Isoquinuclidine Mimics of β-D-Glucopyranosides: Differences and Similarities in the Mechanism of Action of some β-D-Glucosidases and a β-D-Mannosidase

Article abstract of DOI:10.1002/hlca.200390321

The D-gluco-isoquinuclidines 3 and 4 were prepared and tested as inhibitors of the β-glucosidases from Caldocellum saccharolyticum and from sweet almonds; the results are compared to the inhibition of snail β-mannosidase by the D-manno-isoquinuclidines 1

Full text of DOI:10.1002/hlca.200390321

3818
Helvetica Chimica Acta Vol. 86 (2003)
Isoquinuclidine Mimics of b-d-Glucopyranosides: Differences and
Similarities in the Mechanism of Action of some b-d-Glucosidases and a
b-d-Mannosidase
by Matthias Bˆhm, Edwige Lorthiois, Muthuppalaniappan Meyyappan, and Andrea Vasella*
Laboratorium f¸r Organische Chemie, ETH-Hˆnggerberg, Wolfgang-Pauli Strasse 10, CH-8093 Z¸rich
Duilio Arigoni, in dankbarer und freundschaftlicher Verbundenheit, zum 75. herzlich zugeeignet
The d-gluco-isoquinuclidines 3 and 4 were prepared and tested as inhibitors of the b-glucosidases from
Caldocellum saccharolyticum and from sweet almonds; the results are compared to the inhibition of snail b-
mannosidase by the d-manno-isoquinuclidines 1 and 2. Exploratory experiments in the racemic series showed
that treatment of the ester epoxide 6 with benzyl alcoholates leads only to epimerisation, transesterification, and
formation of the cyclopropane 9. Ring opening of the reduced epoxide 13 by NaN3 proceeded regioselectively to
provide 14. Treatment of the C(6)ÀO-triflate 16 with AcOCs induced a rearrangement; the reaction with NaN3
gave the C(5)-azido derivative 14. The acetoxy triflate 18, however, reacted with AcOCs to provide the desired
gluco-isoquinuclidine 19. Similarly, the enantiomerically pure acetoxy triflate 22 provided the d-gluco-
isoquinuclidine 24, which was reduced and deprotected to provide 3 and 4. The deoxy analogues 30 and 31 were
obtained by reductive deiodination of the iodide 27, derived from 22. The d-gluco-isoquinuclidines 3, 4, 30, and
31 are much weaker inhibitors of b-glucosidases than the d-manno-analogues 1 and 2 of snail b-mannosidase.
The N-benzyl derivative 3 is a weaker inhibitor than the N-unsubstituted analogue in the gluco-series, while it is
a much stronger inhibitor in the manno-series. A consideration of the pKHA values of the isoquinuclidines 1
4
and the pH value of the enzyme assays suggests that the d-gluco-isoquinuclidines are poor mimics of the shape
of a reactive, enzyme-bound gluco-conformer, while the d-manno-analogues are reasonably good mimics of a
reactive, enzyme-bound manno-conformer. The inhibition results may also suggest that the glycosidase induced
lengthening of the scissile bond and rehybridisation of the anomeric centre are more strongly correlated with the
change of the ground-state conformation during hydrolysis of b-d-glucopyranosides than of b-d-mannopyrano-
sides.
Introduction. Stereoelectronic control of glycoside hydrolysis requires a coplanar
arrangement of the scissile bond and a doubly occupied, nonbonding orbital of the ring
O-atom, i.e., a pseudoaxial orientation of the scissile bond [1]. b-d-Glycopyranosides
possessing an equatorial glycosidic bond must, therefore, change conformation to
undergo hydrolysis. This aspect and the cationic character of the transition state are
expected to be common to all b-d-glycopyranosidases1). One must, however, expect
differences for b-d-glycosidases belonging to different families2) and hydrolyzing
different substrates. Glycosidase inhibitors3) may contribute to an analysis of
similarities and differences of the mechanism of action of individual b-glycosidases.
Recently, we became interested in the difference between the mechanism of action of
1
)
)
For review articles on the mechanism of glycoside hydrolysis, see [2 4].
A regularly updated database with over 6000 glycosidases classified in 89 families [5] is available on the
internet (http://afmb.cnrs-mrs.fr/CAZY/).
2
3
)
For review articles on glycosidase inhibitors, see [4][6].

Helvetica Chimica Acta Vol. 86 (2003)
3819
b-d-glucosidases and snail b-d-mannosidase. The catalytic nucleophile of retaining b-d-
glucosidases of family 1[7] and 3 [8] strongly interacts with the C(2)OH group of the
substrate, an interaction that is geometrically not feasible for b-d-mannosidases. We
speculated about the effect of this interaction [2], concluding that the mechanism of
action of b-d-glucosidases and b-d-mannosidases should differ at the beginning of the
reaction, but lead to similar cationic intermediates. Not too surprisingly then, a
comparison of the inhibition by gluco- and manno-configured tetrahydroimidazopyr-
idine-type inhibitors did not point to differences between the mechanism of action of
the b-glucosidases from Caldocellum saccharolyticum (family 1) and from sweet
almonds [9], and the b-d-mannosidase from snail [10]. These inhibitors mimic shape
and charge of an oxycarbenium cation, but also the cooperative interaction of the
substrate with the catalytic acid and nucleophile [11] that is typical for an early stage of
the hydrolysis. Inhibitors mimicking the shape of an early stage of the reaction may,
however, evidence differences in the mechanism of action of these glycosidases not
visible for inhibitors with a trigonal anomeric centre.
We found that the manno-configured isoquinuclidines 1 and 2 [12] mimicking the
1,4B conformer of a b-d-mannopyranoside are selective inhibitors of snail b-
mannosidase (at pH 4.5: Ki ˆ 0.001m m, a ˆ 1.9 for 1; Ki ˆ 1.2 mm, a ˆ 1.1 for 2), the
strength and type of inhibition depending on the pH value4). An evaluation of the main
factors contributing to the inhibition (shape, pKHA value, and N-substituent) suggested
that the inhibition results agree with the hypothesis of a conformational change of the
substrate, as discussed above. We wondered if the analogous d-gluco-isoquinuclidines 3
and 4 would similarly inhibit b-glucosidases, or if this inhibition would differ from the
one of snail b-mannosidase by the d-manno-isoquinuclidines 1 and 2. In the following,
we describe the synthesis and inhibitory effect of the d-gluco-isoquinuclidines 3 and 4
on several glucosidases and on snail b-mannosidase, and compare the inhibition of
glucosidases by the d-gluco-isoquinuclidines 3 and 4 to the inhibition of snail b-
mannosidase by the d-manno-isoquinuclidines 1 and 2.
Synthesis. We have recently described an efficient synthesis of the d-manno-
configured isoquinuclidines 1 and 2 via the alkene 20 (cf. Scheme 2). We planned to
synthesise the d-gluco-configured isoquinuclidines 3 and 4 by a regioselective ring
opening of an epoxide derived from 20, or by an inverting substitution of the cis-diol 21.
Exploratory experiments were performed with the racemic alkene 5 [12][14]
(Scheme 1). Epoxidation of 5 with meta-chloroperbenzoic acid (mCPBA) in the
presence of 4,4'-thiobis[6-(tert-butyl)-3-methylphenol] [15] yielded 85% of the epoxide
4
)
A B2,5 conformer has been postulated by Davies and co-workers [13] as reactive conformation of the
substrate of the Pseudomonas cellulosa mannanase Man26A.

3820
Helvetica Chimica Acta Vol. 86 (2003)
Scheme 1
a) mCPBA, 4,4'-thiobis[6-(tert-butyl)-3-methylphenol], ClCH2CH2Cl, D; 85%. b) CsF, BnOH, DMF, D; ((Æ)-7/
(Æ)-8 7:3 (33%) and (Æ)-9 (10%). c) LiBH4, THF, D; 78%. d) 1. MeLi, THF, 08; 2. tBuMe2SiCl, THF, 08; 87%.
e) Dimethyldioxirane, acetone, 08; 71%. f) NaN3, DMF, D; (Æ)-13 (44%) and (Æ)-14 (20%). g) N-
Methylmorpholine N-oxide monohydrate (NMO ¥ H2O), OsO4, acetone/H2O; 81%. h) (CF3SO2)2O (Tf2O),
pyridine, À 108; (Æ)-16 (81%). i) NaN3, DMF, D; 78%. j) 1 . TfO, pyridine; 2. Ac2O, pyridine; 66%. k) AcOCs,
2
DMF, D; 84%.
65). Its structure was established by X-ray crystal-structure analysis6) (Fig.). Treat-
ment of 6 with BnOH in the presence of CsF gave the epimeric benzyl esters 7/8 7:3
(33%) and the cyclopropane 9 (10%). No benzyl ether was observed. BnOLi and
BnONa led only to transesterification and epimerisation. We, therefore, transformed
the ester 5 into the silyl ether 12 before epoxidation or dihydroxylation. Dealkylative
epimerisation of the ester 5, followed by benzylation, provided the benzyl ester 10
(46% [12]). Its reduction with LiBH4 in boiling THF yielded 78% of the alcohol 11.
Silylation gave the silyl ether 12 (87%). Its oxidation with dimethyldioxirane led to the
epoxide 13 (71%). All attempts to substitute 13 with AcOCs failed to provide the
5
)
)
For a similar diastereoselective epoxidation of an isoquinuclidine lactam, see [16].
6
The crystallographic data have been deposited with the Cambridge Crystallographic Data Centre as
deposition No. CCDC-217684. Copies of the data can be obtained, free of charge, on application to the
CCDC, 12 Union Road, Cambridge CB2 1EZ, UK (fax: ‡44(1223)336033; e-mail: deposit@ccdc.cam.ac.uk).

Helvetica Chimica Acta Vol. 86 (2003)
3821
Figure. ORTEP Representation of the Crystal Structure a) of the Epoxide (Æ)-6 and b) of the Diacetoxy
8-Phenylmenthyl Ester 24.
desired trans-acetoxy alcohol, and treatment of 13 with NaN3 gave only the azido
alcohol 14 (20%). The regioselectivity of the ring opening shows that the effect of the
electronegative C(5)ÀN-substituent is more important than the combined effects of
the (unfavourably oriented) C(1)-alkoxy group and of the steric hindrance of C(2).
Dihydroxylation of the alkene 12 led diastereoselectively to the diol 15 (81%). It
was O-sulfonylated with trifluoromethanesulfonic acid anhydride (Tf2O) in pyridine at
À 108 to give regioselectively the triflate 16 (81%). Treatment of 16 with AcOCs in
DMFat 1208 did not lead to substitution; IR and 1H-NMR analysis of the crude product
suggested formation of a deoxy ketone (cf. [17]). Surprisingly, treatment of 16 with
NaN3 in DMF at 858 led to substitution at C(5), providing the azido alcohol 14 (78%).
This result suggests a migration of the Tf group from OÀC(6) to OÀC(5) and a more
facile substitution of the resulting triflate 17. Acetylation of OÀC(5) should prevent
this migration and allow substitution at C(6). The acetoxy triflate 18 was prepared by
treating the diol 15 with Tf2O in pyridine and then with Ac2O (66%). Substitution of 18
with AcOCs in DMF at 608 gave indeed 84% of the desired gluco-configured diacetate
19.
The configurations of the epoxides 6 and 13 were deduced from the coupling
between HÀC(4) and the bridgehead HÀC(5) (J(4,5) ˆ 4.4 and 4.5 Hz, resp.) and
from the W-coupling between HÀC(4) and HendoÀC(9) (J(4,6endo) ˆ 1.3 and < 2.2 Hz,
resp.; Table 2 in Exper. Part). This interpretation is in keeping with the crystal structure
of 66). The structure of the cyclopropane 9 agrees with the multiplicity and chemical
shifts of the signals for C(1) (s at 32.01ppm), C(2) ( s at 68.18 ppm), and C(7) (d at
37.54 ppm). The constitution of the azido alcohol 14 was assigned on the basis of a
decoupling experiment; irradiation of the HÀC(6) signal affected the signals for
HÀC(1) (J(1,6)ˆ 2.1), HÀC(5) (J(5,6) ˆ 0.8), and OH (J(6,OH) ˆ 6.2 Hz). The
configuration of 14 was deduced from the small coupling between HÀC(5) and
HÀC(6) (J(5,6) ˆ 0.8 Hz) and by comparing the coupling of HÀC(6) with the
bridgehead HÀC(1) (J(1,6)ˆ 2.1Hz) to the coupling of H ÀC(1) with HendoÀC(7) and

