Highly stereoselective synthesis of aminoglycosides via rhodium-catalyzed and substrate-controlled aziridination of glycals

Article abstract of DOI:10.1039/b823099b

The flexible installations of a sulfamate ester on a glycal scaffold at C3, C4, or C6 approaching α- or β-aminoglycosides is communicated. A variety of glycal acceptors (O, S, and N) were applied, enhancing the utility of this method as an operationally s

Full text of DOI:10.1039/b823099b

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/obc | Organic & Biomolecular Chemistry
Highly stereoselective synthesis of aminoglycosides via rhodium-catalyzed and
substrate-controlled aziridination of glycals†
Rujee Lorpitthaya,a K. B. Sophy,b Jer-Lai Kuob and Xue-Wei Liu*a
Received 22nd December 2008, Accepted 29th January 2009
First published as an Advance Article on the web 11th February 2009
DOI: 10.1039/b823099b
The flexible installations of a sulfamate ester on a glycal
scaffold at C3, C4, or C6 approaching a- or b-aminoglycosides
is communicated. A variety of glycal acceptors (O, S, and
N) were applied, enhancing the utility of this method as an
operationally simple protocol for the stereoselective synthesis
of polyfunctionalized a- or b-aminosaccharides.
Carbohydrates feature prominently in many biological processes
and in the progression of diseases. In recent years, much research
Scheme 1 Three models for generation of 1,2-aminoglycosides.
has been done to develop new methods for the synthesis of
lation was started on course from 1, 2, to 3, according to the
biologically relevant oligosaccharides, glycoconjugates, and their
sulfamate position and the ring size of intermediates. The catalytic
analogues.1 In particular, 2-amino-2-deoxy glycosides, a major
reaction underwent a one-step process under treatment of the
class of carbohydrates, are vital constituents of glycosaminogly-
substrate with a mixture of 5 mol% Rh2(OAc)4, PhI(OAc)2 and
cans, peptidoglycans, and blood group antigens.2 These com-
pounds serve as challenging synthetic targets to prepare and
transform. In one study, the direct introduction of a nitrogen
substituent at C2 of a glycal held considerable promise to access to
2-aminosugars and other complex 2-aminoglycoside conjugates.3,4
Many research groups had developed the use of metal-catalyzed
nitrene delivery to a double bond of glycal derivatives in inter-
and intramolecular fashions,5 but most of the common methods
require subsequent manipulation to generate a useful glycal donor
and complement with a variety of glycal acceptors.
We have pursued an intramolecular strategy that utilizes an
iminoiodinane-derived sulfamate ester as an electrophilic nitrogen
source6 and generate a transient aziridine intermediate as a highly
activated donor for selective 1,2-trans aminoglycosylation.7,8 Hav-
ing a sulfamate-ester group on the C3, C4, or C6 position of
the glycal allows us to investigate the reactivity, facial preference
and steric hindrance of the aziridine formation. The aziridine
intermediates serve as regio- and stereoselective donors in the
glycosylation step. Conceptually, sulfamate glycals 1 and 3 would
offer the opportunity for the synthesis of a-2-aminosugars; in
turn, a nitrogen atom delivery from the bottom face of 2 would
give b-linked 2-aminosugars (Scheme 1).
MgO in dry dichloromethane.9 The mechanistic pathway for the
formation of 6 is proposed in Scheme 2. The rhodium-catalyzed
aziridination on the top face of molecule formed the transient
aziridine 5, which was regioselectively attacked by an acetate anion
at the anomeric carbon. After 24 h, the suspension was filtered
through celite and purified by silica gel column chromatography
furnishing an inseparable mixture of a- and b-anomers (2 : 1) and
only 75% conversion was observed. The loss of stereospecificity
and long reaction time of 1 is supposedly caused by the strain
of forming aziridine intermediate 5. This result is comparable to
prior studies by Rojas et al.8a,b The rhodium-catalyzed aminogly-
cosylation of D-allal-derived carbamate resulted in an anomeric
mixture (3 : 1).
Scheme 2 Rhodium-catalyzed aminoglycosylation of 1.
