Design and synthesis of polymeric hydrogen sulfide donors

Article abstract of DOI:10.1021/bc500150s

Hydrogen sulfide (H₂S) is a gaseous signaling molecule that has several important biological functions in the human body. Because of the difficulties of handling H₂S gas, small organic compounds that release H₂S under physiological conditions have been developed. The observed bioactivities of these H₂S donors have generally been directly correlated with their H₂S release properties. However, apart from H₂S release, these H₂S donors also exert biological effects by direct interaction with intracellular components within the cytoplasm after passive diffusion across cellular membranes. Here we report polymeric H₂S donors based on ADT-OH which would alter cellular trafficking of ADT-OH to minimize the unfavorable interactions with intracellular components. We designed and synthesized a poly(ethylene glycol)-ADT (PEG-ADT) conjugate having ADT linked via an ether bond. Whereas ADT-OH significantly reduced cell viability in murine macrophages, the PEG-ADT conjugate did not show obvious cytotoxicity. The PEG-ADT conjugate released H₂S in murine macrophages but not in the presence of serum proteins. The PEG-ADT conjugate was taken up by the cell through the endocytic pathway and stayed inside endolysosomes, which is different from the small amphiphilic donor ADT-OH that can directly enter the cytoplasm. Furthermore, PEG-ADT was capable of potentiating LPS-induced inflammation. This polymeric H₂S donor approach may help to better understand the H₂S bioactivities of the H₂S donor ADT-OH.

Full text of DOI:10.1021/bc500150s

Article
pubs.acs.org/bc
Design and Synthesis of Polymeric Hydrogen Sulfide Donors
Urara Hasegawa*,‡,∥ and Andre J. van der Vlies†,‡
́
‡Department of Applied Chemistry, Graduate School of Engineering, Osaka University, Osaka 565-0871, Japan
∥Frontier Research Base for Young Researchers, Graduate School of Engineering, Osaka University, Osaka 565-0871, Japan
†Frontier Research Center, Graduate School of Engineering, Osaka University, Osaka 565-0871, Japan
S
* Supporting Information
ABSTRACT: Hydrogen sulfide (H2S) is a gaseous signaling
molecule that has several important biological functions in the
human body. Because of the difficulties of handling H2S gas,
small organic compounds that release H2S under physiological
conditions have been developed. The observed bioactivities of
these H2S donors have generally been directly correlated with
their H2S release properties. However, apart from H2S release,
these H2S donors also exert biological effects by direct
interaction with intracellular components within the cytoplasm
after passive diffusion across cellular membranes. Here we
report polymeric H2S donors based on ADT−OH which
would alter cellular trafficking of ADT−OH to minimize the
unfavorable interactions with intracellular components. We designed and synthesized a poly(ethylene glycol)−ADT (PEG−
ADT) conjugate having ADT linked via an ether bond. Whereas ADT−OH significantly reduced cell viability in murine
macrophages, the PEG−ADT conjugate did not show obvious cytotoxicity. The PEG−ADT conjugate released H2S in murine
macrophages but not in the presence of serum proteins. The PEG−ADT conjugate was taken up by the cell through the
endocytic pathway and stayed inside endolysosomes, which is different from the small amphiphilic donor ADT−OH that can
directly enter the cytoplasm. Furthermore, PEG−ADT was capable of potentiating LPS-induced inflammation. This polymeric
H2S donor approach may help to better understand the H2S bioactivities of the H2S donor ADT−OH.
of donor molecules that can release H2S under physiological
conditions. For delivering H2S to cells, most studies have used
salts like NaHS/Na2S that produce H2S when dissolved in
aqueous solutions. Whereas these salts show low cytotoxicity at
relatively high dose, the drawbacks in using them as H2S donors
is their uncontrolled release profile. This is especially a problem
as it has been suggested that a slow and continuous release is
important.5,12
For these reasons, small organic compounds have been
reported that can release H2S under physiological conditions. In
general, the observed biological activities of these H2S donors
have been directly correlated with their H2S release properties.
However, what is generally overlooked is that these H2S donors
can directly interact with intracellular biomolecules and
organelles, which can affect cellular functions. Therefore, the
results obtained with these donor molecules may not be
necessary due to H2S.
INTRODUCTION
■
Over the past decade, hydrogen sulfide (H2S) has emerged as
the third endogenously produced gaseous transmitter besides
nitric oxide (NO) and carbon monoxide (CO).1 For a long
time considered as a toxic compound,2,3 currently it has
become clear to affect several organ systems and biological
processes in the human body.4,5 For example, H2S has been
shown to be a regulator of inflammation, promote angiogenesis,
modulate blood pressure, and protect cells during ischemia
reperfusion and neuro-inflammation. Despite the many
publications dealing with H2S as an important gaseous signaling
molecule, it has been recently suggested that persulfides of
cysteine and glutathione6 or polysulfides (H2S2n)7 are the actual
bioactive species. It was proposed that H2S is generated from
cysteine and glutathione persulfides and that H2S is a marker
for their production. This is different from polysulfides that are
reported to form by H2S oxidation.
5-(4-Hydroxyphenyl)-3H-1,2-dithiole-3-thione (ADT−OH,
Scheme 1), one of the most commonly used H2S donors,
belongs to the class of compounds having the 3H-1,2-dithiole-
3-thione group in their structure. Members of this family like 5-
Along with the understanding of the different roles of H2S, its
therapeutic potential has also been realized.8−10 This
exogeneously delivered H2S may act directly or indirectly in
the form of cysteine and glutathione persulfides or polysulfides
because it is well-known that both readily form from H2S under
oxidative conditions.11
Received: April 6, 2014
Revised: May 26, 2014
Published: June 19, 2014
To study the biological effects and to solve the difficulty in
handling H2S gas, much research has gone in the development
© 2014 American Chemical Society
1290
dx.doi.org/10.1021/bc500150s | Bioconjugate Chem. 2014, 25, 1290−1300

Bioconjugate Chemistry
Article
effects were ascribed to the anti-inflammatory effect of released
H2S, which was supported by the increase of H2S plasma levels
in rat for mesalamine36 and diclofenac.34 Because these papers
used an ester bond to prepare hybrid molecules of ADT−OH
and other drugs, we initially prepared the polymer−ADT
conjugates using the same approach. To prepare the polymer−
ADT conjugate, we used polyethylene glycol (PEG) that has
been widely used as a biocompatible polymer in the field of
biomaterials. Hydroxyl-terminated methoxy PEG with a
nominal molecular weight of 5000 was activated with
carbonyldiimidazole and reacted with either glycine (Gly) or
isoleucine (Ile) to introduce a carboxylic acid group that was
reacted with the phenol group of ADT−OH. Coupling was
achieved using dicyclohexylcarbodiimide to afford the PEG−
Gly−ADT (7) and PEG−Ile−ADT (8) conjugates (Scheme 2).
Hydrolytic Stability of the PEG−ADT Ester Conju-
gates. To gain information on the hydrolytic stability of the
PEG−Gly−ADT conjugate, we prepared solutions in PBS
(pH7.4) and incubated at 37 °C. At different time points, the
mixture was acidified with acetic acid and lyophilized. 1H NMR
of the freeze-dried samples was measured to quantify the extent
of hydrolysis by comparing the integral values of the olefinic
CH proton of ADT−OH and ADT linked by the ester bond at
7.82 and 7.68 ppm, respectively. As can be seen in Figure 1, the
Scheme 1. Chemical Structures of 5-(4-Hydroxyphenyl)-3H-
1,2-dithiole-3-thione (ADT−OH, 1) and 5-(4-
Methoxyphenyl)-3H-1,2-dithiole-3-thione (ADT−OCH3, 2)
(4-methoxyphenyl)-3H-1,2-dithiole-3-thione (ADT−OCH3)
and Oltipraz have been shown to have anticancer properties
by activating phase II enzymes.13,14 ADT−OCH3 is a choleretic
drug marketed as Sulfarlem15 that has been investigated in
chemoprevention in human lung cancer16 and colon cancer17
and in rat cancer.18 This drug also has antioxidative properties
by increasing glutathione levels19 and the hepatoprotective
properties due to induction of the phase II detoxification
enzymes, in particular NAD(P)H:quinone oxidoreductase and
glutathione-S-reductase enzymes.20−22 Furthermore, both
ADT−OH and ADT−OCH3 have been shown to inhibit
NF-κB activation in human T cells23 and human estrogen
receptor-negative breast cancer via chemical modification.24
This is different from the reported NF-κB activation by H2S.
where sulfhydration is generally believed to cause activation.25
In the field of drug delivery, polymeric therapeutics have
been extensively studied.26 In this approach, drugs are
conjugated to hydrophilic polymers (polymer−drug conju-
gates).27 It has been shown that polymeric therapeutics not
only improves drug solubility and stability but also prolongs
circulation time, reduces side effects, and alters pharmacoki-
netics and intracellular trafficking.28
In this study, we prepared a polymeric H2S donor which
minimizes the side effects caused by the small donor molecule.