3822
Helvetica Chimica Acta Vol. 86 (2003)
HexoÀC(7) (J(1,7endo) ˆ 2.5, J(1,7exo) ˆ 4.4 Hz). This deduction was corroborated by
NOE experiments confirming that HÀC(6), HendoÀC(7), and HÀC(8) are on the
same, and HÀC(5) and HendoÀC(7) on opposite sides of the cyclohexane ring of 14.
The configuration of the diol 15 was deduced by comparing the coupling of HÀC(6)
with the bridgehead HÀC(1) (J(1,6)ˆ 1.9 Hz) to the coupling of HÀC(1) with
HendoÀC(7) and HexoÀC(7) (J(1,6endo) ˆ 2.2; J(1,6exo) ˆ 3.7 Hz). This deduction was
corroborated by NOE experiments confirming that HÀC(6), HendoÀC(7), and
HÀC(8) are on the same side of the cyclohexane ring of 15. The constitution of the
triflate 16 was assigned on the basis of decoupling experiments confirming the coupling
between OH and HÀC(5) (J(5,OH) ˆ 3.7 Hz), HÀC(5) and HÀC(6) (J(5,6) ˆ
7.9 Hz), and HÀC(6) and HÀC(1) (J(1,6)ˆ 1.7 Hz). The configuration of the gluco-
configured diacetate 19 was deduced by comparing the coupling of HÀC(6) with the
bridgehead HÀC(1) (J(1,6)ˆ 4.4 Hz) to the coupling of HÀC(1) with HendoÀC(7) and
HexoÀC(7) (J(1,7endo) ˆ 2.2, J(1,7exo) ˆ 3.0 Hz); as expected, the coupling between
HÀC(5) and HÀC(6) is small (J(5,6) ˆ 1.6 Hz). Also this deduction was corroborated
by NOE experiments, confirming that HÀC(5) and HÀC(8) are on the same, and
HÀC(6) and HÀC(8) on opposite sides of the cyclohexane ring of 19.
On the basis of these results, we treated the enantiomerically pure diol 21 with Tf2O
and then with Ac2O in pyridine to yield the acetoxy triflate 22 (81%) and the bis-triflate
23 (11%; Scheme 2). The crystalline diol 21 was obtained in 77% yield by
dihydroxylation of the alkene 20 [12]. Substitution of 22 with AcOCs yielded 94%
of the d-gluco-configured diacetate 24. Reduction of 24 (LiAlH4) and acetylation of
the crude product gave the O-acetylated N-benzyl amine 25 (88%). It was deprotected
by cleaving the methoxymethyl acetal with Me3SiBr, followed by deacetylation, to yield
87% of the N-benzyl-d-gluco-isoquinuclidine 3. Catalytic hydrogenolysis of 3 under
acidic conditions led to the N-unsubstituted d-gluco-isoquinuclidine 4 (95%). This
synthesis provided the enantiomerically pure N-unsubstituted d-gluco-isoquinuclidine
4 in eight steps and 55% yield from the diol 21.
The X-ray crystal-structure analysis of the enantiomerically pure diacetoxy 8-
phenylmenthyl ester 247) (Fig.) established the d-gluco-configuration of 24 and
thereby also of 3 and 4. Selected chemical shifts and coupling constants of the
isoquinuclidines 3, 4, and 22 25 are given in Table 3 (Exper. Part).
To study the interaction of glucosidases with the isoquinuclidine OH group
mimicking C(2)OH of the substrate, we also synthesised the 2-deoxy-d-gluco-
isoquinuclidines 30 and 318). Treatment of the acetoxy triflate 22 with Bu4NI gave
the iodide 27 (95%) that was deiodinated with Bu3SnH in toluene at 438 to provide the
lactam 28 in 90% yield. Selected chemical shifts and coupling constants of 27 and 28 are
given in Table 3 (Exper. Part). Reduction of 28 with LiAlH4, followed by acetylation of
the crude product, gave the O-acetylated N-benzyl amine 29 (91%) that was
deprotected by treatment with ethanolic HCl to yield 89% of the N-benzyl-2-deoxy-
7
)
)
The crystallographic data have been deposited with the Cambridge Crystallographic Data Centre as
deposition No. CCDC-217685. Copies of the data can be obtained, free of charge, on application to the
CCDC, 12 Union Road, Cambridge CB2 1EZ, UK (fax: ‡44(1223)336033; e-mail: deposit@ccdc.cam.ac.uk).
The isoquinuclidines 30 and 31 were prepared by Corinne Baumgartner as part of her undergraduate
training.
8

Helvetica Chimica Acta Vol. 86 (2003)
3823
Scheme 2
a) NMO ¥ H2O, OsO4, THF/acetone/H2O; 77%. b) 1 . Tf O, pyridine; 2. Ac2O, pyridine; 22 (81%) and 23 (11%).
2
c) CsOAc, DMF, D; 94%. d) 1. LiAlH4, THF, D; 2. 4-(Dimethylamino)pyridine (DMAP), Ac2O, pyridine;
88%. e) Me3SiBr, molecular sieves (4 ä), CH2Cl2, D; 88%. f) NH3/MeOH; 99%. g) H2, Pd(OH)2ÀC, MeOH/
H2O/HCl; 95%. h) Bu4NI, DMF, D; 95%. i) 2,2'-Azobis(isobutyronitrile) (AIBN), Bu3SnH, toluene, D; 90%. j)
1. LiAlH4, THF, D; 2. DMAP, Ac2O, pyridine; 91%. k) EtOH/H2O/HCl; 89%. l) H2, Pd(OH)2/C, MeOH/H2O/
HCl; 92%.
d-gluco-isoquinuclidine 30. Catalytic hydrogenolysis of 30 under acidic conditions led
to the N-unsubstituted 2-deoxy-d-gluco-isoquinuclidine 31 (92%). This synthesis
provided 31 in eight steps and 52% yield from the diol 21.
Inhibition studies. The d-gluco-isoquinuclidines 3 and 4 are both weak inhibitors
of the b-glucosidases from C. saccharolyticum and from sweet almonds, and of the
cellulase Cel7A from Trichoderma reesei9); the N-unsubstituted d-gluco-isoquinucli-
dine 4 is a weak inhibitor of the b-d-glucan glucohydrolase from barley, whereas the N-
benzyl d-gluco-isoquinuclidine 3 did not inhibit this enzyme10) (Table 1). The d-gluco-
9
)
We thank Prof. Dr. Jerry Stahlberg, Swedish University of Agricultural Sciences, Uppsala, for a generous
sample of T. reesei Cel7A.
Compound 3 did not inhibit barley b-d-glucan glucohydrolase up to a concentration of 4 mm, when it
started to precipitate. We thank Dr. M. Hrmova, University of Adelaide, for testing 3 and 4 as barley b-d-
glucan glucohydrolase inhibitors.
10
)

3824
Helvetica Chimica Acta Vol. 86 (2003)
isoquinuclidines 3 and 4 do neither inhibit the endo-1,3-glucanase from soybean11) nor
the endoglucanase NtEg1from Nasutitermes takasagoensis12). Surprisingly, the N-
benzyl-d-gluco-isoquinuclidine 3 is a stronger inhibitor of snail b-mannosidase than of
the glucosidases from C. saccharolyticum and sweet almonds. The 2-deoxy-d-gluco-
isoquinuclidines 30 and 31 are weak inhibitors of the glucosidase from C. saccha-
rolyticum; the b-glucosidases from sweet almonds are neither inhibited by 30 nor by 31
(Table 1).
Table 1. KM and IC50 Values [mm] for the Inhibition of C. saccharolyticum and Sweet Almond b-Glucosidases,
Barley b-d-Glucan Glucohydrolase, T. reesei Cellulase Cel7A, and Snail b-Mannosidase by the Isoquinuclidines
3, 4, 30, and 31 at the Given Temperature and pH Value
Enzyme
Temp.
pH
KM
IC50 Values
3
4
30
31
b
C. saccharolyticum b-glucosidase
Sweet almond b-glucosidases
Barley b-d-glucan glucohydrolase
T. reesei Cel7A
558
378
378
308
258
6.80
6.80
5.25
5.72
4.50
0.80
3.4
1.7
0.39
0.58
2.1a)
3.3
1
2.7
)
.9 9.4
4.8
n.i.e)
n.i.e)
n.i.c)
6.0
2.2d)
12.6
2.0
Snail b-mannosidase
0.15
a) Ki ˆ 1.5 mm (a ˆ 5.7). b) Ki ˆ 1.3 mm (a ˆ 9.0). c) n.i. ˆ no inhibition up to a concentration of [3] ˆ 4 mm.
d) Ki ¥ 2. e) n.i. ˆ no inhibition up to a concentration of [30] ˆ [31] ˆ 4 mm.
To compare the inhibition of the C. saccharolyticum and sweet almond b-
glucosidases at pH 6.8 by the d-gluco-isoquinuclidines 3 (pKHA ˆ 6.6) and 4 (pKHA
ˆ
7.8) with the inhibition of snail b-mannosidase at pH 4.5 by the d-manno-isoquinucli-
dines 1 (pKHA ˆ 7.5) and 2 (pKHA ˆ 8.4) [12], we considered the difference between the
pKHA value of the isoquinuclidines 3 and 4, and the pH value of the enzyme assays, as
we did for 1 and 2 [12]. On the basis of the pH dependence of the inhibition of snail b-
mannosidase by the d-manno-isoquinuclidines 1 and 2, we concluded that the amines
rather than the ammonium salts act as inhibitors. An extrapolation of the pH-
dependent inhibition constants (at pH 4.5: Ki ˆ 1.0 mm, a ˆ 1.9 for 1 and Ki ˆ 1.2 mm,
a ˆ 1.1 for 2) suggested that d-manno-isoquinuclidines corresponding to 1 and 2, but
with a (hypothetical) pKHA value equivalent to the pH optimum of the b-mannosidase
(pH ca. 4.5) would show an inhibition constant of ca. 2 nm and ca. 2 mm, respectively.
The assumption of a similar dependence on the pH of the inhibition constants for the b-
glucosidase from C. saccharolyticum by the d-gluco-isoquinuclidines 3 and 4 suggests
that d-gluco-isoquinuclidines with a (hypothetical) pKHA value equivalent to the pH
optimum of the b-glucosidase (pH ca. 6.8) would show inhibition constants of ca. 2 mm
and ca. 0.3 mm, respectively, i.e., larger or similar to the MichaelisÀMenten constant
(KM ˆ 0.60 mm). The same assumption suggests IC50 values of ca. 5 mm and ca. 0.5 mm
for the inhibition of the b-glucosidases from sweet almonds by d-gluco-isoquinuclidines
11
)
)
We thank Prof. Dr. J. Ebel, Ludwig-Maximilian-Universit‰t, M¸nchen, for testing 3 and 4 as inhibitors of
the b-d-glucanase from soybean.
We thank Prof. E. Meyer, Texas A&M University, for a generous sample of the endoglucanase NtEg1from
the termite N. takasagoensis.
12

Helvetica Chimica Acta Vol. 86 (2003)
3825
corresponding to 3 and 4, but with a (hypothetical) pKHA value equivalent to the pH
optimum of the b-glucosidase (pH ca. 6.8).
These considerations confirm that isoquinuclidines are much stronger inhibitors of
snail b-mannosidase than of the b-glucosidases from C. saccharolyticum and from sweet
almonds. They evidence that shape, and not charge, is at the origin of this difference13).
The enzymatic hydrolysis of b-d-mannopyranosides may proceed via a conformer that
is mimicked (more or less well, as indicated by the mixed-type inhibition) by the d-
manno-isoquinuclidines 1 and 2, while the enzymatic hydrolysis of the b-d-glucopy-
ranosides proceeds via a significantly different conformer. The weakening effect of the
N-Bn substituent on the inhibition by the gluco-isoquinuclidines 3 and 4 is significant,
since the b-glucosidases from sweet almonds show a strong affinity for hydrophobic
aglyca [9][19][20]. The destabilising interaction with the N-Bn group contrasts with its
effect on the inhibition of the b-mannosidase. This may reflect a location of the N-Bn
group of 3, as imposed by the inappropriate structure of the d-gluco-isoquinuclidine,
that significantly deviates from the location of the aglycon part of the substrate on the
way to the transition state. One also has to consider the possibility that the change of
the ground-state conformation precedes lengthening of the scissile bond and
rehybridisation of the anomeric centre during hydrolysis of b-d-mannopyranosides,
but not of b-d-glucopyranosides; an incipient conformational change that is strongly
correlated with bond breaking would also lead to structures not mimicked by an
isoquinuclidine. This, too, would constitute a significant difference to the mechanism of
action of snail b-mannosidase.
We thank Dr. B. Schweizer for the determination of the X-ray crystal structures, D. Manser and M.
Schneider for the pKHA determinations, Dr. B. Bernet for checking the Exper. Part, and the Swiss National
Science Foundation and Oxford Glycosciences Ltd., Abingdon, UK, for generous support.
Experimental Part
General. Solvents were distilled before use. If not specified otherwise, all reactions were carried out under a
N2 atmosphere. Normal workup implies pouring the reaction mixture into the indicated sat. aq. soln., extracting
into the mentioned org. solvent, if necessary washing with the indicated sat. aq. soln., drying of the org. layer
(Na2SO4 or MgSO4), filtration, and evaporation of the volatiles. TLC: Merck silica-gel 60F-254 plates; detection
with I2, by heating with Mostain (400 ml of 10% H2SO4 soln., 20 g of (NH4)6Mo7O24 ¥ 6 H2O, 0.4 g of Ce(SO4)2)
or with KMnO4 soln. (3 g of KMnO4, 20 g of K2CO3, 0.25 ml of AcOH in 300 ml of H2O). Flash
chromatography (FC): silica gel Fluka 60 (0.04 0.063 mm) if not indicated otherwise. IR Spectra: 0.5% KBr
suspension, 3% CHCl3 or CH2Cl2 soln. 1H- (300 MHz, if not indicated otherwise) and 13C-NMR (75 MHz, if not
indicated otherwise): d in ppm and J in Hz.
Methyl rac-(1R,2R,4R,5R,8R)-1-(Methoxymethoxy)-7-oxo-3-oxa-6-azatricyclo[3.2.2.02,4]octane-8-carbox-
ylate ((Æ)-6). A soln. of (Æ)-5 (1.60 g, 6.63 mmol), mCPBA (2.86 g, 16.6 mmol), and 4,4'-thiobis[6-(tert-butyl)-3-
methylphenol] (42 mg, 0.117 mmol) in ClCH2CH2Cl (50 ml) was heated to reflux for 4.5 h. Normal workup
(CH2Cl2/Na2SO3 soln., MgSO4) and FC (cyclohexane/AcOEt/MeOH 1:1:0.1) gave ( Æ)-6 (1.46 g, 85%). White
solid. Rf (AcOEt/hexane/MeOH 1:1:0.1) 0.29. M.p. 122 8 (CH2Cl2/Et2O). IR (CHCl3): 3424w, 3223m, 3007m,
2901w, 1706s, 1437m, 1364m, 1271m, 1175m, 1152m, 1096w, 1060m, 984m, 908w. 1H-NMR (CDCl3): see Table 2;
additionally, 5.86 (br. d, J ꢀ 0.9, HN); 5.16, 4.94 (2d, J ˆ 7.5, OCH2O); 3.72 (s, CO2Me); 3.39 (s, MeOCH2); 2.94
(dd, J ˆ 10.3, 5.8, HÀC(8)); 2.33 (ddd, J ˆ 13.2, 10.3, 4.1, HexoÀC(9)); 1.87 (ddt, J ˆ 13.2, 5.8, 1.3, HendoÀC(9)).
13C-NMR (CDCl3): 173.18, 171.06 (2s, 2 CˆO); 95.05 (t, OCH2O); 81.76 (s, C(1)); 56.95 (q, MeOCH2); 52.58
13
)
We consider the possibility of an unfavourable interaction of CH2(3) of the d-gluco-isoquinuclidines 3 and
4 (corresponding to an axial C(4)-substituent in a b-d-glucoside) with the b-glucosidase from C.
saccharolyticum as improbable, since this enzyme cleaves both b-d-glucosides and b-d-galactosides [18].

