Syntheses and structures of anomeric quaternary ammonium β-glucosides and comments on the anomeric C-N bond lengths

Article abstract of DOI:10.1016/j.tet.2009.05.086

We report convenient syntheses of anomeric quaternary ammonium β-glucosides in both protected and deprotected forms, together with X-ray structural investigations of two of the products. Both the quaternisation of a protected anomeric N,N-(dimethylamino)g

Full text of DOI:10.1016/j.tet.2009.05.086

Tetrahedron 65 (2009) 6396–6402
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Syntheses and structures of anomeric quaternary ammonium
and comments on the anomeric C–N bond lengths
b-glucosides
Lisa Iddon a, Ryan A. Bragg b, John R. Harding b, Chandrakala Pidathala a, John Bacsa a, Anthony J. Kirby c,
,
Andrew V. Stachulski a
*
a Robert Robinson Laboratories, Department of Chemistry, University of Liverpool, Liverpool L69 7ZD, UK
b Clinical Pharmacology and DMPK, Astra Zeneca Pharmaceuticals, Mereside, Alderley Park, Macclesfield, Cheshire SK10 4TG, UK
c University Chemical Laboratory, University of Cambridge, Cambridge CB2 1EW, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
We report convenient syntheses of anomeric quaternary ammonium b-glucosides in both protected
Received 18 February 2009
Received in revised form 11 May 2009
Accepted 28 May 2009
and deprotected forms, together with X-ray structural investigations of two of the products. Both the
quaternisation of a protected anomeric N,N-(dimethylamino)glucose, derived in two high-yielding
steps from the anomeric azide, and the direct reaction of glucose or 6-O-trityl glucose with a sec-
ondary amine, followed by protection and quaternisation, are viable routes. We also discuss the
significance, in terms of stereoelectronic effects, of the observed C–N bond lengths and the stabilities
of the products.
Available online 6 June 2009
Ó 2009 Elsevier Ltd. All rights reserved.
−
HO
CO2
O
1. Introduction
O
HO
HO
HO
HO
+
+
X−
NR3
NR3
HO
HO
Anomeric quaternary ammonium b-glucosides of type 1 are of
1
2
both theoretical and practical interest: some compounds of this
type appeared in the early carbohydrate literature.1–3 More re-
cently it has been recognised that structurally similar zwitterionic
Nþ-glucuronides, of general structure 2, are important metabolites
for a wide range of drugs that contain a tertiary amino group.4,5
Thus amitriptyline 3,5 imipramine 45 and tamoxifen 56 are all
metabolised at least in part in this way and a very recent report
confirmed the isolation of a similar metabolite of 1-hydroxy-
midazolam.7 N-Glucuronide metabolites are often unique to
humans and are therefore not detected in pre-clinical studies.
N-Glucoside drug metabolites are also known, e.g., of barbitu-
rates,8 though these do not possess quaternary nitrogen. Also, in
the plant kingdom, transferases responsible for N-glucosidation
have been identified.9
In connection with a wider interest in glucuronides and
methods for their synthesis,10,11 we wished to establish general
methods for the synthesis of N-glucuronides, especially the
quaternary ammonium zwitterionic type. Syntheses of some
metabolites of this type were reported by Kaspersen and van
Boeckel.12
O
NMe2
N
NMe2
NMe2
3
4
5
The most obvious access to such metabolites would appear to be
the direct reaction of a tertiary amine with an anomeric bromo-
sugar such as 6. This works well for relatively weak bases, including
heterocycles such as pyridine13 and for anilines14 but is much less
satisfactory for aliphatic tertiary amines. Invariably elimination
predominates with such strong bases, leading to the glycal 7 as the
major product:15 we have obtained similar results in the glucose
series. While this work was being written a new report appeared,16
describing quaternisation via an anomeric mesylate, but this was
still ineffective for aliphatic tertiary amines.
CO2Me
O
CO2Me
O
AcO
AcO
AcO
AcO
AcO
Br
AcO
* Corresponding author. Tel.: þ44 151 794 3542; fax: þ44 151 794 3588.
E-mail address: stachuls@liv.ac.uk (A.V. Stachulski).
6
7
0040-4020/$ – see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2009.05.086

L. Iddon et al. / Tetrahedron 65 (2009) 6396–6402
6397
AcO
AcO
AcO
i)
O
O
O
iii)
O
AcO
AcO
AcO
AcO
AcO
AcO
OAc
N3
NH2
AcO
AcO
AcO
11
8
9
AcO
ii)
AcO
AcO
AcO
Br
10
AcO
AcO
iv)
v)
O
O
AcO
AcO
AcO
NMe2R+ X–
NMe2
HO
AcO
AcO
AcO
12
13a R = Me, X = I
13b R = PhCH2, X = Br
13c R = CH2CH=CH2, X = Br
vi)
O
HO
NMe2R+ X–
HO
HO
14a R = Me, X = I
14b R = PhCH2, X = Br
14c R = CH2CH=CH2, X = Br
Scheme 1. Synthesis of 1
b-quaternary ammonium glucosides: azide route. Reagents: (i) Me3SiN3, SnCl4, CH2Cl2, 95%; (ii) Me3SiN3, TBAF, THF, ca. 80%; (iii) Pd/H2, EtOAc, 75%; (iv)
HCHO aq, i-PrOH/THF, Pd/H2, 92%; (v) MeI, PhCH2Br or CH2]CH$CH2Br, MeCN, 50 ꢀC, 3–4 h, 51–82%; (vi) aq Na2CO3, MeOH, AR-120 (Hþ), 72–92%.
We note that others17 have reported extremely low yields by
this method or a related two-phase procedure.4 Interestingly, the
direct reaction of 5 with bromosugar 6 did give a very modest yield
of tamoxifen Nþ-glucuronide after deprotection.6 This more
favourable outcome is probably due to the relatively weaker base
hemiacetal initially (cf. Scheme 3 below): we have not investigated
the use of higher acyl derivatives in this reaction. A single-crystal
X-ray structure determinationy on compound 13a (Fig. 1, crystals
from MeOH) gave a similar result to the corresponding22,23
bromide.
strength of the b-alkoxyamine 5. We decided it was important to
understand first the synthesis and chemistry of anomeric quater-
nary ammonium glucosides, then later apply this knowledge to Nþ-
glucuronide synthesis. In the following we describe three methods
for the synthesis of different classes of anomeric quaternary Nþ-
glucosides and report on the structures and stabilities of the
products.
