Visible-light photoredox-catalyzed desulfurization of thiol- and disulfide-containing amino acids and small peptides

Article abstract of DOI:10.1002/psc.3016

A scalable protocol for the desulfurization of cysteine by using visible light, the photocatalyst Ir(dF(CF₃)ppy)₂(dtb-bpy)PF₆ and triethylphosphite under biphasic reaction conditions has been developed. The loading of the catalyst can be as low as 0.01?mol%, which can be efficiently removed during the workup (≤0.3?ppm), giving rise to the corresponding desulfurized product in high yields. This method has been applied also to cystine, penicillamine, and reduced and oxidized glutathione. The desulfurization has been found to be pH sensitive, with an optimal pH value of 6.5 and 7.0 for the cysteine derivatives and glutathione, respectively. In addition, during the desulfurization of a decapeptide containing cysteine and methionine, concurrent oxidation of the two sulfur-containing residues to disulfide and sulfoxide has been observed. Therefore, whereas the presented protocol allows a straightforward visible light-mediated desulfurization of simple thiols by using very low catalyst loading and a cost-effective trialkylphosphite as thiyl radical trapping agent, its application to complex substrates needs to be carefully validated. Copyright

Full text of DOI:10.1002/psc.3016

Research Article
Received: 8 March 2017
Revised: 11 May 2017
Accepted: 12 May 2017
Published online in Wiley Online Library
(wileyonlinelibrary.com) DOI 10.1002/psc.3016
Visible-light photoredox-catalyzed
desulfurization of thiol- and disulfide-
containing amino acids and small
peptides
Myungmo Lee,a Saskia Neukirchen,b,c Chiara Cabreleb*
and Oliver Reisera*
A scalable protocol for the desulfurization of cysteine by using visible light, the photocatalyst Ir(dF(CF3)ppy)2(dtb-bpy)PF6 and
triethylphosphite under biphasic reaction conditions has been developed. The loading of the catalyst can be as low as 0.01 mol
%, which can be efficiently removed during the workup (≤0.3 ppm), giving rise to the corresponding desulfurized product in high
yields. This method has been applied also to cystine, penicillamine, and reduced and oxidized glutathione. The desulfurization has
been found to be pH sensitive, with an optimal pH value of 6.5 and 7.0 for the cysteine derivatives and glutathione, respectively. In
addition, during the desulfurization of a decapeptide containing cysteine and methionine, concurrent oxidation of the two sulfur-
containing residues to disulfide and sulfoxide has been observed. Therefore, whereas the presented protocol allows a
straightforward visible light-mediated desulfurization of simple thiols by using very low catalyst loading and a cost-effective
trialkylphosphite as thiyl radical trapping agent, its application to complex substrates needs to be carefully validated.
Copyright © 2017 European Peptide Society and John Wiley & Sons, Ltd.
Additional Supporting Information may be found online in the supporting information tab for this article.
Keywords: desulfurization; photocatalysis; photoredox; iridium; cysteine; glutathione; penicillamine
published a method for the defunctionalization of thiols and
disulfides like cysteine, cystine, reduced and oxidized glutathione
Introduction
Visible-light mediated photoredox catalysis is an active area of
research that aims at the development of safer and
environmentally friendly protocols for radical reactions [1–5].
Photocatalytic defunctionalizations [6–15] represent one important
class of reactions with broad applicability, ranging from biofuel to
(bio)material production [16,17]. We also have contributed to these
areas by developing visible-light mediated protocols for the
deoxygenation of alcohols [13] and halides [18]. Likewise,
desulfurizations are of great importance to remove sulfur from
compounds that would otherwise be harmful for the environment
or act as poison in catalytic transformations. Moreover, since the
introduction in the nineties [19] of the cysteine-mediated native
chemical ligation (NCL) approach for the (semi)synthesis of
proteins, the desulfurization reaction has also become very useful
for peptide and protein chemists [20–23]. Indeed, desulfurization
of cysteine converts this residue into alanine: therefore, besides
natural cysteine residues, also, natural alanine residues along the
sequence can be selected as ligation sites, which are temporarily
replaced by cysteine to perform the ligation and then converted
to alanine by desulfurization [24–26].
(GSH/GSSG), which is based on the use of trialkylborane,
trialkylphosphite and visible light.
For thiol-containing peptidic substrates, the organometallic
desulfurization with Raney-Ni [32] or Pd/Al2O3 [33] has been
reported. In addition, Danishefsky and co-workers [34–36] have
introduced a mild and very effective desulfurization protocol that
relies on the use of VA-044 (2,20-azobis[2-(2-imidazolin-2-yl)
propane]dihydrochloride) as the radical initiator at 37 °C.
Nevertheless, VA-044 requires special safety and storage
precautions, because it decomposes already at room temperature.
Recently, Guo and co-workers have reported on an efficient
desulfurization reaction of cysteine-containing peptides by using
a photocatalyst in the presence of a water-soluble phosphine [37].
