Cysteine as a sustainable sulfur reagent for the protecting-group-free synthesis of sulfur-containing amino acids: Biomimetic synthesis of l-ergothioneine in water

Article abstract of DOI:10.1039/c2gc35367a

Biomass-derived cysteine was used as a sustainable sulfur source for the synthesis of rare sulfur-containing amino acids, such as l-ergothioneine (4), which might be a new vitamin, and various l- or d-2-thiohistidine compounds. Key in this simple, one-pot two-step procedure in water is a bromine-induced regioselective introduction of cysteine followed by a novel thermal cleavage reaction in the presence of thiols, a safer alternative to hazardous red phosphorus. Besides avoiding hazardous sulfur reagents, the new protecting-group-free approach reduces drastically the total number of steps, compared to described procedures. The main drawback, i.e. handling of liquid bromine as an activating and oxidizing reagent in water, was addressed by evaluating four alternative methods using in situ generation of bromine or HOBr, and first encouraging results are described.

Full text of DOI:10.1039/c2gc35367a

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Green Chemistry
Cite this: Green Chem., 2012, 14, 2256
www.rsc.org/greenchem
PAPER
Cysteine as a sustainable sulfur reagent for the protecting-group-free
synthesis of sulfur-containing amino acids: biomimetic synthesis of
L-ergothioneine in water†
Irene Erdelmeier,* Sylvain Daunay, Remi Lebel, Laurence Farescour and Jean-Claude Yadan
Received 12th March 2012, Accepted 31st May 2012
DOI: 10.1039/c2gc35367a
Biomass-derived cysteine was used as a sustainable sulfur source for the synthesis of rare sulfur-
containing amino acids, such as L-ergothioneine (4), which might be a new vitamin, and various L- or
D-2-thiohistidine compounds. Key in this simple, one-pot two-step procedure in water is a bromine-
induced regioselective introduction of cysteine followed by a novel thermal cleavage reaction in the
presence of thiols, a safer alternative to hazardous red phosphorus. Besides avoiding hazardous sulfur
reagents, the new protecting-group-free approach reduces drastically the total number of steps, compared
to described procedures. The main drawback, i.e. handling of liquid bromine as an activating and
oxidizing reagent in water, was addressed by evaluating four alternative methods using in situ generation
of bromine or HOBr, and first encouraging results are described.
Introduction
Sulfur-containing essential amino acids such as cysteine (1) and
methionine (2) are widely distributed in nature, and have impor-
tant functions as building blocks in proteins. The sulfur is
present in these amino acids as a redox-active thiol or thioether
function, which is essential to their physiological role (Fig. 1).1
Another interesting, but less well known class of sulfur-
containing amino acids are thiohistidine compounds, such as
2-thiohistidine (3a) or L-ergothioneine (4), in which the sulfur is
part of a rare imidazole-2-thione structure.2
2-Thiohistidine (3a) is used as an antioxidant, skin whitening3
and photo-protective agent.4,5 L-Ergothioneine (4) is a natural,
Fig. 1 Sulfur-containing amino acids.
non-proteinogenic amino acid. It has been first isolated from
sustainable supply and the interest in further development of
Claviceps purpurea in 1909,6 and is almost ubiquitous in living
thiohistidine compounds, such as 3a and 4.
organisms.7 Biosynthesized in fungi (such as Neurospora
Both amino acids have been synthesized previously. For
crassa)8 and mycobacteria9 it is incorporated in plants and
example, 2-thiohistidine is prepared in 5 steps from histidine,
ingested by animals and humans through the diet, from which it
using phenyl chlorothionoformate12 (5, PCTF, prepared from
is absorbed and concentrated in most tissues through a recently
thiophosgene, see path A, Scheme S1, ESI†) or potassium thio-
discovered specific organic cation transporter (OCTN1 or
cyanate (6) in an acidic medium (see Scheme S2, ESI†)13 as a
ETT10). This discovery, together with the suggestion that
sulfur source (Fig. 2).
L-ergothioneine might be a new vitamin,11 spurred the search for
TETRAHEDRON, Parc Biocitech, 102 Avenue Gaston Roussel, 93230
Romainville, France. E-mail: irene.erdelmeier@tetrahedron.fr;
Fax: +331 41837508; Tel: +331 41837501
Fig. 2 Commonly used sulfur reagents.
†Electronic supplementary information (ESI) available: Complete reac-
tion schemes for the previously described synthesis of 2-thiohistidine
and L-ergothioneine using phenylchlorothionoformate (S1) and KSCN
(S2) as sulfur source, Figures of the 1H NMR-analysis of the re-
action mixture (S3) and copies of spectral data (1H NMR and 13C NMR)
for compounds 3a–3c, 4, 12a–12c, 13, and 16. See DOI:
10.1039/c2gc35367a
Despite the apparently simple structure of 3a, selective sulfur
introduction required protection of competing reactive sites in
histidine, such as the carboxyl and amino-function, increasing
2256 | Green Chem., 2012, 14, 2256–2265
This journal is © The Royal Society of Chemistry 2012

View Article Online
Scheme 2 First example of cysteinyl-histidine synthesis (ref. 26).
Table 1 Reaction parameters for oxidative cysteine introductiona
Scheme 1 Biosynthesis of ergothioneine.
the number of steps, leading to the use of several different
organic solvents, gaseous HCl, and chromatographic separations.
The synthesis of ergothioneine has been achieved using the same
sulfur reagents in 7 and 9 steps, respectively, due to unavoidable
protection of the carboxyl and the amino groups as well as
ultimately of the nucleophilic sulfur function, to establish the
trimethyl ammonium group (see path B, Schemes S1 and S2,
ESI†).
Contrary to these chemical syntheses, the biosynthesis of
ergothioneine in microorganisms such as Neurospora crassa
is straightforward in 4 steps from histidine. In a first step,14
histidine (7a, Scheme 1) is methylated threefold by a specific
N-methyltransferase, using S-adenosyl methionine (SAM) as a
methylating agent, to give L-hercynine (8). Interestingly, sulfur is
introduced selectively in the 2-position of hercynine via an
oxidative, enzyme-catalyzed and iron-dependent reaction with
γ-glutamylcysteine (9)15 (and not cysteine as initially
suggested)16 to yield the key intermediate 10 in two steps.
Finally, 10 is cleaved to L-ergothioneine (4) and pyruvic acid
(11) by a lyase-catalyzed reaction in the presence of pyridoxal
phosphate (PLP).17
The biosynthetic pathway to 4 is appealing from the green-
chemistry viewpoint, as it is a short, protecting-group-free
synthesis, using a non-hazardous sulfur source in water as an
innocuous reaction medium.18 Protecting groups have been intro-
duced to achieve chemo- and regioselective transformations,19 as
exemplified in the previous syntheses of 3a and 4, but substantial
efforts have been developed in the last decade to meet the chal-
lenge of chemo- and regioselectivity without protecting steps.
Protecting-group-free synthesis has a tremendous potential to
improve “the economies of synthesis”,20 as demonstrated in the
synthesis of indole-based marine natural products, alkaloids, ter-
penoids, antibiotics, and azasugars.21 More recently, impressive
examples of protecting-group-free synthesis in water have been
described,22 combining the discovery of unusual reactivities in
water (as for example the rate-accelerating “on-water-effect”23)
with simplified product purification or isolation.
Concn
Entry (M)
HCl
(equiv.)
Br2
(equiv.)
Cys
(equiv.)
