Glycosylated eumelanin building blocks by thioglycosylation of 5,6-diacetoxyindole with an expedient selenium-based dynamic-mixture methodology

Article abstract of DOI:10.1002/ejoc.201200299

A series of 3-thioglycosylated 5,6-diacetoxyindole derivatives, which are important tools for eumelanin research and application, were prepared through a practical and efficient approach exploiting a dynamic mixture of thioglycoside agents. The strategy is feasible for installing both mono- and disaccharide units and relies on the facile in situ conversion of glycosyl disulfides into the corresponding, more reactive, phenylselenenyl sulfides in the presence of diphenyl diselenide, N-bromosuccinimide (NBS) and tetrabutylammonium bromide (TBAB). An expedient thioglycosidation of 5,6-diacetoxyindole with a dynamic mixture of thioglycosides is described. Copyright

Full text of DOI:10.1002/ejoc.201200299

Job/Unit: O20299
/KAP1
Date: 21-06-12 16:02:40
Pages: 7
FULL PAPER
DOI: 10.1002/ejoc.201200299
Glycosylated Eumelanin Building Blocks by Thioglycosylation of
5,6-Diacetoxyindole with an Expedient Selenium-Based Dynamic-Mixture
Methodology
Matteo Adinolfi,[a] Marco d’Ischia,[a] Alfonso Iadonisi,*[a] Loredana Leone,[a]
Alessandro Pezzella,[a] and Silvia Valerio[b]
Keywords: Thioethers / Glycosylation / Glycosides / Selenium / Synthetic methods
A series of 3-thioglycosylated 5,6-diacetoxyindole deriva-
tives, which are important tools for eumelanin research and
application, were prepared through a practical and efficient
approach exploiting a dynamic mixture of thioglycoside
agents. The strategy is feasible for installing both mono- and
disaccharide units and relies on the facile in situ conversion
of glycosyl disulfides into the corresponding, more reactive,
phenylselenenyl sulfides in the presence of diphenyl diselen-
ide, N-bromosuccinimide (NBS) and tetrabutylammonium
bromide (TBAB).
Introduction
Water-soluble dark polymers generated by oxidation of
glycosylated 5,6-dihydroxyindoles provide a novel, powerful
tool in melanin research for studies of visible chromophore
development and aggregation mechanisms underlying the
absorption features and electronic properties of eumelanin
biopolymers.[1] In addition, they hold considerable promise
for cell recognition, surface coating, and other biomedical
and technological applications, by circumventing the no-
torious issues of insolubility and chemical heterogeneity
that has so far slowed down progress in eumelanin research.
A crucial prerequisite for rapid advances in the field is the
availability of suitable glycosylated 5,6-dihydroxyindole pre-
cursors that are amenable to structural and π-electron ma-
nipulation. 3-Thiogalactosylated 5,6-diacetoxyindole (1)
was initially chosen as the glycosylated model precursor of
water-soluble eumelanins, and was polymerized following in
situ deacetylation (Scheme 1).[2]
Compound 1 was prepared by using the selenium-con-
taining thioglycosidation reagent 2a (Scheme 2), which was
coupled with 5,6-diacetoxyindole (DAI, 3) (Scheme 3).
The thioglycosylating agent 2a was easily prepared by
modification of a route previously developed for the synthe-
sis of alkyl thioglycosides.[3] In the modified procedure
(Scheme 2), an inexpensive peracetylated sugar was iodin-
ated at the anomeric position,[4] and the crude product ob-
Scheme 1. Synthesis of water-soluble eumelanin polymers.
tained by a simple extractive workup was directly subjected
to a one-pot, two-step sequence, involving generation of an
isothiouronium intermediate and subsequent conversion
into 2a mediated by triethylamine and diphenyl diselenide.
Although this protocol was more straightforward than a
previously reported approach,[5] because intermediate isola-
tion was avoided, it suffered from poor reproducibility,
leading to 2a in variable yields (up to 45%), with minor
amounts of the corresponding glycosyl thiol 2b (up to 25%)
and disulfide 2c (occasionally the main thioglycosylated de-
rivative).[6]
After chromatographic purification, 2a was treated with
DAI in the presence of stoichiometric amounts of N-
bromosuccinimide (NBS) in acetonitrile at 60 °C to yield 1
in 49% yield. Under these conditions, disulfide 2c proved
again to be the main byproduct, together with diphenyl dis-
elenide and succinimide, whereas DAI was almost totally
consumed due to concurrent bromination and decomposi-
tion processes.
[a] Dipartimento di Scienze Chimiche, Università “Federico II” di
Napoli,
Via Cinthia 4, 80126 Naples, Italy
Fax: +39-081-674393
E-mail: iadonisi@unina.it
[b] International Chemical Industry,
81030 Cellole (CE), Italy
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201200299.
