Attempted synthesis of 2-acetamido and 2-amino derivatives of salacinol. Ring opening reactions

Article abstract of DOI:10.1021/jo060167w

The attempted synthesis of the 2-acetamido and 2-amino derivatives of salacinol, a naturally occurring glycosidase inhibitor, is described. Reaction of the protected acetamidothioarabinitol unit with the cyclic sulfate derived from L-erythritol gave the c

Full text of DOI:10.1021/jo060167w

the Food and Drug Administration for use in the treatment of
type II diabetes.11,12 Swainsonine (4), an inhibitor of Golgi
R-mannosidase II (GMII), a key enzyme in the N-glycosylation
pathway, reduces tumor cell metastasis, enhances cellular
immune responses, and slows tumor cell growth.13
Attempted Synthesis of 2-Acetamido and
2-Amino Derivatives of Salacinol. Ring Opening
Reactions
Niloufar Choubdar and B. Mario Pinto*
Department of Chemistry, Simon Fraser UniVersity, Burnaby,
B.C., Canada V5A 1S6
bpinto@sfu.ca
ReceiVed January 24, 2006
Salacinol (5) and kotalanol (6) are two naturally occurring
glycosidase inhibitors which have been extracted from the roots
and stems of the Sri Lankan plant, Salacia reticulata.14-17 The
extracts from this plant have been traditionally used for the
treatment of the type II diabetes in Sri Lanka and India.18 The
permanent positive charge of the sulfur atom on salacinol and
kotalanol and the shape of the ring are thought to mimic the
oxacarbenium ion transition state in glycosidase-mediated
hydrolysis reactions.19
The attempted synthesis of the 2-acetamido and 2-amino
derivatives of salacinol, a naturally occurring glycosidase
inhibitor, is described. Reaction of the protected acetami-
dothioarabinitol unit with the cyclic sulfate derived from
L-erythritol gave the corresponding sulfonium sulfate, which
underwent ring opening to give an acyclic amido sulfate.
The corresponding reaction of the protected azidothioara-
binitol unit with the cyclic sulfate proceeded to give the
sulfonium sulfate. However, upon reduction of the azido
function to an amine it formed an acyclic ammonium sulfate.
We and others have reported the synthesis of salacinol and
its diastereomers,6,20-26 the nitrogen and selenium analogues
(10) Pistia-Brueggeman, G.; Hollingsworth, R. I. Tetrahedron 2001, 57,
8773-8778.
Glycosidase enzymes are involved in many important bio-
logical processes such as digestion, the biosynthesis of glyco-
proteins, and the catabolism of glycoconjugates.1-5 Inhibition
of glycosidases has potential as a means of therapy for diseases
such as type II diabetes, cancer, and infectious diseases.6 For
example, acarbose (1) and 1-deoxynojirimycin (DNJ, 2) are two
naturally occurring glycosidase inhibitors, of which 1 has been
used for the oral treatment of type II diabetes.7-10 Miglitol (3),
a synthetic R-glycosidase inhibitor, has also been approved by
(11) Taylor, R. H.; Barker, H. M.; Bowey, E. A.; Canfield, J. E. Gut
1986, 27, 1471-1478.
(12) Scott, L. J.; Spencer, C. M. Drugs 2000, 59, 521-549.
(13) Goss, P. E.; Reid, C. L.; Bailey, D.; Dennis, J. W. Clin. Cancer
Res. 1997, 3, 1077-1086.
(14) Yoshikawa, M.; Murakami, T.; Shimada, H.; Matsuda, H.; Yama-
hara, J.; Tanabe, G.; Muraoka, O. Tetrahedron Lett. 1997, 38, 8367-8370.
(15) Yoshikawa, M.; Murakami, T.; Yashiro, K.; Matsuda, H. Chem.
Pharm. Bull. 1998, 46, 1339-1340.
(16) Houston, T. A.; Blanchfield, J. T. Mini-ReV. Med. Chem. 2003, 3,
669-678.
(17) Lillelund, V. H.; Jensen, H. H.; Liang, X. F.; Bols, M. Chem. ReV.
