Visible Light-Mediated Functionalization of Selenocystine-Containing Peptides

Article abstract of DOI:10.1002/adsc.202100373

A straightforward and atom-economic method for the functionalization of short selenocystine-containing peptides is presented. This method is shown to be tolerant to unprotected peptides. The detailed protocol is based on the generation of a selenium radical via visible light-initiated reaction in the presence of transition metal-free photocatalyst. The selenium radical is further oxidized to an electrophile and trapped by N-heterocycles. The mechanism is confirmed by NMR, HRMS, UV, EPR and cyclic voltammetry (CV) experiments and photocatalyst emission quenching studies. A visible light-initiated reaction is employed for the synthesis of selenocysteine-containing indole-based macrocycles via intramolecular Se?C bond formation. (Figure presented.).

Full text of DOI:10.1002/adsc.202100373

RESEARCH ARTICLE
DOI: 10.1002/adsc.202100373
Visible Light-Mediated Functionalization of Selenocystine-
Containing Peptides
Sindija Lapcinska,a Pavels Dimitrijevs,a Linards Lapcinskis,b and
Pavel Arsenyana,*
a
Latvian Institute of Organic Synthesis, Aizkraukles 21, LV-1006, Riga, Latvia
E-mail: pavel@osi.lv
b
Research Laboratory of Functional Materials Technologies Faculty of Materials Science and Applied Chemistry, Riga Technical
University, P. Valdena 3/7, LV-1048, Riga, Latvia
Manuscript received: April 29, 2021; Revised manuscript received: April 29, 2021;
Version of record online: ■■■, ■■■■
Supporting information for this article is available on the WWW under https://doi.org/10.1002/adsc.202100373
Abstract: A straightforward and atom-economic method for the functionalization of short selenocystine-
containing peptides is presented. This method is shown to be tolerant to unprotected peptides. The detailed
protocol is based on the generation of a selenium radical via visible light-initiated reaction in the presence of
transition metal-free photocatalyst. The selenium radical is further oxidized to an electrophile and trapped by
N-heterocycles. The mechanism is confirmed by NMR, HRMS, UV, EPR and cyclic voltammetry (CV)
experiments and photocatalyst emission quenching studies. A visible light-initiated reaction is employed for
the synthesis of selenocysteine-containing indole-based macrocycles via intramolecular SeÀ C bond formation.
Keywords: Macrocycle; N-heterocycle; peptide; photocatalysis; selenocysteine
Introduction
arenes[25] has been established using LiCl as an additive
and a household white LED lamp (λ=430–730 nm) as
Visible light photocatalysis is a rapidly emerging field the light source. The abovementioned methods were
with attractive advantages such as efficiency, sustain- limited solely to the use of diorganyl diselenides,
ability, atom economy, and selectivity.[1–3] Most organic whereas photocatalyst- and additive-free methods for
molecules do not absorb visible light; therefore, the the synthesis of 3-selenyl-, 3-tellanyl-, 3-sulfenyl- and
presence of photosensitizers such as transition metal 3-thiocyanoindoles were achieved by employing 26 W
complexes,[4–6] organic dyes[7] or semiconductors[8] CFL bulbs or sunlight, resulting in similar reaction
facilitates the reaction.
yields.[26] Alternatively, 3-sulfenyl and 3-selenylindoles
A vast array of methods[9] for synthesis of 3- can be obtained by visible light-induced chalcogenyla-
selenylindoles have been demonstrated employing tion of indolines in the presence of graphene oxide.[27]
metal salts (e.g. FeCl3,[10] CuI[11]), bases[12,13] or
An indole ring was formed via 5-endo-dig cycliza-
oxidants.[14–16] More sustainable methods have been tion starting from alkynylanilines and diaryl disele-
developed based on electrochemically-induced nides or disulfides in the presence of H2O2 and blue
process.[17,18]
LED light.[28] A simple method for the construction of
In the last few years, visible light-mediated CÀ H various heterocycles, e.g., oxazoline, isoxazoline,
functionalization of (hetero)arenes with simple, mainly pyrrolidine, and lactone, was established using dio-
diaryl diselenides has been reported. Two catalysts rganyl diselenides and alkene-containing substrates
outperform others in terms of efficiency of selenylation and employing 4-CzIPN and blue LED light to initiate
of indoles and other (hetero)arenes using blue LED the reaction.[29] Catalyst-free synthesis of arylselanyl-
light: FIrPic[19] and Rose Bengal[20,21] (RB). Notably, and arylsulfenyl-3,3-difluoro-γ-lactams was demon-
photocatalyst-free selenylation of indole can be per- strated in the presence of a base (KH2PO4).[30] Notably,
formed by prolonged irradiation in ethanol[22] or in functionalization of styrenes was demonstrated using
methanol under flow conditions.[23] Another green diaryl diselenides in the presence of RB by irradiating
method for selenylation of allenes[24] and (hetero) the mixture with blue LEDs.[31] Visible light-initiated
Adv. Synth. Catal. 2021, 363, 1–12
1
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

