Synthesis of glycosyl amino acids by light-induced coupling of photoreactive amino acids with glycosylamines and 1-C-aminomethyl glycosides

Article abstract of DOI:10.1016/j.carres.2004.12.023

The glycosylamines of O-acetyl-protected GlcNAc and chitobiose, as well as two partially unprotected 1-C-aminomethyl glucosides, were photochemically coupled with orthogonally protected N-aspartyl-5-bromo-7-nitroindoline derivatives. The reactions proceed

Full text of DOI:10.1016/j.carres.2004.12.023

Carbohydrate
RESEARCH
Carbohydrate Research 340 (2005) 557–566
Synthesis of glycosyl amino acids by light-induced coupling of
photoreactive amino acids with glycosylamines and
1-C-aminomethyl glycosides
Ondrej Simo,b Vincent P. Lee,a Alexander S. Davis,a Christian Kreutz,a Paul H. Gross,b
Patrick R. Jonesb and Katja Michaela,*
aDepartment of Chemistry, University of Hawaii, Honolulu, HI 96822, USA
bDepartment of Chemistry, University of the Pacific, Stockton, CA 95211, USA
Received 5 October 2004; accepted 31 December 2004
Available online 29 January 2005
Abstract—The glycosylamines of O-acetyl-protected GlcNAc and chitobiose, as well as two partially unprotected 1-C-aminomethyl
glucosides, were photochemically coupled with orthogonally protected N-aspartyl-5-bromo-7-nitroindoline derivatives. The reac-
tions proceeded under neutral conditions by irradiation with near-UV light. The glycosyl asparagines with N- or C-glycosyl linkages
were afforded in 60–85% yield on a 10–70 mg scale. Moreover, the ability of a highly photoreactive N-glutamyl-4-methoxy-7-nitro-
indoline derivative to acylate amino saccharides was tested. Upon irradiation in the presence of a dimeric 1-C-aminomethyl glyco-
side, or a glycosylamine, the corresponding glycosyl glutamines were obtained in 50% and 30% yield, respectively. Preparations of
the photoreactive aspartates and the 1-C-aminomethyl glycosides are also described.
ꢀ 2005 Elsevier Ltd. All rights reserved.
Keywords: C-Glycosides; Glycosyl amino acids; Photochemical reaction; Transamidation
1. Introduction
to reduce the amount of aspartimide formed in the con-
vergent synthesis of N-glycopeptides.
In the synthesis of complex biopolymers, mild reaction
conditions for the individual manipulation steps are
not only desirable in order to extend the scope of an
orthogonal protecting group strategy, but they may also
be necessary to avoid or minimize irreversible side reac-
tions. For example, in the convergent synthesis of N-gly-
copeptides, the aspartic acid to be coupled with a
glycosylamine can suffer from substantial aspartimide
Our interest in N-glycopeptides led us to the idea to
explore a base-free photochemical coupling method, at
first on the glycosyl amino acid level, in order to assess
its usefulness for N-glycopeptide synthesis. N-Acyl-7-
nitroindolines are attractive photoreactive species that
are known to acylate nucleophiles upon irradiation with
near-UV light in the absence of tertiary amines. Almost
30 years ago it was discovered that N-acyl-5-bromo-7-
nitroindolines undergo photosolvolysis,4 and this was
followed by their application as light-sensitive protect-
ing groups,5 and their use in fragment condensation in
peptide chemistry.6 In recent times, an N-acyl-7-nitro-
indoline was incorporated into a photolabile linker for
solid-phase synthesis,7 and N-acyl-5,7-dinitroindolines
were applied to acylate aliphatic amines, generating
amides, and carbamates.8,9 Our preliminary results
using N-acyl-5-bromo-7-nitroindolines in glycosyl
formation,
a competing intramolecular cyclization
involving, as a nucleophile, the adjacent amide nitrogen
of the amino acid located at the C-terminal side of
Asp.1–3 Since the extent of aspartimide formation de-
pends partially on the amount of base present in the
coupling mixture, neutral reaction conditions may help
*
Corresponding author. Tel.: +1 808 956 5720; fax: +1 808 956 5908;
e-mail: kmichael@hawaii.edu
asparagine synthesis were recently reported in
a
0008-6215/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.carres.2004.12.023

558
O. Simo et al. / Carbohydrate Research 340 (2005) 557–566
aprotic
organic
solvent
X
+
RCONu
X
X
N
H
Nu
NO2
NO
hν
N
N
R
+
O
N
O
O2N
X
aqueous
medium
O
O
+
H
-
-
+ RCO2 +
R
N
H
1
Scheme 1. Proposed photolysis mechanisms of N-acyl-nitroindolines in the presence of a nucleophile in an inert organic solvent, and in water.13
(X = ring substituent, e.g., 4-methoxy, 5-bromo, etc.)
communication.10 Furthermore, a number of N-acyl-7-
nitroindoline derivatives with varying substituents have
been developed as effective caged compounds, which un-
dergo flash photolysis to release carboxylates in aqueous
reaction media.11–14
OAc
O
AcO
AcO
NH2
OAc
NHAc
2
OAc
The light-induced acylation of nucleophiles in aprotic
organic solvents is believed to proceed via a nitronic
anhydride 1 resulting from acyl transfer to the adjacent
nitro group upon excitation.13 In inert organic solvents,
nucleophiles present in the reaction mixture become acyl-
ated by 1, and recyclable 7-nitroindoline is released. Un-
der aqueous conditions, however, intermediate 1 releases
a carboxylic acid and a 7-nitrosoindole (Scheme 1).11,13
Since glycosyl amino acids and glycopeptides may be of
value for medicinal chemistry, another important issue is
their stability toward enzymatic degradation under physio-
logical conditions. Stability toward enzymatic hydrolysis
can be introduced by the incorporation of foreign struc-
tural elements, such as C-glycosides,15–20 that may convert
glycoconjugates into poor substrates for glycoamidases.
For example, a C-glycopentapeptide with an insertion of
a methylene group between the asparagineÕs side chain
amide group and the anomeric center of GlcNAc showed
resistance toward N-glycanase-catalyzed hydrolysis and
was in fact a potent inhibitor of this enzyme.21
Here we report the photochemical acylation of two
glycosylamines and two 1-C-aminomethyl glycosides
with orthogonally protected N-aspartyl-5-bromo-7-
nitroindoline derivatives under neutral reaction condi-
tions, which produces glycosyl asparagines with natural
N- or unnatural C-glycosyl linkages. We also describe
the photoacylation of amino saccharides using a highly
photoreactive glutamate derivative with an incorporated
4-methoxy-7-nitroindoline moiety.12 The syntheses of
the photoreactive asparagines, and the syntheses of the
1-C-aminomethyl b-D-glucosides used in this study are
also described.
