Hydropersulfides: H-Atom Transfer Agents Par Excellence

Article abstract of DOI:10.1021/jacs.7b02571

Hydropersulfides (RSSH) are formed endogenously via the reaction of the gaseous biotransmitter hydrogen sulfide (H₂S) and disulfides (RSSR) and/or sulfenic acids (RSOH). RSSH have been investigated for their ability to store H₂S in vivo and as a line of defense against oxidative stress, from which it is clear that RSSH are much more reactive to two-electron oxidants than thiols. Herein we describe the results of our investigations into the H-atom transfer chemistry of RSSH, contrasting it with the well-known H-atom transfer chemistry of thiols. In fact, RSSH are excellent H-atom donors to alkyl (k ～ 5 × 10⁸ M^-1 s^-1), alkoxyl (k ～ 1 × 10⁹ M^-1 s^-1), peroxyl (k ～ 2 × 10⁶ M^-1 s^-1), and thiyl (k > 1 × 10¹⁰ M^-1 s^-1) radicals, besting thiols by as little as 1 order and as much as 4 orders of magnitude. The inherently high reactivity of RSSH to H-atom transfer is based largely on thermodynamic factors; the weak RSS-H bond dissociation enthalpy (～70 kcal/mol) and the associated high stability of the perthiyl radical make the foregoing reactions exothermic by 15-34 kcal/mol. Of particular relevance in the context of oxidative stress is the reactivity of RSSH to peroxyl radicals, where favorable thermodynamics are bolstered by a secondary orbital interaction in the transition state of the formal H-atom transfer that drives the inherent reactivity of RSSH to match that of α-tocopherol (α-TOH), nature's premier radical-trapping antioxidant. Significantly, the reactivity of RSSH eclipses that of α-TOH in H-bond-accepting media because of their low H-bond acidity (α2^H ～ 0.1). This affords RSSH a unique versatility compared to other highly reactive radical-trapping antioxidants (e.g., phenols, diarylamines, hydroxylamines, sulfenic acids), which tend to have high H-bond acidities. Moreover, the perthiyl radicals that result are highly persistent under autoxidation conditions and undergo very rapid dimerization (k = 5 × 10⁹ M^-1 s^-1) in lieu of reacting with O₂ or autoxidizable substrates.

Full text of DOI:10.1021/jacs.7b02571

Subscriber access provided by UB + Fachbibliothek Chemie | (FU-Bibliothekssystem)
Article
Hydropersulfides: H-Atom Transfer Agents par Excellence
Jean-Philippe R. Chauvin, Markus Griesser, and Derek A. Pratt
J. Am. Chem. Soc., Just Accepted Manuscript • Publication Date (Web): 18 Apr 2017
Downloaded from http://pubs.acs.org on April 18, 2017
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of the American Chemical Society is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 12
Journal of the American Chemical Society
1
2
3
4
5
6
Hydropersulfides: H-Atom Transfer Agents par Excellence
7
8
Jean-Philippe R. Chauvin, Markus Griesser and Derek A. Pratt*
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Department of Chemistry and Biomolecular Sciences, University of Ottawa, Ottawa, Ontario K1N 6N5, Canada
ABSTRACT: Hydropersulfides (RSSH) are formed endogenously via the reaction of the gaseous biotransmitter hydrogen sulfide (H2S)
and disulfides (RSSR) and/or sulfenic acids (RSOH). RSSH have been investigated for their ability to store H2S in vivo and as a line of
defence against oxidative stress, from which it is clear that RSSH are much more reactive to two-electron oxidants than thiols. Herein
we describe the results of our investigations into the H-atom transfer chemistry of RSSH, contrasting it with the well-known H-atom
transfer chemistry of thiols. In fact, RSSH are excellent H-atom donors to alkyl (k ~ 5 ´ 108 M-1 s-1), alkoxyl (k ~ 1 ´ 109 M-1 s-1), peroxyl
(k ~ 2 ´ 106 M-1 s-1) and thiyl (k > 1 ´ 1010 M-1 s-1) radicals, besting thiols by as little as one and as much as four orders of magnitude. The
inherently high reactivity of RSSH to H-atom transfer is based largely on thermodynamic factors; the weak RSS-H BDE (~ 70 kcal/mol)
and associated high stability of the perthiyl radical make the foregoing reactions exothermic by 15 to 34 kcal/mol. Of particular relevance
in the context of oxidative stress is the reactivity of RSSH to peroxyl radicals, where favourable thermodynamics are bolstered by a
secondary orbital interaction in the transition state of the formal H-atom transfer that drives the inherent reactivity of RSSH to match
that of a-tocopherol (a-TOH), Nature’s premier radical-trapping antioxidant. Significantly, the reactivity of RSSH eclipses that of a-
TOH in H-bond accepting media because of their low H-bond acidity (αH2 ~ 0.1). This affords RSSH a unique versatility compared to
other highly reactive radical-trapping antioxidants (e.g. phenols, diarylamines, hydroxylamines, sulfenic acids), which tend to have high
H-bond acidities. Moreover, the perthiyl radicals that result are highly persistent under autoxidation conditions and undergo very rapid
dimerization (k = 6.0 ´ 109 M-1 s-1) in lieu of reacting with O2 or autoxidizable substrates.
n
INTRODUCTION
Considerable interest in hydropersulfides (RSSH) has emerged
in recent years owing to their identification at significant levels
in mammalian tissues.1-3 They, and other polysulfide (RSnR’) spe-
cies, have been proposed to be responsible for the cell signaling
and/or redox modulatory effects of H2S,1,4 which is produced by
cystathionine β-synthase (CBS),5,6 cystathionine γ-lyase (CSE)7
and 3-mercaptopyruvate sulfurtransferase (MST)8,9 in response to
various stimuli. Indeed, multiple biological functions have been
ascribed to H2S, including circulatory regulation,10,11 energy pro-
duction,12 apoptotic gene regulation,13 and protection from oxi-
dative stress.14-16
It is widely believed that the various functions of H2S are me-
diated by RSSH, which arise from its reaction with disulfides or
other sulfur electrophiles.9,17,18 Protein persulfides have been de-
tected using various chemical probes such as cycloalkynes and io-
doacetates.2,3,19-23 Nevertheless, the chemical basis of the putative
physiological effects of RSSH is unclear.24,25 Recent work by Fu-
kuto and co-workers26 has provided insight into the ionic reactiv-
ity of RSSH toward oxidants and electrophiles, revealing that
these are more reactive than the thiols from which RSSH are de-
rived. This provides some support for the suggestion that H2S is
produced to augment the abilities of thiols to sense oxidants and
electrophiles. Given these observations, it figures that the homo-
lytic reactivity of RSSH may also be enhanced relative to that of
thiols.
reaction and as polarity-reversal catalysts.27 There is a widespread
misconception that thiols are also useful as radical-trapping anti-
oxidants (RTAs), which react with autoxidation chain-carrying
peroxyl radicals by formal H-atom transfer (HAT). In fact, thiols
react slowly with peroxyl radicals (<103 M-1 s-1)28 relative to good
RTAs, such as a-tocopherol (a-TOH), the most biologically ac-
tive form of Vitamin E (k > 106 M-1 s-1).29 On the other hand,
despite their apparently weak S-H BDEs (estimated to be ca. 70
kcal/mol – compare to thiols at 87 kcal/mol),30,31 little is known
of the H-atom transfer (HAT) reactivity of RSSH.32 Herein we
describe the results of our efforts to fill this void, and present
data for the reactivity of RSSH toward a variety of oxidants, in-
cluding peroxyl (ROO•), alkoxyl (RO•), alkyl (R•) and thiyl
(RS•) radicals. In line with the trends in their ionic reactivity, we
find the HAT reactivity of RSSH to be analogous – but greatly
accelerated – relative to thiols.
n
RESULTS
Cumylhydropersulfide (cumyl-SSH) and 2-methylundecan-2-per-
sulfide (t-dodecyl-SSH) were chosen for study owing to the greater
persistence of hydropersulfides adjacent to tertiary centres, which
slow their self-reaction to trisulfide and H2S.33 Both compounds
(Figure 1) were synthesized by acidic methanolysis of the corre-
sponding S-acetylalkyldisulfide, which was synthesized from the
corresponding thiol and acetylsulfenyl chloride.33
Thiols are well known to be useful H-atom donors in syn-
thetic contexts, perhaps most prominently in the thiol-ene
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 12
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. PBD-BODIPY serves as the signal carrier in THF autoxidations (A), enabling determination of rate constants (kinh) and reaction
stoichiometries (n) for reactions of inhibitors with chain-carrying peroxyl radicals (B). Co-autoxidations of THF (3.1 M) and PBD-
BODIPY (10 µM) initiated by AIBN (6 mM) in chlorobenzene at 37 °C (dashed black trace) and inhibited by 2 µM, 5 µM and 10 µM
of cumyl-SSH (C) and t-dodecyl-SSH (D). Average inhibition rate constants and stoichiometry summarized in (E). Reaction progress was
monitored by absorbance at 588 nm (e = 128,100 M-1 cm-1).
