Hydrogen Sulfide in Redox Biology Part A

Visit the link below to download the full version of this book:

https://medipdf.com/product/hydrogen-sulfide-in-redox-biology-part-a/

Click Download Now

Contributors xiii

Péter Nagy
Department of Molecular Immunology and Toxicology, National Institute of Oncology,
Budapest, Hungary
Chung-Min Park
Department of Chemistry, Washington State University, Pullman, Washington, USA
Bo Peng
Department of Chemistry, Washington State University, Pullman, Washington, USA
Michael D. Pluth
Department of Chemistry and Biochemistry, Institute of Molecular Biology, Materials
Science Institute, University of Oregon, Eugene, Oregon, USA
Peter Rose
University of Lincoln, Lincoln, Lincolnshire, United Kingdom
Xinggui Shen
Department of Pathology, Louisiana State University Health Sciences Center–Shreveport,
Shreveport, Louisiana, USA
T. Spencer Bailey
Department of Chemistry and Biochemistry, Institute of Molecular Biology, Materials
Science Institute, University of Oregon, Eugene, Oregon, USA
Hidenori Suzuki
Department of Pharmacology, Nippon Medical School, Tokyo, Japan
Victor Vitvitsky
Department of Biological Chemistry, University of Michigan Medical School, Ann Arbor,
Michigan, USA
Ming Xian
Department of Chemistry, Washington State University, Pullman, Washington, USA
Xue Xue
Department of Pharmacology, Yong Loo Lin School of Medicine, National University of
Singapore, Singapore
Pramod K. Yadav
Department of Biological Chemistry, University of Michigan Medical School, Ann Arbor,
Michigan, USA
Shuai Yuan
Department of Pathology, Louisiana State University Health Sciences Center–Shreveport,
Shreveport, Louisiana, USA
PREFACE

Hydrogen sulfide is viewed as the third gasotransmmitter, gaseous signaling

molecule, together with nitric oxide and carbon monoxide. The cellular
sources of hydrogen sulfide involve enzymes of the trans-sulfuration path-
way CBS (cystathionine β-synthase), CSE (cystathioine γ-lyase), and 3MST
(3-mercaptopyruvate sulfur-transferase). Storages of hydrogen sulfide occur
in mitochondria (iron–sulfur clusters of enzymes) and cytosol (bound sulfane
sulfur). Of course, the release of hydrogen sulfide from these storages is
tightly regulated by several pathophysiological processes.
In addition to the myriad of effects arising from hydrogen sulfide as a
signaling molecule, it also protects against oxidative stress and glutamate tox-
icity, inhibits the release of insulin, preserves mitochondrial function, and is a
modulator of inflammatory responses. These pleiotropic effects of hydrogen
sulfide have been the subject of numerous investigations in the last years and
are largely accounted for by as its role as a gaseous signaling molecule.
Hydrogen sulfide may act alone or in conjunction with other gas-
otransmitters and, in doing so, it regulates a number of physiological pro-
cesses and is involved in some stages of the pathogenesis of several diseases.
These volumes of Methods in Enzymology were designed as a compen-
dium for hydrogen sulfide detection methods, the pharmacological activity
of hydrogen sulfide donors, the redox biochemistry of hydrogen sulfide and
its metabolism in mammalian tissues, the mechanisms inherent in hydrogen
sulfide cell signaling and transcriptional pathways, and cell signaling in spe-
cific systems, such as cardiovascular and nervous system as well as its function
in inflammatory responses. Three chapters are also devoted to hydrogen
sulfide in plants and a newcomer, molecular hydrogen, its function as a novel
antioxidant.
In bringing these volumes of Methods in Enzymology to fruition, credit
must be given to the experts in various aspects of hydrogen sulfide research,
whose thorough and innovative work is the basis of these Methods in
Enzymology volumes. We hope that these volumes will be of help to both
new and established investigators in the field.
ENRIQUE CADENAS
LESTER PACKER

xv
 CHAPTER ONE

Mechanistic Chemical Perspective

of Hydrogen Sulfide Signaling
Péter Nagy1
Department of Molecular Immunology and Toxicology, National Institute of Oncology, Budapest, Hungary
1
Corresponding author: e-mail address: [email protected]