RESEARCH ARTICLE
asc.wiley-vch.de
generation of selenyl radicals with subsequent addition observed, although ethyl eosin was capable of inducing
to terminal[32,33] or internal[34] alkynes has also been the formation of 3a (22%). Gratifyingly, RB and
investigated. Significantly, catalyst-free conditions can erythrosin B induced a full conversion of the starting
be used to realize both visible light-induced metathesis materials and selective synthesis of 3-selanyl indole 3a
reactions between diselenides and ditellurides[35] and in just 90 min.
diselenide metathesis between simple diorganyl
The following catalysts were found to be unsuit-
diselenides[36] or SeÀ Se bond-containing peptides.[37] able: 5-carboxytetramethylrhodamine (5-TAMRA),
Strikingly, there has been no profound study on nickel tetraphenyl porphyrin, 4-CzIPN, cresol red,
selenocysteine (Sec)-containing peptide modification chlorophenol red, bromocresol green, methyl orange,
under visible light conditions.
congo red, direct red 81, direct yellow 27, methylene
Late-stage modification of peptides is not an easy blue, basic fuchsine, indigo carmine, alcian blue, 2,4,6-
task; however, photocatalysis can provide a route to triphenylpyrylium tetrafluoroborate, 9-mesityl-10-
achieve chemoselective bioconjugation under mild and methyl acridinium tetrafluoroborate, acridine, and N-
often biocompatible conditions.[38–41] However, other methyl-acridinium iodide. In most cases, the formation
methods of Sec-containing peptide functionalization of 3a was observed; however, the reaction was non-
have been successfully employed, and common mod- selective, and the conversion of starting materials was
ifications of Sec in selenoproteins have been recently less than 50%. Thus, RB was selected as the most
reviewed.[42,43] UV light (254 nm) irradiation of seleno- suitable catalyst. The reduction in the catalyst load to
cystine-based peptides converts Se–Se bridges to 2 mol% resulted in a prolonged reaction time and
selenolanthionine fragments.[44] Deselenylation of Sec- decrease in yield. Unfortunately, the use of greener
containing peptides can be effectively achieved by protic solvents such as MeOH, EtOH, EtOH/H2O, and
reduction of Sec to alanine with TCEP/DTT[45] or iPrOH resulted in a nonselective reaction due to the
oxidation of Sec to dehydroalanine derivatives with fast oxidation and deselenylation of 1a. Changing the
hydrogen peroxide.[46] Notably, an elegant method has light source to a CFL bulb resulted in a significantly
been established for Sec-containing peptides and small slower reaction, whereas a red LED was incapable of
molecule conjugation based on the electrophilic char- initiating the formation of 3a. Control tests showed
acter of (5-nitropyridylthio)-Sec-containing peptides.[47] that the reaction did not occur under daylight or dark
Recently, Sec-containing peptide modification through conditions; thus, the necessity of the LED460 light was
the generation of selenyl electrophiles using weak confirmed. The reaction under an Ar atmosphere did
Lewis acids or oxidants has been demonstrated.[48,49]
not differ from the reaction performed in an open flask,
Here, we report our findings regarding SeÀ Se bond- whereas the reaction in the presence of 4-amino-
containing peptide modification using visible light- TEMPO did not result in the formation of 3a,
initiated reactions, scope and limitation studies, for- confirming the radical mechanism for this process. In
mation of macrocyclic Sec-containing peptides, and summary, the optimized conditions for the synthesis of
mechanistic studies.
3-Sec-indoles were found to be 5 mol% RB, MeCN
and blue LED light for 90 min.
To determine how the substituents in the indole ring
affect the reaction, we tested the reaction of 1a with
Results and Discussion
Dipeptide dimers Boc-Sec-Gly-OBn 1a and 1H-indole various indoles. Notably, the presence of an electron-
(2a) were chosen as model substrates to optimize the donating group (EDG) at the C5 position of indole
reaction conditions. Preliminary screening of a photo- improved the reaction yield; halogen atoms did not
catalyst panel (more than 25, ESI Figure S1) was significantly affect the process, whereas the presence
performed in acetonitrile using 0.5 equiv. 1a, 1 equiv. of an electron-withdrawing group (EWG) diminished
2a, 2 mol% transition metal catalyst or 5 mol% the reactivity and provided only trace amounts of the
organic dye and irradiating the reaction mixture with products. An EDG at the C2 position provided the
blue LED light (max 461 nm, bright blue, x=01440, product in lower yield, but electron-deficient indoles
y=0.0395, >50 000 lx) for 90 min. The absence of a (EWG at the N1 or C2 position) were completely
photocatalyst resulted in no reaction and only a trace unreactive. Tripeptide dimer 1b showed an even better
amount of desired product 3a (Scheme 1). Evaluation ability to react with indoles than 1a, and the
of the obtained data led to the conclusion that the corresponding products 3k–n and 3p were obtained in
popular transition metal catalysts FIrPic and Ru- excellent yields. Importantly, a hydroxy group at the
(bpy)3Cl2 were nonselective and provided a mixture of C5 position of indole was also tolerated under the
products, including 3a. However, a series of fluores- reaction conditions, although the reaction yield was
cein derivatives, which are well-known organic dyes, lower because prolonged irradiation was required for
were more promising. Fluorescein itself has very low full conversion of the starting materials, whereas the
solubility in MeCN; thus, DMF was added. Unfortu- reaction with 5-aminoindole failed, probably due to
nately, the formation of the desired product was not enamine-imino tautomerism. However, the use of tert-
Adv. Synth. Catal. 2021, 363, 1–12
2
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

