2, having two positive
charges at physiological pH (Table 1). Unlike 2, 3a is a selec-
tive and effective Cu2þ chelator (pKuns = 20.4 at pH 14 and
pKuns = 14.2 at pH 7),11 and it is used as an alternative
treatment of Wilson’s disease (hepatolenticular degeneration,
r
pubs.acs.org/jmc
Published on Web 07/20/2010
2010 American Chemical Society

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 15 5739
Table 1. pKa Values of 2, 4a, and Their Charge-Deficient Isosteric
Analogues 3a and 5aa
pKa
1
2
3
4
5
2
8.24
3.27
7.96
3.3
9.81
6.56
8.85
6.3
10.89
9.07
10.02
8.5
3a
4a
5a
9.74
10.80
9.5
10.3
a Table adapted from Weisell et al. (2010).17
a genetic disorder in which copper accumulates in tissues) if
the patient is intolerant to the primary medication with
penicillamine.12 Recently, 3a has gained attention in amelior-
ating left ventricular hypertrophy in diabetic patients.13 It was
also successfully used to regenerate the heart by selective Cu2þ
chelation in a rat model of streptotoxin-induced diabetes.14
Very little is known about its mechanism of action, excluding
its role as a selective Cu2þ chelator, and only a few of these
studies have focused on polyamine metabolism, although 3a
resembles 2 in structure. A recent publication speculated that
SSAT could metabolize 3a, but no data have been published
to support that hypothesis.15
Replacement of the 2 moiety from 4a with 3a residue gives
rise to 5a,16-18 which is an isoster of 4a. 5a exists as non-
symmetrical trivalent cation at physiological pH (Table 1),
and it can induce DNA condensation more efficiently than
2.16 The condensation ability of 5a is more pronounced at
slightly acidic pH where the protonation status of the analo-
gue is close to that of 4a. Therefore, the presence of the 3a
moiety gives good chelating properties for 5a, and the pre-
sence of the aminopropyl group seems to evoke additional
effects on cell growth and polyamine metabolism compared to
3a.17 At the moment, the only known charge-deficient term-
inally bis-N-alkylated polyamine derivative is N1,N12-bis-
(2,2,2-trifluoroethyl)-1,12-diamino-4,9-diazadodecane, but
the electronegative effect of the trifluoroethyl substituent
turns the terminal amino groups nonprotonated at physiolo-
gical pH.2 However, terminally diethylated spermine deriva-
tives with partly protonated central fragment have still remai-
ned unknown.
Figure 1. Structures of the studied compounds.
was elongated with aminoethanol followed by the addition
of the third benzyloxycarbonyl protection group to give the
intermediate 8a, whose synthesis was described earlier in
detail.18 Following the same strategy alcohol 8a was con-
verted via the corresponding mesylate to 9a using N-acetyl-
1,3-diaminopropane19 to give the target compound after the
removal of Cbz groups by catalytic hydrogenation.
Synthesis of 5d (Scheme 1) was started from N-(benzylo-
xycarbonyl)-3-azapentanol-120 using the same sequence of
the reactions as above, but in this case the chain elongation
reagent was N-(2-aminoethyl)aminoethanol, to give the inter-
mediate 8b, which was first mesylated and then aminated
with an excess of N-ethyl-1,3-diaminopropane to give the
protected intermediate 9b. In this case, the reaction mixture
contained, in addition to the target linear compound 9b, a
branched byproduct of the alkylation of the secondary
amino group of N-ethyl-1,3-diaminopropane, which was
hardly separated by flash chromatography on silica gel under
the used conditions. However, the addition of 20 mol % of
salicylic aldehyde to crude 9b resulted in the formation of a
stable Shiff base with the primary amino group of the
byproduct. Thereafter, the formed Schiff base adduct was
separated from target 9b by flash chromatography. Finally,
N-protection groups were removed by catalytic hydrogena-
tion to give target 5d as a pentahydrochloride with 24%
overall yield, calculating from 10.
Here we describe synthesis of N1,N12-Et2-SpmTrien (5d) and
terminally monoacetylated SpmTrien’s, i.e., N1-Ac-SpmTrien
(5c) and N12-Ac-SpmTrien (5b) (Figure 1). We have investi-
gated the uptake of 5c and the earlier synthesized 5a and 3a into
cells, including the impact of AZ in this process, their effects on
cell growth, and their ability to mimic the cellular functions of
the natural polyamines in DU145 prostate carcinoma cells. In
addition, the substrate properties of these analogues for poly-
amine catabolic enzymes SSAT, SMO, and APAO were
studied. The present data demonstrate that not only the proper
geometry but also the proper charge distribution is important
for the analogue to be recognized as a polyamine and to
completely fulfill the cellular functions of polyamines and that
the alteration of the degree of protonation may provide an
analogue with unusual biological activities.
Results and Discussion
Synthesis of 5c was started from N-(benzyloxycarbonyl)-
3-amino-1-propyl methanesulfonate 12,21 used for the chain
elongation, the procedure being close to the above-described
that resulted in the key intermediate 5b with a good yield. To
prepare orthogonally protected 14 (Scheme 2), it was necessary
to discriminate the primary and secondary amino groups of
the intermediate 13. Required selective N-Cbz protection
Chemistry. A rational approach to prepare monoacety-
lated derivatives of 5a and terminally bis-ethylated 5a con-
sisted of the use of orthogonally protected precursors.
Synthesis of 5b (Scheme 1) was started from commercially
available N-(2-aminoethyl)aminoethanol 7, which was con-
verted into bis-Cbz derivative, mesylated, and then the chain

5740 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 15
Scheme 1. Synthesis of 5b and 5da
Weisell et al.
a (a) CbzCl, NaHCO3; (b) MsCl, Et3N; (c) H2N(CH2)2OH; (d) H2N(CH2)2NH(CH2)2OH, THF; (e) H2N(CH2)3NHAc; (f) EtNH(CH2)3NH2, THF;
(g) H2/Pd, AcOH, MeOH.
Scheme 2. Synthesis of 5ca
a (a) H2N(CH2)2NH(CH2)2OH; (b) CbzCl; (c) MsCl, Et3N, NaHCO3; (d) H2N(CH2)2NH2; (e) o-C6H4(OH)CHO/THF; (f) MeONH2; (g) AcCl,
Et3N; (h) H2/Pd, AcOH, MeOH.
Table 2. Effect of 50 μM Polyamines and Charge-Deficient Analogues on Polyamine Metabolizing Enzyme Activities in DU145 Cellsa
metabolic enzymes
((pmol/h)/μg DNA)
ODC
AdoMetDC
SSAT
SMO
APAO
At 24 h
At 48 h
At 72 h
control
2
56 ( 9
236 ( 29
138 ( 21
117 ( 2
132 ( 14
140 ( 22
236 ( 1
537 ( 102
374 ( 33
552 ( 34
509 ( 86
478 ( 32
495 ( 7
334 ( 8
10 ( 1***
13 ( 3***
77 ( 15*
3 ( 0***
53 ( 0
179 ( 3**
108 ( 9***
241 ( 21
52 ( 3***
166 ( 16***
17 ( 3***
277 ( 15
284 ( 14
316 ( 49
274 ( 17
309 ( 10
315 ( 48
4a
3a
5a
5d
6
178 ( 8
22019 ( 3659***
0.5 ( 0.1***
3205 ( 329***
control
2
29 ( 3
223 ( 10
122 ( 4
104 ( 6
330 ( 63
310 ( 40
388 ( 66
298 ( 15
379 ( 26
395 ( 86
2233 ( 15***
274 ( 19
233 ( 20
254 ( 37
231 ( 14
249 ( 18
253 ( 35
278 ( 34
4 ( 0***
6 ( 1***
31 ( 2
1 ( 0***
13 ( 0***
0.1 ( 0.0***
179 ( 7***
112 ( 4***
171 ( 6***
48 ( 1***
143 ( 15***
14 ( 1***
4a
3a
5a
5d
6
125 ( 10
104 ( 5
177 ( 17
129 ( 12
12644 ( 758***
control
DFMO
9 ( 1
174 ( 11
98 ( 13
93 ( 2
312 ( 7
105 ( 11**
450 ( 62
68 ( 27**
284 ( 39
292 ( 62
230 ( 18
237 ( 6
0.3 ( 0.0***
0.2 ( 0.0***
0.3 ( 0.1***
0.2 ( 0.1***
0.3 ( 0.2***
0.2 ( 0.0***
652 ( 25***
105 ( 14*
441 ( 41***
38 ( 1***
69 ( 14***
20 ( 2***
DFMO þ 4a
DFMO þ 3a
DFMO þ 5a
DFMO þ 5d
DFMO þ 6
128 ( 22
85 ( 5
178 ( 14
163 ( 33
9486 ( 1181***
259 ( 26
198 ( 16
206 ( 25
304 ( 42
197 ( 20
1456 ( 108***
a Cells were cultured with 1 mM AG and 50 μM polyamines/analogues with or without 5 mM DFMO. Data (in (pmol/h)/μg DNA) are the mean (
SD, n = 3. /, p < 0.05; //, p < 0.01; ///, p < 0.001 compared to AG-treated control.
