C-glycosylation of oxygenated naphthols with 3-dimethylamino- 2,3,6-trideoxy-L-arabino-hexopyranose and 3-azido- 2,3,6-trideoxy-D-arabino-hexopyranose

Article abstract of DOI:10.1071/CH02236

In connection with studies directed towards the synthesis of the pyranonaphthoquinone antibiotic medermycin, C-aryl glycosides were prepared by C-glycosylation of naphthols with glycosyl donors. Boron trifluoride diethyl etherate proved to be a suitable Lewis acid to promote the C-glycosylation, and use of the azido glycosyl donor proved more successful than using the dimethylamino glycosyl donor. 5-Hydroxy-1,4-dimethoxynaphthalene underwent facile C-glycosylation with two particular glycosyl donors, whereas 3-bromo-5-hydroxy-1,4-dimethoxynaphthalene was not an effective coupling partner with the same glycosyl donors. These studies indicate that subtle steric and electronic effects need to be considered in order to fine-tune C-glycosylations when using highly functionalized glycosyl donors.

Full text of DOI:10.1071/CH02236

Full Papers
CSIRO PUBLISHING
www.publish.csiro.au/journals/ajc
Aust. J. Chem. 2003, 56, 787–794
C-Glycosylation of Oxygenated Naphthols with 3-Dimethylamino-
2,3,6-trideoxy-l-arabino-hexopyranose and 3-Azido-
2,3,6-trideoxy-d-arabino-hexopyranose
Margaret A. Brimble,A,C Roger M. Davey,B Malcolm D. McLeodB and Maureen MurphyB
A Department of Chemistry, University of Auckland, Auckland, New Zealand.
B School of Chemistry, University of Sydney, Sydney 2006, Australia.
C Author to whom correspondence should be addressed (e-mail: m.brimble@auckland.ac.nz).
In connection with studies directed towards the synthesis of the pyranonaphthoquinone antibiotic medermycin,
C-aryl glycosides were prepared by C-glycosylation of naphthols with glycosyl donors. Boron trifluoride diethyl
etherateprovedtobeasuitableLewisacidtopromotetheC-glycosylation, anduseoftheazidoglycosyldonorproved
more successful than using the dimethylamino glycosyl donor. 5-Hydroxy-1,4-dimethoxynaphthalene underwent
facile C-glycosylation with two particular glycosyl donors, whereas 3-bromo-5-hydroxy-1,4-dimethoxynaphthalene
was not an effective coupling partner with the same glycosyl donors. These studies indicate that subtle steric and
electronic effects need to be considered in order to fine-tune C-glycosylations when using highly functionalized
glycosyl donors.
Manuscript received: 15 November 2002.
Final version: 24 April 2003.
Introduction
more recently, Wyeth researchers[5] presented sophisticated
NMR evidence in support of the original structure for
medermycin/lactoquinomycin A (1).
Medermycin, an antibiotic isolated from Streptomyces
tanashiensis,[1] is a unique member of the pyranonaph-
thoquinone family of antibiotics[2] in that it contains a
β-C-glycoside linkage to the amino sugar d-angolosamine.
Medermycin was found to be identical to an anticancer
agent, lactoquinomycin A, that was isolated from the same
source by Tanaka et al.[3] ten years later. The point of attach-
ment of the carbohydrate component (d-angolosamine) of
medermycin was initially assigned as ortho to the hydroxyl
group at C8 in structure (1) (see Diagram 1). Subsequent
chemical and spectroscopic investigations by Morin and
coworkers[4] resulted in temporary revision of the structure
of medermycin/lactoquinomycin A to structure (2), where
the C-glycoside is attached to the kalafungin (3) skeleton
at the C10 position* para to the hydroxyl group. However,
To date, only one lengthy synthesis of medermycin (1)
has been reported[6] in which the C-glycoside in the pyra-
nonaphthalene skeleton was assembled by addition of a
sulfonylphthalide to an enone. The C-glycosidic linkage was
introduced as a d-olivoside early in the synthetic sequence,
before the phthalide annulation step, which rendered it neces-
sary to manipulate the functional groups on the sugar moiety
to give the desired d-angolosaminide amino sugar at the
conclusion of the synthesis.
Our synthetic effort[7,8] towards medermycin (1) focussed
on attachment of a carbohydrate that already bears an amino
or azido group to the ortho position of a naphthol pre-
cursor. We, therefore, herein report our detailed studies on the
Me
O
6
OH
7
Me
Me
5
8
O
O
4
OH
HO
O
Me
O
O
O
H
OH
9
O
7
6
5
8
3
4
H
10
O
Me2N
H
H
O
O
O
Me
2
9
1
O
H
3
O
10
H
O
O
2
1
HO
O
NMe2
(1)
(2)
kalafungin (3)
Diagram 1.
∗ The numbering system of the pyranonaphthoquinone skeleton is used rather than the numbering system used by Morin and coworkers.[4]
© CSIRO 2003 10.1071/CH02236 0004-9425/03/080787

788
M. A. Brimble et al.
Me
Me
HO
HO
Me
O
HO
H
O
O
O
OH
O
O
OH
O
OH
(1)
Me2N
H
Me2N
Me
O
H
O
O
O
H
Me
Me
HO
PO
O
OMe
OMe
O
O
O
O
OMe
R
6
3
6
3
X
X
Me
+
OSiMe3
O
OMe
(4) X = N3
(6) X = N3, R = H;
(7) X = N3, R = Br
(5) X = NMe2
(8) X = NMe2, R = H; (9) X = NMe2, R = Br
OH
OMe
OMe
Me
R
RO
O
+
X
OMe
(10) X = N3, R = H
(11) X = N3, R = Ac
(12) X = NMe2, R = H
(13) R = H
(14) R = Br
Scheme 1.
Me
O
Me
AcO
N3
AcO
N3
O
+
+
OH
OH
OH
Me
Me
D-(17b)
D-(17a)
Me
AcO
80ºC
AcO
AcO
NaN3
O
O
Me
O
HOAc
H2O
AcO
+
AcO
N3
OH
O
D-(15)
D-(16)
N3
OH
D-(17c)
D-(17d)
Ac2O, py, CH2Cl2
Me
O
Me
Me
Me
O
AcO
N3
AcO
+
AcO
N3
AcO
O
O
+
3
1
N3
OMe
OMe
N3
OAc
OAc
D-(11a)
41%
D-(11b)
17%
MeOH
D-(18a)
α-arabino
D-(18b)
β-ribo
Montmorillonite K-10
benzene, ∆
Me
O
Me
O
Me
O
Me
AcO
N3
AcO
N3
AcO
N3
AcO
+
OMe
O
+
+
OMe
D-(11d)
N3
OAc
OAc
D-(11c)
D-(18c)
D-(18d)
β-arabino
α-ribo
0.5%
(18a) : (18b) : (18c) : (18d)
9.1 : 4.8 : 2.3 : 1
Scheme 2.

C-Glycosylation of Oxygenated Naphthols
789
direct C-glycosylation of oxygenated naphthols with azido
and dimethylamino glycosyl donors (10), (11), and (12).
