Deuterium-isotope study on the reductive ring opening of benzylidene acetals

Article abstract of DOI:10.1039/c1ob06056b

Specific deuterated reference compounds were prepared to probe the stereoselectivity of the reductive ring opening of carbohydrate-based benzylidene-type acetals. AlD₃ revealed a retentive stereoselectivity probably through the rare S_N

Full text of DOI:10.1039/c1ob06056b

View Article Online / Journal Homepage / Table of Contents for this issue
Organic &
Biomolecular
Chemistry
Dynamic Article Links
Cite this: Org. Biomol. Chem., 2011, 9, 7655
www.rsc.org/obc
COMMUNICATION
Deuterium-isotope study on the reductive ring opening of benzylidene acetals†
I-Chi Lee,a Medel Manuel L. Zulueta,a,b Chi-Rung Shie,a Susan D. Arcob and Shang-Cheng Hung*a,c
Received 30th June 2011, Accepted 18th August 2011
DOI: 10.1039/c1ob06056b
Specific deuterated reference compounds were prepared to
probe the stereoselectivity of the reductive ring opening of
carbohydrate-based benzylidene-type acetals. AlD3 revealed
a retentive stereoselectivity probably through the rare SNi
(internal nucleophilic substitution) mechanism. An SN1-
like mechanism occurs in the acid-promoted regioselective
BD3·THF- or Et3SiD-reductive ring opening.
Regioselective protections of multiple hydroxyls are obligatory
steps in the chemical synthesis of complex carbohydrates and
natural products.1 Here, the readily installed and easily manip-
ulated benzylidene-type acetals gained immense importance and
widespread use. These base-resistant yet acid-labile functionalities
are typically used to block 4- and 6-hydroxyls of hexopyranoses as
well as vicinal cis-diols to give fused 1,3-dioxane and 1,3-dioxolane
backbones, respectively. Aside from protecting two hydroxyl
groups simultaneously, they also exhibit significant influence on
the stereochemical outcome of chemical glycosylations.2 Another
highly useful advantage is their capability to be reductively
opened in a regioselective manner exposing a free hydroxyl and
a benzyl-type ether enabling the opportunity for further transfor-
mations.
The LiAlH4 and AlCl3 combination is the first reagent set used
in the reductive ring opening of cyclic acetals.3 They generate,
depending on mixing proportions, the alane species (i.e. AlHCl2,
AlH2Cl and AlH3) that participate in the actual hydride transfer.4
In the case of 4,6-O-benzylidene-type acetals, the ring opening
mediated by alanes occurs at the less hindered and more basic
6-O forming the 4-O-benzyl-type derivative (Scheme 1).5 Not
surprisingly, this regioselectivity is also shared by i-Bu2AlH.6 The
NaCNBH3/HCl tandem, later introduced by Garegg, notably
favoured the opposite regioselectivity—the ring opening at the
4-O position.7 In the case of the fused 1,3-dioxolanes, both
the alanes and NaCNBH3/HCl gave identical regioselectivity in
the benzylidene acetal cleavage that is dependent on the stereo-
chemical orientation of the aromatic group.8 The so-called 3X
(eXo gives aXial hydroXyls) rule of thumb1a was realized as
Scheme 1 Representative examples of the reductive ring opening of
benzylidene acetals.
a general tendency during the openings of these 5-membered
rings. Cleavage of the 1,3-dioxolanes was also found possible in
the presence of the 6-membered counterpart.9 Over the years,
a number of boron-,10 silicon-,10f,10h,11 and tin-based12 hydride
reagents in combination with a similarly diverse set of Lewis
and Brønsted acids were introduced and applied to reductively
open benzylidene-type acetals. Variations in hydride source,
ligand, temperature, solvent and the accompanying acid displayed
profound effects on the regioselectivity of the ring opening.
