A serendipitous discovery of a new C-furanosyl glycine synthesis via thiazole-based aminohomologation of hexopyranoses

Article abstract of DOI:10.1055/s-2007-967993

Ring closure via microwave-assisted intramolecular OMs displacement by a γ-OBn group (O-nucleophilic attack) in protected polyhydroxylated N-Boc-thiazolylalkyl amines afforded C-furanosides (37-81%) featuring a chiral thiazolylmethylamino side chain, whic

Full text of DOI:10.1055/s-2007-967993

LETTER
303
A Serendipitous Discovery of a New C-Furanosyl Glycine Synthesis via
Thiazole-Based Aminohomologation of Hexopyranoses
C-Furanosyl
G
lyc
i
l
nes essandro Dondoni,* Alessandro Massi, Andrea Nuzzi
Dipartimento di Chimica, Laboratorio di Chimica Organica, Università di Ferrara, Via L. Borsari 46, 44100 Ferrara, Italy
Fax +39(053)2291167; E-mail: adn@dns.unife.it
Received 8 September 2006
stereochemically well-defined manner. All methods
suffer from various drawbacks such as the scarce stereo-
chemical efficiency, the numerous steps involved, and the
lack of generality. Hence the introduction of new syn-
Abstract: Ring closure via microwave-assisted intramolecular
OMs displacement by a g-OBn group (O-nucleophilic attack) in
protected polyhydroxylated N-Boc-thiazolylalkyl amines afforded
C-furanosides (37–81%) featuring a chiral thiazolylmethylamino
side chain, which, upon thiazole to carboxylate (through aldehyde) thetic routes to C-glycosyl glycines appears to be of great
transformation, furnished enantiopure C-furanosyl glycines.
utility. Thus, we would like to report here on a new
method that we have discovered in the course of our
research on the synthesis of homoazasugars by the
thiazole-based aminohomologation of pentafuranoses8
and hexopyranoses9 through their open-chain nitrones as
intermediates. In the latter investigation9 we observed that
amino alcohol 1, derived from D-mannose,10 upon activa-
tion as O-mesylate and heating in MeCN afforded both
the target piperidine 2 as major product (52%) via intra-
molecular N-nucleophilic displacement of the OMs group
and furanose 3 (30%) via O-nucleophilic attack
(Scheme 1).
Key words: amino acids, cyclization, glycopeptides, glycosides,
ring closure
The importance of anomerically carbon-linked sugar ami-
no acids, i.e. authentic C-glycosyl amino acids, as key
building blocks for the post-translational modification of
natural glycopeptides is widely recognized.1 This struc-
tural change is known to enhance the stability of glyco-
peptides toward chemical and enzymatic degradation as
well as modulate the conformation and folding with ef-
fects on molecular recognition and interactions. Therefore
various synthetic routes have been developed in recent
years,2 especially those leading to methylene isosteres of
natural O-glycosyl a-amino acids, such as L-serine, and L-
threonine, and ethylene isosteres of N-glycosyl L-aspar-
agines which are the most common components of native
glycopeptides.3 Quite recently attention has been also ad-
dressed in our laboratory to the synthesis of C-glycosyl b-
amino acids2b,4 since these compounds may serve to intro-
duce a new and quite substantial element of diversity in
non-natural C-glycosyl amino acid libraries. C-Glycosyl
glycines featuring the a-amino acid group (glycinyl moi-
ety) directly linked to the sugar fragment have been pre-
pared as well.2 Search for the synthesis of this special
class of sugar amino acids was initially driven by their
potential use as building blocks for C-nucleoside anti-
biotics,5 especially polyoxin and nikkomycin analogues.
They are also of great value as components of inhibitors
of bacterial synthetases.6 In recent work C-ribosyl glycine
derivatives have been employed as platforms for the
preparation of various pyrazine C-nucleosides serving in
a study of hydrogen bonding interactions in DNA and
RNA strands.7 Most of the reported synthetic methods to
C-glycosyl glycines involve the introduction of a glycinyl
group equivalent at the anomeric carbon of a sugar sub-
strate or the construction of the glycinyl moiety by ex-
ploiting a carbon-linked functionality already in place in a
OH OBn NHBn
BnO
S
OBn OBn
N
1
a
N-nucleophilic
attack
O-nucleophilic
attack
OBn
S
OBn
BnO
NHBn
S
Bn
N
H
O
N
N
BnO
OBn
OBn
OBn
3
2
Scheme 1 Reagents and conditions: a) MsCl, Et3N, TMEDA,
anhyd toluene, 0–5 °C, 2 h; then heating in refluxing MeCN.