*
Correspondence to: Oliver Reiser, Institute of Organic Chemistry, University of
Regensburg, D-93053 Regensburg, Germany. Chiara Cabrele, Department of
Molecular Biology, University of Salzburg, A-5020 Salzburg, Austria. E-mail:
oliver.reiser@chemie.uni-regensburg.de; chiara.cabrele@sbg.ac.at
a Institute of Organic Chemistry, University of Regensburg, D-93053, Regensburg,
The first radical desulfurization of mercaptans was reported in
1956 by Hoffmann and co-workers [27], who used trialkyl
phosphites at reflux temperatures or UV-light irradiation. Shortly
after, Walling and co-workers [28,29] proposed that the role of
phosphite was that of capturing the thiyl radical in form of
thiophosphate. In the late nineties, Gonzales and Valencia [30,31]
Germany
b Department of Molecular Biology, University of Salzburg, A-5020, Salzburg,
Austria
c
Department of Chemistry and Biochemistry, Ruhr-University Bochum,
Universitätsstrasse 150, 44801, Bochum, Germany
J. Pept. Sci. 2017
Copyright © 2017 European Peptide Society and John Wiley & Sons, Ltd.

M. LEE ET AL.
This reaction relies on the formation of the thiyl radical by single
electron transfer (SET) to the photocatalyst in the excited state, a
mechanism that has been exploited in several visible-light
mediated photocatalytic reactions, for example in the synthesis of
2-substituted benzothiazoles [38–40], or in the thiol-ene
reaction [41–45]. The authors have found that GSH can be
desulfurized in 5 h in the presence of 5 mol% Ru(bpy)3Cl2 (B,
bpy = bipyridine; Figure 1) and of large excess (50 equiv.) 4,40,4″-
phosphanetriyltribenzenesulfonic acid trisodium salt (TPPTS) in
phosphate buffer (pH 7.4). Notably, this desulfurization protocol
was also applicable to a large variety of peptides. In contrast,
(fac)-Ir(ppy)3 (C, ppy = 2-phenylpyridine; Figure 1) as photocatalyst
or 3,30,3″-phosphanetriyltripropionic acid (TCEP) as phosphine
source gave poor yields [37].
3.31 ppm, CHD2C(O)CD3 at 2.05 ppm and HOD at 4.8 ppm.
Chemical shifts for 13C-NMR were reported as δ, parts per million,
relative to the signal of CDCl3 at 77.0 ppm, CD3OD at 49.1 ppm
and (CD3)2CO at 29.9 ppm. Coupling constants (J) are given in Hertz
(Hz). The following notations indicate the multiplicity of the signals:
s = singlet, d = doublet, t = triplet, q = quartet, dd = doublet of
doublets, td = triplet of doublets and m = multiplet. Mass
spectra were recorded on Agilent Technologies 6540 UHD
Accurate-Mass Q-TOF LC/MS and Bruker Daltonics Autoflex
Speed MALDI-TOF-MS (matrix: α-cyano-4-hydroxycinnamic
acid). Analytical/semipreparative high-performance
liquid
chromatography (HPLC) was performed on a Thermo Fisher
Scientific Ultimate 3000 system. The yields reported are referred
to the isolated compounds unless otherwise stated. The trace
iridium metal was determined by inductively coupled plasma
optical emission spectrometry on Spectro Analytical Instruments
ICP Modula EOP (λ = 212 nm).
We report here an alternative protocol that allows a scalable,
visible-light photocatalytic approach for the desulfurization of
cysteine derivatives and glutathione using Ir(dF(CF3)ppy)2(dtb-
bpy)PF6 (E, Figure 1) as photocatalyst with loadings as low as
0.01 mol% in the presence of triethylphosphite.
General Desulfurization Procedure for Cysteine Derivatives,
Glutathione and Penicillamine
Materials and Methods
The thiol-containing substrate (0.25 mmol) and Ir(dF(CF3)ppy)2(dtb-
bpy)PF6 (E, 2.8 mg, 0.0025 mmol, 1 mol%) were placed into a 9-ml
vial. MeCN (0.75 ml) and phosphate buffer (0.75 ml, pH 6.5, 1.0 M)
were added, followed by triethylphosphite (0.13 ml, 0.75 mmol, 3
equiv.). The biphasic reaction mixture was irradiated with 455-nm
blue LED from the bottom of the vial at room temperature with
rapid stirring. After 30 min, the aqueous phase was analyzed by
1H- and 13C-NMR. After full consumption of the substrate, the
aqueous phase was separated and purified by column
chromatography with DOWEX® 50WX8-100 ion-exchange resin to
afford the desulfurized product.
All commercial chemical materials (Aldrich, Fluka, Merck,
Novabiochem) were used as received without further purification.
Irradiation was performed using blue light emitting diodes (3 W,
λmax = 455 nm). Analytical thin layer chromatography was
performed on Merck thin layer chromatography aluminum sheets
silica gel 60 F 254. Flash column chromatography was performed
using Merck flash silica gel 60 (0.040–0.063 mm). Nuclear magnetic
resonance (NMR) spectra were recorded on Bruker Avance 300
spectrometer. Chemical shifts for 1H-NMR were reported as δ, parts
per million, relative to the signal of CHCl3 at 7.26 ppm, CHD2OD at
Figure 1. Structure of ruthenium and iridium photocatalysts.
wileyonlinelibrary.com/journal/jpepsci
Copyright © 2017 European Peptide Society and John Wiley & Sons, Ltd.