The mechanistic aspects of the acid-mediated nucleophilic
addition to cyclic acetals resulting in ring opening have been widely
investigated. Regioselectivity rests on the stability of the initial
complex between the acid and the acetal oxygens and the relative
ease with which the nucleophile approaches the acetal carbon.
Relating to the C–O bond that would be broken, the nucleophile
may form the new covalent bond in either an inversive or retentive
manner. It was suggested13 that the initial complex promotes the
formation of ion pairs that vary from being intimate leading to
inversive stereoselectivity via an SN2-like mechanism to being
separated, presumably through an oxocarbenium ion, enabling
stereorandomization in an SN1-like manner. However, alanes,
which are themselves weak Lewis acids, were earlier speculated to
resort to internal hydride transfer following initial complexation
with the acetal oxygen.14 This mechanism, which results in stere-
ochemical retention, was dismissed13 due to insufficient evidence
at the time. Recent reports highlighted the influence of solvent
and ligand on the relative acid strengths of the Lewis acid and
the reducing agent, which subsequently determines the regio- and
stereoselectivity of the ring opening.15
aGenomics Research Center, Academia Sinica, 128, Section 2, Academia
Road, Taipei 115, Taiwan. E-mail: schung@gate.sinica.edu.tw
bInstitute of Chemistry, University of the Philippines, Diliman, Quezon City
1101, Philippines
cDepartment of Applied Chemistry, National Chiao Tung University, 1001,
Ta-Hsueh Road, Hsinchu, 300, Taiwan
† Electronic supplementary information (ESI) available: Experimental
procedures and NMR spectra. See DOI: 10.1039/c1ob06056b
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 7655–7658 | 7655
©

View Article Online
In our preliminary isotope study, the benzylidene acetal in
compound 1 was regioselectively opened using either BD3·THF
or Et3SiD both catalysed by Cu(OTf)2.10f Based on the NMR
analysis of the products, BD3·THF generated a deuterated 4-O-
benzyl derivative with a 5 : 1 diastereomeric ratio whereas Et3SiD
formed the deuterated 6-O-benzyl counterpart with apparent
stereorandomization (i.e. 1 : 1 ratio of diastereomers). The NMR
data gathered, however, was inadequate to fully distinguish the
configuration of the isomers. Overall, the stereoselective inclina-
tions of the reductive ring opening of benzylidene-type acetals
remain highly speculative without the full confirmation of the
structure of the deuterated products.
Analysis of the NMR spectra alone cannot be used to ascertain
the absolute structure designation of deuterated products in the
reductive ring opening of carbohydrate-based benzylidene acetals.
We, therefore, decided to synthesize deuterium-labelled reference
compounds to probe the stereoselectivity of the nucleophilic attack
on the acetal carbon (Scheme 2). The preparations made use of
the deuterated tosylate 6 as benzylating reagent via Williamson
etherification. An approximately 30% ee compound 6 has been
reported16 which is not appropriate for our study. Alternatively,
(S)-(+)-benzyl-a-d1 alcohol, generated by the enantioselective
reduction of PhCDO with (R)-alpine borane (reportedly attainable
in 96% ee17), was tosylated18 to provide the desired 6. Considering
the possible ring openings of a 4,6-O-benzylidene acetal and
the ring cleavage of a 2,3-O-benzylidene with an exo-oriented
phenyl group, the alcohols 7,19 8,4b and 9 were treated with 620
and NaH to generate the deuterated benzyl ethers 10 (64%), 11
(78%), and 12 (86%), respectively. Desilylation of 10 with tetra-
n-butylammonium fluoride (TBAF) furnished the target 6-O-
ring opening reference compound 13R in 95% yield. Subsequent
cleavage of the 2-naphthylmethyl (2-NAP) and p-methoxybenzyl
(PMB) groups of compounds 11 and 12 gave the individual
reference alcohols 14R (95%) and 15R (93%), respectively. All the
reference compounds are expected to carry the deuterated carbon
in the R-configuration.