Yieldb
(%)
1
2
3
4
5
6
0.1
0.1
0.5
0.5
0.5
0.5
1
1
1
0.5
1.5
1
1
1
1.3
1.3
1.3
1.3
1
3
3
3
3
5
9
42
50/(46)c
30
32
62/(58)c
a Reaction conditions: hercynine (8, 5 mmol, 1.0 equiv.) and HCl
(2.5–7.5 mmol, 0.5–1.5 equiv.) at 0–2 °C in water (10–50 mL),
dropwise Br2-addition in 1–2 min (5.0–6.5 mmol, 1–1.3 equiv.),
cysteine (5–25 mmol, 1–5 equiv.) after 5–7 min. b Calculated yields by
1H NMR of the crude reaction mixture. c Isolated yields after
chromatography of the crude reaction mixture on a DOWEX 50WX2
200–400 (H+-form), 13 is obtained as tri-hydrochloride hydrate.
attractive features of the protecting-group-free biosynthesis in
water with the advantageous use of cysteine as a renewable and
readily accessible sulfur source.25 The strategy could be general-
ized to the synthesis of various L- and D-thiohistidines, rendering
this poorly explored class of sulfur-containing amino acids
readily accessible in multi-gram quantities for further biological
investigations.
Results and discussion
Our starting point was an intriguing reactivity of histidine (7a)
observed by Ito.26 After activation with bromine and addition of
cysteine, a new histidine derivative 12a was formed in low yield,
characterized by an unusual thioether bond in the 2-position of
the imidazole (Scheme 2).
Recognizing the close relationship to the in vivo formation of
intermediate 10 in the biosynthesis of ergothioneine, we studied
the feasibility to activate hercynine (8) and to induce a selective
In the present study, we describe a biomimetic approach to
ergothioneine from easily available hercynine,24 combining the
This journal is © The Royal Society of Chemistry 2012
Green Chem., 2012, 14, 2256–2265 | 2257

View Article Online
Scheme 3 Optimized synthesis of cysteinyl-histidine thioethers.
reaction with cysteine, without protection of competing reactive
sites.
In a first experiment (Table 1, entry 1), one equivalent
of bromine was added at 0 °C to an acidic aqueous solution of
hercynine (0.1 M). Rapid consumption of bromine was indicated
by discoloration of the reaction mixture, accompanied by the
formation of an orange-red gummy solid. Analysis of the reac-
tion mixture after addition of L-cysteine (1.0 equiv.) indicated the
formation of the desired cysteinyl thioether 13, but only in traces
(9%)27 besides a multitude of non-identified products. As further
investigations showed that the major part (88%) of hercynine
was converted, we speculated that the reactive intermediate
formed by the reaction of bromine with hercynine was unstable,
and degraded before reacting to a significant extent with
cysteine. Therefore, to favor the reaction with cysteine vs.
competitive degradation, several parameters were studied, such
as cysteine, acid, and bromine equivalents, and hercynine
concentration.
Scheme 4 Reaction pathways for the cleavage of cysteinyl thioethers.
Using an excess of 50 equivalents of red phosphorus in 100
equivalents of hydroiodic acid for 18 h at 100 °C, 12a is cleaved
to 2-thiohistidine (3a) and D,L-alanine (14, Scheme 4).26
These conditions appeared unacceptable not only due to the
multiple chromatographic purifications which are resource,
energy and time-consuming, but also to the use of hazardous red
phosphorus, in particular in such a high excess. Red phosphorus
is classified as highly toxic and highly inflammable,28 requiring
a high level of risk management.
Interestingly, using three equivalents of cysteine instead of
one increased the yield of thioether 13 in the reaction mixture to
42% (Table 1, entry 2). Slower dropwise addition of bromine in
slight excess (1.3 equiv.) and a higher hercynine concentration
(0.5 M instead of 0.1 M) were found to be beneficial, and pro-
vided 13 in 50% yield (Table 1, entry 3). The strong influence of
the amount of acid was revealed by comparing the reaction in
the presence of 1.0, 0.5, and 1.5 equivalents (Table 1, entries
3–5), one equivalent leading to the highest yield of 13. Finally,
the best result was obtained with 5 equivalents of cysteine,
providing 13 in 62% yield (58% isolated yield, Table 1, entry 6)
in the crude reaction mixture.
In order to study the general scope of the reaction, the opti-
mized reaction conditions were applied to L-histidine (7a), as
well as N-methyl- and N,N-dimethylhistidine24 (7b and 7c,
respectively). To our delight, using an excess of 3 equivalents of
cysteine, the known compound 12a could be obtained from 7a
with an isolated yield of 69% (compared to 19% reported pre-
viously26). Similarly, the new N-methylated cysteinyl-histidine
thioethers 12b and 12c were obtained with an isolated yield of
68 and 63% (Scheme 3), from 7b and 7c, respectively.
Searching for safer alternatives, we discovered that, under
heating for 18–24 h at 90 °C in the presence of thiols
(10 equiv.), 13 cleaved slowly to ergothioneine (80% isolated
yield) and pyruvic acid (11), the latter being converted in situ to
the corresponding dithioketal.29 By choosing an appropriate and
easily available thiol, such as 3-mercaptopropionic acid (15),30
the dithioketal 16 is formed smoothly in this new thermal
cleavage reaction (Scheme 4 and Fig. S4, ESI†).31
With regard to industrial production, another important advan-
tage of this safer alternative to the red phosphorus pathway
became evident upon treatment of the aqueous reaction mixture:
despite being a tricarboxylic acid, 16, together with the excess of
15, can be easily removed by extraction with ethyl acetate after
cleavage, therefore facilitating the purification of the very
polar ergothioneine without chromatography. In comparison,
D,L-alanine, obtained as a side product in the red phosphorus
cleavage, cannot be extracted, and has to be separated by other
means from the desired amino acid, as in Ito’s example for
the cleavage of thioether 12a via several chromatographic
separations and crystallizations.
Taking advantage of the clean and complete conversion of
thioether 13, the two steps – formation of thioether 13 and
cleavage to ergothioneine – could be integrated in a streamlined
one-pot procedure without isolation of the intermediate 13
(Scheme 5).
Carrying a zwitterionic betaine function, ergothioneine is
highly water-soluble (>300 g L−1), and could not be preferen-
tially crystallized in satisfying yield from the crude mixture,
which contained in particular cysteine (used in excess) and salts.
Striving towards a biomimetic synthesis of ergothioneine, we
searched for a direct conversion of cysteinyl thioether 13 to
ergothioneine itself, knowing that 13 is structurally related but
not identical to the biosynthetic sulfoxide intermediate 10.
While the biological transformation of thioether 10 is PLP-
dependent and catalyzed by a β-lyase (Scheme 1), the sole
related chemical transformation described in the literature is the
hydrolysis of the cysteinyl thioether 12a, derived from histidine.
2258 | Green Chem., 2012, 14, 2256–2265
This journal is © The Royal Society of Chemistry 2012

View Article Online
Fig. 3 Desalination via electrodialysis: simultaneous exchange of
cations and anions (CEM = cation exchange membrane, AEM = anion
exchange membrane).
migrate in ED through the membranes and collect in a “concen-
trate”, which can be diluted or exchanged conveniently against
water. The membranes can be used indefinitely in theory without
saturation, but process impurities can induce scaling and occlud-
ing. In short, compared to classical ion exchange resin chromato-
graphy, the process density of exchange via electrodialysis is
higher by orders of magnitude, combined with less material use
(resin vs. membranes) and lower processing costs. This makes
electrodialysis an interesting technology for green process
design, but despite its powerful advantages, it has been only
scarcely applied in organic chemistry until now.40
Scheme 5 One-pot synthesis of L-ergothioneine.
Scheme 6 Cysteine scavenging.
Implementing this efficient and environmentally friendly
technique for the desalination of the reaction mixture with a
20 membrane-pair cell following cysteine scavenging finally
allowed the crystallization of the desired highly polar ergothio-
neine, and its isolation in high purity (>99%) and a 40% global
yield over the two steps (Table 2, entry 1).
While there is room for improvement, in particular to increase
the overall yield, this novel synthetic approach to ergothioneine
via hercynine, readily available from histidine, constitutes a
robust framework for further optimisations. It is not only 3–5
steps shorter than the previously described chemical synthesis,41
but amazingly also one step shorter than the biosynthesis itself;
since cysteine can be used as a sulfur source instead of the
dipeptide γ-glutamyl cysteine.