J. Pept. Sci. 2017

PHOTOCATALYZED CYSTEINE DESULFURIZATION
Synthesis and Desulfurization of Ac-YNMNDCYSKL-NH2
Table 1. Ruthenium-catalyzed desulfurization of L-cysteine (L-Cys) and
glutathione (GSH)a
The cysteine-containing peptide (1e, Ac-YNMNDCYSKL-NH2) was
prepared by Fmoc-based solid-phase methodology on the Rink
Amide MBHA resin using the automated peptide synthesizer
Syro I from Biotage. The peptide was cleaved from the resin
and fully deprotected by treatment with trifluoroacetic
acid/triisopropylsilane/ethanedithiol/H2O/thioanisole
80:5:5:5:5
Entry
Substrate
Catalystb
Solventc
Time
(h)
Conversiond
(%)
(v/v) for 3 h and recovered by precipitation from diethylether,
centrifuged and washed several times with diethylether. The crude
peptide was purified by semi-preparative HPLC to afford the pure
peptide that was lyophilized and characterized by analytical HPLC
and MALDI-TOF-MS (supplementary Figures S1 and S2).
1
2
3
4
L-Cys
L-Cys
GSH
A
B
B
B
THF/H2O
THF/H2O
THF/H2O
THF/PB
pH 6.5
6
6
63
99
2
18
1
L-Cys
99
Desulfurization of the peptide was carried out by using a
phosphite-based (a) and a phosphine-based (b) procedure: the
peptide (1.2 mg, 0.001 mmol) was added to a 9-ml vial with a
magnetic stirring bar. Ir(dF(CF3)ppy)2(dtb-bpy)PF6 photocatalyst E
(0.011 mg, 0.00001 mmol, 1 mol%) was added by using the stock
solution of the catalyst (0.1 ml of the stock solution: 1.1 mg catalyst
in 10 ml MeCN), followed by phosphate buffer (0.1 ml, pH 6.5, 1.0 M)
and (a) triethylphosphite (0.1 ml, 0.58 mmol, 580 equiv.) or (b) TCEP
(25 mg, 0.1 mmol, 100 equiv. in 0.8 ml MeCN/buffer 1:1, v/v). The
biphasic reaction mixtures were irradiated with 455-nm blue LED
from the bottom of the vial and stirred for 1 h at room temperature,
then evaporated in vacuo and analyzed by analytical HPLC and
MALDI-TOF-MS (supplementary Figure S3). The crude product
obtained by the procedure (a) was purified by HPLC to afford
the alanyl peptide (Ac-YNM(O)NDAYSKL-NH2) in 12% yield
(supplementary Figure S4).
5
6
GSH
GSH
B
B
THF/PB
pH 6.5
19
19
17
52
THF/PB
pH 7.0
aGeneral procedure: Substrate (0.25 mmol), catalyst (0.0025 mmol), P
(OEt)3 (0.75 mmol), THF (0.75 ml) and aqueous solvent (0.75 ml) at
room temperature.
bA: Ru(bpz)3Cl2; B: Ru(bpy)3Cl2.
cPhosphate buffer (PB, 1.0 M) was prepared by NaH2PO4 and Na2HPO4 in
distilled water.
dDetermined by 1H-NMR analysis based on ratio of starting material and
product.
phosphine source instead of triethylphosphite. Nevertheless, given
the more than 50-fold price/mol of TPPTS, for large-scale
applications, it would be clearly advantageous to employ
triethylphoshite as the thiyl radical trapping reagent. Therefore,
we carried out further experiments to optimize the protocol based
on the use of triethylphosphite.
General Desulfurization Procedure for Cystine and GSSG
The disulfide substrate (0.10 mmol), Ir(dF(CF3)ppy)2(dtb-bpy)PF6 (E)
(1.0 mg, 0.001 mmol, 1 mol%) and TCEP (75 mg, 0.30 mmol, 3
equiv.) were added to a 10-ml Schlenk tube. MeCN (0.5 ml) and
phosphate buffer (0.5 ml, pH 6.5, 1.0 M) were added. The Schlenk
tube was closed by the screw cap with Teflon sealed inlet for a glass
rod, then the biphasic reaction mixture was irradiated with 455-nm
blue LED from the top and stirred for 1 h at 37 °C using an oil bath.
First, we decided to replace water with phosphate buffer (1.0 M)
at pH 6.5, in order to work at pH conditions similar to those
previously used by Danishefsky and co-workers [34]: notably, the
buffer system accelerated the reaction significantly (1 h instead of
6 h, Table 1, entry 4). Thus, we repeated the desulfurization of
GSH in the same buffer, but, unfortunately, with scarce result
(Table 1, entry 5). However, we found that raising the pH value from
6.5 to 7.0 favored the desulfurization process of GSH (Table 1, entry
6). We confirmed the pH dependency of the GSH desulfurization by
carrying out the reaction at different pH values over the range 6.5–
7.5 (supplementary Figure S5): in fact, the reaction was efficient
only above pH 6.9. This may also explain why the attempt to
desulfurize GSH dissolved in water in the absence of a buffer was
unsuccessful (Table 1, entry 3), as the acidifying property of GSH
may lower the water pH down to 3 at a concentration of about
0.4 M [47]. In the case of cysteine, which has a weaker acidifying
effect on water than GSH (pH 4.5–5.5, as measured for cysteine
concentrations in the range of about 0.1–0.3 M), the desulfurization
occurred both in the absence and in the presence of a buffer at
pH 6.5; however, the benefit of the less acidic pH value on the
reaction time further points out that the desulfurization process is
pH dependent.
