Noeuromycin,1 a glycosyl cation mimic that strongly inhibits glycosidases

Full text of DOI:10.1021/ja010240u

5116
J. Am. Chem. Soc. 2001, 123, 5116-5117
Noeuromycin,1 A Glycosyl Cation Mimic that
Strongly Inhibits Glycosidases
Huizhen Liu, Xifu Liang, Helmer Søhoel, Anne Bu¨low, and
Mikael Bols*
Department of Chemistry, Aarhus UniVersity
DK-8000 Aarhus C, Denmark
ReceiVed January 29, 2001
Interest in design and synthesis of inhibitors of glycoside
cleavage has been intense during the last years.2-4 Such inhibitors
are not only useful as potential drugs against a number of diseases
such as diabetes or influenza but can also provide new insight in
the widespread and important glycoside cleavage/formation
process.
Some time ago it was found that a subtle change in the classical
glycosidase inhibitor 1-deoxynojirimycin (1a, Figure 1), by mov-
ing the nitrogen to the anomeric position to form isofagomine
(2a), gave a much more potent â-glucosidase inhibitor.5 This ap-
pears to be general, and some of the strongest â-glycosidase inhib-
itors have been found among the isofagomines (such as 3a and
4a)6-10 (see Figure 1). It is obvious that protonated 2a mimics
the cation A, an intermediate in the glycoside cleavage process,
and this has been suggested to be the basis of its potent inhibition
particularly if the transition state resembles A. Interestingly the
transition state is not believed to resemble A but rather a resonance
form of A the oxocarbenium ion B according to the Phillips
mechanism for lysozyme action. Furthermore, recent work on
acid-catalyzed glycoside hydrolysis concludes that an oxocarbe-
nium ion transition state (like B) is much more important than
charge development on anomeric carbon (like A).11 These
apparent inconsistencies are worth investigating.
Figure 1. The chemical structures of 1-deoxynojirimycin (1a), nojiri-
mycin (1b), isofagomine (2a), noeuromycin (2b), analogues 3a-4b, and
the resonance forms A and B of the glucosyl cation.
Scheme 1. Synthesis of Noeuromycin and Analoguesa
It may be argued that isofagomine (2) is not a perfect mimic
of A as it lacks the 2-hydroxyl group, which was omitted in the
design because of the reported instability of hemiaminals,12 and
attempts to mimic the 2-hydroxyl group met limited succes.13,14
However a recent study, where the thermodynamic functions of
binding between 1a or 2a and â-glucosidase were examined,
suggested that the 2-OH of 1a contributed significantly to binding
enthalpy, while a similar interaction was lacking for 2a.15 This
accords with the 2-OH being crucial for binding of both the
inhibitor 1a and the substrate to many glycosidases.16 It was
therefore decided to attempt to synthesize an analogue of 2a where
the 2-hydroxyl group was present.
Our synthesis started from the known17 2-hydroxymethylglu-
cose derivative 5 (Scheme 1). This compound was subjected to
(1) We propose the name noeuromycin to 2b due to its resemblance to
nojirimycin 1b and to its discovery on the day of the Danish Euro Referendum
(September 28, 2000).
(2) Stu¨tz, A. E. Iminosugars As Glycosidase Inhibitors: Nojirimycin and
Beyond; Wiley-VCH: Weinheim, 1999.
(3) Heightman, T. D.; Vasella, A. T. Angew. Chem., Int. Ed. Engl. 1999,
38, 750-770.
(4) Zechel, D. L.; Withers, S. G. Acc. Chem. Res. 2000, 33, 11-18.
(5) Jespersen, T. M.; Dong, W.; Sierks, M. R.; Skrydstrup, T.; Lundt, I.;
Bols, M. Angew. Chem., Int. Ed. Engl. 1994, 33, 1778-9.
(6) Ichikawa, Y.; Igarashi, Y. Tetrahedron Lett. 1995, 36, 4585-6.