Eur. J. Org. Chem. 0000, 0–0
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1

Job/Unit: O20299
/KAP1
Date: 21-06-12 16:02:40
Pages: 7
A. Iadonisi et al.
FULL PAPER
trophilic substitution mechanism via sulfonium-like inter-
mediates, suggesting, rather, that critical free-radical steps
were involved in the thioglycosylation process
(Scheme 4).[10]
Scheme 4. Possible key free-radical steps in the NBS-induced thio-
glycosidation of DAI with β-phenylselenothiogalactoside.
Scheme 2. Sequential approach to thioglycosidation agents.
Although available evidence does not allow a detailed
mechanistic picture to be assembled, it seems likely that
thioglycosidation of DAI is initiated by the generation of
small amounts of Br· from NBS, which would trigger re-
lease of glycosylthiyl radical 4 from 2a. This latter would
then attack DAI to generate the β-glycosylthioalkyl radical
5.[11,12]
In this scheme, the inhibitory effects of TEMPO or oxy-
gen could be interpreted as being due to the trapping of
Br·, 5, or related free-radical intermediates that sustain the
process. Clearly, ionic and free-radical mechanisms are not
mutually exclusive and may both be involved in the indole
thioglycosidation process.
Whatever the actual mechanism, this working hypothesis
provided the rationale for subsequent efforts aimed at iden-
tifying alternative thioglycosidation agents potentially more
prone to generate thiyl radicals. Because disulfide 2c was
obtained in high amounts in the preparation of 2a
(Scheme 2), indole thioglycosidation with 2c was investi-
gated. Direct treatment of 3 with 2c did not result in ap-
preciable coupling either in the presence or in the absence
of NBS. This observation was taken to mean that the S–
Scheme 3. Thioglycosidation of DAI.
In view of the emerging interest of soluble glycosylated
eumelanins in materials science[7] and in glycobiology as an
assembly with a multiple exposure of sugars,[8] we have been
prompted to embark on a systematic search for more ef-
ficient, scalable routes to thioglycosylated indoles involving
diverse saccharide precursors.
Results and Discussion
In addition to the synthesis of the thioglycosylating agent SePh functionality was essential for indole thioglycosid-
of type 2a, realization of the proposed goal required the ation. Consistent with this conclusion, treatment of disulf-
installation of a highly functionalized alkylthio moiety onto ide 2c with (PhSe)2 in the presence of NBS, i.e., under the
the 3-position of a labile and oxidizable indole, which was same conditions adopted for indole thioglycosidation, re-
expected to be a difficult task based on the paucity of litera- sulted in moderate yields of 2a (ca. 35%). In the presence
ture data[9] and the limitations inherent to state-of-the-art of DAI, in addition to some 2a, low amounts of 1 were
technologies for indole functionalization with sulfur rea- detected. These results discouraged use of disulfide 2c as a
gents.
single thioglycosidation reagent, but the significant conver-
Clues for improving the previously reported protocol sion of 2c in situ into the actual thioglycosidation agent 2a
were gained from the observation that the coupling of DAI inspired a more expeditious and simpler strategy in which
with the galactosylthio donor 2a (Scheme 3) was signifi- the whole mixture of thiogalactose compounds (2a, 2b and
cantly slowed down in the presence of oxygen and was in- 2c), generated as depicted in Scheme 2, was employed for
hibited by catalytic amounts of the nitroxide spin trap indole functionalization under NBS activation. The poten-
2,2,6,6-tetramethylpiperidin-1-oxyl (TEMPO; 0.2 equiv.). tial of this strategy was supported by the observation that,
This observation was not compatible with the expected elec- in the presence of NBS, 2a promoted thioglycosidation of
2
www.eurjoc.org
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 0000, 0–0

Job/Unit: O20299
/KAP1
Date: 21-06-12 16:02:40
Pages: 7
Synthesis of Glycosylated Eumelanin Building Blocks
indole 3 with concomitant formation of diphenyl diselen- Table 1. Coupling of DAI (In) with thioglycoside mixtures (entries
in italics refer to experiments on the effect of 0.4 equiv. of TBAB).
ides, which, in turn, could induce further conversion of di-
sulfide 2c into 2a. Moreover, thiol 2b, the third component
of the mixture, was expected to serve as a similar or even
better thioglycosidation precursor, in view of its expected
higher facility to produce thiyl radicals through hydrogen
transfer. To test this hypothesis, a mixture of 2a–c was pre-
pared as shown in Scheme 2 by utilizing 1 equiv. of diphenyl
diselenide in the last step of the sequence. After removal of
diphenyl diselenide and triethylsilane apolar byproducts on
silica,[13] the thiogalactose components were collected to-
gether, and the composition of the crude mixture was as-
sessed by NMR spectroscopic analysis. Integration of the
corresponding anomeric resonances indicated a 2a/2b/2c ra-
tio of 35:25:40 (expressed in terms of thiogalactosyl units).