Several of the C-glycosides prepared in the present work are
suitable for further elaboration to analogues of medermycin
in which the carbohydrate is linked to the C8 position, and a
recent synthesis[8] of an azido analogue of medermycin was
based on this study.
ratio of d-arabino to d-ribo isomers to be approximately 2 : 1.
Conversion of the isomeric mixture of 1-O-acetylpyranoses
(18) to methyl glycosides (11) was then achieved using
montmorillonite K-10 resin and methanol in benzene, which
allowed separation of the major α-d-arabino azide (11a) (41%
yield) from its β-d-ribo counterpart (11b) (17% yield) by
careful flash chromatography. Purification of the β-d-arabino
(11c) and α-d-ribo isomer (11d) proved more difficult and
they were afforded as an inseparable mixture in 0.5% yield.
Of interest is the low proportion of α-d-ribo isomer (18d)
obtained from the initial addition of hydrazoic acid and
acetylation of pseudo-glycal (16).This is rationalized by con-
sidering that in a chair-like conformation, the C3 and C1
substituents must both assume an axial orientation, which
would give rise to severe 1,3-diaxial interactions that oppose
the stability offered by the anomeric effect associated with the
α-acetoxy group at C1. The overall predominance of the α-d-
arabino isomer (18a) is due to the favourable combination of
an equatorial substituent at C3 and the stabilizing anomeric
effect contributed by the axial C1 acetoxy group (Fig. 1).
Given that 3,4-di-O-acetyl-l-rhamnal is more readily
available than 3,4-di-O-acetyl-d-rhamnal, optimization of
the glycosyl donor synthesis and C-glycosylation studies was
performedonthel-series.Tothisend, azidoglycosyldonor l-
(11a) was converted into methyl α-l-angolosaminide, l-(12),
in a single step, albeit in 44% yield, by hydrogenation over
10% palladium on charcoal in methanol in the presence
of excess formaldehyde. Alternatively, the transformation
could be achieved in two steps in higher yield. Thus, hydro-
genation of l-(11a) over 10% palladium on charcoal in
methanol afforded methyl α-l-acosaminide l-(19) in 86%
yield (Scheme 3). The reductive methylation was then con-
ducted using similar conditions except that excess aqueous
formaldehyde was also present.After chromatography, l-(12)
was obtained in 79% yield. Whereas cleavage of the acetate
took only a few hours to go to completion, reductive methy-
lation of the azide group required reaction overnight.
Results and Discussion
Our synthetic approach to medermycin (1) hinged on
the furofuran annulation of azido C-glycosyl-3-acetyl-
1,4-naphthoquinone (4) or dimethylamino C-glycosyl-3-
acetyl-1,4-naphthoquinone (5) with 2-trimethylsilyloxyfuran
followed by oxidative rearrangement of the resultant adduct
(Scheme 1).This strategy has been successfully applied to the
synthesis of a 2-deoxyglucosylpyranonaphthoquinone[7] and
the spiroacetal-containing pyranonaphthoquinone griseusin
A.[9] Itwas, therefore, envisagedthatC-glycosylnaphthalenes
(6)–(9) would be suitable precursors to the key C-
glycosylnaphthoquinones (4) and (5), and that they, in turn,
would be available by treatment of azido glycosyl donors
(10) or (11), or dimethylamino glycosyl donor (12) with
naphthols (13) and (14). The bromine substituent at C3 in
C-glycosylnaphthalenes (7) or (9) is required in order to pro-
vide functionality for the introduction of the C3 acetyl group
in the C-glycosylnaphthoquinones (4) and (5). Our synthetic
efforts wereinitially directed towardsestablishingan efficient
protocol for C-glycosylation of naphthols (13) and (14) with
a glycosyl donor that could be elaborated to d-angolosamine
which is present in medermycin.
Synthesis of the glycosyl donors (10), (11), and (12) was
based on existing methodology reported by Monneret and
coworkers[10,11] for the l-isomers (Scheme 2). The synthe-
sis could be adapted to both enantiomeric series in that the
starting material for the d-series, 3,4-di-O-acetyl-d-rhamnal
(15), can be prepared from 3,4,6-tri-O-acetyl-d-glucal,[12]
whereas the l-isomer of this compound, as used by Monneret
and coworkers,[10,11] is commercially available. We herein
provide full experimental details for the preparation of the
d-isomers of the azido glycosyl donors (10) and (11) that
have the correct absolute configuration for the synthesis of
medermycin.
Initial C-glycosylation studies were carried out using the
l-series.The naphthols (13)[7] and (14)[13] used in the present
study were prepared as described previously and our attention
first focussed on the use of the methyl α-l-angolosaminide
l-(12)astheglycosyldonor, giventhepresenceofadimethyl-
amino group on the C-glycoside in medermycin. Based on
3,4-Di-O-acetyl-d-rhamnal (15) was heated with water to
afford pseudo-glycal (16). Subsequent treatment of pseudo-
glycal (16) with sodium azide and acetic acid for 24 h gave an
inseparable mixture of 3-azidopyranoses (17a)–(17d) in 93%
yield, which upon standard acetylation also gave an insepa-
rable mixture of 1-O-acetylpyranoses (18a)–(18d) in 95%
yield. The ratio of diastereomers (18a) : (18b) : (18c) : (18d)
obtained was 9.1 : 4.8 : 2.3 : 1, which thereby established the
Me
Me
O
O
AcO
AcO
1
N3
1
3
3
OAc
N3
OAc
(18d)
(18a)
Fig. 1.
Me
Me
Me
HO
HO
AcO
N3
O
H2, Pd/C
36% H2CO, MeOH
H2, Pd/C
O
O
MeOH, Et3N
Me2N
OMe
H2N
OMe
OMe
L-(12)
L-(11a)
L-(19)
Scheme 3.

790
M. A. Brimble et al.
Table 1. C-Glycosylation of naphthols (13), (14), and (20)
Entry
Glycosyl Donor
Naphthol
Conditions
C-Glycoside
Me
Me
OH OMe
CH3CN, 0ºC
HO
HO
OMe
OMe
OH
O
O
1
2
3
BF3.Et2O (2 eq.)
BF3.Et2O (4 eq.)
BF3.Et2O (6 eq.)
<5%
31%
52%
Me N
2
Me N
2
OMe
OMe
(13)
L-(12)
(8)
Me
OH OMe
Me
HO
Br
HO
OMe
OH
CH3CN, 0ºC
O
O
BF3.Et2O (2 eq.)
<5%
4
Br
N
3
N
3
OMe
OMe
(14)
L-(10)
(7)
OMe
OMe
Me
Me
O
OH OMe
HO
HO
OH
O
CH3CN, 0ºC
5
BF3.Et2O (2 eq.)
34%
N
3
Br
N
3
OMe
OMe
(20)
Br
L-(10)
(21)
O
OMe
OMe
Me
O
OH OMe
Me
HO
HO
OH
CH3CN, 0ºC
BF3.Et2O (2 eq.)
42%
6
N
3
N
3
OMe
OMe
(13)
L-(10)
Me
(6)
OMe
OMe
OH OMe
Me
AcO
N
AcO
OH
O
O
CH3CN, 0ºC
7
66%
BF3.Et2O (2 eq.)