O
O
AcO
AcO
O
NH2
AcO
NHAc
NHAc
3
O
Ph
O
O
HO
NH2
NH
OH
5
O
O
Ph
O
HO
OH
2
9
Figure 1. Amino saccharides used for light-induced glycosyl amino
acid synthesis.
glycosyl amino acid synthesis and N-glycopeptide
synthesis by conventional methods. Recently, we have
subjected 2 and 3, as well as the C-glycosidic amino sac-
charide 5 and its dimeric derivative 9 (Fig. 1) to light-
induced glycosyl asparagine synthesis.10 These glycosyl
asparagines may be useful building blocks for the syn-
thesis of N-glycopeptides and neoglycopeptides. This
was recently demonstrated by means of the N-chitobio-
syl asparagine 14, which was deprotected and used as a
building block in the synthesis of a glycododecapeptide
partial structure of a human cell-adhesion protein.28
2.2. Generation of the 1-C-aminomethyl b-D-gluco-
pyranosides
2. Results and discussion
2.1. General considerations
In an earlier report the reduction of the known 1-C-
nitromethyl b-D-glucopyranoside 4 with elemental iron
in aqueous tetrahydrofuran under a carbon dioxide
atmosphere generated the 1-C-aminomethyl 4,6-O-benz-
ylidene-b-D-glucopyranoside (5) in 78% yield.29,30 We
found that hydrogenation of 4 in the presence of Raney
The O-acetylated GlcNAc and chitobiose derived glyco-
sylamines 222–24 and 325–27 are common reactants in N-

O. Simo et al. / Carbohydrate Research 340 (2005) 557–566
559
nickel furnishes 5 in 83% yield (Scheme 2). Asufficiently
Br
COOAll
NHR
HOOC
pure product that can be used for further synthetic
manipulations without purification is obtained by a sim-
pler work-up.
+
NH
SOCl2
NO2
toluene
DMF
"Bni"
R = Cbz or Fmoc
Br
The dimeric C-glycoside 9 was generated by serendip-
ity in a high-yielding reaction sequence (Scheme 2). In
an attempt to reduce 1-C-nitromethyl b-D-glucopyrano-
side 4 by palladium-catalyzed hydrogenation to 5,
hydroxylamine 6 was obtained instead. Oxidation of 6
with air produced cis/trans oxime 7. Hydrogenation of
7 in the presence of Raney nickel did not afford C-glyco-
side 5, but unexpectedly, the dimeric aminal 8.
60 oC
19-24 h
N
COOAll
O
NHR
NO2
10: R = Cbz (70%)
11: R = Fmoc (81%)
The structure of 8 is supported by mass spectrometry,
elemental analysis, and NMR spectroscopy. The aminal
proton has a chemical shift of 4.29 ppm, and the HSQC
spectrum reveals its direct connectivity to the aminal
carbon at 84.9 ppm. In the HMBC spectrum, which pro-
duces heteronuclear cross peaks between protons and
carbon atoms two and sometimes three bonds apart,
H-1 of glucose ring B (Scheme 3) shows only two cross
peaks, that is, one to C-2, and another one to the aminal
carbon. The only other proton displaying an HMBC
cross peak with the aminal carbon is a proton of the
methylene group of the 1-oxa-3-azacyclohexane ring.
We did not observe a homonuclear 3J coupling between
Scheme 3. Synthesis of photoreactive and orthogonally protected
asparagines.
the two glucose rings lie in approximately perpendicular
planes under the conditions used for this NMR analysis.
Reduction with sodium borohydride converted ami-
nal 8 almost quantitatively into the secondary amine 9.
Even though initially our synthetic efforts were directed
toward a different product, we felt that this dimeric 1-C-
aminomethyl glucoside could be useful in syntheses of
neoglycopeptides mimicking branched disaccharides.
1
the aminal proton and H-1 in the one-dimensional H
NMR spectrum, and in the HOHAHA spectrum only
a weak cross peak between these two protons appears.
3
The small J1,aminal coupling of <1 Hz indicates that
2.3. Generation of the photoreactive asparagine
derivatives
Following a procedure similar to the known amidation
of 5-bromo-7-nitroindoline (Bni),6 the commercially
available Bni was condensed with Cbz-Asp-OAll or
Fmoc-Asp-OAll in the presence of thionyl chloride
and a small amount of DMF to give the photoreactive
amino acids Cbz-Asp(Bni)OAll (10) and Fmoc-As-
p(Bni)OAll (11) in 70% and 81% yield, respectively
(Scheme 3). In organic solvents all N-acyl-5-bromo-7-
nitroindolines have an absorption band in the near-UV
region at ꢀ350 nm, and they clearly show luminescence,
a useful property that facilitates monitoring their forma-
tion or their consumption, as well as their purification
by chromatography. This luminescence is characteristic
for the N-acyl-7-nitroindolines with a 5-bromo substit-
uent. We did not observe it in derivatives in which
the 5-bromo substituent is replaced by a 5-cyano, 5-
carboxymethyl, or a 4-methoxy group.
O
Ph
O
O
R
HO
OH
4
5
6
7
R = CH2NO2
R = CH2NH2
R = CH2NHOH
R = CH=NOH
a
c
b
d
O
O
HO
NH
HO
Ph
O
Ph
2
O
HO
O
O
O
1
2.4. Generation of N- and C-glycosyl amino acids by
phototransamidation
ring B
ring A
8
9
e
The ability of the two photoreactive asparagines 10 and
11 to acylate the weakly nucleophilic glycosylamines 2
and 3, as well as the more nucleophilic and partially
unprotected glucose-derived 1-C-aminomethyl glucosides
5 and 9 was tested. All photochemical experiments were
conducted in THF and/or N,N,N0,N0-tetramethylurea
(TMU) according to Scheme 4. Both solvents are stable
Scheme 2. Synthesis of monomeric and dimeric C-glycosides 5 and 9
used for photoacylation: (a) H2, Raney nickel, MeOH, rt, 15 h; (b) H2,
Lindlar catalyst, MeOH, rt, 8–10 h; (c) air, NH4OH, MeOH, rt, 10–
16 h; (d) H2, Raney nickel, py, H2O, rt, 2–2.5 d; (e) 1. NaBH4, AcOH,
py, MeOH; 2. NH4Cl aq, 3. NaOH, MeOH, H2O (pH 9).

560
O. Simo et al. / Carbohydrate Research 340 (2005) 557–566
Br
O
N
COOAll
NHR
+
NHR'
n
O
NO2
hν (λ > 310 nm)
inert organic
solvent
NHR
COOAll
O
O
rt, 13-24 h
N
n
R'
Scheme 4. General scheme for the synthesis of N- and C-glycosyl asparagines by phototransamidation. R = Cbz or Fmoc, R0 = H or another
C-glycoside, n = 0 or 1.
under the photochemical reaction conditions applied,
and TMU increases the solubility of the polar reactants.
The inexpensive equipment needed comprises a mer-
cury lamp and a Pyrex reaction vessel allowing for water
cooling and inert gas purging. Small-scale reactions can
also be performed by irradiating the reactants in a stan-
dard 5-mm NMR tube. It is important to filter out the
high-energy wavelengths of mercury lamps in these pho-
totransamidations, since otherwise the reactants undergo
photodecomposition. Therefore, quartz glass reactors
are to be avoided, but Pyrex is well suited for this pur-
pose due to its natural cutoff with an onset of 310 nm.
Table 1 summarizes the phototransamidations per-
formed. The amino acids 10 and 11 successfully undergo
phototransamidation with glycosylamines 2 and 3, and
with the 1-C-aminomethyl b-D-glucosides 5 and 9. The
two starting materials were employed in nearly equimo-
lar amounts with one of the two reactants, typically the
amino saccharide component, in 10% excess. Similar to
the condensation of glycosylamines with acid chlorides
of aspartic acid,31 the N-glycosyl Cbz- and Fmoc-pro-
tected allyl asparaginates 12–17 were produced in 60–
85% yield. We did not observe any anomerization in
the sugar or epimerization in the amino acid moiety.
