Stereoelectronic substituent effects in polyhydroxylated piperidines and hexahydropyridazines

Article abstract of DOI:10.1002/1521-3765(20020301)8:5<1218::AID-CHEM1218>3.0.CO;2-X

From the pK_a values of the conjugate acids of a large series of hydroxylated piperidines and hexahydropyridazines, a consistent difference in basicity was found between stereo-isomers having an axial or equatorial hydroxyl (OH) group either β o

Full text of DOI:10.1002/1521-3765(20020301)8:5<1218::AID-CHEM1218>3.0.CO;2-X

FULL PAPER
Stereoelectronic Substituent Effects in Polyhydroxylated Piperidines
and Hexahydropyridazines
Henrik Helligs˘ Jensen, Laila Lyngbye, Astrid Jensen and Mikael Bols*[a]
Abstract: From the pKa values of the
conjugate acids of a large series of
hydroxylated piperidines and hexahy-
dropyridazines, a consistent difference
in basicity was found between stereo-
isomers having an axial or equatorial
hydroxyl (OH)group either b or g to the
amine. Compounds with an equatorial
OH group in the 3-position were 0.8 pH
units more acidic than otherwise identi-
cal compounds with an axial OH group,
whilst compounds with an equatorial
OH group in the 4-position relative to
the amine were 0.4 pH units more acidic
than the corresponding compound with
an axial OH. A similar effect was
observed for the COOMe substituent.
The difference in electron-withdrawing
power of axial and equatorial substitu-
ents was explained by a difference in
charge dipole interactions in the two
systems. Since this stereoelectronic sub-
stituent effect causes differences in ba-
sicity in different conformers, certain
piperidines and hexahydropyridazines
were found to change conformation
upon protonation. A method for pre-
dicting the pKa of piperidines which
takes stereochemistry into account is
described.
Keywords: amines
¥
basicity
¥
glycosidase inhibitors ¥ piperidines
¥ stereoelectronic effects
Introduction
MeOOC
HO
HO
N
N
The importance of being able to predict the effect of
substitution on the reactivity of organic molecules is well
recognized and various substituent constants of the Hammett
or Taft type can be used to achieve this goal. The prediction of
the basicity of amines using Taft substituent constants has
been refined by Clark and Perrin.[1] In the Taft approach the
pKa of a protonated amine is calculated from Equation (1),
where A and B are constants that depend on whether the
amine is primary, secondary, or tertiary; s* is a substituent
constant. In Clark and Perrin×s approach the BSs* term is
merged into one base-weakening substituent constant. A
modified version of this method has been used by Inouye to
predict the pKa of aminosugar derivatives.[2]
COOMe
2 (pKa 8.2)
1 (pKa 9.2)
effect was particularly unpredictable and that its base-weak-
ening effect could vary from 0.4 to 1.2 pH units when b to the
amine. They suggested intramolecular hydrogen bonding to
be the cause of such anomalies.
Thus in the series of b- and g-hydroxyamines 3 14 a
systematic differences in pKa between stereoisomers was
observed, but only in the compounds 6, 12, and 14 was
intramolecular hydrogen bonding observed in the IR spec-
trum.[4, 5]
These methods take through-space effects into account, but
not the configuration and conformation of the molecule. Thus
the methyl esters of both ecgonine (1)and pseudoecgonine
(2)are predicted to have the same p Ka (7.8)even though the
OH
HO
HO
N
N
H
N
H
N
H
H
OH
pKa ˆ A À BSs*
(1)
6 (pKa 9.51)
4 (pKa 9.19)
5 (pKa 8.42)
3 (pKa 8.74)
observed pKa values differ by 1 unit (pKa(1) ˆ 9.2, pKa(2) ˆ
OH
OH
8.2).[3] Clark and Perrin noted that the OH group substituent
HO
HO
N
H
N
H
N
H
N
H
10 (pKa 9.53)
[a] Prof. M. Bols, H. H. Jensen, L. Lyngbye, A. Jensen
Department of Chemistry, University of Aarhus
Langelandsgade 140, 8000 Aarhus (Denmark)
Fax : (‡45)8619-6199
9 (pKa 9.2)
7 (pKa 7.87)
8 (pKa 8.44)
HO
HO
N
N
N
H
N
H
HO
H
E-mail: mb@chem.au.dk
H
OH
Supporting information for this article is available on the WWW under
http://www.wiley-vch.de/home/chemistry/ or from the author.
14 (pKa 10.2)
13 (pKa 8.73)
11 (pKa 8.6)
12 (pKa 9.87)
1218
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0805-1218 $ 17.50+.50/0
Chem. Eur. J. 2002, 8, No. 5

1218 1226
Table 1. pKa values of protonated piperidines at 258C.
It is well known that substituents on a six-membered ring in
the chair conformation prefer the equatorial position, as there
would be an unfavorable 1,3-diaxial steric interaction between
axial substituents. It is much less recognized that polar
Compound
Structure
pKa
Obs.[a]
11.2[2]
Calcd[b]
Calcd[c]
11.2
piperidine
10.7
NH2
NH2
substituents prefer an axial orientation in piperidinium
12]
ions.[6
Thus 3-hydroxypiperidine, upon protonation, shifts
10.5[8]
10.3[8]
10.5
10.2
OOC
isonipecotic acid
from a predominantly equatorial to a predominantly axial
orientation.[6] An even more spectacular effect is seen with
5-fluoropiperidine-3-carboxylic acid 15 (Scheme 1). Upon
OOC
nipecotic acid
NH2
NH2
4
1
protonation 15 flips from the C1 conformation to the C4
OOC
16
17
18
10.0[8]
9.3[8]
10.0
9.2
OH
OOC
F
COO
F
OOC
HO
NH2
NH2
NH
HO
COO
15
10.0[8]
10.0
Scheme 1. Conformational change of 15 upon protonation.
NH2
NH2
OOC
HO
19
20
21
9.4[8]
8.9[8]
9.7[8]
9.6
8.9
9.7
conformation, whilst the corresponding nonfluoro derivative
piperidine-3-carboxylic acid is predominantly in the 4C1
conformation regardless of pH.[11] The axial preference of
polar substituents has been explained by electrostatic inter-
actions, charge dipole interactions, the gauche effect,[13] or
even a fluoro-directing effect. However, no connection
between the conformation and basicity of a piperidine has
apparently been made.
We recently observed differences in the basicity of diaster-
eomeric azasugars much like those in 1 14.[14, 15] Since these
differences in basicity appear to be systematic and since the
examples of 3 14 suggest that hydrogen bonding alone
cannot account for the difference, we have studied the basicity
of a series of polyhydroxylated or otherwise substituted
piperidines and hexahydropyridazines, to investigate to what
extent a substituent×s electronic effect is related to its
stereochemistry. We found that the pKa differences for 1 14
can be explained to a large extent by a difference in the
electron-withdrawing power of axial and equatorial substitu-
ents. We also observed that the conformational changes in
protonated piperidines are related to this dependence of
electron withdrawal on geometry, and that a conformational
change will occur when the difference in charge stabilization
of axial and equatorial substituents is large enough to
overcome the steric bias associated with axial substituents.
COO
COO
COO
HO
NH2
NH2
NH2
AcHN
22
23
24
25
8.6[8]
9.6
8.6
9.7
AcHN
HO
OH
NH2
OH
HO
10.1
9.0
10.1
9.0
NH2
OH
HO
NH2
HO
HO
26
27
28
9.3
8.8
9.3
9.2
8.8
9.3
NH2
NH2
HO
HO
HO
OH
NH2
NH2
OH
OH
29
30
31
8.9
9.5
8.2
8.9
9.6
8.1
MeOOC
OH
OH
NH2
Results and Discussion
OOC
OH
COOMe
We have measured the pKa of a large series of polyhydroxy-
lated piperidines (Table 1, which includes some literature pKa
values)and hexahydropyridazines (Table 2,) some of which
had to be synthesized.