OMe
3
N
3
OMe
(13)
L-(11a)
Me
L-(22)
OMe
OH OMe
Me
AcO
AcO
OMe
OMe
CH3CN, 0ºC
OH
O
O
8
BF3.Et2O (2 eq.)
60%
N
3
N
3
OMe
OMe
(13)
D-(11a)
D-(22)
similar studies using 2-deoxy-d-glucosyl donors,[13,14] naph-
thol (13) was reacted with methyl glycoside l-(12) using
boron trifluoride diethyl etherate as the Lewis acid in ace-
tonitrile at 0◦C (entries 1–3, Table 1). In this case, it was
found that donor l-(12) required six molar equivalents of
boron trifluoride etherate to give an optimum yield of 52%.
This is possibly due to the Lewis basic dimethylamino and
hydroxyl sites on donor l-(12), which can coordinate to the
boron trifluoride and, thereby, inhibit formation of the reac-
tive oxonium ion intermediate. Only one C-glycoside product
was formed and that was established to be the β-C-glycoside
(8). O-Glycoside products were not observed.
1
In the H NMR spectrum for (8), H2ꢀax resonated as a
doublet of doublet of doublets (ddd) at δ 1.49, with large
axial–axial couplings to H1ꢀ and H3ꢀ (both 10.8 Hz), and
a large geminal coupling to H2ꢀeq (12.5 Hz). The anomeric

C-Glycosylation of Oxygenated Naphthols
791
Me
Me
Our efforts were next focussed on the C-glycosylation
of naphthol (13) with azido glycosyl donor l-(10) to afford
C-glycoside (6) in 42% yield using boron trifluoride ether-
ate (2.0 equiv.) in acetonitrile at 0◦C (entry 6, Table 1). In
this case, increasing the ratio of Lewis acid did not improve
the yield. The yield of C-glycoside l-(22) improved to 66%
when azido glycosyl donor l-(11), in which the hydroxyl
group was protected as an acetate, was used (entry 7,Table 1).
Protection of the azido glycosyl donor l-(10) as the acetate
derivative l-(11) may improve the yield obtained in the
C-glycosylation because by masking the secondary hydroxyl
group, unproductive oligomerization of the glycosyl donor is
prevented.
HO
N3
AcO
O
O
K2CO3, MeOH
OMe
N3
OMe
L-(11a)
L-(10)
Scheme 4.
proton, H1ꢀ, resonated as a doublet of doublets at δ 5.02, with
ꢀ
ꢀ
ꢀ
ꢀ
coupling constants J1 ,2 ax 10.8 Hz and J1 ,2 eq 1.8 Hz. The
magnitude of the larger coupling constant is consistent with
an axial–axial coupling, thereby supporting the formation of
a β-glycoside. The observation of four aromatic hydrogens
and a naphthol OH proton at δ 9.76 provided evidence for the
formation of a C-glycoside rather than an O-glycoside.
A setback to the use of C-glycoside (8) as a key intermedi-
ate in the synthesis of medermycin occurred when attempts
to methylate the hydroxyl group resulted in formation of a
trimethylammonium salt. This was due to quaternization of
the amine function occurring preferentially over methyl ether
formation. In order to bypass this problem it was decided to
return to the C-glycosylation reaction, this time using a donor
in which the amine functionality was suitably masked as an
azide.
For the purpose of synthesizing medermycin, use of the
correct azido C-glycosyl donor d-(11) was also investigated.
This reaction was conducted on a larger scale (0.5–1.5 g)
using two equivalents of boron trifluoride diethyl etherate and
consistently gave yields of C-glycoside d-(22) in the range
59–62%(entry8,Table1).Thematerialobtainedwasspectro-
scopically identical to the l-isomer l-(22) and exhibited an
22
optical rotation of [α]D +70.0◦. The 1H NMR spectrum for
d-(22) was consistent with the proposed structure, featuring
expected signals arising from both the azido sugar and the
naphthalene moieties. A three-proton singlet at δ 2.16 was
assigned to the 4ꢀ-O-acetyl protons, and a three-proton dou-
blet at δ 1.29 was assigned to the C6ꢀ methyl protons. The
axial proton H2ꢀax gave rise to a doublet of doublet of doublets
at δ 1.71, with large axial–axial couplings, both 11.3 Hz, to
Our attention next turned to the coupling of glycosyl donor
l-(10)with3-bromonaphthol(14)(entry4,Table1). Donor l-
(10) was readily prepared by deacetylation of the 4-O-acetyl
sugar l-(11) (Scheme 4). When trimethylsilyl triflate was
used as the Lewis acid in dichloromethane no C-glycoside
was isolated from the reaction despite the use of three molar
equivalents of the promoter. Use of boron trifluoride was also
disappointing in that the desired C-glycoside (7) was only
formed in less than 5% yield.
This latter result was in direct contrast to a similar exper-
iment using 2-bromonaphthol (20)[13] in which C-glycoside
(21) was obtained in 34% yield (entry 5, Table 1). These
results reflect those obtained when using a tri-O-benzyl-
2-deoxy-d-glucosyl donor[13] in that 3-bromonaphthol (14)
proved more unreactive towards Lewis acid mediated
C-glycosylation than 2-bromonaphthol (20). In light of this
situation we were forced to carry out the C-glycosylation
using 5-hydroxy-1,4-dimethoxynaphthalene (13) and to
introduce the bromine at C3 on the naphthalene ring after
the C-glycosylation step.
We can offer some suggestions for the differences in reac-
tivity observed for glycosyl acceptors (13), (14), and (20).
Bromine substitution on the aromatic ring is observed to
reduce the yield of C-glycosylation dramatically for (14)
but also for (20). This observation is consistent with the
electron-withdrawing nature of this substituent, which deac-
tivates the aromatic ring towards electrophilic substitution.
This may be attenuated, in part, for (20) by the ability of the
C2 bromine to stabilize the relevant cationic intermediate by
resonance, which leads to higher yields for this derivative.
The influence of steric effects could also reduce the reac-
tivity of the aromatic coupling partner. In particular, the C3
bromine substituent of (14) would give rise to interactions
with the adjacent C4 methoxy group, which in turn, would
afford greater peri-interactions between this substituent and
the C5 hydroxyl that directs the C-glycosylation.
H1ꢀ and H3ꢀ. The geminal coupling constant J2 ax,2 eq was
13.1 Hz. The equatorial proton H2ꢀeq, at δ 2.46, also res-
onated as a doublet of doublet of doublets but with small
equatorial–axial coupling constants to H3ꢀ and H1ꢀ (4.9 and
1.9 Hz, respectively).The anomeric proton appeared as a dou-
blet of doublets at δ 5.06 with coupling constants of 11.3
and 1.9 Hz. The larger coupling constant corresponds to an
axial–axial relationship between H1ꢀ and H2aꢀ x and is, there-
fore, consistent with the formation of a β-glycoside. In the
naphthol portion of the molecule, H7 and H8 resonated as
doublets at δ 7.74 and 7.56, respectively, with a coupling
constant of J7,8 8.7 Hz. A singlet at δ 9.79 was assigned to
the naphthol OH proton and an OH stretch at 3356 cm−1 was
present in the infrared spectrum. A very strong band was also
observed at 2096 cm−1, which confirmed the presence of an
azide functional group.