Hydropersulfides Are Excellent Radical-Trapping Antioxi-
dants. The reactivity of the hydropersulfides toward peroxyl rad-
icals was determined using the venerable inhibited autoxidation
approach.34 The addition of PBD-BODIPY as co-substrate ena-
bles reaction monitoring simply by loss of the absorbance at 588
nm due to the addition of peroxyl radicals to the 1-phenylbuta-
diene moiety (Figure 1A).35 THF is present in order to maintain
a radical chain reaction, ensuring a steady-state concentration of
peroxyl radicals and enabling derivation of the rate constant for
the reaction of the hydropersulfide with peroxyl radicals (kinh)
from the initial rate of the inhibited reaction (Figure 1B). The
stoichiometry (n) of the RTA-peroxyl reaction is determined from
the duration of the inhibited period (tinh, also in Figure 1B).35
Representative results are shown for cumyl-SSH and t-dodecyl-
SSH in Figures 1C and 1D, respectively.
peroxyl radicals, the resultant perthiyl radicals do not contribute
to either chain-propagating or chain-breaking events. Indeed, we
have previously shown that perthiyl radicals combine to give
tetrasulfides with k = 6 ´ 109 M-1 s-1, and tetrasulfides are unreac-
tive to peroxyl radicals under these conditions.36
In order to provide direct evidence that the reaction between
the hydropersulfides and peroxyl radicals is a HAT process, cor-
responding inhibited autoxidations were carried out in the pres-
ence of 1% H2O or 1% D2O. Whilst the kinetic data in the pres-
ence of water were indistinguishable from those obtained in its
absence, in the presence of D2O the rate constants were 2.2 to
2.5-fold lower. This primary kinetic isotope effect suggests that
an (exchangeable) H-atom is transferred in the reaction of the
hydropersulfides with peroxyl radicals. The relatively small ki-
netic isotope effects are consistent with the significant exergon-
icity of the reaction (ca. 16 kcal/mol) and suggest a relatively early
transition state (corroborated by computation, vide infra).
The traces in Figures 1C and 1D clearly indicate that hy-
dropersulfides are excellent inhibitors of autoxidation. The inhi-
bition rate constants derived from the initial rates in the presence
of each compound are essentially indistinguishable, with kinh
=
H-Atom Transfer from Hydropersulfides is Relatively Insensi-
tive to H-Bonding Interactions. The kinetics of HAT reactions
are known to vary systematically as a function of the H-bond do-
nating ability of the H-atom donor and the H-bond accepting
ability of the solvent according to the scheme in Figure 2A and
associated kinetics in Figure 2B.37 Given that thiols are poor H-
bond donors whose HAT kinetics are insensitive to changes in
reaction medium,38 we wondered if hydropersulfides would be
(8.0 ± 0.8) ´ 105 M-1 s-1 and (8.3 ± 0.5) ´ 105 M-1 s-1 for cumyl-SSH
and t-dodecyl-SSH, respectively. For comparison, a-TOH is char-
acterized by kinh = 7.1 ´ 105 M-1 s-1 under the exact same condi-
tions (see Supporting Information for the data). The observed
inhibition periods correspond to stoichiometries of n = 1.0 ± 0.1
and n = 0.9 ± 0.1 for the reactions of cumyl-SSH and t-dodecyl-
SSH, respectively, suggesting that following H-atom transfer to
2
ACS Paragon Plus Environment

Page 3 of 12
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. The kinetics of H-atom transfer reactions are slowed by H-bonding interactions with solvent, illustrated for hydropersulfides
in (A), which can be quantified according to the expression in (B). The H-bonding equilibrium can be expressed in terms of Abraham’s
H-bond acidity and basicity parameters αH2 and β2H as in (C). The inhibition rate constants for t-dodecyl-SSH, determined as in Figure 1
(see Supporting Information for the raw data) plotted as a function of medium βH2 ; the corresponding relationship for a-TOH is represented
by the dashed red line (D). The H-bond acidity parameters for cumyl-SSH and t-dodecyl-SSH determined from the line of best fit of the
data in (D) fit to the expression in (C), and estimated rate constants for the reaction in the absence of any H-bonding interactions (E).
The calculated (CBS-QB3) transition state structures for HAT between a model hydropersulfide (MeSSH) and model peroxyl radical
(MeOO•) and associated free energy barriers and rate constants estimated using transition state theory (F). The SOMO and HOMO of
the syn transition state structure (G).
similarly poor H-bond donors with corresponding medium-inde-
pendent HAT kinetics.
while a-TOH is a better radical-trapping antioxidant than the hy-
dropersulfides in simple hydrocarbons, the trend is reversed once
H-bond acceptors are introduced to the medium.
Using Abraham’s NMR method,39 we determined the H-bond
acidities (on the αH2 scale) of cumyl-SSH and t-dodecyl-SSH to be
0.10 and 0.09, respectively – slightly larger than those deter-
mined for the corresponding thiols (0.02 and 0.08, respectively)
determined under identical conditions. For comparison, alco-
hols have αH2 values that range from 0.4 to 0.8.40 Even a some-
what hindered and electron rich phenol such as a-TOH has an
αH2 value of 0.37,41 and since there is a logarithmic relationship
between this parameter and the HAT kinetics (Figure 2C), its kinh
varies considerably with the H-bond accepting basicity of the sol-
Previous work by our group has revealed the importance of
secondary orbital interactions in reactions between H-atom do-
nors and peroxyl radicals. CBS-QB3 calculations indicate that
this contributes to the reactivity of hydropersulfides as well. As is
the case with phenols,42 diphenylamines,43 and the more struc-
turally-related sulfenic (RSOH)44 and selenenic acids (RSeOH)45,
the lowest energy transition state (TS) structure for the reaction
of hydropersulfides with peroxyl radicals is characterized by a syn
relationship between the substituents on the atoms between
which the H-atom is being transferred (Figure 2F). This maxim-
izes overlap between the singly-occupied p* of the peroxyl radical
and the lone-pair of the ‘outer’ sulfur atom of the hydropersul-
fide, as illustrated in Figure 2G. The free energy barrier calcu-
lated as this level of theory is 11.1 kcal/mol, which corresponds
to a rate constant of k= 2.3 ´ 106 M-1 s-1 at 37 °C upon application
of TS theory. This value is in excellent agreement with the exper-
imental values derived in the absence of H-bonding (k0 = 1.8-2.1
´ 106 M-1 s-1 at 37 °C). The barrier to reaction via the anti TS
structure is 2.3 kcal/mol higher. The TS is expectedly early (con-
sistent with the KIEs, vide supra), with little S-H bond breakage
vent. For example, on going from pure chlorobenzene (β2H
=
0.09) to 3:1 chlorobenzene:THF (βH2 = 0.39), kinh drops from 3.6
´ 106 M-1 s-1 to 7.1 ´ 105 M-1 s-1. In sharp contrast, and consistent
with the values of αH2 for cumyl-SSH and t-dodecyl-SSH we deter-
mined by NMR, autoxidations inhibited by the hydropersulfides
were much less affected by changes in medium, as shown for t-
dodecyl-SSH in Figure 2D. The slope of the correlation between
logkinh and the βH2 of the media yielded an αH2 value of 0.12 for t-
dodecyl-SSH – in excellent agreement with the NMR data. Thus,
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 12
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. STY-BODIPY serves as the signal carrier in THF autoxidations (A), enabling determination of rate constants (kinh) and reaction
stoichiometries (n) for reactions of inhibitors with chain-carrying peroxyl radicals (B). Co-autoxidations of THF (3.1 M) and STY-
BODIPY (10 µM) initiated by AAPH (1 mM) in phosphate-buffered saline (pH 7.4, 100 mM) at 37 °C (dashed black trace) and inhibited
by 10 µM of cumyl-SSH (red), t-dodecyl-SSH (blue) and 4 µM of Trolox (black) (C). Inhibition rate constants and reaction stoichiometries
determined as a function of pH (D). Reaction scheme illustrating the change in mechanism of the hydropersulfide/peroxyl radical
reaction from H-atom transfer to sequential proton loss electron transfer (SPLET) as a function of pH (E). Inhibition rate constants
determined as a function of pD and corresponding deuterium kinetic isotope effects (F).
