Chemoenzymatic syntheses of (-)-1-deoxymannojirimycin (DMJ) and its naturally occurring 6-O-alpha-L-rhamnopyranosyl glycoside.

Article abstract of DOI:10.1039/b304063a

The naturally occurring sugar mimetic alkaloids 1-deoxymannojirimycin (DMJ, 1) and 6-O-alpha-L-rhamnopyranosyl-DMJ (2) have each been prepared in a completely stereoselective manner from the cis-1,2-dihydrocatechol 3, itself obtained in enantiomerically p

Full text of DOI:10.1039/b304063a

View Article Online / Journal Homepage / Table of Contents for this issue
Chemoenzymatic syntheses of (؊)-1-deoxymannojirimycin (DMJ)
and its naturally occurring 6-O-ꢀ-L-rhamnopyranosyl glycoside
Martin G. Banwell,*a Xinghua Ma,a Naoki Asano,b Kyoko Ikedab and John N. Lambertc
a Research School of Chemistry, Institute of Advanced Studies,
The Australian National University, Canberra, ACT 0200, Australia.
E-mail: mgb@rsc.anu.edu.au
b Department of Biochemistry, Faculty of Pharmaceutical Sciences, Hokuriku University,
Kanagawa-machi, Kanazawa 920-1181, Japan
c Biota Chemistry Laboratory, Chemistry Department, Monash University, Clayton,
Victoria 3168, Australia
Received 16th April 2003, Accepted 13th May 2003
First published as an Advance Article on the web 19th May 2003
The naturally occurring sugar mimetic alkaloids 1-deoxy-
mannojirimycin (DMJ, 1) and 6-O-ꢀ-L-rhamnopyranosyl-
DMJ (2) have each been prepared in a completely
stereoselective manner from the cis-1,2-dihydrocatechol 3,
itself obtained in enantiomerically pure form by microbial
oxidation of chlorobenzene.
The alkaloid 1-deoxymannojirimycin (DMJ, 1) has been
isolated from the seeds of Lonchocarpus sericeus1 and L.
costaricensis (Leguminosae)2 as well as from cultures of
Streptomycese lavendulae.3 Like a number of other naturally
occurring polyhydroxylated piperidines,4 this compound
displays various rather important biological properties. In
particular, it is a potent inhibitor of several α-mannosidases.5
Interestingly, derivatisation of DMJ can result in significant
enhancement of its enzyme inhibiting properties. For example,
one of us has recently reported6 the isolation of 6-O-α--
rhamnopyranosyl-DMJ (Rha-DMJ, 2), as well as DMJ itself,6b
from the bark of Anglyocalyx pynaertii (Leguminosae) and
shown that this first naturally occurring glycoside of DMJ is
a significantly more potent inhibitor of α--fucosidase than
parent 1. Given the therapeutic potential arising from a
capacity to control glycosylation,7 DMJ and its derivatives have
attracted attention as agents for the treatment of, inter alia,
cancer, diabetes and viral infections.7,8 As a result various
ingenious synthetic routes to polyhydroxylated piperidines,
including DMJ, have been developed.4,9 One of the most
elegant and effective methods has been described by Hudlicky
and co-workers10 who were able to convert the cis-1,2-dihydro-
catechol 3 into various endo-nitrogenous aza sugars including
(ϩ)-kifunensine and mannojirimycin. A great attraction of this
approach is that compound 311 is available in large quantity
and enantiomerically pure form via microbial oxidation of
chlorobenzene and it can be readily elaborated to various
differentially protected and, therefore, synthetically valuable
forms of several aza sugars. Here we report on the application
of this approach to the preparation of DMJ and a differentially
protected derivative that has been exploited in the (first) total
synthesis of 6-O-α--rhamnopyranosyl-DMJ (2). Since various
disaccharide derivatives of DMJ have been identified as poten-
tial cancer remedies12 the present work should provide useful
methods for accessing members of this interesting class of
pseudo-trisaccharide.