We designed and synthesized a poly(ethylene glycol)−ADT
(PEG−ADT) conjugate having ADT linked via an ether bond.
The cellular toxicity of the PEG−ADT conjugate was compared
with ADT−OH and ADT−OCH3 in murine macrophages. Its
H2S release property was also investigated. Furthermore, the
effect on lipopolysaccharide (LPS)-induced inflammation was
also evaluated.
Figure 1. Hydrolysis kinetics of the PEG−Gly−ADT and PEG−Ile−
ADT conjugates in PBS (pH 7.4) at 37 °C as measured by 1H NMR.
PEG−Gly−ADT conjugate hydrolyzed quickly with an
estimated half-life of about 30 min. This hydrolysis was also
macroscopically visible by the occurrence of an orange
precipitate due to the poor solubility of ADT−OH in PBS.
With such a fast chemical hydrolysis rate, the conjugate is also
expected to be cleaved in serum due to the presence of esterase.
Indeed, the UV/vis spectrum of the PEG−Gly−ADT
conjugate in 10% fetal bovine serum (FBS) after 2 h shows
spectral changes due to the partial hydrolysis of the conjugate
(Supporting Information Figure S1). Because hydrolysis of the
PEG−Gly−ADT conjugate would be too fast for our
application, we investigated whether we could slow down
hydrolysis by changing the chemical environment around the
RESULTS AND DISCUSSION
■
Synthesis of PEG−ADT Conjugates Linking ADT via
an Ester Bond. Biological activities of ADT−OH in
connection with H2S release has been reported for so-called
hybrid molecules in which the ADT−OH moiety is coupled to
known drugs like nonsteroidal anti-inflammatory drugs
(NSAIDs).29,30 The first example of such a compound is
ATB-429, linking ADT−OH via an ester bond with mesal-
amine, which was shown to have reduced toxicity in a murine
model of colitis compared to mesalamine alone.31,32 Similar
reduction of gastrointestinal damaging of NSAIDs were
reported for aspirin,33 diclofenac,34 and naproxen.35 These
a
Scheme 2. Preparation of the PEG−Gly−ADT and PEG−Ile−ADT Conjugates Having ADT Linked via an Ester Bond
a
(a) CDI, DMSO, (b) glycine or isoleucine, Et3N, DMSO, 80 °C, (c) ADT−OH, DCC, DMAP, CH2Cl2.
1291
dx.doi.org/10.1021/bc500150s | Bioconjugate Chem. 2014, 25, 1290−1300

Bioconjugate Chemistry
Article
a
Scheme 3. Preparation of the PEG−ADT Conjugate and TEG−ADT Having ADT Linked via an Ether Bond
a
(a) Methanesulfonyl chloride, Et3N, toluene; (b) ADT−OH, K2CO3, DMF, 80 °C.
ester. We therefore tested the hydrolysis of the PEG−Ile−ADT
conjugate with the idea that the bulky and hydrophobic sec-
butyl group would avoid ester hydrolysis. Although much
slower, hydrolysis still occurred and about 50% was hydrolyzed
after 30 h as shown in Figure 1.
conjugate with an unstable linkage was more toxic than the
PEG−Ile−ADT conjugate which showed much slower release
of ADT−OH. A similar trend was observed in the live/dead
assay as shown in Figure 3. For the nonhydrolyzable PEG−
ADT conjugate, no obvious cellular toxicity was observed in the
live/dead assay unlike the PEG−Gly−ADT and PEG−Ile−
ADT conjugates (Figure 3). On the other hand, a slight
reduction in metabolic activity was observed at concentrations
above 200 μM in the MTT assay (Figure 2).
To confirm that ADT−OH released from the conjugates
exerted the toxicity in the cells, the UV/vis spectrum of the
culture medium containing the PEG−Gly−ADT and PEG−
ADT conjugates was measured after incubation with RAW-Blue
macrophages (Supporting Information Figure S1). The typical
UV/vis spectrum of ADT−OH (Supporting Information
Figure S1) was observed for the PEG−Gly−ADT conjugate,
while the spectrum of ADT−OH was not observed in the case
of the PEG−ADT conjugate, indicating that hydrolysis did not
occur. To show chemical evidence for the release of ADT−OH
in culture, cells with medium that had been incubated with
PEG−Gly−ADT and PEG−Ile−ADT were lyophilized. After
extraction and aqueous workup, 1H NMR showed the presence
of ADT−OH (Supporting Information Figure S2). Therefore,
the cellular toxicities of the PEG−Gly−ADT and PEG−Ile−
ADT conjugates were due to released ADT−OH after
hydrolysis.
Synthesis PEG−ADT Conjugate Linking ADT via an
Ether Bond. In our concept of a polymeric ADT-based H2S
donor, the ester conjugates were not suited due to rapid ester
hydrolysis. We therefore focused our attention on the
preparation of a PEG conjugate linking ADT via a noncleavable
ether bond as shown in Scheme 3. Nucleophilic substitution of
ADT−OH with PEG mesylate afforded the PEG−ADT (11)
conjugate. To confirm the structure of the PEG−ADT
conjugate, we also prepared a small molecule linking ADT to
three ethylene glycol units (TEG−ADT, 12) and characterized
this compound by 1H NMR and high resolution mass
spectrometry. The chemical shifts of the signals due to the
ADT moiety of this model compound were identical with that
of the PEG−ADT conjugate confirming its structure.
Cellular Toxicity of the PEG−Gly−ADT, PEG−Ile−ADT,
and PEG−ADT Conjugates. ADT−OH is known to reduce
cell viability by inhibiting histone deacetylase37,38 and NF-κB
activation.23 Here we tested the effect of PEG conjugation on
the cytotoxicity of ADT−OH. As shown in Figure 2, ADT−
It is interesting to note that most studies using ADT−OH
linked with another molecule via an ester bond do not mention
this hydrolysis. Our data of the PEG−Gly−ADT and PEG−
Ile−ADT conjugates suggests that this is a point to consider in
evaluating the bioactivity of such compounds.
H2S Release from the PEG−ADT Conjugate. Contrary
to other H2S donors allyldisulfide/allyltrisulfide,39 thiobenza-
mides,40 and GYY 4137,12 the precise mechanism for H2S
release from ADT−OH is not clear. A direct hydrolysis
mechanism, in which the thiocarbonyl is converted to a
carbonyl group with concomitant H2S release, has been
suggested.41 However, such a mechanism seems unlikely
because a derivative of ADT−OH was only completely
hydrolyzed after 48 h in PBS/DMSO at 120 °C. It is therefore
more likely that an enzymatic step is involved. It has been
reported that ADT−OH released H2S in rat liver homogenate
and rat plasma but significantly less in buffer alone.34 In
addition, it has been also reported that isolated mitochondria
from U373 cells were able to induce H2S release, suggesting
that reductive enzymes as present in mitochondria are
responsible.42 To see whether the PEG−ADT conjugate was
capable of releasing H2S, we used a H2S-specific electrode to
monitor H2S release. Because ADT−OH and ADT−OCH3
contains a disulfide linkage that can be reduced electrochemi-
cally,43 we investigated if a simple chemical, i.e., nonenzymatic,
reduction could induce H2S release. We first used the reducing
Figure 2. Cell viability of RAW-Blue cells treated with PEG−Gly−
ADT, PEG−Ile−ADT, and PEG−ADT. ADT−OCHH3, and ADT−
OH for 24 h (MTT assay).