The 6-O-ring opening of compound 1 was carried out using
AlD3 or BD3·THF as deuteride source. Inversion of configuration
at the acetal carbon would result in the 6-alcohol 13R whereas
1
retention would lead to 13S (Fig. 1). Comparison with the H
NMR spectrum of the undeuterated 6-alcohol 2 identified the 4-
O-benzylic proton of our reference compound (13R) at around
4.61 ppm. The product of AlD3 treatment only showed a very
minor signal for 13R with a significant peak, ascribed to the 4-
O-benzylic proton of 13S, appearing at 4.86 ppm. The reaction
is highly stereoselective (13R/13S = 6/94) providing compelling
evidence for a retentive deuteride addition. In previous papers,10f,10h
we reported the benzylidene 6-O-ring opening in compound 1 with
BH3·THF catalysed by various metal triflates in excellent yields.
Examination of the stereoselectivity of the BD3·THF-mediated
transformation in tandem with some selected metal triflates as well
as TMSOTf revealed notable similarities in the 13R/13S ratios
(approx. 1/5). Stereochemical retention is favoured but inversion
is more pronounced in contrast to that of AlD3. The proportion of
the diastereomers also appears to be independent of the quantity
of added Lewis acid as shown for TMSOTf (0.5 equiv versus 1.1
equiv).
Fig. 1 1H NMR spectra for the 6-O-ring opening of compound 1.
The 4-O-ring opening of the benzylidene acetal in compound 1
was then carried out using Et3SiD in the presence of Cu(OTf)2,10f
BF3·Et2O,11c or CF3COOH.11a In this transformation, the reference
compound 14R corresponds to retentive substitution whereas 14S
represents inversion. The identified signal relating to the 6-O-
benzylic proton in the 1H NMR spectrum of 14R in comparison
to that of the 4-alcohol 3 appears at 4.50 ppm. As illustrated in
Fig. 2, the diastereomers 14R and 14S are generated in nearly
equal amounts in all cases (14S is measured from the resonance
Scheme 2 Preparations of the reference compounds.
7656 | Org. Biomol. Chem., 2011, 9, 7655–7658
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
was found to agree with our result; the original interpretation was
based on a wrong stereochemical assignment. The highly retentive
deuteride addition by AlD3 led us to suggest the mechanism
following the rarely observed internal nucleophilic substitution
(SNi). As depicted in Scheme 3, AlD3 weakly coordinates with
the less hindered and more nucleophilic 6-O and subsequently
transfers a deuteride to the acetal carbon on the same side of
the departing 6-O. The same mechanism is also expected for the
reductive ring opening by i-Bu2AlH. The BD3·THF cleavage of
the benzylidene acetal also displayed a preference for a retentive
deuteride addition with relatively minor inversion. Together with
the requirement for a Lewis acid, the NMR data precludes the
SN2-like substitution suggested for intimate ion pairs13,15c or a
mechanism similar to that of AlD3. We propose that the acid
complexation with 6-O generated a separated ion pair with
an oxocabenium ion involving O-4. The stereochemical bias is
probably caused by asymmetric induction of the nearby chiral
center (4-C of the sugar) leading to more retention rather than
inversion. The reduction carried out by Et3SiD together with an
acid appears to include an oxocarbenium ion intermediate. Full
stereorandomization suggests deuteride addition in an SN1-like
fashion. The acid may in principle interact with either acetal
oxygen granting more preference for 6-O; but, in this case, the
resultant 4-O-benzyl cation cannot be reduced as Et3SiD is bulky
and not very reactive. On the other hand, the silane can readily
approach either side of the 6-O-benzyl cation formed by the initial
acid complexation with 4-O.
Fig. 2 1H NMR spectra for the 4-O-ring opening of compound 1.
at 4.55 ppm). The data clearly demonstrates stereorandomization
that is independent of the nature of the acid used.