2. Results and discussion
2.1. Syntheses and structures
One of the best ways of introducing anomeric nitrogen into
monosaccharides is via an azide. Reaction (Scheme 1) of glucose
18
b
-pentaacetate 8 with Me3SiN3 and SnCl4 (the most effective
Lewis acid, we have found) affords the b-azide 9 only, in excellent
yield. Alternatively, reaction of acetobromoglucose 10 with
Me3SiN3 and TBAF19 also affords 9, but in lower yield.
Figure 1. X-ray structure of anomeric quaternary ammonium glucoside 13a. The
anomeric C–Nþ length is 1.523 Å (cf. 1.534 Å for the bromide).
Reduction of 9 by catalytic hydrogenation in EtOAc affords the
primary amine 1120 as a single
b-anomer in excellent yield. Re-
ductive N,N-dimethylation of anomeric pyranose amines has not
previously been reported, though it has been used for other ring
positions.21 After much experimentation, we found that Pd-cata-
lysed hydrogenation of a solution of 11 in THF/i-PrOH and aq
formaldehyde afforded the desired tertiary amine 12 in very good
yield; this material slowly crystallised and was quite stable at 20 ꢀC
in neutral conditions. The use of Eschweiler–Clarke conditions or
HCHO/NaCNBH3/AcOH was unsuccessful, almost certainly owing to
the acid instability of the product.
It was crucial to establish reliable conditions for deprotection of
13a without loss of the þNR3 group. Reports in earlier literature
2,3,5
suggested that such compounds are unstable in strong base,
leading to the generation of anhydro-sugars, but we obtained clean
deacetylation of 13a using aq Na2CO3/MeOH, giving the free sugar
14 in good yield. This product crystallised from MeOH, allowing
Quaternisation of 12 worked well for highly electrophilic ha-
lides, and derivatives 13a–c were obtained using MeI, PhCH2Br and
CH2]CHCH2Br, respectively: the reaction was accompanied by
y
Crystal data for 13a: C17H28INO9, M¼517.30, colourless prism, 0.44ꢁ0.42ꢁ0.
29 mm3, orthorhombic, space group P212121 (No. 19), a¼8.6160(17), b¼11.244(2),
c¼24.390(5) Å, V¼2362.9(8) Å3, Z¼4, Dc¼1.454 g/cm3, F000¼1048, STOE IPDS, Mo K
a
radiation,
l
¼0.71073 Å, T¼173(2) K, 2qmax¼48.4ꢀ, 18,276 reflections collected, 3557
a characteristic downfield shift of H(1), e.g., from
d 4.05 (1H, d,
unique (Rint¼0.0456). Final GooF¼0.854, R1¼0.0274, wR2¼0.0519, R indices based
J¼9.3 Hz) to 5.30 (1H, d, J¼8.4 Hz) for the transformation 12/13a.
However, the use of less reactive halides, including primary ha-
lides, led to decomposition, probably via formation of the
on 2812 reflections with I>2
and absorption corrections applied,
s
(I) (refinement on F2), 260 parameters, 0 restraints. Lp
m
¼1.397 mmꢂ1. Absolute structure parame-
ter¼ꢂ0.030(17) (Flack, H. D. Acta Cryst. 1983, A39, 876–881).

6398
L. Iddon et al. / Tetrahedron 65 (2009) 6396–6402
a further single-crystal X-ray structure determinationz: two aspects
are shown in Figure 2.24
This is the first X-ray structure determination of such a carbo-
hydrate in unprotected form, and the pattern of bond lengths at the
anomeric centre is of special interest. The anomeric C–Nþ bond
lengths for 13a (1.523 Å; cf. 1.534 Å for the corresponding bromide)
and 14, at 1.522 Å are the same within experimental error
(ꢃ0.005 Å), and fall into the region expected for a trimethy-
lammonium group attached to secondary carbon. Thus the length
of the Nþ–CH bond in 29 structures in the Cambridge Crystallo-
graphic database with the substructure Me3Nþ–CHR2 (R¼alkyl) is
1.537ꢃ0.008 Å. A discussion of the implications of the observed
bond lengths is given at the end of this section.
We have developed an alternative synthesis of anomeric qua-
ternary ammonium glucosides (Scheme 2), giving access to struc-
tures other than the N,N-dimethyl type. It is well known that free
sugars will react directly with amines to form glycosylamines,25,26
and these can be valuable intermediates for the quaternary com-
pounds provided the Amadori rearrangement can be avoided. The
reaction is particularly efficient for cyclic secondary amines, e.g.,
piperidine with glucose26 affords the glycosylamine 15 in high yield
as mainly the
b product without acid catalysis. For such cyclic
amines, the Amadori rearrangement27 is relatively slow: the de-
rivative remains in the pyranose form long enough to allow per-
acetylation, giving solely the known b
-product 1626 (NMR) in good
yield. We obtained similar reactions using morpholine, pyrrolidine
and N-methylpiperazine.28
Quaternisation of the tertiary amine 16 to give 1729 required
MeOTf (MeI gave only slow decomposition: nor was direct qua-
ternisation of 15 possible) but a good yield of product was obtained.
Here again mild deprotection, using catalytic NaOMe/MeOH,
afforded the free sugar 18 in excellent yield.
Figure 2. Two views of the structure of anomeric quaternary ammonium glucoside 14:
the intermolecular hydrogen bonding involving the iodide ions (plan), and the con-
formation of the ring system (elevation).
HO
HO
AcO
O
i)
O
ii)
O
HO
HO
HO
HO
AcO
AcO
N
N
OH
HO
HO
AcO
15
iv)
16
AcO
HO
O
iii)
O
AcO
AcO
HO
HO
+
+
N
N
TfO−
TfO−
AcO
HO
17
18
Scheme 2. Direct synthesis of anomeric glycosylamines and their quaternary derivatives from glucose. Reagents: (i) piperidine, 80 ꢀC; (ii) Ac2O, pyridine, CH2Cl2 [74% for steps (i)
and (ii)]; (iii) MeOTf, CH2Cl2, 62%; (iv) NaOMe, MeOH, 20 ꢀC, 100%.
RO
RO
H
O
RO
H+
slow
O
+
OH
N+
RO
RO
RO
RO
RO
RO
N
N
RO
RO
RO
fast
H2O fast
RO
O
RO
RO
RO
N
O
RO
RO
+ HNMe2
RO
OH
RO
anomerisation
hydrolysis
Scheme 3. Hydrolysis of glycosylamines, illustrated for an N,N-dimethyl example.