OH significantly reduced metabolic activity in RAW-Blue
macrophages as measured by the MTT assay. The live/dead
assay showed that the majority of cells apoptosed and detached
from the glass surface (Figure 3). On the other hand,
conjugation of ADT−OH to PEG reduced cellular toxicity
(Figure 2). This reduced toxicity is not due to alkylation of the
phenol group as shown by the toxicity profile of ADT−OCH3
(Figure 2). The hydrolysis rate of the PEG−Gly−ADT and
PEG−Ile−ADT conjugates significantly affects toxicity in the
cells. The MTT assay showed that the PEG−Gly−ADT
1292
dx.doi.org/10.1021/bc500150s | Bioconjugate Chem. 2014, 25, 1290−1300

Bioconjugate Chemistry
Article
Figure 3. CLSFM images showing live (green) and dead (red) RAW-Blue macrophages after incubating for 24 h with (A) PEG−ADT, (B) PEG−
Ile−ADT, (C) PEG−Gly−ADT, (D) ADT−OH, and (E) control (without treatment). Concentration of PEG−ADT, PEG−Ile−ADT, PEG−Gly−
ADT, and ADT−OH: 200 μM.
agents such as coenzyme NADH, L-cysteine, and glutathione,
which are present within the cell. Treating the PEG−ADT
conjugate with these compounds at pH 7.4 did not induce H2S
release (data not shown), showing its stability against reduction
but also hydrolysis. Other reducing agents such as hydrazine
and dithiothreitol were also not able to induce H2S release
(data not shown). Only when TCEP, a water-soluble
phosphine reducing agent, was used, H2S release was detected
(Figure 4). Running the experiment on a larger scale, the
To see whether serum proteins could induce H2S release, we
incubated the PEG−ADT conjugate, ADT−OH, and ADT−
OCH3 in 10% fetal bovine serum (FBS). Because it was
reported that ADT−OH releases H2S in the presence of FBS,44
we expected that the PEG−ADT conjugate would release H2S.
Contrary to our expectation, no H2S release was observed for
the PEG−ADT conjugate and ADT−OCH3 in 10% FBS
(Figure 5).
Only for ADT−OH, H2S release was observed as shown in
Supporting Information Figure S4. In addition, a rapid H2S
release was also measured when the experiment was repeated in
PBS. Strangely, we noticed that the signal was increasing
continuously and that it took a long time to equilibrate the
electrode after removing the reaction mixture. We therefore
suspected that the increase of the signal was caused by the
interaction of ADT−OH with the electrode. To confirm this,
we used lead acetate to detect released H2S from a solution of
ADT−OH as described for the PEG−ADT conjugate. After 1
week, only the ADT−OH solution containing TCEP showed a
black color due to the formation of PbS while no obvious color
change was observed for ADT−OH in PBS (Supporting
Information Figure S5). This suggests that H2S release from
ADT−OH in PBS is not significant or absent. On the basis of
these observations, we believe that the H2S release profile of
ADT−OH in PBS and 10% FBS is an artifact.
Figure 4. H2S release from the PEG−ADT conjugate (20 μM) in the
presence of TCEP·HCl (500 μM) in PBS (pH7.4) as measured by the
change in the current using a H2S electrode sensor. The arrow
indicates the addition of TCEP.
We next investigated whether intracellular enzymes could
induce H2S release. The PEG−ADT conjugate and ADT−
OCH3 were added to cell lysate from RAW-Blue macrophages
and H2S release was measured by the H2S electrode sensor. As
can be seen from Figure 5A, the PEG−ADT conjugate released
H2S in the presence of cell lysate. ADT−OCH3 also showed
H2S release, and the amount of released H2S was 10 times
higher than that of the PEG−ADT conjugate as shown in
Figure 5B. To show that the H2S release profiles for ADT−
OCH3 and PEG−ADT were not due to an artifact as was
observed for ADT−OH, two additional experiments were done.
We used the fluorescent H2S detection dye, WSP-1
(Supporting Information Scheme S1)45 and measured the
release of H2S was also indicated qualitatively by the typical
smell of H2S and the formation of brown lead sulfide (PbS)
after exposing the head space to filter paper impregnated with
lead acetate (Supporting Information Figure S3). ADT−OH
and ADT−OCH3 also released H2S in response to TCEP as
confirmed by the formation of PbS (Supporting Information
Figure S3). Because TCEP readily reduces disulfide bonds
within proteins, it is likely that the first step is reduction of the
−S-S− of the trithione followed up by another reaction that
finally releases H2S. Because L-cysteine and glutathione are not
able to induce H2S release from the PEG−ADT conjugate, it
seems unlikely that a simple chemical reduction can induce H2S
release in the body.
1293
dx.doi.org/10.1021/bc500150s | Bioconjugate Chem. 2014, 25, 1290−1300

Bioconjugate Chemistry
Article
Figure 5. H2S release from (A) the PEG−ADT conjugate and (B) ADT−OCH3 (50 μM) in 10% FBS and cell lysate as measured by the change in
the current using a H2S electrode sensor. The arrow indicate when PEG−ADT and ADT−OCH3 were added.
increase in fluorescence of ADT−OCH3 and PEG−ADT
solutions in cell lysate. For both ADT−OCH3 and PEG−ADT,
a higher fluorescence was measured compared to cell lysate
alone (Supporting Information Figure S6). The fluorescence
experiments showed a similar trend in which ADT−OCH3
showed higher release. Furthermore, we also detected H2S in
the gas phase using a similar setup as that used in the PbS
experiments (Supporting Information Figure S7). To increase
sensitivity, we used dansylazide instead of Pb(OAc)2 to detect
H2S. Dansylazide forms a fluorescent product with H2S and has
been used to quantify H2S in blood samples.46 As shown in
Supporting Information Figure S7, both ADT−OCH3 and
PEG−ADT showed higher H2S release compared to cell lysate
alone. These results suggest that an intracellular component(s),
presumably an enzyme, induces H2S release. In the case of
PEG−ADT, the ADT moiety is sterically hindered by PEG
from interacting with proteins, which may reduce the amount
of H2S released. Furthermore, it should be noted that the
experiments were not carried out under strict anaerobic
conditions. Thus, released H2S would be partially oxidized in
this experiment. It can be expected that oxidation of H2S might
become a significant competing reaction when the rate of H2S
generation is slow. Thus, the lower amount of H2S release from
the PEG−ADT conjugate is partially attributed to the slower
H2S generation for the PEG−ADT conjugate compared to
ADT−OCH3.
of ADT−OH. In case of ADT−OCH3, the H2S concentration
steadily increased for the first 3 h and then declined. For the
PEG−ADT conjugate, the increase of H2S concentration was
much lower compared to the small H2S donors, which is
consistent with the H2S release profile in the presence of cell
lysate (Figure 5).
Intracellular Distribution of the PEG−ADT Conjugate.
To see the intracellular distribution of H2S donors in RAW-
Blue macrophages, we prepared FITC-labeled ADT−OH and
PEG−ADT conjugate (for details, see Schemes S2 and S3 in
the Supporting Information). To visualize the endolysosomes,
Rhodamine-labeled dextran was used as a marker. While ADT−
OH was present homogeneously in the cytoplasm, the PEG−
ADT conjugate colocalized with the endolysosomes that
appeared as yellow spots in the merged images (Figure 7). It
is well-documented that amphiphilic small molecules can
diffuse across cellular membranes and directly enter the
cytoplasm.47 In contrast, macromolecules which normally
cannot pass through the lipid bilayer enter cells by endocytosis
and remain inside the endolysosome compartment.48 There-
fore, the different distribution between ADT−OH and the
PEG−ADT conjugate can be attributed to these size effects on
the cellular uptake behavior.
This altered intracellular distribution of the PEG−ADT
conjugate seems to correlate with the reduced toxicity relative
to ADT−OH. It has been reported that besides being a H2S
donor, ADT−OH as well as other trithione derivatives can
directly interact with intracellular enzymes and organelles and
exert biological activities including antiproliferative, chemo-
preventive, and anticarcinogenic properties.16−18,49 As these
interactions with intracellular components occur in the
cytoplasm, it can be expected that the side effects of ADT−
OH could be minimized by preventing its passive diffusion
through the cellular membranes to enter the cytoplasm. As
shown in Figure 7B, the PEG−ADT conjugate stayed inside the
endolysosomes without escaping to other cellular compart-
ments. Therefore, together with the cytotoxicity data, the
trapping of the ADT structure in the endolysosomes by the
PEG−ADT conjugate seems to be a key to prevent unfavorable
side effects of ADT−OH.
To observe H2S release under more relevant conditions, we
incubated the PEG−ADT conjugate with RAW-Blue macro-
phages. To determine the H2S concentration in the culture
medium, we used WSP-1. As can be seen in Figure 6, the H2S
concentration gradually increased up for 1 h upon the addition
Furthermore, the intracellular distribution of these polymeric
formulations also offers insight into the H2S release mechanism.
As shown in Figure 5, it is suggested that the H2S release was
induced by intracellular enzymes. Also, we showed that the
PEG−ADT conjugate was capable of releasing H2S in the
presence of RAW-Blue macrophages as shown in Figure 6.
