A Multi-responsive Fluorescent Probe Reveals Mitochondrial Nucleoprotein Dynamics with Reactive Oxygen Species Regulation through Super-resolution Imaging

Article abstract of DOI:10.1002/anie.202005959

Understanding the biomolecular interactions in a specific organelle has been a long-standing challenge because it requires super-resolution imaging to resolve the spatial locations and dynamic interactions of multiple biomacromolecules. Two key difficulties are the scarcity of suitable probes for super-resolution nanoscopy and the complications that arise from the use of multiple probes. Herein, we report a quinolinium derivative probe that is selectively enriched in mitochondria and switches on in three different fluorescence modes in response to hydrogen peroxide (H₂O₂), proteins, and nucleic acids, enabling the visualization of mitochondrial nucleoprotein dynamics. STED nanoscopy reveals that the proteins localize at mitochondrial cristae and largely fuse with nucleic acids to form nucleoproteins, whereas increasing H₂O₂ level leads to disassociation of nucleic acid–protein complexes.

Full text of DOI:10.1002/anie.202005959

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Super-Resolution Imaging to Reveal Mitochondrial Nucleoprotein
Dynamics with Reactive Oxygen Species Regulation
Authors: Guanqing Yang, Zhengjie Liu, Ruilong Zhang, Xiaohe Tian,
Juan Chen, Guangmei Han, Bianhua Liu, Xinya Han, Yao Fu,
Zhangjun Hu, and Zhongping Zhang
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.202005959
Link to VoR: https://doi.org/10.1002/anie.202005959

10.1002/anie.202005959
Angewandte Chemie International Edition
RESEARCH ARTICLE
Super-Resolution Imaging to Reveal Mitochondrial Nucleoprotein
Dynamics with Reactive Oxygen Species Regulation
[a]
[a]
+
Guanqing Yang , Zhengjie Liu , Ruilong Zhang*[a],[b], Xiaohe Tian[a], Juan Chen[a], Guangmei Han[a],
+
Bianhua Liu[c], Xinya Han[d], Yao Fu[e], Zhangjun Hu[f], Zhongping Zhang*[a],[b],[c]
Abstract: Understanding the biomolecular interactions in a specific
organelle has been a long-standing challenge because it requires
super-resolution imaging to resolve the spatial locations and dynamic
interactions of multiple biomacromolecules. Two key difficulties,
however, are that molecular probes suitable for super-resolution
nanoscopy are rather rare, and more adversely the use of multiple
probes complicates their cellular uptakes, biodistributions, and
spectral distinctions. Here, we report a quinolinium derivative probe
that is selectively enriched in mitochondria, and switches on in three
different fluorescence modes in response to hydrogen peroxide
(H2O2), protein, and nucleic acid, enabling the visualization of
mitochondrial nucleoprotein dynamics. Stimulated emission depletion
(STED) nanoscopy reveals that the proteins localize at mitochondrial
cristae and largely fuse with nucleic acids to form nucleoproteins,
whereas increasing H2O2 level leads to disassociation of nucleic acid-
protein complexes.
generally recognized as the cells “power plants” supplying energy
by oxidizing glucose, fat, and other biomolecules. Meanwhile,
they also produce reactive oxidative species (ROS) that activate
a number of signaling pathways and regulate the functions within
the mitochondria.[1] Interestingly, although the majority of
mitochondrial functions are realized through proteins and
enzymes, mitochondria have their own nucleic acids, which
independently undergo fission and fusion, playing a critical role in
controlling mitochondrial replication, aging, and protein
expression.[2] Similar to the genic materials in the cell nucleus, the
behaviors of mitochondrial nucleic acids are strictly regulated by
a serial of proteins.[2a,3] In the normal state, DNAs and RNAs often
form complexes with proteins (the complex known as
nucleoprotein) for improving stability in biological environments.
The association and dissociation between nucleic acids and
proteins are highly dynamic to realize the biological functions such
as replication, transcription, translation, and transportation.[4]
Therefore, understanding the behaviors of this dynamic duo as
well as its environmental controlling factors inside live cells is of
great significance in cell biology and physiology.