3826
Helvetica Chimica Acta Vol. 86 (2003)
Table 2. Selected 1H-NMR Chemical Shifts [ppm] and Coupling Constants [Hz] of the Racemic Isoquinuclidines
(Æ)-6 to (Æ)-9, (Æ)-13 to (Æ)-16, (Æ)-18, and (Æ)-19. For an easier comparison, all isoquinuclidines are
numbered in the same way as (Æ)-14 to (Æ)-19.
(Æ)-6
(Æ)-7
(Æ)-8
(Æ)-9
(Æ)-13
(Æ)-14
(Æ)-15
(Æ)-16
(Æ)-18
(Æ)-19
HÀC(1)
HÀC(5)
HÀC(6)
J(1,5)
3.90
4.06
3.60
0.6
4.4
1
3.90
4.09
3.613.50
0.9
3.90
3.78
3.54
2.88
3.66
3.69
2.80
3.82
3.46
3.97
3.63
4.08
3.71
5.34
3.62
5.05
4.27
3.40
3.28
3.95
4.98
4.98
4.52
a
)
0.8
3.4
J(1,6)
4.3
4.1
.2 2.11
3.13.7
3.0
2.1
2.5
4.4
1.9
2.2
3.7
1.7
2.2
4.0
1.6
2.2
4.13.0
7.8 .6 1
4.4
2.2
b
c
J(1,7
)
.3
)
4.3
.3
)
)
endo
c
J(1,7
)
4.14.3
4.4
exo
J(5,6)
J(5,OH)
J(6,7endo
J(6,7exo
J(6,OH)
4.1
4.5
0.8
8.4
7.9
2.5
3.7
c
)
1
)
)
1
.6
6.2
8.4
a) Broad signal; J(1,5) < 1.5 Hz is deduced from the width of the signal at the half of the height. b) Broad signal;
J(1,7endo) < 2.5 Hz is deduced from the width of the signal at the half of the height. c) Broad signals; J(1,7endo) ꢀ
J(1,7exo) ꢀ 2.9 Hz, J(6,7endo) < 2.2 Hz are deduced from the width of the signal at the half of the height.
(q, CO2Me); 50.06 (d, C(5)); 48.91( d, C(8)); 45.58, 44.15 (2d, C(2), C(4)); 31.34 (t, C(9)). CI-MS: 258 (57,
‡
[M ‡ H] ), 226 (95), 214 (14), 197 (29), 153 (19), 45 (100). Anal. calc. for C11H15NO6 (257.24): C 51.36, H 5.88,
N 5.44; found C 51.37, H 5.83, N 5.48.
X-Ray Crystal-Structure Analysis of (Æ)-6. The crystals were obtained by recrystallisation from CH2Cl2/
Et2O. Triclinic, P1, a ˆ 6.9317(12) ä, b ˆ 9.5462(13) ä, c ˆ 9.625(2) ä, a ˆ 105.880(12)8, b ˆ 91.401(13)8, g ˆ
104.468(12)8, Vˆ 590.2(2) ä3, Z ˆ 1, Dcalc. ˆ 1.447 Mg/m3. From a crystal of size 0.40  0.40  0.20 mm, 1988
reflexions were measured on an Enraf Nonius CAD-4 diffractometer (graphite monochromator, CuKa
radiation, l ˆ 1.54184 ä) at 293(2) K, by Dr. B. Schweizer (ETH-Z¸rich). R ˆ 0.0610, Rw ˆ 0.2324. The
structure was solved by direct methods with SHELXS-96 [21]. The non-H-atoms were refined anisotropically
with SHELXL-97 [22]. The H-atoms were calculated and included in the structure factor calculation. Drawings
of the molecule were made with ORTEP [23].
Benzyl rac-(1R,2R,4R,5R,8R)- and rac-(1R,2R,4R,5R,8S)-1-(Methoxymethoxy)-7-oxo-3-oxa-6-azatricy-
clo[3.2.2.02,4]octane-8-carboxylate ((Æ)-7 and (Æ)-8, resp.), and Benzyl rac-(1R,2S,5R,6S,7S)-6-Hydroxy-2-
(methoxymethoxy)-3-oxo-4-azatricyclo[3.2.1.02,7]octane-1-carboxylate ((Æ)-9).
A
soln. of (Æ)-6 (30 mg,
0.12 mmol) and dry CsF (90 mg, 0.58 mmol) in DMF (1 ml) was treated with BnOH (0.06 ml, 0.58 mmol) at
238 and stirred at 1008 for 72 h. Normal workup (Et2O/ice-water; MgSO4) and FC (cyclohexane/AcOEt/MeOH
1:1:0.1) gave ( Æ)-7/(Æ)-8 7:3 (13 mg, 33%) and (Æ)-9 (4 mg, 10%).
Data for (Æ)-7/(Æ)-8 7:3: Rf (AcOEt/hexane/MeOH 1:1:0.1) 0.33. 1H-NMR (CDCl3): see Table 2;
additionally, 7.38 7.28 (m, 5 H); 6.38 (br. d, J ˆ 5.6, 0.7 H), 6.00 (br. s, 0.3 H) (HN); 5.29, 4.98 (2d, J ˆ 7.5,
0.6 H), 5.19, 4.93 (2d, J ˆ 7.5, 1.4 H) (OCH2O); 5.21, 5.14 (2d, J ˆ 12.1, 1.4 H), 5.21, 5.10 (2d, J ˆ 12.5, 0.6 H)
(PhCH2); 3.48 (s, 0.9 H), 3.28 (s, 2.1H) (MeO); 3.09 ( dd, J ˆ 9.8, 4.8, 0.3 H), 3.01( dd, J ˆ 10.4, 5.9, 0.7 H)
(HÀC(8)); 2.36 (ddd, J ˆ 13.2, 10.4, 4.3, 0.7 H), 2.18 (ddd, J ˆ 13.4, 4.8, 3.1, 0.3 H) (HexoÀC(9)); 2.09 (ddd, J ˆ
13.4, 9.8, 2.1, 0.3 H), 1.88 (br. dd, J ˆ 13.2, 5.9, w1/2 ˆ 2.5, 0.7 H) (HendoÀC(9)).
Data for (Æ)-9: Rf (AcOEt/hexane/MeOH 1 :1:0.1) 0.11. IR (CHCl3): 3426s, 2957s, 2927s, 2871s, 2858s,
1731s, 1685s, 1458s, 1379m, 1359m, 1270m, 1177m, 1167m, 1109w, 1089w, 998w, 978w. 1H-NMR (CDCl3): see
Table 2; additionally, 7.36 7.30 (m, 5 arom. H, HN); 5.20, 5.12 (2d, J ˆ 12.5, PhCH2); 4.99, 4.77 (2d, J ˆ 6.9,
OCH2O); 4.41 4.07 ( m, exchange with D2O, HO); 4.27 (addn. of D2O ! d, J ˆ 3.0, HÀC(6)); 3.33 (s, MeO);
2.12 (dd, J ˆ 12.8, 3.7, HexoÀC(8)); 2.00 (dd, J ˆ 12.8, 1.2, HendoÀC(8)). 13C-NMR (CDCl3): 169.66, 167.43
(2s, 2 CˆO); 135.78 (s); 128.92 (2d); 128.71 (2d); 128.58 (d); 97.66 (t, OCH2O); 70.04 (d, C(6)); 67.55
(t, PhCH2); 68.18 (s, C(2)); 57.10 (q, MeO); 51.21 (d, C(5)); 37.54 (d, C(7)); 33.24 (t, C(8)); 32.01( s, C(1)). CI-
‡
MS: 334 (65, [M ‡ H] ), 302 (48), 184 (10), 91 (100), 45 (30).

Helvetica Chimica Acta Vol. 86 (2003)
3827
rac-(1R,4S,8R)-2-Benzyl-8-(hydroxymethyl)-4-(methoxymethoxy)-2-azabicyclo[2.2.2]oct-5-ene-3-one
((Æ)-11). A boiling soln. of LiBH4 (207 mg, 9.42 mmol) in THF (4.8 ml) was treated dropwise with a soln. of
(Æ)-10 (640 mg, 1.57 mmol) in THF (4.8 ml), heated to reflux for 90 min, cooled to 238, treated dropwise with
MeOH (5 ml), and stirred at 238 for 1h. Normal workup (AcOEt/H 2O; Na2SO4) and FC (AcOEt/hexane/
MeOH 1:1:0.1) gave ( Æ)-11 (374 mg, 78%). Colourless oil. Rf (AcOEt/hexane/MeOH 1:1:0.1) 0.27. IR
1
(CH2Cl2): 3453m, 3068w, 2932m, 1679s, 1452m, 1421m, 1226w, 1162m, 1057m, 993m, 91 9w. H-NMR (CDCl3):
7.34 7.15 (m, 5 H); 6.75 (dd, J ˆ 8.1, 1.9, irrad. at 3.95! d, J ˆ 8.1, HÀC(5)); 6.29 (dd, J ˆ 8.1, 5.9, irrad. at
3.95 ! d, J ˆ 8.1, HÀC(6)); 5.16, 5.05 (2d, J ˆ 7.5, OCH2O); 4.59, 4.40 (2d, J ˆ 14.8, PhCH2); 3.95 (ddt, J ˆ 5.3,
3.5, 1.9, HÀC(1)); 3.70 3.56 (m, CH2ÀC(8)); 3.49 (s, MeO); 2.92 (br. s, exchange with D2O, HO); 2.18
(ddt, J ˆ 10.0, 7.2, 5.0, HÀC(8)); 1.70 (ddd, J ˆ 12.5, 10.3, 2.2, irrad. at 3.95 ! br. dd, J ˆ 12.5, 10.6, HendoÀC(7));
1.27 (ddd, J ˆ 12.8, 5.0, 3.4, irrad. at 3.95 ! dd, J ˆ 12.1, 4.7, HexoÀC(7)). 13C-NMR (CDCl3): 171.64 (s, C(3));
136.84 (s); 135.75, 131.92 (2d, C(5), C(6)); 129.14 (2d); 128.69 (2d); 128.15 (d); 96.0 (t, OCH2O), 86.98 (s, C(4));
64.95 (t, CH2ÀC(8)); 56.13 (q, MeO), 52.41( d, C(1)); 48.64 (d, C(8)); 42.38 (t, PhCH2); 31.27 (t, C(7)). CI-MS:
‡
304 (3, [M ‡ H] ), 272 (10), 138 (100), 91 (42), 45 (51). Anal. calc. for C17H21NO4 (303.96): C 67.31, H 6.98,
N 4.62; found: C 67.26, H 7.21, N 4.63.
rac-(1R,4S,8R)-2-Benzyl-8-{[(tert-butyl)dimethylsilyloxy]methyl}-4-(methoxymethoxy)-2-azabicyclo[2.2.2]-
oct-5-en-3-one ((Æ)-12). A cooled (08) soln. of (Æ)-11 (285 mg, 0.94 mmol) in THF (5 ml) was treated with a
1.5m soln. of MeLi in Et2O (0.72 ml, 1.15 mmol), stirred for 15 min, treated with a soln. of tBuMe2SiCl (174 mg,
1.15 mmol) in THF (3 ml), and stirred at 08 for 1h. Normal workup (AcOEt/NH 4Cl soln.; Na2SO4) and FC
(hexane/AcOEt 3 :7) gave (Æ)-12 (341mg, 87%). Colourless oil. Rf (AcOEt/hexane 7 :3) 0.60. 1H-NMR
(CDCl3): 7.32 7.16 (m, 5 H); 6.70 (dd, J ˆ 8.1, 1.9, irrad. at 3.94! d, J ˆ 8.1, HÀC(5)); 6.28 (dd, J ˆ 8.1, 5.6,
irrad. at 3.94 ! d, J ˆ 8.4, HÀC(6)); 5.13, 5.01 (2d, J ˆ 7.2, OCH2O); 4.57, 4.37 (2d, J ˆ 14.9, PhCH2); 4.02
(dd, J ˆ 9.7, 4.9, irrad. at 2.12 ! d, J ˆ 10.0, CHaÀC(8)); 3.94 (ddt, J ˆ 5.0, 3.0, 1.9, HÀC(1)); 3.47 (s, MeO); 3.30
(t, J ˆ 9.7, irrad. at 2.12 ! dd, J ˆ 9.3, 2.0, CHbÀC(8)); 2.12 (tt, J ꢀ 9.5, 4.3, HÀC(8)); 1.70 (ddd, J ˆ 13.1, 4.7, 3.1,
irrad. at 3.94 ! dd, J ˆ 12.8, 4.7, irrad. at 2.12 ! change, HexoÀC(7)); 1.62 (ddd, J ˆ 12.5, 10.3, 1.9, irrad. at
3.94 ! dd, J ˆ 12.8, 9.6, irrad. at 2.12 ! change, HendoÀC(7)); 0.84 (s, t-Bu); 0.02, 0.00 (2s, Me2Si). 13C-NMR
(CDCl3): 171.04 (s, C(3)); 136.71 (s); 135.43, 131.87 (2d, C(5), C(6)); 128.65 (2d); 128.20 (2d); 127.57 (d); 95.17
(t, OCH2O); 84.05 (s, C(4)); 63.73 (t, CH2ÀC(8)); 55.56 (q, MeO); 52.44 (t, PhCH2); 48.25 (d, C(1)); 42.46
(d, C(8)); 31.39 (t, C(7)); 25.91( q, Me3CSi); 18.26 (s, Me3CSi); À 5.34, À 5.41(2 q, Me2Si). CI-MS: 418 (7, [M ‡
‡
H] ), 386 (27), 360 (41), 258 (33), 145 (35), 115 (100), 91 (90), 45 (57). Anal. calc. for C23H35NO4Si (417.62):
C 66.15, H 8.45, N 3.35; found: C 65.14, H 8.45, N 3.42.
rac-(1R,2R,4R,5R,8R)-6-Benzyl-8-{[(tert-butyl)dimethylsilyloxy]methyl}-1-(methoxymethoxy)-3-oxa-6-
azatricyclo[3.2.2.02,4]octan-7-one ((Æ)-13). A cooled (08) soln. of (Æ)-12 (44 mg, 0.10 mmol) in CH2Cl2 (2 ml)
was treated with a freshly prepared ca. 0.1m soln. of dimethyldioxirane in acetone [24] (2.1ml, ca. 0.21mmol)
and stirred at 08 for 3 h, when TLC showed an incomplete reaction. The mixture was treated with a ca. 0.1m soln.
of dimethyldioxirane in acetone (1.0 ml, ca. 0.10 mmol) and stirred at 08 for 20 min. Normal workup (CH2Cl2,
Na2S2O3 soln.; Na2SO4) and FC (cyclohexane/EtOAc 4 :1) gave (Æ)-13 (32 mg, 71%). Rf (cyclohexane/AcOEt
4 :1) 0.22. IR (CH2Cl2): 2955s, 2930s, 2857m, 1687s, 1495m, 1463m, 1350w, 1160s, 1106s, 1059s, 996s, 920w, 839s.
1H-NMR (200 MHz, CDCl3): see Table 2; additionally, 7.38 7.26 (m, 5 H); 5.25, 5.02 (2d, J ˆ 7.0, OCH2O); 4.83,
4.23 (2d, J ˆ 15.0, PhCH2); 3.92 (dd, J ˆ 9.6, 3.6, CHaÀC(8)); 3.48 (s, MeO); 3.33 (t, J ꢀ 9.6, CHbÀC(8)); 2.29
(tdd, J ꢀ 8.5, 5.0, 3.7, HÀC(8)); 1.81 (dd, J ꢀ 13.8, 10.4, w1/2 ˆ 2.9, HendoÀC(9)); 1.75 (br. dd, J ꢀ 13.7, 5.8, w1/2
ˆ
‡
2.9, HexoÀC(9)); 0.86 (s, t-Bu); 0.04, 0.02 (2s, Me2Si). CI-MS: 434 (41, [M ‡ H] ), 402 (52), 376 (59), 316 (100),
202 (26), 91(84), 45 (44).
rac-(1R,4R,5S,6R,8R)-5-Azido-2-benzyl-8-{[(tert-butyl)dimethylsilyloxy]methyl}-6-hydroxy-4-(methoxy-
methoxy)-2-azabicyclo[2.2.2]octan-3-one ((Æ)-14). A soln. of (Æ)-13 (32 mg, 0.073 mmol) and NaN3 (12 mg,
0.18 mmol) in DMF (3 ml) was stirred at 1208 for 20 h. Normal workup (AcOEt/H2O; Na2SO4) and FC
(cyclohexane/AcOEt 7 :3) gave (Æ)-13 (14 mg, 44%) and (Æ)-14 (7 mg, 20%).
Data of (Æ)-14: Rf (cyclohexane/AcOEt 13 :7) 0.46. IR (CH2Cl2): 3595w, 2955s, 2950s, 2856m, 2114s, 1683s,
1462m, 1156m, 1096m, 1044m, 994m, 840s. 1H-NMR (200 MHz, C6D6): see Table 2; additionally, 7.22 7.10
(m, 5 H); 5.70 (d, J ˆ 8.0, irrad. at 3.82 ! NOE of 3.1%), 5.00 (d, J ˆ 8.0) (OCH2O); 4.84, 4.10 (2d, J ˆ 14.5,
PhCH2); 4.01( dd, J ˆ 9.5, 4.2, CHaÀC(8)); 3.82 (irrad. at 3.28 ! NOE of 1.9%, HÀC(5)); 3.56 (t, J ꢀ 9.5,
CHbÀC(8)); 3.37 (s, MeO); 3.28 (irrad. at 3.82 ! dd, J ꢀ 4.6, 0.8, irrad. at 2.80 ! br. d, J ꢀ 4.6, irrad. at 3.82 !
NOE of 2.0%, HÀC(6)); 2.80 (irrad. at 3.28 ! NOE of 5.7%, HÀC(1)); 2.59 (tt, J ꢀ 9.4, 4.4, HÀC(8)); 1.54
(dt, J ˆ 14.1, 4.5, irrad. at 2.80 ! dd, J ꢀ 14.1, 4.4, HexoÀC(7)); 1.26 (ddd, J ˆ 14.1, 10.4, 2.5, irrad. at 2.80 !
br. dd, J ꢀ 14.1, 10.4, irrad. at 3.28 ! NOE of 2.6%, HendoÀC(7)); 1.12 (d, J ˆ 6.2, HOÀC(6)); 0.95 (s, t-Bu);
0.06, 0.04 (2s, Me2Si). 13C-NMR (125 MHz, C6D6): 168.89 (s, C(3)); 137.71 (s); 137 125 (several d); 94.74