was achieved in one-pot with 75% yield by using salicylic
aldehyde22 as protection agent for terminal primary amino
group followed by the addition of a Cbz group to the
remaining NH group, and in the last step the salicylidene
group was removed with MeONH2 base to give 14, which
was acetylated to 15 and deprotected to target 5c with 51%
overall yield, calculating from compound 13.
growth and polyamine metabolism, DU145 prostate carci-
noma cells were incubated with 50 μM natural polyamines,
charge-deficient analogues, or the reference compound 6 for
24 or 48 h. There were dramatic differences between the
natural polyamines and their charge-deficient isosters in
their effects on polyamine metabolism. 5a was a very efficient
down-regulator of ODC and AdoMetDC and moderate
inductor of SSAT, while 3a had very little, if any, effect
toward polyamine metabolizing enzymes (Table 2). 5a was
Effects on Cell Growth and Polyamine Metabolism. To
investigate the metabolism and effect of the analogues on cell

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 15 5741
almost as potent an ODC down-regulator as 6, which
encouraged us to synthesize 5d, which is a charge-deficient
analogue of N1,N12-diethylspermine (DESpm). Surprisingly,
5d exerted only modest effects on polyamine metabolism. All
charge-deficient analogues showed slight (up to 25%) cyto-
static effect after 24-48 h of culture, in the order of 3a < 5a
< 5d (Figure 2A). When incubated in the presence of the
ODC inhibitor R-difluoromethylornithine (DFMO) for 72
h, 5a was able to support cell growth for about 75%
(Figure 2B), while in the absence of DFMO after 48 h of
incubation with 5a the growth was 80% (Figure 2A), com-
pared with nontreated control. When 3a was incubated in the
presence of DFMO for 72 h, it was able to restore cell growth
for about 75% (Figure 2B), while in the absence of DFMO
after 48 h of incubation with 3a the growth was 90%
(Figure 2A), compared to growth of control cells. Alto-
gether, the data show that charge-deficient 5a can fulfill
some but not all cellular functions of polyamines, whereas
diethylation of 5a (resulting in the molecule 5d) abolished
biological activity toward regulatory enzymes.
Cellular Uptake and Its Regulation by Antizyme. 5a and 5d
were accumulated intracellularly in the same concentration
range as the natural polyamines, whereas the intracellular
concentration of 3a was 8-10 times lower (Table 3). Com-
pared to the natural polyamines and 6, compounds 3a, 5a,
and 5d were moderately competing for uptake with [14C]-
labeled 1 (10 μM) but very poorly with 2 or 4a, as measured
by 10 min experiments in DU145 cells (Figure 3). A previous
report using rat intestinal brush-border vesicles determined
Km of 1.13 mM for the uptake of 3a, which is much higher
than Km of 1.44 μM for 4a.23,24 Accordingly, the tested
analogues competed efficiently only with 1 for uptake,
suggesting that 3a, 5a, and 5d are taken up by polyamine
transport but their affinity is low.
It is known that polyamine uptake is negatively regulated
by antizyme (AZ), a small protein also binding to ODC and
targeting it for degradation.25,26 To further investigate
whether the charge-deficient analogues are taken up via
AZ-regulated polyamine transporter, we measured the abil-
ity of AZ to decrease the uptake of the analogues. For that
purpose, we constructed doxycycline (Dox) inducible AZ-
overexpressing DU145 cell line. The AZ frameshift site was
removed in order to get high expression of AZ regardless of
the intracellular polyamine level. As shown in Figure 4A, in
the presence of Dox (250 ng/mL, 24 h), DU145 cells synthe-
sized a high amount of AZ protein. Forced overexpression
of AZ significantly decreased the level of 5d and the refer-
ence compound 6 (Figure 4B), which is in accordance with
Figure 2. Growth of DU145 cells incubated with 50 μM poly-
amines/analogues (A) without or (B) with 5 mM DFMO. All
plates contained 1 mM AG. Results are the mean ( SD, n = 3,
compared to (A) untreated control, (B) DFMO-treated group:
Ctrl, control.
Table 3. Effect of 50 μM Polyamines and Charge-Deficient Analogues on Polyamine Pools in DU145 Cellsa
polyamines
(pmol/μg DNA)
1
2
4a
4b
analog
metabolites
At 24 h
control
2
18 ( 5
nd***
nd***
27 ( 6*
6 ( 1**
19 ( 1
nd***
277 ( 37
245 ( 37
nd
294 ( 11***
142 ( 10**
206 ( 20***
88 ( 4***
166 ( 20***
25 ( 5***
174 ( 8**
273 ( 26
263 ( 14
38 ( 2***
197 ( 6
11 ( 2
nd
nd
4a
3a
5a
5d
6
66 ( 14
40 ( 3***
15 ( 2
29 ( 17**
475 ( 38
642 ( 77
921 ( 100
44 ( 7,b 19 ( 0c
36 ( 8***
At 48 h
232 ( 6
control
2
12 ( 2
206 ( 4
nd
3 ( 0***
2 ( 1***
13 ( 2
2 ( 1***
12 ( 2
nd***
250 ( 13***
127 ( 9***
132 ( 6***
58 ( 2***
135 ( 7***
11 ( 2***
160 ( 13**
272 ( 24
207 ( 22
24 ( 1***
184 ( 28
18 ( 2***
6 ( 0***
nd
nd
4a
3a
5a
5d
6
38 ( 7
21 ( 2***
10 ( 1***
nd
351 ( 39
617 ( 33
578 ( 44
19 ( 2,b 10 ( 1c
At 72 h
196 ( 13
148 ( 14
control
DFMO
7 ( 1
161 ( 7
nd
nd
nd***
nd***
nd***
nd***
nd***
nd***
11 ( 1***
132 ( 21*
10 ( 2***
24 ( 5***
40 ( 12***
nd***
DFMO þ 4a
DFMO þ 3a
DFMO þ 5a
DFMO þ 5d
DFMO þ 6
251 ( 45
nd
nd
126 ( 6**
14 ( 1***
67 ( 17***
13 ( 1***
115 ( 8
11 ( 0***
nd
nd
304 ( 22
1365 ( 271
413 ( 48
16 ( 3c
a Cells were cultured with 1 mM AG and 50 μM polyamines/analogues with or without 5 mM DFMO. Data (in (pmol/μg DNA)) are the mean ( SD,
n = 3. nd, not detectable. /, p < 0.05; //, p < 0.01; ///, p < 0.001 compared to AG-treated control. b 5a is metabolized to 3a. c 5a is metabolized to 5c.

5742 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 15
Weisell et al.
Figure 4. (A) AZ expression in doxycycline-inducible AZ-over-
expessing cells, showing Western blot of AZ in mock- or AZ-
transfected DU145 cells, with or without 24 h of preincubation
with Dox. (B) Effect of forced AZ overexpression on accumulation
of the analogues in DU145 cells. Doxycycline-inducible AZ-over-
expressing cells were incubated with 50 μM analogues in the
presence or absence of doxycycline (Dox, 250 ng/mL, preincubation
24 h) for 6 h. (C) Effect of CHX on the intracellular accumulation of
the analogues in DU145 cells. Cells were incubated in the presence
or absence of CHX (10 μg/mL) for 1 h after which the incubation
was continued with 50 μM analogue for 4 h, in the presence or
absence of CHX. Results show the intracellular analogue concen-
trations, mean ( SD, n = 3. No intracellular metabolites of the
analogues were detected by HPLC. /, //, and /// refer to p values of
<0.05, < 0.01, and <0.001, respectively, compared to correspond-
ing without-CHX or without-Dox group.
Figure 3. Uptake competition of the analogues with [14C]-labeled
(A) 1, (B) 2, or (C) 4a in DU145 cells. Compounds were tested for
10 min at 10, 100, and 1000 μM against 10 μM labeled 1, 2, or 4a.
Results are the mean ( SD, n = 3.
the well-known competition of 6 with 4a/2 for transporta-
tion and its ability not only to induce SSAT but also to
down-regulate ODC AZ-dependently.27,28 In contrast to 6
and 5d, the intracellular levels of 3a and 5a were slightly
but not significantly decreased by AZ overexpression and
the observed difference between 5a and 5d is of special
interest.