Most rewardingly, treatment of DAI with the above mix-
ture (1.5 equiv. of thiogalactosyl units) in the presence of
NBS (1.1 equiv.) afforded 1 in much better yield (almost
70%) even at room temperature (Table 1, Entry 1). This re-
sults should be compared with a 49% yield obtained with
2a as the sole thioglycosylation agent at 60 °C within com-
parable reaction times. Another advantage of the new meth-
odology is the remarkably improved yield in the thioglycos-
idation pool (nearly 70%, determined as the overall content
of thiogalactosyl units in 2a + 2b + 2c), far exceeding that
with 2a alone (less than 45%). In addition, utilization of
2a–c as a mixture led to a substantial simplification of the
synthetic protocol, because previous isolation of 2a as a
single component was not straightforward due to almost
coelution with side-product 2b.
In a subsequent series of experiments, the scope of the
synthetic sequence was assessed with a group of structurally
diverse monosaccharides, i.e., d-glucose, d-mannose, l-fuc-
ose, and a representative disaccharide, d-lactose. With all
sugars, the recovery of the resulting mixtures of thioglycos-
ylated components RSX (R: glycosyl residue; X: SePh, H,
or SR) was in the 61–72% range, as determined by NMR
spectroscopic analysis. As expected,[3] a large predominance
[a] 1 equiv. of TBAB.
of 1,2-trans-thioglycosides (β-gluco- and -galactosides, α-
mannosides) was observed for all sugars. Interestingly, in all
these cases the glycosyl thiols were formed in small or trace
amounts, and the glycosyl disulfide/glycosyl-SSePh ratios
(expressed in terms of thioglycosyl units) varied in the range
of 1.5:1 to 3:1. Subsequent indole thioglycosidations
(Table 1, Entries 1, 3, 5, 7, 9, and 10) occurred stereoselec-
tively (except for Entry 5) to yield 1,2-trans-thioglycosides
in satisfactory to good yields (33–68%). Depending on the
reactivity of the thio-sugar reactants, different temperature
conditions were adopted with the aim of performing the
coupling within a few hours. A peculiar result was observed observed lower yield of 7 due to the larger occurrence of
in the synthesis of thiomannoside 7 (Table 1, Entry 5), concurrent degradation processes (observed by TLC). Be-
which was obtained as a mixture of anomeric products in cause Br·, a putative initiator in the reaction process
an approximate 1:1 ratio, despite the almost exclusive α- (Scheme 4), is known to form adducts with Br– anions,[14]
configuration of the thiomannoside components in the re- the effect of tetrabutylammonium bromide (TBAB) was ex-
actant mixture. It may be speculated that the 1,2-trans-diax- amined in separate experiments. As shown in Table 1 (italic-
ial relationship in the expected product 7α might favor an ized entries), addition of TBAB (0.4–1 equiv.) resulted, in
anomerization mechanism based on the expulsion of the almost all cases, in a 10–20% yield enhancement and, most
thioindole moiety. This event might also account for the noteworthy, caused suppression of the observed anomeri-
Eur. J. Org. Chem. 0000, 0–0
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3

Job/Unit: O20299
/KAP1
Date: 21-06-12 16:02:40
Pages: 7
A. Iadonisi et al.
FULL PAPER
Scheme 5. Use of NBS for the oxidative ring opening of a 4,6-benzylidene probe.
zation in the DAI thiomannosylation reaction (compare steps under mild experimental conditions. The effect of
Table 1, Entries 5 and 6). The protocol also proved effective TBAB on the NBS-based procedure is also highlighted.
for the attachment of N-acetylglucosamine to DAI (Table 1,
Entry 12). In this specific case, the mixture of thioglycosyl-
Experimental Section
Preparation of Mixtures of the Thioglycosylation Agents. General
Procedure: To a solution of a per-O-acetylated sugar (390 mg,
1 mmol) in anhydrous CH2Cl2 (4 mL) was added I2 (356 mg,
1.4 mmol), and Et3SiH (225 μL, 1.4 mmol) (Caution: exothermic
reaction). The mixture was heated to reflux until complete con-
sumption of the starting material (5–10 min as monitored by TLC).
