Novel bisbenzimide-nitroxides for nuclear redox imaging in living cells

Article abstract of DOI:10.1016/j.bmcl.2012.01.042

Nuclear oxidative stress damages genomic DNA and may lead to cell death, leading to aging and aging-related disorders. Though it is important to measure the nuclear oxidative stress separately, there are still little examples that applicable to living cells. We have designed and synthesized three bisbenzimide-nitroxides as probes to selectively visualize nuclear redox changes in terms of fluorescence. Compound 3, containing two radical moieties, showed the largest reduction-induced fluorescence change, with good localization in nuclei. RAW264.7 murine macrophage cells were loaded with compound 3 and then treated with 100 μM hydrogen peroxide for 5 min to show the fluorescence increase. This fluorescence increase was inhibited by pretreatment of 1 mM ascorbic acid. These results show that compound 3 was suitable for nuclear-specific redox imaging in murine macrophages.

Full text of DOI:10.1016/j.bmcl.2012.01.042

Bioorganic & Medicinal Chemistry Letters 22 (2012) 1949–1952
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Novel bisbenzimide-nitroxides for nuclear redox imaging in living cells
⇑
⇑
Mamiko Ikeda, Hidehiko Nakagawa , Takayoshi Suzuki, Naoki Miyata
Graduate School of Pharmaceutical Sciences, Nagoya City University, Tanabe-dori 3-1, Mizuho-ku, Nagoya, Aichi 467-8603, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Nuclear oxidative stress damages genomic DNA and may lead to cell death, leading to aging and aging-
related disorders. Though it is important to measure the nuclear oxidative stress separately, there are still
little examples that applicable to living cells. We have designed and synthesized three bisbenzimide-nitr-
oxides as probes to selectively visualize nuclear redox changes in terms of fluorescence. Compound 3,
containing two radical moieties, showed the largest reduction-induced fluorescence change, with good
localization in nuclei. RAW264.7 murine macrophage cells were loaded with compound 3 and then treated
with 100 lM hydrogen peroxide for 5 min to show the fluorescence increase. This fluorescence increase
was inhibited by pretreatment of 1 mM ascorbic acid. These results show that compound 3 was suitable
for nuclear-specific redox imaging in murine macrophages.
Received 5 December 2011
Revised 11 January 2012
Accepted 13 January 2012
Available online 24 January 2012
Keywords:
Fluorophore-nitroxide
Nuclei
Redox
ESR
Ó 2012 Elsevier Ltd. All rights reserved.
Fluorescence
Oxidative stress is usually caused by overproduction or accumu-
lation of reactive oxygen species (ROS). Although ROS plays a vari-
ety of beneficial roles in biological processes, including signaling
and immune response,1–4 an inappropriate concentration or loca-
tion of ROS may result in oxidative modification of biomolecules
such as membrane components,5 proteins6 and DNA.7 Accumula-
tion of such oxidative damage is considered to be associated with
aging8 and aging-related disorders, including carcinogenesis,9,10
ischemia11 and neurological disorders.12–15 Therefore, determina-
tion of subcellular oxidative stress levels has potential value for
the diagnosis and treatment of such disorders.
In particular, oxidative stress in nuclei is critical, because it may
induce DNA damage.16 The relationship between genomic DNA
damage and ROS-mediated disorders has usually been studied by
extracting and quantitating oxidatively modified nucleic acids
from cells, serum or urine.17–19 However, oxidized nucleic acids
are also produced from the cytoplasmic nucleotide pool,20,21 so it
is difficult to selectively estimate nuclear oxidative stress in the
living body.
or nuclei.27 These probes are nitroxide derivatives bearing a local-
izing moiety and a fluorescent tag allows visualization of each
probe’s distribution in cells by microscopic imaging. Those ESR
probes can be used to estimate the organelle-specific redox state
in living cells by observing the ESR signal decay rate. As a next step,
we aimed to use a similar strategy to develop fluorophore-nitrox-
ide probes.