(1.472 Å compared to 1.349 Å in the starting hydropersulfide
and little O-H bond formation (1.402 Å compared to 0.967 Å in
the product hydroperoxide). This is consistent with the signifi-
cant thermodynamic driving force (∆Go = -15.4 kcal/mol) owing
to the difference in the MeSS-H and MeOO-H bond dissociation
enthalpies (BDEs), which were calculated to be 70.8 and 86.2
kcal/mol, respectively.
carboxylic acid. The inhibition rate constants determined for
cumyl-SSH and t-dodecyl-SSH were 3.7 ´ 106 and 3.2 ´ 106 M-1 s-
1, respectively – almost an order of magnitude larger than that
determined for Trolox (4.1 ´ 105 M-1 s-1).46 Despite this exciting
reactivity, the hydropersulfides suffer from poor stability in wa-
ter, which facilitates their self-reactions (to mixtures of polysul-
fides and thiols). As such, the inhibition periods were much
shorter than those observed in organic solution, corresponding
to much smaller reaction stoichiometries of n = 0.15 ± 0.03 and
0.12 ± 0.05 for cumyl-SSH and t-dodecyl-SSH, respectively, as op-
posed to n ~ 1 as above.
Since cumyl-SSH has slightly better stability in aqueous solu-
tion than t-dodecyl-SSH, we expanded our studies with it to in-
clude inhibited autoxidations in buffers of varying pH ranging
from 4 to 9. Interestingly, not only were the kinh values dependent
on pH, but a sigmoidal relationship was observed (Figure 3D).
The reaction stoichiometries were also pH-dependent, increasing
at lower pH presumably due to a slowing of their self-reactions,
which are known to be faster at basic pH.33 The sigmoidal curve
was fit using a Boltzmann function, yielding a pKa value of 7.0
corresponding to the inflection point at pH = 7, strikingly similar
to pKas estimated for hydropersulfides.47,48
Hydropersulfides React with Peroxyl Radicals via Two Distinct
Mechanisms in Water. The HAT reactivity of the hydropersul-
fides was also investigated in aqueous solution. THF has been
shown to be a useful autoxidizable substrate for monitoring per-
oxyl radical reactions in buffered solutions (pH 2-12),46 and we
have shown that STY-BODIPY, the more slowly oxidized ana-
logue of PBD-BODIPY, is an effective signal carrier for monitor-
ing reaction progress under these conditions (Figure 3A,B).35
Representative results from co-autoxidations of THF and STY-
BODIPY carried out at pH 7.4 are shown in Figure 3C.
Following the trend established in the previous section, the
hydropersulfides were significantly more reactive to peroxyl radi-
cals than a-TOH – or at least Trolox, a water-soluble version of
a-TOH in which the phytyl sidechain has been replaced with a
4
ACS Paragon Plus Environment

Page 5 of 12
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4. Photolytic generation of the 6,6-diphenylhexenyl radical and its use as a radical clock to obtain the rate constant for HAT
from hydropersulfides to alkyl radicals (A). Ratio of linear and cyclized products as a function of the concentration of cumyl-SSH (red)
or t-dodecyl-SSH (blue) during photocleavage of the radical precursor in benzene at 25 °C (B). Photolytic generation of cumyloxyl
radicals from dicumylperoxide (C). Dependence of the pseudo-first order rate on the concentration of cumyl-SSH (red) or t-dodecyl-SSH
(blue) upon photolysis of dicumylperoxide in benzene at 25 °C; inset: example decay of the cumyloxyl radical (D). Photolytic generation
of phenylthiyl radicals from diphenyldisulfide (E). Dependence of the pseudo-first order rate on the concentration of cumyl-SSH (red)
or t-dodecyl-SSH (blue) upon photolysis of diphenyldisulfide in benzene at 25 °C; inset: example decay of the phenylthiyl radical (F).
Calculated (CBS-QB3) transition state structures and associated free energy barriers, estimated rate constants and reaction free energies
for HAT between a model hydropersulfide (MeSSH) and a model alkyl radical (n-Bu•) (G), a model alkoxyl radical (n-BuO•) (H), and
phenylthiyl radical (I).
Accordingly, we wondered if the pH dependence of kinh could be
attributed to a change in mechanism: as the pH of the medium
increases and the concentration of the persulfide anion in-
creases, the reactions takes place by electron transfer as opposed
to the HAT that takes place from the hydropersulfide at lower
pH (Figure 3E). This mechanism is sometimes referred to as se-
quential proton loss electron transfer or SPLET.37 To provide ad-
ditional insight on this apparent change in mechanism, deuter-
ium kinetic isotope effects were determined at acidic pH (4), neu-
tral pH (7) and basic pH (9) by running the inhibited autoxida-
tions in buffers made in D2O as opposed to H2O (Figure 3F). In
the event, the DKIE decreased from 2.5 at pH 4 (similar to the
value obtained in chlorobenzene, vide supra) to 1.8 at pH 7, and
then vanished at pH 9 (0.8 ± 0.2). These trends are fully con-
sistent with an increasing contribution of the SPLET mechanism
in the reaction of the hydropersulfide with peroxyl radicals on
going from pH 4 to 7 and this mechanism operating exclusively
at basic pH.
Hydropersulfides Have Enhanced Reactivity to Alkyl, Alkoxyl
and Thiyl Radicals Compared to Thiols. The H-atom transfer
chemistry of hydropersulfides was further investigated with alkyl,
alkoxyl and thiyl radicals. Alkyl radicals were considered first, as
they are known to react very quickly with thiols (k ~ 106 M-1 s-
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 12
1
38,49
)
in one of the two propagating steps in the thiol-ene reac- HAT between thiols and thiyl radicals, despite being a ther-
1
2
3
4
5
6
7
8
tion. To do so, a classic alkyl radical clock approach was em-
ployed which utilized the 5-exo trig cyclization of the 6,6-diphe-
nylhexenyl radical (kc = 5.0 ´ 107 s-1) as the reference reaction
(Figure 4A).50 The PTOC ester precursor to this radical was syn-
thesized as described by Newcomb,50,51 who had used it to ‘clock’
the reactions of selenols with alkyl radicals. Similarly, we photo-
lyzed the ester (at 25 °C in benzene) and determined the ratio of
the linear and cyclized hydrocarbons (by GC) as a function of
hydropersulfide concentration in order to obtain the second or-
der rate constant for trapping the primary alkyl radical (Figure
4B) using Eq. (1):
moneutral process, is a fast reaction. Literature reports offer rate
constants that span 104 to 107 M-1 s-1, and our own CBS-QB3
calculations suggest a rate constant of 2.2 ´ l07 M-1 s-1 for the reac-
tion between n-butylthiyl radical and n-butylthiol.54 Thus, the anal-
ogous reaction between hydropersulfides and thiyl radicals,
which would have a considerable driving force, could be diffu-
sion-controlled. Despite the fact that thiyl radicals are quite easily
generated (by photolysis of disulfides), they are challenging spe-
cies to follow by optical spectroscopy due to their lack of absorb-
ance > 300 nm (aromatic thiyls being the exception, vide infra).
Fortunately, the perthiyl radical has an absorbance at 375 nm –
enabling monitoring of the reaction by growth of this signal ra-
ther than consumption of the thiyl. Regardless, when n-butyl di-
sulfide was photolyzed in the presence of cumyl-SSH or t-dodecyl-
SSH, we were unable to fully resolve the growth of the corre-
sponding perthiyl radical. Considering the experimental condi-
tions and the time resolution of our instrument, we estimate a
lower-bounds for the rate constant of 1 ´ 1010 M-1 s-1. Efforts to
compute a rate constant using the same CBS-QB3 approach that
has yielded highly accurate predictions with peroxyl, alkyl and
alkoxyl radicals gave inconsistent results. Initially, when MeSSH
was used as a simplified model, a predicted rate constant of 4.2
´ 109 M-1 s-1 was obtained, but the larger t-BuSSH yielded kH =
1.2 ´ 1010 M-1 s-1 (∆G‡ = 5.6 kcal/mol). The origins of this differ-
ence are unclear. Nevertheless, it seems reasonable to suggest that
these reactions are indeed diffusion-controlled.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
[Linear]
k
k
=
H [RSSH]
(1)
[Cyclized]
C
The reactions of the two hydropersulfides yielded essentially
indistinguishable rate constants, viz. kH = (5.2 ± 0.9) ´ 108 and
(5.6 ± 2.0) ´ 108 M-1 s-1, for cumyl-SSH and t-dodecyl-SSH, respec-
tively. Corresponding calculations of the HAT transition state
(Figure 4G) yielded a free energy barrier of 7.8 kcal/mol, which
corresponds to a rate constant of 3.0 ´ 108 M-1 s-1, in excellent
agreement with experiment. Once again, the driving force for
HAT is considerable (∆G0 = -30.2 kcal/mol), as a result of the
weak RSS-H bond.
