Stereoselective synthesis of Cα-tetrasubstituted azabicyclo[X.3.0]alkane amino acids

Article abstract of DOI:10.1016/j.tetlet.2004.06.077

a- a- a- By a stereoselective alkylation approach a synthesis of eight enantiopure, sterically constrained C ^?±-tetrasubstituted azabicycloalkane amino acids was carried out.

Full text of DOI:10.1016/j.tetlet.2004.06.077

Tetrahedron
Letters
Tetrahedron Letters 45 (2004) 6311–6315
Stereoselective synthesis of Ca-tetrasubstituted
azabicyclo[X.3.0]alkane amino acids
b
b,ꢀ
Leonardo Manzoni,a, Laura Belvisi, Matteo Colombo, Eliana Di Carlo,c
*
Alessandra Fornia and Carlo Scolasticob,
*
aCNR––Istituto di Scienze e Tecnologie Molecolari (ISTM) and Centro Interdisciplinare Studi bio-molecolari e applicazioni
Industriali (CISI), Via Venezian 21, I-20133 Milano, Italy
bDipartimento di Chimica Organica e Industriale and Centro Interdisciplinare Studi bio-molecolari e applicazioni Industriali (CISI),
`
Universita degli Studi di Milano, Via Venezian 21, I-20133 Milano, Italy
cCentro Interdisciplinare Studi bio-molecolari e applicazioni Industriali (CISI), Via Venezian 21, I-20133 Milano, Italy
Received 26 April 2004; revised 17 June 2004; accepted 18 June 2004
Available online 10 July 2004
Abstract—By a stereoselective alkylation approach a synthesis of eight enantiopure, sterically constrained Ca-tetrasubstituted
azabicycloalkane amino acids was carried out.
ꢀ 2004 Elsevier Ltd. All rights reserved.
Biologically active peptides are involved in a great num-
ber of physiological processes through their interaction
with receptors and enzymes. Amino acids possessing
alkyl substituents have emerged as important tools for
controlling peptide conformation because their steric
interactions can restrict the motion between backbone
and side-chain within a peptide and may promote par-
ticular peptide secondary structures. Furthermore, the
Figure 1. Bicyclic lactams.
hydrophobic nature of the alkyl substituent may en-
hance the affinity between ligand and receptor. Peptides
themselves are not ideal drug candidates due to their low
formationally constrained substitutes for Ala-Pro dipep-
tide units. Functionalizing these molecules with
lipophilic appendages is a very attractive possibility be-
cause peptide–receptor affinity could be improved by
interaction of the substituent with hydrophobic pockets
in the receptor.
metabolic stability, rapid excretion and lack of selectiv-
ity towards a specific receptor. An attractive alternative
lies in peptide analogues or de novo designed molecules
that mimic the action of the native peptides at the recep-
tor level (peptidomimetics).1
In the course of our studies on peptide secondary struc-
ture mimics, we synthesized several 5,5- 6,5-, 7,5- and
8,5-fused 1-aza-2-oxobicyclo[X.3.0]alkane amino acids2
(Fig. 1, general formula I), that can be regarded as con-
In this paper, we report on a convenient method for the
synthesis of trans-fused bicyclic lactams substituted with
a benzyl or allyl group at the C3 position (Scheme 1).
In previous works we have synthesized benzyl substi-
tuted lactams both via radical3 or nonradical ap-
proaches.4 However, the lack of versatility of these
two synthetic pathways and the difficulty encountered
extending these procedures to large scale synthesis per-
suaded us to investigate a new approach for the synthe-
sis of benzyl or allyl bicyclic lactams.
Keywords: Peptidomimetics; Alkylation; Lactams; Bicyclic aliphatic
compounds; Enolates.
*
Corresponding authors. Tel.: +39-0250314088; fax: +39-0250314072
(L.M.); tel.: +39-0250314090; fax: +39-0250314072 (C.S.); e-mail
addresses: leonardo.manzoni@istm.cnr.it; carlo.scolastico@unimi.it
ꢀ Present address: Nikem Research, via Zambeletti 25, I-20021,
Baranzate di Bollate, Milano, Italy.