We also investigated whether the analogues are able to
induce AZ expression. Previous reports have indicated that
compounds bearing even some resemblance to polyamines
can induce AZ.29 Since AZ is an inducible protein with a very
short half-life and regulates polyamine uptake, we utilized an
indirect method where we measured the intracellular accu-
mulation of the analogues in the presence and absence of
protein synthesis inhibitor cycloheximide (CHX) (Figure 4C).
As expected, CHX treatment (prevention of AZ induction)
increased the intracellular accumulation of 6. However, it did
not affect the intracellular level of 3a or 5d. Since the level of
5d was not increased by CHX treatment but was decreased
by forced AZ expression, it is possible that 5d is not able to
induce AZ. This view is supported by the finding that 5d did
not down-regulate ODC (Table 2). In contrast, 5a did down-
regulate ODC, and its accumulation was significantly in-
creased by CHX treatment, suggesting that it induces AZ. As
AZ production is regulated by a unique mechanism invol-
ving polyamine-induced þ1 ribosomal frameshift on AZ
mRNA, the ability of 5a and 5d to induce AZ frameshifting
should be studied in more detail. Altogether, the results from
the uptake experiments suggest that the charge-deficient
analogues 5a, 5d, and 3a have low affinity for the polyamine
transporter, but they readily accumulate in DU145 cells at
the same concentration range as natural polyamines with the
exception of 3a.
Catabolism of 5a. The analysis of the polyamine pool in 5a-
treated DU145 cells demonstrated that the drug was meta-
bolized to 3a and independently to 5c (based on HPLC reten-
tion time, i.e., coelution with authentic reference compound),

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 15 5743
Table 4. Kinetic Values for Mouse Recombinant SSATa
compd
Km (μM)
Vmax ((μmol/min)/mg SSAT)
5a
5c
5b
3a
3b
2
106 ( 6
250 ( 18
435 ( 30
144 ( 6
79 ( 6
1.17 ( 0.02
0.44 ( 0.01
1.18 ( 0.03
1.36 ( 0.02
0.57 ( 0.02
8.85 ( 0.5b
6.39 ( 0.6b
151 ( 15
nd
4a
a Mouse recombinant SSAT (20 or 2 ng) was incubated with supple-
mented analogue at 50, 100, 250, 500, and 2500 μM substrate concen-
tration range in triplicate. Blank reaction mixtures were done in
triplicate and treated identically as earlier samples but without SSAT
and by using 2500 μM concentration of each analogue. Values are the
mean ( SD, n = 3. nd, not determined. Kinetic values for 3b were
determined by using 50, 100, and 250 μM becauseof substrateinhibition.
b Reference Vmax of 2 and 4a has been determined by using 2500 μM and
2 ng of mouse recombinant SSAT.
while no intracellular metabolites were detectable for 3a or
5d (Table 3). Obviously, 3a can be formed from 5a only either
by the result of direct oxidation of 5a by SMO or APAO or
via the SSAT/APAO pathway. The latter pathway requires
an intermediate 5b to be formed first. To study whether the
enzymes of polyamine metabolism are involved in biotrans-
formation of 5a into 3a, we investigated the interaction of 5a
with recombinant SMO, SSAT, and APAO.
5a exhibited very minor substrate properties with SMO or
APAO; HPLC analysis of the 200 μM 5a substrate mixtures
showed less than 5% metabolism in the presence of 1 μg of
SMO or APAO after 30 min of incubation at 37 ꢀC (data not
shown). Therefore, a direct intracellular transformation of
5a into 3a by APAO or SMO is unlikely. However, 5a proved
to be a substrate of SSAT (Table 4), but in contrast to 4a, 5a
is an asymmetric molecule. To determine the site of its
acetylation, two earlier unknown monoacetylated deriva-
tives, i.e., 5c and 5b (Figure 1), were synthesized and used as
reference compounds in HPLC coelution analysis as de-
scribed above. 5b was a good substrate of APAO, thus
Figure 5. Competition of the analogues with (A) 2 for mouse
recombinant SSAT, (B) 4b for human recombinant APAO, and
(C) 4a for human recombinant SMO. Analogues were tested at 100,
500, and 1000 μM (or 2000 μM) against (A) 150 μM 2, (B) 10 μM 4b,
or (C) 10 μM 4a. Results are the mean ( SD, n = 3.
producing 3a (Km = 382 ( 23 μM, kcat = 6.17 ( 0.15 s-1
;
reference value for 4b was kcat = 3.15 ( 0.1 s-1, determined
under similar conditions at 0.5 mM; Km of 4b is 14 μM30).
Taken together, these data indicate that the biotransforma-
tion of 5a into 5c and into 3a via 5b can take place with the
consequent action of SSAT/APAO enzymes. Furthermore,
we studied the metabolism of 50 μM 5a for 24 h in SSAT-
deficient mouse fetal fibroblasts and did not detect any
conversion of 5a to 3a or 5b or 5c, while syngenic fibroblast
did metabolize 5a to 3a and 5c (data not shown). It is
noteworthy that the amount of 3a produced from supple-
mentation of DU145 cells with 50 μM 5a is about the same as
after the treatment of the cells with 50 μM 3a itself (Table 3),
demonstrating that efficiently transported 5a is a bioactive
precursor of 3a.
incorporation of [14C]-label from Ac-CoA to the compounds
tested (data not shown). We found that 5a, 3a, and 5b
showed similar maximal reaction velocity of about 1.2
(μmol/min)/mg enzyme, but their affinities for SSAT were
different (Table 4). Kinetic values for N1-Ac-Trien (3b) were
determined by using 50, 100, and 250 μM substrate ranges.
With higher concentration we detected increasing substrate
inhibition retarding observed maximal reaction, since we
excluded product (N1,N8-Ac2-Trien, 3c) feedback inhibition
(Figure 5A). 5c was acetylated with reasonable velocity but
less efficiently than 5a itself. However, 5d proved to be a
relatively inefficient inhibitor of SSAT in comparison with 6
(Figure 5A).
We did not detect 3b in DU145 cells supplemented with
50 μM 3a (Table 3). However, prior induction of SSAT by a
24 h treatment with compound 6 in mouse fetal fibroblasts
induced a detectable accumulation of 3b (data not shown),
suggesting that SSAT may be a functional acetylator of 3a.
3b elutes just after 2 in the used HPLC method, and small
amounts of 3b remain undetectable in the presence of 2.
However, treatment with compound 6 induced SSAT, and
furthermore, it depleted the 2 pool that facilitated detect-
ing 3b. Charge-deficiency of the central part of this type of
Interaction of Charge-Deficient Analogues with SSAT,
SMO, and APAO. Typically SSAT favors charged amino
groups separated by three carbons,31 but under certain
conditions SSAT can also acetylate compounds having
charged amino groups separated by two-carbon-atom back-
bone and even N8-aminoterminus of spermidine.32,33 All
tested compounds having a free aminoterminal moiety (3a,
5a, 5b, and 5c) proved to be substrates of SSAT under the
standard assay conditions (Table 4). Therefore, we ruled out
the nonenzymatic acetylation by using a reaction mixture
without recombinant SSAT and did not detect significant

5744 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 15
Weisell et al.
Table 5. SSATX/Total SSAT mRNA Ratio and Polyamine Levels in Mouse Primary Fetal Fibroblasts Treated with CHX and Polyamine Analoguesa
level (pmol/μg DNA)
treatment
SSATX/total SSAT ratio (fold)
1
2
4a
analogue
total PA
control
CHX
1.00 ( 0.08***
9.51 ( 2.58
7.92 ( 1.15
4.46 ( 0.33***
5.87 ( 0.27**
4.92 ( 0.74***
72 ( 6***
10 ( 8
<10
411 ( 4
88 ( 11
89 ( 15
92 ( 2
95 ( 13
106 ( 20
90 ( 15
499 ( 15
479 ( 47
658 ( 9**
666 ( 53**
841 ( 63***
647 ( 41 **
390 ( 33
389 ( 25
307 ( 23
380 ( 15
203 ( 21
CHX þ 3a
CHX þ 5a
CHX þ 5d
CHX þ 6
176 ( 31***
264 ( 31***
354 ( 41***
352 ( 17***
<10
<10
<10
a Cells were cultured with CHX (10 μg/mL) for 8 h, of which the last 7 h were with polyamine analogues (50 μM). The ratio of SSATX to total SSAT
mRNA (SSATX þ SSAT) was analyzed by quantitative RT-PCR. Data are the mean ( SD, n = 3. /, //, and /// refer to p values of <0.05, < 0.01, and
<0.001, respectively, compared to CHX-treated group. “Total PA” indicates the sum of 2, 4, and analogue. No metabolism of the analogues was
detectable by HPLC at 8 h.
