A straightforward route to enantiopure 2-substituted-3,4-dehydro-β- proline via ring closing metathesis

Article abstract of DOI:10.1007/s00726-011-0921-3

The synthesis of unusual cyclic amino acids, that may be envisaged as proline analogs, is an area of great interest for their potential applications as scaffolds for the design of bioactive peptidomimetics or units for the creation of novel foldamers. We have carried out the preparation of cyclic dehydro-β-amino acids starting from allylic carbonates via a two-step allylic amination/ring closing metathesis (RCM) protocol. The introduction of the allylamino moiety has been carried out either without a catalyst, through an S_N2′ reaction, or in the presence of iridium complexes. The backbone of the allylamino intermediates contains two unsaturations, thus suggesting that RCM could be a valuable tool for the preparation of dihydropyrrole scaffolds. A similar reaction has been already reported in the literature for racemic aromatic-substituted substrates, but no examples of enantiopure derivatives bearing aliphatic chains have been reported. The reaction was optimized by testing different Grubbs' catalysts and carbamate nitrogen protecting groups. Moreover, in view of a future application of these dehydro-β-amino acids as central core of peptidomimetics, the malonate chain was also used to protect nitrogen prior to RCM.

Full text of DOI:10.1007/s00726-011-0921-3

Amino Acids (2011) 41:575–586
DOI 10.1007/s00726-011-0921-3
ORIGINAL ARTICLE
A straightforward route to enantiopure 2-substituted-3,4-
dehydro-b-proline via ring closing metathesis
•
Alessandra Tolomelli Luca Gentilucci
•
•
Elisa Mosconi Angelo Viola Enrico Paradisi
•
Received: 31 January 2011 / Accepted: 16 April 2011 / Published online: 4 May 2011
Ó Springer-Verlag 2011
Abstract The synthesis of unusual cyclic amino acids,
that may be envisaged as proline analogs, is an area of
great interest for their potential applications as scaffolds for
the design of bioactive peptidomimetics or units for the
creation of novel foldamers. We have carried out the
preparation of cyclic dehydro-b-amino acids starting from
allylic carbonates via a two-step allylic amination/ring
closing metathesis (RCM) protocol. The introduction of the
allylamino moiety has been carried out either without a
catalyst, through an SN20 reaction, or in the presence of
iridium complexes. The backbone of the allylamino inter-
mediates contains two unsaturations, thus suggesting that
RCM could be a valuable tool for the preparation of di-
hydropyrrole scaffolds. A similar reaction has been already
reported in the literature for racemic aromatic-substituted
substrates, but no examples of enantiopure derivatives
bearing aliphatic chains have been reported. The reaction
was optimized by testing different Grubbs’ catalysts and
carbamate nitrogen protecting groups. Moreover, in view
of a future application of these dehydro-b-amino acids as
central core of peptidomimetics, the malonate chain was
also used to protect nitrogen prior to RCM.
Introduction
Heterocyclic rings represent the most common scaffolds
for the preparation of bioactive small molecules (Schuf-
fenhauer et al. 2007) and for this reason novel synthetic
pathways leading to their synthesis are always of great
interest. These structures are often the focal point of more
complex molecules, since they are efficiently recognized
by proteic receptors through the formation of hydrogen
bonds (Abell 2002). Moreover, the addition of functional-
ized side chains to the heterocycle opens the easy access to
small libraries of derivatives, differing for the substitution
pattern. Recently, the term ‘‘privileged structure’’ has
appeared frequently in the literature to define recurring
structural elements that are likely to facilitate binding with
biological targets, inducing ordered spatial disposition of
side chains (Mu¨ller 2003). The cyclic structure of proline
gives to this residue a particular conformational rigidity
due to the reduced allowed rotations (McDonald and Still
1996), and for this reason it may be considered a privileged
amino acid, where a turned peptide backbone is required
for bioactivity. The involvement of peptides in physio-
logical processes fundamental to life makes these mole-
cules particularly attractive as starting point for drug
discovery programs dealing with the preparation of bio-
active mimetics. For this reason, the use of proline analogs
as conformational restraints or rigid cores in peptidomi-
metics has been extensively explored. Among the number
of modified structures, b-proline has received great atten-
tion since the substitution of an a-amino acid with the
corresponding b-analog in key positions of a bioactive
sequence represents a successful approach to obtain mi-
metics overcoming the low bioavailability limitations of
natural peptides (Cheng et al. 2001). For instance, endo-
morphin-1 analogs containing b-proline, synthesized by
Keywords Dehydro-b-proline ꢀ Allylic amination ꢀ
Iridium catalysis ꢀ Ring closing metathesis ꢀ
Grubbs’ catalyst
Electronic supplementary material The online version of this
article (doi:10.1007/s00726-011-0921-3) contains supplementary
material, which is available to authorized users.
A. Tolomelli (&) ꢀ L. Gentilucci ꢀ E. Mosconi ꢀ A. Viola ꢀ
E. Paradisi
Department of Chemistry, ‘‘G. Ciamician’’, University
of Bologna, Via Selmi 2, 40126 Bologna, Italy
e-mail: alessandra.tolomelli@unibo.it
123

576
A. Tolomelli et al.
our group, showed good receptor affinity together with
enhanced stability towards enzymatic hydrolysis (Cardillo
et al. 2003). Pyrrolidine-3-carboxylic acid (Pca) and nipe-
cotic acid (Nip), b-substituted equivalents of Pro and the
six membered ring pipecolinic acid, appear to adopt well-
defined conformations, and dipeptide sequences incorpo-
rating Nip strongly stabilize reverse turn formation in a
manner similar to the stabilization of reverse turns by Pro-
containing dipeptides (Abele and Seebach 1999). On the
other hand, unsaturated a-amino acids have been also
explored and 3,4-dehydroproline (Dhp), the analog of
proline having a double bond between Cb and Cc, showed
to induce changes in the structure and biological properties
of collagen, proteins and peptides (Kang and Park 2009). In
the last few years, we have been involved in the synthesis
of small libraries of avb3- and a5b1-integrin ligands, with
the aim to obtain anticancer molecules for therapeutical as
well as for diagnostic use. To this purpose, we needed to
prepare polyfunctionalized heterocyclic scaffolds to be
linked to the proper pharmacophoric groups and submitted
to bioassays (Benfatti et al. 2007a) .
enantiopure structures may be applied as rigid scaffolds for
the preparation of bioactive peptidomimetics.
Experimental
General
All chemicals were purchased from commercial suppliers
and used without further purification. Anhydrous solvents
were purchased in sure seal bottles over molecular sieves
and used without further drying. Flash chromatography
was performed on silica gel (230–400 mesh). DOWEXÒ
50WX2-200(H) ion exchange resin was used for purifica-
tion of free amino acids. NMR Spectra were recorded with
Varian Gemini 200, Mercury Plus 400 or Unity Inova
600 MHz spectrometers. Chemical shifts were reported as
d values (ppm) relative to the solvent peak of CDCl3 set at
d = 7.27 (1H NMR) or d = 77.0 (13C NMR), CD3OD set
at d = 3.31 (1H NMR) or d = 49.0 (13C NMR), D2O set at
d = 4.79 (1H NMR). Coupling constants are given in
Hertz. The enantiomeric excesses of products were deter-
mined by HPLC analyses performed on an HP1100
instrument with UV–VIS detector and equipped with
Chiralpak IC column (25 9 0.46 cm) or Chiralpak IA
column (25 9 0.46 cm), eluted with n-hexane/2-propanol
(typical program: from 96/4 to 90/10 in 30 min, flux
0.6 mL/min). Optical rotations were measured in a Perkin
Elmer 343 polarimeter using a 1 dm cuvette and are ref-
erenced to the Na-D line value. Melting points were
determined on a Stuart Scientific SMP3 apparatus and are
uncorrected. LC–MS analyses was performed on a HP1100
liquid chromatograph coupled with an electrospray ioni-
zation-mass spectrometer (LC–ESI–MS), using H2O/
CH3CN as solvent at 25°C (positive scan 100–500 m/z,
fragmentor 70 V). Elemental analyses have been per-
formed using a Thermo Flash 2000 CHNS/O analyzer.
Compounds 1a–f were prepared following the procedure
already reported (Cardillo et al. 2010).
We have recently reported the preparation of enantio-
pure allylic alcohols via lipase catalyzed enzymatic reso-
lution and their use as starting material for the synthesis of
dehydro-b-amino esters (Benfatti et al. 2007b). The trans-
formation of alcohols into amines was achieved via allylic
amination of the corresponding carbonates in the presence
of palladium catalysts or without catalysis. The process
was completely regio- and stereoselective, allowing to
obtain exclusively one of the two possible regioisomers
depending on the reaction conditions. In particular, the
uncatalyzed reaction proceeds via SN20 mechanism, while
under palladium-catalyzed conditions, the reaction shows a
strong solvent-dependent regiocontrol. The regioisomer
deriving SN20 is predominant when the reaction is per-
formed in THF and the opposite one is exclusively
obtained from the reaction in CH3CN. Anyway, in both
processes complete transfer of chirality from the substrates
to the products was observed, with retention of the ste-
reocenter configuration (Benfatti et al. 2008).
