New rhodamine nitroxide based fluorescent probes for intracellular hydroxyl radical identification in living cells

Article abstract of DOI:10.1021/ol202816m

The synthesis, characteristics, and biological applications of a series of new rhodamine nitroxide fluorescent probes that enable imaging of hydroxyl radicals (?OH) in living cells are described. These probes are highly selective for ?OH in aqueous solution, avoiding interference from other reactive oxygen species (ROS), and they facilitate ?OH imaging in biologically active samples. The robust nature of these probes (high specificity and selectivity, and facile synthesis) offer distinct advantages over previous methods for ?OH detection.

Full text of DOI:10.1021/ol202816m

ORGANIC
LETTERS
2012
Vol. 14, No. 1
50–53
New Rhodamine Nitroxide Based
Fluorescent Probes for Intracellular
Hydroxyl Radical Identification in Living
Cells
Nazmiye B. Yapici,† Steffen Jockusch,§ Alberto Moscatelli,§
Srinivas Rao Mandalapu,† Yasuhiro Itagaki,§ Dallas K. Bates,† Sherri Wiseman,‡
K. Michael Gibson,*,‡ Nicholas J. Turro,*,§ and Lanrong Bi*,†
Department of Chemistry and Department of Biological Sciences, Michigan
Technological University, Houghton, Michigan 49931, United States, and Department
of Chemistry, Columbia University, 3000 Broadway, New York, New York 10027,
United States
*lanrong@mtu.edu (L.B.); kmgibson@mtu.edu (K.M.G.); njt3@columbia.edu (N.J.T.)
Received October 19, 2011
ABSTRACT
The synthesis, characteristics, and biological applications of a series of new rhodamine nitroxide fluorescent probes that enable imaging of
hydroxyl radicals (•OH) in living cells are described. These probes are highly selective for •OH in aqueous solution, avoiding interference from
other reactive oxygen species (ROS), and they facilitate •OH imaging in biologically active samples. The robust nature of these probes (high
specificity and selectivity, and facile synthesis) offer distinct advantages over previous methods for •OH detection.
Reactive oxygen species (ROS) are associated with
numerous pathologies including neurodegenerative dis-
eases, ischemic or traumatic brain injury, cancer, dia-
betes, liver injury, and AIDS.1 The primary ROS include
the hydroxyl radical, superoxide anion, and hypochlorite
radical. Of the preceding, the hydroxyl radical (•OH)
holds an important role in relation to its high reactivity.
Essentially all biomacromolecules are susceptible to dam-
age from •OH, including carbohydrates, nucleic acids,
lipids, and amino acids2 making it one of the most
deleterious intracellular ROS. The ability to detect and
quantify intra- and extracellulary generated •OH yields
key information about the location and extent of redox
damage and provides clues to the resulting cellular re-
sponse. Accordingly, methods that are sensitive, accu-
rate, and reproducible for the detection of •OH provide
an important tool for understanding pathophysiology in
a host of inherited and acquired diseases. A frequently
(3) (a) Yang, X. F.; Guo, X. Q. Analyst 2001, 126, 1800. (b) Li, B. B.;
Gutierrez, P. L.; Blough, N. V. Anal. Chem. 1997, 69, 4295. (c) Tang, B.;
Zhang, L.; Geng, Y. Talanta 2005, 65, 769. (d) Ischiropoulos, H.; Gow,
A.; Thom, S. R.; Kooy, N. W.; Royall, J. A.; Crow, J. P. Methods
Enzymol. 1999, 301, 367. (e) Lebel, C. P.; Ischiropoulos, H.; Bondy, S. C.
Chem. Res. Toxicol. 1992, 5, 227. (f) Ohashi, T.; Mizutani, A.; Murakami,
A.; Kojo, S.; Ishii, T.; Taketani, S. FEBS Lett. 2002, 511, 21. (g) Ward-
man, P.; Burkitt, M. J.; Patel, K. B.; Lawrence, A.; Jones, C. M.; Everett,
S. A.; Vojnovic, B. J. Fluoresc. 2002, 12, 65. (h) Myhre, O.; Andersen,
J. M.; Aarnes, H.; Fonnum, F. Biochem. Pharmacol. 2003, 65, 1575. (i)
Tang, B.; Li, P.; Xie, T.; Duan, X.; Yu, F. B.; Wang, X. Chem.;Eur. J.
2010, 16, 1834. (j) Setsukinai, K.; Urano, Y.; Kakinuma, K.; Majima,
H. J.; Nagano, T. J. Biol. Chem. 2003, 278, 3170. (k) Soh, N.; Makihara,
K.; Ariyoshi, T.; Seto, D.; Maki, T.; Nakajima, H.; Nakano, K.; Imato, T.
Anal. Sci. 2008, 24, 293. (l) Lin, W. Y.; Yuan, L.; Song, J. Z. Chem.
Commun. 2010, 46, 7930.
† Department of Chemistry, Michigan Technological University
§ Department of Chemistry, Columbia University.
‡ Department of Biological Sciences, Michigan Technological University.
(1) Dalle-Donne, I.; Rossi, R.; Colombo, R.; Giustarini, D.; Milzani,
A. Clin. Chem. 2006, 52, 601.
(2) Halliwell, B.; Gutteridge, J. M. C. Free Radicals in Biology and
Medicine, 2nd ed.; Oxford University Press: New York, 1989.
r
10.1021/ol202816m
Published on Web 12/16/2011
2011 American Chemical Society