Spm/Spd analogues has little affect in productive binding at
the SSAT active center.
more efficient reduction in SSATX/total SSAT ratio. Since
both compounds reduced SSATX mRNA level but only 6
caused SSAT superinduction (Table 2), it is likely that 5d is
not able to prolong the SSAT protein half-life, like 6,38
because of its low affinity toward SSAT (Figure 5A). The
fact that 5a and 5d, but not 3a, were able to regulate the
alternative splicing of SSAT implies that the charge of the
polyamine is important in the alternative splicing process.
These charge-deficient analoguesmay therefore help todissect
the mechanism of polyamines’ action in alternative splicing.
SMO is known to prefer 4a and then N1-Ac-Spm (4b) as its
substrates, while APAO has been shown to strongly prefer
acetylated polyamines.34,35 Such a substrate specificity of SMO
in conjunction with our recent finding that both oxidases
metabolize some N-alkylated polyamine analogues36 vali-
dated the testing of synthesized analogues for their substrate
properties. However, none of the charge-deficient analogues
exhibited substrate properties in a SMO-catalyzed reaction
(less than 5% product formation in 30 min in the presence of
1 μg of SMO; data not shown). Hence, in contrast with
SSAT, the adequate protonation of the central part of
pseudosubstrate molecules is essential for SMO-catalyzed
transformations of these analogues because in competition
studies all the analogues and their metabolites showed some
degree of inhibition (Figure 5C). Both APAO and SMO are
FAD-dependent oxidases, but unlike SMO, APAO does not
have such precise requirements for the degree of protonation
of the inner part of the substrate molecule(s). Thus, 5b was a
substrate of APAO (see above), while 5c turned out to be an
effective inhibitor of the enzyme (Figure 5B).
Conclusion
The obtained data clearly demonstrate that charge defi-
ciency of the central moiety of the 4a isoster, i.e., 5a, has in
some cases practically no effect on the biochemical properties
of the analogue:(i) 5a was a relativelygoodsubstrate of mouse
recombinant SSAT; (ii) 5b was a substrate of human recom-
binant APAO; (iii) 5a very effectively down-regulated ODC;
(iv) both 5a and 5d positively controlled the productive
splicing of SSAT mRNA.
In other cases, charge deficiency of the central moiety
restricted or diminished the recognition of the analogues as
polyamines: (i) 5d did not induce SSAT or SMO in DU145
cells; (ii) 5a and 5d accumulated efficiently inside DU145 cells,
although their affinity for polyamine transporter was low; (iii)
none of the compounds tested, except for 5b, was a substrate
of human recombinant SMO or APAO.
Finally, an effective intracellular conversion of 5a into 3a
makes it possible to consider 5a as a bioactive precursor of 3a
withenhancedbioavailability. Hence, the synthesizedisosteric
analogues of 4a having charge-deficient central moieties are
promising tools for investigating the cellular functions of the
natural polyamines and also the peculiarities of their recognition
by intracellular binding sites and their metabolic enzymes.
Our data also imply that SSAT could be a physiological acety-
lator of 3a.
Taken together, these data clearly indicate that the addi-
tion of the 3-aminopropyl group to 3a provides novel
properties to the analogue and converts it into a rather good
mimetic of 4a, i.e., 5a, which provides additional information
about the differences between SMO and APAO, especially in
the recognition of 4a derivatives with substrate properties.
Ability To Regulate the Alternative Splicing of SSAT. To
further study whether the charge-deficient analogues differ
from the natural polyamines in their cellular functions, their
ability to modulate the alternative splicing of SSAT was
investigated. Our previous work demonstrated that the
intracellular polyamine level regulates the alternative spli-
cing of SSAT; high polyamine level promotes the production
of normal SSAT mRNA, and low polyamine level promotes
the generation of an unproductive splice variant (SSATX),
which is subsequently targeted to protein-synthesis-depen-
dent degradation pathway called nonsense-mediated mRNA
decay (NMD).37 Primary fetal fibroblasts were treated with
CHX to shut down the rapid degradation of SSATX via
NMD. Then the effect of the analogues on the abundance of
SSATX mRNA relative to total SSAT was measured by
quantitative RT-PCR. As indicated in Table 5, 5a, 5d, and 6
decreased the SSATX/total SSAT ratio, whereas 3a did not
show statistically significant effect. In the case of the charge-
deficient analogues, SSATX/total SSAT mRNA ratio did
not correlate with polyamine concentration (Table 5).
Although the total polyamine level in 5d-treated cells was
much higher than that of 6-treated cells, the latter showed
Experimental Section
Materials. The human prostate carcinoma cell line DU145
was obtained from American Type Culture Collection. Non-
transgenic mouse primary fetal fibroblasts were prepared as
previously published.39 DFMO was obtained from ILEX On-
cology Inc., and putrescine dihydrochloride, spermidine trihy-
drochloride, and spermine tetrahydrochloride were from Sigma
Aldrich. [14C]-Labeled putrescine dihydrochloride (specific ac-
tivity 107 mCi/mmol), spermine tetrahydrochloride (specific
activity 112 mCi/mmol), spermidine trihydrochloride (specific
activity 113 mCi/mmol), L-ornithine (specific activity 57 mCi/mmol),
acetyl-CoA (specific activity 60 mCi/mmol), and S-adenosyl-L-
methionine (specific activity 54 mCi/mmol) were obtained from
GE Healthcare.

Article
Syntheses. N-(2-Aminoethyl)aminoethanol, ethylenediamine,
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 15 5745
dry DCM (100 mL) containing Et3N (4.23 mL, 30.3 mmol) and
was used (without purification) to alkylate N-(2-aminoethyl)-
aminoethanol (31.7 mL, 0.318 mol) in dry THF (40 mL) for 20 h
at 20 ꢀC. THF and unreacted amine were removed in vacuo
(0.1 mbar). To the residue 2 M NaOH (20 mL) was added, and
the resulting mixture was extracted with DCM (2 ꢀ 20 mL).
Extracts were washed with H2O (5 mL), brine (10 mL), dried
(K2CO3), and DCM was evaporated in vacuo. The residue was
purified on a silica gel column (125 g) using a mixture of
MeOH-Et3N (95-5) as an eluent to give 11 (3.1 g, 50%) as a
viscous oil. TLC, NMR, and ESI-MS were conducted.
methanesulfonyl chloride (MsCl), and acetyl chloride (AcCl)
were obtained from Fluka (Switzerland); triethylene tetraamine
(3a), benzyl chloroformate (CbzCl), salicylic aldehyde, O-meth-
ylhydroxylamine hydrochloride were from Aldrich. N-(Benzyl-
oxycarbonyl)-3-azapentanol-1 was synthesized as described
earlier.20 N-(Benzyloxycarbonyl)-3-amino-1-propyl methane-
sulfonate was prepared following a published procedure.21 N-
Ethyl-1,3-diaminopropane was synthesized by LiAlH4 reduc-
tion of corresponding nitrile as described earlier,40 and physi-
cochemical properties of the obtained product were identical
with the data published earlier.41 N-Acetyl-1,3-diaminopropane
hydrochloride was synthesized as described previously.19 6 was
synthesized essentially as described previously.42 5a was synthe-
sized as described previously.17,18 8a and 13 were synthesized as
described previously.18 3b and 3c were synthesized as described
earlier.43
Flash chromatography was performed on Kieselgel (40-
63 μm, Merck, Germany), and systems for elution are indicated
in the text. 1H and 13C NMR spectra were measured on a Bruker
Avance 500 DRX (Germany) using tetramethylsilane (TMS) in
CDCl3 or sodium 3-(trimethylsilyl)-1-propanesulfonate (TSP)
in D2O as internal standards. Chemical shifts are given in ppm.
