On the electronic effects of OH groups. Synthesis and investigation of tetrahydroxylated azabicycloheptanes

Article abstract of DOI:10.1039/b419154d

Two stereoisomeric 2,3,5,6-tetrahydroxyazabicyclo[2.2.1]heptanes were synthesised and their base strengths determined. The 2,3,5,6-exo-isomer 1 and the 2,3-exo-5,6-endo-isomer 2 were prepared from the Diels-Aldcr adduct of Boc-pyrrole and tosylacetylene b

Full text of DOI:10.1039/b419154d

View Article Online / Journal Homepage / Table of Contents for this issue
A R T I C L E
On the electronic effects of OH groups. Synthesis and investigation
of tetrahydroxylated azabicycloheptanes†
Anette Gregersen, Christian Marcus Pedersen, Henrik Helligsø Jensen and Mikael Bols*
Department of Chemistry, University of Aarhus, DK-8000, Aarhus C, Denmark.
E-mail: mb@chem.au.dk
Received 22nd December 2004, Accepted 22nd February 2005
First published as an Advance Article on the web 17th March 2005
Two stereoisomeric 2,3,5,6-tetrahydroxyazabicyclo[2.2.1]heptanes were synthesised and their base strengths
determined. The 2,3,5,6-exo-isomer 1 and the 2,3-exo-5,6-endo-isomer 2 were prepared from the Diels–Alder adduct
of Boc-pyrrole and tosylacetylene by a route involving osmium catalyzed dihydroxylation and protection, tosyl group
reduction and repeated dihydroxylation. Deprotection gave 1, while 2 was obtained by conversion of the diol into the
ditriflate, followed by nucleophilic inversion with KNO2 and deprotection. Synthesis of the 2,3,5,6-endo-isomer by a
similar strategy was attempted but failed. The pKa of 1 and 2 was determined to be 7.0 and 6.4 respectively. This
means that the change in base strength as a result of stereoisomerism of an OH is smaller in the [2.2.1]-azabicyclic
system than in the piperidines. This is explained by a difference in charge–dipole interactions in the two systems.
Introduction
Being able to predict the effect of substitution on the acid–base
strength of organic molecules is well recognised as being crucial
for predetermining their suitability as drugs or their reactivity.
From the work of Hammett and others much is known about
substituent effects on acidity in both aromatic and aliphatic
systems.1–4 Generally, substitution effects are divided into the
three components: 1) field effects that act through space and
are the sum of coulombic interactions between substituent and
acid; 2) inductive effects that act through bonds by polaration
of the r-bond framework; and 3) conjugative effects that act
through resonance. Naturally, only the former two are relevant
to aliphatic systems. Recent computational work indicates that
the field effects are the dominant factor in the substituents effect
on the acidity of many aliphatic and aromatic acids.5–7
In general, the influence of a substituent’s stereochemistry on
its effects has received relatively little attention. However, field
effects from a substituent will depend on the stereochemistry
and conformation of the molecule. Likewise, inductive effects
may also be stereoelectronic due to different possibilities of hy-
perconjugation depending on the molecule’s overall geometry.8
In a recent series of papers we have investigated the stereochem-
ical effect of polar substituents and, in particular, systematic
variations in the effect of the hydroxyl group in six-membered
rings when placed axially or equatorially.9–17 We found that the
equatorial hydroxyl group in the b- or c-position was three-
times more electronwithdrawing than their axial counterparts
in reducing the acidity of a piperidine10 or in reducing the rate
of glycoside hydrolysis.12,13,16 The cause of this difference may be
associated with field effects or possibly hyperconjugation, but
appear to be very consistent within six-membered rings in a chair
conformation. Our favored explanation is that the substituent
effect difference is caused by different charge–dipole interactions
in the axial and equatorial case (Fig. 1): When the positive end
of the C–O bond dipole is much closer to the positive charge
than the negative end we have an unfavorable interaction. The
system can stabilize itself by losing the charge, hence a stronger
acid/weaker base. In contrast, the more perpendicular the C–O
dipole, the weaker the interaction. The present work constitutes
an effort to learn more about these effects by studying them in
a different system. It was anticipated that similar differences
Fig. 1 Charge–dipole interactions in hydroxylated cyclic amines.
in the charge–dipole interactions might be obtained in the
azabicyclo[2.2.1]heptane system (Fig. 1). In this system the
five-membered rings are in rigid envelope conformations and
hydroxyl substituents, being endo- or exo-configured, will be
positioned similar to an equatorial or axial OH-group in the
b-position of a piperidine. Thus, the distance between oxygen–
˚
˚
nitrogen has been measured to be 3.8 A and 2.9 A in the
b-hydroxypiperinium ion in the equatorial and axial case,18
respectively, while in the bicyclic[2.2.1] system O–N distance
for endo- and exo-alcohols was found to be 3.6 and 3.0. Thus,
the tetrahydroxyamines 1, 2 and 3 were envisaged as amines that
might display a large variation in base strength simply as a result
of stereochemical inversions (Scheme 1). In 1, four perpendicular
charge–dipole interactions, a pseudo-axial polar effect, should
render this compound the most basic isomer. In contrast, the all
endo-isomer 3 would, with four pseudo-equatorial interactions,
have the lowest basicity in the series. The exo–endo-isomer 2
would be predicted to possess an intermediate basicity. As none
of these compounds were known and no pKa data for any
hydroxylated azabicycloheptane were available, we decided to
try to synthesise these three compounds and determine their
base strength.
Scheme 1 Target molecules 1–3 and their precursors.