Interestingly, cysteine can be scavenged directly from an
aqueous mixture at room temperature with benzaldehyde
(Scheme 6). An optimisation study showed that a slight excess
of benzaldehyde32 (1.16 equiv. compared to cysteine) was suffi-
cient to remove cysteine completely after pH adjustment to 4–5.
Moreover, omitting ethanol as the usually employed organic co-
solvent resulted in precipitation of the corresponding phenylthia-
zolidine-carboxylic acid 1733 (Scheme 6), which was isolated by
simple filtration in a purity >95%. Besides improving the iso-
lation of ergothioneine, this operation allows for a potential
recovery and recycling of cysteine used in excess, as the trans-
formation of 17 to cystine (the industrial cysteine precursor)34 is
known.35
For the final isolation of ergothioneine, the salt content was
significantly reduced to allow for its direct crystallization from
the aqueous mixture. For optimal process and material efficiency,
electrodialysis (ED) as a powerful desalination technology was
explored, instead of ion-exchange chromatography requiring two
consecutive cation- and anion-exchange columns. ED is a high-
throughput membrane-based technique which is easily adaptable
on an industrial scale, and commonly used in water treatment
for desalination of seawater,36,37 removal of inorganic trace con-
taminants from bore water sources,38 or tartrate stabilization in
wine.39
The general feasibility of the two-step one-pot strategy was
demonstrated by synthesizing a series of 2-thiohistidine-derived
compounds 3a to 3c, which could be isolated in 46 to 49%
global yield (Table 2, entries 2–4), respectively. Depending on
the water solubility of the corresponding final compound,42 the
treatment could be significantly simplified compared to the iso-
lation procedure for ergothioneine. Interestingly, also the non-
natural enantiomers D-ergothioneine (D-4) and D-thiohistidine
(D-3a) are easily accessible from D-hercynine and D-histidine,
respectively (entries 5 and 6), using L-cysteine as a sulfur source
and the same reaction sequence.
In a typical set-up, as shown in Fig. 3, the aqueous
reaction mixture is cycled repeatedly through the system (about
100–150 L h−1), consisting of an electrical cell containing a
membrane pack with multiple layers of alternating anion- and
cation-selective membranes. Anions and cations are exchanged
simultaneously against OH− and H+, respectively, driven by low
electrical tension (10–20 V). Contrary to ion-exchange column
chromatography, where the ions are absorbed on a resin, the ions
Noteworthy is in particular the rapidity and ease of preparation
of 2-thiohistidine (3a). As the least water-soluble compound in
this series, it can be isolated directly from the crude aqueous
reaction mixture in 49% global yield, after extraction of the
dithioketal 16, despite the presence of salts and excess cysteine.
Cysteine used in excess can then be recovered from the filtrate as
cystine via room-temperature oxidation using H2O2 and filtration
of the precipitate. To illustrate the potential for up-scaling,
This journal is © The Royal Society of Chemistry 2012
Green Chem., 2012, 14, 2256–2265 | 2259

View Article Online
Table 2 One-pot synthesis of ergothioneine and 2-thiohistidine
compounds
Evaluation of alternatives to liquid bromine
The here presented biomimetic approach represents major
improvements in the spirit of green chemistry by (i) using water
as a reaction medium, (ii) reducing the total number of steps by
avoiding protection, (iii) integrating two steps in one-pot, (iv)
avoiding hazardous red phosphorus in the cleavage step and in
particular (v) using cysteine as a renewable sulfur reagent. The
main drawback of this method is the use of elemental bromine,
which is toxic, corrosive and volatile.
The role of bromine in this reaction sequence is twofold. First
it reacts with the aromatic imidazole ring as an electrophile,
forming a reactive intermediate. We have not yet been able to
identify unambiguously its structure, but by analogy to previous
studies,44 formation of a labile bromolactone 18 can be assumed
(Scheme 7). 18 could react with cysteine as a nucleophile,
yielding the desired cysteinyl thioether after aromatization. In an
overall redox and mass balance, bromine serves furthermore as
an oxidant, which is reduced to hydrogen bromide, the hydro-
gens originating formally from the thiol function of cysteine and
the 2-position of the imidazole ring as a consequence of the
overall oxidative cysteine introduction.
Entry Substrate R/R′
Product
Yielda (%)
−
1
2
3
8
7a
7b
NMe3+/CO2
NH2/CO2H
NHMe/CO2H
4
3a
40
49
48
4
7c
NMe2/CO2H
46
−
5
6
D-8
NMe3+/CO2
37
51
D-7a
NH2/CO2H
a Isolated yield.
enantiomerically pure L-2-thiohistidine was prepared in two days
on a hundred-gram scale, without any isolation of the intermedi-
ate, chromatography or recrystallization. Using a renewable and
non-toxic sulfur source, this new manufacturing process for 3a
provides not only a significant gain in time, but also compares
favorably with the previously described 5-step processes in
terms of process mass intensity (PMI, eqn (1)). PMI is a green
chemistry “yardstick” used in the pharmaceutical industry
to drive more sustainable processes, and related to Sheldon’s
E-factor (E = PMI − 1) (Table 3).43
Scheme 7 A possible mechanism of oxidative cysteine introduction.
Numerous studies described in this journal45 and others46
investigated safer alternatives for the use of elemental bromine.
Hutchinson and coworkers47 proposed the use of HBr–H2O2 as a
greener bromination procedure, generating bromine in situ. Other
alternative procedures are based on the use of carrying agents
supporting bromine. Eissen and Lenoir48 evaluated 24 alternative
methods under environmental, health and safety aspects, and
in particular the mass index and E-factor, concluding that the
avoidance of molecular bromine by most of the investigated
methods is not a significant improvement due to higher waste
production and solvent demand. Nonetheless, the most competi-
tive alternatives considered are variations of the in situ gener-
ation of bromine, as for example by reaction of HBr with H2O2
(method A, eqn (2)), or also to a minor extent with Oxone
(method B, eqn (3)).49 Recently, the use of a bromide/bromate
couple50 in an acidic medium was introduced as another alterna-
tive to generate bromine (method C, eqn (4)) or HOBr (method
D, eqn (5)), depending on the ratio used.51
Table 3 PMI-comparison of three different processes for 2-thio-
histidine
raw materials ðkgÞ
Process Mass Intensity ðPMIÞ ¼
ð1Þ
product ðkgÞ
Cysteine-
process
KSCN-
process
PCTF-
process
PMI excluding water
PMI including water
104
133
257
416
>500
>500
2260 | Green Chem., 2012, 14, 2256–2265
This journal is © The Royal Society of Chemistry 2012

View Article Online
Table 4 Oxidative cysteine introduction using alternative methods for
bromine/HOBr generation
2-thiohistidine (3a), which could be isolated in 48% yield (using
method C, Table 4, entry 8).
These encouraging examples illustrate first attempts to avoid
the handling and use of liquid bromine for the oxidative sulfur
introduction. In our view, a detailed case-to-case assessment is
necessary to evaluate, in a holistic approach, which method is
the best for each specific amino acid by considering environ-
mental, safety and health data of the reagents together with
reaction efficiency and reagent prices. Further issues to be con-
sidered are the manufacturing scale, the treatment and isolation
procedure, and the fact that more waste (in the form of inorganic
salts) is generated by the alternative methods.
Entry
Substrate
Methoda
Product
Yieldb (%)
1
2
3
4
5
6
7
8
7a
7a
7a
7a
8
8
8
A
B
C
D
B
C
D
C
12a
12a
12a
12a
13
13
13
50
54
70
64
34
53
65
48
7a
3a
a Method A: H2O2 was stirred with HBr for 18 h before addition of
histidine/hercynine followed by cysteine at 0 °C. Method B: HBr was
added to an aqueous Oxone-solution and stirred for 15 min before
addition of histidine/hercynine followed by cysteine at 0 °C. Method C:
HCl was added dropwise at 0 °C to an acidic solution of histidine or
hercynine, NaBr and NaBrO3 (5 : 1) and stirred for 5 min before the
addition of cysteine. Method D: HCl was added dropwise at 0 °C to an
acidic solution of histidine or hercynine, NaBr and NaBrO3 (2 : 1) and
stirred for 5 min before the addition of cysteine. b Calculated yields by
1H NMR of the crude reaction mixture.
