Stereoselective synthesis of 2,4-methanoproline homologues

Article abstract of DOI:10.1016/j.tetasy.2005.12.009

The stereoselective synthesis of 2-azabicyclo[2.2.1]heptane-1-carboxylic acid and 6-azabicyclo[3.2.1]octane-5-carboxylic acid, novel rigid bicyclic proline analogues, is reported. The synthesis was performed in five steps from the corresponding 2-cycloalk

Full text of DOI:10.1016/j.tetasy.2005.12.009

Tetrahedron:
Asymmetry
Tetrahedron: Asymmetry 17 (2006) 252–258
Stereoselective synthesis of 2,4-methanoproline homologues
Oleksandr O. Grygorenko,a Oleksiy S. Artamonov,a Gennady V. Palamarchuk,b
Roman I. Zubatyuk,b Oleg V. Shishkinb and Igor V. Komarova,c,
*
aDepartment of Chemistry, Kyiv Taras Shevchenko University, Volodymyrska Street, 64, Kyiv 01033, Ukraine
bSTC ‘Institute for Single Crystals’ National Academy of Science of Ukraine, 60 Lenina ave., Kharkiv 61001, Ukraine
cEnamine Ltd, Alexandra Matrosova Street, 23, Kyiv 01103, Ukraine
Received 14 November 2005; accepted 13 December 2005
Abstract—The stereoselective synthesis of 2-azabicyclo[2.2.1]heptane-1-carboxylic acid and 6-azabicyclo[3.2.1]octane-5-carboxylic
acid, novel rigid bicyclic proline analogues, is reported. The synthesis was performed in five steps from the corresponding 2-cyclo-
alken-1-ones. A known approach to 2,4-methanoproline is improved. The three amino acids constitute a library of conformationally
constrained proline analogues, which can be used for the design of peptidomimetics and peptide models.
Ó 2006 Elsevier Ltd. All rights reserved.
1. Introduction
Herein we report the design and synthesis of a library,
which consists of three amino acids—bicyclic proline
analogues. Proline is often found in those secondary
structure elements of peptides which are responsible
for their biological action.2,5 One of the compound in
the library is well-known 2-azabicyclo[2.1.1]hexane-1-
carboxylic acid (trivial name 2,4-methanoproline) 1,
which was found in Atheleia herbert smithii in 1980.6
Several syntheses of this compound have been reported
since then.7,8 Two other amino acids, the methanopro-
line homologues 2 and 3, are novel.
Conformationally restricted molecules have become
increasingly popular in those areas of chemistry where
efficient intermolecular interaction is important. In
medicinal chemistry, conformationally restricted drug
molecules often exhibit more efficient interactions with
the corresponding biological targets than the flexible
analogues.1,2 For example, many peptidomimetics
reported to date contain rigid fragments; this is usually
achieved by incorporation of conformationally restricted
amino acid residues.3
Libraries of conformationally constrained or rigid
amino acids, designed to vary torsion angles in peptido-
mimetics in a systematic manner, are of particular
interest. Such libraries can be composed of either
isomers or homologues of amino acids. While the fixed
torsion angles in the common rigid fragments are
different within the set of compounds in such a library,
other properties (steric bulkiness, lipophilicity, etc.) are
similar. Therefore, the libraries could be used to study
the structure-property relationships. However, there
are not many examples of the libraries of isomeric and
especially homologic amino acids in the literature.4
6
4
7
8
1
COOH
COOH
COOH
5
1
1
6
NH
NH
NH
3
4
2
5
6
2
4
5
2
7
3
3
2
1
3
Analysis of molecular models for the N-acetyl N0-meth-
ylamide derivatives of compounds 1–3 showed that tor-
sion angle u at the residues of these amino acids in
peptides should change in consequential manner when
going from 1 to 3. This difference is achieved by minimal
structural changes within the set, allowing one to
assume that the observed regularities in the properties
of the peptides are caused at first approximation by
changes of u.
*
Corresponding author. Tel.: +38 044 239 33 15; fax: +38 044 235 12
73; e-mail: ik214@yahoo.com
0957-4166/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2005.12.009

O. O. Grygorenko et al. / Tetrahedron: Asymmetry 17 (2006) 252–258
253
2. Results and discussion
of which has already been reported.10 However, the
synthesis involves the use of a rather expensive chiral
auxiliary—1,4-bis(benzyloxy)butane-2,3-diol (1,4-di-O-
benzylthreitol), therefore we abandoned this approach.
Our strategy was based on the separation of stereoiso-
mers rather than on chiral induction. The use of racemic
7 or 8 and enantiomerically pure (e.g., S) a-phenylethyl-
amine in reactions analogous to those shown in Scheme
1 should yield equimolar mixtures of diastereomers 11a
and 11b (or 12a and 12b, respectively), which could
hopefully be separated.