The most related compound reported in the literature was a
protected derivative of 1 that was prepared by the Prinzbach
group starting from cyclitol precursors.19 While this compound
† Electronic supplementary information (ESI) available: A table of
calculated angles, distances and charge dipole energies in compounds
1, 2 and 27. See http://www.rsc.org/suppdata/ob/b4/b419154d/
1 5 1 4
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 5 1 4 – 1 5 1 9
T h i s j o u r n a l i s
T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 5
©

View Article Online
could be prepared and deprotected, we selected a route that
would allow us to obtain the molecules also possessing 2,3-
endo-diols from the same precursor. We chose the Diels–Alder
adduct 4 (Scheme 1), which was obtained from the reaction
of Boc-pyrrole (5) with tosylacetylene (6).20,21 Several recent
syntheses22,23 have employed this strategy effectively to obtain
substituted azabicycloheptanes, even with hydroxyl groups.23
Upon removal of the sulfone the exo-selectivity in the rigid
bicyclic system ensures control of an osmium catalyzed dihy-
droxylation reaction of the double bond. Nucleophilic inversion
reactions might be employed to obtain the pseudoequatorial
compounds.
Scheme 3 Synthesis of 2.
Results and discussion
led mainly to decomposition or multiple product formation.
This failure may be caused by the many possibilities of side-
reactions of the intermediate alcohols with triflates such as
epoxide formation etc.
Synthesis
Synthesis of known 8 was carried out essentially as has been
previously described:21 In our hands, the solventless reaction of
An attempt to carry out the nucleophilic substitutions in
two-steps was also made. This required, however, some more
protection group chemistry to be able to deprotect one diol in
the presence of the other. Therefore, 7 was converted into the
benzylidene derivative 17 by reaction with dimethoxytoluene
and toluene sulfonic acid (TsOH) in 58% yield (Scheme 5).
Two-step reduction of the sulfone was carried out by radical
addition of Bu3SnH to the conjugated system, giving stannane
18 in 79% yield. Treatment of 18 with TBAF gave the alkene 19
in 82% yield. OsO4 catalyzed dihydroxylation gave the diol 20 in
a quantitative yield, which was then converted to the ditriflate
21 with Tf2O–pyridine in 82% yield. Substitution with KNO2–
crown ether gave an excellent 90% yield of the endo-diol 22.
Protection of this diol with dimethoxypropane–TsOH, giving
23 in 91% yield, was followed by dissolving metal reduction of
the benzylidene group to the exo-diol 24, being obtained in 88%
yield. The exo-diol 24 was converted to the ditriflate 25 in 80%
yield and several attempts to transform this ditriflate to the all
endo-hydroxylated tetrol was made. However, disappointingly
neither KNO2–crown ether nor KOAc in DMF was able to
provide the desired all endo-configured compound. It is likely
these difficulties are due to crowding on the endo-face of the
molecule.
◦
6 in two eq. of 5 for 48 h at 85 C gave adduct 4 in 87% yield
(Scheme 2) and OsO4 catalyzed dihydroxylation of 4 gave the diol
7 in 70% yield. Subsequent isopropylidene protection resulted
in 8 in 73% yield. Compound 8 could be reduced using sodium
amalgam giving the alkene 10 sometimes in yields as high as 84%
yield. However, the reaction depended strongly on the quality of
the reagent, saturation of the double bond was often observed
as a byproduct and yields were frequently lower. Therefore, a
two-step protocol22,24 involving radical addition of tributyltin
hydride to give 9 was preferred, followed by elimination using
tetrabutylammonium fluoride (TBAF). This gave the alkene 10
in 83% yield over two-steps. Dihydroxylation of the alkene 10
with OsO4–NMO gave the exo-diol 11, exclusively, in 72% yield.
Finally, treatment with hydrochloric acid led to quantitative
deprotection to the tetrol 1. Tetrol 1 has two signals in a NMR
spectrum, which is consistent with the assigned stereochemistry
only.
Scheme 4 Attempted synthesis of 3 through tetratriflate 16.
Scheme 2 Synthesis of 1.
Base strength
From the intermediate 11 the endo–exo-tetrol 2 was prepared
(Scheme 3). The diol was converted to the ditriflate 12 with Tf2O
in CH2Cl2–pyridine, giving 85% yield. Treatment of this triflate
with KNO2 in DMF in the presence of crown ether25 led to clean
di-inversion of configuration, giving 48% yield of the diol 13.
Acidic hydrolysis of 13 with hydrochloric acid gave a quantitative
yield of tetrol 2. Tetrol 2 has three signals in a NMR spectrum,
which is consistent with the assigned stereochemistry only.
Several attempts to synthesise the all endo-tetrol 3 were also
made. Conversion of 1 to the pentaacetate 14 with acetic an-
hydride in pyridine gave 68% yield (Scheme 4). Quantitative O-
selective deacetylation gave tetrol 15, which was then converted
to the tetratriflate 16 in 83% yield. Reaction of 16 with KNO2–
crown ether however failed to give the desired product, but
The base strength of the amines 1 and 2 was determined by
titration of their hydrochlorides with NaOH. From the tritration
curves the pKa values were determinedtobe 7.0 for 1 and 6.4 for 2
(Fig. 2). For comparison the unhydroxylated azabicycloheptane
26 has a pKa of 10.8.26 This gives an average base lowering
effect of the exo-hydroxyl group of (10.8 − 7.0)/4 = 0.95 ≈ 1.0,
corresponding to 0.59 kcal mol−1. The base lowering effect of
the endo-OH will be (10.8 − 6.4 − (2 × 0.95))/2 = 1.25 ≈ 1.3,
corresponding to 0.77 kcal mol−1. In hydroxypiperidines an axial
b-hydroxyl group was found to have a base lowering effect of 0.5
pH units (0.30 kcal mol−1), while an equatorial group had a base
lowering effect of 1.3 (0.77 kcal mol−1). Thus, it is clear from
the above data that the endo-hydroxyl group has a base lowering
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 5 1 4 – 1 5 1 9
1 5 1 5

View Article Online
Scheme 5 Attempted synthesis of 3 through a protected derivative of 2.
effect that is very similar to the equatorial OH in piperidines.