These observations together with the intracellular distribution
data indicate that enzymes locating inside the endolysosomes
Figure 6. H2S release from ADT−OH, ADT−OCH3, and the PEG−
ADT conjugate in the presence of RAW-Blue macrophages.
Concentration: 100 μM.
1294
dx.doi.org/10.1021/bc500150s | Bioconjugate Chem. 2014, 25, 1290−1300

Bioconjugate Chemistry
Article
Figure 7. Intracellular distribution of fluorescently labeled ADT−OH (A) and the PEG−ADT conjugate (B) in RAW-Blue macrophages.
Concentration: 70 μM for ADT−OH, 100 μM for the PEG−ADT conjugate.
are probably responsible for H2S release from the PEG−ADT
conjugate although further investigation is necessary to
elucidate the exact mechanism.
compared to Na2S. This result may be due to the different H2S
release profiles of these donors. Furthermore, ADT−OH also
seems to promote inflammation upon LPS stimulation,
although its effect was not statistically significant. Because
ADT−OH is known to inhibit the transcriptional factor NF-kB
which plays a key role in immunological response to
pathogens,23 the reduced activity may be caused by the
biological effect of ADT−OH itself.
Effect of the PEG−ADT Conjugate on LPS-Induced
Inflammation. Several studies suggested that endogenously
produced H2S enhance pro-inflammatory responses in acute
inflammation such as sepsis although the role of H2S in
inflammation is complex, with both pro- and anti-inflammatory
effects reported.30 We investigated the effect of the PEG−ADT
conjugate in lipopolysaccharide (LPS)-induced inflammation.
LPS is a ligand for Toll-like receptor 4 (TLR4) found on
immune cells that initiate an immunological reaction upon
bacterial infection. RAW-Blue macrophages were stimulated
with LPS in the presence of the PEG−ADT conjugate, ADT−
OH, and Na2S, a sulfide salt known as an rapid H2S releaser.
The production of the pro-inflammatory cytokine tumor
necrosis factor (TNF-α) was quantified after 2 h as shown in
Figure 8.
CONCLUSION
■
We have designed and synthesized polymeric H2S donors based
on the small H2S donor ADT−OH. The PEG−Gly−ADT and
PEG−Ile−ADT conjugates readily hydrolyzed and released
ADT−OH, resulting in reduced cellular viability as shown by
the MTT and live/dead assay. On the other hand, the PEG−
ADT conjugate having a noncleavable ether bond did not show
obvious cellular toxicity. This reduced toxicity seems to
correlate with the colocalization of the PEG−ADT conjugate
with the endolysosomes without escaping to the cytoplasm.
The PEG−ADT conjugate was capable of releasing H2S in the
presence of murine macrophages. Furthermore, PEG−ADT
was capable of potentiating LPS-induced inflammation. The
PEG−ADT conjugate may provide a tool that can give
complementary information for a better understanding of the
H2S donor ADT−OH within cells.
EXPERIMENTAL PROCEDURES
■
Materials. Methoxy-polyethylene glycol with a nominal
molecular weight of 5000 (mPEG114OH), isoleucine (Ile),
NADH, L-cysteine, (reduced) glutathione, hydrazine, TCEP·
HCl, and (2-(2-(2-methoxyethoxy)ethoxy)ethoxy)p-toluenesul-
fonate were from Sigma-Aldrich. Carbonyldiimidazole (CDI),
dimethyl sulfoxide (DMSO), triethylamine (Et3N), glycine
(Gly), sodium hydrogensulfate hydrate (NaHSO4·H2O),
anhydrous sodium sulfate (Na2SO4), anhydrous potassium
carbonate (K2CO3), dicyclohexylcarbodiimide (DCC), ninhy-
drin, potassium hydroxide (KOH), diethyl ether (Et2O), and
anhydrous dimethylformamide (DMF) were from Wako Pure
Chemical Industry. 4-Dimethylaminopyridine (DMAP) and
thiazolyl Blue tetrazolium bromide (MTT) were from Tokyo
Chemical industry (TCI). Dichloromethane (CH2Cl2), chloro-
form (CHCl3), ethyl acetate (EtOAc), and molecular sieves 3A
were from Nacalai Tesque. LPS-free water was purchased from
Figure 8. Effect of the different H2S donors (50 μM) on TNF-α
production upon lipopolysaccharide (LPS) stimulation. Solid column,
LPS (+); empty column, LPS (−). ** p < 0.01 versus control, LPS
(+).
Upon the addition of Na2S and the PEG−ADT conjugate,
the TNF-α production was significantly enhanced in RAW-Blue
macrophages stimulated with LPS, whereas the H2S donors by
themselves (without LPS stimulation) did not increase TNF-α
levels compared to the control. In addition, a much higher
TNF-α level was observed for the PEG−ADT conjugate
1295
dx.doi.org/10.1021/bc500150s | Bioconjugate Chem. 2014, 25, 1290−1300

Bioconjugate Chemistry
Article
Otsuka Pharmaceutical Factory. DMEM GlutaMax, fetal bovine
serum (FBS), penicillin−streptomycin, and live/dead cell
viability assay kit were purchased from Invitrogen. Lipo-
polysaccharide (LPS) was purchased from Invivogen. Passive
lysis buffer was purchased from Promega. WSP-1 was
purchased from Cayman Chemical. RAW-Blue macrophages,
QUANTI-Blue, and Zeocin were purchased from Invivogen.
Triple-well glass-based dishes and 96-well microplates were
purchased from Iwaki. Water was freed from salt using the
Milli-Q water system.
calculated from the integral values of the methyl and methoxy
signal, was 93%.
PEG−Gly−ADT (7). First, 513 mg (0.1 mmol) of (5) and
24.6 mg (0.11 mmol, 1.1 equiv) of (1) were dissolved in 10 mL
of CH2Cl2. To the solution was added 22.9 mg (0.11 mmol,
0.11 equiv) of DCC in 1 mL of CH2Cl2 followed by 1.3 mg
(0.011 mmol, 0.11 equiv) of DMAP in 109 μL of CH2Cl2. The
mixture was stirred for 28 h and concentrated under reduced
pressure. The residue was suspended in 25 mL of 1 M NaHSO4
(aq) and filtered to remove DCU. The aqueous phase was
extracted with CH2Cl2 (3 × 25 mL) and after combining dried
over Na2SO4. The clear solution was concentrated to about 10
mL and diluted with Et2O (400 mL). The orange precipitate
was filtered, washed with Et2O (100 mL), and dried under
reduced pressure to yield 414 mg (81%) of an orange solid. 1H
NMR (DMSO-d6) δ = 7.99 (d, 2H, 2 × CHaromat ADT), 7.85
(bs, 1H, NH), 7.83 (s, 1H, CHC ADT), 7.31 (d, 2H, 2 ×
CHaromat ADT), 4.11 (s, 2H, NH-CH2-C(O)), 4.08 (m, 2H,
CH2-C(O)-NH), 3.76−3.53 (m, CH2CH2O), 3.24 (s, 3H,
OCH3). The degree of functionalization, calculated from the
integral values of the ADT and methoxy signals, was 95%.
PEG−Ile−ADT (8). First, 507 mg (0.1 mmol) of (6) and 33.1
mg (0.15 mmol, 0.15 equiv) of (1) were dissolved in 10 mL of
CH2Cl2 and 1.8 mg (0.015 mmol, 0.15 equiv) of DMAP in 375
μL of CH2Cl2 and 33.7 mg (0.16 mmol, 0.16 equiv) of DCC in
1 mL of CH2Cl2 were added. After stirring, the mixture for 19
h, the solution was concentrated and the residue suspended in
25 mL of 1 M NaHSO4 (aq). After filtering, the clear solution
was extracted with CH2Cl2 (3 × 25 mL), dried over Na2SO4
concentrated to a small volume, and diluted with 400 mL Et2O.