(7) Bols, M. Acc. Chem. Res. 1998, 31, 1-8.
a i) CH2dCHCH2NH2, NaCNBH3, AcOH, 25 °C, 83%. ii) (Ph3P)3RhCl,
MeCN/H2O, reflux followed by TFA/H2O, 25 °C. iii) (t-BuOCO)2O,
NaHCO3, Me2CO/H2O, 25 °C, 87% from 6. iv) NaIO4. v) H2 (1 atm.),
Pd/C, MeCOOH, 25 °C. vi) TFA/H2O, 25 °C, R:â 1:2, 55% from 8. vii)
CH2dCHCH2NH2, NaCNBH3, AcOH, 25 °C, 57%. viii) (Ph3P)3RhCl,
MeCN/H2O, reflux. ix) (t-BuOCO)2O, NaHCO3, Me2CO/H2O, 25 °C, 64%
from 12. x) TEMPO, MCPBA, 41%. xi) HCl, H2O, R:â:pyranose ≈ 6:0:
1, 99%. xii) CH2dCHCH2NH2, NaCNBH3, AcOH, 25 °C, 86%. xiii)
(Ph3P)3RhCl, MeCN/H2O, reflux. xiv) (t-BuOCO)2O, NaHCO3, MeCN,
25 °C, 86%. xv) TEMPO, MCPBA, Bu4NBr, CH2Cl2, 25 °C, 99%. xvi)
HCl, H2O, R:â ≈ 1:0, 99%.
(8) Ichikawa, Y. Igarashi, Y. Ichikawa, M. Suhura, Y. J. Am. Chem. Soc.
1998, 120, 5854.
(9) Williams, S. J. Hoos, R. Withers, S. G. J. Am. Chem. Soc. 2000, 122,
2223-5.
(10) Nishimura, Y.; Shitara, E.; Adachi, H.; Toyoshima, M.; Nakajima,
M.; Okami, Y.; Takeuchi, T. J. Org. Chem. 2000, 65, 2-11.
(11) Namchuk, M. N., McCarter, J. D.; Becalski, A.; Andrews, T.; Withers,
S. G. J. Am. Chem. Soc. 2000, 122, 1270-7.
(12) Paulsen, H. Liebigs Ann. Chim. 1965, 683, 187-198.
(13) Schuster, M. B. Bioorg. Med. Chem. Lett. 1999, 9, 615-618.
(14) Williams, S. J.; Notenboom. V.; Wicki, J.; Rose, D. R.; Withers, S.
G. J. Am. Chem. Soc. 2000, 122, 4229-4230.
reductive amination with allylamine to 6, deallylation using
Wilkinson’s catalyst followed by acidic hydrolysis to free amine
7 and Boc-protection to give the partially protected aminoglucitol
derivative 8. The glycol of 8 was subjected to periodate cleavage
(15) Bu¨low, A.; Plesner, I.; Bols M. J. Am. Chem. Soc. 2000, 122, 8567-
8568.
(16) Panday, N.; Meyyappan, M.; Vasella, A. HelV. Chim. Acta 2000, 83,
513-538.
10.1021/ja010240u CCC: $20.00 © 2001 American Chemical Society
Published on Web 05/03/2001

Communications to the Editor
J. Am. Chem. Soc., Vol. 123, No. 21, 2001 5117
Table 1. Inhibition Constants Ki in nM for Binding of Inhibitors
1-2, 9, and 10 to Various Enzymesa
Table 2. Inhibition Constants Ki in nM for Binding of 3-4, 19,
and 20 to Various Enzymesa
enzyme
1ad
1bd
2ad
2b
9d
10d
enzyme
3a
3b
19
4a
4b
20
29j
R-glucosidaseb
12600
11000
47000
6300
-
890
86000
7200
110
22
25
69
59000
-
100
180e
-
R-fucosidaseb
-
-
-
-
4000
6400k
-
4.7
3.2
isomaltaseb
R-fucosidasec
-
50000
328
-
742
91
1.3j
â-glucosidasec
200000e
R-galactosidased
â-galactosidasee
â-galactosidasef
â-galactosidaseg
1.6j
-
-
81000
-
-
-
-
-
-
-
-
a -: inhibition not measured. b From yeast. c From almonds. d From
4h
35
397
-
ref 2. e IC50 value.