Contents
1. Introduction 5
2. Bioavailability of Sulfide—The Signal 5
2.1 Endogenous sulfide production 6
2.2 Sulfide catabolism 8
2.3 Endogenous sulfide buffers 9
3. Inorganic Polysulfides 9
3.1 Biological relevance 9
3.2 Speciation and redox capacity of polysulfides 10
3.3 Polysulfide formation by sulfide oxidation 11
3.4 Stability of polysulfides 12
4. Sulfide Signaling Via Protein Sulfhydration 13
4.1 Mechanisms of persulfide formation 14
5. Sulfide Signaling via Sulfide–Hemeprotein Interactions 18
5.1 Sulfide mediates heme protein functions 19
5.2 Heme proteins generate sulfide oxidation products 21
5.3 Antioxidant properties of sulfide via reduction of metal
centers with higher oxidation states 22
6. Conclusions 23
Acknowledgments 24
References 24

Abstract
Hydrogen sulfide is now a well-appreciated master regulator in a diverse array of phys-
iological processes. However, as a consequence of the rapid growth of the area, sulfide
biology suffers from an increasing number of controversial observations and interpre-
tations. A better understanding of the underlying molecular pathways of sulfide's
actions is key to reconcile controversial issues, which calls for rigorous chemical/
biochemical investigations.
Protein sulfhydration and coordination/redox chemical interactions of sulfide with
heme proteins are the two most extensively studied pathways in sulfide biochemistry.
These pathways are important mediators of protein functions, generate bioactive

Methods in Enzymology, Volume 554 # 2015 Elsevier Inc. 3

ISSN 0076-6879 All rights reserved.
http://dx.doi.org/10.1016/bs.mie.2014.11.036
4 Péter Nagy

sulfide metabolites, contribute to sulfide storage/trafficking and carry antioxidant func-

tions. In addition, inorganic polysulfides, which are oxidative sulfide metabolites, are
increasingly recognized as important players in sulfide biology.
This chapter provides an overview of our mechanistic perspective on the reactions
that govern (i) sulfide's bioavailability (including the delicate enzyme machineries that
orchestrate sulfide production and consumption and the roles of the large sulfide-
storing pools as biological buffers), (ii) biological significance and mechanisms of per-
sulfide formation (including the reduction of disulfides, condensation with sulfenic
acids, oxidation of thiols with polysulfides and radical-mediated pathways),
(iii) coordination and redox chemical interactions of sulfide with heme proteins (includ-
ing cytochrome c oxidase, hemoglobins, myoglobins and peroxidases), and (iv) the
chemistry of polysulfides.

ABBREVIATIONS
3MST 3-mercaptopyruvate sulfurtransferase
AAT aspartate/cysteine aminotransferase
CBS cystathionine-β-synthase
CcO cytochrome C oxidase
CSE cystathionine-γ-lyase
CySOH Cys sulfenic acids
CySSH and GSSH, respectively Cys- and GSH-persulfides
DTNB 5,50 -dithiobis-(2-nitrobenzoic acid)
ER endoplasmic reticulum
ERK1/2 extracellular signal-regulated kinases ½
Ero1 ER oxidoreductin
Erv1 essential for respiration and vegetative growth sulfhydryl oxidase
GAPDH glyceraldehyde-3-phosphate dehydrogenase
GOR glutathione oxido-reductase
Hb Hemoglobin
HSSH• disulfide anion radical species
Keap1 Kelch-like ECH-associated protein 1
LPO lactoperoxidase
Mb myoglobin
MEK1 mitogen-activated protein kinase kinase
MPO myeloperoxidase
NfκB nuclear factor kappa-light-chain-enhancer of activated B cells
Nrf2 nuclear factor (erythroid-derived 2)-like 2
PARP-1 poly [ADP-ribose] polymerase 1
PDI protein disulfide isomerase
PLP pyridoxal phosphate
PTEN phosphatase and tensin homolog
PTP1B protein-tyrosine phosphatase 1B
roGFP reduction–oxidation sensitive green fluorescent protein
ROS reactive oxygen species
RPS3 ribosomal protein S3
SAM S-adenosylmethionine
Mechanistic Chemistry of Sulfide Signaling 5

SDO sulfide dioxigenase

SO sulfite oxidase
SQR sulfide quinone reductase
SUMO small ubiquitin-like modifier
TNB 5-thio-2-nitrobenzoic acid
TPO thyroid peroxidase
Trx thioredoxin
TST thiosulfate:glutathione sulfurtransferase