The mixture was then diluted with CH2Cl2, and the organic phase
ating agents was prepared by starting from the correspond-
ing glycosyl chloride[6] rather than the glycosyl iodide,
which is more difficult to synthesize.[15]
In an attempt to clarify the role of NBS as a promoter
in the overall process, a test reaction was performed to as-
certain whether this reagent is capable of generating radical
species under the experimental conditions adopted for DAI
thioglycosidation. For this purpose 4,6-benzylidene deriva- was washed with aqueous sodium carbonate containing sodium
thiosulfate (the latter was added portionwise until consumption of
residual iodine in the organic phase was observed). The organic
phase was then washed with water, dried, and concentrated. Thio-
urea (114 mg, 1.5 mmol) was added to the crude residue, and the
mixture was suspended in acetonitrile (4 mL), and then heated to
60 °C. After 15–30 min, TLC analysis evidenced complete con-
sumption of the glycosyl iodide and the generation of a polar prod-
uct. The vessel was cooled to room temp., and then phenyl diselen-
ide (1 mmol, 312 mg) and triethylamine (0.55 mL, 4 mmol) were
sequentially added. The mixture was stirred until TLC analysis evi-
tive 11 was exposed to a slight stoichiometric excess of NBS
in acetonitrile under argon at room temperature. It is well
established that thermal or photochemical activation of
NBS in halogenated solvents causes oxidative opening of
benzylidene derivatives through the generation of a benzyl
radical from the acetal carbon atom.[16] Under these condi-
tions, the corresponding 6-bromo-4-O-benzoyl derivative
(i.e., 14 in Scheme 5) is predominantly formed under anhy-
drous conditions, whereas, in the presence of water, the ini-
tial intermediate of benzylic bromination undergoes hydrol- denced consumption of the thiouronium derivative (25 min for ga-
lactose, 60 min for glucose, 60 min for mannose, 15 min for lactose,
45 min for fructose); thereafter, the mixture was concentrated in
vacuo, the residue was dissolved in CH2Cl2, and the organic phase
was washed with water. The aqueous phase was then re-extracted
with CH2Cl2 and the combined organic phases were dried and con-
centrated. Silica gel flash chromatography of the residue (hexane/
ethyl acetate) yielded sequentially diphenyl diselenide, the corre-
sponding (phenylselenyl)thioglycoside, the corresponding thiol,
minor amounts of the sugar hemiacetal (occasionally), and the
symmetric glycosyl disulfides. All recovered thioglycoside agents
ysis and then evolves to mono-O-benzoylated carbinols
such as 12 and 13.[17] This latter process occurs with poor
regioselectivity when the procedure is applied to six-mem-
bered benzylidenes.[18]
In support of the Br· generating pathway, exposure of
11 to NBS (1 equiv.) in acetonitrile under argon at room
temperature for 5 h provided a mixture enriched in mono-
O-benzoylated products 12–14 (Scheme 5). Activation of
NBS occurred in acetonitrile under very mild conditions
entailing neither high-temperature activation nor specific were collected together and the composition of the mixture was
evaluated by NMR spectroscopic analysis. In the case of the N-
acetylglucosamine, the protocol was applied by starting from the
corresponding glycosyl chloride, and acetone was used as the sol-
vent for the synthesis of the thiouronium intermediate (3 h under
reflux) and its final exposure to phenyl diselenide (1 equiv.) and
triethylamine (4 equiv.) (45 min, at room temp.). Overall yield and
composition (expressed as the ratio of thioglycosyl units present in
each component evaluated by NMR spectroscopic analysis) of the
obtained mixtures: d-galactose, overall yield 68%, sugar disulfide/
sugar thiol/sugar–SSePh 40:25:35; d-glucose: overall yield 66%,
sugar disulfide/sugar–SSePh, 3:1; d-mannose: overall yield 61%,
sugar disulfide/sugar–SSePh, 1.5:1; l-fucose, overall yield 68%,
sugar disulfide/sugar–SSePh, 3:1; d-lactose: overall yield 72%,
sugar disulfide/sugar–SSePh, 2.5:1; N-acetyl-d-glucosamine, overall
yield 45%, sugar disulfide/sugar–SSePh, 2.2:1.
photochemical initiation, the process being effective at
room temperature under standard laboratory illumination.
Conclusions
In this paper we have described a convenient methodol-
ogy for thioglycosidation of DAI. The procedure is appli-
cable to a variety of saccharide units, including disac-
charides, deoxy and amino sugars, and is characterized by
optimized utilization of the thioglycosyl pool, easy workup
with no need to isolate intermediate products, and im-
proved thioglycoside yields with respect to state-of-the-art
methodologies. Although some mechanistic facets of the
overall process require elucidation, the present data under-
DAI Thioglycosidation. General Procedure: DAI (30 mg,
score the role of NBS as a trigger for the critical free-radical 0.13 mmol), the appropriate amount of TBAB (see Table 1), and a
4
www.eurjoc.org
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 0000, 0–0

Job/Unit: O20299
/KAP1
Date: 21-06-12 16:02:40
Pages: 7
Synthesis of Glycosylated Eumelanin Building Blocks
mixture of thioglycosides (0.19 mmol of thioglycosyl units) were
dissolved under argon in acetonitrile (2.5 mL) (acetonitrile was
deoxygenated with a stream of argon for 15 min prior to use). To
the resulting solution was added NBS (25 mg, 0.15 mmol) under
argon, and the mixture was stirred under laboratory illumination
at the temperature indicated in Table 1. At the end of the reaction,
the mixture was diluted with CH2Cl2, and the organic phase was
washed with aqeous thiosulfate. The aqueous phase was re-ex-
tracted with CH2Cl2, and the combined organic phases were dried
and concentrated under vacuum to yield a residue that was purified
by silica gel flash chromatography (hexane/ethyl acetate).