Following the same synthetic pathway, we envisaged
N-allyl-dehydro-b-amino esters as suitable intermediates for
the preparation of enantiopure substituted a-alkyl-dehydro-
b-prolines via ring closing methatesis (RCM) (Fig. 1). These
General procedure for allylic amination without catalyst
To a stirred solution of 1 (1 mmol) in CH3CN (5 mL)
under nitrogen atmosphere, allylamine (9 mmol, 9 equiv)
was added in one portion and the mixture was wormed to
reflux. After 3 days, the solvent and the excess of allyla-
mine were removed under reduced pressure and the residue
was diluted with ethyl acetate (10 mL). The organic solu-
tion was extracted three times with 0.1 M HCl (10 mL),
then it was dried over Na2SO4, and solvent was removed
under reduced pressure to recover unreacted 1. NaOH (1 M
solution) was added dropwise to the acid aqueous layer
until basic pH was achieved and then the solution was
Fig. 1 Retrosynthetic pathway to enantiopure 2-substituted-3,4-
dehydro-b-proline
123

Enantiopure 2-substituted-3,4-dehydro-b-proline via ring closing metathesis
Table 1 Optimization of allylic amination reaction conditions
577
Entry
Carbonate
Solvent
Allylamine (equiv)
Catalyst
Time (h)
Yielda (%)
2:3
b
1
2
3
4
5
6
7
8
9
rac-Z-1a
(S)-Z-1ac
(R)-Z-1ac
(S)-Z-1bc
(R)-Z-1bc
rac-Z-1e
rac-Z-1f
rac-Z-1a
rac-Z-1a
CH3CN
CH3CN
CH3CN
CH3CN
CH3CN
CH3CN
CH3CN
THF
9
9
/
72
72
72
72
72
72
72
72
10
86
85
[99:1
[99:1b
[99:1b
[99:1b
[99:1b
[99:1b
[99:1b
[99:1
[99:1
/
9
/
85
9
/
65
9
/
68
9
/
50
9
/
85
15
3
Pd2(dba)3CHCl3
[Ir(COD)Cl]2 (2%)
P(OPh)3 (8%)
/
35
EtOH
[99
10
11
12
rac-Z-1a
rac-Z-1a
rac-Z-1c
EtOH
EtOH
EtOH
3
3
3
16
16
8
52
60
[99:1
[99:1
[99:1d
[Ir(COD)Cl]2 (2%)
[Ir(COD)Cl]2 (2%)
P(OPh)3 (8%)
[Ir(COD)Cl]2 (2%)
P(OPh)3 (8%)
[Ir(COD)Cl]2 (2%)
P(OPh)3 (8%)
[Ir(COD)Cl]2 (2%)
P(OPh)3 (8%)
[99
13
14
15
a
rac-Z-1d
rac-Z-1e
rac-Z-1f
EtOH
EtOH
EtOH
3
3
3
10
10
10
95
[99
90
[99:1e
75:25
60:40
Yields were calculated after purification of the products by flash chromatography on silica gel. The unreacted carbonate 1 was completely
recovered after work-up as pure Z-isomer
b
Compound 2 was always obtained as 4/1 mixture of Z/E isomers
c
Enantiomeric excess of substrates and products was determined by HPLC on chiral column (always [95% ee)
In this case only the one product is formed for the symmetry of the molecule
d
e
Obtained as a 5/1 mixture of Z/E isomers, each as 1:1 diastereomeric mixture
extracted three times with CH2Cl2 (10 mL). The organic
layers were dried over Na2SO4 and solvent was removed
under reduced pressure to afford allylamino derivative 2 as
a yellow oil (65–86%, see Table 1), Compound 2 was
purified by flash chromatography on silica gel (cyclohex-
ane/ethyl acetate 70:30 as eluant).
acid aqueous layer until basic pH was achieved and then
the solution was extracted three times with CH2Cl2
(10 mL). The organic layers were dried over Na2SO4 and
solvent was removed under reduced pressure to afford al-
lylamino derivative 2 in quantitative yield as a yellow oil.
Compound 2 was purified by flash chromatography on
silica gel (cyclohexane/ethyl acetate 70:30 as eluant).
Z-tert-Butyl 3-(allylamino)-2-ethylidene-4-methylpent-
anoate 2a: Yellow oil; 1H NMR (200 MHz, CDCl3) d 0.68
(d, J = 6.6 Hz, 3H), 1.02 (d, J = 6.6 Hz, 3H), 1.41 (s, 9H),
1.69 (d, J = 7.4 Hz, 3H), 1.97 (m, 1H), 2.91 (dd, J = 14.0,
6.2 Hz, 1H), 3.05 (d, J = 9.2 Hz, 1H), 3.16 (dd, J = 14.0,
5.4 Hz, 1H), 5.03–5.18 (m, 2H), 5.89 (m, 1H), 6.91 (q, J =
7.2 Hz, 1H); 13C NMR (75 MHz, CDCl3) d 13.9, 20.0, 21.0,
28.1, 32.0, 49.7, 61.4, 80.4, 115.0, 133.3, 134.3, 138.4, 166.7.
LC–MS–ESI rt 14.97 min, 254 (M?1). (S)-Z-2a [a]D =
? 19.5 (c 1 in CHCl3); (R)-Z-2a [a]D = -20.6 (c 1 in
CHCl3); Anal. cald. for C15H27NO2 (253.20): C 71.10, H
10.74, N 5.53; found C 70.84, H 10.65, N 5.57.
General procedure for allylic amination in the presence
of Iridium catalyst
To a stirred solution of [Ir(COD)(Cl)]2 (0.02 mmol) and
phosphorous ligand (0.08 mmol) in EtOH (5 mL) under
nitrogen atmosphere, carbonate 1 (1 mmol) and allylamine
(3 mmol, 3 equiv.) were added in one portion at room
temperature. The solution was stirred under refluxing
conditions and monitored by TLC. After removal of the
solvent and the excess of allylamine under reduced pres-
sure, the residue was diluted with ethyl acetate (10 mL).
The organic solution was extracted three times with 0.1 M
HCl (10 mL), then was dried over Na2SO4 and solvent was
removed under reduced pressure to recover unreacted
starting carbonate. 1 M NaOH was added dropwise to the
E-tert-Butyl 3-(allylamino)-2-ethylidene-4-methylpent-
anoate 2a: Yellow oil; 1H NMR (200 MHz, CDCl3) d 0.77
(d, J = 6.6 Hz, 3H), 0.89 (d, J = 6.6 Hz, 3H), 1.40 (s,
123

578
A. Tolomelli et al.
9H), 1.84 (d, J = 7.0 Hz, 3H), 1.97 (m, 1H), 2.72 (d,
J = 8.2 Hz, 1H), 2.91 (dd, J = 14.0, 6.2 Hz, 1H), 3.16
(dd, J = 14.0, 5.4 Hz, 1H), 5.03-5.18 (m, 2H), 5.68 (q,
J = 7.0 Hz, 1H); 5.89 (m, 1H); 13C NMR (75 MHz,
CDCl3) d 14.8, 20.0, 21.0, 28.0, 32.1, 49.8, 61.4, 80.0,
115.0, 133.3, 134.8, 137.8, 167.8. LC–MS–ESI rt
15.84 min, 254 (M?1). Anal. cald. for C15H27NO2
(253.20): C 71.10, H 10.74, N 5.53; found C 71.22, H
10.68, N 5.50.
167.4. LC–MS–ESI rt 11.44 min, 226 (M?1). Anal. cald.
for C13H23NO2 (225.17): C 69.29, H 10.29, N 6.22; found
C 69.58, H 10.24, N 6.25.
Z-tert-Butyl 3-(allylamino)-2-ethylidene-4-methylhexe-
1
noate 2d: Yellow oil; H NMR (400 MHz, CDCl3) d 0.70
(d, J = 7.2 Hz, 3H), 1.05 (d, J = 7.2 Hz, 3H), 1.00–1.82
(m, 3H), 1.46 (s, 9H), 1.74 (d, J = 7.2 Hz, 3H), 2.82–3.28
(m, 3H), 4.99–5.16 (m, 2H), 5.82 (m, 1H), 6.87 (q,
J = 7.2 Hz, 1H); 13C NMR (75 MHz, CDCl3) d 11.6,
13.9,16.2, 26.5, 28.4, 38.7, 50.4, 59.6, 80.8, 115.8, 132.6,
134.9, 137.8, 164.1. LC–MS–ESI rt 16.11 min, 268
(M?1). Anal. cald. for C16H29NO2 (267.22): C 71.86, H
10.93, N 5.24; found C 71.64, H 10.95, N 5.25.
