A novel hydrazino-substituted naphthalimide-based fluorogenic probe for tert-butoxy radicals

Article abstract of DOI:10.1039/c3cc42052c

A novel fluorogenic probe for tert-butoxy radicals based on the hydrazino-naphthalimide system is reported. Interestingly, different regioisomers exhibited significantly different optical properties toward ROS, which suggested 4-hydrazinyl naphthalimide as a potential new platform for the in vitro and in cellulo detection of alkoxyl radicals.

Full text of DOI:10.1039/c3cc42052c

ChemComm
View Article Online
View Journal
COMMUNICATION
A novel hydrazino-substituted naphthalimide-based
fluorogenic probe for tert-butoxy radicals†
Cite this: DOI: 10.1039/c3cc42052c
a
a
a
a
a
a
Zhao Ma, Wei Sun, Laizhong Chen, Jing Li, Zhenzhen Liu, Haixiu Bai,
Received 21st March 2013,
Accepted 23rd May 2013
a
a
b
a
Mengyuan Zhu, Lupei Du, Xiaodong Shi and Minyong Li*
DOI: 10.1039/c3cc42052c
www.rsc.org/chemcomm
A novel fluorogenic probe for tert-butoxy radicals based on
the hydrazino-naphthalimide system is reported. Interestingly,
different regioisomers exhibited significantly different optical
properties toward ROS, which suggested 4-hydrazinyl naphthalimide
as a potential new platform for the in vitro and in cellulo detection of
alkoxyl radicals.
2 2
Reactive oxygen species (ROS), including peroxides (H O ), hydroxyl
ꢀ
ꢀ
ꢀ
radicals (HO ), alkoxyl radicals (RO ), peroxyl radicals (ROO ),
1
ꢀ
ꢁ
Scheme 1 The structures of 1a, 2a, probe 1 and probe 2.
singlet oxygen ( O ), and superoxide radicals (O ), play important
2
2
roles in a variety of biological phenomena, such as aging, cancer and
1–3
many other disease processes. Doctors and scientists have paid
much attention to this area and various effective fluorescent probes
have been reported for selective detection of specific ROS species.
According to literature reports, the probes reported so far are mainly
It is known that the amino-substituted 1,8-naphthalimide can
emit strong fluorescence through a typical ICT (Intramolecular
4,5
Charge Transfer) process in the presence of the electron-donating
14–16
amino group (1a and 2a in Scheme 1).
Replacing the amino
ꢀ ꢀ
ꢁ
1
ꢁ
4,6–8
directed towards H
Successful examples of probes for alkoxyl radicals are rare.
Alkoxyl radicals are usually generated from the reaction of lipid
2
O
2
, O
2
, HO , O , ONOO and so on.
2
group with hydrazine caused the fluorescence intensity to decline
due to the decrease of the electron-donating ability of the sub-
stituent group (Fig. S1 and S2, ESI†). Meanwhile, hydrazine and its
9–11
peroxidation (LPO) in vivo.
Nevertheless, due to their high
17
derivatives can rapidly react with oxidants, giving the N–N bond
reactivity, they could lead to sequential peroxidation of lipids and
fatty acid decomposition, which dissociate the membrane to kill
18–20
cleavage products (amine).
Considering the weak electron-
donating ability and strong redox ability of hydrazine, we postulated
that hydrazine modified naphthalimide could be used as a potential
fluorogenic switch for ROS detection and two probes were designed
as shown in Scheme 1.
12
the cells. RO radicals are also predominantly related to the
protein damage by degradation, fragmentation, peroxidation,
12,13
modification, and inactivation.
Indeed, alkoxyl radicals are
devastating to the cell, tissue, and even the organism. However,
few methods have been reported that specifically target the alkoxyl
radical detection. Thus, it is highly desired to develop fluorescent
probes for selectively detecting alkoxyl radicals in vitro or in vivo.
With strong interest, we focused on the tert-butoxy radical,
which is a rather stable and accessible alkoxyl radical. Herein,
we report the hydrazino containing naphthalimide chromo-
The syntheses of probes 1 and 2 are shown in Scheme 2, and the
end products were stored as dihydrochloride salts to avoid oxidation
in the air. The N,N-dimethylamino group was introduced into both
ꢀ
phore as an effective probe against tBuO with high efficiency
and excellent selectivity.
a
Department of Medicinal Chemistry, Key Laboratory of Chemical Biology (MOE),
School of Pharmacy, Shandong University, Jinan, Shandong 250012, China.