1. INTRODUCTION
Sulfide biology has expanded very rapidly in the past decade with a
great number of groundbreaking fundamental discoveries (see, e.g.,
Abe & Kimura, 1996; Blackstone, Morrison, & Roth, 2005; Coletta
et al., 2012; Elrod et al., 2007; Flannigan et al., 2014; Mustafa et al.,
2009; Papapetropoulos et al., 2009; Suzuki et al., 2011; Szabo et al.,
2013; Yang et al., 2008). However, the field suffers from a side effect com-
mon to many extensively growing areas, that of generating a large number of
controversial reports. Better understanding of the underlying molecular
mechanisms of sulfide actions is a fundamental requirement for reconciling
these controversial physiological observations. Therefore, this review is
intended to provide our mechanistic chemical perspective on what are cur-
rently considered to be the most important reactions in sulfide biology.

2. BIOAVAILABILITY OF SULFIDE—THE SIGNAL

In considering the mechanistic details of sulfide signaling, the first
obvious parameter to address is the physiological concentration of sulfide,
i.e., the signal. Moreover, in dynamic, living systems, it is rather the change
in sulfide concentration over time that triggers the cascade of biochemical
reactions that constitute the signaling process.
Sulfide concentrations can change: (i) as a result of physiological
or pathophysiological events (for example, in some cancer cells the
sulfide-producing enzymes, cystathionine-β-synthase (CBS), and/or
cystathionine-γ-lyase (CSE) are overexpressed and could increase sulfide
levels, but in inflammation or at sites of oxidative stress, sulfide levels could
be significantly diminished), (ii) due to administration of authentic sulfide or
slow releasing sulfide donor molecules, or (iii) simply as a result of normal
cellular functions. The latter can be demonstrated by the fact that in cellular
6 Péter Nagy

systems metabolic pathways work in parallel to strictly govern free sulfide

levels, and a small perturbation of sulfide-producing or consuming reactions
can have a profound effect on steady-state concentrations (Vitvitsky,
Kabil, & Banerjee, 2012).

2.1. Endogenous sulfide production

Endogenous sulfide production via cysteine metabolism is catalyzed
by at least three different enzymatic systems, the main ones being the
two pyridoxal phosphate (PLP) dependent CBS and CSE enzymes and
the cooperative actions of aspartate/cysteine aminotransferase (AAT) and
3-mercaptopyruvate sulfurtransferase (3MST) (reviewed recently in Kabil &
Banerjee, 2014; Nagahara, 2013). Although these enzymes all use cysteine
as their biological substrate, their individual enzyme kinetic properties are
very different. Differences include the involvement of (i) a variety of
co-substrates, (ii) parallel enzymatic activities, and (iii) mechanisms of inhi-
bition/potentiation (Fig. 1).
Therefore, the dominant cysteine metabolism-mediated sulfide-producing
pathway in a certain biological situation depends not only on the relative CSE:
CBS:AAT/3MST concentrations on site, but also on many other components
including the bioavailability of cysteine, homocysteine, α-ketoglutarate, and
other enzyme activity modulating factors (Chiku et al., 2009; Yadav,
Yamada, Chiku, Koutmos, & Banerjee, 2013). The primary sulfide-producing
reactions catalyzed by these enzymes are also very different chemically: CSE and
CBS catalyze α,β-elimination and β-replacement reactions of cysteine, respec-
tively, while 3MST produces sulfide via reductive elimination from a pref-
ormed cysteine-persulfide intermediate species (see Fig. 1).
Two additional mechanisms have the potential to produce sulfide indi-
rectly, from Cys- and GSH-persulfides (CySSH and GSSH, respectively).
These are CSE (Yamanishi & Tuboi, 1981) and CBS (Ida et al., 2014) cat-
alyzed beta-elimination from cystine and subsequent transsulfuration
between CySSH and GSH. Note that these pathways exclude the direct par-
ticipation of sulfide in the persulfide formation process, which is an impor-
tant difference from sulfide-mediated persulfide generation reactions (see
later). Based on the measured KM and kcat values, Ida et al. proposed that
under biologically relevant Cys:CySSCy ratios, CySSCy could be the pre-
ferred substrate for CSE and CBS. Hence, these reactions could provide the
majority of the large amounts (tens to hundreds of μM-s) of persulfides
(mostly as GSSH) that were detected in cellular systems (Ida et al., 2014).
Mechanistic Chemistry of Sulfide Signaling 7