The stereospecific formation of C-furanosides via in-
tramolecular nucleophilic displacement of an activated
OH group by a g-benzyloxy group, was earlier reported by
Martin and Yang and their co-workers.11 However, exam-
ples of C-furanosides bearing chiral side chains whose
substituents were prone to undergo synthetic transforma-
tions, were not described. On the other hand, in the light
of the amply documented one-pot thiazole-to-formyl
transformation and oxidation of the aldehyde to carboxy-
lic acid,12 compound 3 can be regarded as a potential pre-
cursor to an optically pure C-furanosyl glycine. This
synthetic route was attractive because it overcame the
SYNLETT 2007, No. 2, pp 0303–0307
0
1.
0
2.
2
0
0
7
Advanced online publication: 24.01.2007
DOI: 10.1055/s-2007-967993; Art ID: D26906ST
© Georg Thieme Verlag Stuttgart · New York

304
A. Dondoni et al.
LETTER
main problem afflicting other approaches to these sugar sional NMR analysis and ROESY experiments. Given this
amino acids, namely the control of the configuration of successful result, the above reaction sequence was extend-
the anomeric carbon of the sugar moiety as well as the ed to alcohol 6a derived from D-mannose as well and to
configuration of the asymmetric carbon bearing the amino the pairs of stereoisomers 4b/6b and 4c/6c (Table 1)
group in the glycinyl residue.13 However, in order to whose hexopyranose progenitors were D-glucose and D-
optimize the method to a synthetically relevant level, the galactose, respectively.9 With one exception, i.e. deriva-
formation of piperidine 2 via the N-nucleophilic displace- tive 4c, these compounds were transformed into the corre-
ment needed to be inhibited. We envisaged achieving this sponding optically pure C-furanosyl glycinates 10b and
crucial objective by decreasing the nucleophilic power of 11a–c in variable yet fair yields.20 The missing product 8c
the nitrogen atom of the amino alcohol. To this aim the from D-galactose was due to the unsuccessful cyclization
transformation of the NHBn group into NHBoc was con- of N-Boc-amino alcohol 5c to C-furanoside 8c under the
sidered. Hence the N-Boc-protected amino alcohol 5a standard conditions adopted for the other stereoisomers.
(Scheme 2) was prepared by reductive dehydroxylation This observation together with the exceptionally scarce
and debenzylation of the N(OH)Bn group14 of the known efficiency registered in the conversion of 7c to 9c indi-
thiazolylalkylhydroxylamine 4a and treatment of the cates the high sensitivity of the intramolecular O-nucleo-
crude amine thus formed with Boc2O.15 The mesylation of philic substitution process to the structure of the
alcohol 5a followed by heating the crude material in re- substrate.21 It is worth noting that the C-glycosyl glycines
fluxing MeCN for 12 hours afforded the C-furanoside 8a, 10 and 11 prepared belong to the L-series of carbo-
still in modest yield (35%), while no piperidine formation hydrates. However, a similar approach starting from L-
was detected as evidenced by NMR analysis. An unsatis- hexopyranose would lead to the diastereoisomers of the D-
factory result (40% of 8a) was also obtained under harsher series. Nevertheless, we envisaged the approach to this
thermal condition (DMF, 150 °C, 12 h). Therefore, given class of compounds by exploiting the C2-epimers of the
the well established beneficial effects of microwave irra- alcohols employed above. While attempts to epimerize
diation on rates and yields of organic reactions,16 we de- the D-glucose derived amino alcohol 7b via Mitsunobu re-
cided to apply this technique to our system. To our delight action [p-nitrobenzoic acid, DEAD (diethyl azodicarbox-
microwave heating of the crude mesylate at 150 °C for 20 ylate), PPh3, THF, 0 °C] met with failure, the oxidation–
minutes in DMF (sealed tube) increased the yield of 8a to reduction sequence earlier applied to similar compounds
the acceptable value of 73%.17
in our laboratory22 gave satisfactory results (Scheme 3).
NHBoc
Th
NHBoc
O
OBn
OH OBn NR1R2
H
O
a
b
BnO
b
BnO
S
7b
BnO
Th
95%
40%
OBn OBn
OBn OBn
N
BnO
OBn
8a (73%)
12
4a R1 = OH, R2 = Bn (see ref. 9)
5a (56%) R1 = H, R2 = Boc
NHBoc
Th
H
a
NHBoc
Th
OH OBn
O
c
BnO
BnO
56%
NHBoc
BnO
H
OBn
OBn OBn
O
c
BnO
CO2Me
13
14
NHBoc
CO2Me
BnO
H
O
OBn
10a (67%)
d
BnO
47%
BnO
OBn
Scheme 2 Reagents and conditions: a) TiCl3, MeOH, r.t., 15 min;
then Boc2O, NaHCO3, dioxane, r.t., 24 h; b) MsCl, pyridine, 0 °C to
r.t., 2 h; then DIPEA, DMF, microwave; c) MeOTf, MeCN, r.t., 15
min; then NaBH4, MeOH, 0 °C to r.t., 15 min; then CuO/CuCl2·2H2O,
10:1 MeCN–H2O, r.t., 15 min; then NaClO2, H2O2, NaH2PO4, MeCN,
r.t., 2 h; then CH2N2, Et2O, 0 °C.
15
Scheme 3 Reagents and conditions: a) Ac2O, DMSO, r.t., over-
night; b) NaBH4, MeOH, 0 °C to r.t., 15 min; c) and d) see Scheme 2.
Hence, 7b was transformed into ketone 12 via Moffat
oxidation and this was reduced by NaBH4 at room tem-
perature to give a mixture of 7b and the desired epimer 13
in an almost 1:1 ratio.23 Fortunately these diastereoiso-
mers were easily separable by column chromatography,
thus allowing the recycling of 7b for further epimeriza-
tion. Finally, derivative 13 underwent the expected O-nu-
cleophilic substitution leading to the C-furanoside 14 in
fair yield. Thereafter this intermediate was transformed
into the target C-glycosyl glycinate 15 featuring the D-
arabinosyl ring. This result may serve to broaden the
The substitution of the NHBn for the NHBoc group not
only favored the O-nucleophilic pathway, but made also
the final step readily feasible since the latter group is well
known to be tolerated by the thiazole-to-formyl transfor-
mation protocol (N-methylation, reduction, hydrolysis)
while the former is not.18 This allowed for the effective
elaboration of 8a into aldehyde which upon oxidation
with NaClO2 and treatment with CH2N2 afforded C-fura-
nosyl glycinate 10a in very good yield.19 The structure of
both 8a and 10a was supported by mono- and bidimen-
Synlett 2007, No. 2, 303–307 © Thieme Stuttgart · New York

LETTER
C-Furanosyl Glycines
305
Table 1 Other C-Furanosyl Glycines 10 and 11 and their Intermediates Prepared by Aminohomologation of Hexopyranoses
Hexopyranose
D-Mannose
Alcohol (yield, %)
C-Furanoside (yield, %)b
C-Furanosyl glycine (yield, %)b
OH OBn NR1R2
NHBoc
OBn
OBn
NHBoc
H
O
H
BnO
O
S
6
7
6
1
2
Th
2
CO2Me
1
1
3
OBn OBn
N
BnO
OBn
BnO
OBn
6aa R1 = OH, R2 = Bn
9a (73)
11a (34)
7a (68)b R1 = H, R2 = Boc
OH OBn NR1R2
BnO
NHBoc
NHBoc
D-Glucose
D-Glucose
D-Galactose
D-Galactose
a See ref. 9.
OBn
OBn
O
H
H
H
H
H
O
S
S
S
S
Th
CO2Me
OBn OBn
N
BnO
BnO
OBn
OBn
4ba R1 = OH, R2 = Bn
8b (81)
10b (61)
5b (52)b R1 = H, R2 = Boc
OH OBn NR1R2
BnO
OH OBn NR1R2
OBn
NHBoc
Th
O
BnO
S
OBn OBn
N
OBn OBn
N
BnO
OBn
6ba R1 = OH, R2 = Bn
9b (85)
11b (67)
7b (20)b R1 = H, R2 = Boc
OH OBn NR1R2
BnO
OBn
NHBoc
Th
O
OBn OBn
N
BnO
OBn
4ca R1 = OH, R2 = Bn
8c (0)
5c (43)b R1 = H, R2 = Boc
OH OBn NR1R2
BnO
OH OBn NR1R2
OBn
NHBoc
Th
O
BnO
S
OBn OBn
N
OBn OBn
N
BnO
OBn
6ca R1 = OH, R2 = Bn
9c (37)
11c (29)
7c (47)b R1 = H, R2 = Boc
b Same reagents and conditions as in Scheme 2.
potential of the method as it allows introducing a new utility. For instance a service can be foreseen in the
element of diversity in libraries of this class of sugar preparation of libraries of special and stereodiversified
amino acids.
glycoconjugates for biological testing.
In conclusion, the unexpected O-nucleophilic substitution
reaction occurring in heavily functionalized chiral amino
alcohols generated from hexopyranoses has been convert-
ed into a stereospecific synthesis of a class of amino acids
of a rather difficult access such as enantiopure C-furano-
syl glycines. It is worth noting that all the amino acids pre-
pared are orthogonally protected building blocks, which
are therefore well suited for peptide and nucleoside syn-
thesis.
Acknowledgment
We gratefully acknowledge the University of Ferrara and MIUR
(COFIN 2004) for financial support.
References and Notes
(1) Imperiali, B. Acc. Chem. Res. 1997, 30, 452.
(2) (a) Dondoni, A.; Marra, A. Chem. Rev. 2000, 100, 4395.
(b) For recent references (2000–2005), see: Dondoni, A.;
Massi, A.; Sabbatini, S. Chem. Eur. J. 2005, 11, 7110.
(3) (a) Arsequell, G.; Valencia, G. Tetrahedron: Asymmetry
1997, 8, 2839. (b) Arsequell, G.; Valencia, G. Tetrahedron:
Asymmetry 1999, 10, 3045.
However, while this synthetic method may suffer for the
number of steps, its stereospecificity, the simple yet effi-
cient chemistry involved, and the broad scope, are all fa-
vorable features that concur positively to some synthetic
Synlett 2007, No. 2, 303–307 © Thieme Stuttgart · New York

306
A. Dondoni et al.
LETTER
(4) (a) Dondoni, A.; Massi, A.; Sabbatini, S.; Bertolasi, V. Adv.
Synth. Catal. 2004, 46, 1355. (b) Dondoni, A.; Massi, A.;
Minghini, E. Synlett 2006, 539.
(5) (a) Bischofberger, K.; Hall, R. H.; Jordaan, A. J. Chem. Soc.,
Chem. Commun. 1975, 806. (b) Hall, R. H.; Bischofberger,
K.; Eitelman, S. J.; Jordaan, A. J. Chem. Soc., Perkin Trans.
1 1977, 743. (c) Zhang, D.; Miller, M. J. J. Org. Chem.
1998, 63, 755.
(6) Jarvest, R. L.; Berge, J. M.; Brown, P.; Hamprecht, D. W.;
McNair, D. J.; Mensah, L.; O’Hanlon, P. J.; Pope, A. J.
Bioorg. Med. Chem. Lett. 2001, 11, 715.
(7) von Krosigk, U.; Benner, S. A. Helv. Chim. Acta 2004, 87,
1299.
Hz, J3,4 = 3.5 Hz, H-3), 3.72 (dd, 1 H, J1a,1b = 9.5 Hz,
1a,2 = 3.0 Hz, H-1a), 3.64 (dd, 1 H, J1b,1a = 9.5 Hz, J1b,2 = 5.5
J
Hz, H-1b), 2.65 (s, 1 H, OH), 1.45 (s, 9 H, t-Bu). 13C NMR
(CDCl3): d = 173.9, 155.4, 143.2, 138.5, 138.2, 137.9, 137.5,
128.6, 128.5, 128.4, 128.3, 127.9, 127.8, 127.7, 127.5,
127.0, 118.9, 80.9, 80.1, 78.4, 78.1, 74.9, 73.8, 73.4, 72.8,
71.2, 70.0, 60.4, 53.5, 28.3, 21.1, 14.2. MALDI-TOF MS
(724.9): m/z 747.1 [M + Na], 763.1 [M + K]. Anal. Calcd for
C42H48N2O6S: C, 71.16; H, 6.82; N, 3.95. Found: C, 71.17;
H, 6.79; N, 3.97.
(16) (a) Microwaves in Organic Synthesis; Loupy, A., Ed.;
Wiley-VCH: Weinheim, 2002. (b) Kappe, C. O. Angew.
Chem. Int. Ed. 2004, 43, 6250.
(8) Dondoni, A.; Giovannini, P. P.; Perrone, D. J. Org. Chem.
2002, 67, 7203.
(17) Typical Procedure for Entry C-Furanosides 8a,b, 9a–c,
14.
(9) Dondoni, A.; Nuzzi, A. J. Org. Chem. 2006, 71, 7574.
(10) The open-chain amino alcohol 1 was obtained by reduction
of the minor product that was formed by addition of
thiazolylmagnesium bromide to the sugar hydroxylamine–
nitrone mixture derived from the reaction of N-
To a cooled (0 °C), stirred solution of N-Boc-protected
amines 5a–c, 7a–c, 13 (175 mg, 0.24 mmol) in pyridine (6.0
mL) was added mesyl cloride (56 mL, 0.72 mmol) in one
portion. The mixture was stirred for an additional 2 h, then
some drops of MeOH were added and the solvent was
removed in vacuo. The residue was taken up in EtOAc (10
mL) and washed with a sat. solution of NaHCO3 (2 × 10
mL). The organic layer was dried over Na2SO4, filtered, and
concentrated. A 0.5–2.0 mL process vial was filled with the
above crude mesylated intermediate, DIPEA (83 mL, 0.48
mmol), and anhyd DMF (2.0 mL). The vial was sealed with
the Teflon septum and aluminium crimp by using an
appropriate crimping tool. The vial was then placed in its
correct position in the Biotage Initiator cavity where
irradiation for 20 min at 150 °C was performed. After the
full irradiation sequence was completed, the vial was cooled
to r.t. and then opened. The solution was transferred into a
round-bottomed flask and the solvent was removed in vacuo.
The residue was taken up in EtOAc (10 mL), and washed
with a sat. solution of NaHCO3 (2 × 15 mL). The organic
phase was dried over Na2SO4, filtered and concentrated. The
residue was eluted from a column of silica gel with the
suitable elution system to give the C-furanosides 8a,b, 9a–c,
14. Column chromatography with 3:1 cyclohexane–EtOAc
afforded 8a (112 mg, 73%) as a syrup. [a]D 42.8 (c 1.0,
CHCl3). 1H NMR (C6D6): d = 7.52 (d, 1 H, J = 3.2 Hz, Th),
7.20–7.00 (m, 15 H, 3 Ph), 6.50 (d, 1 H, J = 3.2 Hz, Th), 6.36
(d, 1 H, JNH,2 = 7.5 Hz, NH), 5.69 (br s, 1 H, H-2), 4.90 (br s,
1 H, H-3), 4.42 and 4.18 (2 d, 2 H, J = 12.0 Hz, CH2Ph), 4.26
(s, 2 H, CH2Ph), 4.21 and 4.14 (2 d, 2 H, J = 11.0 Hz,
CH2Ph), 4.14 (dd, 1 H, J4,3 = 1.5 Hz, J4,5 = 4.0 Hz, H-4), 4.11
(ddd, 1 H, J6,5 = 1.5 Hz, J6,7a = 6.5 Hz, J6,7b = 5.0 Hz, H-6),
3.83 (dd, 1 H, J5,4 = 4.0 Hz, J5,6 = 1.5 Hz, H-5), 3.66 (dd, 1
H, J7a,6 = 6.5 Hz, J7a,7b = 9.5 Hz, H-7), 3.52 (dd, 1 H,
benzylhydroxylamine with tetra-O-benzyl-D-
mannopyranose.
(11) (a) Martin, R. R.; Yang, F.; Xie, F. Tetrahedron Lett. 1995,
36, 47. (b) Yang, B.-H.; Jiang, J.-Q.; Ma, K.; Wu, H.-M.
Tetrahedron Lett. 1995, 36, 2831.
(12) Dondoni, A.; Junquera, F.; Merchan, F. L.; Merino, P.;
Scherrmann, M.-C.; Tejero, T. J. Org. Chem. 1997, 62,
5484.
(13) For a critical discussion on this issue, see section IIA in ref.
2a.
(14) (a) A warning on this transformation (see ref. 14b) is
worthwhile since in our hand the reduction occurred
efficiently only by using samples of the TiCl3 reagent
withdrawn from freshly opened commercially available
vials (Fluka 14010). (b) Murahashi, S.-I.; Kodera, Y.
Tetrahedron Lett. 1985, 26, 4633; and ref. 12.
(15) Typical Procedure for Entry N-Boc-Amines 5a–c, 7a–c.
To a stirred solution of N-benzyl-N-glycosylhydroxylamines
4a–c, 6a–c (0.50 g, 0.69 mmol) in MeOH (9.0 mL) was
added a 12% HCl solution of TiCl3 (1.22 mL, 1.71 mmol).
Stirring was manteined for an additional 15 min at r.t., then
5 M NaOH was added dropwise until pH 7 (the solution
became white). After stirring for an additional 5 min, the
primary amine was extracted with EtOAc (3 × 25 mL). The
organic phase was washed with brine (2 × 20 mL), dried
over Na2SO4, and concentrated. Crude primary amine was
dissolved in dioxane (6.0 mL) and Boc2O (330 mg, 1.50
mmol) was added in one portion. Then some drops of a sat.
solution of NaHCO3 were added to the mixture until pH 7.5.