The requisite glycal starting materials were constructed to incor-
porate a variety of commonly employed carbohydrate protecting
groups. Investigation of the substrate-controlled aminoglycosy-
In view of the stereoselectivity, the installation of sulfamate ester
or carbamate at C3 is unlikely to be a good model to achieve pure
aminoglycosides. We suggested that this could be solved by either
increasing the ring size of the intermediate or elongating the arm
of the sulfamate moiety. Consequently, the introduction of the
sulfamate moiety on C4 and C6 of the glycal molecule was carried
out.
To examine the nitrogen insertion on the bottom face of
the glycal molecule, compound 7 was prepared. The initial
investigation was made by subjecting sulfamate glycal 7 to the
optimized catalytic conditions. Instead of nitrogen atom insertion
to the double bond, C–H insertion and cleavage of a silyl protective
group spontaneously took place, giving rise to compound 10 as
aDivision of Chemical and Biological Chemistry, School of Physical
and Mathematical Sciences, Nanyang Technological University, Singapore
637371, Singapore. E-mail: xuewei@ntu.edu.sg; Fax: +65 6791 1961;
Tel: +65 6316 8901
bDivision of Physics and Applied Physics, School of Physical and Math-
ematical Sciences, Nanyang Technological University, Singapore 637371,
Singapore
† Electronic supplementary information (ESI) available: Synthetic, spec-
tral and crystallographic details. CCDC reference number 708553. For
ESI and crystallographic data in CIF or other electronic format see DOI:
10.1039/b823099b
1284 | Org. Biomol. Chem., 2009, 7, 1284–1287
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
Table 1 Stereoselective b- and a-aminoglycosylation of 2 and 3
a side product (Scheme 3). Investigation of the formation of the
unexpected product 10, without addition of a rhodium catalyst,
found that the reaction occurred in the same manner but needed
longer reaction time and gave a lower conversion. Based on this
result, the rhodium complex seems to play a crucial role in the
catalytic system by increasing the rate of reaction, but it is still
premature to conclude that having the O-silyl group adjacent to
the allylic proton promotes N-insertion on the C–H rather than
=
the C C bond. Structure of compound 10 was confirmed by X-ray
crystallography and identified as a hemi(ethanol) solvate.
Isolated yield
Entry
Nucleophile
Conditionsa
11
12
1
2
3
-OAc
-OAc
A
A
B
73b
68c
83
—
82
94
4
5
B
B
80
90
89
95
6
7
Bd
Bd
79
86
84
83
8
Bd
—
78
Scheme 3 Predictive pathway for C–H insertion on sulfamate glycal 7.
9
-N3 [from
B
76
80
However, our aim to succeed in b-aminoglycosylation was
continued by removing a labile silane group from the C3 position.
Replacement of a benzoyl protecting group in 2a–b gave stable
substrates, which could be stored at -4 ◦C for more than a month.
(CH3)3Si-N3]e
a Condition A: 5 mol% of [Rh2(OAc)4], PhI(OAc)2 (1.5 equiv), MgO
(5 equiv) at RT. Condition B: 5 mol% of [Rh2(tfacam)4], PhIO (1.5 equiv),
MgO (5 equiv). Nucleophile (2 equiv) was added in 15 min for glycal
=
2 and 30 min for glycal 3, respectively. b Yield for P1 = C(O)Ph, P2
=
Moreover, the nitrogen atom insertion to the C C bond could
be accomplished within 1–2 hours and with virtually complete
1,2-trans stereochemistry (entries 1–2, Table 1). No C–H insertion
products were seen for both 2a and b.
OSiMe2tBu. c Yield for P1 = P2 = C(O)Ph. d Nucleophile (1.2 equiv) and
BF3·OEt2 (0.1 equiv). e N3 was generated by treatment of (CH3)3Si-N3
-
(2 equiv) and TBAF (2 equiv) at 0 ◦C for 30 min.
