Rational Design of Specific Recognition Molecules for Simultaneously Monitoring of Endogenous Polysulfide and Hydrogen Sulfide in the Mouse Brain

Article abstract of DOI:10.1002/anie.201907210

A biosensor was created for the simultaneous monitoring of endogenous H₂S_n and H₂S in mouse brains and exploring their roles in activation of the TRPA1 channel under two types of brain disease models: ischemia and Alzheimer's disease (AD). Based on DFT calculations and electrochemical measurements, two probes, 3,4-bis((2-fluoro-5-nitrobenzoyl)oxy)-benzoic acid (M_PS-1) and N-(4-(2,5-dinitrophenoxy) phenyl)-5-(1, 2-dithiolan-3-yl)pentanamide (M_HS-1), were synthesized for specific recognition of H₂S_n and H₂S. Through co-assembly of the two probes at the mesoporous gold film with good anti-biofouling ability and electrocatalytic activity, this microsensor showed high selectivity for H₂S_n and H₂S against potential biological interferences. The biosensor can simultaneously determine the concentration of H₂S_n from 0.2 to 50 μm, as well as that of H₂S from 0.2 to 40 μm. The expression of TRPA1 protein positively correlated with levels of H₂S_n under both ischemia and AD.

Full text of DOI:10.1002/anie.201907210

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Rational Design of Specific Recognition Molecules for
Simultaneously Monitoring of Endogenous Polysulfide and
Hydrogen Sulfide in the Mouse Brain.
Authors: Hui Dong, Qi Zhou, Limin Zhang, and Yang Tian
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.201907210
Angew. Chem. 10.1002/ange.201907210
Link to VoR: http://dx.doi.org/10.1002/anie.201907210
http://dx.doi.org/10.1002/ange.201907210

10.1002/anie.201907210
Angewandte Chemie International Edition
COMMUNICATION
Rational Design of Specific Recognition Molecules for Simultaneously
Monitoring of Endogenous Polysulfide and Hydrogen Sulfide in the
Mouse Brain
Hui Dong, Qi Zhou, Limin Zhang,* and Yang Tian*
Abstract: The correlation and distinction between H2Sn and H2S playing in
brain function remain the subject of considerable debate. Herein, an
efficient biosensor was created for the simultaneous monitoring of
endogenous H2Sn and H2S in mouse brains and exploring their roles in
activation of the TRPA1 channel under two types of brain disease models-
ischemia and Alzheimer’s disease (AD). Integrated with density functional
theory (DFT) calculation and electrochemical measurements, the novel
probes, 3,4-bis((2-fluoro-5-nitrobenzoyl)oxy)-benzoic acid (MPS-1) and N-
inert, and generates a Faradic signal beyond the potential window of
water splitting. This characteristic leads to great difficulty in direct
determination of H2Sn in aqueous solution, depending on the Faradic
signal of H2Sn. On the other hand, H2S in principle can be oxidized at
the working electrode with the transfer of two electrons (H2S →
S+2H++2e-, ES/H2S at 0.144 V (vs NHE, pH 7.4) at a high overpotential.
As a result, the detection of H2S often suffers from great interferences
of other electroactive species in the central nervous system.7 Although
H2S could be electrochemically oxidized at special metal electrodes
such as Ni, V2O5 at low potentials, the electrochemical reaction
generally occurs in acid (pH <2) or basic media (pH >12).7 Recently,
Zhang et al. developed an electrochemical sensor for in vivo
monitoring of H2S at a low potential of -0.19 V (vs. Ag/AgCl) in the
brain by applying ammineruthenium (III) to catalyze the
electrochemical oxidation of H2S in a neutral solution.8 However, a key
issue about electrode contamination should be carefully considered
based on the electrochemical oxidation of H2S because the oxidation
product of H2S, sulfur (S), is strongly adsorbed onto the electrode
surface to poison the electrode, resulting in low reproducibility and big
determination error. Therefore, it still renders a great challenge to
design an efficient sensing strategy to simultaneous assay of H2Sn and
H2S without cross-talk and electrode contamination.
(4-(2,5-dinitrophenoxy) phenyl)-5-(1, 2-dithiolan-3-yl)pentanamide (MHS-1
)
were rationally designed and synthesized for specific recognition of H2Sn
and H2S. Through co-assembly of the two probes at the mesoporous gold
film (mAu) with good anti-biofouling ability and high electrocatalytic
activity, this single microsensor showed high selectivity for H2Sn and H2S
against potential biological interferences including sulfur-containing
species. The present biosensor can simultaneously determine the
concentration of H2Sn from 0.2 to 50 μM, as well as that of H2S from 0.2 to
40 μM. Finally, using this useful tool, the expression of TRPA1 protein was
found to be positively correlated with the levels of H2Sn under both
ischemia and AD.
