Selective, Modular Probes for Thioredoxins Enabled by Rational Tuning of a Unique Disulfide Structure Motif

Article abstract of DOI:10.1021/jacs.1c03234

Specialized cellular networks of oxidoreductases coordinate the dithiol/disulfide-exchange reactions that control metabolism, protein regulation, and redox homeostasis. For probes to be selective for redox enzymes and effector proteins (nM to μM concentrations), they must also be able to resist non-specific triggering by the ca. 50 mM background of non-catalytic cellular monothiols. However, no such selective reduction-sensing systems have yet been established. Here, we used rational structural design to independently vary thermodynamic and kinetic aspects of disulfide stability, creating a series of unusual disulfide reduction trigger units designed for stability to monothiols. We integrated the motifs into modular series of fluorogenic probes that release and activate an arbitrary chemical cargo upon reduction, and compared their performance to that of the literature-known disulfides. The probes were comprehensively screened for biological stability and selectivity against a range of redox effector proteins and enzymes. This design process delivered the first disulfide probes with excellent stability to monothiols yet high selectivity for the key redox-Active protein effector, thioredoxin. We anticipate that further applications of these novel disulfide triggers will deliver unique probes targeting cellular thioredoxins. We also anticipate that further tuning following this design paradigm will enable redox probes for other important dithiol-manifold redox proteins, that will be useful in revealing the hitherto hidden dynamics of endogenous cellular redox systems.

Full text of DOI:10.1021/jacs.1c03234

pubs.acs.org/JACS
Article
Selective, Modular Probes for Thioredoxins Enabled by Rational
Tuning of a Unique Disulfide Structure Motif
Jan G. Felber, Lukas Zeisel, Lena Poczka, Karoline Scholzen, Sander Busker, Martin S. Maier,
Ulrike Theisen, Christina Brandstädter, Katja Becker, Elias S. J. Arnér, Julia Thorn-Seshold,
and Oliver Thorn-Seshold*
Cite This: J. Am. Chem. Soc. 2021, 143, 8791−8803
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: Specialized cellular networks of oxidoreductases
coordinate the dithiol/disulfide-exchange reactions that control
metabolism, protein regulation, and redox homeostasis. For probes
to be selective for redox enzymes and effector proteins (nM to μM
concentrations), they must also be able to resist non-specific
triggering by the ca. 50 mM background of non-catalytic cellular
monothiols. However, no such selective reduction-sensing systems
have yet been established. Here, we used rational structural design
to independently vary thermodynamic and kinetic aspects of
disulfide stability, creating a series of unusual disulfide reduction
trigger units designed for stability to monothiols. We integrated the
motifs into modular series of fluorogenic probes that release and
activate an arbitrary chemical cargo upon reduction, and compared their performance to that of the literature-known disulfides. The
probes were comprehensively screened for biological stability and selectivity against a range of redox effector proteins and enzymes.
This design process delivered the first disulfide probes with excellent stability to monothiols yet high selectivity for the key redox-
active protein effector, thioredoxin. We anticipate that further applications of these novel disulfide triggers will deliver unique probes
targeting cellular thioredoxins. We also anticipate that further tuning following this design paradigm will enable redox probes for
other important dithiol-manifold redox proteins, that will be useful in revealing the hitherto hidden dynamics of endogenous cellular
redox systems.
redox networks are tightly regulated by reaction kinetics and
compartmentalization.10
1. INTRODUCTION
Controlled dithiol/disulfide-exchange reactions are essential to
cellular metabolism, protein folding, protein regulation, and
diverse aspects in cellular homeostasis and stress response.1,2
Two ubiquitous networks are particularly important: the
thioredoxin (Trx)−thioredoxin reductase (TrxR) system and
the glutaredoxin (Grx)−glutathione (GSH)−glutathione re-
ductase (GR) systems (Figure 1a).3,4 These systems are driven
by catalytically active enzymes, GR and isozymes of TrxR (low
nM cellular concentrations), which pass reducing equivalents
from NADPH into a wide range of dithiol/disulfide-type
reactions via effector proteins, especially several isozymes of
Trx and Grx (μM); in the case of GR, this transfer goes over
the redox-active peptide GSH (mM concentration).5,6 The
networks are interlinked and can function as backup systems
for each other.4,7 The Trx and Grx systems catalyze two-
electron dithiol/disulfide-exchange reactions proceeding via
polar ionic mechanisms with high substrate specificities and
precise biological control.8,9 Their scope of substrates is
controlled by chemocompatibility as well as protein−substrate
binding, and the relative turnover rates of each step in these
The biochemistry and isoforms of the Trx and GSH/Grx
networks have been excellently reviewed;11 relevant aspects are
summarized here. (i) TrxRs and GRs use similar FAD-
containing domains to harvest electrons from NADPH, passing
them to similar structurally restricted dithiol/disulfide active
sites (CVNVGC motif).12 Mammalian TrxR also has a C-
terminal selenolthiol/selenenylsulfide active site (CU motif)
on a flexible surface-exposed loop, to relay electrons from the
NADPH-driven dithiol site onto its substrates.13 These include
Trxs and related redox proteins, e.g., TRP14 (also known as
TXNDC17), though it can also reduce non-physiological small
molecules (Figure 1b).14,15 Due to its rare selenolthiol, TrxR
has unusual redox properties compared to dithiols: with both
Received: March 26, 2021
Published: June 1, 2021
https://doi.org/10.1021/jacs.1c03234
J. Am. Chem. Soc. 2021, 143, 8791−8803
© 2021 American Chemical Society
8791

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
readouts for the reduced-to-oxidized state ratio of the target
species. However, as ratiometric readouts of redox state, they
do not reveal turnover dynamics. Also, genetic approaches are
monitoring-only systems that cannot respond to redox
reactions with an action, such as delivering a therapeutic cargo.
To quantify turnover, selective turn-on probes are needed,
whereby enzymatic turnover events can be measured by
integrating cumulative and irreversible signal accumulation in
the form of an activated probe. In this context, small-molecule
probes that employ disulfide-based trigger−cargo designs
(Figure S1) are promising alternatives to protein sensors,
and are potentially chemocompatible with the redox enzymes.
Such probes are ideally created so that their cargo is
deactivated in the oxidized state, while reduction irreversibly
activates the cargo, resulting in a signal. Typically, activation
proceeds by a fragmentation reaction, exposing a key structural
element on the cargo, often simply by cleaving it from the
trigger. This is a conceptually simple design, and it is
additionally attractive because, unlike genetic probes, it can
act upon a stimulus: modular designs can enable not only
probes for detection and quantification of enzyme activity but
also prodrugs delivering drug cargos conditional upon this
activity.
In the thiol redox context, trigger−cargo designs have been
used to release a range of cargo types upon reduction,
including phenol- and aniline-type fluorophores and drugs
(Figure S1a,b). For such designs to be selective, a probe must
only commit to the first irreversible step on the pathway to turn-
on after it reacts with the targeted enzymeand not with any
undesired species (reducing or otherwise). Therefore, while
analyzing thiol/disulfide exchanges in light of equilibrium
thermodynamic parameters (e.g., reduction potential E°) is
generally informative, to design selective cumulative-release
probes for enzyme turnover requires considering the kinetics
and reversibility of each step in the reaction sequence. These,
in turn, depend on the chemistry of the redox trigger and its
reaction with target enzymes and reductants. For this analysis,
we must distinguish the reaction of a disulfide with a single
thiolate to liberate a thiolate, which we refer to as exchange,
from the net reduction of a disulfide to liberate two thiolates
by two stepwise exchanges, which we will refer to as reduction.
Linear disulfides are the most-used redox triggers in cargo-
releasing designs (Figure S1a,b).27 However, linear disulfides
cannot selectively detect or respond to enzyme activities in the
cellular context.28−30 This is because thiol−disulfide exchange
at either sulfur irreversibly and non-selectively commits linear
disulfides to turn-on (Figure 2a), and the rapidity of even
uncatalyzed exchange with the high intracellular monothiol
concentrations (ca. 5 mM GSH, ca. 50 mM protein thiols
(PRSH)) causes a high degree of monothiol-based cargo
release that likely drowns out enzyme-specific components.28
Therefore, at best, linear disulfides report non-specifically on
initial thiol−disulfide exchange rates.31
Figure 1. Key players in dithiol/disulfide homeostasis. (a) NADPH-
driven cellular signal transduction and redox repair through the GSH/
Grx and Trx systems. (b) Active-site structures of Trx18 and
TrxR:12,14,19,20 pre-organized cyclic dichalcogenides.
enhanced reaction kinetics and lowered reduction potential.16
GR is unusual in that it is highly specific for the small molecule
glutathione disulfide (GSSG), so offering little scope for other
substrates. (ii) Trxs and dithiol Grxs (includes Grx1 and Grx2)
are dithiol/disulfide redox-active proteins with surface-exposed
dithiol redox-active sites (CxxC motifs) that recognize and
reduce a variety of protein substrates in pro- and eukaryotic
organisms (Figure 1b). Trxs have lower reduction potentials
and are reduced by selenolthiol TrxR; Grxs have higher
reduction potentials and use monothiol GSH for their
reduction.17
In all these enzymes, the spatial pre-organization and
flexibility of the dichalcogenide redox-active sites are key to
their kinetics and thermodynamics (mirroring what is known
for small-molecule reductants).21 These factors, plus others
such as protein recognition, redox potentials, and non-
equilibrium metabolic flows, contribute to their substrate
scopes in cellsthough our understanding of the determinants
for cellular activities of these enzymes remains poor.22
The lack of techniques to measure or respond to the
turnover dynamics of these redox proteins in situ greatly limits
our understanding of redox biology, because it is turnover
dynamics that are key to understanding network function and
homeostasis.2 So far, the best-developed probes for use in
redox biology are genetically engineered ratiometric fluores-
cent sensors of redox-active protein or small-molecule redox
poise.23−26 These give specific, spatiotemporally resolved
Cyclic disulfide redox triggers of type A (Figure 2b) instead
allow kinetic reversibility in both exchange steps: their first
irreversible step is the actual cargo release itself (krel).
Reversion to the cyclic disulfide A after exchange to B or D
(kT1 or kG1 in Figure 2b) is a unimolecular reaction that may
be kinetically rapid (intramolecular cyclization that can also be
pre-organized) as well as thermodynamically favored by
enthalpic (pre-organized disulfide geometry) and entropic
effects (molecular cleavage). These factors can be amenable to
tuning by structural design of the disulfide trigger. However,
8792
https://doi.org/10.1021/jacs.1c03234
J. Am. Chem. Soc. 2021, 143, 8791−8803

