β-Glycosidase inhibitors mimicking the pyranoside boat conformation

Article abstract of DOI:10.1039/b005578f

Polyhydroxylated isoquinuclidines mimicking the boat conformation of pyranosides are strong and selective inhibitors of a retaining β-mannosidase.

Full text of DOI:10.1039/b005578f

b-Glycosidase inhibitors mimicking the pyranoside boat conformation
Edwige Lorthiois, Muthuppalaniappan Meyyappan and Andrea Vasella*
Laboratorium für Organische Chemie, ETH-Zentrum, Universitätstrasse 16, CH-8092 Zürich, Switzerland.
E-mail: vasella@sugar.org.chem.ethz.ch
Received (in Cambridge, UK) 17th July 2000, Accepted 14th August 2000
First published as an Advance Article on the web 11th September 2000
Polyhydroxylated isoquinuclidines mimicking the boat con-
formation of pyranosides are strong and selective inhibitors
of a retaining b-mannosidase.
correct orientation of its lone pairs may possess a more strongly
pronounced character as transition state analogues than in-
hibitors mimicking the cationic reactive intermediate. Inhibitors
mimicking the shape of an oxycarbenium cation appear indeed
to be only partial transition state analogues.4 Since the
conformational change from a chair to a twist-boat or boat
should be induced by all b-glycosidases, we wanted to
synthesize and evaluate a mimic of such a conformer of a
glycoside that is cleaved by an exo-glycosidase. The cyclohex-
ane ring of 14 and 15 (Fig. 1) mimics the tetrahydropyran ring
of a b-mannoside in a 1,4B conformation; it is substituted by a
pseudoaxial amino group that should allow lateral protonation.5
Only one6 of the known bicyclic glycosidase inhibitors7
mimicks a tetrahydropran ring in a (2,5B) boat conformation. It
has been designed in another conceptual context, and was not
tested against b-mannosidases; it is a poor inhibitor of other
glycosidases.
According to the principle of stereoelectronic control,1 hetero-
lytic cleavage of an acetal C–O bond requires an antiperiplanar
orientation of a doubly occupied, non-bonding orbital. This
antiperiplanar lone pair hypothesis (ALPH) means that hydroly-
sis of b- -pyranosides involves a conformational change of the
D
tetrahydropyran ring from a chair to a twist-boat or boat
resulting in a pseudoaxial orientation of the aglycon (Fig. 1).2 A
The isoquinuclidines 14 and 15 were synthesised in 14 and 15
steps and in 6.6 and 5.4% overall yields, respectively, from
3-hydroxypyridone via the known tosylate 18 as shown in
Scheme 1. Notable features are the diastereoselective high
yielding Michael addition and aldolisation of 2 to 4, the
epimerisation and hydrolysis (dealkylation?) of the ester 6 to 7a,
and the two step reduction of 11a via 12 to 13.†
Fig. 1 Skew boat (1S3, a) and boat (1,4B, b) conformers of a b-
mannopyranoside and isoquinuclidines 14 and 15.
