Esterase-Sensitive Glutathione Persulfide Donor

Article abstract of DOI:10.1021/acs.orglett.8b02611

An esterase-sensitive glutathione persulfide (GSSH) donor (BW-GP-401) is described. The release profile was studied by monitoring the formation of lactone and direct trapping of GSSH with 1-fluoro-2,4-dinitrobenzene (DNFB). The donor was examined for its

Full text of DOI:10.1021/acs.orglett.8b02611

Letter
Cite This: Org. Lett. XXXX, XXX, XXX−XXX
pubs.acs.org/OrgLett
Esterase-Sensitive Glutathione Persulfide Donor
†
†
Zhengnan Yuan, Yueqin Zheng, Bingchen Yu, Siming Wang, Xiaoxiao Yang, and Binghe Wang*
Department of Chemistry and Center for Diagnostics and Therapeutics, Georgia State University, Atlanta, Georgia 30303, United
States
*
S Supporting Information
ABSTRACT: An esterase-sensitive glutathione persulfide (GSSH) donor (BW-GP-401) is described. The release profile was
studied by monitoring the formation of lactone and direct trapping of GSSH with 1-fluoro-2,4-dinitrobenzene (DNFB). The
donor was examined for its inhibitory effect toward glyceraldehyde 3-phosphate dehydrogenase (GAPDH). Under highly
oxidative conditions, the donor also shows cytoprotective effects in H9c2 cardiomyocytes.
lutathione (GSH) is known to be an essential
physiological molecule for cellular survival. Beyond
and carboxylic group. Thus, we first built a reactive disulfide
bond by treating GSH with 2,2′-dipyridyldisulfide, which was
1
9
G
the simple role of serving as an antioxidant in the cellular
environment, GSH also plays critical roles in signaling
processes, especially in redox signaling, which responds to
changes in oxidative stress and in the production of reactive
further treated with thiol acid BW-HP-101 to afford GSSH
donor BW-GP-401 (401), a stable odorless white powder
(Scheme 1, see Supporting Information (SI)). Using this route,
we protected the outer sulfur of the persulfide group and caged
it as a thioester, which has an esterase-sensitive trigger using a
2
oxygen species. For example, thioredoxin-glutathione signal-
3
10
ing in the CNS is regulated through a redox process. Among
“trimethyl lock” system. After the removal of protecting
the oxidized form of GSH, its disulfide (GSSG) and persulfide
group by an esterase, the phenol hydroxyl group is expected to
undergo lactonization to release GSSH (Scheme 1). Several
reasons were considered for choosing this system. Previously,
4
(
GSSH) were said to play dominant roles in signaling. As a
result, GSSH and GSSG are also important probes for
mechanistic studies. GSSH was said to be the most abundant
persulfide species in the physiological environment and was
Scheme 1. Synthesis and Release Mechanism of BW-GP-401
4b
shown to have strong H O -scavenging ability. It might
2
2
imply the importance of GSSH in cellular antioxidant defense.
The preparation of GSSG is relatively easy. Further GSSG is a
stable species, allowing for easy storage and application in
various studies. However, GSSH and RSSH in general are
5
unstable under physiological conditions and are hard to store
and deliver for mechanistic studies. Currently, the generation
of RSSH depends on inorganic polysulfide salts or redox
6
reactions involving H S with RSSR. However, the presence of
2
H S in the mixture limits their applications for exploring
2
functions of RSSH other than H S. Recently, our laboratory
2
has developed two series of esterase-sensitive prodrugs that
controllably release pure RSSH or hydrogen persulfide (H S )
2
2
7
under physiological conditions. Shortly thereafter, pH-, F-,
and H O -sensitive RSSH donors were reported. As follow-up
8
2
2
work, we are interested in developing donors or prodrugs of
GSSH to allow for in-depth mechanistic understandings of its
role in redox signaling. Hence, we undertook the effort to
develop a chemical strategy that can directly store and release
unprotected GSSH using a physiological trigger.
To prepare a donor of unprotected GSSH, one key issue is
developing a synthetic route, which can tolerate the free amino
Received: August 14, 2018
©
XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.8b02611
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
we have used this strategy to release H S and H S , and the
the trapping method has been successfully applied to benzyl
persulfide with high yield, the low trapping efficiency might
2
2 2
7b,11
7a
chemistry is well understood.
control the release rates have been extensively studied.