3828
Helvetica Chimica Acta Vol. 86 (2003)
(t, OCH2O); 81.16 (s, C(4)); 76.56 (d, C(6)); 69.74 (d, C(5)); 63.83 (t, CH2ÀC(8)); 56.18 (d, C(1)); 56.02
(q, MeO); 49.75 (t, PhCH2); 36.64 (d, C(8)); 26.76 (t, C(7)); 26.17 (q, Me3CSi); 18.58 (s, Me3CSi); À 5.28, À5.40
‡
(2q, Me2Si). CI-MS: 477 (6, [M ‡ H] ), 449 (25), 417 (38), 391 (61), 359 (86), 202 (34), 91 (100), 75 (28), 45
(19).
rac-(1R,4R,5R,6R,8R)-2-Benzyl-8-{[(tert-butyl)dimethylsilyloxy]methyl}-5,6-dihydroxy-4-(methoxymeth-
oxy)-2-azabicyclo[2.2.2]octan-3-one ((Æ)-15). A soln. of (Æ)-12 (640 mg, 1.53 mmol) and N-methylmorpholine
N-oxide ¥ H2O (NMO ¥ H2O; 310 mg, 2.29 mmol) in acetone/H2O 3 :2 (20 ml) was treated with OsO4 (ca. 20 mg,
ca. 0.08 mmol) and stirred at 258 for 2 h. Normal workup (AcOEt/Na2SO3 soln.; Na2SO4) and FC (hexane/
AcOEt 1:1) gave ( Æ)-15 (565 mg, 81%). Rf (hexane/AcOEt 1:1) 0.51. IR (CH 2Cl2): 3500s, 3366s, 2933m,
2855m, 1678s, 1461m, 1094s, 1011m, 839s. 1H-NMR (CDCl3): see Table 2; additionally, 7.35 7.27 (m, 5 H); 5.77,
4.78 (2d, J ˆ 7.8, OCH2O); 5.35 (d, J ˆ 2.5, HOÀC(5)); 5.12, 4.13 (2d, J ˆ 14.6, PhCH2); 3.97 (irrad. at 1.66!
NOE of ca. 2%, HÀC(5)); 3.95 (irrad. at 3.46 ! t, J ꢀ 8.4, irrad. at 1.66! NOE of ca. 2%, HÀC(6)); 3.87
(dd, J ˆ 9.7, 4.0, CHaÀC(8)); 3.73 (br. d, J ꢀ 5.5, HOÀC(6)); 3.52 (s, MeO); 3.46 (irrad. at 1.66! NOE of 4.9%,
HÀC(1)); 3.27 (t, J ˆ 9.5, CHbÀC(8)); 2.06 (tdd, J ˆ 9.7, 5.6, 4.0, irrad. at 1.66! NOE of 7.6%, HÀC(8)); 1.66
(ddd, J ˆ 14.0, 10.0, 2.2, irrad. at 3.46 ! dd, J ˆ 14.0, 10.0, HendoÀC(7)); 1.39 (ddd, J ˆ 14.0, 5.9, 3.7, irrad. at
3.46 ! dd, J ˆ 14.0, 5.6, irrad. at 1.66 ! NOE of 24.6%, HexoÀC(7)); 0.84 (s, t-Bu); À 0.02, À 0.03 (2s, Me2Si).
13C-NMR (CDCl3): 168.79 (s, C(3)); 137.53 (s); 128.91 (2d); 128.86 (2d); 127.84 (d); 93.58 (t, OCH2O); 81.27
(s, C(4)); 71.43, 68.84 (2d, C(5), C(6)); 63.89 (t, CH2ÀC(8)); 57.10 (d, C(1)); 55.88 (q, MeO), 49.70 (t, PhCH2);
39.58 (d, C(8)); 28.09 (t, C(7)); 25.92 (q, Me3CSi); 18.29 (s, Me3CSi); À 5.51, (q, Me2Si). CI-MS: 452 (6, [M ‡
‡
H] ), 420 (39), 362 (66), 334 (40), 91(100), 75 (77), 49 (57). Anal. calc. for C 23H37NO6Si (451.63): C 61.17,
H 8.26, N 3.10; found: C 61.17, H 8.06, N 3.14.
rac-(1R,4R,5R,6R,8R)-2-Benzyl-8-{[(tert-butyl)dimethylsilyloxy]methyl}-5-hydroxy-4-(methoxymeth-
oxy)-3-oxo-2-azabicyclo[2.2.2]oct-6-yl Trifluoromethanesulfonate ((Æ)-16) and Its Transformation to (Æ)-14. A
cooled (À108) soln. of (Æ)-15 (1.01 g, 2.24 mmol) in pyridine (13 ml) was treated with Tf2O (0.48 ml, 2.92 mmol)
and stirred for 30 min. Normal workup (AcOEt/NaHCO3 soln.; Na2SO4) and FC (cyclohexane/AcOEt 7:3)
gave (Æ)-16 (1.05 g, 81%). Pale white foam. Rf (hexane/AcOEt 7:3) 0.35. 1H-NMR (CDCl3): see Table 2;
additionally, 7.33 7.29 (m, 5 H); 5.78, 4.75 (2d, J ˆ 8.0, OCH2O); 5.21( d, J ˆ 3.7, HOÀC(5)); 4.95, 4.25 (2d, J ˆ
14.7, PhCH2); 4.90 (irrad. at 3.63 ! d, J ˆ 7.9, HÀC(6)); 4.08 (irrad. at 5.21 ! d, J ˆ 7.9, HÀC(5)); 3.82 (dd, J ˆ
9.8, 3.9, CHaÀC(8)); 3.55 (s, MeO); 3.40 (dd, J ˆ 9.7, 8.4, CHbÀC(8)); 2.06 1.98 (m, HÀC(8)); 1.72 (ddd, J ˆ
14.3, 9.6, 2.2, irrad. at 3.63 ! dd, J ꢀ 14.3, 9.6, HendoÀC(7)); 1.64 (ddd, J ˆ 14.3, 6.5, 4.0, irrad. at 3.63 ! dd, J ꢀ
14.3, 6.5, HexoÀC(7)); 0.85 (s, t-Bu); 0.03, 0.01(2 s, Me2Si). 13C-NMR (CDCl3): 168.05 (s, C(3)); 136.25 (s); 129.0
(2d); 128.96 (2d); 128.15 (d); 118.20 (q, J(C,F) ꢀ 320, CF3); 93.87 (t, OCH2O); 82.88 (d, C(6)); 80.82 (s, C(4));
72.19 (d, C(5)); 63.13 (t, CH2ÀC(8)); 56.15 (q, MeO); 54.33 (d, C(1)); 49.17 (t, PhCH2); 39.42 (d, C(8)); 27.66
(t, C(7)); 26.08 (q, Me3CSi); 18.51 (s, Me3CSi); À 5.26, À 5.38 (2q, Me2Si).
A soln. of (Æ)-16 (21mg, 0.035 mmol) and NaN 3 (35 mg, 0.53 mmol) in DMF (1ml) was stirred at 85 8 for
4 h. Normal workup (AcOEt/H2O; Na2SO4) and FC (cyclohexane/AcOEt 8 :2) gave (Æ)-14 (13 mg, 78%).
rac-(1R,4R,5R,6R,8R)-2-Benzyl-8-{[(tert-butyl)dimethylsilyloxy]methyl}-4-(methoxymethoxy)-3-oxo-6-
[(trifluoromethyl)sulfonyloxy]-2-azabicyclo[2.2.2]oct-5-yl Acetate ((Æ)-18). A cooled (08) soln. of (Æ)-15
(210 mg, 0.47 mmol) in pyridine (3 ml) was treated with Tf2O (0.17 ml, 0.61 mmol), stirred at 258 for 7 h,
treated with Ac2O (0.3 ml, 1.7 mmol), and stirred at 258 for 12 h. Normal workup (AcOEt/H2O; Na2SO4) and
FC (hexane/AcOEt 17:3) gave (Æ)-18 (180 mg, 66%). Rf (hexane/AcOEt 7:3) 0.85. IR (CHCl3): 3004m,
2968m, 1727s, 1687s, 1473m, 1418s, 1120s, 1009m. 1H-NMR (CDCl3): see Table 2; additionally, 7.36 7.26
(m, 5 H); 5.39, 4.81(2 d, J ˆ 7.5, OCH2O); 5.34 (irrad. at 4.98 ! s, HÀC(5)); 5.03, 4.23 (2d, J ˆ 14.8, PhCH2);
4.98 (irrad. at 3.71 ! d, J ˆ 8.1, HÀC(6)); 3.83 (dd, J ˆ 10.0, 3.7, CHaÀC(8)); 3.71(irrad. at 4.98
!
br. s, HÀC(1)); 3.42 (dd, J ˆ 10.0, 8.1, CHbÀC(8)); 3.36 (s, MeO); 2.26 2.14 (m, HÀC(8)); 2.11 (s, Ac); 1.76
(ddd, J ˆ 12.3, 10.0, 2.2, irrad. at 3.71 ! dd, J ˆ 12.1, 10.0, HendoÀC(7)); 1.65 (ddd, J ˆ 10.3, 6.2, 4.1, irrad. at
3.71 ! dd, J ˆ 10.3, 6.2, HexoÀC(7)); 0.85 (s, t-Bu); 0.03, 0.00 (2s, Me2Si). 13C-NMR (CDCl3): 170.13, 167.88
(2s, 2 CˆO); 136.16 (s); 129.07 (2d); 128.94 (2d); 128.33 (d); 118.45 (q, J(C,F) ˆ 321, CF3); 93.90 (t, OCH2O);
80.76 (s, C(4)); 77.74, 71.37 (2d, C(5), C(6)); 62.92 (t, CH2ÀC(8)); 55.98 (q, MeO); 54.86 (d, C(1)); 49.21
(t, PhCH2); 38.88 (d, C(8)); 27.49 (t, C(7)); 25.89 (q, Me3CSi); 20.32 (s, Me3CSi); 18.32 (q, MeCˆO); À 5.51,
À5.61(2 q, Me2Si). 19F-NMR (282 MHz, CDCl3): À 74.77 (s). HR-MALDI-MS: 648.1883 (100, [M ‡ Na] ,
‡
‡
‡
‡
[C26H38F3NNaO9SSi] ; calc. 648.1886), 594.1796 (94, [M À MeO] , [C25H35F3NO8SSi] ; calc. 594.1805).
rac-(1R,4R,5R,6S,8R)-2-Benzyl-8-{[(tert-butyl)dimethylsilyloxy]methyl}-4-(methoxymethoxy)-3-oxo-2-
azabicyclo[2.2.2]octane-5,6-diyl Diacetate ((Æ)-19). A soln. of (Æ)-18 (20 mg, 0.032 mmol) in DMF (0.5 ml) was
treated with AcOCs (9 mg, 0.048 mmol) and stirred at 608 for 6 h. Normal workup (AcOEt/H2O; Na2SO4) and
FC (hexane/AcOEt 1:4) gave (Æ)-19 (15 mg, 84%). Colourless solid. Rf (hexane/AcOEt 17:3) 0.33. M.p.