The correlation and distinction between hydrogen sulfide (H2S),
and its potential oxidized forms, hydrogen polysulfide (H2Sn, n>1), in
physiological and pathological processes in the brain remain the
subject of considerable debate.1 In the past decades, H2S has been
extensively studied as a signalling molecule.2 However, recent
investigation has revealed that H2Sn might act as the real regulators in
cell signalling transduction. Some biological activities initially
attributed to H2S were hypothesized to be mediated by H2Sn.3 An
important example is protein S-sulfhydration, which has proven to be a
major signalling pathway involving sulfur. From a chemical reactivity
point-of-view, H2S cannot simply “persulfidate” a thiol group. By
contrast, H2Sn possesses stronger oxidation ability than H2S. Thus,
H2Sn should be more active in S-sulfydration than H2S.4 A few studies
have given the support that H2Sn is the actual signalling molecule in
protein S-sulfhydration.4 Unfortunately, the lack of efficient strategies
for simultaneous detection of H2Sn and H2S in the brain has been a
bottleneck to better understanding of the pathophysiological roles of
H2Sn and H2S.
Motivated by the above challenges, we rationally designed and
synthesized the specific recognition molecules for H2Sn and H2S, 3,4-
bis((2-fluoro-5-nitrobenzoyl)oxy)-benzoic acid (MPS-1) and N-(4-(2,5-
dinitro-phenoxy) phenyl)-5-(1,2-dithiolan-3-yl) pentanamide (MHS-1
)
(Scheme 1) by integrating density functional theory (DFT) calculation
with electrochemical measurements. Through co-assembly of MPS-1 and
MHS-1 molecules at the microelectrode decorated by mesoporous gold
film (mAu), we successfully evaluated the levels of H2Sn and H2S in
To date, only a few fluorescent probes have been reported for the
determination of H2Sn.4a, 5 However, to the best of our knowledge, no
efficient approach has been reported for in vivo determination of H2Sn.
In vivo electrochemical analysis of chemical signals in the brain using
implanted microsensors is a vital way to study brain function due to
high temporal and spatial resolutions.6 But H2Sn is electrochemically
[*]
Hui Dong, Qi Zhou, Prof. Limin Zhang, Prof. Yang Tian
Shanghai Key Laboratory of Green Chemistry and Chemical
Processes, School of Chemistry and Molecular Engineering, East
China Normal University, Dongchuan Road 500, Shanghai 200241
(China)
Scheme 1. (A) Probes designed in the proposed sensing strategy. (B)
Working principle of MPS-1 and MHS-2 for specific recognition of H2S and H2Sn.
(C) Simultaneous assaying of H2S and H2Sn in a live mouse brain and further
investigation of the activation of the TRPA1 channel by H2Sn and H2S.
E-mail: ytian@chem.ecnu.edu.cn; lmzhang@chem.ecnu.edu.cn
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.

10.1002/anie.201907210
Angewandte Chemie International Edition
COMMUNICATION
energies than their oxidized states, which suggested that they were
redox characteristics. With comparison of the Gibbs free energy (ΔG0)
and oxidation potential (E0), MHS-1 and MPS-1 were optimized for
simultaneous detection of H2Sn and H2S because the Faradic signals of
PHQ (E0 = 0.16 V) and AP (E0 = 0.32 V) were well separated
(Supporting Information, Table S1).
This design concept was then confirmed by a series of experiments.
All the probes were synthesized and characterized (Figure S1-S4). The
sensing strategy was based on the typical proceeding chemical
reactions. FTIR spectroscopy and LC-MS were used to confirm the
chemical products of four probes with targets. Combined with the peak
at ~3273 cm-1 (-OH group), new m/z peaks of 154.03 and 138.03
confirmed that the chemical products of MPS-1 and MPS-2 with H2Sn
were PHQ and PN (Figure S5-S7), respectively. Similarly, the products
of MHS-1 and MHS-2 with H2S, AP and NP, were confirmed by the m/z
peaks of 298.09 and 336.07 (Figure S8-S9). According to the results,
the chemical reactions of four probes were summarized (Scheme S2),
and the corresponding equilibrium constants (k) were estimated (Figure
S10). The four designed probes showed similar k values, which
provided a similar response time during simultaneous determination.
