Polyamine modification by acrolein exclusively produces 1,5-diazacyclooctanes: A previously unrecognized mechanism for acrolein-mediated oxidative stress

Article abstract of DOI:10.1039/c4ob00761a

Acrolein, a toxic unsaturated aldehyde generated as a result of oxidative stress, readily reacts with a variety of nucleophilic biomolecules. Polyamines, which produced acrolein in the presence of amine oxidase, were then found to react with acrolein to produce 1,5-diazacyclooctane, a previously unrecognized but significant downstream product of oxidative stress. Although diazacyclooctane formation effectively neutralized acrolein toxicity, the diazacyclooctane hydrogel produced through a sequential diazacyclooctane polymerization reaction was highly cytotoxic. This study suggests that diazacyclooctane formation is involved in the mechanism underlying acrolein-mediated oxidative stress.

Full text of DOI:10.1039/c4ob00761a

Organic &
Biomolecular Chemistry
View Article Online
View Journal
PAPER
Polyamine modification by acrolein exclusively
produces 1,5-diazacyclooctanes: a previously
unrecognized mechanism for acrolein-mediated
oxidative stress†
Cite this: DOI: 10.1039/c4ob00761a
Ayumi Tsutsui,a Rie Imamaki,b Shinobu Kitazume,b Shinya Hanashima,b
Yoshiki Yamaguchi,b Masato Kaneda,c Shinya Oishi,c Nobutaka Fujii,c
Almira Kurbangalieva,d Naoyuki Taniguchib and Katsunori Tanaka*a,d
Acrolein, a toxic unsaturated aldehyde generated as a result of oxidative stress, readily reacts with a variety
of nucleophilic biomolecules. Polyamines, which produced acrolein in the presence of amine oxidase,
were then found to react with acrolein to produce 1,5-diazacyclooctane, a previously unrecognized but
significant downstream product of oxidative stress. Although diazacyclooctane formation effectively
neutralized acrolein toxicity, the diazacyclooctane hydrogel produced through a sequential diazacyclo-
octane polymerization reaction was highly cytotoxic. This study suggests that diazacyclooctane formation
is involved in the mechanism underlying acrolein-mediated oxidative stress.
Received 11th April 2014,
Accepted 16th May 2014
DOI: 10.1039/c4ob00761a
www.rsc.org/obc
Acrolein conjugates are currently used as biomarkers of oxi-
Introduction
dative stress7 in the context of a variety of diseases. Acrolein–
Acrolein is a highly toxic unsaturated aldehyde1 that can be amino conjugates involving, for example, lysine δ-amino
produced during the burning of oils, charcoal, wood, or groups, 3-formyl-3,4-dehydropiperidine (FDP),7 or 3-methyl-
plastic. It can also be generated by cells under oxidative stress pyridinium (MP) derivatives10 have been described. Antibodies11
conditions (through the enzymatic oxidation of threonine to these conjugates are widely used for the immunochemical
or polyamines2–4) or during reactive oxygen species (ROS)- detection of various disease states, such as arteriosclerosis,11–14
mediated oxidation of highly unsaturated lipids.5 The unsub- Alzheimer’s disease,15,16 tumors,17–21 diabetes,22–26 autoimmune
stituted and most reactive 2-alkenal produced through the
latter pathway can react with nearby thiol, hydroxyl, or amino
functional groups on DNA,6 proteins,7 or phosphatidyl
ethanolamines to accelerate the oxidative stress processes
associated with various disease states.8,9 Studies of acrolein
conjugates could, therefore, contribute to an understanding of
the relationship between acrolein and oxidative stress and,
hence, disease at a molecular level.
aBiofunctional Synthetic Chemistry Laboratory, RIKEN, 2-1 Hirosawa, Wako-shi,
Saitama 351-0198, Japan. E-mail: kotzenori@riken.jp; Fax: +(81) 48-467-9379;
Tel: +(81) 48-467-9405
bSystems Glycobiology Research Group, RIKEN-Max Planck Joint Research Center for
Systems Chemical Biology, RIKEN Global Research Cluster, RIKEN, 2-1 Hirosawa,
Wako-shi, Saitama 351-0198, Japan
cGraduate School of Pharmaceutical Sciences, Kyoto University,
46-29 Yoshidashimoadachi-cho, Sakyo-ku, Kyoto 606-8501, Japan
dBiofunctional Chemistry Laboratory, A. Butlerov Institute of Chemistry,
Scheme 1 1,5-Diazacyclooctanes are previously unrecognized pro-
ducts of the reaction of polyamines with acrolein. Proposed reaction
mechanism governing the OH- and amine-mediated “head-to-tail”
imino [4 + 4] reaction of unsaturated imines.