Conclusion
Applying a biomimetic approach, various sulfur-containing
amino acids such as L-ergothioneine (4)53 and 2-thiohistidine
(3a)54 have been synthesized in a simple, one-pot two-step pro-
cedure in water. It encompasses oxidative regioselective intro-
duction of cysteine as a renewable and non-toxic sulfur reagent
followed by a novel thermal cleavage reaction in the presence of
thiols as a safer alternative to a reaction implying hazardous red
phosphorus. Inspired by the biosynthesis, this new chemical
approach is one step shorter, as the histidine–sulfur bond can be
formed directly with cysteine (instead of γ-glutamyl-cysteine in
the biosynthesis).55
ðAÞ 2HBr þ H2O2 ! Br2 þ 1H2O
ð2Þ
ð3Þ
ðBÞ
ðCÞ 5NaBr þ 1NaBrO3 þ 6HCl ! 3Br2 þ 6NaCl þ 3H2O ð4Þ
Addressing the main drawback of this approach, i.e., the hand-
ling of liquid bromine as an activating and oxidizing reagent in
water, first encouraging results were obtained in the evaluation
of four alternative methods, using in situ generation of bromine
or HOBr directly in the reaction vessel.
ðDÞ 2NaBr þ 1NaBrO3 þ 3HCl ! 3HOBr þ 3NaCl ð5Þ
To evaluate the alternatives to liquid bromine in the here
described oxidative sulfur introduction, the four procedures A–D
were compared (Table 4). The in situ formation of bromine by
oxidation of HBr with H2O2 is slow. Hutchinson’s procedure is
performed by heating the reaction mixture for several hours,
using an excess of H2O2 and HBr. Since we previously observed
the thermal instability of the bromine-activated histidine inter-
mediate, bromine was therefore generated in situ prior to the
addition of histidine at 0 °C, to avoid decomposition. As shown
in Table 4, the desired thioether 12a could indeed be obtained
following the in situ generation of bromine with H2O2, but only
with a comparatively lower yield, due to incomplete conversion
(see Table 4, entry 1). This can be explained by competitive
decomposition of H2O2 catalyzed by bromine or HBr, in an
acidic medium,52 which renders complete oxidation of HBr
difficult to achieve. While the in situ generation of bromine
using Oxone is significantly quicker, the reaction yields of the
desired thioether 12a (Table 4, entry 2) or 13 (Table 4, entry 5)
are still lower than upon oxidative cysteine introduction using
liquid bromine.
Besides the benefit of avoiding hazardous sulfur reagents,
taking advantage of the unusual reactivity of the bromine acti-
vated histidine derivatives towards cysteine in water drastically
reduces the total number of steps compared to classical
approaches. As derivatization with protecting groups becomes
unnecessary, generation of waste as well as the use of organic
solvents are minimized.56
L-Ergothioneine is a highly polar and water-soluble betaine
amino acid, and to improve its isolation from the aqueous
reaction mixture in high purity, electrodialysis was used for
desalination prior to crystallization. As a time-, resource- and
material-efficient alternative to ion-exchange chromatography,
electrodialysis was shown to be an interesting technology for
organic synthesis and green process design, and may have
broader potential for the treatment and purification of reactions
designed in water.
Experimental section
Conversely, good reaction yields ranging from 64 to 70% of
the desired thioether 12a (Table 4, entries 3 and 4) and 53–65%
of the thioether 13 (Table 4, entries 6 and 7) could be obtained
with the alternative protocols C and D, comparable or even
slightly better than the result obtained by direct addition of
liquid bromine.
General
Chemicals were purchased from commercial suppliers and used
as received. H NMR spectra were recorded at 400 MHz with
D2O (4.79 ppm) or D2O–DCl as the solvent and internal stan-
dard (‘NH’ protons did not appear in H NMR spectra because
1
1
The use of in situ generated bromine is also compatible
with the two-step one-pot procedure for the synthesis of
of D2O exchange, for D2O–DCl-solutions, 2-(trimethylsilyl)-
1-propanesulfonic acid sodium salt57 was used as the internal
This journal is © The Royal Society of Chemistry 2012
Green Chem., 2012, 14, 2256–2265 | 2261

View Article Online
2-Amino-3-{[4-(2-amino-2-carboxyethyl)-1H-imidazol-2-yl]thio}-
propanoic acid (12a)
standard). Chemical shifts are reported in parts per million
(δ) downfield from TMS. Spin multiplicities are indicated by the
following symbols: s (singlet), d (doublet), t (triplet), q (quartet),
and m (multiplet). Coupling constants are reported in hertz (Hz).
13C NMR spectra were obtained at 75 MHz with D2O or D2O–
DCl as the solvent. Mass spectra were recorded by electrospray
ionization mass-spectrometry (ESI-MS) or atmospheric pressure
chemical ionization (APCI-MS). Melting points were determined
on a Stuart Scientific apparatus. Purifications of compounds
13 and 12a–c were performed by chromatography on a
DOWEX 50WX2 200–400 (H+-form). Desalination was
performed via electrodialysis (10 V) using a cell package with
20 membrane-pairs (cation-exchange PC-SK, anion-exchange
PC-SA, ED200-020, PC Cell, Germany). N-Methyl-histidine
hydrochloride (7b·HCl), N,N-dimethyl-histidine hydrochloride
hydrate (7c·HCl·H2O) and L-hercynine (8) were prepared accord-
ing to ref. 24. Other reagents, such as L-histidine hydrochloride
hydrate (7a·HCl·H2O), cysteine (1), 3-mercaptopropionic acid
(15), and benzaldehyde, were commercially available and used
without further purification.
Compound 12a was isolated as tri-hydrochloride hydrate follow-
ing general procedure A as a pale yellow hygroscopic solid; mp
(dec) 108 °C; δH (400 MHz, D2O) 3.39 (1H, dd, J 15.9, 6.9 Hz),
3.46 (1H, dd, J 15.9, 6.9 Hz), 3.68 (1H, dd, J 15.4, 4.7 Hz),
3.83 (1H, dd, J 15.4, 4.7 Hz), 4.32 (1H, t, J 6.9 Hz) and 4.34
(1H, t, J 4.7 Hz), 7.51 (1H, s); δC (75 MHz, D2O) 25.9, 35.6,
52.5, 53.1, 121.3, 129.9, 138.0, 169.9 and 170.8; m/z (ESI) 275
(MH+). Found: C, 26.80; H, 4.71; N, 13.79. Calc. for
C9H14N4O4S × 3HCl × H2O: C, 26.91; H, 4.77; N, 13.95%.
2-Amino-3-({4-[2-carboxy-2-(methylamino)ethyl]-1H-imidazol-
2-yl}thio)propanoic acid (12b)
Compound 12b was isolated as tri-hydrochloride hydrate follow-
ing general procedure A as a pale yellow hygroscopic solid; mp
(dec) 110 °C; δH (400 MHz, D2O) 2.83 (3H, s), 3.40 (1H, dd,
J 15.8, 7.8 Hz), 3.50 (1H, dd, J 15.8, 5.4 Hz), 3.70 (1H, dd,
J 15.4, 4.8 Hz), 3.83 (1H, dd, J 15.4, 4.8 Hz), 4.16 (1H, dd,
J 7.8, 5.4 Hz), 4.36 (1H, t, J 4.8 Hz) and 7.52 (1H, s); δC
(75 MHz, D2O) 24.6, 32.1, 35.4, 52.8, 60.0, 121.3, 129.5,
137.7, 169.4 and 169.9; m/z (APCI) 290. Found: C, 29.30; H,
4.80; N, 13.57%. Calc. for C10H16N4O4S × 3HCl × H2O: C,
28.89; H, 5.09; N, 13.48%.
