NitroxylFluor: A Thiol-Based Fluorescent Probe for Live-Cell Imaging of Nitroxyl

Article abstract of DOI:10.1021/jacs.7b11471

Detection of nitroxyl (HNO), the transient one-electron reduced form of nitric oxide, is a significant challenge owing to its high reactivity with biological thiols (with rate constants as high as 10⁹ M^-1 s^-1). To address this, we report a new thiol-based HNO-responsive trigger that can compete against reactive thiols for HNO. This process forms a common N-hydroxysulfenamide intermediate that cyclizes to release a masked fluorophore leading to fluorescence enhancement. To ensure that the cyclization step is rapid, our design capitalizes on two established physical organic phenomena; the alpha-effect and the Thorpe-Ingold effect. Using this new trigger, we developed NitroxylFluor, a selective HNO-responsive fluorescent probe. Treatment of NitroxylFluor with an HNO donor results in a 16-fold turn-on. This probe also exhibits excellent selectivity over various reactive nitrogen, oxygen, and sulfur species and efficacy in the presence of thiols (e.g., glutathione in mM concentrations). Lastly, we successfully performed live cell imaging of HNO using NitroxylFluor.

Full text of DOI:10.1021/jacs.7b11471

Communication
Cite This: J. Am. Chem. Soc. XXXX, XXX, XXX−XXX
pubs.acs.org/JACS
NitroxylFluor: A Thiol-Based Fluorescent Probe for Live-Cell Imaging
of Nitroxyl
Nicholas W. Pino, Jerome Davis, III, Zhengxin Yu, and Jefferson Chan*
Department of Chemistry and Beckman Institute for Advanced Science and Technology, University of Illinois at Urbana−Champaign,
Urbana, Illinois 61801, United States
*
S Supporting Information
ABSTRACT: Detection of nitroxyl (HNO), the transient
one-electron reduced form of nitric oxide, is a significant
challenge owing to its high reactivity with biological thiols
9
−1 −1
(
with rate constants as high as 10 M s ). To address
this, we report a new thiol-based HNO-responsive trigger
that can compete against reactive thiols for HNO. This
process forms a common N-hydroxysulfenamide inter-
mediate that cyclizes to release a masked fluorophore
leading to fluorescence enhancement. To ensure that the
cyclization step is rapid, our design capitalizes on two
established physical organic phenomena; the alpha-effect
and the Thorpe-Ingold effect. Using this new trigger, we
developed NitroxylFluor, a selective HNO-responsive
fluorescent probe. Treatment of NitroxylFluor with an
HNO donor results in a 16-fold turn-on. This probe also
exhibits excellent selectivity over various reactive nitrogen,
oxygen, and sulfur species and efficacy in the presence of
thiols (e.g., glutathione in mM concentrations). Lastly, we
successfully performed live cell imaging of HNO using
NitroxylFluor.
Figure 1. (a) HNO reacts with protein thiols to form an N-
hydroxysulfenamide intermediate, which rearranges to a sulfinamide or
is attacked by another thiol to form a disulfide. (b) HNO reacts with
our new thiol-based trigger in the same fashion to afford a common N-
hydroxysulfenamide intermediate, which cyclizes to release the masked
dye.
low thiol pK (e.g., glyceraldehyde 3-phosphate dehydrogen-
a
9
ase). Once formed, the N-hydroxysulfenamide intermediate
can rearrange to a sulfinamide product or react with another
thiol to yield a disulfide (Figure 1a). Thus, any detection
strategy must proceed with comparable kinetics to intercept
HNO before it is metabolized.
itroxyl (HNO), the one-electron reduced derivative of
N
nitric oxide (NO), has emerged as a potential therapeutic
agent for the treatment of cardiovascular disorders due to its
unique ability to increase cardiac output by decreasing venous
A promising approach to detect analytes in general is to
utilize reaction-based probes that couple fluorescent enhance-
1
resistance. HNO is also recognized to protect against
2
10
myocardial ischemia-reperfusion injury and has gained
ment with analyte-associated reactivity. In the context of
3
attention as a possible anticancer agent. The cellular
HNO sensing, a variety of reaction-based HNO probes for
1
1,12
production of HNO is proposed to involve either non-
enzymatic or enzymatic mechanisms. For example, HNO has
fluorescence imaging have been developed.