used approach for detecting •OH, electron spin resonance
(ESR) spectroscopy is limited by low spatial resolution
and, thus, cannot be applied to real-time imaging of •OH
in individual cells. To overcome this limitation, other
investigators have employed fluorescent probes coupled
to ESR in order to enhance both sensitivity and spatial
resolution needed for intracellular detection of •OH.3
molecular probes IꢀIII was accomplished using sequen-
tial “click” reactions under mild conditions (Scheme S1,
Supporting Information). In order to prevent fluoro-
phore/nitroxide cleavage, the nitroxide is introduced into
the molecular probe via the “click” reaction. To avoid
reduction of the nitroxide to the nonparamagnetic hydro-
xylamine derivative by sodium L-ascorbate during the
click reaction, copper(I) iodide (10 mol %) has been
employed as the copper(I) source instead of the more
common Cu(II)SO4/sodium L-ascorbate system. All in-
termediates were characterized by 1H and 13C NMR, and
the structural characteristics of all target compounds
were confirmed by high-resolution mass spectrometry.
Because of the observed paramagnetic broadening caused
by the free radical, probes IꢀIII were additionally treated
with sodium L-ascorbate to generate the corresponding
hydroxylamines (nonradical), which facilitated charac-
1
terization by H and 13C NMR. The nitroxide-labeled
probes were also analyzed by EPR spectroscopy to con-
firm the intact nitroxide label. Probes IꢀIII are stable and
were stored at room temperature in sealed and light-
protected bottles until use.
Spectroscopic Properties and Response to •OH. We
employed the Fenton reaction to generate •OH, at pH
4.0, a pH supporting optimal Fenton oxidation.4 Spectral
characteristics of probes IꢀIII were tested following
reaction with •OH generated from the Fenton reaction.
Figures 2 and S1 describe typical absorbance/fluores-
cence spectra for probes IꢀIII in the absence and presence
of a Fenton reagent. Probes IꢀIII in their lactam form
show only weak absorbance and almost no fluorescence.
After addition of the Fenton reagent •OH are generated,
which can react with DMSO, which is present in the
solution, to generate methyl radicals (•CH3). Subse-
quently, •CH3 can react efficiently with the nitroxide
radical of the probes, which leads to a diamagnetic
compound, thereby eliminating intramolecular quench-
ing and resulting in enhanced fluorescence. In addition,
the low pH (pH = 4) opens the lactam leading to strongly
absorbing rhodamine chromophores.
Next, we subsequently monitored the changes in fluo-
rescence spectra of our probes in the presence of differing
•OH concentrations. The concentration of •OH is directly
proportional to the Cu2þ concentration. Figure S2 de-
picts the fluorescence responses of probes IꢀIII to the
presence of •OH. When simultaneously added to probes
IꢀIII, the Fenton reagent and DMSO yielded an im-
mediate increase in fluorescence intensity. For example,
incubation of probe III with Fenton reagent/DMSO
led to an ∼400-fold increase in fluorescence intensity,
and this fluorescence intensity increased in parallel with
the increase of the Fenton reagent concentration. When
the •OH concentration exceeds 33 μM, however, the
fluorescence intensity displayed saturation indicating that
essentially all of probe III (1 μM) had reacted with •OH.
Figure 1. Rhodamine nitroxide based fluorescent probes IꢀIII.
Nonetheless, the practical applications for fluorescent-
ESR probes in biological systems are limited by the co-
occurrence of cell damage, autofluorescence, and inter-
ference from other ROS. We sought to overcome these
limitations through the development of a nonredox mech-
anism for •OH detection that eliminates ROS interfer-
ences and facilitates high specificity. Our novel approach
for •OH imaging employs fluorogenic spin probes that
can be detected by both fluorescence and EPR spectro-
scopy. The rhodamine nitroxide probe represents a non-
fluorescent species that quantitatively reacts with •OH in
DMSO solution to produce highly fluorescent products
with absorbance/emission maxima at 560/588 nm. Because
the reaction mechanism is not redox-based, we hypothesized
that these probes would be selective. Moreover, the long
wavelength emission spectrum of the fluorescent product
minimizes interference from autofluorescence that exists in
the majority of biological samples and mostly appears at
shorter wavelengths. The current report describes the synth-
esis, spectroscopic characteristics, and imaging features of a
new series of highly selective •OH-responsive fluorescent
probes (IꢀIII) (Figure 1) that we predict will significantly
advance the field of intracellular •OH imaging.
(4) (a) Kwon, B. G.; Lee, D. S.; Kang, N.; Yoon, J. Water Res. 1999,
33, 2110. (b) Nogueira, R. F. P.; Guimaraes, J. R. Water Res. 2000, 34,
895. (c) Kang, Y. W.; Hwang, K. Y. Water Res. 2000, 34, 2786.
Design and Synthesis of the Rhodamine Nitroxide Based
Dual Probes for Hydroxyl Radical Detection. Synthesis of
Org. Lett., Vol. 14, No. 1, 2012
51