The solid was collected, washed with 100 mL of Et2O, and
dried under reduced pressure to yield 486 mg (96%) of orange
solid. 1H NMR (DMSO-d6) δ = 7.99 (d, 3H, 2 × CHaromat ADT
+ NH), 7.83 (s, 1H, CHC ADT), 7.31 (d, 2H, 2 × CHaromat
ADT), 4.19 (m, 1H, NH-CH-C(O)O), 4.12 (m, 2H, CH2-
C(O)-NH), 3.76−3.35 (m, CH2CH2O), 3.24 (s, 3H, OCH3),
1.95 (m, 1H, CHCH3), 1.70 (m, 1H, CH-CH2-CH3), 1.52 (m,
1H, CH-CH2-CH3), 0.98 (d, 3H, CHCH3), 0.89 (t, 3H, CH2-
CH3). The degree of functionalization, calculated from the
integral values of ADT and the methoxy signals, was 63%.
mPEG114−Mesylate (9). 9 was prepared according to that
reported.50 First, 21.0 g (4.2 mmol) of (3) was dissolved in 600
mL of toluene and dried by azeotropic distillation using a
Dean−Stark trap. After 80 mL of toluene/H2O was collected,
the mixture was cooled to 30 °C. To the clear solution was
added 1.6 mL (21 mmol, 5 equiv) of methanesulfonyl chloride
and 2.9 mL (21 mmol, 5 equiv) of Et3N. The mixture became
hazy within 1 min. After 19 h, the mixture was filtered through
SiO2 (2.5 cm diameter × 2.0 cm height) and concentrated
under reduced pressure. The oily residue was dissolved in 25
mL of CH2Cl2 and added to 1 L of Et2O. The white precipitate
was collected, washed with Et2O (2 × 100 mL), and dried
Synthesis. All reactions were done in an inert argon
atmosphere, and chemicals were used as received unless stated
otherwise. Molecular sieves were dried at 180 °C under
reduced pressure. Et3N was distilled from ninhydrin and KOH
and stored over molecular sieves. CH2Cl2 was dried over
molecular sieves.
ADT−OH (1). ADT−OH (1) was prepared as reported.34
mPEG114−CDI (4). First, 10.9 g (2.2 mmol) of mPEG114
−
OH (3) was heated at 80 °C under reduced pressure to remove
water. After cooling down to RT, the solid was dissolved in 40
mL of DMSO and 1.68 g (10.4 mmol, 4.8 equiv) of CDI was
added to the clear solution. After stirring for 24 h, the solution
was diluted with 500 mL of EtOAc and put at −20 °C. After 2
h, a white solid precipitated that was filtered and washed with
Et2O (2 × 100 mL). After drying under reduced pressure, this
1
yielded 10.2 g (94%) of a white solid. H NMR (DMSO-d6) δ
= 8.25 (s, 1H, CHimidazole), 7.59 (s, 1H, CHimidazole), 7.08 (s, 1H,
CHimidazole), 4.50 (m, 2H, CH2-CO-N), 3.76−3.35 (m,
CH2CH2O), and 3.24 (s, 3H, OCH3). The degree of
functionalization, calculated from the integral values of the
imidazole and the methoxy signals, was 100%.
mPEG114−Gly (5). First, 1.0 g (0.21 mmol) (4) was dissolved
in 5 mL of DMSO and to the clear solution was added 75.9 mg
(1.0 mmol, 4.9 equiv) of glycine followed by 141 μL (1.0
mmol, 4.9 equiv) of Et3N and the mixture was heated at 80 °C.
After 29 h, the yellow mixture was diluted with 25 mL of 1 M
NaHSO4 (aq) and extracted with CH2Cl2 (3 × 25 mL). After
drying over Na2SO4, the solution was concentrated to a small
volume, diluted with 200 mL of EtOAc, and kept at −20 °C for
2 h. The precipitate was filtered, washed with EtOAc (25 mL)
and Et2O (100 mL) and dried under reduced pressure to yield
897 mg (87%) of a white solid. 1H NMR (DMSO-d6) δ = 12.5
(bs, 1H, CO2H), 7.41 (bs, 1H, NH), 4.05 (m, 2H, CH2-C(O)),
3.76−3.35 (m, CH2CH2O + NH-CH2-CO2H), 3.24 (s, 3H,
OCH3). The degree of functionalization, calculated from the
integral values of the CH2-C(O) and methoxy signals, was 91%.
mPEG114−Ile (6). First, 1.0 g (0.21) mmol (4) was dissolved
in 10 mL of DMSO, and then to the clear solution was added
262 mg (2.0 mmol, 9.5 equiv) of isoleucine followed by 279 μL
(2.1 mmol, 10 equiv) of Et3N and the mixture heated at 80 °C.
After 23 h, the yellow mixture was diluted with 25 mL of 1 M
NaHSO4 (aq) and extracted with CH2Cl2 (3 × 25 mL). After
drying over Na2SO4, the clear solution was concentrated under
reduced pressure to give a liquid that was diluted with 200 mL
of EtOAc and placed at −20 °C for 3 h. The precipitate was
filtered and washed with 200 mL of Et2O and dried under
reduced pressure and yielded 925 mg (88%). 1H NMR
(DMSO-d6) δ = 12.5 (bs, 1H, CO2H), 7.43 (bs, 1H, NH), 4.05
(m, 2H, CH2-C(O)), 3.85 (m, 1H, NH-CH-CO2H), 3.76−3.53
(m, CH2CH2O), 3.24 (s, 3H, OCH3), 1.76 (m, 1H, CHCH3),
1.38 (m, 1H, CH-CH2-CH3), 1.24 (m, 1H, CH-CH2-CH3),
0.84 (m, 6H, 2 × CH3). The degree of functionalization,
1
under reduced pressure to yield 19.5 g (3.9 mmol, 92%). H
NMR (CDCl3): δ = 4.36 (m, 2H, CH2OSO2), 3.81−3.43 (m,
CH2CH2O PEG), 3.36 (s, 3H, OCH3 PEG), 3.07 (s, 3H,
SO2CH3). The degree of functionalization, calculated from the
integral values of the mesylate and the methoxy signals, was
100%.
PEG−ADT (11). First, 1.0 g (0.2 mmol) of (9), 91 mg (0.4
mmol, 2 equiv) of (1), and 54 mg (0.4 mmol, 2 equiv) of
K2CO3 were dissolved in 10 mL of DMF and heated at 80 °C
for 24 h. The mixture was then cooled to RT and concentrated
under reduced pressure. The residue was partitioned in 50 mL
1296
dx.doi.org/10.1021/bc500150s | Bioconjugate Chem. 2014, 25, 1290−1300

Bioconjugate Chemistry
Article
of CH2Cl2 and 50 mL of 1 M NaHSO4 (aq). After separating
both layers, the aqueous phase was extracted with CH2Cl2 (2 ×
50 mL) and the combined extracts were dried over Na2SO4 and
concentrated under reduced pressure to 20 mL. The solution
was then added to 300 mL of Et2O, and the orange solid was
collected. After washing with Et2O (4 × 50 mL), the solid was
dried under reduced pressure, yielding 817 mg (0.16 mmol,
82%). 1H NMR (CDCl3): δ = 7.60 (d, 2H, 2 × CHaromat ADT),
7.38 (s, 1H, CHC ADT), 6.99 (d, 2H, 2 × CHaromat ADT),
4.18 (t, 2H, CH2OPh), 3.88−3.47 (m, CH2CH2O PEG), 3.37
(s, 3H, OCH3 PEG). The degree of functionalization,
calculated from the integral values of the ADT and methoxy
signals, was 89%.
placed on top. Solutions containing ADT−OH and ADT−
OCH3 at 50 μM in 1% DMSO/PBS and without H2S donor
were reacted under the same conditions. After 1 h, H2S release
was confirmed by the formation of PbS on the filter paper. To
confirm the absence of H2S generation from ADT−OH in PBS,
the reaction was repeated on a larger scale. A 15 mL Falcon
tube was filled with 10 mL of a 50 μM solution in PBS
containing ADT−OH or ADT−OCH3 and reacted with or
without TCEP at 500 μM final concentration. Upon mixing, the
tubes were closed immediately so that any released H2S could
be directly trapped with a Pb(OAc)2-impregnated filter paper
inserted at the top and not in contact with the solution. After 1
week, H2S release was confirmed by the formation of PbS on
the filter paper.
Cell Culture. RAW-Blue macrophages were cultured in
DMEM GlutaMAX supplemented with 10% heat-inactivated
fetal bovine serum, 50 U/mL to 50 μg/mL penicillin−
streptomycin and 200 mg/mL Zeocin in a CO2 incubator at
37 °C. Cells were passaged when reaching 70−80% confluency.
Preparation of Cell Lysate. RAW-Blue macrophages at 80%
confluency were scraped off and centrifuged at 500 rpm for 5
min. The cell pellet was washed with cold Dulbecco’s
phosphate buffered saline (PBS) three times. The cell
suspension was centrifuged at 500 rpm for 5 min, resuspended
in passive lysis buffer at 1 × 107 cells/mL, and vortexed at RT
for 10 min. The suspension was centrifuged, and the clear
supernatant was collected and stored at −20 °C.