Such probes are
or phosphine-
1
1,13−18
based on either redox coordination
4
19−21
been generated via reduction of NO by thiols, or enzymatic
mediated Staudinger chemistry.
However, metal-based
reduction of L-arginine by nitric oxide synthase (NOS) under
HNO fluorescent probes are limited by slow reaction kinetics
4
tetrahydrobiopterin-free conditions. It is also speculated that
relative to thiols and exhibit small dynamic range (typically less
11,14,16,18
hydroxylamine and N-hydroxy-L-arginine may be oxidized to
than 5-fold fluorescent turn-on).
Phosphine-based
produce HNO in reactions mediated by heme proteins, such as
peroxidases.
probes suffer from reduced sensitivity because for every
equivalent of HNO, two probes are required for sensing
because one equivalent is consumed to generate an
5
,6
The detection of HNO in a biological context is an immense
challenge due to its highly reactive chemical nature. For
instance, HNO can spontaneously dimerize and dehydrate in a
2
0
unproductive phosphine oxide byproduct. Additionally,
some phosphine-based probes also exhibit cross-reactivity
12
6
−1 −1
7
with NO donors. Herein, we report the rational design of a
new bioinspired thiol-based trigger for the development of a
series of HNO-responsive fluorescent probes (Figure 1b). We
sequence that affords N O (k = 8 × 10 M s ). HNO can
2
also act as a potent electrophile in a reaction with nucleophilic
biological thiols to form N-hydroxysulfenamide intermediates
(
1
(
Figure 1a). The rate constants for this reaction range from 2 ×
6
−1 −1
0 M s , for low molecular weight thiols (e.g., glutathione
Received: October 27, 2017
8
9
−1 −1
GSH)) to greater than 10 M
s
for proteins exhibiting a
©
XXXX American Chemical Society
A
DOI: 10.1021/jacs.7b11471
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
demonstrate excellent selectivity against a panel of reactive
oxygen species, reactive sulfur species and reactive nitrogen
species, as well as efficacy for HNO detection in competition
experiments in the presence of biological thiols. In this study,
we employ our best probe, NitroxylFluor, to successfully image
HNO in living MDA-MB-231 human breast adenocarcinoma
cells.
We propose that for any HNO-responsive trigger to be
competent in a biological setting, it must exhibit sufficient
reactivity with HNO to outcompete abundant biological thiols
present in the cellular milieu. Our initial design featured a 2-
mercaptoacetate trigger that can be used to cap the hydroxyl
group on various dye platforms to quench its fluorescence
reagent by blocking the reactivity of the thiol via methylation
of NB-2. No hydrolysis was observed after incubation in
aqueous media, providing compelling evidence that the release
of the coumarin was indeed due to α-thiolactonization (Figure
S2). To address this instability issue we sought to further
stabilize the ester moiety by exploiting X···CO n→π*
interactions (where X = halogen) by installing an ortho-chloro
25
substituent to give rise to NitroxyBlue-3 (NB-3) (Figure 2a).
We reasoned this modification would be sufficient to prevent
intramolecular attack by the thiol, yet would allow for
cyclization of the more reactive N-hydroxysulfenamide α-
nucleophile. Indeed, when NB-3 was treated with a vehicle
control, we did not observe any fluorescence enhancement,
indicating the ortho-chloro substituent was sufficient to
stabilize the ester from α-thiolactonization (Figure S1). On
the other hand, the addition of Angeli’s salt resulted in a 3.6-
fold turn-on response (Figure S1).