The electrochemical behaviors of all the probes were investigated by
cyclic voltammetry (CV). MPS-1 and MPS-2 were assembled at
cysteamine (Cyst)-decorated CFME/mAu to form CFME/mAu/MPS-1
and CFME/mAu/MPS-2, respectively. Meanwhile, MHS-1 and MHS-2 were
modified at CFME/Au to fabricate CFME/Au/MHS-1 and
CFME/Au/MHS-2, respectively. Each modification step was recorded by
CV and XPS (Figure S11). No obvious peak was obtained at
CFME/mAu and CFME/mAu/MPS-1 in blank aCSF. However, a pair of
redox peaks with a half-wave potential (E1/2) of ~ 0.25 ± 0.03 V were
observed at CFME/mAu/MPS-1 in aCSF containing 30 μM Na2S2
(Figure 2I). This peak was ascribed to the redox process of the formed
well-separated peak potentials, 3, 4-bis((2-fluoro-5-nitrobenzoyl)-
oxy)benzoic acid (MPS-1), and N-(4-(2,5-dinitrophenoxy phenyl)-5
Figure 1. (A) SEM image of CFME coated by mAu. (B-C) Top-surface images,
(D) X-ray spectroscopy analysis, (E-F) High-resolution TEM images, (G) XRD
pattern, and (H) Atomic force microscopy (AFM) image of the mAu.
the mouse brains with ischemia and AD, and further explored the
relationship between the activation of transient receptor potential
ankyrin 1 (TRPA1) channel and levels of H2Sn and H2S.
As a starting point for this work, mAu was in-situ synthesized at
carbon fibre microelectrode (CFME) (Figure 1A).9 Mesopores with
averaged diameter of 11 ± 7 nm were distributed in the entire area
(Figure 1B-1D). Moreover, several crystal domains were connected
continuously to create a concave surface, which exposed the high index
plane (Figure 1E). The lattice spacing of 0.235 nm was attributed to the
Au (111) plane of fcc crystal, suggesting the formation of crystalline
Au at CFME (Figure 1F-1G) with an averaged pore depth of 24 ± 1 nm,
as indicated by AFM (Figure 1H).
H2Sn is a combination of polysulfide species. The dissolution of any
polysulfide salts results in a similar distribution of these species. H2S2
is a typical active species of H2Sn, and remains in dynamic equilibrium
with other H2Sn.10 Thus, H2S2 was taken as a representative target for
H2Sn detection. The compound containing bis-electrophilic groups was
envisoned to selectively capture H2S2. One of the electrophilic groups
was designed to be a latent electroactive group and was released under
nucleophilic reaction. Based on this strategy, MPS-1 and 4-((2-fluoro-5-
nitrobenzoyl)oxy)benzoic acid (MPS-2) were designed (Scheme 1A).
They underwent nucleophilic cyclization to form benzodithiolone
products after reacted with H2S2, resulting in electroactive pyrocatechol
(PHQ) released from MPS-1, as well as phenol (PN) from MPS-2 (Figure
2E-2F). Additionally, MHS-1 and N-(3,4-bis(2,5-dinitrophenoxy)-
phenyl)-5-(1,2-dithiolan-3-yl)pentanamide (MHS-2), were designed for
recognition of H2S based on the mechanism in which thiolysis of the
dinitrophenyl ether reaction was chemoselective for H2S.11 A -NH
group was devised at the 4’-site of the phenol derivatives to increase
the electron cloudy density of the aromatic ring to tune the response
signal for H2S at a relatively low potential. Once MHS-1 and MHS-2
reacted with H2S, nucleophilic substitution occurred, and 5-(1,2-
dithiolan-3-yl)-N-(4-hydroxy phenyl) pentanamide (AP) and N-(3,4-
dihydroxyphenyl)-5-(1,2-dithiolan-3-yl) pentanamide (NP) were
released from MHS-1 and MHS-2, respectively. The DFT calculation was
performed to rationalize the design of the electrochemical probes. For
the oxidation of PHQ, the prevailing mechanism consisting of a 2e--
2H+ process was considered. The higher HOMO energy of PHQ than
its oxidized state represented a typical characteristic of redox couples
(Figure 2A). In contrast, the oxidation process of PN was deduced to
be irreversible because the PN radical cation generated by PN
spontaneously convert to PHQ due to the small HOMO-LUMO energy
gap (0.6 eV) (Figure 2B, 2F). For the specific probes for H2S, the
chemical products, AP and NP, possessed higher HOMO and LOMO
PHQ moiety at CFME/mAu/MPS-1
. In comparison, only one
irreversible anodic peak with EP of 0.61 V was obtained at CFME/-
mAu/MPS-2 due to the oxidation of PN (Figure 2J). However, this peak
rapidly decreased to the basal level during the second cycle of scanning.