ꢀ
ꢀ
In summary, an effective method for the key C-
glycosylation of azido glycosyl donor d-(11) with naphthol
(13) has been developed.The final C-glycoside thus obtained,
d-(22), does require further functionalization at C3 in order
for the synthetic strategy proposed in Scheme 1 to be pur-
sued, however, prior functionalization of the naphthol with a
bromine substituent at C3 afforded poorer yields in the key
C-glycosylation step. Useof anazido glycosyl donorwasalso
more successful than using a dimethylamino glycosyl donor.
Experimental
Melting points were determined using a Kofler hot-stage apparatus and
are uncorrected. Optical rotations were measured using a PolAAR 2001

792
M. A. Brimble et al.
polarimeter at room temperature in the indicated solvent. Infrared spec-
tra were recorded on a Perkin-Elmer 1600 Fourier-transform infrared
spectrophotometer as thin films or Nujol mulls between sodium chlo-
ride plates. Proton (1H) NMR spectra were recorded on a BrukerAC 200
(200 MHz) or a Bruker DRX 400 (400 MHz) spectrometer at ambient
temperature. Carbon NMR spectra were recorded on a Bruker AC 200
(50.3 MHz) or a Bruker DRX 400 (100.5 MHz) spectrometer at ambi-
ent temperature with complete proton decoupling. Low-resolution mass
spectra were recorded on a VG70-250S, a VG70-s.d. or a AEI model
MS902 double focussing magnetic sector mass spectrometer operating
with an ionization potential of 70 eV (EI, DEI, CI and DCI). High-
resolution mass spectra were recorded at a nominal resolution of 5000
or 10 000 as appropriate. Major fragments are given as percentages rela-
tivetothebasepeakandareassignedwherepossible. Ionizationmethods
employed were either electron impact or chemical ionization (CI) with
ammonia or methane as reagent gas. Flash chromatography was per-
formed using Merck Kieselgel 60 or Riedel-de-Haen Kieselgel S silica
gel (both 230–400 mesh) with the solvents indicated. Compounds were
visualized under ultraviolet light or by staining with iodine or vanillin in
methanolic sulfuric acid.Acetonitrile was distilled from calcium hydride
before use.
(18a) : (18b) : (18c) : (18d) was determined to be 9.1 : 4.8 : 2.3 : 1 by 1H
NMR analysis. νmax (film)/cm−1 2100, 1745. δH (200 MHz; CDCl3)
‡
6.16 (br d, J1,2ax 2.6, H1α-arabino), 6.05 (app br s, H1α-ribo), 5.94 (dd,
J1,2eq 2.6 and J1,2ax 8.9, H1β-ribo), 5.73 (dd, J1,2eq 2.3 and J1,2ax 10.1,
H1β-arabino), 4.64–4.72 (m, H4), 3.49–4.24 (m, H3 and H5), 1.71–2.22
(m, H2eq and H2ax), 2.06–2.14 (m, OAc × 2), 1.23 (d, J 6.3, H6β-ribo),
1.21 (d, J6,5 6.2, H6β-arabino), 1.16 (d, J6,5 6.2, H6α-arabino).
Methyl 4-O-acetyl-3-azido-2,3,6-trideoxy-D-hexopyranosides
(11a)–(11d)[10]
To a solution of 1-O-acetyl-d-azidopyranoses (18a)–(18d) (1.93 g,
7.50 mmol) in dry benzene (60 mL) under an atmosphere of nitrogen
was added montmorillonite K-10 (9.04 g) and dry methanol (1.6 mL,
39.6 mmol).The mixture was stirred vigorously and heated under reflux
overnight. The mixture was then allowed to cool to room temperature
and was filtered through a pad of celite, washing with ethyl acetate
(100 mL). The solution was concentrated under vacuum and the residue
was purified by flash column chromatography (hexanes/ethyl acetate,
92 : 8) to afford the following:
Methyl4-O-acetyl-3-azido-2,3,6-trideoxy-α-D-arabino-hexopyrano-
22
side (11a). Clear oil (705 mg, 41%), [α]D +168.9◦ (c 1.2 in CH2Cl2)
22
(lit.[10] [α]D −171◦ (c 1 in CHCl3) for l-isomer). νmax (film)/cm−1
4-O-Acetyl-3-azido-2,3,6-trideoxy-D-hexopyranoses (17a)–(17d)[10]
2099, 1748. δH (400 MHz; CDCl3) 4.73 (1 H, d, J1,2ax 3.3, H1), 4.65
(1 H, dd, J4,5 = J4,3 9.7, H4), 3.71–3.87 (2 H, m, H3 and H5), 3.32
(3 H, s, OMe), 2.14 (1 H, ddd, J2eq,1 1.0, J2eq,2ax 13.1 and J2eq,3 5.1,
H2eq), 2.11 (3 H, s, OAc), 1.71 (1 H, ddd, J2ax,1 3.4 and J2ax,2eq = J2ax,3
13.1, H2ax), 1.15 (3 H, d, J6,5 6.3, H6). δC (100 MHz; CDCl3) 170.7,
98.0, 76.2, 66.4, 58.4, 55.4, 35.8, 21.5, 18.1. The 1H NMR data were in
agreement with the data reported for the l-isomer.[10]
To 3,4-di-O-acetyl-d-rhamnal (15)[12] (5.00 g, 23.3 mmol) was added
water (35 mL) and the mixture was stirred at 80◦C for 2 h. The reaction
mixture was allowed to cool to room temperature then glacial acetic acid
(5.20 mL) and sodium azide (2.44 g, 37.3 mmol) were added in single
portions. The reaction mixture quickly turned yellow and was stirred at
room temperature overnight.The mixture was then poured into saturated
aqueous sodium bicarbonate solution (50 mL) and stirred until bubbling
ceased.The product was extracted into ethyl acetate (3 × 50 mL) and the
combined organic phases were dried over anhydrous sodium sulfate.
Filtration and concentration under vacuum gave 4-O-acetyl-3-azido-
2,3,6-trideoxy-D-hexopyranoses (17a)–(17d) (4.67 g, 93%) as a yellow
syrup that was not purified further. νmax (film)/cm−1 3360, 2104, 1743.