5,6-Diacetoxyindol-3-yl 2,3,4-Tri-O-acetyl-1-thio-β-D-fucopyran-
oside (8): Rf = 0.43 (ethyl acetate/hexane, 7:3); [α]D = +83.5 (c =
0.8, CHCl3). 1H NMR (500 MHz, CDCl3): δ = 8.67 (br. s, 1 H.
NH), 7.55 (s, 1 H), 7.28 (d, J = 2.0 Hz, 1 H), 7.14 (s, 1 H), 5.15
(br. d, J = 3.0 Hz, 1 H, Fuc 4-H), 5.11 (t, J = 10.0 Hz, 1 H, Fuc
2-H), 5.00 (dd, J = 3.0, 9.5 Hz, 1 H, Fuc 3-H), 4.43 (d, J = 9.5 Hz,
1 H, Fuc 1-H), 3.70 (br. q, J = 6.5 Hz, 1 H, Fuc 5-H), 2.32 (ϫ2),
2.15, 1.95, 1.93 (5 s, 18 H, 6 COCH3), 1.13 (d, J = 6.5 Hz, 3 H,
Fuc 6-H3) ppm. 13C NMR (125 MHz, CDCl3): δ = 170.7, 170.1,
169.6, 169.2 (ϫ2, COCH3), 138.4, 136.8, 133.3, 132.9, 127.8, 114.0,
105.9, 101.7 (indole C atoms), 86.2 (Fuc C-1), 72.9, 72.5, 70.2, 67.3,
21.0, 20.7, 20.5; 16.3 (Fuc C-6) ppm. MALDI-MS: calcd. for [M
+ Na]+ 559.97; found 560.12. C24H27NO11S (537.54): calcd. C
53.63, H 5.06; found C 53.45, H, 5.00.
5,6-Diacetoxyindol-3-yl 2,3,4,6-Tetra-O-acetyl-1-thio-β-D-galacto-
pyranoside (1): Rf = 0.40 (ethyl acetate/hexane, 3:2). [α]D = –70.7 (c
1
= 0.6, CHCl3). H NMR (400 MHz, CDCl3): δ = 8.81 (br. s, 1 H,
5,6-Diacetoxyindol-3-yl 2,3,4,6-Tetra-O-acetyl-β-
D-galacto-pyran-
NH), 7.51 (s, 1 H), 7.29 (d, J = 2.8 Hz, 1 H), 7.14 (s, 1 H), 5.28
(br. d, J = 2.8 Hz, 1 H, Gal 4-H), 5.07 (t, J = 10.0 Hz, 1 H, Gal
2-H), 4.98 (dd, J = 2.8, 10.0 Hz, 1 H, Gal 3-H), 4.44 (d, J =
10.0 Hz, 1 H, Gal 1-H), 4.10 (dd, J = 6.8, 10.8 Hz, 1 H, Gal 6a-
H), 3.97 (dd, J = 6.8, 11.2 Hz, 1 H, Gal 6b-H), 3.81 (br. t, J =
6.8 Hz, 1 H, Gal 5-H), 2.32, 2.31, 2.14, 2.01, 1.93, 1.85 (6 s, 18 H,
6 COCH3) ppm. 13C NMR (100 MHz, CDCl3): δ = 170.4, 170.3,
170.1, 169.5, 169.4 (ϫ2, COCH3), 138.1, 136.6, 133.5, 133.2, 127.8,
113.7, 105.9, 100.4 (indole Cs), 86.1 (Glc C-1), 74.1, 72.1, 67.1
(ϫ2), 61.6, 20.9, 20.7, 20.6, 20.5 (ϫ2), 20.2 ppm. MALDI-TOF
osyl-(1Ǟ4)-2,3,6-tri-O-acetyl-1-thio-β- -glucopyranoside (9): Rf =
D
1
0.43 (ethyl acetate/hexane, 4:1). [α]D = –76.5 (c = 1.0, CHCl3). H
NMR (500 MHz, CDCl3): δ = 8.61 (br. s, 1 H, NH), 7.50 (s, 1 H),
7.21 (d, J = 2.5 Hz, 1 H), 7.15 (s, 1 H), 5.29 (br. d, J = 3.0 Hz, 1
H, Gal 4-H), 5.14 (t, J = 9.0 Hz, 1 H, Glc 3-H), 5.05 (dd, J = 8.0,
10.5 Hz, 1 H, Gal 2-H), 4.89 (dd, J = 3.0, 10.0 Hz, 1 H, Gal 3-H),
4.75 (t, J = 10.0 Hz, 1 H, Glc 2-H), 4.42 (br. d, J = 11.5 Hz, 1 H,
Glc 6a-H), 4.40 (d, J = 10.0 Hz, 1 H, Glc 1-H), 4.35 (d, J = 8.0 Hz,
1 H, Gal 1-H), 4.10–4.00 (overlapped signals, 2 H, Gal 6-H2), 3.97
(dd, J = 4.0, 11.5 Hz, 1 H, Glc 6b-H), 3.78 (br. t, J = 7.0 Hz, 1 H,
Gal 5-H), 3.56–3.48 (m, 2 H, Glc 4-H and 5-H), 2.32 (ϫ2), 2.12
(ϫ2), 2.04, 2.01, 2.00, 1.99, 1.95 (5 s, 27 H, 9 COCH3) ppm. 13C
NMR (125 MHz, CDCl3): δ = 170.4 (ϫ2), 170.3, 170.1, 169.8.
