On the basis of Hodosi and Kova´c’s work, a triflate leaving
group was the obvious choice. Unfortunately, the triflate of
Boc-Hyp-OMe has been described as unstable,11 so we had
some reservations about the viability of 5 (Scheme 5). The
Table 1. Glycosylation Reactionsa
Scheme 5. Synthesis of Hyp and hyp Derivatives
ratio
6:5
additives
(equiv)
yield 4
(%)
entry
solvent
1
2
3
4
5
6
7
8
9
10
11
12
13
1:1
1:1
2:1
6:1
2:1
2:1
2:1
2:1
2:1
2:1
2:1
2:1
2:1
DMF
b
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CHCl3
37
45
44
b
nBu4NI (2)
nBu4NBr (2)
nBu4NF (1)
nBu4NF (2)
CsF (1)
CsF (2)
CsF (2)
CsF (2)
CsF (2)
b
37
27
55
60
59
b
DMSO
CH2Cl2
37
18-crown-6 (2)
other issue that needed to be addressed was that, in fact, we
required a derivative of cis-4-hydroxyproline (hyp) since C4
was expected to undergo inversion of configuration during
the glycosylation reaction. We prepared hyp derivative cis-5
by analogy to the work of Gomez-Vidal and Silverman.12
Glycosylation reactions were conducted under a variety
of conditions (Table 1). We began with DMF since this had
been the solvent of choice for Hodosi and Kova´c.6 We
isolated no glycoside, but obtained the corresponding formate
ester 10 in good yield.13 Dichloromethane ultimately gave
the best results since both reaction partners were soluble and
stable. Employing 2 equiv of the stannylene acetal 6 relative
to triflate cis-5 was beneficial, but increasing the stoichi-
ometry further gave no advantage.
In the alkylation and acylation of carbohydrate-derived
stannylene acetals, it is conventional to include a tetrabutyl-
ammonium halide salt.5 We were reluctant to include such
a nucleophilic species since it might displace the triflate.
Indeed, inclusion of tetrabutylammonium iodide and tetra-
butylammonium bromide led to formation of the correspond-
ing 4-halo-prolines 11 and 12, respectively (entries 5 and
6). Fortunately, better results were forthcoming with the
inclusion of fluoride salts. Fluoride is less nucleophilic than
the larger halide ions, and we believe it does more than
“solubilize the tin complex”, as stated by Hodosi and Kova´c.6
In fact, it may serve as a fifth ligand in the monomeric
complex14 or a more reactive dimeric complex.15 That it does
a 4 Å molecular sieves added to all reactions. b Other products isolated:
not harm the triflate, or the Fmoc protecting group,16 is
circumstantial evidence that it is actively complexed with
the tin. Replacing the tetrabutylammonium counterion with
cesium improved the yield. Disappointingly, inclusion of a
crown ether, in a bid to make the fluoride anion more
“naked”, did not help.
Earlier reports indicated that the rate of reaction is faster
in more polar solvents.6 Chloroform led to no improvement,
and DMSO resulted in the isolation of 13, presumably
formed via a Swern-type mechanism.17
The NMR spectra of glycoside trans-4 were very complex.
To fully assign the spectra and unambiguously confirm
stereochemical issues, we removed the Fmoc protecting
group (Scheme 6). NMR spectra of secondary amine, trans-
14, were of a single species (not a mixture of rotamers as
was the case for trans-4). The anomeric proton in trans-14
showed nOe’s to both H3 and H5, indicating that all three
protons are on the same face of the pyranose ring (Figure
2). This is good evidence for the â-stereochemistry of the
glycosidic linkage.
It was also necessary to address the stereochemistry at Cγ
of the Hyp residue. Since trans-5 was readily available, we
glycosylated it, presumably forming cis-4, the glycoside of
hyp (Scheme 6). Following N-deprotection, a single com-
pound was formed (cis-14) which was distinct from trans-
(11) Lowe, G.; Vilaivan, T. J. Chem. Soc., Perkin Trans. 1 1997, 539-
546.
(12) Gomez-Vidal, J. A.; Silverman, R. B. Org. Lett. 2001, 3, 2481-
2484.
(13) DMF has been shown to behave as a formate anion equivalent and
displace tosylates: Suri, S. C.; Rodgers, S. L.; Radhakrishnam, K. V.; Nair,
V. Synth. Commun. 1996, 26, 1031-1039.
(14) David, S. In PreparatiVe Carbohydrate Chemistry; Hanessian, S.,
Ed.; Marcel Dekker: New York, 1997; pp 69-86.
(15) Bredenkamp, M. W.; Spies, H. S. C.; van der Merwe, M. J.
Tetrahedron Lett. 2000, 41, 547-550.
(16) The Fmoc group can be cleaved by dilute TBAF in DMF: Ueki,
M.; Amemiya, M. Tetrahedron Lett. 1987, 28, 6617-6620.
(17) Tidwell, T. T. Org. React. 1990, 39, 297-572.
Org. Lett., Vol. 8, No. 18, 2006
3961