Methyl4-O-acetyl-3-azido-2,3,6-trideoxy-β-D-ribo-hexopyranoside
22
(11b). Clear oil (299 mg, 17%), [α]D +42.2◦ (c 1.0 in CH2Cl2)
(lit.[10] [α]D −41◦ (c 1 in CHCl3) for l-isomer). δH (400 MHz; CDCl3)
22
4.65 (1 H, dd, J4,3 3.4 and J4,5 9.1, H4), 4.62 (1 H, dd, J1,2eq 2.0 and
J1,2ax 8.9, H1), 4.10–4.19 (1 H, m, H3), 3.92–3.97 (1 H, m, H5), 3.47
(3 H, s, OMe), 2.13 (3 H, s, OAc), 2.05 (1 H, ddd, J2eq,1 2.0, J2eq,3
3.8 and J2eq,2ax 13.8, H2eq), 1.80 (1 H, ddd, J2ax,1 8.9, J2ax,3 3.4 and
J2ax,2eq 13.8, H2ax), 1.24 (3 H, d, J6,5 6.3, H6). The 1H NMR data were
in agreement with the data reported for the l-isomer.[10]
†
δH (400 MHz; CDCl3) 5.26 (br d, J1,2ax 3.1, H1α-arabino), 5.07 (br m,
H1α-ribo), 4.98 (dd, J1,2eq 2.0 and J1,2ax 9.2, H1β-ribo), 4.79 (dd, J1,2eq
1.9 and J1,2ax 9.5, H1β-arabino), 4.63 (dd, J4,3 3.3 and J4,5 9.6, H4α-ribo),
4.59 (dd, J4,3 9.7 and J4,5 9.7, H4α-arabino and H4β-arabino), 4.58 (dd,
J4,3 3.6 and J4,5 9.7, H4β-ribo), 4.27–3.31 (m, H3 and H5), 3.21 (br s,
OH), 2.24 (ddd, J2eq,1 2.0, J2eq,2ax 13.0 and J2eq,3 4.8, H2eq β-arabino),
2.12 (ddd, J2eq,1 1.2, J2eq,2ax 13.2 and J2eq,3 4.9, H2eq α-arabino), 2.10
(s, OAcα-ribo), 2.08 (s, OAcβ-arabino), 2.06 (s, OAcα-arabino), 2.06 (s,
OAcβ-ribo), 1.93–2.05 (m, H2eq α-ribo and H2eq β-ribo), 1.47–1.74 (m,
H2ax), 1.15 (d, J6,5 6.2, H6β-ribo), 1.15 (d, J6,5 6.2, H6β-arabino and
H6α-ribo), 1.08 (d, J6,5 6.4 Hz, H6α-arabino).
Methyl4-O-acetyl-3-azido-2,3,6-trideoxy-β-D-arabino-hexopyrano-
side (11c) and methyl 4-O-acetyl-3-azido-2,3,6-trideoxy-α-D-ribo-
hexopyranoside (11d). Yellow oil (8 mg, 0.5%) identified on the basis
of data extracted from the 1H NMR spectrum. For β-d-arabino isomer
(11c): δH (400 MHz; CDCl3) 4.63 (1 H, dd, J4,5J4,3 9.4, H4), 4.40 (1 H,
dd, J1,2eq 1.9 and J1,2ax 9.4, H1), 3.40–3.62 (2 H, m, H3 and H5), 3.50
(3 H, s, OMe), 2.20 (1 H, ddd, J2eq,1 1.9, J2eq,3 5.0, and J2eq,2ax 12.8,
H2eq), 2.13 (3 H, s, OMe), 1.63 (1 H, ddd, J2ax,1 9.4 and J2ax,3 = J2ax,2eq
12.8, H2ax), 1.22 (3 H, d, J6,5 6.2, H6). For α-d-ribo isomer (11d): δH
(400 MHz; CDCl3) 4.60–4.66 (2 H, m, H1 and H4), 4.05–4.17 (2 H, m,
H3 and H5), 3.35 (3 H, s, OMe), 2.13 (3 H, s, OAc), 1.94–2.07 (2 H, m,
H2), 1.17 (3 H, d, J6,5 6.5, H6).
1,4-Di-O-acetyl-3-azido-2,3,6-trideoxy-D-hexopyranoses
(18a)–(18d)[10]
To a stirred, ice-cooled solution of pyranoses (17a)–(17d) (4.33 g,
20.1 mmol) in anhydrous pyridine (40 mL) under an atmosphere of
nitrogen were added acetic anhydride (15 mL, 150 mmol) and a cata-
lytic quantity of 4-dimethylaminopyridine. The reaction mixture was
then allowed to warm to room temperature and was stirred under nitro-
gen overnight. The mixture was then cooled to 0◦C with an ice-bath,
and water (25 mL) was added with stirring. The mixture was poured
into a conical flask then saturated aqueous sodium bicarbonate solu-
tion (50 mL) and dichloromethane (50 mL) were added with vigorous
stirring. When bubbling had ceased, the phases were separated and the
organic phase was treated with further portions of saturated aqueous
sodium bicarbonate solution (2 × 50 mL) before being dried over anhy-
drous sodium sulfate and concentrated under vacuum. Residual pyridine
was removed by heating under high vacuum to afford a mixture of 1,4-di-
O-acetyl-3-azido-2,3,6-trideoxy-D-hexopyranoses (18a)–(18d) (4.94 g,
95%) as a pale-yellow syrup which was not purified further. The ratio
Methyl 4-O-acetyl-3-azido-2,3,6-trideoxy-L-hexopyranosides L-(11a)
and L-(11b)[10]
The above procedure was repeated using 1-O-acetyl-l-azidopyranoses
l-(18a)–(18d) (4.94 g, 19.2 mmol), which had been prepared from 3,4-
di-O-acetyl-l-rhamnal, to afford the following:
Methyl 4-O-acetyl-3-azido-2,3,6-trideoxy-α-L-arabino-hexopyrano
22
side L-(11a). Clear oil (1.80 g, 41%), [α]D −172.2◦ (c 2.1 in CH2Cl2)
22
[lit.[10] [α]D −171◦ (c 1 in CHCl3)].
Methyl 4-O-acetyl-3-azido-2,3,6-trideoxy-β-L-ribo-hexopyranoside
22
L-(11b). Clear oil (614 mg, 14%), [α]D −40.3◦ (c 0.8 in CH2Cl2)
22
[lit.[10] [α]D −41◦ (c 1 in CHCl3)].
Methyl 4-O-acetyl-3-azido-2,3,6-trideoxy-β-L-arabino-hexopyrano
side L-(11c) and methyl 4-O-acetyl-3-azido-2,3,6-trideoxy-α-L-ribo-
hexopyranoside L-(11d). Yellow oil (58 mg, 1.3%).
† Due to the extensive overlapping of signals in the 1H NMR spectrum, integrations of the individual resonances for the four diastereomers are not reported.
‡ Only selected data is reported due to extensive overlap of signals. Signals for the minor α-ribo isomer (18d) were partially obscured.

C-Glycosylation of Oxygenated Naphthols
793
Methyl 3-azido-2,3,6-trideoxy-α-L-arabino-hexopyranoside L-(10)[10]
for 2 h then quenched with saturated sodium hydrogen carbonate solu-
tion (3 mL), diluted with ethyl acetate (5 mL), stirred an additional
10 min and then extracted with ethyl acetate (3 × 5 mL). The com-
bined organic phases were washed with brine (10 mL), dried over anhy-
drous sodium sulphate, and concentrated under vacuum. Purification by
flash column chromatography (dichloromethane/methanol, 4 : 1) gave
6-(3ꢀ-dimethylamino-2ꢀ,3ꢀ,6ꢀ-trideoxy-β-L-arabino- hexopyranosyl)-5-
hydroxy-1,4-dimethoxynaphthalene (8) (44 mg, 52%) as a pale foam,
Methyl4-O-acetyl-3-azido-2,3,6-trideoxy-α-l-arabino-hexopyranoside
l-(11a) (260 mg, 1.14 mmol) in methanol (10 mL) was treated with
potassium carbonate (500 mg) and the mixture was stirred at room tem-
perature for 16 h. Neutralization with Amberlite IR-115 resin (H+)
followed by evaporation under reduced pressure and flash chromatogra-
phy afforded methyl 3-azido-2,3,6-trideoxy-α-arabino-hexopyranoside
22
25
L-(10) as a syrup, [α]D −122.1◦ (c 1.2 in CHCl3) [lit.[10] [α]D −125.0◦
(c 0.86 in CHCl3)]. The 1H NMR data were in agreement with the
literature.[10]
[α]D −35.2◦ (c 0.9 in CH2Cl2) (Found: [M + H]+•, 362.1978.