E-mail: mli@sdu.edu.cn; Fax: +86-531-8838-2076; Tel: +86-531-8838-2076
b
C. Eugene Bennett Department of Chemistry, West Virginia University,
Morgantown, WV 26506, USA
†
Electronic supplementary information (ESI) available. See DOI: 10.1039/c3cc42052c
Scheme 2 The syntheses of probe 1 and probe 2.
This journal is c The Royal Society of Chemistry 2013
Chem. Commun.

View Article Online
Communication
ChemComm
Fig. 1 Time course of the interaction of two probes (50 mM) with tert-butoxy
radicals (100.0 mM) in PBS (pH 6.0). Probe 1: lex = 440 nm, lem = 550 nm; probe 2:
l
ex = 405 nm, lem = 590 nm.
compounds to improve the lipid–water partition coefficient. These
two compounds were first evaluated in a fluorescence study (Fig. S1
and S2, ESI†) under the mild acidic conditions (pH 6.0) to reduce the
auto-oxidation of arylhydrazines. As expected, although the absor-
bance spectra of arylhydrazines are similar to corresponding amides
with the absorption band centered at the same wavelength, the
emission of probe 1 is very weak while probe 2 basically gave no
fluorescent emission, which is in stark contrast to the high fluores-
cence intensity of 1a (lem = 550 nm) and 2a (lem = 590 nm). Thus, the
substitution of amino with hydrazino has no measurable impact on
the absorption spectra, but quenches the fluorescence greatly. After
Scheme 3 Selective N–N and C–N bond dissociation.
ꢀ
2+
addition of tBuO generated from the Fenton reagent (Fe :TBHP,
21,22
mol/mol = 10 : 1),
probe 1 exhibited a significant increase in
fluorescence, but probe 2 exhibited no fluorescence (Fig. 1).
The acquired fluorescence intensity–time curve indicated a
significant difference between these two probes. Initially, probes 1
Fig. 2 Fluorescence response of probe 1 (50 mM) in 50 mM PBS (pH = 6.0) to
various analytes at 550 nm (lex = 440 nm). This figure is shown for 50 mM of ROS,
2+
2+
3+
RNS and RSS; 0.5 mM of Fe , Fe (0.1% DMSO) and Fe . Data were obtained
and 2 themselves are almost non-fluorescent. The fluorescence after incubation with the appropriate analytes at room temperature for 0.5 min,
intensity of probe 1 increased sharply at short time after the addition 10.5 min and 20.5 min.
ꢀ
of tBuO , and reached a maximum value (more than 60-fold) at
about 900 s. However, the fluorescent intensity of probe 2 did not
increase at all throughout the reaction.
fluorescence changes of probe 1 at 0.5, 10.5, 20.5 min after the
To verify the mechanism we examined the reactions of probe 1 addition of various species containing ROS (1 equiv.) and some
ꢀ
and probe 2 with tert-butoxy radicals, respectively (Scheme S2, ESI†). inorganic ions (10 equiv.). Interestingly, introduction of t-BuO
Interestingly, compound 1a was isolated from the reaction of probe urges probe 1 to emit higher fluorescence more quickly than
ꢀ
1
, while the corresponding compound 2a was not observed in the others. The response to tBuO increases 30-fold compared to
ꢀ
reaction of probe 2. These results strongly suggested that the N–N hydrogen peroxide, alkyl hydroperoxides, ROO , superoxide,
ꢁ
ꢁ
ꢁ
3+
bond cleavage in probe 1 (formation of NH
2
) is much faster than peracetic acid, Cys, PhSH, NO, NO
3
, ONOO , NO
2
and Fe
that in probe 2 under the reaction conditions. Thus, effective (control experiments). The emission from the tert-butoxy radi-
fluorescence emission was observed, which allowed probe 1 to serve cal is about 5–10-fold compared to hydroxyl, alkoxyl and methyl
2
+
as a novel turn-on sensor.