1
The aqueous phase was analyzed by H- and 13C-NMR. After full
consumption of the substrate, the aqueous phase was separated
and purified by column chromatography with DOWEX® 50WX8-
100 ion-exchange resin to afford the desulfurized product.
Results and Discussion
We began our study by testing the desulfurization of cysteine in
THF/H2O in the presence of triethylphosphite (3 equiv.) and 1 mol
% Ru(bpz)3Cl2 (A, bpz = bipyrazine, E(2+*/+) = +1.35 V) as
photocatalyst, using a blue LED (455 nm) as a visible-light source.
After 6 h, 63% of cysteine was consumed with concurrent formation
of alanine as the sole product (Table 1, entry 1). In a second
experiment, we replaced Ru(bpz)3Cl2 (A) with the commercially
available Ru(bpy)3Cl2 (B) that displays a smaller oxidation potential
(E(2+*/+) = +0.77 V) [46]. Gratifyingly, a nearly quantitative
consumption of cysteine was achieved with this catalyst (Table 1,
entry 2). Furthermore, we treated glutathione (GSH) under the latter
reaction conditions: surprisingly, only trace amounts of the
desulfurized product were formed despite longer reaction times
(Table 1, entry 3), contrasting the report by Guo and co-workers
who obtained good results in this transformation using 5 mol%
Ru(bpz)3Cl2 (A) or Ru(bpy)3Cl2 (B) with 2.5 equiv. TPPTS as
Second, we tested various iridium photocatalysts (C–E, Figure 1)
for the cysteine desulfurization in MeCN/buffer (pH 6.5). The results
obtained were significantly better than those with the ruthenium
photocatalysts A and B: indeed, using 1 mol% catalyst (Table 2,
entries 2 and 3), the reactions were completed in only 0.5 h giving
J. Pept. Sci. 2017
Copyright © 2017 European Peptide Society and John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/jpepsci

M. LEE ET AL.
Table 2. Iridium-catalyzed desulfurization of L-cysteine derivativesa
Entry
Substrate
Catalystb
Solventc
Time (h)
Yield (%)
1
2
3
4
L-Cys
L-Cys
L-Cys
L-Cys
C
D
E
MeCN/PB pH 6.5
MeCN/PB pH 6.5
MeCN/PB pH 6.5
MeCN/PB pH 6.5
0.5
0.5
0.5
48
86d
99d
96 (99d)
96 (99d)
E
(25 mmol)
GSH
5
6
E
MeCN/PB pH 7.0
MeCN/PB pH 6.5
2
99
L-Cys
E
0.5
—
(dark)
9
Boc-L-Cys-OMe
E
E
E
E
MeCN/PB pH 6.5
MeCN/PB pH 6.5
MeCN/PB pH 6.5
MeCN/PB pH 6.5
0.5
0.5
1
97 (99d)
84 (99d)
12f
10
11e
12
Fmoc-L-Cys-OH
Ac-YNMN-DCYSKL-NH2
D-Penicillamine (R3 = Me)
0.5
96 (99d)
aGeneral procedure: Substrate (0.25 mmol), catalyst (0.0025 mmol), P(OEt)3 (0.75 mmol), MeCN (0.75 ml) and aqueous PO4 buffer (0.75 ml) at room
temperature.
bC: fac-Ir(ppy)3 (ppy = 2-phenylpyridyl); D: Ir(ppy)2(dtb-bpy)PF6 (dtb-bpy = 4,40-di-tert-butyl-2,20-dipyridyl); E: Ir((dF(CF3)ppy)2(dtb-bpy)PF6 (dF(CF3)ppy = 2-
(2,4-difluorophenyl)-5-trifluoromethylpyridyl).
cPhosphate buffer (1.0 M) was prepared by NaH2PO4 and Na2HPO4 in distilled water.
dConversion determined by 1H-NMR analysis based on ratio of starting material and product.
e1 μmol peptide and 0.58 mmol P(OEt)3 were used.
fThe methionine residue was quantitatively oxidized to methionine sulfoxide.
rise to alanine in quantitative yield. Also, GSH was completely
desulfurized in 2 h (Table 2, entry 5), dramatically demonstrating
in this case the far superior properties of the Ir3+ over the Ru2+
catalysts. Moreover, in an attempt to reduce the amount of the
catalyst, we found that 0.01 mol% Ir(dF(CF3)ppy)2(dtb-bpy)PF6 E
was sufficient to convert cysteine to alanine again in a quantitative
and clean manner within 48 h on a multigram scale (Table 2, entry
4). It should be pointed out that photoredox catalyzed reactions
with the iridium complex E in biphasic solvents such as
MeCN/H2O are well known to proceed efficiently [48,49]. Thus,
the biphasic systems employed in this study have the advantage
that the photocatalyst as well as excess triethylphosphite remains
in the organic phase, while the product resides in the aqueous
phase. Especially significant, the iridium contamination of the
desulfurized product is below 0.3 ppm as determined by ICP-OES
analysis.