Table 1. Glycosyl asparagines synthesized
Photoreactive
asparagine
Amino
saccharide
Amino acid/sugar
ratio
Solvent
THF
Glycosyl asparagine
Yielda
74%
OAc
H
N
O
10
2
1:1.10
AcO
AcO
COOAll
NHCbz
NHAc
O
1210
OAc
H
O
AcO
AcO
COOAll
N
11
11
2
3
1:1.10
1:1.14
THF
61%
63%
NHAc
O
NHFmoc
13
OAc
OAc
O
O
H
N
AcO
AcO
O
COOAll
THF/TMU 5:1
AcO
NHAc
NHAc
O
NHFmoc
1428
O
NHCbz
COOAll
Ph
O
O
O
O
HO
10
11
10
5
5
9
1.11:1
1:1.11
1:1.13
THF
TMU
THF
70%
60%
85%
N
OH
15
H
O
O
NHFmoc
COOAll
Ph
Ph
O
HO
N
H
OH
16
O
O
NHCbz
COOAll
O
O
HO
N
OH
17
2
a All yields refer to chromatographically purified products.

O. Simo et al. / Carbohydrate Research 340 (2005) 557–566
561
MeO
cally all of the photoreactive starting material 18 had
been consumed, which indicates that due to its high
reactivity, one or more unproductive side reactions were
competing with the desired phototransamidations.
NHBoc
COOtBu
+
9
N
O
NO2
18
hν (λ > 310 nm)
THF-d8
3. Conclusions
rt, 15 h
The N-aspartyl-5-bromo-7-nitroindolines 10 and 11 are
latent nitronic anhydrides, which become activated
upon irradiation with near-UV light. In inert organic
solvents (THF and/or TMU) amino saccharides, includ-
ing the relatively weakly nucleophilic glycosylamines 2
and 3 and the partially unprotected 1-C-aminomethyl
b-D-glucopyranosides 5 and 9, become acylated. The
overall reaction is a phototransamidation, which gener-
ates glycosyl asparagines in fair-to-good yields (60–
85%). Carbohydrate derivative 9 is a novel dimeric
C-glycoside, similar to the one produced in the L-fucose
series.32 It can be efficiently synthesized starting from
the known 4,6-O-benzylidene-protected 1-C-nitromethyl
b-D-glucopyranoside 4 via the novel aminal 8.
In the glycosyl amino acid syntheses shown, the two
reactants are economically utilized in nearly equimolar
amounts, and the 5-bromo-7-nitroindoline liberated is
recyclable. The major strength of this method lies in
its mild activation by light under neutral reaction condi-
tions. Therefore, together with the ability of the 5-bro-
mo-7-nitroindoline group to function as a protecting
group for carboxyl groups in the dark, the phototrans-
amidations demonstrated here pave the way for the
convergent synthesis of glycopeptides with N- and C-
glycosyl linkages.
The known highly photoreactive 4-methoxy-7-nitro-
indoline derivative of glutamic acid (18) produces glyco-
syl glutamines in lower yield (30–50%), especially when
the weakly nucleophilic glycosylamine 2 is utilized.
The fact that 18 is completely consumed upon
irradiation indicates that one or more unproductive side
reactions also take place. We conclude that the 4-meth-
oxy-7-nitroindoline derivatives are too reactive to effi-
ciently perform phototransamidations under our
reaction conditions. Apparently, the N-acyl-5-bromo-
7-nitroindolines do not show these side reactions, at
least not to a large extent, and the natural and unnatural
glycosyl amino acids are generated much more effi-
ciently. Adetailed study of indoline ring substituent
effects on phototransamidations is currently underway.
NHBoc
COOtBu
O
N
O
Ph
O
HO
OH
O
2
19 (50%)
Scheme 5. Light-induced coupling of Boc-Gln(4-methoxy-7-nitroind-
oline)-OtBu with the C-glycosidic secondary amine 9.
Our results indicate that this photochemical method
may be promising for the convergent synthesis of N-gly-
copeptides, that is, the light-induced coupling of nitro-
indoline derivatized peptides with aminosaccharides
under neutral conditions.
In the phototransamidation of Fmoc-protected aspar-
tate derivative 11 and C-glycoside 5, we observed the
presence of approximately 10% dibenzofulvene. Possi-
bly, this partial loss of the Fmoc group is due to the ba-
sicity of the primary amino group in 5. Fmoc cleavage
was not apparent when the less basic glycosyl amines 2
and 3 were used.
The known N-glutamyl-4-methoxy-7-nitroindoline 18
is the protected precursor of a compound that can effi-
ciently release glutamate upon flash photolysis in aque-
ous media.11–13 Its high efficiency in flash photolysis
experiments has been attributed to the electron-releasing
effect of the 4-methoxy substituent para to the 7-nitro
group, which increases the extinction coefficient and
the quantum yield.12 In order to look into the influence
of the ring substituent of N-acyl-7-nitroindolines in pho-
totransamidations, we were interested in studying gluta-
mate derivative 18, which has a 4-methoxy instead of a
5-bromo substituent in the indoline ring. The ability of
this electron-rich photoreactive glutamate to acylate
amino saccharides was tested by irradiation of 18 in
the presence of glycosylamine 2, and in the presence of
1-C-aminomethyl b-D-glucoside 5 in THF. Surprisingly,
the photochemical experiment with glycosylamine 2
showed only 30% of N-glycosyl glutamine formation
based on NMR analysis of the crude mixture. Due to
the inefficient conversion to the N-glycosyl glutamine,
we decided not to isolate any photoproducts from this
reaction. Similarly, the photochemical reaction of 18
with the more nucleophilic amino saccharide 9 gave
the C-glucopyranosyl glutamine 19 in only 50% yield
after chromatography (Scheme 5). In both cases, practi-
4. Experimental
4.1. General methods
1H and 13C NMR spectra were taken at 300 MHz using
a Varian Mercury spectrometer or at 500 MHz using a

562
O. Simo et al. / Carbohydrate Research 340 (2005) 557–566
Varian Unity Inova spectrometer. Protons of diastereo-
topic methylene groups were assigned with or without
ÔprimeÕ, that is, H-6 (downfield) and H-60 (upfield).
Thin-layer chromatography (TLC) was performed on
aluminum or glass plates coated with Silica Gel 60
(2.34 g, 83%). All analytical data concurred with the
ones published earlier.29
4.3. Aminal 8
F
245. TLC plates were developed with a staining solution
Asuspension of 4 (9.34 g, 30 mmol) and Lindlar catalyst
(700 mg) in MeOH (150 mL) was flushed with H2 three
times and stirred under H2 using a rubber balloon at
ambient temperature until silica TLC indicated the con-
sumption of the fast-migrating compound 4 (8–10 h).
The precipitated gray crude product 6, 8.5 g, was filtered
off and used without purification. Air was bubbled
through a stirred suspension of crude 6 (6.0 g) in MeOH
(240 mL) and NH4OHconcd (100 mL) until silica TLC
indicated the consumption of 6 as the slow-migrating
spot (10–16 h). The reaction mixture was filtered and
concentrated to a small volume in vacuo. The white pre-
cipitate (oxime 7, cis/trans mixture) was filtered off and
was recrystallized from H2O to give pure 7 (4.4 g, 70%
over two steps): mp 227–230 ꢁC (dec), Rf 0.34, 0.39
(SiO2; 10:1 CHCl3–EtOH). Raney nickel (0.7 g) was
added to a suspension of 7 (2 g, 6.77 mmol) in pyridine
(10 mL) and H2O (20 mL), and the reaction flask was
evacuated and filled with H2 three times. Hydrogenation
using a rubber balloon at ambient temperature for 50–
60 h produced a precipitate that was filtered off, and
was digested with pyridine (35 mL). The pyridine solu-
tion was filtered from the catalyst and evaporated
in vacuo to a syrup. MeOH precipitated aminal 8, which
[P2O5 · 24MoO3 · xH2O (10 g), (NH4)2Ce(NO3)6 (5 g),
H2O (450 mL), and H2SO4 (50 mL)] and heat. Mass
spectra were measured on a VG Analytical 70SE or a
JEOL LCmate mass spectrometer. Elemental analyses
were carried out at Desert Analytics, Tucson, AZ. Opti-
cal rotations were measured with an O.C. Rudolph &
Sons polarimeter, model 956, or with a Jasco DIP-370
polarimeter. Toluene and THF were dried by standard
methods and distilled. N,N,N0,N0-Tetramethylurea was
freshly distilled. All other solvents were reagent grade.