radicals and Fe , and about 10-fold compared to hypochlorite
ꢀ
ꢁ
During the reaction, tBuO will likely capture an electron from (ClO ). Specifically, probe 1 shows obvious response (Fig. S6,
the hydrazine group, which will lead to either N–N bond dissociation ESI†) to the tBuO radical generated from the pyrolysis of
or C–N bond dissociation. For probe 1, the stable intermediate tBuONO at 115 1C, which is a good source for obtaining tBuO
ꢀ
2
3
radical favored the N–N bond homolysis, which will lead to the due to the stablizing function of the NO species. In addition,
formation of fluorescence active compound 1a. A similar resonance as a known scavenger of radicals, TEMPO could quench the
ꢀ
does not exist for the N-radical produced from probe 2. Thus, the definite fluorescence of the mixture of probe 1 and tBuO
C–N bond dissociation occurred, giving the corresponding carbon without reacting with probe 1 or 1a directly (Fig. S7, ESI†),
radical through the release of N gas as shown in Scheme 3. As the which strongly confirms the radical process. Thus, probe 1 has
2
ꢀ
result, fluorescence inactive compound 2b is formed, instead of the huge potential to detect tBuO , which suggests a promising
desired compound 2a (not observed at all). These results provided a future for the development of alkoxyl radical sensors.
good example of subtle structural changes that might account for a
significant reactivity difference.
The response capability of probe 1 to different concentrations of
tert-butoxy radicals via fluorescence changes was also investigated.
ꢀ
Various analytes were then charged with probe 1 to evaluate As shown in Fig. 3, upon addition of tBuO (from 0 to 320 mM), the
the efficiency and selectivity of this new probe. Fig. 2 shows the fluorescence intensity increases more than 10-fold. An excellent
Chem. Commun.
This journal is c The Royal Society of Chemistry 2013

View Article Online
ChemComm
Communication
treated with tert-butoxy radicals, a medium fluorescence was
recorded as shown in Fig. 4h which was stronger than that in
Fig. 4b and weaker than that in Fig. 4e. The fluorescence imaging of
ꢀ
Fenton reagent induced tBuO production in live PC-3 cells gave
24
the same results (Fig. S9, ESI†). These results demonstrated that
ꢀ
probe 1 could exhibit fluorescence response to tBuO in living cells.
Significantly, to the best of our knowledge, this is the first report on
the detection of alkoxyl radicals in living cells.
In conclusion, based on the hybrid hydrazino-naphthalimide
Fig. 3 Concentration-dependent emission intensity changes of probe1 at room platform, a new effective fluorescent probe for tert-butoxy radicals
temperature in 50 mM PBS, pH = 6.0. The emission spectra were obtained at
was developed. Upon the incorporation of the hydrazino functional
1
5 min after the addition of tert-butoxy radicals (0–320 mM) to 50 mM probe 1
group, this new fluorescence probe was able to produce 4-amino-
naphthalimide as well as a significant fluorescence increase. In
addition, the fluorescent probe 1 was tested in conventional aspects
and fluorescence imaging of tBuO radicals in living cells. Specific
and rapid fluorescent responses were obtained in vitro. As a result,
the first fluorescent probe for tert-butoxy radicals, probe 1, was
identified. It is our expectation that, based on the reported strategy,
more hydrazine derivatives linked to fluorophores will be developed
to become new fluorogenic probes and further effort should be
made to develop an excellent probe for other alkoxyl radicals.
The present work was supported by the Program for New
Century Excellent Talents in University (No. NCET-11-0306), the
Shandong Natural Science Foundation (No. JQ201019) and the
Independent Innovation Foundation of Shandong University,
IIFSDU (No. 2010JQ005).
with lex = 440 nm. Inset: fluorescence intensity at 550 nm as a function of
concentration of tert-butoxy radicals.
Fig. 4 Fluorescence imaging of the heated tBuONO induced tBuO production in Notes and references
live PC-3 cells. (a) Bright-field image of cells incubated with only probe 1 (10 mM)
1
2
3
J. S. Valentine and E. B. Gralla, Proc. Natl. Acad. Sci. U. S. A., 2008,
05, 8178.
J. T. Landrum, in Carotenoids and Human Health, ed. S. A. Tanumihardjo,
Humana Press, 2013, pp. 57–101.
for 2 min; (b) fluorescence image of (a) with green light excitation; (c) overlap of
a) and (b); (d) bright-field image of cells preprocessed using tBuONO (3.8 mM)
1
(
for 5 min and then incubated with probe 1 (10 mM) for 2 min; (e) fluorescence
image of (d) with green light excitation; (f) overlap of (d) and (e); (g) bright-field
image of cells incubated with tBuONO (3.8 mM) for 5 min first, and then with
S. Wang, N. Li, W. Pan and B. Tang, Trends Anal. Chem., 2012, 39, 3–37.
4 X. Chen, X. Tian, I. Shin and J. Yoon, Chem. Soc. Rev., 2011, 40,
4783–4804.