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
TrxR and were applied in the fluorogenic “TRFS” probes33−35
and prodrugs.36 However, 1,2-dithiolanes undergo strain-
promoted, kinetically irreversible, facile, and non-specific
exchange with cell-surface monothiols,37,38 as known from
disulfide-mediated cellular uptake studies.39−41 Our work
confirmed that 1,2-dithiolane probes are, correspondingly,
non-specifically activated by monothiols.31
To resist monothiol reduction by minimizing the concen-
tration of C under monothiol challenge requires a high
reversion rate kG1 (and optionally kG2) compared to kG3. We
thus focused on unstrained cyclic disulfides, which are in fact
the most poorly explored as triggers of irreversibly activated
redox probes (Figure S1c,d, and further discussion in the
Supporting Information). The few notable studies concern the
6-membered cyclic disulfide 1,2-dithiane. As far as we know,
only three concepts of 1,2-dithiane designs intended to release
a general molecular cargo after reduction by biological species
have been disclosed. First, Butora used DTT phosphoesters to
release phosphate cargos after reduction, which was stated to
occur by reaction with GSH;30,42 this approach was since used
by, e.g., Urata.43 Second, Xu and Tang used an aniline
carbamate of DTT to release aniline cargos after reduction,
which was stated to occur by reaction with H2Se;44 Fang also
applied this approach, but contradicting Butora, stated that
neither TrxR nor GSH caused release.35 Third, Ziv disclosed
phenolic carbamate prodrugs of 4-amino-1,2-dithiane, in-
tended to release the phenol cargo after reduction, though
this was only described as occurring through the “ambient
reductive environment” of the cell.45,46 Relatedly, Kohn
worked on 6-membered cyclic disulfide derivatives of porfiro-
mycin and mitomycin, though not releasing a cargo;47,48 like
Butora, their reports cited non-specific dithiane exchange/
reduction by monothiols (Figure S1c,d).
We consider these reports both contradictory and limited in
their assessment of which cellular reductants are, or are not,
effective for exchange/reduction of 1,2-dithiane-based trigger−
cargo probes: so we set out to assess this comprehensively.
We also believed that tuning rates throughout the turn-on
process would prove critical for achieving protein selectivity.
Particularly, we wished to explore pre-organization by
annelating the cyclic disulfide triggers (dotted ring in Figure
2b). This modification adds minimal steric pressure at the
redox-active -SS- site, so it might not suppress enzyme docking,
but it could bring two favorable effects into play: (i) increasing
thiol cyclization rate of C (krel), to accelerate signal generation
without changing the reductant selectivity profile, and so
improve probe performance (N.B.: krel is equally increased by
both cis- and trans-annelation); (ii) stabilizing the cyclic
disulfide A, to increase kT1 and kG1 and so potentially change
the selectivity profile (N.B.: these effects should depend on the
cis-/trans-stereochemistry). Such annelated disulfides have not
yet been reported, so we wished to develop scalable and
convenient syntheses, enabling their general application to
redox research.
Figure 2. Mechanistic aspects of (a) irreversible monothiol reduction
of linear disulfide-type reduction-sensing units and (b) thermody-
namically reversible dithiol reduction of cyclic disulfide systems as
part of irreversible release probes.
for dithiol proteins, completing the reduction (B to C) is a pre-
organized, unimolecular reaction that reductases are evolved to
perform, and which could therefore be competitive to or faster
than kT1. By contrast, the reaction of D to C is both
bimolecular and not kinetically favored, so reversion by kG1 is
the most probable fate of D under monothiol challenge.
Therefore, A should be much more rapidly reduced to the
signal-generating intermediate C by dithiol proteins than by
monothiol background. Reversion of C to A is also well
known: it underlies the use of dithiothreitol (DTT) as a rapid,
stoichiometric, non-specific reductant particularly of linear
disulfides GSSG, PRSSG, and PRSSPR (kG2 in Figure 2b); this
pathway could also prove relevant in suppressing signal
generation by monothiol challenge. Therefore, we concluded
that stably pre-organized cyclic disulfides A would be suitable
starting points for triggers in enzyme-selective probes (Figure
2b), and we determined to explore their synthesis and use.
Cyclic disulfide trigger substrates for trigger−cargo probes
have been sparsely reported, and structures that are known
have rarely been assessed for redox selectivity and kinetic
performance. The most studied are strained cyclic 6-membered
disulfide epidithiodiketopiperazines (ETPs), but their strain-
promoted reactivity with monothiols is non-specific and
significant, so they are not suitable for selective probes.32
Moderately strained 5-membered cyclic disulfides (1,2-
dithiolanes) had been reported to be selectively reduced by
Lastly, we noted that, according to our proposed
mechanism, as long as hydrolytic cargo release is blocked,
the choice of cargo only influences krel and should not affect a
probe’s reductant selectivity. We were therefore curious to use
the same trigger with different cargos, to test the bioreductant
selectivities, and to rationally tune probe performance.
In this work we address the lack of information on cyclic
disulfide suitability for probes. We test our design principles by
stepwise exploration of the features predicted to determine
8793
https://doi.org/10.1021/jacs.1c03234
J. Am. Chem. Soc. 2021, 143, 8791−8803

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Figure 3. Trigger-and-cargo design for redox probes. (a) Stable cyclic disulfide-based reduction sensors and their control and comparison systems.
(b) Complete off→on fluorogenic cargos based on phenol mono-unmasking. (c) Representatives of the 20 fluorogenic probes used in this study,
derived by trigger−cargo combinations.
cargo release selectivity and rate. We scalably synthesize novel
cyclic disulfide-based triggers and apply them in robust,
irreversible-release probes suitable for generalized cargos. We
comprehensively evaluate the probes’ selectivity for key redox-
active proteins and their resistance to a high monothiol
background, also benchmarking previously reported but poorly
characterized disulfides. This will reveal the potential of pre-
organized disulfides as enzyme-selective reduction sensors and
identify a promising Trx-selective probe design.
conditions.35 Alternative anilide designs such as amides and
ureas are not cleavable and so do not represent modular
response systems as we sought.35,50
By comparison, phenol-releasing systems could benefit from
superior leaving-group nature to improve release kinetics.
Several phenolic systems featuring true off-to-on fluorogenicity
are known, and phenols are found as key activity-determining
groups in a broad range of diagnostic or therapeutic cargos,
meaning that modular designs in this work could be extended
into a useful series of probes and prodrugs. Stability against
non-reductive hydrolysis is the major challenge for most
phenols. For example, disulfide-trigger phenolic esters and
carbonates have been reported,51,52 but non-reductive phenol
release by spontaneous and enzymatic hydrolysis is certain to
prevent them achieving any selectivity in biological settings.
Phenolic carbamates have previously been reported to be
unstable.35 However, we interpreted this instability as the
spontaneous E1cB elimination pathway that affects primary
carbamates,53 so we designed all triggers for secondary
carbamate linkages to avoid this (Figure 3a),54,55 guided by
precepts from our earlier work.31,56
2.3. Cargo Choice and Use. We selected two major series
of fluorescent cargos to be masked by mono-O-carbamylation
(Figure 3b), requiring that this should entirely suppress their
fluorescence by mechanism-based quenching. (i) We used
xanthene-based mono-O-alkylated fluorescein MF-OH57 and
its acidified photostable congener IG-OH58 (as its iPr might
resist nucleophilic attack better than MF-OH’s methyl). Their
mono-O-carbamylation mechanistically quenches fluorescence
by locking the central carbon as sp3, so destroying conjugation.
Conjugation is restored by cleaving the carbamate substrate
(simplifying kinetics compared to typical double-substrate
probes such as fluorescein diacetate), giving soluble,
biocompatible, bright fluorophores (ex/em 485/515 nm).
(ii) We also used the precipitating fluorophore PQ-OH,59,60
whose high-quantum-yield, large-Stokes-shift fluorescence (ex/
em 360/520 nm) depends on excited-state intramolecular
proton transfer (ESIPT) of the phenolic hydrogen. Carbamy-
lating the phenol entirely and mechanistically abolishes ESIPT
fluorescence of the probe. The electron-withdrawn PQ should
also have fast release. To compare our phenolic designs to
2. RESULTS AND DISCUSSION
2.1. Trigger Design. We designed a broad panel of
disulfide-based redox triggers for this study (Figure 3a). To
avoid unspecific monothiol-triggered cargo release, we used
stable cyclic 6-membered disulfide triggers: most simply,
disulfide trigger SS60. To study how pre-organizing annelation
affects both cargo release and selectivity profiles, we designed
cis- and trans-fused triggers SS66C and SS66T. As linear
disulfide controls, we used simple trigger SS0027 and annelated
SS06 (which, when compared, should test for pre-organization
only of cargo release krel). Where needed, we compare the
results with those of the strained and/or polymerizable
monocyclic disulfide controls SS5031,33 and SS7.49 This
broad panel of disulfide structure motifs would allow us to
compare reductant selectivity and cargo release rates, across
stable or pre-organized or strained, cyclic or linear disulfide
redox probes, on a level that to our knowledge has not yet been
attempted. Lastly, we introduced O56 as a similarly polar but
non-reducible negative control trigger to check stringently for
stability against non-reductive chemical or enzymatic release of
probe cargos.
2.2. Probe Design Considerations. To create modular
probe designs, we chose to work primarily with phenolic
fluorophores. Anilines are simpler to adapt into fluorogenic
probes, as acylation usually substantially suppresses fluores-
cence (though only in special cases does it do so completely),
and since acylated anilines have good hydrolytic stability.35
Aniline carbamates are well represented in trigger−probe
designs. However, the poor leaving-group nature of anilides
results in slow cargo release for 5-exo-trig thiol cyclization, with
physiological half-lives of >3 h, even under fully reduced
8794
https://doi.org/10.1021/jacs.1c03234
J. Am. Chem. Soc. 2021, 143, 8791−8803