The letter “J” indicates normal 3JHH couplings, and all J values
are given in Hz. High-resolution electrospray mass spectra were
obtained on Applied Biosystems/MDS Sciex QSTAR XL. All
final compounds were g98% pure, which was confirmed with
analytical HPLC used according to the published method.44
N12-Acetyl-N1,N3,N6-tris(benzyloxycarbonyl)-1,12-diamino-
3,6,9-triazadodecane (9a). Methanesulfonyl chloride (0.163 mL,
2.1 mmol) in dry DCM (3 mL) was added dropwise with stirring
to a cooled (0 ꢀC) solution of 8a18 (1.1 g, 2.0 mmol) and Et3N
(0.39 mL, 2.8 mmol) in dry DCM (10 mL). Stirring was
continued for 1 h at 0 ꢀC and 1 h at 20 ꢀC, and the reaction
mixture was poured into a 1 M NaHCO3 (5 mL) solution. The
organic layer was separated, washed with H2O (3 mL), 0.5 M
H2SO4 (3 ꢀ 4 mL), H2O (3 mL), 1 M NaHCO3 (3 mL), H2O
(3 mL), brine (5 mL) and dried (MgSO4). The solvent was eva-
porated in vacuo to give intermediate N3,N6,N8-tris(benzyloxy-
carbonyl)-8-amino-3,6-diaza-1-octyl methanesulfonate, which
was used (without purification) to alkylate N-(3-aminopro-
pyl)acetamide (2.3 g, 20 mmol) in dry THF (10 mL) first for
6 h at 20 ꢀC and then for 48 h at 37 ꢀC. The reaction mixture was
evaporated to dryness in vacuo, NaOH (1M, 5 mL) was added,
and the resulting mixture was extracted with DCM (2 ꢀ 3 mL).
Solvent was evaporated in vacuo and the residue was purified on
a silica gel column (120 g) using a mixture of dioxane-25%
NH4OH (100-0.7) as an eluent to give 9a (0.97 g, 75%) as a
viscous oil. TLC, NMR, and HRMS were conducted.
N3,N6,N9-Tris(benzyloxycarbonyl)-3,6,9-diazaundecanol-1 (8b).
Benzyl chloroformate (1.7 mL, 11.46 mmol) was added in three
portions within 15 min intervals to a cooled (0 ꢀC) and vigor-
ously stirred mixture of 11 (1.69 g, 5.5 mmol), Na2CO3 (2M,
6 mL), NaHCO3 (1.0 g, 11.9 mmol), THF (15 mL), and H2O
(5 mL). Stirring was continued for 1 h at 0 ꢀC and 4 h at room
temperature. The organic layer was separated, and the water
layer was extracted with DCM (2 ꢀ 10 mL). The combined
organic extracts were evaporated to dryness in vacuo. The
residue was dissolved in EtOAc (20 mL), washed with 1 M
HCl (3 ꢀ 5 mL), H2O (5 mL), 1 M NaHCO3 (5 mL), H2O (5 mL),
brine (10 mL) and dried (MgSO4). Solvent was distilled off in
vacuo and the residue was purified on a silica gel column (60 g)
using first CHCl3 followed by a mixture of CHCl3-MeOH
(95-5) as eluents to give 8b (2.8 g, 88%) as a viscous oil. TLC,
NMR, and ESI-MS were conducted.
N10,N13,N16-Tris(benzyloxycarbonyl)-3,7,10,13,16-pentaaza-
octadecane (9b). An intermediate N3,N6,N9-tris(benzyloxycarb-
onyl)-3,6,9-diaza-1-undecanoyl methanesulfonate was prepared
as N3,N6,N8-tris(benzyloxycarbonyl)-8-amino-3,6-diaza-1-octyl
methanesulfonate (synthesis of 9a) from 8b (2.74 g, 4.8 mmol)
and MsCl (0.4 mL, 5.2 mmol) in dry DCM (25 mL) containing
Et3N (1.0 mL, 7.1 mmol) and was used (without purification) to
alkylate N-ethyl-1,3-diaminopropane (5.1 g, 50 mmol) in THF
(10 mL) first for 6 h at 0 ꢀC and then for 24 h at 20 ꢀC. THF and
an excess of N-ethyl-1,3-diaminopropane were distilled off in
vacuo and the residue was purified on a silica gel column (60 g)
using a mixture of MeOH-Et3N (9-1) as an eluent to give crude
9b (2.0 g). To remove the product of the alkylation of the
secondary amino group of N-ethyl-1,3-diaminopropane, crude
9b was treated with salicylic aldehyde (0.074 g, 0.61 mmol) in
MeOH (12 mL) for 3 h at 20 ꢀC. The reaction mixture was
evaporated to dryness in vacuo and the residue was purified on a
silica gel column (110 g) using a mixture of MeOH-Et3N (9-1)
as an eluent to give 9b (1.2 g, 32%) as a viscous oil. TLC, NMR,
and ESI-MS were conducted.
3,7,10,13,16-Pentaazaoctadecane Pentahydrochloride, N1,N11-
Et2-SpmTrien (5d). The compound was prepared as 5b from 9b
(1.15 g, 1.7 mmol) to give 5d (0.54 g, 71%) as colorless crystals,
which did not melt below 250 ꢀC. Rf = 0.26; (E) 1H NMR (D2O)
δ 3.42-3.34 (10H, m, CH2NH), 3.34-3.28 (2H, m, CH2NH),
3.10 (2H, t, J = 8.0 Hz, CH2NH), 3.07-2.93 (6H, m, CH2NH),
2.04-1.94 (2H, m, CCH2C), 1.16 (3H, t, J = 7.2 Hz, CH3), 1.14
(3H, t, J = 7.3 Hz, CH3); 13C NMR (D2O) δ 48.05, 46.79, 46.74,
46.70, 46.55 (2 C), 46.49, 46.38, 46.07, 45.55, 25.63, 13.45 (2 C).
HRMS (ESIþ) calcd for C13H34N5 [M þ H]þ 260.2814, found
260.2812. Anal. C, H, N.
N12-Acetyl-1,12-diamino-3,6,9-triazadodecane Tetrahydroch-
loride, N12-Ac-SpmTrien (5b). Pd black in MeOH (∼0.5 mL) was
added to a solution of 9a (0.9 g, 1.39 mmol) in a mixture of AcOH-
MeOH (1:1, 20 mL), and hydrogenation was carried out at
atmospheric pressure. Solids were filtered off, washed with MeOH
(5 mL), and combined filtrates were evaporated to dryness in
vacuo. The residue was dissolved in MeOH (5 mL), diluted with
5 M HCl (2 mL), evaporated to dryness in vacuo and the residue
was recrystallized from a mixture H2O-MeOH-EtOH to give 7
(0.42 g, 77%) as colorless crystals, which did not melt below
N3,N6,N9,N12-Tetrakis(benzyloxycarbonyl)-1,12-diamino-3,6,
9-triazadodecane (14). A solution of 1318 (1.30 g, 2.15 mmol) and
salicylic aldehyde (0.275 g, 2.26 mmol) in THF (4 mL) was kept
for 1 h at 20 ꢀC followed by adding H2O (1.5 mL), NaHCO3
(0.200 g, 2.4 mmol), and Na2CO3 (2 M, 1.2 mL). The reaction
mixture was cooled to 0 ꢀC, and benzyl chloroformate (0.35 mL,
2.38 mmol) was added with vigorous stirring. After the mixture
was stirred at 0 ꢀC for 1 h and at 20 ꢀC for 3 h, the organic phase
was separated and a mixture of CH3ONH2 (0.425 g, 9 mmol)
and AcOH (30 μL, 0.5 mmol) was added. After being stirred for
2 h at 20 ꢀC, the reaction mixture was evaporated to dryness in
vacuo. The residue was dissolved in DCM (20 mL), washed with
1
250 ꢀC. Rf = 0.27 (E); H NMR (D2O) δ 3.44-3.26 (12H, m,
NHCH2CH2NH), 3.21-3.14 (2H, m, CH2NHAc), 3.07-2.99
(2H, m, AcNH(CH2)2CH2), 1.88 (3H, s, CH3), 1.86-1.78 (2 H,
m, CCH2C); 13C NMR (D2O) δ 174.99, 45.80, 44.76, 43.74 (2 C),
43.70, 43.32, 36.01, 35.56, 25.80, 21.95. HRMS (ESIþ) calcd for
C11H28N5O [M þ H]þ 246.2294, found 246.2298. Anal. C, H, N.
N9-(Benzyloxycarbonyl)-3,6,9-triazaundecanol-1 (11). An int-
ermediate N-(benzyloxycarbonyl)-3-aza-1-pentyl methanesul-
fonate was prepared as N3,N6,N8-tris(benzyloxycarbonyl)-8-
amino-3,6-diaza-1-octyl methanesulfonate (synthesis of 9a)
from 10 (4.5 g, 20.2 mmol) and MsCl (1.72 mL, 22.2 mmol) in

5746 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 15
Weisell et al.