To investigate the importance of visible light on the reaction, we
performed a dark experiment with cysteine (Table 2, entry 6): no
conversion was observed, indicating that visible light is required
for the reaction to occur. Furthermore, we successfully applied
the optimized desulfurization procedure to N-Boc-protected
cysteine methylester and N-Fmoc-protected cysteine (Table 2,
entries 9 and 10).
found that methionine was converted to methionine sulfoxide
(supplementary Figures S3 and S4). It is known that exposure of
proteins to daylight induces methionine oxidation [50]. Therefore,
it is plausible to assume that the iridium photocatalyst triggered
the formation of methionine sulfoxide, which is known to occur
with other photocatalysts as well [51]. However, such side reaction
was not reported to occur when methionine-containing peptides
were desulfurized with Ru(bpy)3Cl2 (B) in the presence of a sulfur
additive like t-butylmercaptan [37]. In addition to methionine
oxidation, we obtained several minor byproducts containing
disulfide bonds and other unidentified modifications, which
decreased the yield of the alanine-containing product 2e
(supplementary Figure S4).
Next, we selected penicillamine to test the effect of the steric
hindrance of the thiol group on the desulfurization: the substrate
was photochemically desulfurized in excellent yield (Table 2,
entry 12).
Finally, we investigated the behavior of disulfides under our
desulfurization conditions. For this purpose, we used cystine as
substrate (Table 3, entry 1-4). Due to the insolubility of cystine
between pH 6 and 7, besides phosphate buffer we also evaluated
1 M KOH (entry 1). However, no reaction was observed, indicating
that the disulfide bond was stable under the photochemical
desulfurization protocol. Based on this observation, we adapted
our protocol to a domino disulfide-reduction/desulfurization in a
one-pot reaction. A common water-soluble disulfide reducing
agent in peptides and proteins is TCEP [34,52]. Therefore, we
replaced triethylphosphite, which is apparently a too weak
reductant for S―S bonds (Table 3, entry 2 vs 3), with TCEP and
increased the reaction temperature to 37 °C to allow the efficient
generation of the thiol substrate in situ (Table 3, entry 3 vs 4). Under
To assess the compatibility of the protocol with the presence of
sensitive amino acids in peptides, we used the peptide sequence
YNMNDCYSKL that, besides cysteine, contains the (photo)redox-
reactive tyrosine and methionine residues as well as the
deamidation/dehydration-prone residues asparagine, aspartic acid
and serine. It was necessary to use a large excess of phosphite
to obtain the corresponding alanine-containing peptide
(Table 2, entry 11). However, besides Cys-to-Ala conversion, we also
wileyonlinelibrary.com/journal/jpepsci
Copyright © 2017 European Peptide Society and John Wiley & Sons, Ltd.
J. Pept. Sci. 2017

PHOTOCATALYZED CYSTEINE DESULFURIZATION
Table 3. Iridium-catalyzed desulfurization of disulfidesa,b
mixture was irradiated for 1 min, followed by 9 min of dark reaction.
The substrate consumption was 11%, as obtained in the 1 min of
light reaction. Thus, the hydrogen atom source for the capture of
the alkyl radical may be hydroperoxyl radicals (HO2•) that are
formed in the course of the iridium catalyst regeneration by oxygen
[53,54]. However, as it has been previously reported that the radical
propagation should be considered even if the reaction stops in the
dark [55,56], the thiol substrate might transfer a hydrogen atom to
initiate a chain process (dark reaction).
Entry Substrate
PR3
Temperature (°C) Time (h) Yield (%)
1c
2
L-Cystine
L-Cystine
L-Cystine
L-Cystine
GSSG
P(OEt)3
P(OEt)3
TCEP
r.t.
37
37
r.t.
37
18
18
1
—
—
85 (99d)
18d
75 (96d)
3
Conclusions
4
5e
TCEP
16
1
TCEP
In summary,
a
scalable visible-light photoredox-catalyzed
desulfurization of cysteine derivatives was developed. Only
catalytic amounts of an iridium photocatalyst down to 0.01 mol %
are necessary, which is readily removed by the use of a biphasic
aqueous organic system, followed by the chromatographic workup
of the products in the aqueous phase. In this manner, less than
0.3 ppm iridium could be detected on the desulfurized products,
as determined by ICP-OES analysis. Triethylphosphite serves as
the trapping reagent for sulfur, but no other additives such as
hydrogen sources are required. Disulfides were also desulfurized
in a one-pot process employing TCEP as S―S bond breaker at
37 °C. This protocol may be applied to cysteine-containing
peptides, provided methionine is not present. However, the
photocatalyzed ruthenium-based protocol recently developed by
Guo and coworkers [37] for medium-sized peptides is superior to
the one reported here for small-scale desulfurization of cysteine-
containing peptides.
aGeneral procedure: Substrate (0.1 mmol), catalyst (0.001 mmol),
phosphine source (0.3 mmol), MeCN (0.5 ml) and aqueous phosphate
buffer (0.5 ml, pH 6.5).
bPhosphate buffer was prepared by NaH2PO4 and Na2HPO4 in distilled
water.
c1 M KOH was used in place of the phosphate buffer, to dissolve the
substrate.
dConversion determined by 1H-NMR analysis based on ratio of starting
material and product.
ePhosphate buffer (pH 7.0) was used.
these conditions, cystine was desulfurized completely in 1 h, and
alanine was isolated in 85% yield (Table 3, entry 3). Likewise, GSSG
could be desulfurized under these conditions in a highly efficient
manner (Table 3, entry 5).