The stereoselectivity of the ring opening of the 1,3-dioxolane in
compound 16 was next probed (Fig. 3). With the exo orientation
of the phenyl ring, the reductive ring opening is predicted to
favour the formation of the 2-alcohol 171b instead of the 3-alcohol
counterpart. Thus, the diastereomers 15R (retentive reduction)
and 15S (inversive reduction) are the possible products upon
reduction with AlD3. The resonance at 4.71 ppm, indicated by the
reference compound (15R), fully coincides with the major peak of
the ring-opening product. Comparison with the minor signal at
around 4.86 ppm attributed to 15S (in reference to the 1H NMR
spectrum of compound 17) gave a 91/9 ratio (15R/15S) clearly
favouring the retentive introduction of the deuteride.
Scheme 3 Mechanisms of the benzylidene ring opening.
In conclusion, the deuterated reference compounds we prepared
were successfully utilized to clarify the stereoselectivity of the
reductive benzylidene ring opening. The data we presented can
supplement further selectivity and mechanistic analysis of this
type of reaction.
Fig. 3 1H NMR spectra for the 2-O-ring opening of compound 16.
With the reference compounds and their corresponding 1H
NMR data, the stereoselectivity of the reductive ring opening
of benzylidene acetals is now evident. The AlD3-mediated ring
opening possesses retentive stereoselectivity for acetals with both
1,3-dioxane and 1,3-dioxolane backbones. By applying our NMR
data, the previously reported15c inversive ring opening of a deuter-
ated derivative of compound 1 with the LiAlH4/AlCl3 tandem
Acknowledgements
This work was supported by the National Science Council
(NSC 97-2113-M-001-033-MY3, NSC 98-2119-M-001-008-MY2,
NSC 98-3112-B-007-005, NSC 98-2627-B-007-008) and Academia
Sinica.
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 7655–7658 | 7657
©

View Article Online
Tzeng, S. S. Kulkarni, B.-J. Uang, C.-Y. Hsu and S.-C. Hung, Angew.
Chem., Int. Ed., 2005, 44, 1665; (g) A. I. Zinin, N. N. Malysheva, A. M.
Shpirt, V. I. Torgov and L. O. Kononov, Carbohydr. Res., 2007, 342,
627; (h) C.-R. Shie, Z.-H. Tzeng, C.-C. Wang and S.-C. Hung, J. Chin.
Chem. Soc., 2009, 56, 510.
Notes and references
1 (a) A. Lipta´k, A. Borba´s and I. Bajza, in Comprehensive Glycoscience:
From Chemistry to Systems Biology, ed. J. P. Kamerling, Elsevier,
Amsterdam, 2007, vol. 1, pp. 203–259; (b) C.-C. Wang, J.-C. Lee, S.-Y.
Luo, S. S. Kulkarni, Y.-W. Huang, C.-C. Lin, K.-L. Chang and S.-C.
Hung, Nature, 2007, 446, 896; (c) C.-C. Wang, S. S. Kulkarni, J.-C.
Lee, S.-Y. Luo and S.-C. Hung, Nat. Protoc., 2008, 3, 97; (d) X. Zhu
and R. R. Schmidt, Angew. Chem., Int. Ed., 2009, 48, 1900; (e) K.-L.
Chang, M. M. L. Zulueta, X.-A. Lu, Y.-Q. Zhong and S.-C. Hung, J.
Org. Chem., 2010, 75, 7424.
11 (a) M. P. DeNinno, J. B. Etienne and K. C. Duplantier, Tetrahedron
Lett., 1995, 36, 669; (b) S. Chandrasekhar, Y. R. Reddy and C. R.
Reddy, Chem. Lett., 1998, 27, 1273; (c) S. D. Debenham and E. J.
Toone, Tetrahedron: Asymmetry, 2000, 11, 385; (d) M. Sakagami and
H. Hamana, Tetrahedron Lett., 2000, 41, 5547.
12 B.-Z. Zheng, M. Yamauchi, H. Dei, S.-I. Kusaka, K. Matsui and O.
Yonemitsu, Tetrahedron Lett., 2000, 41, 6441.