3MST
3-Mercapto pyruvate

Glutamate 2R-SH R-S-S-R

Pyruvate
AAT
α-Ketoglutarate

L-Homocysteine Cystathionine

Cystathionine L-Cysteine H2S

CBS
L-Cysteine Lanthionine
2 L-Homocysteine
SAM CO, NO, SUMO
SUMO CSE H2O

Homolanthionine
H2S

α-Ketobutyrate
L-Cysteine L-Serine Pyruvate + NH3
NH3
Figure 1 The main sulfide-producing enzymatic pathways via cysteine metabolism. The
figure demonstrates (with examples) that beside the dominant sulfide-producing
pathways CSE and CBS catalyze many other reactions using different substrates that
not necessarily produce sulfide. Furthermore, both CSE and CBS are inhibited by small
ubiquitin-like modifier (SUMO) catalyzed sumoilation. CO and NO were also shown to be
inhibitory on CBS activity, but S-adenosylmethionine (SAM) is an allosteric activator of
the enzyme (Kabil & Banerjee, 2014).

Furthermore, a newly discovered thiosulfate:glutathione sulfurtransferase (TST) in

the transsulfuration pathway of sulfide catabolism to sulfite, sulfate, and thio-
sulfate (see later), could also generate substantial concentrations of GSH-
persulfide (and potentially CoA-SSH and other persulfide derivatives),
which may have a sulfide trafficking function particularly at sites with
decreased sulfide dioxygenase (SDO) activities (Melideo, Jackson, &
Jorns, 2014).
Importantly, the subsequent liberation of sulfide from persulfides could
also be a controlling factor in sulfide production where the thermodynamic
(reducing vs. oxidizing environment) and kinetic (rate of persulfide reduc-
tion) barriers should be highly dependent on the biological environment and
the chemical nature of the apparent reducing reaction partner. Hence, path-
ways that produce sulfide via persulfide intermediates are not only orches-
trated by persulfide-generating enzymatic activities, but also via persulfide
8 Péter Nagy

reducing reactions (Kabil & Banerjee, 2014; Melideo et al., 2014; Mikami
et al., 2011; Westrop, Georg, & Coombs, 2009; Yadav et al., 2013). Fur-
thermore, a concept is emerging, that persulfides have direct biological func-
tions rather than just being sulfide-storing molecules (Nielsen, Tachibana,
Hansen, & Winther, 2011; Ono et al., 2014; Paul & Snyder, 2012;
Toohey, 1989).

2.2. Sulfide catabolism

Sulfide catabolism mostly occurs in mitochondria via oxidative processes
(Szabo et al., 2014) driven primarily by the sulfide quinone reductase (SQR)
enzyme (see Fig. 2) (Hildebrandt & Grieshaber, 2008; Theissen,
Hoffmeister, Grieshaber, & Martin, 2003). The molecular mechanism of
this pathway involves initial reduction of an intramolecular disulfide moiety
of SQR to produce an SQR-persulfide intermediate (Cherney, Zhang,
Solomonson, Weiner, & James, 2010). Subsequently, this persulfide func-
tional group is transferred catalytically on to GSH by TST to produce
GSSH, which is used as a substrate by a sulfur dioxygenase to give sulfite
(Melideo et al., 2014). Sulfite is than utilized by either sulfite oxidase (SO)
or SQR to give sulfate or thiosulfate, respectively.
Despite the major mechanism of sulfide toxicity being inhibition of
mitochondrial respiration via interaction with cytochrome C oxidase (CcO)
(see later), at low concentrations sulfide can also serve as a stimulator of
ATP production by the mitochondrial electron transport chain (Goubern,
Andriamihaja, Nubel, Blachier, & Bouillaud, 2007; Helmy et al., 2014;
Lagoutte et al., 2010; Modis, Panopoulos, Coletta, Papapetropoulos, &

O2 GSH
SSO32–
Mitochondrion 2–
SO3 GSSH SO3 2–
H2S SQR-SSH
SQR-SSH SDO
SSO32– TST SO
S S
GSH SO42–

SQR

Figure 2 Proposed sulfide catabolism model in mitochondria. Reduction of an intramo-

lecular disulfide bond in SQR gives an SQR-SSH intermediate species. TST-mediated
transsulfuration from SQR-SSH to GSH gives GSSH. GSSH was proposed to engage in
sulfide trafficking, but its SDO catalyzed oxidation generates sulfite. Sulfite can either
be further oxidized by SO to sulfate or feed back to the catalytic cycle and react with
another SQR-SSH (Melideo et al., 2014).
Mechanistic Chemistry of Sulfide Signaling 9

Szabo, 2013; Szabo et al., 2014). Hence, regulation of mitochondrial ener-

getic functions coupled with sulfide catabolism is another prime example of
a fine-tuned sulfide-mediated physiological process.