This unstable Faradic response was hard to apply in the brain. On the
other hand, a couple of redox peaks with good stability were both
observed at CFME/mAu/MHS-1 and CFME/mAu/MHS-2 after the
addition of 30 μM H2S, which were attributed to the redox processes of
AP (E1/2 = 0.35 V) and NP moieties (E1/2 = 0.08 V) (Figure 2K, 2L).
The heterogeneous electron transfer constants (ks) of four generated
electroactive probes at electrodes were estimated by using the Laviron
method (Figure S12, and Table S2). Larger ks values of PHQ (5.06 ±
0.12 s-1) and AP (4.91 ± 0.17 s-1) indicate their faster kinetic processes
than those of PN (0.52 ± 0.30 s-1) and NP (4.17 ± 0.09 s-1).
Accordingly, it was rational to select MPS-1 and MHS-1 as the probes to
construct a single sensor for simultaneous detection of H2Sn and H2S
because of well-separated Faradic signals, good stability, and fast
electrochemical kinetic processes.
Subsequently, Cyst and MHS-1 were co-assembled at CFME/mAu to
form CFME/mAu/Cyst+MHS-1. MPS-1 was further confined to result in
CFME/mAu/MPS-1+MHS-1. The surface coverages of MPS-1 and MHS-1
were estimated to be 31.8 pmol cm-2 for MPS-1, 21.6 pmol cm-2 for MHS-
1. Each modification step was characterized by differential pulse
voltammetry (DPV) because of its high sensitivity (Figure S13). No
Faradic current was observed at CFME/mAu, CFME/mAu/Cyst+MPS-1
,
and CFME/mAu/MPS-1+MHS-1 in blank aCSF (Curves a-c). Only one
cathodic peak, with Ep values of 0.21 V and 0.39 V, were observed at
This article is protected by copyright. All rights reserved.

10.1002/anie.201907210
Angewandte Chemie International Edition
COMMUNICATION
Figure 2. Optimized ground-state geometries of PHQ (A), PN (B), AP (C) and NP (D) in its oxidized and reduced forms calculated using DFT methods. From top
to bottom, images correspond to top view, side view, molecular orbital energy level, and HOMO and LUMO shapes. Electrochemical reactions (E-H) and CVs
obtained at (I) CFME/mAu/MPS-1, (J) CFME/mAu/MPS-2 in aCSF (pH 7.4) in the absence (a) and presence of 30 µM Na2S2 before (b) and after (c) scanning for 100
cycles, and CVs obtained at (K) CFME/mAu/MHS-1, (L) CFME/mAu/MHS-2 in the absence (a) and presence of 30 µM Na2S before (a) and after (b) 100 cycles of
scanning. Scan rate: 100 mV s-1.
CFME/mAu/MPS-1+MHS-1 after the addition of 30 μM Na2S2 and 30 μM
Na2S, respectively. These peaks were ascribed to the reduction of PHQ
and AP (Curve f and g). As expected, two well-separated cathodic
peaks were obtained at 0.22 V and 0.39 V after the simultaneous
addition of 30 μM Na2S2 and 30 μM Na2S (Curve h). Both of MPS-1 and
MHS-1 exhibited fast responses for Na2S2 and Na2S within 40 s (Figure
S14). On the one hand, jp value at 0.21 V (denoted as jp(PHQ)) gradually
increased with an increasing concentration of Na2S2, exhibiting a good
linearity from 0.2 to 60 μM (Figure 3A, 3A’), while jp value at 0.39 V
(jp(AP)) kept almost constant. On the other hand, jp(AP) showed a good
linearity with the concentration of Na2S ranging from 0.2 to 50 μM in
the presence of 30 μM Na2S2 (Figure 3B, B’). During simultaneous
measurements, negligible cross-talk was observed between the
responsive channels of Na2S2 and Na2S. When the concentrations of
Na2S2 and Na2S were simultaneously increased, good linearities were
obtained between jp(PHQ) with the concentration of Na2S2 from 0.2 to 50
μM, as well as between jp(AP) and the concentration of Na2S from 0.2 to
40 μM (Figure 3C, 3C’). The detection limits were estimated to be 40 ±
3 nM for H2Sn, and 47 ± 4 nM for H2S. Derived from the high
roughness of mAu, jp(PHQ) and jp(AP) at CFME/mAu/MPS-1+MHS-1 was
5.1-fold greater than obtained at the gold electrode modified with MPS-1
and MHS-1 (Au/MPS-1+MHS-1) (Figure S15). Moreover, the developed
biosensor exhibited high selectivity. Interferences smaller than 4.0%
and 5.1% were observed for jp(PHQ) and jp(AP) upon the addition of
potential interferences, particularly sulfur-containing species (Figure
S16), even though several interferences had relatively high
concentrations much larger than those of H2Sn and H2S. Besides, the
negligible current fluctuation as pH shifting from 6.0 to 8.0 suggested
the pH stability of measurement signal (Figure S17). Similar
measurements for detection of Na2S4 using MPS-1 probe were also
examined (Figure S18), indicating that MPS-1 was available for
determination of other polysulfides.