compound with lithium azide resulted in an SN2-type displace-
ment and formation of the previously reported10 cis-azido-
alcohol 6 (91%, mp 93–94 ЊC) which was O-benzylated under
standard conditions to give the expected ether 7† {100%, [α]D
Ϫ111.0 (c 0.8)‡}. Ozonolytic cleavage of a methanolic solution
of this last compound followed by reductive work-up with
sodium borohydride and subsequent protection of the ensuing
1Њ-alcohol using TBS-Cl then afforded the azido-ether 8 {89%,
[α]D Ϫ10.2 (c 1.3)}. Two-fold hydrogenolysis of this last
compound using 5% Pd on C as catalyst resulted in formation
of lactam 913 {86%, mp 110–111 ЊC, [α]D ϩ15.3 (c 1.0)} which
was converted into the corresponding piperidine 1013 {73%, mp
86–87 ЊC, [α]D Ϫ66 (c 0.6)} upon exposure to the borane–
dimethyl sulfide complex followed by cleavage of the resulting
borane–amine complex with methanol in the presence of 10%
palladium on carbon.14 Treatment of compound 10 with 80%
v/v trifluoroacetic acid (TFA)–H2O at 18 ЊC for 18 h then gave
the TFA salt, 1ؒTFA, of DMJ. The spectral and physical data
derived from this material proved an excellent match with those
obtained on an authentic sample.
The differentially protected DMJ derivative 10 served as
an effective precursor to 6-O-α--rhamnopyranosyl-DMJ
(2) (Scheme 2). Thus, the former compound was N- and O-
benzylated under standard conditions and the TBS-ether
moiety then cleaved with tetra-n-butylammonium fluoride
(TBAF) to give the 1Њ-alcohol 1115 {89% from 10, [α]D ϩ10.5
(c 1.0)} which could be coupled with the -rhamnopyranosyl
bromide 1216 in the presence of silver triflate.17 Coupling
product 13 {87%, [α]D Ϫ42.5 (c 1.0)} was deprotected in a step-
wise fashion by sequential exposure to TFA (resulting in
removal of the acetonide group), methanolic sodium methoxide
(to remove the acetate groups) and dihydrogen in the presence
of 5% Pd on C. The sample of Rha-DMJ (2) {80% from 13, [α]D
Ϫ51.7 (c 0.6, H2O)} thus obtained after this final deprotection
step gave spectral data in full accord with the assigned structure
and in excellent agreement with those derived from the natural
material.6a
The synthesis of DMJ from cis-1,2-dihydrocatechol 3 is
shown in Scheme 1 and follows the early steps of Hudlicky’s
mannojirimycin synthesis. Thus, the readily available acetonide
derivative10 of diol 3 was subject to epoxidation with m-CPBA
and the resulting epoxide 410 (81% from 3) converted, in a
completely regioselective fashion, into the chlorohydrin 5 (98%)
upon reaction with lithium chloride. Treatment of the last
Synthetic 2 was evaluated for its glycosidase inhibitory
activity (Table 1) and shown to be 5–10 times less active than
T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 3
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 2 0 3 5 – 2 0 3 7
2035

View Article Online
Scheme 1 Reagents and conditions: (i) 2,2-DMP, p-TsOHؒH2O (cat.), 18 ЊC, 1 h; (ii) m-CPBA (1.0 mole equiv.), CH2Cl2, 0–18 ЊC, 11 h; (iii) LiCl
(5 mole equiv.), HOAc (3 mole equiv.), THF, 18 ЊC, 17 h; (iv) LiN3 (3 mole equiv.), DMF, 18 ЊC, 72 h; (v) BnBr (3.0 mole equiv.), KI (0.04 mole
equiv.), NaH (1.6 mole equiv.), THF, 0–18 ЊC, 24 h; (vi) O3 (excess), pyridine (4.9 mole equiv.), MeOH, Ϫ78 ЊC, 1 h then NaBH4 (11 mole equiv.),
Ϫ10 ЊC, 3 h; (vii) TBS-Cl (1.5 mole equiv.), imidazole (5 mole equiv.), CH2Cl2, 18 ЊC, 2 h; (viii) H2 (1 atm), 5% Pd on C (cat.), EtOAc, 18 ЊC, 36 h; (ix)
BH3ؒDMS (10 mole equiv.), THF, 18 ЊC, 4.5 h then 10% Pd on C, MeOH, 18 ЊC, 38 h; (x) 80% v/v TFA–H2O, 18 ЊC, 20 h.
Table 1 Comparison of the glycosidase inhibitory properties of
synthetic and natural samples of compound 2a
IC50/µM
Enzyme
Synthetic
Natural
ꢀ-Fucosidase
Bovine epididymis
2.8
23
28
0.5
4.6
3.2
Human placenta
Rat epididymis
ꢀ-Mannosidase
Rat epididymis
ꢀ-L-Rhamnosidase
Penicillium decumbens
>1000
>1000
460
740
a The enzymes α--fucosidases from bovine epididymis and human
placenta, and α--rhamnosidase from Penicillium decumbens were
purchased from Sigma Chemical Co. The rat epididymal fluid prepared
from rat epididymis according to the method of Skudlarek et al.18 was
used as the enzyme source of rat epididymis glycosidases. The glyco-
sidase activities were determined using an appropriate p-nitrophenyl
glycoside as substrate at the optimum pH of each enzyme. The reaction
was stopped by adding 400 mM Na2CO3. The released p-nitrophenol
was measured spectrometrically at 400 nm.