2.3. Endogenous sulfide buffers

The bioavailability of sulfide is not only governed by endogenous sulfide-
producing and catabolizing reactions. It is now clear that a large amount
of sulfide is stored in biomolecule-sulfide adducts. Due to the reversible
nature of sulfide binding, this pool could serve as a buffer of free biological
sulfide concentrations (Nagy et al., 2014). In fact, based on a large number of
investigations (summarized in the SI of Nagy et al., 2014) we now know that
in most cells, tissues, and biological fluids, free sulfide represents less than 1%
of the potentially available sulfide, indicating that endogenous sulfide pools
should have large buffering capacities. This sulfide buffer system is under
strict thermodynamic control (constituted by the actual equilibrium con-
stants describing the relationships between the dissociated and bound
sulfide-biomolecule complexes and their interrelations), which is crucial
for maintaining sulfide levels below the toxic threshold. In addition, the
kinetics of sulfide binding and release are likely to mediate sulfide signaling,
because most reported physiological sulfide functions required relatively
high concentrations. In vivo these could only be provided by the sulfide
buffer system. This notion is corroborated by the fact that many of sulfide’s
physiological actions are better represented in vitro using a slow releasing
donor molecule (Whiteman, Le Trionnaire, Chopra, Fox, & Whatmore,
2011), most likely because the former could better mimic the sulfide-
donating capacity of the endogenous sulfide pool. The rate of sulfide release
is highly dependent on the binding partner and could show orders of mag-
nitude differences among the different sulfide-biomolecule adducts. The
large variation in on–off rates could provide selectivity and potentially serve
as a mediator of time-resolved sulfide bioavailability.

3. INORGANIC POLYSULFIDES
3.1. Biological relevance
It is increasingly appreciated that some of sulfide’s biological actions are due
to partially oxidized sulfane-sulfur (sulfur with an oxidation state of 0) con-
taining molecules (Greiner et al., 2013; Ida et al., 2014; Kimura, 2013, 2014;
Kimura & Kimura, 2004; Kimura et al., 2013; Koike, Ogasawara, Shibuya,
10 Péter Nagy

Kimura, & Ishii, 2013; Nielsen et al., 2011; Ono et al., 2014; Toohey,
1989). Major players in sulfane biochemistry may be inorganic polysulfides
(Greiner et al., 2013; Kimura, 2014; Nagy, Jameson, & Winterbourn, 2009;
Toohey, 2011). In fact, many of the reported controversies in sulfide biology
may be due to different polysulfide contents in “sulfide donor” solutions
(Nagy et al., 2014). Polysulfide contamination in sulfide (and hydrosulfide)
salts shows large variations. In addition, the extent and rate of air oxidation of
sulfide to polysulfides in a particular experimental setup, largely depends on
components of the applied medium such as oxygen, metal ions, and pH
(Nagy et al., 2014). Furthermore, very small amounts of polysulfide even
in the presence of a large excess of sulfide can serve as the actual regulators
of enzymatic functions (Greiner et al., 2013). Based on these observations,
the reported controversies and irreproducible results in sulfide biology are
not at all surprising, and it is likely that a deeper insight into polysulfide-
mediated processes will reconcile many conflicting observations. Moreover,
polysulfides are likely to be major endogenous oxidation products of sulfide
in different biological situations and participate in the cross-talk of sulfide
and redox signaling events (see later) with important physiological conse-
quences (Greiner et al., 2013; Kimura, 2014).