Additional factors that
11
be due to the unstable nature of GSSH under physiological
conditions and/or further scrambling by other thiol species
present. It has been well accepted that the general stability of
Different masking groups on the phenol hydroxyl group should
allow for tuning the release rates and on-demand and selective
7
b,11
14
prodrug activation.
They can also be used to improve the
persulfide species is poor. Only a limited number of small-
7b
stability and water solubility. The byproduct lactone is a
nontoxic and stable molecule that ensures its application in
molecule persulfide species have been obtained and partially
15
characterized. In the trapping process, there is a competition
between trapping and decomposition of the persulfide species
through disproportionation. To further identify the products
from GSSH, we used mass spectrometry to analyze the
incubation solution and found the increasing relative intensity
of a peak attributed to GSSSG as the release reaction
progressed (Figures S1, S2). Such results further suggest
GSSH disproportionation as a contributing factor for the low
yield in the trapping experiments. As an effort to minimize the
bimolecular disproportionation reaction, we conducted addi-
tional release experiments using a lower level of 401 (5 μM)
and PLE (1 unit/mL) and were able to achieve 34 ± 3% yield
of the trapped experiments with the remaining mass balanced
to be predominately GSSSG.
7
b
biological systems.
We first examined the release of GSSH from 401. Since
GSSH is unstable for detection under physiological conditions,
we studied the release kinetics by using HPLC to monitor
lactone formation. Specifically, 100 μM 401 was incubated
with 2 unit/mL of porcine liver esterase (PLE) at 37 °C. At
different time points, the generation of the lactone was
analyzed by HPLC. After 10 min, the donor peak became
almost undetectable. As shown in Figure 1B, at the 30 min
H S was considered to be one of the products from
2
1
4,15e
persulfide through disproportionation (Figure 2A).
Figure 1. (A) Activation of 401 and subsequent GSSH trapping by
DNFB. (B) Released lactone was detected by HPLC. n = 3. (C)
Trapped product A was monitored by HPLC. n = 3.
time point, the lactone peak corresponded to over 80%
conversion. To verify that triggered release only happens in the
presence of PLE, we also incubated the GSSH donor in
phosphate-buffered saline (PBS) without introducing PLE: no
lactone was detected, and the donor peak remained at about
the same area over a period of 2 h, the maximal time of
incubation period studied.
Figure 2. (A) H S was released from 401 with or without NAC. (B)
H S detection by the MB method. n = 3.
2
2
To further validate the release, we used a trapping reagent to
12
directly capture GSSH from 401. Unlike H S and H S , no
2
2 2
specific probe has been developed to detect GSSH. Previously,
our laboratory used 2,4-dinitrofluorobenzene (DNFB) to trap
Compound 401 provides a chance to study the H S release
2
from pure GSSH. In this case, 200 μM of 401 was incubated in
PBS containing 10 unit/mL of PLE. The methylene blue (MB)
method was used to probe H S generation. 50 μM of H S was
7a,13
benzyl persulfide with a yield of over 80%.
also used DNFB to trap GSSH (Figure 1A). To fully consume
01 within a short period of time, 10 unit/mL of PLE was
In this study, we
2
2
4
detected from 200 μM of 401 between 2 and 10 min (Figure
2B). In the control group, without PLE, no generation of H2S
was observed within 10 min (Figure 2B, S3). In a biological
environment, the level of cysteine or other free thiol groups is
added to the solutions of 401 at 100, 200, and 400 μM with 2
mM DNFB. The resulting mixture was allowed to be incubated
for 30 min at 37 °C. The generation of the trapped product A
was analyzed by HPLC (Figure 1C). For the purpose of
quantitative analysis of the trapped product A, the standard
compound was synthesized (see SI). Different concentrations
of 401 gave various amounts of trapped product A; specifically,
about 12 μM, 18 μM, and 45 μM GSSH (∼10%) were
detected by treating with 100 μM, 200 μM, and 400 μM of
4b
high. GSSH is known to quickly react with thiol and yield the
14,16
disulfide byproduct and H S (Figure 2A).