MeOOC
NH2
OH
COOMe
32
9.1
9.1
NH2
COOMe
HO
Syntheses: Compounds (Æ)-23,[16] 28,[17] 31,[18] (Æ)-32,[18] (Æ)-
34,[19] (Æ)-35,[19] (Æ)-36,[19] (Æ)-37,[19] 38,[20] 40,[36] 41,[18] (Æ)-
42,[22] (Æ)-43,[22] 44,[19] 45,[19] (Æ)-46,[19] (Æ)-65,[23] 66,[14] and
(Æ)-67[23] were synthesized by previously published methods.
Compounds 47, 48, 52, and 54[24] were kindly provided by
Professor Asano, and 53 by Professor St¸tz. The synthesis of
compounds 61 63 and 84 will be published later.
COO
33
34
9.1
9.0
8.8
9.0
MeOOC
NH2
OH
HO
H2N
NH2
Chem. Eur. J. 2002, 8, No. 5
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0805-1219 $ 17.50+.50/0
1219

M. Bols et al.
FULL PAPER
Table 1. (cont.)
Table 1. (cont.)
Compound
Compound
Structure
pKa
Structure
pKa
Obs.[a]
9.1
Calcd[b]
Calcd[c]
Obs.[a]
Calcd[b]
Calcd[c]
7.7
OH
OH
HO
HO
NH2
1-deoxymannonojiri-
mycin (54)
7.5, 7.2[28]
7.6
HO
OH
HO
35
9.1
NH2
NH2
H2N
HO
HO
HO
5.7[2]
5.3[2]
5.5
4.8
5.9
5.0
OH
55
HO
NH2
O
NH
36
37
6.9
7.4
HO
O
OH
NH2
HO
nojirimycin (56)
OH
O
HO
NH2
HO
O
NH
HO
HO
OH
57
58
5.1[26]
5.6[27]
5.6
5.6
5.0
5.0
NH2
HO
HO
OH
HO
38
39
7.9
7.9
8.3
HO
NH2
NH2
OH
NH2
HO
OH
HO
HO
HO
8.3[27]
HO
[a] See ref. or this paper. [b] pKa values have been calculated from the
formula: pKa ˆ 10.7 À Sss (ss values from Table 4). [c] Calculated from the
formula in ref. [2].
OH
HO
HO
isofagomine (40)
8.4
8.8
8.4
8.8
9.3
9.3
NH2
NH2
NH2
OH
HO
HO
41
Table 2. pKa values of protonated hexahydropyridazines at 258C deter-
mined from titration curves. The site of protonation is shown arbitrarily.
OH
[a]
[a]
[b]
Compound
Structure
pKa pKaN1
pKaN2
Calcd pKa
42
43
44
HO
9.2
9.4
9.3
9.2
9.6
9.3
9.3
9.3
NH
NH2
HO
59
7.9 7.3
7.3
7.6
OH
HO
H
N
HO
HO
60
61
5.5 5.4
6.4 6.3
5.4
5.3
5.7
6.3
NH2
NH2
HO
HO
HO
H
OH
N
HO
HO
10.0
NH2
NH2
NH2
OH
H
HO
OH
OH
HO
N
62
6.8 6.7
6.1
6.8
NH2
45
46
9.6
9.2
9.7
9.2
10.0
10.0
HO
HO
HO
HO
H
N
63
6.2 5.6
5.5 5.5
5.3 5.0
6.0
5.5
4.7
6.1
5.8
5.2
NH2
NH2
NH2
H
N
HO
HO
HO
HO
64
NH2
HO
fagomine (47)
8.1
8.1
8.5
8.4
5.8
8.7
8.7
8.6
HO
HO
H
N
HO
azafagomine (65)
HO
NH2
HO
HO
48
49
50
8.5
OH
H
HO
HO
NH2
NH2
N
HO
66
67
5.7 5.4
6.0 5.8
5.5
5.1
5.8
5.9
NH2
HO
HO
8.4[28]
5.85[29]
OH
HO
HO
NH
NH2
F
HO
OH
HO
NH2
NH2
HO
HO
[a] Calculated from Equation (5). [b] Calculated from Equation (4).
HO
F
51
5.75[29]
5.7
6.8
7.6
HO
Compound (Æ)-24 was made by a simple modification of
the synthesis of (Æ)-23.[16] Thus, commercially available (Æ)-
68 was converted into (Æ)-69 by global reduction as previ-
ously described and then deprotected with aqueous HCl to
provide (Æ)-24 (Scheme 2). Synthesis of (Æ)-25 was analogous
to that of 20 by the Krogsgaard Larsens group:[8] 70 was
substituted with amine and esterified to produce 71 (overall
yield 60%; Scheme 2). The ester was reduced with LiAlH4 to
HO
HO
1-deoxynojirimy-
cin (52)
6.7, 6.7[30]
6.3[31], 6.6[32]
,
7.7
7.7
HO
NH2
NH2
HO
HO
HO
galactostatin (53)
7.5, 7.1[26]
HO
HO
1220
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0805-1220 $ 17.50+.50/0
Chem. Eur. J. 2002, 8, No. 5

Stereoelectronic Substituent Effects
1218 1226
diacetate (Æ)-75 with NaOMe/MeOH and hydrogenolysis
gave (Æ)-27.
OEt
OH
OH
HO
HO
O
O
HCl
N
NH
NH
Boc
Compounds 29, 30, and 33 were made by deprotection of
the known N-Boc-protected derivatives.[18]
68
24
69
Compounds (Æ)-60 and (Æ)-64 were synthesized by an
adaptation of the synthesis of (Æ)-azafagomine (65;
Scheme 3).[23] Thus the known Diels Alder adduct 76[25]
was converted to (Æ)-60 by epoxidation with trifluoromethyl-
methyldioxirane, which gave the epoxide 76a in 86% yield,
which was treated with acetic anhydride/acetic acid in the
presence of BF3 ¥ Et2O to yield (Æ)-78 (94%). This diacetate
was deprotected by deacetylation with NaOMe/MeOH, then
hydrazinolyzed with neat hydrazine hydrate at 1008C, giving
(Æ)-60 in 56% yield over two steps.
Br
COOH
H2N
COOMe
1) NH3
LiAlH4
2) MeOH
HCl, 2 d
N
N
70
71
H2N
HO
OH
OH
NaNO2
H2SO4
N
N
The known adduct (Æ)-77[23] was subjected to a similar
sequence of reactions to obtain (Æ)-64 (Scheme 3). In this
case epoxidation gave a 2:1 mixture of the trans- and cis-
epoxides (Æ)-77a and (Æ)-77b in 93% yield. This mixture was
unseparable, but since cis- and trans-epoxides of this type
appear to be preferentially attacked at different carbon atoms
(C3 and C4, respectively)the mixture was believed to yield
mainly the desired isomer. Indeed, acetolysis of the mixture of
(Æ)-77a and (Æ)-77b with Ac2O/BF3 ¥ Et2O in acetic acid gave
mainly diacetate (Æ)-79 containing some of the 2,3-cis-3,4-
trans isomer (Æ)-79a, in a ratio of 6:1. After deprotection as
above, (Æ)-64 was obtained.
72
73
OH
H2
HO
NH
Rh/C
40 atm
50 o
C
25
Scheme 2. Synthesis of piperidines 24 and 25.
give 72 (yield 93%), which was diazotized with NaNO2 and
H2SO4 to give 73 (yield 83%). This pyridine was hydro-
genated at 40 atm and 508C with a Rh catalyst, giving 25
(yield 77%)as a racemate.
Compounds (Æ)-26 and (Æ)-27 were both made from the
known alkene 74. cis-Diol (Æ)-26 was obtained by dihydrox-
ylation of 74 with OsO4/NMO followed by conversion to (Æ)-
26 by hydrogenolysis (Scheme 3). trans-Diol (Æ)-27 was
pKa measurements: The pKa values were obtained from pH
curves made by titration at 258C (representative examples
have been included in the Supporting Information). The
uncertainty in our determinations by titration at 258C is about
Æ0.1 pH units; therefore the pKa values are given to only one
decimal place. The literature pKa values for galactostatin (53)
and 1-deoxymannonojirimycin (54)(7.1 and 7.2, respective-
ly)[26] did not fit the values we would expect (see below)so
they were remeasured; a pKa of 7.5 was obtained for both
compounds. The pKa of 1-deoxynojirimycin (52)was also
ambiguous since three different values (6.3, 6.6, and 6.7)have
been reported; our measurement confirmed it to be 6.7
(Table 1).