1 M NaHCO3 (5 mL), H2O (5 mL), brine (10 mL), and dried
(K2CO3). The residue after evaporation was purified on a silica
gel column (65 g) using a mixture CHCl3-MeOH (85-15) as an
eluent to give 14 (1.25 g, 75%) as a viscous oil. TLC, NMR, and
HRMS were conducted.
ing polyamine/analogues (10, 100, 1000 μM). After a 10 min
incubation, the plates were washed twice with ice-cold PBS and
the cells lysed in 500 μL of 0.1 M NaOH. Then an amount of 200
μL of the lysate was mixed with 3 mL of Optiphase “HiSafe”
scintillation cocktail (PerkinElmer) and counted with liquid
scintillation counter (1450 Microbeta PLUS, Wallac). For anti-
zyme-depletion experiments, the cells were incubated in the
presence or absence of CHX (10 μg/mL) for 1 h after which
the incubation was continued with 50 μM analogue for 4 h in the
presence or absence of CHX before the determination of the
intracellular concentrations of analogues.
Experiments with Antizyme-Overexpressing DU145 Cells. A
mutant mouse AZ cDNA construct lacking T at site 205 was
used to get efficient AZ expression without the need for
frameshifting.47 For doxycycline (Dox) inducible expression, a
novel system containing a reverse transactivator, the gene
construct of choice under tetracycline responsive element, and
a selection marker in one lentiviral vector was used.48 The
lentiviral particles were produced as described earlier.48
DU145 cells were infected at MOI 1 and placed under neomycin
selection (0.5 mg/mL Geneticin (Sigma Aldrich)) for 2 weeks.
For uptake experiments, the cells were plated onto six-well
plates at a density of 1 ꢀ 106 cells/well and incubated for 24 h
in the presence or absence of Dox (250 ng/mL). Medium
containing analogues (50 μM) with or without doxycycline
was then changed, and the cells were harvested after 6 h of
incubation. The intracellular concentrations of analogues were
measured with HPLC.
Experiments with Recombinant Proteins. Plasmids encoding
human recombinant SMO or APAO were a kind gift from Dr.
Carl Porter, Roswell Park Cancer Institute, Buffalo, NY. The
enzymes were produced as described previously.49 Recombinant
SSAT was purified, and kinetic measurements were done as
described previously.6 Initial studies with APAO and SMO were
performed using fixed analogue concentration (200 μM) with
0.1 or 1.0 μg of recombinant protein and incubated for 30 min in
a 37 ꢀC water bath. Samples were analyzed with HPLC with
postcolumn o-phthaldialdehyde derivatization to determine the
concentrations of the reaction products. 5b proved to be a
substrate of APAO in initial screening; therefore, a more
detailed kinetic was carried out in triplicate using a 50, 100,
250, 500, and 2500 μM substrate concentration range with 0.2 μg
of APAO incubated for 5 or 10 min to verify the linearity of the
observed reaction velocity. Kinetic values were determined by
Lineweaver-Burk plotting using Graph Pad Prism 4.03 soft-
ware with nonlinear fitting. Kcat values were determined using
an Mw of 55 382 for recombinant human APAO as monomer
consisting of one catalytically active center.
APAO and SMO inhibition experiments were measured as
follows. Competitor (50 μL) was pipetted in black 96-well plates
containing assay buffer (50 mM potassium phosphate buffer,
pH 7.8, 1 mM EDTA, 0.02 U/mL horseradish peroxidase, and
10 μM 4a (SMO) or 10 μM 4b (APAO)). The reaction was
initiated by the addition of 50 μL of enzyme mixture (50 μM
Ampliflu Red, 0.1% Triton X-100, and 0.05 μg of recombinant
SMO/APAO) to a total reaction volume of 200 μL. Fluores-
cence was measured for 15-30 min at excitation 540 nm and
emission 590 nm with a Victor2 counter (PerkinElmer) at 1 min
intervals. A standard curve (0-3175 pmol) was prepared im-
mediately before measurement using dilutions of H2O2 from
fresh 30% stock solution.
N1-Acetyl-N3,N6,N9,N12-Tetrakis(benzyloxycarbonyl)-1,12-
diamino-3,6,9-triazadodecane (15). To a cooled (0 ꢀC) and stirred
mixture of 14 (1.2 g, 1.62 mmol) and Et3N (0.28 mL, 2 mmol) in
dry THF (8 mL) was added dropwise within 5 min a solution of
AcCl (0.13 mL, 1.86 mmol) in dry THF (2 mL). Stirring was
continued for 30 min at 0 ꢀC, MeOH (10 mL) was added, and
after 10 min of stirring the mixture was evaporated to dryness in
vacuo. The residue was dissolved in CHCl3 (20 mL), washed
with NaHCO3 (1 M, 3 ꢀ 5 mL), H2O (5 mL), H2SO4 (0.5 M, 3 ꢀ
5 mL), H2O (5 mL), brine (10 mL), and dried (MgSO4). The
solvent was evaporated in vacuo to give 15 (1.22 g, 95%) as a
viscous oil. TLC, NMR, and HRMS were conducted.
N1-Acetyl-1,12-diamino-3,6,9-triazadodecane Tetrahydrochlo-
ride, N1-Ac-SpmTrien (5c). The compound was prepared as 5b
from 15 (1.17 g, 1.5 mmol) to give 5c (0.424 g, 72%) as colorless
1
crystals, which did not melt below 250 ꢀC. Rf = 0.27; (E) H
NMR (D2O) δ 3.55-3.40 (10H, m, CH2NH), 3.26-3.16 (4H, m,
CH2NHAc þ CCCH2), 3.09-3.01 (2H, m, CH2NH2), 2.11-
2.01 (2H, m, CCH2C), 1.95 (3H, s, CH3); 13C NMR (D2O) δ
177.24, 51.11, 48.14, 46.55, 46.43, 46.15 (2 C), 39.45, 38.78,
26.69, 24.90. HRMS (ESIþ) calcd for C11H28N5O [M þ H]þ
246.2294, found 246.2297. Anal. C, H, N.
Cell Culture. Cells were cultured in Dulbecco’s modified
Eagle’s medium (DMEM, Sigma Aldrich) supplemented with
10% heat-inactivated fetal bovine serum (Sigma Aldrich) and 50
μg/mL gentamycin (Sigma Aldrich). The cells were incubated in
a humidified atmosphere at þ37 ꢀC, 10% CO2. The cells were
detached using a solution containing 0.25% trypsin and 1 mM
EDTA. The pelleted cells were washed with phosphate-buffered
saline (PBS), pelleted, and stored at -70 ꢀC. The cell number
was measured with the Coulter Counter model Z1. Cell pellets
were lysed in buffer (50 mM potassium phosphate buffer, pH
7.2, 1 mM EDTA, 0.1% Triton X-100, 0.1 mM dithiothreitol,
complete EDTA-free protease inhibitor cocktail (Roche Diag-
nostics)) and incubated for 20 min on ice. Samples for polyamine
measurement were taken and mixed 9:1 with 50% sulfosalicylic
acid containing 100 μM diaminoheptane. The rest of the lysate
was centrifuged at 13 000 rpm for 20 min at þ4 ꢀC. The super-
natant fraction was used for the assay of SSAT, ODC, Ado-
MetDC, SMO, and APAO activities.
Enzyme Activities, Polyamines, and Analogues. Intracellular
polyamines and polyamine analogues were measured with
HPLC according to published methods.44 The amount of
DNA was measured from the pellets of polyamine samples
using PicoGreen (Invitrogen) according to the manufacturer’s
instructions. Dilutions of calf thymus DNA (Sigma Aldrich)
were used as standards. ODC, AdoMetDC, and SSAT activities
were measured as described earlier.7,45,46 SMO and APAO
activities were measured as follows. Cell samples (20 μL) were
preincubated in black 96-well plates for 15 min in an assay buffer
containing 50 mM potassium phosphate buffer, pH 7.8, 1 mM
pargyline, 0.1 mM semicarbazide, and 0.02 U/mL horseradish
peroxidase (Roche Diagnostics). Then a mixture of 50 μM
Ampliflu Red (Sigma Aldrich) and 1 mM 4a (SMO) or 4b
(APAO) was added to a total volume of 200 μL. Fluorescence
was measured for 30-60 min at excitation 540 nm and emission
590 nm with a Victor2 counter (PerkinElmer) at 2 min intervals.
A standard curve (0-2500 pmol) was prepared immediately
before measurement using dilutions of H2O2 from fresh 30%
stock solution.