Z-tert-Butyl 2-[(allylamino)(cyclohexyl)methyl]but-2-e-
1
noate 2b: Yellow oil; H NMR (400 MHz, CDCl3) d 0.78
(m, 1H), 0.92 (m, 1H), 1.11–1.25 (m, 4H), 1.42 (s, 9H),
1.49 (m, 4H), 1.76 (d, J = 6.8 Hz, 3H), 2.28 (m, 1H), 2.96
(dd, J = 17.6, 6.6 Hz, 1H), 3.20 (d, J = 9.2 Hz, 1H), 3.22
(dd, J = 17.6, 5.6 Hz, 1H), 5.04 (dd, J = 1.4, 10.0 Hz,
1H), 5.13 (dd, J = 1.4, 17.2 Hz, 1H), 5.88 (m, 1H), 6.89
(q, J = 7.2 Hz, 1H); 13C NMR (75 MHz, CDCl3) d 14.0,
26.3, 26.7, 28.2, 30.7, 31.7, 41.6, 60.1, 80.2, 115.2, 134.0,
137.9, 138.6, 166.9. LC–MS–ESI rt 16.21 min, 294
(M?1). (S)-Z-2a [a]D = ? 14.3 (c 1 in CHCl3); (R)-Z-2a
[a]D = -16.6 (c 1 in CHCl3). Anal. cald. for C18H31NO2
(293.24): C 73.67, H 10.65, N 4.77; found C 73.91, H
10.61, N 4.78.
E-tert-Butyl 3-(allylamino)-2-ethylidene-4-methylhexe-
1
noate 2d: Yellow oil; H NMR (400 MHz, CDCl3) d 0.77
(d, J = 7.0 Hz, 3H), 0.84 (d, J = 6.8 Hz, 3H), 1.00–1.82
(m, 3H), 1.48 (s, 9H), 1.90 (d, J = 7.2 Hz, 3H), 2.82–3.28
(m, 3H), 4.99–5.16 (m, 2H), 5.71–5.96 (m, 2H); 13C NMR
(75 MHz, CDCl3) d 12.2, 13.3,18.2, 26.1, 28.3, 36.2, 50.0,
60.1, 81.6, 118.0, 132.1, 133.7, 138.1, 167.3. LC–MS–ESI
rt 13.34 min, 268 (M?1). Anal. cald. for C16H29NO2
(267.22): C 71.86, H 10.93, N 5.24; found C 72.00, H
10.88, N 5.21.
E-tert-Butyl 2-[(allylamino)(cyclohexyl)methyl]but-2-e-
1
Z-tert-Butyl 2-[(allylamino)(phenyl)methyl]but-2-enoate
2e: Yellow oil; 1H NMR (400 MHz, CDCl3) d 1.31 (s, 9H),
1.93 (d, J = 7.6 Hz, 3H), 3.21 (dd, J = 14.0, 6.0 Hz, 1H),
3.32 (dd, J = 14.0, 5.6 Hz, 1H), 4.83 (s, 1H), 5.11 (dd,
J = 1.6, 10.4 Hz, 1H), 5.22 (dd, J = 1.6,16.8 Hz, 1H),
5.99 (m, 1H), 7.01 (q, J = 7.6 Hz, 1H), 7.18-7.43 (m, 5H);
13C NMR (75 MHz, CDCl3) d 14.5, 28.8, 49.8, 60.6, 81.6,
116.8, 127.4, 127.9, 128.5, 132.1, 133.9, 136.0, 139.2,
166.4. LC–MS–ESI rt 20.22 min, 288 (M?1). Anal. cald.
for C18H25NO2 (287.19): C 75.22, H 8.77, N 4.87; found C
75.40, H 8.72, N 4.84.
noate 2b: Yellow oil; H NMR (400 MHz, CDCl3) d 0.78
(m, 1H), 0.92 (m, 1H), 1.11–1.25 (m, 4H), 1.42 (s, 9H),
1.49 (m, 4H), 1.90 (d, J = 6.8 Hz, 3H), 2.28 (m, 1H), 2.82
(d, J = 8.6 Hz, 1H), 2.96 (dd, J = 17.6, 6.6 Hz, 1H), 3.20
(d, J = 9.2 Hz, 1H), 3.22 (dd, J = 17.6, 5.6 Hz, 1H), 5.04
(dd, J = 1.4, 10.0 Hz, 1H), 5.13 (dd, J = 1.4, 17.2 Hz,
1H), 5.78(q, J = 7.2 Hz, 1H); 5.89 (m, 1H); 13C NMR
(75 MHz, CDCl3) d 15.0, 26.4, 26.7, 28.2, 30.6, 31.7, 41.8,
68.1, 80.6, 115.2, 133.6, 137.8, 138.6, 166.9. LC–MS–ESI
rt 16.95 min, 294 (M?1). Anal. cald. for C18H31NO2
(293.24): C 73.67, H 10.65, N 4.77; found C 73.39, H
10.60, N 4.75.
E-tert-Butyl 2-[(allylamino)(phenyl)methyl]but-2-enoate
2e: Yellow oil; 1H NMR (400 MHz, CDCl3) d 1.30 (s, 9H),
1.94 (d, J = 7.2 Hz, 3H), 3.12 (dd, J = 14.0, 6.4 Hz, 1H),
3.20 (dd, J = 14.0, 6.0 Hz, 1H), 4.45 (s, 1H), 5.05 (dd,
J = 1.6, 10.4 Hz, 1H), 5.15 (dd, J = 1.6,17.2 Hz, 1H), 5.88
Z-tert-Butyl 3-(allylamino)-2-ethylidene-butenoate 2c:
Yellow oil; 1H NMR (400 MHz, CDCl3) d 1.33 (d,
J = 7.2 Hz, 3H), 1.50 (s, 9H), 1.79 (d, J = 7.2 Hz, 3H), 3.07
(dd, J = 14.0, 6.0 Hz, 1H), 3.18 (dd, J = 14.0, 6.0 Hz, 1H),
3.75 (q, J = 7.2 Hz, 1H), 5.06–5.17 (m, 2H), 5.92 (m, 1H),
6.81 (q, J = 7.2 Hz, 1H); 13C NMR (75 MHz, CDCl3) d
13.8, 20.5, 28.3, 50.1, 59.8, 81.4, 117.4, 132.3, 135.8, 137.6,
166.5. LC–MS–ESI rt 10.32 min, 226 (M?1). Anal. cald. for
C13H23NO2 (225.17): C 69.29, H 10.29, N 6.22; found C
69.03, H 10.10.27, N 6.22.
(m, 1H), 6.09 (q, J = 7.2 Hz, 1H), 7.18–7.43 (m, 5H); 13
C
NMR (75 MHz, CDCl3) d 16.1, 28.7, 50.1, 60.2, 83.4, 117.9,
127.5, 127.8, 128.2, 132.6, 134.6, 137.2, 138.9, 167.9.
LC–MS–ESI rt 21.56 min, 288 (M?1). Anal. cald. for
C18H25NO2 (287.19): C 75.22, H 8.77, N 4.87; found C
75.01, H 8.75, N 4.86.
E-tert-Butyl 3-(allylamino)-2-benzylidene-butanoate 3e:
Yellow oil; 1H NMR (400 MHz, CDCl3) d 1.38 (d,
J = 6.8 Hz, 3H), 1.49 (s, 9H), 2.83 (dd, J = 14.0, 6.4 Hz,
1H), 3.03 (dd, J = 14.0, 7.6 Hz, 1H), 3.84 (q, J = 6.8 Hz,
1H), 4.83–4.90 (m, 2H), 5.71 (m, 1H), 7.18-7.31 (m, 5H),
7.61 (s, 1H); 13C NMR (75 MHz, CDCl3) d 21.2, 28.4,
51.3, 58.2, 80.9, 117.4, 127.5, 127.7, 128.0, 129.1, 134.9,
136.7, 138.0, 165.9. LC–MS–ESI rt 18.44 min, 288
E-tert-Butyl 3-(allylamino)-2-ethylidene-butenoate 2c:
Yellow oil; 1H NMR (400 MHz, CDCl3) d 1.44 (d,
J = 6.8 Hz, 3H), 1.53 (s, 9H), 1.91 (d, J = 6.8 Hz, 3H),
3.11 (dd, J = 13.8, 6.0 Hz, 1H), 3.24 (dd, J = 13.8,
6.0 Hz, 1H), 3.36 (q, J = 6.8 Hz, 1H), 5.06-5.17 (m, 2H),
5.86-5.96 (m, 2H); 13C NMR (75 MHz, CDCl3) d 14.1,
20.2, 28.3, 51.1, 60.0, 81.2, 116.9, 132.2, 135.6, 137.8,
123

Enantiopure 2-substituted-3,4-dehydro-b-proline via ring closing metathesis
579
(M?1). Anal. cald. for C18H25NO2 (287.19): C 75.22, H
8.77, N 4.87; found C 75.33, H 8.77, N 4.85.