HO
1) OsO4
NMO
HO
N
NH
CO2Bn
2)
H2/Pd
74
26
1) NaOMe
HO
HO
1)
AcO
AcO
MCPBA
NH
N
74
2)
H2/Pd
CO2Bn
2) Ac2O
BF3•Et2O
27
75
For epimeric pairs of 4-hydroxypiperidines where the 4-OH
group is unquestionably either axial or equatorial, one finds
that the axial epimer is more basic. Thus in the epimeric pairs
3/4, 7/8, 9/10, 18/19, 23/24, 40/41, 44/45, and 47/48 the pKa of
the axial isomer is 0.3 0.6 pH units higher than the equatorial
isomer with an average difference of about 0.4 pH units. This
corresponds to a difference in protonation energies (DDG)of
2.3 kJmolÀ1. Thus, the basicity of the axial isomer appears to
be increased consistently. Intramolecular hydrogen bonding is
unlikely to cause this effect in this case, in view of the
remoteness of the OH group and since it was not observed in
the IR spectra of 3 4 and 7 10.[4, 5] Similarly in the epimeric
pairs of b-hydroxypiperidines 5/6, 11/12, 13/14, 16/17, 40/42,
52/53, and 52/54 the axial epimer is 0.7 1.47 pH units more
basic than the equatorial one with a typical difference of
0.8 pH units. This corresponds to an average DDG of
4.6 kJmolÀ1. In this case intramolecular hydrogen bonding
could be affecting the basicity of the axial isomers since it has
been observed in 6, 12, and 14.
R
R
O
N
O
EtOAc
25 oC
1) Me(CF3)CO2
N
N
N
Ph
N Ph
+
N
2) Ac2O,
BF3•Et2O
O
O
76: R = H
77: R = Me
R
O
R
AcO
AcO
H
1) NaOMe
2) NH2NH2
N
N
HO
HO
N
Ph
NH
N
O
60: R = H
64: R = Me
78: R = H
79: R = Me
Scheme 3. Synthesis of piperidines 26 and 27 and hexahydropyridazines 60
and 64.
obtained by epoxidation of 74 with m-chloroperbenzoic acid
(MCPBA); the epoxide obtained was opened with Ac2O/
BF3 ¥ Et2O in acetic acid; deacetylation of the resulting
Chem. Eur. J. 2002, 8, No. 5
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0805-1221 $ 17.50+.50/0
1221

M. Bols et al.
FULL PAPER
Table 3. Heat of formation of protonated and unprotonated piperidines obtained from AM1 or PM3 calculations
using MOPAC in CS6Chem3D Pro, Version 3.5.1.
The basicity of amines is
influenced
by
inductive
À1
À1
Compound
DHf (AM1)[kJmol
]
DHf (PM3)[kJmol
]
(through-bond)effects, field
(through-space)effects, reso-
nance, solvation, steric hin-
drance, and internal hydrogen
bonding.[1] However, the basic-
ity differences observed here
cannot be caused by through-
bond or resonance effects, and
steric hindrance is expected to
2-hydroxypiperidine (axial 2-OH)
2-hydroxypiperidine (equat. 2-OH)
À 280.7
À 277.8
À 261.9
À 260.2
‡
2-hydroxypiperidine, H (axial 2-OH)349.9
392.1
387.6
‡
2-hydroxypiperidine, H (equat. 2-OH)346.5
3-hydroxypiperidine (axial 3-OH)
3-hydroxypiperidine (equat. 3-OH)
À 280.7
À 280.7
À 251.8
À 253.1
‡
3-hydroxypiperidine, H (axial 3-OH)347.8
399.7
413.6
‡
3-hydroxypiperidine, H (equat. 3-OH)364.9
4-hydroxypiperidine (axial 4-OH)
4-hydroxypiperidine (equat. 4-OH)
À 279.4
À 281.1
À 253.5
À 253.9
be equal in the compounds
being compared. Solvation nor-
‡
4-hydroxypiperidine, H (axial 4-OH)354.9
406.0
414.4
‡
4-hydroxypiperidine, H (equat. 4-OH)365.4
mally affects amine basicity
when the solvent shell round
the amine is changed by sub-
stitution. Since the substituent changes we have studied in
many cases are relatively remote from the amine, solvation
does not appear to be the cause of the differences in base
strength. As explained above, internal hydrogen bonding may
play a role in some but not all of the basicity differences;
therefore electronic effects must account for a significant
fraction of them. They can be explained as a charge dipole
effect of the polar substituents (Figure 1). Each OH group (or
À
piperidine with the OH group in either the axial or the
equatorial position (Table 3). Regardless of the position of the
OH group, the difference in predicted heat of formation
(DHf)of the unprotonated piperidine with an axial or
equatorial OH group is relatively small. However, for the
3-hydroxypiperidinium ion the axial OH is 14 17 kJmolÀ1
more stable than the equatorial 3-OH, and for the 4-hydroxy-
piperidinium ion the axial OH is preferred to the equatorial
one by 8 10 kJmolÀ1. This shows that a difference in electro-
static interactions in the two molecules is likely to be the cause
of the basicity differences observed experimentally. Hydrogen
bonding and solvation effects must therefore be expected to
play a minor role in governing the basicity differences.
other polar substituent)has a dipole moment along the C
O
bond that will interact with the protonated amine. If the OH
group is directed away from the amine the C O dipole will
destabilize the positive charge, whereas if it is perpendicular it
will not affect the amine. In a piperidine axial OH groups in
À
the 3- and 4-positions have C O bonds that are close to
perpendicular to the amine, whilst equatorial OH groups have
À
Prediction of pKa: From the pKa of a series of compounds
with and without OH groups in various positions (Table 1)it
was possible to determine the average regio- and stereo-
chemical effect on piperidine basicity by introducing an OH
group in a given position. Thus introduction of an equatorial
OH group in the b position decreases the pKa of the
piperidinium ion by 1.3 units, whereas introducing an axial
b-OH decreases the pKa by 0.5 units. Similarly the average
pKa values of other substituents were determined (Table 4).
From them it is possible to predict the pKa of a substituted
piperidine from pKa ˆ 10.7 À Sss, where ss is the substituent
constant. The pKa was predicted for piperidine, isonipocetic
acid, nipocetic acid, and 16 58 to check the consistency of the
ss values assigned (Table 1).
C O bonds directed away from it.
Alternatively the difference in electron-withdrawing effect
may be explained by differences in orbital overlap between
the polar substituent and the bond in the piperidine (Fig-
ure 1). In the equatorial isomer 40 the antibonding orbital of
OH
OH
HO
HO
HO
HO
NH2
NH2
40
OH
41
OH
HO
HO
HO
HO
NH2
NH2
The ss values can also be used to predict the pKa values of
hexahydropyridazines. However, to predict the overall pKa it
is necessary to take into account the two ways in which these
compounds can become protonated. In this case Equa-
H
H
41
40
Figure 1. Explanations for the different substituents effect of axial and
equatorial 4-OH groups.
tions (2) (4) apply, where HA and HA2 are the products
1
of protonation at either nitrogen; Ka1 and Ka2 are the acidity
constants of the two nitrogen atoms; and pKa1 and pKa2 were
predicted from Equation (5)using substituent values from
Table 4.
the 4-C O may overlap with the s orbital of the C2 C3 bond
À
or the C5 C7 bond, thereby drawing electrons away from
nitrogen. In the axial isomer 41 the antibonding orbital will
À
À
overlap with the orbitals of the C3 H and C5 H bonds,
thereby not reducing the electron density around nitrogen.