RESEARCH ARTICLE
asc.wiley-vch.de
Scheme 1. Sec-indole formation: scope and limitation studies. Reaction conditions: indole 2 (1 equiv.), 1 (0.5 equiv.), RB
(0.05 equiv.), blue LED460, MeCN. 1a (Boc-Sec-Gly-OBn)2, 1b (Boc-Glu(OtBu)-Sec-Gly-OBn)2, 1c (Boc-Sec)2, 1d (H2N-Sec-Gly-
OBn·TFA)2, 1e (H2N-Glu(OH)-Sec-Gly-OBn·TFA)2, 1f (H2N-Sec-Lys-Arg-Phe-OPEG3OMe·TFA)2, 1g (H2N-Sec-His-Phe-
OPEG3OMe·TFA)2, 1h (H2N-Sec-Tyr-Phe-OPEG3OMe·TFA)2, 1i (H2N-Sec-Trp-Phe-OPEG3OMe·TFA)2; 2a 1H-indole, 2b 5-
(benzyloxy)indole, 2c 5-bromoindole, 2d 5-cyanoindole, 2e 2-methylindole, 2f 1-methyl-2-phenylindole, 2g 1-Boc-2-phenyl-
indole, 2h 1H-indole-2-carboxylic acid, 2i ethyl 5-chloro-1H-indole-2-carboxylate, 2j 5-hydroxyindole, 2k 5-aminoindole, 2l tert-
butyl (1H-indol-5-yl)carbamate, 2m 5-((tert-butyldimethylsilyl)oxy)-1H-indole, 2n 2-(trimethylsilyl)ethyl (1H-indol-5-yl)
carbamate, 2o pindolol.
butyl (1H-indol-5-yl)carbamate 2l resulted in the pindolol – a nonselective β-adrenergic antagonist[50]
formation of desired product 3o in high yield. was successful.
–
Remarkably, we discovered that Boc-selenocystine 1c
Next, unprotected Sec-containing peptides were
can also be employed in visible light-mediated applied as starting materials to clearly facilitate the use
selenylation, leading to a selenium analog of trypto- of the developed protocol. Consequently, utilization of
phan 3p as well as products 3q–s, which can serve as unprotected dipeptide dimer 1d and tripeptide dimer
valuable building blocks. Notably, selenylation of 1e led to the formation of desired products 4a–e.
Encouraged by these results, we decided to test more
Adv. Synth. Catal. 2021, 363, 1–12
3
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

RESEARCH ARTICLE
asc.wiley-vch.de
sophisticated Sec-containing peptides with “sensitive” unambiguously confirmed by X-ray analysis
amino acid residues to determine their tolerance under (Figure 1A and 1B).
the developed reaction conditions. Thus, pegylated
tetra- and tripeptide dimers 1f–i were prepared and protocol can be applied for intramolecular indole
employed in visible light-mediated reactions. selenylation to form Sec-containing peptides with
Next, we were interested in whether the developed
We successfully obtained products 4f–h, confirm- indole-embedded macrocycles. Furthermore, three
ing that the Lys, Arg, His and Tyr moieties are well strategies (Figure 2) were proposed for the formation
tolerated, whereas the reaction with Trp-containing of macrocycles: (A) visible light-mediated SeÀ C
peptide 1i was nonselective.
formation or, in other words, intramolecular indole
To extend the method’s scope, the possibilities for selenylation; (B) selenylation of an indole attached to
selenylation of indole ring bioisosteres – azaindoles 5 the peptide and subsequent intramolecular amide bond
– were evaluated. These substrates proved to be formation; and (C) a reaction between Boc-Sec and
significantly less reactive than 1H-indole. The forma- protected 5-hydroxy- or aminoindole, coupling with a
tion of product 6a (8%), along with side products, was small peptide, deprotection and intramolecular amide
observed when 7-azaindole was used; only trace bond formation.
amounts of product 6b were formed when 4-azaindole
First, we intended to prepare Boc-Sec-containing
was used, whereas 5-azaindole did not lead to product dipeptide and tripeptide dimers (8,9) attached to the C4
6c. Azaindoles[51] are electron-deficient heterocycles or C5 position of indole through an ester or amide
and are thus less reactive than indoles; moreover, these bond. Successive use of 8 and 9 for visible light-
substrates might become deactivated due to excited- mediated intramolecular SeÀ C bond formation (Ap-
state tautomerization.[52]
proach A) could result in the synthesis of macrocyclic
However, protonation of azaindoles changes their structures 10 (Scheme 3). 1-Methyl-1H-indol-4-ol (2p)
electronic properties and promotes the formation of 3- was coupled with Boc-Phe and then deprotected,
Sec-azaindoles 6a–c (Scheme 2). We observed signifi- affording 7a, which was next coupled with Boc-Sec,
cant differences in the reaction rate for different yielding 8a. Gratifyingly, substrate 8a in the presence
isomers: 1.5 h was needed for 7-azaindole and 24 h of RB under visible light irradiation provided macro-
was needed for 5-azaindole to achieve 100% con- cycle 10a in 60% yield. The intramolecular selenyla-
version.
These results correlate with the pKa of azaindoles:
tion proceeded slightly slower than the intermolecular
7-azaindole has the highest acidity, whereas 5-azain-
dole has the highest basicity.[53] Notably, protonated
imidazo[1,2-a]pyridine and imidazo[1,2-a]pyrimidine
were also successfully employed in visible light-
induced selenylation, and the structures of products 6d
(CCDC 2054758) and 6e (CCDC 2054757) were
Figure 1. ORTEP molecular structures of 6d (A), 6e (B), 10a
(C), 10f (D).
Scheme 2. Synthesis of Sec-azaindoles. Reaction conditions:
azaindole
5 (1 equiv.), 1c (0.5 equiv.), RB (0.05 equiv.),
LED460, MeCN.
Figure 2. Proposed strategies for Sec-macrocycle formation.
Adv. Synth. Catal. 2021, 363, 1–12
4
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