3.2. Speciation and redox capacity of polysulfides

From a chemical perspective, polysulfides are thermodynamically unstable
sulfide oxidation products, with a general formula of H2Sx (x ¼ 2–9). Their
acid dissociation constants (pKa) decrease with an increase in chain length,
but for all polysulfides it is likely to be <7, the pKa of HS (Hoffmann,
1977). Therefore, at physiological pH, polysulfides are expected to be in
their mono-anionic forms and it is more appropriate to use the general for-
mula HSx  instead of H2Sx. For the sake of simplicity, they can be consid-
ered as molecules that contain one sulfide and 1–8 sulfane-sulfurs, which
nicely demonstrate that their oxidizing capacity increases with the number
of sulfur atoms in the chain.
The speciation of polysulfides is described by a large number of interre-
lated equilibrium processes, exemplified by the following reactions:
2HS2  Ð HS3  + HS (1)
HS2  + HS3  Ð HS4  + HS (2)
HS3  + HS3  Ð HS4  + HS2  (3)
HS4  + HS2  Ð HS5  + HS (4)
Mechanistic Chemistry of Sulfide Signaling 11

This complexity together with (i) the difficulty of achieving aqueous

media in which one or other species dominates and (ii) polysulfides being
thermodynamically unstable and kinetically labile molecules, make the sys-
tematic investigation of polysulfide distribution extraordinary challenging.
Nevertheless, Giggenbach attempted to study polysulfide speciation under
different conditions based on UV-characteristics, (Giggenbach, 1972). His
data indicate that polysulfide distribution is largely dependent on pH and
the sulfane-sulfur–sulfide ratio. Generally, higher pH and a more reducing
environment (i.e., high sulfide to polysulfide ratios) favor the formation of
polysulfides with lower numbers of zerovalent sulfur atoms (i.e., shorter
chain length). It has also been demonstrated that the redox potential of poly-
sulfide solutions strictly correlates with the sulfide–sulfane-sulfur ratio and is
largely influenced by the pH (Lessner, Mclarnon, Winnick, & Cairns, 1993),
suggesting that the redox potential of polysulfides increases with chain
length.

3.3. Polysulfide formation by sulfide oxidation

Polysulfides can be generated in both 1 and 2 electron oxidation reactions of
sulfide. Importantly, some oxidants react at moderate rates (e.g.,
H2O2 kapppH 6:8
¼ 0:2M 1 s1 , Hoffmann, 1977), whereas other reactions
(e.g., with HOCl kapp pH 7:4
¼ 2:3  109 M 1 s1 , Nagy & Winterbourn,
2010a) are almost diffusion controlled. Although a number of reports suggest
an antioxidant role for sulfide via neutralizing reactive oxygen species
(ROS) (Fu, Liu, Geng, Fang, & Tang, 2008; Jha, Calvert, Duranski,
Ramachandran, & Lefer, 2008; Kimura & Kimura, 2004; Laggner et al.,
2007; Whiteman et al., 2004, 2005; Yonezawa et al., 2007), in my view,
because of the orders of magnitude lower free sulfide levels compared to
Cys thiol concentrations, direct scavenging of oxidants is unlikely to be
responsible for the observed antioxidant properties in biological systems.
Instead, antioxidant activity may be better explained by modulation of
pro- and antioxidant enzyme functions via reactions with their Cys residues
(such as the activation of the nuclear factor (erythroid-derived 2)-like 2 (Nrf2)
pathway via Kelch-like ECH-associated protein 1 (Keap1) inhibition at its
active site Cys residue (Hourihan, Kenna, & Hayes, 2013; Yang et al.,
2013)) or metal centers (see later).
Two-electron oxidation is likely to be the dominant pathway in sulfide’s
reactions with closed shell oxidants such as HOCl, HOSCN, H2O2 and other
lipid, peptide, or protein peroxides. These reactions, in analogy with Cys, are
12 Péter Nagy

expected to produce the inorganic disulfide species via highly reactive sulfenic
acid (or sulfenyl halide) intermediate species (see reactions 5 and 6).
HS + Ox ! HSX + Red ðX ¼ Cl, SCN, or OHÞ (5)
  