H-atom transfer kinetics to alkoxyl radicals were determined
by transient absorption spectroscopy. Cumyloxyl radicals were
generated by photolysis of dicumylperoxide (Figure 4C) with the
308 nm emission of a nanosecond-pulsed XeCl excimer laser and
the rates of decay of their absorbance at 485 nm were determined
as a function of hydropersulfide concentration (Figure 4D). The
rate constants were expectedly greater than those measured
above, at (8.3 ± 0.4) ´ 108 and (1.1 ± 0.1) ´ 109 M-1 s-1 for cumyl-
SSH and t-dodecyl-SSH, respectively, consistent with the greater
driving force for the H-atom transfer of -33.9 kcal/mol (calcu-
lated by CBS-QB3, Figure 4H). Again, computation predicts a
barrier to reaction (∆G‡ = 7.1 kcal/mol) that yields a calculated
rate constant in excellent agreement with experiment (9.5 ´ 108
M-1 s-1). These rate constants are all significantly faster than those
reported for the reactions of thiols with tertiary alkoxyl radicals
(e.g. 6.5 ´ 107 M-1 s-1 for t-BuO• + n-C6H13SH).52
Transient absorption spectroscopy was also used to obtain ki-
netics for the reaction of the hydropersulfides with the triplet
state of benzophenone (corresponding data can be found in the
Supporting Information). We expected similar results to those
obtained with cumyloxyl radicals since the ꢀ → ꢁ∗ state of the
carbonyl has an effectively unpaired electron on the oxygen
atom. As for the reaction with cumyloxyl radicals, the experiment
was performed under pseudo-first order conditions and the ben-
zophenone triplet state monitored at 530 nm following photoly-
sis at 308 nm. The resulting rate constants were determined to
be (1.1 ± 0.1) ´ 109 and (1.5 ± 0.1) ´ 109 M-1s-1 for cumyl-SSH and
t-dodecyl-SSH, respectively. Once again, the results of CBS-QB3
calculations are in excellent agreement with experiment, predict-
ing a slightly faster reaction than with alkoxyl radicals (k = 1.3 ´
109 M-1 s-1, ∆G‡ = 6.9 kcal/mol) and a driving force of ∆G0 =
-36.5 kcal/mol. For comparison, thiols are reported to react
with the benzophenone triplet much more slowly (e.g. 8.8 ´ 106
M-1 s-1 for n-C5H12SH).53
Phenyldisulfide was also used as a thiyl radical precursor (Fig-
ure 4E). We reasoned that the lower S-H BDE in thiophenol (83
kcal/mol)55 would render the reaction of phenylthiyl radicals
with the hydropersulfides less exergonic and this may afford the
opportunity to measure its kinetics. Moreover, the phenylthiyl
radical is characterized by a broad absorption band centered
around 460 nm such that its decay in the presence of hydroper-
sulfides could be followed directly by LFP (Figure 4F). Indeed,
the rate constants for HAT could be determined as (2.4 ± 0.1) ´
108 and (3.7 ± 0.1) ´ 108 M-1 s-1 for cumyl-SSH and t-dodecyl-SSH,
respectively. (Similar data were obtained for reactions of the hy-
dropersulfides with 2-pyridinethiyl radicals, see the Supporting
Information for the data). In this case it is difficult to compare
these results to the kinetics of the corresponding reaction of a
thiol since it would be endothermic. Interestingly, the calculated
transition state structure for the reaction between the model hy-
dropersulfide (MeSSH) and phenylthiyl radical (Figure 4I) sug-
gests some interaction between the SOMO of the phenylthiyl
radical and the HOMO of the hydropersulfide – a secondary or-
bital interaction not unlike that seen in the reaction with peroxyl
radicals (cf. Figure 2G). The associated free energy barrier is com-
puted to be ∆G‡ = 7.7 kcal/mol, equivalent to a rate constant of
3.6 ´ 108 M-1 s-1, in excellent agreement with experiment. Expect-
edly, the driving force for HAT from the hydropersulfide to phe-
nylthiyl (∆G0 =
to n-butylthiyl (∆G0 =
-
11.2 kcal/mol) is predicted to be smaller than
15.4 kcal/mol).
-
n
DISCUSSION
Thiols are often described as ‘antioxidants’, but this can be mis-
leading. Their ionic reactions with H2O2 and hydroperoxides are
not particularly fast (k ~ 1 M-1 s-1 at neutral pH),56 so they are used
6
ACS Paragon Plus Environment

Page 7 of 12
Journal of the American Chemical Society
instead as reducing co-factors for enzymes (e.g. glutathione perox- reaction stoichiometry decreases sharply in aqueous solution (0.3
1
2
3
4
5
6
7
8
idases) that catalyze reactions with these oxidants with kinetics
that are greater by several orders of magnitude. Likewise, HAT
from thiols to peroxyl radicals, the key chain-carrying radicals in
lipid peroxidation, are quite slow (<103 M-1 s-1),28,57 making it im-
possible for them to compete with propagation of the peroxida-
tion chain reaction. In a biological context, this task is largely the
responsibility of a-tocopherol (a-TOH), owing to its much
greater reactivity to peroxyl radicals (> 106 M-1 s-1).29
at pH 4) and is further eroded with increasing pH where decom-
position is accelerated.33 Hydropersulfides are known to decom-
pose to yield mixtures of polysulfides and thiols,33 which are un-
reactive to peroxyl radicals.36
In contrast to the insignificant role of HAT reactions between
thiols and peroxyl radicals, thiols are important reagents in a va-
riety of alkyl radical-mediated reactions, perhaps most im-
portantly in the thiol-ene reaction (Scheme 1).
9
According to the Evans-Polanyi principle and estimates of the
RSS-H bond strength (70 kcal/mol),30 hydropersulfides should
undergo HAT reactions more readily than thiols (87 kcal/mol).31
The foregoing results establish that this is indeed true for HAT
to peroxyl radicals as well as alkyl, alkoxyl and thiyl radicals. The
inherent reactivity of hydropersulfides to peroxyl radicals is par-
ticularly striking because it is 4 orders of magnitude greater than
for thiols, and essentially the same as that of a-TOH, long con-
sidered the gold-standard against which other radical-trapping
antioxidants are measured.58,59 In fact, upon transitioning from
hydrocarbons to media containing H-bond accepting functional-
ity, hydropersulfides become superior to a-TOH, owing to the
fact that they do not engage in strong H-bonds. The weak H-bond
donating acidities of hydropersulfides (αH2 ~ 0.1) make them ideal
HAT agents to peroxyl radicals, as all other classes that have sim-
ilarly high reactivity (kinh > 106 M-1 s-1), which include some phe-
nols,42 diarylamines,60 sulfenic acids45,61 and hydroxylamines62 are
characterized by significantly higher H-bond acidities (generally
αH2 ³ 0.4). Moreover, the perthiyl radicals that arise upon HAT
from hydropersulfides persist until they encounter another
perthiyl radical with which they form tetrasulfides rather than
carry on the oxidation via reactions with either O2 or the sub-
strate. These results suggest that unlike thiols, hydropersulfides,
produced when cells synthesize H2S, prevent lipid peroxidation.
In order to operate effectively in this capacity, the disulfides
or other electrophilic thiol derivatives from which the hydroper-
sulfides are produced should be relatively lipophilic. This ensures
they have access to lipid regions where they are more likely to
encounter lipid-derived peroxyl radicals, and because the hy-
dropersulfides are much less effective as H-atom donors in aque-
ous media. It is important to make the distinction between effec-
tive and reactive, since the hydropersulfides are highly reactive to
peroxyl radicals in aqueous solution. In fact, over a range of pH
from 4 to 10, kinh varied from 0.5 to 4.4 ´ 106 M-1 s-1 – greatly
exceeding the corresponding rate constants obtained with
Trolox, a water-soluble derivative of a-TOH (k = 4.1 ´ 105 M-1 s-
1), which varies little in the same pH range (k = 2.1 ´ 105 M-1 s-1
and k = 6.8 ´ 105 M-1 s-1 at pH 4 and 9, respectively).46 The pH
dependence of hydropersulfide reactivity results from the increas-
ing contribution of a faster electron transfer reaction between the
persulfide anion and a peroxyl radical as the pH approaches –
and then exceeds – the estimated pKa of the hydropersulfide (ca.