RESEARCH ARTICLE
asc.wiley-vch.de
Scheme 3. Synthesis of macrocycles: approach A. Reaction conditions: (a) 8 or 9 (1 equiv.), RB (0.1 equiv.), LED460, MeCN.
*Fmoc cleavage, cyclization, **cyclization, Fmoc cleavage.
reaction, but it was selective. Furthermore, the strategy was unsuitable for the synthesis of unprotected
structure of 10a (CCDC 2054762) was unambiguously macrocycles due to the instability of indoles under
confirmed by X-ray analysis (Figure 1C). Macrocycle acidic conditions; after protonation, they formed
with N(ɛ)-protected lysine 10b was successfully dimers and trimers. However, the preparation of
synthesized, as was the tryptophan moiety containing unprotected macrocycle 10j was accomplished by
macrocycle 10c, although the yield was slightly lower employing a Fmoc protection strategy. Irradiation of
due to the formation of side products. Next, we tested 8g with LED460 light in the presence of RB resulted in
whether a Sec-containing peptide derivative with 1H- the formation of the corresponding macrocycle, and
indole 8d can be used to prepare macrocycles. subsequent cleavage of the Fmoc group yielded
Fortunately, the desired products were obtained starting unprotected macrocycle 10j. The same product was
from 4-hydroxyindole (2q) and 4-amino-1H-indole also obtained by deprotection of 8g followed by a
(2r). Compounds 10a–e exhibited relatively low macrocyclization reaction.
solubility; therefore, PEGylated glutamic acid was
In this case, 10j was obtained in lower yield;
employed to resolve this issue. The molecular structure therefore, the advisable sequence for the preparation of
of 10f (CCDC 2054761) was also confirmed by X-ray unprotected Sec macrocycles is macrocyclization and
analysis (Figure 1D). Next, we investigated the prepa- then deprotection.
ration of macrocycles that contained tripeptides
Next, the synthesis of Sec macrocycles with a short
attached to the indole. Product 10g containing a Boc- peptide at the C5 of indole was evaluated. For this
Sec-Lys-Phe moiety was successfully obtained, where- purpose, substrate 8h was irradiated with LED460 light.
as Arg-containing substrate 9b was insoluble in Unfortunately, the formation of product 10k was not
MeCN; consequently, the reaction did not occur, but observed. However, we confirmed that intermolecular
the addition of DMF resulted in a nonselective reaction selenylation can be performed using 8h. The addition
that yielded only trace amounts of the product 10h. of 1H-indole to the reaction mixture of 8h and RB
Fortunately, the introduction of tyrosine (substrate 9c) resulted in the formation of indole derivative 10l,
was well tolerated under the reaction conditions, along with many side products. The results led us to
resulting in the successful isolation of Tyr-containing conclusion that strategy A is not suitable for macro-
macrocycle 10i. Unfortunately, the Boc protection cycle formation starting from indoles with peptides
Adv. Synth. Catal. 2021, 363, 1–12
5
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

RESEARCH ARTICLE
asc.wiley-vch.de
attached to its C5 position, probably due to conforma- unstable product 11b. However, indole 7j was
tional restrictions. successfully selenylated, and product 11c was further
Approach B to prepare Sec-containing macrocycles used for intramolecular amide bond formation. Surpris-
relied on visible light-mediated selenylation of indoles ingly, only bis-macrocyclization product 13b was
7 that contained the amino acid at the C4 or C5 formed in the reaction.
position of the indole ring and subsequent intra-
We conceded that if a small peptide instead of a
molecular peptide bond formation using standard single amino acid residue was attached to the C5
coupling conditions (Scheme 4). We were keen to position of indole, it would allow the formation of a
investigate whether this synthesis pathway could lead macrocycle rather than a bis-macrocyclization product.
to the formation of macrocycles in which amino acids Thus, selenylation of indole 7l with a Phe-Lys-Phe
are attached to the C5 position of indole due to the moiety resulted in formation of 11d, and subsequent
failed attempt to prepare these macrocycles through intramolecular amide bond formation provided bis-
approach A.
First, a mixture of Boc-Sec and indole 7h with
macrocyclization product 13c as a single product.
The third approach (Approach C) involved the use
PEGylated Glu at the C4 position was irradiated in the of products 3s and 3t that were obtained by visible
presence of RB. The synthesis resulted in isolation of light-mediated selenylation of protected 5-hydroxy-
Boc-Sec-containing indole 11a in good yield. Next, a and 5-aminoindoles and proved to be excellent build-
routine EDC/HOBt method was used for intramolecu- ing blocks for macrocycle formation (Scheme 5).
lar amide bond formation. Macrocycle 12a was Product 3s was coupled with dipeptide 14a and treated
isolated in moderate yield along with bis-macrocycli- with TBAF to obtain the corresponding indole-contain-
zation product 13a as the minor product. Then, we ing peptide 15a. Subsequent intramolecular amide
moved on to selenylation of indoles that contained Phe bond formation resulted in a mixture of compounds,
at the C5 position of indole. We observed that the including bis-macrocyclization product 16. Apparently,
attachment of Phe to the C5 position of indole through this product was also unstable, similar to the other
ester bonds and the subsequent reaction with 1c led to indoles that contained amino acids attached at the C5
Scheme 4. Synthesis of macrocycles: approach B. (a) indole 7 (1 equiv.), 1c (0.5 equiv.), RB (0.05 equiv.), LED460, MeCN; (b) 11
(1 equiv.), HOBt (1 equiv.), EDC (1.5 equiv.), DMF.
Adv. Synth. Catal. 2021, 363, 1–12
6
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