22
C20H28NO5 requires [M + H]+•, 362.1967). νmax (film)/cm−1 3355,
2937, 1614, 1517, 1454, 1391, 1251, 1097. δH (400 MHz; CDCl3) 9.76
(1 H, s,ArOH), 7.73(1 H, d, J7,8 8.7, H7), 7.59(1 H, d, J8,7 8.7, H8), 6.66
Methyl 3-amino-2,3,6-trideoxy-α-L-arabino-hexopyranoside
(Methyl α-L-acosaminide) L-(19)[10]
ꢀ
ꢀ
(1 H, d, J2,3 8.4, H2), 6.60 (1 H, d, J3,2 8.4, H3), 5.02 (1 H, dd, J1 ,2 ax
10.8 and J1 ,2 eq 1.8, H1ꢀ), 3.99 (3 H, s, OMe), 3.96 (3 H, s, OMe), 3.54–
ꢀ
ꢀ
To a solution of methyl pyranoside l-(11a) (260 mg, 1.14 mmol) in
methanol (10 mL) was added triethylamine (0.2 mL) and 10% palla-
dium on charcoal (262 mg). The vessel was evacuated using a water
aspirator and hydrogen was introduced from a balloon. This sequence
was repeated twice more before the reaction mixture was allowed to stir
for 2 h under an atmosphere of hydrogen.The reaction slurry was filtered
through celite, washed with methanol (50 mL), and the pale solution was
concentrated under vacuum. Purification through a small plug of silica
and trituration of the residue with diethyl ether gave methyl 3-amino-
2,3,6-trideoxy-α-L-arabino-hexopyranoside L-(19) (156 mg, 86%) as a
3.58 (1 H, m, H5ꢀ), 3.45 (1 H, br s, OH), 3.25 (1 H, dd, J4 ,5 = J4 ,3
ꢀ
ꢀ
ꢀ
ꢀ
9.6, H4ꢀ), 2.81–2.88 (1 H, m, H3ꢀ), 2.38 (6 H, s, NMe2), 2.17 (1 H, br d,
J2 eq,2 ax 12.5, H2ꢀeq), 1.49(1 H, ddd, J2 ax,1 = J2 ax,3 10.8andJ2 ax,2 eq
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
12.5, H2ax), 1.44 (3 H, d, J6 ,5 6.1, H6ꢀ). δC (100 MHz; CDCl3) 151.0,
150.8, 150.4, 128.2, 125.3, 124.6, 115.9, 113.8, 104.5, 103.6, 78.1, 73.3,
72.3, 68.4, 57.1, 56.4, 40.9, 28.9, 19.4. Mass spectrum (CI) m/z 362
(100%, [M + H]+•), 311 (17), 132 (49).
ꢀ
ꢀ
6-(4ꢀ-O-Acetyl-3ꢀ-azido-2ꢀ,3ꢀ,6ꢀ-trideoxy-β-D-arabino-hexopyranosyl)-
5-hydroxy-1,4-dimethoxynaphthalene D-(22)
22
white solid, mp 130–131◦C (lit.[10] mp 132◦C), [α]D −134.3◦ (c 2.2 in
A solution of methyl 4-O-acetyl-3-azido-2,3,6-trideoxy-α-d-arabino-
hexopyranoside d-(11) (1.36 g, 5.93 mmol) and naphthol (13)[7] (1.00 g,
4.90 mmol) was stirred in dry acetonitrile (30 mL) at 0◦C under an
atmosphere of nitrogen. Boron trifluoride diethyl etherate (1.3 mL,
10.2 mmol) was added dropwise. The mixture was stirred for 2 h then
quenched with water (10 mL) and extracted with dichloromethane
(3 × 25 mL). The combined organic phases were washed with water
(50 mL), dried over anhydrous sodium sulphate, and concentrated
under vacuum. Purification by flash column chromatography (hexanes/
ethyl acetate, 4 : 1) gave 6-(4ꢀ-O-acetyl-3ꢀ-azido-2ꢀ,3ꢀ,6ꢀ-trideoxy-
β-D-arabino-hexopyranosyl)-5-hydroxy-1,4-dimethoxynaphthalene D-
22
MeOH) [lit.[10] [α]D −135.0◦ (c 0.5 in MeOH)]. δH (200 MHz; CDCl3)
4.67 (1 H, d, J1,2ax 2.8, H1), 3.59 (1 H, dq, J5,4 9.1 and J5,6 6.0, H5),
3.31 (3 H, s, OMe), 2.91–3.07 (1 H, m, H3), 2.88 (1 H, dd, J4,3 = J4,5
9.1, H4), 2.72 (3 H, app s, OH and NH2), 1.94–2.02 (1 H, m, H2eq), 1.58
(1 H, ddd, J2ax,1 2.8 and J2ax,2eq = J2ax,3 12.5, H2ax), 1.26 (3 H, d, J6,5
6.0, H6). The 1H NMR data were in agreement with the literature.[10]
Methyl 3-dimethylamino-2,3,6-trideoxy-α-L-arabino-hexopyranoside
(Methyl α-L-angolosaminide) L-(12)[15]
From azido sugar L-(11a). To a solution of methyl pyranoside l-(11a)
(1.08 g, 4.72 mmol) in methanol (100 mL) was added formaldehyde
(36% aqueous solution, 6.50 mL, 86.7 mmol) and 10% palladium
on charcoal (0.9 g). The vessel was evacuated using a water aspira-
tor and hydrogen was introduced from a balloon. The reaction
mixture was allowed to stir overnight under an atmosphere of
hydrogen. Workup as described above followed by purification of the
residue by flash column chromatography (dichloromethane/methanol,
4 : 1) gave methyl 3-dimethylamino-2,3,6-trideoxy-α-L-arabino-
22
(22) (1.19 g, 60%) as a pale foam, [α]D +70.0◦ (c 0.6 in CH2Cl2)
(Found: C, 59.6; H, 5.7; N, 10.6%. C20H23N3O6 requires C, 59.8; H,
5.8; N, 10.5%). νmax (film)/cm−1 3356, 2936, 2096, 1744, 1614, 1518,
1390, 1251, 1055. δH (400 MHz; CDCl3) 9.79 (1 H, s, OH), 7.74 (1 H, d,
J7,8 8.7, H7), 7.56 (1 H, d, J8,7 8.7, H8), 6.69 (1 H, d, J2,3 8.4, H2), 6.63
(1 H, d, J3,2 8.4, H3), 5.06 (1 H, dd, J1 ,2 ax 11.3 and J1 ,2 eq 1.9, H1ꢀ),
ꢀ
ꢀ
ꢀ
ꢀ
4.80 (1 H, dd, J4 ,3 = J4 ,5 9.5, H4ꢀ), 4.02 (3 H, s, OMe), 3.93 (3 H,
ꢀ
ꢀ
ꢀ
ꢀ
s, OMe), 3.62–3.78 (2 H, m, H3ꢀ and H5ꢀ), 2.46 (1 H, ddd, J2 eq,1 1.9,
ꢀ
ꢀ
22
J2 eq,2 ax 13.1 and J2 eq,3 4.9, H2ꢀeq), 2.16 (3 H, s, OAc), 1.71 (1 H, ddd,
hexopyranoside L-(12) (396 mg, 44%) as a pale-yellow syrup, [α]D
ꢀ
ꢀ
ꢀ
ꢀ
22
−86.1◦ (c 1.2 in MeOH) (lit.[15] [α]D +87.0◦ (c 1.8 in MeOH) for
J2 ax,1 = J2 ax,3 11.3 and J2 ax,2 eq 13.1, H2ꢀax), 1.29 (3 H, d, J6 ,5 6.2,
H6ꢀ). δH (100 MHz; CDCl3) 170.3, 150.3, 150.1, 149.9, 127.7, 124.4,
122.5, 115.1, 113.3, 103.8, 103.0, 75.6, 74.8, 71.9, 61.7, 56.5, 55.8,
36.9, 21.0, 18.0. Mass spectrum (EI) m/z 401 (55%, M+•), 256 (15),
255 (11), 241 (23), 230 (52), 218 (34), 215 (26), 43 (100).