(Figure 2a). In the absence of HNO, we anticipated that the
Having established a new thiol-based HNO-responsive
trigger on a commercially available coumarin scaffold, we
sought to develop a custom HNO probe with an absorbance
and emission profile in the visible spectrum. Visible wavelength
probes are more desirable than their UV counterparts for
biological studies because visible light is less phototoxic. This
led to the development of our final probe in this series,
NitroxylFluor, which is based on a chlorinated hydroxymethyl
fluorescein platform. Synthesis of NitroxylFluor was achieved
by diallylation of fluorescein with allyl bromide to afford 1 in
3
7% yield. Lithium aluminum hydride reduction of the allyl
Figure 2. Chemical structures of (a) the NitroxylBlue series and (b)
NitroxylFluor and the fluorescent turned over product.
ester also resulted in reduction of the xanthene moiety to give
the hydroxymethyl intermediate 2 in 83% yield. Disruption of
the xanthene π-system facilitated regioselective ortho,ortho-
dichlorination of the resulting phenol moiety under basic
conditions using sodium hypochlorite to yield 3. Of note, we
did not exhaustively try to obtain the monochloro product
since dichlorination would likely lead to greater stabilization
against α-lactonization. The chromophore was reestablished via
chloranil oxidation to afford 4, which was subsequently capped
using a tritylated 2-mercaptoisobutyric acid building block
under Steglich coupling conditions to give 5 in 60% yield over
cyclization to release the dye would be unlikely since the
product would be a strained α-thiolactone. However, we
hypothesized that reaction with HNO would yield a highly
reactive nucleophile (N-hydroxysulfenamide intermediate) that
could readily cyclize due to the lone-pair electrons on the
nitrogen atom, which enhances the nucleophilicity of the
2
2,23
adjacent −OH group (alpha-effect).
With this design in mind, we developed NitroxylBlue-1 (NB-
), a coumarin-based HNO probe featuring a 2-mercaptoace-
1
2
-steps. The O-allyl functionality was removed under Tsuji-
tate trigger (Figure 2a). The maximum absorbance and
emission of NB-1 were centered at 350 and 448 nm,
respectively. The probe itself was weakly fluorescent indicating
efficient fluorescent quenching. Ten minutes after treatment
with Angeli’s salt, an HNO donor, we observed insignificant
fluorescence enhancement over the background (Figure S1).
We speculate that the lack of response was due to cyclization
kinetics that were too slow. Additionally, we observed that the
ester moiety was prone to hydrolysis (∼2-fold turn-on after 10
min at 37 °C), which can lead to false positives (Figure S1). To
address these shortcomings, we installed an α-geminal dimethyl
group to afford NitroxylBlue-2 (NB-2) (Figure 2a). This
modification was made to increase stability by shielding the
ester from hydrolysis with greater steric bulk. We also aimed to
leverage the Thorpe-Ingold effect, which describes the
acceleration of cyclization from the substitution of hydrogen
atoms for alkyl groups on the carbons tethering two reaction
Trost deallylation conditions to afford 6 in near quantitative
yields. Finally, the trityl protecting group was removed under
acidic conditions to give NitroxylFluor in 13% yield (Scheme
1
).
With NitroxylFluor in hand, we began to evaluate its
response to HNO in vitro. NitroxylFluor displays an absorbance
3
−1
−1
maximum centered at 505 nm (ε = 1.3 × 10 M cm ) and an
Scheme 1. Synthesis of NitroxylFluor
2
4
centers. Although incubating NB-2 with Angeli’s salt did
result in a more significant 3.5-fold turn-on response, we were
surprised to find that NB-2 was not stable in aqueous media as
the background enhancement at 37 °C was 1.6-fold after 10
min (Figure S1). To determine whether this was due to
hydrolysis or α-thiolactonization, we developed a control
B
DOI: 10.1021/jacs.7b11471
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
emission maximum at 525 nm. Although the fluorescence was
initially attenuated, reaction with HNO afforded a 16-fold
fluorescent turn-on response in a matter of seconds (Figure
function in the presence of a large excess GSH, we synthesized
a control compound where we replaced the dye with a methoxy
nucelofuge (Scheme S5). We did not anticipate that this will
3a,b). Given the rapid reaction of HNO with thiols, we
significantly alter the pK of the thiol. However, the poor
a
leaving group ability of the methoxy suppresses α-thiolactoni-
zation, which allowed us to measure a pK value of 8.01 by UV/
a
vis spectroscopy (Figure S3). In comparison, the reported pK
a
27
value of GSH is 9.42. The lower pK of our probe renders it
a
more reactive toward HNO.