169.7, 169.4, 169.2, 169.0 (COCH3), 138.5, 136.9, 133.7, 133.3,
128.4, 114.2, 105.9, 100.9 (indole C atoms and Gal C-1), 84.8 (Glc
C-1), 75.9, 74.1, 70.9, 70.5, 69.8, 69.0, 66.5, 62.2, 60.6, 60.4, 20.9,
20.8, 20.6, 20.5, 20.4 ppm. MALDI-MS: calcd. for [M + Na]+
906.60; found 906.81. C38H45NO21S (883.82): calcd. C 51.64, H
5.13; found C 51.80, H 5.05.
MS: calcd. for [M + Na]+ 617.90; found 618.13. C26H29NO13
(595.57): calcd. C 52.43, H 4.91; found C 52.25, H 4.95.
S
5,6-Diacetoxyindol-3-yl 2,3,4,6-Tetra-O-acetyl-1-thio-β-D-glucopyr-
anoside (6): Rf = 0.40 (ethyl acetate/hexane, 3:2). [α]D = –70.8 (c =
0.5, CHCl3). 1H NMR (500 MHz, CDCl3): δ = 8.63 (br. s, 1 H,
NH), 7.49 (s, 1 H), 7.28 (d, J = 2.5 Hz, 1 H), 7.16 (s, 1 H), 5.17 (t,
J = 9.5 Hz, 1 H, Glc 3-H), 4.91 and 4.90 (overlapped t, J = 9.5,
10.0 Hz, 2 H, Glc 2-H and 4-H), 4.43 (d, J = 10.0 Hz, 1 H, Glc 1-
H), 4.15–4.05 (m, 2 H, Glc 6-H2), 3.65–3.60 (m, 1 H, Glc 5-H),
2.32, 2.31, 2.13, 1.99, 1.98, 1.96 (6 s, 18 H, 6 COCH3) ppm. 13C
NMR (125 MHz, CDCl3): δ = 170.7, 170.2, 169.4, 169.3, 169.2,
169.1 (COCH3), 138.5, 136.9, 133.4, 133.2, 128.6, 113.8, 105.9,
100.1 (indole C atoms), 85.3 (Glc C-1), 75.6, 74.2, 69.7, 68.0, 62.0,
20.8–20.5 ppm. MALDI-MS: calcd. for [M + Na]+ 618.13; found
617.97. C26H29NO13S (595.57): calcd. C 52.43, H 4.91; found: C
52.30, H 4.85.
5,6-Diacetoxyindol-3-yl 3,4,6-Tri-O-acetyl-2-deoxy-2acetamido-1-
thio-β-D-glucopyranoside (10): Rf = 0.30 (ethyl acetate). [α]D = –67.4
1
(c = 0.5, CHCl3). H NMR (500 MHz, CDCl3): δ = 9.04 (br. s, 1
H, NH), 7.46 (s, 1 H), 7.19 (d, J = 2.4 Hz, 1 H), 7.08 (s, 1 H), 6.08
(d, J = 9.2 Hz, 1 H, 2-NHAc), 5.18 (t, J = 9.6 Hz, 1 H, GlcNAc
3-H), 4.91 (t, J = 9.6 Hz, 1 H, GlcNAc 4-H), 4.56 (d, J = 10.0 Hz,
1 H, GlcNAc 1-H), 4.15–4.05 (m, 2 H, GlcNAc 6-H2), 3.86 (q, J
= 9.2 Hz, 1 H, GlcNAc 2-H), 3.65–3.55 (m, 1 H, Glc 5-H), 2.32,
2.31, 2.02 (ϫ2), 1.97 (ϫ2, 4 s, 18 H, 6 COCH3) ppm. 13C NMR
(125 MHz, CDCl3): δ = 170.8 (ϫ2), 170.5, 169.7, 169.5, 169.4
(COCH3), 138.2, 136.6, 133.2 (ϫ2), 127.9, 113.6, 106.2, 100.5 (in-
dole C atoms), 86.5 (GlcNAc C-1), 75.4, 73.8, 68.5, 62.3, 52.9 (C-
2), 23.2, 20.7–20.5 ppm. MALDI-MS: calcd. for [M + Na]+ 617.16;
found 617.27. C26H30N2O12S (594.59): calcd. C 52.52, H 5.09;
found C 52.30, H 4.95.