In contrast, the exo-OH has a much greater base lowering effect
than a piperidine axial hydroxyl group.
groups the base lowering effect is calculated to 0.87 kcal mol−1
and 0.50 kcal mol−1, which are also in relatively close agreement
with the values measured in this work. The calculated ratio
between exo : endo becomes 1.7, while the observed ratio was
1.3. Though not a perfect fit, charge–dipole theory nevertheless
explains to a large extent the smaller difference in substituent
effect between exo–endo isomers than between equatorial–axial
isomers. In the piperidine system, when the C–O dipole is axial
it is able to adopt a more perpendicular position than can
be accommodated in the azabicyclic system. The remaining
discrepancy between observed and calculated values must be
caused by other factors. One reasonable explanation to 1 being
less basic than predicted could be steric hindrance against
protonation. Hydrogen bonding may or may not occur but this
probably has little impact on base strength, as was demonstrated
from our earlier work with piperidines.9,10,12
Why are these exo-groups so electron withdrawing when
compared to axial groups? To answer this question we carried
out a detailed analysis of the angles and bond lengths using
computer models of the conjugate acid forms of 1, 2 and trans-
3,5-dihydroxypiperidine 27.† The computer models were made
in Chem3D Pro and energy minimized using MOPAC PM3.
In the model of 27 the angle between the C–O bond and the
line between the C–O bond middle and nitrogen atom was
145.8◦ in the equatorial case and 104.7◦ in the axial case. In
1 the average angle between C–O dipoles to the charge was
114◦ for the exo-bonds, while the average angle was 151◦ for
the endo-dipoles in 2. The distances between the middle of the
dipole and the nitrogen atom is however virtually identical when
comparing the equatorial OH in 27 and the endo-OH’s in 2,
and when comparing the axial OH in 27 with the exo-OH’s
in 1. Charge–dipole interactions can be calculated using the
Conclusions
Kirkwood–Westheimer equation.5,27 This equation estimates the
We have investigated the effect of configuration on the base
lowering effect of hydroxyl groups in the azabicycloheptane
system. We find that the difference in base strength of the endo-
and exo-hydroxyl groups is relatively small, with the endo-OH
group lowering pKa of the conjugate acid by 0.3 pH units more
than the exo-OH. The higher substituent effect of the endo-OH
is consistant with a less favourable charge–dipole interaction
when in the conjugate acid form.
69.13l cos a
change in free energy as DD ≈
where l is the dipole
2
r
moment, r is the distance between DtEhe centre of the dipole and
the charge, a is the angle between dipole and the line to the charge
from the middle of the dipole and DE is the effective dielectric
constant. The dielectric constant depends on how much of the
interaction is going through the molecule or the solvent. In
the present case DE is not known, but must be between the
DE of the part of molecule the interaction is through, which
is usually set to 2 (the dielectric constant of cyclohexane), and
the DE of water, which is 78. Reasonably, the DE must be low
since most of the interaction takes place through an alkane-like
part of the molecule. When DE is set arbitrarily to 13 and a
bond dipole moment of 1.7 for the C–O bond is used in the
Kirkwood–Westheimer equation, a base lowering effect of the
equatorial OH of 0.78 kcal mol−1 is obtained, while the axial OH
gives 0.31 kcal mol−1. These values are close to those observed
experimentally in water. The ratio between the equatorial and
axial charge dipole interaction is calculated to 2.5, while the
measured substituent effect ratio is 2.6. For the exo/endo-OH
Experimental
General
1
13C-, H- and H,H-COSY NMR were recorded on a Varian
Gemini 200 (200 MHz) NMR Instrument and, when specifically
noted, on a Mercury 400 (400 MHz) NMR Instrument. The
spectra were referenced to solvent residues. MS was recorded
on a Micromass LC-TOF instrument. Chromatography was
performed in Merck 60 silica. TLC was performed on Merck
silica 60 E254 coated glass plates and developed using either
vanillin (3 g in 100 mL EtOH with 1 mL H2SO4 added),
potassium permanganate (aqueous), Ce–mol (10 g Ce(IV)SO4
and 15 g (NH4)2MoO4 in 1 L 10% H2SO4) or ninhydrin (2% in
n-BuOH) and subsequent heating.
( )-7-tert-Butoxycarbonyl-2-(4-toluenesulfonyl)-7-azabicyclo
[2.2.1]hepta-2,5-diene (4)21. A mixture of ethynyl p-tolyl
sulfone (6, 1.00 g, 5.55 mmol) and 1-tert-butoxycarbonylpyrrole
(5, 2.0 mL, 12.0 mmol) was heated under an inert atmosphere
at 85 ◦C for 48 h. The compound was purified by column
chromatography (EtOAc–pentane, 1 : 4, Rf 0.3) and afforded
Fig. 2 pKa values of the conjugated acids 1 and 2.
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 5 1 4 – 1 5 1 9
1 5 1 6

View Article Online
1.68 g of 4 (87%) as a light yellow powder. 1H-NMR (CDCl3):
d = 7.76 (d, 2H, Ts, J = 4.0 Hz), 7.58 (br s, 1H, H3), 7.35 (d,
2H, Ts, J = 4.0 Hz), 6.90 (d, 1H, H5, J5,6 = 2.6 Hz), 6.87 (d, 1H,
H6, J6,5 = 2.6 Hz), 5.38 (br s, 1H, H4), 5.19 (br s, 1H, H1), 2.42
(s, 3H, Me), 1.24 (s, 9H, Boc). 13C-NMR (CDCl3): d = 152.8
NMR (CDCl3): d = 6.30 (d, 2H, H2, H3, J = 6.2 Hz), 4.66
(d, 2H, H5, H6, J = 15.4 Hz), 4.26 (s, 2H, H1, H4), 1.46, 1.40
(s, 2 × 3H, 2 × Me acetonide), 1.30 (s, 9H, Boc). 13C-NMR
=
(CDCl3): d = 155.1 (C O, Boc), 136.8, 135.7 (C2,C3), 116.0
(C(Me)2 acetonide), 80.1 (C(Me)3 Boc), 62.9, 62.1, 61.1 (C1, C4,
C5, C6), 28.6, 26.5 (2 × Me acetonide), 25.4 (3 × Me Boc).
HRMS (ES): m/z = 290.1368 (M + Na)+, calculated 290.1369.