Optimization study: preparation of 3-{2-[(2-amino-2-
carboxyethyl)thio]-1H-imidazol-4-yl}-2-(trimethylammonio)-
propanoate (13)
For the optimization of reaction conditions, hercynine (8,
986 mg, 5 mmol) and concd hydrochloric acid (2.5–7.5 mmol,
0.5–1.5 equiv.) were dissolved in water (10–50 mL) and cooled
in an ice-bath. Br2 (800–1040 mg, 5.0–6.5 mmol, 1.0–1.3
equiv.) was added dropwise, and cysteine (0.61–3.03 g,
5–25 mmol, 1–5 equiv.) after 5–7 min. After stirring for 1 h at
0 °C, 200 μL of the reaction mixture were diluted in 500 μL
2-Amino-3-({4-[2-carboxy-2-(dimethylamino)ethyl]-1H-
imidazol-2-yl}thio)propanoic acid (12c)
Compound 12c was isolated as tri-hydrochloride hemi-hydrate
following general procedure A as a pale yellow hygroscopic
solid; mp (dec) 102 °C; δH (400 MHz, D2O) 3.01 (6H s), 3.42
(1H, dd, J 15.3, 9.8 Hz), 3.54 (1H, dd, J 15.3, 4.4 Hz), 3.70
(1H, dd, J 15.5, 4.8 Hz), 3.81 (1H, dd, J 15.5, 4.8 Hz), 4.30
(1H, dd, J 9.8, 4.4 Hz), 4.46 (1H, t, J 4.8 Hz) and 7.52 (1H, s);
δC (75 MHz, D2O) 22.4, 35.5, 41.3 (broad), 52.7, 66.6, 121.2,
130.1, 137.4 and 169.4;58 m/z (ESI) 303 (MH+). Found: C,
31.40; H, 5.12; N, 13.12%. Calc. for C11H18N4O4S × 3HCl ×
0.5H2O: C, 31.40; H, 5.27; N, 13.32%.
1
D2O, filtered, and the solution analyzed by H NMR (400 MHz,
see Fig. S3, ESI†).
Column chromatography (gradient 0.5–2 N HCl, DOWEX
50WX2-400) of the filtered reaction mixture provided compound
13 in the form of a slightly yellow hygroscopic solid as tri-
hydrochloride hydrate (1.295 g, 58% using 5 equiv. cysteine;
1.065 g, 46% using 3 equiv. cysteine); mp (dec) 97 °C; δH
(400 MHz, D2O) 3.36 (9H, s), 3.50 (1H, dd, J 14.2 and
11.8 Hz), 3.63 (1H, dd, J 14.2 and 3.8 Hz), 3.73 (1H, dd, J 15.5
and 4.7 Hz), 3.84 (1H, dd, J 15.5 and 4.7 Hz), 4.31 (1H, dd,
J 11.8 and 3.8 Hz), 4.49 (t, J 4.7 Hz, 1H) and 7.51 (1H, s); δC
(75 MHz, D2O) 23.0, 35.4, 52.7, 53.0, 74.5, 121.3, 129.1,
138.0, 168.6 and 169.4; m/z (ESI) 317 (MH+). Found: C, 32.10;
H, 5.59; N, 12.22. Calc. for C12H20N4O4S × 3HCl × H2O: C,
32.48; H, 5.68; N, 12.63%.
L-Ergothioneine (4)
(A) Formation of 4 and 2,2-bis[(2-carboxyethyl)thio]propa-
noic acid (16) by cleavage of compound 13 in the presence of
15. Compound 13·3HCl·H2O (2.22 g, 5 mmol) and 3-mercapto-
propionic acid (15, 4.4 mL, 5.31 g, 50 mmol, 10 equiv.) were
dissolved in water (20 mL) and heated under stirring for 18 h at
90–100 °C (see Fig. S4, ESI†). The solution was cooled to rt,
extracted with ethyl acetate (20 mL × 4, to remove the excess of
15 and dithioketal 16), and the pH adjusted to 6 by addition of a
20% ammonia solution. After lyophilization, the powder
obtained was dissolved in an ethanol–water mixture and treated
with charcoal. After filtration, crystallization provided
L-ergothioneine (4) as a white powder (917 mg, 80%).
General procedure A: synthesis of compounds 12a–12c
Following the procedure described for 13, using 1.3 equiv. Br2 at
0 °C and 3 equiv. cysteine, compounds 12a (69%), 12b (68%),
and 12c (63%) were obtained from L-histidine (7a), α-N-methyl-
L-histidine (7b), and α-N,N-dimethyl-L-histidine (7c), respect-
ively, after column chromatography (gradient 0.5–2.0 N HCl,
DOWEX 50WX2-400) of the filtered reaction mixture.
Compound 16 was synthesized independently to confirm
the structure: a mixture of pyruvic acid (11, 4 mL, 5.0 g,
55.6 mmol), 15 (9.86 mL, 11.9 g, 111.2 mmol) and water
2262 | Green Chem., 2012, 14, 2256–2265
This journal is © The Royal Society of Chemistry 2012

View Article Online
(2 mL) was heated under stirring at 90 °C for 18 h. After the
addition of water (20 mL), filtration and washing with water,
ethyl acetate and pentane, a white solid was obtained, which was
recrystallized in acetonitrile (containing 1% dithiothreitol). Com-
pound 16 was obtained as fine colorless crystals (6.13 g; 37%);
mp 139 °C; δH (400 MHz, D2O + NaOD (pH 7)) 1.74 (3H, s),
2.48 (4H, m) and 2.82 (4H, m); δC (75 MHz, D2O + NaOD
(pH 7)) 27.0, 27.5, 37.5, 64.8, 178.0 and 181.6; m/z (APCI) 305
(M + Na+). Found: C, 38.40; H, 4.99. Calc. for C9H14O6S2: C,
38.28; H, 4.99%.
at 0 °C. 3-Mercaptopropionic acid (960 g, 9 mol, 6 equiv.) was
added, and the mixture heated for 18 h at 95 °C. After cooling to
rt, the clear brown solution was extracted with ethyl acetate. The
aqueous phase was placed in a water bath at 40 °C under nitro-
gen, and the pH of the solution adjusted to 6.5. L-2-Thiohistidine
(3a) precipitated as a white solid (135.0 g, 49%), which was
filtered, rinsed several times with water and finally ethanol, and
dried; mp (dec) 314 °C; [α]D −10.7 (c 1.0 in HCl 1 N) (lit.,26
mp > 280 °C; [α]D −10.6 (c 0.5 in HCl 1 N)); δH (400 MHz,
D2O + DCl) 3.22 (1H, dd, J 15.9, 7.3 Hz), 3.30 (1H, dd, J 15.9,
5.9 Hz), 4.36 (1H, dd, J 7.3, 5.9 Hz) and 6.92 (1H, s); δC
(75 MHz, D2O + DCl) 25.7, 52.2, 116.5, 123.6, 156.8, and
170.7; m/z (ESI) 188 (MH+).
(B) One-pot-procedure for the synthesis of L-ergothioneine
(4). Hercynine (183.0 g, 0.9 mol) and concd HCl (37%,
75.1 mL, 0.9 mol) were dissolved in water (1.8 L), and the solu-
tion cooled to 0 °C. Under vigorous stirring and continued
cooling, Br2 (60.1 mL, 186.0 g, 1.17 mol, 1.3 equiv.) was added
rapidly, followed by cysteine (556.3 g, 4.5 mol, 5 equiv.)
5–7 min later, and the mixture stirred for 1 h at 0 °C. 3-Mercap-
topropionic acid (579 g, 5.4 mol, 6 equiv.) was added, and the
mixture heated for 18 h at 95 °C. After cooling to rt, the clear
orange solution was extracted with ethyl acetate, and the pH of
the aqueous phase adjusted to 4.5 by addition of a 20%
ammonia solution. Under stirring and cooling with a water bath,
benzaldehyde (482.3 g, 4.5 mol, 5 equiv.) was added rapidly.