‡
‡
Kaoverall ˆ [H ][A]/[HA] ˆ [H ][A]/([HA1] ‡ [HA2])(2)
1/Ka ˆ 1/Ka1 ‡ 1/Ka2
Theoretical calculations: To gain support for the theory that
electrostatic interactions were causing the stereoelectronic
effects, we carried out semiempirical calculations on the
protonated and unprotonated forms of 2-, 3-, and 4-hydroxy-
(3)
(5)
pKa ˆ log(1/Ka1 ‡ 1/Ka2)(4)
pKa ˆ 7.3 À Sss
1222
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0805-1222 $ 17.50+.50/0
Chem. Eur. J. 2002, 8, No. 5

Stereoelectronic Substituent Effects
1218 1226
[a]
Table 4. Average effect (ss [pH units])of various substituents on the p Ka
of a piperidine.
and between 81 and 83. This suggests that a change in the
conformation may occur when a piperidine is protonated,
provided that the stereoelectronic substituent effects are
powerful enough to overcome the unfavorable steric inter-
actions associated with 1,3-diaxial substituents, which would
explain the conformational change of 3-fluoropiperidine-5-
carboxylic acid 15 upon protonation.[11] Fluorine is strongly
electronegative and a strong charge dipole interaction will
occur in the protonated piperidine, which will tend to be
eliminated by moving the fluorine to the axial position. At the
same time fluorine is small and has almost no steric
preference for the equatorial position (Table 5).[33]
Substituent
a-position
b-position
g-position
H
0
ꢀ 2
ꢀ 2
0
1.3
0.5
0
0.6
0.2
OH (eq)
OH (ax)
CH2OH (eq)0.7
CH2OH (ax)
Me (eq)
F (eq)
NH2 (eq)
NH2 (ax)
COOÀ(eq)
COOMe (eq)
COOMe (ax)
NHAc (eq)
NHAc (ax)
0.4
0.5
[b]
À 0.1[b]
À 0.1[b]
[b]
2.3
1.7[b]
[b]
[b]
0.7
0.6
0.5
1.2
0.2
1.6
0.2
[b]
[b]
Table 5. Energy difference between equatorial and axial substituents in a
cyclohexane (DGsteric)and a protonated piperidine ( DGelectrostatic).
[b]
0.5
[a]
[a] Determined from the formula pKa ˆ 10.7 À Sss. [b] Based on the pKa of
Group
DGsteric [kJmolÀ1
]
DGelectrostatic [kJmolÀ1
]
a single compound.
3-OH
4-OH
Me
3-COOMe
F
ꢀ 4
ꢀ 4
7.3
5.4
1
4.6
2.3
ꢀ 0
It is clear from Tables 1 and 2 that the predicted pKa values
of all the substituted piperidines and most of the hexahydro-
pyridazines fit remarkably well with experimental values. The
difference between prediction and measured value is nor-
mally within 0.1 pH units, which is also the experimental
uncertainty on the pKa measurements with the titration
method used. The success of these predictions supports the
hypothesis that electronic effects and not hydrogen bonding
or solvation are the cause of the observed changes in basicity.
Only the unsubstituted compounds piperidine and hexahy-
dropyridazine are poorly estimated by this method of
calculation.
5.6
[a] From ref. [33].
One may predict that certain hydroxylated piperidines
change chair conformation upon protonation. The free energy
difference between an equatorial and an axial OH in a
cyclohexane is about 4 kJmolÀ1.[33] This is less than the
electrostatic free energy difference (4.6 kJmolÀ1)between
an axial and an equatorial b-OH group in a piperidinium ion,
but more than the electrostatic energy difference (2.3 kJ
molÀ1)for a g-OH (Table 5). Thus 4-hydroxypiperidine should
be predominantly in the 4C1 conformation regardless of
protonation, whereas 3-hydroxypiperidine should have a
small preference (0.6 kJmolÀ1)for 1C4 conformation when
protonated, and this has indeed been observed.[6] Also, some
of the compounds in Tables 1 and 2 might be expected to
change conformation. Thus 60 changes from predominantly
Effect on conformation: If a polar substituent destabilizes a
piperidinium ion more when it is equatorial than when it is
axial, this must influence the conformational equilibrium of
the molecule. This can be seen in the conformational
equilibria and acid base reactions of 4-hydroxypiperidine
(Scheme 4). 4-Hydroxypiperidine in the 4C1 conformation
4C1 conformation to predominantly C4 upon protonation of
1
+ H+
the amine (Scheme 5). The coupling constant J3ax,4,in water,
changed from 9.6 Hz to 4.5 Hz upon protonation, from which
it can be estimated[37] that the conformer ratio goes from 9:1
to 1:4. Piperidine 27, on the other hand, does not change its
predominant conformation in water, although the equilibrium
HO
HO
NH2
Kc'
NH
Ka
80
81
Kc
1
is shifted toward more C4 conformer. The similar coupling
+ H+
constant, J2ax,3, changed from 8.4 Hz (pH 11)to 6.2 Hz (pH 1),
from which it can be estimated[37] that the conformer ratio
goes from 4:1 to 1:1. In these cases the predicted values of
DGsteric for two equatorial OH, and DGelectrostatic for a 3-OH and
a 4-OH interaction, would be 8 kJmolÀ1 and 6.9 kJmolÀ1,
NH
NH2
Ka'
HO
HO
82
83
Scheme 4. Protonation and conformational change of 4-hydroxypiperi-
dine. Kc ˆ [82]/[80], K'c ˆ [83]/[81], Ka ˆ [80][H ]/[81], K'a ˆ [82][H ]/[83].
‡
‡
4
respectively. Therefore the predicted preference for the C1
conformation after protonation would still be 1.1 kJmolÀ1.
The more profound change in conformation of 60 than of 27 is
probably due to DGsteric being smaller in the case of
hexahydropyridazines than the values in Table 5, as one 1,3-
diaxial interaction is absent because of the presence of the
extra nitrogen atom. Thus one of the nitrogen atoms in 60 will
not have an axial H atom, whereas protonated 27 will have a
CH2 in that position.
(80)can be converted into the 1C4 piperidinium ion 83 in two
ways, through either 81 or 82 depending on the sequence of
conformational change and protonation. Since both routes
result in the same equilibrium, Kc'/Ka ˆ Kc/Ka' and therefore
Ka'/Ka ˆ Kc/Kc', which shows that a difference in the acidity
constant between 80 and 82 will be reflected in similar
differences in conformational equilibria between 80 and 82
Chem. Eur. J. 2002, 8, No. 5
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0805-1223 $ 17.50+.50/0
1223

M. Bols et al.
FULL PAPER
H+
conformers to have different basicity, which can cause them to
change conformation upon protonation. The effect must also
be expected to play a role in many other systems. Thus it can
explain the difference in basicity of cocaine derivatives 1 and
2. It should also affect the reactivity of compounds in
reactions where positive charge is developed in the transition
state. Thus as we demonstrated previously,[15] the rate of acidic
glycoside hydrolysis is correlated with piperidine basicity; that
is, glycosides with axial OH groups are hydrolyzed faster than
glycosides with equatorial OH groups because the equatorial
OH makes it more difficult to build positive charge on C1.
Other reactions may be found to be affected similarly.
H
HO
NH
N
HO
HO
NH2
NH
HO
60
H+
HOOC
HO
HOOC
HO
HO
OH
NH
NH
NH
NH2
84
HOOC
HO
HO
HO
HO
NH2
NH2
27
85
H+
MeOOC
MeOOC
HO
NH2
Experimental Section
NH
COOMe
COOMe
HO
General: All reactions were carried out under an inert atmosphere in
preheated glass equipment. Solvents were distilled under anhydrous
conditions. Thus THF was distilled from sodium/benzophenone and used
directly. All reagents were used as purchased, without further purification.
Columns were packed with silica gel 60 (230–400 mesh)as the stationary
phase. TLC glass plates (Merck 60, F254)were visualized by spraying with
cerium ammonium sulfate (1%)and phosphomolybdic acid (1.5%)in
H2SO4 (10%)and heating until colored spots appeared.
32
Scheme 5. Conformational change of 32, 60 and 84 upon protonation.
The same phenomenon was seen with the carboxylic acids
84 and 85. Compound 84 was observed to flip to predom-
inantly C4 conformation when the nitrogen was protonated,
1
regardless of whether the carboxylate was protonated or not.
It had an estimated conformer ratio in water of 9:1 at pH 11
and 1:4 at pH 1. However, the known compound 85 is
predominantly in the 4C1 conformation as the hydrochlor-
ide.[20]
(Æ)-(3,4-cis)-3-Hydroxymethyl-4-hydroxypiperidine (24): Piperidone (Æ)-
68 was converted to N-tert-butoxycarbonyl-3-ethoxycarbonyl-4-piperidone
[21]
(68a)as previously described.