2 D. Crich and L. Li, J. Org. Chem., 2007, 72, 1681.
3 (a) E. L. Eliel, V. G. Badding and M. N. Rerick, J. Am. Chem. Soc.,
1962, 84, 2371; (b) B. E. Leggetter and R. K. Brown, Can. J. Chem.,
1964, 42, 990; (c) B. E. Leggetter, U. E. Diner and R. K. Brown, Can.
J. Chem., 1964, 42, 2113.
13 S. E. Denmark and N. G. Almstead, J. Am. Chem. Soc., 1991, 113,
8089.
14 (a) H. A. Davis and R. K. Brown, Can. J. Chem., 1971, 49, 2563; (b) W.
J. Richter, J. Org. Chem., 1981, 46, 5119.
4 E. C. Ashby and J. Prather, J. Am. Chem. Soc., 1966, 88, 729.
5 (a) P. Fu¨gedi, A. Lipta´k, P. Na´na´si and J. Szejtli, Carbohydr. Res.,
1982, 104, 55; (b) A. Lipta´k, A. Borba´s, L. Ja´nossy and L. Szila´gyi,
Tetrahedron Lett., 2000, 41, 4949.
6 T. Mikami, H. Asano and O. Mitsunobu, Chem. Lett., 1987, 16,
2033.
7 (a) P. J. Garegg and H. Hultberg, Carbohydr. Res., 1981, 93, C10; (b) P.
J. Garegg, H. Hultberg and S. Wallin, Carbohydr. Res., 1982, 108, 97.
8 P. J. Garegg, Pure Appl. Chem., 1984, 56, 845.
15 (a) R. Johnsson, R. Cukalevski, F. Dragen, D. Ivanisevic, I. Johansson,
L. Petersson, E. E. Wettergren, K. B. Yam, B. Yang and U. Ellervik,
Carbohydr. Res., 2008, 343, 2997; (b) R. Johnsson, D. Olsson and U.
Ellervik, J. Org. Chem., 2008, 73, 5226; (c) R. Johnsson, M. Ohlin and
U. Ellervik, J. Org. Chem., 2010, 75, 8003.
16 C. Fisher, E. Morse, B. Romer, T.-P. You, C. W. Mosher and H. S.
Mosher, Tetrahedron, 1992, 48, 2993.
17 D. W. Barnett, M. J. Panigot and R. W. Curley, Tetrahedron: Asymme-
try, 2002, 13, 1893.
9 A. Lipta´k, J. Imre, J. Harangi, P. Na´na´si and A. Neszme´lyi, Tetrahedron,
1982, 38, 3721.
18 F. Kazemi, A. R. Massah and M. Javaherian, Tetrahedron, 2007, 63,
5083.
10 (a) M. Ek, P. J. Garegg, H. Hultberg and S. Oscarson, J. Carbohydr.
Chem., 1983, 2, 305; (b) S. Saito, A. Kuroda, K. Tanaka and R. Kimura,
Synlett, 1996, 231; (c) M. Oikawa, W.-C. Liu, Y. Nakai, S. Koshida, K.
Fukase and S. Kusumoto, Synlett, 1996, 1179; (d) L. Jiang and T.-H.
Chan, Tetrahedron Lett., 1998, 39, 355; (e) C.-C. Wang, S.-Y. Luo, C.-
R. Shie and S.-C. Hung, Org. Lett., 2002, 4, 847; (f) C.-R. Shie, Z.-H.
19 G. D. K. Kumar and S. Baskaran, J. Org. Chem., 2005, 70, 4520.
20 Subsequent NMR analysis of the corresponding benzylic proton of
compound 14R gave a diastereomeric ratio of 97/3. This would
translate to an enantiomeric excess of 94% for compound 6 ignoring
any possible racemization during the predicted inversive etherification
step.
7658 | Org. Biomol. Chem., 2011, 9, 7655–7658
This journal is
The Royal Society of Chemistry 2011
©