2002, 102, 515-553.
* To whom correspondence should be addressed. Phone: (604) 291-4152.
Fax: (604) 291-5424. http://www.sfu.ca/chemistry/faculty/pinto.htm.
(1) ComprehensiVe Natural Products Chemistry; Pinto, B. M., Barton,
D. H. R., Nakanishi, K., Meth-Cohn, O., Eds.; Elsevier: New York, 1999;
Vol. 3.
(2) McCarter, J. D.; Withers, S. G. Curr. Opin. Struct. Biol. 1994, 4,
885-892.
(3) Ly, H. D.; Withers, S. G. Annu. ReV. Biochem. 1999, 68, 487-522.
(4) Dwek, R. A. Chem. ReV. 1996, 96, 683-720.
(5) Varki, A.; Cumming, R.; Esko, J.; Freeze, H.; Hart, G.; Marth, J.
Essentials of Glycobiology; Cold Spring Harbor Laboratory Press: New
York, 1999.
(18) Matsuda, H.; Morikawa, T.; Yoshikawa, M. Pure Appl. Chem. 2002,
74, 1301-1308.
(19) Kuntz, D. A.; Ghavami, A.; Johnston, B. D.; Pinto, B. M.; Rose,
D. R. Tetrahedron: Asymmetry 2005, 16, 25-32.
(20) Yuasa, H.; Takada, J.; Hashimoto, H. Tetrahedron Lett. 2000, 41,
6615-6618.
(21) Yuasa, H.; Takada, J.; Hashimoto, H. Bioorg. Med. Chem. Lett.
2001, 11, 1137-1139.
(22) Yoshikawa, M.; Morikawa, T.; Matsuda, H.; Tanabe, G.; Muraoka,
O. Bioorg. Med. Chem. 2002, 10, 1547-1554.
(23) Ghavami, A.; Johnston, B. D.; Maddess, M. D.; Chinapoo, S. M.;
Jensen, M. T.; Svensson, B.; Pinto, B. M. Can. J. Chem.-ReV. Can. Chim.
2002, 80, 937-942.
(24) Ghavami, A.; Sadalapure, K. S.; Johnston, B. D.; Lobera, M.; Snider,
B. B.; Pinto, B. M. Synlett 2003, 1259-1262.
(25) Danquigny, A.; Benazza, M.; Protois, S.; Demailly, G. Tetrahedron
Lett. 2004, 45, 4365-4369.
(6) Ghavami, A.; Johnston, B. D.; Pinto, B. M. J. Org. Chem. 2001, 66,
2312-2317.
(7) Holman, R. R.; Cull, C. A.; Turner, R. C. Diabetes Care 1999, 22,
960-964.
(8) Jacob, G. S. Curr. Opin. Struct. Biol. 1995, 5, 605-611.
(9) Dong, W.; Jespersen, T.; Bols, M.; Skrydstrup, T.; Sierks, M. R.
Biochemistry 1996, 35, 2788-2795.
10.1021/jo060167w CCC: $33.50 © 2006 American Chemical Society
Published on Web 05/16/2006
J. Org. Chem. 2006, 71, 4671-4674
4671

34-38
of salacinol,27-33 chain-extended analogues,
and other
SCHEME 1
analogues.34-38 Some of these derivatives exhibited interesting
biological properties.34,35,39,40
It was of interest to investigate the effect of introducing an
acetamido or an amine function at the C-2 position of salacinol
to broaden the scope of glycosidase inhibitory activities, in
particular, to target the hexosaminidase enzymes. Therefore, we
report herein the attempted synthesis of the 2-acetamido- (7)
and 2-amino- (8) substituted analogues of salacinol. These
derivatives were found to undergo ring opening reactions by
nucleophilic participation of the amide or amine moieties to
give acyclic amido or ammonium sulfates, respectively.
to give 11, and the 2-OH group of compound 11 was then
mesylated (12). The substitution of the mesylate in 12 by an
azide group did not proceed in MeOH; however, at 75 °C in
DMF, the substitution proceeded with inversion of configuration
at C-2 (13). The azide group in compound 13 was then reduced
to an amine by using triphenylphosphine (PPh3) in dioxane:
MeOH:H2O (10:3:1) to afford compound 14 in 70% yield;
acetylation then afforded compound 15 in 77% yield. The
alkylation reaction of compound 15 with 2,4-O-benzylidene-L-
erythritol-1,3-cyclic sulfate (9) did not proceed as expected.