(2-(2-(2-Methoxyethoxy)ethoxy)ethoxy)−ADT (TEG−ADT,
12). First, 129 mg (0.41 mmol) (2-(2-(2-methoxyethoxy)-
ethoxy)ethoxy)p-toluenesulfonate, 102 mg (0.45 mmol) (1),
and 63 mg (0.45 mmol) K2CO3 were dissolved in 5 mL of
DMF and heated at 80 °C for 18 h. After cooling down to RT,
the mixture was then concentrated and 50 mL of CHCl3 added.
The suspension was filtered, and the residue was washed with
CHCl3 (2 × 50 mL). The filtrate was concentrated and the
residue purified by SiO2 column chromatography (Rf = 0.46,
EtOAc/hexane 4:1). This yielded 125 mg (0.34 mmol, 82%) of
1
a red oil. H NMR (CDCl3): δ = 7.60 (d, 2H, 2 × CHaromat
ADT), 7.39 (s, 1H, CHC ADT), 6.99 (d, 2H, 2 × CHaromat
ADT), 4.19 (t, 2H, CH2O), 3.88 (t, 2H, CH2), 3.74 (t, 2H,
CH2), 3.68 (t, 2H, CH2), 3.65 (m, 2H, CH2), 3.50 (t, 2H,
CH2), 3.38 (s, 3H, OCH3). MS (m/z, M + Na+, ESI-TOF-MS):
calculated for C16H19O3S3 395.0415, found 395.0406.
Measurement of H2S Release in Cell Lysate by H2S
Electrode Sensor. Degassed PBS containing 10 vol % FBS or
20 vol % cell lysate (1 mL) was placed in a four-port closed
chamber and incubated at 37 °C. After the sensor signal
became stable, 50 μL of the PEG−ADT conjugate solutions
was added (final concentration: 50 μM) using a Hamilton
syringe. Released H2S was recorded on a four-channel free
radical analyzer.
Dansylazide. Dansylazide was prepared as reported.46
Time Course Hydrolysis Experiment. Stock solutions of
PEG−Gly−ADT and PEG−Ile−ADT at 420 μM were
prepared in 10 mM PBS (pH 7.4). The stock solutions were
then split into several 15 mL of Falcon tubes and put at 37 °C
in a bioshaker. At different time points, a tube was taken out,
acidified with acetic acid 1% (v/v), frozen, and stored at −20
°C. After the last sample was collected, all samples were
lyophilized. To each tube was then added 800 μL of DMSO-d6
and put on a sample rotator for 6 h. After spinning down
insoluble salts, the clear supernatant was measured by 1H NMR
to determine the extent of hydrolysis.
Chemical Detection of H2S with Dansylazide. To detect
H2S release qualitatively from cell lysate, dansylazide was used.
The azide is reduced by H2S and generates a fluorescent
product that can be readily visualized with an UV lamp.46 To
validate this method, the experiment of H2S detection from
ADT−OCH3 in the presence of TCEP was repeated with filter
paper impregnated with 100 μL of saturated solution of
dansylazide in 0.5% Tween/PBS pH 7.4. For detection of H2S
from ADT−OCH3 and PEG−ADT in the presence of cell
lysate, the following setup was used. A 2 mL glass HPLC vial
with a stirring bar was filled with 670 μL (ADT−OCH3) or 900
μL (PEG−ADT) of 20 vol % cell lysate/PBS pH 7.4 and 6.7 μL
of ADT−OCH3 in DMSO or 9.0 μL of PEG−ADT in H2O was
added (final concentration 50 μM). As control, 670 and 900 μL
of 20 vol % cell lysate/PBS pH 7.4 solutions without donor
were prepared. The Teflon seal of the sample vial was replaced
with a filter paper impregnated with 25 μL of a saturated
solution of dansylazide in 0.5% Tween/PBS pH 7.4. After
attaching the screw cap, the vial was transferred to a larger vial
and capped. The mixture was stirred for 24 h at RT. The filter
papers were removed, put in small transparent plastic bags, and
visualized with a Biorad Gel Doc EZ imager.
Extraction of ADT−OH from Cell Cultures. RAW-Blue were
seeded at 100000 cells/well with 200 μL of medium and
cultured overnight. The supernatant was replaced with 200 μL
of medium containing PEG−Gly−ADT and PEG−Ile−ADT at
1 mM. The cells were cultured for 24 h, after which the well
plate was frozen and lyophilized. For each conjugate, three
wells were extracted with 3 mL of MeOH. The yellow mixture
was concentrated under reduced pressure, and the orange
residue acidified with 5 mL of 1 M NaHSO4 (aq) and extracted
with CH2Cl2 (3 × 5 mL). After drying over Na2SO4, the yellow
solution was concentrated under reduced pressure. The residue
1
was dissolved in 700 μL of DMSO-d6 and the H NMR
spectrum recorded (Supporting Information Figure S2).
Chemical Detection of H2S with Lead Acetate. To get
qualitative chemical evidence for H2S released in the presence
of TCEP, we used the qualitative test for H2S with lead acetate
(Pb(OAc)2). H2S reacts with Pb2+ to form PbS, which has a
brownish to black color depending on the amount of H2S. To a
1.5 mL Eppendorf tube with 1.0 mL of PBS (pH 7.4)
containing 50 μM PEG−ADT was added TCEP (aq) to give a
final concentration of 500 μM. Immediately, a Whatman paper
impregnated with 100 μL of a 50 mM Pb(OAc)2 solution was
Measurement of H2S Release in Cell Lysate by the
Fluorescent Dye WSP-1. To a black well polystyrene plate
containing 100 μL of 20 vol % cell lysate/PBS pH 7.4 were
added 1 μL of ADT−OCH3 in DMSO or 1 μL of PEG−ADT
in H2O (final concentration 50 μM) and 10 μL of WSP-1 in
acetonitrile (final concentration 100 μM). As a control, cell
1297
dx.doi.org/10.1021/bc500150s | Bioconjugate Chem. 2014, 25, 1290−1300

Bioconjugate Chemistry
Article
lysate without donor was measured as well. Immediately after
adding, the fluorescence was measured on a Tecan well plate
reader (λex = 465 nm, λem = 515 nm).
UV/Vis Spectroscopy. Absorbance and spectra were
measured on a Tecan infinite M200 in transparent polystyrene
96-well plates.
H2S Electrode Sensor. H2S was measured by monitoring the
change in current in pA with a World Precision Instruments
ISO-H2S-2 hydrogen sulfide sensor coupled to a four-channel
free radical analyzer. The measurement chamber was kept at 37
°C using a thermostatic bath, and the temperature was
measured with a temperature sensor.
High Resolution Mass Spectrometry. ESI-TOF MS analyses
were performed on a Bruker microTOFII mass spectrometer.
Confocal Laser Scanning Fluorescence Microscopy
(CLSFM). Fluorescent images were acquired on an Olympus
FluoView FV1000-D confocal microscope equipped with 405,
473, 559, and 635 nm lasers.
Measurement of H2S Concentration in Cell Culture
Medium by the Fluorescent Dye WSP-1. RAW-Blue macro-
phages were seeded in a 96-well plate (5 × 104 cells/well) and
cultured for 1 d. The medium was replaced with 100 μL/well of
fresh medium and 1 μL/well of ADT−OH in DMSO, 5 μL/
well of the PEG−ADT conjugate in water were added (final
concentration of ADT: 100 μM). At different time points, 5 μL
of medium was withdrawn and diluted with 45 μL of PBS. This
sample solution was immediately mixed with 50 μL of WSP-1/
DMSO (100 μM). A serial dilution of Na2S in PBS was used as
a standard. Fluorescence intensity (λex = 465 nm, λem = 515
nm) was measured on a Tecan well plate reader.
Cell Viability Assay by the MTT Assay. RAW-Blue
macrophages were seeded in a 96-well plate (1 × 105 cells/
well) and cultured for 1 d. The medium was replaced with 100
μL/well of fresh medium and 1 μL/well of ADT−OH in
DMSO and 5 μL/well of the PEG−ADT conjugate in LPS-free
water with different concentrations were added. Cells were
cultured for 1 d in a CO2 incubator at 37 °C. Thereafter, the
medium was replaced with 100 μL/well fresh medium and 10
μL of MTT solution (5 mg/mL in PBS) was added to each
well. The plate was incubated for 2 h at 37 °C. Then, 100 μL of
0.1 g/mL sodium dodecyl sulfate in 0.01 M HCl (aq) was
added to each well to lyse cells and solubilize formazan crystals.
The OD at 570 nm was measured.