Introduction
Probing the nucleoprotein dynamics, however, has been
technically challenging due to multiple cellular barriers. In addition
to the cell plasma membrane, mitochondria are also enclosed by
another lipid bilayer, isolating the functional biomolecules from the
rest of the cell space. A traditional compromised approach is
transmission electron microscopy (TEM), which is great for
outlining the ultrafine structures of cell organelles, but incapable
of identifying multiple biological species and depicting the
dynamics of biomolecular interactions. In this context, advanced
super-resolution fluorescence nanoscopy has recently become
commercially available, providing biologists with a powerful tool
to peek inside living cells with unprecedented spatial and
temporal resolutions.[5] For example, super-resolution
nanoscopy in combination with fluorescent proteins allows
imaging of single functional protein on the cell membrane.[6]
More recently, super-resolution nanoscopy aided by membrane-
penetrating inorganic carbon dots has achieved dynamical
tracking of nucleic acids in cells and C. elegans.[7] In parallel, a
large number of fluorescent probes have been developed to
detect ROS,[8] enzymes[9] and other biological analytes[10] in live
cells, but most of them exhibit optical properties undesirable for
multicolor super-resolution imaging such as limited by small
Stokes shift and photostability.[11] Furthermore, these imaging
agents all have different chemical properties and intracellular
biodistributions, making multi-analyte imaging in the same cell
compartment extremely difficult.
Mitochondria are one of the most intriguing and sophisticated
intracellular organelles owing to their unique structures, biological
components, and physiological functions. Mitochondria are
[+]
[+]
[a]
G. Yang , Dr. Z. Liu , Dr. R. Zhang, Dr. X. Tian, J. Chen, Dr. G.
Han, Prof. Z. Zhang
School of Chemistry and Chemical Engineering
and Institute of Physical Science and Information Technology,
Anhui University,
Hefei, Anhui 230601 (China)
E-mail: zrl@ahu.edu.cn
zpzhang@iim.ac.cn
[b]
[c]
[d]
[e]
[f]
Dr. R. Zhang, Prof. Z. Zhang
Key Laboratory of Structure and Functional Regulation of Hybrid
Materials (Anhui University), Ministry of Education,
Hefei, Anhui 230601 (China)
Dr. B. Liu, Prof. Z. Zhang
Institute of Intelligent Machines
Chinese Academy of Sciences
Hefei, Anhui 230031 (China)
Dr. X. Han
School of Chemistry and Chemical Engineering
Anhui University of Technology
Ma’anshan, Anhui 243032 (China)
Prof. Y. Fu
Department of Chemistry
University of Science and Technology of China
Hefei, Anhui 230026 (China)
Dr. Z. Hu
Department of Physics, Chemistry and Biology
Linköping University
Here, we report a new class of super-resolution imaging probe
to address the aforementioned issues. We engineer one
fluorescent probe with triple spectrally distinctive switch-on
modes in response to different targets of interest. This one
molecule-three mode design completely avoids the difficulty of
Linköping 58183 (Sweden)
These authors contributed equally to this work.
+
[ ]
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.
1

10.1002/anie.202005959
Angewandte Chemie International Edition
RESEARCH ARTICLE
transporting three different probes to an identical intracellular
organelle, enabling the study of dynamic interactions between
multiple analytes. We demonstrate this concept through imaging
of nucleoproteins in mitochondria as well as the effect of an
upstream signaling molecule, hydrogen peroxide (H2O2).
probe fluorescence, but also their fluorescence is of different
colors (spectrally distinguishable: orange for H2O2, red for
proteins, and deep red for nucleic acids).
First, we characterized the fluorescence emission behaviors
of QVD-B in the presence of individual analytes in PBS buffer.
Upon the addition of H2O2, the non-fluorescent probe emitted
orange fluorescence at 580 nm under excitation of 405 nm (see
the excitation spectra in Figure S4a) and reached a 44-fold
fluorescence enhancement with 100 μM H2O2 (Figure 1a and
inset). The fluorescence intensity showed a linear relationship
against H2O2 concentration in the range of 5-50 μM (Figure S5).
As illustrated in Scheme 1a, H2O2 caused the cleavage of the
boronate group followed by decomposition of quinolinium to
eliminate the intramolecular rotation that drains the excitation
energy, leading to fluorescence emission.[8e,12] The reaction
product, QVD, was confirmed using 1H NMR, 13C NMR, and HR-
MS (Figure S6-S8).
Results and Discussion
b)
a)
c)
e)
1600
1200
800
400
0
600
400
200
0
H2O2
BSA
580 nm
632 nm
100
3.0
ex= 530 nm

ex= 405 nm
0 μM
0 mg/mL
600
700
800
900
500
600
700
800
900
Wavelength (nm)
Wavelength (nm)
d)
1000
800
600
400
200
0
800
600
400
200
0
RNA
DNA
688 nm
688 nm
100
100
ex= 580 nm
Scheme 1. Schematic illustrations for (a) the chemical structure of the
mitochondria-targeting probe and its three modes of fluorescence in response
to H2O2, protein binding, and nucleic acid (DNA and RNA) binding, and (b)
simultaneous imaging of H2O2, proteins and nucleic acids in a mitochondrion.