Then (Æ)-68a (300 mg, 1.1 mmol)was
dissolved in absolute ethanol (30 mL), and DIEA (140 mg, 1.1 mmol) was
added. A flow of nitrogen was bubbled through the solution for 5 min, then
10% Pd on carbon (200 mg)was added. The mixture was hydrogenated for
24 h (40 atm, 508C)before filtering through Celite and evaporating to
dryness. To get rid of the DIEA the remaining oil was dissolved in ether
Clear evidence for the change of piperidine 32 from
1
predominantly 4C1 to predominantly C4 upon protonation
(20 mL)and a saturated solution of KHSO (20 mL)was added. The
4
in water was seen from the coupling constant J4,5, which was
7.6 Hz at pH 11 but 4.8 Hz at pH 2. Based on these values the
conformer ratios 4C1/1C4 for 32 were estimated[37] to be 2:1 at
pH 11 and 1:2 at pH 2.
Another way of explaining the observed conformational
changes is that the two conformers of a hydroxylated
piperidine or hexahydropyridazine have different basicity,
and protonation will cause the amine to prefer the less acidic
conformation.
Recently it was reported by Pinto×s group that a trihy-
droxylated six-membered sulfonium salt adopted a confor-
mation with all three OH groups axial, contrary to expect-
ations.[38] This observation can also be explained using the
stereoelectrostatic substituent effects postulated in this paper.
The gauche effect has been used to explain why 3-hydrox-
ypiperidinium salts prefer a conformation with the O-
substituent axial.[13] It is possible that the gauche effect can
account for some of the conformational changes observed
here, but it cannot explain that of 32, which does not have two
heteroatoms gauche.
aqueous phase was extracted with diethyl ether (3 Â 20 mL), then the
combined organic phases were dried (MgSO4)and evaporated to crude 3,4-
cis-N-tert-butoxycarbonyl-3-ethoxycarbonyl-4-hydroxypiperidine (68b).
Without further purification the resulting alcohol (Æ)-68b (300 mg,
1.1 mmol)was dissolved in dry THF (50 mL)and treated with LiBH
4
(24 mg, 1.1 mmol). The solution was refluxed for 10 min and cooled to 08C
before a solution of saturated KHSO4 (40 mL)was added slowly. The two
layers were separated, and the aqueous phase was extracted with AcOEt
(2 Â 30 mL). The combined organic phases were washed with brine
(40 mL), dried over MgSO4 and evaporated to dryness, leaving crude
(Æ)-69.
The diol 69 was deprotected by stirring it in hydrochloric acid (4m, 10 mL)
overnight. Removal of the solvent at reduced pressure gave (Æ)-24.[39]
1H NMR (D2O): d ˆ 4.17 (q, J ˆ 3.0 Hz, 1H, H4), 3.67 (dd, J(5,5'a) ˆ 6.6,
J(5'a,5'b) ˆ 11.0 Hz, 1H, H5'a), 3.54 (dd, J(5,5'b) ˆ 7.4 Hz, 1H, H5'b), 3.06 3.20
(m, H2eq, H2ax, H6eq), 2.93 (t, J(6ax,6eq;6ax,5) ˆ 11.8 Hz, 3H, H6ax), 1.80
2.14 (m, 3H, H3eq, H3ax, H5); 13C NMR (D2O): d ˆ 62.8 (C4), 60.4 (C5'),
40.9, 40.0 (C2, C6), 38.8 (C5), 29.4 (C3). These NMR data were essentially
identical to those in DMSO published previously.[39]
Methyl 5-aminonicotinate (71): 5-Bromonicotinic acid (70, 4.04 g,
20 mmol)and CuSO 4 ¥ 5H2O (1 g, 4 mmol)were dissolved in concentrated
aqueous ammonia (18 mL)and heated in a sealed container to 180 8C for
24 h. The solvent was then carefully removed, and the residue was dissolved
in methanol (175 mL). Acetyl chloride (10 mL) and trimethyl orthoacetate
(30 mL)were added to this solution, and the mixture was refluxed for 48 h.
Evaporation under reduced pressure and addition of aqueous sodium
carbonate solution (10%, 50 mL)followed by extraction with CH 2Cl2
(10 Â 50 mL)yielded 71 as a brown powder (1.82 g, 60%). The product
was sufficiently pure for further reaction.[40] 1H NMR (CD3C(O)CD3): d ˆ
8.39 (s, 1H, ArH), 8.25 (s, 1H, ArH), 7.54 7.56 (m, 1H, ArH), 5.2 (brs, 2H,
NH2), 3.87 (s, 3H, OCH3); 13C NMR (CD3C(O)CD3): d ˆ 166.8 (CO2CH3),
145.3, 141.3, 139.2, 126.8, 120.8 (Ar), 52.3 (CO2CH3).
Conclusion
We have shown that an equatorial OH group is more strongly
electron-withdrawing than an axial OH group in the 3- and
4-positions of piperidines and hexahydropyridazines. This
stereoelectronic effect affects the basicity of these compounds
and also causes some piperidine or hexahydropyridazine
3-Hydroxy-5-hydroxymethylpyridine (73): Methyl ester 71 (584 mg,
3.84 mmol)was dissolved in dry THF (12 mL)and the solution was cooled
1224
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0805-1224 $ 17.50+.50/0
Chem. Eur. J. 2002, 8, No. 5

Stereoelectronic Substituent Effects
1218 1226
to 08C. After addition of lithium aluminum hydride (584 mg, 15.4 mmol)
the cooling bath was removed, and the mixture stirred for 21 h at room
temperature. The solution was then carefully acidified (aqueous HCl)to
pH 3 and thereafter alkalized (solid Na2CO3)to pH 8. The solvents were
removed under reduced pressure and the residue was filtered through a
column of silica gel (CH2Cl2/CH3OH 9:1). This gave the aminopyridine 72
as a yellow oil (428 mg, 93%). 1H NMR (CD3C(O)CD3): d ˆ 7.92 (s, 1H,
ArH), 7.80 (s, 1H, ArH), 7.04 (m, 1H, ArH), 4.53 (s, 2H, CH2), 4.36 (brs,
3H, OH, NH2); 13C NMR (CD3C(O)CD3): d ˆ 144.5, 137.9, 136.4, 135.2,
119.1 (Ar), 61.4 (CH2).
and hydrogenated (1 atm, 2 h)using Pd/C (25 mg, 10%)as catalyst. The
resulting solution was filtered and concentrated to give 27 (50 mg, 65%)as
the hydrochloride. The NMR spectrum was identical to previously
published spectra of the optically pure hydrochloride.[20]
(3R,4S,5R)-4-Hydroxy-5-hydroxymethyl-3-piperidinecarboxylic methyl es-
ter hydrochloride (29): Prepared by treating N-tert-butoxycarbonyl-
(3R,4S,5R)-4-hydroxy-5-hydroxymethyl-3-piperidinecarboxylic methyl es-
ter[18] with hydrochloric acid, and subsequent evaporation. 1H NMR (D2O):
d ˆ 4.5 (brs, 1H, H4), 3.7 (s, 3H, OMe), 3.4 3.6 (m, 3H, H5'a, H5'b, H2eq),
3.2 (m, 2H, H2ax, H6eq), 3.0 (ddd, J3,4 ˆ 2.5, J2eq,3 ˆ 4, J2ax,3 ˆ 13 Hz, 1H,
H3), 2.9 (t, J ˆ 13 Hz, 1H, H6ax), 2.1 (m, 1H, H5).