RESEARCH ARTICLE
asc.wiley-vch.de
Scheme 5. Synthesis of macrocycles: approach C. Reaction conditions: (a) 3s or 3t (1 equiv.), 14 (1 equiv.), HOBt (0.5 equiv.),
EDC (1.5 equiv.), DMF; (b) TBAF (3 equiv.), THF; (c) 15 (1 equiv.), HOBt (1 equiv.), EDC (1.5 equiv.), DMF.
position of the indole through an ester bond. Fortu- (bpy)3Cl2 also resulted in the formation of 20. Other
nately, stable macrocycles 17a and 17b were isolated catalysts, namely, FIrPic, ethyl eosin, and 4-CzIPN,
by employing compounds 15b and 15c for intra- induced the formation of 18 and 20 as well but were
molecular amide bond formation.
not able to achieve full consumption of the starting
Mechanistic studies. To gain insight into visible material. 2,4,6 Triphenyl-pyrylium tetrafluoroborate
light-mediated selenylation, we first examined the and 9-mesityl-10-methyl-acridinium tetrafluoroborate
possible degradation of Sec-containing peptide 1a provided a less selective reaction, whereas DDQ and
under irradiation. Preliminary tests were performed in TAMRA were not efficient. Reactions performed in
acetonitrile, and a solution of 1a was irradiated with MeOH, EtOH or DMF did not reach full conversion of
LED460 light for 90 min with and without a photo- 1a in the given time.
catalyst. No changes were observed in the absence of
the photocatalyst, whereas the formation of two formation of 20; however, the reaction performed in
products was detected in the presence of RB. dry and degassed MeCN under an Ar atmosphere
The addition of water did not interfere with the
Analysis of the LC/MS data showed that the provided a considerably lower conversion of 1a, thus
products were seleninic acid 18 and a dehydroalanine confirming that the presence of water and oxygen in
(Dha) derivative 20, respectively (Scheme 6). On the the solvent is necessary for the reaction to occur.
basis of 77Se NMR spectroscopy data[54] (Figure 3A), Under irradiation conditions, oxygen is converted to
*
1
À
alkyl seleninic acid 18 (RSeO2H 1217.5 ppm, [M+ short-living O2, returning RB to ground state by the
Na]=471.0633) was formed after 1 h of irradiation. energy transfer.[55–58] Singlet oxygen reacts with water
Our attempts to isolate oxidized form of 1a failed due generating hydroperoxyl radical that produces hydro-
to product lability. Furthermore, storage of an NMR gen peroxide,[22] which is trapped by 1a, resulting in
tube for 24 h led to the disappearance of the 18 signal, oxidation and deselenylation with the formation of a
confirming the formation of seleninic acid (H2SeO3 19) double bond. The control reaction with TEMPO did
(1302.7 ppm) and Dha-peptide 20. The use of Ru- not lead to the formation of 20, thus verifying the
Adv. Synth. Catal. 2021, 363, 1–12
7
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

RESEARCH ARTICLE
asc.wiley-vch.de
Scheme 6. Proposed mechanism of visible light mediated reaction.
initial radical pathway. On the basis of previously oxidized with oxygen to selenyl electrophile II,[19]
published reports,[19–22,25,26,59] due to differences in which is further trapped by 2a, resulting in the
reaction conditions, it was problematic to unambigu- formation of 3a (Scheme 6). Utilization of Ru-
ously specify whether the selenyl radical I attacks the (bpy)3Cl2 ·6H2O and FIrPic leads to the formation of
indole at C3 position, or the indolyl radical is formed both selenyl and indolyl radicals, and consequently, an
first or both selenyl and indolyl radicals are formed unspecific reaction with the formation of undesired
simultaneously. To answer this question, the following byproducts occurs.
experiments were performed.
EPR studies. To confirm directly the generation of
Optical absorption properties. UV spectra were selenyl radical, EPR studies were conducted. We
recorded for 1a, 2a, RB, ethyl eosin, erythrosin B, examined three solutions – RB (50 μM), 1a (50 μM),
FIrPic, Ru(bpy)3Cl2 ·6H2O, and 4-CzIPN in dry mixture of RB and 1a (1:5, 50 μM) in acetonitrile/
acetonitrile solutions (Figure 3B and 3 C). The photo- water (9:1) under LED460 light irradiation. Thus, in the
chemical reaction between 2a and diselenide 1a was absence of light source, no radical formation was
not effective in the absence of a catalyst under LED460 detected in the case of RB, 1a and the mixture of both
light because indole has an absorption band from 200 (Figure 3G). However, radical formation was con-
to 305 nm and 1a exhibits absorption until 430 nm, firmed during irradiation of RB in solution (g=2.008,
albeit with low intensity. Notably, the absorption ΔHpp =6G). The amount of RB* is constant in time.
shoulder at 275–430 nm in the 1a UV spectrum is Notably, switching off the light led to disappearance of
characteristic of SeÀ Se bonds, which facilitate the RB radical. Received data are in agreement with
formation of selenyl radicals in the presence of a already published research.58 RB* is not stable and
*
*
À
photocatalyst.
recombines to RB*+RB!RB + +RB . Irradiation
Photocatalyst emission quenching. The photo- of 1a without photocatalyst did not produce a selenyl
luminescence quenching of RB, Ru(bpy)3Cl2 ·6H2O radical (Figure 3H). However, the presence of disele-
and FIrPic was performed using 1a or 2a in degassed nide 1a during irradiation, reduced the intensity of
acetonitrile. The quenching rate constant of RB in the RB* signal twice allowing to confirm the establish-
experiment with 1a was determined to be 10.42× ment of dynamic equilibrium under LED460 light: RB*
$
10À 3 l/mol (Figure 3D). In contrast, 2a does not +1a (RB+1a*).
quench the fluorescence of RB to any significant
Cyclic voltammetry (CV) studies. To investigate
extent. Notably, quenching experiments with Ru- the redox behavior of 1a and N-heterocycles, as well
(bpy)3Cl2 ·6H2O and FIrPic showed the opposite as the photocatalysts (RB, FIrPic, Ru(bpy)3Cl2 ·6H2O),
pattern: the 2a-quenched fluorescence of both catalysts CV studies were performed in dry degassed
had a higher Stern-Volmer constant than 1a (2a: 5.2× acetonitrile (details are presented in Table S1 and
10À 3 l/mol, 1a: 3.6×10À 3 l/mol of Ru(bpy)3Cl2 ·6H2O Figure S2).
(Figure 3E); 2a: 0.9×10À 3 l/mol, 1a: 0.5×10À 3 l/mol
The onset oxidation potentials (Eox.vsFc/Fc*onset) versus
of FIrPic) (Figure 3F). Based on the obtained results, it Fc were measured by CV from the first redox cycle.
can be concluded that upon excitation of RB, only The CV of 2a shows Eox.vsFc/Fc*onset =0.43 V, that of
selenyl radical is formed. Then, selenyl radical I is dipeptide 1a shows Eox.vsFc/Fc*onset =0.76 V, that of RB
Adv. Synth. Catal. 2021, 363, 1–12
8
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