In the above sequence, and elsewhere, the acid instability of
tertiary amines such as 12 and 16 frequently makes their handling
and chromatographic purification difficult. This is understood to
be a consequence of reversible protonation on the pyranose oxygen
followed by ring opening and rapid hydrolysis or anomerisation of
the resulting iminium species, Scheme 3.30 In contrast, quaternary
ammonium species show greater acid stability, as do the HCl salts of
the tertiary amines, since no N lone pair is available to participate:
thus the HCl salt of 15 is stable in aqueous solution for 18 h at 20 ꢀC.
z
Crystal data for 14: C9H20INO5, M¼349.16, colourless prism, 0.32ꢁ0.19ꢁ0.
15 mm3, orthorhombic, space group P212121 (No. 19), a¼9.3032(19), b¼9.985(2),
c¼14.663(3) Å, V¼1362.0(5) Å3, Z¼4, Dc¼1.703 g/cm3, F000¼696, Bruker D8 dif-
fractometer with APEX detector, Mo Ka radiation, l¼0.71073 Å, T¼100(2) K,
2qmax¼55.0ꢀ, 8145 reflections collected, 3038 unique (Rint¼0.0177). Final GooF¼1.
026, R1¼0.0182, wR2¼0.0477, R indices based on 3024 reflections with I>2
s(I)
(refinement on F2), 189 parameters, 8 restraints. Lp and absorption corrections
applied, m¼2.358 mmꢂ1. Absolute structure parameter¼0.02(2) (Flack, H. D. Acta
Cryst.. 1983, A39, 876–881).

L. Iddon et al. / Tetrahedron 65 (2009) 6396–6402
6399
Ph3CO
Ph3CO
O
i)
O
HO
HO
HO
+
N
HO
OH
NHMe
HO
HO
21
20
19
Ph3CO
HO
ii)
iii)
O
O
AcO
AcO
AcO
N
N
+
AcO
–OTf
AcO
AcO
22
23
Scheme 4. Quaternary ammonium glucosides via 6-O-trityl glucose. Reagents: (i) 19þ20, CH2Cl2, 40 ꢀC, 49%; (ii) Ac2O, pyridine, CH2Cl2, 79%; (iii) MeOTf, CH2Cl2, 59%.
AcO
O2CR
A more general procedure for less reactive (acyclic) secondary
amines, which can also avoid difficulties arising from Amadori
rearrangement, uses 6-O-trityl glucose 19 (Scheme 4).31 In this way
both the sugar and the amine (the example shown employs nor-
triptyline 20,32 the secondary amine related to amitriptyline 2) may
be kept in solution using an organic solvent (CH2Cl2 is suitable).
Even at a 1:1 ratio of 19/20, the equilibrium is strongly (>9:1) in
favour of glycosylamine 21 by NMR spectroscopy, though again
there are losses on chromatography (cf. Scheme 3). It is important
to purify 21 rapidly to forestall rearrangement, but after tri-
acetylation a stable pyranose product 22 is obtained. Similarly to
Scheme 2, quaternisation of 21 is not possible, but 22 may be
quaternised using MeOTf, with concomitant loss of the trityl group,
to give 23 in good yield.
O2CR
Cl
O
O
O
RCO2
RCO2
AcO
AcO
n
Cl
Cl
O
x
n
x
AcO
Cl
25 R = Me or Ph
24
26
Similar effects are observed at the anomeric centres of glyco-
pyranosides, where the electrophilicity of Cl can elicit a substantial
anomeric effect even from the ring oxygen of tetraacyl derivatives.
Thus the bond lengths x and n of the tetraacyl
a-D-mannosyl
chlorides 25 are 1.859 and 1.364 Å (R¼Me), and 1.812 and 1.398 Å
(R¼Ph), respectively.34,35 By comparison, those of the equatorial
tetraacetyl
b
-D
-glucosyl chloride 26, with nO–
s C–O overlap ‘turned
*
off’, are 1.767 and 1.429 Å.36,37 These values are not very different
from those for the equatorial centre in 24.
2.2. Observed bond lengths: stereoelectronic implications
In summary, the limited amount of data available suggest that
the exocyclic C–Nþ bond lengths at the anomeric centres of sys-
tems, like 13a and 14, with equatorial trialkylammonium groups,
are indeed subject to a combination of effects, which largely cancel
out. Systems with the trialkylammonium group axial are expected
to show a more substantial anomeric effect, and will be of con-
siderable interest if the substantial synthetic difficulties can be
overcome.
The pattern of bond lengths at the anomeric centre of a gly-
coside is subject to a combination of effects, which however are
minimised in systems like 13a and 14 with equatorial trialkyl-
ammonium groups. Electron withdrawal by the four OH or OAc
groups of a hexapyranoside is sufficient to make the ring oxygen
more electronegative than the oxygen of an alkoxyl group, so that
the endocyclic C–O bond n at the glycosidic centre (cf. structures
24 and 25 below) is typically longer than the exocyclic bond x for
an O-glycoside; though both are shortened, compared to alkyl
3. Conclusions
ether/alcohol C–O bonds, by mutual nO–
the Me3Nþ group can only act as a
-acceptor, because there is no
lone pair on N for nN– C–O donation. On the other hand, nO–
s
*
overlap.33 However,
C–O
In summary, we have demonstrated general methods for the
syntheses of tertiary glucosylamines and their quaternisation. Both
the quaternisation of a protected anomeric N,N-dimethylamino-
glucoside and the direct reaction of glucose or 6-O-trityl glucose
with a secondary amine, followed by protection and quaternisation,
are viable routes. While the tertiary glycosylamines are rather un-
stable, particularly in acid conditions, after quaternisation the
products may be deprotected by mild base-catalysed hydrolysis to
deliver the free glucosides, in good yields. These quaternary de-
rivatives are in general more stable over the intermediate (3–10) pH
range. Both protected and deprotected quaternary ammonium
derivatives exhibit well-defined crystal structures with essentially
undistorted chair conformations and C–Nþ bond lengths within the
previously observed range.
s
s
*
s
*
C–N
donation would be possible if the exocyclic C–N were axial but is
not expected for equatorial C–N. In this respect the Me3Nþ group
is best compared with a halogen, except that the halogen atom of
a glycosyl halide is generally axial, while in the handful of known
examples the Me3Nþ group in systems like 13a and 14 is always
equatorial. The electronic anomeric effect of a Me3Nþ group
would be expected to be at least as large as that of Cl; but the
combined steric effects of equatorial Me3Nþ and the 6-CH2OR
group of a hexapyranoside are decisive, as are the combined
anomeric effect of an axial halogen and the steric effect of equa-
torial 6-CH2OR.