Fluorophores lose their fluorescence upon interaction with an
adjacent radical species28 via electron transfer from the excited
state of the fluorophore to the radical moiety. On the other hand,
nitroxide is gradually reduced by reductants such as ascorbic acid
in biological systems, forming hydroxylamine, and this results in
loss of the fluorescence-quenching effect of the radical.
Based on these phenomena, we designed novel bisbenzimide-
nitroxide derivatives 1–3 (Fig. 1) as candidate nuclear-localizing
redox probes. Bisbenzimide is the fundamental structure of the
widely used nuclear stain Hoechst 33342, which has both DNA-
binding ability and blue fluorescence.29
Syntheses of these compounds are summarized in Schemes 1
and 2. Compound 1 was prepared by repeated cyclization of alde-
hyde with phenylenediamine derivatives, followed by coupling
with 4-aminoTEMPO. Compounds 2 and 3 were prepared by
heterocyclization of the common intermediate 13 with each
phenylenediamine.
Recently, several groups have developed functionalized nucle-
ar-localizing molecules22–24 by utilizing a nuclear localization
signal peptide or genetic engineering, but there are still few exam-
ples of theoretically designed small molecular organelle-specific
probes.
We have recently reported organelle-specific electron spin res-
As these bisbenzimide-nitroxides contain a radical moiety, we
were not able to obtain 1H or 13C NMR spectra due to shortening
of relaxation time. Therefore, nitroxide-containing compounds
were characterized by NMR after addition of phenylhydrazine as
a reductant.
We evaluated the spectral characteristics and responses of
10 lM bisbenzimide-nitroxides to nitroxide reduction by 100 lM
onance (ESR) probes localizing to membrane25 or mitochondria26
⇑
Corresponding authors. Tel./fax: +81 052 836 3408 (H.N.); tel./fax: +81 052 836
3407 (N.M.).
E-mail addresses: deco@phar.nagoya-cu.ac.jp (H. Nakagawa), miyata-n@phar.
nagoya-cu.ac.jp (N. Miyata).
0960-894X/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2012.01.042

1950
M. Ikeda et al. / Bioorg. Med. Chem. Lett. 22 (2012) 1949–1952
Figure 2. Fluorescence and ESR responses of 10
lM bisbenzimide-nitroxides to
100 M ascorbic acid (AsA) in DMSO/D-PBS (9/1) solution. Values are presented as
l
the means SD of three experiments.
Subsequent reduction induced a fluorescence recovery of 180–
490% (Fig. 2).
The double-nitroxide compound 3 showed the largest fluores-
cence recovery, despite having the smallest reduction range in
ESR. It is possible that the multiple nitroxide structures work syn-
ergistically to induce a stronger fluorescence-quenching effect.
The nuclear stain Hoechst 33342 shows an increase of fluores-
cence on binding to DNA. Thus, our bisbenzimide-nitroxides,
which are based on the same structure, may also exhibit changes
of fluorescence intensity in response to factors other than reduc-
tion of nitroxide. Therefore, we investigated the influence of DNA
on bisbenzimide-nitroxide fluorescence.
Figure 1. Structures of novel bisbenzimide-nitroxides 1, 2 and 3.
ascorbic acid in DMSO/D-PBS (9/1) solution. The fluorescence char-
acteristics of these compounds were similar to those of Hoechst
33342 (kex = 352 nm, kem = 451 nm; 1: kex = 350 nm, kem = 378 nm;
2: kex = 320 nm, kem = 390 nm; 3: kex = 350 nm, kem = 425 nm). Dis-
appearance of the radical was confirmed by ESR observation. The
ESR signal intensity of these compounds decreased with time,
becoming constant within 15 min. The intensity was decreased
A buffered solution of 3 (10
taining 0.1% DMSO) was subjected to fluorescence spectroscopy
in presence of 0–1000 M ascorbic acid or 1 g/mL DNA (Fig. 3).
lM in TNE-buffered solution con-
to 6–14% of the initial level in presence of 100 lM ascorbic acid.