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
prior art in aniline disulfide redox probes, we also prepared 3-
O-methyl-rhodol MR-NH2 for kinetic comparisons.61 Thus, all
our probes feature a true off-to-on mechanistic switch of signal
upon disulfide trigger cleavage, thereby maximizing their
signal-to-noise ratio and simplifying analysis (details in Figure
S2).
We combined triggers freely with cargos to create a panel of
probes for analysis. We name the resulting probes by the
combination used; e.g., in Figure 3c, SS66C-MF is an SS66C-
trigger MF-OH-releasing probe, etc. (further details and
discussion in the Supporting Information).
2.4. Diastereomerically Pure Synthesis of Annelated
SS66 Disulfides. The simpler disulfide redox triggers were
accessed straightforwardly. Monocyclic disulfide SS60 was
obtained using the approach reported by Raines62 to give the
protected primary amine, then by N-methylation and
deprotection to prepare the novel secondary amine SS60.
Known polymerization-prone38,63 unstable cyclic disulfides
SS50 and SS7 were accessed from aminopropanediol31 and
from a nitrogen mustard precursor,49 respectively, as described.
Symmetric linear disulfides SS00 and SS06 were accessed by
oxidative dimerization of the respective thiols; as far as we
know, cyclization-pre-organized linear disulfides like SS06 have
not been studied before (Figure 3a). The secondary amine of
these symmetric disulfides proved inconvenient for yields and
purification, with double-reaction side products being formed
during later carbamate coupling. Unsymmetric linear disulfides
of the SS00 type are rarely reported in the literature, and we
found no convenient methods to synthesize them. We
developed a versatile batch synthesis strategy for unsymmetric
linear disulfides by coupling easily prepared, isolable precursors
(one primary S-acetyl and one primary S-tosyl), achieving high
chemoselectivity and isolating the unsymmetric morpholino-
substituted SS00M (Figure 3a) in good yields. For details, see
the Supporting Information.
The bicyclic redox triggers were more challenging (Figure
4). We first focused on the cis-fused SS66C system, setting the
stereochemistry by heterogeneous hydrogenation of pyridine 1.
Boc-protection to amine 2 and then ester reduction gave cis-
diol 3, which we could not, however, convert into a bicyclic
disulfide. During Mitsunobu reaction with thioacetic acid
under Raines’s conditions,62 no bis(thioacetate) but rather
undesired cis-ether 5 was isolated, with X-ray confirming its
structure. Fortunately, 5 provided ideal non-reducible hydrol-
ysis controls, as the O56 series. Other strategies to activate the
hydroxy groups for S-nucleophile substitution were unsuccess-
ful; e.g., attempted bis(sulfonylation)s gave intramolecularly
cyclized carbamates, e.g., 4, as known for Boc-/Cbz-methanol-
piperidines.64−66 To avoid carbamate cyclizations, we switched
to PMB as the N-protecting group. Diastereomerically pure N-
PMB-diol 10 was accessed in decent yield by hydrogenation of
1, ester reduction to 9, and N-alkylation. It was crucial to
reduce the esters before nitrogen protection, to avoid partial
cis−trans-isomerization at the diester stage. Mesylation of diol
10 and nucleophilic substitution with KSAc gave bis-
(thioacetate) 11, and tandem basic acetate cleavage/oxidation
gave novel annelated disulfide 12. PMB was deprotected
without reductive or oxidative conditions by using 2-
chloroethyl chloroformate (ACE-Cl),67 giving the precursor
to all SS66C probes (Figure 4).
Figure 4. Diastereomerically pure syntheses of SS66T- and SS66C-
series bicyclic disulfide redox probes. Reagents and conditions: (i) (1)
H2, H2SO4, Pd/C, MeOH, 80 bar, r.t., 15 h; (2) Boc2O, NEt3,
dioxane/H2O, r.t., 15 h (98%). (ii) LiAlH4, Et2O, 0 °C to r.t., 3 h.
(88%). (iii) MsCl, DCM, NEt3, r.t., 15 h (89%). (iv) Cis-selective
tetrahydrofuran formation: (1) PPh3, DIAD, HSAc, THF, r.t., 15 h
(90%); (2) HCl, MeOH, 80 °C, 4 h (quant.). (v) (1) H2, H2SO4, Pd/
C, MeOH, 80 bar, r.t., 15 h; (2) LiAlH4, THF, 0 °C to r.t., 3 h. (vi)
PMB-Cl, K2CO3, EtOH, r.t., 15 h (v−vi: 42% over 3 steps). (vii) (1)
MsCl, NEt3, DCM, r.t., 1 h; (2) KSAc, DMF, r.t., 15 h. (viii) KOH,
MeOH, r.t., 15 h (vii−viii: 51% over 3 steps). (ix) (1) ACE-Cl, DCE,
80 °C, 5 h; (2) MeOH, 80 °C, 2 h; (3) carbamate coupling. (x) (1)
NaOMe, MeOH, r.t., 1 h; then AcOH (2:1 trans:cis); (2) LiAlH4,
Et2O, 0 °C to r.t., 2 h. (72% over 2 steps). (xi) Trans-selective
bis(thioacetylation): HSAc, PPh3, DIAD, THF, r.t., 15 h. (xii) (1)
KOH, MeOH, r.t., 15 h; (2) HCl, r.t., 4 h (xi−xii: 45% over 3 steps).
(xiii) Carbamate coupling.
reaching an inseparable 2:1 trans:cis epimerized mixture by
NaOMe treatment and then quenching by AcOH. Ester
reduction gave 2:1 cis:trans-diol 6, still unseparated. Then, we
could selectively destroy the residual cis while carrying the
trans through its next step of thio-Mitsunobu reaction: trans-
diol 6 converted smoothly to the trans-bis(thioacetate) 7,
while cis-diol 6 converted to cis-ether 5, separable by
chromatography. Thioester hydrolysis of 7, dithiol oxidation,
and deprotection gave diastereomerically pure trans-8, the key
SS66T precursor, in decent yields (Figure 4). (This also
suggested that the cis-fused system’s greater tendency to 5-exo-
tet cyclization than the trans might be reflected in a greater
kinetic and/or thermodynamic stability of its cis-disulfide,
which we soon examined, as described below.)
For the trans-fused SS66T, we took a route exploiting the
previous setbacks to our advantage. First, we could use basic
cis−trans-epimerization to recycle cis-diester 2, after some trials
Deprotected secondary amines of the disulfide triggers were
reacted with O-chloroformates of phenolic fluorophores to
deliver 19 fluorogenic MF, IG, and PQ series probes (Figure
8795
https://doi.org/10.1021/jacs.1c03234
J. Am. Chem. Soc. 2021, 143, 8791−8803