(293.14): C 65.49, H 7.90, N 4.77, S 10.93; found C 65.44,
H 7.94, N 4.77, S 10.90.
Z-tert-Butyl 3-(allylamino)-2-benzylidene-butanoate 3e:
1
Synthesis of N-BOC derivative 4a: Boc2O (1 mmol,
1 eq.) was added to a solution of 2a in EtOH (5 mL) at
room temperature and the reaction mixture was left stirring
for 3 h. Then, the solvent was removed under reduced
pressure and the residue was diluted with EtOAc (10 mL).
After washing three times with water (5 mL), the organic
layer was dried over Na2SO4 and solvent was removed
under reduced pressure. Compound 4a was purified by
flash chromatography on silica gel (cyclohexane/ethyl
acetate 70:30 as eluant).
Yellow oil; H NMR (400 MHz, CDCl3) d 1.43 (s, 9H),
1.45 (d, J = 6.8 Hz, 3H), 2.84 (dd, J = 14.0, 6.4 Hz, 1H),
3.02 (dd, J = 14.0, 7.6 Hz, 1H), 4.59 (q, J = 6.8 Hz, 1H),
4.83–4.90 (m, 2H), 5.78 (m, 1H), 6.82 (s, 1H), 7.15–7.40
(m, 5H); 13C NMR (75 MHz, CDCl3) d 19.2, 28.3, 51.3,
58.6, 81.2, 116.9, 127.5, 127.6, 128.1, 129.1, 134.6, 135.9,
138.6, 167.0. LC–MS–ESI rt 17.10 min, 288 (M?1). Anal.
cald. for C18H25NO2 (287.19): C 75.22, H 8.77, N 4.87;
found C 75.11, H 8.81, N 4.87.
Z-tert-Butyl 2-[(allylamino)(thiophenyl)methyl]but-2-e-
1
Z-tert-Butyl 3-(N-(tert-butoxycarbonyl)allylamino)-2-
ethylidene-4-methylpentanoate 4a: Yellow oil; 1H NMR
(400 MHz, CDCl3) d 0.79 (d, J = 6.4 Hz, 3H), 0.90 (d,
J = 6.4 Hz, 3H), 1.48 (s, 9H), 1.53 (s, 9H), 1.92 (d,
J = 7.4 Hz, 3H), 2.62 (m, 1H), 3.82-4.01 (m, 2H), 4.81(d,
J = 11.2 Hz, 1H), 4.94-5.04 (m, 2H), 5.74 (m, 1H), 6.83
(q, J = 7.2 Hz, 1H); 13C NMR (75 MHz, CDCl3) d 14.2,
19.7, 27.7, 27.9, 28.1, 45.6, 58.4, 78.8, 79.9, 114.4, 133.9,
136.1, 142.4, 155.8, 166.7. LC–MS–ESI rt 16.6 min 354
(M?1), 729 (2M?23). Anal. cald. for C20H35NO4
(353.26): C 67.95, H 9.98, N 3.96; found C 67.77, H 10.00,
N 3.94.
noate 2f: Yellow oil; H NMR (400 MHz, CDCl3) d 1.42
(s, 9H), 1.97 (d, J = 7.2 Hz, 3H), 3.26 (dd, J = 14.0,
5.6 Hz, 1H), 3.38 (dd, J = 14.0, 5.6 Hz, 1H), 4.89 (s, 1H),
5.16–5.31 (m, 2H), 6.03 (m, 1H), 7.09 (q, J = 7.2 Hz, 1H),
7.10 (m, 1H), 7.24–7.35 (m, 2H); 13C NMR (75 MHz,
CDCl3) d 13.9, 28.4, 49.9, 57.6, 81.3, 115.9, 120.3, 125.4,
128.1, 132.4, 134.4, 136.5, 139.0, 168.0. LC–MS–ESI rt
19.02 min, 294 (M?1). Anal. cald. for C16H23NO2S
(293.14): C 65.49, H 7.90, N 4.77, S 10.93; found C 65.64,
H 7.91, N 4.76, S 10.98.
E-tert-Butyl 2-[(allylamino)(thiophenyl)methyl]but-2-e-
1
noate 2f: Yellow oil; H NMR (400 MHz, CDCl3) d 1.44
E-tert-Butyl 3-(N-(tert-butoxycarbonyl)allylamino)-2-
ethylidene-4-methylpentanoate 4a: Yellow oil; 1H NMR
(400 MHz, CDCl3) d 0.92(d, J = 6.8 Hz, 6H), 1.48 (s,
9H), 1.53 (s, 9H), 1.92 (d, J = 7.4 Hz, 3H), 2.62 (m, 1H),
3.82–4.01 (m, 2H), 4.62(d, J = 10.8 Hz, 1H), 4.94-5.04
(m, 2H), 5.48 (q, J = 7.2 Hz, 1H), 5.76 (m, 1H); 13C NMR
(75 MHz, CDCl3) d 14.2, 20.1, 27.7, 27.9, 29.4, 45.3, 59.6,
78.8, 79.4, 114.4, 133.9, 135.5, 141.6, 155.4, 166.7 LC–
MS–ESI rt 14.9 min 354 (M?1), 729 (2M?23). Anal. cald.
for C20H35NO4 (353.26): C 67.95, H 9.98, N 3.96; found C
68.26, H 9.95, N 3.96.
(s, 9H), 2.05(d, J = 7.2 Hz, 3H), 3.22–3.34 (m, 2H), 4.57
(s, 1H), 5.16–5.31 (m, 2H), 5.98–6.15 (m, 2H), 7.12 (m,
1H), 7.24–7.35 (m, 2H); 13C NMR (75 MHz, CDCl3) d
14.6, 28.3, 49.7, 57.0, 80.9, 117.1, 122.2, 125.6, 128.0,
132.1, 135.0, 136.9, 140.2, 166.5. LC–MS–ESI rt
16.43 min, 294 (M?1). Anal. cald. for C16H23NO2S
(293.14): C 65.49, H 7.90, N 4.77, S 10.93; found C 65.77,
H 7.88, N 4.74, S 10. 94.
E-tert-Butyl 3-(allylamino)-2-(thiophenyl)methylidene-
1
butanoate 3f: Yellow oil; H NMR (400 MHz, CDCl3) d
1.57 (s, 9H), 1.61 (d, J = 6.8 Hz, 3H), 2.81 (dd, J = 14.0,
6.0 Hz, 1H), 3.02 (dd, J = 14.0, 5.6 Hz, 1H), 4.02 (q,
J = 6.8 Hz, 1H), 4.90-5.10 (m, 2H), 5.91 (m, 1H), 7.09
8 m, 1H), 7.24–7.40 (m, 2H), 7.68 (s, 1H); 13C NMR
(75 MHz, CDCl3) d 20.5, 28.4, 49.9, 50.7, 81.0, 116.6,
125.7, 126.3, 128.3, 130.3, 133.5, 135.6, 136.2, 167.2. LC–
MS–ESI rt 19.00 min, 294 (M?1). Anal. cald. for
C16H23NO2S (293.14): C 65.49, H 7.90, N 4.77, S 10.93;
found C 65.27, H 7.93, N 4.76, S 10.98.
Synthesis of N-CBz derivative 5a: Benzyloxycarbonyl
chloride (1.5 mmol, 1.5 equiv.) and K2CO3 (2 mmol, 2
equiv) were added to a solution of 2a (1 mmol) in dioxane/
water (1:1 solution, 5 mL). The reaction mixture was
stirred at room temperature overnight. The solution was
diluted ethyl acetate (10 mL) and washed three times with
water (10 mL). The organic layer was dried over Na2SO4
and solvent was removed under reduced pressure. Com-
pound 5a was purified by flash chromatography on silica
gel (eluant cyclohexane/EtOAc 97:3).