After demonstrating excellent responsiveness to HNO,
exceptional selectivity over a range of biological analytes, and
effective competition with thiols, we tested the ability of
NitroxylFluor to visualize HNO in living cells. Specifically, we
stained MDA-MB-231 cells, a human breast adenocarcinoma
cell line, with NitroxylFluor for 15 min before removing the
probe solution and replacing with fresh media. Addition of
Angeli’s salt at various concentrations (250 to 1000 μM)
resulted in a dose-dependent fluorescence turn-on response
after 15 min (Figure 4a,b). A statistically significant enhance-
Figure 3. (a) Fluorescence spectra of NitroxylFluor (2 μM) upon
addition of AS (100 μM). (b) Fluorescence kinetic trace of
NitroxylFluor upon addition of 1 mM Angeli’s salt, arrow indicates
time of addition. (c) Response of NitroxylFluor to various reactive
oxygen, sulfur, and nitrogen species at concentrations of 100 μM
(
GSH was tested at 1 mM). Measurements were taken 15 min into
treatment. (d) Competition assays against various thiols. Control
indicates no addition of Angeli’s salt. Statistics are compared to the
control. *, p < 0.05; ***, p < 0.001; p < 0.0001. (n = 3).
Figure 4. (a) Confocal microscopy images acquired by irradiation of
MDA-MB-231 cells treated with 0 μM (vehicle control), 250 μM, 500
μM, 1000 μM Angeli’s salt for 15 min at 25 °C with a 488 nm laser.
Images taken through a 40× oil immersion objective. Scale bar
represents 20 μm. (b) Quantification of imaging data. *, p < 0.05; ***,
p < 0.001; p < 0.0001. (n = 5).
speculate the rate-limiting step is not formation of the N-
hydroxysulfenamide intermediate, but rather cyclization.
Having established excellent responsiveness to HNO, we
turned our attention to determining the selectivity profile of
the trigger. In particular, NitroxylFluor was treated with various
oxidants, including H O because it is known that thiols can be
ment was observed even for the lowest Angeli’s salt
concentration. We also performed time-lapse imaging by
treating NitroxylFluor stained cells with 500 μM Angeli’s salt
and recorded images every 1 min after addition for 20 min. We
noted a gradual increase in the fluorescence during this time
course. Representative images at 0, 10, and 20 min are shown in
Figure S5. We also evaluated the cytotoxicity of NitroxylFluor
by performing a MTT assay where MDA-MB-231 cells were
stained with 10 μM probe for 3, 6, and 24 h. We did not
observe cytotoxicity at any time point (Figure S6). The
subcellular localization of NitroxylFluor in MDA-MB-231 cells
appeared to be lysosomal. We confirmed this by using a
commercial lysosome stain (LysoTracker Red) and observed
significant overlap in signal with a Pearson’s coefficient of 0.932
± 0.01 (Figure S7).
In conclusion, we have harnessed the known reactivity of
thiols with HNO to develop a novel HNO-reactive trigger. We
then called upon physical organic phenomena to improve upon
this reactivity to develop a probe that competes with the native
chemistry, even when known scavengers are in excess. In doing
this, we have created a selective fluorescent probe for HNO
that avoids inefficient byproducts and the lack of selectivity
suffered by previously reported technologies. Our results
indicate that the cyclization of the trigger outcompetes
foreseeable side reactions. Owing to the culmination of this
work, we predict that this technology will allow us and others to
2
2
oxidized to sulfenic acids (RSOH), sulfinic acids (RSO H), and
2
sulfonic acids (RSO H), which can potentially cyclize to release
3
2
6
the dye. However, we did not observe cross-reactivity with
any of the oxidants tested (Figure 3c). Likewise, when
NitroxylFluor was incubated with various reactive sulfur species
(
e.g., H S) and reactivate nitrogen species (e.g., NO) there was
2
no change in fluorescence after a 15 min incubation at 37 °C
(
Figure 3c). Taken together, this demonstrates high selectivity
of the trigger for HNO.