l
l
Scheme 1. Synthesis of 1. Reagents and conditions: (a) 2-(Boc-amino)ethanol, DEAD, PPh3, THF, rt, 12 h, 52%; (b) 3,4-diaminobenzoic acid N,O-dimethylhydroxamide,
Na2S2O5, EtOH, reflux, 2 h, 95%; (c) DIBAL, THF, À40 to 0 °C, 69%; (d) 3,4-diaminobenzoic acid, Na2S2O5, EtOH, reflux, 12 h, 99%; (e) 4-amino-TEMPO, EDCI, HOBt, DMF, rt, 30 h,
64%.
Scheme 2. Syntheses of 2 and 3. Reagents and conditions: (a) oxalyl chloride, cat. DMF, CH2Cl2, rt, 1 h, then N,O-dimethylhydroxylamine, Et3N, CH2Cl2, rt, 30 min, 98%; (b) H2,
5% Pd/C, MeOH, rt, 40 h; (c) terephthalaldehydic acid, Na2S2O5, EtOH, reflux, 6 h, 81%; (d) 4-amino-TEMPO, EDCI, HOBt, DMF, rt, 5 h, 95%; (e) LiAlH4, THF, À78 °C to rt, 19 h,
54%; (f) 4-(4-methylpiperazin-1-yl)-1,2-phenylenediamine, EtOH, reflux, 38 h, 18%; (g) 4-(3,4-diaminobenzoylamino)-TEMPO, EtOH, reflux, 6.5 h, 48%.

M. Ikeda et al. / Bioorg. Med. Chem. Lett. 22 (2012) 1949–1952
1951
created by merging of images of the same area in different focal
planes. The cells may not move only horizontally but also change
its shape vertically during the observation. To avoid the fluores-
cence changes due to such vertical movement of the cells, we used
total fluorescence intensity from the stacked images of different fo-
cus planes.
Cell nuclei were counterstained with 5 lM SYTO Green 11
Nucleic Acid Stain. Compound 3 was reduced by intracellular
ascorbic acid, resulting in an increase of the fluorescence level,
which allowed distribution of the dye to be observed. Blue fluores-
cence of compound 3 was colocalized with green fluorescence of
SYTO Green 11 (Fig. 4). It was shown from the results in Figure 3
that the fluorescence of compound 3 did not affected by DNA. This
means that the compound 3 shows DNA-independent fluorescence
although Hoechst 33342 shows DNA-dependent fluorescence. We
only observed nuclear fluorescence staining with compound 3,
meaning that the compound 3 is distributed predominantly in nu-
clei in Figure 4.
Figure 3. Fluorescence responses of 10
lM compound 3 to various concentrations
of ascorbic acid or 1 g/mL DNA in TNE-buffered solution. Values are presented as
l
the means SD of three experiments.
The fluorescence intensity of 3 showed a concentration-dependent
To observe the fluorescence change in living cells, the cells were
increase of up to 2.8-fold in the range of 0–500 lM ascorbic acid,
treated with 50
by treatment with 100
l
M compound 3 for 3 h, then were exposed to ROS
M hydrogen peroxide (H2O2). Blue fluores-
while the fluorescence change induced by DNA was negligible,
indicating that existence of DNA does not contribute to the fluores-
cence increase of 3.
Next, we examined the suitability of compound 3 for nuclear-
specific redox imaging in living cells. Cells of the RAW264.7 murine
l
cence of compound 3 was increased after 5 min exposure to H2O2
(Fig. 5a) and this increase was suppressed by pretreatment with
1 mM ascorbic acid (Fig. 5b). The frequency–brightness relation-
ship shows that the number of bright cells increased with a
corresponding decrease of dark cells (Fig. 5c). The median bright-
ness of H2O2 treated cells were raised from 76.7 to 106.6. On the
macrophage cell line were treated with 50
lM compound 3 for
24 h, and then observed by confocal microscopy. All images were
Figure 4. Distribution of 50 lM compound 3 in RAW 264.7 cells (left). The cell nuclei were stained with 5 lM SYTO Green 11 Nucleic Acid Stain (center).