ex= 580 nm

0 μM
0 μM
600
700
800
900
600
700
800
900
Wavelength (nm)
Wavelength (nm)
For mitochondria-targeting, we designed
a
cationic
quinolinium-vinyl-N,N-dimethylaniline boronate derivative (QVD-
B) probe (Scheme 1a, see the synthetic route and
characterizations in Scheme S1 and Figure S1-S3, respectively).
In this structure, a benzeneboronate ester moiety is linked to a
800
I580 with ex=405 nm (for detection H2O2)
I632 with ex=530 nm (for detection protein)
I688 with ex=580 nm (for detection nucleic acid)
600
400
200
0
quinoline-vinyl-N,N-dimethylaniline (QVD) through
a single
carbon atom spacer to form a positively charged N-alkyl-
quinolinium. The as-synthesized probe is non-fluorescent due to
the intramolecular single-bond rotations that deplete the
excitation energy (indicated by the arrow in Scheme 1a).[12]
Moreover, the positively charged group endows QVD-B with the
capability of specific mitochondrial enrichment (Scheme 1b),
driven by the large negative mitochondrial transmembrane
electric potential (negative inside). Similar probe accumulation in
the mitochondria promoted by electrostatic interactions has been
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Figure 1. (a-d) Fluorescence spectra of the QVD-B probe upon addition of H2O2
(excitation 405 nm), BSA (excitation 530 nm), RNA (20 bp), and DNA (20 bp)
(excitation 580 nm), respectively. The probe concentration was fixed at 10 μM,
and the inset images were taken with a 365 nm UV lamp excitation. (e) The
selectivity of QVD-B (10 μM in PBS) responses to various analytes: 1. blank; 2.
H2O2; 3. Ca2+; 4. Mg2+; 5. 1O2; 6. O2•−; 7. HClO; 8. NO (NONOate); 9. ONOO−;
10. •OH (FeSO4 and H2O2); 11. cysteine; 12. serine; 13. lysine; 14. glutathione;
15. BSA; 16. HSA; 17. lysozyme; 18. hemoglobin; 19. trypsin; 20.
immunoglobulin G; 21. pepsin; 22. RNA; and 23. DNA. The sample
concentrations are 100 μM, except samples 15-21 (1 mg/mL). The error bars
represent the mean errors from the results of 5 measurements.
observed
previously
using
other
positively
charged
compounds.[8c,8f] The QVD-B molecule interacts with H2O2,
proteins and nucleic acids in three different modes: (i) chemical
reaction with H2O2 removes the boronate moiety, transforming the
quinolinium into a quinoline derivative (QVD);[8c,8e] (ii) the
hydrophobicity of QVD moiety (log P = 2.995) allows the probe to
bind with proteins with exposed hydrophobic pockets; (iii) the
positive charge and the -conjugates of QVD-B provide strong
interactions with negative-charged nucleic acids (DNA and RNA)
which leads to the condensation and aggregation of the probe.[7]
Remarkably, not only all three modes of interactions switch on the
In the presence of bovine serum albumin (BSA), the probe
instantly showed strong red fluorescence (emission peak at 632
nm) under 530 nm excitation (Figure 1b and inset). At a BSA
concentration of 3.0 mg/mL, the fluorescence enhancement
reached 582 folds. To investigate the switch-on mechanism, a
binding competition experiment was performed by the addition of
warfarin, a classic molecule widely used to examine the
2
This article is protected by copyright. All rights reserved.

10.1002/anie.202005959
Angewandte Chemie International Edition
RESEARCH ARTICLE
interaction with BSA.[13] It has been well documented that warfarin
binds to BSA through hydrophobic-hydrophobic interactions
because BSA molecules have hydrophobic pockets exposed on
the surface.[14] The addition of warfarin into the probe-BSA
complex resulted in a continuous reduction of the probe
fluorescence at 632 nm because the probe was displaced by
warfarin (Figure S9). The binding stoichiometry between the
probe and BSA was close to 1:1 (molar ratio) as shown in the
Job's plot diagram (Figure S10). These results suggest that i) the
switch-on red fluorescence is due to spatial confinement of the
boronate moiety which restricts the energy-dissipating
intramolecular rotation (also see the molecular docking in Figure
S11), and ii) BSA molecules have a hydrophobic pocket on the
surface accessible to QVD-B. In contrast, the addition of proteins
without exposed hydrophobic pockets did not switch on the
fluorescent emission of QVD-B (see the Selectivity Test in Figure
1e).