0040-4039/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2004.06.077

6312
L. Manzoni et al. / Tetrahedron Letters 45 (2004) 6311–6315
Scheme 1. Synthesis of Ca-tetrasubstituted azabicycloalkane amino acids.
The method here described is based on the stereoselec-
tive alkylation of a Shiff base amide enolate. Thus, start-
ing from the known lactams 1 and 22a (Fig. 1),
N-deprotection and treatment with benzaldehyde
gave the imines 3 and 4 (Fig. 1). Reaction of 3 and 4
with a base and benzyl bromide or allyl bromide yielded
the alkyl derivatives 5–8 (Scheme 1).
We anticipated that due to the steric and electronic fac-
tors in compounds 3 and 4 the reaction with the base
was regioselective, only the proton at C3 was removed,
and no epimerization at C9 or C10, respectively, has
been observed.
As it can be seen from the results collected in Table 1,
formation of the enolate from 3 with LiHMDS followed
by alkylation with benzyl bromide proceeded with mod-
erate yields to afford the (3R) isomer 5a as the major
product (Table 1, entry 1).
It is well documented that the outcome of alkylation
reactions depends on a series of factors such as solvent,5
base counterion6 and temperature. These can dramati-
cally affect both the yield and the stereoselectivity of
the process.
The reactivity of an alkali metal enolate is very sensitive
to its state of aggregation and the maximum reactivity
would be expected in a medium where the cation is
strongly solvatated. So, with the aim of decreasing the
steric demand of enolate, the reaction was conducted
in presence of a polar aprotic solvent such as DMPU;
The purpose of this work was to explore how the reac-
tion conditions can affect yields and steric course of
the alkylation reaction performed at the position C3 of
bicyclic lactams.
Table 1. Reaction conditions for alkylation of compounds 3 and 4
Entry
Imine
Base
T (ꢁC)a
R
Prod.
Yield (%)
Ratio (3R)a/(3S)b
1
3
3
3
3
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
LiHMDS
À78firt
À78firt
À50
–CH2Ph
–CH2Ph
5a,b
5a,b
5a,b
5a,b
5a,b
5a,b
5a,b
5a,b
5a,b
6a,b
6a,b
6a,b
6a,b
6a,b
7a,b
7a,b
7a,b
7a,b
7a,b
7a,b
8a,b
8a,b
8a,b
56
86
89
55
81
24
43
43
33
75
90
78
55
45
53
82
81
73
59
68
67
67
20
92:8b
2
LiHMDS+DMPU
LiHMDS
NaHMDS
82:18b
90:10b
75:25b
81:19b
31:69b
5:95b
3
–CH2Ph
–CH2Ph
4
À78firt
À78firt
À78firt
À78firt
À50fiÀ20
À78firt
À78firt
À50
5
NaHMDS+DMPU
KHMDS
LiHMDS+Mg++
LiHMDS+Mg++
LiHMDS+Sn++
LiHMDS
–CH2Ph
–CH2Ph
6
7
–CH2Ph
–CH2Ph
–CH2Ph
8
>2:98b
21:79b
89:11c
84:16c
84:16c
7:93c
<2:>98c
21:79c
40:60c
10:90c
20:80c
9:91c
9
10
11
12
13
14
15
16
18
19
20
17
21
22
23
–CH2CH@CH2
–CH2CH@CH2
–CH2CH@CH2
–CH2CH@CH2
–CH2CH@CH2
–CH2Ph
LiHMDS
LiHMDS+DMPU
LiHMDS+Mg++
LiHMDS+Mg++
LiHMDS
À78firt
À78firt
À50fiÀ20
À78firt
À50
LiHMDS
NaHMDS
NaHMDS
–CH2Ph
–CH2Ph
À78firt
À50fiÀ20
À78firt
À78firt
À78firt
À50
–CH2Ph
–CH2Ph
–CH2Ph
NaHMDS+DMPU
LiHMDS+Mg++
LiHMDS
<2:>98c
54:46c
55:45c
6:94c
–CH2CH@CH2
–CH2CH@CH2
–CH2CH@CH2
LiHMDS
LiHMDS+Mg++
À78firt
a Base was added at À78ꢁC, then bromide was added at reported temperature.
b Ratio determined by 1H NMR.
c Ratio determined by HPLC.