Instead, the silyl protecting group was removed under the
reaction conditions. Therefore, we chose to replace this group
by benzyl ethers. The silyl protecting group in 15 was thus
removed, using tetra-n-butylammonium fluoride (TBAF), and
the crude product obtained was then subjected to benzylation
by using benzyl bromide and a mixture of BaO and Ba(OH)2‚
8H2O to afford compound 16 in 78% yield. A 1D-NOESY
experiment performed on compound 16 confirmed the stereo-
chemistry at C-2 by showing a correlation between H-2 and
H-4 of the five-membered ring. The alkylation reaction between
compound 16 and 2,4-O-benzylidene-L-erythritol-1,3-cyclic
sulfate (9) gave compound 18c in 87% yield. A 1D-NOESY
experiment performed on the coupled product showed no
correlation between H-1′ and H-1 (as would be expected for
the target compound 17); the latter result also confirms that the
five-membered ring had been opened to give an acyclic
thioether. The 13C NMR spectrum of the coupled product
indicated the presence of a carbonyl group, and a g-HMBC
spectrum showed a correlation between this carbonyl carbon
and H-2, and no correlation between the carbonyl carbon and
H-1, thus suggesting 18c as the product of the reaction. The IR
spectrum of this product suggested the presence of an OH group
as well as an amide carbonyl group, further corroboration of
the structure of 18c. The high-resolution mass spectrum
confirmed the molecular formula of compound 18c. The
microanalysis of compound 18c also confirmed the presence
of potassium as the counterion (Scheme 2).
The target compounds 7 and 8 could be obtained by
hydrogenolysis of the coupled product A, which has an
acetamide or an azide group at C-2 (Scheme 1). Compound A
could be synthesized, in turn, by the alkylation of a 2-azido- or
2-acetamido-1,4-anhydro-4-thio-D-arabinitol derivative (B) with
2,4-O-benzylidene-L-erythritol-1,3-cyclic sulfate (9). Compound
B could, in turn, be synthesized from the reaction of sodium
azide with a selectively protected 1,4-anhydro-4-thio-D-ribose
(C) unit in which C-2 bears a good leaving group. Compound
C could be obtained from 1,4-anhydro-4-thio-D-ribose (10),
which could be synthesized, in turn, from commercially
available D-ribose.41
1,4-Anhydro-4-thio-D-ribose (10) was synthesized from com-
mercially available D-ribose in 11 steps according to a literature
procedure.41 To introduce an azide group selectively at C-2 of
compound 10, the 3-OH and 5-OH groups were first protected
by using 1,3-dichloro-1,1,3,3-tetraisopropyldisiloxane in pyridine
(26) Kumar, N. S.; Pinto, B. M. Carbohydr. Res. 2005, 340, 2612-
2619.
(27) Ghavami, A.; Johnston, B. D.; Jensen, M. T.; Svensson, B.; Pinto,
B. M. J. Am. Chem. Soc. 2001, 123, 6268-6271.
(28) Johnston, B. D.; Ghavami, A.; Jensen, M. T.; Svensson, B.; Pinto,
B. M. J. Am. Chem. Soc. 2002, 124, 8245-8250.
(29) Liu, H.; Pinto, B. M. J. Org. Chem. 2005, 70, 753-755.
(30) Liu, H.; Pinto, B. M. Can. J. Chem. 2006, in press.
(31) Liu, H.; Pinto, B. M. Can. J. Chem. 2005, in press.
(32) Muraoka, O.; Ying, S.; Yoshikai, K.; Matsuura, Y.; Yamada, E.;
Minematsu, T.; Tanabe, G.; Matsuda, H.; Yoshikawa, M. Chem. Pharm.