In the case of nucleic acids (single-strand RNA or double-
strand DNA, main forms of nucleic acids in cells), the probe
solution also lit up, but the emission peak exhibited a large
bathochromically shift to 688 nm (deep red under excitation light
of 580 nm, Figure 1c and d). At the concentration of 100 M
single-strand RNA, the maximal fluorescent enhancement was as
high as 172 folds. The spectral red-shift suggests a fluorescence
enhancement mechanism different from those for H2O2 and BSA.
Because of the positive charge and the -conjugation, QVD-B has
strong affinities to negative-charged nucleic acids. Electrostatic
interactions lead to the condensation and aggregation of QVD-B
along nucleic acid strands. UV measurements revealed that the
clusters have a molar ratio of 1:7 between the nucleic acid chain
(20 bp) and QVD-B (See item 2 in Supporting Experimental). To
better understand the aggregation-induced emission red-shift, we
measured the fluorescence spectra of a serial of QVD-B solutions
(Figure S12). Increasing concentration of QVD-B resulted in
larger red-shift, similar to the formation of J-aggregate crystals for
organic dye molecules.[15]
showed the high selectivity towards H2O2, proteins with exposed
hydrophobic pockets, and nucleic acids.
To investigate whether probe binding to proteins and nucleic
acids affect cell normal physiology, we evaluated the cell viability,
adenosine triphosphate production, and mitochondrial membrane
potential (Figure S14-S16), important indicators of cell health. The
results showed that QVD-B has virtually no effect on cell
physiology.
Next, we proceeded with triple-color imaging of live cells: (1)
H2O2 channel: excitation at 405 nm, collection in 500-560 nm; (2)
protein channel: excitation at 530 nm, collection in 550-590 nm;
(3) nucleic acid channel: excitation at 610 nm, collection in 710-
770 nm. Note that the nucleic acid channel excitation was set to
610 nm rather than the optimal wavelength of 580 nm to avoid
overlap with the protein channel emission band (Figure S17).
Fortunately, the QVD-B probe has a sufficiently strong absorption
at 610 nm in the presence of nucleic acids (Figure S18). It is also
worth mentioning that the product QVD resulting from the H2O2
reaction could also enter proteins’ hydrophobic pockets as well as
other hydrophobic environments, but the QVD emission blue-
shifts significantly under 405 nm excitation (Figure S4c-e), further
minimizing spectral crosstalk. More importantly, the competition
experiments certified that the binding ability of QVD-B with BSA
was prior to that of QVD (Figure S19), and thereby QVD had no
interference in protein and nucleic acid imaging.
Interestingly, the QVD-B emissions in response to H2O2, BSA,
and nucleic acids are spectrally separated by approximately 50
nm (Figure 1a-d). To directly demonstrate simultaneous imaging,
we measured QVD-B fluorescence in the mixture of H2O2, BSA,
and RNA. Under excitations of 405, 530 and 580 nm (Figure S13),
the emissions peaks centered at 580, 632 and 688 nm were easily
detected, respectively, and the spectral profiles were the same as
solutions containing single analytes (Figure 1a-d), proving that
H2O2, proteins, and nucleic acids can be simultaneously imaged
by our single probe.
We further tested the probe selectivity before moving to
cellular imaging. In the selectivity test, we added multiple
molecules commonly found in live cells to QVD-B buffer. The
cellular compounds included inorganic salts, ROS, nucleic acids,
amino acids, peptides, and proteins/enzymes (Figure 1e). Under
the excitation of 405 nm, only H2O2 induced strong emission at
580 nm; whereas under 530 nm excitation, only proteins with
exposed hydrophobic pockets such as BSA and human serum
albumin (HSA) switched on fluorescence at 632 nm. In
comparison, only DNA and RNA induced fluorescence
enhancement at 688 nm under light excitation of 580 nm. In
addition, we also investigated the spectral behavior of QVD-B in
Figure 2. Intracellular imaging by using QVD-B. (a) Hep G2 cells were
incubated with 10 μM QVD-B for 30 min. (b) Negative control adding
dithiothreitol (DTT) and nucleases to cells. Upper: Hep G2 cells were first
treated with 100 M DTT for 3 h, then incubated with QVD-B for 30 min. Bottom:
Hep G2 cells were incubated with 10 μM QVD-B for 30 min, fixed with
paraformaldehyde and permeabilized with Triton X-100, and further incubated
with nucleases (10 mg/mL) for 2 h.