Figure 5. (a) Fluorescence changes of 50 lM compound 3 before (left) and after (right) addition of 100 lM H2O2 in RAW 264.7 cells. (b) Same as (a) except for pretreatment
with 1 mM AsA. (c) Relationship between fluorescence brightness of the cell and frequency. (d) Same as (c) except for pretreatment with 1 mM AsA. The median brightness
was shown as vertical solid/dashed line in (c) and (d). Frequency was calculated from cell number in arbitrary areas of three dishes.

1952
M. Ikeda et al. / Bioorg. Med. Chem. Lett. 22 (2012) 1949–1952
other hand, ascorbic acid pretreated cells did not show significant
fluorescence increase (Fig. 5b). The difference of frequency-bright-
ness relationship before and after addition of H2O2 was not notice-
able, showing the slight decrease of the median brightness from
98.2 to 85.6 by addition of H2O2.
References and notes
1. Lander, H. M. FASEB J. 1997, 11, 118.
2. Finkel, T. J. Leukoc. Biol. 1999, 65, 337.
3. Griendling, K. K.; Harrison, D. G. Circ. Res. 1999, 85, 562.
4. Sauer, H.; Wartenberg, M.; Hescheler, J. Cell. Physiol. Biochem. 2001, 11, 173.
5. Poli, G.; Biasi, F.; Chiarpotto, E.; Dianzani, M. U.; De Luca, A. Free Radic. Biol. Med.
1989, 6, 167.
6. Berlett, B. S.; Stadtman, E. R. J. Biol. Chem. 1997, 272, 20313.
7. Meneghini, R. Free Radic. Biol. Med. 1997, 23, 783.
8. Best, B. P. Rejuvenation Res. 2009, 12, 199.
9. Farinati, F.; Cardin, R.; Degan, P.; Rugge, M.; Di Mario, F.; Bonvicini, P.;
Naccarato, R. Gut 1998, 42, 351.
10. Shinmura, K.; Tao, H.; Goto, M.; Igarashi, H.; Taniguchi, T.; Maekawa, M.;
Takezaki, T.; Sugimura, H. Carcinogenesis 2004, 25, 2311.
11. Flamm, E. S.; Demopoulos, H. B.; Seligman, M. L.; Poser, R. G.; Ransohoff, J.
Stroke 1978, 9, 445.
H2O2 is known to generate hydroxyl radical by Fenton reaction
with biological Fe(II) ion.30 Nitroxide can be oxidized by hydroxyl
radical, forming oxoammonium.31 This oxoammonium form is eas-
ily reduced by two-electron reductants. Compound 3 was consid-
ered to have been oxidized by hydroxyl radical generated from
H2O2, and subsequent two-electron reduction results in loss of
the paramagnetic property. In the case of pretreatment with ascor-
bic acid, H2O2 or other ROS are trapped by the excess amount of
ascorbic acid, so that the redox level does not change. Even though
TEMPO can also lose its paramagnetic property by the reduction
with cellular reductant such as ascorbic acid directly, TEMPO is rel-
atively rapidly reduced by such reductants and reached its redox
equilibrium status. In the experiments, the ESR signal intensity of
compound 3 was found to become stable before the oxidant treat-
ment. It was considered that the effect of the initial reduction by
intracellular ascorbic acid and other reductants was canceled in
our experimental conditions.
12. Gabbita, S. P.; Lovell, M. A.; Markesbery, W. R. J. Neurochem. 1998, 71, 2034.
13. Wang, J.; Xiong, S.; Xie, C.; Markesbery, W. R.; Lovell, M. A. J. Neurochem. 2005,
93, 953.
14. Smith, M. A.; Harris, P. L.; Sayre, L. M.; Perry, G. Proc. Natl. Acad. Sci. U.S.A. 1997,
94, 9866.