5,6-Diacetoxyindol-3-yl 2,3,4,6-Tetra-O-acetyl-1-thio-D-mannopyr-
anoside (7): Rf = 0.40 (ethyl acetate/hexane, 3:2). 1H NMR
(500 MHz, CDCl3): δ (signals of α-anomer) = 8.61 (br. s, 1 H, NH),
7.42 (s, 1 H), 7.30 (d, J = 2.5 Hz, 1 H), 7.16 (s, 1 H), 5.51 (dd, J =
1.5, 3.0 Hz, 1 H, Man 2-H), 5.40 (dd, J = 3.0, 9.5 Hz, 1 H, Man
3-H), 5.30 (t, J = 10.0 Hz, 1 H, Man 4-H), 5.11 (br. s, 1 H, Man
1-H), 4.50–4.58 (m, 1 H, Man 5-H), 4.28 (dd, J = 5.0, 12.0 Hz, 1
H, Man 6a-H), 4.08 (dd, J = 2.5, 12.0 Hz, 1 H, Man 6b-H), 2.31,
2.30, 2.09, 2.07, 2.05, 2.00 (6 s, 18 H, 6 COCH3) ppm; δ (significant
signals of the β-anomer) = 8.59 (br. s, 1 H, NH), 7.51 (s, 1 H), 7.24
(d, J = 2.5 Hz, 1 H), 7.16 (s, 1 H), 5.64 (d, J = 3.0 Hz, 1 H, Man
2-H), 5.27 (t, J = 10.0 Hz, 1 H, Man 4-H), 4.96 (dd, J = 3.5,
10.0 Hz, 1 H, Man 3-H), 4.59 (s, 1 H, Man 1-H), 3.45–3.55 (m, 1
H, Man 5-H), 2.32, 2.30, 2.12, 2.04, 2.00, 1.96 (6 s, 18 H, 6 COCH3)
ppm. 13C NMR (125 MHz, CDCl3): δ = 170.8, 170.7, 170.4, 170.1,
170.0, 169.8, 169.7, 169.6, 169.5, 169.4 (COCH3), 138.4 (ϫ2), 136.9
(ϫ2), 133.2 (ϫ2), 132.5 (ϫ2), 127.2, 126.7, 113.1, 112.3, 106.6,
106.4, 102.3, 101.4 (indole C atoms), 86.2, 85.8 (2 C-1), 76.2, 72.0,
70.6, 70.4, 69.6, 69.3, 66.5, 65.6, 62.6 (ϫ 2), 20.7–20.5 ppm.
MALDI-MS: calcd. for [M + Na]+ 618.13; found 617.92.
C26H29NO13S (595.57): calcd. C 52.43, H 4.91; found C 52.55, H
4.85.
Supporting Information (see footnote on the first page of this arti-
1
cle): Copies of the H and 13C NMR spectra of all products.
Acknowledgments
This work carried out in the frame of the Programmi di Ricerca di
Interesse Nazionale (PRIN) 2008 project (M. d’I.).
[1] a) M. d’Ischia, A. Napolitano, A. Pezzella, E. J. Land, C. A.
Ramsden, P. A. Riley, Adv. Heterocycl. Chem. 2005, 89, 1; b)
M. d’Ischia, A. Napolitano, A. Pezzella, Eur. J. Org. Chem.
2011, 5501.
Eur. J. Org. Chem. 0000, 0–0
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5

Job/Unit: O20299
/KAP1
Date: 21-06-12 16:02:40
Pages: 7
A. Iadonisi et al.
FULL PAPER
[2]
[3]
[4]
[5]
[10] L. Tebben, A. Studer, Angew. Chem. 2011, 123, 5138; Angew.
Chem. Int. Ed. 2011, 50, 5034.
[11] For examples of indole functionalization by initial addition of
electrophilic radicals, see: a) D. A. Androsov, T. L. S. Kish-
baugh, G. W. Gribble, Tetrahedron Lett. 2008, 49, 6621; b) J. W.
Tucker, J. M. R. Narayanam, S. W. Krabbe, C. R. J. Stephen-
son, Org. Lett. 2010, 12, 368.