It was realized that an increase in flexibility for the formation
of the aziridine intermediate led to success in the synthesis of
2-aminoglycosides. With this concept, we then moved on to
the a-linked aminoglycoside syntheses. The introduction of the
sulfamate moiety on C6 would produce an 8-membered ring
fused with an aziridine ring, which is more flexible than previous
cases. The catalytic reaction of sulfamate glycal 3 could be
completed within 5 hours affording a-linked 2-aminoglycoside
in 82% yield (entry 2, Table 1). Reaction of 3 showed a promising
result that could be applied to the stereoselective synthesis of
a-aminosugars.10
The intramolecular rhodium-catalyzed aminoglycosylation was
generally accomplished in one pot because of the instability of the
aziridine intermediates. Trapping such aziridine species became
a challenging problem in the synthesis of functionalized 1,2-
aminoglycosides. Though a Rh2(OAc)4 catalyst can be widely used
with a variety of functionalized alcohols, none of desired products
was detected for thiols, and amines. Those compounds presumably
reduce the catalytic capability of [Rh2(OAc)4]. In contrast, the
employment of [Rh2(tfacam)4] and PhIO showed modest to good
results in all cases.11 Alcohol nucleophiles could be neatly intro-
duced into the catalytic reactions without adding any promoters,
but thiols required activation by a Lewis acid, BF3·OEt2. Another
application of this methodology is to synthesize diamino sugars
-
(entry 9, Table 1). An azide anion, N3 , was generated by treatment
of trimethysilyl azide and TBAF at 0 ◦C and then slowly injected
into the aziridine intermediate. The sulfamate ester-derived glycals
2 and 3 selectively furnished b- and a-aminoglycosides.
To investigate the cause of the different amination patterns of 1,
2, and 3, we utilized the hybrid density functional theory (DFT) to
gain a deeper understanding of the mechanistic reaction as well as
to determine the energy order of the corresponding intermediates.
The mechanism of all three series involved in the aziridination
reaction is similar and presented in Scheme 4 using sulfamate 2 as
a reference. Initially, the intermediate 13 was generated, followed
by the rhodium-catalyzed elimination of the PhI moiety from 14
giving rise to the rhodium–nitrene complex 15. Prior to aziridine
ring formation in 19, our calculations show the rapid relaxation of
the singlet state 15 to the triplet state 16, which such a reaction is
likely to undergo from the ground triplet state. The transition state
17 could be located on the triplet potential energy surface (PES)
and its N–C1 proximity was much less than the distance between
˚
˚
N and C2 (2.035 A and 2.611 A, respectively). The N–C1 single
˚
bond was then formed in 18 (1.497 A), whereas N and C2 kept the
˚
distance at 2.509 A. These led us to believe that a nitrogen atom
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 1284–1287 | 1285
©

View Article Online
Scheme 4 Rhodium catalytic cycle for aziridination.
delivery to the double bond is taking place in a stepwise diradical
process.
state 17 for all three series agrees well with the corresponding
experimental reaction time. The reaction time for sulfamate glycal
1 was slower than 3 and 2, respectively (see ESI†). This convinces
us that the delivery of a nitrene species from C3 to the double
bond of glycal was a laborious route, which was different from a
nitrene on C4 and C6. The reactions proceed fast and smooth for
glycals 2 and 3.
In summary, facial preference, which is controlled by substrate,
was found to be the key factor that determines stereoselec-
tivity. The experimental and computational data show that a
sulfamate ester on C3 could not impact on the stereoselective
The energy diagram for the aziridination step is presented in
Scheme 5. We observed that the relative energies of 14, 15, and
16 for all three series are close to each other with less than 1 kcal
mol-1 difference. This indicated that the position of the sulfamate
ester on the sugar ring does not effect those transformations. On
the other hand, a large change in energies was obviously observed
in the nitrene transfer step. The free energy value of 17 from glycal
1 is higher than that of 17 from 2 and 3 by 8.7 and 5.2 kcal mol-1,
respectively. The variation in the Gibbs free energies of transition
Scheme 5 Relative energy diagram of the aziridination step of 1, 2, and 3.
1286 | Org. Biomol. Chem., 2009, 7, 1284–1287 This journal is The Royal Society of Chemistry 2009
©

View Article Online
aminoglycosylation. On the contrary, installation of a sulfamate
ester on the C6 position of a glycal substrate approaches
a-aminoglycosides selectively. Also, introduction of the sulfamate
moiety on C4 gives b-linked 2-amino sugars. This series of glycal
glycosylations can be applied in the chemical synthesis of highly
complex aminosaccharide conjugates.