The consequent effects on bond lengths are illustrated by results
with simpler systems. The crystal structure of the cis-dichloro-
dioxan 24 shows a classical chair conformation, with one Cl atom
necessarily equatorial, and thus antiperiplanar to ring bonds rather
than lone pair electron density. As a result the C–Cl bond length
(x¼1.781 Å) and that of the corresponding endocyclic C–O bond
(n¼1.425 Å) are normal for alkyl halides or ethers. The corre-
sponding bond lengths at the centre with axial Cl, by contrast, at
1.820 and 1.394 Å, show the marked lengthening and shortening,
respectively, characteristic of the anomeric effect.33
4. Experimental
4.1. General experimental methods
All organic solvents were anhydrous and of AR grade. Vacuum
rotary evaporation was carried out at <30 ꢀC. Analytical thin-layer
chromatography was performed using Merck Kieselgel 60 F254 sil-
ica plates; preparative column chromatography was performed on

6400
L. Iddon et al. / Tetrahedron 65 (2009) 6396–6402
Merck 938S silica gel. Infra-red spectra were obtained using an FT/
IR-4100 type A instrument. Both 1H and 13C NMR spectra were
recorded for the solvents noted using either Bruker 250 MHz or
400 MHz instruments (the latter operating at 100 MHz for 13C ob-
servation) with tetramethylsilane as internal standard. Mass spec-
tra in the chemical ionisation (CI) mode were obtained using
a VG7070E mass spectrometer. Both low and high resolution elec-
trospray mode (ES) mass spectra were obtained using a Micromass
LCT mass spectrometer operating in the þve or ꢂve ion mode as
indicated. Elemental microanalysis was performed by Mr. Steve
Apter (Liverpool). 3-[2,2,3,3-2H4] Trimethylsilyl propionate sodium
salt (TSP), sodium dihydrogen phosphate and disodium hydrogen
phosphate, were purchased from Sigma–Aldrich Company, Ltd
(Gillingham, Dorset, UK). HPLC–NMR grade deuterium oxide
(2H2O) was obtained from Goss Scientific Instruments (Essex, UK).
(1H, t, J¼9.3 Hz, 4-H), 5.49 (1H, t, J¼9.2 Hz, 3-H), 5.78 (1H, t,
J¼8.4 Hz, 2-H) and 7.65–7.75 (5H, m, ArH); 13C NMR:
d (D2O) 20.5,
20.6, 20.8, 21.1, 47.6, 49.1, 62.3, 67.5, 68.0, 68.8, 74.2, 74.9, 91.1, 126.7,
130.1, 132.0, 133.4, 172.2, 173.0, 173.2 and 174.2; m/z (ES þve mode)
466 (cation Mþ, 100%).
4.3.3. 2,3,4,6-Tetra-O-acetyl-b-D-glucopyranosyl(dimethyl-2-
propenyl)ammonium bromide (13c)
Yield of 13c, 82%; amorphous glass. Found: C, 44.6; H, 5.9; N,
2.25; m/z, 416.1900. C19H30BrNO9$H2O requires C, 44.4; H, 6.2; N,
2.7%; C19H30NO9 requires m/z, 416.1921; 1H NMR:
d [(CD3)2CO] 1.98,
2.03, 2.06 and 2.17 (12H, 4s, 4ꢁCH3CO), 3.42 (two signals, 6H, 2s,
2ꢁCH3Nþ), 4.26 (1H, dd, J¼12.5 and 5.6 Hz, H-6a), 4.38 (1H, dd,
J¼12.5 and 2.4 Hz, H-6b), 4.47–4.65 (2H, 2 m, NþCH2CH]CH2), 4.70
(1H, approx. dd, J¼10.1 and 3.2 Hz, 5-H), 5.22 (1H, dd, J¼10.1 and
9.3 Hz, 4-H), 5.52 (1H, t, J¼9.1 Hz, H-3), 5.67 (1H, t, J¼8.9 Hz, H-2),
5.72 (1H, dd, J¼10.1 and 1.6 Hz, CH]CH2 cis-olefinic H), 5.89 (1H,
dd, J¼16.8 and 1.4 Hz, CH]CH2 trans-olefinic H), 6.26 (1H, d,
J¼9 Hz, 1-H) and 6.30–6.35 (1H, m, CH2CH]CH2); 13C NMR:
4.2. 2,3,4,6-Tetra-O-acetyl-b-D-glucopyranosyl-N,N-
dimethylamine (12)
To a solution of 11 (0.397 g, 1.15 mmol) in THF (15 mL) was added
Pd/C (20 mol %) and isopropanol (15 mL), followed by the addition of
37% aqueous formaldehyde solution (11.5 mmol). The flask was then
placedundervacuumbeforepurgingwithhydrogen.Thereactionwas
stirred at room temperature under a positive pressure of hydrogen for
5 h. The solutionwasfilteredthrougha pad of Celite, and washed with
ethyl acetate. The solvent was removed in vacuo and the residue was
chromatographed using a gradient from 30% to 70% EtOAc/hexane, to
afford the product 12 (0.395 g, 92%) as a hygroscopic solid. Found: C,
46.5; H, 6.6; N, 3.4; m/z, 398.1424. C16H25NO9$2H2O requires C, 46.7;
H, 7.05; N, 3.4%; C16H25NO9$Na requires m/z, 398.1427; 1H NMR:
d [(CD3)2CO] 20.9, 21.0, 21.2, 21.5, 48.6, 49.4, 62.6, 67.0, 68.6, 69.1,
74.5, 75.9, 91.9, 127.1, 129.7, 170.5 (ꢁ2), 170.7 and 171.2; m/z (ES þve
mode) 416 (cation Mþ, 100%).
4.4. General procedure for the hydrolysis of peracetates
(13a)–(13c)
To a solution of perester 13a, 13b or 13c (0.058 mmol) in MeOH
(1 mL) was added solid anhydrous Na2CO3 (0.17 mmol) at ambient
temperature. The reaction was left stirring at room temperature for
3 h, followed by quenching with Amberlite Hþ resin (0.17 mmol)
until pH 7 was reached. The resin was then filtered and washed
with methanol, followed by evaporation in vacuo: the residue was
recrystallised from ethanol to afford the product 14a, 14b or 14c.
d
(CDCl3) 2.00, 2.02, 2.04, 2.08 (12H, 4s, 4ꢁCH3CO), 2.40 (6H, s,
2ꢁCH3N), 3.59 (1H, ddd, J¼9.9, 4.9 and 2.7 Hz, 5-H), 4.05 (1H, d,
J¼9.3 Hz, 1-H), 4.12 (1H, dd, J¼12.1 and 2.5 Hz, H-6a), 4.23 (1H, dd,
J¼12.1 and 4.9 Hz, H-6b), 5.00 (1H, t, J¼9.9 Hz, H-4), 5.12 (1H, t,
J¼9.3 Hz, H-2), 5.20 (1H, t, J¼9.3 Hz, H-3); 13C NMR:
d
(CDCl3) 21.0,
4.4.1. b-D-Glucopyranosyl(trimethylammonium)iodide (14a)
21.1, 21.2, 23.0, 39.9 (ꢁ2), 62.8, 68.7, 69.4, 73.3, 74.5, 93.8,169.9,170.0,
White solid, yield, 87%. Found: C, 30.88, H, 5.78, N, 4.00; m/z,
170.6 and 171.0; m/z (ES þve mode) 398 (MNaþ, 100%).
222.1349. C9H20INO5 requires C, 30.94, H, 5.78, N, 4.01%; C9H20NO5
requires m/z, 222.1341; 1H NMR: [(CD3)2SO] 3.11 [9H, s, (CH3)3Nþ],
d
4.3. General procedure for the quaternisation of (12)
3.10–3.40 (3H, m, 2-Hþ3-Hþ4-H), 3.45–3.50 (2H, m, 2ꢁ6-H), 3.70–
3.75 (1H, m, 5-H) and 4.54 (1H, d, J¼8.9 Hz, 1-H); 13C NMR:
To a solution of 12 (0.13 mmol) in acetonitrile (1 mL) under ar-
gon was added MeI, PhCH2Br or CH2]CHCH2Br (1.33 mmol), and
then the reaction was heated to 50 ꢀC for 5 h. The acetonitrile was
removed in vacuo and the solid residue was recrystallised from
ethanol to give the product 13a, b or c.
d
[(CD3)2SO] 50.7, 60.8, 69.2, 70.5, 77.3, 80.6 and 95.4; m/z (ES þve
mode) 222 (cation Mþ, 100%).
4.4.2. -Glucopyranosyl(benzyldimethylammonium)-
b-D
bromide (14b)
Yield, 72%; amorphous glass. Found: m/z, 298.1660. C15H24NO5
4.3.1. 2,3,4,6-Tetra-O-acetyl-
ammonium)iodide (13a)
Yield of 13a, 0.034 g (51%); off-white solid, mp 173–174 ꢀC.
Found: C, 39.2; H, 5.5; N, 2.6; m/z, 390.1765. C17H28INO9 requires C,
39.5; H, 5.4; N, 2.7%; C17H28NO9 requires m/z, 390.1764; nmax (di-
amond) cmꢂ1 2974, 2920, 1740, 1612, 1446, 1376 and 1211; 1H NMR:
b
-
D
-glucopyranosyl(trimethyl-
requires m/z, 298.1654; 1H NMR:
d [(CD3)2SO] 3.10, 3.12 (6H, 2s,
2ꢁCH3Nþ), 3.16 (1H, t, J¼9.4 Hz, 4-H), 3.28 (1H, t, J¼8.7 Hz, 3-H),
3.39 (1H, m, 5-H), 3.56 (1H, dd, J¼12.3 and 6.7 Hz, 6-Ha; other 6-H
is below CH3Nþby COSY), 3.65 (1H, t, J¼8.7 Hz, 2-H), 4.24 (1H, d,
J¼8.9 Hz, 1-H), 4.56, 4.88 (2H, ABq, ArCH2Nþ), 7.49–7.59 (3H, m,
ArH) and 7.63–7.68 (2H, m, ArH); 13C NMR:
d [(CD3)2SO] 46.9, 48.2,
d
(D2O) 2.20, 2.22, 2.25 and 2.28 (12H, 4s, 4ꢁCH3CO), 3.32 [9H, s,
61.1, 66.1, 69.2, 70.1, 77.2, 80.5, 92.9, 127.9, 129.4, 130.8 and 133.5; m/z
(CH3)3Nþ], 4.38–4.42 (1H, m, 5-H), 4.45–4.65 (2H, m, 2ꢁ6-H), 5.30
(ES þve mode) 298 (cation Mþ, 100%).
(1H, d, J¼8.4 Hz, 1-H), 5.36 (1H, t, J¼9.2 Hz, 4-H), 5.58 (1H, t,
J¼8.8 Hz, 3-H) and 5.75 (1H, t, J¼8.4 Hz, 2-H); 13C NMR:
d
(D2O) 20.5,
4.4.3. b-D-Glucopyranosyl(dimethyl-2-propenyl)-
ammonium bromide (14c)
Yield, 98%; amorphous glass. Found: C, 38.1; H, 6.9; N, 3.4; m/z,
248.1492. C11H22BrNO5.H2O requires C, 38.15; H, 6.9; N, 4.0%;
C11H22NO5 requires m/z, 248.1498; nmax (diamond) cmꢂ1 3700–
2700 (br, vs), 2980, 2927, 1612, 1442, 1095 and 1060; 1H NMR:
20.6, 20.7, 20.8, 51.7, 62.1, 67.6, 68.4, 74.2, 75.2, 93.8, 100.0, 172.2,
173.0, 173.3 and 174.2; m/z (ES þve mode) 390 (cation Mþ, 100%).
4.3.2. 2,3,4,6-Tetra-O-acetyl-b-D-glucopyranosyl(benzyldimethyl-
ammonium)bromide (13b)
Yield of 13b, 0.068 g (64%): solid, mp 160–162 ꢀC. Found: m/z,
d
[(CD3)2SO] 2.97, 3.02 (6H, 2s, 2ꢁCH3Nþ), 3.16 (1H, t, J¼9.4 Hz, 4-
466.2072. C23H32NO9 requires m/z, 466.2077; 1H NMR:
d
(D2O) 2.18,
H), 3.32 (1H, t, J¼8.7 Hz, 3-H), 3.30–3.40 (1H, m, 5-H, partly ob-
scured), 3.49 (1H, dd, J¼12.4 and 6.1 Hz, 6a-H), 3.56 (1H, t, J¼8.8 Hz,
2-H), 3.72 (1H, dd, J¼12.4 and 1.9 Hz, 6b-H), 3.90 (1H, dd, J¼12.9
and 7.2 Hz, one of CHCH2Nþ), 4.16 (1H, dd, J¼12.8 and 7.7 Hz, one of
2.22, 2.29 (12H, 3s, 4ꢁCH3CO; two signals coincident), 3.21, 3.33
(6H, 2s, 2ꢁCH3Nþ), 4.41–4.45 (1H, m, 5-H), 4.55–4.70 (2H, m, 2ꢁ6-
H), 4.85–5.00 (2H, ABq, PhCH2Nþ), 5.11 (1H, d, J¼8.7 Hz, 1-H), 5.37

L. Iddon et al. / Tetrahedron 65 (2009) 6396–6402
6401
CHCH2Nþ), 4.39 (1H, d, J¼8.9 Hz, 1-H), 5.62 (1H, dd, J¼10.0 and
1.5 Hz, CH]CH2 cis-olefinic H), 5.66 (1H, dd, J¼17.2 and 1.5 Hz,
The reactants were stirred at rt for 24 h: the DCM was then re-
moved in vacuo. Column chromatography was carried out using
EtOAc to afford the product 21 (49%) as a solid, mp 91–92 ꢀC. Found:
m/z, 668.3407. C44H46NO5 (MHþ) requires m/z, 668.3376; 1H NMR:
CH]CH2 trans-olefinic H), and 5.90–6.00 (1H, m, CH2CH]CH2); 13
C
NMR:
d [(CD3)2SO] 47.1, 48.3, 60.5, 65.9, 68.9, 70.1, 76.9, 80.1, 93.3,
125.4 and 129.0; m/z (ES þve mode) 248 (cation Mþ, 100%).
d (CD3CN) 2.40–2.47 (2H, m, ]CHCH2), 2.40 (3H, s, NCH3), 2.90–
3.00 (2H, m, NCH2CH2), 2.95–3.15 (4H, br m, ArCH2CH2Ar), 3.18 (1H,
dd, J¼10.0 and 5.5 Hz, 6a-H), 3.25–3.32 (3H, m, 4-Hþ5-Hþ6b-H),
3.33 (1H, t, J¼9 Hz, 3-H), 3.40 (1H, t, J¼8.5 Hz, 2-H), 3.91 (1H, d,
J¼8.8 Hz, 1-H), 5.92 (1H, t, J¼7.6 Hz, ]CHCH2), 7.05–7.33 (17H, m,
4.5. N-(2,3,4,6-Tetra-O-acetyl-b-D-glucopyranosyl)-
piperidine (16)
This was prepared essentially according to the reported pro-
cedure, Ref. 26 in the text as a white solid. Found: C, 55.1; H, 7.1; N,
3.2; m/z, 438.1718. C19H29NO9 requires C, 54.9; H, 7.0; N, 3.4%;
ArH) and 7.49–7.53 (6H, m, ArH); 13C NMR:
d (CD3CN) 27.6, 31.2,
33.0, 34.3, 53.3, 63.4, 69.5, 70.1, 76.1, 77.4, 85.7, 93.6, 125.4, 125.5,
126.7, 126.8, 127.1,127.4, 127.6, 127.7,128.0, 128.3, 129.0, 129.7, 136.6,
139.1, 139.8, 140.8, 143.1 and144.0; m/z (ES þve mode) 668 (MHþ,
100%).
C19H29NO9Na requires m/z, 438.1740; 1H NMR:
d (CDCl3) 1.40–1.55
(6H, m, 3ꢁCCH2C), 2.02, 2.04, 2.07, 2.16 (12H, 4s, 4ꢁCH3CO), 2.49–
2.56 and 2.88–2.96 (4H, 2m, 2ꢁNCH2), 3.59 (1H, ddd, J¼10.0, 4.8
and 2.6 Hz, 5-H), 3.98 (1H, d, J¼8.9 Hz,1-H), 4.11 (1H, dd, J¼12.1 and
2.6 Hz, 6a-H), 4.23 (1H, dd, J¼12.1 and 4.8 Hz, 6b-H), 4.99 (1H, t,
J¼9.4 Hz, 4-H), 5.17 (1H, t, J¼9.3 Hz, 2-H) and 5.20 (1H, t, J¼8.5 Hz,
4.9. [N-[(6-O-Trityl)-2,3,4-tri-O-acetyl]-b-D-glucopyranosyl]-
N-methyl-3-(10,11-dihydro-5H-dibenzo[a,d]cyclohepten-5-
ylidene)-1-propanamine (22)
3-H); 13C NMR:
d
(CDCl3) 21.0, 21.1 (ꢁ2), 21.2, 24.9, 26.7, 49.4, 62.8,
67.9, 69.4, 73.4, 74.3, 94.9, 169.9, 170.0, 170.7 and 171.0; m/z (ES þve
mode) 438 (MNaþ, 100%).
To a solution of 21 (1.5 mmol) in pyridine (15 mL) at 0 ꢀC under
nitrogen was added acetic anhydride (9 mmol). The reaction was
allowed to reach room temperature on addition of the acetic anhy-
dride and left stirring for 4 h. The reaction was basified using satd
NaHCO3 until pH 7 was reached, then the solution was diluted with
DCM (60 mL) and the organic layer was separated. The aqueous layer
was then extracted with DCM (3ꢁ50 mL) and the combined DCM
layers were washed with water (100 mL), brine (100 mL) and dried
over Na2SO4. After evaporation, column chromatography was carried
out using 50:50 ether/hexane to afford the product 22 (79%) as an
amorphous foam. Found: m/z, 794.3710. C50H52NO8 (MHþ) requires
4.6. [N-(2,3,4,6-Tetra-O-acetyl-b-D-glucopyranosyl)-N-
methyl]piperidinium trifluoromethanesulfonate (17)
To a solution of 16 (0.100 g, 0.24 mmol) in DCM (2 mL) under
nitrogen was added MeOTf (0.48 mmol). The reaction was left
stirring for 4 h at room temperature, and then the DCM was re-
moved in vacuo. Column chromatography was carried out using
100% EtOAc, then 50:50 ethanol/DCM. The ethanol/DCM fractions
were combined and solvent removed to give 17 in 62% yield as
a foam. Found: m/z, 430.2064. C20H32NO9 requires m/z, 430.2077;
m/z, 794.3693; 1H NMR:
d (CD3CN) 1.73, 1.77 and 1.95 (9H, 3s,
1H NMR:
d
(CDCl3) 1.70–1.80 (3H, m, CCH2C), 1.98, 2.03, 2.05 and
3ꢁCH3CO), 2.39 (5H, m, ]CHCH2þNCH3), 2.75–3.20 (6H, m,
ArCH2CH2ArþNCH2CH2), 3.25–3.60 (3H, m, 2ꢁ6-Hþ5-H), 4.15–4.35
(1H, br d, 1-H), 5.10–5.25 (3H, br m, 2-Hþ3-Hþ4-H), 5.87 (1H, t,
]CHCH2), 7.04–7.06 (1H, m, ArH), 7.09–7.17 (3H, m, ArH), 7.22–7.34
(13H, m, ArH) and 7.42–7.51 (6H, m, ArH); the 1H NMR spectrum
showed very broad signals and meaningful coupling constants can-
2.12 (12H, 4s, 4ꢁCH3CO), 2.00–2.10 (3H, m, CCH2C), 3.30 (3H, s,
CH3Nþ), 3.60–3.95 (4H, m, 2ꢁCH2Nþ), 4.27 (1H, dd, J¼12.6 and
5.9 Hz, 6a-H), 4.36–4.44 (2H, m, 6b-Hþ5-H), 5.23 (1H, t, J¼9.7 Hz,
4-H), 5.49 (1H, dd, J¼9.4 and 8.3 Hz, 3-H), 5.69–5.78 (2H, m,1-Hþ2-
H); 13C NMR:
d
(CDCl3) 20.7, 20.9, 21.0 (ꢁ2), 21.3, 21.8, 61.8, 62.2,
62.6, 68.4, 68.9, 74.6, 76.5, 90.9, 122.5 (CF3, q, J¼320 Hz), 170.5 (ꢁ2),
not be quoted; 13C NMR:
d
(CD3CN) 19.4 (ꢁ2), 19.6, 22.0, 27.4, 31.0,
170.6 and 171.3; m/z (ES þve mode) 430 (cation Mþ, 100%).
31.3, 33.0, 33.4, 61.6, 67.9, 68.1, 73.4, 73.6, 85.7, 92.2, 125.3, 125.7,
126.7, 127.1, 127.5, 127.7, 128.1, 128.2, 129.2, 129.7, 136.7, 139.1, 139.8,
141.0, 142.9, 143.6, 168.6, 169.0 and 169.6; m/z (ES þve mode) 794
(MHþ, 100%).
4.7. [N-(b-D-Glucopyranosyl)-N-methyl]piperidinium
trifluoromethanesulfonate (18)
To a solution of 17 (0.532 g, 0.92 mmol)inMeOH(6 mL)wasadded
NaOMe (0.046 mmol). The reaction was left stirring at room tem-
perature for 5 h, then quenched with Amberlite Hþ resin
(0.046 mmol) to reach pH 7. The resin was filtered off and washed
withmethanol, and then thesolventwas removed invacuotogive the
product 18 in quantitative yield as a non-crystalline foam. Found: m/z,
4.10. [N-[2,3,4-Tri-O-acetyl]-b-D-glucopyranosyl]-N,N-
dimethyl-3-(10,11-dihydro-5H-dibenzo[a,d]cyclohepten-5-
ylidene)-1-propanammonium trifluoromethanesulfonate (23)
To a solution of 22 (0.38 mmol) in DCM (8 mL) under nitrogen
was added MeOTf (0.76 mmol). The reaction was left stirring at
20 ꢀC for 17 h. The DCM was removed in vacuo and column chro-
matography was carried out using 100% EtOAc, and then 20% EtOH/
DCM to afford the product 23 (59%) as an amorphous foam. Found:
m/z, 566.2740. C32H40NO8 requires m/z, 566.2754; nmax (diamond)
262.1661. C12H24NO5 requires m/z, 262.1654; 1H NMR:
d [(CD3)2CO]
1.69–1.78,1.91–2.03, 2.05–2.12 (6H, m, 3ꢁCCH2C), 3.27 (3H, s, CH3Nþ),
3.48 (1H, t, J¼9.5 Hz, 4-H), 3.45–3.75 (2H, m, CH2Nþ), 3.65–3.75 (2H,
m, 2ꢁ6-H), 3.67 (1H, t, J¼9.3 Hz, 3-H), 3.87–3.98 (2H, m, CH2Nþ), 3.90
(1H, t, J¼8.8 Hz, 2-H), 4.10–4.20(1H, m, 5-H) and 5.00 (1H, d, J¼8.8 Hz,
1-H); 13C NMR:
d
[(CD3)2CO] 20.7, 20.8, 22.0, 45.5, 61.1, 61.8, 61.9, 70.2,
cmꢂ1 3347 (br s), 1751, 1604, 1520 and 1452; 1H NMR:
d (CD3CN,
71.6, 78.7, 81.7, 93.5, 119.9 (CF3, q, J¼320 Hz); only the central two
peaks of the CF3 quartet can be assigned confidently as this is a weak
spectrum; m/z (ES þve mode) 262 (cation Mþ, 100%).
343 K) 2.01, 2.03 and 2.09 (9H, 3d, 3ꢁCH3CO; fluxional effects), 2.64
(2H, m, ]CHCH2), 3.02, 3.04 (6H, 2d, 2ꢁCH3Nþ), 2.95–3.30 (4H, br
m, ArCH2CH2Ar), 3.30–3.45 (1H, br m, one NþCH2), 3.50–3.70 (4H,
br m, 2ꢁ6-Hþ5-Hþone NþCH2), 4.86 (1H, approx. d, J¼6.5 Hz, 1-H),
5.08 (1H, approx. t, J¼9 Hz, 4-H), 5.26 (1H, approx. t, J¼7.8 Hz, 3-H),
5.41 (1H, dd, J¼9.2 and 7.5 Hz, 2-H), 5.79 (1H, t, J¼6.4 Hz, ]CHCH2)
4.8. [N-(6-O-Trityl)-b-D-glucopyranosyl]-N-methyl-3-(10,11-
dihydro-5H-dibenzo[a,d]cyclohepten-5-ylidene)-1-
propanamine (21)
and 7.05–7.45 (8H, m, ArH); 13C NMR:
d (CD3CN) 19.5, 19.6, 19.8,
22.5, 31.2, 32.9, 48.3, 59.2, 63.6, 66.3, 67.0, 72.8, 77.5, 91.0, 120.5
(CF3, q, J¼310 Hz), 122.8, 125.9, 127.1, 127.4, 127.7, 128.1, 130.0, 136.9,
138.6, 139.1, 169.3, 169.6 and 169.8; only the central two peaks of
To a solution of 6-O-trityl glucose 19 (2.83 mmol) in DCM
(15 mL) under nitrogen was added nortriptyline 20 (2.83 mmol).

6402
L. Iddon et al. / Tetrahedron 65 (2009) 6396–6402
14. Salekh, M. A.; Krasavina, L. S.; Vigdorchik, M. M.; Turchin, K. F.; Kuleshova, E. F.;
Suvorov, N. N. Zh. Org. Khim. 1989, 25, 2613–2619; (Chem. Abstr. 1990, 113,
115694).
the CF3 quartet can be assigned confidently as this is a weak
spectrum (cf. 18); m/z (ES þve mode) 566 (cation Mþ, 100%).
15. (a) Collins, P. M.; Ferrier, R. J. Monosaccharides; Wiley: Chichester, UK and
New York, NY, 1995; pp 318–319; (b) Ref. 15a, p 175; cf. Baker, J. W. J. Chem.
Soc., Chem. Commun. 1929, 1205–1210; This elimination may be due in part to
the acetate groups: thus a trimethyl ether anomeric chloroglucose derivative
was successfully quaternised using Me3N: Freudenberg, K.; Braun, E. Justus
Liebigs Ann. Chem. 1928, 460, 288–304.
16. Araya, I.; Akita, H. Heterocycles 2008, 75, 1213–1223.
17. Mey, U.; Wachsmuth, H.; Breyer-Pfaff, U. Drug Metab. Dispos. 1999, 27, 1281–
1292.
18. (a) Bianchi, A.; Bernardi, A. J. Org. Chem. 2006, 71, 4565–4577; (b) Paulsen, H.;
Gyorgydeak, Z.; Friedman, M. Chem. Ber. 1974, 107, 1590–1613.
19. Marmuse, L.; Nepogodiev, S. A.; Field, R. A. Org. Biomol. Chem. 2005, 3, 2225–
2227; Unverzagt, C.; Kunz, H. J. Prakt. Chem. 1992, 335, 570–578.
20. Shiozaki, M.; Mochizuki, T.; Hanazawa, H.; Haruyama, H. Carbohydr. Res. 1996,
288, 99–108.
21. Liberek, B.; Melcer, A.; Osuch, A.; Wakiec, R.; Milewski, S.; Wisniewski, A.
Carbohydr. Res. 2005, 340, 1876–1884.
22. Skorupowa, E.; Kurszewska, M.; Konitz, A.; Wojnowski, W.; Wisniewski, A.
Carbohydr. Res. 2001, 331, 343–346. The authors used the procedure of Karrer
and Smirnoff, Ref. 1.
Acknowledgements
We are grateful to Astra Zeneca plc and the EPSRC for funding
(CASE award to L.I.): also to a referee for pointing out further ref-
erences in early literature, which are now cited in Refs. 3 and 15
below.
Supplementary data
Cif files for structures 13a and 14. Supplementary data associ-
ated with this article can be found in the online version, at
doi:10.1016/j.tet.2009.05.086.
23. An earlier report described the crystallisation of both quaternary halides, viz.
13a and the corresponding bromide, and noted that they both adopted an
orthorhombic space group but without structural detail: Krouvine, Y.; Stora, C.
Comput. Rend. 1945, 220, 328–330; (Chem. Abstr. 1945, 41, 3555).
24. Compounds13b and 13c were also successfully deprotected to give the corre-
sponding free sugars; see Experimental.
25. Paulsen, H.; Pflughaupt, K.-W. In The Carbohydrates, 2nd ed.; Pigman, W.,
Horton, D., Eds.; Academic: New York, NY, 1980; Vol. 1B, pp 881–927.
26. Hodge, J. E.; Rist, C. E. J. Am. Chem. Soc. 1952, 74, 1494–1497.
27. Hodge, J. E. Adv. Carbohydr. Chem. 1955, 10, 169–205. The first product from the
Amadori rearrangement of a glucopyranosylamine is usually a fructosylamine.
28. Iddon, L. Ph. D. thesis, University of Liverpool, in preparation.
29. A similar intermediate was used in an intramolecular glycosidation study: Bols,
M. Acta Chem. Scand. 1993, 47, 829–834; (Chem. Abstr. 1994, 120, 77532).
30. David, S. The Molecular and Supramolecular Chemistry of Carbohydrates; OUP:
Oxford, 1997; pp 60–62.
31. cf. De Vries, N. K.; Buck, H. M. Carbohydr. Res. 1987, 165, 1–16; (Chem. Abstr. 1988,
108, 187084).
32. Available as its HCl salt from Sigma–Aldrich Co.
33. Briggs, A. J.; Glenn, R.; Jones, P. G.; Kirby, A. J.; Ramaswamy, P. J. Am. Chem. Soc.
1984, 106, 6200–6206.
34. Romers, C.; Altona, C. Acta Crystallogr. 1963, 16, 1225–1232.
35. Herpin, P.; Famery, R.; Auge´, J.; David, S.; Guibe´, L. Acta Cryst., Sect. B: Struct. Sci.
1976, 32, 209–215.
References and notes
1. The original preparation was given by Karrer, P.; Smirnoff, A. P. Helv. Chim. Acta
1921, 4, 817–821.
2. Micheel, F. Ber. Dtsch. Chem. Ges. 1929, 62B, 687–693.
3. McCloskey, C. M.; Coleman, G. H. J. Org. Chem. 1945, 10, 184–193; Also a qua-
ternisation of a protected glycosylamine disaccharide (cellobiose) has been
reported: Karrer, P.; Harloff, J. C. Helv. Chim. Acta 1933, 16, 962–968 and here
again (cf. above) vigorous hydrolysis led to an anhydrosugar, with no isolation
of the free quaternary ammonium salt.
4. Luo, H.; Hawes, E. M.; McKay, G.; Midha, K. K. J. Pharm. Sci. 1992, 81, 1079–1083.
5. Hawes, E. M. Drug Metab. Dispos. 1998, 26, 830–837.
6. Kaku, T.; Ogura, K.; Nishiyama, T.; Ohnuma, T.; Muro, K.; Hiratsuka, A. Biochem.
Pharmacol. 2004, 67, 2093–2102.
7. Zhu, B.; Bush, D.; Doss, G. A.; Vincent, S.; Franklin, R. B.; Xu, S. Drug Metab.
Dispos. 2008, 36, 331–338.
8. Palbir, S. G.; Soine, H. J. Chromatogr., B 1997, 691, 111–117.
9. Hou, B. K.; Lim, E. K.; Higgins, G. S.; Bowles, D. J. J. Biol. Chem. 2004, 279, 47822–
47832.
10. Stachulski, A. V.; Jenkins, G. N. Nat. Prod. Rev. 1998, 15, 173–188.
11. Stachulski, A. V.; Harding, J. R.; Lindon, J. C.; Maggs, J. L.; Park, B. K.; Wilson, I. D.
J. Med. Chem. 2006, 49, 6931–6945.
12. Kaspersen, F. M.; van Boeckel, C. A. A. Xenobiotica 1987, 17, 1451–1471.
13. Aboul-Enein, H. Y. J. Carbohydr., Nucleosides, Nucleotides 1977, 4, 77–81; (Chem.
Abstr. 1977, 87, 85188).
36. Lichtenthaler, F. W.; Sakakibara, T.; Oeser, E. Carbohydr. Res. 1977, 59, 47–61.
37. Lichtenthaler, F. W.; Ronninger, S.; Kreis, U. Liebigs Ann. Chem. 1990, 1001–1006.