After incubation with QVD-B, Hep G2 cells did not show
fluorescent signals in the H2O2 channel due to the extremely low
level of endogenous H2O2 in cells (Figure 2a). However, the
signals for proteins and nucleic acids were readily detected using
confocal microscopy. To confirm the origins of switched-on
fluorescence, negative control experiments were carried out using
dithiothreitol (DTT) and nucleases, which have proven abilities in
degrading protein structures[16] and nucleic acids,[17] respectively.
As shown in Figure 2b, DTT completely suppressed fluorescence
from the protein channel, but the fluorescence from the nucleic
acid channel remained. Similarly, nucleases led to the elimination
of fluorescence in the nucleic acid channel, whereas the
cell
membrane-mimicking
environments
such
as
dodecylphosphocholine (DPC) micelles and ethyl oleate (Figure
S4b). No switch-on fluorescence was observed for QVD-B in the
presence of these phospholipid structures. Combined, the probe
3
This article is protected by copyright. All rights reserved.

10.1002/anie.202005959
Angewandte Chemie International Edition
RESEARCH ARTICLE
fluorescence from the protein channel remained. The above
imaging results indicate that the two switched-on fluorescent
signals are indeed from proteins and nucleic acids.
experiment, we treated the cells with exogenous QVD and found
that QVD only accumulated in hydrophobic lipid droplets in the
cytosol (Figure S24).
To characterize the intracellular distribution of QVD-B, Hep
G2 cells were treated with phorbol 12-myristate 13-acetate (PMA)
to stimulate the production of endogenous H2O2.[8a,8b] As a result,
fluorescence signals from all three channels were switched on
(Figure 3a-d, see the negative control experiment by adding H2O2
scavenger in Figure S20). Interestingly, the images obtained from
the three channels appear to have identical localization patterns
in cells. To understand the unique co-localization pattern, a
commercially available spectrally distinguishable mitochondria-
tracker was added to the cell imaging experiment (Figure 3e-h),
and the fluorescence signals from the three channels almost
completely overlapped with the mitochondrial tracker (overlap
coefficients: 0.90 for H2O2, 0.93 for protein, and 0.90 for nucleic
acid). Similar results were also observed in two additional cell
lines (Figure S21 and S22). These results show that the QVD-B
probe is specific for imaging of H2O2, proteins and nucleic acids
in mitochondria.
To characterize the intracellular distribution with higher
precision, super-resolution nanoscopy is needed. Due to the large
Stokes shifts (Figure 1a-d) and photostability (Figure S25), the
QVD-B probe not only minimizes the autofluorescence
background, but also helps realize multicolor stimulated emission
depletion (STED) imaging.[11] Using standard confocal
microscopy, the imaging resolution was insufficient to reach the
sub-organelle scale (Figure 4a). STED, in comparison, revealed
that the proteins with exposed hydrophobic pockets were mainly
situated at the mitochondrial cristae (Figure 4b and c).[18] It
appears that the probe first concentrated in the mitochondrial
matrix followed by interactions with protein hydrophobic pockets,
nucleic acids, and H2O2. The co-localization experiment using a
mitochondrial inner-membrane dye was carried out on the STED
nanoscope.[19] Super-resolution images revealed a major overlap
between the mitochondrial inner-membrane and the protein
channel (Figure S26), suggesting that these proteins were bound
to the mitochondrial inner-membrane.
To further confirm the charge-induced enrichment of QVD-B
in mitochondria, a well-established dissipating mitochondrial-
membrane
potential
reagent,
carbonyl
cyanide
4-
(trifluoromethoxy) phenylhydrazone (FCCP) was employed.
When first treated with FCCP followed by treatment with QVD-B,
no appreciable fluorescence was detected in cells (Figure S23),
whereas reversing the treatment sequence (QVD-B then FCCP)
led to strong fluorescence emissions from the probe. These
results strongly suggest that QVD-B accumulation in the
mitochondria was associated with the probe’s positive charges.
Unlike conventional mitochondrial dyes, QVD-B stayed in the
mitochondria after the transmembrane potential was dissipated
by FCCP, due to its strong interactions with proteins' hydrophobic
pockets and nucleic acids.
a)
b)
c)
d)
10 μm
H2O2
protein
nucleic acid
BF
e)
f)
g)
h)
mito-tracker
overlay (b+e)
overlay (d+e)
overlay (c+e)
Figure 3. Multicolor confocal imaging of H2O2, proteins and nucleic acids in
mitochondria. Hep G2 cells were incubated with QVD-B (10 M) for 30 min,
followed by the incubations with PMA (2 μg/mL) to produce more H2O2, and with
Mito Tracker Green (2 M). (a) Bright-field image. (b-d) Fluorescent images
from the H2O2, protein, and nucleic acid channels, respectively. (e) Fluorescent
image of Mito-tracker (excitation at 490 nm, collection in 500-520 nm). (f-h)
Overlays of (b), (c) and (d) with (e), respectively.