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Figure 5. Redox thermodynamics of cyclic 6-membered disulfide triggers. (a) Single-crystal X-ray structures of 1,2-dithianes: CSS bond angles θ;
CSSC dihedral angles φ; disulfide bond distance Δ. (b) HPLC-based E° measurement protocol. Disulfides are equilibrated against DTTred, and
then dithiols are quantitatively re-oxidized by iodine as a control. (c) HPLC time courses, equilibrium ratios, and resulting disulfide E° values. (d)
Graphical depiction of E° for this study’s 1,2-dithianes against known (bio)reductants.11
4) and one MR reference compound.68 These were all
colorless and non-fluorescent both as solids and in solution, as
expected for their mechanistically quenched emission. To
assess disulfide thermodynamics, we also benzoylated the
amines to give “SS-Bz” compounds that mimic the geometry of
the triggers in the carbamate probes but can be reversibly
equilibrated as well as crystallized (see below).
2.5. Thermodynamic Analyses: Redox Potentials Are
Lowest for Annelated Cyclic Disulfides (SS66). Redox
potentials depend on structural/geometric features that
(de)stabilize the disulfide relative to the dithiol. Bond distance
(SS), bond angle (CSS/SSC), and dihedral angle (CSSC),40
together with strain factors, typically affect disulfide stability.
We could acquire X-ray structures of all three SS-Bz 1,2-
dithianes (Figure 5a), showing their disulfide rings in chairlike
conformations with minimal eclipsing and CSSC dihedral
angles close to that of the model 1,2-dithiane DTTox (56°).69
But, while a double-chair conformation was found for SS66C-
Bz (similar to a cis-decalin), against our expectations SS66T-
Bz formed a more strained boat-chair system unlike a trans-
decalin (for further discussion, see Raines70 and references
therein, and Supporting Information).
We then experimentally examined the thermodynamic
reduction potential E° of the cyclic disulfides. We had
anticipated that thermodynamic stability (low E°) would be
important to resist reduction by monothiols and so to achieve
selectivity for dithiol species. Whitesides calculated and
experimentally determined E° values of uncommon disulfides
in the 1980s−90s.63,71 The lowest E° values they reported were
indeed for 2,3-dithiadecalin-type disulfides,72 which under
non-standard conditions (50:50 buffer:cosolvent) were −0.35
V (trans) and −0.34 V (cis). This extraordinary stability
compared to linear disulfides (E° ca. −0.22 V)63 was promising
for our aims in this study. However, while we expected the E°
of the tetrahedral amine triggers SS66T and SS66C to match
these, we believed that the restrained geometries of the
carbamates would modify the probes’ potentials. To mimic
these while avoiding the influence of kinetic considerations
that also determine irreversible probe performance (trigger
reduction and thiol cyclization rates), we again used the
reversible SS-Bz series; these also allowed us to benchmark our
reduction potentials under normalized conditions.
To determine reduction potentials, we equilibrated the SS-
Bz series against chemocompatible reductant standards,
analyzing oxidized:reduced compound ratios by HPLC. We
did this similarly to methods used by Raines62 and others63,73
though with three additional features. First, the benzamide is a
convenient UV-active tag for HPLC signal integration. Second,
we detected oxidized:reduced ratios for both benzamide probe
and reductant standard (e.g., DTT); this controls for parasitic
oxidation by adventitious oxygen (which would be reflected by
an increasing total of oxidized species). Third, we performed
time course measurements to follow the progress of
equilibration, and upon reaching stable plateaus we controlled
for reversibility by re-oxidizing all dithiols to disulfides using I2
(Figure 5b). This controls for polymerization of strained
disulfides, which risks being misinterpreted. By titrating
reductants, we determined reduction potentials of −0.276 V
for SS60-Bz, −0.317 V for SS66T-Bz, and −0.339 V for
SS66C-Bz (Figure 5c,d), whereas the strained analogues SS50-
Bz and SS7-Bz were not equilibrating systems. We estimate
0.01 V precision in our experiment. The lower reduction
8796
https://doi.org/10.1021/jacs.1c03234
J. Am. Chem. Soc. 2021, 143, 8791−8803

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Figure 6. Reductant-resistance profiling and cargo-release kinetics. (a) Fluorescence response of representative linear and 1,2-dithiane probes to
monothiol reductants (1 mM), benchmarked to quantitative reductant TCEP (100 μM) and to dithiol DTT (1 mM); t = 2 h. (b) Probe activation
kinetics with TCEP (100 μM). (c) GSH challenge titrations (0.01−10 mM GSH, 10 μM probe, t = 6 h data). (d) Dithiol challenge titrations
(0.01−10 mM DTT, 10 μM probe, t = 6 h data). (e, f) Modular probe design allows comparing cargo release kinetics.
potentials of the annelated SS66 compared to the monocyclic
SS60 matched expectations. Potentially due to the benzamide
strain, SS66T-Bz indeed had 0.02 V higher E° than SS66C-Bz.
While Whitesides gave a 0.01 V lower potential for the trans-
2,3-dithiadecalin,63 this may only hold true for the SS66
amines (Figure 5d; see also Supporting Information).
2.6. Kinetic Analyses: Cyclic 6-Membered Disulfides
Resist GSH Reduction but Are Triggered by Vicinal
Dithiols. We now turned to the PQ- and MF/IG-based
fluorogenic probes, to test their activation by, or resistance to,
small-molecule reductants.
Calibrations on the cargos MF-OH, IG-OH, and PQ-OH
confirmed large linear concentration/signal ranges
(20 nM−30 μM for xanthenes, >1 μM for PQ-OH; Figure
S2a,b). Assays of the probes’ fluorogenic character highlighted
their true off-to-on fluorescence switching, based on complete
mechanistic quenching in the intact probes (Figure S2c).
Throughout the reductant challenge assays, we controlled
against fluorophore release by non-reductive degradation
mechanisms (such as carbamate hydrolysis, or carbamate
thiolysis by thiol reductants) using the O56 probe series.
Matching our design requirements, the O56 series showed in
all cases that the secondary amine carbamates were entirely
robust, thereby solving the problems with prior phenolic
carbamates35 and indicating that disulfide reduction and
intramolecular thiol cyclization would be the only pathway
for signal generation with our disulfide probe series.
First, to report selectively on enzyme activity, a probe must
resist reduction by the cellular monothiol background (ca.
50 mM, of which up to 5 mM GSH).11 We therefore began by
incubating probes with monothiols (1 mM) to profile their
monothiol resistance. Linear SS00 and SS06 probes and
strained 5-membered cyclic SS50 probes were rapidly triggered
by all monothiols: GSH, cysteine, cysteamine (CA), N,N-
dimethyl cysteamine (MEDA), and N-acetylcysteine (NAC).
In contrast, 1,2-dithiane SS60, SS66C, and SS66T as well as 7-
8797
https://doi.org/10.1021/jacs.1c03234
J. Am. Chem. Soc. 2021, 143, 8791−8803

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
membered cyclic SS7 probes resisted monothiol reduction
(Figure 6a, Figures S4 and S5 for PQ series, Figure S6 for MF
series, and Figure S7 for IG and MR series).
mechanistic difference would allow monothiol-stable cyclic
disulfide probes to be selectively activated by vicinal dithiols.
Dose−response titrations showed that the GSH-resistant
SS60, SS66T, and SS66C cyclic disulfide probes were indeed
fully activated by DTT (Figure 6d). The linear and SS50
probes were quantitatively triggered by even equimolar DTT,
and the control O56 probes were again fully resistant (Figures
S4−S7). Pleasingly, the ordering of DTT resistance (SS00/
SS06 < SS60 < SS66T/SS66C) was common to all cargos,
again supporting the modularity of the probes’ design.
As for the monothiols, we noted apparent differences of
probe sensitivity to DTT depending on the cargo nature, but
again we attribute this to leaving-group kinetics (PQ-OH ca.
10-fold faster-releasing than MF-OH/IG-OH; we had
expected that the acidification of IG-OH would make it a
faster cargo than MF-OH77 but we did not see this; Figure 6e,f,
Figures S4−S7). The design control probe SS60-MR had poor
activation (ca. 5% at 10 mM DTT at 6 h, Figure 6f),
highlighting a general benefit of our phenol-releasing design
over standard aniline-releasing probes (further detailed
evaluation in the Supporting Information).
Taken together, this systematic comparison strongly showed
that only 6-membered disulfides resist uncatalyzed reduction
by monothiols at physiological concentrations, but that such
motifs can still be reduced by vicinal dithiols. To the best of
our knowledge, dose−response assays evaluating, e.g., GSH
and DTT stability across wide concentration ranges to test for
trigger reductant preferences have not been used in this field
before, nor have previous reports established trigger or cargo
selectivities on the basis of independently varying both triggers
and cargos within the same probe series. In our opinion, both
would be useful additions to the toolbox of standard assays for
cumulative-release turnover probes.
2.7. Cyclic 6-Membered Disulfides Are Selectively
Reduced by Trx Rather than by Other Oxidoreductases.
Having established the selectivity of the 6-membered cyclic
disulfide probes for reduction by vicinal dithiols and their
resistance to monothiols, we next proceeded to profile the
probes’ reduction by biological vicinal dithiol/disulfide-type
proteins, focusing on the main components of the TrxR/Trx
and GR/GSH/Grx systems. In cells, the effector proteins Trx
and Grx (ca. 10 μM each) have orders of magnitude higher
cellular concentrations than their upstream reductases TrxR
and GR (ca. 20 nM each). To design cell-free assays to predict
cellular enzyme selectivities, we ensured that the assays were
performed (a) with cellular reductant concentrations, (b) using
the catalytically powered redox systems rather than only pre-
reduced effector proteins, and (c) examining a range of protein
isoforms, both isolated from primary tissues as well as
recombinantly expressed. The last point is particularly relevant
for TrxR since the key selenocysteine (Sec, U) residue in its
active site is highly present in isolates of native enzymes78 or
recombinant forms made with novel production method-
ologies (up to 100% Sec contents),79 which we employed in
this study, rather than the ca. 30% Sec content from standard
expression methods.80 Accordingly, our tests used recombinant
human TrxRs, both cytosolic TrxR1 and mitochondrial TrxR2;
recombinant human Trxs, both cytosolic Trx1 and mitochon-
drial Trx2; and as a comparison, recombinant human
thioredoxin-related protein TRP14,81 which also features a
vicinal dithiol/disulfide active site that is reduced by TrxR. Grx
variants were examined using recombinant human vicinal
Next, we examined the kinetic performance of each probe,
using the rapid and quantitative disulfide reducing agent tris(2-
carboxyethyl)phosphine (TCEP). In these settings where no
dithiol re-oxidation is possible (cf. kT2 and kG2, Figure 2),
probe signal generation depends on the rates of reduction by
the phosphine, 5-exo-trig thiol cyclization, and cargo expulsion.
Under such special conditions, the cargo nature rather than the
disulfide structure became the major determinant of signal
kinetics (Figure 6b; PQ-OH release approximately 10 times
faster than that of MF-OH). While some reports claim that
distal cargos substantially affect the reductant selectivity
profile,74 we attribute the faster release of PQ-OH than MF-
OH simply to its better leaving-group character.
Pleasingly, the relative rates of probes within each series were
almost identical for the MF as for the PQ series (Figure 6b; fits
in Figure S3), revealing trigger-based performance features. (1)
While annelations have been used for amine cyclization
promotion,56,75 this has not yet been studied for thiols. The ca.
3-fold rate enhancement of annelation in the linear disulfides
was significant (SS06 > SS00, also suggesting that their thiol
cyclization is more rate-determining than their phosphine
reduction). Annelation in the 1,2-dithianes was interesting:
while the cis-fused SS66C was faster than SS60 as expected,
the trans-fused SS66T was slower than SS60. (2) Thiol
cyclization/cargo elimination (C → PhOH, Figure 2) occurs
on a scale of minutes. As intramolecular thiol−disulfide
exchanges (B → A, B → C, and D → A) typically take <10
ms,76 we propose that “on-reductant” cyclization of the
exchange intermediates B and D to release PhOH before
their full reduction to C (Figure 2) is not a major contributor
to cargo release. (3) The rates of bisthiolate cyclization (from
SS50) were twice those of the correponding monothiolate
(SS00), a satisfying match to theory, though the SS7 probes
were sluggish and behaved irreproducibly. (We believe
oligomerization of SS737,63,71 causes its poor performance,
paralleling the behavior of polymerizable31 SS50 probes.
Additionally, neither SS7-Bz nor SS50-Bz performed reliably
in the equilibration/reoxidation assay. So, while we carried
these probes through all assays, we focus on interpreting only
the reliable 1,2-dithiane and linear disulfide results; see
Supporting Information.)
We next aimed to profile GSH sensitivity more meaningfully
than by single-concentration assays, to reach results
predictively applicable to diverse settings. Therefore, we
titrated the probes with GSH over a wide concentration
range (0.01−10 mM GSH, 10 μM probe) to build logarithmic
dose−response curves profiling GSH sensitivity with long-term
challenge (fluorescence at t = 6 h; Figure 6c, Figures S4−S7).
These showed that linear disulfide (SS00/SS06) and 5-
membered cyclic SS50-type probes are indeed quantitatively
activated by physiological GSH concentrations (Figure 6c).
The novel 1,2-dithiane probes, however, gave either zero
response to GSH up to 10 mM (SS66C, SS60) or else very
low signal from 1 to 3 mM GSH (SS66T; Figure 6c). As the
1,2-dithiane probes are monothiol-resistant, this indicated they
might have potential for protein-selective monitoring.
We then explored a simple model system for vicinal dithiol
reactivity using DTT. Vicinal dithiols reduce cyclic disulfide
triggers via a net bimolecular pathway, in contrast to
monothiols (net trimolecular; Figure 2). We hoped that this
8798
https://doi.org/10.1021/jacs.1c03234
J. Am. Chem. Soc. 2021, 143, 8791−8803