200i
12500j
-
a -: Inhibition not measured. b From bovine kidney. c From human
placenta. d From Green Coffee Bean. e From Saccharomyces Fragilis.
f From Aspargillus Oryzae. g From E. Coli. h From ref 6. i Measured
on the racemic compound, from ref 21. j From ref 2. k From ref 20.
followed by deprotection to give noeuromycin1 2b (R:â 1:2).18 It
was isolated as a hydrotrifluoroacetate and could as such be kept
in solution for days without apparent decomposition.
Noeuromycin (2b) was tested for inhibition of glycosidases
(Table 1) and was found to be a remarkably strong glucosidase
inhibitor. The Ki values were all in the nanomolar range and
between 2 and 4000 times smaller than those of 2a. Evidently
the incorporation of the 2-hydroxyl group increases inhibition
profoundly; therefore, the inhibition of glucosidases is presumably
caused mainly by the equatorial â-anomer. The inhibitor 2b was
also considerably more potent against glucosidases than the
inhibitors 1a, 1b, 9, and 10 resembling oxocarbenium ion
intermediate B (Table 1). It is also remarkable that in contrast to
2a the inhibitor 2b inhibits both â- and R-glucosidases strongly.
The study was extended to synthesis of the D-galacto- and
L-fuco-isomers of 2b (Scheme 1). The galacto-isomer was
obtained from the known derivative 11.19 Reductive amination
with allylamine to 12, deallylation with Wilkinsons catalyst to
13 and Boc-protection gave 14. Alcohol 14 was oxidized with
TEMPO and MCPBA to the lactol, which was finally deprotected
to the D-galacto-noeuromycin 3b (R:â: pyranose ≈ 6:0:1, Scheme
1).18 By a similar sequence (Scheme 1) the known derivative 15,20
was converted to L-fuco-noeuromycin 4b (R:â ≈ 1:0)18 in good
overall yield. Both compounds were only in R-form, which has
the 2-hydroxyl group equatorial. In the case of 3b a small amount
of a pyranose form was present.
Figure 2. The chemical structure of the glycosidase inhibitors 9, 10,
19, and 20.
3b and 4b were, like 2b, glycosidase inhibitors in the nano-
molar range (Table 2). While the galactose analogue 3b was 50
times more potent R-galactosidase inhibitor than isogalactofago-
mine 3a, its inhibition of three â-galactosidases varied from 4
times more potent to 9 times less potent than 3a. The contribution
of the 2-hydroxyl group to binding varies obviously between
â-galactosidases. The L-fucose analogue 4b was an R-fucosidase
inhibitor 1000 times more powerful than isofucofagomine 4a. The
comparison of 3b and 4b to the galacto and fuco analogues of
1-deoxynojirimycin, 19 and 20 (Figure 2) is also interesting.
Compound 20 is known to be an extremely potent R-fucosidase
inhibitor, but it is clear that 4b is equally potent (Table 2). On
the other hand compound 19 is much more potent than 3b versus
Figure 3. Proposed binding of 2b in the active site of a â-glucosidase
(C), in this case from white clover and a retaining R-glucosidase (D).
R-galactosidase, while the reverse is true for â-galactosidase much
like what has been observed for 3a6 but less extreme.
It is obvious that incorporation of a 2-hydroxyl group in 2a-
4a creates very tight binding inhibitors of both R- and â-glyco-
sidases. The increase in binding to R-glycosidases of 2b-4b is
particularly remarkable compared with 2a-4a. It shows that, con-
trary to previous beliefs, the anomeric nitrogen atom can interact
effectively with these enzymes as well. It is likely that a salt bridge
is formed between this group of 2b and the nucleophilic carboxyl-
ate of the enzyme as suggested for binding of 2a to â-glycosidases
(C, Figure 3).3,9 A similar interaction of 2b with the nucleophilic
carboxylate of an R-glycosidase can be imagined (D, Figure 3).
