Lactosamine from lactulose via the Heyns rearrangement: A practical protocol

Article abstract of DOI:10.1016/j.tetlet.2013.05.086

The Heyns rearrangement is possibly one of the oldest, easiest and most economic ways to synthesise 2-deoxy-2-amino sugars. Initially reported yields were disappointingly low, but in the late 90s Wrodnigg and Stütz discovered modified reaction conditions

Full text of DOI:10.1016/j.tetlet.2013.05.086

Tetrahedron Letters 54 (2013) 3960–3961
Contents lists available at SciVerse ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Lactosamine from lactulose via the Heyns rearrangement: a practical
protocol
⇑
Yulong Shan, Farah Oulaidi, Martina Lahmann
School of Chemistry, University of Bangor, Alun Roberts Building, Deiniol Road, Bangor, Gwynedd LL57 2UW, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 20 February 2013
Revised 7 May 2013
Accepted 17 May 2013
Available online 23 May 2013
The Heyns rearrangement is possibly one of the oldest, easiest and most economic ways to synthesise 2-
deoxy-2-amino sugars. Initially reported yields were disappointingly low, but in the late 90s Wrodnigg
and Stütz discovered modified reaction conditions that gave substantially increased yields, making the
reaction viable on preparative scale. Requiring larger amounts of N-acetyl lactosamine, we utilised the
reported reaction conditions and found that additional modifications, especially in the work-up proce-
dure—for example employing rapid filtration of crude diethyl ether precipitates, were necessary to make
this approach robust in the research laboratory environment.
Keywords:
Heyns rearrangement
N-Acetyllactosamine lactulose
2-Deoxy-2-amino sugar
Ó 2013 Elsevier Ltd. All rights reserved.
2-Deoxy-2-amino sugars are undoubtedly important in living
organisms and widely distributed as subunits in glycoconjugates
such as glycosaminoglycans and peptidoglycans.1 While N-acety-
lated glucosamine is a readily accessible bulk chemical, generally
produced by hydrolysis of the second most abundant biopolymer
in the world, chitin,2 N-acetyl lactosamine is a common disaccha-
ride epitope in complex glycans such as in the Lewis x family,
but is rather expensive.
Many alternative routes for the preparation of 2-deoxy-2-amino
sugars have been described, including the slightly hazardous but
well-applied azidonitration3 and azidoselenation4,5 of glycals. Re-
cent achievements in the preparation and use of 2-deoxy-2-amino
sugars have been thoroughly reviewed.6,7
Thus, only very few studies have included the Heyns rearrange-
ment.13 In 1999, Wrodnigg and Stütz revisited this neglected reac-
tion by focusing on benzylamine as the nitrogen source, producing
N-acetyl-D-lactosamine in 38–45% yield with respect to lactulose
using a ‘one-pot’ protocol.14 This study was followed up by apply-
ing their elaborated methodology to a range of ketuloses producing
various N-substituted sugar amine derivatives.8,15
To our knowledge two companies are producing lactosamine on
industrial scale in 300 kg batches, and giving the prospect of access
to a variety of rare and expensive sugar derivatives. However, de-
spite the apparent ease of the protocol this development seemed to
have difficulties in entering the carbohydrate community.
To our knowledge, only one research paper16 on orthogonally
protected Lewis x derivatives, and a patent17 on novel lactosamine
derivatives have utilised the Heyns rearrangement for lactosamine.
Possibly the oldest, and maybe the easiest way to access 2-
deoxy-2-amino sugars is via the Heyns rearrangement (Scheme 1).
As summarised earlier,8 the observation that ketoses react with
amines to afford ketosyl amines, followed by isomerisation into the
corresponding 2-amino-2-deoxy-1-aldoses, was first reported by
Fischer et al.9 This type of reaction was further investigated by
CH2OH
CH2OH
N
H
OH
OH
CHOH
NH
H
OH
OH
CHO
NH R
H
OH
C
O
R
R
H
HO
H
cat H+
RNH2
Heyns and Koch in 1952.10 They found that
ammonia to form -glucosamine. Subsequently, it was demon-
D-fructose reacts with
HO
H
H
H
OH
OH
HO
H
H
HO
H
H
D
H
OH
strated that numerous organic amines react in a similar fash-
ion.11,12 The major drawback of this reaction sequence is the
generally low to moderate isolated yields (about 20%). Competing
Amadori rearrangement, especially when using primary amines,
and the ease of hydrolysis of the initial condensation product are
possibly the major factors accounting for this.
CH2OH
CH2OH
CH2OH
CH2OH
D-fructose
HO
O
OH
O
H
N
HO
HO
R
OH
R
OH
HN
OH
HO
ketosamine
⇑
Corresponding author. Tel.: +44 (0)1248 38 2390; fax: +44 (0)1248 37 0528.
E-mail address: m.lahmann@bangor.ac.uk (M. Lahmann).
Scheme 1. Heyns rearrangement of D-fructose shown with intermediate steps.
0040-4039/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.tetlet.2013.05.086