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Figure 7. Cyclic 6-membered disulfides are Trx-selective substrates. (a) Bioreductant selectivity (single-concentration profiling; t = 4 h and t = 6 h;
see also Figures S8 and S9). (b) Response kinetics of SS60-PQ (10 μM) and SS60-MF (1 μM) compared to their linear disulfide analogues. (c)
Kinetic response of SS60-MF (1 μM) to titrated redox effectors Trx1, Trx2, Grx1, and Grx2. (d) Dose−response profiles of the 1,2-dithiane probes
to redox effector proteins (t = 6 h). See Supporting Information for details (Figures S8−S13).
dithiol glutaredoxins Grx1 and Grx2 together with human
recombinant GR.
approach has not been reported with other compounds
described as redox turnover probes, but it proved to be a
powerful technique to illustrate probe selectivities. All assays
were performed with NADPH; GR assays optionally included
a low concentration of GSH to allow Grx reduction without
significantly reducing probe directly; see Supporting Informa-
tion for further details.
The 6-membered cyclic disulfide probes based on SS60,
SS66C, and SS66T were selectively activated by thioredoxins
Trx1 and Trx2 (Figure 7a), with the SS60 series the most
sensitive. The SS66C series had the highest Trx selectivity,
We evaluated all probes in fluorescence time courses. To
conclude on whether probes were reduced by the effectors Trx
and Grx, and/or by direct reaction with the upstream
reductants TrxR and GR, we compared assays using both
effectors and upstream reductants against assays with only
upstream reductants, controlling against the effectors only. We
also confirmed these interpretations by titrating effectors (e.g.,
Trx1) in the presence of a fixed concentration of upstream
reductants (e.g., TrxR1) or vice versa. To our knowledge, this
8799
https://doi.org/10.1021/jacs.1c03234
J. Am. Chem. Soc. 2021, 143, 8791−8803

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
with almost zero detected signal under challenge by the Grx
+GSH+GR and TrxR+TRP14 cascades over 4 or 6 h,
respectively (Figure 7b,c, Figures S8−S13). Both Trx1 and
Trx2 dose-dependently activated the dithiane probes, though
pleasingly, under relevant concentrations, these probes were
non-responsive to any of the TrxR isoforms in the absence of
the effector Trx. It is notable that the 1,2-dithiane-type probes
were not affected by the Trx-analogue vicinal dithiol protein
TRP14, which likely reflects the different active-site structure
and/or electrostatic surface charge of TRP14 compared to
either Trx1 or Trx2.82 We note that the general order of Trx
selectivity in the dithiane series (SS66C ≳ SS60 > SS66T)
matches the order of stability against the supraphysiological
monothiol levels (e.g., GSH at 10 mM, Figure 6c), supporting
the earlier interpretation of structural influences on the relative
kinetic instability of the trans-fused system. Matching expect-
ations, linear disulfide reference probes SS00(M) were
activated completely non-specifically by Trx1, Trx2, TRP14,
Grx1, Grx2, and TrxR1 (Figure 7a,b, Figures S8, S10, and S11)
in addition to the monothiols previously shown. This
highlights the importance of the dithiane for achieving Trx
selectivity (Figure 2). The signal kinetics for all PQ probes
were again faster than for the corresponding MF probes
(Figures S8−S13).
We again performed dose−reponse titrations to test the
sensitivity profiles of all probes to varying concentrations of the
redox-active proteins, which should be a useful predictive tool
to estimate performance in complex applications. Figure 7d
shows these dose−response plots for the two major dithiane
triggers, the more-Trx-sensitive SS60 and the most-Trx-
selective SS66C, as applied to both PQ and MF probes
(other probes in Figures S8−S13). This highlights the highly
preferred processing of the novel dithiane probes by the Trx
isoforms, regardless of their cargo (Figure 7d)an excellent
result for the mechanism-based design.
In summary, this systematic exploration of disulfide probe
reductive triggering has relied on time course rather than
endpoint data, and featured systematic variations and titrations
of all reducing species. The results indicate that regardless of the
cargos with which they are used, the hitherto-reported linear and
cyclic 5-membered disulfide triggers do not withstand non-
selective and non-enzymatic triggering by GSH, monothiols,
and a range of cellular reductants, and are therefore non-
selective reduction sensors. In contrast, the new 6-membered
cyclic disulfide probes display excellent selectivity and steep
dose−response curves for reduction by vicinal dithiols,
particularly by Trx1 and Trx2. SS60 and SS66C were thus
identified as promising disulfide motifs for sensitive, modular
probes selectively reduced by thioredoxins.
both probes and prodrugs) with (ii) a disulfide/dithiol-type
redox trigger that resists the high cellular background of
monothiols yet can be triggered by specific redox-active
dithiol-type enzymes.
In this work we used mechanism-based analysis to design
novel disulfide-based trigger−cargo probes which unite both
features. The 6-membered cyclic disulfide probes characterized
here feature rational tuning of thermodynamic and kinetic
parameters of disulfide reduction as well as of cargo release, to
achieve higher stability to the monothiol background than any
probes previously reported. The trigger motifs were synthe-
sized scalably, with the novel bicyclic disulfide structures being
accessed by divergent diastereomerically pure routes, and then
integrated into modular phenol-releasing probes which feature
stability to cleavage by non-reductive mechanisms. These
probes have been evaluated through a rigorous methodology
combining time course and dose−response screening of
chemical and biological reductants. The disulfides we report,
and the probe designs we apply them to, fill a gap in the
literature by being the first systematic exploration of the
synthesis and reductant selectivity of non-strained and non-
polymerizing disulfides, with the outlook toward adapting
these modular phenol-releasing designs for a range of purposes.
Pleasingly, the design logic of achieving selectivity by matching
probe energetics and mechanism to those of the desired
reducing species was supported by the selectivity in reduction
of the 6-membered disulfides by the vicinal dithiol Trx, but not
by the other redox-active enzymes analyzed here.
Further developing the Trx-selective 1,2-dithiane chemo-
types for cellular monitoring of endogenous Trx is a major aim
of our ongoing research. This faces challenges including (1)
ensuring reduction is competitive with native Trx substrates, to
reach high turnover or signal in the cellular context, and (2)
favoring the post-reduction cyclization krel relative to the re-
oxidation pathways (particularly kG2; Figure 2), to maximize
the proportion of Trx-reduced probe which releases its cargo.
These are non-trivial challenges. We also expect that tuning the
probes’ intracellular localization will prove crucial for perform-
ance.
Diversifying disulfide trigger structures to address other key
oxidoreductases is another goal of our research. The scope of
redox biology would be greatly expanded by developing probes
that are GSSG-competitive substrates of GR, or Trx-
competitive substrates of TrxR. This would allow researchers
to tap into the turnover dynamics that drive the major dithiol/
disulfide-manifold processes throughout the cell. This work has
demonstrated that disulfides are a still-untapped well of
potential for redox probes. We therefore hope that further
developments of probes with increased selectivity for specific
redox enzymes will facilitate progress in the many areas and
applications of research in redox biology.
3. CONCLUSIONS
Specific dithiol−disulfide-type reactions underlie a multitude
of biological pathways, and engineered disulfides are recently
finding applications from probes in chemical biology to uses in
biophysics and materials chemistry. Linear disulfides have been
used for decades as non-specifically and irreversibly intra-
cellularly cleaved substrates, often for intracellular release of
appended cargos. While strained cyclic disulfides have found
several applications, it has remained unproven whether any
disulfide could be used reliably for creating robust, modular,
enzyme-selective small-molecule probes.31 The challenge
posed has been to unite (i) a broadly applicable, robust design
that is suitable for efficiently releasing arbitrary cargos (toward
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/jacs.1c03234.
■
sı
Synthesis, analysis, and biochemistry, including Figures
S1−S13 (PDF)
Accession Codes
CCDC 2072438−2072442 contain the supplementary crys-
tallographic data for this paper. These data can be obtained
free of charge via www.ccdc.cam.ac.uk/data_request/cif, or by
8800
https://doi.org/10.1021/jacs.1c03234
J. Am. Chem. Soc. 2021, 143, 8791−8803