Z-tert-Butyl 3-(allylamino)-2-(thiophenyl)methylidene-
1
butanoate 3f: Yellow oil; H NMR (400 MHz, CDCl3) d
Z-tert-Butyl
3-(N-(benzyloxycarbonyl)allylamino)-2-
1
1.56 (s, 9H), 1.52 (d, J = 6.8 Hz, 3H), 2.77 (dd, J = 14.0,
6.2 Hz, 1H), 2.98 (dd, J = 14.0, 5.6 Hz, 1H), 4.65 (q,
J = 6.8 Hz, 1H), 4.90–5.10 (m, 2H), 5.92 (m, 1H), 6.80 (s,
1H), 7.09 (m, 1H), 7.24–7.40 (m, 2H); 13C NMR (75 MHz,
CDCl3) d 20.0, 28.2, 50.1, 52.4, 80.4, 117.9, 125.7, 126.0,
128.2, 130.7, 134.5, 135.9, 137.0, 165.1. LC–MS–ESI rt
23.11 min, 294 (M?1). Anal. cald. for C16H23NO2S
ethylidene-4-methylpentanoate 5a: Yellow oil; H NMR
(400 MHz, CDCl3) d: 0.79 (d, J = 6.8 Hz, 3H), 0.92 (d,
J = 6.8 Hz, 3H), 1.46 (s, 9H), 1.90 (d, J = 7.2 Hz, 3H),
2.68 (m, 1H), 3.94-4.06 (m, 2H), 4.83 (d, J = 9.8 Hz, 1H),
4.90–5.18 (m, 4H), 5.67 (m, 1H), 6.85 (q, J = 7.2 Hz, 1H)
7.20–7.40 (m, 5H); 13C NMR (75 MHz, CDCl3) d 14.5,
19.8, 20.0, 27.8, 28.0, 45.7, 59.7, 67.0, 80.3, 115.0, 127.6,
123

580
A. Tolomelli et al.
128.2, 128.5, 133.6, 135.8, 137.0, 143.0, 156.8, 166.8. LC–
MS–ESI rt 14.7 min 388 (M?1), 410 (M?23). Anal. cald.
for C23H33NO4 (387.24): C 71.29, H 8.58, N 3.61; found C
71.43, H 8.56, N 3.96.
J = 17.2 Hz, 1H), 5.13 (d, J = 10.4 Hz, 1H), 5.15 (d,
J = 10.0 Hz, 1H), 5.70 (m, 1H), 6.18 (q, J = 7.2 Hz, 1H);
13C NMR (CDCl3, 75 MHz) d 14.3, 19.8, 27.8, 29.2, 41.0,
46.9, 52.1, 57.3, 80.2, 115.0, 133.1, 135.0, 142.7, 166.6,
167.1, 168.4. LC–MS–ESI rt 10.1 min 376 (M?23), 729
(2M?23). Chiral HPLC analysis (Chiralpak IC, eluted with
n-hexane/2-propanol, typical program: from 96/4 to 90/10
in 30 min, flux 0.6 mL/min) rt 48.40 min for (R)-E-6a and
59.28 min for (S)-E-6a. Anal. cald. for C19H31NO5
(353.22): C 64.56, H 8.84, N 3.96; found C 64.58, H 8.82,
N 3.98.
E-tert-Butyl
3-(N-(benzyloxycarbonyl)allylamino)-2-
1
ethylidene-4-methylpentanoate 5a: Yellow oil; H NMR
(400 MHz, CDCl3) d: 0.91 (d, J = 6.8 Hz, 6H), 1.44 (s,
9H), 1.90 (d, J = 7.2 Hz, 3H), 2.68 (m, 1H), 3.94–4.06 (m,
2H), 4.62 (d, J = 9.2 Hz, 1H), 4.90–5.18 (m, 4H), 5.68 (m,
1H), 6.01 (q, J = 7.2 Hz, 1H), 7.20–7.40 (m, 5H); 13C
NMR (75 MHz, CDCl3) d 14.1, 19.9, 20.3, 27.8, 29.7,
46.2, 59.9, 67.3, 80.4, 115.3, 127.6, 128.2, 128.5, 133.5,
135.2, 136.6, 142.0, 156.0, 166.8. LC–MS–ESI rt 13.5 min
388 (M?1), 410 (M?23). Anal. cald. for C23H33NO4
(387.24): C 71.29, H 8.58, N 3.61; found C 71.07, H 8.62,
N 3.95.
General procedure for the ring closing metathesis
(RCM) with ruthenium catalyst: Catalyst A–D (0.03 mmol)
was added to a solution of 4a, 5a or 6a (1 mmol) in MTBE
and the reaction mixture was stirred at reflux for 3 h. Then
the reaction cooled to room temperature and was quenched
by addition of ethyl vinyl ether (0.1 mL). The solvent was
removed under reduced pressure and the crude residue was
purified through by chromatography on silica gel (cyclo-
hexane/EtOAc 95:5).
Synthesis of N-malonamido derivative 6a: To a stirred
solution of 2a (1 mmol) in dry CH2Cl2 (5 mL), methyl
malonyl chloride (1.5 mmol, 1.5 equiv.) and triethylamine
(1.5 mmol, 1.5 equiv.) were added at room temperature.
After stirring for 3 h, the reaction mixture diluted with
CH2Cl2 (10 mL) and washed twice with water (10 mL).
The organic layer was dried over Na2SO4 and solvent was
removed under reduced pressure. Purification of the crude
residue by flash chromatography on silica gel (eluant
cyclohexane/EtOAc 9:1) afforded compound 6a. The
desired product was obtained in yields up to 95% as an
yellow oil.
tert-Butyl N-(tert-butoxycarbonyl)2-isopropyl-2,5-dihy-
dro-1H-pyrrole-3-carboxylate 7a: Yellow oil (1:1 trans/cis
conformers mixture); 1H NMR (400 MHz, CDCl3) d Trans
conformer: 0.82(d, J = 6.6 Hz, 3H), 0.94 (d, J = 6.6 Hz,
3H), 1.46 (s, 18H), 2.21 (m, 1H), 3.98 (bd, J = 18.6 Hz,
1H), 4.31 (bd, J = 18.6 Hz, 1H), 4.91 (bm, 1H), 6.76 (bm,
1H); Cis conformer: 0.84 (d, J = 7.0 Hz, 3H), 0.86 (d,
J = 7.0 Hz, 3H), 1.43 (s, 9H), 2.21 (m, 1H), 4.08 (bd,
J = 18.0 Hz, 1H), 4.42 (bd, J = 18.0 Hz, 1H), 4.78 (bm,
1H), 6.63 (bm, 1H); 13C NMR (CDCl3, 100 MHz) d Trans
conformer: 17.0, 19.1, 25.0, 27.4, 31.7, 53.5, 65.3, 78.6,
80.2, 135.3, 136.0, 153.7, 161.9. Cis conformer: 16.6, 19.2,
27.2, 27.9, 32.1, 53.5, 67.2, 79.1, 82.4, 135.7, 136.1, 154.0,
161.1. LC–MS–ESI rt 12.6 min, 334 (M?23). Anal. cald.
for C17H29NO4 (311.21): C 65.57, H 9.39, N 4.50; found C
65.80, H 9.35, N 4.47.
Z-tert-Butyl 3-(N-allyl-3-methoxy-3-oxopropanamido)-
1
2-ethylidene-4-methylpentanoate 6a: Yellow oil; H NMR
(400 MHz, CDCl3) d: 0.81(d, J = 6.6 Hz, 3H), 0.92 (d,
J = 6.6 Hz, 3H), 1.45 (s, 9H), 1.95 (d, J = 7.4 Hz, 3H), 2.67
(m, 1H), 3.38 (d, J = 15.4 Hz, 1H), 3.53 (d, J = 15.4 Hz,
1H), 3.74 (s, 3H), 3.94 (dd, J = 18.2, 5.2 Hz, 1H), 4.44 (dd,
J = 18.2, 2.4 Hz, 1H), 5.04 (d, J = 17.2 Hz, 1H), 5.13 (d,
J = 10.4 Hz, 1H), 5.33 (d, J = 11.6 Hz, 1H), 5.70 (m, 1H),
6.91 (q, J = 7.2 Hz, 1H); 13C NMR (CDCl3, 75 MHz) d
14.6, 19.7, 27.8, 28.0, 41.2, 46.8, 52.1, 57.0, 80.4, 115.7,
132.9, 135.0, 143.9, 166.6, 167.0, 168.4. LC–MS–ESI rt
10.7 min 376 (M?23), 729 (2M?23). Chiral HPLC analysis
(Chiralpak IC, eluted with n-hexane/2-propanol, typical
program: from 96/4 to 90/10 in 30 min, flux 0.6 mL/min) rt
33.24 min for (R)-Z-6a and 36.62 min for (S)-Z-6a; (R)-Z-6a
[a]D = ?26.7 (c 1 in CHCl3). (S)-Z-6a [a]D = -30.8 (c 1 in
CHCl3); Anal. cald. for C19H31NO5 (353.22): C 64.56, H
8.84, N 3.96; found C 64.89, H 8.90, N 3.95.