Bull. 2001, 49, 1503-1505.
(33) Gallienne, E.; Gefflaut, T.; Bolte, J.; Lemaire, M. J. Org. Chem.
2006, 71, 894-902.
(34) Johnston, B. D.; Jensen, H. H.; Pinto, B. M. J. Org. Chem. 2006,
71, 1111-1118.
Scheme 3 shows the proposed mechanism for the formation
of 18c. Nucleophilic attack of the amide oxygen on C-1 of
compound 17 would result in opening of the five-membered
ring to give (18a), which could then react with water during
the processing and purification to give 18b. Further rearrange-
ment of the orthoaminal (18b) would result in 18c.
(35) Liu, H.; Sim, L.; Rose, D. R.; Pinto, B. M. J. Org. Chem. Submitted
for publication.
(36) Muraoka, O.; Yoshikai, K.; Takahashi, H.; Minematsu, T.; Lu, G.
X.; Tanabe, G.; Wang, T.; Matsuda, H.; Yoshikawa, M. Bioorg. Med. Chem.
2006, 14, 500-509.
(37) Gallienne, E.; Benazza, M.; Demailly, G.; Bolte, J.; Lemaire, M.
Tetrahedron 2005, 61, 4557-4568.
(38) Szczepina, M. G.; Johnston, B. D.; Yuan, Y.; Svensson, B.; Pinto,
B. M. J. Am. Chem. Soc. 2004, 126, 12458-12469.
(39) Pinto, B. M.; Johnston, B. D.; Ghavami, A.; Szczepina, M. G.; Liu,
H.; Sadalapure, K. U.S. patent application, filed June 25, 2004.
(40) Li, Y. J.; Scott, C. R.; Chamoles, N. A.; Ghavami, A.; Pinto, B.
M.; Turecek, F.; Gelb, M. H. Clin. Chem. 2004, 50, 1785-1796.
(41) Naka, T.; Minakawa, N.; Abe, H.; Kaga, D.; Matsuda, A. J. Am.
Chem. Soc. 2000, 122, 7233-7243.
To prevent the intramolecular nucleophilic participation of
the amide group observed in compound 18a, we next examined
the alkylation reaction of a 2-azido arabinitol derivative (Scheme
4). Starting from compound 13, the silyl protecting group was
substituted by benzyl protecting groups (19). The alkylation
reaction of compound 19 with 2,4-O-benzylidene-L-erythritol-
4672 J. Org. Chem., Vol. 71, No. 12, 2006

SCHEME 2a
SCHEME 4a
a Reagents and conditions: (a) tetra-n-butylammonium fluoride (TBAF),
THF; (b) BnBr, NaH, DMF; (c) 2,4-O-benzylidene-L-erythritol-1,3-cyclic
sulfate, 1,1,1,3,3,3-hexafluoro-2-propanol (HFIP), 64 °C; (d) H2, Pd/C.
free amine (21a) formed during the hydrogenolysis reaction
would result in the formation of the aziridinium compound 21b.
Further reduction of compound 21b during the hydrogenolysis
reaction would then afford 21c.
a Reagents and conditions: (a) 1,3-dichloro-1,1,3,3-tetraisopropyldi-
siloxane, pyridine (TIPSCl2), rt; (b) methanesulfonyl chloride, pyridine; (c)
NaN3, DMF, 85 °C; (d) PPh3, dioxane:MeOH:H2O (10:3:1), 55 °C; (e)
Ac2O, pyridine; (f) tetra-n-butylammonium fluoride (TBAF), THF; (g) BnBr,
BaO (9.3 equiv), Ba(OH)2‚8H2O (1.75 equiv), DMF; (h) 2,4-O-benzylidene-
L-erythritol-1,3-cyclic sulfate, K2CO3, 1,1,1,3,3,3-hexafluoro-2-propanol
(HFIP), 60 °C.