7).47,48 It is noteworthy that the position of the inflection point
in the sigmoidal relationship between the inhibition rate con-
stant and the pH nicely corroborate estimated pKa values re-
ported in the literature.
kadd
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
RS
RS
+
R'
R'
kHAT
RS
RS
+
RSH
RS
+
R'
R'
Scheme 1. The radical chain propagation steps of a generalized
thiol-ene reaction.
A corresponding hydropersulfide-ene reaction would, in prin-
ciple, provide a route to unsymmetric disulfides. Although hy-
dropersulfides are two orders of magnitude more reactive to (pri-
mary) alkyl radicals than thiols, and it is often the HAT step that
is rate-controlling in the propagation sequence, it is likely that
the perthiyl addition would be too thermodynamically unfavour-
able (CBS-QB3 predicts the reaction with styrene to have DG° =
+9.4 kcal/mol). Indeed, while we have estimated that styrene re-
acts with t-butylperthiyl radicals generated by laser flash photoly-
sis of t-BuSSSSt-Bu with k = 1.1 ´ 104 M-1 s-1 (see Supporting In-
formation) this was only observed in the presence of O2, which
presumably trapped the intermediate benzylic radical precluding
the reverse reaction (fragmentation).63 This can be compared to
the addition of thiyl radicals to styrene, which proceeds 200-fold
faster, and is slightly exothermic (CBS-QB3 predicts the reaction
to have DG° = -4.5 kcal/mol).64 Although it seems unlikely that
disulfides can be prepared via a hydropersulfide-ene reaction, hy-
dropersulfides may yet find use in place of thiols for other trans-
formations.
Our expectations for the greater HAT reactivity of hydroper-
sulfides compared to thiols are based on the S-H BDE estimates
that were made on the basis of Benson’s group additivity
scheme.30,47,65 These estimates have been corroborated by the
high accuracy quantum chemical calculations (CBS-QB3) we
have made use of here, which suggest BDEs of 70.8, 70.4 and
70.4 kcal/mol in MeSSH, t-BuSSH and cumyl-SSH, respectively.
Given the similarity of the hydropersulfide S-H BDE with that of
the hydroxylamine derived from the persistent nitroxide
TEMPO (69.6 kcal/mol),66 we attempted experimental confirma-
tion of a hydropersulfide S-H BDE by measuring the equilibrium
constant for eq 2 by EPR – an approach used to obtain many X-
H BDEs (making the reasonable assumption that DS ~ 0 for the
HAT process).67,68
RSSH + TEMPO• D RSS• + TEMPOH
(2)
Sadly, despite extensive efforts, we were unable to resolve a
signal for the perthiyl radical. Expecting that this may be due to
its rapid dimerization to the tetrasulfide (k = 6.0 ´ 109 M-1 s-1),69
we also carried out the reaction under continuous photolysis, but
to no avail (see Supporting Information). It is important to note
that Fukuto, Toscano and co-workers recently reported similar
The difficulty associated with measuring a reliable pKa value
for hydropersulfides underlies why they are relatively ineffective
radical-trapping antioxidants in aqueous solutions: they decom-
pose too readily. As such, where the hydropersulfides each
cleanly trap one peroxyl radical by HAT in organic solution, the
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 12
results.70 This led us to suspect that the nitroxide and perthiyl
measurements were carried out using an Accument Electrode ref-
1
2
3
4
5
6
7
8
may react with each other. Indeed, perthiyl radicals generated by
laser flash photolysis of solutions of di-tert-butyl tetrasulfide gave
clean pseudo-first order decay kinetics in the presence of
TEMPO that afforded a second order rate constant of (1.9 ± 0.1)
´ 108 M-1 s-1. The formation of the perthiyl-TEMPO adduct is
readily reversible since the central S-O bond is very weak (DG° =
6.0 kcal/mol, by CBS-QB3), but competitive N-O bond cleavage
of the adduct (which requires only 4.2 kcal/mol more free en-
ergy, by CBS-QB3) leads to reactive products, precluding the re-
verse reaction. This would account for the inability to observe
the perthiyl radical in the radical equilibration experiment.71
erenced to Ag/AgCl. pD were measured by adding a correction
factor of 0.4 to the pH readings (pD = pHread + 0.4).78 Unless oth-
erwise indicated, the errors which are reported were obtained
from the standard deviation of at least three experimental trials.
Synthetic Procedures. The syntheses of cumyl-SSH and t-do-
decyl-SSH have been achieved following a similar approach to
Pluth’s reported methods.33 For the complete synthetic proce-
dure of the intermediates see supporting information.
Cumylhydropersulfide. Cumylacetyldisulfide (0.10 g, 0.44
mmol) was dissolved in methanol under nitrogen at 0 °C. In a
separate flask, a 5 M methanolic solution of HCl was made by
the addition of acetyl chloride (0.89 mL) to methanol (1.61 mL)
at 0 °C. The HCl solution (0.30 mL, 1.50 mmol) was added drop-
wise to the cumylacetyldisulfide solution after which the flask was
equipped with a nitrogen balloon directly on the septum, with-
out the use of a needle to avoid any metal from making contact
in the flask. The reaction was stirred overnight at room tempera-
ture and the solvent was evaporated by rotary evaporation with a
water bath at room temperature to yield the product as a clear
colorless oil (67 mg, 83%). 1H NMR (400 MHz; CDCl3): δ 7.52-
7.49 (m, 2H), 7.38-7.33 (m, 2H), 7.28-7.23 (m, 1H), 2.62 (s, 1H),
1.76 (s, 6H). 13C NMR (101 MHz; CDCl3): δ 144.4, 128.4, 127.2,
126.8, 51.3, 28.0. HRMS (ESI, [M-H+]): m/z calc. for C9H11S2:
183.0307, found 183.0302.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
n
CONCLUSIONS
Hydropersulfides are excellent H-atom donors towards alkyl (k
~ 5 ´ 108 M-1 s-1), alkoxyl (k ~ 1 ´109 M-1s-1), peroxyl (k ~ 2 ´ 106
M-1 s-1) and thiyl (k > 1 ´ 1010 M-1 s-1) radicals – besting thiols in
all cases by at least one, and up to four, orders of magnitude. The
inherently high reactivity of hydropersulfides to H-atom transfer
is based largely on thermodynamic factors; the weak RSS-H BDE
(~ 70 kcal/mol) and associated high stability of the perthiyl radi-
cal makes the foregoing reactions exothermic by 15 to 34
kcal/mol. Given the implication of hydropersulfides in redox bi-
ology, their (peroxyl) radical-trapping antioxidant activity may be
the most relevant reactivity studied here. The favourable thermo-
dynamics of the reaction are bolstered by secondary orbital inter-
actions in the HAT transition state that enhance the kinetics.
Furthermore, the persistence of the product perthiyl radicals –
toward both O2 and the substrate undergoing autoxidation – is
key, as is their near-diffusion-controlled combination to give in-
nocuous tetrasulfide products. A highly unique attribute of hy-
dropersulfides as radical-trapping antioxidants is their low H-
bond acidity, which makes their reactivity relatively insensitive to
changes in solvent/medium. The reactivity of well-established
radical-trapping antioxidants, such as phenols and diarylamines,
are highly medium-dependent owing to strong H-bond interac-
tions of the H-atom undergoing HAT. The radical-trapping anti-
oxidant activity of hydropersulfides may be exploited by am-
phiphilic disulfides (thiols) in small molecules and/or proteins
exposed to exogenous H2S, which is known to readily transverse
and accumulate in lipid bilayers.72-74 Whether this reactivity
could be exploited in the design of small molecule preventive
and/or therapeutic agents for diseases wherein lipid peroxida-
tion (ferroptosis)75-77 has been implicated remains to be explored.
t-Dodecylhydropersulfide. t-Dodecylacetyldisulfide (0.10 g, 0.36
mmol) was dissolved in isopropanol under nitrogen at 0 °C. In a
separate flask, a 5 M methalonic solution of HCl was made by
the addition of acetyl chloride (0.89 mL) to methanol (1.61 mL)
at 0 °C. The HCl solution (0.25 mL, 1.24 mmol) was added drop-
wise to the t-butyldodecylacetyldisulfide solution after which the
flask was equipped with a nitrogen balloon directly on the sep-
tum, without the use of a needle to avoid any metal from making
contact in the flask. The reaction was stirred overnight at room
temperature and the solvent was evaporated by rotary evapora-
tion with a water bath at room temperature to yield the product
1
as a clear colorless oil (74 mg, 87%). H NMR (300 MHz;
CDCl3): δ 2.63 (s, 1H), 1.57-1.52 (m, 2H), 1.37-1.32 (m, 3H),
1.27 (br s, 17H), 0.88 (t, J = 6.7 Hz, 3H). 13C NMR (76 MHz;
CDCl3): δ 48.9, 40.5, 32.0, 30.2, 29.8, 29.5, 26.8, 24.8, 22.8,
14.3. HRMS (ESI, [M-H+]): m/z calc. for C12H25S2: 233.1398,
found 233.1394.