HS + HSX ! HS2 + H + X +
(6)
In contrast to cysteine, this inorganic disulfide is thermodynamically
unstable and subsequent disproportionation gives a mixture of polysulfide
species (see Section 3.2). Alternatively, HS2  could be reduced before it
has time to disproportionate, e.g., by Cys thiolates in biological systems.
Hence, it could potentially serve as a sulfhydrating agent, but further kinetic
information is required to establish whether this is the case.
One electron oxidation of sulfide by radicals (such as superoxide,
Wedmann et al., 2014) or metal centers (such as myeloperoxidase (MPO),
Palinkas et al., 2014) generate polysulfides via the intermediate production
of the thiyl radical (HS•). The reaction of HS• with another sulfide (reaction 7)
is diffusion controlled and gives the disulfide anion radical species (HSSH• )
(Das, Huie, Neta, & Padmaja, 1999). HSSH• is then expected to be further
oxidized (e.g., by oxygen) to give HS2  (reaction 8) and eventually poly-
sulfides (and superoxide if the reaction partner is oxygen).
HS• + HS ! HSSH• (7)
HSSH• + O2 ! HSSH + O2 • (8)
Polysulfides generated by sulfide oxidation can be viewed as secondary
reactive oxidant species. This property is best demonstrated by their reac-
tions with Cys thiols to give persulfides (and possibly Cys-polysulfides;
see later). However, sulfur can also acquire oxidation states larger than
0 and therefore polysulfides can also act as reducing molecules. Persulfides
are likely to be more efficient radical reducing agents than thiols because
the unpaired electron of RSS• is resonance stabilized by the adjacent sulfur
atom (Ono et al., 2014), and this might well be true for inorganic
polysulfides.

3.4. Stability of polysulfides

Polysulfides are thermodynamically unstable species in aqueous media. In
the absence of reducing or oxidizing reaction partners they disproportionate
to eventually give H2S and S8:
HSðn + 1Þ  + H + ! n=8S8 + H2 S ðn ¼ 1  8Þ (9)
Mechanistic Chemistry of Sulfide Signaling 13

In biological systems, polysulfides are expected to be consumed more

rapidly by biomolecules before they have time to disproportionate. How-
ever, it is important to take disproportionation into consideration when
working with polysulfide solutions in experimental systems. Therefore,
the stability of polysulfide species under different experimental conditions
needs to be systematically studied in order to avoid misinterpretation of
results. Furthermore, it is important to note that it is not possible to make
sulfide-free polysulfide solutions, because based on reaction (9), consump-
tion of sulfide (e.g., by volatilization) will facilitate polysulfide decomposi-
tion and sulfur precipitation.

4. SULFIDE SIGNALING VIA PROTEIN SULFHYDRATION

The most frequently quoted mechanism for sulfide signaling is sul-
fhydration of protein cysteine residues. Chemically, this posttranslational
modification results in an analogous functional group to those in the previ-
ously discussed CySSH and GSSH species (note that from a chemical per-
spective the term persulfide formation is preferred over sulfhydration).
Persulfide formation on functional Cys residues can enhance or inhibit enzy-
matic activities either directly or indirectly as indicated by the following
examples:
Examples of enzyme activation via persulfide formation: (1) Up to sevenfold
greater glycolytic activity of glyceraldehyde-3-phosphate dehydrogenase
(GAPDH) was observed when the active site Cys150 was modified to a per-
sulfide (Mustafa et al., 2009). (2) Persulfide formation on Cys38 of the p65
subunit of nuclear factor kappa-light-chain-enhancer of activated B cells (NfκB)
activated binding to the co-activator ribosomal protein S3 (RPS3) and
triggered antiapoptotic transcriptional activity (Sen et al., 2012). (3) ATP-
sensitive potassium channels were activated via persulfide formation-
induced inhibition of ATP binding (Mustafa et al., 2011). (4) Persulfide
formation on Cys341 of mitogen-activated protein kinase kinase (MEK1) leads
to poly [ADP-ribose] polymerase 1 (PARP-1) activation and DNA damage
repair via facilitation of phosphorylated extracellular signal-regulated kinases
½ (ERK1/2) translocation into the nucleus (Zhao et al., 2014). (5) Nrf2
was activated via Cys151-persulfide formation-induced activity loss of
Keap1 (Yang et al., 2013).
Inhibition of enzymatic activities via persulfide formation: (1) The previous
examples of Nrf2 or PARP-1 activation represent indirect effects, because
they are facilitated via persulfide formation-mediated inactivation of their
14 Péter Nagy