tert-Butyl N-(benzyloxycarbonyl)2-isopropyl-2,5-dihy-
dro-1H-pyrrole-3-carboxylate 8a: Yellow oil (1.1 trans/cis
conformers mixture); 1H NMR (400 MHz, CDCl3) d Trans
conformer: 0.83 (d, J = 7.2 Hz, 3H), 0.96 (d, J = 7.2 Hz,
3H), 1.49 (s, 9H), 2.53 (m, 1H), 4.10 (bd, J = 18.0 Hz, 1H),
4.39 (bd, J = 18.0 Hz, 1H), 4.91 (bm, 1H), 6.64 (bm, 1H),
7.28–7.41 (m, 5H); Cis conformer: 0.88 (d, J = 6.6 Hz, 3H),
0.89 (d, J = 6.6 Hz, 3H), 1.43 (s, 9H), 2.21 (m, 1H), 4.10
(bd, J = 18.0 Hz, 1H), 4.48 (bd, J = 18.0 Hz, 1H), 4.84
(bm, 1H), 6.71 (bm, 1H), 7.28-7.41 (m, 5H); 13C NMR
(CDCl3, 100 MHz) d Trans conformer: 17.6, 19.0, 28.0,
32.3, 54.6, 66.8, 67.9, 81.2, 127.9, 128.0, 128.1, 128.5, 135.7,
136.8, 154.9, 162.6. Cis conformer: 17.4, 19.2, 29.6, 32.7,
54.0, 67.1, 68.5, 81.2, 127.9, 128.0, 128.1, 128.5, 136.1,
136.6, 155.3, 162.6. LC–MS–ESI rt 12.2 min, 368 (M?23).
Anal. cald. for C20H27NO4 (345.19): C 69.54, H 7.88, N 4.05;
found C 69.33, H 7.92, N 4.06.
E-tert-Butyl 3-(N-allyl-3-methoxy-3-oxopropanamido)-
1
2-ethylidene-4-methylpentanoate 6a: Yellow oil; H NMR
(400 MHz, CDCl3) d 0.94 (d, J = 6.6 Hz, 3H), 0.96 (d,
J = 6.6 Hz, 3H), 1.43 (s, 9H), 1.92 (d, J = 7.4 Hz, 3H),
2.67 (m, 1H), 3.44 (d, J = 15.4 Hz, 1H), 3.54 (d,
J = 15.4 Hz, 1H), 3.76 (s, 3H), 4.02 (dd, J = 18.2, 5.0 Hz,
1H), 4.23 (dd, J = 18.2, 2.0 Hz, 1H), 5.04 (d,
123

Enantiopure 2-substituted-3,4-dehydro-b-proline via ring closing metathesis
581
tert-Butyl N-(methylmalonyl)2-isopropyl-2,5-dihydro-
1H-pyrrole-3-carboxylate 9a: Yellow oil (70:30 trans/cis
conformers mixture); 1H NMR (400 MHz, CDCl3) d Trans
conformer: 0.91(d, J = 7.0 Hz, 3H), 0.92 (d, J = 7.0 Hz,
3H), 1.50 (s, 9H), 2.41 (m, 1H), 3.41 (s, 2H), 3.75 (s, 3H),
4.27 (ddd, J = 16.6, 2.0, 4.4 Hz, 1H), 4.35 (dd, J = 16.6,
2.4 Hz, 1H), 5.12 (bt, J = 3.2 Hz, 1H), 6.64 (bm, 1H); Cis
conformer: 0.92 (d, J = 6.6 Hz, 3H), 0.96 (d, J = 6.6 Hz,
3H), 1.51 (s, 9H), 1.96 (m, 1H), 3.45 (d, J = 15.0 Hz, 1H),
3.51 (d, J = 15.0 Hz, 1H), 3.76 (s, 3H), 4.08 (ddd,
J = 18.8, 1.6, 3.0 Hz, 1H), 4.35 (dd, J = 18.8, 2.8 Hz,
1H), 4.77 (bt, J = 3.2 Hz, 1H), 6.76 (bm, 1H); 13C NMR
(CDCl3, 100 MHz) d Trans conformer:18.3, 18.4, 28.0,
31.8, 42.2, 52.5, 54.6, 68.0, 81.6, 134.1, 136.7, 162.5,
164.3, 167.7. Cis conformer:17.5, 18.7, 28.0, 34.5, 40.8,
52.5, 53.8, 68.4, 81.5, 134.9, 137.3, 162.5, 164.8, 167.5.
LC–MS–ESI rt 8.25 min, 312 (M?1), 334 (M?23), 645
(2M?23). (S)-9a [a]D = ? 34.9 (c 1 in CHCl3); (R)-9a
[a]D = -28.7 (c 1 in CHCl3). Anal. cald. for C16H25NO5
(311.17): C 61.72, H 8.09, N 4.50; found C 61.98, H 8.11,
N 4.52.
CH3OH). Anal. cald. for C8H13NO2 (155.09): C 61.91, H
8.44, N 9.03; found C 62.02, H 8.40, N 9.00.
Results and discussion
Amination of racemic and enantiopure allylic
carbonates
Racemic and enantiopure 1a–f, prepared following an
already reported procedure (Benfatti et al. 2007a, b) were
submitted to allylic amination using allylamine as nucle-
ophile under different conditions (Scheme 1).
The regiochemistry and the stereochemistry of the pro-
cesses has been deeply explored by performing the reaction
without catalyst or in the presence of Pd complexes,
selecting reaction conditions on the basis of our previously
reported work. Moreover, although (p-allyl)-palladium
complexes are the most general and versatile catalysts for
the allylic substitution (Pfaltz and Lautens 1999), a wide
variety of transition metal complexes have been reported to
induce the same reaction (Trost and Hachiya 1998; Malkov
et al. 1999; Evans and Nelson 1998). In particular, the
combination of an iridium precursor and a phosphite or
phosphoramidite ligand has been described as excellent
catalyst for allylic substitution with a number of different
nucleophiles (Takeuchi and Kashio 1997; Bartels et al.
2002). Iridium complexes have been also successfully
applied in the stereoselective formation of C–N bonds via
allylic amination (Roggen and Carreira 2010), mostly on
monosubstituted alkenyl carbonates (Takeuchi et al. 2001;
Leitner et al. 2004; Gnamm et al. 2008). Up to our
knowledge, iridium catalysis has been never applied to
allylic carbonates having alkyl substituents on both termi-
nals and a carboxylic group in the central position, as in
compound 1. The regioselectivity and the stereoselectivity
of the substitution strongly depends on the solvent of
choice and on the phosphorous ligand, that influence the
stability and rate of syn-anti isomerization of the Ir-allyl
intermediate. Ethanol has been reported to favor the for-
mation of the product deriving from SN20 mechanism and,
on these bases, this solvent was used to verify the efficacy
Removal of the N-tert-butoxycarbonyl and tert-butyl
ester protections: TFA (7 mmol, 7 equiv.) was added to a
solution of 7a (1 mmol) in CH2Cl2 (5 mL) and the reaction
mixture was stirred at room temperature until complete
consumption of the starting material. The solvent and the
excess of reagent were removed under reduced pressure.
The crude residue was treated with Dowex 50WX2-200 ion
exchange resin, eluting with NH4OH 0.5 M. Compound 10
was isolated in quantitative yield after removal of the
aqueous solvent.
Removal of the N-benzyloxycarbonyl and tert-butyl ester
protections: TMSCl (3 mmol, 3 equiv.) was added to a
solution of NaI (3 mmol, 3 equiv.) in CH3CN (2 mL) at
room temperature. The mixture was stirred for 10 min and
then, after settling of NaCl powder, the solution was
transferred with lower lock syringe into a flask containing a
solution of 8a in CH3CN (3 mL). The reaction mixture was
refluxed and monitored by TLC until complete consump-
tion of the starting material. The solvent was removed
under reduced pressure and the crude residue was treated
with Dowex 50WX2-200 ion exchange resin, eluting with
NH4OH 0.5 M. Compound 10 was isolated in quantitative
yield after removal of the aqueous solvent. Compound 11
was observed as intermediate in the NMR spectrum of the
crude when the reaction mixture was monitored after stir-
ring at room temperature for 6 h, before refluxing.
(2S)-2-Isopropyl-2,5-dihydro-1H-pyrrole-3-carboxylic
acid 10: 1H NMR (400 MHz, D2O) d 0.79 (d, J = 7.2 Hz,
3H), 0.92 (d, J = 7.2 Hz, 3H), 2.25 (m, 1H), 4.01 (m, 2H),
4.45 (m, 1H), 6.25 (m, 1H); 13C NMR (100 MHz, D2O) d
16.3, 20.1, 31.1, 53.8, 72.8, 136.1, 138.0, 171.4. LC–MS–
ESI rt 1.09 min m/z 156 (M?1). [a]D = ? 26.1 (c 0.7 in
Scheme 1 Allylic amination of carbonates 1a–f
123

582
A. Tolomelli et al.
of [Ir(COD)Cl]2-P(OPh)3 as catalyst for allylic amination.