General Procedure for Inhibited Autoxidations. Inhibited au-
toxidations were carried out following our recently published
method.35 Similar procedures were used for all autoxidations re-
ported herein, the procedure for THF autoxidation in chloro-
benzene is given as an example. For detailed autoxidation proce-
dures of other substrates in various solvents, see supporting in-
formation. A 3.5 mL quartz cuvette was loaded with unstabilized
THF (0.625 mL) along with PhCl (1.805 mL), such that the final
reaction volume is 2.50 mL. The cuvette was then pre-heated in
a thermostatted sample holder of a UV-vis spectrophotometer
and allowed to equilibrate to 37 °C for approximately 15
minutes. A small volume (12.5 µL) of a 2.00 mM solution of the
PBD-BODIPY probe in 1,2,4-trichlorobenzene was added, fol-
lowed by 50 µL of a 300 mM solution of AIBN in PhCl. The
solution was thoroughly mixed and the absorbance at 588 nm
was monitored for 10 minutes to ensure that the reaction was
proceeding at a constant rate, after which 10.0 µL of a solution
of the test antioxidant was added. The solution was thoroughly
mixed and the absorbance readings resumed. The resulting data
n
EXPERIMENTAL SECTION
General. Reagents were obtained from commercial suppliers and
used as is, unless indicated otherwise. Column chromatography
was carried out with 40-63 µm, 230-400 mesh silica gel. 1H and
13C NMR spectra were recorded on a Bruker AVANCE spec-
trometer at 400 MHz and 101 MHz, respectively, unless indi-
cated otherwise. High-resolution mass spectra were obtained on
a Kratos Concept Tandem mass spectrometer (EI) and Micro-
mass Q-TOF (ESI). PBD-BODIPY and STY-BODIPY were syn-
thesized following our previously reported procedure.35 Chloro-
benzene was dried over 3 Å molecular sieves before use. UV-
visible spectra were measured with a Cary 100 spectrophotome-
ter equipped with a thermostatted 6 x 6 multi-cell holder. pH
8
ACS Paragon Plus Environment

Page 9 of 12
Journal of the American Chemical Society
The manuscript was written through contributions of all au-
was processed as previously reported.35 The rate of initiation (Ri
= 3.2 ´ 10-9 M s-1) and propagation rate constant for the dye (kPBD-
BODIPY = 2240 M-1 s-1) necessary to derive stoichiometric data and
inhibition rate constants, were determined using PMC as a stand-
ard, which has an established stoichiometry of 2.79
1
thors. All authors have given approval to the final version of the
manuscript.
2
3
4
5
Notes
The authors declare no competing financial interest.
Alkyl Radical Clock Kinetics. Alkyl radical clock kinetics were
determined using a previously reported procedure by New-
comb.50,51 The PTOC ester and the standards were synthesized
according to the procedures described therein. Briefly, vials
equipped with a septum and wrapped in aluminum foil were
loaded with the appropriate amount of hydropersulfides (50-300
mM) and PTOC ester clock (20 mM) in THF (1 mL). The vials
were thoroughly purged with nitrogen for 10 minutes before the
aluminum foil was removed and the vials exposed to a high pres-
sure sodium lamp (400 W) at a distance of 50 cm for 1 h. A 50
µL aliquot was taken from each vial and added to a GC vial con-
taining 50 µL of a hexylbenzene standard solution (40 mM). The
solution was diluted by adding 900 µL of benzene for a total vol-
ume of 1 mL per vial. The samples (4 µL splitless injections) were
analyzed by GC-MS equipped with an Agilent HP-5 ms column
(30 m x 0.25 mm x 0.25 µm) with a constant He flow of 1.2 mL
min-1 using the following temperature profile (inlet temperature
was set at 260 °C): 100 °C hold 3 min, 8 °C min-1 to 260 °C, hold
12 min. The method yielded retention times of 7.1, 15.9, and
16.7 min for hexylbenzene, linear and cyclized products, respec-
tively.
Laser Flash Photolysis. Nanosecond transient absorption exper-
iments were performed on an LFP-112 spectrometer (Luzchem,
Canada). Irradiation source was an EX10 (GAM Laser, USA)
XeCl Excimer laser (308 nm, ca. 10 mJ/pulse, ca. 12 ns pulse
width). The transient absorption data were recorded in a quartz
cuvette (1 cm x 1 cm) equipped with a septum. Sample concen-
trations in benzene/PhCl were adjusted to yield an absorbance
of 0.2-0.3 at 308 nm and the solutions were bubbled with nitro-
gen (or oxygen) for 10 minutes before measurement. The rate
constants for hydrogen abstraction (kH) were determined in
pseudo-first-order according to the following equation
kobs = k0 + kH[RSSH]. Each plotted data point is the average of 6-
12 individual traces (corresponding standard deviation plotted as
error bars).
6
7
8
9
ACKNOWLEDGMENT
This work was supported by grants from the Natural Sciences and
Engineering Research Council of Canada and the Canada Founda-
tion for Innovation, and through generous access to the computa-
tional resources of the Centre for Advanced Computing
(cac.queensu.ca). DAP and JPRC also acknowledge the support of
the Canada Research Chairs and Ontario Graduate Scholarships
programs, respectively. The authors thank Luke Farmer for assis-
tance with the standardization of autoxidation kinetics in diox-
ane/PhCl and dioxane/DMSO/PhCl as well as the βH2 values for
hexadecene/PhCl, THF/PhCl, dioxane/PhCl and diox-
ane/DMSO/PhCl solvent systems.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
n REFERENCES
(1) Ida, T.; Sawa, T.; Ihara, H.; Tsuchiya, Y.; Watanabe, Y.;
Kumagai, Y.; Suematsu, M.; Motohashi, H.; Fujii, S.; Matsunaga, T.;
Yamamoto, M.; Ono, K.; Devarie-Baez, N. O.; Xian, M.; Fukuto, J.
M.; Akaike, T. Proc. Natl. Acad. Sci. U.S.A. 2014, 111, 7606.
(2) Dóka, É.; Pader, I.; Bíró, A.; Johansson, K.; Cheng, Q.; Bal-
lagó, K.; Prigge, J. R.; Pastor-Flores, D.; Dick, T. P.; Schmidt, E. E.;
Arnér, E. S. J.; Nagy, P. Sci. Adv. 2016, 2, e1500968.
(3) Longen, S.; Richter, F.; Köhler, Y.; Wittig, I.; Beck, K.-F.;
Pfeilschifter, J. Sci. Rep. 2016, 6, 29808.
(4) Mustafa, A. K.; Gadalla, M. M.; Sen, N.; Kim, S.; Mu, W.;
Gazi, S. K.; Barrow, R. K.; Yang, G.; Wang, R.; Snyder, S. H. Sci.
Signal. 2009, 2, ra72.
(5) Abe, K.; Kimura, H. J. Neurosci 1996, 16, 1066.
(6) Chen, X.; Jhee, K.-H.; Kruger, W. D. J. Biol. Chem. 2004, 279,
52082.
(7) Swaroop, M.; Bradley, K.; Ohura, T.; Tahara, T.; Roper, M.
D.; Rosenberg, L. E.; Kraus, J. P. J. Biol. Chem. 1992, 267, 11455.
(8) Shibuya, N.; Tanaka, M.; Yoshida, M. Antioxid. Redox. Signal.
2009, 11, 703.
(9) Paul, B. D.; Snyder, S. H. Nat. Rev. Mol. Cell Biol. 2012, 13,
449.
(10) Yang, G.; Wu, L.; Jiang, B.; Yang, W.; Qi, J.; Cao, K.; Meng,
Q.; Mustafa, A. K.; Mu, W.; Zhang, S.; Snyder, S. H.; Wang, R. Sci-
ence 2008, 322, 587.