Helvetica Chimica Acta Vol. 86 (2003)
3829
82 838. 1H-NMR (CDCl3): see Table 2; additionally, 7.34 7.25 (m, 5 H); 5.1 9 (d, J ˆ 7.4), 4.83 (d, J ˆ 7.4, irrad.
at 5.05 ! NOE of 0.6%) (OCH2O); 5.05 (irrad. at 4.52 ! s, HÀC(5)); 4.62, 4.56 (2d, J ˆ 14.6, PhCH2); 4.52
(irrad. at 3.62 ! br. s, irrad. at 5.05 ! NOE of 4.5%, HÀC(6)); 3.86 (dd, J ˆ 10.0, 4.0, irrad. at 2.34 ! d, J ˆ 10.0,
CHaÀC(8)); 3.62 (irrad. at 4.52 ! br. s, irrad. at 4.52 ! NOE of 6.8%, HÀC(1)); 3.41 (t, J ˆ 10.0, irrad. at
2.34 ! d, J ˆ 9.8, CHbÀC(8)); 3.38 (s, MeO); 2.39 2.29 (m, irrad. at 5.05 ! NOE of 7.3%, HÀC(8)); 2.05
(s, irrad. at 4.52 ! NOE of 1.3%, 2 Ac); 2.00 (ddd, J ꢀ 14.0, 10.0, 2.2, irrad. at 2.34 ! br. d, J ꢀ 12.5, irrad. at
3.62 ! dd, J ˆ 14.0, 10.3, HendoÀC(7)); 1.60 1.49 (m, irrad. at 2.34 ! br. d, J ꢀ 14.0, irrad. at 3.62 ! br. dd, J ꢀ
14.0, 5.6, irrad. at 4.52 ! ddd, J ꢀ 14.0, 6.0, 3.0, HexoÀC(7)); 0.86 (s, t-Bu); 0.03, 0.01(2 s, Me2Si). HR-MALDI-
‡
‡
‡
‡
MS: 558.2490 (94, [M ‡ Na] , [C27H41NNaO8Si] ; calc. 558.2499), 504.2408 (100, [M À MeO] , [C26H38NO7Si] ;
calc. 504.2418), 416.1887 (27).
(1R,2S,5R)-5-Methyl-2-(1-methyl-1-phenylethyl)cyclohexyl (1R,4R,5S,7R,8R)-2-Benzyl-7,8-dihydroxy-4-
(methoxymethoxy)-3-oxo-2-azabicyclo[2.2.2]octane-5-carboxylate (21 [12]). A yellow soln. of 20 (3.00 g,
5.64 mmol) and NMO ¥ H2O (1.18 g, 8.46 mmol) in THF/acetone/H2O 3 :1:2 (60 ml) was treated with a 2.5%
soln. of OsO4 in t-BuOH (0.71ml, 0.056 mmol) and stirred at 24 8 for 6 h. Normal workup (AcOEt/Na2SO3 soln.;
brine; Na2SO4) and recrystallisation of the crude product from i-PrOH/hexane gave 21 (2.44 g, 77%). Colourless
crystals. M.p. 1628 (Et2O/hexane; [12]: 1628).
(1R,2S,5R)-5-Methyl-2-(1-methyl-1-phenylethyl)cyclohexyl (1R,4R,5S,7R,8R)-8-Acetoxy-2-benzyl-4-(meth-
oxymethoxy)-3-oxo-7-[(trifluoromethyl)sulfonyloxy]-2-azabicyclo[2.2.2]octane-5-carboxylate (22) and
(1R,2S,5R)-5-Methyl-2-(1-methyl-1-phenylethyl)cyclohexyl (1R,4R,5S,7R,8R)-2-Benzyl-4-(methoxymethoxy)-
3-oxo-7,8-bis[(trifluoromethyl)sulfonyloxy]-2-azabicyclo[2.2.2]octane-5-carboxylate (23). A soln. of 21 [12]
(3.00 g, 5.30 mmol) in CH2Cl2/pyridine 1:1(60 ml) was treated dropwise with Tf 2O (0.96 ml, 5.82 mmol) at 248,
stirred for 12 h, treated with Ac2O (1.0 ml, 10.6 mmol), and stirred at 258 for 4 h. Normal workup (AcOEt/
NaHCO3 soln.; Na2SO4) and FC (cyclohexane/AcOEt 6 :1) gave 23 (0.45 g, 11%) and 22 (3.16 g, 81%) as
colourless foams.
Data of 22: Rf (cyclohexane/AcOEt 3 :2) 0.64. M.p. 86 908 (Et2O/hexane). [a]2D5 ˆ À59.3 (c ˆ 1.0, CHCl3).
IR (CHCl3): 2966m, 2928m, 2852w, 1766m, 1728s, 1701s, 1600w, 1496w, 1454w, 1422m, 1371m, 1322m, 1143s,
1079m, 999m, 961m, 897m, 842w. 1H-NMR (CDCl3): see Table 3; additionally, 7.43 7.16 (m, 9 H); 7.15 7.08
(m, 1H); 5.20, 4.13 (2 d, J ꢀ 14.8, PhCH2); 5.07, 4.64 (2d, J ˆ 7.4, OCH2O); 4.81( td, J ˆ 10.8, 4.4, HÀC(1')); 4.80
(irrad. at 3.62 ! d, J ˆ 7.5, HÀC(7)); 3.30 (s, MeO); 2.08 (td, J ꢀ 12.5, 3.7, irrad. at 4.81 ! m, HÀC(2')); 2.08
(s, Ac); 1.96 (br. d, J ꢀ 10.9, irrad. at 4.81 ! d, J ˆ 10.6, HeqÀC(6')); 1.91 (dd, J ˆ 10.3, 5.9, HÀC(5)); 1.86
(dq, J ꢀ 13.4, 3.7, HeqÀC(3')); 1.72 (ddd, J ˆ 14.0, 5.9, 3.7, irrad. at 3.62 ! dd, J ˆ 14.3, 5.9, HexoÀC(6)); 1.72 1.63
(m, HeqÀC(4')); 1.46 (ddd, J ˆ 14.0, 10.3, 2.2, irrad. at 3.62 ! dd, J ˆ 14.0, 10.0, HendoÀC(6)); 1.52 1.38
(m, HÀC(5')); 1.24, 1.16 (2s, Me2PhC); 1.17 (qd, J ꢀ 12.6, 3.4, HaxÀC(3')); 0.91( qd, J ꢀ 11.6, 3.6, HaxÀC(4'));
0.90 (q, J ꢀ 12.2, irrad. at 4.81 ! m, HaxÀC(6')); 0.88 (d, J ˆ 6.5, MeÀC(5')). 13C-NMR (CDCl3): 170.31, 169.54,
165.77 (3s, 3 CˆO); 152.16, 135.15 (2s); 128.90 (2d); 128.63 (2d); 127.94 (d); 127.60 (2d); 125.08 (2d); 124.74
(d); 116.97 (q, J(C,F) ˆ 316, CF3); 93.38 (t, OCH2O); 79.90 (d, C(8)); 77.87 (s, C(4)); 75.68 (d, C(1')); 70.39
(d, C(7)); 56.05 (q, MeO); 53.40 (d, C(1)); 50.27 (d, C(2')); 49.09 (t, PhCH2); 41.55 (t, C(6')); 41.39 (d, C(5));
39.44 (s, Me2PhC); 34.55 (t, C(4')); 31.26 (d, C(5')); 29.78, 22.75 (2q, Me2PhC); 27.60 (t, C(6)); 26.22 (t, C(3'));
21.83 (q, MeÀC(5')); 20.29 (q, MeCˆO). 19F-NMR (282 MHz, CDCl3): À 74.80 (s). HR-MALDI-MS: 762.2536
‡
‡
‡
(52, [M ‡ Na] , [C36H44F3NNaO10S] ; calc. 762.2536), 548.0892 (100, [M À (8-phenylmenthyl) ‡ H ‡ Na] ,
‡
‡
‡
[C20H22F3NNaO10S] ; calc. 548.0814), 494.0814 (80, [M À (8-phenylmenthyl) À MeO ‡ H] , [C19H19F3NO9S] ;
calc. 494.0733). Anal. calc. for C36H44F3NO10S (739.80): C 58.45, H 5.99, N 1.89, S 4.33; found: C 58.50, H 6.10,
N 1.90, S 4.33.
Data of 23: Rf (cyclohexane/AcOEt 3 :2) 0.76. M.p. 86 908. [a]2D5 ˆ 10.0 (c ˆ 1.0, CHCl3). IR (CHCl3):
2960w, 2928w, 1731s, 1709m, 1600w, 1496w, 1455m, 1428s, 1369w, 1324m, 1139s, 1092m, 1000m, 948m, 923m,
888m, 859m. 1H-NMR (CDCl3): see Table 3; additionally, 7.42 7.28 (m, 5 H); 7.24 7.16 (m, 4 H); 7.1 2 (ddt, J ꢀ
8.1, 5.9, 2.6, 1 H); 5.42, 4.98 (2d, J ˆ 14.8, PhCH2); 5.41, 4.86 (2d, J ꢀ 7.6, OCH2O); 4.80 (td, J ˆ 10.6, 4.4,
HÀC(1')); 3.37 (s, MeO); 2.32 (dd, J ˆ 10.6, 5.3, HÀC(5)); 2.18 (br. d, J ꢀ 11.5, HeqÀC(6')); 2.06 (td, J ˆ 10.6,
3.5, HÀC(2')); 1.67 1.50 (m, HeqÀC(3'), HeqÀC(4')); 1.58 (ddd, J ˆ 14.3, 10.7, 1.9, HendoÀC(6)); 1.50 1.31
(m, HÀC(5')); 1.33 (dt, J ꢀ 14.3, 5.0, HexoÀC(6)); 1.19, 1.12 (2s, Me2PhC); 1.05 (qd, J ꢀ 12.5, 2.5, HaxÀC(3'));
0.95 (q, J ꢀ 11.2, HaxÀC(6')); 0.87 (d, J ˆ 6.5, MeÀC(5')); 0.85 (qd, J ꢀ 11.6, 3.6, HaxÀC(4')). 13C-NMR (CDCl3):
169.91, 164.44 (2s, 2 CˆO); 150.81, 135.15 (2s); 128.49 (2d); 128.45 (2d); 127.81 (d); 127.48 (2d); 124.94 (2d);
124.74 (d); 118.10, 117.94 (2q, J(C,F) ꢀ 312, 2 CF3); 93.51( t, OCH2O); 80.48, 77.79, 76.91(3 d, C(7), C(8),
C(1')); 76.31( s, C(4)); 55.59 (q, MeO); 52.52 (d, C(1)); 49.32 (d, C(2')); 48.70 (t, PhCH2); 42.12 (d, C(5)); 40.63
(t, C(6')); 39.59 (s, Me2PhC); 34.15 (t, C(4')); 31.02 (d, C(5')); 27.67 (t, C(6)); 26.75 (t, C(3')); 26.64, 26.56
(2q, Me2PhC); 21.50 (q, MeÀC(5')). 19F-NMR (282 MHz, CDCl3): À 73.97, À 73.49 (2s). HR-MALDI-MS:

3830
Helvetica Chimica Acta Vol. 86 (2003)
1
Table 3. Selected H-NMR Chemical Shifts [ppm] and Coupling Constants [Hz] of the Enantiomerically Pure
Isoquinuclidines 3, 4, and 22 28.
3
4
22
23
24
25
26
27
28
3.41
HÀC(1)
HendoÀC(7)
HexoÀC(7)
C(8)
4.02
3.78
3.62
4.80
3.74
4.87
3.63
4.47
2.82
2.78
3.51
3.45
1.99
1.41
5.00
2.2
2.712.69
3.48
5.02
4.96
2.5
5.09
4.80
2.6
3.39
2.8
3.13.7
5.20
2.2
4.99
1.9
4.83
2.5
5.13
2.5
J(1,6
)
1.9
endo
a
b
J(1,6
)
)
4.4
4.0
2.8
3.14.0
)
exo
J(1,7
)
)
1
.7
1
.9
1
.8
endo
a
J(1,7
)
)
)
3.6
1
1.2
4.4
1.9
4.0
2.5
3.9
1.9
2.13.8
1
2.2
exo
a
J(3,8)
J(6exo,7exo
.3
.6
a
)
2.2
J(7endo,7exo
J(7endo,8)
J(7exo,8)
)
1
5.6
7.8
7.8
8.1
1.2
a
)
1.5
0.6
1.9
2.2
2.2
a) Broad signals; J(1,6exo) < 3.5, J(1,7exo) ꢀ 3.5, J(3,8) ꢀ J(6exo,7exo) ꢀ J(7,8) < 2.5 Hz deduced from the width of
the signals at the half of their height. b) 3.41(br. s, HÀC(1)); 1.83 1.76 (m, HexoÀC(6)).
‡
‡
852.1914 (88, [M ‡ Na] , [C35H41F6NNaO11S2] ; calc. 852.1923), 638.0170 (100, [M À (8-phenylmenthyl) ‡ H ‡
‡
‡
‡
Na] , [C19H19F6NNaO11S2] ; calc. 638.0201), 584.0105 (94, [M À (8-phenylmenthyl) À MeO ‡ H] ,
‡
[C18H16F6NO10S2] ; calc. 584.0120).
(1R,2S,5R)-5-Methyl-2-(1-methyl-1-phenylethyl)cyclohexyl (1R,4R,5S,7S,8R)-7,8-Diacetoxy-2-benzyl-4-
(methoxymethoxy)-3-oxo-2-azabicyclo[2.2.2]octane-5-carboxylate (24). A soln. of 22 (2.50 g, 3.38 mmol) and
AcOCs (973 mg, 5.07 mmol) in DMF (50 ml) was stirred at 808 for 2.5 h. Normal workup (AcOEt/H2O;
Na2SO4) and FC (cyclohexane/AcOEt 3 :1) gave 24 (2.05 g, > 98%) as a colourless solid. Recrystallisation from
CH2Cl2/hexane (1:3) gave 24 (1.92 g, 94%). Colourless needles. Rf (cyclohexane/AcOEt 3 :2) 0.39. M.p. 174.58
(CH2Cl2/hexane). [a]2D5 ˆ 19.6 (c ˆ 1.0, CHCl3). IR (CHCl3): 2957m, 2927m, 1741s (br.), 1697s, 1600w, 1496m,
1454m, 1373s, 1330m, 1299m, 1159s, 1129m, 1090m, 1056s, 1038s, 978m, 959m, 91 6m, 850w. 1H-NMR (CDCl3):
see Table 3; additionally, 7.40 7.18 (m, 9 H); 7.14 7.04 (m, 1H); 4.99, 4.65 (2 d, J ˆ 7.5, OCH2O); 4.83 (irrad. at
3.63 ! d, J ˆ 0.6, HÀC(8)); 4.78 (td, J ˆ 10.6, 4.3, HÀC(1')); 4.67, 4.55 (2d, J ˆ 14.6, PhCH2); 4.47 (irrad. at
3.63 ! br. s, HÀC(7)); 3.29 (s, MeO); 2.10, 2.09 (2s, 2 Ac); 2.06 (dd, J ꢀ 10.0, 6.5, HÀC(5)); 2.14 2.04
(m, irrad. at 4.78 ! change, HeqÀC(6')); 2.00 (br. d, J ꢀ 13.7, irrad. at 4.81 ! br. s, HÀC(2')); 1.84 1.73
(m, HeqÀC(3')); 1.78 (ddd, J ˆ 13.7, 10.0, 2.5, HendoÀC(6)); 1.70 1.59 (m, HeqÀC(4')); 1.66 (m, irrad. at 3.63 !
ddd, J ꢀ 13.4, 6.7, 1.9, HexoÀC(6)); 1.53 1.35 (m, HÀC(5')); 1.24, 1.16 (2s, Me2PhC); 1.12 (qd, J ꢀ 12.8, 3.1,
HaxÀC(3')); 0.98 0.80 (m, HaxÀC(4')); 0.89 (q, J ꢀ 11.7, irrad. at 4.78 ! m, HaxÀC(6')); 0.87 (d, J ˆ 6.5,
MeÀC(5')). 13C-NMR (CDCl3): 170.96, 169.35, 168.84, 166.62 (4s, 4 CˆO); 152.25, 135.97 (2s); 128.63 (2d);
128.52 (2d); 127.78 (d); 127.56 (2d); 125.10 (2d); 124.55 (d); 93.10 (t, OCH2O); 78.17, 75.41, 74.99 (3d, C(7),
C(8), C(1')); 77.00 (s, C(4)); 55.92 (q, MeO); 51.06 (d, C(1)); 50.30 (d, C(2')); 48.42 (t, PhCH2); 42.45 (d, C(5));
41.51 (t, C(6')); 39.48 (s, Me2PhC); 34.61( t, C(4')); 31.26 (d, C(5')); 29.44, 23.13 (2q, Me2PhC); 26.67 (t, C(6));
‡
26.29 (t, C(3')); 21.83 (q, MeÀC(5')); 20.90, 20.84 (2q, 2 MeCˆO). ESI-MS: 1321 (8, [2M ‡ Na] ), 672 (100,
‡
‡
[M ‡ Na] ), 650 (6, [M ‡ H] ). Anal. calc. for C37H47NO9 (649.77): C 68.39, H 7.29, N 2.16; found: C 68.24,
H 7.30, N 2.30.
X-Ray Crystal-Structure Analysis of 24. The crystals were obtained by azeotropic recrystallisation in
CH2Cl2/hexane. Orthorhombic, P21P21P21, a ˆ 9.575(2) ä, b ˆ 13.034(2) ä, c ˆ 28.313(3) ä, Vˆ 3533.5(10) ä3,
Z ˆ 4, Dcalc. ˆ 1.221 Mg/m3. From a crystal of size 0.40  0.30  0.30 mm, 3644 reflexions were measured on an
Enraf Nonius CAD-4 diffractometer (graphite monochromator, CuKa radiation, l ˆ 1.54184 ä) at 293(2) K, by
Dr. B. Schweizer (ETH-Z¸rich). R ˆ 0.0355, Rw ˆ 0.1094. The structure was solved by direct methods with
SIR97 [25]. The non-H-atoms were refined anisotropically with SHELXL-97 [22]. The H-atoms were calculated
and included in the structure factor calculation. Drawings of the molecule were done with ORTEP [23].
(1R,4R,5R,7S,8R)-7,8-Diacetoxy-2-benzyl-4-(methoxymethoxy)-2-azabicyclo[2.2.2]octane-5-methyl Ace-
tate (25) and (1R,2S,5R)-5-Methyl-2-(1-methyl-1-phenylethyl)cyclohexyl Acetate. A mixture of 24 (250 mg,

Helvetica Chimica Acta Vol. 86 (2003)
3831
0.385 mmol) and LiAlH4 (160 mg, 3.85 mmol) in THF (12.5 ml) was heated to reflux for 5 h, cooled to 08,
treated dropwise with MeOH (3 ml), and evaporated. The residue was suspended in pyridine/Ac2O 2 :1
(4.5 ml), treated with DMAP (5 mg, 0.04 mmol), and stirred at 248 for 12 h. The mixture was filtered through
Celite, and the residue was washed with AcOEt (3 Â 5 ml). Normal workup (CH2Cl2/NaHCO3 soln.; brine;
Na2SO4) and FC (cyclohexane/AcOEt 5 :1 ! 3 :1) gave (1R,2S,5R)-5-methyl-2-(1-methyl-1-phenylethyl)cyclo-
hexyl acetate (97 mg, 92%) as a colourless oil and 25 (152 mg, 88%) as a colourless solid.
Data of 25: Rf (cyclohexane/AcOEt 3 :2) 0.41. [a]2D5 ˆ 21.5 (c ˆ 1.2, CHCl3). IR (CHCl3): 2958w, 2826w,
1
1736s, 1494w, 1453w, 1370m, 1151w, 1121w, 1080w, 1030m, 921w. H-NMR (CDCl3): see Table 3; additionally,
7.34 7.21( m, 5 H); 4.70, 4.60 (2d, J ˆ 7.2, OCH2O); 4.38 (dd, J ˆ 10.9, 4.5, CHaÀC(5)); 4.17 (dd, J ˆ 10.9, 9.0,
CHbÀC(5)); 3.80, 3.75 (2d, J ˆ 13.6, PhCH2); 3.30 (s, MeO); 2.94 (dd, J ˆ 10.0, 1.6, HaÀC(3)); 2.65 (dd, J ˆ 10.0,
1.2, HbÀC(3)); 2.36 2.20 (m, HÀC(5)); 2.13, 2.08, 2.05 (3s, 3 Ac); 1.94 (ddd, J ˆ 13.7, 10.3, 2.5, HendoÀC(6));
1.67 (dddd, J ˆ 14.0, 5.6, 2.8, 2.5, HexoÀC(6)). 13C-NMR (CDCl3): 170.79, 169.65, 169.38 (3s, 3 CˆO); 138.11 (s);
128.11 (2d); 128.05 (2d); 126.91 (d); 91.24 (t, OCH2O); 76.27, 75.16 (2d, C(7), C(8)); 75.54 (s, C(4)); 64.97
(t, CH2ÀC(5)); 59.13 (t, PhCH2); 55.87 (q, MeO); 51.35 (d, C(1)); 48.58 (d, C(3)); 34.91( d, C(5)); 24.34
‡
‡
(t, C(6)); 21.04, 21.01, 20.97 (3q, 3 MeCˆO). HR-MALDI-MS: 450.2118 (83, [M ‡ H] , [C23H32NO8] ; calc.
‡
‡
‡
‡
450.2128), 448.1970 (26, [M À H] , [C23H30NO8] ; calc. 448.1971), 390.1914 (100, [M À OAc] , [C21H28NO6] ;
calc. 390.1917). Anal. calc. for C23H31NO8 (449.50): C 61.46, H 6.95, N 3.12; found: C 61.47, H 7.03, N 2.95.
Data of (1R,2S,5R)-5-Methyl-2-(1-methyl-1-phenylethyl)cyclohexyl Acetate: see [26].
(1R,4R,5R,7S,8R)-7,8-Diacetoxy-2-benzyl-4-hydroxy-2-azabicyclo[2.2.2]octane-5-methyl Acetate (26). A
cooled (08) mixture of 25 (1.20 g, 2.67 mmol) and molecular sieves (4 ä; 0.60 g) in CH2Cl2 (60 ml) was treated
dropwise with Me3SiBr (0.71ml, 5.38 mmol) and heated to reflux for 2 h. Normal workup (AcOEt/NaHCO 3
soln.; brine; Na2SO4) and FC (cyclohexane/AcOEt 2 :1 ! 1:1) gave 26 (955 mg, 88%). Colourless oil. Rf
(cyclohexane/AcOEt 1:2) 0.60. [ a]2D5 ˆ 19.4 (c ˆ 1.0, CHCl3). IR (CHCl3): 3577w, 2961w, 2837w, 1735s, 1494w,
1453w, 1369m, 1125w, 1079w, 1031m, 977w, 91 w4. 1H-NMR (CDCl3): see Table 3; additionally, 7.33 7.21
(m, 5 H); 4.41( dd, J ˆ 11.2, 6.2, CHaÀC(5)); 4.23 (dd, J ˆ 11.2, 7.4, CHbÀC(5)); 3.81, 3.74 (2d, J ˆ 13.4,
PhCH2); 2.74 (dd, J ˆ 10.0, 1.6, HaÀC(3)); 2.65 (dd, J ˆ 10.0, 1.6, HbÀC(3)); 2.31( s, HOÀC(4)); 2.15, 2.06, 2.04
(3s, 3 Ac); 2.15 2.06 (m, HÀC(5)); 1.95 (ddd, J ˆ 13.4, 10.6, 2.6, HendoÀC(6)); 1.58 (dddd, J ˆ 13.6, 5.3, 3.1, 1.9,
HexoÀC(6)). 13C-NMR (CDCl3): 171.37, 170.77, 169.40 (3s, 3 CˆO); 138.22 (s); 128.23 (4d); 127.05 (d); 79.18,
74.34 (2d, C(7), C(8)); 71.75 (t, CH2ÀC(5)); 65.08 (s, C(4)); 59.13 (t, PhCH2); 51.40 (d, C(1)); 50.11 (t, C(3));
‡
‡
37.49 (d, C(5)); 24.45 (t, C(6)); 21.13 (q, 3 MeCO). HR-MALDI-MS: 406.1856 (73, [M ‡ H] , [C21H28NO7] ;
‡
‡
‡
calc. 406.1866), 404.1701 (26, [M À H] , [C21H26NO7] ; calc. 404.1709), 346.1651 (100, [M À OAc] ,
‡
[C19H24NO5] ; calc. 346.1654). Anal. calc. for C21H27NO7 (405.45): C 62.21, H 6.71, N 3.45; found: C 61.94,
H 6.91, N 3.23.
(1R,4R,5R,7S,8R)-2-Benzyl-5-(hydroxymethyl)-2-azabicyclo[2.2.2]octane-4,7,8-triol (3). A cooled (08)
soln. of 26 (50 mg, 0.123 mmol) in MeOH (20 ml) was saturated with NH3 and stirred at 248 for 12 h. Ion-
‡
exchange chromatography (Amberlite CG-120, H form, 0.1m aq. NH3) gave 3 (34 mg, 99%). Colourless solid.
Rf (AcOEt/MeOH 10 :1) 0.30. [a]2D5 ˆ 8.0 (c ˆ 0.5, EtOH). IR (0.5% in KBr): 3374s (br.), 2899m (br.), 1495w,
1
1452m, 1380m, 1150m, 1111m, 1038s, 975m, 926w, 803w. H-NMR (300 MHz, D2O): see Table 3; additionally,
7.44 7.33 (m, 5 H); 3.89 (dd, J ˆ 10.6, 5.6, CHaÀC(5)); 3.83, 3.79 (2d, J ꢀ 13.3, PhCH2); 3.70 (dd, J ˆ 10.9, 7.8,
CHbÀC(5)); 2.73 (br. d, J ˆ 10.3, HaÀC(3)); 2.60 (br. d, J ˆ 10.6, HbÀC(3)); 2.00 (ddd, J ˆ 13.1, 11.2, 1.9
HendoÀC(6)); 1.87 (br. tt, J ˆ 10.9, 6.0, HÀC(5)); 1.63 (br. d, J ꢀ 13.0, w1/2 ˆ 12, HexoÀC(6)). 13C-NMR (75 MHz,
D2O): 139.73 (s); 131.87 (2d); 131.11 (2d); 130.14 (d); 82.21, 76.54 (2d, C(7), C(8)); 74.91( s, C(4)); 64.41
(t, CH2ÀC(5)); 60.96 (t, PhCH2); 55.50 (d, C(1)); 50.77 (t, C(3)); 40.52 (d, C(5)); 25.11 (t, C(6)). HR-MALDI-
‡
‡
‡
‡
MS: 302.1360 (3, [M ‡ Na] , [C15H21NNaO4] ; calc. 302.1368), 280.1541 (100, [M ‡ H] , [C15H22NO4] ; calc.
280.1549). Anal. calc. for C15H21NO4 ¥ 1³2 H2O (288.34): C 62.48, H 7.69, N 4.86; found: C 62.56, H 7.60, N 4.89.
pKHA ˆ 6.6.
(1R,4R,5R,7S,8R)-5-(Hydroxymethyl)-2-azabicyclo[2.2.2]octane-4,7,8-triol (4). A mixture of 3 (31mg,
0.11 mmol), 20% Pd(OH)ÀC (6 mg), MeOH (3 ml), and conc. aq. HCl (0.15 ml) was stirred at 248 for 12 h
under H2 (6 bar). The mixture was filtered through Celite, and the residue was washed with MeOH (3 Â 5 ml).
The filtrate was diluted with toluene (5 ml) and evaporated. Ion-exchange chromatography (Amberlite CG-120,
‡
H
form, 0.1m aq. NH3) gave 4 (21mg, 95%). Colourless solid. Rf (MeOH/25% aq. NH3 5 :2) 0.49. [a]2D5 ˆ ‡61.1
(c ˆ 0.4, H2O). IR (0.5% in KBr): 3391s (br.), 2935m, 2890m, 1636w, 1530m, 1423m, 1384m, 1258w, 1172w,
1112m, 1045m, 1008m, 926w, 808w. 1H-NMR (300 MHz, CD3OD): see Table 3; additionally, 3.90 (dd, J ˆ 10.6,
7.2, CHaÀC(5); 3.62 (dd, J ˆ 10.9, 7.2, CHbÀC(5)); 2.89 (dd, J ˆ 10.9, 1.3, HaÀC(3)); 2.70 (dd, J ꢀ 10.9, 1.2,
HbÀC(3)); 2.16 (ddd, J ˆ 14.0, 11.2, 2.8, HendoÀC(6)); 1.84 (dqd, J ꢀ 11.2, 5.9, 1.3, HÀC(5)); 1.38 (dddd, J ˆ 14.0,
4.7, 3.1, 1.2, HexoÀC(6)). 13C-NMR (75 MHz, CD3OD): 81.59, 72.98 (2d, C(7), C(8)); 78.41( s, C(4)); 63.84