Alternative Splicing of SSAT. Mouse primary fetal fibroblasts
were seeded in six-well plates and incubated overnight. The cells
were incubated in the presence of CHX (10 μg/mL) for 1 h after
which 50 μM analogue was added, and incubation continued for
a further 7 h. SSAT-X/total SSAT ratio was analyzed by
quantitative RT-PCR as described earlier.37
Western Blotting. Total protein concentrations were mea-
sured using Coomassie Brilliant Blue (Bio-Rad) with dilutions
of bovine serum albumin (Bio-Rad) as standards. Equal amounts
Uptake Experiments. For competition experiments, DU145
cells were plated onto six-well plates at a density of 1 ꢀ 106 cells/
well. After overnight incubation, the medium was changed to
prewarmed serum-free medium supplemented with 10 μM [14C]-
labeled 1, 2, or 4a (specific activity 25 mCi/mmol) and compet-

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 15 5747
of proteins (10 μg) were run on 12% SDS-polyacrylamide gel,
and samples were then transferred onto a PVDF membrane
(Immobilon, Millipore). The AZ protein was detected by using 1
μg/mL rabbit anti-mouse AZ antibody (a kind gift from Prof.
copper(II)-selective chelator ameliorates left-ventricular hypertro-
phy in type 2 diabetic patients: a randomised placebo-controlled
study. Diabetologia 2009, 52, 715–722.
(14) Cooper, G. J.; Phillips, A. R.; Choong, S. Y.; Leonard, B. L.;
Crossman, D. J.; Brunton, D. H.; Saafi, L.; Dissanayake, A. M.;
Cowan, B. R.; Young, A. A.; Occleshaw, C. J.; Chan, Y. K.; Leahy,
F. E.; Keogh, G. F.; Gamble, G. D.; Allen, G. R.; Pope, A. J.;
Boyd, P. D.; Poppitt, S. D.; Borg, T. K.; Doughty, R. N.; Baker,
J. R. Regeneration of the heart in diabetes by selective copper
chelation. Diabetes 2004, 53, 2501–2508.
(15) Lu, J.; Poppitt, S. D.; Othman, A. A.; Sunderland, T.; Ruggiero,
K.; Willet, M. S.; Diamond, L. E.; Garcia, W. D.; Roesch, B. G.;
Cooper, G. J. Pharmacokinetics, pharmacodynamics, and meta-
bolism of triethylenetetramine in healthy human participants: an
open-label trial. J. Clin. Pharmacol. 2010, 50, 647–658.
€
Olli Janne, Finland) and anti-rabbit Dylight 800 secondary
antibody (Thermo Fischer Scientific). The antibody-bound
protein was visualized by an infrared scanner (Odyssey Imager,
Li-Cor Biosciences).
Statistical Analysis. Values are the mean ( SD. One-way
analysis of variance (ANOVA) with Tuckey’s or Dunnett’s post
hoc test was used for multiple comparisons with the aid of a
software package, GraphPad Prism 4.03 (GraphPad Software
Inc.).
(16) Khomutov, A. R.; Grigorenko, N. A.; Skuridin, S. G. Novel
approach to design an isosteric charge-deficient analogue of sper-
mine and its biochemically important derivatives. Biochem. Soc.
Trans. 2007, 35, 369–373.
Acknowledgment. The authors warmly thank Sisko Juuti-
nen, Anne Karppinen, Arja Korhonen, Tuula Reponen, and
Maritta Salminkoski for skillful technical assistance. This
work was supported by grants from the Academy of Finland,
the Russian Foundation for Basic Research (Project Nos. 09-
04-01272 and 08-04-91777), and the program Molecular and
Cell Biology of the Presidium of the Russian Academy of
Sciences.
€
€ €
€
(17) Weisell, J.; Hyvonen, M. T.; Vepsalainen, J.; Alhonen, L.; Keinanen,
T. A.; Khomutov, A. R.; Soininen, P. Novel isosteric charge-
deficient spermine analogue;1,12-diamino-3,6,9-triazadodecane:
Synthesis, pKa measurement and biological activity. Amino Acids
2010, 38, 501–507.
(18) Khomutov, A. R.; Grigorenko, N. A.; Skuridin, S. G.; Demin,
€ €
A. V.; Vepsalainen, J.; Casero, R. A.; Woster, P. M. A charge-
deficient analogue of spermine with chelating properties. Russ.
J. Bioorg. Chem. 2005, 31, 271–278.
(19) Tabor, H.; Tabor, C. W.; de Meis, L. Chemical Synthesis of N-
Acetyl-1,4-diamiho-butane, N1-Acetylpermidine, and N8-Acetyl-
spermidine. In Methods in Enzymology; Academic Press: New York,
1971; Vol. 17, pp 829-833.
(20) Wendta, B.;Hab, H. R.;Hessea, M. Synthesisoftwometabolitesofthe
antiarrythmicum amiodarone. Helv. Chim. Acta 2002, 85, 2990–3001.
(21) Khomutov, M. A.; Mandal, S.; Weisell, J.; Saxena, N.; Simonian,
Supporting Information Available: Synthesis and the spectro-
scopic details for compounds 8a and 13; physicochemical data
for compounds 9a, 11, 8b, 9b, and 13-15; and elementary
analysis results of target compounds. This material is available
free of charge via the Internet at http://pubs.acs.org.
€ €
A. R.; Vepsalainen, J.; Madhubala, R.; Kochetkov, S. N. Novel
convenient synthesis of biologically active esters of hydroxylamine.
References
Amino Acids 2010, 38, 509–517.
(22) Khomutov, A. R.; Shvetsov, A. S.; Vepsalainen, J. J.; Kritzyn,
(1) Pegg, A. E.; Michael, A. J. Spermine synthase. Cell. Mol. Life Sci.
2010, 67, 113–121.
(2) Bergeron, R. J.; McManis, J. S.; Weimar, W. R.; Schreier, K. M.;
Gao, F.; Wu, Q.; Ortiz-Ocasio, J.; Luchetta, G. R.; Porter, C.;
Vinson, J. R. The role of charge in polyamine analoguerecognition.
J. Med. Chem. 1995, 38, 2278–2285.
(3) Seiler, N. Thirty years of polyamine-related approaches to cancer
therapy. Retrospect and prospect. Part 2. Structural analogues and
derivatives. Curr. Drug Targets 2003, 4, 565–585.
(4) Casero, R. A., Jr.; Wang, Y.; Stewart, T. M.; Devereux, W.; Hacker,
A.; Wang, Y.; Smith, R.; Woster, P. M. The role of polyamine
catabolism in anti-tumour drug response. Biochem. Soc. Trans.
2003, 31, 361–365.
€ €
A. M. Novel acid-free cleavage of N-(2-hydroxyarylidene) pro-
tected amines. Tetrahedron Lett. 2001, 42, 2887–2889.
(23) Tanabe, R.; Kobayashi, M.; Sugawara, M.; Iseki, K.; Miyazaki, K.
Uptake mechanism of trientine by rat intestinal brush-border
membrane vesicles. J. Pharm. Pharmacol. 1996, 48, 517–521.
(24) Kobayashiþ, M.; Iseki, K.; Sugawara, M.; Miyazaki, K. The diver-
sity of Na -independent uptake systems for polyamines in rat
intestinal brush-border membrane vesicles. Biochim. Biophys. Acta
1993, 1151, 161–167.
(25) Suzuki, T.; He, Y.; Kashiwagi, K.; Murakami, Y.; Hayashi, S.;
Igarashi, K. Antizyme protects against abnormal accumulation
and toxicity of polyamines in ornithine decarboxylase-overprodu-
cing cells. Proc. Natl. Acad. Sci. U.S.A. 1994, 91, 8930–8934.
(26) Kanamoto, R.; Kameji, T.; Iwashita, S.; Igarashi, K.; Hayashi, S.
Spermidine-induced destabilization of ornithine decarboxylase
(ODC) is mediated by accumulation of antizyme in ODC-over-
producing variant cells. J. Biol. Chem. 1993, 268, 9393–9399.
(27) Mitchell, J. L.; Leyser, A.; Holtorff, M. S.; Bates, J. S.; Frydman,
B.; Valasinas, A. L.; Reddy, V. K.; Marton, L. J. Antizyme
induction by polyamine analogues as a factor of cell growth
inhibition. Biochem. J. 2002, 366, 663–671.
(5) Lakanen, J. R.; Coward, J. K.; Pegg, A. E. R-Methyl polyamines:
metabolically stable spermidine and spermine mimics capable of
supporting growth in cells depleted of polyamines. J. Med. Chem.
1992, 35, 724–734.
€
€
(6) Jarvinen, A.; Grigorenko, N.; Khomutov, A. R.; Hyvonen, M. T.;
€ €
€
Uimari, A.; Vepsalainen, J.; Sinervirta, R.; Keinanen, T. A.; Vujcic,
S.; Alhonen, L.; Porter, C. W.; Janne, J. Metabolic stability of
€
R-methylated polyamine derivatives and their use as substitutes for
the natural polyamines. J. Biol. Chem. 2005, 280, 6595–6601.