(11) Morikawa, T.; Kajimura, M.; Nakamura, T.; Hishiki, T.;
Nakanishi, T.; Yukutake, Y.; Nagahata, Y.; Ishikawa, M.; Hattori, K.;
Takenouchi, T.; Takahashi, T.; Ishii, I.; Matsubara, K.; Kabe, Y.;
Uchiyama, S.; Nagata, E.; Gadalla, M. M.; Snyder, S. H.; Suematsu,
M. Proc. Natl. Acad. Sci. U.S.A. 2012, 109, 1293.
Calculations. Calculations were carried out using the CBS-
QB380 complete basis set method as implemented in the Gauss-
ian 0981 suite of programs. Rate constants were calculated via
transition state theory at 25 °C, except for peroxyls (37 °C).
n
ASSOCIATED CONTENT
Supporting Information
The Supporting Information is available free of charge on the ACS
Publications website. NMR spectra of new compounds, KIE experi-
ments, details of LFP experiments, and optimized geometries and
energies for computational results. (PDF)
(12) Fu, M.; Zhang, W.; Wu, L.; Yang, G.; Li, H.; Wang, R. Proc.
Natl. Acad. Sci. USA 2012, 109, 2943.
(13) Sen, N.; Paul, B. D.; Gadalla, M. M.; Mustafa, A. K.; Sen, T.;
Xu, R.; Kim, S.; Snyder, S. H. Molecular Cell 2012, 45, 13.
(14) Kimura, Y.; Kimura, H. FASEB J. 2004, 18, 1165.
(15) Yang, G.; Zhao, K.; Ju, Y.; Mani, S.; Cao, Q.; Puukila, S.;
Khaper, N.; Wu, L.; Wang, R. Antioxid. Redox. Signal. 2013, 18, 1906.
(16) Mishanina, T. V.; Libiad, M.; Banerjee, R. Nat. Chem. Biol.
2015, 11, 457.
(17) Nishida, M.; Sawa, T.; Kitajima, N.; Ono, K.; Inoue, H.;
Ihara, H.; Motohashi, H.; Yamamoto, M.; Suematsu, M.; Kurose, H.;
van der Vliet, A.; Freeman, B. A.; Shibata, T.; Uchida, K.; Kumagai,
Y.; Akaike, T. Nat. Chem. Biol. 2012, 8, 714.
n AUTHOR INFORMATION
Corresponding Author
* dpratt@uottawa.ca
Author Contributions
9
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 12
(18) Yadav, P. K.; Martinov, M.; Vitvitsky, V.; Seravalli, J.; Wed-
mann, R.; Filipovic, M. R.; Banerjee, R. J. Am. Chem. Soc. 2016, 138,
289.
(19) Pan, J.; Carroll, K. S. ACS Chem. Biol. 2013, 8, 1110.
(20) Zhang, D.; Macinkovic, I.; Devarie-Baez, N. O.; Pan, J.; Park,
C.-M.; Carroll, K. S.; Filipovic, M. R.; Xian, M. Angew. Chem. Int. Ed.
2014, 53, 575.
(21) Galardon, E.; Padovani, D. Bioconjugate Chem. 2015, 26,
1013.
(45) Zielinski, Z.; Presseau, N.; Amorati, R.; Valgimigli, L.; Pratt,
D. A. J. Am. Chem. Soc. 2014, 136, 1570.
(46) Amorati, R.; Baschieri, A.; Morroni, G.; Gambino, R.; Val-
gimigli, L. Chem. Eur. J. 2016, 22, 7924.
(47) Everett, S. A.; Folkes, L. K.; Wardman, P.; Asmus, K. D. Free
Radic Res 1994, 20, 387.
(48) Cuevasanta, E.; Lange, M.; Bonanata, J.; Coitiño, E. L.; Fer-
rer-Sueta, G.; Filipovic, M. R.; Alvarez, B. J. Biol. Chem. 2015, 290,
26866.
(49) Newcomb, M.; Glenn, A. G.; Manek, M. B. J. Org. Chem.
1989, 54, 4603.
1
2
3
4
5
6
7
8
(22) Han, X.; Yu, F.; Song, X.; Chen, L. Chemical Science 2016, 7,
5098.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(23) Wedmann, R.; Onderka, C.; Wei, S.; Szijártó, I. A.;
Miljkovic, J. L.; Mitrovic, A.; Lange, M.; Savitsky, S.; Yadav, P. K.;
Torregrossa, R.; Harrer, E. G.; Harrer, T.; Ishii, I.; Gollasch, M.;
Wood, M. E.; Galardon, E.; Xian, M.; Whiteman, M.; Banerjee, R.;
Filipovic, M. R. Chem. Sci. 2016, 7, 3414.
(24) Ono, K.; Akaike, T.; Sawa, T.; Kumagai, Y.; Wink, D. A.;
Tantillo, D. J.; Hobbs, A. J.; Nagy, P.; Xian, M.; Lin, J.; Fukuto, J. M.
Free Radic. Biol. Med. 2014, 77 (C), 82.
(25) Saund, S. S.; Sosa, V.; Henriquez, S.; Nguyen, Q. N. N.;
Bianco, C. L.; Soeda, S.; Millikin, R.; White, C.; Le, H.; Ono, K.;
Tantillo, D. J.; Kumagai, Y.; Akaike, T.; Lin, J.; Fukuto, J. M. Arch.
Biochem. Biophys. 2015, 588 (C), 15.
(26) Millikin, R.; Bianco, C. L.; White, C.; Saund, S. S.; Hen-
riquez, S.; Sosa, V.; Akaike, T.; Kumagai, Y.; Soeda, S.; Toscano, J.
P.; Lin, J.; Fukuto, J. M. Free Radic. Biol. Med. 2016, 97 (C), 136.
(27) Dénès, F.; Pichowicz, M.; Povie, G.; Renaud, P. Chem. Rev.
2014, 114, 2587.
(28) Chenier, J. H. B.; Furimsky, E.; Howard, J. A. Can. J. Chem.
1974, 52, 3682.
(29) Burton, G. W.; Ingold, K. U. J. Am. Chem. Soc. 1981, 103,
6472.
(30) Benson, S. W. Chem. Rev. 1978, 78, 23.
(31) Janousek, B. K.; Reed, K. J.; Brauman, J. I. J. Am. Chem. Soc.
1980, 102, 3125.
(32) H-atom abstraction from RSSH by an alkyl radical (propan-2-
ol-2-yl) generated using pulse radiolysis at pH 4 yielded a rate con-
stant of k = 2.4 ´ 109 M-1 s-1. See Everett, S. A.; Folkes, L. K.; Ward-
man, P.; Asmus, K. D. Free Radic. Res. 1994, 20, 387.
(33) Although these might be the initial reaction products, poly-
sulfides and thiols are generally the end products.33a-b (a) Bailey, T.
S.; Zakharov, L. N.; Pluth, M. D. J. Am. Chem. Soc. 2014, 136, 10573.
(b) Bailey, T. S.; Pluth, M. D. Free Radic. Biol. Med. 2015, 89, 662.
(34) Howard, J. A.; Ingold, K. U. Can. J. Chem. 1962, 40, 1851.
(35) Haidasz, E. A.; Van Kessel, A. T. M.; Pratt, D. A. J. Org. Chem.
2016, 81, 737.
(50) Ha, C.; Horner, J. H.; Newcomb, M.; Varick, T. R.; Arnold,
B. R.; Lusztyk, J. J. Org. Chem. 1993, 58, 1194.
(51) Newcomb, M.; Choi, S.-Y.; Horner, J. H. J. Org. Chem. 1999,
64, 1225.
(52) Denisov, E.; Chatgilialoglu, C.; Shestakov, A.; Denisova, T.
Int. J. Chem. Kinet. 2009, 41, 284.
(53) Inbar, S.; Linschitz, H.; Cohen, S. G. J. Am. Chem. Soc. 1982,
104, 1679.
(54) The rate constant for the reaction of C6H13S• and C8H17SH
has been estimated to be k = 2.7 ´ 104 M-1 s-1,54a while in aqueous
medium the rate constant for the reaction of 2-hydroxyethanethiyl
radical with 1,4-dithiothreitol has been measured at 1.7 ´ l07 M-1 s-
.
1 54b (a) Alnajjar, M. S.; Garrossian, M. S.; Autrey, S. T.; Ferris, K. F.;
Franz, J. A. J. Phys. Chem. 1992, 96, 7037. (b) Akhlaq, M. S.; Sonntag,
von, C. Z. Naturforsch. C Bio. Sci. 1987, 42, 134.