15. Evans, P. H. Br. Med. Bull. 1993, 49, 577.
16. Gensler, H. L.; Bernstein, H. Q Rev. Biol. 1981, 56, 279.
17. Kasai, H.; Crain, P. F.; Kuchino, Y.; Nishimura, S.; Ootsuyama, A.; Tanooka, H.
Carcinogenesis 1986, 7, 1849.
18. Saito, S.; Yamauchi, H.; Hasui, Y.; Kurashige, J.; Ochi, H.; Yoshida, K. Res.
Commun. Mol. Pathol. Pharmacol. 2000, 107, 39.
19. Herbert, K. E.; Evans, M. D.; Finnegan, M. T.; Farooq, S.; Mistry, N.; Podmore, I.
D.; Farmer, P.; Lunec, J. Free Radic. Biol. Med. 1996, 20, 467.
20. Haghdoost, S.; Czene, S.; Näslund, I.; Skog, S.; Harms-Ringdahl, M. Free Radic.
Res. 2005, 39, 153.
21. Cooke, M. S.; Evans, M. D.; Dizdaroglu, M.; Lunec, J. FASEB J. 2003, 17, 1195.
22. Srikun, D.; Albers, A. E.; Nam, C. I.; Iavarone, A. T.; Chang, C. J. J. Am. Chem. Soc.
2010, 132, 4455.
In conclusion, compound 3 provided the largest fluorescence
recovery among the three novel bisbenzimide-nitroxides we syn-
thesized. This fluorescence increase was ascorbic acid concentra-
tion-dependent and was not affected by the presence of DNA.
Compound 3 was localized in cell nuclei and showed s substantial
23. Dickinson, B. C.; Tang, Y.; Chang, Z.; Chang, C. J. Chem. Biol. 2011, 18, 943.
24. Halvey, P. J.; Hansen, J. M.; Johnson, J. M.; Go, Y.-M.; Samali, A.; Jones, D. P.
Antioxid. Redox Signal. 2002, 9, 807.
25. Ban, S.; Nakagawa, H.; Suzuki, T.; Miyata, N. Bioorg. Med. Chem. Lett. 2007, 17,
1451.
fluorescence increase upon exposure to 100 lM H2O2. Our results
indicate that compound 3 can be used as a probe to visualize the
redox state in nuclei of living cells.
26. Ban, S.; Nakagawa, H.; Suzuki, T.; Miyata, N. Bioorg. Med. Chem. Lett. 2007, 17,
2055.
Acknowledgments
27. Ikeda, M.; Nakagawa, H.; Ban, S.; Tsumoto, H.; Suzuki, T.; Miyata, N. Free Radic.
Biol. Med. 2010, 49, 1792.
28. Bystryak, I. M.; Likhtenshtein, G. I.; Kotel’nikov, A. I.; Hankovsky, O. H.; Hideg,
K. Russ. J. Phys. Chem. 1986, 60, 1679.
29. Bazhulina, N. P.; Nikitin, A. M.; Rodin, S. A.; Surovaya, A. N.; Kravatsky, Y. V.;
Pismensky, V. F.; Archipova, V. S.; Martin, R.; Gursky, G. V. J. Biomol. Struct. Dyn.
2009, 26, 701.
30. Imalay, J. A.; Chin, S. M.; Linn, S. Science 1988, 240, 640.
31. Samuni, A.; Goldstein, S.; Russo, A.; Mitchell, J. B.; Krishna, M. C.; Neta, P. J. Am.
Chem. Soc. 2002, 124, 8719.
This work was supported by Grant-in-Aid for JSPS Fellows
(M.I.), JST PRESTO program (H.N.), Grants-in-Aid for Scientific Re-
search on Innovative Areas (Research in a Proposed Research Area)
(No. 21117514 to H.N.), and Grants-in-Aid for Scientific Research
(No. 22590103 to H.N.) from Ministry of Education, Culture, Sports
Science and Technology Japan.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2012.01.042.