Preparation, X-ray structure and reactivity of a stable glycosyl iodide

Article abstract of DOI:10.1039/b302629a

Highly selective reaction of methyl tetra-O-pivaloyl-β-D-glucopyranuronate 2 with iodotrimethylsilane or (Me₃Si)₂ and I₂ affords, in excellent yield, the 'disarmed' glycosyl iodide 1 which has good stability at 20°C and ex

Full text of DOI:10.1039/b302629a

Preparation, X-ray structure and reactivity of a stable glycosyl iodide
a
a
c
b
d
Jamie Bickley, Jennifer A. Cottrell, John R. Ferguson, Robert A. Field, John R. Harding, David
b
b
c
c
L. Hughes, K. P. Ravindanathan Kartha, Jayne L. Law, Feodor Scheinmann and Andrew V.
a
Stachulski*
a
The Robert Robinson Laboratories, Department of Chemistry, University of Liverpool, Liverpool, UK L69
7
ZD. E-mail: stachuls@liv.ac.uk; Fax: 44 (0)151 794 3588; Tel: 44 (0)151 794 3542
b
Centre for Carbohydrate Chemistry, School of Chemical Sciences and Pharmacy, University of East
Anglia, Norwich, UK NR4 7TJ
Ultrafine UFC Ltd., Synergy House, Guildhall Close, Manchester Science Park, Manchester, UK M15 6SY
AstraZeneca UK Ltd., Drug Metabolism and Pharmacokinetics Department, Mereside, Alderley Park,
Macclesfield, Cheshire, UK SK10 4TG
c
d
Received (in Cambridge, UK) 7th March 2003, Accepted 2nd April 2003
First published as an Advance Article on the web 30th April 2003
Highly selective reaction of methyl tetra-O-pivaloyl-b-
glucopyranuronate 2 with iodotrimethylsilane or (Me Si)
and I affords, in excellent yield, the ‘disarmed’ glycosyl
iodide 1 which has good stability at 20 °C and excellent
stability at 0 °C; the X-ray crystal structure of 1 is described,
along with a comparison of its utility as a glycosyl donor to
that of the corresponding bromide.
D
-
and away from light (best stored in a Nalgene bottle); at 20 °C
it remains unchanged after many months.
The glycosyl iodide 1 was readily crystallised from hexane,
mp 104–105 °C and gave crystals suitable for X-ray structure
determination (pale yellow prisms, orthorhombic).17 The struc-
3
2
2
ture of 1 is presented in Fig. 1.
It can be seen that 1 adopts a normal ‘chair’ C
4
1
conforma-
tion: its structure differs in minor respects from that of the
bromosugar 3. The O–C(1) bond shows characteristic bond
shortening (1.398 Å, identical to the O–C(1) bond in 3: typical
Although a very wide range of glycosyl donors are now
1
available, including in particular thioglycosides, anomeric
2
3
4
3
trichloroacetimidates, sulfoxides and pentenyl glycosides,
sp C–O, 1.43Å), and the C(1)–I bond characteristic lengthening
3
anomeric halosugars are still commonly employed in glycosida-
tion reactions. Traditionally glycosyl bromides have been used,
activated by heavy metal salts (usually those of Ag and Hg,
occasionally of Cd), the classical Koenigs–Knorr conditions.5
Within the last decade it has been shown that iodine reagents are
effective alternatives to the heavy metals, for both ‘armed’ and
(2.22 Å: typical sp C–I, 2.16 Å), owing to the anomeric effect.
Table 1 Preparative methods for the anomeric iodosugar 1
a
Reagents, conditions
Time/h
Yield (%)
1
3,14
‘
disarmed’ cases, and are effective for other anomeric leaving
KI, BF
3
·Et
2
O, MeCN, 2, 80 °C
3
2.75
5
67
91
92
53
9a,10
6,7
8
Me SiI, MeCN, 2, 50 °C
groups. It has been traditionally held that glycosyl iodides
are too difficult to prepare, and too unstable, to be useful
glycosyl donors. A number of publications have appeared
regarding their preparation, nevertheless, and others have cited
them as reactive in situ intermediates. In particular Gervay et
al.
3
3
Si)
2
, I
, I
2
, CH
, MeCN, 2, 50 °C1
2 2
Cl
, 2, 20 °C1
5
(Me
6
PhSiMe
3
2
1.7
a
After chromatography and/or crystallisation.
9
10,11
have shown that ‘armed’ glycosyl iodides will react with
a variety of nucleophiles and have reinvestigated ‘disarmed’
iodosugars without isolating them.11
As is widely understood, substitution critically affects
reactivity and stability in all carbohydrate derivatives. We now
report that the pivaloate-protected glycosyl iodide 1, derived
from glucuronolactone, is formed in high yield and under mild
12
conditions from the b-tetrapivaloate 2 and moreover pos-
sesses excellent stability. We also present the X-ray structure of
1
and briefly compare its reactivity to that of the corresponding
12
bromosugar 3.
Various methods for the conversion of anomeric ester 2 to
glycosyl iodide 1 were studied, as summarised in Table 1.
Marginally best was the hexamethyldisilane–I
2
method,15
3 3
which generates Me SiI in situ, though the Me SiI procedure
also afforded a 90%+ yield.10 The chemical stability of 1 is
excellent. At 0 °C it is stable for well over one year if kept dry
Fig. 1 X-Ray structure of iodosugar 1, (top) in plan, (bottom) in elevation
showing the ‘chair’.
1
266
CHEM. COMMUN., 2003, 1266–1267
This journal is © The Royal Society of Chemistry 2003

View Article Online
2
3
4
5
(a) R.R. Schmidt, Angew. Chem., Int. Ed. Engl., 1986, 5, 212; (b) W.
Kinzy and R. R. Schmidt, Adv. Carbohydr. Chem. Biochem., 1994, 50,
1.
(a) D. Kahne, S. Walker, Y. Cheng and D. van Engen, J. Am. Chem.
Soc., 1989, 111, 6881; (b) L. Yan and D. Kahne, J. Am. Chem. Soc.,
2
1
996, 118, 9239.
(a) D. R. Mootoo, V. Date and B. Fraser-Reid, J. Am. Chem. Soc., 1988,
10, 2662; (b) A. J. Ratcliffe, P. Konradsson and B. Fraser-Reid, J. Am.
1
Chem. Soc., 1990, 112, 5665.
(a) W. Koenigs and E. Knorr, Chem. Ber., 1901, 34, 957; (b) B.
Helferich and K. F. Wedeneyer, Liebigs Ann. Chem., 1949, 563, 139; (c)
R. B. Conrow and S. Bernstein, J. Org. Chem., 1971, 36, 863; (d)
Summarised by H. Paulsen, in Modern Methods in Carbohydrate
Synthesis, S. H. Khan and R. A. O’Neill, eds., Harwood Academic
Publishers, The Netherlands, 1996, pp. 1–19.
Scheme 1
6 K. P. R. Kartha, M. Aloui and R. A. Field, Tetrahedron Lett., 1996, 37,
807.
K. P. R. Kartha, T. S. Karkkainen, S. J. Marsh and R. A. Field, Synlett.,
001, 260.
For example P. M. Collins and R. J. Ferrier, Monosaccharides, J. Wiley,
Chichester, 1995, p. 163.
8
7
8
9
2
The conformations of the C(2) and C(5)/C(6) substituents,
however, differ between 1 and 3, probably owing to the relative
sizes of the Br/I atoms and packing effects. The H NMR
spectra of the two halosugars also show some differences. Both
the H(2)–H(3) axial–axial and H(1)–H(2) axial–equatorial
coupling constants in 1, respectively 9.8 and 4.4 Hz,17 fall
within the typical pyranose range, but very interestingly the
chemical shift of the 2-H is significantly lower in 1 than in 3,
whereas the chemical shift of the 1-H is higher in 1 than in 3. A
1
Especially:(a) A. Klemer and M. Bieber, Liebigs Ann. Chem., 1984,
1
052; (b) B. Ernst and T. Winkler, Tetrahedron Lett., 1989, 30, 3081; (c)
U. Schmid and H. Waldmann, Tetrahedron Lett., 1996, 37, 3837; (d) S.
M. Chervin, P. Abada and M. Koreeda, Org. Lett., 2000, 2, 369.
10 J. Gervay and M. J. Hadd, J. Org. Chem., 1997, 62, 6961.
11 (a) J. Gervay, T. N. Nguyen and M. J. Hadd, Carbohydr. Res., 1997,
3
3
00, 119; (b) S. N. Lam and J. Gervay-Hague, Carbohydr. Res., 2002,
37, 1953; (c) J. Gervay, Glycosyl Iodides in Organic Synthesis, in
similar trend has been noted before in anomeric iodosugars of
the galactose series:15 moreover the difference in chemical
Organic Synthesis: Theory and Applications, JAI Press Inc., New York,
998, vol. 4, pp. 121–153.
1
shifts of the anomeric protons of 1 and 3, at 0.36 ppm, is
comparable to that between the anomeric protons of the parent
-bromo- and 2-iodo-tetrahydropyrans (0.46 ppm) as given by
1
1
2 J. Vlahov and G. Snatzke, Liebigs Ann. Chem., 1983, 570.
3 J. R. Ferguson, J. L. Law, F. Scheinmann and A. V. Stachulski, British
Patent Application, No. 9914382.8, to Ultrafine UFC Ltd., July 1999.
2
1
8
Anderson and Sepp.
14 Cf. the corresponding (and much less stable) triacetate: R. T. Brown, F.
Scheinmann and A. V. Stachulski, J. Chem. Res. (S), 1997, 370.
15 K. P. R. Kartha and R. A. Field, Carbohydr. Lett., 1998, 3, 179.
16 T.-L. Ho and G. A. Olah, Synthesis, 1977, 417.
1
Because of the continuing interest in morphine-6-glucur-
onide 4 as an analgesic more potent and better tolerated than
the glycosyl donor ability of 1 was studied
in this context (Scheme 1). In a model experiment, 1 (1 eq.)
morphine itself,1
3,19
1
7 Full characterisation was obtained for 1. H NMR of 1: d
H
(400 MHz,
C), 3.74 (3 H, s, CH O),
.27 (1 H, dd, J = 9.8 and 4.4 Hz, 2-H), 4.37 (1 H, d, J = 10.3 Hz, 5-H),
.31 (1 H, t, 3-H), 5.61 (1 H, t, 4-H) and 7.01 (1 H, d, J = 4.4 Hz, 1-H).
CDCl
3
) 1.13, 1.18 and 1.20 (27 H, 3 s, 3 3 Me
3
3
proved to be a suitable donor for glycosidation of 2-phenyl-
4
5
ethanol (1.5 eq.) using ZnCl
2
(1.1 eq.) as catalyst in 1,2-di-
chloroethane (20 h, 20 °C); glucuronate ester 520 was isolated in
13
C 3
C NMR: d (100 MHz, CDCl ) 27.42, 27.51, 38.95, 39.12, 39.17,
5
6
6% yield (unoptimised). Reaction of 3-O-pivaloyl morphine
with 1 (1.1 eq.) mediated by I
catalyst afforded glucuronate ester 7 (55% after chromatog-
5
3.34, 68.31, 70.37, 70.94, 71.70, 75.49, 167.04, 176.99 and 177.30.
1
9c
2
(2.5 eq.) with no other
+
9
Found: C, 46.4; H, 6.25; MNa , 593.1230. C22H35IO requires C, 46.3;
H, 6.20%; C22H35IO Na requires m/z, 593.1224. Crystal data:
9
1
3,21
raphy) after 60 h.
The yield compares well to that obtained
C22H35IO , M = 570.40, orthorhombic, space group P2 2 2 , a =
9
w
1 1 1
using the trichloroacetimidate method2 and the product was
identical to 7 prepared in that way; moreover only a 10% excess
of 1 is required. By contrast, reaction of 6 with the bromosugar
2
9.6982(16), b = 11.4712(19), c = 22.658(4) Å, T = 100 K, U =
3
21
2
520.7(7) Å , Z = 4, m(Mo-Ka) = 1.317 mm , 4595 reflections
2
3
2
collected, 3911 unique.
Rint = 0.0413, final wR(F ) = 0.0525 (all
24
data). Absolute structure Flack = 0.006(16). CCDC reference number
05813. See http://www.rsc.org/suppdata/cc/b3/b302629a/ for crys-
tallographic data in CIF or other electronic format.
H NMR of 3: d (400 MHz, CDCl ) 1.14, 1.17 and 1.19 (27 H, 3 s, 3 3
Me C), 3.74 (3 H, s, CH O), 4.59 (1 H, d, J = 10.3 Hz, 5-H), 4.85 (1
3
(which afforded typically 80–85% yields in glycosidation of
2
20
primary alcohol acceptors) under comparable conditions gave
only a 20% yield of 7. Finally, base-catalysed hydrolysis of 7
led cleanly to morphine-6-glucuronide 41 in 85% yield.
In summary, we have characterised a highly stable ‘dis-
armed’ glycosyl iodide, namely 1, and shown it to possess a
typical chair structure by X-ray diffraction. This represents the
first crystal structure for this class of molecule. Furthermore, we
have shown glycosyl iodide 1 to be an effective glycosyl donor
in reactions with both a primary and a secondary alcohol
acceptor.
1
H
3
9c
3
3
H, dd, J = 9.9 and 4.1 Hz, 2-H), 5.28 (1 H, t, 3-H), 5.69 (1 H, t, 4-H)
and 6.66 (1 H, d, J = 4.1 Hz, 1-H)..
18 C. B. Anderson and D. T. Sepp, J. Org. Chem., 1967, 32, 607.
1
9 (a) R. J Osborne, S. P. Joel, D. Trew and M. Slevin, Lancet, 1988, i: 828;
b) R. J. Osborne, P. I. Thompson, S. P. Joel, D. Trew, N. Patel and M.
(
Slevin, Br. J. Clin. Pharmacol., 1992, 34, 130; (c) F. Scheinmann, K. W
Lumbard, R. T. Brown and S. P. Mayalarp, International Patent, WO
9
3/3051, 1993, to Ultrafine UFC Ltd.
0 A. V. Stachulski, Tetrahedron Lett., 2001, 42, 6611.
1 It was known that ZnBr in dichloromethane at reflux was a suitable
We are grateful to the EPSRC and Astra Zeneca PLC for an
industrial CASE studentship (to J. A. C.) and to Professor A. J.
Kirby (University of Cambridge) for valuable comments.
2
2
2
catalyst for the glucuronidation of 3-acylmorphines using a glycosyl
bromide: I. Rukhman, G. Nisnevich and A. L. Gutman, Tetrahedron,
2
001, 57, 1083.
2
2
2 R. T. Brown, N. E. Carter, F. Scheinmann and K. W. Lumbard,
Tetrahedron Lett., 1995, 36, 8661.
3 G. M. Sheldrick, Shelx97, Crystal structure determination program,
University of Gottingen, Gottingen, Germany, 1997.
Notes and references
1
(a) P. Fuegedi, P. J. Garegg, H. Loenn and T. Norberg, Glycoconjugate
J., 1987, 4, 97; (b) F. Dasgupta and P. J. Garegg, Carbohydr. Res., 1988,
24 H. D. Flack, Acta Crystallogr. Sect. A Fundam. Crystallogr., 1983, 39,
177, C13.
876.
CHEM. COMMUN., 2003, 1266–1267
1267