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
d-isomer). νmax (film)/cm−1 3461. δH (200 MHz; CDCl3) 4.80 (1 H, d,
J1,2ax 3.1, H1), 3.85 (1 H, s, OH), 3.64 (1 H, dq, J5,4 9.8 and J5,6 6.1,
H5), 3.32 (3 H, s, OMe), 3.12 (1 H, dd, J4,3 = J4,5 9.8, H4), 2.91 (1 H,
ddd, J3,2eq 3.6, J3,2ax 12.5 and J3,4 9.8, H3), 2.29 (6 H, s, NMe2), 1.85
(1 H, dd, J2eq,3 3.6 and J2eq,2ax 12.5, H2eq), 1.58 (1 H, ddd, J2ax,1 3.5
6-(4ꢀ-O-Acetyl-3ꢀ-azido-2ꢀ,3ꢀ,6ꢀ-trideoxy-β-L-arabino-hexopyranosyl)-
5-hydroxy-1,4-dimethoxynaphthalene L-(22)
1
and J2ax,2eq = J2ax,3 12.5, H2ax), 1.30 (3 H, d, J6,5 6.1, H6). The H
NMR data were in agreement with the literature.[15]
From acosaminide L-(19). To a solution of methyl pyranoside l-
(19) (199 mg, 1.23 mmol) in methanol (30 mL) was added formaldehyde
(36% aqueous solution, 1.50 mL, 20.0 mol) and 10% palladium on
charcoal (285 mg). The reaction mixture was allowed to stir overnight
under an atmosphere of hydrogen. Workup as described above fol-
lowed by purification of the residue by flash column chromatography
(dichloromethane/methanol, 4 : 1) gave methyl 3-dimethylamino-2,3,6-
trideoxy-α-L-arabino-hexopyranoside L-(12) (185 mg, 79%) as a pale
syrup, which was identical to the material prepared above.
Theaboveexperimentwasrepeatedinthel-enantiomericseriesbyreact-
ing glycosyl donor l-(11) (136 mg, 0.592 mmol) with naphthol (13)
(100 mg, 0.490 mmol) and boron trifluoride diethyl etherate (125 µL,
0.99 mmol) in acetonitrile (3 mL) to afford, after work up and chro-
22
matography, l-(22) (130 mg, 66%) as a colourless foam, [α]D −74.4◦
(c 0.48 in CH2Cl2). This material was spectroscopically identical to
d-(22) prepared above.
6-(3ꢀ-Azido-2ꢀ,3ꢀ,6ꢀ-trideoxy-β-L-arabino-hexopyranosyl)-3-bromo-
5-hydroxy-1,4-dimethoxynaphthalene (7)
6-(3ꢀ-Dimethylamino-2ꢀ,3ꢀ,6ꢀ-trideoxy-β-L-arabino-hexopyranosyl)-
5-hydroxy-1,4-dimethoxynaphthalene (8)
A mixture of 3-bromo-5-hydroxy-1,4-dimethoxynaphthalene (14)[13]
(63 mg, 0.22 mmol) and methyl 2,3,6-trideoxy-3-azido-α-l-arabino-
hexopyranoside l-(10) (45 mg, 0.24 mmol) was evacuated on a high
vacuum line for 3–4 h. Dry distilled acetonitrile (1 mL) was added by
cannula and the solution was cooled to 0◦C. Distilled boron triflu-
oride diethyl etherate (54 µL, 0.44 mmol) was added dropwise with
To a stirred solution of methyl α-l-angolosaminide l-(12) (53 mg,
0.280 mmol) and naphthol (13)[7] (48 mg, 0.24 mmol) in dry acetonitrile
at 0◦C under an atmosphere of nitrogen was added dropwise boron tri-
fluoride diethyl etherate (178 µL, 1.40 mmol). The mixture was stirred

794
M. A. Brimble et al.
a dry microsyringe. An initial red coloration was observed which
turned to brown within five minutes. After stirring at this temper-
ature for 15 min no further change was observed by TLC and the
reaction mixture was poured into water (1 mL) and extracted with
ethyl acetate (3 × 5 mL). The combined organic extracts were dried
over magnesium sulfate and the solvent was removed under reduced
pressure to give the crude product as a brown oil. Purification by
flash chromatography using hexane/ethyl acetate (4 : 1) as eluent
afforded 6-(3ꢀ-azido-2ꢀ,3ꢀ,6ꢀ-trideoxy-β-L-arabino-hexopyranosyl)-3-
bromo-5-hydroxy-1,4-dimethoxynaphthalene (7) (ca. 5 mg, 5%) as a
colourless oil (Found: M+•, 439.0559 and 437.0581. C18H20BrN3O5
requires M+•, 439.0566 and 437.0586). νmax (film)/cm−1 3251, 2092,
1392, 1248. δH (200 MHz; CDCl3) 9.73 (1 H, s, OH), 7.72 (1 H, d, J7,8
8.8, H7), 7.58 (1 H, d, J8,7 8.8, H8), 6.83 (1 H, s, H2), 5.08 (1 H, dd,
δH (400 MHz; CDCl3) 9.79 (1 H, s, OH), 7.73 (1 H, d, J7,8 8.7, H7),
7.56 (1 H, d, J8,7 8.7, H8), 6.62 (1 H, d, J2,3 8.4, H3), 6.68 (1 H, d, J2,3
8.4, H2), 5.05 (1 H, dd, J1 ,2 ax 11.2 and J1 ,2 eq 2.0, H1ꢀ), 4.02 (3 H, s,
OMe), 3.93 (3 H, s, OMe), 3.54–3.58 (2 H, m, H4ꢀ and H5ꢀ), 3.26 (1H,
m, H3ꢀ), 2.44 (1 H, m, H2eꢀ q), 2.30 (1 H, br s, 4ꢀ-OH), 1.63–1.72 (1 H, m,
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
H2ax), 1.42 (3 H, d, J6 ,5 6.0, H6 ). δC (100 MHz; CDCl3) 150.3, 150.1,
149.9, 127.6, 124.5, 122.9, 115.2, 113.2, 103.8, 103.0, 76.4, 76.0, 71.8,
64.6, 56.4, 55.8, 36.6, 18.3.