Cbz-Asp-OAll was synthesized from Cbz-Asp-anhy-
dride and allyl alcohol as previously described for
Cbz-Asp-OBn,33 and Fmoc-Asp-OAll was either synthe-
sized in the same way or purchased from Novabiochem.
5-Bromo-7-nitroindoline was purchased from Lancaster
synthesis.
Phototransamidation reactions were performed at
ambient temperature under argon either in a microscale
photochemical reaction assembly, model 7880, ACE
Glass, pyrex, (reaction scale 20–50 mg), or in an NMR
tube (reaction scale ꢀ10 mg). For irradiation up to four
mercury gaseous discharge lamps (Pen-Ray, 5.5 W, low
pressure, cold cathode) were used. In the photochemical
reaction assembly, the reaction mixtures were purged
with argon for 20 min prior to irradiation. For small-
scale phototransamidations the NMR tubes containing
the reaction mixtures were thoroughly flushed with
argon and tightly capped. During irradiations the reac-
tion mixtures were kept at ambient temperature by
water cooling. Irradiations were continued until TLC
or NMR spectroscopy indicated the consumption of
the photoreactive amino acids. Upon completion, the
solvents were evaporated under reduced pressure, and
the crude mixtures were chromatographed on silica gel
(230–400 mesh, Natland Int. Corporation).
was filtered off, washed with MeOH, and dried to give
22
D
a white solid (1.43 g, 78%): mp 223–225 ꢁC (dec), ½aꢁ
ꢂ55 (pyridine), Rf 0.41 (SiO2; 10:1 CHCl3–EtOH). Since
the molecule is unsymmetrical, the sugar ring with the
two free hydroxyl groups is designated ÔAÕ, the other
1
one ÔBÕ (Scheme 2) for NMR assignments. H NMR
(500 MHz, Me2SO-d6, 298 K): d 7.47–7.42 (m, 4H,
arom.); 7.39–7.35 (m, 6H, arom.); 5.58 (s, 1H, CH–
3
Ph); 5.57 (s, 1H, CH–Ph); 5.25 [d, 1H, J3,OH-3 4.7 Hz,
3
OH-3 (A)]; 5.18 [d, 1H, J3,OH-3 4.6 Hz, OH-3 (B)];
3
5.16 [d, 1H, J2,OH-2 5.5 Hz, OH-2 (A)]; 4.29 (d, 1H,
3JCH,NH 11.4 Hz, O–CH–N); 4.20–4.14 [m, 2H, H-
3
6(A), H-6(B)]; 3.70 [dd, 1H, J5,6 = 2J6,6 10.1 Hz, H-
0
4.2. Preparation of 1-C-aminomethyl 4,6-O-benzyli-
dene-b-D-glucopyranoside (5)
60(A)]; 3.69–3.58 [m, 2H, H-60(B), H-3(B)]; 3.53 [m,
1H, H-2(A)]; 3.48–3.40 [m, 3H, H-4(B), H-5(B), H-
3
3(A)], 3.37 [dd, 1H, J3,4 = 3J4,5 9.1 Hz, H-4(A)]; 3.33
[m, 1H, H-5(A), overlapped with H2O]; 3.28 [dd, 1H,
Raney nickel (slurry, 3 g) was added to a suspension of
nitro derivative 4 (3.11 g, 10 mmol) in MeOH (40 mL).
The reaction mixture was degassed by evacuating the
flask and filling with hydrogen three times and then stir-
red under H2 using a rubber balloon until silica TLC
(6:1 CHCl3–EtOH) indicated that the reaction was com-
plete (ꢀ15 h). The white product was dissolved in boil-
ing MeOH, the catalyst was filtered off, and the
methanolic filtrate was concentrated in vacuo to a small
volume. The amine 5 was filtered off after cooling,
washed with a minimal amount of MeOH, and dried
3
3J1,2 9.8 Hz; J1,O–CH–N 1.7 Hz, H-1(A)]; 3.20 [m, 1H,
H-1(B)]; 3.16–3.10 [m, 2H, H-2(B), CH (C-1–CH2–N)];
2.64–2.53 [m, 2H, NH, CH0 (C-1–CH2–N)]; 13C NMR:
(125.8 MHz, Me2SO-d6, 298 K): d 137.8, 137.7 (arom.,
quart.); 128.8, 128.7, 128.0, 126.4, 126.3 (arom.);
100.9, 100.5 (CH–Ph); 84.9 (O–CH–N); 82.0, 81.9
[C-2(B), C-4(B)]; 80.7 [C-4(A)]; 80.6 [C-1(A)]; 74.2 [C-
3(A)]; 73.6 [C-1(B)]; 70.8 [C-3(B)]; 70.5 [C-5(A), C-
5(B)]; 69.9 [C-2(A)]; 67.9 [C-6(A)]; 67.8 [C-6(B)]; 46.9
(C-1–CH2–N). ESIMS: (m/z): [M+H]+ calcd for

O. Simo et al. / Carbohydrate Research 340 (2005) 557–566
563
C28H34NO10, 544.2; found, 544.2. Anal. Calcd for
C28H33NO10: C, 61.87; H, 6.12; N, 2.58. Found: C,
61.21; H, 6.31; N, 2.52.
was chromatographed on silica with 1:1 EtOAc–hex-
anes. The desired Cbz-Asp(Bni)-OAll (10) was obtained
as a brown syrup (241 mg, 70%). TLC: Rf 0.31 (SiO2; 1:1
EtOAc–hexanes). The NMR spectrum showed absence
of extraneous lines except for the H2O signal at
3.35 ppm and the Me2SO-d6 signal at 2.50 ppm. 1H
NMR (300 MHz, Me2SO-d6, 298 K): d 7.86, 7.84
4.4. N,N-[Bis-C-(4,6-O-benzylidene-b-D-glucopyranosyl
methyl)]amine (9)
3
NaBH4 (151.3 mg, 4 mmol) was added to a stirred sus-
pension of aminal 8 (1.087 g, 2 mmol) in 2:1 pyridine–
HOAc (10 mL) and MeOH (8 mL) at rt during 20 min.
The resulting solution was stirred for 1 h. Solvents were
removed under reduced pressure. Diethyl ether (40 mL)
was added to the remaining gel, and the suspension was
vigorously stirred for 2 h, filtered, and washed with
ether. The crude product was dissolved in MeOH
(15 mL) under reflux, and NH4Cl (428 mg, 8 mmol) in
H2O (10 mL) was added. Awhite product precipitated
after a few minutes. The suspension was refluxed for
an additional 5 min, cooled, and filtered. The white
hydrochloride was washed with water and dried
(966 mg, 83%). NaOH (2 N, 0.52 mL, 1.04 mmol) was
added to a stirred suspension of the hydrochloride
(582 mg, 1 mmol) in MeOH (20 mL) and H2O (7 mL).