Although H2O2 stimulated by PMA may diffuse throughout the
entire cell, the fluorescence was only seen in the mitochondria
likely due to the high concentration of QVD-B in the mitochondria
and confinement of the resulting hydrophobic QVD (does not
diffuse beyond the barrier of the mitochondrial membrane).
Therefore, we acknowledge a potential limitation of QVD that it
only indicates the quantity of mitochondrial H2O2. To show the
absence of QVD leakage from mitochondria, in a separate
Figure 4. (a) Conventional confocal microscopy, and (b, c) STED super-
resolution nanoscopy of H2O2, proteins and nucleic acids in mitochondria. In (a)
and (b), Hep G2 cells were incubated with QVD-B for 30 min. In (c), Hep G2
cells were first treated with PMA for 2 h to promote H2O2 production, followed
by incubation with the probe for 30 min.
4
This article is protected by copyright. All rights reserved.

10.1002/anie.202005959
Angewandte Chemie International Edition
RESEARCH ARTICLE
Figure 5. STED super-resolution nanoscopy of mitochondrial H2O2, proteins, and nucleic acids in live Hep G2 cells. (a) Fluorescent signals from proteins and nucleic
acids upon the addition of exogenous H2O2 at various doses. (b) Correlation plots between the overlap coefficients (proteins and nucleic acids) and the H2O2 doses
in (a). (c) Fluorescent signals from proteins and nucleic acids upon the addition of PMA of various doses. (d) The correlation plots between the overlap coefficients
and PMA doses in (c). (e) STED imaging after cells sequentially treated with H2O2 scavenger (NAC) and PMA. (f) Fluorescence intensities of the H2O2 channel, and
the overlap coefficients of proteins and nucleic acids under different treatments. (g) Schematic drawings of H2O2 regulating the dissociation of nucleoprotein
complexes. The fluorescent intensities were obtained from 6 images with an identical observation area. The error bars represent the standard deviation (±SD). All
experiments were repeated with 3 batches of cells.
The cristae are the invagination of the inner membrane, and
exist in the form of crista groups in mitochondria, providing a large
surface area hosting functional biomolecules. The distance
between adjacent cristae groups was about 300 nm (the bottom
panels of Figure 4b), in agreement with prior reports.[20] Most
interestingly, the super-resolution images showed significant
overlap between the proteins and nucleic acids with an overlap
coefficient as high as 0.67, suggesting the formation of nucleic
and d). To show the dynamic process of nucleoprotein regulated
by H2O2, we then included N-acetyl-cysteine (NAC), a scavenger
of intracellular H2O2, in the control imaging experiments. When
cells were sequentially treated with NAC, PMA and QVD-B probe,
the fluorescent signal from H2O2 became much weaker than that
without NAC treatment. Meanwhile, the overlap between proteins
and nucleic acids kept virtually the same as no PMA stimulation
(Figure 5e and f). The same results were also confirmed on other
cell lines (Figure S27 and S28). Together all, we have
acid-protein
complexes
known
as
nucleoproteins.[21]
Nucleoproteins are highly dynamic structures. The nucleic acids
and nucleic acid-binding proteins can associate and dissociate in
response to cell needs and external stimulations to realize diverse
biological functionalities.[2c,3,21a,22]
demonstrated that H2O2 plays a critical role to regulate
mitochondrial nucleoprotein dynamics (Figure 5g).
Conclusion
Lastly, we studied the effect of exogenous and endogenous
H2O2 in regulating nucleoprotein dissociation using STED (Figure
5). Here, we directly treated cells with H2O2 of different
concentrations, followed by incubation with the probe QVD-B.
With increasing doses of H2O2, the overlap between the protein
and nucleic acid gradually decreased (Figure 5a). For example,
at the H2O2 concentration of 25 M, the overlap coefficient sharply
decreased from 0.67 to 0.53, suggesting that even a low level of
H2O2 can significantly impact nucleoprotein complexes (Figure
5b). The effect was similar to promoting the production of
endogenous H2O2 in the mitochondria (e.g., using PMA, Figure 5c
In summary, we have developed an intracellular organelle-
specific probe with three distinctive switch-on fluorescence
modes for simultaneous and discriminative imaging of H2O2,
proteins with surface hydrophobic pockets, and nucleic acids.
This probe is also suitable for STED super-resolution nanoscopy.