References
[1] S. Takano, K. Hasuda, A. Ito, Y. Koide, F. Ishii, I. Haneda,
S. Chihara, Y. Koyama, J. Antibiot. 1976, 29, 765.
[2] For a review see: M. A. Brimble, L. J. Duncalf, M. R. Nairn,
Nat. Prod. Rev. 1999, 16, 267.
[3] (a) N. Tanaka, T. Okabe, F. Isono, M. Kashowagi, K. Nomoto,
M. Takahashi, A. Shimazu, T. Nishimura, J. Antibiot. 1985,
38, 1327. (b) T. Okabe, K. Nomoto, H. Funabashi, S. Okuda,
H. Suzuki, N. Tanaka, J. Antibiot. 1985, 38, 1333.
[4] P.-M. Leo, C. Morin, C. Philouze, Org. Lett. 2002, 4, 2711.
[5] R. T. Williamson, L. A. McDonald, L. R. Barbieri, G. T. Carter,
Org. Lett. 2002, 4, 4659.
[6] (a) K. Tatsuta, H. Ozeki, M. Yamaguchi, M. Tanaka, T. Okui,
Tetrahedron Lett. 1990, 31, 5495. (b) K. Tatsuta, H. Ozuki,
M. Yamaguchi, M. Tanaka, T. Okui, M. Nakata, J. Antibiot.
1991, 44, 901.
[7] M. A. Brimble, T. J. Brenstrum, J. Chem. Soc., Perkin Trans. 1
2001, 1624.
[8] M. A. Brimble, R. M. Davey, M. D. McLeod, Synlett 2002, 1318.
[9] M. A. Brimble, M. R. Nairn, J. Park, J. Chem. Soc., Perkin
Trans. 1 2000, 697.
[10] J.-C. Florent, C. Monneret, J. Chem. Soc., Chem. Commun.
1987, 1171.
[11] B. Abbaci, J.-C. Florent, C. Monneret, Bull. Soc. Chim. Fr.
1989, 5.
[12] B. Fraser-Reid, D. R. Kelly, D. B. Tulshian, P. S. Ravi,
J. Carbohydr. Chem. 1983, 2, 105.
J1 ,2 ax 9.3 and J1 ,2 eq 1.9, H1ꢀ), 4.00 (3 H, s, OMe), 3.95 (3 H, s, OMe),
ꢀ
ꢀ
ꢀ
ꢀ
3.52–3.62(2 H, m, H4ꢀ andH5ꢀ), 3.25(1 H, m, H3ꢀ), 2.48(1 H, m, H2eꢀ q),
ꢀ
1.57–1.75 (1 H, m, H2ax), 1.42 (3 H, d, J6 ,5 6.1, H6ꢀ).
ꢀ
ꢀ
6-(3ꢀ-Azido-2ꢀ,3ꢀ,6ꢀ-trideoxy-β-L-arabino-hexopyranosyl)-2-bromo-
5-hydroxy-1,4-dimethoxynaphthalene (21)
Using the procedure described above for C-glycoside (22), 2-bromo-
5-hydroxy-1,4-dimethoxynaphthalene (20)[13] (43 mg, 0.15 mmol) and
methyl 3-azido-2,3,6-trideoxy-α-l-arabino-hexopyranoside l-(10)
(35 mg, 0.19 mmol) were reacted with boron trifluoride diethyl etherate
(38 µL, 0.30 mmol) in acetonitrile (1 mL) to give the 6-(3ꢀ-azido-
2ꢀ,3ꢀ,6ꢀ-trideoxy-β-L-arabino-hexopyranosyl)-2-bromo-5-hydroxy-1,4-
dimethoxynaphthalene (21) (22 mg, 34%) as a pale-yellow oil (Found:
M+•, 439.0553 and 437.0579. C18H20BrN3O5 requires M+•, 439.0566
and437.0586). νmax (film)/cm−1 3245, 2095, 1395, 1246. δH (200 MHz;
CDCl3) 9.51 (1 H, s, OH), 7.64 (1 H, d, J7,8 8.8, H7), 7.58 (1 H, d, J8,7
ꢀ
ꢀ
ꢀ
ꢀ
8.8, H8), 6.86 (1 H, s, H3), 5.02 (1 H, dd, J1 ,2 ax 11.1 and J1 ,2 eq 1.9,
H1ꢀ), 4.05 (3 H, s, OMe), 3.90 (3 H, s, OMe), 3.25–3.72 (3 H, m, H3ꢀ,
H4ꢀ and H5ꢀ), 2.41 (1 H, m, H2eꢀ q), 1.57–1.75 (1 H, m, H2aꢀ x), 1.41 (3 H,
d, J6 ,5 6.2, H6ꢀ).
ꢀ
ꢀ
6-(3ꢀ-Azido-2ꢀ,3ꢀ,6ꢀ-trideoxy-β-L-arabino-hexopyranosyl)-5-hydroxy-
1,4-dimethoxynaphthalene (6)
[13] M. A. Brimble, T. J. Brenstrum, J. Chem. Soc., Perkin Trans. 1
2001, 1612.
[14] D. S. Larsen, F. L. Andrews, L. Larsen, Aust. J. Chem. 2000,
53, 15.
[15] P. Bartner, D. Boxler, R. Brambilla, A. K. Mallams, J. B. Morton,
P. Reichert, F. D. Sancillo, H. Suprenant, G. Tomalesky,
G. Lucaas, A. Olesker, T. T. Thang, L. Valente, J. Chem.
Soc., Perkin Trans. 1 1979, 1600.
Using the procedure described above for C-glycoside (22), 5-hydroxy-
1,4-dimethoxynaphthalene (13)[7] (30 mg, 0.15 mmol) and methyl 3-
azido-2,3,6-trideoxy-β-l-arabino-hexopyranoside l-(10) (33 mg, 0.18
mmol) were reacted with boron trifluoride diethyl etherate (38 µL,
0.30 mmol) in acetonitrile (1 mL) to give 6-(3ꢀ-azido-2ꢀ,3ꢀ,6ꢀ-trideoxy-
α-L-arabino-hexopyranosyl)-5-hydroxy-1,4-dimethoxynaphthalene (6)
(22 mg, 42%) as a colourless gum (Found: M+•, 359.1476. C18H21N3O5
requires M+•, 359.1481). νmax (film)/cm−1 3363, 2097, 1390, 1249.