(55) There has been a range of experimental values reported, from
7755a to 83 kcal/mol.55b High level calculations are in the range of 81
(the CBS-QB3 method employed here) to 83 kcal/mol.56c (a) Lim, I.
S.; Lim, J. S.; Lee, Y. S.; Kim, S. K. J. Chem. Phys. 2007, 126, 034306.
(b) Borges dos Santos, R. M.; Muralha, V. S. F.; Correia, C. F.;
Guedes, R. C.; Costa Cabral, B. J.; Martinho Simões, J. A. J. Phys.
Chem. A 2002, 106, 9883. (c) Cabral do Couto, P.; Costa Cabral, B.
J.; Martinho Simões, J. A. Chem. Phys. Lett. 2006, 421, 504.
(56) Winterbourn, C. C.; Metodiewa, D. Free Radic. Biol. Med.
1999, 27, 322.
(57) Alkylthiols react so slowly that they cannot compete with pro-
pagation, precluding a determination of this rate constant. CBS-
QB3 calculations predict a rate constant of k = 190 M-1 s-1 for the
reaction of n-butylthiol with methylperoxyl radical.
(58) Ingold, K. U.; Pratt, D. A. Chem. Rev. 2014, 114, 9022.
(59) Valgimigli, L.; Pratt, D. A. Acc. Chem. Res. 2015, 48, 966.
(60) Hanthorn, J. J.; Pratt, D. A. J. Org. Chem. 2012, 77, 276.
(61) Amorati, R.; Lynett, P. T.; Valgimigli, L.; Pratt, D. A. Chem.
Eur. J. 2012, 18, 6370.
(62) Amorati, R.; Pedulli, G. F.; Pratt, D. A.; Valgimigli, L. Chem.
Commun. (Camb.) 2010, 46, 5139.
(63) Chatgilialoglu, C.; Ferreri, C. Acc. Chem. Res. 2005, 38, 441.
(64) Northrop, B. H.; Coffey, R. N. J. Am. Chem. Soc. 2012, 134,
13804.
(65) Everett, S. A.; Wardman, P. Meth. Enzymol. 1995, 251, 55.
(66) McGrath, A. J.; Garrett, G. E.; Valgimigli, L.; Pratt, D. A. J.
Am. Chem. Soc. 2010, 132, 16759.
(67) Lucarini, M.; Pedulli, G. F.; Cipollone, M. J. Org. Chem.
1994, 59, 5063.
(36) Chauvin, J.-P. R.; Haidasz, E. A.; Griesser, M.; Pratt, D. A.
Chem. Sci. 2016, 7, 6347.
(37) Litwinienko, G.; Ingold, K. U. Acc. Chem. Res. 2007, 40, 222.
(38) Tronche, C.; Martinez, F. N.; Horner, J. H.; Newcomb, M.;
Senn, M.; Giese, B. Tetrahedron Lett. 1996, 37, 5845.
(39) Abraham, M. H.; Abraham, R. J.; Byrne, J.; Griffiths, L. J.
Org. Chem. 2006, 71, 3389.
(40) Abraham, M. H.; Grellier, P. L.; Prior, D. V.; Duce, P. P.;
Morris, J. J.; Taylor, P. J. J. Chem. Soc., Perkin Trans. 2 1989, 6, 699.
(41) Snelgrove, D. W.; Lusztyk, J.; Banks, J. T.; Mulder, P.; Ingold,
K. U. J. Am. Chem. Soc. 2001, 123, 469.
(42) Valgimigli, L.; Amorati, R.; Petrucci, S.; Pedulli, G. F.; Hu,
D.; Hanthorn, J. J.; Pratt, D. A. Angew. Chem. Int. Ed. 2009, 48, 8348.
(43) Hanthorn, J. J.; Valgimigli, L.; Pratt, D. A. J. Am. Chem. Soc.
2012, 134, 8306.
(44) Vaidya, V.; Ingold, K. U.; Pratt, D. A. Angew. Chem. Int. Ed.
2009, 48, 157.
(68) Lucarini, M.; Pedrielli, P.; Pedulli, G. F.; Cabiddu, S.; Fat-
tuoni, C. J. Org. Chem. 1996, 61, 9259.
(69) The rate constants were determined to be (5.3 ± 0.9) ´ 109 M-
1 s-1 and (3.0 ± 0.3) ´ 109 M-1 s-1 for the cumyl and t-dodecyl perthiyl
radicals, respectively. We previously reported a recombination rate
for t-butyl perthiyl radicals of (6.0 ± 0.5) ´ 109 M-1 s-1. See ref 36.
(70) Bianco, C. L.; Chavez, T. A.; Sosa, V.; Saund, S. S.; Nguyen,
Q. N. N.; Tantillo, D. J.; Ichimura, A. S.; Toscano, J. P.; Fukuto, J.
M. Free Radic. Biol. Med. 2016, 101, 20.
10
ACS Paragon Plus Environment

Page 11 of 12
Journal of the American Chemical Society
(71) The cross-reactivity between TEMPO and perthiyl radicals
also suggests that the results of experiments wherein TEMPO has
1
2
3
4
5
6
7
8
9
been employed as the oxidant to produce perthiyl radicals from hy-
dropersulfides must be interpreted with some caution. In particular,
the lack of reactivity of perthiyl radicals to O2 and NO, which was
inferred on the basis that neither O2 nor NO were consumed when
a hydropersulfide was exposed to an excess of TEMPO could simply
be explained by the very rapid reaction of the perthiyl radical with
the excess TEMPO.
(72) Mathai, J. C.; Missner, A.; Kuegler, P.; Saparov, S. M.; Zeidel,
M. L.; Lee, J. K.; Pohl, P. Proc. Natl. Acad. Sci. USA 2009, 106, 16633.
(73) Cuevasanta, E.; Denicola, A.; Alvarez, B.; Möller, M. N. PLoS
ONE 2012, 7, e34562.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(74) Riahi, S.; Rowley, C. N. J. Am. Chem. Soc. 2014, 136, 15111.
(75) Dixon, S. J.; Lemberg, K. M.; Lamprecht, M. R.; Skouta, R.;
Zaitsev, E. M.; Gleason, C. E.; Patel, D. N.; Bauer, A. J.; Cantley, A.
M.; Yang, W. S.; Morrison, B., III; Stockwell, B. R. Cell 2012, 149,
1060.
(76) Friedmann Angeli, J. P.; Schneider, M.; Proneth, B.; Tyurina,
Y. Y.; Tyurin, V. A.; Hammond, V. J.; Herbach, N.; Aichler, M.;
Walch, A.; Eggenhofer, E.; Basavarajappa, D.; Radmark, O.; Koba-
yashi, S.; Seibt, T.; Beck, H.; Neff, F.; Esposito, I.; Wanke, R.; For-
ster, H.; Yefremova, O.; Heinrichmeyer, M.; Bornkamm, G. W.;
Geissler, E. K.; Thomas, S. B.; Stockwell, B. R.; O’Donnell, V. B.;
Kagan, V. E.; Schick, J. A.; Conrad, M. Nat Cell Biol 2014, 16, 1180.
(77) Zilka, O.; Shah, R.; Li, B.; Friedmann Angeli, J. P.; Griesser,
M.; Conrad, M.; Pratt, D. A. ACS Cent. Sci. 2017,
acscentsci.7b00028.
(78) Glasoe, P. K.; Long, F. A. J. Phys. Chem. 1960, 64, 188.
(79) Burton, G. W.; Doba, T.; Gabe, E.; Hughes, L.; Lee, F. L.;
Prasad, L.; Ingold, K. U. J. Am. Chem. Soc. 1985, 107, 7053.
(80) Montgomery, J. A., Jr; Frisch, M. J.; Ochterski, J. W.; Peters-
son, G. A. J. Chem. Phys. 1999, 110, 2822.
(81) Gaussian 09, Revision E.01. Frisch, M. J.; Trucks, G. W.;
Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scal-
mani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; Nakatsuji, H.;
Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.;
Zheng, G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fu-
kuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao,
O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr; Peralta, J. E.; Ogli-
aro, F.; Bearpark, M. J.; Heyd, J.; Brothers, E. N.; Kudin, K. N.;
Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.;
Rendell, A. P.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.;
Rega, N.; Millam, N. J.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken,
V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev,
O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin,
R. L.; Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.;
Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, Ö.; For-
esman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian, Inc.:
Wallingford, CT, USA 2009.
11
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 12
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
12
ACS Paragon Plus Environment