D-
similar conformational change is required for nucleophilic
assistance of the cleavage according to an intermediate SN1/SN2
mechanism. The relevance of ALPH for enzymic glycoside
cleavage has met with strong scepticism,2c but crystal structures
of three endo-glycosidases in complex with substrate analogues
show a skew boat or a flattened boat conformation of the
tetrahydropyran ring.3 A skew boat or boat-like conformer with
a pseudoaxial (elongated) C(1)–O bond may conceivably be
closer to the transition state of an enzymic b-glycoside cleavage
than a reactive intermediate of an oxycarbenium cation type. If
so, inhibitors mimicking the boat-like conformation of a
glycoside, the proper location of the ‘glycosidic heteroatom’
(i.e. the heteroatom attached to C(1) of the glycoside), and the
Both 14‡ (Ki = 0.17 mM; IC50 = 0.69 mM)§ and 15¶ (Ki =
20 mM; IC50
= 29.4 mM) inhibit snail b-mannosidase
competitively, as determined from a Lineweaver–Burk plot, 14
being about 120 times stronger than 15. Jack bean a-
mannosidase is inhibited ca. 104 times more weakly by 14 (IC50
= 9.6 mM) than snail b-mannosidase and about 700 times more
weakly by 15 (IC50 = 20 mM). Both 14 (IC50 > 5.0 mM) and
15 (IC50 = 4.3 mM) are poor inhibitors of b-glucosidase from
Caldocellum saccharolyticum. The value of IC50 for b-
mannosidase dropped from 4.08 to 0.69 mM for 14 and from 183
to 29.4 mM for 15 as the preincubation was prolonged from 10
Scheme 1 Reagents and conditions: i, BF3·Et2O, CH2Cl2, 98%; ii, CH2NCH–CO2Me, Et3N, 92% from 2; iii, CH2(OMe)2, P2O5, CHCl3, 88%; iv, Na–C10H8,
DME, 278 °C, 88% from 5; v, K2CO3, MeOH, D; vi, BnBr, NaHCO3, DMF, 8a/8b 80+20 (49% from 6), 9a/9b 80+20 (18% from 6); vii, NaH, BnBr, DMF,
89%; viii, OsO4, acetone–H2O, 89%; ix, Me2C(OMe)2, acetone, CSA, 94%; x, THF, D, 12 (57%), 13a (17%), 13b (8%); xi, LiAlH4, dioxane, D, 13a (67%);
xii, TFA, H2O, D, 84%; xiii, H2/Pd(OH)2/C, conc. HCl, MeOH/H2O, 82%.
DOI: 10.1039/b005578f
Chem. Commun., 2000, 1829–1830
This journal is © The Royal Society of Chemistry 2000
1829

irrad. at 1.85 ? d, J = 10.6, CHb–C(8)), 3.73 (dd, J ≈ 8.6, 1.8, irrad. at
2.65? d, J = 8.7, irrad. at 3.99 ? t, J = 1.9, H–C(5)), 3.62 (dd, J ≈ 11.1,
8.0, irrad. at 1.85 ? d, J = 10.6, –CHa–C(8)), 2.88, br. dd, J ≈ 11.2, 2.0,
irrad. at 1.85 ? d, J = 11.2, irrad. at 2.65 ? d, J ≈ 5.0, Ha–C(3)), 2.75 (br
q, J ≈ 2.4, irrad. at 1.52 ? br. t, J ≈ 2.5, irrad. at 1.85 ? t, J = 2.2, irrad.
at 3.99 ? change, H–C(1)); 2.65 (dd, J ≈ 11.4, 1.8, Hb–C(3)), 1.76–1.95
(m, irrad. at 1.52 ? change, irrad. at 2.75 ? change, Hendo–C(7), H–C(8)),
1.52 (dd, J = 7.8, 3.4, irrad. at 1.85 ? d, J = 2.5, irrad. at 2.75 ? d, J =
7.8, Hexo–C(7)). dC(75 MHz, D2O): 73.49 (d, C(6)), 73.35 (s, C(4)), 70.24
(d, C(5)), 65.00 (t, CH2–C(8)), 41.58 (t, C(3)), 40.53 (d, C(1)), 29.01 (t,
C(7)); MALDI-MS: 190 (100, [M + 1]+), 212 (10, [M + Na]+). Anal. calc.
for C8H15NO4·0.5 H2O: C 48.48, H 8.14, N 7.07%. Found: C 48.28, H 7.91,
N 6.77%.