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
emailing data_request@ccdc.cam.ac.uk, or by contacting The
Cambridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK; fax: +44 1223 336033.
scholarship; L.Z. thanks the Fonds der Chemischen Industrie
(FCI) for support through a Ph.D. scholarship; L.P. thanks the
GRK 2338 for support through a Ph.D. scholarship; J.T.-S.
thanks the Joachim Herz Foundation for fellowship support.
AUTHOR INFORMATION
Corresponding Author
Notes
■
The authors declare the following competing financial
interest(s): J.G.F., L.Z., J.T.-S., and O.T.-S. are inventors on
a patent application filed by the LMU Munich in 2021
covering compound structures reported in this paper.
Oliver Thorn-Seshold − Department of Pharmacy, Ludwig
Maximilians University Munich, 81377 Munich, Germany;
orcid.org/0000-0003-3981-651X; Email: oliver.thorn-
seshold@cup.lmu.de
ACKNOWLEDGMENTS
■
Authors
We thank Peter Mayer (LMU) for X-ray crystallography; Qing
Cheng (KI) for production and purification of recombinant
enzymes; Paula Ruppel, Christina Wichmann, and Julia Rauh
(LMU) for synthetic assistance; Reviewer 2 for the helpful
comments regarding the sources of artifactual signal; and Matt
Fuchter (ICL), Klaus T. Wanner (LMU), Kate Carroll
(Scripps), and the attendees of the SPP 1710 conference
Thiol-Based Redox Switches in 2019, for their supportive and
collegial discussions.
Jan G. Felber − Department of Pharmacy, Ludwig
Maximilians University Munich, 81377 Munich, Germany;
orcid.org/0000-0002-5010-9624
Lukas Zeisel − Department of Pharmacy, Ludwig Maximilians
University Munich, 81377 Munich, Germany; orcid.org/
0000-0001-7813-7099
Lena Poczka − Department of Pharmacy, Ludwig
Maximilians University Munich, 81377 Munich, Germany
Karoline Scholzen − Department of Medical Biochemistry,
Karolinska Institutet, 17177 Stockholm, Sweden
Sander Busker − Department of Medical Biochemistry,
Karolinska Institutet, 17177 Stockholm, Sweden;
orcid.org/0000-0002-7069-3864
Martin S. Maier − Department of Pharmacy, Ludwig
Maximilians University Munich, 81377 Munich, Germany;
orcid.org/0000-0002-0409-0539
Ulrike Theisen − Institute of Pharmacology and Toxicology,
Medical Center, University of Rostock, 18057 Rostock,
Germany
Christina Brandstädter − Interdisciplinary Research Centre
(IFZ), Justus Liebig University Giessen, 35392 Giessen,
Germany
Katja Becker − Interdisciplinary Research Centre (IFZ), Justus
Liebig University Giessen, 35392 Giessen, Germany;
orcid.org/0000-0003-4673-3675
Elias S. J. Arnér − Department of Medical Biochemistry,
Karolinska Institutet, 17177 Stockholm, Sweden; Department
of Selenoprotein Research, National Institute of Oncology,
1122 Budapest, Hungary; orcid.org/0000-0002-4807-
6114
DEDICATION
■
This paper is dedicated to George Whitesides, who so greatly
contributed to setting thiol/disulfide redox chemistry on
secure theoretical and practical foundations.
REFERENCES
■
(1) Herrmann, J. M.; Dick, T. P. Redox Biology on the Rise. Biol.
Chem. 2012, 393 (9), 999−1004.
(2) Jones, D. P.; Sies, H. The Redox Code. Antioxid. Redox Signaling
2015, 23 (9), 734−746.
(3) Arnér, E. S. J.; Holmgren, A. Physiological Functions of
Thioredoxin and Thioredoxin Reductase. Eur. J. Biochem. 2000, 267
(20), 6102−6109.
(4) Miller, C. G.; Holmgren, A.; Arnér, E. S. J.; Schmidt, E. E.
NADPH-Dependent and -Independent Disulfide Reductase Systems.
Free Radical Biol. Med. 2018, 127, 248−261.
(5) Pillay, C. S.; Hofmeyr, J.-H. S.; Rohwer, J. M. The Logic of
Kinetic Regulation in the Thioredoxin System. BMC Syst. Biol. 2011,
5 (1), 15.
(6) Deponte, M. Glutathione Catalysis and the Reaction
Mechanisms of Glutathione-Dependent Enzymes. Biochim. Biophys.
Acta, Gen. Subj. 2013, 1830 (5), 3217−3266.
(7) Lu, J.; Holmgren, A. The Thioredoxin Antioxidant System. Free
Radical Biol. Med. 2014, 66, 75−87.
(8) Fernandes, A. P.; Holmgren, A. Glutaredoxins: Glutathione-
Dependent Redox Enzymes with Functions Far Beyond a Simple
Thioredoxin Backup System. Antioxid. Redox Signaling 2004, 6 (1),
63−74.
Julia Thorn-Seshold − Department of Pharmacy, Ludwig
Maximilians University Munich, 81377 Munich, Germany;
orcid.org/0000-0002-4879-4159
Complete contact information is available at:
https://pubs.acs.org/10.1021/jacs.1c03234
(9) Holmgren, A.; Johansson, C.; Berndt, C.; Lönn, M. E.;
Hudemann, C.; Lillig, C. H. Thiol Redox Control via Thioredoxin
and Glutaredoxin Systems. Biochem. Soc. Trans. 2005, 33 (6), 1375.
(10) Hanschmann, E.-M.; Godoy, J. R.; Berndt, C.; Hudemann, C.;
Lillig, C. H. Thioredoxins, Glutaredoxins, and Peroxiredoxins
Molecular Mechanisms and Health Significance: From Cofactors to
Antioxidants to Redox Signaling. Antioxid. Redox Signaling 2013, 19
(13), 1539−1605.
(11) Fridovich, I.; Poole, L. B.; Holmgren, A.; Lou, M. F.; David, S.
S.; Osborne, R. L.; Dawson, J. H.; Copley, S. D.; Kadokura, H.;
Beckwith, J.; Gilbert, H. F.; Ragsdale, S. W.; Gladyshev, V. N.; Becker,
D.; Dickmann, M.; Banerjee, R. Antioxidant Enzymes. Redox
Biochemistry; John Wiley & Sons, Inc.: Hoboken, NJ, 2008; pp 49−
134.
Funding
This research was supported by funds from the German
Research Foundation (DFG: SFB 1032 project B09 number
201269156, SFB TRR 152 project P24 number 239283807,
SPP 1926 project number 426018126, and Emmy Noether
grant 400324123 to O.T.-S.; SPP 1710 project BE 1540/23-2
to K.B. (now transferred to Stefan Rahlfs)); LMUExcellent
(Junior Researcher Fund to O.T.-S.); the Munich Centre for
NanoScience initiative (CeNS) to O.T.-S.; and from
Karolinska Institutet, The Knut and Alice Wallenberg
Foundations, The Swedish Cancer Society, The Swedish
Research Council, Cayman Biomedical Research Institute
(CABRI), and the Hungarian Thematic Excellence Programme
(TKP2020-NKA-26) to E.S.J.A. J.G.F. thanks the Studien-
stiftung des deutschen Volkes for support through a Ph.D.
(12) Cheng, Q.; Sandalova, T.; Lindqvist, Y.; Arnér, E. S. J. Crystal
Structure and Catalysis of the Selenoprotein Thioredoxin Reductase
1. J. Biol. Chem. 2009, 284 (6), 3998−4008.
8801
https://doi.org/10.1021/jacs.1c03234
J. Am. Chem. Soc. 2021, 143, 8791−8803