The resulting solution (pH 9) was then stirred at rt for
20 min. MeOH was removed from the reaction mixture
under reduced pressure, H2O (20 mL) was added, and
(2 · m, 2H, H-4, H-6, indoline); 7.69 (d, 1H, JNH,a
8.1 Hz, NH); 7.37–7.29 (m, 5H, arom., Cbz); 5.89
3
(m, 1H, –CH@CH2); 5.31 (dtd, 1H, Jtrans 17.3 Hz,
3
–CH@CH2, trans); 5.20 (dtd, 1H, Jcis 10.5 Hz,
–CH@CH2, cis); 5.04 (s, 2H, –CH2–Ph); 4.59 (m, 2H,
3
–O–CH2–, allyl); 4.54 (m, 1H, H-a); 4.25 (t, 2H, J2,3
8.1 Hz, H-2, H-20, indoline); 3.20 (m, 2H, H-3, H-30,
2
indoline); 3.07 (dd, 1H, Jb,b 16.8 Hz, Ja,b 5.4 Hz, H-
3
0
3
b); 2.96 (dd, 1H, Ja,b 7.8 Hz, H-b0); 13C NMR
0
(126 MHz, Me2SO-d6, 298 K): d 170.8, 168.1 [2 · C@O
(amide, ester)]; 155.8 [C@O (Cbz)]; 143.7, 136.8, 133.2
(arom.); 132.2 [–CH@CH2 (allyl)]; 132.0 (C-4, indoline);
128.3, 127.8, 127.6 (arom.); 124.3 (C-6, indoline); 117.7
(–CH@CH2, allyl); 115.4 (arom.); 65.6 (CH2, Cbz);
65.2 (O–CH2–, allyl); 50.4 (C-a); 49.4 (C-2, indoline);
36.5 (C-b); 28.3 (C-3, indoline). FABMS: (m/z):
[M+H]+ calcd for C23H23BrN3O7 532.07; found, 532.2.
4.6. N-a-(Fluoren-9-ylmethoxycarbonyl)-b-(5-bromo-7-
nitroindolin-1-yl)-L-aspartamide-a-allyl ester (11)
the product was filtered off, washed with H2O, and dried
22
D
(490 mg, 90%): mp 233–234 ꢁC, ½aꢁ ꢂ44 (DMF), Rf
0.21 (SiO2; 10:1 CHCl3–EtOH). The assignment of the
proton NMR signals were made based on the HOHA-
HAand 1D TOCSY spectra. 1H NMR (500 MHz,
THF-d8, 298 K): d 7.47–7.46 (m, 4H, arom.); 7.31–7.26
(m, 6H, arom.); 5.50 (s, 2H, 2 · CH–Ph); 4.20 (dd, 2H,
Fmoc-Asp-OAll (246 mg, 0.62 mmol) and 5-bromo-7-
nitroindoline (302 mg, 1.25 mmol) were suspended in
anhydrous toluene (6 mL) and DMF (0.01 mL,
0.13 mmol) and warmed to 60 ꢁC under argon. Freshly
distilled SOCl2 (0.081 mL, 0.93 mmol) was injected,
and the mixture was stirred at 60 ꢁC. After 15 h silica
TLC indicated an incomplete consumption of Fmoc-
Asp-OAll (Rf 0.34, SiO2; 9:1 CHCl3–MeOH). More
SOCl2 (0.03 mL, 0.31 mmol) was injected, and the mix-
ture was allowed to react for additional 8 h. The reac-
tion mixture was allowed to cool and was quenched by
addition of H2O (20 mL). The mixture was extracted
3· with EtOAc, and the crude product was chromato-
graphed on silica with 3:4 EtOAc–hexanes. The desired
Fmoc-Asp(Bni)-OAll (11) was obtained as a yellow
foam (311 mg, 81%). TLC: Rf 0.20 (SiO2; 3:4 EtOAc–
hexanes). The NMR assignments were made based on
HOHAHA and HSQC spectra. 1H NMR (500 MHz,
Me2SO-d6, 298 K): d 7.87 (d, 2H, 3J 7.6 Hz, Fmoc);
7.84, 7.82 (2 · s, 2H, H-4, H-6, indoline); 7.72 (d, 1H,
3JNH,a 8.1 Hz, NH); 7.68 (d, 2H, 3J 7.5 Hz, arom.
3
3
0
J6,6 10.3 Hz, J5,6 4.3 Hz, 2 · H-6); 3.62 (2H, m,
2 · H-60); 3.51 (2H, dd, J2,3 = 3J3,4 8.5 Hz, 2 · H-3);
3
3.40 (m, 2H, 2 · H-1); 3.37–3.30 (m, 6H, 2 · H-2,
2
0
2 · H-4, 2 · H-5); 2.96 [dd, 2H, JCH–N,CH –N 12.3 Hz,
3J1,CH–N 3.7 Hz, 2 · CH (C-1–CH2–N)]; 2.79 [dd, 2H,
3
J1,CH –N 6.2 Hz, 2 · CH0 (C-1–CH2–N)] ppm. APC-
0
IMS: (m/z): [M+H]+ calcd for C28H36NO10, 546.2;
found, 546.3. Anal. Calcd for C28H35NO10: C, 61.64;
H, 6.47; N, 2.57. Found: C, 61.71; H, 6.81; N, 2.59.
4.5. N-a-Benzyloxycarbonyl-b-(5-bromo-7-nitroindolin-1-
yl)-L-aspartamide-a-allyl ester (10)
Cbz-Asp-OAll (200 mg, 0.65 mmol) and 5-bromo-7-
nitroindoline (302 mg, 1.25 mmol) were suspended in
anhydrous toluene (6 mL) and DMF (0.01 mL,
0.13 mmol) and warmed to 60 ꢁC under argon. Freshly
distilled SOCl2 (0.081 mL, 0.93 mmol) was injected,
and the mixture was stirred at 60 ꢁC. After 19 h TLC
analysis indicated the consumption of Cbz-Asp-OAll.
The reaction mixture was allowed to cool and was
quenched by addition of H2O (20 mL). The mixture
was extracted 3· with EtOAc, and the crude product
3
Fmoc); 7.40 (t, 2H, J 7.4 Hz, arom. Fmoc); 7.30 (m,
2H, arom. Fmoc); 5.87 (m, 1H, –CH@CH2); 5.29 (dtd,
3
1H, Jtrans 17.2 Hz, –CH@CH2, trans); 5.17 (dd, 1H,
3Jcis 10.5 Hz, –CH@CH2, cis); 4.57 (m, 2H, O–CH2–, al-
lyl); 4.52 (m, 1H, H-a); 4.33–4.28 (m, 2H, CH2, Fmoc);
4.27–4.20 (m, 3H, H-2, H-20, indoline, CH, Fmoc); 3.20
2
(m, 2H, H-3, H-30, indoline); 3.05 (dd, 1H, Jb,b
0

564
O. Simo et al. / Carbohydrate Research 340 (2005) 557–566
3
3
0
16.7 Hz, Ja,b 5.3 Hz, H-b); 2.96 (dd, 1H, Ja,b 7.7 Hz,
H-b0). 13C NMR (126 MHz, Me2SO-d6, 298 K): d
170.8, 168.1 [2 · C@O (amide, ester)]; 155.8 [C@O
(Fmoc)]; 143.7, 140.7, 140.3, 133.2 (4 · arom., no CH);
132.2 (–CH@CH2, allyl); 132.0 (C-4, indoline); 127.6
(CH, arom. Fmoc); 127.08, 127.06 (CH, arom. Fmoc
and arom., no CH); 125.2 (CH, arom. Fmoc); 124.3
(C-6, indoline); 120.1 (CH, arom. Fmoc); 117.7 (–CH@
CH2, allyl); 115.4 (arom.); 65.8 (CH2, Fmoc); 65.2
(–O–CH2–, allyl); 50.4 (C-a); 49.4 (C-2, indoline); 46.6
(CH, Fmoc); 36.5 (C-b); 28.4 (C-3, indoline). FABMS:
(m/z): [M+H]+ calcd for C30H27BrN3O7, 620.1; found,
620.1; [M+Li]+ calcd for C30H26BrLiN3O7, 626.1;
found, 626.1. Anal. Calcd for C30H26BrN3O7: C,
58.07; H, 4.22; N, 6.77; Br, 12.88; O, 18.05. Found: C,
57.65; H, 4.04; N, 6.66; Br, 13.25; O, 18.42.