pigment epithelial cells) to examine the capacity of our
probes to measure •OH generated in an active biological
environment. Incubation of ARPE-19 cells with probe I,
II, or III (2 μM) for 12 h at 37 °C yielded negligible
intracellular background fluorescence (Figure S5). Con-
versely, addition of 10 μM •OH to probe-loaded cells
(12 h, 37 °C) triggers a considerable increase in intracellular
fluorescence. Also, determination of DIC (differential
interference contrast) images and nuclear counter stain-
ing confirmed that our cells were viable throughout the
imaging analyses, all of which indicated that our probes
were cell permeable, nontoxic, and capable of detecting
increased •OH under conditions indicative of oxidative
stress.
Iron or copper can induce production of •OH from
H2O2. Accordingly, we extended the previous studies and
further induced oxidative stress using hydrogen peroxide.
Exposure of probe (IꢀIII)-preloaded ARPE-19 cells re-
sulted in considerable fluorescence induction following
H2O2 exposure (Figure S6), and the time-related increase
in fluoroscence suggested ongoing reaction of the probe
with additional radicals. As a control for this experiment,
cells were exposed to oxidative stress conditions in the
absence of the probes, and the weak autofluorescence
indicated that enhanced fluorescence with peroxide addi-
tion correlated with increased •OH concentration.
Low-Level Detection of •OH with PMA (12-Myristate
13-Acetate) Stimulation. To test the sensitivity of our
probes detecting the concentrations of •OH produced
with physiological stimulation (very low levels expected
in signaling cascades), we employed ARPE-19 cells which
are predicted to generate low micromolar levels of •OH
upon stimulation with PMA.6 Accordingly, ARPE-19
cells were stimulated with PMA (40 ng/mL) and incu-
bated with probes IꢀIII (2 μM) for 12 h (Figure 3).
APRE-19 cells incubated with 2 μM probes IꢀIII re-
vealed minimal background fluorescence. Conversely, a
striking increase in intracellular fluorescence was ob-
served upon treatment of probe IꢀIII- (2 μM) preloaded
cells with PMA. Control experiments with cells in the
absence of either probe or a probe without PMA stimula-
tion yielded negligible fluorescence responses. Moreover,
cellular viability was verified with DIC images. These
findings indicate that probes IꢀIII are both cell mem-
brane-permeable and responsive to intracellular •OH
concentration changes, with probe III producing the
strongest fluorescence signal.
Figure 2. Absorbance (line 1, 3) and fluorescence spectra (line 2,
4) of probe I (1 μM) before (line 1, 2) and after (line 3, 4) addition
of Fenton reagent (Cu2þ: H2O2, mol/mol = 1:30) in the presence
of DMSO (0.1%) at pH = 4.
The highest fluorescence enhancement seen with probe III
was most likely associated with its binitroxide structure.
Unique to our strategy, a dark background from which a
bright signal appears upon reaction with •OH offers a dis-
tinct advantage in comparison to other probes for •OH,3
and the correspondingly decreased background fluorescence
lends itself to a higher signal-to-noise ratio, a highly desirable
end point for in vivo imaging with live cells.
We exposed our probes to a variety of radical species.
Our results revealed that probe III exhibited a >1000-
fold higher response for •OH in comparison to NO,
H2O2, and glutathione and a >40-fold higher response
for •OH over ROO•, ascorbate, SIN-1, and superoxide.
Probes I and II also demonstrated selectivity for •OH
versus other ROS (Figure S3ꢀS4). We suggest that the
selectivity of these probes for •OH results from the high
reactivity of •OH with any organic molecules (such as
DMSO present in our experiments) to generate carbon
centered radicals by H-abstraction, which subsequently
react with the nitroxides. The other ROS have lower rate
constants of direct reaction with nitroxides and lower rate
constants of H-abstraction from organic molecules to
generate carbon centered radicals.
Detection of •OH in Cancer Cells. Expanding evidence
suggests a link between free radicals and cancer.7 Accord-
ingly, to examine the capacity of our probes to quantify
•OH in cancer cells, we characterized these probes in
SW620 (human colon cancer cells), Hela (human cervical
cancer cells), and HepG2 cells (human hepatocellular
liver carcinoma cells). Time-course analyses of probe
uptake revealed rapid cellular access and stability over
In Vivo Detection of •OH during Oxidative Stress. Since
retinal pigment epithelial cells are susceptible to oxidative
damage,5 we employed ARPE-19 cells (human retinal
(6) Takeuchi, T.; Nakajima, M.; Morimoto, K. Carcinogenesis 1996,
(5) (a) Winkler, B. S.; Boulton, M. E.; Gottsch, J. D.; Sternberg, P.
Mol. Vis. 1999, 5. (b) Beatty, S.; Koh, H. H.; Henson, D.; Boulton, M.
Surv. Ophthalmol. 2000, 45, 115. (c) Cai, J. Y.; Nelson, K. C.; Wu, M.;
Sternberg, P.; Jones, D. P. Prog. Retin. Eye Res. 2000, 19, 205.
17, 1543.
(7) Hemnani, T.; Parihar, M. S. Indian J. Physiol. Pharmacol. 1998,
42, 440.
52
Org. Lett., Vol. 14, No. 1, 2012