=
(C O), 145.2, 143.9, 142.0, 140.5 (C3, C4, C6, C7), 139.1, 134.6
(substituted Ar), 129.4, 129.0, 127.4, 127.0 (unsubstituted Ar),
80.4 (C(CH3)3), 66.7, 65.8 (C1,C4) 26.8 (3 × Me), 20.63 (Me).
MS (ES): m/z = 370.2 (M + Na)+, calculated 371.1.
7-tert-Butoxycarbonyl-2,3-O-isopropylidene-2,3,5,6-exo-tetra-
hydroxy-7-azabicyclo[2.2.1]heptane (11). A solution of 10
(0.37 g, 1.38 mmol) in THF (3.3 mL), NMO (1.5 eq., 0.29 g),
OsO4 (0.3 mL of a 2.5 wt% in t-BuOH) were added to a
stirred solution of NaHCO3 (0.134 g, 1.38 mmol) in t-BuOH
(10.7 mL) and water (2.7 mL). The reaction mixture was stirred
at room temperature overnight and quenched with excess 10%
NaHSO3. The aqueous phase was extracted with ethyl acetate,
dried (MgSO4) and concentrated to give crude 11. Purification
was carried out using column chromatography (EtOAc, Rf
( )-7-tert-Butoxycarbonyl-5,6-exo-dihydroxy-2-(4-toluenesul-
fonyl)-7-azabicyclo[2.2.1]hept-2-ene (7)21.
A solution of 4
(1.68 g, 4.83 mmol) in THF (6.3 mL), NMO (1.5 eq., 0.98 g)
and OsO4 (0.63 mL of a 2.5 wt% in tert-butylalcohol) were
added to a stirred solution of NaHCO3 (0.5 g) in t-BuOH–H2O
(4 : 1, 50 mL). The reaction mixture was stirred at room
temperature overnight and quenched with excess 10% NaHSO3.
The compound was extracted into ethyl acetate, dried (MgSO4)
and concentrated to give 7 (1.28 g, 70%), which was used
without further purification. Rf 0.15 (EtOAc–pentane, 1 : 3).
1H-NMR (CDCl3): d = 7.79 (d, 2H, Ar, J = 8.4 Hz), 7.38 (d,
2H, Ar, J = 8.4 Hz), 7.06 (d, 1H, H3, J3,4 = 1.1 Hz), 4.72 (br s,
1H, H1), 4.59 (br s, 1H, H4), 3.97 (d, 1H, H6, J6,5 = 5.8 Hz),
3.89 (d, 1H, H5, J5,6 = 5.8 Hz), 2.39 (s, 3H, Me), 1.13 (s, 9H,
3 × Me) MS (ES): m/z = 404.1 (M + Na)+, calculated 404.1.
1
0.35). Yield of 11: 0.42 g (72%). H-NMR (CDCl3): d = 4.18
(br s, 2H, H2, H3), 4.10 (s, 2H, H5, H6), 3.67 (dd, 2H, H1,
H4, J = 5.9 Hz, 16.8 Hz), 1.40 (s, 9H, Boc), 1.32, 1.19 (s, 6H,
2 × Me acetonide). HRMS (ES): m/z = 324.1430 (M + Na)+,
calculated 324.1423.
2,3,5,6-exo-Tetrahydroxy-7-azabicyclo[2.2.1]heptane (1). 11
(300 mg, 1.0 mmol) was refluxed in 1.0 M HCl for 2 h and
the compound was purified by ion-exchange chromatography
on Amberlite IR-120 (H+). Elution with 5% NH4OH afforded
1 in a quantitative yield (160 mg). 1H-NMR (D2O): d = 4.20 (s,
4H, H2, H3, H5, H6), 4.00 (s, 2H, H1, H4). 13C-NMR (D2O):
d = 70.4 (C1, C4), 66.9 (C2, C3, C5, C6), HRMS (ES): m/z =
162.0769 (M + H)+, calculated 162.0766.
7-tert-Butoxycarbonyl-2,3-di-O-trifluoromethanesulfonyl-5,6-
O-isopropylidene-2,3,5,6-exo-tetrahydroxy-7-azabicyclo[2.2.1]
heptane (12). To a solution of 11 (40 mg, 0.13 mmol) in
dichloromethane (1 mL) and pyridine (0.11 mL) was added
dropwise Tf2O (6 eq., 0.064 mL) at 0 ◦C. The mixture was
stirred for 1 h and quenched with water. After extraction with
EtOAc the collected organic phases were washed with 1.0 M
hydrochloric acid and sat. aqueous NaHCO3. Concentration
gave 12 as a syrup. Yield: 64 mg (85%). 1H-NMR (CDCl3): d =
4.76 (s, 2H, H5, H6), 4.59 (d, 1H, H3, J3,2 = 2.0 Hz), 4.48 (d,
1H, H2, J2,3 = 2.0 Hz), 4.20 (s, 2H, H1, H4), 1.50 (s, 9H, Boc),
1.45, 1.25 (s, 2 × 3H, 2 × Me acetonide). HRMS (ES): m/z =
588.0418 (M + Na)+, calculated 588.0409.
7-tert-Butoxycarbonyl-5,6-exo-dihydroxy-5,6-O-isopropylidene-
2-(4-toluenesulfonyl)-7-azabicyclo[2.2.1]hept-2-ene (8)21. A sol-
ution of 7 (1.28 g, 3.4 mmol) in dimethoxypropane–acetone (1
: 1, 10.8 mL) containing p-toluenesulfonic acid monohydrate
(35 mg, 0.19 mmol) was stirred at room temperature overnight.
The solution was made neutral with a few drops of triethylamine,
concentrated, dissolved in EtOAc and washed with 1.0 M HCl.
The organic layer was dried (MgSO4) and concentrated
to give crude 8. The compound was purified by column
chromatography (EtOAc–pentane, 1 : 3, Rf 0.49). Yield: 1.04 g
1
(73%). H-NMR (CDCl3): d = 7.78 (d, 2H, Ar, J = 8.0 Hz),
7.37 (d, 2H, Ar, J = 8.0 Hz), 7.06 (d, 1H, H3, J3,4 = 2.6 Hz),
4.81 (br s, 1H, H4), 4.65 (br s, 1H, H1), 4.50 (d, 1H, H5, J5,6
=
5.4 Hz), 4.34 (d, 1H, H6, J6,5 = 5.6 Hz), 2.45 (s, 3H, Me), 1.44,
1.30 (2s, 2 × 3H, 2 × Me, acetonide), 1.25 (s, 9H, Boc). MS
(ES): m/z = 443.9 (M + Na)+, calculated 444.1.
7-tert-Butoxycarbonyl-2,3-exo-dihydroxy-2,3-O-isopropylidene-
7-aza-bicyclo[2.2.1]hept-5-ene (10).
A. Addition–elimination of tributylstannane. To a stirred
solution of 8 (234 mg, 0.47 mmol) and AIBN (5 mg) in benzene
was added tributyltin hydride (0.28 g, 0.95 mmol, 0.26 mL) via
a syringe under an inert atmosphere. The reaction mixture was
refluxed for three hours and cooled to room temperature. The
crude mixture was concentrated in vacuo, dissolved in MeCN
and washed with hexanes to remove waste tin compounds.
The MeCN phase was concentrated in vacuo and the product
was purified by column chromatography (gradient: pentane
to EtOAc–pentane, 1 : 9) to give the intermediate 9 (303 mg
(90%)). Stannane 9 (303 mg, 0.43 mmol) was solved in dry THF
and TBAF (1.0 mL of a 1 M solution in THF) was added.
The reaction mixture was refluxed overnight, cooled to room
temperature and concentrated in vacuo. The crude product was
purified using column chromatography (CH2Cl2 to CH2Cl2–
EtOAc, 10 : 1) to give alkene 10. Yield: 105 mg (83%).
7-tert-Butoxycarbonyl-2,3-O-isopropylidene-2,3-exo-5,6-endo-
tetrahydroxy-7-azabicyclo[2.2.1]heptane (13). A solution of
12 (193 mg, 0.34 mmol), potassium nitrite (10 eq., 0.29 g) and
18-crown-6 (1 eq., 0.09 g) in dry DMF (0.5 mL) was heated
at 50 ◦C for 3 days. The reaction mixture was quenched with
water and extracted with ethyl acetate, washed with KCl (sat.),
dried (MgSO4) and concentrated to give 13. Yield: 49 mg (48%).
13
=
C-NMR (CDCl3): d = 155.0 (C O, Boc), 110.0 (C(Me)2,
acetonide), 80.5 (C(Me)3, Boc), 65.5, 65.0, 63.5, 62.0, 60.6,
60.5 (C1, C2, C3, C4, C5, C6), 28.6 (Me, Boc), 25.8, 24.3 (Me
acetonide). MS (ES): m/z = 324.0 (M + Na)+, calculated 324.1.
2,3-endo-5,6-exo-Tetrahydroxy-7-azabicyclo[2.2.1]heptane
(2). Compound 13 (49 mg) was refluxed in 1.0 M hydrochloric
acid for 2 h and the product was purified on ion-exchange resin
(IR-120, H+) eluting with ammonia 5%, giving 2 in a quantitative
yield (26 mg). 1H-NMR (D2O): d = 4.59 (br s, 2H, H2, H3), 4.32
(dd, 2H, J = 2.6 Hz), 4.10 (dd, 2H, J = 2.6 Hz) (H1, H4 and
H5, H6). MS (ES): m/z = 162.0 (M + H)+, calculated 162.0.
HRMS (ES): m/z = 184.0586 (M + Na)+, calculated 184.0583.
2,3,5,6,7-Penta-N,O-acetyl-2,3,5,6-exo-tetrahydroxy-7-aza-
bicyclo[2.2.1]heptane (14). A solution of 1 (150 mg, 1.0 mmol)
in pyridine–acetic acid anhydride (1 : 1, 2 mL) was stirred
overnight at room temperature. The mixture was quenched
with water and concentrated to give crude 14 (254 mg, 68%).
B. Reduction with sodium amalgam. A solution of 8 (1.07 g,
2.54 mmol), Na2HPO4 (1.44 g, 10.1 mmol) and 5% Na(Hg)
(32.3 g, 76.2 mmol) in dry THF–dry MeOH (1 : 1, 30 mL) was
heated under an inert atmosphere to 50 ◦C and stirred for 48 h.
The reaction mixture was quenched with water and the amalgam
was removed by filtration. The aqueous phase was extracted with
ethyl acetate, dried (MgSO4), concentrated in vacuo and finally
purified by column chromatography (EtOAc–pentane, 1 : 3, Rf
0.42). Yield of 10: 0.57 g (84%). Yields were however frequently
1
lower depending on the quality of the sodium amalgam. H-
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 5 1 4 – 1 5 1 9
1 5 1 7

View Article Online
1H-NMR (CDCl3): d = 4.96 (d, 2H, J = 6.0 Hz), 4.92 (d, 2H,
J = 6.0 Hz), 4.68 (d, 1H, J = 2.8 Hz), 4.16 (d, 1H, J = 2.8 Hz)
(H1, H2, H3, H4, H5, H6), 2.06 (s, 3H, NAc), 2.02 (s, 6H, 2 ×
Ac), 1.98 (s, 6H, 2 × Ac).
d = 7.53 (m, 2H, Ar), 7.34 (m, 3H, Ar), 6.40 (bd, 2H, H5, H6),
5.95 (s, 1H, Ph–CH-), 4.87 (bs, 1H, H2, H3) 4.77 (s, 2H, H1,
13
=
H4), 1.42 (s, 9H, Boc). C-NMR (CDCl3): d = 154.7 (C O,
Boc), 137.5, 136.1 (C5, C6), 136.6 (Ar), 130.6 (Ar), 128.5 (Ar),
127.5 (Ar), 109.5 (PhCH-), 80.6, 80.1 (C2, C3), 80.4 (CMe3,
Boc), 63.1, 62.4 (C1, C4), 28.5 (–C(CH3)3, Boc). HRMS (ES):
calculated for (M + Na) = 338.1368, found: 338.1363.
7-Acetyl-7-aza-bicyclo[2.2.1]heptane-2,3,5,6-exo-tetraol (15).
Compound 14 (254 mg) was dissolved in methanol and a
catalytic amount of sodium in methanol was added. The reaction
mixture was stirred at room temperature for 2 h. The reaction
was quenched with dry-ice and the mixture was concentrated to
2,3-O-Benzylidene-7-tert-butoxycarbonyl-2,3,5,6-exo-tetrahy-
droxy-7-azabicyclo[2.2.1]heptane (20). To a solution of the
alkene 19 (140 mg, 0.44 mmol) in THF (1 mL) was added N-
methylmorpholine-N-oxide (67 mg, 0.66 mmol). OsO4 (0.1 mL
of a 2.5 wt% in tert-butylalcohol) was added to a stirred solution
of NaHCO3 (50 mg) in t-BuOH (4 mL) and water (1 mL). The
reaction mixture was stirred at room temperature overnight and
quenched with excess 10% NaHSO3. The aqueous phase was
extracted with ethyl acetate, dried (MgSO4) and concentrated
to give the desired product in a quantitative yield (152 mg).
1H-NMR (CDCl3): d = 7.48 (m, 2H, Ar), 7.36 (m, 3H, Ar), 5.31
(s, 1H, PhCH-), 4.47/4.33 (2 bs, 2H, H5, H6), 4.22 (s, 2H, H1,
H4), 3.82/3.74 (2 bs, 2H, H2, H3), 1.40 (s, 9H, Boc). HRMS
(ES): calculated for (M + Na) = 372.1423, found: 372.1426.
1
give 15 in a quantitative yield (139 mg). H-NMR (D2O): d =
4.21 (s, 1H, H1), 3.98 (s, 1H, H4), 3.91 (s, 4H, H2, H3, H5, H6),
2.08 (s, 3H, NAc).
7-Acetyl-2,3,5,6-tetra-O-trifluoromethanesulfonyl-7-azabicyclo
[2.2.1]heptane-2,3,5,6-exo-tetraol (16). To a solution of 15
(30 mg, 0.137 mmol) in dry dichloromethane (1.5 mL) and
dry pyridine (0.17 mL) at 0 ◦C was added dropwise Tf2O (12
eq., 0.132 mL). The mixture was stirred for 1 h and quenched
with water. After extraction, the organic layer was washed
with 1.0 M hydrochloric acid and sat. aqueous NaHCO3. The
organic layer was dried (MgSO4) and concentrated to give 16
(85 mg, 83%). 1H-NMR (CDCl3): d = 5.42 (d, 2H, H3, H5, J =
4.0 Hz), 5.28 (d, 2H, H2, H6, J = 4.0 Hz), 5.22 (d, 1H, H1, J =
4.0 Hz), 4.63 (d, 1H, H4, J = 4.0 Hz), 2.18 (s, 3H, NAc). MS
(ES): m/z = 754.3 (M + Na)+, calculated 753.9.
5,6-O-Benzylidene-7-tert-butoxycarbonyl-2,3-di-O-trifluoro-
methanesulfonyl-2,3,5,6-exo-tetrahydroxy-7-azabicyclo[2.2.1]
heptane (21). To a solution of 20 (170 mg, 0.48 mmol) in
dichloromethane (4 mL) and pyridine (0.4 mL) was added
dropwise Tf2O (6 eq., 0.24 mL) at 0 ◦C. The mixture was stirred
for 1 h and quenched with water. After extraction the organic
layer was washed with KHSO4 (10%), NaHCO3 (sat.) then
brine and dried over MgSO4. Concentration gave 21 as a syrup.
Yield: 0.245 g (81%). 1H-NMR (CDCl3): d = 7.30–7.46 (m, 5H,
Ar), 5.65 (s, 1H, Ph–CH-), 4.91 (s, 2H, H2, H3), 4.82 (d, J =
( )-5,6-O-Benzylidene-7-tert-butoxycarbonyl-5,6-exo-dihydr-
oxy-2-(4-toluenesulfonyl)-7-azabicyclo[2.2.1]hept-2-ene (17).
A
solution of the diol 7 (310 mg, 0.81 mmol) in dimethoxytoluene
(3 mL) containing a catalytic amount of p-toluenesulfonic
acid monohydrate was stirred at room temperature overnight.
The product 17 was purified by column chromatography
(EtOAc–pentane, 3 : 1, Rf = 0.25). Yield: 172 mg (58%)
1H-NMR (CDCl3): d = 7.80 (d, J = 8.2, 2H, Ts), 7.47 (m, 2H,
Ar), 7.38 (d, J = 8.1, 2H, Ts), 7.31–7.36 (m, 3H, Ar), 7.15 (m,
1H, H3), 5.98 (s, 1H, PhCH-), 4.98 (bs, 1H, H1), 4.81 (bs, 1H,
H4), 4.56 (d, 1H, J = 5.7, H6), 4.43 (d, 1H, J = 5.6, H5), 2.46
(s, 3H, Ts), 1.19 (s, 9H, –C(CH3)3). 13C-NMR (CDCl3): d =
2.0 Hz, 1H, H5), 4.67 (d, J = 2.03 Hz, 1H, H6), 4.26 (s, 2H, H1,
13
=
H4), 1.42 (s, 9H, Boc). C-NMR (CDCl3): d = 152.1 (C O,
Boc), 133.3 (Ar), 129.2 (Ar), 127.4 (Ar), 126.0 (Ar), 117.3 (q,
J = 1280 Hz, Tf), 105.6 (Ph–CH), 80.53 (CMe3, Boc), 79.3 (C2,
C3), 76.6, 76.9 (C5, C6), 63.0, 61.6 (C1, C4), 26.9 (–C(CH3)3,
Boc). HRMS (ES): calculated for (M + Na) = 636.0408, found
636.0330.
=
145.6 (C O), 144.6 (C2), 136.2 (Ts), 135.8 (Ts), 130.4 (Ts),
130.2 (Ar), 130.0 (C3), 128.4 (Ts), 128.3 (Ar), 127.2 (Ar), 109.8
(PhCH-), 81.2 (–CMe3), 80.3 (C5), 79.6 (C6), 65.2 (C1), 64.2
(C4), 27.9 (C(CH3)3), 21.8 (Ts). HRMS (ES): calculated for
(M + Na) = 492.1457, found: 492.1470.
2,3-O-Benzylidene-7-tert-butoxycarbonyl-2,3-exo-5,6-endo-
tetrahydroxy-7-azabicyclo[2.2.1]heptane (22). A solution of 21
(230 mg, 0.375 mmol), potassium nitrite (10 eq., 0.32 g) and
18-crown-6 (1 eq., 0.10 g) in dry DMF (0.5 mL) was heated at
50 ◦C for 5 days. The reaction mixture was quenched with water
and extracted with ethyl acetate, washed with KCl (sat.), dried
(MgSO4) and concentrated to give 22. Yield: 119 mg (90%).
1H-NMR (CDCl3) (rotamers present): d = 7.45 (d, J = 7.2 Hz,
2H, Ar), 7.38(m, 3H, Ar), 5.71 (s, 1H, PhCH-), 4.73 (s, 2H,
H2, H3), 4.60, 4.40 (2 bs, 2H, H1, H4), 4.00 (bs, 2H, H5, H6),
( )-5,6-Benzylidene-7-tert-butoxycarbonyl-5,6-exo-dihydroxy-
2-p-toluenesulfonyl-3-tributylstannyl-7-azabicyclo[2.2.1]hepthane
(18). To a stirred solution of 17 (140 mg, 0.30 mmol) and
AIBN (3 mg) in benzene was added tributyltin hydride (0.17 g,
0.6 mmol, 0.16 mL) via a syringe under an inert atmosphere.
The reaction mixture was refluxed for three hours and cooled
to room temperature. The crude mixture was concentrated
in vacuo, dissolved in MeCN and washed with hexanes. The
MeCN phase was concentrated in vacuo and the product was
purified with column chromatography (gradient: pentane to
EtOAc–pentane, 1:9) to give the product 18 as a thick sirup,
which upon standing became a waxy solid. Yield: 181 mg
1.22–1.35 (m, 9H, Boc). 13C-NMR (CDCl3): d = 154.5 (C O),
=
135.8 (Ar), 129.7 (Ar), 128.3 (Ar), 127.2 (Ar), 103.8 (PhCH-),
80.5 (–CMe3), 77.9 (C2, C3), 65.6, 65.2 (C5, C6), 62.9, 61.8 (C1,
C4), 28.4 (–CCH3). HRMS (ES): calculated for (M + Na) =
372.1423, found 372.1415.
1
(79%). H-NMR (CDCl3): d = 7.79 (d, J = 8.0, 2H, Ts), 7.46
(m, 2H, Ar), 7.39 (d, J = 8.0, 2H, Ts), 7.26–7.32 (m, 3H, Ar),
5.58 (s, 1H, Ph–CH-), 5.27 (d, J = 5.6, 1H), 4.43 (d, J = 5.6,
1H), 4.39 (bs, 2H), 3.60 (bs, 1H), 2.46 (s, 3H, Me), 1.58 (s, 1H,
H3), 1.2–1.5 (m, 21 H, Boc, Bu), 0.90 (t, J = 7.2, 9H, Bu). MS
(ES): calculated for (M + Na) = 784.3, found: 784.2.
2,3-O-Benzylidene-7-tert-butoxycarbonyl-5,6-O-isopropylidene-
2,3-exo-5,6-endo-tetrahydroxy-7-azabicyclo[2.2.1]heptane (23).
A solution of 22 (0.119 g, 0.339 mmol) in dimethoxy-
propane:acetone (1:1, 3 mL) containing p-TsOH (5 mg) was
stirred at room temperature overnight. The solution was
made neutral with a few drops of triethylamine, concentrated,
dissolved in EtOAc and washed with KHSO4 (10%). The
organic layer was dried (MgSO4) and concentrated to give 23.
Yield: 120 mg (91%). 1H-NMR (CDCl3): d = 7.51 (m, 2H, Ar),
7.34 (m, 3H, Ar), 5.60 (s, 1H, PhCH-), 4.83 (bs, 2H, H2, H3),
4.65 (bs, 2H, H1, H4), 3.75, 3.65 (bs, 2H, H5, H6), 1.49, 1.25
(–CH3, acetonide), 1.33 (s, 9H, Boc). HRMS (ES): calculated
for (M + Na) = 412.1735, found 412.1736.
2,3-O-Benzylidene-7-tert-butoxycarbonyl-2,3-exo-dihydroxy-
7-azabicyclo[2.2.1]hept-5-ene (19). Compound 18 (155 mg,
0.20 mmol) was solved in dry THF and TBAF (0.41 mL of
a 1 M solution in THF) was added. The reaction mixture
was refluxed overnight, cooled to room temperature and
concentrated in vacuo. The crude product was purified by
column chromatography (CH2Cl2 to CH2Cl2–EtOAc, 10 : 1)
1
to give the alkene 19. Yield: 75 mg (82%). H-NMR (CDCl3):
1 5 1 8
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 5 1 4 – 1 5 1 9

View Article Online
2,3-O-Isopropylidene-2,3-endo-5,6-exo-tetrahydroxy-7-tert-
butoxycarbonyl-7-azabicyclo[2.2.1]heptane (24). A solution of
23 (120 mg, 0.308 mmol) in dry THF was added to 80 mL
NH3 (l) at −78 ◦C. To the solution was added sodium (45 mg)
and it was stirred until the blue color was persistent. The
reaction mixture was quenced with NH4Cl and the ammonia
was allowed to evaporate. The crude mixture was diluted with
water, extracted with EtOAc, washed (KHSO4), dried and
References
1 L. P. Hammett, Chem. Rev., 1935, 17, 125–136.
2 J. D. Roberts and W. T. Moreland, J. Am. Chem. Soc., 1953, 75,
2167–2173.
3 J. Clark and D. D. Perrin, Quart. Rev. Biophys., 1964, 18, 295–
320.
4 S. Inouye, Chem. Pharm. Bull., 1968, 16, 1134–1137.
5 K. B. Wiberg, J. Org. Chem., 2002, 67, 1613–1617.
6 K. B. Wiberg, J. Org. Chem., 2002, 67, 4787–4794.
7 K. B. Wiberg, J. Org. Chem., 2003, 68, 875–882.
8 I. V. Alabugin and T. A. Zeidan, J. Am. Chem. Soc., 2002, 124, 3175–
3185.
9 H. H. Jensen, L. Lyngbye and M. Bols, Angew. Chem., Int. Ed., 2001,
40, 3447–3449.
10 H. H. Jensen, L. Lyngbye, A. Jensen and M. Bols, Chem.–Eur. J.,
2002, 8, 1218–1226.
11 M. Bols, H. H. Jensen, L. Lyngbye, A. Jensen and I. Thomsen,
Latvian J. Chem., 2002, 60–68.
12 M. Bols, X. Liang and H. H. Jensen, J. Org. Chem., 2002, 67, 8970–74.
13 H. H. Jensen and M. Bols, Org. Lett., 2003, 5, 3201–3203.
14 A. Bu¨low, T. Meyer, T. K. Olszewski and M. Bols, Eur. J. Org. Chem.,
2004, 2, 323–329.
15 H. H. Jensen, L. U. Nordstrøm and M. Bols, J. Am. Chem. Soc.,
2004, 126, 9205–9213.
16 C. McDonnell, O. L. Lopez, P. Murphy, J. F. Bolanos, R. Hazell and
M. Bols, J. Am. Chem. Soc., 2004, 126, 12374–12385.
17 C. M. Pedersen and M. Bols, Tetrahedron, 2005, 61, 115–122.
18 Y. Terui and K. Tori, J. Chem. Soc., Perkin Trans. 2, 1975, 127–
133.
1
concentrated to give crude 24. Yield: 82 mg (88%). H-NMR
(CDCl3): d = 4.54 (s, 2H, H2, H3), 4.43 (s, 2H, H5, H6),
4.27 (b s, 2H, H1, H4), 1.45 (s, 9H, Boc), 1.30, 1.25 (s, 6H,
13
=
–CH3(acetonide)). C-NMR (CDCl3): d = 157.2 (C O),
117.1 (CMe2), 81.3 (–CMe3), 77.1 (C1, C4), 69.3 (C2, C3, C5,
C6), 28.9 (–CCH3), 25.8 (–CH3), 24.7 (-CH3). HRMS (ES):
calculated for (M + Na) = 324.1419, found: 324.1423.
7-tert-Butoxycarbonyl-5,6-di-O-trifluoromethanesulfonyl-2,3-
O-isopropylidene-2,3-endo-5,6-exo-tetrahydroxy-7-azabicyclo
[2.2.1]heptane (25). To a solution of 24 (82 mg, 0.272 mmol)
in dichloromethane (2 mL) and pyridine (4 mL) was added
dropwise Tf2O (6 eq., 0.28 mL) at 0 ◦C. The mixture was stirred
for 1 h and quenched with water. After extraction, the organic
layer was washed with KHSO4 (10%), NaHCO3 (sat.), brine
and dried over MgSO4. Concentration gave 25 as a syrup. Yield:
0.123 g (80%). 1H-NMR (CDCl3): d = 5.51 (s, 2H, H2, H3), 4.73
(bs, 2H, H1, H4), 4.63 (s, 2H, H5, H6), 1.46 (s, 9H, Boc), 1.32
(s, 3H, Me (acetonide)), 1.25 (s, 3H, Me (acetonide)). 13C-NMR
19 J. Schubert, R. Keller, R. Schwesinger and H. Prinzbach, Chem. Ber.,
=
(CDCl3): d = 153.5 (C O), 118.1 (CMe2), 82.7 (–CMe3), 81.7
1983, 116, 2524–2545.
20 H. J. Altenbach, B. Blech, J. A. Marco and E. Vogel, Angew. Chem.,
Int. Ed. Engl., 1982, 21, 778.
(C2, C3), 64.3 (C5, C6), 63.3 (C1, C4), 27.9 (–CCH3), 25.9 (Me
(acetonide)), 24.5 (Me acetonide). HRMS (ES): calculated for
(M + Na) = 588.0409, found 588.0414.
21 R. Leung-Toung, Y. Liu, J. M. Muchowski and Y.-L. Wu, J. Org.
Chem., 1998, 63, 3235–3250.
22 F. I. Carroll, F. Liang, H. A. Navarro, L. E. Brieaddy, P. Abraham,
M. I. Damaj and B. R. Martin, J. Med. Chem., 2001, 44, 2229–
2237.
Determination of pKa of amines
The amine hydrochloride (30 mg) was dissolved in 15–20 ml
distilled water and subjected to titration with 0.1 M NaOH,
following the pH using a pH electrode. The pKa was determined
from the resulting titration curve and was the average of three
determinations (error 0.1).
23 A. J. Moreno-Vargas, C. Schu¨tz, R. Scopelliti and P. Vogel, J. Org.
Chem., 2003, 68, 5632–5640.
24 L. E. Brieaddy, F. Liang, P. Abraham, J. R. Lee and F. I. Carroll,
Tetrahedron Lett., 1998, 39, 5321–5322.
25 R. M. Moriarty, H. Zhuang, R. Penmasta, K. Liu, A. K. Awasthi,
S. M. Tuladhar, M. S. C. Rao and V. K. Shingh, Tetrahedron Lett.,
1993, 34, 8029–8032.
26 L. J. Kricka and J. M. Vernon, Adv. Heterocycl. Chem., 1974, 16,
Acknowledgements
87–120.
We thank SNF and the Lundbeck foundation for financial
support.
27 J. G. Kirkwood and F. H. Westheimer, J. Chem. Phys., 1938, 6, 513–
517.
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 5 1 4 – 1 5 1 9
1 5 1 9