The anti-biofouling ability of the present biosensor modified with
mAu was examined. Obvious adsorption of BSA was observed at
Au/MPS-1+MHS-1
, but only weak fluorescence was obtained at
CFME/mAu/MPS-1+MHS-1 (Figure S19). Additionally, jp/jp0 values (jp0 is
the initial current density, and jp is the current density at different time)
of PHQ and AP obtained at Au/MPS-1+MHS-1 greatly decreased by
~54.0% and 52.7%, but negligible decrease (< 9.2%) was observed at
CFME/mAu/MPS-1+MHS-1 (Figure S20). To further clarify the roles of
3-
3+
the anti-fouling ability of mAu, Fe(CN)6 and Ru(NH3)6 with
opposite charges, which are sufficiently small to transfer into the
mesopores of mAu, were used to estimate the non-specific adsorption
of BSA. As expected, the redox peaks of two probes obviously
decreased at Au/MPS-1+MHS-1 in BSA solution, but changed slightly at
CFME/mAu/MPS-1+MHS-1 (Figure S21). The results revealed that the
adsorption of large biomolecules was restricted by the mesopores of
mAu, instead of electrostatic interaction.
The developed biosensor with high selectivity and sensitivity, as
well as good anti-fouling ability provided a reliable platform for
assaying H2Sn and H2S in the brain. Two brain disease models were
used: middle cerebral artery occlusion (MCAO) and AD. Figure 4A
shows DPVs obtained at CFME/mAu/MPS-1+MHS-1 electrode in the
hippocampus of a normal mouse brain before and after MCAO. Two
cathodic peaks at 0.22 V and 0.39 V were clearly observed (Figure 4A).
The standard deviation of jp(PHQ) and jp(AP) for five biosensors tested in
the same mouse brain did not exceed 4.8% for H2Sn and 5.3% for H2S,
indicating good reproducibility of the present biosensor (Figure S22).
With the ischemia duration prolonged, the infarction area was
gradually increased in the right cerebral hemisphere, indicating the
successful mouse MCAO model (Figure 4B). Correspondingly, the
concentration of H2Sn remarkably increased by 336.1%, 376.5%, and
400.3% in the hippocampus, cortex, and striatum of normal mouse
brain upon MCAO for 2.0 h, respectively (Figure 4C, Figure S23). To
This article is protected by copyright. All rights reserved.

10.1002/anie.201907210
Angewandte Chemie International Edition
COMMUNICATION
A
C
B
D
E
F
G
H
Figure 3. (A) DPVs obtained at the CFME/mAu/MPS-1+MHS-1 in aCSF (pH 7.4)
containing 30 μM Na2S and varied Na2S2 concentrations from 0.2 to 100 μM.
(B) DPVs obtained at the CFME/mAu/MPS-1+MHS-1 in aCSF (pH 7.4) containing
30 μM Na2S2 and varied Na2S concentrations ranging from 0.2 to 100 μM. (C)
DPVs obtained at the CFME/mAu/MPS-1+MHS-1 in aCSF solution (pH 7.4) with
simultaneously varied Na2S2 and Na2S concentrations ranging from 0.2 to 100
μM, and (A’-C’) Calibration plots of j versus Na2S2 and Na2S concentrations
obtained from A-C. The error bars indicate standard deviations (n=3, S.D.).
Figure 4. (A) Typical DPVs obtained at CFME/mAu/MPS-1+MHS-1 in a mouse
brain before (a) and after (b) MCAO for 2.0 h. (B) TTC -staining of ischemic
brain slices upon MCAO for different times, and Western blot analysis of the
level of TRPA1 protein and β-actin. (C) The levels of H2Sn, H2S, and TRPA1
protein in hippocampus upon MACO with different hours. (D) Fold of control
for H2Sn, H2S, and TRPA1 expression obtained in mouse brain slices after AO
treatment, (E) and successive stimulation by Na2S2, (F) Na2S, (G) Na2S and
H2O2, and (H) H2O2. The concentrations of Na2S2, Na2S and H2O2 in
experiments were 10 μM. The Error bars indicate standard deviations (n=3,
S.D.).