4.9. Preparation of N-c-[2-acetamido-3,6-di-O-acetyl-2-
deoxy-4-O-(2-acetamido-3,4,6-tri-O-acetyl-2-deoxy-b-D-
glucopyranosyl)-b-D-glucopyranosyl]-N-a-fluoren-9-
ylmethoxycarbonyl-L-asparagine allyl ester (14)28
The photoreactive amino acid 11 (51 mg, 0.082 mmol)
and glycosylamine 3 (59 mg, 0.093 mmol) were dissolved
in 5 mL of anhyd THF and 1 mL of TMU. The mixture
was irradiated for 13 h under a continuous argon flow.
During the course of the reaction the product partially
precipitated out. After evaporation of the solvent the
crude product was chromatographed on SiO2 with 9:1
CHCl3–MeOH. The purified glycosyl amino acid 14
was obtained as a white solid (52 mg, 63%): TLC: Rf
0.30 (SiO2; 9:1 CHCl3–MeOH); FABMS: (m/z):
[M+H]+ calcd for C48H59N4O20, 1011; found, 1011;
[M+K]+ calcd for C48H58KN4O20, 1049; found, 1049.
1
The H and 13C NMR spectra matched the ones previ-
4.7. Preparation of N-c-(2-acetamido-3,4,6-tri-O-acetyl-
2-deoxy-b-D-glucopyranosyl)-N-a-benzyloxycarbonyl-L-
asparagine allyl ester (12)
ously reported for 14.28
4.10. N-c-(4,6-O-Benzylidene-b-D-glucopyranosylmeth-
yl)-N-a-benzyloxycarbonyl-L-asparagine allyl ester (15)
The synthesis and characterization of this glycosyl
asparagine has been described.10 ½aꢁ +2 (c 1, CHCl3).
20
D
The photoreactive amino acid 10 (71.2 mg, 0.134 mmol)
and
1-C-aminomethyl
glucoside
5
(33.8 mg,
4.8. N-c-(2-Acetamido-3,4,6-tri-O-acetyl-2-deoxy-b-D-
glucopyranosyl)-N-a-fluoren-9-ylmethoxycarbonyl-L-
asparagine allyl ester (13)
0.120 mmol) were dissolved in 10.5 mL of anhyd THF,
and the mixture was irradiated for 24 h under a contin-
uous argon flow. After evaporation of the solvent, the
crude product was chromatographed with 3:1 EtOAc–
The photoreactive amino acid 11 (50 mg, 0.081 mmol)
and glycosylamine 2 (30.7 mg, 0.089 mmol) were dis-
solved in 10 mL of dry THF. The mixture was irradiated
for 14 h under a continuous argon flow. After evapora-
tion of the solvent the crude product was chromato-
graphed on SiO2 with 97.5:2.5 CH2Cl2–MeOH. The
purified glycosyl amino acid 14 was obtained as a white
solid (35 mg, 61%). TLC: Rf 0.61 (SiO2; 9:1 CHCl3–
MeOH). The NMR spectrum showed absence of extra-
neous lines except for the H2O signal at 3.35 ppm and
hexanes. The purified glycosyl amino acid 15 was ob-
20
D
tained as a colorless film (48 mg, 70%): ½aꢁ +38 (c 1,
CHCl3); TLC: Rf 0.16 (SiO2; 3:1 EtOAc–hexanes); Rf
0.40 (SiO2; 9:1 CHCl3–MeOH). The NMR spectrum
showed absence of extraneous lines except for the H2O
signal at 3.30 ppm and the Me2SO-d6 signal at
2.50 ppm. 1H NMR: (300 MHz, Me2SO-d6, 298 K) d
3
7.98 (dd, 1H, Jc,CHN = 3Jc,CH N 4.8 Hz, NH-c); 7.65
0
(m, 1H, H-a); 7.46–7.25 (m, 10H, arom.); 5.86 (m, 1H,
–CH@CH2, allyl); 5.54 (s, 1H, Ph–CH, benzylidene);
5.34–5.22 [m, 3H, OH-2, OH-2, –CH@CH2 (allyl,
1
the Me2SO-d6 signal at 2.50 ppm. H NMR (300 MHz,
3
Me2SO-d6, 298 K): 8.69 (2 · d, 1H, NH-1); 7.88 (2 · d,
1H, NH-1); 7.98–7.84 (m, 3H, NH-2, 2 arom.); 7.78–
7.60 (m, 3H, NH-a, 2 arom.); 7.41 (dd, 2H, 2 arom.);
7.32 (dd, 2H, 2 arom.); 5.86 (m, 1H, –CH2–CH@CH2);
trans)]; 5.18 (d, 1H, Jcis 10.8 Hz, –CH@CH2, allyl,
cis); 5.02 (s, 2H, CH2, Cbz); 4.55 (m, 2H, –O–CH2, al-
lyl); 4.47 (m, 1H, H-a); 4.17 (m, 1H, H-6); 3.70–3.54
[m, 2H, H-60, CH (C-1–CH2–N)]; 3.47–3.20 (m, 4H,
3
5.28 (d, 1H, Jtrans 17.1 Hz, terminal allyl, trans); 5.22–
5.12 (m, 3H, terminal allyl, cis, H-1); 5.10 (ddd, 1H,
H-3, H-4, H-5, H-1, overlapped with H2O); 3.10–2.90
2
[m, H-2, CH0 (C-1–CH2–N)]; 2.64, (dd, 1H, Jb,b
0
3
3J3,4 = 3J4,5 9.9 Hz, H-4); 4.82 (dd, 1H, J2,3 9.9 Hz, H-
16.5 Hz, H-b); 2.55 (m, 1H, H-b0). FABMS: (m/z):
[M+H]+ calcd for C29H35N2O10, 571; found, 571;
[M+K]+ calcd for C29H34KN2O10, 609; found, 609.
3); 4.63–4.42 (m, –CH2–CH@CH2, H-a); 4.38–4.10 (m,
4H, –CH2–CH– of Fmoc, H-6, H-60); 4.00–3.77 (m,
3H, –CH2–CH– of Fmoc, H-2, H-5); 2.75–2.44
(m, 2H, H-b, H-b0); 1.99, 1.97, 1.96 (3 · s, 6H, 2 · Ac);
1.90 (s, 3H, Ac); 1.74, 1.72 (2 · s, 3H, Ac).
FABMS: (m/z): [M+H]+ calcd for C36H42N3O13, 724;
found, 724; [M+Li]+ calcd for C36H41LiN3O13,
730; found, 730; [M+K]+ calcd for C36H41KN3O13,
762; found, 762.