In principle, the synthesis of novel amino acids 2 and 3
could be performed using a strategy reported by de
Kimpe for 2,4-methanoproline 1.8 The key step in the
synthesis of 1 is tandem cyanide addition—intramole-
cular cyclization of the imine 4, which was prepared from
3-chloromethylcyclobutanone 5. Compound 6 was then
converted to 1 by the hydrolysis of the nitrile group
and subsequent Pd-catalysed hydrogenolysis (Scheme 1).
The corresponding chloromethylcycloalkanones 7 and
8, which can be used as the starting compounds for
the synthesis of 2 and 3 using de Kimpe’s strategy, are
known compounds. They can be obtained from 2-cyclo-
alken-1-ones in two steps in 45–50% overall yield
(Scheme 2).9
CN
N
CN
N
N
N
NC
NC
Ph
Ph
Ph
(1S,5R)-12a
Ph
(1R,4S)-11a
(1R,5S)-12b
(1S,4R)-11b
O
O
O
Me3SOI
.
py HCl
The second problem can be solved by avoiding imine
isolation during the synthesis of 11 and 12. We achieved
this using a tandem Strecker-intramolecular cyclization
reaction. Only a few papers on the synthesis of a-amino-
nitriles directly from c-chloromethylketones using this
strategy have been published, all of them exploiting
achiral a-aminonitrile 13 as a cyanide source (Scheme
3).11
CH3CN
65-75%
DMSO, NaH
70%
n
n
n
Cl
9, n = 1
10, n = 2
7, n = 1
8, n = 2
Scheme 2.
However, it was necessary to modify the reported stra-
tegy for at least two reasons. First, while the 2,4-methano-
proline is achiral, the novel amino acids here contain
two stereogenic centers. A stereoselective synthesis has
to be developed in order to obtain nonracemic com-
pounds 2 and 3. Another problem is related to the diffi-
culties in imine formation from 7 and 8, and following
cyclization under treatment of the acetone cyano-
hydrine. In our hands, the reported procedures led to
complex mixtures of unidentified compounds, contain-
ing only minor amounts of the desired aminonitriles.
An analogous reaction of 7 or 8 with chiral a-amino-
nitrile 14 (instead of 13) leads to the desirable nitriles
11a, 11b or 12a, 12b. Surprisingly, compound 14 was
not described in the literature, but was found to be easily
formed from a-phenylethylamine and acetone cyano-
hydrine in quantitative yield (Scheme 4).
NH2
HO
CN
MeOH
+
Ph
N
H
CN
Ph
24 h, r. t.
The stereoselectivity problem could be solved, if non-
racemic chloromethylcycloalkanones 7 and 8 were used.
These could be prepared from the corresponding non-
racemic cyclopropane derivatives 9 and 10, the synthesis
14 (100%)
Scheme 4.
Ph
Me2C(OH)CN
O
CN
N
PhCH2NH2
N
1
Ph
TiCl4
Cl
Cl
6 (65%,
calc.on 5)
5
4
Scheme 1.
O
N
Cl
H2N CN
MeOH
24 h, r. t.
Cl
n
+
n
R
NC
R
R
R
n = 1, R = H,CH3
n = 2, R = H
13
Scheme 3.

254
O. O. Grygorenko et al. / Tetrahedron: Asymmetry 17 (2006) 252–258
The reaction of 14 with 3-chloromethylcyclohexanone 8
in methanol proceeded uneventfully; compounds 12a
and 12b were formed in 80% total yield. Column chro-
matography on silica gel turned out to be very efficient
for their separation. It has been found that the prelimin-
ary isolation of 14 is not necessary: the reaction was
equally efficient when chloromethylketone 8, a-phenyl-
ethylamine and acetone cyanohydrin were mixed simul-
taneously in methanol (Scheme 5).
The structure of 15 was also confirmed by its transfor-
mation to the known 7-azabicyclo[2.2.1]heptane-1-car-
boxylic acid 16 (Scheme 8).12
To determine the absolute configuration of compounds
11a, 11b and 12a, 12b, X-ray analysis of aminonitriles
(1S,4R)-11b and (1S,5R)-12a was performed (Fig. 2).
The absolute configuration of these compounds was
established using the known (S)-configuration of the
stereogenic center of the a-phenylethylamine moiety.
The asymmetric part of the crystal unit cell contains
one molecule in the case of 11b and two molecules
Analogous reaction proceeded smoothly with 3-chloro-
methylcyclobutanone 5 (achiral benzylamine was taken
in this case, Scheme 6). This approach to 2,4-methano-
proline precursor 6 is simpler than reported earlier.8
˚
(A and B) for 12a. The N(2)–C(9) bond (1.496(2) A
˚
˚
11b, 1.496(3) A, and 1.492(3) A for molecules A and B
of 12a) is slightly longer compared to average value13
1.475 A. The N(2) atom has pyramidal configuration,
˚
When we applied the reaction conditions described
above to 3-chloromethylcyclopentanone 7, which has a
less reactive carbonyl group when compared to com-
pounds 5 and 8, none of the expected products were
obtained. Optimization of the reaction conditions was
performed, and it was found that a mixture of 11a,
11b, and 15 was formed in acetonitrile as the solvent
in 38% overall yield (Scheme 7). Chromatographic sepa-
ration of the isomers was analogous to that for 12.