min to 2 h. The inhibition by 14 and 15, as represented by
1/IC50, shows a linear dependence on pH, revealing inhibition
by the free amines rather than by the ammonium salts. The
inhibition is ca. 4–6 times stronger at pH 5.6 than at pH 4.5 and
again ca. 4–6 times weaker at pH 3.6.∑
Since 15 (pKHA = 8.4) is a stronger base than 14 (pKHA
=
7.5), one expects 14 to be a stronger inhibitor than 15 at pH 4.5
by a factor of about 10. That 14 is 120 times stronger than 15
evidences a hydrophobic interaction of the benzyl group of 14
with the aglycon binding site. Such interactions are well
precedented.10,11
The pH dependence of the inhibition by 14 and 15 evidences
the essential interaction with the catalytic acid, and confirms its
flexibility.12 In contrast to the inhibition by the azole type
inhibitors,4,13–15 that is characterised by a cooperative inter-
action of the inhibitor with the catalytic acid and the catalytic
nucleophile2b the interaction of 14 and 15 with the catalytic
nucleophile appears to play at best a minor role. The inhibitory
activity of 14 and 15 is in agreement with the postulate that a
conformational change of the pyranose ring precedes or
accompanies the enzymatic cleavage of b-glycosides.
∑ The enzyme loses ca. 50% activity at pH 5.5 and ca. 10% at pH 3.5.9
1 (a) C. L. Perrin, R. E. Engler and D. B. Young, J. Am. Chem. Soc., 2000,
122, 4877; (b) A. J. Kirby, The Anomeric Effect and Related
Stereoelectronic Effects at Oxygen, Springer–Verlag, Berlin, 1983; (c)
P. Deslongchamps, Stereoelectronic Effects in Organic Chemistry,
Pergamon Press, Oxford, 1983.
2 Recent reviews on glycosidase mechanisms: (a) D. L. Zechel and S. G.
Withers, Acc. Chem. Res., 2000, 33, 11; (b) T. D. Heightman and A. T.
Vasella, Angew. Chem., Int. Ed., 1999, 38, 750; (c) G. Davies, M. L.
Sinnott and S. G. Withers, in Comprehensive Biochemical Catalysis, ed.
M. Sinnott, Academic Press, London, 1998, 1, p. 119 and references
cited there.
Notes and references
† The esters 9a and 9b were isolated in yields of 41 and 10% from 5.
Selected H-NMR data for 8a: 4.00 (ddt, J = 5.6, 3.4, 2.1, H-C(1)), 2.93
(dd, J ≈ 10.5, 4.9, H-C(5)), 1.95 (ddd, J = 12.5, 5.0, 3.4, Hexo-C(6)), 1.79
(ddd, J = 12.5, 10.6, 2.2, Hendo-C(6)); selected 1H-NMR data for 8b: 4.05
(ddt, J ≈ 5.3, 3.6, 1.4, H-C(1)), 3.06 (ddd, J = 10.0, 5.1, 0.9, H-C(5)), 2.09
3 (a) G. J. Davies, L. Mackenzie, A. Varrot, M. Dauter, A. M.
Brzozowski, M. Schülein and S. G. Withers, Biochemistry, 1998, 37,
11707; (b) I. Tews, A. Perrakis, A. Oppenheim, Z. Dauter, K. S. Wilson
and C. E. Vorgias, Nat. Struct. Biol., 1996, 3, 638; (c) G. Sulzenbacher,
H. Driguez, B. Henrissat, M. Schülein and G. Davies, Biochemistry,
1996, 35, 15280.
1
(ddd, J = 12.8, 10.0, 3.7, Hexo-C(6)), 1.56 (ddd, J = 12.5, 5.0, 1.6, Hendo
-
C(6)). The exo configuration of the diols 10 (exo refers to the face syn to the
C(1)–N bond) was assigned on the basis that J1,6 = 2.1 Hz (10a and 10b)
4 P. Ermert, A. Vasella, M. Weber, K. Rupitz and S. G. Withers,
Carbohydr. Res., 1993, 250, 113.
is smaller than J1,7exo = 3.6 (10a) and 4.4 Hz (10b) and closer to J1,7endo
=
5 T. D. Heightman, M. Locatelli and A. Vasella, Helv. Chim. Acta, 1996,
79, 2190.
2.0 (10a) and 1.7 Hz (10b).8 An attempt to reduce 11a to 13 in one step was
not successful.