The proposed, broadly accepted mechanism of the
photocatalyzed desulfurization involves the formation of an
alkylthiyl radical (Scheme 1, light reaction), which is trapped to a
phosphoranyl radical, followed by fragmentation to an alkyl radical.
A rapid hydrogen transfer from the thiol to the alkyl substrate
would provide the desulfurized product by propagation of the
radical chain pathway (Scheme 1, dark reaction). To clarify whether
the following radical propagation is visible-light dependent or not,
we performed ‘light–dark’ experiments. One reaction mixture was
irradiated for 1 min only, resulting in 11% substrate consumption,
whereas a second mixture was irradiated for 10 min, which led to
82% substrate consumption. In the third experiment, the reaction
Acknowledgments
Financial support from the Deutsche Forschungsgemeinschaft
(GRK 1626) is gratefully acknowledged.
References
1 Zeitler K. Photoredox catalysis with visible light. Angew. Chem. Int. Ed.
2009; 48: 9785–9789.
2 Shaw MH, Twilton J, MacMillan DWC. Photoredox catalysis in organic
chemistry. J. Org. Chem. 2016; 81: 6898–6926.
3 Prier CK, Rankic DA, MacMillan DWC. Visible light photoredox catalysis
with transition metal complexes: applications in organic synthesis.
Chem. Rev. 2013; 113: 5322–5363.
4 Schultz DM, Yoon TP. Solar synthesis: prospects in visible light
photocatalysis. Science 2014; 343: 985–994.
5 Shi L, Xia WJ. Photoredox functionalization of C―H bonds adjacent to a
nitrogen atom. Chem. Soc. Rev. 2012; 41: 7687–7697.
6 Bordoni A, de Lederkremer RM, Marino C. Photoinduced electron-
transfer alpha-deoxygenation of aldonolactones. Efficient synthesis of
2-deoxy-D-arabino hexono-1,4-lactone. Carbohydr. Res. 2006; 341:
1788–1795.
7 Kachkovskyi G, Faderl C, Reiser O. Visible light-mediated synthesis of
(spiro)anellated furans. Adv. Synth. Catal. 2013; 355: 2240–2248.
8 Lackner GL, Quasdorf KW, Overman LE. Direct construction of
quaternary carbons from tertiary alcohols via photoredox-catalyzed
fragmentation of tert-alkyl N-phthalimidoyl oxalates. J. Am. Chem. Soc.
2013; 135: 15342–15345.
9 Nguyen JD, D’Amato EM, Narayanam JMR, Stephenson CRJ. Engaging
unactivated alkyl, alkenyl and aryl iodides in visible-light-mediated free
radical reactions. Nat. Chem. 2012; 4: 854–859.
10 Nguyen JD, Reiss B, Dai C, Stephenson CRJ. Batch to flow deoxygenation
using visible light photoredox catalysis. Chem. Commun. 2013; 49:
4352–4354.
Scheme 1. Proposed mechanism for the desulfurization by iridium
photocatalyst in the presence of triethylphosphite.
11 Okada K, Okamoto K, Morita N, Okubo K, Oda M. Photosensitized
decarboxylative Michael addition through N-(acyloxy)phthalimides via
J. Pept. Sci. 2017
Copyright © 2017 European Peptide Society and John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/jpepsci

M. LEE ET AL.
an electron-transfer mechanism. J. Am. Chem. Soc. 1991; 113:
9401–9402.
39 Yu C, Lee K, You Y, Cho EJ. Synthesis of 2-substituted benzothiazoles by
visible light-driven photoredox catalysis. Adv. Synth. Catal. 2013; 355:
1471–1476.
40 Zhang G, Liu C, Yi H, Meng Q, Bian C, Chen H, Jian J-X, Wu L-Z, Lei A.
External oxidant-free oxidative cross-coupling: a photoredox cobalt-
catalyzed aromatic C–H thiolation for constructing C―S bonds. J. Am.
Chem. Soc. 2015; 137: 9273–9280.
41 Tyson EL, Farney EP, Yoon TP. Visible light-promoted [2+2] radical anion
cycloadditions with cleavable redox auxiliaries. Abstr. Pap. Am. Chem. S
2012; 243.
12 Prudhomme DR, Wang ZW, Rizzo CJ. An improved photosensitizer for
the photoinduced electron-transfer deoxygenation of benzoates and
m-(trifluoromethyl)benzoates. J. Org. Chem. 1997; 62: 8257–8260.
13 Rackl D, Kais V, Kreitmeier P, Reiser O. Visible light photoredox-catalyzed
deoxygenation of alcohols. Beilstein J. Org. Chem. 2014; 10: 2157–2165.
14 Saito I, Ikehira H, Kasatani R, Watanabe M, Matsuura T. Selective
deoxygenation of secondary alcohols by photosensitized electron-
transfer reaction
– a general procedure for deoxygenation of
ribonucleosides. J. Am. Chem. Soc. 1986; 108: 3115–3117.
15 Speckmeier E, Padié C, Zeitler K. Visible light mediated reductive
cleavage of C―O bonds accessing α-substituted aryl ketones. Org.
Lett. 2015; 17: 4818–4821.