We then applied
2
2 mM NAC into the same system used above. At the 2 min
point, the H S level reached a peak concentration of 115 μM
2
and then decreased to 28 μM at the 10 min point (Figure 2B).
It should be noted that the peak concentration for a
gasotransmitter generated from a prodrug is a different
4
01 with 10 unit/mL of PLE, respectively (Figure 1C). Since
B
DOI: 10.1021/acs.orglett.8b02611
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
concept as compared to the maximal concentration for a
a thiol residue, was added to each group. A significant increase
in activity was observed after incubation for another 2 h at
room temperature (Figure 3). According to the above results,
GSSH can also inhibit the activity of GAPDH and might act as
a redox regulator participating in regulating cellular metabo-
lism.
1
7
nonvolatile drug generated from a prodrug. We have
demonstrated in the past that a peak concentration at 50%
of the prodrug concentration was about the highest that one
11
can achieve from a H S donor. Thus, it is reasonable to
2
assume that the generation of 115 μM of H S represents a near
2
quantitative reaction. The G-SS-NAC disulfide byproduct was
also quantified by mass spectrometry, and the yield was 38%
GSSH was found to be more potent in providing
cytoprotective effects in highly oxidative cellular environment
4b
(
see SI). There is a clear discrepancy between H S production
than GSH or H S. In an effort to probe this point, we used
2
2
and G-SS-NAC formation in quantity. This is easy to
understand because of the scrambling effect of the existing
thiol (excessive NAC) species, which can react with G-SS-
NAC to give GSH and NAC disulfide. In this case, we also
quantified the GSH in the system, and the yield was about
401 to rescue cells in highly oxidative environments. We first
tested the cytotoxicity of 401 on H9c2 cell line; no obvious
toxicity was observed at 200 μM after 24 h of incubation
(Figure S4). We next treated cells with 50, 100, and 150 μM
401 before exposure to H O (Figure 4). Without treatment
2 2
7
7%. All these explain the mass balance issue. However, in the
absence of PLE, H S was not detectable at the 10 min point
2
with exposure to a high concentration of NAC (Figure 2B, S3).
The results further demonstrate the need for an esterase to
trigger the release of GSSH from 401, supporting the
instability of GSSH under physiological conditions. Further,
the relatively low trapping yield of GSSH and the high
conversion yield in reaction with NAC give a further hint of
why nature “designs” GSSH as a reagent for protein thiol
modification.
It has been reported that RSSH can regulate the activity of
glyceraldehyde 3-phosphate dehydrogenase (GAPDH)
18
through S-persulfidation. We previously applied our H2S2
donor on GAPDH and confirmed the correlation between
elevated protein sulfhydration levels and decreased enzyme
Figure 4. Effects of different concentrations of 401 or various controls
and relevant compounds on the H9c2 cell line in the presence of
H O . Prior to incubation with H O (450 μM) for 3 h at 37 °C
2
2
2
2
7
b
activity. In the physiological environment, GSSH was said to
(except for vehicle group), H9c2 cells were treated with (1) blank
be the most abundant small-molecule persulfide and may also
control; (2) 150 μM Na S; (3) 150 μM lactone; (4) 150 μM GSH;
(5) 50 μM 401; (6) 100 μM 401; and (7) 150 μM 401 for 1 h at 37
2
4b,18
be involved in the regulation of GAPDH activity.
explore this aspect, GAPDH was incubated with 25, 50, and
00 μM of 401, respectively, for 30 min in the presence of 10
To
°
C. Cell viability was measured by Cell Counting Kit-8 (CCK-8). n =
3. *P < 0.01 for the comparison with group 1. #P < 0.01 for the
1
comparison with group 5.
unit/mL of PLE. Then its catalytic activity was determined.
Compared to the vehicle group, the catalytic ability decreased
to 62 ± 1%, 44 ± 9%, and 37 ± 8% of the maximal activity,
respectively (Figure 3). A control group treated with lactone
did not show any effect on GAPDH activity. To see if the
decreased activity was caused by S-persulfidation, 2 mM
dithiothreitol (DTT), which can reduce the persulfide group to
with 401, after the incubation with 450 μM H O , cell viability
2
2
was reduced by 60% compared with the vehicle group. In the
presence of 401, this reduction in viability was less than 40%
and was concentration dependent (401). Meanwhile, the
byproduct lactone failed to rescue cells from H O . H9c2 cells
2
2
were also treated with GSH and Na S before adding H O .