ꢁ1
catalase (1 mg mL ) for 5 min, and finally with probe 1 (10 mM) for 2 min; (h)
fluorescence image of (g) with green light excitation; (i) overlap of (g) and (h).
5 T. Nagano, Proc. Jpn. Acad., Ser. B, 2010, 86, 837–847.
6
7
Z. Chen, Q. Li, Q. Sun, H. Chen, X. Wang, N. Li, M. Yin, Y. Xie, H. Li
and B. Tang, Anal. Chem., 2012, 84, 4687–4694.
K. Xu, L. Wang, M. Qiang, L. Wang, P. Li and B. Tang, Chem.
Commun., 2011, 47, 7386–7388.
ꢀ
linear correlation between FI and [tBuO ] was obtained with
r > 0.99. This result demonstrates that the fluorescence intensity
of probe 1 increases as a linear function of tBuO concentration.
Furthermore, the kinetic data for the reactions of probe 1 with tert-
8 K. Xu, H. Chen, J. Tian, B. Ding, Y. Xie, M. Qiang and B. Tang, Chem.
Commun., 2011, 47, 9468–9470.
ꢀ
9
O. Panasenko, A. Chekanov, J. Arnhold, V. Sergienko, A. Osipov and
Y. A. Vladimirov, Biochemistry, 2005, 70, 998–1004.
butoxy radicals were also collected (Fig. S8, ESI†). The apparent rate 10 J. M. McCord, Nutr. Rev., 1996, 54, 85–88.
1
1
1
1
1 B. Halliwell, FASEB J., 1987, 1, 358–364.
constant for the reaction is given based on the pseudo-first-order
equation (eqn (1)):
2 R. Kohen and A. Nyska, Toxicol. Pathol., 2002, 30, 620–650.
3 M. Logani and R. Davies, Lipids, 1980, 15, 485–495.
4 B. Zhu, X. Zhang, Y. Li, P. Wang, H. Zhang and X. Zhuang, Chem.
Commun., 2010, 46, 5710–5712.
ln(F /Fmax) = ꢁk0t
(1)
t
1
5 W. Sun, W. Li, J. Li, J. Zhang, L. Du and M. Li, Tetrahedron, 2012, 68,
where F
t
and Fmax are the fluorescence intensities at 550 nm at
5363–5367.
time t and the maximum value obtained after the reaction was 16 J. Jiang, W. Liu, J. Cheng, L. Yang, H. Jiang and D. Bai, Chem.
0
Commun., 2012, 48, 8371–8373.
complete. The second-order rate constant k calculated from k
1
1
7 F. D. Chattaway, J. Chem. Soc., Trans., 1908, 93, 270–280.
8 S. E. Denmark, O. Nicaise and J. P. Edwards, J. Org. Chem., 1990, 55,
6219–6223.
9 D. M. Stanbury, in Progress in Inorganic Chemistry, ed. K. D. Karlin,
John Wiley & Sons, 2007, vol. 47, pp. 511–561.
0 E. Fischer, Ber. Dtsch. Chem. Ges., 1877, 10, 1331–1338.
ꢁ
1
ꢁ1
is 21.56 M
s .
Finally, this new tert-butoxy radical sensing was evaluated in
living cells through the fluorescent imaging. Human prostate
cancer cells (PC-3) incubated with tBuO radicals from heated
1
2
tBuONO and probe 1 exhibited bright fluorescence in the green 21 J. K. Kochi, J. Am. Chem. Soc., 1962, 84, 1193–1197.
2
2
2 L. Du, N. Ni, M. Li and B. Wang, Tetrahedron Lett., 2010, 51, 1152–1154.
3 Z. Sun, S. Zheng, J. Wang, M. F. Ge and D. Wang, Chem.–Eur. J.,
channel (Fig. 4e), whereas PC-3 cells treated with only probe 1
showed modest fluorescence in the green channel (Fig. 4b). If cells
2001, 7, 2995–2999.
were treated with catalase (a free radical scavenger), after being 24 H. H. Szeto, AAPS J., 2006, 8, 277–283.
This journal is c The Royal Society of Chemistry 2013
Chem. Commun.