16 Christensen CH, Rass-Hansen J, Marsden CC, Taarning E, Egeblad K. The
renewable chemicals industry. ChemSusChem 2008; 1: 283–289.
17 Dodds DR, Gross RA. Chemicals from biomass. Science 2007; 318:
1250–1251.
42 Tyson EL, Niemeyer ZL, Yoon TP. Redox mediators in visible light
photocatalysis: photocatalytic radical thiol-ene additions. J. Org. Chem.
2014; 79: 1427–1436.
43 Bhat VT, Duspara PA, Seo S, Abu Bakar NSB, Greaney MF. Visible light
promoted thiol-ene reactions using titanium dioxide. Chem. Commun.
2015; 51: 4383–4385.
44 Keshari T, Yadav VK, Srivastava VP, Yadav LDS. Visible light
organophotoredox catalysis:
a general approach to beta-keto
18 Maji T, Karmakar A, Reiser O. Visible-light photoredox catalysis:
dehalogenation of vicinal dibromo-, α-halo-, and α,α-dibromocarbonyl
compounds. J. Org. Chem. 2011; 76: 736–739.
19 Dawson PE, Muir TW, Clarklewis I, Kent SBH. Synthesis of proteins by
native chemical ligation. Science 1994; 266: 776–779.
sulfoxidation of alkenes. Green Chem. 2014; 16: 3986–3992.
45 Keylor MH, Park JE, Wallentin C-J, Stephenson CRJ. Photocatalytic
initiation of thiol–ene reactions: synthesis of thiomorpholin-3-ones.
Tetrahedron 2014; 70: 4264–4269.
46 Ross HB, Boldaji M, Rillema DP, Blanton CB, White RP. Photosubstitution
in tris chelate complexes of ruthenium(II) containing the ligands 2,20-
bipyrazine, 2,20-bipyrimidine, 2,20-bipyridine, and 4,40-dimethyl-2,20-
bipyridine – energy-gap control. Inorg. Chem. 1989; 28: 1013–1021.
47 Scoppola E, Sodo A, McLain SE, Ricci MA, Bruni F. Water-peptide site-
specific interactions: a structural study on the hydration of glutathione.
Biophys. J. 2014; 106: 1701–1709.
48 Pagire SK, Paria S, Reiser O. Synthesis of β-hydroxysulfones from sulfonyl
chlorides and alkenes utilizing visible light photocatalytic sequences.
Org. Lett. 2016; 18: 2106–2109.
49 Rackl D, Kreitmeier P, Reiser O. Synthesis of a polyisobutylene-tagged
fac-Ir(ppy)3 complex and its application as recyclable visible-light
photocatalyst in a continuous flow process. Green Chem. 2016; 18:
214–219.
20 Kent SBH. Total chemical synthesis of proteins. Chem. Soc. Rev. 2009; 38:
338–351.
21 Ma JM, Zeng J, Wan Q. Postligation–desulfurization: a general approach
for chemical protein synthesis. Top. Curr. Chem. 2015; 363: 57–101.
22 Monbaliu JCM, Katritzky AR. Recent trends in Cys- and Ser/Thr-based
synthetic strategies for the elaboration of peptide constructs. Chem.
Commun. 2012; 48: 11601–11622.
23 Wu B, Hua ZH, Warren JD, Ranganathan K, Wan Q, Chen G, Tan ZP,
Chen JH, Endo A, Danishefsky SJ. Synthesis of the fucosylated
biantennary N-glycan of erythropoietin. Tetrahedron Lett. 2006; 47:
5577–5579.
24 Wong CTT, Tung CL, Li XC. Synthetic cysteine surrogates used in native
chemical ligation. Mol. BioSyst. 2013; 9: 826–833.
25 Thapa P, Zhang RY, Menon V, Bingham JP. Native chemical ligation: a
boon to peptide chemistry. Molecules 2014; 19: 14461–14483.
26 Malins LR, Payne RJ. Recent extensions to native chemical ligation for the
chemical synthesis of peptides and proteins. Curr. Opin. Chem. Biol. 2014;
22: 70–78.
27 Hoffmann FW, Ess RJ, Simmons TC, Hanzel RS. The desulfurization of
mercaptans with trialkyl phosphites. J. Am. Chem. Soc. 1956; 78:
6414–6414.
50 Liu H, Gaza-Bulseco G, Zhou L. Mass spectrometry analysis of photo-
induced methionine oxidation of a recombinant human monoclonal
antibody. J. Am. Soc. Mass Spectrom. 2009; 20: 525–528.
51 Neveselý T, Svobodová E, Chudoba J, Sikorski M, Cibulka R. Efficient
metal-free aerobic photooxidation of sulfides to sulfoxides mediated
by a vitamin B2 derivative and visible light. Adv. Synth. Catal. 2016;
358: 1654–1663.
52 Burns JA, Butler JC, Moran J, Whitesides GM. Selective reduction of
disulfides by tris(2-carboxyethyl)phosphine. J. Org. Chem. 1991; 56:
2648–2650.
28 Walling C, Rabinowitz R. The reaction of thiyl radicals with trialkyl
phosphites. J. Am. Chem. Soc. 1957; 79: 5326–5326.
29 Walling C, Rabinowitz R. The reaction of trialkyl phosphites with thiyl and
alkoxy radicals. J. Am. Chem. Soc. 1959; 81: 1243–1249.