€
(7) Hyvonen, M. T.; Howard, M. T.; Anderson, C. B.; Grigorenko, N.;
€ €
€
€
Khomutov, A. R.; Vepsalainen, J.; Alhonen, L.; Janne, J.; Keinanen,
T. A. Divergent regulation of the key enzymes of polyamine
metabolism by chiral alpha-methylated polyamine analogues. Bio-
chem. J. 2009, 422, 321–328.
(28) Matsufuji, S.; Matsufuji, T.; Miyazaki, Y.; Murakami, Y.; Atkins,
J. F.; Gesteland, R. F.; Hayashi, S. Autoregulatory frameshifting
in decoding mammalian ornithine decarboxylase antizyme. Cell
1995, 80, 51–60.
(8) Baillon, J. G.; Mamont, P. S.; Wagner, J.; Gerhart, F.; Lux, P.
Fluorinated analogues of spermidine as substrates of spermine
synthase. Eur. J. Biochem. 1988, 176, 237–242.
(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.
€ €
(9) Khomutov, A. R.; Shvetsov, A. S.; Vepsalainen, J.; Kramer, D. L.;
€
€
Hyvonen, T.; Keinanen, T.; Eloranta, T. O.; Porter, C. W.; Khomutov,
R. M. New aminooxy analogs of biogenic polyamines. Bioorg.
Khim. 1996, 22, 557–559.
€
€
(30) Jarvinen, A.; Keinanen, T. A.; Grigorenko, N. A.; Khomutov, A. R.;
€ €
€
€
€
Uimari, A.; Vepsalainen, J.; Narvanen, A.; Alhonen, L.; Janne, J.
Guide molecule-driven stereospecific degradation of alpha-methylpo-
lyamines by polyamine oxidase. J. Biol. Chem. 2006, 281, 4589–4595.
(31) Della Ragione, F.; Pegg, A. E. Studies of the specificity and kinetics
of rat liver spermidine/spermine N1-acetyltransferase. Biochem.
J. 1983, 213, 701–706.
(10) Lin, P. K. T.; Maguire, N. M.; Brown, D. M. Synthesis of novel
oxa- isosteres of spermidine and spermine. Tetrahedron Lett. 1994,
35, 3605–3608.
€
(11) Schwarzenbach, G. Metallkomplexe mit polyaminen III: mit tria-
thylen-tetramin = “trien”. Helv. Chim. Acta 1950, 33, 974–985.
(12) Dixon, H. B.; Gibbs, K.; Walshe, J. M. Preparation of triethyle-
netetramine dihydrochloride for the treatment of Wilson’s disease.
Lancet 1972, 7755, 853.
(13) Cooper, G. J.; Young, A. A.; Gamble, G. D.; Occleshaw, C. J.;
Dissanayake, A. M.; Cowan, B. R.; Brunton, D. H.; Baker, J. R.;
Phillips, A. R.; Frampton, C. M.; Poppitt, S. D.; Doughty, R. N. A
(32) Hegde, S. S.; Chandler, J.; Vetting, M. W.; Yu, M.; Blanchard, J. S.
Mechanistic and structural analysis of human spermidine/sper-
mine N1-acetyltransferase. Biochemistry 2007, 46, 7187–7195.
(33) Montemayor, E. J. Biochemical Studies of Spermidine/Spermine
N1-Acetyltransferase, an Important Regulator of Cellular Polyamines.
Ph.D. Dissertation, University of Texas at Austin, TX, 2008.

5748 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 15
Weisell et al.
(34) Vujcic, S.; Diegelman, P.; Bacchi, C. J.; Kramer, D. L.; Porter,
C. W. Identification and characterization of a novel flavin-contain-
ing spermine oxidase of mammalian cell origin. Biochem. J. 2002,
367, 665–675.
(42) Rehse, K.; Puchert, E.; Leissring, S. Platelet aggregation inhibiting
and anticoagulant effects of oligoamines, XII: alkyl and arylalkyl
derivatives of putrescine, spermidine, and spermine. Arch. Pharm.
(Weinheim, Ger.) 1990, 323, 287–294.
(35) Seiler, N. Polyamine oxidase, properties and functions. Prog. Brain
(43) Kuswik-Rabiega, G.; Bruenger, F. W.; Miller, S. C. Regiospecific
synthesis of mono- and diacetylated polyethylenepolyamine deri-
vatives. Synth. Commun. 1992, 22, 1307–1318.
Res. 1995, 106, 333–344.
€
(36) Hakkinen, M. R.; Hyvonen, M. T.; Auriola, S.; Casero, R. A., Jr.;
€ €
€
€
€
Vepsalainen, J.; Khomutov, A. R.; Alhonen, L.; Keinanen, T. A.
Metabolism of N-alkylated spermine analogues by polyamine and
spermine oxidases. Amino Acids 2010, 38, 369–381.
(44) Hyvonen, T.; Keinanen, T. A.; Khomutov, A. R.; Khomutov, R. M.;
Eloranta, T. O. Monitoring of the uptake and metabolism of ami-
nooxy analogues of polyamines in cultured cells by high-performance
liquid chromatography. J. Chromatogr. 1992, 574, 17–21.
€
€
(37) Hyvonen, M. T.; Uimari, A.; Keinanen, T. A.; Heikkinen, S.;
€
Pellinen, R.; Wahlfors, T.; Korhonen, A.; Narvanen, A.; Wahlfors,
J.; Alhonen, L.; Janne, J. Polyamine-regulated unproductive spli-
€
€
(45) Janne, J.; Williams-Ashman, H. G. On the purification of L-
€
ornithine decarboxylase from rat prostate and effects of thiol
compounds on the enzyme. J. Biol. Chem. 1971, 246, 1725–1732.
(46) Libby, P. R. Calf liver nuclear N-acetyltransferases. Purification
and properties of two enzymes with both spermidine acetyltrans-
ferase and histone acetyltransferase activities. J. Biol. Chem. 1978,
253, 233–237.
cing and translation of spermidine/spermine N1-acetyltransferase.
RNA 2006, 12, 1569–1582.
(38) Fogel-Petrovic, M.; Kramer, D. L.; Vujcic, S.; Miller, J.; McManis,
J. S.; Bergeron, R. J.; Porter, C. W. Structural basis for differential
induction of spermidine/spermine N1-acetyltransferase activity by
novel spermine analogs. Mol. Pharmacol. 1997, 52, 69–74.
(39) Alhonen, L.; Karppinen, A.; Uusi-Oukari, M.; Vujcic, S.; Korhonen,
€
(47) Kankare, K.; Uusi-Oukari, M.; Janne, O. A. Structure, organiza-
tion and expression of the mouse ornithine decarboxylase antizyme
gene. Biochem. J. 1997, 324, 807–813.
€
V. P.; Halmekyto, M.; Kramer, D. L.; Hines, R.; Janne, J.; Porter,
€
C. W. Correlation of polyamine and growth responses to N1,N11
-
(48) Shin, K. J.; Wall, E. A.; Zavzavadjian, J. R.; Santat, L. A.; Liu, J.;
Hwang, J. I.; Rebres, R.; Roach, T.; Seaman, W.; Simon, M. I.;
Fraser, I. D. A single lentiviral vector platform for microRNA-
based conditional RNA interference and coordinated transgene
expression. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 13759–13764.
diethylnorspermine in primary fetal fibroblasts derived from trans-
genic mice overexpressing spermidine/spermine N1-acetyltransfer-
ase. J. Biol. Chem. 1998, 273, 1964–1969.
€
€
(40) Hakkinen, M. R.; Keinanen, T. A.; Khomutov, A. R.; Auriola, S.;
€
€ €
€
Weisell, J.; Alhonen, L.; Janne, J.; Vepsalainen, J. Synthesis of
novel deuterium labelled derivates of N-alkylated polyamines.
Tetrahedron 2009, 65, 547–562.
(49) Jarvinen, A. J.; Cerrada-Gimenez, M.; Grigorenko, N. A.; Khomutov,
€ €
A. R.; Vepsalainen, J. J.; Sinervirta, R. M.; Keinanen, T. A.; Alhonen,
L. I.; Janne, J. E. Alpha-methyl polyamines: efficient synthesis and
€
€
(41) Tarbell, D. S.; Shakespeare, N.; Claus, C. J.; Bunnett, J. F. The
synthesis of some 7-chloro-4-(3-alkylaminopropylamino)-quino-
lines. J. Am. Chem. Soc. 1946, 68, 1217–1219.
tolerance studies in vivo and in vitro. First evidence for dormant
stereospecificity of polyamine oxidase. J. Med. Chem. 2006, 49,
399–406.