Live/Dead Assay. RAW-Blue macrophages were seeded in a
triple-wells glass-based dish (5 × 104 cells/well) and cultured
for 1 d. The medium was replaced with 200 μL/well of fresh
medium containing 2 μL of ADT−OH in DMSO and 20 μL of
the PEG−Gly−ADT, PEG−Ile-ADT, and PEG−ADT con-
jugates in LPS-free water were added (final concentration: 200
μM). After 1 d of culture, cells were washed with PBS and
stained using the live/dead cell vitality assay kit according to
the manufacturer’s procedure. Cells were observed with an
Olympus FluoView FV1000-D confocal microscope.
Observation of Intracellular Distribution of H2S Donors.
RAW-Blue macrophages were seeded in a triple-wells glass-
based dish (5 × 104 cells/well) and cultured for 1 d. The
medium was replaced with 200 μL/well of fresh medium
containing 1 mg/mL rhodamine-labeled dextran (10 kDa) and
2 μL of FITC-labeled ADT−OH in DMSO (final concen-
tration: 70 μM) and 20 μL of FITC-labeled PEG−ADT
conjugate in water (final concentration: 100 μM) were added.
After 4 h of culture, cells were washed with PBS and 100 μL/
well of fresh medium was added. Cells were observed with an
Olympus FluoView FV1000-D confocal microscope.
ASSOCIATED CONTENT
■
S
* Supporting Information
UV/vis spectra of PEG−Gly−ADT, PEG−ADT, and ADT−
1
OH, H NMR spectra of hydrolysis PEG−Gly−ADT, and
PEG−Ile−ADT in cell culture, qualitative detection of H2S by
lead acetate and dansylazide, structure of WSP-1, synthesis of
FITC-labeled ADT−OH and PEG−ADT conjugate. This
material is available free of charge via the Internet at http://
pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*Phone: +81-6-6879-7365. Fax: +81-6-6879-7367. E-mail:
urara.hasegawa@chem.eng.osaka-u.ac.jp. Address: 2-1, Yama-
daoka, Suita, Osaka 565-0871, Japan.
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank Prof. Hiroshi Uyama as well as Prof. Takashi Hayashi
and Prof. Koji Oohora (Osaka University, Japan) for ESI-TOF
MS analysis. This work was supported by Grant-in-Aid for
Young Scientists (B), no. 24700482, from the Japan Society for
the Promotion of Science.
REFERENCES
■
(1) Wang, R. (2002) Two’s company, three’s a crowd: can H2S be
the third endogenous gaseous transmitter? FASEB J. 16, 1792−1798.
(2) Beauchamp, R., Bus, J., Popp, J., Boreiko, C., and Andjelkovich,
D. (1984) A critical review of the literature on hydrogen sulfide
toxicity. Crit. Rev.Toxicol. 13, 25−97.
(3) Reiffenstein, R., Hulbert, W., and Roth, S. (1992) Toxicology of
hydrogen sulfide. Annu. Rev. Pharmacol. Toxicol. 32, 109−134.
(4) Qu, K., Lee, S. W., Bian, J. S., Low, C. M., and Wong, P. T. H.
(2008) Hydrogen sulfide: neurochemistry and neurobiology. Neuro-
chem. Int. 52, 155−165.
(5) Caliendo, G., Cirino, G., Santagada, V., and Wallace, J. L. (2010)
Synthesis and biological effects of hydrogen sulfide (H2S): Develop-
ment of H2S-releasing drugs as pharmaceuticals. J. Med. Chem. 53,
6275−6286.
Lipopolysaccharide-Induced Inflammation. RAW-Blue
cells were seeded at 5 × 104 cells/well in a 96-well plate and
cultured for 1 d. The medium was replaced with 100 μL/well
fresh medium containing 1 μL of ADT−OH in DMSO, 10 μL
of Na2S, and 10 μL of the PEG−ADT conjugate in LPS-free
water were added (final concentration: 50 μM). Cells were
stimulated with 1 μg/mL LPS for 2 h, and the supernatant was
analyzed by ELISA. Statistical analysis was performed using the
Student t test.
(6) 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., and Akaike,
T. (2014) Reactive cysteine persulfides and S-polythiolation regulate
oxidative stress and redox signaling. Proc. Natl. Acad. Sci. U.S.A. 111,
7606−7611.
1
Instrumentation. NMR Spectroscopy. H NMR spectra
were measured with a Bruker DPX400 NMR spectrometer. A
total of 32 scans were collected, and the D1 was set to 10 s.
Chemical shifts are referenced to residual undeuterated NMR
solvent signal at 2.50 ppm (DMSO-d6) and 7.26 ppm (CDCl3).
(7) Kimura, H. (2014) The physiological role of hydrogen sulfide
and beyond, Nitric Oxide, doi: 10.1016/j.niox.2014.01.002.
1298
dx.doi.org/10.1021/bc500150s | Bioconjugate Chem. 2014, 25, 1290−1300

Bioconjugate Chemistry
Article
(8) Wagner, F., Asfar, P., Calzia, E., Radermacher, P., and Szabo, C.
(2009) Bench-to-bedside review: hydrogen sulfide, the third gaseous
transmitter: applications for critical care. Crit. Care 13, 213.
(9) Szabo, C. (2007) Hydrogen sulphide and its therapeutic
potential. Nature Rev. Drug Discovery 6, 917−935.
(27) Ringsdorf, H. (1975) Structure and properties of pharmaco-
logically active polymers. J. Polym. Sci., Polym. Symp. 51, 135−153.
(28) Maeda, H. (2001) The enhanced permeability and retention
(EPR) effect in tumor vasculature: the key role of tumor-selective
macromolecular drug targeting. Adv. Enzyme Regul. 41, 189−207.
(29) Kashfi, K., and Olson, K. R. (2013) Biology and therapeutic
potential of hydrogen sulfide and hydrogen sulfide-releasing chimeras.
Biochem. Pharmacol. 85, 689−703.
(10) Olson, K. R. (2011) The therapeutic potential of hydrogen
sulfide: separating hype from hope. Am. J. Physiol.: Regul., Integr. Comp.
Physiol. 301, R297−R312.
(30) Whiteman, M., and Winyard, P. G. (2011) Hydrogen sulfide
and inflammation: the good, the bad, the ugly and the promising.
Expert Rev. Clin. Pharmacol. 4, 13−32.
́ ́
(11) Greiner, R., Palinkas, Z., Basell, K., Becher, D., Antelmann, H.,
̈
Nagy, P., and P, D. T. (2013) Polysulfides link H2S to protein thiol
oxidation. Antioxid. Redox. Signal. 19, 1749−1765.
(31) Fiorucci, S., Distrutti, E., Cirino, G., and Wallace, J. L. (2006)
The emerging roles of hydrogen sulfide in the gastrointestinal tract
and liver. Gastroenterology 131, 259−271.
(12) Li, L., Whiteman, M., Guan, Y. Y., Neo, K. L., Cheng, Y., Lee, S.
W., Zhao, Y., Baskar, R., Tan, C.-H., and Moore, P. K. (2008)
Characterization of a novel, water-soluble hydrogen sulfide-releasing
molecule (GYY4137): new insights into the biology of hydrogen
sulfide. Circulation 117, 2351−2360.
(32) Fiorucci, S., Orlandi, S., Mencarelli, A., Caliendo, G., Santagada,
V., Distrutti, E., Santucci, L., Cirino, G., and Wallace, J. L. (2007)
Enhanced activity of a hydrogen sulphide-releasing derivative of
mesalamine (ATB-429) in a mouse model of colitis. Br. J. Pharmacol.
150, 996−1002.
(13) Long, M. J. D., Dolan, P., Santamaria, A. B., and Bueding, E.
(1986) 1, 2-Dithiol-3-thione analogs: effects on NAD(P)H: quinone
reductase and glutathione levels in murine hepatoma cells. Carcino-
genesis 7, 977−980.
(14) Zhang, Y., and Munday, R. (2008) Dithiolethiones for cancer
chemoprevention: where do we stand? Mol. Cancer Ther. 7, 3470−
3479.
(33) Sparatore, A., Perrino, E., Tazzari, V., Giustarini, D., Rossi, R.,
Rossoni, G., Erdman, K., Schroder, H., and Soldato, P. D. (2009)
̈
Pharmacological profile of a novel H2S-releasing aspirin. Free Radical
Biol. Med. 46, 586−592.
(34) Li, L., Rossoni, G., Sparatore, A., Lee, L. C., Del Soldato, P., and
Moore, P. K. (2007) Anti-inflammatory and gastrointestinal effects of
a novel diclofenac derivative. Free Radical Biol. Med. 42, 706−719.