our surprise, the concentration of H2S first increased in the initial 0.5 h
MCAO, but fell down in the following 1.5 h MCAO. This observation
was completely different from previous report.12 The initial increase of
H2S was possibly due to the remarkedly elevated level of cysteine.13
However, H2S was gradually consumed by the produced ROS,
resulting in the decrease in the level of H2S with the ischemia duration
prolonged. After reperfusion, both of the peak currents returned to
nearly basal levels (Figure S24). Similar measurements were
conducted to track the variation of H2Sn and H2S in the mouse brain
with AD. Compared with those in the normal brain, the levels of H2Sn
increased to 2.65 ± 0.33 μM, 2.77 ± 0.31 μM, and 2.81± 0.24 μM in the
hippocampus, cortex, and striatum of APP/PS1 Tg mice, and the H2S
levels decreased to 0.98± 0.17 μM, 0.68 ± 0.16 μM, and 0.69 ± 0.13
μM (Table S3).
Recent investigation has revealed that TRPA1 channel, a Ca2+
permeable channel, plays important roles in ischemia and AD.14
Meanwhile, H2S and H2Sn are two potential molecules to activate
TRPA1 channel via S- sulfhydration. Therefore, it is meaningful to
explore the relationship between TRPA1 channel, and H2Sn and H2S in
the brain with ischemia and AD. Western blot analysis was first used to
track the expression of TRPA1 proteins in the mouse brain slices.
Compared with that of the normal brain, the level of TRPA1 protein
upregulate by 5.0% to 55.3% as MCAO time increased from 0.5 h to
2.0 h (Figure 4C). Interestingly, different from H2S, the variation of
TRPA1 protein exhibited positive correlation with the level of H2Sn in
the mouse brain under ischemia (Figure 4C). In addition, similar
positive correlation between TRPA1 protein and the concentration of
H2Sn was observed in brain slices from wild mice and APP/PS1 Tg
mice (Figure S25-S26).4c Previous studies have mainly focused on the
activity of H2S on S -sulfhydration of protein,15 even if H2Sn has been
reported to be generated through oxidation of H2S, and shows high
oxidation ability.1b,16 Our observation suggested that H2Sn was possibly
the actual molecule to activate TRPA1 channel.
cysteine), the generation of H2S was efficiently blocked by reducing
the amount of cysteine (Figure 4D). As expected, the amounts of H2S
and H2Sn in the brain slice were apparently decreased by 49.0% and
68.0%, and the TRPA1 protein downregulated by 62.1% (Figure 4D).
As demonstrated above, the levels of H2S and H2Sn in the brain slices
were significantly decreased after the slices were treated by AO, which
were considered as basal conditions. Then, the treated brain slices were
exogenously stimulated by Na2S2 (10 μM). The level of H2Sn
dramatically increased to 425.0%, but the level of H2S slightly
increased by 5.9% (Figure 4E). Meanwhile, a remarkable increase in
TRPA1 protein of 415.8% was observed. By contrast, in the treated
brain slices stimulated by Na2S (10 μM), the level of H2S increased
apparently, while negligible variation was obtained for both of H2Sn
and TRPA1 protein (Figure 4F). The results suggested that H2S failed
to generate H2Sn and to elevate the protein expression of TRPA1
channel, possibly due to lack of sufficient oxidant.1c Thus, followed by
the experiments above, we next stimulated the brain slices by H2O2
after addition of H2S, because H2O2 has been previously reported to
facilitate the endogenous generation of H2Sn from H2S.4c As expected,
the amount of H2S was gradually reduced with increasing
concentration of H2O2, but remarkable enhancement of H2Sn amount
was observed (Figure 4G). Meanwhile, it was noteworthy that the
protein expression of TRPA1 channels was increased with increasing
amount of H2Sn upon H2O2 stimulation (Figure S27). However, for the
control experiment, negligible changes were obtained for TRPA1
protein upon stimulation by only H2O2 using the slices treatment by
AO (Figure 4H) without addition of H2S and H2S2, demonstrating that
H2O2 cannot activate TRPA1 channel. Thus, the activation of TRPA1
channel were more positively correlated with H2Sn, which was
consistent with our results observed in the brain under ischemia and
AD (Figure 4C, Figure S26). These findings provided an evidence that
To further verify the roles of H2S and H2Sn in the activation of
TRPA1 channel, our useful tool was extended to monitor the levels of
H2Sn and H2S in the hippocampal mouse brain slices under different
stimulation. Cysteine is the main substance to generate H2S.17 Through
exogenous addition of 10 μM amino-oxyacetate (AO) (the inhibitor of
This article is protected by copyright. All rights reserved.