RESEARCH ARTICLE
asc.wiley-vch.de
Figure 3. (A) 77Se NMR spectrum of 1a degradation products after 1 h of irradiation and after storage of the reaction mixture in
NMR tube for 24 h; (B) UV spectra of 1a, 2a and photocatalysts; (C) Magnified UV spectra of 250–500 nm region (D)
Photoluminescence quenching of RB with 1a and 2a; (E) Photoluminescence quenching of Ru(bpy)3Cl2 ·6H2O with 1a and 2a; (F)
Photoluminescence quenching of FIrPic with 1a and 2a; (G, H) EPR spectra of RB and RB + 1a with and without LED460
irradiation.
Adv. Synth. Catal. 2021, 363, 1–12
9
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

RESEARCH ARTICLE
asc.wiley-vch.de
shows Eox.vsFc/Fc*onset =À 0.20 V and Ered.vsFc/Fc*
=
purified by reverse phase flash chromatography (C-18, MeCN/
H2O, 10–85%) to give the product 3.
onset
À 1.39 V, that of FIrPic shows Eox.vsFc/Fc*onset =0.73 V
and Ered.vsFc/Fc*onset =0.94 V, and that of Ru-
(bpy)3Cl2 ·6H2O shows Eox.vsFc/Fc*onset =0.75 V and
Ered.vsFc/Fc*onset =À 1.42 V. Our experimental data are
similar to those in a previously published report,[19]
CCDC 2054758 (6a), CCDC 2054757 (6e) CCDC 2054762
(10a), CCDC 2054761 (10f) contains the supplementary
crystallographic data for this paper. These data can be obtained
free of charge from The Cambridge Crystallographic Data
indicating that the photoreaction is mainly initiated by Centre via www.ccdc.cam.ac.uk/data_request/cif.
the interaction between the excited photocatalyst and
diselenide 1a. The oxidation potential of 1a is much
higher than the reduction potential of excited RB and
Acknowledgements
Ru(bpy)3Cl2 ·6H2O; however, in contrast to that of
diphenyl diselenide,[19] the oxidation potential of 1a
matches the reduction potential of FIrPic.[60,61] This
result helps explain the low efficiency of FIrPic
utilization in light-induced reactions with Sec-contain-
ing peptides under the developed reaction conditions.
In addition, the redox potentials measured for the
studied indoles did not provide clear insight into their
reactivity, providing additional evidence that N-hetero-
cycles do not involve an electron transfer step from
exited RB* and participate solely in the reaction with
selenyl electrophile II, yielding Sec-containing pep-
tides.
Financial support from Latvian Institute of Organic Synthesis is
gratefully acknowledged (internal grant: IG-2021-01). Authors
would like to thank Dr. Sergey Belyakov for X-ray studies and
Dr. Larisa Baumane for EPR experiments.
References
[1] X.-Y. Yu, J.-R. Chen, W.-J. Xiao, Chem. Rev. 2021, 121,
506–561.
[2] J. Xie, H. Jin, A. S. K. Hashmi, Chem. Soc. Rev. 2017,
46, 5193–5203.
[3] L. Marzo, S. K. Pagire, O. Reiser, B. König, Angew.
Chem. Int. Ed. 2018, 57, 10034–10072; Angew. Chem.
2018, 130, 10188–10228.
[4] C. K. Prier, D. A. Rankic, D. W. C. MacMillan, Chem.
Rev. 2013, 113, 5322–5363.
[5] T. P. Yoon, M. A. Ischay, J. Du, Nat. Chem. 2010, 2,
527–532.
[6] F. Strieth-Kalthoff, F. Glorius, Chem. 2020, 6, 1888–
1903.
[7] N. A. Romero, D. A. Nicewicz, Chem. Rev. 2016, 116,
10075–10166.
[8] H. Kisch, Angew. Chem. Int. Ed. 2013, 52, 812–847;
Angew. Chem. 2013, 125, 842–879.
[9] A. Ivanova, P. Arsenyan, Coord. Chem. Rev. 2018, 370,
55–68.
[10] E. Q. Luz, D. Seckler, J. S. Araújo, L. Angst, D. B.
Lima, E. A. M. Rios, R. R. Ribeiro, D. S. Rampon,
Tetrahedron 2019, 75, 1258–1266.
[11] B. M. Vieira, S. Thurow, M. Costa, A. M. Casaril, M.
Domingues, R. F. Schumacher, G. Perin, D. Alves, L.
Savegnago, Eder J. Lenardão, Asian J. Org. Chem. 2017,
6, 1635–1646.
Conclusion
A straightforward, atom-economic method for the
modification of selenocystine peptides was developed.
The mechanism of the visible light-mediated reaction
was confirmed by NMR, HRMS, UV, EPR and CV
experiments and photocatalyst emission quenching
studies. The novel method is based on a visible light-
initiated reaction for the generation of selenium
radical, which is then converted to selenium electro-
phile that is trapped by electron-rich N-heterocyles,
thus providing Sec-containing indoles in good yields.
Notably, because of initial homolytic cleavage of
SeÀ Se bond the current method allows to utilize both
parts of diselenide. Both protected and unprotected
Sec-containing peptides can be successfully employed
with excellent tolerance for sensitive amino acids (Lys,
Arg, His, Glu, Tyr). Furthermore, three approaches
were established for the synthesis of Sec-containing
indole-based macrocycles. The utilization of visible
light provides easy access to simple and sophisticated
functionalized Sec-containing peptides, opening the
way to a broad application for various types of
reactions.
[12] Y. Yu, Y. Zhou, Z. Song, G. Liang, Org. Biomol. Chem.
2018, 16, 4958–4962.
[13] Z. Gao, X. Zhu, R. Zhang, RSC Adv. 2014, 4, 19891–
19895.
[14] J. Rafique, S. Saba, M. S. Franco, L. Bettanin, A. R.
Schneider, L. T. Silva, A. L. Braga, Chem. A Eur. J.
2018, 24, 4173–4180.
[15] C. Ding, Y. Yu, Q. Yu, Z. Xie, Y. Zhou, J. Zhou, G.
Liang, Z. Song, ChemCatChem 2018, 10, 5397–5401.
[16] J. B. Azeredo, M. Godoi, G. M. Martins, C. C. Silveira,
A. L. Braga, J. Org. Chem. 2014, 79, 4125–4130.
[17] X. Zhang, C. Wang, H. Jiang, L. Sun, Chem. Commun.
2018, 54, 8781–8784.
Experimental Section
Representative procedure for visible light-mediated indole
selenylation: To a solution of Sec peptide 1a (100 mg,
0.12 mmol, 0.5 equiv.) and indole 2 (0.24 mmol, 1 equiv.) in
MeCN (5 ml) Rose Bengal (12 mg, 0.012 mmol, 0.05 equiv.)
was added and the reaction mixture was irradiated by 36 W
blue LEDs for 90 minutes. After evaporation the residue was
Adv. Synth. Catal. 2021, 363, 1–12
10
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