Together, they revealed that H2O2 is a critical upstream signaling
transductor to regulate the disassociation of nucleoproteins in
mitochondria. This imaging probe can become a useful tool for
studies of mitochondrial biology and inspire the design of other
Swiss army knife-type sensors for other intracellular organelles.
5
This article is protected by copyright. All rights reserved.

10.1002/anie.202005959
Angewandte Chemie International Edition
RESEARCH ARTICLE
[16] Q. Sun, W. Wang, Z. Chen, Y. Yao, W. Zhang, L. Duan, J. Qian, Chem.
Commun. 2017, 53, 6432−6435.
Acknowledgments
[17] M. D. Miller, J. Tanner, M. Alpaugh, M. J. Benedik, K. L. Krause, Nat.
Struct. Biol. 1994, 1, 461−468.
This work was supported by the National Natural Science
Foundation of China (21775001, 21705001, 21974001 and
21976183), Anhui Provincial Natural Science Foundation of China
(1808085MB32), and Hefei Physical Science Center Foundation
(2017FXZY001).
[18] (a) S. Jakobs, C. A. Wurm, Curr. Opin. Chem. Biol. 2014, 20, 9−15; (b)
B. G. Kopek, G. Shtengel, C. S. Xu, D. A. Clayton, H. F. Hess, Proc. Natl.
Acad. Sci. U. S. A. 2012, 109, 6136−6141.
[19] X. Huang, J. Fan, L. Li, H. Liu, R. Wu, Y. Wu, L. Tang, Nat. Biotechnol.
2018, 36, 451−459.
[20] T. Stephan, A. Roesch, D. Riedel, S. Jakobs, Sci. Rep. 2019, 9, 1−6.
[21] (a) N. Kunová, G. Ondrovičová, J. A. Bauer, J. Bellová, Ľ. Ambro, L.
Martináková, V. Pevala, Sci. Rep. 2017, 7, 631−643; (b) J. N. Spelbrink,
F. Y. Li, V. Tiranti, K. Nikali, Q. P. Yuan, M. Tariq, L. Santoro, Nat. Genet.
2001, 28, 223−231.
Keywords: super-resolution • nucleic acids • proteins • imaging
probe • reactive oxygen species
[1]
(a) S. G. Rhee, S. W. Kang, W. Jeong, T. S. Chang, K. S. Yang, H. A.
Woo, Curr. Opin. Cell Biol. 2005, 17, 183−189; (b) A. C. Nulton-Persson,
L. I. Szweda, J. Biol. Chem. 2001, 276, 23357−23361; (c) N. R. Sims, M.
F. Anderson, L. M. Hobbs, J. Y. Kong, S. Phillips, J. A. Powell, E. Zaidan,
Brain Res. Mol. Brain Res. 2000, 77, 176−184.
[22] (a) A. Choumar, A. Tarhuni, P. Lettéron, F. Reyl-Desmars, N. Dauhoo, J.
Damasse, A. Mansouri, Antioxid. Redox Signal. 2011, 15, 2837−2854;
(b) T. Ozawa, Y. Natori, M. Sato, Y. Umezawa, Nat. Methods 2007, 4,
413−419.
[2]
(a) N. G. Larsson, J. Wang, H. Wilhelmsson, A. Oldfors, P. Rustin, M.
Lewandoski, D. A. Clayton, Nat. Genet. 1998, 18, 231−236; (b) A. Bratic,
N. G. Larsson, J. Clin. Invest. 2013, 123, 951−957; (c) J. He, H. M.
Cooper, A. Reyes, M. Di Re, H. Sembongi, T. R. Litwin, J. E. Walker,
Nucleic Acids Res. 2012, 40, 6109−6121.
[3]
[4]
[5]
(a) O. Rackham, T. R. Mercer, A. Filipovska, Wiley Interdiscip. Rev. RNA
2012, 3, 675−695; (b) X. J. Chen, X. Wang, B. A. Kaufman, R. A. Butow,
Science 2005, 307, 714−717.
(a) M. Falkenberg, N. G. Larsson, C. M. Gustafsson, Annu. Rev.
Biochem. 2007, 76, 679−699; (b) D. A. Clayton, Cell 1982, 28, 693−705;
(c) D. A. Clayton, Int. Rev. Cytol. 1992, 141, 217−232.
(a) S. Sreedharan, M. R. Gill, E. Garcia, H. K. Saeed, D. Robinson, A.
Byrne, J. Bernardino de la Serna, J. A. Thomas, J. Am. Chem. Soc. 2017,
139, 15907−15913; (b) E. Betzig, G. H. Patterson, R. Sougrat, O. W.