The mechanism of formation of 15 is currently under
investigation.
the sum of bond angles centered at this atom is
334.7(4)° for 11b and 339.5(6)° (A) and 338.7(6)° (B)
for 12a. The a-phenylethyl substituent possesses endo
orientation relative to the polycyclic fragment (the
C(10)–C(9)–N(2)–C(7) torsion angle is ꢀ133.0(2)° for
11b, ꢀ140.5(2)° (A) and ꢀ141.6(2)° (B) for 12a). Mole-
cules 11b and 12a differ by orientation of the a-phenyl-
ethylamine substituent with respect to the nitrile group.
The C(8) methyl group adopts ap conformation with
respect to the N(2)–C(9) bond in 11b and –sc conforma-
tion in 12a (the C(9)–N(2)–C(7)–C(8) torsion angle is
175.3(2)° for 11b, ꢀ54.9(2)° (A) and ꢀ57.3(2)° (B) for
12a). The phenyl subtituent has almost orthogonal ori-
entation relative to the C(7)–C(8) bond in all molecules
(the C(8)–C(7)–C(1)–C(2) torsion angle is 93.2(2)° in
11b, ꢀ111.6(3)° (A) and ꢀ109.9(3)° (B) in 12a).
All three isomers 11a, 11b, and 15 were indistinguishable
1
by H and 13C NMR spectroscopy. Structural assign-
ments were performed using 13C DEPT-135 experi-
ments. There are three carbon atoms connected to
nitrogen in each isomer; secondary, tertiary, and quater-
nary in the case of 11a(11b) or tertiary, tertiary, and
quaternary in the case of 15. Therefore, the DEPT-135
spectra were different and characteristic for 11a, 11b,
and 15 in the region of 55–65 ppm (Fig. 1).
Hydrolysis of the cyano group in compounds 11 and 12
was accompanied by partial N-deprotection. This does
not pose a problem since after hydrogenation and
O
CN
N
NH2
MeOH
HO CN
Ph
N
+
+
+
NC
reflux, 30h
Ph
Ph
12a (40%)
Cl
8
12b (40%)
Scheme 5.
O
CN
N
HO CN
MeOH
PhCH2NH2
+
+
reflux, 72 h
Ph
Cl
5
6 (70%)
Scheme 6.
Ph
O
CN
N
CH3CN
N
+
N
+
+
Ph
Ph
N
H
CN
NC
reflux, 30 h
Cl
Ph
11a (8%)
NC
15 (24%)
7
11b (8%)
14
Scheme 7.

O. O. Grygorenko et al. / Tetrahedron: Asymmetry 17 (2006) 252–258
255
Figure 1.
dard procedures.14 3-Chloromethylcyclobutanone 5,15
3-chloromethylcyclopentanone 7,9 and 3-chloromethyl-
cyclohexanone 89 were prepared according to the
procedures described in the literature. All other starting
materials were purchased from Acros, Merck, Aldrich,
and Fluka chemicals. Melting points are uncorrected.
Analytical TLC was performed using Polychrom SI
F254 plates. Column chromatography was performed
using Kieselgel Merck 60 (230–400 mesh) stationary
Ph
NH+2
1. HCl, ∆
N
2. H2, Pd-C
-OOC
16 (39%)
NC
15
Scheme 8.
1
phase. H NMR and 13C NMR spectra were recorded
on Varian Unity Plus 400 spectrometer at 400.4 and
100.7 MHz, respectively. Chemical shifts are reported
in ppm downfield from TMS as an internal standard.
IR spectra were obtained on Nicolet Nexus 470 spec-
trometer. mmax (cmꢀ1) values in IR spectra are given
for the main absorption bands. Mass spectra were
recorded on Agilent 1100 LCMSD SL instrument with
chemical ionization (CI). Optical rotation values were
measured on Perkin–Elmer 341 polarimeter.
recrystallization, pure N-deprotected amino acids 2 and
3 were obtained as hydrochlorides in moderate yields
(35–50% over two steps) (Scheme 9).
CN
N
COOH
.
1. HCl, ∆
2. H2, Pd-C
HCl
NH
Ph
n
n
11a,b, n = 1
12a,b, n = 2
35-50%
2a,b, n = 1
3a,b, n = 2
4.1. General procedure for the cyclopropanation
(compounds 9 and 10)
Scheme 9.
The procedure is a modification of the method reported
by Gassman and Atkins.16 The synthesis of bicyclo-
[4.1.0]heptan-2-one 10 is representative.
3. Conclusions
A library, consisting of three amino acids—bicyclic pro-
line analogues (compounds 1–3) has been designed. A
simple and convenient method for the synthesis of 2,4-
methanoproline homologues as pure enantiomers was
developed. A convenient synthetic route to 2,4-methano-
proline has been proposed.