Figure 4. Fluorescence images of probe III in HepG2 cells. Probe
III (30 μM, green, A) was incubated in non-FBS media for 3 h and
counterstained with Hoechst 3342 (1 μg/mL, blue, B); mitotracker
green (80 nM, red, C); and an overlay (D). All images were
acquired with 40ꢁ objective.
Figure 3. Fluorescence and DIC images of PMA (40 ng/mL)
induced •OH production in ARPE-19 cells using 2 μM probe
(IꢀIII) for 12 h (pH 6) and 37 °C. Images for probe I alone (A
and B); Images for probe I þ PMA (C and D); Images for probe II
alone (E and F); Images for probe II þ PMA (G and H); Images for
probe III alone (I and J); Images for probe III þ PMA (K and L).
All images were acquired with 20ꢁ objective lens.
In conclusion, we have developed new rhodamine ni-
troxide based fluorescent probes for intracellular •OH
imaging. The probes (IꢀIII) exhibit excellent selectivity for
•OH, with good dynamic response ranges and micromolar
detection limits for •OH. Further, we have documented the
utility of these probes for diverse applications in biological
imaging, where our probes demonstrate high sensitivity and
selectivity, good photostability, and cell membrane perme-
ability. Our studies will facilitate further development of
fluorescent probes that will aid in investigating the patho-
physiology of •OH toxicity in living cells and tissue.
time (Figures S7ꢀS12). There were no indications of cell
damage induced by probe loading. Along these lines,
HepG2 cells display an epithelial-like morphology, and
this morphology was maintained even after probe incu-
bation for >24 h. Similar results were observed with
spindle-shaped SW 620 cells; DIC imaging confirmed
cellular viability, at up to a 32 μM probe concentration.
Counterstaining of cells (HepG2, SW-620 and Hela cells)
with mitotracker and Hoechst 3342/DAPI (40,6-diamidi-
no-2-phenylindole dihydrochloride) revealed that the pre-
dominant probe distribution was in the mitochondria,
although some localization to the nucleus was also ob-
served (Figure 4). These results suggest that the predom-
inant site of •OH generation is intramitochondrial, with
only a slight production in the nuclear organelle.
Acknowledgment. L.B. is grateful to NASA (Grant
MSGC R85197), Research Excellence Fund (Grant
R01121) for support of this project. The authors at Columbia
thank the National Science Foundation for support through
Grant NSF-CHE-07-17518.
Supporting Information Available. Experimental de-
tails of synthesis, fluorescence spectroscopy, NMR data,
and fluorescence imaging are provided as Supporting
Information. This material is available free of charge
via the Internet at http://pubs.acs.org.
Org. Lett., Vol. 14, No. 1, 2012
53