negative regulators, Keap1 and MEK1, respectively (Yang et al., 2013). (2)
Polysulfide-induced persulfide generation on the active site Cys124 and/or
Cys71 residues efficiently inactivated the phosphatase activity of phosphatase
and tensin homolog (PTEN) (Greiner et al., 2013). (3) Persulfide formation on
Cys215 had a similar inactivating effect on protein-tyrosine phosphatase 1B
(PTP1B) as the well-established redox switch of the enzyme via reversible
cyclic sulfenamide formation between the Cys215 thiolate and its backbone
amide nitrogen (Krishnan, Fu, Pappin, & Tonks, 2011).
Persulfide formation has recently been proposed to serve as a potential
protecting mechanism of thiol residues toward oxidative stress (Greiner
et al., 2013; Ida et al., 2014; Ono et al., 2014). The driving force of
ROS scavenging was proposed to be preferential oxidation of the persulfide
moiety compared to the corresponding thiol. The protecting effect was
explained by the chemical notion that the perthiosulfenic (P-CySSOH),
perthiosulfinic (P-CySSO2H), and perthiosulfonic acid (P-CySSO3H) oxi-
dation products of persulfides can efficiently be recycled via reduction of
their -S-S- moieties, in contrast to the sulfinic (CySO2H) or sulfonic acid
(CySO3H) products of thiol oxidation (Ono et al., 2014).
From a chemical perspective, it is important to emphasize that persulfide
formation mostly inhibits the function of the modified Cys residue, with the
only reported example of GAPDH, where activation is solicited by the
persulfidation of the actual catalytic thiol (Mustafa et al., 2009). All the other
examples of enzyme activations represent indirect effects via persulfide
formation-induced inhibition of the respective negative regulator. Further-
more, the observed enhanced peroxide-scavenging potential of GSSH (Ida
et al., 2014) could be simply due to a lower pKa of GSSH versus GSH, resulting
in an increase in the proportion of the more nucleophilic deprotonated form of
the peroxide-scavenging sulfhydryl moiety (as in GSS vs. GS). It will be
interesting to see how the reactivities of the deprotonated forms, i.e., the
Cys thiolates versus the corresponding perthiolates compare to each other.

4.1. Mechanisms of persulfide formation

An early misconception that the direct reaction of sulfide with reduced
Cys residues can give persulfides has been addressed (Greiner et al., 2013;
Nagy & Winterbourn, 2010a; Toohey, 2012) and it is now widely
appreciated that sulfide-mediated persulfide formation occurs either via
the reactions of oxidized Cys derivatives with sulfide or sulfide oxidation
products reacting with Cys thiols.
Mechanistic Chemistry of Sulfide Signaling 15

4.1.1 Persulfide formation via disulfide reduction

A potential mechanism for persulfide formation is the reduction of biolog-
ically abundant Cys-disulfide bonds by sulfide (Cavallini, Federici, &
Barboni, 1970; Francoleon, Carrington, & Fukuto, 2011; Nielsen et al.,
2011). However, the physiological relevance of this pathway has been
questioned by many investigators based on either thermodynamic or kinetic
considerations (Kabil & Banerjee, 2010; Paulsen & Carroll, 2013; Toohey,
2011, 2012; Zhang et al., 2014). In a comprehensive kinetic study on using
model disulfides (Vasas, Doka, Fabian, & Nagy, 2014), we found that the
rate of sulfide-mediated disulfide reduction is highly dependent on the reac-
tivity of the disulfide moiety. This introduces orders of magnitude differ-
ences in second-order rate constants. On this basis, it would be predicted
that activated functional disulfides, such as in SQR, protein disulfide isomerase
(PDI), Mia40, thioredoxin (Trx), ER oxidoreductin (Ero1), essential for respira-
tion and vegetative growth sulfhydryl oxidase (Erv1), glutathione oxido-reductase
(GOR), or the disulfide bond protein family are likely to be rapidly reduced
by sulfide. Corroborating this notion, the first proposed reaction step in
SQR-mediated sulfide catabolism is the reduction of the intramolecular
Cys160–Cys356 disulfide bond by sulfide to give a persulfide intermediate
species (Cherney et al., 2010; Kabil & Banerjee, 2014).
Our experimental data are consistent with a kinetic model which
includes a direct reaction between the disulfide and sulfide to give the
corresponding persulfide and thiol derivatives (reaction 10 in Scheme 1).
The persulfide can be further reduced by sulfide to give inorganic poly-
sulfides and another thiol (reactions 12–13 in Scheme 1). However, at high
cellular thiol concentrations, persulfide is likely to be reduced by another

Scheme 1 Proposed pathways for the reactions of Cys-disulfides with sulfide (Vasas
et al., 2014).