Engl., 1996, 35, 1380; (c) T. K. Park, I. J. Kim, S. Hu, M. T. Bilodeau,
J. T. Randolph, O. Kwon and S. J. Danishefsky, J. Am. Chem. Soc.,
1996, 118, 11488; (d) J. M. Owens, B. K. S. Yeung, D. C. Hill and P. A.
Petillo, J. Org. Chem., 2001, 66, 1484.
4 (a) J. Du Bois, C. S. Tomooka, J. Hong and E. M. Carreira, J. Am.
Chem. Soc., 1997, 119, 3179; (b) R. S. Dahl and N. S. Finney, J. Am.
Chem. Soc., 2004, 126, 8356.
5 (a) C. G. Espino, J. Du Bois in Modern Rhodium-Catalyzed Organic
Reaction, ed. P. A. Evans, Wiley-VCH, Weinheim, 2005, p. 379; (b) P. M.
Wehn, J. Lee and J. Du Bois, Org. Lett., 2003, 5, 4823; (c) J. L. Liang,
J. S. Huang, X. Q. Yu, N. Zhu and C. M. Che, Chem.–Eur. J., 2002,
8, 1563; (d) J. L. Liang, S. X. Yuan, J. S. Huang, W. Y. Yu and C. M.
Che, Angew. Chem., Int. Ed., 2002, 41, 3465; (e) J. Liu and D. Y. Gin,
J. Am. Chem. Soc., 2002, 124, 9789; (f) P. Muller and C. Fruit, Chem.
Rev, 2003, 103, 2905.
6 (a) E. Kozlowska-Gramsz and G. Descotes, Tetrahedron Lett., 1981,
22, 563; (b) E. Kozlowska-Gramsz and G. Descotes, Can. J. Chem.,
1982, 60, 558.
7 (a) S. C. Bergmeier and D. M. Stanchina, J. Org. Chem., 1999, 64, 2852;
(b) T. Bach, B. Schlummer and K. Harms, Chem. Commun., 2000, 287.
8 (a) R. Bodner, B. K. Marcellino, A. Severino, A. L. Smenton and
C. M. Rojas, J. Org. Chem., 2005, 70, 3988; (b) E. Levites-Agababa, E.
Menhaji, L. N. Perlson and C. M. Rojas, Org. Lett., 2002, 4, 863; (c) C.
Kan, C. M. Long, M. Paul, C. M. Ring, S. E. Tully and C. M. Rojas,
Org. Lett., 2001, 3, 381.
Acknowledgements
We thank Dr Yong-Xin Li for X-ray analyses and gratefully
acknowledge Nanyang Technological University and Ministry of
Education, Singapore for financial support.
Notes and references
1 (a) A. Varki, R. Cummings, J. Esko, H. Freeze, G. Hart, J. Marth,
in Essentials of Glycobiology, Cold Spring Harbor Laboratory Press,
Cold Spring Harbor, NY, 1999; (b) For reviews: R. A. Dwek and
T. D. Butters, Chem. Rev., 2002, 102, 283; (c) C. R. Bertozzi and L. L.
Kiessling, Science, 2001, 291, 2357; (d) A. Dove, Nat. Biotechnol., 2001,
19, 913.
2 (a) R. A. Dwek, Chem. Rev., 1996, 96, 683; (b) Synthetic Oligosac-
charides, Indispensable Probes for the Life Sciences ed. P. Kovac, ACS
Symposium Series 560, American Chemical Society, Washington DC,
1994.
3 (a) D. A. Griffith and S. J. Danishefsky, J. Am. Chem. Soc., 1990, 112,
5811; (b) S. J. Danishefsky and M. T. Bilodeau, Angew. Chem., Int. Ed.
9 The roles of additive, temperature and solvent were further investigated,
see ESI.
10 R. Lorpitthaya, Z. Z. Xie, J. L. Kuo and X. W. Liu, Chem.–Eur. J.,
2008, 14, 1561.
11 The comparison results of the rhodium(II)-catalyzed aziridination by
Rh2(OAc)4 and Rh2(tfacam)4 are shown in the ESI.
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 1284–1287 | 1287
©