Some selected results are reported in Table 1.
The reaction was first performed on the Z carbonate,
without any catalyst in refluxing acetonitrile (entries 1–7).
Under these conditions good yields were achieved only
after 3 days, using a large excess of the amine (9 equiva-
lents). The regioisomer 2, deriving from SN20 mechanism,
was exclusively obtained, as 4/1 Z/E isomers mixture, and
no traces of the product 3 could be detected, even in the
case of phenyl and thiophenyl substituted derivatives 1e–
f (entries 6–7). As already observed for allylic substitution
with different amines, the enantiomeric excess of the
starting carbonate was completely retained and enantiopure
2a–b was obtained starting from the corresponding chiral
carbonates 1a–b (entries 2–5). On the basis of our previous
experience, the reaction was then performed in the pres-
ence of 3% amount of Pd2(dba)3CHCl3 in THF. The
reaction on rac-1a afforded rac-2a in modest yield, even in
the presence of 15 equivalents of allylamine (entry 8).
Iridium catalyzed substitution was then performed on
racemic carbonate Z-1a (entry 9). Under these conditions,
the regioselective conversion of the starting carbonate into
amine 2a, isolated as an 87/13 mixture of Z/E isomers, was
observed in 10 h. To verify whether the shorter reaction
time and the enhanced yield had to be ascribed only to the
change in solvent or also to the formation of the iridium-
phosphite complex, the reaction was repeated in refluxing
EtOH without any catalyst (entry 10) and in the presence of
[Ir(COD)Cl]2, without additional phosphine ligand (entry
11). In both cases, a lower yield was observed even after
several hours, thus confirming the catalytic activity of the
complex. The reaction on carbonate 1c afforded the prod-
uct in quantitative yield after shorter reaction times (entry
12). In this case, due to the symmetry of the molecule, the
same product is obtained from SN2 or SN20 mechanism.
Change of regioselectivity depending on reaction condi-
tions had been already observed in allylic amination of
these substrates for palladium-catalyzed reactions (Benfatti
et al. 2008). The preference toward one or the other reg-
ioisomer resulted to be a general phenomenon and similar
results were observed for differently substituted substrates.
To confirm a general behavior also in this case, we per-
formed the Ir-catalyzed substitution also on rac-Z-1d-f.
The results obtained for 1d were in complete agreement
with previous results (entry 13), being 2d obtained as
exclusive regioisomer, as a 5/1 mixture of Z/E isomers.
When the reaction was performed on aromatic and het-
eroaromatic substituted compounds, the regioisomers 2e–
f were still the major ones, but the formation of a consistent
amount of 3e–f was observed (entries 14 and 15). The
different behavior of these substrates is probably due to the
conjugation of the double bond with the aromatic system in
the regioisomers 3e–f.
Fig. 2 Phosphoramidite ligands L1 and L2 used for Ir-complex
formation
Table 2 Iridium-promoted allylic substitution on (S)-Z-1a repeated
in the presence of different phosphorous ligands
Entrya
Carbonate
Ligand
(R)-Z-2a/(S)-E-2a
1
2
3
4
a
(S)-Z-1a
(S)-Z-1a
(S)-Z-1a
(S)-Z-1a
/
86/14
87/13
95/5
P(OPh)3
L1
L2
87/13
Enantiomeric excess of substrates and products was determined by
HPLC on chiral column
The reaction was then repeated on enantiopure (S)-Z-1a.
The regio- and stereoselectivity of allylic substitution,
leading to conservation of the E/Z ratio and enantiomeric
purity, depends on the rate of syn-anti isomerization of
metal-allyl complex. The selection of the proper phos-
phorous ligand may influence the product distribution
(Bartels et al. 2002). According to these observation, the
Z-isomer must have the opposite configuration of the
E-isomer (Pfaltz and Lautens 1999). Therefore, the reaction
was repeated in the presence of different catalysts. The
results obtained with the selected phosphoramidite ligands
(Fig. 2) are reported in Table 2.
As expected, all of the reactions were completely reg-
ioselective, affording exclusively amine 2a. Comparison of
(R)-Z-2a and (S)-E-2a enantiomeric excesses with those of
the starting carbonates, evaluated by HPLC on chiral col-
umn, confirmed that only slight epimerization occurred
(\5%). The result obtained without phosphorous ligand
(entry 1) was quite similar to those obtained with trip-
henylphosphite or phosphoramidite L2 (entries 2 and 4).
The best result was observed for the reaction performed
with ligand L1, that allowed to enhance the Z/E ratio up to
95/5.
Ring closing methatesis to 2-substituted-
dihydropyrroles
The presence of two unsaturations in 2, suggested its
possible transformation into polyfunctionalized dihydro-
pyrrole via ruthenium catalyzed ring closing metathesis
(Grubbs 2004). The metathesis of olefinic amines has been
problematic and limited examples were reported in the
literature using tungsten catalysts (Edwige et al. 1980;
123

Enantiopure 2-substituted-3,4-dehydro-b-proline via ring closing metathesis
583
performed in the presence of acids, able to involve nitrogen
lone pair into protonation (Fu et al. 1993; Wright et al.
2000). On this basis, we first tested the reaction on
unprotected amine 2a in the presence of p-TSA, but we did
not observe any transformation. Thus, 2a was protected as
carbamate or amide, optimizing reaction conditions for
each protecting group (Scheme 2). The N-BOC protected
derivative 4a was obtained in quantitative yield by treat-
ment of 2a with BOC2O in ethanol at room temperature for
3 h. Introduction of benzyloxycarbonyl group was
accomplished by addition of the corresponding chloride to
a solution of 2a and potassium carbonate in water/dioxane
(1/1 ratio) at room temperature overnight, to afford 5a in
80% yield. Moreover, we also synthesized the N-malona-
mide derivative 6a, that we envisaged as a useful inter-
mediate for the preparation of bioactive integrin ligands.
Compound 6a was obtained in 90% yield, by treatment of
2a with methylmalonyl chloride and triethylamine in
dichloromethane at room temperature for 3 h.
Although the RCM has been extensively explored in the
last two decades, it is commonly accepted that a screening
of catalysts should be done to determine which is the best
for each specific transformation. A strong effect on catalyst
efficacy is linked to substrate’s steric hinderance and to the
electronic nature of the olefins. When steric bulk and
electron withdrawing character of the olefin substituents
are combined in the same substrate, the rate of RCM is
strongly reduced and the metal-complex may decompose
before reacting (Kirkland and Grubbs 1997). On these
bases, we tested, for the transformation of 4a, 5a and 6a,
four different catalysts as reported in Scheme 3. The
reaction was performed in the presence of 3% of the cat-
alyst in refluxing methyl tert-butyl ether (MTBE). Selected
results are reported in Table 3.
Scheme 2 Conversion of 2a into N-protected derivatives 4a, 5a, 6a
Levisalles et al. 1984). The first successful use of olefin
metathesis for the formation of N-containing heterocycles
was reported by Grubbs (Fu and Grubbs 1992) and was
then applied to the preparation of various compounds in the
field of medicinal chemistry and material science (Com-
pain 2007). Synthesis of pyrrole derivatives starting from
aza-Baylis–Hillman products has been already reported by
Lamaty’s group on a variety of racemic aromatic-substi-
tuted substrates (Benakki et al. 2008, Declerck et al. 2004,
2007), but to our knowledge no examples of enantiopure
derivatives bearing aliphatic chains have been reported.
The main issue with the RCM of amines is the possible
poisoning of the ruthenium catalyst by coordination of the
nitrogen electron pair to the metal–alkylidene complex. To
avoid this problem, RCM of amines has been successfully
As expected, first generation Grubbs’ catalyst A did not
catalyze the ring closing metathesis on both carbamate
compounds 4a and 5a (entries 1 and 3) allowing to recover
Scheme 3 Screening of
catalysts A–D for the RCM of
compounds 4a, 5a, 6a
123

584
A. Tolomelli et al.
through RCM of the corresponding optically active starting
materials.
Table 3 RCM on compounds 4a, 5a, 6a
Entrya
Starting material
Catalyst
t (h)
Yieldb (%)
It is well known that the amide bond preceding the
proline nitrogen may adopt either the cis or trans confor-
mation (Christoph and Wuthrich 1981). While for all the
other proteinogenic amino acids the trans conformation is
by far energetically more favorable than the cis, for proline
the difference is much smaller. For this reason, the less
stable cis conformer is always present in peptide samples,
in a percentage depending on the adjacent residues and on
the polarity of the solvent (Sugarawa et al. 2001). Two
distinct sets of resonances are commonly detected in the
NMR spectrum of these compounds, as a consequence of
the slow rate of interconversion (Deber et al. 1970). This
effect is enhanced for b-proline oligomers, whose cis and
trans conformers are well populated at room temperature
(Sandvoss and Carlson 2003). Similar considerations have
been reported also for dehydroproline (Kang and Park
2009).