SCHEME 5
SCHEME 3
In conclusion, the synthesis of the 2-acetamido and 2-amino
derivatives of salacinol was attempted. However, the 2-acet-
amido analogue underwent ring opening of the five-membered
ring by nucleophilic participation of the amide. In the case of
the 2-amino derivative, the amine group participated to open
the five-membered ring via formation of an aziridinium ion.
These are the first examples of which we are aware of the
instability of salacinol analogues. Our synthetic efforts to date
have not revealed any ring-opened, epoxide products stemming
from the participation of O-2 in salacinol analogues (Scheme
6a). Furthermore, enzyme inhibition studies with a variety of
glycosidase enzymes have been consistent with a model of
reversible, competitive inhibition,27-32 and have not involved
ring-opening reactions by nucleophilic groups in the enzyme
active sites (Scheme 6b).
1,3-cyclic sulfate (9) gave the coupled product 20 in 76% yield.
A 1D-NOESY experiment of the coupled product 20 confirmed
the formation of the sulfonium salt by showing a correlation
between H-1, H-1′, and H-4 in this product. Hydrogenolysis of
compound 20 afforded the acyclic thioether 21c in 53% yield,
and not the expected compound 21a (Scheme 4). A 1D-NOESY
experiment showed no correlation between H-1 and H-1′ as one
would have expected for compound 21a. g-HMBC data
confirmed that compound 21c was the likely product of the
reaction. Thus, no correlation between H-5′ and C-2′ or between
H-5′ and C-4 was observed. In addition, no correlation between
H-2′ and C-5′ or H-4 and C-5′ was observed. These results
confirm that the ring is opened. Correlations between H-2′ and
C-4 and also between H-4 and C-2′ (alternative numbering; see
Scheme 4) were observed, indicating that the side chain is on
C-2′. A DEPT experiment showed the existence of a CH3 group,
which corroborated the structure 21c, and the high-resolution
mass spectral data were also consistent with the molecular
formula of compound 21c.
SCHEME 6
Experimental Section
Potassium (2S,3S)-1,3-Benzylidenedioxy-4-[(2R,3S,4R)-4′-
acetamido-1′,3′-bisbenzyloxy-5′-hydroxy-2′-pentylthio]butane-
2-sulfate (18c). A mixture of compound 16 (66.6 mg, 0.18 mmol)
and 2,4-O-benzylidene-L-erythritol-1,3-cyclic sulfate (9) (63.5 mg,
Scheme 5 shows the proposed mechanism for the formation
of compound 21c from 21a. Nucleophilic participation of the
J. Org. Chem, Vol. 71, No. 12, 2006 4673

1.3 equiv) was dissolved in 1,1,1,3,3,3-hexafluoro-2-propanol
(HFIP) (1.5 mL), and K2CO3 (40 mg) was added. The mixture was
stirred in a sealed tube in an oil bath (60 °C) for 13 h. The solvent
was removed under reduced pressure, and flash chromatography
[CH2Cl2/MeOH, 10:1] of the crude product gave compound 18c
(100 mg, 87%) as an oil. [R]D -28 (c 0.03, MeOH); 1H NMR (CD3-
OD) δ 7.4-7.1 (15H, m, Ar), 5.46 (1H, s, CHPh), 4.58, 4.45 (2H,
2d, JA,B ) 10.9 Hz, CH2Ph), 4.51 (1H, ddd, J5′b,4′ ) 7.5 Hz, J5′a,4′
) 6.8 Hz, J3′,4′ ) 1.5 Hz, H-4′), 4.48 (1H, dd, J2,1 ) 5.4 Hz, H-1a),
carbon (100 mg) under H2 (90 psi). After 72 h, the reaction mixture
was filtered. The filtrate was concentrated and the residue was
purified by column chromatography [EtOAc/MeOH/H2O, 10:3:1]
to give compound 21c (50 mg, 53%) as a colorless oil. [R]D -0.05
1
(c 0.22, MeOH); H NMR (D2O) δ 4.23 (1H, ddd, J3,2 ) 6.9 Hz,
H-2), 3.93 (1H, ddd, H-3), 3.80 (1H, dd, J1b,1a ) 12.7 Hz, J2,1a
)
3.2 Hz, H-1a), 3.74 (1H, dd, J1′b,1′a ) 12 Hz, J2′,1′a ) 5.53 Hz,
H-1′a), 3.70 (1H, dd, H-1b), 3.67 (1H, dd, J2′,3′ ) 5.8 Hz, H-3′),
3.63 (1H, dd, J2′,1′b ) 5.9 Hz, H-1′b), 3.45 (1H, dt, J3′,4′ ) 6.1 Hz,
H-4′), 2.92 (1H, ddd, H-2′), 2.88 (1H, dd, J3,4a ) 3.8 Hz, H-4a),
2.64 (1H, dd, J4a,4b ) 14.0 Hz, J3,4b ) 8.2 Hz, H-4b), 1.15 (3H, d,
H-5′); 13C NMR (D2O) δ 81.2 (C-2), 73.7 (C-3′), 69.4 (C-3), 60.4
(C-1′), 59.8 (C-1), 49.8 (C-2′), 49.5 (C-4′), 34.3 (C-4), 16.1 (C-
5′). HRMS Calcd for C9H20NO8S2 (M - H) 334.06249, found
334.06248.