We next evaluated whether NitroxylFluor possessed requisite
reactivity to compete against biologically relevant thiols for
HNO. In these experiments, we treated 10 μM solutions of
NitroxylFluor and various thiols such as L-cysteine (100 μM),
N-acetyl-L-cysteine (100 μM), and GSH (1 mM) with Angeli’s
salt (200 μM) (Figure 3d). In the case of L-cysteine, the
fluorescence response was not attenuated (16.2-fold increase)
indicating NitroxlFluor was the more reactive species. Similarly,
competition with N-acetyl-L-cysteine resulted in a 9.8-fold
fluorescence enhancement. GSH, on the other hand, is highly
abundant in cells (millimolar concentrations) and reacts rapidly
6
−1 −1
with HNO (k = 2 × 10 M s ). Thus, interference from GSH
presents a significant challenge for any HNO detection strategy.
Even in the presence of 1 mM GSH, treatment of NitroxylFluor
with Angeli’s salt still gave rise to a 3.9-fold fluorescent turn-on
response. To determine the origin of why NitroxylFluor can
C
DOI: 10.1021/jacs.7b11471
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
study HNO with enhanced rate and selectivity to elucidate its
biological roles and pharmacological potential.
(14) Rosenthal, J.; Lippard, S. J. J. Am. Chem. Soc. 2010, 132, 5536.
(
15) Zhou, Y.; Liu, K.; Li, J. Y.; Fang, Y.; Zhao, T. C.; Yao, C. Org.
Lett. 2011, 13, 1290.
16) Apfel, U.-P.; Buccella, D.; Wilson, J. J.; Lippard, S. J. Inorg.
Chem. 2013, 52, 3285.
17) Royzen, M.; Wilson, J. J.; Lippard, S. J. J. Inorg. Biochem. 2013,
18, 162.
18) Wrobel, A. T.; Johnstone, T. C.; Deliz Liang, A.; Lippard, S. J.;
Rivera-Fuentes, P. J. Am. Chem. Soc. 2014, 136, 4697.
19) Reisz, J. A.; Klorig, E. B.; Wright, M. W.; King, S. B. Org. Lett.
009, 11, 2719.
20) Kawai, K.; Ieda, N.; Aizawa, K.; Suzuki, T.; Miyata, N.;
(
ASSOCIATED CONTENT
Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/jacs.7b11471.
■
*
S
(
1
(
Experimental details, including synthetic procedures for
NB-1, NB-2, NB-3, and NitroxylFluor (PDF)
(
2
(
AUTHOR INFORMATION
Corresponding Author
jeffchan@illinois.edu
■
Nakagawa, H. J. Am. Chem. Soc. 2013, 135, 12690.
(21) Miao, Z.; Reisz, J. A.; Mitroka, S. M.; Pan, J.; Xian, M.; King, S.
B. Bioorg. Med. Chem. Lett. 2015, 25, 16.
*
(22) Jencks, W. P.; Carriuolo, J. J. Am. Chem. Soc. 1960, 82, 1778.
(23) Jencks, W. P.; Carriuolo, J. J. Am. Chem. Soc. 1960, 82, 675.
(24) Jung, M. E.; Piizzi, G. Chem. Rev. 2005, 105, 1735.
ORCID
Jefferson Chan: 0000-0003-4139-4379
Notes
(25) Chyan, W.; Kilgore, H. R.; Gold, B.; Raines, R. T. J. Org. Chem.
2017, 82, 4297.
The authors declare no competing financial interest.