3832
Helvetica Chimica Acta Vol. 86 (2003)
‡
(t, CH2ÀC(5)); 50.29 (t, C(3)); 40.65, 39.95 (2d, C(1), C(5)); 26.19 (t, C(6)). ESI-MS: 401(18, [2 M ‡ Na] ), 360
‡
‡
‡
(43, [2M À H2O] ), 250 (18), 212 (30, [M ‡ Na] ), 190 (38, [M ‡ H] ), 181 (43), 149 (100). Anal. calc. for
C8H15NO4 ¥ 3³4 H2O (202.72): C 47.40, H 8.20, N 6.91; found: C 47.26, H 8.36, N 6.75. pKHA ˆ 7.80.
(1R,2S,5R)-5-Methyl-2-(1-methyl-1-phenylethyl)cyclohexyl (1R,4R,5R,7S,8S)-8-Acetoxy-2-benzyl-4-(meth-
oxymethoxy)-7-iodo-3-oxo-2-azabicyclo[2.2.2]octane-5-carboxylate (27). A soln. of 22 (348 mg, 0.47 mmol)
and Bu4NI (1.74 g, 4.70 mmol) in DMF (5 ml) was stirred at 808 for 6 h. Normal workup (Et2O/10% aq. Na2S2O3
soln.; Na2SO4) and FC (cyclohexane/AcOEt 4 :1) gave 27 (322 mg, 95%). Colourless foam. Rf (cyclohexane/
AcOEt 2 :3) 0.48. M.p. 56 608. [a]2D5 ˆ 35.8 (c ˆ 0.7, CHCl3). IR (CHCl3): 3011m, 2959s, 2928s, 1750s, 1728s,
1697s, 1610w, 1496m, 1453m, 1371s, 1324m, 1133m, 1077m, 1045s, 992m, 926w, 849w. 1H-NMR (CDCl3): see
Table 3; additionally, 7.37 7.17 (m, 10 H); 5.13 (irrad. at 3.51 ! s, HÀC(8)); 5.01, 4.59 (2d, J ˆ 7.8, OCH2O);
4.78 (td, J ˆ 10.6, 4.4, HÀC(1')); 4.72, 4.48 (2d, J ˆ 14.9, PhCH2); 3.45 (irrad. at 3.51 ! br. s, HÀC(7)); 3.28
(s, MeO); 2.47 (ddd, J ˆ 13.4, 10.0, 2.5, HendoÀC(6)); 2.10 (td, J ꢀ 11.5, 3.4, irrad. at 4.78 ! m, HÀC(2')); 2.09
(dd, J ˆ 10.6, 7.5, HÀC(5)); 2.04 (s, Ac); 2.04 1.92 (m, irrad. at 3.51 ! change, irrad. at 4.78 ! change,
HexoÀC(6), HeqÀC(6')); 1.89 (dq, J ꢀ 12.8, 3.3, HeqÀC(3')); 1.70 (br. d, J ꢀ 12.5, HeqÀC(4')); 1.50 1.37
(m, HÀC(5')); 1.25, 1.18 (2s, Me2PhC); 1.17 (qd, J ꢀ 12.8, 3.4, HaxÀC(3')); 0.93 (qd, J ꢀ 12.5, 3.4, HaxÀC(4'));
0.89 (q, J ꢀ 12.0, irrad. at 4.78 ! t, J ꢀ 12.8, HaxÀC(6')); 0.88 (d, J ˆ 6.5, MeÀC(5')). 13C-NMR (CDCl3): 170.89,
169.26, 166.15 (3s, 3 CˆO); 152.15, 135.95 (2s); 128.63 (2d); 128.49 (2d); 127.93 (d); 127.73 (2d); 125.17 (2d);
124.98 (d); 93.18 (t, OCH2O); 81.50 (d, C(8)); 77.91( s, C(4)); 75.48 (d, C(1')); 55.93 (q, MeO); 55.92 (d, C(1));
50.39 (d, C(2')); 48.08 (t, PhCH2); 41.50 (d, C(5)); 41.50 (t, C(6')); 39.37 (s, Me2PhC); 34.64 (t, C(4')); 31.28
(d, C(5')); 29.95, 22.44, 21.85; 20.62 (4q, Me2PhC, MeÀC(5'), MeCˆO); 29.63 (t, C(6)); 26.22 (t, C(3')); 24.10
‡
‡
(d, C(7)). HR-MALDI-MS: 740.2059 (69, [M ‡ Na] , [C35H44INNaO7] ; calc. 740.2060), 526.0340 (100, [M À 8-
‡
‡
phenylmenthyl ‡ H ‡ Na] , [C19H22INNaO7] ; calc. 526.0339). Anal. calc. for C35H44INO7 (717.64): C 58.58,
H 6.18, N 1.95; found: C 58.63, H 6.37, N 1.88.
(1R,2S,5R)-5-Methyl-2-(1-methyl-1-phenylethyl)cyclohexyl (1R,4R,5S,8S)-8-Acetoxy-2-benzyl-4-(methoxy-
methoxy)-3-oxo-2-azabicyclo[2.2.2]octane-5-carboxylate (28). A soln. of 27 (250 mg, 0.35 mmol) and 2,2'-
azobis(2-methylpropionitrile) (AIBN; 0.62 mg, 0.004 mmol) in toluene (1ml) was treated with Bu 3SnH
(0.19 ml, 0.70 mmol), stirred at 438 for 2 h, and evaporated. FC (cyclohexane/AcOEt 3 :1) gave 28 (185 mg,
90%). Colourless foam. Rf (cyclohexane/AcOEt 2 :3) 0.32. M.p. 59 658. [a]2D5 ˆ 9.2 (c ˆ 0.8, CHCl3). IR
(CHCl3): 3013m, 2959m, 2928m, 1728s, 1690s, 1601w, 1496w, 1456m, 1180m, 1159m, 1092m, 1051m, 1017m,
1
977m, 960m, 924w, 847w. H-NMR (CDCl3): see Table 3; additionally, 7.36 7.20 (m, 9 H); 7.11 (dddd, J ˆ 9.0,
7.8, 4.3, 3.5, 1H); 5.08, 4.68 (2 d, J ˆ 7.5, OCH2O); 4.86, 4.42 (2d, J ˆ 14.9, PhCH2); 4.81( td, J ˆ 10.8, 4.4,
HÀC(1')); 3.32 (s, MeO); 2.11 1.98 (m, HeqÀC(6')); 2.07 (td, J ꢀ 10.6, 3.5, irrad. at 1.79 ! t, J ꢀ 10.6, HÀC(2'));
2.05 (br. dd, J ˆ 10.0, 6.2, HÀC(5)); 2.02 (s, Ac); 1.99 (irrad. at 5.00! br. d, J ꢀ 15.6, irrad. at 3.41 ! dd, J ˆ
15.6, 8.7, HendoÀC(7)); 1.83 1.76 (m, irrad. at 3.41 ! change, HexoÀC(6)); 1.79 (dq, J ꢀ 12.8, 3.2, HeqÀC(3'));
1.67 (br. ddd, J ꢀ 12.8, 3.4, 2.8, HeqÀC(4')); 1.52 (ddd, J ˆ 12.8, 10.3, 2.2, irrad. at 3.41 ! dd, J ˆ 13.1, 10.4, irrad.
at 2.05 ! dd, 10.3, 2.2, HendoÀC(6)); 1.48 1.35 (m, HÀC(5')); 1.41 (irrad. at 5.00! dt, J ꢀ 15.6, 4.3, irrad. at
3.41 ! dt, J ꢀ 14.6, 2.2, irrad. at 2.05 ! br. dd, J ꢀ 15.6, 3.8, HexoÀC(7)); 1.25, 1.17 (2s, Me2PhC); 1.14 (q, J ꢀ 12.7,
irrad. at 1.79! td, J ꢀ 12.1, 2.2, HaxÀC(3')); 0.91( q, J ꢀ 12.1, HaxÀC(6')); 0.90 (qd, J ꢀ 12.8, 3.4, irrad. at 1.79 !
q, J ꢀ 12.6, HaxÀC(4')); 0.88 (d, J ˆ 7.8, MeÀC(5')). 13C-NMR (CDCl3): 171.30, 169.84, 166.88 (3s, 3 CˆO));
152.07, 136.56 (2s); 128.42 (2d); 128.33 (2d); 127.54 (d); 127.48 (2d); 125.06 (2d); 124.64 (d); 93.11 (t, OCH2O);
78.61( s, C(4)); 75.24 (d, C(1')); 72.96 (d, C(8)); 55.78 (q, MeO); 50.25 (d, C(2')); 48.95 (d, C(1)); 48.01
(t, PhCH2); 42.53 (d, C(5)); 41.53 (t, C(6')); 39.47 (s, Me2PhC); 36.72 (t, C(7)); 34.57 (t, C(4')); 32.20 (t, C(6));
31.23 (d, C(5')); 26.30 (t, C(3')); 29.18, 23.36, 21.80; 21.07 (4q, Me2PhC, MeÀC(5'), MeCˆO). HR-MALDI-
‡
‡
MS: 614.3082 (100, [M ‡ Na] , [C35H45NNaO7] ; calc. 614.3094), 400.1360 (98, [M À 8-phenylmenthyl ‡ H ‡
‡
‡
Na] , [C19H23NNaO7] ; calc. 400.1372), 346.1286 (55). Anal. calc. for C35H45NO7 (591.73): C 71.04, H 7.66,
N 2.37; found: C 71.11, H 7.74, N 2.25.
(1R,4R,5R,8R)-8-Acetoxy-2-benzyl-4-(methoxymethoxy)-2-azabicyclo[2.2.2]octane-5-methyl Acetate (29)
and (1R,2S,5R)-5-Methyl-2-(1-methyl-1-phenylethyl)cyclohexyl Acetate. A mixture of 28 (150 mg, 0.25 mmol)
and LiAlH4 (85 mg, 2.23 mmol) in THF (10 ml) was heated to reflux for 90 min, cooled to 08, treated dropwise
with MeOH (20 ml), and evaporated. The residue was suspended in pyridine/Ac2O 2 :1(30 ml), treated with
DMAP (2 mg, 0.03 mmol), and stirred at 258 for 12 h. The mixture was filtered through Celite, and the residue
was washed with AcOEt (3 Â 5 ml). Normal workup (CH2Cl2/NaHCO3 soln.; brine; Na2SO4) and FC
(cyclohexane/AcOEt 3 :1 ! 1:1) gave (1 R,2S,5R)-5-methyl-2-(1-methyl-1-phenylethyl)cyclohexyl acetate
(70 mg, > 98%) and 29 (90 mg, 91%) as colourless oils.
Data of 29: Rf (cyclohexane/AcOEt/MeOH 1:1:0.1) 0.30. [ a]2D5 ˆ 13.2 (c ˆ 0.9, CHCl3). IR (CHCl3):
1
2951m, 2925w, 1731s, 1602w, 1494w, 1453m, 1369s, 1149s, 1126m, 1045s, 922m, 830w. H-NMR (CDCl3): 7.34