After 5–10 min, phenylthiazolidine-4-carboxylic acid (17, δH
(400 MHz, d6-DMSO)33 3.00–3.17 (1H, m), 3.25–3.42 (1H, m),
3.89 (0.45H, t, J 7.8 Hz), 4.22 (0.55H, dd, J 6.7, 4.7 Hz), 5.50
(0.45H, s), 5.67 (0.55H, s) and 7.20–7.54 (5H, m)) precipitated
as a white solid, which was filtered, and rinsed several times
with water. The combined filtrates were washed with ethyl
acetate, and desalinated by electrodialysis. The obtained slightly
yellow solution was concentrated under vacuum. Crystallization
provided L-ergothioneine (4) in the form of white crystals
(83.0 g, 40%); mp (dec) 251 °C; [α]D +125.2 (c 1.0 in H2O)
(lit.,12 mp 262–263 °C; [α]D +126.6 (c 1.0 in H2O)); δH
(400 MHz, D2O) 3.18 (2H, m), 3.28 (9H, s), 3.90 (1H, dd,
J 11.7, 4.0 Hz), 6.80 (1H, s); δC (75 MHz, D2O) 23.4, 52.7,
77.5, 115.7, 124.3, 156.6, and 170.5; m/z (ESI) 230 (MH+).
Found: C, 47.30; H, 6.59; N, 18.29%. Calc. for C9H15N3O2S: C,
47.14; H, 6.59; N, 18.32%.
D-(+)-2-Thiohistidine (D-3a)
Following the same protocol as described for the synthesis of
L-2-thiohistidine, D-2-thiohistidine (D-3a, 7.6 g, 51%) was
obtained starting from D-histidine·HCl·H2O (16.77 g, 80 mmol);
mp (dec) 316–318 °C; [α]D +11.3 (c 1.0 in HCl 1 N). Found: C,
38.10; H, 4.88; N, 22.45 Calc. for C6H9N3O2S: C, 38.49; H,
4.84; N, 22.44%. Spectral data were identical in all respects with
those of the above enantiomer.
2-(Methylamino)-3-(2-thioxo-2,3-dihydro-1H-imidazol-4-yl)-
propanoic acid (3b)
Following the same protocol as described for the synthesis of
L-ergothioneine (without desalination), compound 3b·H2O
(5.28 g, 48%) was obtained in the form of a white powder
starting from compound 7b·HCl (10.28 g, 50 mmol), Br2
(3.34 mL, 10.39 g, 65 mmol, 1.3 equiv.) and L-cysteine (18.17 g,
150 mmol, 3 equiv.), after cleavage of intermediate 12b in the
presence of 3-mercaptopropionic acid (31.8 g, 300 mmol,
6 equiv.), scavenging of the excess of cysteine with benz-
aldehyde (15.9 g, 150 mmol, 3 equiv.) and crystallization; mp
(dec) 234 °C; [α]D +36.7 (c 1.0 in HCl 1 N); δH (400 MHz, D2O
+ DCl) 2.81 (3H, s), 3.31 (2H, m), 4.27 (1H, t, J 6.2 Hz) and
6.93 (1H, s); δC (75 MHz, D2O + DCl) 24.8, 32.2, 60.0, 116.4,
123.2, 156.9 and 170.1; m/z (APCI) 202 (MH+). Found: C,
38.30; H, 5.82; N, 18.87%. Calc. for C7H11N3O2S × H2O: C,
38.34; H, 5.98; N, 19.16%.
D-(−)-Ergothioneine (D-4)
Following the same protocol, D-ergothioneine (D-4, 1.7 g, 37%)
was obtained starting from D-hercynine (4.0 g, 20 mmol), Br2
(1.34 mL, 4.17 g, 26 mmol, 1.3 equiv.) and L-cysteine (12.41 g,
100 mmol, 5 equiv.), via desalination with Amberlite IRA 410
resin (HCO3−-form); mp (dec) 250 °C; [α]D −124.6 (c 1.0 in
H2O). Found: C, 47.10; H, 6.55; N, 18.40%. Calc. for
C9H15N3O2S: C, 47.14; H, 6.59; N, 18.32%. Spectral data were
identical in all respects with those of the above enantiomer.
2-(Dimethylamino)-3-(2-thioxo-2,3-dihydro-1H-imidazol-4-yl)-
propanoic acid (3c)
Following the same protocol as described for the synthesis of
L-ergothioneine (without desalination), compound 3c·0.5H2O
(10.25 g, 46%) was obtained in the form of a white powder start-
ing from compound 7c·HCl·H2O (23.76 g, 100 mmol), Br2
(6.68 mL, 20.7 g, 130 mmol, 1.3 equiv.) and L-cysteine (37.0 g,
300 mmol, 3 equiv.), after cleavage of intermediate 12c in the
presence of 3-mercaptopropionic acid (64.3 g, 600 mmol, 6
equiv.), scavenging of the excess of cysteine with benzaldehyde
(30.7 g, 300 mmol, 3 equiv.) and crystallization; mp (dec)
200 °C; [α]D +5.9 (c 1.0 in HCl 1 N); δH (400 MHz, D2O +
DCl) 3.02 (6H, s), 3.30 (1H, dd, J 15.6, 8.5 Hz), 3.40 (1H, dd,
J 15.6, 5.5 Hz), 4.32 (1H, dd, J 8.5, 5.5 Hz) and 6.94 (1H, s); δC
L-(−)-2-Thiohistidine (3a)
Compound 7a·HCl·H2O (317.6 g, 1.5 mol) was dissolved in
water (3 L), and the solution cooled to −2–0 °C. Under vigorous
stirring and continued cooling, Br2 (100 mL, 311 g, 1.95 mol,
1.3 equiv.) was added rapidly, followed by cysteine (562 g,
4.5 mol, 3 equiv.) 5–7 min later, and the mixture stirred for 1 h
This journal is © The Royal Society of Chemistry 2012
Green Chem., 2012, 14, 2256–2265 | 2263

View Article Online
(75 MHz D2O + DCl) 22.9, 40.9, 43.1, 66.7, 116.2, 123.8,
156.3 and 169.5; m/z (APCI) 216 (MH+). Found: C, 42.80; H,
6.18; N, 18.65%. Calc. for C8H13N3O2S × 0.5H2O: C, 42.84; H,
6.29; N, 18.74%.
Notes and references
1 See for example: J. T. Brosnan and M. E. Brosnan, J. Nutr., 2006, 135,
1636S–1640S; A. Agbas and J. Moskovitz, Curr. Signal Transduction
Ther., 2009, 4, 46–50; C. Klomsiri, P. A. Karplus and L. B. Poole, Anti-
oxid. Redox Signaling, 2011, 14, 1065–1077.
2 An imidazole-thione group is in tautomeric thione/thiol equilibrium, but
at physiological pH, the thione form predominates strongly, see:
N. Motohashi, I. Mori and Y. Sugiura, Chem. Pharm. Bull., 1976, 24,
1737–1741. This explains the striking air-stability of aqueous solutions of
2-thiohistidine compounds such as ergothioneine compared to other bio-
logical important thiols, such as cysteine or glutathione
Oxidative cysteine introduction using alternative methods
(Table 4)
3 P. André, C. Marteau, I. Renimel and M. Marielle, US Pat., 7 935 359,
2011.
Method A (Table 4, entry 1): Hydrogen peroxide (30%, 736 mg,
6.5 mmol, 1.3 equiv.) was added dropwise to a solution of
hydrobromic acid (2.19 g, 1.47 mL, 13 mmol, 2.6 equiv.) in
water (10 mL). After stirring for 18 h at room temperature in a
closed vessel, the orange-red solution was cooled to 0 °C. Histi-
dine hydrochloride hydrate (1.05 g, 5 mmol, 1 equiv.) was added
and, after 5 min at 0 °C, cysteine (1.85 g, 15 mmol, 3 equiv.).