Y. Shan et al. / Tetrahedron Letters 54 (2013) 3960–3961
3961
HO
O
depend on the quality of the catalyst. At this point, it was also pos-
sible to remove remaining lactulose 1 from the reaction mixture by
cation exchange chromatography giving D-lactosamine hydrochlo-
ride 4 in an overall 60% yield starting from (1). However, it was
found that this extra purification step did not increase significantly
the overall yield.
HO
O
HO
HO
OH
O
OH
O
O
NHBn
O
OH
i
HO
HO
OH
OH
HO
HO
HO
HO
2
1
N-acetylation was achieved by adding sodium methoxide and
subsequently acetic anhydride to the crude material 4 providing
crude 5 (yield, 82%). The crude N-acetyl lactosamine 5 was then
peracetylated using acetic anhydride in pyridine to form amino su-
gar 6 (Scheme 2). After this step, the material was purified by flash
column chromatography providing the target compound 6 as an
anomeric mixture in a total yield of 48% on a 25 g scale with re-
spect to the starting material 1.
In summary, we improved the protocol for the Heyns rearrange-
ment giving straightforward access to N-acetyl lactosamine from
lactulose. Similar to the previously reported procedures the same
material can be prepared without chromatography and other N-
protecting groups can be used.
HO
HO
OH
OH
OH
OH
ii
O
O
O
iii
O
O
O
HO
HO
OH
OH
Cl
HO
HO
OH
OH
NHBn
NH3
3
4
AcO
OAc
HO
HO
OAc
OH
O
v
iv
O
OH
O
O
O
AcO
O
OAc
AcO
OH
AcO
HO
AcHN
OH
AcHN
6
5
Scheme 2. Reagents and conditions: (i) BnNH2, 40 °C, 3 d; (ii) MeOH, glacial AcOH,
rt, 2 h; (iii) HCl, pH 1–2, Pd(OH)2/C, H2, 6 bar, 40 °C, 60%; (iv) MeOH, NaOMe, Ac2O,
40 °C, 82%; (v) Ac2O, pyridine, rt, overnight, 67%.
Acknowledgments
Previously, in 1996, the Cytel Corporation submitted a patent
application on sialyl Lewis x analogues detailing Heyns rearrange-
ment reaction conditions producing successfully the desired
derivatives, but encountering similar problems linked to the inter-
mediate work-up procedures as we experienced when we started
our investigations.18 Requiring larger amounts of N-acetyl lactos-
amine,19 we set out to utilise the reported reaction conditions
and found, after numerous unsuccessful attempts, that additional
modifications, especially in the work-up procedure, were neces-
sary to provide the desired product in acceptable yields. This com-
munication discusses the additional details of this practical
protocol.
The authors thank Dr. Viacheslav Tverezovskiy for access and
help with the WHXINGYU jacket hydrogenation apparatus (Bio-
composites Centre, Bangor University). Financial support from
the European Commission (INTERREG IVA WINSS, project
F002726) is gratefully acknowledged.
Supplementary data
Supplementary data (a detailed experimental description and
selected NMR spectra are reproduced and compiled) associated
with this article can be found, in the online version, at http://
dx.doi.org/10.1016/j.tetlet.2013.05.086.
Lactulose (1) (25 g) was reacted with excess benzylamine to
produce ketosyl amine 2 as previously described,15 monitoring
the reaction progress by TLC. Ketosyl amine 2 was produced as
the major product while some unreacted lactulose remained
(Scheme 2). NMR sampling can be used to follow the reaction.
The crude mixture was then precipitated from diethyl ether at