10.1002/anie.201907210
Angewandte Chemie International Edition
COMMUNICATION
Anal. Chem. 2016, 88, 2113; j) L. Zhang, F. Liu, X. Sun, G. Wei, Y. Tian, Z. Liu,
R. Huang, Y. Yu, H. Peng, Anal. Chem. 2017, 89, 1831.
H2Sn might a more efficient molecule for activation of TRPA1 channel
than H2S.
[7] a) D. Giovanelli, N. S. Lawrence, L. Jiang, T. G. J. Jones, R. G. Compton, Sens.
Actuators B 2003, 88, 320; b) J. A. Bennett, J. E. Pander, M. A. Neiswonger, J.
Electroanal. Chem. 2011, 654, 1.
In summary, using DFT calculation and electrochemical
measurements, we have rationally designed and synthesized MPS-1 and
MHS-1 as specific probes for the chemical recognition of H2Sn and H2S,
respectively. Based on MPS-1 and MHS-1, a single electrochemical
biosensor was created to simultaneously determine H2Sn and H2S levels
in vivo in the mouse brains with MCAO and AD. In addition, mAu
synthesized at CFME can improve the sensitivity, and presents good
anti-biofouling capability. This biosensor not only showed high
selectivity for H2Sn and H2S against potential interferences in the brain,
but was also able to avoid electrode contamination from sulfur
absorption. Using this in vivo biosensor, we observed the variation of
H2Sn and H2S in mouse brains under ischemia and AD. More
interestingly, H2Sn was found to exhibit a stronger ability to activate
TRPA1 channel than H2S. This investigation gives impetus to
reacquaintance of the molecular mechanism of H2S and H2Sn. More
importantly, this study also supplies an approach for designing single
biosensors for the determination of multiple species closely related to
physiological and pathological events in the brain.
[8] S. Wang, X. Liu, M. Zhang, Anal. Chem. 2017, 89, 5382.
[9] C. Li, O. Dag, T. Dao, T. Nagao, Y. Sakamoto, T. Kimura, O. Terasaki, Y.
Yamauchi, Nat. Commun. 2015, 6, 6608.
[10] P. Nagy, Z. Palinkas, A. Nagy, B. Budai, I. Toth, A. Vasas, Biochim. Biophys.
Acta 2014, 1840, 876.
[11] C. Liu, B. Peng, S. Li, C. Park, A. R. Whorton, M. Xian, Org. Lett. 2012, 14,
2184.
[12] a) K. Qu, C. P. Chen, B. Halliwell, P. K. Moore, P. T-H. Wong, Stroke 2006, 37,
889; b) Z. Li, Y. Wang, Y. Xie, Z. Yang, T. Zhang, Neurochem. Res. 2011, 36,
1840.
[13] a) K. Abe, H. Kimura, J. Neurosci. 1996, 16, 1066; b) C. Ren, A. Du, D. Li, J.
Sui, W. Mayhan, H. Zhao, Brain Res. 2010, 1345, 197.
[14] a) K. Lee, H. Lee, H. Lin, H. Tsay, F. Tsai, S. Shyue, T. Lee, J.
Neuroinflammation. 2016, 13, 92; b) P. W. Pires, S. Earley, Elife 2018, 7, e35316.
[15] A. K. Mustafa, M. M. Gadalla, N. Sen, S. Kim, W. Mu, S. K. Gazi, R. K. Barrow,
G. Yang, R. Wang, S. H. Snyder, Sci. Signal. 2009, 2, ra72.
[16] Y. Kimura, Y. Mikami, K. Osumi, M. Tsugane, J. Oka, H. Kimura, FASEB J.
2013, 27, 2452.
[17] R. d'Emmanuele di Villa Bianca, R. Sorrentino, P. Maffia, V. Mirone, C.
Imbimbo, F. Fusco, R. De Palma, L. J. Ignarro, G. Cirino, Proc. Natl. Acad. Sci.
2009, 106, 451.
Acknowledgements. Support by National Natural Science Foundation
of China (Nos. 21635003, 21827814 for Y. T. and 21675054 for L. Z.),
Innovation Program of Shanghai Municipal Education Commission
(201701070005E00020).
Keyword: Brain chemistry • Biosensors • Analytical Methods •
Polysulfide • Hydrogen sulfide
[1] a) T. V. Mishanina, M. Libiad, R. Banerjee, Nat. Chem. Bio. 2015, 11, 457; b) H.