2
2
2
However, no obvious increase in viability was observed (Figure
). The above results further support the fact that GSSH is
4
more effective than GSH and H S in protecting cells from
2
highly oxidative environment, which might indicate its unique
property in defending cellular oxidative stress.
In conclusion, we have successfully developed the first
esterase-sensitive GSSH. The release of GSSH was validated by
the formation of byproduct lactone and direct trapping with
DNFB. Using this donor, we successfully addressed the
stability issue of GSSH. In a biological context, the donor
inhibited the activity of GAPDH presumably through S-
persulfidation and protected H9c2 cells against H O -induced
2
2
highly oxidative stress. Taken together, we provide a robust
research tool to study the biological relevance of GSSH and
also the chemistry that helps store and deliver other important
persulfide species. However, we should also note that the easy
scrambling process involving persulfide (Figure 2) means that
the concentration and time scale make a difference in the
experimental outcome when designing biological studies.
Figure 3. Inhibition of GAPDH activity by 401. 2 μg/mL of GAPDH
was incubated with various concentration of 401 or 100 μM lactone
for 0.5 h with 10 unit/mL of PLE and 1% DMSO, after which the
enzyme activity was measured. Each group was then incubated with 2
mM DTT at rt for 2 h before the measurement of GAPDH activity. n
=
3. *P < 0.01 for the comparison with the vehicle group. #P < 0.05
for the comparison with the group before DTT treatment.
C
DOI: 10.1021/acs.orglett.8b02611
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
(
12) (a) Peng, H.; Chen, W.; Burroughs, S.; Wang, B. Curr. Org.
ASSOCIATED CONTENT
Supporting Information
■
Chem. 2013, 17, 641−653. (b) Lin, V. S.; Chen, W.; Xian, M.; Chang,
*
S
C. J. Chem. Soc. Rev. 2015, 44, 4596−4618.
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.8b02611.
(13) Sawahata, T.; Neal, R. A. Anal. Biochem. 1982, 126, 360−364.
(14) Park, C. M.; Weerasinghe, L.; Day, J. J.; Fukuto, J. M.; Xian, M.
Mol. BioSyst. 2015, 11, 1775−1785.
(
15) (a) Nakabayashi, T.; Tsurugi, J.; Yabuta, T. J. Org. Chem. 1964,
Experimental procedure and characterization data
29, 1236−1238. (b) Kawamura, S.; Otsuji, Y.; Nakabayashi, T.; Kitao,
(PDF)
T.; Tsurugi, J. J. Org. Chem. 1965, 30, 2711−2714. (c) Chatterji, T.;
Keerthi, K.; Gates, K. S. Bioorg. Med. Chem. Lett. 2005, 15, 3921−
3924. (d) Bailey, T. S.; Zakharov, L. N.; Pluth, M. D. J. Am. Chem.
Soc. 2014, 136, 10573−10576. (e) Bailey, T. S.; Pluth, M. D. Free
Radical Biol. Med. 2015, 89, 662−667.
AUTHOR INFORMATION
Corresponding Author
E-mail: wang@gsu.edu.
■
*
(16) Pan, J.; Carroll, K. S. ACS Chem. Biol. 2013, 8, 1110−1116.
(17) (a) Zheng, Y.; Yu, B.; La Cruz, L. K.; Roy Choudhury, M.;
ORCID
Anifowose, A.; Wang, B. Med. Res. Rev. 2018, 38, 57−100.
b) Martelli, A.; Testai, L.; Breschi, M. C.; Blandizzi, C.; Virdis, A.;
Binghe Wang: 0000-0002-2200-5270
(
Author Contributions
Taddei, S.; Calderone, V. Med. Res. Rev. 2012, 32, 1093−1130.
(18) (a) Mustafa, A. K.; Gadalla, M. M.; Sen, N.; Kim, S.; Mu, W.;
Gazi, S. K.; Barrow, R. K.; Yang, G.; Wang, R.; Snyder, S. H. Sci.
Signaling 2009, 2, ra72. (b) Jarosz, A. P.; Wei, W.; Gauld, J. W.; Auld,
†Z.Y. and Y.Z. contributed equally to this work.