L. Manzoni et al. / Tetrahedron Letters 45 (2004) 6311–6315
6313
Figure 2. (a) RHF/3-21+G minimum energy conformation of the enolate derived from 6,5-fused bicyclic lactam 3. (b) RHF/3-21+G minimum
energy conformation of the enolate derived from the 7,5-fused bicyclic lactam 4.
a dramatic effect was observed on the yield, which in-
creased until 86% (entry 2) while only a slight effect
can be observed on the diastereisomeric ratio. The same
result has been reached performing the reaction at
À50ꢁC instead of À78ꢁC (entry 3).
As known in the literature, the conformation of the
reactive enolate could drastically change if coordinative
effects are present. A totally reversed stereochemically
outcome can be obtained if the enolate is generated with
LiHMDS and a bicoordinating Lewis acid such as
MgBr2ÆEt2O is added (entries 7 and 8). The same result
can be obtained also using SnCl2 as LA even if with low
yield (entry 9).
The change from lithium to sodium counterion showed
little effect on the yield while the diastereoselectivity
was considerably decreased (entry 4). Again the presence
of DMPU enhanced the yield of the alkylation (entry 5).
On the contrary switching to KHMDS had a dramatic
effect and the preferred product obtained, even if with
moderate yield, was the (3S) isomer 5b (Table 1, entry 6).
The data suggest that there could be different possible
conformations and aggregations of the enolate in
solution, and the equilibrium between chelated and
nonchelated system can favour one or the other diaste-
reoisomer. This was investigated using metal additives
in the alkylation reaction.
The allylation of 3 gave the same results, the tempera-
ture or the presence of DMPU showed influence on
the yield leaving almost unchanged the diastereoiso-
meric ratio (entries 10–12). The presence of MgBr2ÆEt2O
reverses completely the stereochemistry outcome of
the reaction (entries 13 and 14).
By contrast, the benzylation of 4 turned out to afford
selectively the (3S) isomer 7b, independently of the
reaction conditions. In this case the preferred reactive
Figure 3. ORTEP plot of isomer 5b with atom numbering scheme.
Displacement ellipsoids at 30% probability level.
Figure 4. ORTEP plot of isomer 6a with atom numbering scheme.
Displacement ellipsoids at 30% probability level.

6314
L. Manzoni et al. / Tetrahedron Letters 45 (2004) 6311–6315
The pure isomers 5b and 6a were obtained by recrystal-
ization from ethylic ether and their absolute configura-
tion was assigned by X-ray structure analysis.9 On the
basis of the known stereogenic centres of the molecules,
as shown in Figure 3, the X-ray structure of 5b clearly
indicated that the 3-benzyl group is cis to the tert-but-
oxycarbonyl group, which implied a 3S configuration.
By contrast, the X-ray structure of 6a (Fig. 4) indicated
that the 3-allyl group was trans to the tert-butoxycarbo-
nyl group, which implied a 3R configuration. Com-
pound 7b (Fig. 5) was also recrystallized from ethylic
ether and X-ray analysis showed a cis relation between
3-benzyl group and the tert-butoxycarbonyl group une-
quivocally assigning the S configuration at C3 for lac-
tam 7b.
For allyl substituted compounds 8a,b the stereochemis-
try of the newly formed stereocentre was unequivocally
determined by NOE experiments (Fig. 6). In compound
8a a NOE between the protons of benzylamine and H7
assigns the relative configuration at C3 as 3R.
Figure 5. ORTEP plot of isomer 7b with atom numbering scheme.
Displacement ellipsoids at 30% probability level.
In conclusion we have developed a new and versatile
method for the preparation of Ca-substituted peptidomi-
metics based on stereoselective alkylation of a Shiff base
amide enolate. The use of these scaffolds in the synthesis
of biologically active molecules, as well as the applica-
tion of this methodology to the preparation of other
C3-substituted lactams are in progress and will be re-
ported in due course.
Acknowledgements
Figure 6. NOE in compound 8a.
The authors thank CNR and MIUR (COFIN and FIRB
research programs) for financial support.
conformation was not affected by the presence of Lewis
acid or at least was the same with or without coordinat-
ing metals; the enhancing of the diastereoselectivity was
the only effect observed.
Supplementary material
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.tetlet.
2004.06.077.
The allylation on compound 4 afforded 8b as the major
product (entry 23), even if with a very low yield, only
when a LA was present, otherwise low diastereoselectiv-
ity can be reached.
References and notes
The stereochemical outcome of the alkylation reaction
could be rationalized by invoking a pseudo-axial attack7
onto the lowest energy conformers obtained from ab ini-
tio calculations8 for the bicyclic intermediate enolates
(Fig. 2). The preferred geometry of the 6,5-fused enolate
derived from 3 features a pseudo-chair conformation of
the lactam ring (Fig. 2a), that should lead to the 3R 5a
or 6a diastereoisomer if the pseudo-axial attack of the
alkylating reagent is hypothesized. On the contrary,
the same attack to the preferred lactam ring geometry
of the 7,5-fused enolate derived from 4 (Fig. 2b) should
favour the 3S 7b or 8b diastereoisomer. Some other fac-
tors as coordination phenomena, become important
when the alkylation of the bicyclic lactam 3 is conducted
in the presence of bicoordinating metals (Mg, Zn) or
KHMDS; in this case a switching of the favoured face
of attack of the incoming alkylating reagent is observed.
1. (a) Kahn, M. Ed.; Peptide Secondary Structure Mimetics.
Tetrahedron Symposia-in-Print No. 50, 1993, 49, 3433–
3689; (b) Gante, J. Angew. Chem., Int. Ed. Engl. 1994, 33,
1699–1720; (c) Olson, G. L.; Bolin, D. R.; Bonner, M. P.;
Bo¨s, M.; Cook, C. M.; Fry, D. C.; Graves, B. J.; Hatada,
M.; Hill, D. E.; Kahn, M.; Madison, V. S.; Rusiecki, V. K.;
Sarabu, R.; Sepinwall, J.; Vincent, G. P.; Voss, M. E.
J. Med. Chem. 1993, 36, 3039–3049; (d) Kitagawa, O.;
Velde, D. V.; Dutta, D.; Morton, M.; Takusagawa, F.;
`
Aube, J. J. Am. Chem. Soc. 1995, 117, 5169–5178; (e)
Giannis, A.; Kolter, T. Angew. Chem., Int. Ed. Engl. 1993,
32, 1244–1267; (f) Aube, J. Tetrahedron Symposia-in-Print
No. 50, 2000, 56, 9725–9842.
2. (a) Angiolini, M.; Araneo, S.; Belvisi, L.; Cesarotti, E.;
Checchia, A.; Crippa, L.; Manzoni, L.; Scolastico, C. Eur.
J. Org. Chem., 2000, 2571–2581; (b) Belvisi, L.; Bernardi,
A.; Manzoni, L.; Potenza, D.; Scolastico, C. Eur. J. Org.

L. Manzoni et al. / Tetrahedron Letters 45 (2004) 6311–6315
6315
Chem. 2000, 2563–2569; (c) Manzoni, L.; Belvisi, L.;
Scolastico, C. Synlett 2000, 1287–1288.
deposition numbers CCDC 229008–229010, the names of
the authors and the journal citation (fax: +44-1223-336-
033; e-mail: deposit@ccdc.cam.ac.uk; web site: http://
www.ccdc.cam.ac.uk). X-ray data were collected on a
Bruker Smart Apex CCD area detector using graphite-
3. Colombo, L.; Di Giacomo, M.; Belvisi, L.; Manzoni, L.;
Scolastico, C. Gazz. Chim. Ital. 1996, 126, 543–554.
4. (a) Colombo, L.; Di Giacomo, M.; Brusotti, G.; Sardone,
N.; Angiolini, M.; Belvisi, L.; Maffioli, S.; Manzoni, L.;
Scolastico, C. Tetrahedron 1998, 54, 5325–5336; (b)
Colombo, L.; Di Giacomo, M.; Vinci, V.; Colombo, M.;
Manzoni, L.; Scolastico, C. Tetrahedron 2003, 59,
4501–4513.
˚
monochromated Mo-Ka radiation (k=0.71073A). Data
reductions were made using SAINT programs. The struc-
tures were solved by SIR-92 and refined on F2 by full-
matrix least-squares using SHELXL-97. Crystal data for
5b:
C27H34N2O3,
Mr=434.56,
colourless
plate
5. (a) For reviews, see: Parker, A. J. Chem. Rev. 1969, 69,
1–32; (b) Jackman, L. M.; Lange, B. C. Tetrahedron 1977,
33, 2737–2769.
6. Caine, D. In Carbon–Carbon Bond Formation; Augustine,
R. L., Ed.; Dekker: New York, 1979; Vol. 85, and
references cited therein.
7. (a) House, H. O.; Umen, M. J. J. Org. Chem. 1973, 38,
1000–1003; (b) Houk, K. N.; Paddon-Row, M. N. J. Am.
Chem. Soc. 1986, 108, 2659–2662; (c) Johnson, F. Chem.
Rev. 1968, 68, 375–413.
8. Ab initio geometry optimizations were performed at the
RHF/3-21+G level using Gaussian 98 (Frisch, M. J. et al.,
Gaussian, Inc.: Pittsburgh, PA, 1998). Starting conforma-
tion for the 6,5-trans and 7,5-trans fused enolates deriving
from 3 and 4 were generated by molecular mechanics
calculations using the Macromodel version 5.5 program
and the Macromodel (Mohamadi, F.; Richards, N. G. J.;
Guida, W. C.; Liskamp, R.; Lipton, M.; Caufield, C.;
Chang, G.; Hendrickson, T.; Still, W. C. J. Comput. Chem.
1990, 11, 440–467) implementation of the AMBER force
field (Weiner, S. J.; Kollman, P. A.; Nguyen, D. T.; Case,
D. A. J. Comput. Chem. 1986, 7, 230–252).
0.33·0.20·0.08mm, orthorhombic, P212121, a=6.2365(10),
3
˚
˚
b=18.367(2), c=21.042(2)A, V=2410.2(5)A , Z=4,
T=293(2)K, l=0.078mmÀ1. 41452 measured reflections,
2459 independent reflections, 2413 reflections with I>2r(I),
2.94<2h<50.00ꢁ, Rint =0.066. Refinement on 2459 reflec-
tions, 292 parameters. Final R=0.0591, wR=0.1309 for
data with F2>2r(F2), (D/r)max =0.001, Dqmax =0.16,
Dqmin =À0.17eAÀ3. Crystal data for 6a: C28H36N2O3,
˚
Mr =448.59, colourless prism 0.40·0.34·0.20mm, mono-
clinic, P21, a=6.5233(5), b=11.5837(8), c=17.0095(10)A,
˚
3
˚
b=95.96(1)ꢁ, V=1278.36(15)A , Z=2, T=293(2)K,
l=0.075mmÀ1. 35023 measured reflections, 3091 inde-
pendent reflections, 2556 reflections with I>2r(I),
2.40<2h<55.00ꢁ, Rint =0.058. Refinement on 3091 reflec-
tions, 301 parameters. Final R=0.0461, wR=0.1240 for
data with F2>2r(F2), (D/r)max =0.000, Dqmax =0.20,
Dqmin =À0.23eAÀ3. Crystal data for 7b: C23H32N2O3ÆH2O,
˚
Mr =402.52, colourless prism 0.46·0.18·0.08mm, mono-
clinic, P21, a=6.6256(6), b=16.9585(15), c=19.8997(17)A,
˚
3
˚
b=92.43(2)ꢁ,
V=2233.9(3)A ,
Z=4,
T=150(2)K,
l=0.081mmÀ1. 27304 measured reflections, 6724 inde-
pendent reflections, 4007 reflections with I>2r(I),
4.74<2h<60.00ꢁ, Rint =0.094. Refinement on 6724 reflec-
tions, 727 parameters. Final R=0.0415, wR=0.0649 for
data with F2>2r(F2), (D/r)max =0.002, Dqmax =0.21,
9. Tables of atomic coordinates, anisotropic thermal para-
meters, bond lengths and angles of isomers 5b, 6a and 7b
may be obtained free of charge from The Director CCDC,
12 Union Road, Cambridge CB2 1 EZ, UK, on quoting the
À3
˚
Dqmin =À0.17eA
.