Scheme 2 Reagents and conditions: (i) BnBr (4.3 mole equiv.), KI
(0.5 mole equiv.), NaH (6.8 mole equiv.), THF, 18 ЊC, 24 h; (ii) TBAF
(1.2 mole equiv.), THF, 18 ЊC, 2 h; (iii) AgOTf (1.2 mole equiv,), 4 Å
molecular sieves, CH2Cl2, Ϫ10 ЊC, 0.5 h; (iv) 1 : 5 v/v TFA–CH2Cl2, 18
ЊC, 24 h; (v) NaOMe (0.2 M solution in MeOH), 18 ЊC, 2 h; (vi) H2
(1 atm), 5% Pd on C (cat.), EtOH, H2O (trace), 18 ЊC, 48 h.
yellow oil thus obtained was subjected to flash chromatography
(silica, 25–35% v/v ethyl acetate–hexane gradient elution) and
concentration of the appropriate fractions (Rf 0.5 in 4 : 2.5 : 5.5
v/v/v ethyl acetate–CH2Cl2–hexane) afforded glycoside 13
(377 mg, 87%) as a white foam (Found: Mϩ , 655.2971.
ؒ
the natural product. These results call into question the
previously reported6 glycosidase inhibitory activity of the
naturally occurring 2 which may have been compromised by
contamination of the assay sample with a more powerful
inhibitor like β--homofuconojirimycin. Despite this, com-
pound 2 must still be regarded as a rather potent inhibitor
of bovine epididymis-derived α-fucosidase.
C35H45NO11 requires Mϩ , 655.2993). νmax (NaCl)/cmϪ1 2985,
ؒ
2936, 1750, 1371, 1223, 1053, 735, 699; δH (300 MHz, CDCl3)
7.40–7.10 (10 H, complex m), 5.32–5.18 (3 H, complex m), 5.04
(1 H, t, J 9.8 Hz), 4.72 (1 H, d, J 11.7 Hz), 4.64 (1 H, broad s),
4.63 (1 H, d, J 11.7 Hz, partially obscured), 4.30 (2 H, broad s),
4.04–3.90 (2 H, complex m), 3.81 (2 H, m), 3.62 (1 H, d, J 14.0
Hz), 3.54 (1 H, dd, J 10.0 and 4.8 Hz), 2.88 (1 H, m), 2.74 (1 H,
broad d, J 14.0 Hz), 2.13 (3 H, s), 2.04 (3 H, s), 1.99 (3 H, s),
1.55 (3 H, s), 1.34 (3 H, s), 1.14 (3 H, d, J 6.3 Hz); δC (75 MHz,
CDCl3) 169.9, 169.8, 169.7, 139.0, 138.2, 128.4, 128.3, 128.2,
127.8, 127.6, 126.8, 109.0, 97.5, 76.5, 75.2, 72.8, 72.1, 71.0, 69.9,
69.2, 66.7, 66.5, 60.8, 58.7, 49.9, 27.4, 25.5, 21.1, 21.0, 20.9,
Experimental
Compound 13
A solution of silver triflate (203 mg, 0.79 mmol) in toluene
(5 mL) was added dropwise to a magnetically stirred solution
of alcohol 11 (253 mg, 0.66 mmol) and bromide 12 (297 mg,
0.84 mmol) in CH2Cl2 (20 mL) containing activated 4 Å
molecular sieves (2 g) and maintained under a nitrogen
atmosphere at Ϫ10 ЊC. After 0.5 h the reaction mixture was
treated with Et3N (1.5 mL) then CH2Cl2 (50 mL). The ensuing
mixture was filtered through a 2 cm deep pad of Celite and
the filtrate concentrated under reduced pressure. The light-
17.5; m/z (EI) 655 (<1%, Mϩ ), 640 (5), 352 (100), 91 (60).
ؒ
Acknowledgements
Dr Gregg Whited and Professor Tomas Hudlicky are warmly
thanked for providing generous quantities of the diol 3. XM
is the grateful recipient of an Australian Postgraduate Award
(Industry).