Notes
̈
J.; Ozcan, F.; Aslan, M.; Mutus, B. Free Radical Biol. Med. 2015, 89,
12−521.
The authors declare no competing financial interest.
5
ACKNOWLEDGMENTS
■
We would like to express our appreciation to the GSU B&B
program for a fellowship to B.Y., Georgia State University and
the Georgia Research Alliance for an Eminent Scholar
endowment (B.W.), and the Chemistry Department for their
partial support of the students (Y.Z., Z.Y.).
REFERENCES
■
(
1) (a) Townsend, D. M.; Tew, K. D.; Tapiero, H. Biomed.
Pharmacother. 2003, 57, 145−155. (b) Trachootham, D.; Lu, W.;
Ogasawara, M. A.; Valle, N. R.-D.; Huang, P. Antioxid. Redox Signaling
2
(
008, 10, 1343−1374.
2) (a) Rahman, I.; Biswas, S. K.; Jimenez, L. A.; Torres, M.;
Forman, H. J. Antioxid. Redox Signaling 2004, 7, 42−59. (b) Hancock,
J.; Whiteman, M. React. Oxyg. Species 2018, 5, 78−85.
(
3) (a) Ren, X.; Zou, L.; Zhang, X.; Branco, V.; Wang, J.; Carvalho,
C.; Holmgren, A.; Lu, J. Antioxid. Redox Signaling 2017, 27, 989−
1
2
(
010. (b) Aquilano, K.; Baldelli, S.; Ciriolo, M. R. Front. Pharmacol.
014, 5, 196−207.
4) (a) Kimura, Y.; Koike, S.; Shibuya, N.; Lefer, D.; Ogasawara, Y.;
Kimura, H. Sci. Rep. 2017, 7, 10459−10470. (b) Ida, T.; Sawa, T.;
Ihara, H.; Tsuchiya, Y.; Watanabe, Y.; Kumagai, Y.; Suematsu, M.;
Motohashi, H.; Fujii, S.; Matsunaga, T.; Yamamoto, M.; Ono, K.;
Devarie-Baez, N. O.; Xian, M.; Fukuto, J. M.; Akaike, T. Proc. Natl.
Acad. Sci. U. S. A. 2014, 111, 7606−7611.
(
5) Barker, P. L.; Gendler, P. L.; Rapoport, H. J. Org. Chem. 1981,
6, 2455−2465.
6) (a) Rao, G. S.; Gorin, G. J. Org. Chem. 1959, 24, 749−753.
b) Smith, D. J.; Venkatraghavan, V. Synth. Commun. 1985, 15, 945−
50.
7) (a) Zheng, Y.; Yu, B.; Li, Z.; Yuan, Z.; Organ, C. L.; Trivedi, R.
K.; Wang, S.; Lefer, D. J.; Wang, B. Angew. Chem., Int. Ed. 2017, 56,
1749−11753. (b) Yu, B.; Zheng, Y.; Yuan, Z.; Li, S.; Zhu, H.; De La
Cruz, L. K.; Zhang, J.; Ji, K.; Wang, S.; Wang, B. J. Am. Chem. Soc.
018, 140, 30−33.
8) (a) Kang, J.; Xu, S.; Radford, M. N.; Zhang, W.; Kelly, S. S.; Day,
4
(
(
9
(
1
2
(
J. J.; Xian, M. Angew. Chem., Int. Ed. 2018, 57, 5893−5897. (b) Powell,
C. R.; Dillon, K. M.; Wang, Y.; Carrazzone, R. J.; Matson, J. B. Angew.
Chem., Int. Ed. 2018, 57, 6324−6328.
(
9) Navath, R. S.; Kurtoglu, Y. E.; Wang, B.; Kannan, S.; Romero, R.;
Kannan, R. M. Bioconjugate Chem. 2008, 19, 2446−2455.
10) Amsberry, K. L.; Borchardt, R. T. J. Org. Chem. 1990, 55,
867−5877.
11) Zheng, Y.; Yu, B.; Ji, K.; Pan, Z.; Chittavong, V.; Wang, B.
Angew. Chem., Int. Ed. 2016, 55, 4514−4518.
(
5
(
D
DOI: 10.1021/acs.orglett.8b02611
Org. Lett. XXXX, XXX, XXX−XXX