All operations were performed under a constant argon
flow through the reaction flask. To a flame-dried
three-necked flack, equipped with an efficient stirrer,
inert gas inlet, thermometer, and dropping funnel,
250 ml of dry dimethylsulfoxide were placed. Vigorous
stirring was started and 9.0 g (0.356 mol) of sodium
hydride (95%) carefully added in small portions (less
than 1 g) in such a manner that each next portion was
only added after gas evolution from the previous
addition ceased. The temperature of the reaction
mixture had to be kept within the range of 20–35 °C
4. Experimental
All air- and moisture-sensitive reactions were performed
under an argon atmosphere using standard Schlenk
technique. Solvents were purified according to stan-
Figure 2. Molecular structure of compounds 11b and 12a.

256
O. O. Grygorenko et al. / Tetrahedron: Asymmetry 17 (2006) 252–258
by external cooling. After the addition of NaH, 78.4 g
(0.356 mol) of trimethylsulfoxonium iodide17 was
added, also in small portions. Temperature control
(20–35 °C) is particularly critical at this stage. The white
milky suspension obtained was stirred for an additional
30–45 min to complete the ylide formation. The reaction
mixture was cooled to 20 °C, and a solution of 30 ml
(0.324 mol) of 2-cyclohexen-1-one in 50 ml of dimethyl-
sulfoxide was slowly added dropwise under vigorous
stirring. When all the cycloalkenone was added, the
reaction mixture was stirred for 0.5 h at an ambient tem-
perature and for an additional 2 h at 50 °C, then cooled
and poured onto 300 g of ice. The suspension formed
was filtered, and the filtrate thoroughly extracted with
ether. The combined extracts were dried (MgSO4) and
evaporated at a reduced pressure (temperature of the
bath was kept not higher than 30 °C) to give 25.2 g
(0.229 mol, 70%) of bicyclo[4.1.0]heptan-2-one 10, suffi-
ciently pure for the next step. Further purification can
be performed as described elsewhere.16
(dt, J = 9.0 Hz and 3.2 Hz, 1H), 2.47 (m, 1H), 2.40 (s,
1H), 2.33 (d, J = 8.5 Hz, 1H), 2.14 (d, J = 9.0 Hz, 1H),
1.6–1.8 (m, 3H), 1.49 (m, 1H), 1.43 (d, J = 6.5 Hz, 3H,
CH(CH3)). 13C NMR (CDCl3), d: 143.2, 128.5, 128.4,
128.1, 60.1, 58.9, 57.5, 45.5, 37.2, 30.3, 29.9, 22.6. IR
(KBr): 2236 (m C„N). MS (m/z): 227 (M+1), 123, 105
(PhCH(CH3)).
4.2.3. (1S,4R)-2-((1S)-1-Phenylethyl)-2-azabicyclo[2.2.1]-
heptane-1-carbonitrile 11b. (1S,4R)-2-((1S)-1-Phenyl-
ethyl)-2-azabicyclo[2.2.1]heptane-1-carbonitrile was
prepared from 7 by method B in 8% yield. Mp 65–
66 °C. [a]D = ꢀ43.3 (c 0.60, MeOH). 1H NMR (CDCl3),
d: 7.3 (m, 5H, C6H5), 3.69 (q, J = 6.5 Hz, 1H,
CH(CH3)), 2.93 (dt, J = 9.5 and 3.2 Hz, 1H), 2.45 (m,
1H), 2.3 (s, 1H), 2.27 (d, J = 9.0 Hz, 1H), 1.8–1.9 (m,
4H), 1.54 (d, J = 6.5 Hz, CH(CH3)), 1.27 (m, 1H). 13C
NMR (CDCl3), d: 145.2, 128.4, 127.1, 127.0, 122.3,
62.7 (CH(CH3)), 60.1 (3-CH2), 58.1 (1-C), 40.6 (CH2),
37.3 (4-CH), 29.3 (CH2), 29.3 (CH2), 24.2 (CH3). IR
(KBr): 2240 (m C„N). MS (m/z): 227 (M+1), 123, 105
(PhCH(CH3)). Anal. Calcd for C15H18N2: C, 79.61; H,
8.02; N, 12.38. Found C, 79.72; H, 8.06; N, 12.42.
4.2. General procedure for the synthesis of a-aminonitriles
11, 12, and 15
Synthesis of 6-[(1S)-1-phenylethyl]-6-azabicyclo[3.2.1]-
octane-5-carbonitriles 12a and 12b is representative.
4.2.4. 7-((1S)-1-Phenylethyl)-7-azabicyclo[2.2.1]heptane-
1-carbonitrile 15. 7-((1S)-1-Phenylethyl)-7-azabicyclo-
[2.2.1]heptane-1-carbonitrile was prepared by method
B from 7 in 24% yield. Mp 64–65 °C. [a]D = ꢀ37.6 (c
0.84, MeOH). 1H NMR (CDCl3), d: 7.41 (d,
J = 7.0 Hz, 2H, o-C6H5), 7.33 (m, 3H, C6H5), 3.55 (q,
J = 6.5 Hz, 1H, CH(CH3)), 3.14 (t, J = 4.0 Hz, 1H, 4-
CH), 2.20 (m, 2H), 1.84 (m, 4H), 1.50 (d, J = 6.5 Hz,
3H, CH(CH3)), 1.36 (m, 2H). 13C NMR (CDCl3), d:
145.25, 128.4, 127.5, 127.2, 120.5, 59.5 (CH), 57.1 (C-
4), 56.6 (CH), 35.3 (CH2), 35.2 (CH2), 28.0 (CH2),
28.0 (CH2), 23.9 (CH(CH3)). IR (KBr): 2230 (m
C„N). MS (m/z): 227 (M+1), 123, 105 (PhCH(CH3)).
Anal. Calcd for C15H18N2: C, 79.61; H, 8.02; N, 12.38.
Found C, 79.68; H, 7.97; N, 12.34.
Method A: To a solution of 25.6 g (0.174 mol) of 3-chlo-
romethylcyclohexanone 8 in 150 ml of dry methanol
47.7 ml (0.522 mol) of acetone cyanohydrin and
23.6 ml (0.183 mol) of (S)-a-phenylethylamine were
added. The mixture obtained was refluxed under an
argon atmosphere for 30 h, then poured into 500 ml of
10% sodium hydroxide solution and extracted with
dichloromethane. The combined extracts were dried
over MgSO4 and evaporated under a reduced pressure.
The residue was chromatographed (gradient hexane–
ethylacetate, 5:1–2:1 as an eluent) to give 16.6 g
(0.069 mol, 40%) of 12b first, and then 16.9 g (0.070
mol, 40%) of 12a.
4.2.5. (1S,5R)-6-((1S)-1-Phenylethyl)-6-azabicyclo[3.2.1]-
12a. (1S,5R)-6-((1S)-1-Phenyl-
Method B: To a solution of 0.39 g of 3-chloromethyl-
cyclohexanone 8 in 10 ml of dry acetonitrile, 0.53 g of
aminonitrile 14 (see the synthesis below) was added.
The mixture obtained was refluxed under an argon
atmosphere for 30 h and then worked up as described
above.
octane-5-carbonitrile
ethyl)-6-azabicyclo[3.2.1]octane-5-carbonitrile was pre-
pared by both A and B methods. Mp 176–178 °C.
[a]D = ꢀ18.6 (c 0.16, MeOH). 1H NMR (CDCl3), d:
7.44 (d, J = 8 Hz, 2H, o-C6H5), 7.33 (m, 3H, C6H5),
4.12 (q, J = 6.5 Hz, 1H, CH(CH3)), 3.34 (dd, J = 9.2
and 6.2 Hz, 1H, exo-7-CH), 2.79 (d, J = 9.0 Hz, 1H,
endo-7-CH), 2.39 (s, 1H), 2.33 (d, J = 11 Hz, 1H, 8-
CH), 2.24 (d, J = 11.5 Hz, 1H), 1.81 (d, J = 11 Hz,
1H, 8-CH), 1.70 (m, 3H), 1.5–1.6 (m, 2H), 1.50 (d,
J = 7.0 Hz, 3H, CH(CH3)). 13C NMR (CDCl3), d:
143.4, 128.4, 128.3, 127.8, 121.4, 57.9, 57.6, 53.6, 45.4,
33.4, 33.0, 30.0, 22.6, 19.6. IR (KBr): 2232 (m C„N).
MS (m/z): 241 (M+1), 137, 105 (PhCH(CH3)). Anal.
Calcd for C16H20N2: C, 79.96; H, 8.39; N, 11.66. Found
C, 79.81; H, 8.48; N, 11.59.
4.2.1. 2-Benzyl-2-azabicyclo[2.1.1]hexane-1-carbonitrile 6.
2-Benzyl-2-azabicyclo[2.1.1]hexane-1-carbonitrile was
prepared by method A [using benzylamine instead of
(S)-a-phenylethylamine] in 70% yield. The reaction time
was increased to 72 h. There was no need in chromato-
graphic purification of crude product (more than 90% of
the main compound). For spectral and physical data see
Ref. 8.
4.2.2. (1R,4S)-2-((1S)-1-Phenylethyl)-2-azabicyclo[2.2.1]-
heptane-1-carbonitrile 11a. (1R,4S)-2-((1S)-1-Phenyl-
ethyl)-2-azabicyclo[2.2.1]heptane-1-carbonitrile was
prepared by method B from 7 in 8% yield. [a]D =ꢀ5.1
4.2.6. (1R,5S)-6-((1S)-1-Phenylethyl)-6-azabicyclo[3.2.1]-
octane-5-carbonitrile
ethyl)-6-azabicyclo[3.2.1]octane-5-carbonitrile
12b. (1R,5S)-6-((1S)-1-Phenyl-
was
1
(c 0.66, MeOH). H NMR (CDCl3), d: 7.41 (m, 2H),
prepared by both A and B methods. [a]D = ꢀ26.2 (c
7.33 (m, 3H), 3.75 (q, J = 6.5 Hz, 1H, CH(CH3)), 3.30
0.54, MeOH). 1H NMR (CDCl3), d: 7.38 (d,

O. O. Grygorenko et al. / Tetrahedron: Asymmetry 17 (2006) 252–258
257
J = 7.5 Hz, 2H, o-C6H5), 7.31 (t, J = 7.0 Hz, m-C6H5),
7.23 (t, J = 7.5 Hz, 1H, p-C6H5), 4.13 (q, J = 6.7 Hz,
1H, CH(CH3)), 3.05 (dd, J = 9.2 and 5.2 Hz, 1H, exo-
7-CH), 2.42 (m, 1H, 8-CH), 2.36 (dd, J = 13 and
5.2 Hz, 2-CH), 2.28 (br s, 1H, 1-CH), 2.24 (d,
J = 10.0 Hz, endo-7-CH), 1.90 (d, J = 11.5 Hz, 8-CH),
1.82 Hz (td, J = 12.6 and 6.0 Hz, 1H, 2-CH), 1.70 (m,
1H, 3-CH), 1.61 (m, 1H, 3-CH), 1.61 (d, J = 6.5 Hz,
3H, CH(CH3)), 1.49 (m, 2H, 4-CH2). 13C NMR
(CDCl3), d: 145.7, 128.4, 127.1, 127.0, 123.9 (CN), 59.9
(CH(CH3)), 56.9 (5-C), 56.7 (2-CH2), 45.9 (CH2), 33.3
(1-CH), 32.9 (CH2), 30.1 (CH2), 24.1 (CH(CH3)), 19.2
(CH2). IR (KBr): 2240 (m C„N). MS (m/z): 241
(M+1), 137.
69.1 (5-C), 48.9 (7-CH2), 39.2(1-CH), 34.3, 31.6, 28.8,
17.8 (3-CH2). IR (KBr): 1731 (m C@O). Anal. Calcd
for C8H14ClNO2: C, 50.14; H, 7.36; N, 7.31. Found C,
50.09; H, 7.30; N, 7.35.
4.3.1.
(1S,5R)-6-Azabicyclo[3.2.1]octane-5-carboxylic
acid hydrochloride 3a. (1S,5R)-6-Azabicyclo[3.2.1]oct-
ane-5-carboxylic acid hydrochloride was prepared ana-
logously to 3b from 12a in 38% yield. Mp >250 °C.
[a]D = +17.5 (c 0.95, MeOH). NMR and IR spectra
were identical to that for 3b enantiomer. Anal. Calcd
for C8H14ClNO2: C, 50.14; H, 7.36; N, 7.31. Found C,
50.19; H, 7.41; N, 7.29.
4.3.2. (1R,4S)-2-Azabicyclo[2.2.1]heptane-1-carboxylic
acid 2a. (1R,4S)-2-Azabicyclo[2.2.1]heptane-1-carbox-
ylic acid was prepared analogously to 3b from 11a,
except the purification was performed using ion-
exchange chromatography (strong cationite, 10% aque-
ous pyridine as an eluent), in 33% yield. Mp >230 °C
(dec.). [a]D = +10.5 (c 0.28, MeOH). 1H NMR
(CD3OD), d: 3.22 (d, J = 10.0 Hz, 1H, 3-CH), 3.01 (d,
J = 9.5 Hz, 1H, 3-CH), 2.69 (s, 1H, 4-CH), 2.19 (m,
1H), 1.9–2.0 (m, 4H), 1.63 (m, 1H). 13C NMR
(CD3OD), d: 171.2 (COOH), 71.9 (1-C), 49.1 (3-CH2),
38.5, 35.9, 28.7, 25.9. IR (KBr): 1640 (mas COOꢀ),
1600 (d NH2+), 1409 (ms COOꢀ). Anal. Calcd for
C7H11NO2: C, 59.56; H, 7.85; N, 9.92. Found C,
59.50; H, 7.79; N, 9.90.
4.2.7. 2-Methyl-2-(((1S)-1-phenylethyl)amino)propane-
nitrile 14. 2-Methyl-2-(((1S)-1-phenylethyl)amino)pro-
panenitrile 14.5 ml (0.159 mol) of acetone cyano-
hydrine and 20 ml (0.155 mol) of (S)-a-phenylethyl-
amine were mixed in 50 ml of absolute methanol and
allowed to stand overnight. Then the mixture was evapo-
rated to dryness to give 29.1 g (0.155 mol, 100%) of pure
14. An analytical sample of 14 was obtained by recrys-
tallization from hexane. Mp 77–78 °C. [a]D = ꢀ146.3
(c 0.64, MeOH). 1H NMR (CDCl3), d: 7.41 (d,
J = 7.0 Hz, 2H, o-C6H5), 7.32 (t, J = 7.5 Hz, 2H, m-
C6H5), 7.23 (t, J = 7.2 Hz), 4.11 (q, J = 6.6 Hz, 1H,
CHCH3), 1.53 (s, 3H, CH3), 1.41 (d, J = 6.5 Hz, 3H,
CHCH3), 1.12 (s, 3H, CH3). 13C NMR (CDCl3), d:
146.6 (CN), 128.5, 127.0, 126.5, 125.8, 55.6, 52.2, 29.2,
27.8, 26.3. IR (KBr): 3323 (m N–H), 2223 (m C„N).
Anal. Calcd for C12H16N2: C, 76.56; H, 8.57; N, 14.88.
Found C, 76.48; H, 8.61; N, 14.84.
4.3.3. (1S,4R)-2-Azabicyclo[2.2.1]heptane-1-carboxylic
acid 2b. (1S,4R)-2-Azabicyclo[2.2.1]heptane-1-carbox-
ylic acid was prepared analogously to 2a from 11b in
35% yield. Mp >230 °C (dec.). [a]D = ꢀ13.7 (c 0.26,
MeOH). NMR and IR spectra were identical to that
for 2a enantiomer. Anal. Calcd for C7H11NO2: C,
59.56; H, 7.85; N, 9.92. Found C, 59.51; H, 7.89, N,
9.88.
4.3. General procedure for the synthesis of amino acids
2, 3, and 16
The synthesis of (1R,5S)-6-azabicyclo[3.2.1]octane-5-
carboxylic acid 3b hydrochloride is representative. To
16.6 g (0.069 mol) of compound 12b, 60 ml of concen-
trated hydrochloric acid was added. The resulting
mixture was refluxed for 72 h, then cooled, filtered and
evaporated under reduced pressure. Water (20 ml) was
added and subsequently removed by evaporation under
reduced pressure. This operation was repeated twice to
remove the excess of HCl. To the obtained solid residue,
70 ml of absolute ethanol were added, the resulting mix-
ture was refluxed, then cooled and left in an ice bath for
3 h. Ammonium hydrochloride was filtered off, the fil-
trate was evaporated to dryness. The solid obtained
was dissolved in 80 ml of methanol and hydrogenated
(10% Pd/C, 35 atm, 35 °C) for 36 h. The catalyst was fil-
tered off, and the solvent was removed by rotary evapo-
ration. The crude product was recrystallized from 10 ml
of concentrated hydrochloric acid to obtain 4.52 g of
pure 3b. The additional 1.84 g of pure 3b were obtained
from the mother liquor by further crystallization. Total
yield—6.36 g (0.033 mol, 48%). Mp >250 °C. [a]D =
4.3.4. 7-Azabicyclo[2.2.1]heptane-1-carboxylic acid 16.
7-Azabicyclo[2.2.1]heptane-1-carboxylic acid was pre-
pared analogously to 2a from 15 in 39% yield. For spec-
tral and physical data see Ref. 12.
4.4. X-ray analysis
Crystals for X-ray diffraction studies were obtained
by slow evaporation of hexane–ethyl acetate (4:1) solu-
tion of 11b and heptane solution of 12a, respectively.
The crystals of 11b (C15H18N2) are orthorhombic.
At
100 K
a = 6.3333(3),
b = 10.7795(8),
c =
3
˚
˚
18.511(1) A, V = 1263.8(1) A , Mr = 226.31, Z = 4,
space group P212121, dcalcd = 1.189 g/cm3, l (Mo Ka) =
0.071 mmꢀ1, F(000) = 488. Intensity of 16,391 reflec-
tions (2095 independent, Rint = 0.088) were measured
on an automatic «Xcalibur 3» diffractometer (graphite
monochromated Mo Ka radiation, CCD-detector
H/2H scanning, 2Hmax = 60°).
1
ꢀ19.9 (c 1.30, MeOH). H NMR (DMSO-d6), d: 10.20
(br s, 1H, NH), 9.24 (br s, 1H, NH), 3.32 (br s, 1H, 3-
CH), 3.14 (br s, 1H, 3-CH), 2.56 (s, 1H, 4-CH), 1.97
(m, 2H), 1.85 (m, 3H), 1.65 (m, 1H), 1.56 (m, 1H),
1.50 (m, 2H). 13C NMR (DMSO-d6), d: 171.8 (COOH),
The crystals of 12a (C16H20N2) are monoclinic. At
˚
293 K, a = 8.4708(5), b = 14.8404(6), c = 11.2447(5) A,
3
˚
b = 98.950(4)°, V = 1396.36(12) A , Mr = 240.34, Z =
4, space group P21, dcalcd = 1.143 g/cm3, l (Mo Ka) =

258
O. O. Grygorenko et al. / Tetrahedron: Asymmetry 17 (2006) 252–258
0.068 mmꢀ1, F(000) = 520. Intensity of 7128 reflections
(4607 independent, Rint=0.024) were measured on an
automatic «Xcalibur 3» diffractometer (graphite mono-
chromated Mo Ka radiation, CCD-detector H/2H scan-
ning, 2Hmax = 50°).
3. Gante, J. Angew. Chem., Int. Ed. Engl. 1994, 33, 1699–
1720.
`
4. (a) Pelliciari, R.; Marinozzi, M.; Camaioni, E.; Nunez, M.
del C.; Costantino, G.; Gasparini, F.; Giorgi, G.; Macch-
iarulo, A.; Subramanian, N. J. Org. Chem. 2002, 67, 5497–
´
5507; (b) Cativiela, C.; Dıaz-de-Villegaz, M. D.; Mayoral,
J. A. Tetrahedron 1993, 49, 677–684.
Both structures were solved by direct method using
SHELX97 package.18 Positions of hydrogen atoms were
located from electron density difference maps and
refined by ‘riding’ model with Uiso = nUeq of nonhydro-
gen atom is bonded with given hydrogen atom (n = 1.5
for methyl group and n = 1.2 for other hydrogen
atoms). Full-matrix least-squares refinement against F2
in anisotropic approximation was converged to
wR2 = 0.116 for 2095 reflections (R1 = 0.049 for 2011
reflections with I > 4r(I), S = 1.05) for 11b and to
wR2 = 0.120 for 4607 reflections (R1 = 0.046 for 3788
reflections with I > 4r(I), S = 1.08) for 12a. Final ato-
mic coordinates, geometrical parameters and crystallo-
graphic data have been deposited with the Cambridge
Crystallographic Data Centre, 12 Union Road, Cam-
bridge, CB2 1EZ, UK (fax: +44 1223 336033; e-mail:
deposit@ccdc.cam.ac.uk) and are available on request
by quoting the deposition number CCDC 289018 (11b)
and CCDC 289081 (12a).
5. (a) Gunasekaran, K.; Ramakrishnan, C.; Balaram, P.
Protein Eng. 1997, 10, 1131–1141; (b) Choi, S. H.; Yu, Y.
J.; Shin, K. I.; Jhon, M. S. J. Mol. Struct. 1994, 323, 233–
242.
6. Bell, E. A.; Qureshi, M. Y.; Pryce, R. J.; Janzen, D. H.;
Lemke, P.; Clardy, J. J. Am. Chem. Soc. 1980, 102, 1409–
1412.
7. Hughes, P.; Clardy, G. J. Org. Chem. 1988, 53, 4793–4796.
8. Rammeloo, T.; Stevens, C. V.; de Kimpe, N. J. Org.
Chem. 2002, 67, 6509–6513.
9. di Beelo, N.; Pellacani, L.; Tardella, P. A. Synthesis 1978,
227–228.
10. (a) Mash, E. A.; Nelson, K. A. Tetrahedron 1987, 43, 679–
692; (b) Johnson, C. R.; Barbachyn, M. R. J. Am. Chem.
Soc. 1982, 104, 4290–4291.
11. (a) Oka, M.; Matsumoto, Y.; Unno, R. Heterocycles 1997,
45, 1447–1450; (b) Oka, M.; Baba, K.; Nakamura, K.;
Dong, L.; Hamajima, H.; Unno, R.; Matsumoto, Y. J.
Heterocycl. Chem. 2003, 40, 177–180.
´
12. Avenoza, A.; Cativiela, C.; Busto, J. H.; Fernandez-Recio,
´
M. A.; Peregrina, J. M.; Rodrıguez, F. Tetrahedron 2001,
57, 545–548.
13. Burgi, H.-B.; Dunitz, J. D. In Structure Correlation; VCH:
¨
Acknowledgements
Weinheim, 1994; Vol. 2, pp 741–784.
14. Armarego, W. L. F.; Chai, C. L. L. Purification of
Laboratory Chemicals; Elsevier: Oxford, 2003; pp 1–609.
15. Stevens, C.; de Kimpe, N. J. Org. Chem. 1996, 61, 2174–
2178.
Financial support of Medivir is gratefully acknow-
ledged. The authors are also thankful to Dr. Marcel
Linschoten, for encouragement and helpful discussion.
16. Gassman, P. G.; Atkins, T. J. J. Am. Chem. Soc. 1972, 94,
7748–7756.
References
17. Corey, E. J.; Chaykowsky, M. J. Am. Chem. Soc. 1965, 87,
1353–1364.
18. Sheldrick, G. M. SHELX97. Programs for Crystal Struc-
1. Silverman, R. B. The Organic Chemistry of Drug Design
and Drug Action; Elsevier: Oxford, 2004; pp 1–617.
2. Komarov, I. V.; Grygorenko, O. O.; Turov, A. V.; Khilya,
V. P. Russ. Chem. Rev. 2004, 73, 785–810.
ture Analysis (Release 97-2). Institut fur Anorganische
¨
Chemie der Universita¨t Go¨ttingen, Tammanstrasse 4,
D-3400 Go¨ttingen, Germany; 1998.