(
26) van Bergen, L. A.; Roos, G.; De Proft, F. J. Phys. Chem. A 2014,
18, 6078.
27) Tajc, S. G.; Tolbert, B. S.; Basavappa, R.; Miller, B. L. J. Am.
Chem. Soc. 2004, 126, 10508.
1
(
ACKNOWLEDGMENTS
■
N.W.P. was supported by the Chemistry-Biology Interface
Training Grant (T32 GM070421), Robert C. and Carolyn J.
Springborn Graduate Fellowship, and Alfred P. Sloan
Foundation MPHD Fellowship. J.D. III was supported by an
Undergraduate Research Scholarship from the Department of
Chemistry at UIUC. Z.Y. was supported by Department of
Chemistry at Nankai University. J.C. acknowledges the Alfred
P. Sloan Foundation for an Alfred P. Sloan Fellowship. Major
funding for the 500 MHz Bruker CryoProbe was provided by
the Roy J. Carver Charitable Trust (Muscatine, Iowa; Grant
#15-4521) to the School of Chemical Sciences NMR Lab. The
Q-Tof Ultima mass spectrometer was purchased in part with a
grant from the National Science Foundation, Division of
Biological Infrastructure (DBI-0100085). We also acknowledge
the Core Facilities at the Carl R. Woese Institute for Genomic
Biology for access to the Zeiss LSM 700 Confocal Microscope
and corresponding software.
REFERENCES
■
(
1) Fukuto, J. M.; Dutton, A. S.; Houk, K. N. ChemBioChem 2005, 6,
612.
(
2) Irvine, J. C.; Ritchie, R. H.; Favaloro, J. L.; Andrews, K. L.;
Widdop, R. E.; Kemp-Harper, B. K. Trends Pharmacol. Sci. 2008, 29,
01.
6
(
(
3) Shoman, M. E.; Aly, O. M. Curr. Top. Med. Chem. 2016, 16, 2464.
4) Suarez, S. A.; Munoz, M.; Alvarez, L.; Venancio, M. F.; Rocha, W.
R.; Bikiel, D. E.; Marti, M. A.; Doctorovich, F. J. Am. Chem. Soc. 2017,
39, 14483.
1
(
5) Liochev, S. I.; Fridovich, I. Free Radical Biol. Med. 2003, 34, 1399.
(
6) Miranda, K. M.; Paolocci, N.; Katori, T.; Thomas, D. D.; Ford, E.;
Bartberger, M. D.; Espey, M. G.; Kass, D. A.; Feelisch, M.; Fukuto, J.
M.; Wink, D. A. Proc. Natl. Acad. Sci. U. S. A. 2003, 100, 9196.
(
9
(
7) Shafirovich, V.; Lymar, S. V. Proc. Natl. Acad. Sci. U. S. A. 2002,
9, 7340.
8) Miranda, K. M.; Paolocci, N.; Katori, T.; Thomas, D. D.; Ford, E.;
Bartberger, M. D.; Espey, M. G.; Kass, D. A.; Feelisch, M.; Fukuto, J.
M.; Wink, D. A. Proc. Natl. Acad. Sci. U. S. A. 2003, 100, 9196.
(
9) Flores-Santana, W.; Switzer, C.; Ridnour, L. A.; Basudhar, D.;
Mancardi, D.; Donzelli, S.; Thomas, D. D.; Miranda, K. M.; Fukuto, J.
M.; Wink, D. A. Arch. Pharmacal Res. 2009, 32, 1139.
(
(
(
(
10) Chan, J.; Dodani, S. C.; Chang, C. J. Nat. Chem. 2012, 4, 973.
11) Rivera-Fuentes, P.; Lippard, S. J. Acc. Chem. Res. 2015, 48, 2927.
12) Miao, Z.; King, S. B. Nitric Oxide 2016, 57, 1.
13) Tennyson, A. G.; Do, L.; Smith, R. C.; Lippard, S. J. Polyhedron
2
007, 26, 4625.
D
DOI: 10.1021/jacs.7b11471
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX