Carbohydrate mimic of 2-deoxystreptamine for the preparation of conformationally constrained aminoglycosides

Article abstract of DOI:10.1016/j.tet.2007.02.006

The synthesis of a carbohydrate mimic of 2-deoxystreptamine (2-DOS) is described starting from d-ribose. Crucial steps of the synthesis involve a stereoselective nitroaldol condensation and deoxygenation via elimination-in situ reduction. Moreover, glycos

Full text of DOI:10.1016/j.tet.2007.02.006

Tetrahedron 63 (2007) 3183–3188
Carbohydrate mimic of 2-deoxystreptamine for the preparation
of conformationally constrained aminoglycosides
Guuske F. Busscher, Sebastiaan (Bas) A. M. W. van den Broek, Floris P. J. T. Rutjes
*
and Floris L. van Delft
Institute for Molecules and Materials, Organic Chemistry, Radboud University Nijmegen, Toernooiveld 1,
6525 ED Nijmegen, The Netherlands
Received 14 November 2006; revised 18 January 2007; accepted 1 February 2007
Available online 7 February 2007
Abstract—The synthesis of a carbohydrate mimic of 2-deoxystreptamine (2-DOS) is described starting from D-ribose. Crucial steps of the
synthesis involve a stereoselective nitroaldol condensation and deoxygenation via elimination–in situ reduction. Moreover, glycosylation of
the carbohydrate 2-DOS derivative with a phenyl thioglycoside donor in the presence of TTBP and AgOTf followed by ring-closing meta-
thesis yielded a conformationally restricted aminoglycoside analogue.
Ó 2007 Elsevier Ltd. All rights reserved.
1. Introduction
may be worthy alternatives for 2-DOS has also been pursued
by Boons and co-workers (vide infra).4,5
Aminoglycosides form a large class of clinically important
antibiotics with a broad antibacterial spectrum and proven
efficacy, particularly against Gram-negative bacteria.1 The
key structural feature of nearly all aminoglycosides is a
1,3-diaminocyclohexanetriol termed 2-deoxystreptamine
(2-DOS). The central position of 2-deoxystreptamine sug-
gests that 2-DOS is a crucial scaffold for correct positioning
of the aminoglycosides during binding to bacterial RNA for
interference with protein translation. Not surprisingly, there-
fore, the vast majority of novel RNA-binding ligands re-
ported in literature are structures derived from 2-DOS. It is
obvious that such an approach requires that the availability
of substantial amounts of 2-DOS is secured. However,
although many synthetic routes toward 2-DOS have been
developed,2 the majority of these routes necessitate a large
number of synthetic steps. We therefore reasoned that a car-
bohydrate-derived entity like compound A (Fig. 1) could be
an interesting alternative for 2-DOS, taken into consider-
ation that the amino group at C-1 is modified in several clini-
cally used aminoglycosides and the C-6 hydroxyl is not
essential for antibacterial activity (2-DOS numbering). Fur-
thermore, starting from a (cheap) carbohydrate precursor
leads to enantiomerically pure product. To date, two
syntheses of such carbohydrate analogues are known. A car-
bohydrate analogue was synthesized starting from benzyl-a-
D-glucopyranoside by Meyer zu Reckendorf and co-workers
(not depicted), in a 10-step synthesis leading to a compound
with structural resemblance to 2-DOS, but as a mixture of
anomers.3 Secondly, the assumption that carbohydrates
Retrosynthetic analysis of b-configured carbohydrate ana-
logue A suggests a synthesis from nitroglucopyranoside B,
by means of deoxygenation at C-4, nitro reduction, and azide
introduction at C-6 (carbohydrate numbering). The nitroglu-
copyranoside in turn was projected to result from D-ribose, a
cheap and enantiopure carbohydrate, via oxidative glycol
cleavage followed by tandem Henry reaction with nitrometh-
ane. Although it is clear that the final carbohydrate analogue
lacks the C-6 hydroxyl, the fact that the 6-OH of 4,5-linked
aminoglycoside antibiotics is only involved in indirect bind-
ing events with the RNA backbone (Fig. 2)6,7 led us to assume
thatitsomissionwouldnotnecessarilyresultinalargepenalty
in binding enthalpy. Furthermore, such a synthetic strategy
would result in a compound with a homologated amino group,
as in several semi-synthetic aminoglycoside analogues.
* Corresponding author. Tel.: +31 (0) 24 365 2373; fax: +31 (0) 24 365
3393; e-mail: f.vandelft@science.ru.nl
Figure 1. Retrosynthesis of carbohydrate derivative (A) as a mimic for
2-deoxystreptamine.
0040–4020/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2007.02.006

3184
G. F. Busscher et al. / Tetrahedron 63 (2007) 3183–3188
led to unacceptable amounts of fully hydrolyzed product.8–11
However, the more labile cyclopentylidene group12,13 was
chemoselectively removed with acetic acid and water (80/
20), leading to allyl b-D-ribofuranoside (3) in reasonable
yield, although in this case it was mandatory to quench the
reaction after 48 h to avoid hydrolysis of the anomeric pro-
tective group.14 Next, a challenging synthetic transformation
was undertaken, involving oxidative cleavage of the diol,
followed by nitroaldol condensation with nitromethane.15
The fact that the desired stereoisomer 4 has the thermo-
dynamically most favorable all-equatorial configuration
was an important stimulus to undertake this endeavor and
we were delighted to find that the diastereomerically pure
product 3-nitro-b-D-glucopyranoside 4 was isolated after
acidification and crystallization from EtOAc (Scheme 1).16
Two other isomers, formed in smaller proportions, presum-
ably the galacto and manno configured sugars,16,17 were
not further characterized. Having the requisite nitrosugar 4
in hand, the desired regioselective deoxygenation at C-4
could be investigated. To this end, the 6-OH was first
converted into an azide by treatment of 4 with tosyl chloride
followed by displacement with sodium azide to give the
6-azidopyranoside 5 in good yield (Scheme 2). Much to
our satisfaction, subsequent protection of the 2-hydroxyl
with a triethylsilyl group proceeded with good regioselectiv-
ity to yield monosilylated glucopyranoside 6.18 Subsequent
acetylation of O-4 with acetyl chloride19 and addition of
sodium borohydride to the crude reaction product led to a
one-step elimination–reduction affording compound 7.20
Finally, removal of the TES group in 2 M HCl in EtOAc
afforded the desired structure 8 in a total of eight synthetic
steps. As projected, compound 8 possesses the correct
stereochemistry to function as a carbohydrate mimic of
2-deoxystreptamine, with the 2-OH (5-OH in 2-DOS num-
bering) selectively free for glycosylation. Before such a gly-
cosylation of 8 was undertaken, reduction of the nitro or
azido group was investigated. Indeed, in the presence of
Raney-Ni, hydrogenation of nitroglucopyranoside 8 led to
full reduction of the azido, nitro, and allyl groups. Subse-
quent removal of the TES group under aforementioned con-
ditions yielded the carbohydrate derivative 9 quantitatively.
Figure 2. Neomycin complexed to nucleobases and the phosphate backbone
of 16S rRNA.
2. Preparation of 2-DOS carbohydrate mimic
The first synthetic step in the preparation of carbohydrate
analogue A is triple protection of D-ribose. In a one-pot reac-
tion, both C-2 and C-3 hydroxyls were protected with a
cyclopentylidene group and the C-1 hydroxyl with an allyl
group yielding the protected allyl b-D-ribofuranoside 2. A
cyclopentylidene function is employed instead of the more
common isopropylidene because various acidic conditions
3. Conformationally constrained aminoglycoside ligand
NMR spectroscopy and X-ray crystallography studies have
indicated that neamine is the minimal motif for selective
binding to the A-site of 16S rRNA.6,7 It has also been shown
that 2,6-diamino-2,6-dideoxyglucopyranoside is responsible
Scheme 1. Reagents and conditions: (a) CuSO4, cyclopentanone, allyl alco-
hol, H2SO4, 40 ꢀC, 48 h; (b) CH3COOH 80%, 40 ꢀC, 48 h; (c) NaIO4, H2O,
0 ꢀC to rt, 30 min, then NaOMe, NO2Me, MeOH, rt, 45 min.
Scheme 2. Reagents and conditions: (a) TsCl, CH2Cl2/pyridine (1/1), 0 ꢀC to rt, o.n.; (b) NaN3, DMF, 55 ꢀC, o.n.; (c) TESOTf, Et3N, CH2Cl2, 0 ꢀC, 30 min; (d)
Ac2O, Et3N, DMAP, CH2Cl2, 0 ꢀC to rt, 30 min, then NaBH4, EtOH, rt, 2 h; (e) 2 M HCl, EtOAc, rt, 10 min; (f) Ra-Ni, MeOH, H2, 3 bar, rt, 24 h; (g) 2 M HCl,
EtOAc, rt, 10 min.

G. F. Busscher et al. / Tetrahedron 63 (2007) 3183–3188
3185
ꢀ
˚
Scheme 3. Reagents and conditions: (a) AgOTf, 4 A MS, TTBP, CH2Cl2, 0 C to rt, o.n. (a/bz3/1); (b) Hoveyda–Grubbs catalyst, toluene, rt, 4 h.
for most of the key interactions with RNA. Studies by Wong
and co-workers have demonstrated that substitution patterns
other than 2,6-diamines lead to loss of activity of these anti-
biotics.21 Studies by Boons and co-workers4,5 on a library
of 24 disaccharide mimics of neamine including a(1–3) or
b(1–3) and a(1–4) or b(1–4) linked dimers containing 2–4
amino groups indicated that the compound with the best
binding affinity was the one with four amino groups (C-2,
C-20, C-6, C-60), displaying an affinity similar to that of
neamine. On the downside of aminoglycosides, extensive
clinical use is limited due to the associated nephro- and oto-
toxicities,22 as well as the global development of microbial
resistance as the result of structural modification by bacterial
enzymes.23,24 In this respect, it is known that most aminogly-
cosides bind rather promiscuously to a variety of RNA
targets. The NMR structure derived from paromomycin
complexed to 16S rRNA revealed that several intermolecular
contacts between the aminoglycosides and RNA recognition
sites are important for binding.6a Our attention was drawn
to a hydrogen bond between N-20 and O-400 (Fig. 2) and it was
reasoned that a neamine analogue covalently linked between
C-20 and O-5 (O-2 in 2-DOS analogue 8) would be locked in
the appropriate conformation for binding to rRNA and may
therefore show enhanced binding specificity (Scheme 3). A
product, due to incompatibility of azide functionality with
the phosphine ligand of the catalyst.30 Consequently, phos-
phine-free Hoveyda–Grubbs catalyst was applied instead
and much to our satisfaction, the cyclic product 13 was
formed in excellent yield (99%). We are currently applying
the methodology for the preparation of a library of neamine
analogues for application in novel antimicrobial therapy.
4. Conclusion
A carbohydrate analogue of 2-DOS was synthesized success-
fully in eight steps and an overall yield of 6.6%. The analogue
is enantiopure and orthogonally protected. Model reduction
and deprotection of the analogue proceeded in only two steps
in a high yield. The carbohydrate precursor is conveniently
protected for incorporation in new aminoglycoside entities,
suitable for the synthesis of 4,5-linked aminoglycoside
antibiotics. Moreover, the presence of an allyl moiety at
O-1 makes the carbohydrate mimic highly suitable for the
synthesis of conformationally restricted RNA binders as
was illustrated in the successful coupling of the carbohydrate
derivative to thioglycoside 10, to form neamine mimic 11. A
conformationally restricted analogue was synthesized by
ring-closing metathesis with the Hoveyda–Grubbs catalyst.
Further application of the thus developed route, to prepare a
library of conformationally restricted neamine analogues is
currently under investigation.
ꢀ
similar reasoning was followed by Jimenez-Barbero and
co-workers, who recently prepared a conformationally re-
stricted neomycin B analogue and found it to be less suscep-
tible to resistance.25 Another report on conformationally
restricted aminoglycoside analogues was recently published
by Blount et al.26
5. Experimental
5.1. General
These observations stimulated us to investigate incorpora-
tion of nitroanalogue 8 in a conformationally restricted ana-
logue of neamine via coupling to a thioglycoside followed
by macrocycle formation. Thus, glycosylation of the carbo-
hydrate acceptor 8 with thioglycoside donor 1027 was car-
ried out in the presence of 2,4,6-tri-tert-butylpyrimidine
(TTBP) and AgOTf (Scheme 3).28 The product was formed
in a yield of 45% and the a/b ratio was 3:1.29 Apart from
that, formation of a substantial amount (32%) of an undesir-
able side-product was observed, NMR analysis of which re-
vealed that elimination product 12 was formed. Next, after
separation of the two products (11 and 12) ring-closing
metathesis was attempted. It was found that exposure of di-
saccharide 11 to Grubbs’ catalyst failed to give the desired
All solvents were distilled from appropriate drying agents
prior to use. Chemicals were purchased from Sigma–Aldrich
and used as such. Reactions were carried out under argon
atmosphere. Standard syringe techniques were applied to the
transfer of dry solvents and air- or moisture sensitive re-
agents. Rf values were obtained using thin layer chromato-
graphy (TLC) on silica gel-coated plates (Merck 60 F254
with the indicated solvent mixture. Melting points were an-
)
€
alyzed with a Buchi melting point apparatus B-545. IR spec-
tra were recorded on an ATI Mattson Genesis Series FTIR
spectrometer; absorption in cmꢁ1. GC was performed on a

3186
G. F. Busscher et al. / Tetrahedron 63 (2007) 3183–3188
Hewlett Packard 5890, containing an HP1 column
(25 mꢂ0.32 mmꢂ0.17 mm), with FID detection and
HP3393A integrator. NMR spectra were recorded on a
slowly warmed to room temperature. After stirring for
30 min, TLC analysis (EtOAc) indicated complete disap-
pearance of starting material. EtOH was added to the reac-
tion mixture before cooling to 0 ꢀC and filtration. The
filtrate was evaporated, the residue dissolved in EtOH, and
filtered again. The resulting dialdehyde was co-evaporated
twice with MeOH, dissolved in MeOH, and cooled to
0 ꢀC. Nitromethane (2.0 mL, 37.34 mmol) was added, fol-
lowed by freshly prepared NaOMe (2.48 g, 35.9 mmol)
and the reaction was stirred at room temperature for
45 min. To the reaction mixture was added a cold suspension
of Amberlite IR-120 (H+) in MeOH. The yellow reaction
mixture became colorless after 10 min. The reaction mixture
was passed through a column of Amberlite IR-120 and thor-
oughly rinsed with MeOH. After evaporation of the solvent
EtOAc was added and evaporated, EtOAc was added again,
and the solution was stored in the fridge for one night leading
to precipitation of 4. EtOAc was filtered off to yield com-
pound 4 as a white powder (2.54 g, 30%). Rf 0.62 (EtOAc).
Bruker DMX 300 (300 MHz), and
a Varian 400
(400 MHz) spectrometer in CDCl3 solutions (unless other-
wise reported) using TMS as internal standard; chemical
shifts are given in parts per million. Coupling constants are
reported as J-values in hertz. Peak assignment in 13C spectra
is based on 2D-GHSQC and GHMBC spectra. Enantiomeric
purities were determined on a Shimadzu HPLC, with indi-
cated column and solvent mixture. Column or flash chroma-
tography was carried out using ACROS silica gel (0.035–
0.070 mm and ca. 6 nm pore diameter). Optical rotations
were determined with a Perkin Elmer 241 polarimeter.
Mass spectra were obtained with a Fisons (VG) Micromass
7070E apparatus and/or a Finnigan MAT900S. MALDI-
TOF-MS spectra were obtained on a Bruker Biflex III
machine, with dihydroxybenzoic acid (DHB) as matrix.
Elemental analyses were carried out using a Carlo Erba
Instruments CHNS-O EA 1108 element analyzer.
20
Mp: 158.6 ꢀC. [a]D ꢁ24.9 (c 1.0, CH2Cl2). IR nmax film:
(cmꢁ1) 3285, 1556, 1124, 1079, 1042, 1012. 1H NMR
(CD3OD, 400 MHz, ppm): d 7.09–5.96 (m, 1H, CH), 5.37
(ddt, J¼1.6, 1.8 Hz, 1H, CH), 5.31 (ddt, J¼1.3, 1.5 Hz,
1H, CH), 4.46 (t, J¼10.1 Hz, 1H, H-3), 4.37 (d, J¼7.7 Hz,
1H, H-1), 4.41–4.33 (m, 1H, CH2), 4.13 (ddt, J¼13.0, 6.0,
1.4 Hz, 1H, CH2), 3.96 (t, J¼10.0 Hz, 1H, H-4), 3.86 (dd,
J¼12.0, 2.3 Hz, 1H, H-6a), 3.78 (dd, J¼10.2, 7.8 Hz, 1H,
H-2), 3.71 (dd, J¼8.0, 5.2 Hz, 1H, H-6b), 3.27 (ddd, J¼9.7,
5.1, 2.3 Hz, 1H, H-5). 13C NMR (CDCl3, 75 MHz, ppm): d
135.1 (CH), 117.3 (CH2), 102.7 (CH), 95.8 (CH), 78.1 (CH),
72.3 (CH), 71.1 (CH2), 69.2 (CH), 62.0 (CH2). HRMS (ESI)
m/z calcd for C9H15O7N (M+Na)+: 272.0746, found:
272.0745. Elemental analysis: calculated for C9H15O7: C
43.37, H 6.07, N 5.62, found C 43.00, H 6.10, N 5.51.
5.1.1. Allyl 2,3-O-cyclopentanone-b-D-ribofuranoside
(2). Powdered D-ribose (1, 5.85 g, 38.9 mmol) and anhy-
drous cuprous sulfate (12.4 g) were suspended in a mixture
of cyclopentanone (110 mL) and allyl alcohol (32 mL) con-
taining a catalytic amount of H2SO4 (0.2 mL). The resulting
mixture was stirred at 40 ꢀC for 48 h and then neutralized
with NaHCO3, filtered, and the solvents were evaporated.
The crude product was extracted with EtOAc and washed
with brine, dried (Na2SO4), and the solvent was evaporated
to give the crude product, which was purified by flash chro-
matography (EtOAc/n-heptane, 1/10), to obtain 2 (3.32 g,
20
53%) as a colorless oil.13 [a]D ꢁ70.8 (c 0.35, CH2Cl2).
IR nmax film: (cmꢁ1) 3475, 2964, 1338, 1102, 1043, 1001,
914, 742. H NMR (CDCl3, 400 MHz, ppm): d 5.90 (m,
1
1H, allyl), 5.30 (m, 2H, allyl), 5.13 (s, 1H, CH), 4.78 (d,
1H, J¼6.0 Hz, CH), 4.55 (d, 1H, J¼6.0 Hz, CH), 4.41 (br
s, 1H, OH), 4.21 (m, 1H, CH), 4.08 (m, 1H, CH), 3.69 (m,
2H, CH), 3.22 (m, 1H, CH), 1.68 (m, 8H, CH2). 13C NMR
(CDCl3, 75 MHz, ppm): d 132.9, 121.6, 118.2, 107.6, 88.1,
85.6, 81.3, 68.8, 63.9, 35.6, 35.5, 23.5, 23.0. HRMS (EI)
m/z calcd for C13H20O5 (M)+: 256.1311, found: 256.1312.
5.1.4. Allyl 3-nitro-3-deoxy-6-O-tosyl-b-D-glucopyrano-
side. Compound 4 (1.07 g, 5.13 mmol) was dissolved in
a mixture of CH2Cl2 and pyridine (1/1, 0.2 mL). After cool-
ing the mixture to 0 ꢀC tosyl chloride (1.17 g, 6.16 mmol)
was added. The reaction was stirred overnight. The reaction
mixture was quenched with water, extracted with EtOAc,
and dried with Na2SO4 to yield after flash column chromato-
graphy (EtOAc/n-heptane, 1/10) the compound as a colorless
5.1.2. Allyl b-D-ribofuranoside (3). Compound 2 (2.05 g,
8.00 mmol) was dissolved in 80% acetic acid and heated to
40 ꢀC for 48 h. The reaction mixture was neutralized with
NaHCO3 and evaporated. Elution by flash column chromato-
graphy (EtOAc/n-heptane, 1/5) gave first recovered starting
material (0.41 g, 25%) followed by 3 as a colorless oil
oil (1.74 g, 91%). Rf 0.78 (EtOAc). IR nmax film: (cmꢁ1
)
3482, 1559, 1173, 1039, 981, 814, 730, 533. 1H NMR
(CD3OD, 400 MHz, ppm): d 7.79, (d, J¼8.2 Hz, 2H,
arom), 7.35 (d, J¼8.1 Hz, 2H, arom), 5.94–5.82 (m, 1H),
5.33–5.19 (m, 2H), 4.54 (t, J¼10.1 Hz, 1H), 4.41–4.33 (m,
2H), 4.34–4.25 (m, 2H), 4.18–4.03 (m, 3H), 3.94 (dq,
J¼9.7, 4.1, 2.1 Hz, 1H), 3.57–3.51 (m, 1H), 1.54 (s, 3H).
13C NMR (CDCl3, 75 MHz, ppm): d 145.3, 132.9, 130.0,
128.0, 118.6, 101.2, 92.2, 74.1, 70.9, 67.7, 22.1.
20
(604 mg, 40%). Rf 0.38 (EtOAc). [a]D ꢁ57.1 (c 0.77,
CH2Cl2). IR nmax film: (cmꢁ1) 3347, 2926, 1640, 1084,
1033, 993, 931, 633, 559. 1H NMR (CDCl3, 400 MHz,
ppm): d 6.04–5.97 (m, 1H, allyl), 5.28 (dd, J¼1.6, 1.8 Hz,
1H, allyl), 5.20 (dd, J¼1.8, 1.6 Hz, 1H, allyl), 4.97 (s, 1H,
CH), 4.20–3.91 (m, 7H, CH), 3.72–3.59 (m, 2H, OH), 3.67
(br s, 1H, OH). 13C NMR (CD3OD, 75 MHz, ppm): d
135.4, 116.8, 107.7, 84.8, 76.2, 72.7, 69.1, 65.0. HRMS (CI)
m/z calcd for C8H15O5 (M+H)+: 191.0919, found: 191.0920.
5.1.5. Allyl 6-azido-3-nitro-3,6-dideoxy-b-D-glucopyr-
anoside (5). Tosyl glucopyranoside (33 mg, 0.082 mmol)
was dissolved in DMF (1 mL). The reaction mixture was
heated to 55 ꢀC and NaN3 (10 mg, 0.16 mmol) was added.
After stirring overnight the reaction mixture was quenched
with water and extracted with EtOAc, to yield after flash col-
umn chromatography (EtOAc/n-heptane, 2/3) compound 5
5.1.3. Allyl 3-nitro-3-deoxy-b-D-glucopyranoside (4).
NaIO4 (7.68 g, 35.9 mmol) was dissolved in H2O (90 mL).
After cooling the reaction mixture to 0 ꢀC, compound 3
(6.83 g, 35.9 mmol) was added. The reaction mixture was
as a white solid (17 mg, 84%). Rf 0.28 (EtOAc/n-heptane,
)
3426, 2103, 1559, 1372, 1281, 1065. H NMR (CDCl3,
20
2/3). [a]D ꢁ16 (c 0.45, CH2Cl2). IR nmax film: (cmꢁ1
1

G. F. Busscher et al. / Tetrahedron 63 (2007) 3183–3188
3187
400 MHz, ppm): d 6.04–5.81 (m, 1H), 5.52–5.20 (m, 2H),
4.54 (t, J¼10.0 Hz, 1H), 4.42 (d, J¼7.8 Hz, 1H), 4.36 (d,
J¼7.5 Hz, 1H), 4.21–4.30 (m, 2H), 4.18 (t, J¼9.5 Hz, 1H),
3.63 (br s, 1H, OH), 3.59–3.38 (m, 3H), 3.24 (br s, 1H,
OH). 13C NMR (CDCl3, 75 MHz, ppm): d 132.9, 118.9,
100.9, 92.8, 75.5, 70.9, 70.8, 69.0, 51.2. HRMS (CI) m/z
calcd for C9H15O6N4 (M+H)+: 275.0992, found: 275.0981.
17.1 Hz, 2H), 4.59 (ddd, J¼5.0, 10.0, 12.5 Hz, 1H), 4.38
(d, J¼6.6 Hz, 1H), 4.51–4.31 (m, 1H), 4.14 (ddd, J¼1.1,
6.5, 12.6 Hz, 1H), 4.01 (dd, J¼8.0, 9.7 Hz, 1H), 3.75–3.64
(m, 1H), 3.47 (dd, J¼7.2, 13.0 Hz, 1H), 3.21 (dd, J¼3.4,
13.0 Hz, 1H), 2.71 (br s, 1H), 2.41–2.19 (m, 1H), 2.02 (q,
J¼12.1 Hz, 1H). 13C NMR (CDCl3, 75 MHz, ppm):
d 133.3, 118.8, 101.3, 85.7, 71.8, 71.2, 70.5, 53.8, 33.1.
HRMS (CI) m/z calcd for C9H14O5N4 (M+H)+: 259.1042,
found: 259.1049.
5.1.6. Allyl 6-azido-3-nitro-3,6-dideoxy-2-O-triethylsilyl-
b-D-glucopyranoside (6). The glucopyranoside 5 (13 mg,
0.047 mmol) was dissolved in CH2Cl2 (1 mL). After cooling
the reaction mixture to 0 ꢀC, triethylsilyl trifluoromethane-
sulfonate (16 mL, 0.071 mmol) and Et3N (17 mL, 1.2 mmol)
were added and the reaction mixture was stirred for 30 min
before quenching with satd aq NH4Cl, extraction with
CH2Cl2, and drying with Na2SO4. Flash column chromato-
graphy (EtOAc/n-heptane, 2/3) gave compound 6 (18 mg,
78%) as a colorless oil. Rf 0.52 (EtOAc/n-heptane, 1/10).
5.1.9. Propyl 3,6-diamino-3,4,6-trideoxy-2-O-triethyl-
silyl-b-D-glucopyranoside.
Compound
8
(30 mg,
0.081 mmol) was dissolved in MeOH, Raney-Ni in MeOH
(7 mL) was added and the hydrogenation was performed in
a Parr apparatus overnight at 3 bar, followed by filtration
over Hyflo-supercel, and evaporation of the solvent to yield
the compound as a white powder. IR nmax film: (cmꢁ1
)
3369, 2955, 1582, 1113, 1078, 823, 741. 1H NMR
(CD3OD, 400 MHz, ppm): d 4.13 (br d, J¼7.4 Hz, 1H),
3.77 (dd, J¼7.3, 16.4 Hz, 1H), 3.57–3.34 (m, 2H), 3.27 (br
s, 1H), 3.12–2.94 (m, 1H), 2.79–2.58 (m, 2H), 1.75 (dd,
J¼3.0, 12.8 Hz, 1H), 1.19 (dq, J¼7.2, 14.3 Hz, 2H), 0.92
(dd, J¼11.9, 24.2 Hz, 1H), 1.04–0.83 (m, 12H), 0.77–0.58
(m, 6H). 13C NMR (CD3OD, 75 MHz, ppm): d 104.6, 79.0,
75.0, 72.0, 54.7, 46.7, 36.5, 24.3, 11.1, 7.4, 6.3. HRMS
(ESI) m/z calcd for C15H35O3N2Si (M+H)+: 319.2417, found:
319.2404.
20
[a]D ꢁ14 (c 0.46, CH2Cl2). IR nmax film: (cmꢁ1) 3425,
2876, 2099, 1557, 1067, 815, 745. 1H NMR (CDCl3,
400 MHz, ppm): 6.05–5.90 (m, 1H), 5.42–5.18 (m, 2H),
4.49 (t, J¼6.4 Hz, 1H), 4.45–4.31 (m, 1H, CH), 4.37 (d,
J¼7.6 Hz, 1H), 4.19–3.92 (m, 3H), 3.52–3.49 (m, 3H),
1.14–0.79 (m, 9H, TES), 072–0.45 (m, 6H, TES). 13C
NMR (CDCl3, 75 MHz, ppm): d 133.1, 118.5, 101.6, 95.0,
75.3, 72.3, 70.9, 69.4, 51.4, 6.9, 5.1. HRMS (FAB) m/z calcd
for C15H29O6N4Si (M+H)+: 389.1856, found: 389.1863.
5.1.7. Allyl 6-azido-3-nitro-3,4,6-trideoxy-2-O-triethyl-
silyl-b-D-glucopyranoside (7). Compound 5 (100 mg,
0.26 mmol) was dissolved in CH2Cl2 (2 mL) and cooled to
0 ꢀC. Et3N (64 mL, 0.47 mmol), Ac2O (49 mL, 0.51 mmol),
and DMAP (cat.) were added and the reaction was stirred for
30 min. The reaction mixture was quenched with a 0.1 M
HCl solution and extracted with CH2Cl2 and dried with
Na2SO4. The compound was dissolved in EtOH (2 mL)
and NaBH4 (20 mg, 0.51 mmol) was added to the reaction
mixture. The reaction mixture was stirred at room tempera-
ture for 2 h and quenched with acetone and evaporated. Flash
column chromatography (EtOAc/n-heptane, 1/5) gave the
compound 7 (87 mg, 91%) as colorless oil. Rf 0.60 (EtOAc/
5.1.10. Propyl 3,6-diamino-3,4,6-trideoxy-b-D-glucopyr-
anoside (9). The diamino glucopyranoside (26 mg,
0.081 mmol) was dissolved in a 2 M HCl solution in EtOAc
(5 mL). The reaction mixture was stirred for 1.5 h and the
solvent was evaporated and the residue was dissolved in
t-BuOH and evaporated. This was repeated two times, to
yield compound 9 as a white powder (22 mg, quant, two
20
steps). [a]D ꢁ3.5 (c 1.1, H2O). 1H NMR (D2O, 300 MHz,
ppm): d 4.53 (d, 1H, J¼7.2 Hz), 4.10–3.87 (m, 2H), 3.71–
3.60 (m, 1H), 3.51–3.42 (m, 2H), 3.40–2.30 (m, 1H),
3.22–3.05 (m, 1H), 2.32–2.15 (m, 1H), 1.81–1.58 (m, 3H),
0.95 (t, J¼7.2 Hz, 3H). 13C NMR (D2O, 75 MHz, ppm):
d 103.1, 73.3, 71.5, 69.5, 52.3, 43.2, 32.2, 23.2, 10.7.
HRMS (CI) m/z calcd for C9H20O3N2 (M+H)+: 205.1539,
found: 205.1552.
20
n-heptane, 2/3). [a]D ꢁ29 (c 0.32, CH2Cl2). IR nmax film:
(cmꢁ1) 2954, 2877, 2099, 1558, 1130, 1006, 743. 1H
NMR (CDCl3, 400 MHz, ppm): d 6.10–5.82 (m, 1H,
allyl), 5.49–5.18 (m, 2H, allyl), 4.64–4.54 (m, 1H, H-3),
4.46–4.38 (m, 1H, H-7a), 4.32 (d, 1H, J¼7.6 Hz, H-1),
4.15–4.00 (m, 2H, H-7b and H-2), 3.70–3.65 (m, 1H, H-5),
3.62–3.39 (m, 1H, H-6a), 2.20 (dd, 1H, J¼3.5 Hz, H-6b),
2.20 (ddd, 1H, J¼1.9 Hz, H-4a), 2.04 (q, 1H, J¼5.9 Hz,
H-4b), 1.01–0.81 (m, 9H, TES), 0.71–0.51 (m, 6H, TES). 13C
NMR (CDCl3, 75 MHz, ppm): d 133.2, 118.4, 101.8, 88.1,
72.5, 71.7, 70.7, 54.0, 33.6, 6.9, 5.2. HRMS (FAB) m/z calcd
for C15H29O5N4Si (M+H)+: 373.1907, found: 373.1904.
5.1.11. Allyl (20-C-allyl-30,40-di-O-benzyl-20,60-dideoxy-
a/b-D-glucopyranosyl)-(10,2)-6-azido-3,4,6-trideoxy-
3-nitro-b-D-glucopyranoside (11). AgOTf (142 mg,
0.552 mmol) was co-evaporated three times with toluene
and activated molecular sieves (50 mg) were added. In a
separate flask, compounds 8 (45 mg, 0.18 mmol) and 10
(93 mg, 0.19 mmol) were co-evaporated three times with
toluene after which DCE (3 mL) and TTBP (137 mg,
0.552 mmol) were added and the mixture was stirred for
30 min, before transferring the solution under argon to the
flask containing AgOTf (at 0 ꢀC). The reaction mixture
was stirred overnight at room temperature. The reaction
mixture was quenched with 50 equiv pyridine and filtered
through a layer of Hyflo-supercel. The solvents were evapo-
rated and flash column chromatography (EtOAc/n-heptane,
1/20 to 1/2) gave product 11 (54 mg, 45%) as colorless oil
and 12 (23 mg, 32%) as well as remaining starting material
(4 mg, 9%). Rf 0.45 (EtOAc/n-heptane, 1/1). IR nmax film:
(cmꢁ1) 2915, 2098, 1557, 1453, 1371, 1283, 1066, 919,
5.1.8. Allyl 3-nitro-3,4,6-trideoxy-6-azido-b-D-glucopyr-
anoside (8). Compound 7 (28 mg, 0.075 mmol) was dis-
solved in a 2 M HCl solution and stirred for 10 min at
room temperature. The reaction mixture was evaporated
and flash column chromatography (EtOAc/n-heptane, 1/5)
gave 8 (12 mg, 63%) as colorless oil. Rf 0.55 (EtOAc/n-hep-
)
3431, 2924, 2099, 1556, 1065, 1036. H NMR (CDCl3,
20
tane, 1/1). [a]D ꢁ26 (c 1.4, CH2Cl2). IR nmax film: (cmꢁ1
1
400 MHz, ppm): d 6.07–5.84 (m, 1H), 5.32 (dd, J¼35.2,

3188
G. F. Busscher et al. / Tetrahedron 63 (2007) 3183–3188
1
Q.; Westhof, E. Structure 2001, 9, 647–658; (c) Vicens, Q.;
Westhof, E. Chem. Biol. 2002, 9, 747–755; (d) Vicens, Q.;
Westhof, E. J. Mol. Biol. 2003, 326, 1175–1188.
8. Conditions explored include CF3COOH/CH2Cl2/H2O (10/10/
2), and CH3COOH/H2O (80/20), CH3COOH/H2O/(HOCH2)2
(14/6/3), 2 M HCl in MeOH, p-TsOH in MeOH, FeCl3$SiO2
in CHCl3, FeCl3$SiO2, and CH3COOH in CHCl3.
9. Greene, T. W.; Wuts, P. G. M. Protective Groups in Organic
Synthesis; Wiley: New York, NY, 1999; pp 211–215.
10. Kim, K. S.; Song, Y. H.; Lee, B. H.; Hahn, C. S. J. Org. Chem.
1986, 51, 404–407.
740, 699. H NMR (CDCl3, 400 MHz, ppm): d 7.42–7.21
(m, 10H), 6.09–5.69 (m, 2H), 5.51–5.35 (m, 1H), 5.27–
5.21 (m, 1H), 5.15–4.95 (m, 3H), 4.92–4.81 (m, 3H),
4.72–4.65 (m, 4H), 4.49–4.31 (m, 2H), 4.32–4.04 (m, 2H),
3.82–3.62 (m, 1H), 3.51–3.35 (m, 6H), 2.52–2.32 (m, 1H),
2.28–2.10 (m, 2H), 1.91–1.82 (m, 1H). HRMS (FAB) m/z
calcd for C32H40O8N7 (M+H)+: 650.2938, found: 650.2918.
5.1.12. 1,5-Anhydro-2-allyl-2,6-dideoxy-D-arabino-hex-l-
enitol (12). Yield 23 mg, 32%: Rf 0.75 (EtOAc/n-heptane,
1/2). IR nmax film: (cmꢁ1) 3336, 2070, 1120, 1090, 973,
1
11. Phillips, D.; Chamberlin, A. R. J. Org. Chem. 2002, 67, 3194–
3204.
12. van Heeswijk, W. A. R.; Goedhart, J. B.; Vliegenthart, J. F. G.
Carbohydr. Res. 1977, 58, 337–344.
823, 697. H NMR (CD3OD, 400 MHz, ppm): d 7.39–7.23
(m, 10H), 6.23 (s, 1H), 5.81–5.52 (m, 1H), 5.11–4.91 (m,
2H), 4.66 (s, 2H), 4.52 (dd, J¼11.5, 45.9 Hz, 2H), 4.22–
4.07 (m, 1H), 3.94 (d, J¼4.3 Hz, 1H), 3.84 (dd, J¼4.3,
5.5 Hz, 1H), 3.65 (dd, J¼7.2, 13.2 Hz, 1H), 3.45–3.22 (m,
1H), 2.72 (dq, J¼6.3, 15.4 Hz, 2H). 13C NMR (CD3OD,
75 MHz, ppm): d 139.4, 135.8, 114.6, 110.5, 75.0, 73.3,
73.3, 71.8, 70.9, 49.4, 32.4. HRMS (EI) m/z calcd for
C23H26O3N3 (M+H)+: 392.1974, found: 392.1979.
13. Ghosh, A. K.; Liu, W. J. Org. Chem. 1996, 61, 6175–6182.
14. Investigations on the synthesis of the allyl b-D-ribofuranoside 3
in a single synthetic step by Fisher glycosidation of D-ribose
with allyl alcohol activated by CuSO4 and H2SO4, SnCl4 or
Amberlite IR-120 (H+, ion-exchange resin) did not improve
matters since the a-ribofuranoside was always obtained as
the main product.
15. (a) Baer, H. H. Chem. Ber. 1960, 93, 2865–2870; (b) Baer,
H. H.; Kienzle, F. Can. J. Chem. 1963, 410, 1606–1607.
16. The large 1H NMR coupling constants of ring protons of 4 are
highly indicative of an all-equatorial substitution pattern.
Furthermore, spectral data are nearly identical to those reported
for the methyl glycoside: Sofia, M. J.; Hunter, R.; Chan, T. Y.;
Vaughan, W.; Dulina, R.; Wang, H.; Gange, D. J. Org. Chem.
1998, 63, 2802–2803.
5.1.13. (2R,3S,4R,4R,9R,11S,13S,13R)-2,11-Bis(azido-
methyl)-3,4-bis(benzyloxy)-13-nitro-2,3,4,4,5,8,9,11,12,
13,13,14-dodecahydrodipyrano[2,3-20,30][1,4]dioxecine
(13). Compound 11 (6 mg, 910ꢁ3 mmol) was co-evaporated
´
three times with toluene and dissolved in toluene (0.5 mL)
again. Argon was bubbled through the solution for 10 min.
To the reaction mixture was added Hoveyda–Grubbs cata-
lyst31 (9.8 mg, 1510ꢁ3 mmol). The mixture was stirred for
´
4 h at room temperature and evaporated. Column chromato-
graphy (EtOAc/n-heptane, 1/10 to 1/1) gave 13 (6 mg, 99%)
as a white solid. Rf 0.34 (EtOAc/n-heptane, 1/1). IR nmax
17. (a) Baer, H. H. Meth. Carbohydr. Chem. 1972, 6, 245–249;
(b) Lichtenthaler, F. W. Meth. Carbohydr. Chem. 1972, 6,
250–260.
film: (cmꢁ1) 2924, 2100, 1557, 1067, 1028, 916. H NMR
1
18. It was found that silylation of C-4 could not be fully suppressed
in larger scale reactions.
19. This compound is very unstable and was used in the next step
without purification.
20. Yoshikawa, M.; Ikeda, Y.; Kayakiri, H.; Kitagawa, I.
Heterocycles 1982, 17, 209–214.
(CDCl3, 400 MHz, ppm): d 7.39–7.21 (m, 10H), 5.41–4.32
(m, 12H), 4.30–4.04 (m, 2H), 3.81–3.62 (m, 1H),
3.51–3.32 (m, 6H), 2.52–2.32 (m, 1H), 2.28–2.10 (m, 2H),
1.91–1.82 (m, 1H). HRMS (ESI) m/z calcd for
C30H35O8N7 (M+Na)+: 644.2445, found: 644.2450.
21. Greenberg, W. A.; Priestley, E. S.; Sears, P. S.; Alper, P. B.;
Rosenbohm, C.; Hendrix, M.; Hung, S.-C.; Wong, C.-H.
J. Am. Chem. Soc. 1999, 121, 6527–6541.
References and notes
1. Beaucaire, G. J. Chemother. 1995, 7, 111–123.
2. Busscher, G. F.; Rutjes, F. P. J. T.; van Delft, F. L. Chem. Rev.
2005, 105, 775–791.
3. Meyer zu Reckendorf, W. Dtsch. Apoth. Ztg. 1972, 41,
1617–1622.
22. (a) Zembower, T. R.; Noskin, G. A.; Postelnick, M. J.; Nguyen,
C.; Peterson, L. R. Int. J. Antimicrob. Agents 1998, 10, 95–105;
(b) Nagai, J.; Takano, M. Drug Metab. Pharmacokinet. 2004,
19, 159–170.
23. Neonakis, I.; Gikas, A.; Scoulica, E.; Manios, A.;
Georgiladakis, A.; Tselentis, Y. Int. J. Antimicrob. Agents
2003, 22, 526–531.
4. Venot, A.; Swayze, E. E.; Griffey, R. H.; Boons, G.-J.
ChemBioChem 2004, 5, 1228–1236.
5. Rao, Y.; Venot, A.; Swayze, E. E.; Griffey, R. H.; Boons, G.-J.
Org. Biomol. Chem. 2006, 4, 1328–1337.
24. Haddad, J.; Vakulenko, S.; Mobeshary, S. J. Am. Chem. Soc.
1999, 121, 1922–1923.
25. Asensio, J. L.; Hidalgo, A.; Bastida, A.; Torrado, M.; Corzana,
6. NMR spectroscopy: (a) Fourmy, D.; Recht, M. I.; Blanchard,
S. C.; Puglisi, J. D. Science 1998, 274, 1367–1371; (b)
Recht, M. I.; Fourmy, D.; Blanchard, S. C.; Dahlquist, K. D.;
Puglisi, J. D. J. Mol. Biol. 1996, 262, 421–436; (c) Fourmy,
D.; Recht, M. I.; Puglisi, J. D. J. Mol. Biol. 1998, 277, 333–
345; (d) Fourmy, D.; Recht, M. I.; Puglisi, J. D. J. Mol. Biol.
1998, 277, 347–362; (e) Lynch, S. R.; Puglisi, J. D. J. Mol.
Biol. 2001, 306, 1037–1058; (f) Jiang, L. C.; Patel, D. J. Nat.
Struct. Biol. 1998, 5, 769–774.
ꢀ
~
ꢀ
F.; Luis Chiara, J. L.; Garcıa-Junceda, E.; Canada, J.; Jimenez-
Barbero, J. J. Am. Chem. Soc. 2005, 127, 8278–8279.
26. Blount, K. F.; Zhao, F.; Hermann, T.; Tor, Y. J. Am. Chem. Soc.
2005, 127, 9818–9829.
27. Unpublished results.
28. Lear, M. J.; Yoshimura, F.; Hirama, M. Angew. Chem., Int. Ed.
2001, 40, 946–949.
29. Determined by 1H NMR.
7. X-ray crystallography: (a) Carter, A. P.; Clemons, W. M.;
Brodersen, D. E.; Morgan-Warren, R. J.; Wimberly, B. T.;
Ramakrishnan, V. Nature 2000, 407, 340–348; (b) Vicens,
30. Kanemitsu, T.; Seeberger, P. H. Org. Lett. 2003, 5, 4541–4544.
31. 1,3-Bis-(2,4,6-trimethylphenyl)-(2-imidazolidene)dichloro(O-
isopropoxyphenylmethylene)ruthenium.