After stirring for 1 h at 0 °C, 200 μL of the reaction mixture
were diluted in 500 μL D2O, filtered, and the solution analyzed
by 1H NMR (400 MHz).
4 K. K. Makinen and P. L. Makinen, Biosci. Rep., 1982, 2, 169–175.
5 Up to now, 2-thiohistidine (3a) has not yet been identified as a naturally
occurring amino acid, but it is found in the peptide sequence of important
enzymes such as tyrosinase, where it is located near the metal-containing
active site (K. Lerch, J. Biol. Chem., 1982, 257, 6414–6419). In peptide
synthesis, it is applied to introduce conformational rigidity (C. Gielens,
K. Idakieva, M. De Maeyer, V. Van Den Bergh, N. I. Siddiqui and
F. Compernolle, Peptides, 2007, 28, 790–797) and in drug discovery to
systematically study the role of histidine and cysteine residues by repla-
cing them with the thiohistidine fragment (see for example W. Shen,
K. J. Barr, J. D. Oslob and M. Zhong, WO2005/044817)
Method B (Table 4, entries 2 and 5): Hydrobromic acid
(2.19 g, 1.47 mL, 13 mmol, 2.6 equiv.) was added dropwise to a
cooled solution of Oxone (2.0 g, 6.5 mmol, 1.3 equiv.) in water
(10 mL) at 0 °C. After stirring for 15 min at room temperature
in a closed vessel, histidine (5 mmol, 1 equiv. for entry 2) or
hercynine (8, 986 mg, 5 mmol, 1 equiv. for entry 5) was
added and, after 5 min at 0 °C, cysteine (1.85 g, 15 mmol,
3 equiv. for entry 2; 3.09 g, 25 mmol, 5 equiv. for entry 5). After
stirring for 1 h at 0 °C, 200 μL of the reaction mixture were
diluted in 500 μL D2O, filtered, and the solution analyzed by
1H NMR.
Method C (Table 4, entries 3 and 6): Hydrochloric acid (37%,
1.2 g, 1.1 mL, 13 mmol, 2.6 equiv.) was added dropwise to a
cooled solution of histidine hydrochloride hydrate (5 mmol,
1 equiv. for entry 3) or hercynine hydrochloride (5 mmol,
1 equiv. for entry 6), sodium bromide (1.14 g, 11 mmol) and
sodium bromate (0.33 g, 2.2 mmol) in water (10 mL) at 0 °C.
Cysteine (1.85 g, 15 mmol, 3 equiv. for entry 3; 3.09 g,
25 mmol, 5 equiv. for entry 6) was added after 5 min at 0 °C.
After stirring for 1 h at 0 °C, 200 μL of the reaction mixture
were diluted in 500 μL D2O, filtered, and the solution analyzed
by 1H NMR.
6 C. Tanret, C. R. Acad. Sci., 1909, 149, 222–224.
7 I. K. Cheah and B. Halliwell, Biochim. Biophys. Acta, Mol. Basis Dis.,
2012, 1822, 784–793.
8 S. D. Genghof and O. Van Damme, J. Bacteriol., 1970, 103, 475–478.
9 S. D. Genghof, J. Bacteriol., 1964, 87, 852–862.
10 D. Grundemann, S. Harlfinger, S. Golz, A. Geerts, A. Lazar, R. Berkels,
N. Jung, A. Rubbert and E. Schornig, Proc. Natl. Acad. Sci. U. S. A.,
2005, 102, 5256–5261.
11 B. D. Paul and S. H. Snyder, Cell Death Differ., 2009, 17, 1134–1140.
12 J. Xu and J.-C. Yadan, J. Org. Chem., 1995, 60, 6296–6301.
13 M. Trampota, US Pat., 7 767 826, 2010.
14 Y. Ishikawa and D. B. Melville, J. Biol. Chem., 1970, 245, 5967–5973.
15 F. P. Seebeck, J. Am. Chem. Soc., 2010, 132, 6632–6633.
16 Y. Ishikawa, S. E. Israel and D. B. Melville, J. Biol. Chem., 1974, 249,
4420–4427.
17 Another difference between the biosynthesis and a common feature of
both chemical preparations is the fate of the imidazole-cycle during sulfur
introduction: while the imidazole moiety of hercynine is completely con-
served during the enzymatic sulfur-introduction, it is cleaved in the
chemical processes, either via a Bamberger-type or by an ANRORC-
sequence (see ref. 12)
18 M.-O. Simon and C.-J. Li, Chem. Soc. Rev., 2012, 41, 1415–1427.
19 For numerous examples see: Protective Groups in Organic Synthesis, ed.
T. W. Greene and P. G. M. Wuts, John Wiley & Sons, New York, 4th
Revised edn, 2006; Protecting Groups, ed. P. J. Kocienski, Georg Thieme
Verlag, Stuttgart, 3rd edn, 2005.
20 T. Newhouse, P. S. Baran and R. W. Hoffmann, Chem. Soc. Rev., 2009,
38, 3010–3021.
21 R. W. Hoffmann, Synthesis, 2006, 3531–3541; I. S. Young and
P. S. Baran, Nat. Chem., 2009, 1, 193–205; R. A. Shenvi, D. P. O’Malley
and P. S. Baran, Acc. Chem. Res., 2009, 42, 530–541; E. M. Dangerfield,
C. H. Plunkett, B. L. Stocker and M. S. M. Timmer, Molecules, 2009, 14,
5298–5307.
22 See for example: J. McNulty, P. Das and D. McLeod, Chem.–Eur. J.,
2010, 16, 6756–6760; J. Liu, X. Deng, A. E. Fitzgerald, Z. S. Sales,
H. Venkatesan and N. S. Mani, Org. Biomol. Chem., 2011, 9, 2654–
2660.
23 J. E. Klijn and B. F. N. Engberts, Nature, 2005, 435, 746–748; C.-J. Li,
Chem. Rev., 2005, 105, 3095–3165; U. M. Lindström and F. Andersson,
Angew. Chem., Int. Ed., 2006, 45, 548–551.
24 L-Hercynine can be prepared in two steps and 73% global yield from
L-histidine: V. N. Reinhold, Y. Ishikawa and D. B. Melville, J. Med.
Chem., 1968, 11, 258–260.
Method D (Table 4, entries 4 and 7): Hydrochloric acid (37%,
0.64 g, 0.55 mL, 6.6 mmol, 1.3 equiv.) was added dropwise to a
cooled solution of histidine hydrochloride hydrate (5 mmol,
1 equiv. for entry 4) or hercynine hydrochloride (5 mmol,
1 equiv. for entry 7), sodium bromide (0.45 g, 4.33 mmol) and
sodium bromate (0.33 g, 2.2 mmol) in water (10 mL) at 0 °C.
Cysteine (1.85 g, 15 mmol, 3 equiv. for entry 4; 3.09 g,
25 mmol, 5 equiv. for entry 7) was added after 5 min at 0 °C.
After stirring for 1 h at 0 °C, 200 μL of the reaction mixture
were diluted in 500 μL D2O, filtered, and the solution analyzed
by 1H NMR.
25 Cysteine is produced in several thousand tons per year by extraction or
fermentation (W. Leuchtenberger, K. Huthmacher and K. Drauz, Appl.
Microbiol. Biotechnol., 2005, 69, 1–8; A. Ault, J. Chem. Educ., 2004,
81, 347–355)
Acknowledgements
26 S. Ito, J. Org. Chem., 1985, 50, 3636–3638.
The authors would like to thank Dr Marc Moutet (Tetrahedron)
for helpful discussions.