4.11. N-c-(4,6-O-Benzylidene-b-D-glucopyranosyl
methyl)-N-a-fluoren-9-ylmethoxycarbonyl-L-asparagine
allyl ester (16)
Photoreactive amino acid 11 (50 mg, 0.080 mmol) and
C-glycoside 5 (25 mg, 0.089 mmol) were dissolved in

O. Simo et al. / Carbohydrate Research 340 (2005) 557–566
565
10 mL of freshly distilled TMU and irradiated for 24 h.
After evaporation of the solvent under reduced pressure,
the crude product was chromatographed on SiO2 with
97:3 CHCl3–MeOH. The purified C-glycosyl amino acid
16 was isolated as a white solid: 32 mg, 60% yield. TLC:
Rf 0.5 (SiO2; CHCl3–MeOH 9:1). Compound 16 showed
two equally populated conformers in Me2SO, most
likely due to cis/trans isomerization of the carbamate.
The NMR spectrum showed the absence of extraneous
lines except for the H2O signal at 3.30 ppm and the
(Cbz)]; 4.62–4.51 [m, 3H, –O–CH2– (allyl), H-a]; 4.19–
4.08 (m, 2H, 2 · H-6); 4.02, 3.98 [2 · d, 1H, CH (C-1–
CH2–N)]; 3.81, 3.74 [2 · d, 1H, CH (C-1–CH2–N)];
3.69–3.58 (m, 2H, 2 · H-60); 3.58–3.33 [m, 7H, 2 · H-
1, 2 · H-3, CH0 (C-1–CH2–N), 2 · H-4, overlapped with
H2O]; 3.32–3.12 [m, 3H, 2 · H-5, CH0 (C-1–CH2–N)];
3.07–2.94 (m, 2H, 2 · H-2); 2.94–2.81 (m, 2H, H-b, H-
b0); 13C NMR (126 MHz, Me2SO-d6, 298 K): d 171.4,
171.2 (C@O, a); 170.0, 169.8 (C@O, b); 155.80, 155.77
(C@O, Cbz); 137.81, 137.76, 136.9, 136.8 (arom.);
132.4 [–CH@CH2 (allyl)]; 128.8, 128.4, 128.3, 128.0,
127.9, 127.8, 127.7, 126.3 (arom.); 117.5, 117.4
(–CH@CH2, allyl); 100.64, 100.61 [Ph–CH (benzylid-
ene)]; 80.9, 80.8, 80.7 (C-4); 80.0, 79.9, 79.2, 79.1 (C-
1); 74.0, 73.82, 73.77 (C-3); 73.2, 73.1, 72.5, 72.3 (C-2);
70.1, 70.0, 69.9 (C-5); 67.94, 67.87 (C-6); 65.6 [Ph–
CH2–O– (Cbz)]; 64.9 [–O–CH2– (allyl)]; 50.8 (C-a);
50.27 (C-1–CH2–N); 48.8 (C-1–CH2–N); 34.8 (C-b).
HRFABMS: (m/z): [M+H]+ calcd for C43H51N2O15,
835.3289; found, 835.3336.
Me2SO-d6 signal at 2.50 ppm. 1H NMR (300 MHz,
3
Me2SO-d6, 298 K)
d
7.99 (dd, 1H, JNH-c,CHN
=
JNH-c,CH N 5.4 Hz, NH-c); 7.88 (d, 2H, 3J 7.5 Hz, arom.
Fmoc); 7.76–7.65 (m, 3H, NH-a, arom. Fmoc); 7.46–
7.27 [m, 9H, arom. Fmoc, Ph]; 5.86 [m, 1H, –CH@CH2
(allyl)]; 5.55, 5.53 (2 · s, 1H, benzylidene); 5.34–5.23 [m,
3H, CH@CH2 (allyl, trans), OH-3, OH-2]; 5.17 [dd,
3
0
3
1H, Jcis 10.5 Hz, CH@CH2 (allyl, cis)]; 4.53 [d, 2H,
–O–CH2– (allyl)]; 4.47 (m, 1H, H-a); 4.33–4.11 [m, 4H,
CH, CH2 (Fmoc), H-6]; 3.71–3.58 [m, 2H, H-60, CH
(C-1–CH2–N)]; 3.46–3.22 (m, 4H, H-3, H-4, H-5, H-1,
overlapped with H2O); 3.11–2.90 [m, 2H, H-2, CH0
(C-1–CH2–N)]; 2.72–2.48 (m, 2H, H-b, H-b0). FABMS:
(m/z): [M+H]+ calcd for C36H39N2O10, 559.26; found,
559.3. [M+Li]+ calcd for C36H38LiN2O10, 665.27; found,
665.3.
4.13. N-d,N-d-[Bis-C-(4,6-O-benzylidene-b-D-glucopyr-
anosylmethyl)]-N-a-tert-butoxycarbonyl-L-glutamine
tert-butyl ester (19)
The photoreactive glutamate 18 (11.6 mg, 0.024 mmol)
and
1-C-aminomethyl
glucoside
9
(19.8 mg,
4.12. N-c,N-c-[Bis-(4,6-O-benzylidene-b-D-glucopyrano-
syl methyl)]-N-a-benzyloxycarbonyl-L-asparagine allyl
ester (17)
0.036 mmol) were dissolved in THF-d8 in an NMR tube
under argon. The mixture was irradiated for 15 h with
four mercury lamps. The solvent was evaporated under
reduced pressure, and the crude product was chromato-
graphed on SiO2 with 95:5 CHCl3–MeOH. The C-glyco-
syl glutamine 19 was obtained as a white solid (10 mg,
50%): TLC: Rf 0.46 (SiO2; 95:5 CHCl3–MeOH). The
NMR spectrum showed the absence of extraneous lines
except for the THF signals at 3.58 and 1.73 ppm. The
two diastereotopic glucose spin systems, designated a
and b, were identified based on 1D TOCSY and HOHA-
The photoreactive amino acid 10 (9 mg, 0.017 mmol)
and the dimeric C-glycoside 9 (10.4 mg, 0.019 mmol)
were dissolved in 0.6 mL THF-d8 in a 5-mm NMR tube
under argon. The mixture was irradiated with four mer-
cury lamps for 20 h. The solvent was evaporated, and
the crude product was flash chromatographed on SiO2
with 95:5 CHCl3–MeOH. The C-glycosyl amino acid
was obtained as a white film (12 mg, 85%). TLC: Rf
0.20 (SiO2; 95:5 CHCl3–MeOH). The NMR spectrum
showed the absence of extraneous lines except for the
H2O signal at 3.35 ppm and the Me2SO-d6 signal at
2.50 ppm. The NMR spectra of 17 are quite complex
due to two diastereotopic glucose rings, as well as the
existence of two equally populated conformers. Due to
partial signal overlap in the HOHAHA, HSQC, and
HMBC spectra, the two conformers and the signals of
the diastereotopic glucose rings were not separately as-
signed. 1H NMR (500 MHz, Me2SO-d6, 298 K): d
7.39–7.28 (m, 16H, NH, arom.); 5.88 [m, 1H, –CH@
CH2 (allyl)]; 5.56, 5.55 [2s, 2H, 2 · Ph–H (benzylidene)];
5.44 (m, 1H, OH-2); 5.35 (d, 1H, 3JOH-3,3 4.9 Hz, OH-3);
HAspectra.