4.42, 4.39 (2H, 2d, JA,B ) 11.9 Hz, CH2Ph), 4.26 (1H, ddd, J3,2
)
9.8 Hz, H-2), 3.88 (1H, dd, J2′,3′ ) 8.7 Hz, H-3′), 3.84 (1H, ddd,
J4,3 ) 1.9 Hz, H-3), 3.78 (1H, dd, J1′b,1′a ) 9.8 Hz, J2′,1′a ) 4.1 Hz,
H-1′a), 3.70 (1H, dd, J2′,1′b ) 4.5 Hz, H-1′b), 3.68 (1H, dd, J2,1b
)
1.4 Hz, J1a,1b ) 10.8 Hz, H-1b), 3.6 (1H, dd, J5′b,5′a ) 10.7 Hz,
J4′,5′a ) 6.9 Hz, H-5′a), 3.4 (1H, dd, J4′,5′b ) 7.6 Hz, H-5′b), 3.2
(1H, ddd, H-2′), 3.16 (1H, dd, J4b,4a ) 15.3 Hz, J3,4a ) 1.9 Hz,
H-4a), 2.93 (1H, s, CH3), 2.74 (1H, dd, J3,4b ) 8.1 Hz, H-4b); 13
C
Acknowledgment. We thank the Natural Sciences and
Engineering Research Council of Canada for financial support
and B. D. Johnston for helpful discussion.
NMR (CD3OD) δ 172.9 (CO), 138.4, 138.3, 137.9, 128.7, 128.2,
128.1, 127.9, 127.8, 127.6, 127.5, 126.2 (18C, Ph), 101.3 (CHPh),
81.6 (C-3), 77.2 (C-3′), 74.4, 73.0 (2 × CH2Ph), 70.3 (C-1′), 69.9
(C-2), 69.0 (C-1), 61.8 (C-5′), 52.3 (C-4′), 47.9 (C-2′), 34.8 (C-4),
21.8 (CH3). Anal. Calcd for C32H38NO10S2: C, 54.91; H, 5.48; N,
2.00. Found: C, 54.89; H, 5.50; N, 2.10. HRMS Calcd for C32H38-
NO10S2 (M - H) 660.19426, found 660.19407.
Supporting Information Available: General experimental and
complete experimental details for the synthesis of 18c and 21c from
10 and 13, respectively, and 1H and 13C NMR spectra for compound
21c. This material is available free of charge via the Internet at
http://pubs.acs.org.
(2S,3S)-1,3-dihydroxy-4-[(2R,3S,4S)-4′-amino-1′,3′-dihydroxy-
2′-pentylthio]butane-2-sulfate (21c). The protected compound 20
(178 mg, 0.28 mmol) was dissolved in AcOH:H2O (4:1, 10 mL)
and the solution was stirred with palladium hydroxide catalyst on
JO060167W
4674 J. Org. Chem., Vol. 71, No. 12, 2006