[12] A similar radical intermediate was proposed for the thiocyan-
ation of indoles promoted by oxone, see: a) G. Wu, Q. Liu, Y.
Shen, W. L. Wu, Tetrahedron Lett. 2005, 46, 5831; b) A. Pez-
zella, A. Palma, A. Iadonisi, A. Napolitano, M. d’Ischia, Tetra-
hedron Lett. 2007, 48, 3883.
[13] Diphenyl diselenide can be recovered by crystallization.
[14] a) J. Lind, X. Shen, T. E. Eriksen, G. Merenyi, L. Eberson, J.
Am. Chem. Soc. 1991, 113, 4629; b) D. I. Pattison, R. J. O’Re-
illy, O. Skaff, L. Radom, R. F. Anderson, M. J. Davies, Chem.
Res. Toxicol. 2011, 24, 371.
A. Pezzella, A. Iadonisi, S. Valerio, L. Panzella, A. Napolitano,
M. Adinolfi, M. d’Ischia, J. Am. Chem. Soc. 2009, 131, 15270.
S. Valerio, A. Iadonisi, M. Adinolfi, A. Ravidà, J. Org. Chem.
2007, 72, 6097.
M. Adinolfi, A. Iadonisi, A. Ravidà, M. Schiattarella, Tetrahe-
dron Lett. 2003, 44, 7863.
D. P. Gamblin, P. Garnier, S. van Kasteren, N. J. Oldham, A. J.
Fairbanks, B. G. Davis, Angew. Chem. 2004, 116, 846; Angew.
Chem. Int. Ed. 2004, 43, 828.
[6] Use of catalytic diphenyl diselenide (3%) results in a high-
yielding synthesis of the symmetrical glycosyl disulfides, see:
M. Adinolfi, D. Capasso, S. Di Gaetano, A. Iadonisi, L. Leone,
A. Pastore, Org. Biomol. Chem. 2011, 9, 6278.
[7] M. d’Ischia, A. Napolitano, A. Pezzella, P. Meredith, T. Sarma,
Angew. Chem. 2009, 121, 3972; Angew. Chem. Int. Ed. 2009,
48, 3914.
[8] C.-Y. Wu, C.-H. Wong, Chem. Commun. 2011, 47, 6201.
[9] a) J. A. Campbell, C. A. Broka, L. Gong, K. A. M. Walker, J.-
H. Wang, Tetrahedron Lett. 2004, 45, 4073; b) J. S. Yadav,
B. V. S. Reddy, Y. J. Reddy, Tetrahedron Lett. 2004, 45, 7034; c)
K. M. Schlosser, A. P. Krasutsky, J. E. Reed, H. W. Hamilton,
K. Sexton, Org. Lett. 2004, 6, 819; d) Y. Maeda, M. Koyabu,
T. Nishimura, S. Uemura, J. Org. Chem. 2004, 69, 7688; e) J. S.
Yadav, B. V. S. Reddy, Y. J. Reddy, K. Praneeth, Synthesis 2009,
1520; f) X. L. Fang, R. Y. Tang, P. Zong, J. H. Li, Synthesis
2009, 4183.
[15] N. Miquel, S. Vignando, G. Russo, L. Lay, Synlett 2004, 341.
[16] S. Hanessian, N. R. Plessas, J. Org. Chem. 1969, 34, 1045.
[17] R. W. Binkley, G. S. Goewey, J. C. Johnston, J. Org. Chem.
1984, 49, 992.
[18] a) M. Adinolfi, G. Barone, L. Guariniello, A. Iadonisi, Tetrahe-
dron Lett. 1999, 40, 8439; b) P. M. Senthilkumar, A. Aravind,
S. Baskaran, Tetrahedron Lett. 2007, 48, 1175; c) A. Stevenin,
F.-D. Boyer, J.-M. Beau, J. Org. Chem. 2010, 75, 1783.
Received: March 12, 2012
Published Online: ᭿
6
www.eurjoc.org
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 0000, 0–0

Job/Unit: O20299
/KAP1
Date: 21-06-12 16:02:40
Pages: 7
Synthesis of Glycosylated Eumelanin Building Blocks
Thioglycosidation
An expedient thioglycosidation of 5,6-di-
acetoxyindole with a dynamic mixture of
thioglycosides is described.
M. Adinolfi, M. d’Ischia, A. Iadonisi,*
L. Leone, A. Pezzella, S. Valerio ...... 1–7
Glycosylated Eumelanin Building Blocks
by Thioglycosylation of 5,6-Diacetoxy-
indole with an Expedient Selenium-Based
Dynamic-Mixture Methodology
Keywords: Thioethers
/ Glycosylation /
Glycosides / Selenium / Synthetic methods
Eur. J. Org. Chem. 0000, 0–0
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
7