Stirring was mantained for 18 h, then the reaction was
quenched with a 10% citric acid solution (4.0 mL). The
protected amine was extracted with EtOAc (2 × 20 mL). The
combined organic layers were washed with brine (3 × 20
mL) and H2O (10 mL). The organic phase was dried over
Na2SO4, filtered, and concentrated. The residue was eluted
from a column of silica gel with the suitable elution system
to give the N-Boc-protected amines 5a–c,7a–c. Column
chromatography with 3:1 cyclohexane–EtOAc afforded 5a
(278 mg, 56%) as a syrup. [a]D 5.9 (c 0.7, CHCl3). 1H NMR
(CDCl3): d = 7.78 (d, 1 H, J = 3.2 Hz, Th), 7.42–7.00 (m, 21
H, 4 Ph and Th), 5.86 (d, 1 H, JNH,6 = 9.0 Hz, NH), 5.59 (dd,
1 H, J6,5 = 1.0 Hz, J6,NH = 9.0 Hz, H-6), 4.72 and 4.62 (2 d, 2
H, J = 11.0 Hz, CH2Ph), 4.72 and 4.59 (2 d, 2 H, J = 11.0 Hz,
CH2Ph), 4.58 (dd, 1 H, J5,4 = 8.0 Hz, J5,6 = 1.0 Hz, H-5), 4.54
and 4.51 (2 d, 2 H, J = 11.0 Hz, CH2Ph), 4.23 and 3.99 (2 d,
2 H, J = 11.0 Hz, CH2Ph), 4.10 (br s, 1 H, H-2), 3.97 (dd, 1
H, J4,3 = 3.5 Hz, J4,5 = 8.0 Hz, H-4), 3.88 (dd, 1 H, J3,2 = 8.0
J
7b,6 = 5.0 Hz, J7b,7a = 9.5 Hz, H-7b), 1.35 (s, 9 H, t-Bu). 13
C
NMR (C6D6): d = 172.1, 155.6, 142.9, 138.5, 137.9, 128.3,
128.2, 128.1, 127.9, 127.7, 127.4, 118.4, 110.3, 98.2, 85.4,
83.0, 82.4, 80.0, 73.0, 71.6, 71.3, 67.4, 55.3, 35.5, 28.0.
MALDI-TOF MS (616.7): m/z 655.2 [M + K]. Anal. Calcd
for C35H40N2O6S: C, 68.16; H, 6.54; N, 4.54. Found: C,
68.22; H, 6.60; N, 4.59.
(18) (a) Dondoni, A.; Marra, A.; Perrone, D. J. Org. Chem. 1993,
58, 275. (b) Dondoni, A.; Perrone, D. Synthesis 1993, 1162.
(c) Dondoni, A.; Perrone, D. Aldrichimica Acta 1997, 30,
35. (d) Dondoni, A. Synthesis 1998, 1681. (e) Dondoni, A.;
Marra, A. Chem. Rev. 2004, 5, 2557.
(19) Typical Procedure for Entry C-Glycosyl Glycines 10a,b,
11a–c, 15.
A mixture of the thiazole derivative 8a,b, 9a–c, 14 (164 mg,
0.26 mmol), activated 4 Å MS (400 mg) and MeCN (4.0 mL)
was stirred at r.t. for 10 min then methyl triflate (32 mL, 0.28
mmol) was added in one portion. The suspension was stirred
Synlett 2007, No. 2, 303–307 © Thieme Stuttgart · New York

LETTER
for an additional 15 min then the solvent was removed under
C-Furanosyl Glycines
307
Compound 10b: [a]D 10.9 (c 1.0, CHCl3). 1H NMR (C6D6):
d = 7.25–7.00 (m, 15 H, 3 Ph), 5.42 (d, 1 H, JNH,2 = 8.5 Hz,
NH), 4.92 (dd, 1 H, J2,NH = 8.5 Hz, J2,3 = 4.5 Hz, H-2), 4.75
(t, 1 H, J3,2 = J3,4 = 4.5 Hz, H-3), 4.32 (ddd, 1 H, J6,5 = 4.0
Hz, J6,7a = 6.5 Hz, J6,7b = 5.0 Hz, H-6), 4.28 and 4.19 (2 d, 2
H, J = 11.5 Hz, CH2Ph), 4.26 and 4.21 (2 d, 2 H, J = 11.0 Hz,
CH2Ph), 4.21 and 4.12 (2 d, 2 H, J = 12.0 Hz, CH2Ph), 4.01
(dd, 1 H, J4,3 = 4.5 Hz, J4,5 = 2.0 Hz, H-4), 3.87 (dd, 1 H,
J5,4 = 2.0 Hz, J5,6 = 4.0 Hz, H-5), 3.71 (dd, 1 H, J7a,6 = 6.5
Hz, J7a,7b = 9.5 Hz, H-7a), 3.58 (dd, 1 H, J7b,6 = 5.0 Hz,
J7b,7a = 9.5 Hz, H-7b), 3.20 (s, 3 H, CH3), 1.40 (s, 9 H, t-Bu).