1H NMR (500 MHz, THF-d8 + D2O,
298 K): d 7.47–7.45 (m, 4H, arom.); 7.35–7.24 (m, 6H,
arom.); 6.38 (d, NH, not fully deuterium exchanged);
5.51 (s, 1H, CH–Ph); 5.49 (s, 1H, CH–Ph); 4.24 [dd,
3
3
0
1H, J6,6 10.4 Hz, J5,6 4.7 Hz, H-6 (a)]; 4.19 [dd, 1H,
3
3
J6,6 10.2 Hz, J5,6 4.8 Hz, H-6 (b)]; 3.97 (dd, 1H, Ja,b
3
0
3
0
9.3 Hz, Ja,b 4.7 Hz, H-a); 3.90 [m, 1H, CH (C-1–
CH2–N) (a)]; 3.82 [d, 1H, JCHN,CH N 13.8 Hz, CH (C-
2
0
1–CH2–N) (b)]; 3.69–3.61 [m, 3H, H-60 (a), H-60 (b),
CH0 (C-1–CH2–N) (b)]; 3.60–3.51 [m, 5H, CH0 (C-1–
CH2–N) (a), H-3 (a), H-3 (b), H-1 (a), H-1 (b)]; 3.41–
3.27 [m, 4H, H-4 (a), H-4 (b), H-5 (a), H-5 (b)], 3.17
3
[m, 1H, H-2 (a)]; 3.12 [dd, 1H, J1,2 = 3J2,3 9.0 Hz, H-2
(b)]; 2.65–2.51 (m, 2H, H-c, H-c0); 2.03 (m, 1H, H-b);
1.85 (m, 1H, H-b0); 1.45 (s, 9H, t-Bu), 1.43 (s, 9H, t-
Bu) ppm. FABMS: (m/z): [M+H]+ calcd for
C42H59N2O15, 831.3915; found, 831.3923.
3
5.30 [d, 1H, Jtrans 17.3 Hz, –CH@CH2 (allyl, trans)];
5.27 (1H, d, JOH-3,3 4.6 Hz, OH-3); 5.22–5.16 [m, 2H,
3
–CH@CH2 (allyl, cis), OH-2]; 5.09–5.00 [m, 2H, CH2

566
O. Simo et al. / Carbohydrate Research 340 (2005) 557–566
12. Papageorgiou, G.; Corrie, J. E. T. Tetrahedron 2000, 56,
8197–8205.
Acknowledgements
13. Morrison, J.; Wan, P.; Corrie, J. E. T.; Papageorgiou, G.
Photochem. Photobiol. Sci. 2002, 1, 960–969.
14. Papageorgiou, G.; Lukeman, M.; Wan, P.; Corrie, J. E. T.
Photochem. Photobiol. Sci. 2004, 3, 366–373.
15. Levy, D. E.; Tang, C. The Chemistry of C-Glycosides, 1st
ed.; Pergamon: Oxford, 1995.
16. Meo, P.; Osborn, H. M. I. The Synthesis of C-Linked
Glycosides. In Carbohydrates; Osborn, H. M. I., Ed.;
Elsevier Science: Oxford, 2003; pp 337–384.
We are grateful to Wesley Yoshida for the recording of
TOCSY, HMBC and HSQC NMR spectra, and to Mike
Burger for mass spectrometric analysis. We also thank
Dr. John Corrie for supplying the glutamine derivative
18 and for helpful discussions. Acknowledgment is made
to the Donors of the American Chemical Society Petro-
leum Research Fund for support of this research given
to K.M. (PRF Type G 36892-G4).
´ ´
17. Gyo¨rgydeak, Z.; Pelyvas, I. F. Glycoscience 2001, 1, 691–
747.
18. Gross, P. H.; Franz, A. H.; Samoshin, V. V. Recent Res.
Dev. Org. Chem. 2004, 8, 255–280.
Supplementary data
19. Gross, P. H. Carbohydr. Polym. 1998, 37, 215–217.
20. Gross, P. H.; Samoshin, V. V.; Franz, A.; Gallagher, J.;
Gremyachinskij, D.; Lu, J.; Phiasivongsa, P.; Simo, O.;
Wei, Y. Q. Glycobiology 2001, 11, 907–908.
21. Wang, L.-X.; Tang, M.; Suzuki, T.; Kitajima, K.; Inoue,
Y.; Inoue, S.; Fan, J.-Q.; Lee, Y. C. J. Am. Chem. Soc.
1997, 119, 11137–11146.
1H NMR spectra for compounds 10, 12–17, and 19 are
provided in the Supplemental data section, which is
available in the electronic version of this paper. Supple-
metary data associated with this article can be found,
in the online version, at doi:10.1016/j.carres.2004.12.023.
22. Bolton, C. H.; Hough, L.; Khan, M. Y. Biochem. J. 1966,
101, 184–190.
References
´
´
23. Szilagyi, L.; Gyo¨rgydeak, Z. Carbohydr. Res. 1985, 143,
21–41.
24. Waldmann, H.; Kunz, H. Liebigs Ann. Chem. 1983, 1712–
1725.
25. Spinola, M.; Jeanloz, R. W. J. Biol. Chem. 1970, 245,
4158–4162.
1. Bodanszky, M.; Natarajan, S. J. Org. Chem. 1975, 40,
2495–2499.
2. Bodanszky, M.; Kwei, J. Z. Int. J. Pept. Protein Res. 1978,
12, 69–74.
3. Tam, J. P.; Riemen, M. W.; Merrifield, R. B. Pept. Res.
1988, 1, 6–18.
26. Kunz, H.; Waldmann, H. Angew. Chem., Int. Ed. Engl.
1985, 24, 883–885.
4. Amit, B.; Ben-Efraim, D. A.; Patchornik, A. J. Am. Chem.
Soc. 1976, 98, 843–844.
27. Kunz, H.; Waldmann, H.; Marz, J. Liebigs Ann. Chem.
1989, 45–49.
¨
5. Goissis, G.; Erickson, B. W.; Merrifield, R. B. Pept., Proc.
Am. Pept. Symp., 5th 1977, 559–561.
28. Wagner, M.; Dziadek, S.; Kunz, H. Chem. Eur. J. 2003, 9,
6018–6030.
6. Pass, S.; Amit, B.; Patchornik, A. J. Am. Chem. Soc. 1981,
103, 7674–7675.
29. Drew, K. N.; Gross, P. H. Tetrahedron 1991, 47, 6113–
6126.
7. Nicolaou, K. C.; Safina, B. S.; Winssinger, N. Synlett
2001, SI, 900–903.
30. Wang, X.; Gross, P. H. J. Org. Chem. 1995, 60, 1201–
1206.
8. Helgen, C.; Bochet, C. G. Synlett 2001, 1968–1970.
9. Helgen, C.; Bochet, C. G. J. Org. Chem. 2003, 68, 2483–
2486.
10. Vizvardi, K.; Kreutz, C.; Davis, A. S.; Lee, V. P.; Philmus,
B. J.; Simo, O.; Michael, K. Chem. Lett. 2003, 32, 348–349.
11. Papageorgiou, G.; Ogden, D. C.; Barth, A.; Corrie, E. E.
T. J. Am. Chem. Soc. 1999, 121, 6503–6504.
31. Christiansen-Brams, I.; Meldal, M.; Bock, K. J. Chem.
Soc., Perkin Trans. 1 1993, 1461–1471.
32. Phiasivongso, P.; Gallagher, J.; Chen, C.-N.; Jones, P. R.;
Samoshin, V. V.; Gross, P. H. Org. Lett. 2002, 4, 4587–
4590.
33. Bergmann, M.; Zervas, L.; Salzmann, L. Ber. Dtsch.
Chem. Ges. 1933, 66, 1288–1290.