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
(13) Fritz-Wolf, K.; Urig, S.; Becker, K. The Structure of Human
Thioredoxin Reductase 1 Provides Insights into C-Terminal
Rearrangements During Catalysis. J. Mol. Biol. 2007, 370 (1), 116−
127.
(14) Gromer, S.; Johansson, L.; Bauer, H.; Arscott, L. D.; Rauch, S.;
Ballou, D. P.; Williams, C. H.; Schirmer, R. H.; Arner, E. S. J. Active
Sites of Thioredoxin Reductases: Why Selenoproteins? Proc. Natl.
Acad. Sci. U. S. A. 2003, 100 (22), 12618−12623.
(32) Cherblanc, F. L; Chapman, K. L; Brown, R.; Fuchter, M. J
Chaetocin Is a Nonspecific Inhibitor of Histone Lysine Methyl-
transferases. Nat. Chem. Biol. 2013, 9, 136−137.
(33) Zhang, L.; Duan, D.; Liu, Y.; Ge, C.; Cui, X.; Sun, J.; Fang, J.
Highly Selective Off−On Fluorescent Probe for Imaging Thioredoxin
Reductase in Living Cells. J. Am. Chem. Soc. 2014, 136 (1), 226−233.
(34) Ma, H.; Zhang, J.; Zhang, Z.; Liu, Y.; Fang, J. A Fast Response
and Red Emission Probe for Mammalian Thioredoxin Reductase.
Chem. Commun. 2016, 52 (81), 12060−12063.
(35) Li, X.; Zhang, B.; Yan, C.; Li, J.; Wang, S.; Wei, X.; Jiang, X.;
Zhou, P.; Fang, J. A Fast and Specific Fluorescent Probe for
Thioredoxin Reductase That Works via Disulphide Bond Cleavage.
Nat. Commun. 2019, 10 (2745), 1−12.
(36) Li, X.; Hou, Y.; Meng, X.; Ge, C.; Ma, H.; Li, J.; Fang, J.
Selective Activation of a Prodrug by Thioredoxin Reductase Providing
a Strategy to Target Cancer Cells. Angew. Chem., Int. Ed. 2018, 57
(21), 6141−6145.
(37) Houk, J.; Whitesides, G. M. Structure-Reactivity Relations for
Thiol-Disulfide Interchange. J. Am. Chem. Soc. 1987, 109 (22), 6825−
6836.
(38) Singh, R.; Whitesides, G. M. Degenerate Intermolecular
Thiolate-Disulfide Interchange Involving Cyclic Five-Membered
Disulfides Is Faster by Approximately 103 than That Involving Six-
or Seven-Membered Disulfides. J. Am. Chem. Soc. 1990, 112 (17),
6304−6309.
(15) Arnér, E. S. J. Focus on Mammalian Thioredoxin Reductases 
Important Selenoproteins with Versatile Functions. Biochim. Biophys.
Acta, Gen. Subj. 2009, 1790 (6), 495−526.
(16) Maroney, M. J.; Hondal, R. J. Selenium versus Sulfur:
Reversibility of Chemical Reactions and Resistance to Permanent
Oxidation in Proteins and Nucleic Acids. Free Radical Biol. Med. 2018,
127, 228−237.
(17) Sagemark, J.; Elgán, T. H.; Burglin, T. R.; Johansson, C.;
̈
Holmgren, A.; Berndt, K. D. Redox Properties and Evolution of
Human Glutaredoxins. Proteins: Struct., Funct., Genet. 2007, 68 (4),
879−892.
(18) Weichsel, A.; Gasdaska, J. R.; Powis, G.; Montfort, W. R.
Crystal Structures of Reduced, Oxidized, and Mutated Human
Thioredoxins: Evidence for a Regulatory Homodimer. Structure 1996,
4 (6), 735−751.
(19) Johansson, L.; Arscott, L. D.; Ballou, D. P.; Williams, C. H., Jr.;
Arnér, E. S. J. Studies of an Active Site Mutant of the Selenoprotein
Thioredoxin Reductase: The Ser-Cys-Cys-Ser Motif of the Insect
Orthologue Is Not Sufficient to Replace the Cys-Sec Dyad in the
Mammalian Enzyme. Free Radical Biol. Med. 2006, 41 (4), 649−656.
(20) Fritz-Wolf, K.; Kehr, S.; Stumpf, M.; Rahlfs, S.; Becker, K.
Crystal Structure of the Human Thioredoxin Reductase−Thioredoxin
Complex. Nat. Commun. 2011, 2 (383), 1−8.
(21) Pan, J. L.; Bardwell, J. C. A. The Origami of Thioredoxin-like
Folds. Protein Sci. 2006, 15 (10), 2217−2227.
(22) Palde, P. B.; Carroll, K. S. A Universal Entropy-Driven
Mechanism for Thioredoxin−Target Recognition. Proc. Natl. Acad.
Sci. U. S. A. 2015, 112 (26), 7960−7965.
(23) Meyer, A. J.; Dick, T. P. Fluorescent Protein-Based Redox
Probes. Antioxid. Redox Signaling 2010, 13 (5), 621−650.
(24) Lukyanov, K. A.; Belousov, V. V. Genetically Encoded
Fluorescent Redox Sensors. Biochim. Biophys. Acta, Gen. Subj. 2014,
1840 (2), 745−756.
(25) Schwarzländer, M.; Dick, T. P.; Meyer, A. J.; Morgan, B.
Dissecting Redox Biology Using Fluorescent Protein Sensors.
Antioxid. Redox Signaling 2016, 24 (13), 680−712.
(26) Sallin, O.; Reymond, L.; Gondrand, C.; Raith, F.; Koch, B.;
Johnsson, K. Semisynthetic Biosensors for Mapping Cellular
Concentrations of Nicotinamide Adenine Dinucleotides. eLife 2018,
7, e32638.
(27) Lee, M. H.; Yang, Z.; Lim, C. W.; Lee, Y. H.; Dongbang, S.;
Kang, C.; Kim, J. S. Disulfide-Cleavage-Triggered Chemosensors and
Their Biological Applications. Chem. Rev. 2013, 113 (7), 5071−5109.
(28) Pires, M. M.; Chmielewski, J. Fluorescence Imaging of Cellular
Glutathione Using a Latent Rhodamine. Org. Lett. 2008, 10 (5), 837−
840.
(29) Lim, C. S.; Masanta, G.; Kim, H. J.; Han, J. H.; Kim, H. M.;
Cho, B. R. Ratiometric Detection of Mitochondrial Thiols with a
Two-Photon Fluorescent Probe. J. Am. Chem. Soc. 2011, 133 (29),
11132−11135.
(30) Butora, G.; Qi, N.; Fu, W.; Nguyen, T.; Huang, H.-C.; Davies,
I. W. Cyclic-Disulfide-Based Prodrugs for Cytosol-Specific Drug
Delivery. Angew. Chem., Int. Ed. 2014, 53 (51), 14046−14050.
(31) Felber, J. G.; Poczka, L.; Busker, S.; Theisen, U.; Zeisel, L.;
Maier, M. S.; Brandstädter, C.; Scholzen, K.; Becker, K.; Arnér, E. S.
J.; Ahlfeld, J.; Thorn-Seshold, O. Cyclic 5-Membered Disulfides Are
Not Selective Substrates of Thioredoxin Reductase, but Are Opened
Nonspecifically by Thiols. ChemRxiv 2020, DOI: 10.26434/
chemrxiv.13483155.v1 (accessed May 9, 2021).
(39) Abegg, D.; Gasparini, G.; Hoch, D. G.; Shuster, A.; Bartolami,
E.; Matile, S.; Adibekian, A. Strained Cyclic Disulfides Enable Cellular
Uptake by Reacting with the Transferrin Receptor. J. Am. Chem. Soc.
2017, 139 (1), 231−238.
(40) Gasparini, G.; Sargsyan, G.; Bang, E.-K.; Sakai, N.; Matile, S.
Ring Tension Applied to Thiol-Mediated Cellular Uptake. Angew.
Chem., Int. Ed. 2015, 54 (25), 7328−7331.
(41) Cheng, Y.; Pham, A.-T.; Kato, T.; Lim, B.; Moreau, D.; López-
Andarias, J.; Zong, L.; Sakai, N.; Matile, S. Inhibitors of Thiol-
Mediated Uptake. Chem. Sci. 2021, 12, 626.
(42) Butora, G. Nucleoside Kinase Bypass Compositions and
Methods. WO2014088923A1, 2014.
(43) Hayashi, J.; Samezawa, Y.; Ochi, Y.; Wada, S.; Urata, H.
Syntheses of Prodrug-Type Phosphotriester Oligonucleotides Re-
sponsive to Intracellular Reducing Environment for Improvement of
Cell Membrane Permeability and Nuclease Resistance. Bioorg. Med.
Chem. Lett. 2017, 27 (14), 3135−3138.
(44) Kong, F.; Zhao, Y.; Liang, Z.; Liu, X.; Pan, X.; Luan, D.; Xu, K.;
Tang, B. Highly Selective Fluorescent Probe for Imaging H2 Se in
Living Cells and in Vivo Based on the Disulfide Bond. Anal. Chem.
2017, 89 (1), 688−693.
(45) Ziv, I. Compounds and Methods for Trans-Membrane Delivery
of Molecules. US9687556B2, 2017.
(46) Ziv, I. Pro-Drugs and Related Methods. WO2017017669A1,
2017.
(47) Na, Y.; Kohn, H. Quinone-Cyclized Porfiromycins. Heterocycles
2001, 55 (7), 1347−1364.
(48) Lee, S. H.; Kohn, H. Cyclic Disulfide C(8) Iminoporfiromycin:
Nucleophilic Activation of a Porfiromycin. J. Am. Chem. Soc. 2004,
126 (13), 4281−4292.
(49) Ollivier, N.; Dheur, J.; Mhidia, R.; Blanpain, A.; Melnyk, O.
Bis(2-Sulfanylethyl)Amino Native Peptide Ligation. Org. Lett. 2010,
12 (22), 5238−5241.
(50) Zhu, B.; Zhang, X.; Li, Y.; Wang, P.; Zhang, H.; Zhuang, X. A
Colorimetric and Ratiometric Fluorescent Probe for Thiols and Its
Bioimaging Applications. Chem. Commun. 2010, 46 (31), 5710.
(51) Hong, K.-H.; Kim, D. I.; Kwon, H.; Kim, H.-J. A
Fluoresceinylcarbonate-Based Fluorescent Probe for the Sensitive
Detection of Biothiols in a HEPES Buffer and Its Cellular Expression.
RSC Adv. 2014, 4 (2), 978−982.
(52) Dan, K.; Veetil, A. T.; Chakraborty, K.; Krishnan, Y. DNA
Nanodevices Map Enzymatic Activity in Organelles. Nat. Nanotechnol.
2019, 14 (3), 252−259.
8802
https://doi.org/10.1021/jacs.1c03234
J. Am. Chem. Soc. 2021, 143, 8791−8803