Compound 11b: [a]D 2.3 (c 1.0, CHCl3). 1H NMR (C6D6):
d = 7.22–7.00 (m, 15 H, 3 Ph), 5.77 (d, 1 H, JNH,2 = 10.0 Hz,
NH), 5.33 (dd, 1 H, J2,NH = 10.0 Hz, J2,3 = 5.0 Hz, H-2), 4.76
(t, 1 H, J3,2 = J3,4 = 5.0 Hz, H-3), 4.49 (ddd, 1 H, J6,5 = 4.0
Hz, J6,7a = 7.0 Hz, J6,7b = 5.5 Hz, H-6), 4.32 and 4.25 (2 d, 2
H, J = 12.0 Hz, CH2Ph), 4.20 and 4.15 (2 d, 2 H, J = 11.5 Hz,
CH2Ph), 4.09 (s, 2 H, CH2Ph), 4.00 (dd, 1 H, J4,3 = 5.0 Hz,
reduced pressure. The residue was taken up in MeOH (4.0
mL), cooled to 0 °C and treated with NaBH4 (22 mg, 0.56
mmol). The mixture was stirred at r.t. for 15 min, diluted
with acetone, filtered through a pad of Celite®, and
concentrated in vacuo. The residue was taken up in 10:1
MeCN–H2O (4.0 mL) and then treated with CuO (60 mg,
0.78 mmol), and CuCl2·2H2O (48 mg, 0.28 mmol). The
resulting suspension was stirred at r.t. for 10 min, then
filtered through a pad of Celite® and concentrated in vacuo
at a temperature below 30 °C. The residue was partitioned
between brine (30 mL) and Et2O (30 mL). The organic layer
was separated and the aqueous layer was extracted with Et2O
(2 × 30 mL). The combined organic extracts were washed
with sat. aq EDTA (disodium salt), brine, dried over Na2SO4,
filtered, and concentraed to give the essentially pure a-
amino aldehyde which was taken up in MeCN (2 mL). The
resulting solution was treated with 35% aq H2O2 (40 mL),
1.2 M aq KH2PO4 (0.2 mL) and 0.17 M aq NaClO2 (1.4 mL).
After 2 h the reaction was acidified with 1 N aq HCl till
pH = 2 and the resulting mixture was extracted with EtOAc
(3 × 20 mL). The organic extracts were dried over Na2SO4,
filtered, and concentrated. The crude carboxylic acid was
taken up in Et2O, the solution was kept at 0 °C and treated
with an ethereal solution of CH2N2 until a pale yellow colour
persisted. The solution was stirred for an additional 15 min,
then the residue was dried in vacuo. The residue was eluted
from a column of silica gel with the suitable elution system
to give the corresponding C-glycosylglycines 10a,b, 11a–c,
15. Column chromatography with 3:1 cyclohexane–EtOAc
afforded 10a (104 mg, 67%) as a syrup. [a]D 32.1 (c 1.0,
CHCl3). 1H NMR (C6D6): d = 7.20–7.00 (m, 15 H, 3 Ph),
5.96 (d, 1 H, JNH,2 = 9.0 Hz, NH), 4.85 (dd, 1 H, J2,NH = 9.0
Hz, J2,3 = 2.5 Hz, H-2), 4.65 (dd, 1 H, J3,2 = 2.5 Hz, J3,4 = 4.0
Hz, H-3), 4.42 and 4.12 (2 d, 2 H, J = 12.0 Hz, CH2Ph), 4.26
and 4.22 (2 d, 2 H, J = 11.0 Hz, CH2Ph), 4.21 (s, 2 H,
CH2Ph), 4.14 (t, 1 H, J4,3 = J4,5 = 4.0 Hz, H-4), 4.12 (br s, 1
H, H-6), 3.82 (dd, 1 H, J5,4 = 4.0 Hz, J5,6 = 1.5 Hz, H-5), 3.64
(dd, 1 H, J7a,6 = 6.5 Hz, J7a,7b = 10.0 Hz, H-7a), 3.54 (dd, 1 H,
J7b,6 = 5.0 Hz, J7b,7a = 10.0 Hz, H-7b), 3.25 (s, 3 H, CH3),
1.37 (s, 9 H, t-Bu). 13C NMR (C6D6): d = 170.2, 156.0,
138.4, 138.0, 137.9, 128.3, 128.2, 128.1, 127.9, 127.7,
127.6, 127.4, 127.3, 83.7, 83.2, 82.5, 80.2, 79.1, 73.0, 71.9,
71.3, 67.4, 55.7, 53.0, 51.6, 28.0. MALDI-TOF MS (591.7):
m/z = 614.6 [M + Na], 630.3 [M + K]. Anal. Calcd for
C34H41NO8: C, 69.02; H, 6.98; N, 2.37. Found: C, 69.04; H,
6.97; N, 2.37.
J
4,5 = 1.5 Hz, H-4), 3.84 (dd, 1 H, J5,4 = 1.5 Hz, J5,6 = 4.0 Hz,
H-5), 3.81 (dd, 1 H, J7a,6 = 7.0 Hz, J7a,7b = 9.5 Hz, H-7a),
3.68 (dd, 1 H, J7b,6 = 5.5 Hz, J7b,7a = 9.5 Hz, H-7b), 3.15 (s, 3
H, CH3), 1.40 (s, 9 H, t-Bu).