RESEARCH ARTICLE
asc.wiley-vch.de
[18] A. G. Meirinho, V. F. Pereira, G. M. Martins, S. Saba, J. [39] J.-Q. Liu, A. Shatskiy, B. S. Matsuura, M. D. Kärkäs,
Rafique, A. L. Braga, S. R. Mendes, Eur. J. Org. Chem.
2019, 6465–6469.
[19] Q.-B. Zhang, Y.-L. Ban, P.-F. Yuan, S.-J. Peng, J.-G.
Fang, L.-Z. Wu, Q. Liu, Green Chem. 2017, 19, 5559–
5563.
[20] S. Saba, J. Rafique, M. S. Franco, A. R. Schneider, L.
Espíndola, D. O. Silva, A. L. Braga, Org. Biomol. Chem.
2018, 16, 880–885.
[21] A. Srivastava, P. K. Singh, A. Ali, P. P. Singh, V.
Srivastava, RSC Adv. 2020, 10, 39495–39508.
Synthesis 2019, 51, 2759–2791.
[40] X. Chen, F. Ye, X. Luo, X. Liu, J. Zhao, S. Wang, Q.
Zhou, G. Chen, P. Wang, J. Am. Chem. Soc. 2019, 141,
18230–18237.
[41] H. Choi, M. Kim, J. Jang, S. Hong, Angew. Chem. Int.
Ed. 2020, 59, 22514–22522; Angew. Chem. 2020, 132,
22703–22711.
[42] R. Mousa, R. N. Dardashti, N. Metanis, Angew. Chem.
Int. Ed. 2017, 56, 15818–15827; Angew. Chem. 2017,
129, 16027–16037.
[22] I. D. Lemir, W. D. Castro-Godoy, A. Heredia, L. C. [43] E. S. J. Arnér, Essays Biochem. 2019, 64, 45–53.
Schmidt, J. E. Argüello, RSC Adv. 2019, 9, 22685– [44] M. Waliczek, Ö. Pehlivan, P. Stefanowic, New J. Chem.
22694.
2020, 44, 11433–11436.
[23] A. A. Heredia, S. M. Soria-Castro, W. D. Castro-Godoy, [45] X. Wang, A. S. Ashhurst, L. J. Dowman, E. E. Watson,
I. D. Lemir, M. López-Vidal, F. R. Bisogno, J. E.
Argüello, G. Oksdath-Mansilla, Org. Process Res. Dev.
2020, 24, 540–545.
H. Y. Li, A. J. Fairbanks, M. Larance, A. Kwan, R. J.
Payn, Org. Lett. 2020, 22, 6863–6867.
[46] K. M. Reddy, G. Mugesh, Chem. Eur. J. 2019, 25, 8875–
8883.
[24] G. Kumaraswamy, S. Vijaykumar, K. Ankammaa, V.
Narayanaraoa, Org. Biomol. Chem. 2016, 14, 11415– [47] D. T. Cohen, C. Zhang, C. M. Fadzen, A. J. Mijalis, L.
11425.
Hie, K. D. Johnson, Z. Shriver, O. Plante, S. J. Miller,
S. L. Buchwald, B. L. Pentelute, Nat. Chem. 2019, 11,
78–85.
[25] G. Kumaraswamy, V. Ramesh, M. Gangadhar, S.
Vijaykumar, Asian J. Org. Chem. 2018, 7, 1689–1697.
[26] V. Rathore, S. Kumar, Green Chem. 2019, 21, 2670– [48] P. Arsenyan, S. Lapcinska, A. Ivanova, J. Vasiljeva, Eur.
2676. J. Org. Chem. 2019, 4951–4961.
[27] C. Liu, X. Peng, D. Hu, F. Shi, P. Huang, J. Luo, Q. Liu, [49] S. Lapcinska, P. Arsenyan, Eur. J. Org. Chem. 2020,
L. Liu, New J. Chem. 2020, 44, 17245–17251. 784–795.
[28] Q. Shi, Y. Zhang, L. Wang, Org. Chem. Front. 2017, 4, [50] P. C. Stafylas, P. A. Sarafidis, Vasc. Health Risk. Manag.