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 2 0 3 5 – 2 0 3 7
2036

View Article Online
9 For a comprehensive listing of previous syntheses of DMJ see (a)
Notes and references
J. Spreitz, A. E. Stütz and T. M. Wrodnigg, Carbohydr. Res., 2002,
337, 183; (b) O. V. Singh and H. Han, Tetrahedron Lett., 2003, 44,
2387.
† All new and stable compounds had spectroscopic data (IR, NMR,
mass spectrum) consistent with the assigned structure. Satisfactory
combustion and/or high-resolution mass spectral analytical data were
obtained for new compounds and/or suitable derivatives.
‡ Unless specified otherwise, all optical rotations were determined in
chloroform solution at 18–26 ЊC.
10 T. Hudlicky, J. Rouden, H. Luna and S. Allen, J. Am. Chem. Soc.,
1994, 116, 5099 . Closely related methodology has been described
by Johnson et al. (C. R. Johnson, A. Golebiowski, H. Sunddram,
M. W. Miller and R. L. Dwaihy, Tetrahedron Lett., 1995, 36,
653).
1 L. E. Fellows, E. A. Bell, D. G. Lynn, F. Pilkiewicz, I. Miura and
K. Nakanishi, J. Chem. Soc., Chem. Commun., 1979, 977.
2 A. F. Magalhaes, C. C. Santos, E. G. Magalhaes and M. A.
Nogueira, Phytochem. Anal., 2002, 13, 215.
3 Y. Ezure, N. Ojima, K. Konno, K. Miyazaki, N. Yamada,
M. Sugiyama, M. Itoh and T. Nakamura, J. Antibiot., 1988, 41,
1142.
11 For an excellent review on the production and general synthetic
utility of these types of compounds see T. Hudlicky, D. Gonzalez
and D. T. Gibson, Aldrichimica Acta, 1999, 32, 35.
12 (a) T. Ohgi, Y. Kyotani, H. Ogawa and N. Taniguchi, PCT Int. Appl.,
1997, WO 97 31,928 A1 19970904 (Chem. Abstr., 1997, 127,
234558f ); (b) H. Furui, K. Ando-Furui, H. Inagaki, T. Ando,
H. Ishida and M. Kiso, J. Carbohydr. Chem., 2001, 20, 789.
13 G. W. J. Fleet, N. G. Ramsden and D. R. Witty, Tetrahedron, 1989,
45, 319.
4 H. Yoda, Curr. Org. Chem., 2002, 6, 223.
5 A. D. Elbein and R. J. Molyneux, Iminosugars as Glycosidase
Inhibitors, in Inhibitors of Glycoprotein Processing, ed. A. E. Stütz,
Wiley-VCH, Weinheim, 1999, pp. 216–251.
14 M. Couturier, J. L. Tucker, B. M. Andresen, P. Dubé and J. T. Negri,
Org. Lett., 2001, 3, 465.
6 (a) N. Asano, K. Yasuda, H. Kizu, A. Kato, J.-Q. Fan, R. J. Nash,
G. W. J. Fleet and R. J. Molyneux, Eur. J. Biochem., 2001, 268, 35;
(b) K. Yasuda, H. Kizu, T. Yamashita, Y. Kameda, A. Kato, R. J.
Nash, G. W. J. Fleet, R. J. Molyneux and N. Asano, J. Nat. Prod.,
2002, 65, 198.
7 R. A. Dwek, T. D. Butters, F. M. Platt and N. Zitzmann, Nat. Rev.
Drug Discovery, 2002, 1, 65.
15 G. W. J. Fleet, N. G. Ramsden, R. J. Nash, L. E. Fellows, G. S. Jacob,
R. J. Molyneux, I. C. di Bello and B. Winchester, Carbohydr. Res.,
1990, 205, 269.
16 G. M. Bebault, G. G. S. Dutton and C. K. Warfield, Carbohydr. Res.,
1974, 34, 174.
17 H. Yamada, H. Takimoto, T. Ikeda, H. Tsukamoto, T. Harada and
T. Takahashi, Synlett, 2001, 1751 and references cited therein.
18 M. D. Skudlarek, D. R. Tulsiani and M. C. Orgebin-Crist, Biochem.
J., 1992, 286, 907.
8 L. Cipolla, A. Palma, B. La Ferla and F. Nicotra, J. Chem. Soc.,
Perkin Trans. 1, 2002, 2161 and references cited therein.
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 2 0 3 5 – 2 0 3 7
2037