Kimura, Proc. Jpn. Acad. Ser. B 2015, 91, 131; c) R. Greiner, Z. Palinkas, K.
Basell, D. Becher, H. Antelmann, P. Napy, T. P. Dick, Antioxid. Redox Signaling
2013, 19, 1749.
[2] a) M. Eberhardt, M. Dux, B. Namer, J. Miljkovic, N. Cordasic, C. Will, T. I.
Kichko, J. de la Roche, M. Fischer, S. A. Suarez, D. Bikiel, K. Dorsch, A. Leffler,
A. Babes, A. Lampert, J. K. Lennerz, J. Jacobi, M. A. Marti, F. Doctorovich, E.
D. Hogestatt, P. M. Zygmunt, I. Ivanovic-Burmazovic, K. Messlinger, P. Reeh,
M. R. Filipovic, Nat. Commun. 2014, 5, 4381; b) E. Blackstone, M. Morrison, M.
B. Roth, Science 2005, 308, 518; c) C. Szabo, Nat. Rev. Drug Discovery 2007, 6,
917.
[3] a) P. K. Yadav, M. Martinov, V. Vitvitsky, J. Seravalli, R. Wedmann, M. R.
Filipovic, R. Banerjee, J. Am. Chem. Soc. 2016, 138, 289; b) B. Tan, P. T.-H.
Wong, J. S. Bian, Neurochem. Int. 2010, 56, 3; c) J. W. Elrod, J. W. Calvert, J.
Morrison, J. E. Doeller, D. W. Kraus, L. Tao, X. Jiao, R. Scalia, L. Kiss, C.
Szabo, H. Kimura, C. W. Chow, D. J. Lefer, Pro. Natl. Acad. Sci. U. S. A. 2007,
104, 15560.
[4] a) B. Yu, Y. Zheng, Z. Yuan, S. Li, H. Zhu, L. K. De La Cruz, B. Wang, J. Am.
Chem. Soc. 2018, 140, 30; b) Y. Kimura, Y. Mikami, K. Osumi, M. Tsugane, J.
Oka, H. Kimura, FASEB J. 2013, 27, 2451.
[5] a) C. Liu, W. Chen, W. Shi, B. Peng, Y. Zhao, H. Ma, M. Xian, J. Am. Chem. Soc.
2014, 136, 7257; b) V. S. Lin, W. Chen, M. Xian, C. J. Chang, Chem. Soc. Rev.
2015, 44, 4596.
[6] a) A. G. Ewing, J. C. Bigelow, R. M. Wightman, Science 1983, 221, 169; b) G. S.
Wilson, M. C. Johnson, Chem. Rev. 2008, 108, 2462; c) D. L. Robinson, A.
Hermans, A. T. Seipel, R. M. Wightman, Chem. Rev. 2008, 108, 2554; d) W. Liu,
H. Dong, L. Zhang, Y. Tian, Angew. Chem. Int. Ed. 2017, 56, 16328; e) L. Liu, F.
Zhao, W. Liu, T. Zhu, J. Zhang, C. Chen, Z. Dai, H. Peng, J. Huang, Q. Hu, W.
Bu, Y. Tian, Angew. Chem. Int. Ed. 2017, 56, 10471; f) Y. Luo, L. Zhang, W. Liu,
Y. Yu, Y. Tian, Angew. Chem. Int. Ed. 2015, 54, 14053; g) X. Chai, X. Zhou, A.
Zhu, L. Zhang, Y. Qin, G, Shi, Y. Tian, Angew. Chem. Int. Ed. 2013, 52, 8129; h)
L. Zhang, Y. Tian, Acc. Chem. Res. 2018, 51, 688; i) J. Zhou, L. Zhang, Y. Tian,
This article is protected by copyright. All rights reserved.

10.1002/anie.201907210
Angewandte Chemie International Edition
COMMUNICATION
The correlation and distinction between
H2Sn and H2S playing in brain function
remain the subject of considerable
debate. A novel single electrochemcial
sensor was created for simultaneous
detection of H2Sn and H2S in brain. With
our sensor, the expression of TRPA1
channel was found to be positively
correlated with the levels of H2Sn in two
brain disease models under ischemia and
Alzheimer’s diseases, and H2Sn is more
acitive for expression of TRPA1 protein
than that of H2S.
Hui Dong, Qi Zhou, Limin Zhang,*
Yang Tian*
Page 1. – Page 5.
Rational Design of Specific Recognition
Molecules
for
Simultaneously
Monitoring of Endogenous Polysulfide
and Hydrogen Sulfide in the Mouse
Brain
This article is protected by copyright. All rights reserved.