1322–1330. 2008, 4, 23–30.
[29] Q.-B. Zhang, P.-F. Yuan, L.-L. Kai, K. Liu, Y.-L. Ban, [51] P. Kannaboina, K. Mondal, J. K. Laha, P. Das, Chem.
X.-Y. Wang, L.-Z. Wu, Q. Liu, Org. Lett. 2019, 21, 885–
889.
[30] Z.-P. Ye, P.-J. Xia, F. Liu, Y.-Z. Hu, D. Song, J.-A. Xiao,
Commun. 2020, 56, 11749–11762.
[52] M. T. Cash, P. R. Schreinera, R. S. Phillips, Org. Biomol.
Chem. 2005, 3, 3701–3706.
P. Huang, H.-Y. Xiang, X.-Q. Chen, H. Yang, J. Org. [53] K. Belasri, F. Fülöp, I. Szatmári, Molecules 2019, 24,
Chem. 2020, 85, 5670–5682. 3578–3587.
[31] J. Chen, R. Chen, L. Mei, S. Yan, Y. Wu, Q. Li, B. Yuan, [54] M. S. Silva, D. Alves, D. Hartwig, R. G. Jacob, G. Perin,
Asian J. Org. Chem. 2020, 9, 181–184. E. J. Lenardão, Asian J. Org. Chem. 2021, 10, 91–128.
[32] A. C. H. Weber, F. L. Coelho, R. F. Affeldt, P. H. [55] W. Guo, W. Tan, M. Zhao, K. Tao, L. Zheng, Y. Wu, D.
Schneider, Eur. J. Org. Chem. 2018, 6738–6742. Chen, X. Fan, RSC Adv. 2017, 7, 37739–37742.
[33] H. Chen, R. Ding, H. Tang, Y. Pan, Y. Xu, Y. Chen, [56] G. Brahmachari, I. Karmakar, J. Org. Chem. 2020, 85,
Chem. Asian J. 2019, 14, 3264–3268. 8851–8864.
[34] X.-L. Ma, Q. Wang, X.-Y. Feng, Z.-Y. Mo, Y.-M. Pan, [57] T. Sabri, P. D. Pawelek, J. A. Capobianco, ACS Appl.
Y.-Y. Chen, M. Xin, Y.-L. Xu, Green Chem. 2019, 21,
3547–3551.
[35] C. Liu, J. Xia, S. Ji, Z. Fan, H. Xu, Chem. Commun.
2019, 55, 2813–2816.
Mater. Interfaces 2018, 10, 26947–26953.
[58] J. AlÀ Nu’airat, B. Z. Dlugogorski, X. Gao, N. Zeinali, J.
Skut, P. R. Westmoreland, I. Oluwoyeaand, M. Altaraw-
neh, Phys. Chem. Chem. Phys. 2019, 21, 171–183.
[36] S. Ji, W. Cao, Y. Yu, H. Xu, Angew. Chem. Int. Ed. 2014, [59] W. Zhang, S. Li, X. Tang, J. Tang, C. Pan, G. Yu, Appl.
53, 6781–6785; Angew. Chem. 2014, 126, 6899–6903. Catal. B 2020, 272, 118982.
[37] M. Waliczek, Ö. Pehlivan, P. Stefanowicz, Chemis- [60] Y. Zhou, W. Li, Y. Liu, L. Zeng, W. Su, M. Zhou, Dalton
tryOpen 2019, 8, 1199–1203.
Trans. 2012, 41, 9373–9381.
[38] C. Bottecchia, T. Noël, Chem. Eur. J. 2019, 25, 26–42.
[61] H. Zhou, P. Lu, X. Gu, P. Li, Org. Lett. 2013, 15, 5646–
5649.
Adv. Synth. Catal. 2021, 363, 1–12
11
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

RESEARCH ARTICLE
Visible Light-Mediated Functionalization of Selenocystine-
Containing Peptides
Adv. Synth. Catal. 2021, 363, 1–12
S. Lapcinska, P. Dimitrijevs, L. Lapcinskis, P. Arsenyan*