Helvetica Chimica Acta Vol. 86 (2003)
3833
7.20 (m, 5 H); 5.02 (ddd, J ˆ 9.3, 3.1, 1.8, HÀC(8)); 4.75, 4.61(2 d, J ˆ 7.5, OCH2O); 4.42 (dd, J ꢀ 10.7, 4.6,
CHaÀC(5)); 4.21( dd, J ˆ 10.6, 9.4, CHbÀC(5)); 3.69 (s, PhCH2); 3.32 (s, MeO); 2.98 (dd, J ˆ 9.9, 1.6,
HaÀC(3)); 2.73 (dd, J ˆ 10.0, 1.9, HbÀC(3)); 2.65 (quint., J ꢀ 2.5, HÀC(1)); 2.22 (br. tt, J ꢀ 9.7, 4.3,
HÀC(5)); 2.13 (ddd, J ꢀ 14.9, 9.7, 2.0, HendoÀC(7)); 2.12, 2.08 (2s, 2 Ac); 1.89 (dq, J ꢀ 14.3, 3.0, HexoÀC(7));
1.81 (ddd, J ꢀ 13.4, 4.8, 2.8, HexoÀC(6)); 1.71 (ddd, J ꢀ 13.4, 10.6, 2.5, HendoÀC(6)). 13C-NMR (CDCl3): 171.01,
170.21 (2s, 2 CˆO); 138.84 (s); 128.19 (2d); 128.13 (2d); 126.81 (d); 91.21 (t, OCH2O); 75.14 (s, C(4)); 71.67
(d, C(8)); 65.57 (t, CH2ÀC(5)); 59.49 (t, C(3)); 55.92 (q, MeO); 49.70 (t, PhCH2); 48.46 (d, C(1)); 35.11
‡
(d, C(5)); 33.48 (t, C(7)); 29.80 (t, C(6)); 21.39, 21.09 (2q, 2 MeCˆO). HR-MALDI-MS: 414.1883 ([M ‡ Na] ,
‡
‡
‡
[C21H30NNaO6] ; calc. 414.1893), 392.2064 ([M ‡ H] , [C21H30NO6] ; calc. 392.2073).
Data of (1R,2S,5R)-5-Methyl-2-(1-methyl-1-phenylethyl)cyclohexyl Acetate: see [26].
(1R,4R,5R,8R)-2-Benzyl-5-(hydroxymethyl)-2-azabicyclo[2.2.2]octane-4,8-diol (30). A soln. of 29 (38 mg,
0.10 mmol) and 37% aq. HCl (0.4 ml) in EtOH (4 ml) was stirred at 258 for 24 h and evaporated. Ion-exchange
chromatography (Amberlite CG-120, 0.1m aq. NH3) and FC (Macherey-Nagel Lichoprep-NH2, AcOEt/MeOH
10 :1) gave 30 (23 mg, 89%). Colourless solid. Rf (Macherey-Nagel Nanosil-NH2; AcOEt/MeOH 6 :1) 0.39.
[a]2D5 ˆ À14.6 (c ˆ 0.6, MeOH). IR (0.5% in KBr): 3400m, 3083w, 3028w, 2937m, 2892m, 2866m, 1956w, 1883w,
1810w, 1740w, 1602w, 1494w, 1453w, 1384w, 1351w, 1119m, 1031m, 990w, 956w, 91 6w, 81 7w. 1H-NMR (CD3OD):
7.35 7.26 (m, 4 H); 7.21( tt, J ꢀ 6.9, 1.9, 1 H); 3.90 (dd, J ꢀ 10.5, 6.5, CHaÀC(5)); 3.73 (dd, J ꢀ 10.5, 5.9,
CHbÀC(5)); 3.71, 3.67 (2d, J ꢀ 13.4, PhCH2); 3.72 3.67 (m, HÀC(8)); 2.78 (dd, J ˆ 10.0, 1.5, HaÀC(3)); 2.61
(dd, J ˆ 10.3, 2.2, HbÀC(3)); 2.62 2.58 (m, HÀC(1)); 2.00 (ddd, J ˆ 14.0, 9.3, 2.8, HendoÀC(7)); 1.92 (br. d, J ꢀ
13.7, HexoÀC(7)); 1.78 (dqd, J ꢀ 13.7, 6.9, 1.0, HÀC(5)); 1.70 1.66 (m, 2 HÀC(6)). 13C-NMR (CD3OD): 139.80
(s); 129.76 (2d); 129.15 (2d); 127.95 (d); 73.32 (s, C(4)); 72.70 (d, C(8)); 63.93 (t, CH2ÀC(5)); 60.52 (t, C(3));
51.33 (t, PhCH2); 50.55 (d, C(1)); 40.78 (d, C(5)); 34.08 (t, C(7)); 30.33 (t, C(6)). HR-MALDI-MS: 264.1596
‡
‡
([M ‡ H] , [C15H22NO3] ; calc. 264.1600). Anal. calc. for C15H21NO3 ¥ 1³2 H2O (272.16): C 68.15, H 8.39, N 5.30;
found: C 67.92, H 8.46, N 5.26.
(1R,4R,5R,8R)-5-(Hydroxymethyl)-2-azabicyclo[2.2.2]octane-4,8-diol (31). A mixture of 30 (23 mg,
0.09 mmol), 20% Pd(OH)ÀC (6 mg), MeOH (1ml), and 37% aq. HCl (0.15 ml) was stirred at 24 8 for 16 h
under H2 (6 bar). The mixture was filtered through Celite, and the residue was washed with MeOH (3 Â 5 ml).
The filtrate was diluted with toluene (5 ml) and evaporated. Ion-exchange chromatography (Amberlite CG-120,
0.1m aq. NH3) gave 31 (14 mg, 92%). Colourless solid. Rf (MeOH/25% aq. NH3 soln. 3 :1) 0.25. [a]2D5 ˆ ‡25.3
(c ˆ 0.8, H2O). IR (0.5% in KBr): 3293m, 3221m, 3045m, 2910m, 2776m, 2482w, 2138w, 1610m, 1480w, 1410m,
1337m, 1262w, 1155m, 1111m, 993m, 958w, 901w, 846w. 1H-NMR (CD3OD): 3.85 (dd, J ˆ 10.9, 4.7, CHaÀC(5));
3.79 (br. d, J ˆ 9.7, HÀC(8)); 3.76 (dd, J ˆ 10.9, 5.0, CHbÀC(5)); 3.45 (quint., J ꢀ 2.8, HÀC(1)); 3.31 (dd, J ꢀ
11.5, 1.5, HaÀC(3)); 3.23 (dd, J ˆ 11.5, 1.6, HbÀC(3)); 2.36 (ddd, J ˆ 14.9, 9.7, 1.9, HendoÀC(7)); 1.99 (br. d, J ꢀ
15.5, HexoÀC(7)); 1.95 1.85 (m, HÀC(5), 2 HÀC(6)). 13C-NMR (CD3OD): 70.20 (s, C(4)); 70.05 (d, C(8));
61.41 (t, CH2ÀC(5)); 46.82 (d, C(1)); 39.82 (t, C(3)); 38.72 (d, C(5)); 35.20 (t, C(7)); 27.87 (t, C(6)). ESI-MS:
‡
174 (45, [M ‡ H] ), 163 (54), 153 (26), 145 (35), 103 (100), 97 (54).
Inhibition of Glycosidases. The IC50 values were determined at a substrate concentration corresponding to
the KM of each enzyme. IC50 Values were calculated by plotting the inhibitor concentration vs. the rate of
hydrolysis. Determination of the inhibition constants Ki was performed at five different concentrations of the
inhibitor bracketing the IC50 value. Ki Values and a values were determined from the replot of the slopes and the
replot of the 1/v axis intercepts of LineweaverÀBurk plots [27].
a) Inhibition of C. saccharolyticum b-Glucosidase. KM ˆ 0.80 mm ([28]: KM ˆ 0.50 0.78 mm; [1 8]:KM
ˆ
0.51m m). The inhibition studies were carried out at pH 6.8 (KH2PO4/K2HPO4 buffer, 0.06m) and 558. The
reaction was started by the addition of 4-nitrophenyl b-d-glucopyranoside after pre-incubating the enzyme in
the presence of the inhibitor for 30 min at 558. After 15 min, the enzyme reaction was quenched by the addition
of borate buffer (pH 9.2, 0.2m). The rate of hydrolysis was determined by measuring the absorption at l ˆ
405 nm and subsequently substracting the absorption of a blank probe (H2O, buffer, substrate).
b) Inhibition of Sweet Almonds b-Glucosidases. KM ˆ 3.4 mm ([28]: KM ˆ 2.9 3.1m m). The inhibition
studies were carried out at pH 6.8 (KH2PO4/K2HPO4 buffer, 0.06m) and 378. The reaction was started by the
addition of 4-nitrophenyl b-d-glucopyranoside after pre-incubating the enzyme in the presence of the inhibitor
for 30 min at 378. The velocity of the substrate hydrolysis was determined by measuring the increase of
absorption per min at l ˆ 405 nm during 10 min.
c) Inhibition of T. reesei Cellulase Cel7A. KM ˆ 0.39 mm ([29]: KM ˆ 0.406 mm). The inhibition studies were
carried out at pH 5.7 (KH2PO4/Na2HPO4 buffer, 0.05m) and 308. The reaction was started by the addition of 4-
nitrophenyl b-lactopyranoside after pre-incubating the enzyme in the presence of the inhibitor for 30 min at 308.

3834
Helvetica Chimica Acta Vol. 86 (2003)
After 10 min, the enzyme reaction was quenched by the addition of borate buffer (pH 9.2, 0.2m). The rate of
hydrolysis was determind by measuring the absorption at l ˆ 405 nm and subsequently substracting the
absorption of a blank probe (H2O, buffer, substrate).
d) Inhibition of N. takasagoensis Endoglucanase NtEg1. KM ˆ 8.7 mg/ml (corresponding to ca. 0.032 mm of
−sodium carboxymethyl cellulose×). According to a personal communication from Prof. E. Meyer, Texas A&M
University, the inhibition studies were carried out at pH 5.6 (AcONa buffer, 0.04m containing CaCl2, 0.8 mm)
and 378. The reaction was started by the addition of a 2.3 (w/w) −sodium carboxymethyl cellulose× soln. (average
molecular weight ca. 250000 Da, degree of carboxymethyl substitution 0.7 (w/v)) after pre-incubating the
enzyme in the presence of the inhibitor for 30 min at 378. After 30 min, the mixture was added to a soln. of
tetrazolium blue reagent soln. [30] and heated to 1008 for 3 min. After cooling to 258, the rate of hydrolysis was
determined by measuring the absorption at l ˆ 660 nm and subsequently substracting the absorption of a blank
probe (H2O, buffer, substrate).
e) Inhibition of Snail b-Mannosidase. KM ˆ 0.58 mm ([10]: KM ˆ 0.42 0.80 mm). The inhibition studies
were carried out at pH 4.5 (AcONa buffer, 0.04m) and 258. The reaction was started by the addition of 4-
nitrophenyl b-d-mannopyranoside after pre-incubating the enzyme in the presence of the inhibitor for 120 min
at 258. After 5 min, the enzyme reaction was quenched by the addition of borate buffer (pH 9.2, 0.2m). The rate
of hydrolysis was determined by measuring the absorption at l ˆ 405 nm and subsequently substracting the
absorption of a blank probe (H2O, buffer, substrate).
REFERENCES
[1] P. Deslongchamps, −Stereoelectronic Effects in Organic Chemistry×, Pergamon Press, Oxford, 1983; A. J.
Kirby, −The Anomeric Effect and Related Stereoelectronic Effects at Oxygen×, Springer-Verlag, Berlin,
1983; P. Deslongchamps, Pure Appl. Chem. 1993, 65, 1161.
[2] A. Vasella, G. J. Davies, M. Bˆhm, Curr. Opin. Chem. Biol. 2002, 619.
[3] P. J. Berti, H. S. E. Tanaka, Adv. Phys. Org. Chem. 2002, 37, 239; S. G. Withers, Carbohydr. Polym. 2001, 44,
325; D. L. Zechel, S. G. Withers, Acc. Chem. Res. 2000, 33, 11; C. S. Rye, S. G. Withers, Curr. Opin. Chem.
Biol. 2000, 4, 573; T. D. Heightman, A. T. Vasella, Angew. Chem., Int. Ed. 1999, 38, 750; H. Ly, S. G.
Withers, Annu. Rev. Biochem. 1999, 68, 487; D. L. Zechel, S. G. Withers, in −Comprehensive Natural
Product Chemistry×, Ed. C. D. Poulter, Elsevier, New York, 1999, Vol. 5, p. 279; G. Davies, M. L. Sinnott,
S. G. Withers, in −Comprehensive Biological Catalysis×, Academic Press, 1998, Vol. 1, p. 119; A. White,
D. R. Rose, Curr. Opin. Biol. Chem. 1997, 7, 645; G. Davies, B. Henrissat, Structure 1995, 3, 853; J. D.
McCarter, S. G. Withers, Curr. Opin. Struct. Biol. 1994, 4, 885.
[4] A. E. St¸tz, −Iminosugars as Glycosidase Inhibitors: Nojirimycin and Beyond×, Wiley-VCH, Weinheim,
1999.
[5] P. M. Coutinho, B. Henrissat, in −Recent Advances in Carbohydrate Bioengineering×, Eds. H. J. Gilbert, G.
Davies, B. Henrissat, and B. Svensson, The Royal Society of Chemistry, Cambridge, 1999, p. 3.
[6] V. H. Lillelund, H. H. Jensen, X. Liang, M. Bols, Chem. Rev. 2002, 102, 515; E. S. H. El Ashry, N. Rashed,
A. H. S. Shobier, Pharmazie 2000, 55, 251, 331, 403; N. Asano, J. Enzym. Inhib. 2000, 15, 215; N. Asano,
R. J. Nash, R. J. Molyneux, G. W. J. Fleet, Tetrahedron: Asymmetry 2000, 11, 1645; G. Legler, Adv.
Carbohydr. Chem. Biochem. 1990, 48, 319.
[7] M. N. Namchuk, S. G. Withers, Biochemistry 1995, 34, 16194.
[8] K.-R. Roeser, G. Legler, Hoppe-Seyler×s Z. Physiol. Chem. 1980, 361, 321; K.-R. Roeser, G. Legler,
Biochim. Biophys. Acta 1981, 657, 321.
[9] N. Panday, Y. Canac, A. Vasella, Helv. Chim. Acta 2000, 83, 58.
[10] M. Terinek, A. Vasella, Helv. Chim. Acta 2003, 86, 3482.
[11] T. D. Heightman, A. Vasella, K. E. Tsitsanou, S. E. Zographos, V. T. Skamnaki, N. G. Oikonomakos, Helv.
Chim. Acta 1998, 81, 853.
[12] M. Bˆhm, E. Lorthiois, M. Meyyappan, A. Vasella, Helv. Chim. Acta 2003, 86, 3787.
¬
[13] V. M.-A. Ducros, D. L. Zechel, G. N. Murshudov, H. J. Gilbert, L. Szabo, D. Stoll, S. G. Withers, G. J.
Davies, Angew. Chem., Int. Ed. 2002, 41, 2824.
[14] E. Lorthiois, M. Meyyappan, A. Vasella, Chem. Commun. 2000, 1829.
[15] Y. Kishi, M. Aratani, H. Tanino, T. Fukuyama, T. Goto, S. Inoue, S. Sugiura, H. Kakoi, J. Chem. Soc., Chem.
Commun. 1972, 64.

Helvetica Chimica Acta Vol. 86 (2003)
3835
[16] J. P. Kutney, G. H. Bokelman, M. Ichikawa, E. Jahngen, A. V. Joshua, P.-H. Liao, B. R. Worth, Can. J.
Chem. 1977, 55, 3227.
[17] M. Kawana, S. Emoto, Tetrahedron Lett. 1975, 39, 3395.
[18] A. R. Plant, J. E. Oliver, M. L. Patchett, R. M. Daniel, H. W. Morgan, Arch. Biochem. Biophys. 1988, 262,
181.
[19] G. Legler, Biochim. Biophys. Acta 1978, 524, 94.
[20] M. Kleban, P. Hilgers, J. N. Greul, R. D. Kugler, J. Li, S. Picasso, P. Vogel, V. J‰ger, ChemBioChem 2001, 2,
365.
[21] G. M. Sheldrick, −SHELXS-96, Program for the Solution of Crystal Structures×. Universit‰t Gˆttingen,
Germany, 1996.
[22] G. M. Sheldrick, −SHELXL-97, Program for the Refinement of Crystal Structures×. Universit‰t Gˆttingen,
Germany, 1997.
[23] C. K. Johnson, −ORTEP II, A Fortran Thermal-Ellipsoid Plot Program. Report ORNL-5138×, Oak Ridge
National Laboratory, Oak Ridge, Tennessee, USA, 1976.
[24] W. Adam, J. Bialas, L. Hadjiarapoglou, Chem. Ber. 1991, 124, 2377.
[25] A. Altomare, B. Carrozzini, G. L. Cascarano, C. Giacovazzo, A. Guagliardi, A. G. Moliterni, R. Rizzi, −A
package for crystal structure solution by direct methods and refinement×, Istituto di Ricerca per lo Sviluppo
di Metodologie Cristallografiche, CNR, Campus Universitario, Via Orabona 4, 70125 Bari, Italia, 1997.
[26] S. Aoyagi, R. Tanaka, M. Naruse, C. Kibayashi, J. Org. Chem. 1998, 63, 8397.
[27] I. H. Segel, −Enzyme Kinetics×, Wiley, New York, 1975.
[28] C. Bl¸chel, C. V. Ramana, A. Vasella, Helv. Chim. Acta 2003, 86, 2998.
[29] S. Vonhoff, K. Piens, M. Pipelier, C. Braet, M. Claeyssens, A. Vasella, Helv. Chim. Acta 1999, 82, 963.
[30] C. K. Jue, P. N. Lipke, J. Biochem. Biophys. Methods 1985, 11, 109.
Received August 22, 2003