4 °C without the reported addition of methanol.8,16 Possibly due
to the large amount of benzylamine, the precipitate showed a ten-
dency to convert into a syrup-like mass as described in one of the
patents.18 This could be avoided by rapid suction filtration using a
small amount of cold diethyl ether (4 °C) to wash out excess ben-
zylamine. The rearrangement reaction was then carried out with
the crude yellow solid 2 (35 g) using glacial acetic acid in methanol
at room temperature. During this step, N-benzyl lactosamine 3 was
formed within a short time (2 h). Isolation of this crude material 3,
still containing some unreacted lactulose 1 and remaining benzyl-
amine, was difficult, as the material refused to solidify after slow
addition to diethyl ether and subsequent cooling. However, using
a similar work-up procedure as applied for the first reaction pro-
vided crude solid 3 (ca. 40 g). It turned out to be advantageous
not to allow the precipitation of 3 from diethyl ether in a refriger-
ator because, after extended times, the semi-crystalline material
started to convert into a syrup-like mass that could not be filtered.
The best way was to filter the precipitate immediately. Although
the hydrogenolysis can be carried out at ambient pressure, applica-
tion of 6 bar hydrogen pressure increased the overall efficiency of
the reaction. According to our results, the reaction times seem to
References and notes
1. Varki, A.; Cummings, R. D.; Esko, J. D.; Freeze, H. H.; Stanley, P.; Bertozzi, C. R.;
Hart, G. W.; Etzler, M. E. Essentials of Glycobiology, 2nd ed.; Cold Spring Harbor
Laboratory Press: Cold Spring Harbor, NY, 2009.
2. Chen, J. K.; Shen, C. R.; Liu, C. L. Mar. Drugs 2010, 8, 2493–2516.
3. Lemieux, R. U.; Ratcliffe, R. M. Can. J. Chem. 1979, 57, 1244–1251.
4. Santoyo-Gonzalez, F.; Calvo-Flores, F. G.; Garcia-Mendoza, P.; Hernandez-
Mateo, F.; Isac-Garcia, J.; Robles-Diaz, R. J. Org. Chem. 1993, 58, 6122–6125.
5. Czernecki, S.; Ayadi, E.; Randriamandimby, D. J. Org. Chem. 1994, 59, 8256–
8260.
6. Bongat, A. F. G.; Demchenko, A. V. Carbohydr. Res. 2007, 342, 374–406.
7. Schmidt, R. R.; Vankar, Y. D. Acc. Chem. Res. 2008, 41, 1059–1073.
8. Stütz, A. E.; Dekany, G.; Eder, B.; Illaszewicz, C.; Wrodnigg, T. M. J. Carbohydr.
Chem. 2003, 22, 253–265.
9. Fischer, E. Ber. Dtsch. Chem. Ges. 1884, 17, 579–584.
10. Heyns, K.; Koch, W. Z. Naturforsch. (B) 1952, 7, 486–488.
11. Carson, J. F. J. Am. Chem. Soc. 1955, 77, 1881–1884.
12. Heyns, K.; Meinecke, K. H. Chem. Ber. Recl. 1953, 86, 1453–1462.
13. Mossine, V. V.; Barnes, C. L.; Glinsky, G. V.; Feather, M. S. Carbohydr. Res. 1996,
284, 11–24.
14. Wrodnigg, T. M.; Stütz, A. E. Angew. Chem., Int. Ed. 1999, 38, 827–828.
15. Wrodnigg, T. M.; Lundt, I.; Stütz, A. E. J. Carbohydr. Chem. 2006, 25, 33–41.
16. Ohmae, M.; Takada, J.; Murakami, H.; Kimura, S. Chem. Lett. 2011, 40, 438–439.
17. Dekany, G.; Agoston, K.; Bajza, I.; Bøjstrup, M.; Kröger, L. Lactosamine
derivatives. App. No: 12/225,094, US patent 8,207,328 B2, 26 June, 2012;
Chem. Abstr. 2007, 147, 386185.
18. DeFrees, S. A. App. No: 08/730,553, US patent 5,854,218, 29 December, 1998;
Chem. Abstr. 1999, 130, 66730.
19. André, S.; Lahmann, M.; Gabius, H.-J.; Oscarson, S. Mol. Pharm. 2010, 7, 2270–
2279.