27 As 13 had been synthesized in small amounts in the study of the bio-
synthesis of ergothioneine (see ref. 16) from chloroalanine and
2264 | Green Chem., 2012, 14, 2256–2265
This journal is © The Royal Society of Chemistry 2012

View Article Online
ergothioneine, its described 1H NMR-spectrum could be compared to the
1H NMR of the crude reaction mixture (Fig. S3, ESI†). Besides non-
converted hercynine (about 12%) and other unidentified hercynine-
derived products with aromatic protons, a signal at 7.43 ppm, attributed
to H-5 in 13, could indeed be detected, its integration accounting for
ca. 9% of the total aromatic protons
44 G. L. Schmir and L. A. Cohen, Biochemistry, 1965, 4, 533–538, and
ref. 26
45 (a) G. Rothenberg and H. H. Clark, Green Chem., 2000, 2, 248–251;
(b) D. Bogdal, M. Lukasiewicz and J. Pielichowski, Green Chem., 2004,
6, 110–113; (c) A. Podgorsek, M. Eissen, J. Fleckenstein, S. Stavber,
M. Zupan and J. Iskra, Green Chem., 2009, 11, 120–126; (d) J. Wang,
W. Wang and J.-H. Li, Green Chem., 2010, 12, 2124–2126;
(e) D. Wischang, J. Hartung, T. Hahn, R. Ulber, T. Stumpf and C. Fecher-
Trost, Green Chem., 2011, 13, 102–108.
46 A. Podgorsek, M. Zupan and J. Iskra, Angew. Chem., Int. Ed., 2009, 48,
8424–8450.
47 L. C. McKenzie, L. M. Huffman and J. E. Hutchinson, J. Chem. Educ.,
2005, 82, 306–310.
48 M. Eissen and D. Lenoir, Chem.–Eur. J., 2008, 14, 9830–9841.
49 K. M. Kim and I. H. Park, Synthesis, 2004, 2641–2644.
50 (a) F. Fujisaki, S. Hamura, H. Eguchi and A. Nishida, Bull. Chem. Soc.
Jpn., 1993, 66, 2426–2428; (b) S. Adimurthy, G. Ramachandraiah, A.
V. Bedekar, S. Ghosh, B. C. Ranu and P. K. Ghosh, Green Chem., 2006,
8, 916–922; (c) S. Adimurthy, S. Ghosh, P. U. Patoliya,
G. Ramachandraiah, M. Agrawal, M. R. Gandho, S. C. Upadhyay,
P. K. Ghosh and B. C. Ranu, Green Chem., 2008, 10, 232–237.
51 NaBrO3 is preferable compared to KBrO3, as potassium bromate
is classified by IARC as possibly carcinogenic to humans (class 2B), see:
28 (a) European Chemical Substances Informations System (ESIS). 2000.
IUCLID Dataset for phosphorus. European Commission Joint Research
Centre. http://esis.jrc.ec.europa.eu/doc/IUCLID/data̲sheets/7723140.pdf,
last accessed May 2012; (b) Hazardous Substances Data Bank (HSDB).
2010. Entry for phosphorus. United States National Library of Medicine
http://toxnet.nlm.nih.gov/, last accessed May 2012.
29 The transient formation of 11, which is also a byproduct in the biosyn-
thesis of ergothioneine, is probably due to immediate hydrolysis of
dehydroalanine as the primary cleavage product, which could not be
detected in the reaction mixture
30 3-Mercaptopropionic acid (CAS-Nr. 107-96-0) is registered in the
context of the REACH-regulation, and the dossier available at: http://
apps.echa.europa.eu/registered/data/dossiers/DISS-97d8e287-1c3b-58ba-
e044-00144f67d031/DISS-97d8e287-1c3b-58ba-e044-00144f67d031_
DISS-97d8e287-1c3b-58ba-e044-00144f67d031.html, last accessed May
2012.
31 While thermal elimination reactions of aryl-alkyl-sulfoxides are well
known (see for example: S. I. Goldberg and M. S. Sahli, J. Org. Chem.,
1967, 32, 2059–2062), the mechanism and driving force of this cysteinyl-
thioether cleavage in the presence of thiols needs further investigation.
A first suspected intermediate oxidation of 13 to sulfoxide 10 by ambient
oxygen can be excluded, as this cleavage proceeds smoothly also in the
absence of oxygen and using degassed thioether-solutions
32 For a summary of data, see OECD SIDS-document for benzaldehyde,
CAS-Nr. 100-52-7, http://www.inchem.org/documents/sids/sids/100527.
pdf, last accessed May 2012.
http://monographs.iarc.fr/ENG/Monographs/vol73/mono73-22.pdf,
accessed May 2012.
last
52 A. Podgorsek, S. Stavber, M. Zupan and J. Iskra, Green Chem., 2007, 9,
1212–1218.
53 I. Erdelmeier, WO2011042480, 2011.
54 I. Erdelmeier and S. Daunay, WO2011042478, 2011.
55 Theoretically, a shorter reaction sequence yielding ergothioneine in
one step and with higher atom economy could be imagined by
using Na2S instead of cysteine, comparable to the one-step synthesis
of 2-thiohistidine as reported by Ito in 12% yield (ref. 26). In
a comparative evaluation of the Na2S reaction with bromine-activated
33 M. Seki, Y. Mori, M. Hatsuda and S. Yamada, J. Org. Chem., 2002, 67,
5527–5536.
34 K. Drauz, I. Grayson, A. Kleemann, H.-P. Krimmer, W. Leuchtenberger
and C. Weckbecker, Amino Acids, Ullmann’s Encyclopedia of Industrial
Chemistry, Wiley-VCH, 2007.
35 M. P. Schubert, J. Biol. Chem., 1936, 114, 341–350.
36 T. Xu and C. Huang, AIChE J., 2008, 54, 3147–3159.
37 H. Strathmann, Desalination, 2010, 264, 268–288.
38 L. J. Banasiak and A. I. Schäfer, Desalination, 2009, 248, 48–57.
39 F. Goncalves, C. Fernandes, P. Carreira dos Santos and M. N. de Pinho,
J. Food Eng., 2003, 59, 229–235.
hercynine on a 5 mmol-scale, we detected small amounts of
ergothioneine (2–4%). As in an industrial perspective the use of Na2S
in an acidic medium (generating H2S in situ) or the direct use of
gaseous H2S is not recommendable and sustainable for safety and
environmental reasons (see for example: hydrogen sulfide: human
health aspects. (Concise International Chemical Assessment Document;
53). World Health Organization: Geneva, 2003), further optimization
of this reaction pathway was not considered, despite its high atom
economy
40 A. Aghajanyan, A. A. Hambardzumyan, A. A. Vardanyan and
A. S. Saghiyan, Desalination, 2008, 228, 237–244.
56 The use of organic solvents impacts strongly the environmental footprint
of a process, see for a quantitative assessment: R. K. Henderson,
C. Jimenez-Gonzalez, D. J. C. Constable, S. R. Alston, G. G. A. Inglis,
G. Fisher, J. Sherwood, S. P. Binks and A. D. Curzons, Green Chem.,
2011, 13, 854–862.
41 Four steps starting from L-histidine (2 steps for the synthesis of hercynine
according to ref. 24) and 2 steps in the here described one-pot sequence)
vs. 7 steps for the PCTF procedure and 9 steps for the synthesis using
KSCN, see schemes S1 and S2, ESI.†
57 The use of this and other internal references related to 1H NMR studies
of histidine-derivatives is mentioned in this reference: T. Tynkkynen,
M. Tiainen, P. Soininen and Laatikainen, Anal. Chim. Acta, 2009, 648,
105–112.
42 The water-solubility ranges from >300 g L−1 for 4 to ∼2 g L−1 for 3a as
the least water-soluble compound in this series
43 C. Jimenez-Gonzalez, C. S. Ponder, Q. B. Broxterman and J. B. Manley,
Org. Process Res. Dev., 2011, 15, 912–917; R. A. Sheldon, Green
Chem., 2007, 9, 1273–1283.
58 Apparently, the carbons of both carboxyl groups have the same NMR
resonance in this compound, contrary to those of 12a and 12b
This journal is © The Royal Society of Chemistry 2012
Green Chem., 2012, 14, 2256–2265 | 2265