30 Cuesta J, Arsequell G, Valencia G, Gonzalez A. Photochemical
desulfurization of thiols and disulfides. Tetrahedron-Asymmetry 1999;
10: 2643–2646.
53 Bielski BHJ, Arudi RL, Sutherland MW. A study of the reactivity of HO2/O2ꢀ
with unsaturated fatty-acids. J. Biol. Chem. 1983; 258: 4759–4761.
54 Corey EJ, Mehrotra MM, Khan AU. Water induced dismutation of
superoxide anion generates singlet molecular-oxygen. Biochem.
Biophis. Res. Commun. 1987; 145: 842–846.
31 Gonzalez A, Valencia G. Photochemical desulfurization of L-cysteine
derivatives. Tetrahedron-Asymmetry 1998; 9: 2761–2764.
32 Pentelute BL, Kent SBH. Selective desulfurization of cysteine in the
presence of Cys(Acm) in polypeptides obtained by native chemical
ligation. Org. Lett. 2007; 9: 687–690.
55 Cismesia MA, Yoon TP. Characterizing chain processes in visible light
photoredox catalysis. Chem. Sci. 2015; 6: 5426–5434.
56 Karkas MD, Matsuura BS, Stephenson CRJ. Enchained by visible light-
mediated photoredox catalysis. Science 2015; 349: 1285–1286.
33 Yang YY, Ficht S, Brik A, Wong CH. Sugar-assisted ligation in glycoprotein
synthesis. J. Am. Chem. Soc. 2007; 129: 7690–7701.
Supporting information
34 Wan Q, Danishefsky SJ. Free-radical-based, specific desulfurization of
cysteine: a powerful advance in the synthesis of polypeptides and
glycopolypeptides. Angew. Chem. Int. Ed. 2007; 46: 9248–9252.
35 Wang P, Dong SW, Brailsford JA, Iyer K, Townsend SD, Zhang Q,
Hendrickson RC, Shieh J, Moore MAS, Danishefsky SJ. At last:
erythropoietin as a single glycoform. Angew. Chem. Int. Ed. 2012; 51:
11576–11584.
36 Wang P, Dong SW, Shieh JH, Peguero E, Hendrickson R, Moore MAS,
Danishefsky SJ. Erythropoietin derived by chemical synthesis. Science
2013; 342: 1357–1360.
Additional Supporting Information may be found online in the
supporting information tab for this article.
Scheme S1. Synthesis of the Ir(dF(CF3)ppy)2(dtb-bpy)PF6 (C6 = E)
photocatalyst.
Scheme S2. Desulfurization of thiols in amino acids and
glutathione.
37 Gao XF, Du JJ, Liu Z, Guo J. Visible-light-induced specific desulfurization
of cysteinyl peptide and glycopeptide in aqueous solution. Org. Lett.
2016; 18: 1166–1169.
38 Cheng Y, Yang J, Qu Y, Li P. Aerobic visible-light photoredox radical C–H
functionalization: catalytic synthesis of 2-substituted benzothiazoles.
Org. Lett. 2012; 14: 98–101.
Scheme S3. Desulfurization of disulfide compounds.
Figure S1. Analytical HPLC of the purified peptide Ac-
YNMNDCYSKL-NH2.
Figure S2. LC-HRMS (ESI) of the purified peptide Ac-YNMNDCYSKL-
NH2. m/z calculated for C55H83N14O18S2 [M + H]+ 1291.5373, found
wileyonlinelibrary.com/journal/jpepsci
Copyright © 2017 European Peptide Society and John Wiley & Sons, Ltd.
J. Pept. Sci. 2017

PHOTOCATALYZED CYSTEINE DESULFURIZATION
1291.5444; m/z calculated for C55H84N14O18S2 [M
m/z = 646.2687, found 646.2769.
+
2H]2+
Ac-YNMNDCYSKL-NH2 with P(OEt)3 (Figure S3A) and lyophilization.
The purity of the product is 94% (yield of the desulfurization
reaction after purification: 12%). m/z calculated for C55H83N14O19S
[M + H]+ 1275.57, found 1275.72.
Figure S5. Visible light-mediated desulfurization of glutathione
(GSH) at different pH values. GSH (0.25 mmol), 5 mol% catalyst B
(0.0125 mmol), P(OEt)3 (0.75 mmol), THF (1.5 ml) and phosphate
buffer (1.0 M) (1.5 ml), 19-h reaction time at room temperature. B:
Ru(bpy)3Cl2. The GSH conversion was determined by 1H-NMR
analysis based on ratio of starting material and product.
Figure S3. (A) HPLC of the desulfurization mixture of Ac-
YNMNDCYSKL-NH2 with P(OEt)3 after 1-h reaction. (B) HPLC of the
desulfurization mixture of Ac-YNMNDCYSKL-NH2 with TCEP after
1-h reaction. (C) Overlay of the HPLC profiles of the cysteine-
containing peptide Ac-YNMNDCYSKL-NH2 and of the two
desulfurization mixtures.
Figure S4. (A) HPLC and (B) MALDI-TOF-MS of the Ala-peptide with
Met(O) after HPLC separation from the desulfurization mixture of
J. Pept. Sci. 2017
Copyright © 2017 European Peptide Society and John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/jpepsci