(3R,4S,5R)-4-Hydroxy-5-hydroxymethyl-3-piperidinecarboxylic acid hy-
drochloride (30): Prepared by treating N-tert-butoxycarbonyl-(3R,4S,5R)-
4-hydroxy-5-hydroxymethyl-3-piperidinecarboxylic acid[18] with hydro-
chloric acid, and subsequent evaporation. The NMR spectrum was
identical to that previously published for the racemic compound.[19]
The aminopyridine 72 (424 mg, 3.53 mmol)was dissolved in sulfuric acid
(2.5m, 5 mL)and put on an ice bath. Small portions of NaNO (268 mg,
2
3.89 mmol)were added over a 10 min period. The solution was then stirred
for a further 30 min, whereafter sulfuric acid (1m, 15 mL)was added and
the mixture heated to 708C for 1 h. After ion exchange (Amberlite IR-120,
‡
H )the product mixture was released from the resin with 5% NH 4OH,
which gave 73 (357 mg, 83%).[41] 1H NMR (D2O): d ˆ 7.86 7.82 (m, 2H,
ArH), 7.30 (t, J ˆ 2.0 Hz, 1H, ArH), 4.58 (s, 2H, CH2); 13C NMR (D2O):
d ˆ 157.3, 138.3, 133.4, 132.9, 126.1 (Ar), 60.3 (CH2).
(3R,4S,5S)-3-Carboxymethyl-4-hydroxy-5-piperidinecarboxylic acid hy-
drochloride (33): Prepared by treating N-tert-butoxycarbonyl-(3R,4S,5S)-
3-carboxymethyl-4-hydroxy-5-piperidinecarboxylic acid[18] with hydro-
chloric acid, and subsequent evaporation. 1H NMR (D2O): d ˆ 4.8 (m,
1H, H4), 3.7 (s, 3H, OMe), 3.5 (dt, Jgem ˆ 14 Hz, 2H, H2eq, H6eq), 3.0 3.3
(m, 4H, H2ax, H3, H5, H6ax).
cis-3-Hydroxy-5-hydroxymethylpiperidine (25): Pyridine 73 (107 mg,
0.86 mmol)was dissolved in ethanol (10 mL)and Rh/C was added (5%,
100 mg). A hydrogen pressure of 40 atm was applied and the reaction vessel
was heated to 508C for 48 h. The reaction mixture was filtered through
Celite and flash chromatography of the residue with ethanol/CH2Cl2/conc.
NH4OH 10:9:1 (Rf ˆ 0.19)gave 25 (86 mg, 77%). 1H NMR (D2O): d ˆ 3.68
(tt, Jax,ax ˆ 10.6, Jax,eq ˆ 4.6 Hz, 1H, H3), 3.45 (d, J5,5' ˆ 6.2 Hz, 2H, H5'a,
(Æ)-8-Phenyl-1,6,8-triazabicyclo[4.3.0]non-3-ene-7,9-dione (76): Prepared
as previously described.[25] Yield: 57% (Lit.:[25] 62%); 1H NMR (CDCl3):
d ˆ 7.5 (m, 5H, Ar), 5.97 (s, 2H, H3, H4), 4.19 (s, 4H, H2a, H2b, H5a, H5b);
13
ˆ
C NMR (CDCl3): d ˆ 152.6 (C O), 131.4, 128.3, 129.3, 125.6 (Ar), 121.0
H5'b), 3.11 (dd, J2eq,3 ˆ 3.8, J2eq,2ax ˆ 11.8 Hz, 1H, H2eq), 2.98 (dd, J5,6eq
ˆ
‡
(C3, C4), 43.6 (C2, C5); MS (ES): m/z: 230.0925 [M‡H ]; calcd for
3.6, J6eq,6ax ˆ 12.4 Hz, 1H, H6eq), 2.26 (t, 1H, H2ax), 2.16 (t, 1H, H6ax),
1.94 2.10 (m, 1H, H4eq), 1.64 1.88 (m, 1H, H5), 1.02 (q, 1H, H4ax);
13C NMR (D2O): d ˆ 66.3 (C3), 63.8 (C5'), 50.5, 46.5 (C2, C6), 37.3, 35.0
(C4, C5). Acetylation of a sample with pyridine and acetic anhydride
followed by evaporation gave the triacetate which was characterized by
C12H11O2N3H: 230.0930.
(Æ)-(3,4-trans)-3,4-Epoxy-8-phenyl-1,6,8-triazabicyclo[4.3.0]nonane-7,9-
dione (76a): Alkene 76 (20 mg, 0.87 mmol)was dissolved in CH 3CN
(6.8 mL)and water (4.5 mL)in a two-necked flask equipped with a
dropping funnel with a dry ice/acetone condenser, and the solution was
cooled to 08C. 1,1,1-Trifluoroacetone (0.9 mL)and NaHCO 3 (0.59 g)were
added, followed by Oxone (2.8 g)in small portions over a 5 min period. The
reaction mixture stirred at room temperature for 20 h. An extra amount of
NaHCO3 (0.29 g)and Oxone (1.39 g)were added and after reaction for
another 2.5 h the mixture was worked up. Water (45 mL)was added and the
mixture was extracted with CHCl3 (8 Â 25 mL). The combined organic
layers were dried with MgSO4, filtered, and concentrated to give 76a
‡
MS: MS (ES): m/z: 280.1161 [M‡Na ]; calcd for C12H19NO5‡Na: 280.1161.
(Æ)-3,4-cis-3,4-Dihydroxypiperidine (26): N-Benzyloxycarbonyl-1,2,5,6-
tetrahydropyridine[34] (74; 0.10 g, 0.45 mmol)was dissolved in acetone/
H2O (1:1, 0.4 mL)and N-methylmorpholine N-oxide (80 mg)was added,
followed by a solution (0.2 mL)of OsO 4 in tert-butanol (10 gLÀ1), then the
mixture stirred for five days at 258C. A saturated solution of Na2S2O5
(6 mL)was added and the mixture was extracted with EtOAc (5 Â 5 mL).
The organic phase was dried over MgSO4, filtered, and concentrated to
give the diol product (Rf ˆ 0.29, pentane/EtOAc 5:1). This compound was
dissolved in EtOH (3 mL), adding one drop of HCl (conc.), and hydro-
genated (1 atm, 2 h)using Pd/C (25 mg, 10%)as a catalyst. The resulting
solution was filtered and concentrated to give (Æ)-26 (38 mg, 56%)as a
hydrochloride. The free amine could be obtained by ion exchange
chromatography in Amberlite IR-120. The NMR spectrum of the free
amine was identical to the previously published spectra of the optically
pure compound.[35]
(0.185 g, 86%). 1H NMR (CDCl3): d ˆ 7.4 (m, 5H, Ar), 4.18 (d, J2,2
ˆ
13.6 Hz, 2H, H2a, H5a), 3.90 (d, 2H, H2b, H5b), 3.57 (s, 2H, H3, H4);
13
ˆ
C NMR (CDCl3): d ˆ 151.6 (C O), 129.9, 128.1, 127.3, 124.4 (Ar), 47.4
‡
(C3, C4), 41.7 (C2, C4); MS (ES): m/z: 268.0703 [M‡Na ]; calcd for
C12H11O2N3Na: 268.0698.
(Æ)-(3,4-trans)-3,4-Diacetoxy-8-phenyl-1,6,8-triazabicyclo[4.3.0]nonane-
7,9-dione (78): Epoxide 76a (0.64 g, 2.6 mmol)was dissolved in acetic
anhydride (2.49 mL)and dry acetic acid (35 mL)was added. Boron
trifluoride etherate (0.77 mL)was added carefully at 25 8C, and the reaction
mixture was kept at 258C for 2.5 h. After neutralization with a saturated
NaHCO3, the solution was extracted with EtOAc (4 Â 200 mL). The
combined organic layers were washed with saturated NaHCO3 solution and
with saturated NaCl solution. The combined organic layers were dried with
MgSO4, filtered, and concentrated to give 78 (0.86 g, 94%). 1H NMR
(CDCl3): d ˆ 7.4 (m, 5H, Ar), 5.11 (d, J3,2a ˆ 1.2 Hz, 2H, H3, H4), 4.07 (dd,
J2a,2b ˆ 13.2, J2a,3 ˆ 1.2 Hz, 2H, H2a, H5a), 3.69 (d, 2H, H2b, H5b), 2.14 (s,
3,4-trans-3,4-Dihydroxypiperidine (27): N-Benzyloxycarbonyl-1,2,5,6-tet-
rahydropyridine[34] (74, 0.28 g, 1.27 mmol)was dissolved in dichlorome-
thane (4 mL)at 0 8C, and a solution of m-chloroperbenzoic acid (0.34 g,
1.78 mmol)was added in CH 2Cl2 (8 mL). Cooling was stopped and the
mixture was stirred for 4 h at 258C, then washed with 5% aqueous K2CO3
solution (10 mL)followed by a saturated aqueous NaCl solution (10 mL).
The organic phase was dried over MgSO4, filtered, and concentrated. The
product was purified by flash chromatography (pentane/EtOAc 5:1, Rf ˆ
0.21)giving N-benzyloxycarbonyl-3,4-epoxypiperidine (92%, 0.27 g).
6H, 2CH3); 13C NMR (CDCl3): d ˆ 168.3 (O C O), 151.4 (N-C O), 129.9,
À ˆ
ˆ
128.2, 124.5, 127.3 (Ar), 64.0 (C3, C4), 42.9 (C2, C5), 19.9 (CH3); MS (ES):
‡
This epoxide was dissolved in acetic acid (7 mL)and acetic anhydride
(0.47 mL)was added. Boron trifluoride etherate (0.15 mL, 1.15 mmol)was
slowly added, and the mixture was stirred at room temperature for 3 h. An
aqueous saturated solution of NaHCO3 (3 mL)was added, and the mixture
extracted with EtOAc (10 mL). The organic layer was washed with
saturated aqueous Na2CO3 solution (10 mL)followed by saturated aqueous
NaCl solution (10 mL), and was dried over MgSO4. After filtration crude
75 was obtained in quantitative yield.
m/z: 370.1019 [M‡Na ]; calcd for C16H17O6N3Na: 370.1015.
(Æ)-(4,5-trans)-4,5-Dihydroxyhexahydropyridazine (60): Diacetate 78
(0.198 g, 0.57 mmol)was dissolved in methanol (10 mL)containing a
catalytic amount of Na and was kept at room temperature for 1 h 15 min.
The solution was concentrated to give the crude diol. 1H NMR (CD3OD):
d ˆ 7.35 (m, 5H, Ar), 3.88 (m, 2H, H3, H4), 3.77 (dd, J2a,2b ˆ 12.4, J2a,3
ˆ
2.3 Hz, 2H, H2a, H5a), 3.68 (m, 2H, H2b, H5b). This diol (0.57 mmol) was
dissolved in hydrazine hydrate (10 mL)and kept at 100 8C for 18 h. The
hydrazine was removed by evaporation and the crude product was
dissolved in water and poured onto an ion exchange resin column
This product was dissolved in MeOH (2 mL), and a solution of a catalytic
amount of NaOMe in MeOH (1 mL)was added. The mixture was stirred
for 1.5 h, then dry ice was added and the solution was concentrated. The
product diol was purified by flash chromatography (pentane/EtOAc 4:1,
Rf ˆ 0.35), then dissolved in EtOH (3 mL, adding one drop of HCl (conc.),
‡
(Amberlyst 15, H ). The column was washed with water and eluted with
NH4OH (2.5%). The eluent was removed at reduced pressure and the
product was further purified by EtOH/25% NH4OH 99:1 to give 60[42]
Chem. Eur. J. 2002, 8, No. 5
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0805-1225 $ 17.50+.50/0
1225

M. Bols et al.
FULL PAPER
(38 mg, 56%); 1H NMR (D2O): d ˆ 3.44 (m, 2H, H4, H5), 3.10 (dd, J3eq,4
ˆ
[6] C. Y. Chen, R. J. Lefevre, Tetrahedron Lett. 1965, 45, 4057 4063.
[7] Y. Terui, K. Tori, J. Chem. Soc. Perkin Trans. 2 1975, 127 133.
[8] P. Jacobsen, I. M. Labouta, K. Schaumburg, E. Falch, P. Krogsgaard-
Larsen, J. Med. Chem. 1982, 25, 1157 1162.
[9] H. Booth, J. M. Bailey, H. A. R. Y. Shirayda, M. L. Trimble, J. Chem.
Soc. Perkin Trans. 2 1984, 737 743.
4.4, J3eq,3ax ˆ 12.8 Hz, 2H, H3eq, H6eq), 3.07 (dd, J3ax,4 ˆ 9.6 Hz, 2H, H3ax,
H6ax); 13C NMR (D2O): d ˆ 51.2 (C3, C6), 70.8 (C4, C5).
(Æ)-(2,3-trans-3,4-cis)- and (2,3-cis-3,4-cis)-3,4-Epoxy-2-methyl-8-phenyl-
1,6,8-triazabicyclo[4.3.0]nonane-7,9-dione (77a and 77b): Alkene 77[23]
(1.04 g, 4.3 mmol)was dissolved in CH 3CN (33.3 mL)and water
(22.2 mL)in a two-necked flask equipped with a dropping funnel and a
dry ice/acetone condenser, and the solution was cooled to 08C. 1,1,1-
[10] R. J. Abraham, C. J. Medforth, P. E. Smith, J. Comput.-Aided Mol.
Des. 1991, 5, 205 212.
[11] D. C. Lankin, N. S. Chandrakumar, S. N. Rao, D. P. Spangler, J. P.
Snyder, J. Am. Chem. Soc. 1993, 115, 3356 3357.
[12] J. P. Snyder, N. S. Chandrakumar, H. Sato, D. C. Lankin, J. Am. Chem.
Soc. 2000, 122, 544 545.
[13] B. Bernet, U. Piantini, A. Vasella, Carbohydr. Res. 1990, 204, 11 25.
[14] H. H. Jensen, M. Bols, J. Chem. Soc. Perkin Trans. 1 2001, 905 909.
[15] H. H. Jensen, L. Lyngbye, M. Bols, Angew. Chem. 2001, 113, 3555
3557; Angew. Chem. Int. Ed. 2001, 40, 3447 3449.
[16] H. H. Jensen, A. Lohse, B. O. Petersen, J. È. Duus, M. Bols, J. Chem.
Soc. Perkin Trans. 1 2000, 667 670.
[17] A. Hansen, T. M. Tagmose, M. Bols, Tetrahedron 1997, 53, 697 706.
[18] X. Liang, A. Lohse, M. Bols, J. Org. Chem. 2000, 65, 7432 7437.
[19] A. Lohse, H. H. Jensen, P. Bach, M. Bols, J. Chem. Soc. Perkin Trans. 1
2000, 659 665.
[20] Y. Ichikawa, Y. Igarashi, M. Ichikawa, Y. Suhura, J. Am. Chem. Soc.
1998, 120, 3007 3018.
[21] M. Willert, M. Bols, Acta Chem. Scand. 1998, 52, 461 468.
[22] S. U. Hansen, M. Bols, J. Chem. Soc. Perkin Trans. 1 2000, 911 915.
[23] M. Bols, R. G. Hazell, I. B. Thomsen, Chem. Eur. J. 1997, 3, 940 947.
[24] N. Asano, K. Oseki, H. Kizu, K. Matsui J. Med. Chem. 1994, 37, 3701
3706.
[25] R. C. Cookson, S. S. H. Gilani, I. D. R. Stevens, J. Chem. Soc. (C) 1967,
1905 1909.
[26] G. Legler, S. Pohl, Carbohydr. Res. 1986, 155, 119 129.
[27] G. Legler, E. J¸lich, Carbohydr. Res. 1984, 128, 61 72.
[28] B. Winchester, C. Barker, S. Baines, G. Jacob, S. K. Nagoong, G. W. J.
Fleet, Biochem. J. 1990, 256, 277 285.
[29] S. M. Andersen, M. Ebner, C. W. Ekhart, G. Gradnig, G. Legler, I.
Lundt, A. E. Stutz, S. G. Withers, T. Wrodnigg, Carbohydr. Res. 1997,
301, 155 166.
[30] M. P. Dale, H. E. Ensley, K. Kern, K. A. R. Sastry, L. D. Byers,
Biochemistry 1985, 24, 3530 3539.
[31] D. D. Schmidt, W. Frommer, L. M¸ller, E. Truscheit, Naturwissen-
schaften 1979, 66, 584 585.
[32] S. Inouye, T. Tsurouka, T. Ito, T. Niida, Tetrahedron 1968, 23, 2125
2144.
[33] a)J. March, Advanced Organic Chemistry, 3rd ed., Wiley, New York,
1985; b)F. R. Jensen, C. H. Bushweller, Adv. Alicyclic Chem. 1971, 3,
139 194.
[34] S. V. D×Andrea, E. T. Michalson, J. P. Freeman, C. G. Chidester,
J. Szmuszkovicz, J. Org. Chem. 1991, 56, 3133 3137.
[35] L. Chen, D. P. Dumas, C. -H. Wong, J. Am. Chem. Soc. 1992, 114, 741
748.
[36] T. M. Jespersen, M. Bols, M. R. Sierks, T. Skrydstrup, Tetrahedron
1994, 50, 13449 13460.
[37] The conformer ratios were calculated in the usual way[13] from the
following Equation: equatorial conformer ˆ (Jax,ax À J)/(Jax,ax À Jax,eq),
with Jax,ax ˆ 10 Hz and Jeq,ax ˆ 2.5 Hz, which are typical values for
similar compounds.
Trifluoroacetone (4.5 mL)and NaHCO (2.89 g)was added, followed by
3
Oxone (13.7 g)in small portions over a 5 min period. The reaction mixture
was stirred at room temperature for 1 h, and the mixture was worked up.
Water (200 mL)was added and the mixture was extracted with CHCl 3 (8 Â
100 mL). The combined organic layers were dried with MgSO4, filtered,
and concentrated to give a 2:1 mixture of 77a and 77b as a yellow solid
(1.03 g, 93%). 77a: 1H NMR (CDCl3): d ˆ 7.4 (m, 5H, Ar), 4.58 (dq, J2,2'
ˆ
6.8, J2,3 ˆ 1.6 Hz, 1H, H2), 4.00 (d, J5a/5b,4 ˆ 2.4 Hz, 2H, H5a, H5b), 3.53 (dt,
J4,3 ˆ 4.0 Hz, 1H, H4), 3.28 (dd, 1H, H3), 1.39 (d, 3H, H2'); 13C NMR
(CDCl3): d ˆ 129.9, 128.1, 127.2, 124.3 (Ar), 52.0, 48.0 (C3, C4), 46.8 (C2),
41.2 (C5), 13.3 (C2'); 77b: 1H NMR (CDCl3): d ˆ 7.4 (m, 5H, Ar), 4.40 (dq,
1H, J2,2' ˆ 6.0, J2,3 ˆ 4.4 Hz, H2), 4.27 (dd, J5b/5a ˆ 13.6, J5b,4 ˆ 1.2 Hz, 1H,
H5b), 3.58 (dd, J5a,4 ˆ 2.0 Hz, 1H, H5a), 3.53 (m, 1H, H4), 3.47 (t, J3,2/4
ˆ
4.2 Hz, H3, 1H), 1.42 (d, 3H, H2'); 13C NMR (CDCl3): d ˆ 129.9, 128.1,
127.2, 124.3 (Ar), 51.6, 49.1 (C3, C4), 47.7 (C2), 42.7 (C5), 11.9 (C2') ; MS
‡
(ES): m/z: 282.0854 [M‡Na ]; calcd for C13H13N3O3Na: 282.0855.
(Æ)-(2,3-trans-3,4-trans)- and (2,3-cis-3,4-trans)-3,4-Diacetoxy-2-methyl-
8-phenyl-1,6,8-triazabicyclo[4.3.0]nonane-7,9-dione (79 and 79a): Epoxides
77a and 77b (1.04 g, 4.0 mmol)were dissolved in acetic anhydride
(3.79 mL)and dry acetic acid (50 mL)was added. Boron trifluoride
etherate (1.14 mL)was added carefully at 0 8C, and the reaction mixture
was kept at 258C temperature for 2 h. After addition of water (100 mL)and
neutralization with
a saturated NaHCO3 solution, the solution was
extracted with CHCl3 (3 Â 100 mL). The combined organic layers were
washed with saturated NaHCO3 solution (200 mL)and with saturated
NaCl solution (200 mL). The combined organic layers were dried with
MgSO4, filtered, and concentrated to give a 6:1 ratio of 79 and 79a (1.26 g,
87%). 79: 1H NMR (CDCl3): d ˆ 7.4 (m, 5H, Ar), 5.06 (m, J4,3 ˆ 2.6 Hz, H4,
1H), 4.96 (t, J3,4/2 ˆ 1.8 Hz, H3, 1H), 4.34 (dq, J2,2× ˆ 7 Hz, H2, 1H), 4.12 (dd,
J5b,5a ˆ 13.4, J5b,4 ˆ 2.2 Hz, H5b, 1H), 3.61 (dd, J5a,4 ˆ 2.6 Hz, H5a, 1H), 2.13
(2s, 6H, CH3CO), 1.43 (d, J2',2 ˆ 6.8 Hz, H2', 3H); 13C NMR (CDCl3): d ˆ
169.4 (CH3CO), 131.2, 129.4, 128.5, 125.6 (Ar), 68.8, 66.2 (C3, C4), 53.1
(C2), 44.6 (C5), 21.1 (CH3CO), 13.7 (C2'); MS (ES): m/z: 384.1175
‡
[M‡Na ]; calcd for C17H19N3O6Na: 384.1172.
(Æ)-(3,4-trans-4,5-trans)-4,5-Dihydroxy-3-methylhexahydropyridazine
(64): Diacetates 79 and 79a (1.24 g, 3.4 mmol)were dissolved in methanol
(50 mL)containing a catalytic amount of Na, and kept at room temperature
for 20 min. The solution was concentrated to give the crude diol. The diol
(3.4 mmol)was dissolved in hydrazine hydrate (45 mL)and kept at 100 8C
for 18 h. The hydrazine was removed by evaporation and the crude product
was purified by flash chromatography in EtOH/25% NH4OH 99:1 to give
64 as a white solid (0.158 g, 35%). The product contained approximately
6% cis isomer. 1H NMR (D2O): d ˆ 3.51 (ddd, J5,4 ˆ 9.2, J5,6eq ˆ 5.0, J5,6ax
ˆ
10.6 Hz, H5, 1H), 3.15 (dd, J6eq,6ax ˆ 12.8, J6eq,5 ˆ 5.2, H6eq, 1H), 3.06 (t,
J4,5/3 ˆ 9.2 Hz, H4, 1H), 2.61 (dq, J3,3' ˆ 6.6, J3,4 ˆ 9.2 Hz, H3, 1H), 2.55 (dd,
J6ax,5 ˆ 11 Hz, H6ax, 1H), 1.09 (d, H3', J3',3 ˆ 6.2 Hz, 3H); 13C NMR (D2O):
d ˆ 76.8, 70.7 (C4, C5), 57.2 (C3), 51.4 (C6), 14.2 (C3').
[38] L. Svansson, B. D. Johnston, J. -H. Gu, B. Patrick, B. M. Pinto, J. Am.
Chem. Soc. 2000, 112, 10769 10775.
[39] M. D. S˘rensen, N. M. Khalifa, E. B. Pedersen, Synthesis 1999, 1937
1943.
[40] C. Gandolfo, G. Buemi, G. Favini, Gazz. Chim. Ital. 1971, 101, 946
955.
Acknowledgements
This work was supported financially by the Danish National Research
Council through the THOR programme and by the Lundbeck Foundation.
[41] Y. V. Morozov, V. P. Almazov, F. A. Savin, N. P. Bazhulina, Bioorg.
Khim. 1982, 8, 1119 1132.
[42] R. Gabler, H. R. Meyer, Angew. Chem. 1962, 74, 942; Angew. Chem.
Int. Ed. Engl. 1962, 1, 660.
[1] J. Clark, D. D. Perrin, Q. Rev. 1964, 18, 295 320.
[2] S. Inouye, Chem. Pharm. Bull. 1968, 16, 1134 1137.
[3] J. Chilton, J. B. Stenlake, J. Pharm. Pharmacol. 1955, 7, 1004 1007.
[4] H. S. Aaron, G. E. Wicks Jr., C. P. Rader, J. Org. Chem. 1964, 29,
2248 2252.
[5] H. O. House, H. C. M¸ller, C. G. Pitt, P. P. Wickham, J. Org. Chem.
1963, 28, 2407 2416.
Received: September 7, 2001 [F3537]
1226
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0805-1226 $ 17.50+.50/0
Chem. Eur. J. 2002, 8, No. 5