Lindwasser, S. Olenych, J. S. Bonifacino, H. F. Hess, Science 2006, 313,
1642−1645. (c) C. Wang, M. Taki, Y, Sato, Y. Tamura, H. Yaginuma, Y.
Okada, S. Yamaguchi, Proc. Natl. Acad. Sci. U. S. A. 2019, 116,
15817−15822.
[6]
(a) S. Wang, M. Ding, X. Chen, L. Chang, Y. Sun, Biomed. Opt. Express
2017, 8, 3119−3131; (b) M. Ishigaki, M. Iketani, M. Sugaya, M. Takahashi,
M. Tanaka, S. Hattori, I. Ohsawa, Mitochondrion 2016, 28, 79−87.
G. Han, J. Zhao, R. Zhang, X. Tian, Z. Liu, A. Wang, Z. Zhang, Angew.
Chem., Int. Ed. 2019, 58, 7087−7091.
[7]
[8]
(a) D. Srikun, E. W. Miller, D. W. Domaille, C. J. Chang, J. Am. Chem.
Soc. 2008, 130, 4596−4597; (b) M. Abo, Y. Urano, K. Hanaoka, T. Terai,
T. Komatsu, T. Nagano, J. Am. Chem. Soc. 2011, 133, 10629−10637; (c)
Y. Shen, X. Zhang, Y. Zhang, Y. Wu, C. Zhang, Y. Chen, H. Li, Sens.
Actuators B Chem. 2018, 255, 42−48; (d) R. Zhang, J. Zhao, G. Han, Z.
Liu, C. Liu, C. Zhang, B. Liu, C. Jiang, R. Liu, T. Zhao, M. Y. Han, Z.
Zhang, J. Am. Chem. Soc. 2016, 138, 3769−3778; (e) F. Hu, Y. Huang,
G. Zhang, R. Zhao, D. Zhang, Tetrahedron Lett. 2014, 55, 1471−1474.
(f) N. Kumar, V. Bhalla, M. Kumar, Chem. Commun. 2015, 51,
15614−15628.
[9]
(a) X. Wang, P. Li, Q. Ding, C. Wu, W. Zhang, B. Tang, J. Am. Chem.
Soc. 2019, 141, 2061−2068; (b) P. Ou, R. Zhang, Z. Liu, X. Tian, G. Han,
B. Liu, Z. Zhang, Angew. Chem., Int. Ed. 2019, 58, 2261−2265; (c) Z. Liu,
J. Zhao, R. Zhang, G. Han, C. Zhang, B. Liu, Z. Zhang, M. Y. Han, X.
Gao, ACS Nano 2018, 12, 3629−3637.
[10] (a) K. Hanaoka, K. Kikuchi, H. Kojima, Y. Urano, T. Nagano, Angew.
Chem., Int. Ed. 2003, 42, 2996−2999; (b) Y. Li, X. Zhang, B. Zhu, J. Xue,
Z. Zhu, W. Tan, Analyst 2011, 136, 1124−1128; (c) A. R. Lippert, E. J.
New, C. J. Chang, J. Am. Chem. Soc. 2011, 133, 10078−10080.
[11] G. Vicidomini, P. Bianchini, A. Diaspro, Nat. Methods 2018, 15, 173−182.
[12] Y. Hong, J. W. Lam, B. Z. Tang, Chem. Commun. 2009, 29, 4332−4353.
[13] X. L. Jin, X. Wei, F. M. Qi, S. S. Yu, B. Zhou, S. Bai, Org. Biomol.
Chem. 2012, 10, 3424−3431.
[14] C. A. Lelis, G. M. D. Ferreira, G. M. D. Ferreira, M. do Carmo Hespanhol,
M. S. Pinto, L. H. M. da Silva, A. C. dos Santos Pires, Food
Hydrocoll. 2017, 70, 29−35.
[15] (a) M. Reers, T. W. Smith, L. B. Chen, Biochemistry 1991, 30,
4480−4486; (b) V. Grande, B. Soberats, S. Herbst, V. Stepanenko, F.
Würthner, Chem. Sci. 2018, 9, 6904−6911.
6
This article is protected by copyright. All rights reserved.

10.1002/anie.202005959
Angewandte Chemie International Edition
RESEARCH ARTICLE
Entry for the Table of Contents
A mitochondria-enriched probe switches on in three different fluorescence modes in response to H2O2, protein, and nucleic acid, enabling the
visualization of mitochondrial nucleoprotein dynamics. Super-resolution imaging reveals that the proteins localize at mitochondrial cristae and
largely fuse with nucleic acids to form nucleoproteins, whereas increasing H2O2 level leads to the disassociation of nucleoproteins.
This article is protected by copyright. All rights reserved.
7