Kazan Federal University, 18 Kremlyovskaya street, Kazan 420008, Russia
†Electronic supplementary information (ESI) available. See DOI:
10.1039/c4ob00761a
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
disease,27,28 high blood pressure,29 and so on.30–34 Alter- reaction readily provided eight-membered heterocycles, the
natively, acrolein can react with polyamines to produce FDP 1,5-diazacyclooctanes, at micromolar concentrations. The
conjugates that modulate the cytotoxicity of acrolein.9
nucleophilic hydroxyl groups on the imino nitrogen (X = OH in
In our research program, which explores the novel reactiv- Scheme 1) accelerated and stabilized 1,5-diazacyclooctane for-
ities of imines,35–38 we recently found by chance that the un- mation. Based on the reactivity profiles, we hypothesized that
saturated N-alkylimines participated in the hitherto unknown amino groups (X = RNH in Scheme 1) could similarly form diaza-
“head-to-tail” [4 + 4] dimerization in the presence of hydroxyl- cyclooctanes. For example, these results suggested that poly-
ated alkyl groups on the imino nitrogen (Scheme 1).39 The amines could react with acrolein to provide diazacyclooctanes
through the [4 + 4] dimerization of the imines, as illustrated in
Scheme 2. Considering that acrolein is produced as a poly-
amine metabolite by amine oxidase2–4 during oxidative stress
processes, the diazacyclooctanes (acrolein-modified polyamines),
if formed, may be involved in the mechanisms underlying
acrolein-mediated oxidative stress.
In this study, we investigated the reactions of the poly-
amines with acrolein. We revealed for the first time that 1,5-di-
azacyclooctanes are the exclusive and previously unrecognized
products of polyamine modification by acrolein. The cyto-
toxicity, effects on oxidative stress, and oxidative metabolism of
the acrolein-modified polyamines were evaluated in an effort
to elucidate the mechanism underlying acrolein-mediated
oxidative stress.
Results and discussion
The reaction of spermine 1, N1-Ac-spermine 2, and spermidine
3 with 1 equivalent acrolein at room temperature smoothly
produced the corresponding eight-membered diazacyclooctane
derivatives 1a–3a within 15 min (Scheme 2(a)–(d)). These pro-
ducts could exist as a 1 : 1 mixture of the diastereomers at the
acetal stereogenic centers, which are interconvertible with
each other under equilibrium conditions. It should be noted
that the reaction proceeded in both chloroform and PBS buffer
with similar efficiencies and at a concentration of 100 μM; that
is, the reactivity was tested under biologically relevant concen-
trations of polyamines present in and on cells.40–42 Although
an MS analysis of the crude mixture detected small quantities
of the FDP derivatives among the starting polyamines, the 1,5-
diazacyclooctane derivatives were generated exclusively, i.e., in
more than 80% yield, and very rapidly under mild conditions,
as shown in Scheme 2. The structures of 1,5-diazacyclooctane
Scheme 2 Reaction of various polyamines, at concentrations ranging
from 10 µM to 100 mM, with acrolein in either chloroform (CHCl3) or
PBS buffer at room temperature. (a) Reaction of spermine 1 with 1 equi-
were unambiguously determined for the spermine derivative
1a using MS, 1H, 13C, and various 2D NMR techniques. The
spectra of the derivatives 2a and 3a were compared with those
2 equivalents acrolein gradually produced the of 1a. The structurally more simple diamine, putrescine 4, on
valent acrolein to provide the 1,5-diazacyclooctane 1a. (b) The reaction
of spermine with
1
1,5-diazacyclooctane polymers 1aa, i.e., the polymers with molecular
weights of ca. 6000 after 5 h, 15 000 after 10 h, and an insoluble hydro-
gel that adsorbed 16 g water g−1 material (w/w) after 24 h. The mole-
cular weight was estimated using diffusion ordered NMR spectroscopy
(DOSY) techniques. (c) The reaction of N1-acetyl spermine 2 with 1 equi-
the other hand, gave a mixture of products, approximately
50% of which comprised the bridged irreversible 2,6,9-triazabi-
cyclo[3.3.1]nonane structure 4a (Scheme 2(e)). The reactivity of
acrolein toward the simple diamine was similar to that of
valent acrolein yielded 1,5-diazacyclooctane 2a. (d) The reaction of sper- N1-Me-propanediamine 4′ (Scheme 2(f)), which smoothly pro-
midine 3 with 1 equivalent acrolein provided 1,5-diazacyclooctane 3a.
(e) The reaction of putrescine 4 with 1 equivalent acrolein provided a
mixture of products that contained the 2,6,9-triazabicyclo[3.3.1]nonane
duced the caged product 4a′ in quantitative yield, as reported
previously.43
The reaction of spermine 1 with 2 equivalents acrolein in
PBS (Scheme 2(b)) resulted in a [4 + 4] polymerization reaction
between the two imines generated at the two amino termini of
4a as the main product (in approximately 50% yield, based on the
1H NMR results). (f) Quantitative reaction of N1-Me-propanediamine
with 0.7 equivalents acrolein to provide the caged compound 4a’.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
Fig. 1 Cytotoxic activities and effects on oxidative stress of polyamines and their acrolein-modified products. (a) Cytotoxic activities of the poly-
amines, the 1,5-diazacyclooctanes, and the 2,6,9-triazabicyclo[3.3.1]nonane derivatives 1–4a’. HeLa cells were treated with compounds 1–4a’ for
72 h at 37 °C, and the cytotoxicities were evaluated using the MTS method. For the amine oxidase inhibition experiments, 1 mM of the aminoguani-
dine was applied. The acrolein–putrescine conjugate, i.e., the caged compound 4a, was obtained as a mixture with other byproducts, and the pure
4a was not isolated; therefore, the cytotoxicity and the oxidative stress activities were evaluated using a similar caged compound, 4a’, which was
obtained as a single product from the corresponding N1-Me-propanediamine. Reference IC50 of putrescine 4: >50 μM both in the presence and
absence of aminoguanidine; IC50 of 4a (mixtures of the products): 17.6 and 28.2 μM, respectively, in the absence and presence of aminoguanidine.
(b) HeLa cells treated with the diazacyclooctane hydrogel 1aa in the medium at 37 °C. The cells immediately adhered to the hydrogel and were lysed
within a few minutes. The dashed line shows the interface between the cells in the medium and those that adhered to the hydrogel. The scale bars
indicate 10 μm for the images collected at 30 s or 1 min and 100 μm for the image collected at 3 min. (c) Western blots of the cell lysates using
anti-HO-1 (upper) and anti-GAPDH (lower) antibodies after treatment with 1–4a’ in the absence of aminoguanidine. HO-1: heme oxygenase-1.
GAPDH: glyceraldehyde-3-phosphate dehydrogenase.
spermine. This reaction produced the diazacyclooctane poly- data could not be obtained, the enzymatic oxidation reaction
mers 1aa with an average molecular weight of 6000 after 5 h with polyamine mediated the polymerization reaction.
and of 15 000 after 10 h (estimated based on diffusion ordered
The cytotoxicities of the 1,5-diazacyclooctanes products
NMR spectroscopy (DOSY) methods,44 ESI Fig. S2†). Interest- 1a–4a′ and their effects on oxidative stress were evaluated in
ingly, the polymerization reaction ultimately produced an inso- cell-based assays using HeLa and A549 (human lung adeno-
luble hydrogel that contained 16 water molecules in the carcinoma epithelial) cells (Fig. 1). The putrescine-derived 4a
material (w/w) (see the pictures in Fig. 1(b)).
was obtained in a mixture with other byproducts and could
Acrolein may be generated along the polyamine metabolic not be further purified; therefore a similar caged product 4a′
pathway by amine oxidase.2–4 Polyamines are also degraded to was used in place of 4a. The compounds exhibited similar
acrolein when they are released from the cells and exposed to effects on both cell types. The data from the HeLa cell assay
serum amine oxidase. The treatment of spermine 1 with are provided in Fig. 1 (cytotoxic activities on A549 cells are pro-
serum amine oxidase4 did not produce detectable levels of the vided in ESI Table S1†). All of the diazacyclooctanes exhibited
[4 + 4] dimerization products, i.e., diazacyclooctanes, because cytotoxicity profiles with IC50 values in the micromolar range
these products were further oxidized by the enzyme (vide infra). based on the MTS method (Fig. 1(a)). These IC50 values were
Nevertheless, the enzymatic reaction ultimately provided a lower than those of acrolein (17.2 μM) and, in most cases, than
suspension of the polymers over 24 h that displayed physio- the IC50 values of the starting polyamines. All products 1a–4a′
logical properties that were very similar to those of the hydro- notably increased the cellular levels of heme oxygenase-1
gel 1aa. Although conclusive solid-phase NMR spectroscopic (HO-1), which is a marker for oxidative stress (Fig. 1(c)).45
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
On the other hand, the cytotoxic activities of the eight-membered
The diazacyclooctane hydrogel 1aa, produced through the
diazacyclooctanes 1a–3a, except for the caged compound 4a′, [4 + 4] polymerization of the bis-imine derivative of spermine
could be markedly reduced in the presence of aminoguani- 1, on the other hand, exhibited notable toxicity toward the
dine, an amine oxidase inhibitor (Fig. 1(a)).8 The data indi- cells, regardless of the presence or absence of the aminoguani-
cated that the cytotoxicities of the diazaoctanes 1a–3a were dine (Fig. 1(b)). Thus, upon introduction of the hydrogel into a
most likely derived from their oxidation by the serum amine suspension of HeLa cells in DMED medium, the cells immedi-
oxidase, which was included in the assay media.
ately adhered to the cationic hydrogel and were lysed within a
The oxidation products generated by amine oxidase were few minutes.
examined by treating one of the diazacyclooctanes, the acro-
Acrolein is the most cytotoxic unsaturated aldehyde yet
lein-modified spermidine 3a, with serum amine oxidase characterized. It readily reacts with a variety of functional groups
(Fig. 2).4 The oxidation reaction was directly monitored by common to certain biomolecules (i.e., amines, thiols, or
1H NMR (Fig. 2(a)). We clearly detected proton signals corres- hydroxyls), the modification of which is correlated with a variety
ponding to putrescine 4 over 24 h, together with other insoluble of disease states.1 Piperidine and pyridine-type derivatives, i.e.,
polymeric products, presumably the mixed diazacyclooctane 3-formyl-3,4-dehydropiperidine (FDP) and 3-methylpyridinium
and triazabicyclo[3.3.1]nonane polymers, suggesting that (MP) derivatives,7,9,10 were previously known to be the major pro-
4 molecules of the cytotoxic acrolein were generated during ducts resulting from reactions with the amino groups of, e.g.,
the oxidation process of the diazacyclooctane 3a (Fig. 2(b)).
lysine or polyamines. Some FDP and MP derivatives are used as
biomarkers. The present study clearly showed that the main acro-
lein-modified products of the polyamines were the diazacyclo-
octanes. The facile formation of eight-membered heterocycles as
the exclusive acrolein–polyamine conjugates, i.e., the formation
at μM concentrations under physiological conditions, has been
overlooked until now. This could be because (i) the 1,5-diaza-
cyclooctanes existed in equilibrium with the starting imines39 or
(ii) these heterocycles are unstable under the biological con-
ditions, i.e., in the presence of amine oxidase, which facilely oxi-
dized and metabolized these products, as demonstrated
experimentally in Fig. 2. Furthermore, (iii) these products were
transformed into a variety of unidentified compounds upon
exposure to acidic, basic, and heated conditions;38,39,43 hence,
they would have been undetectable using standard analytical
methods (chromatographic separation conditions, including
exposure to silica gel or LC-MS techniques). The FDP and MP
derivatives are, therefore, the only derivatives that have previously
been examined in depth, as these compounds are stable under
the standard analytical conditions used previously to examine
both chemical and biological samples.7,9,10 Even if the bypro-
ducts had been identified previously, the simple 1H NMR spectra
of the symmetrical structures, such as 1a–3a (see ESI Fig. S2†),
which displayed the marked absence of vinyl protons, may have
been assumed to be a polymerized byproduct due to the pre-
sumed instabilities of the alkylamine-derived unsaturated
imines. We succeeded in identifying a novel polyamine modifi-
cation by acrolein during a basic research study in which the
reactivity profiles of the unsaturated imines were explored.35–39
Polyamines, i.e., spermine and spermidine, are present in
cells at concentrations of a few mM and mostly bind to
RNA.42,46–48 Polyamines are essential for cell growth, and their
expression levels inside mammalian cells are precisely regu-
lated by biosynthesis, degradation, and transport;49 however,
Fig. 2 Oxidation of the 1,5-diazacyclooctane 3a with the serum amine
oxidase. (a) The reaction was performed in PBS-d at 37 °C, pH 7.2, using
10 μM 3a and 3.8 units bovine serum amine oxidase and was directly
monitored using 1H NMR techniques. Diazacyclooctane 3a was gradually once cells are damaged, e.g., under oxidative stress conditions,
polyamines are released from RNA and then from the cells,40
and they are oxidized by the serum amine oxidase to gen-
transformed into the putrescine 4 over 24 h. Lower signal-to-noise
ratios of the spectra in the process of the reaction are due to the pro-
duction of other insoluble polymeric products, e.g., eight-membered
polymers. (b) Possible reaction mechanism for the enzyme-catalyzed
erate acrolein. The reactive and, therefore, toxic acrolein in
turn further accelerate cell damage. Comprehensive toxicity
investigations by Igarashi and co-workers revealed that acrolein
oxidation of the diazacyclooctane 3a and the production of 4 molecules
of acrolein.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
Scheme 3 A new proposed mechanism for the generation of acrolein-mediated oxidative stress, based on the reactivity profiles of the polyamines
with acrolein and the cytotoxicity of their products. (a) 1,5-Diazacyclooctanes, e.g., 1a, were obtained smoothly by the reaction of polyamines with
acrolein, which neutralized the toxic acrolein. (b) Large quantities of acrolein, generated by amine oxidase during oxidative stress and/or being
present in cigarette and other environmental sources, produced a highly toxic diazacyclooctane hydrogel. The 2,6,9-triazabicyclo[3.3.1]nonane
structure efficiently linked the diazacyclooctane polymers to form a polymer lattice structure with the characteristic structural features of a hydro-
gel. The resulting cationic hydrogel adhered to the cell surfaces, leading to immediate cell death and further accelerating oxidative stress processes.
Both the oxidase-mediated production of acrolein and the oxidase-independent cytotoxicity of 2,6,9-triazabicyclo[3.3.1]nonane compounds were
responsible for the high toxicity.
is more toxic to cells9 than reactive oxygen species (ROS) such as Although a detailed spectroscopic analysis using solid-phase
hydrogen peroxide (H2O2) or hydroxy radical (·OH), the major NMR techniques could not precisely elucidate the structural
oxidative stress factors that led to a variety of disorders.5
constituents of the hydrogel due to the similarities between
Considering that acrolein is produced from polyamines by the proton and carbon signals of the polyamine-derivatized
amine oxidase and that its production is involved in the pro- products, the reactivity profiles presented in Scheme 1 strongly
gression of disorders, our finding of the facile production of suggest that the 2,6,9-triazabicyclo[3.3.1]nonane structure 4a
1,5-diazacyclooctanes,
modified polyamine, suggests a new mechanism for acrolein- important role in hydrogel formation (see the structure 1aa in
mediated oxidative stress (Scheme 3). Scheme 3(b)). Thus, (i) the spermine amino terminus of the
a previously unrecognized acrolein- constitutes a ridged and irreversible product and could play an
Once acrolein was produced through the oxidation of poly- developing diazacyclooctane polymers and/or (ii) the putrescine
amine or by the other sources, the diazacyclooctanes were produced during the enzyme-catalyzed oxidation of the poly-
immediately produced through a reaction with nearby poly- amines and diazacyclooctanes could spatially link the diaza-
amines (Scheme 3(a)). The cytotoxicity test revealed that the cyclooctane polymers by forming 2,6,9-triazabicyclo[3.3.1]nonane
diazacyclooctanes themselves were not toxic; hence the forma- lattice polymer structures characteristic of a hydrogel. This
tion of the eight-membered product via the newly identified enzymatic process efficiently produced the hitherto unrecog-
chemical reaction effectively neutralized the toxic acrolein. The nized hydrogel.
acrolein-neutralizing effects of FDP derivatives have previously
Finally, the resulting diazacyclooctane hydrogel, which is a
highly cationic and conformationally flexible material, strongly
been suggested.9
When large quantities of polyamines are released from dis- adhered to the negatively charged cell surfaces (Scheme 3(b))
ordered cells, polyamines as well as cyclic products are readily and led to immediate cell death. The presumed oxidase-
oxidized by serum amine oxidase to continuously produce mediated production of large quantities of acrolein near cell
acrolein (Scheme 3(b)). The excess acrolein in turn mediates adhesion regions on the diazacyclooctane polymeric materials,
a sequential [4 + 4] imino-polymerization reaction with together with the oxidase-independent cytotoxic properties of
polyamines, yielding a diazacyclooctane hydrogel. Polymer the 2,6,9-triazabicyclo[3.3.1]nonane structures present in the
formation under physiological conditions was strongly sup- partial hydrogel constituents, may account for the high cellular
ported by chemical and enzymatic reactions with spermine 1. toxicity of acrolein.
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
It should be noted that the acrolein is not only generated
3 B. M. Kimes and D. R. Morris, Biochim. Biophys. Acta, 1971,
228, 223–234.
4 G. Houen, K. Bock and A. L. Jensen, Acta Chem. Scand.,
1994, 48, 52–60.
5 K. Hensley, K. A. Robinson, S. P. Gabbita, S. Salsman
and R. A. Floyd, Free Radicals Biol. Med., 2000, 28,
1456–1462.
6 J. D. West and L. J. Marnett, Chem. Res. Toxicol., 2006, 19,
173–194.
7 K. Uchida, M. Kanematsu, Y. Morimitsu, T. Osawa,
N. Noguchi and E. Niki, J. Biol. Chem., 1998, 273,
16058–16066.
8 S. Sharmin, K. Sakata, K. Kashiwagi, S. Ueda, S. Iwasaki,
A. Shirahata and K. Igarashi, Biochem. Biophys. Res.
Commun., 2001, 282, 228–235.
extracellularly by the serum amine oxidase as discussed above,
but also found in cigarette and other environmental sources.1
The excess acrolein being produced during the combustion of
organic materials or associated with smoking could produce the
toxic diazacyclooctane hydrogel, leading to cell damage. Alter-
natively, it is known that the amine oxidase levels are also elevated
inside diseased cells under oxidative stress. In fact, increased
intracellular levels of polyamine oxidase and acrolein are con-
sistently evaluated as good markers of chronic failure and
brain stroke.29,50,51 Although we have not examined the intra-
cellular effects of the diazacyclooctane or the hydrogel, these
previously unrecognized products could pivotally damage the
cells, if formed inside the cells, based on the high reactivity
and toxicity profiles of the cyclic products reported in this paper.
9 M. Yoshida, H. Tomitori, Y. Machi, M. Hagihara,
K. Higashi, H. Goda, T. Ohya, M. Niitsu, K. Kashiwagi and
K. Igarashi, Biochem. Biophys. Res. Commun., 2009, 378,
313–318.
10 A. Furuhata, T. Ishii, S. Kumazawa, T. Yamada,
T. Nakayama and K. Uchida, J. Biol. Chem., 2003, 278,
48658–48665.
11 K. Uchida, M. Kanematsu, K. Sakai, T. Matsuda, N. Hattori,
Y. Mizuno, D. Suzuki, T. Miyata, N. Noguchi, E. Niki and
T. Osawa, Proc. Natl. Acad. Sci. U. S. A., 1998, 95, 4882–4887.
12 C. Vindis, I. Escargueil-Blanc, M. Elbaz, B. Marcheix,
M.-H. Grazide, K. Uchida, R. Salvayre and A. Negre-Salvayre,
Clic. Res., 2006, 98, 785–792.
13 C. Vindis, I. Escargueil-Blanc, K. Uchida, M. Elbaz,
R. Salvayre and A. Negre-Salvayre, Redox Rep., 2007, 12,
96–100.
Conclusions
In conclusion, we discovered that the exclusive products of the
acrolein modification of polyamines were 1,5-diazacyclo-
octanes, rather than the previously reported piperidine and
pyridine-type derivatives. The previously unrecognized eight-
membered heterocycles were rapidly produced under mild
aqueous physiological conditions. Although diazacyclooctane
formation effectively neutralized acrolein toxicity, the diaza-
cyclooctane polymers produced in the presence of excess acro-
lein, such as is present under oxidative stress conditions, were
highly toxic to cells. This study strongly suggests that diaza-
cyclooctane formation is involved in the mechanism underlying
acrolein-mediated oxidative stress.
14 B. Shao, X. Fu, T. O. McDonald, P. S. Green, K. Uchida,
K. D. O’Brien, J. F. Oram and J. W. Heinecke, J. Biol. Chem.,
2005, 280, 36386–36396.
Acknowledgements
15 N. Y. Calingasan, K. Uchida and G. E. Gibson, J. Neuro-
chem., 1999, 72, 751–756.
16 K. Hamann, A. Durkes, H. Ouyang, K. Uchida, A. Pond and
R. Shi, J. Neurochem., 2008, 107, 712–721.
We thank Dr Hiroyuki Koshino and Dr Takashi Nakamura
at RIKEN for the NMR structural analysis and Dr Takashi
Fujimoto and Prof. Tomoya Machinami at Meisei University
for assistance with the HRMS analysis. We are grateful to
Dr Kazuhiro Nishimura and Dr Kazuei Igarashi at Chiba
University for their helpful discussions, which initiated this
research. This work was supported in part by Grants-in-Aid for
Scientific Research from the Japan Society for the Promotion
of Science, 23681047 and 25560410, by a Research Grant from
the Mizutani Foundation for Glycoscience, by a MEXT Grant-
in-Aid for Scientific Research on Innovative Areas “Chemical
Biology of Natural Products: Target ID and Regulation of Bio-
activity” (24102518 and 26102743), and by a grant for Incentive
Research Projects, RIKEN. This work was also performed
according to the Russian Government Program of Competitive
Growth of Kazan Federal University.
17 K. Satoh, S. Yamada, Y. Koike, Y. Igarashi, S. Toyokuni,
T. Kumano, T. Takahata, M. Hayakari, S. Tsuchida and
K. Uchida, Anal. Biochem., 1999, 270, 323–328.
18 Y. Kawai, A. Furuhata, S. Toyokuni, Y. Aratani and
K. Uchida, J. Biol. Chem., 2003, 278, 50346–50354.
19 S. Akatsuka, T. T. Aung, K. K. Dutta, L. Jiang, Y. Liu,
W. Lee, J. Onuki, T. Shirase, Y. Naito, T. Yoshikawa,
K. Uchida and S. Toyokuni, Free Radical Res., 2006, 40,
S129–S129.
20 K. Zarkovic, K. Uchida, D. Kolenc, L. Hlupic and
N. Zarkovic, Free Radical Res., 2006, 40, 543–552.
21 C. K. Glass and J. L. Witztum, Cell, 2001, 104, 503–516.
22 D. Suzuki, T. Miyata, N. Saotome, K. Horie, R. Inagi,
Y. Yasuda, K. Uchida, Y. Izuhara, M. Yagame, H. Sakai and
K. Kurokawa, J. Am. Soc. Nephrol., 1999, 10, 822–832.
23 S. Schaefer, M. Kajimura, S. Tsuyama, K. Uchida, E. Sato,
M. Inoue, M. Suematsu and K. Watanabe, Antioxid. Redox
Signaling, 2003, 5, 457–465.
Notes and references
1 J. P. Kehrer and S. S. Biswal, Toxicol. Sci., 2000, 57, 6–15.
2 R. A. Alarcon, Arch. Biochem. Biophys., 1970, 137, 365–372.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
24 Y. Iuchi, T. Kaneko, S. Matsuki, T. Ishii, Y. Ikeda, K. Uchida 37 K. Tanaka, K. Fukase and S. Katsumura, Synlett, 2011,
and J. Fujii, Cell. Biol., 2004, 121, 123–130. 2115–2139.
25 S. Cameron-Schaefer, K. Kondo, A. Ishige, S. Tsuyama, 38 K. Tanaka, E. R. O. Siwu, S. Hirosaki, T. Iwata, R. Matsumoto,
K. Uchida, T. Hanawa, M. Suematsu and K. Watanabe,
Y. Kitagawa, A. R. Pradipta, M. Okumura and K. Fukase,
Ophthalmic Res., 2006, 38, 95–104.
Tetrahedron Lett., 2012, 53, 5899–5902.
26 P. H. Yong, H. Zong, R. J. Medina, G. A. Limb, K. Uchida, 39 K. Tanaka, R. Matsumoto, A. R. Pradipta, Y. Kitagawa,
A. W. Stitt and T. M. Curtis, Mol. Vision., 2010, 16, 2524–
2538.
27 Y. Iuchi, F. Okada, R. Takamiya, N. Kibe, S. Tsunoda,
M. Okumura and K. Fukase, Synlett, 2014, 1026–1030.
40 S. Watanabe, K. Kusama-Eguchi, H. Kobayashi and
K. Igarashi, J. Biol. Chem., 1991, 266, 20803–20809.
O. Nakajima, K. Toyoda, R. Nagae, M. Suematsu, T. Soga, 41 S. Miyamoto, K. Kashiwagi, K. Ito, S. Watanabe and
K. Uchida and J. Fujii, Biochem. J., 2009, 422, 313–320. K. Igarashi, Arch. Biochem. Biophys., 1993, 300, 63–68.
28 Y. Iuchi, N. Kibe, S. Tsunoda, S. Suzuki, T. Mikami, 42 K. Igarashi and K. Kashiwagi, Biochem. Biophys. Res.
F. Okada, K. Uchida and J. Fujii, Free Radicals Biol. Med.,
2010, 48, 935–944.
29 H. Tomitori, T. Usui, N. Saeki, S. Ueda, H. Kase,
Commun., 2000, 271, 559–564.
43 A. Tsutsui and K. Tanaka, Org. Biomol. Chem., 2013, 41,
7208–7211.
K. Nishimura, K. Kashiwagi and K. Igarashi, Stroke, 2005, 44 Y. Cohen, L. Avram and L. Frish, Angew. Chem., Int. Ed.,
36, 2609–2613. 2005, 44, 520–554.
30 N. Tanaka, S. Tajima, A. Ishibashi, K. Uchida and 45 S. Kitazume, R. Oka, K. Ogawa, S. Futakawa, Y. Hagiwara,
T. Shigematsu, Arch. Dermatol. Res., 2001, 293, 363–367.
H. Takikawa, M. Kato, A. Kasahara, E. Miyoshi, N. Taniguchi
31 N. Shibata, A. Hirano, E. T. Hedley-Whyte, M. C. Dal Canto,
and Y. Hashimoto, Glycobiology, 2009, 19, 479–487.
R. Nagai, K. Uchida, S. Horiuchi, M. Kawaguchi, T. Yamamoto 46 S. S. Cohen, A guide to the polyamines, Oxford University
and M. Kobayashi, Acta. Neuropathol., 2002, 104, 171–178. Press, 1998.
32 E. Noiri, S. Yamada, A. Nakao, M. Tsuchiya, I. Masaki, 47 C. W. Tabor and H. Tabor, Annu. Rev. Biochem., 1984, 53,
K. Fujino, K. Nosaka, T. Ozawa, T. Fujita and K. Uchida,
Free Radicals Biol. Med., 2003, 33, 1651–1656.
33 A. Kudo, S. Kashiwagi, M. Kajimura, Y. Yoshimura,
749–790.
48 K. Igarashi and K. Kashiwagi, J. Biochem., 2006, 139,
11–16.
K. Uchida, S. Arii and M. Suematsu, Hepatology, 2004, 39, 49 T. Ogasawara, K. Ito and K. Igarashi, J. Biochem., 1989, 105,
1099–1109. 164–167.
34 J. Luo, K. Uchida and R. Shi, Neurochem. Res., 2005, 30, 50 K. Sakata, K. Kashiwagi, S. Sharmin, S. Ueda, Y. Irie,
1099–1109.
N. Murotani and K. Igarashi, Biochem. Biophys. Res.
Commun., 2003, 305, 143–149.
51 M. Yoshida, H. Tomitori, Y. Machi, D. Katagiri, S. Ueda,
K. Horiguchi, E. Kobayashi, N. Saeki, K. Nishimura, I. Ishii,
K. Kashiwagi and K. Igarashi, Atherosclerosis, 2009, 19,
479–487.
35 K. Tanaka and S. Katsumura, J. Am. Chem. Soc., 2002, 124,
9660–9661.
36 K. Tanaka, T. Masuyama, K. Hasegawa, T. Tahara, H. Mizuma,
Y. Wada, Y. Watanabe and K. Fukase, Angew. Chem., Int.
Ed., 2008, 47, 102–105.
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem.