One pot/two donors/one diol give one differentiated trisaccharide: Powerful evidence for reciprocal donor-acceptor selectivity (RDAS)

Article abstract of DOI:10.1039/b206598c

Three component, one-pot reactions involving equimolar amounts of the acceptor diol and both armed and disarmed donors presented simultaneously, produce a single double-differential glycosidation product; this phenomenon provides evidence for Reciprocal D

Full text of DOI:10.1039/b206598c

One pot/two donors/one diol give one differentiated trisaccharide:
powerful evidence for reciprocal donor–acceptor selectivity (RDAS)
Bert Fraser-Reid,*a J. Cristóbal López,*b K. V. Radhakrishnan,a M. V. Nandakumar,a Ana M.
Gómezb and Clara Urielb
a Natural Products and Glycotechnology Research Institute Inc, (NPG), 4118 Swarthmore Road, Durham,
North Carolina, USA 27706. E-mail: Dglucose@aol.com; Fax: 1 919 4936113
b Instituto de Química Orgánica General, Juan de la Cierva 3, 28006 Madrid, Spain.
E-mail: clopez@iqog.csic.es; Fax: 34 915644853
Received (in Cambridge, UK) 9th July 2002, Accepted 8th August 2002
First published as an Advance Article on the web 21st August 2002
Three component, one-pot reactions involving equimolar
amounts of the acceptor diol and both armed and disarmed
donors presented simultaneously, produce a single double-
differential glycosidation product; this phenomenon pro-
vides evidence for Reciprocal Donor Acceptor Selectivity
(RDAS).
The traditional protocol for differential, double-glycosidation of
an acceptor diol requires a series of programmed protection/
deprotection steps to ensure that only one of the acceptor-OHs
is presented to each donor at each coupling event.1,2 Thus, for
the case illustrated in equation (i), a minimum of four steps
would be needed to obtain compound 3 from diol 1, without
contamination of the diastereomer 4 (and avoidance of the
symmetrical competitors 5 and 6). The direct process, 1?3, is
generally thought to require exquisite regioselective finesse and
therefore is best left to enzymatic procedures.3
However, in this manuscript, we report the development of
differential, double-glycosidations of diols in which donors are
virtually ‘told’ where to go, thereby enabling direct conversions
of the type 1?3, without contamination of regioisomer 4.
We recently reported that a variety of diols, including 7, 8 and
9, underwent regioselective glycosidations upon treatment with
n-pentenyl glycosyl donors.4 Thus, a disarmed donor (NPGBZ),
e.g. 10, and/or its orthoester (NPOE) equivalent, e.g. 11,
glycosidated the bold OH overwhelmingly (and frequently
exclusively)5 whereas the armed donor, e.g. 12, was pro-
miscuous, reacting substantially with both italic and bold-
OHs.
Clearly, these examples indicate that each donor expresses
preference for one of the diol –OHs and vice versa, the resulting
‘match’ being evidence for Reciprocal Donor Acceptor Se-
6
lectivity (RDAS).† A rationalization for these selectivities
awaits further insight; but their reality invites immediate
practical exploitation.
For in-depth studies, we chose to first examine altroside 7.
This diol had given a 92% yield of 13 as the exclusive product
with NPOE 11a, but a 2+1 mixture of 13b and 14 with the armed
NPGALK 124 [Scheme 1 (a and b)]. Notably the preferred site for
both donors is the bold (C3)-OH which, on the basis of
conventional wisdom, did not augur well for differential,
double-glycosidation experiments.
Nevertheless, when a 1+1.3+1.3 mixture of diol 7, and both
donors, (11a and 12) was treated with 2.5 equivalents of NIS
and a catalytic amount of BF3·Et2O for 10 min, trisaccharide 15
and disaccharide 13a, were obtained as the only products in 37
and 37% yields respectively.
Obviously, the yield of trisaccharide 15, would be improved
if glycosidation of disaccharide 13a by NPGALK donor 12 could
be enhanced. Indeed, entries i?iv in Table 1 show that
increasing the concentration of the armed donor 12, had a
salutary effect on the yield of 15.
An even more interesting set of results arises from our studies
on diol 16 (Scheme 2). Differential 3,6-dimannosylation of this
mannoside is of interest since the resulting trimannan occurs as
Scheme 1 Glycosidation of altroside 7 with glycosyl donors 11a and 12.
2104
CHEM. COMMUN., 2002, 2104–2105
This journal is © The Royal Society of Chemistry 2002

Table 1 One-pot glycosidation of altroside7 with donors 11a and 12.
Table 2 One-pot glycosidation of mannoside 16 with donors 10 and 12.
Entry
7 (equiv.) 11a (equiv.) 12 (equiv.) 15a (%) 13a (%)
Entry
16 (equiv.)
10 (equiv.)
12 (equiv.)
17a (%) 21a (%)
i
1
1
1
1
1.3
1
1
1.3
1.6
2
37
43
52
57
37
31
19
16
i
ii
1
1
1
1
1
2
36
27
21
25
ii
iii
iv
a Compound 22 is the only trisaccharide (of four possibilities) observed in
these optimizations.
1
3
a Compound 15 is the only trisaccharide (of four possibilities) observed in
these optimizations.
a repetitive, interlocking motif in high mannose glycopro-
teins.7
The RDAS preferences of 16 were determined by the
previously described equimolar two-component reactions‡
shown in Scheme 2 (a and b). With the disarmed donor 10,
mannosylation occurred at the bold (C6)-OH only to give 17 in
53% yield, and also the symmetrical trisaccharide 18 in 13%
yield—but with no evidence (TLC nor NMR) for the dimannan
resulting from glycosidation of the italic (C3)-OH. By contrast,
the ‘armed’ donor 12 gave a 38% yield of the O6 product, 19,
but also 11% of the O3 regioisomer 20.
Analysis of the results in Scheme 2 (a and b) according to
conventional wisdom, dictates that the preference of both
donors, 10 and 12, for the primary –OH ‘is to be expected’ on
the grounds of steric hindrance, and so in contemplating a
differential, double-glycosidation experiment, the obvious
question was: What will happen when 10 and 12 compete for
diol 16? Our calculations8 showed that the relative reactivity of
these donors (k12/k10) is 3.2. Hence, it was expected that O6
mannosylation by the ‘armed’ donor, 12, would predominate in
any trimannan produced.
Scheme 3
reaction of 10, 16 and 12 (entry i) gave a single trimannan 22,
in which the less reactive donor 10 ended up at O6. But even
more surprisingly, the same held true for the single disaccha-
ride, 17, obtained.As with the altroside study in Scheme 1, an
increase in the ratio of the armed donor 12 (entry ii) led to an
increase (albeit modest) of trisaccharide 21—but still none of
the symmetrical trisaccharide.
In reviewing the above data, we regard it as simply
astonishing that even with the audacious disparity in the ratio of
donors 12 and 11a (Table 1, equation iv) or 12 and 10 (Table 2,
equation ii), there was absolutely no evidence for trisaccharides
other than 15 and 21. In view of the excess of the ‘armed’ donor
12 in these experiments, symmetrical dimannosylation was
expected in both Schemes 1 and 2.
Surprising disagreement with conventional wisdom is de-
picted in Scheme 2(c). Thus the 1+1+1 three-component
A study8 of the three types of n-pentenyl donors used above
indicate that their relative reactivities are in the order NPOE >
armed > disarmed (e.g. 11 > 12 > 10). The most and least
stable donors therefore give rise to the highly delocalised, more
stable intermediate 22, while the armed donor gives the less
stable oxocarbenium ion 239 (Scheme 3). The conclusion from
Scheme 1(c) and Scheme 2(c) is that in competitive glycosida-
tions, the more stable donor/intermediate (not the most reactive
donor) controls regioselectivity, resulting in the formation of
the single trisaccharides 15 and 21 and the single disaccharides
13a and 17. How the competing OH groups play into this
phenomenal regioselectivity awaits clarification.
B. F. R. is grateful to the Human Frontier Science Program
Organization and Synthon Chiragenics of Monmouth Junction,
NJ, USA for partial support of this work. Funding from the
DGICYT (grants: PPQ2000-1330, and BQU2001-0582).
Notes and references
† The regioselectivities are the same whether the NPOE or NPGBZ is used,
although yields and side-products may differ.
‡ The structure of the regioisomers was determined, as previously
described.4
1 H. M. I. Osborn and T.H. Khan, Oligosaccharides. Their synthesis and
biological roles, Oxford University Press, Oxford, 2000, ch. 3.
2 H. Paulsen, Angew. Chem., Int. Ed. Engl., 1982, 21, 152.
3 C-H. Wong, R. L. Halcomb, Y. Ichikawa and T. Kajimoto, Angew.
Chem., Int. Ed. Engl., 1995, 34, 521; S. David, C. Auge and C.
Gautheron, Adv. Carbohydr. Chem. Biochem., 1991, 49, 176.
4 B. Fraser-Reid, J. C. Lopez, K. V. Radhakrishnan, M. Mach, U.
Schlueter, A. M. Gomez and C. Uriel, J. Am. Chem. Soc., 2002, 124,
3198.
5 G. Anilkumar, L. G. Nair and B. Fraser-Reid, Org. Lett., 2000, 2,
2587.
6 B. Fraser-Reid, J. C. Lopez, K. V. Radhakrishnan, M. Mach, U.
Schlueter, A. M. Gomez and C. Uriel, Can. J. Chem., in press.
7 J. Montruiel, Adv. Carbohydr. Chem. Biochem., 1980, 37, 157.
8 B. G. Wilson and B. Fraser-Reid, J. Org. Chem., 1995, 60, 317.
9 With S. Grimme and M. Piacenza, unpublished results.
Scheme 2 Glycosidation of mannoside 16 with donors 10 and 12.
CHEM. COMMUN., 2002, 2104–2105
2105