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
(53) Mata, G.; do Rosário, V. E.; Iley, J.; Constantino, L.; Moreira,
R. A Carbamate-Based Approach to Primaquine Prodrugs: Anti-
malarial Activity, Chemical Stability and Enzymatic Activation. Bioorg.
Med. Chem. 2012, 20 (2), 886−892.
(54) Ghosh, A. K.; Brindisi, M. Organic Carbamates in Drug Design
and Medicinal Chemistry. J. Med. Chem. 2015, 58 (7), 2895−2940.
(55) Vacondio, F.; Silva, C.; Mor, M.; Testa, B. Qualitative
Structure-Metabolism Relationships in the Hydrolysis of Carbamates.
Drug Metab. Rev. 2010, 42 (4), 551−589.
(56) Thorn-Seshold, O.; Vargas-Sanchez, M.; McKeon, S.;
Hasserodt, J. A Robust, High-Sensitivity Stealth Probe for Peptidases.
Chem. Commun. 2012, 48 (50), 6253−6255.
(57) Tian, L.; Yang, Y.; Wysocki, L. M.; Arnold, A. C.; Hu, A.;
Ravichandran, B.; Sternson, S. M.; Looger, L. L.; Lavis, L. D. Selective
Esterase-Ester Pair for Targeting Small Molecules with Cellular
Specificity. Proc. Natl. Acad. Sci. U. S. A. 2012, 109 (13), 4756−4761.
(58) Orte, A.; Crovetto, L.; Talavera, E. M.; Boens, N.; Alvarez-Pez,
J. M. Absorption and Emission Study of 2‘,7‘-Difluorofluorescein and
Its Excited-State Buffer-Mediated Proton Exchange Reactions. J. Phys.
Chem. A 2005, 109 (5), 734−747.
(59) Machida, T.; Dutt, S.; Winssinger, N. Allosterically Regulated
Phosphatase Activity from Peptide-PNA Conjugates Folded Through
Hybridization. Angew. Chem., Int. Ed. 2016, 55 (30), 8595−8598.
(60) Rando, G.; Winssinger, N.; Lindberg, E.; Anzola, M. New
Compounds and Uses Thereof for Detection of Target Molecules in a
Sample. US 2019/0315713 A1.
(74) Jia, H.; Hu, G.; Shi, D.; Gan, L.; Zhang, H.; Yao, X.; Fang, J.
Fluorophore-Dependent Cleavage of Disulfide Bond Leading to a
Highly Selective Fluorescent Probe of Thioredoxin. Anal. Chem. 2019,
91 (13), 8524−8531.
(75) Dal Corso, A.; Borlandelli, V.; Corno, C.; Perego, P.; Belvisi, L.;
Pignataro, L.; Gennari, C. Fast Cyclization of a Proline-Derived Self-
Immolative Spacer Improves the Efficacy of Carbamate Prodrugs.
Angew. Chem., Int. Ed. 2020, 59 (10), 4176−4181.
(76) Laurent, Q.; Sakai, N.; Matile, S. The Opening of 1,2-
Dithiolanes and 1,2-Diselenolanes: Regioselectivity, Rearrangements,
and Consequences for Poly(Disulfide)s, Cellular Uptake and Pyruvate
Dehydrogenase Complexes. Helv. Chim. Acta 2019, 102 (2),
e1800209.
(77) Chyan, W.; Kilgore, H. R.; Gold, B.; Raines, R. T. Electronic
and Steric Optimization of Fluorogenic Probes for Biomolecular
Imaging. J. Org. Chem. 2017, 82 (8), 4297−4304.
(78) Cheng, Q.; Arnér, E. S. J. Selenocysteine Insertion at a
Predefined UAG Codon in a Release Factor 1 (RF1)-Depleted
Escherichia Coli Host Strain Bypasses Species Barriers in Recombi-
nant Selenoprotein Translation. J. Biol. Chem. 2017, 292 (13), 5476−
5487.
(79) Cheng, Q.; Arnér, E. S. J. Overexpression of Recombinant
Selenoproteins in E. Coli. In Selenoproteins; Chavatte, L., Ed.;
Methods in Molecular Biology; Springer: New York, 2018; Vol.
1661, pp 231−240. DOI: 10.1007/978-1-4939-7258-6_17.
(80) Lacey, B. M.; Eckenroth, B. E.; Flemer, S.; Hondal, R. J.
Selenium in Thioredoxin Reductase: A Mechanistic Perspective.
Biochemistry 2008, 47 (48), 12810−12821.
(81) Espinosa, B.; Arnér, E. S. J. Thioredoxin-Related Protein of 14
KDa as a Modulator of Redox Signalling Pathways: TRP14 in Redox
Signalling. Br. J. Pharmacol. 2019, 176 (4), 544−553.
(61) Fu, Q.; Li, H.; Duan, D.; Wang, C.; Shen, S.; Ma, H.; Liu, Z.
External-Radiation-Induced Local Hydroxylation Enables Remote
Release of Functional Molecules in Tumors. Angew. Chem., Int. Ed.
2020, 59 (48), 21546−21552.
(62) Lukesh, J. C.; Palte, M. J.; Raines, R. T. A Potent, Versatile
Disulfide-Reducing Agent from Aspartic Acid. J. Am. Chem. Soc. 2012,
134 (9), 4057−4059.
(82) Woo, J. R.; Kim, S. J.; Jeong, W.; Cho, Y. H.; Lee, S. C.; Chung,
Y. J.; Rhee, S. G.; Ryu, S. E. Structural Basis of Cellular Redox
Regulation by Human TRP14. J. Biol. Chem. 2004, 279 (46), 48120−
48125.
(63) Lees, W. J.; Whitesides, G. M. Equilibrium Constants for Thiol-
Disulfide Interchange Reactions: A Coherent, Corrected Set. J. Org.
Chem. 1993, 58 (3), 642−647.
̂
(64) Chenevert, R.; Dickman, M. Enzymatic Route to Chiral,
Nonracemic Cis-2,6- and Cis,Cis-2,4,6-Substituted Piperidines. Syn-
thesis of (+)-Dihydropinidine and Dendrobate Alkaloid (+)-241D. J.
Org. Chem. 1996, 61 (10), 3332−3341.
(65) Le Corre, L.; Kizirian, J.-C.; Levraud, C.; Boucher, J.-L.;
Bonnet, V.; Dhimane, H. Diastereoselective Functionalizations of
Enecarbamates Derived from Pipecolic Acid towards 5-Guanidinopi-
pecolates as Arginine Mimetics. Org. Biomol. Chem. 2008, 6 (18),
3388−3398.
(66) Vartak, A. P.; Gabriela Deaciuc, A.; Dwoskin, L. P.; Crooks, P.
A. Quinlobelane: A Water-Soluble Lobelane Analogue and Inhibitor
of VMAT2. Bioorg. Med. Chem. Lett. 2010, 20 (12), 3584−3587.
(67) Yang, B. V.; O’Rourke, D.; Li, J. Mild and Selective
Debenzylation of Tertiary Amines Using α-Chloroethyl Chlorofor-
mate. Synlett 1993, 1993 (03), 195−196.
(68) Thorn-Seshold, O.; Felber, J.; Thorn-Seshold, J.; Zeisel, L.
Disulfide-Based Prodrug Compounds. EP1163944.8.
(69) Capasso, S.; Zagari, A. 1,2-Dithiane-4,5-Diol. Acta Crystallogr.,
Sect. B: Struct. Crystallogr. Cryst. Chem. 1981, 37 (7), 1437−1439.
(70) Kilgore, H. R.; Raines, R. T. Disulfide Chromophores Arise
from Stereoelectronic Effects. J. Phys. Chem. B 2020, 124 (19), 3931−
3935.
(71) Burns, J. A.; Whitesides, G. M. Predicting the Stability of Cyclic
Disulfides by Molecular Modeling: Effective Concentrations in Thiol-
Disulfide Interchange and the Design of Strongly Reducing Dithiols. J.
Am. Chem. Soc. 1990, 112 (17), 6296−6303.
(72) Luttringhaus, A.; Brechlin, A. Cyclische Disulfide, III. cis- und
̈
trans-2.3-Dithia-decalin. Chem. Ber. 1959, 92 (9), 2271−2277.
(73) Rothwarf, D. M.; Scheraga, H. A. Equilibrium and Kinetic
Constants for the Thiol-Disulfide Interchange Reaction between
Glutathione and Dithiothreitol. Proc. Natl. Acad. Sci. U. S. A. 1992, 89
(17), 7944−7948.
8803
https://doi.org/10.1021/jacs.1c03234
J. Am. Chem. Soc. 2021, 143, 8791−8803