(35) Wallace, J. L., Caliendo, G., Santagada, V., and Cirino, G. (2010)
Markedly reduced toxicity of a hydrogen sulphide-releasing derivative
of naproxen (ATB-346). Br. J. Pharmacol. 159, 1236−1246.
(36) Distrutti, E., Sediari, L., Mencarelli, A., Renga, B., Orlandi, S.,
Russo, G., Caliendo, G., Santagada, V., Cirino, G., Wallace, J. L., and
Fiorucci, S. (2006) 5-Amino-2-hydroxybenzoic acid 4-(5-thioxo-5H-
[1,2]dithiol-3yl)-phenyl ester (ATB-429), a hydrogen sulfide-releasing
derivative of mesalamine, exerts antinociceptive effects in a model of
postinflammatory hypersensitivity. J. Pharmacol. Exp. Ther. 319, 447−
458.
(37) Perrino, E., Cappelletti, G., Tazzari, V., Giavini, E., Soldato, P.
D., and Sparatore, A. (2008) New sulfurated derivatives of valproic
acid with enhanced histone deacetylase inhibitory activity. Bioorg. Med.
Chem. Lett. 18, 1893−1897.
(38) Tazzari, V., Cappelletti, G., Casagrande, M., Perrino, E., Renzi,
L., Del Soldato, P., and Sparatore, A. (2010) New aryldithiolethione
derivatives as potent histone deacetylase inhibitors. Bioorg. Med. Chem.
18, 4187−4194.
(15) Nagano, T., and Takeyama, M. (2001) Enhancement of salivary
secretion and neuropeptide (substance P, α-calcitonin gene-related
peptide) levels in saliva by chronic anethole trithione treatment. J.
Pharm. Pharmacol. 53, 1697−1702.
(16) Lam, S., MacAulay, C., le Riche, J. C., Dyachkova, Y., Coldman,
A., Guillaud, M., Hawk, E., Christen, M.-O., and Gazdar, A. F. (2002)
A randomized phase IIb trial of anethole dithiolethione in smokers
with bronchial dysplasia. J. Natl. Cancer Inst. 94, 1001−1009.
(17) Reddy, B. S., Rao, C. V., Rivenson, A., and Kelloff, G. (1993)
Chemoprevention of colon carcinogenesis by organosulfur com-
pounds. Cancer Res. 53, 3493−3498.
(18) Lubet, R. A., Steele, V. E., Eto, I., Juliana, M. M., Kelloff, G. J.,
and Grubbs, C. J. (1997) Chemopreventive efficacy of anethole
trithione, N-acetyl-L-cysteine, miconazole and phenethylisothiocyanate
in the DMBA-induced rat mammary cancer model. Int. J. Cancer 72,
95−101.
(19) Drukarch, B., Schepens, E., Stoof, J. C., and Langeveld, C. H.
(1997) Anethole dithiolethione prevents oxidative damage in
glutathione-depleted astrocytes. Eur. J. Pharmacol. 329, 259−262.
(20) Mansuy, D., Sassi, A., Dansette, P. M., and Plat, M. (1986) A
new potent inhibitor of lipid peroxidation in vitro and in vivo, the
hepatoprotective drug anisyldithiolthione. Biochem. Biophys. Res.
Commun. 135, 1015−1021.
(21) Warnet, J. M., Christen, M. O., Thevenin, M., Biard, D.,
Jacqueson, A., and Claude, J. R. (1989) Protective effect of anethol
dithiolthione against acetaminophen hepatotoxicity in mice. Pharma-
col. Toxicol. 65, 63−64.
(22) Chen, P., Luo, Y., Hai, L., Qian, S., and Wu, Y. (2010) Design,
synthesis, and pharmacological evaluation of the aqueous prodrugs of
desmethyl anethole trithione with hepatoprotective activity. Eur. J.
Med. Chem. 45, 3005−3010.
(23) Sen, C. K., Traber, K. E., and Packer, L. (1996) Inhibition of
NF-κB activation in human T-cell lines by anetholdithiolthione.
Biochem. Biophys. Res. Commun. 218, 148−153.
(39) Benavides, G. A., Squadrito, G. L., Mills, R. W., Patel, H. D.,
Isbell, T. S., Patel, R. P., Darley-Usmar, V. M., Doeller, J. E., and Kraus,
D. W. (2007) Hydrogen sulfide mediates the vasoactivity of garlic.
Proc. Natl. Acad. Sci. U. S. A. 104, 17977−17982.
(40) Zhao, Y., Wang, H., and Xian, M. (2010) Cysteine-activated
hydrogen sulfide (H2S) donors. J. Am. Chem. Soc. 133, 15−17.
(41) Zanatta, S. D., Jarrott, B., and Williams, S. J. (2010) Synthesis
and preliminary pharmacological evaluation of aryl dithiolethiones
with cyclooxygenase-2-selective inhibitory activity and hydrogen
sulfide-releasing properties. Aust. J. Chem. 63, 946−957.
(42) Lee, M., Tazzari, V., Giustarini, D., Rossi, R., Sparatore, A., Del
Soldato, P., McGeer, E., and McGeer, P. L. (2010) Effects of hydrogen
sulfide-releasing L-DOPA derivatives on glial activation: potential for
treating parkinson disease. J. Biol. Chem. 285, 17318−17328.
(43) Sugii, A., Kabasawa, Y., and Akamatsu, T. (1968) Polarographic
studies on pharmaceuticals. II: Polarography of the anethole trithione.
J. Pharm. Soc. Jpn. 88, 1371−1374.
(44) Yu, F., Li, P., Song, P., Wang, B., Zhao, J., and Han, K. (2012)
An ICT-based strategy to a colorimetric and ratiometric fluorescence
probe for hydrogen sulfide in living cells. Chem. Commun. 48, 2852−
2854.
(45) Liu, C., Pan, J., Li, S., Zhao, Y., Wu, L. Y., Berkman, C. E.,
Whorton, A. R., and Xian, M. (2011) Capture and visualization of
hydrogen sulfide by a fluorescent probe. Angew. Chem., Int. Ed. 50,
10327−10329.
(24) Switzer, C. H., Cheng, R. Y.-S., Ridnour, L. A., Murray, M. C.,
Tazzari, V., Sparatore, A., Del Soldato, P., Hines, H. B., Glynn, S. A.,
Ambs, S., and Wink, D. A. (2012) Dithiolethiones inhibit NF-κB
activity via covalent modification in human estrogen receptor−
negative breast cancer. Cancer Res. 72, 2394−2404.
(25) Sen, N., Paul, B. D., Gadalla, M. M., Mustafa, A. K., Sen, T., Xu,
R., Kim, S., and Snyder, S. H. (2012) Hydrogen sulfide-linked
sulfhydration of NF-κB mediates its antiapoptotic actions. Mol. Cell 45,
13−24.
(26) Duncan, R. (2003) The dawning era of polymer therapeutics.
Nature Rev. Drug Discovery 2, 347−360.
1299
dx.doi.org/10.1021/bc500150s | Bioconjugate Chem. 2014, 25, 1290−1300

Bioconjugate Chemistry
Article
(46) Peng, H., Cheng, Y., Dai, C., King, A. L., Predmore, B. L., Lefer,
D. J., and Wang, B. (2011) A fluorescent probe for fast and
quantitative detection of hydrogen sulfide in blood. Angew. Chem., Int.
Ed. 50, 9672−9675.
(47) Luxenhofer, R., Sahay, G., Schulz, A., Alakhova, D., Bronich, T.
K., Jordan, R., and Kabanov, A. V. (2011) Structure−property
relationship in cytotoxicity and cell uptake of poly(2-oxazoline)
amphiphiles. J. Controlled Release 153, 73−82.
(48) Conner, S. D., and Schmid, S. L. (2003) Regulated portals of
entry into the cell. Nature 422, 37−44.
(49) Scholz, M., Ulbrich, H. K., Soehnlein, O., Lindbom, L., Mattern,
A., and Dannhardt, G. (2009) Diaryl-dithiolanes and -isothiazoles:
COX-1/COX-2 and 5-LOX-inhibitory, OH scavenging and anti-
adhesive activities. Bioorg. Med. Chem. 17, 558−568.
(50) van der Vlies, A. J., Hasegawa, U., and Hubbell, J. A. (2012)
Reduction-sensitive tioguanine prodrug micelles. Mol. Pharmaceutics 9,
2812−2818.
1300
dx.doi.org/10.1021/bc500150s | Bioconjugate Chem. 2014, 25, 1290−1300