The present work shows that 1-azasugars are extremely potent
inhibitors of glycosidases. It also suggests that charge development
at the anomeric center is involved in the glycosidase-catalyzed
reaction. The very tight binding competitive inhibition observed
suggests that positive charge is present at the anomeric carbon at
the transition state or in an intermediate on the reaction trajectory
that is close to the transition state.
Acknowledgment. We thank the Danish Research Councils for finan-
cial support through the THOR program, Vinni Andreassen for prelimi-
nary experiments, and Professor A. E. Stu¨tz for kindly providing us a
sample of 19.
Supporting Information Available: Experimental procedures for the
preparation and characterization of 2b, 3b and 4b, and NMR spectra of
2b, 3b and 4b (PDF). This material is available free of charge via the
Internet at http://pubs.acs.org.
(17) Jespersen, T. M.; Bols, M.; Sierks, M. R.; Skrydstrup, T. Tetrahedron
1994, 50, 13449-13460.
(18) Compounds 2b, 3b, and 4b were isolated as hydrochlorides or
hydrotrifluoroacetates. NMR data of the noeuromycins (R is R-anomer, â is
â-anomer, p is pyranose form): 2b, 13C NMR (50 MHz, D2O): δ 162.9 (m,
CF3COO-), 115.7 (q, CF3COO-), 81.1 [C-2 (â)], 78.1 [C-2 (R)], 74.3 [C-3
(â)], 71.4 [C-3 (R)], 69.8 [C-4 (â)], 66.8 [C-4 (R)], 59.2 [C-6 (â)], 58.9 [C-6
(R)], 41.4 [C-5′ (â)], 41.0 [C-5′ (R)], 40.8 [C-5 (â)], 38.5 [C-5 (R)].1H NMR
(200 MHz, D2O): δ 4.96 [d, J2,3 3.0, H-2 (â)], 4.30 [d, J2,3 9.0, H-2(R)],
3.6-2.6 (m, H-3, H-4, H-5′a, H-5′b, H-6a, H-6b), 2.1-1.5 (m, H-5). 3b, 13
C
NMR (50 MHz, D2O): δ 81.2 (C-2), 73.9 (C-3), 70.1 (C-4), 62.4 (C-6), 42.0
1
(C-5), 41.8 (C-5′). H NMR (400 MHz, D2O): δ 5.08 [d, J1,2 2.4, H-1 (p)],
4.76-4.82 [H-2 (R)], 4.15 [bs, H-4 (R)], 3.78 [bs, H-3 (p)], 3.61-3.71 [m,
H-3 (R), H-6 (R), H-2 (p), H-5a (p), H-5b (p)], 3.54 [dd, J5,6a 7.2, J6a,6b 11
Hz, H-6a (R)], 3.23 [dd, J5,5′eq 3.8, J5′ax,5′eq 13 Hz, H-5′eq (R)], 3.06-3.14 [m,
H-4′a (p)], 3.03 [t, J5,5′ax 13.0 Hz, H-5′ax (R)], 2.86-2.96 [m, H-4′b (p)],
2.19-2.28 [m, H-4 (p)], 2.05-2.16 (m, H-5 (R)]. 4b, 13C NMR (50 MHz,
D2O): δ 77.7 (C-2), 71.0 (C-3), 69.8 (C-4), 42.3 (C-5′), 31.3 (C-5), 12.7
(C-6). 1H NMR (200 MHz, D2O): δ 4.70 (d, J2,3 9.1, H-2), 3.97 (br s, H-4),
3.68 (dd, J3,4 2.5 Hz, H-3), 3.09 (dd, J5,5′eq 5.1, J5′ax,5′eq 12.8 Hz, H-5′eq), 2.95
(t, J5,5′ax 12.8 Hz, H-5′ax), 2.06 (m, H-5), 0.98 (d, J5,6 7.3 Hz, CH3).
(19) Søhoel, H.; Liang, X.; Bols, M. Synlett 2000, 347-348.
(20) Hansen, A.; Tagmose, T. M.; Bols, M. Tetrahedron 1997, 53, 697-
706.
(21) Jensen, H. H.; Bols, M. J. Chem. Soc., Perkin Trans. 1 2001, 905-
909.
JA010240U