Compound 11c: [a]D –28.5 (c 0.6, CHCl3). 1H NMR
(CDCl3): d = 7.40–7.20 (m, 15 H, 3 Ph), 6.07 (d, 1 H,
JNH,2 = 9.0 Hz, NH), 4.83 (t, 1 H, J3,2 = J3,4 = 5.0 Hz, H-3),
4.74 (dd, 1 H, J2,NH = 9.0 Hz, J2,3 = 5.0 Hz, H-2), 4.64 and
4.59 (2 d, 2 H, J = 10.5 Hz, CH2Ph), 4.56 and 4.39 (2 d, 2 H,
J = 12.0 Hz, CH2Ph), 4.54 and 4.47 (2 d, 2 H, J = 11.0 Hz,
CH2Ph), 4.28 (br s, 2 H, H-4 and H-6), 4.03 (t, 1 H,
J5,4 = J5,6 = 5.0 Hz, H-5), 3.54 (s, 3 H, CH3), 3.53 (dd, 1 H,
J
7a,6 = 4.0 Hz, J7a,7b = 10.5 Hz, H-7a), 3.46 (dd, 1 H,
J7b,6 = 3.0 Hz, J7b,7a = 10.5 Hz, H-7b), 1.40 (s, 9 H, t-Bu).
Compound 15: [a]D –6.9 (c 0.7, CHCl3). 1H NMR (C6D6):
d = 7.20–7.00 (m, 15 H, 3 Ph), 5.79 (d, 1 H, JNH,2 = 9.0 Hz,
NH), 5.30 (dd, 1 H, J2,NH = 9.0 Hz, J2,3 = 6.0 Hz, H-2), 4.47
(dd, 1 H, J3,2 = 6.0 Hz, J3,4 = 4.0 Hz, H-3), 4.31 and 4.24 (2
d, 2 H, J = 12.0 Hz, CH2Ph), 4.30 and 4.29 (2 d, 2 H, J = 12.0
Hz, CH2Ph), 4.21 (br s, 1 H, H-6), 4.14 (s, 2 H, CH2Ph), 4.01
(br s, 1 H, H-5), 3.95 (br s, 1 H, H-4), 3.59 (dd, 1 H,
J7a,6 = 5.0 Hz, J7a,7b = 10.0 Hz, H-7a), 3.51 (dd, 1 H,
J
7b,6 = 7.5 Hz, J7b,7a = 10.0 Hz, H-7b), 1.39 (s, 9 H, t-Bu).
(21) Cyclization of the N-Boc amino alcohol 7c required two
cycles of microwave irradiation to get an acceptable yield of
the C-furanoside 9c. On the other hand, microwave
irradiation of 5c for prolonged time periods (single or
repeated cycles) did not afford the corresponding furanoside
8c.
(20) Compound 11a: [a]D 27.9 (c 0.3, CHCl3). 1H NMR (C6D6):
d = 7.25–7.00 (m, 15 H, 3 Ph), 5.94 (d, 1 H, JNH,2 = 6.5 Hz,
NH), 4.92 (dd, 1 H, J2,NH = 6.5 Hz, J2,3 = 5.0 Hz, H-2), 4.45
and 4.15 (2 d, 2 H, J = 12.0 Hz, CH2Ph), 4.35–4.20 (m, 7 H,
2 CH2Ph, H-3, H-4 and H-6), 3.83 (br s, 1 H, H-5), 3.79 (dd,
1 H, J7a,6 = 6.5 Hz, J7a,7b = 10.0 Hz, H-7a), 3.68 (dd, 1 H,
J7b,6 = 5.5 Hz, J7b,7a = 10.0 Hz, H-7b), 3.22 (s, 3 H, CH3),
1.35 (s, 9 H, t-Bu).
(22) (a) Dondoni, A.; Fantin, G.; Fogagnolo, M.; Medici, A.;
Pedrini, P. J. Org. Chem. 1989, 54, 702. (b) Dondoni, A.;
Orduna, J.; Merino, P. Synthesis 1992, 201.
(23) Reduction of 12 by NaBH4 at lower temperatures and by L-
Selectride (THF, –78 °C) afforded the undesired alcohol 7b
as major product.
Synlett 2007, No. 2, 303–307 © Thieme Stuttgart · New York

Copyright of Synlett is the property of Georg Thieme Verlag Stuttgart and its content may not
be copied or emailed to multiple sites or posted to a listserv without the copyright holder's
express written permission. However, users may print, download, or email articles for
individual use.