6 K. S. E. Tanaka and A. J. Bennet, Can. J. Chem., 1998, 76, 431.
7 (a) K. H. Smelt, Y. Bleriot, K. Biggadike, S. Lynn, A. L. Lane, D. J.
Watkin and G. W. J. Fleet, Tetrahedron Lett., 1999, 40, 3255; (b) K. H.
Smelt, A. J. Harrison, K. Biggadike, M. Müller, K. Prout, D. J. Watkin
and G. W. J. Fleet, Tetrahedron Lett., 1999, 40, 3259; (c) A. Stütz,
Iminosugars as Glycosidase Inhibitors: Nojirimycin and Beyond,
Wiley–VCH, Weinheim, 1999.
‡ Data for 14: Rf (AcOEt–MeOH 5+1), 0.40; dH(300 MHz, CD3OD):
7.40–7.13 (m, arom. H), 3.88 (dd, J = 12.1, 6.5, CH2OH), 3.85 (dd, J = 8.4,
2.1, irrad. at 2.66 ? d, J = 8.4, H-C(6), 3.81, 3.72 (2d, J = 13.1, N-CH2Ph),
3.65 (dd, J = 10.6, 6.5, CH2OH), 3.60 (dd, J = 8.4, 1.3, H-C(5)); 2.89 (dd,
J ≈ 9.5, 1.8, Hb-C(3)), 2.66 (br q, J ≈ 2.6, H-C(1)), 2.47 (dd, J ≈ 9.6, 1.6,
irrad. at 2.89 ? d, J = 4.4, Ha-C(3)), 1.84–1.61 (m, irrad. at 2.66 ? change,
irrad. at 2.89 ? change, Hexo-C(7), H-C(8)), 1.59 (ddd, J ≈ 13.7, 10.6, 2.8,
irrad. at 2.66 ? change, Hendo-C(7)); dC(75 MHz, CD3OD): 140.54 (s),
130.22 (d), 129.60 (d); 128.38 (arom. C); 73. 56 (s, C(4)); 73.45 (d, C(6));
70.33 (d, C(5)); 63.81 (t, CH2OH), 61.29 (t, N-CH2Ph), 56.82 (d, C(1)),
51.32 (t, C(3)), 40.14 (d, C(8)), 24.89 (d, C(7)); ESI–MS: 280 ([M + 1]+),
302 ([M + Na]+). Anal. calc. for C15H21NO4·0.5H2O: C 62.48, H 7.69, N
4.86%. Found: C 62.24, H 7.47, N 4.83%.
8 G. H. Posner, V. Vinader and K. Afarinkia, J. Org. Chem., 1992, 57,
4088.
9 B. V. McCleary, Carbohydr. Res., 1983, 111, 297.
10 (a) A. Blaser and J.-L. Reymond, Org. Lett., 2000, 2, 1733; (b) A. M.
Davis and S. J. Teague, Angew. Chem., Int. Ed., 1999, 38, 737.
11 N. Panday, Y. Canac and A. Vasella, Helv. Chim. Acta, 2000, 83, 58.
12 S. L. Lawson, W. W. Wakarchuk and S. G. Withers, Biochemistry,
1997, 36, 2257.
13 N. Panday and A. Vasella, Synthesis, 1999, 1459.
14 K. Tatsuta, Y. Ikeda and S. Miura, J. Antibiot., 1996, 49, 836.
15 K. Tatsuta, S. Miura, S. Ohta and H. Gunji, J. Antibiot., 1995, 48,
286.
§ Snail b-mannosidase: at 25 °C and pH 4.5; jack bean a-mannosidase: at
37 °C and pH 4.5; b-glucosidase from Caldocellum saccharolyticum: at
55 °C and pH 6.8.
¶ Data for 15: Rf (AcOEt–MeOH 5:1): 0.40; dH(300 MHz, D2O): 3.99 (dd,
J ≈ 8.6, 2.1, irrad. at 2.75 ? d, J = 8.7, H-C(6)), 3.85 (dd, J = 10.9, 5.3,
1830
Chem. Commun., 2000, 1829–1830