1
2
3
4
5
6
7
8
9
a
4a
4ac
5a
5a
5a
5a
6ac
6a
6a
A
B
A
B
C
D
B
C
D
16
2
/
[98
/
16
1.5
1
[98
[98
[98
54
1
16
12
12
[98
[98
All experiments were performed on E/Z mixtures of starting ally-
lamino derivatives, in the presence of 3% of the catalyst in refluxing
methyl tert-butyl ether (MTBE)
b
Yields were calculated after purification of the products by flash
chromatography on silica gel
c
The reaction was performed on racemic and enantiopure substrate
As expected, the products of the RCM were all isolated
1
as mixtures of two conformers. The H NMR spectrum in
the unreacted starting material without any trace of the
heterocyclic compound, even after long reaction times. On
the basis of these results, catalyst A was not tested on the
less reactive amide 6a. In contrast, when the reaction was
performed in the presence of 3% amount of the more
reactive second generation Grubbs’ catalyst B, both car-
bamates 4a and 5a were converted into the corresponding
dihydropyrroles 7a and 8a respectively, in quantitative
yields (entries 2 and 4). The presence of Z/E mixtures of
isomers did not represent a limitation, since both com-
pounds showed the same reactivity. The reaction was
monitored by TLC and complete disappearance of the
starting material could be observed after only 2 h for 4a
and 1.5 h for 5a. To complete the screening of catalysts,
we performed the RCM on compound 5a using Hoveyda–
Grubbs second generation catalyst C and its analog D,
bearing 2-methylphenyl substituents on the NHC ligand in
place of the 2,4,6-trimethylphenyl groups. Under these
conditions, the reaction occurred slightly faster, since in
both cases compound 8a could be obtained in quantitative
yield after 1 h (entries 5, 6). Comparison of the results
obtained on the carbamate derivatives 4a and 5a with
different catalysts, showed that no great advantages could
be observed in the use of expensive complexes C and D,
since the cheaper catalyst B induced the complete con-
version into the desired products quite rapidly. The RCM
was then performed on malonamide 6a, using the more
efficient catalysts B, C and D. Using Grubbs’s II catalyst B,
the heterocyclic product could be obtained after 16 h only
in 54% yield. Enhanced reactivity could be observed in the
presence of catalysts C and D that quantitatively converted
6a into the corresponding dihydropyrrole 9a in 12 h.
Enantiopure 7a and 9a were obtained with the same yield
CDCl3 of N-BOC protected 7a and N-CBz derivative 8a
showed the presence of the two forms as 1:1 mixtures. For
compound 9a, where dehydro-b-proline ring is linked to
malonamide, the ratio between the conformers was 70:30.
In order to define the conformation of amide bond, the two
distinct sets of resonances were attributed by means of
COSY analysis. Then, spatial arrangement of the malonic
chain for the two conformers was explored through RO-
ESY experiments (Fig. 3). The presence of the cross peak
between malonic chain CH2 and the d-CH2 protons for the
major conformer allowed to attribute the trans conforma-
tion to the amide bond. On the contrary, the cross peak
between malonic chain CH2 and the a-CH proton con-
firmed the cis conformation of the minor isomer.
Removal of protections to obtain dehydro-b-proline
Finally, to simultaneously remove N-BOC protecting group
and t-butyl ester, 7a was treated with an excess of TFA in
dichloromethane at room temperature for 3 days, to afford
2-isopropyl-dehydro-b-proline 10 in quantitative yield
(Scheme 4). Removal of the Cbz protection resulted more
Fig. 3 Conformer attribution via ROESY experiment
123

Enantiopure 2-substituted-3,4-dehydro-b-proline via ring closing metathesis
585
and electron withdrawing olefin substituents could dra-
matically decrease the rate of RCM. To achieve our goal,
different catalysts and conditions were tested and, after
removal of the protecting groups (S)-2-alkyl-dehydro-b-
proline was achieved in pretty good yield.
Acknowledgments This study has been carried out with the fun-
damental contribution of ‘‘Fondazione del Monte di Bologna e Rav-
enna’’. We also thank MAE (Italian Minister for Foreign Affair,
General Direction for the Cultural Promotion and Cooperation) for
financial support to a bilateral projects of between Italy and Mexico.
MIUR (PRIN 2008) and University of Bologna for financial support.
Mr. Andrea Garelli is gratefully acknowledged for the LC–ESI–MS
analysis. We thank C.I.N.M.P.I.S (Consorzio Interuniversitario Naz-
ionale ‘‘Metodologie e Processi Innovativi di Sintesi’’ for a fellowship
to Angelo Viola.
Scheme 4 Removal of protections to afford (S)-2-isopropyl-dehydro-
b-proline 10
difficult, since both hydrogenolysis and treatment with
hydrochloric acid leaded to complete degradation of the
molecule. By treatment of (S)-8a (95% ee) with TMSI,
generated in situ from TMSCl and NaI, in acetonitrile for
6 h at room temperature, N-Cbz-derivative was observed in
the 1H NMR spectrum of the crude reaction, together with
a 25% amount of the free amino acid (S)-10. Further
treatment with the same reagent at 45°C for two days
afforded in almost quantitative yield 2-isopropyl-dehydro-
b-proline 10, that was purified on Dowex 50WX2-200 ion
exchange resin, eluting with 0.5 M NH4OH (Scheme 4).
References
Abele S, Seebach D (1999) Oligomers of b2- and of b3-homoproline:
what are the secondary structures of b-peptides lacking
H-bonds? Helv Chim Acta 82:1539–1558
Abell AD (2002) Heterocyclic-based peptidomimetics. Lett Pept Sci
8:267–272
Bartels B, Garcia-Yebra C, Rominger F, Helmchen G (2002) Iridium-
catalysed allylic substitution: stereochemical aspects and isola-
tion of IrIII complexes related to the catalytic cycle. Eur J Inorg
Chem 2569–2586
`
Benakki H, Colacino E, Andre C, Guenoun F, Martinez J, Lamaty F
(2008) Microwave-assisted multi-step synthesis of novel pyrrol-
o-[3, 2]quinoline derivatives. Tetrahedron 64:5949–5955
Benfatti F, Cardillo G, Fabbroni S, Galzerano P, Gentilucci L, Juris R,
Tolomelli A, Baiula M, Sparta A, Spampinato S (2007a)
Synthesis and biological evaluation of non-peptide a vb 3/a 5b
1 integrin dual antagonists containing 5, 6-dihydropyridin-2-one
scaffolds. Bioorg Med Chem 15:7380–7390
Conclusion
A novel synthetic pathway leading to the preparation of
dehydro-b-amino acids has been developed starting from
enantiopure allylic carbonate 1, obtained from the enzy-
matic resolution of the corresponding alcohol racemate.
The first step of the protocol, the allylic amination, was
performed following two different procedures. Treatment
of 1 without catalyst in refluxing acetonitrile allowed to
obtain regioselectively the allylamino derivative 2. Under
these conditions, the enantiomeric excess of the starting
material was completely maintained, but long reaction
times and the use of a large excess of the amine were
required. The same transformation performed using
[Ir(COD)Cl]2-complexes with phosphorous ligands, affor-
ded a very fast conversion of the carbonate to the allyla-
mino derivative. The ring closure to dihydropyrrole was
performed via Grubbs’ catalyst induced metathesis. This
reaction was already reported in the literature on racemic
aromatic-substituted substrates (Benakki et al. 2008;
Declerck et al. 2004, 2007), but no examples of enantio-
pure derivatives bearing aliphatic chains have been repor-
ted. Moreover, it is well known that the presence of bulky
Benfatti F, Cardillo G, Gentilucci L, Mosconi E, Tolomelli A (2007b)
Enzymatic resolution of ethyl 3-hydroxy-2-(10-substituted-meth-
ylidene) butyrate by Pseudomonas cepacia lipase catalyzed
acetylation. Tetrahedron Asymmetr 18:2227–2232
Benfatti F, Cardillo G, Gentilucci L, Mosconi E, Tolomelli A (2008)
Synthesis of dehydro-b-amino esters via highly regioselective
amination of allylic carbonates. Org Lett 12:2425–2428
Cardillo G, Gentilucci L, Tolomelli A, Calienni M, Qasem AR,
Spampinato S (2003) Stability against enzymatic hydrolysis of
endomorphin-1 analogues containing b-proline. Org Biomol
Chem 1:1498–1502
Cardillo G, Gentilucci L, Mosconi E, Tolomelli A, Troisi S, Juaristi E
(2010) Highly regio- and stereoselective palladium-catalyzed
allylic carbonate amination. A practical route to dehydro-b-
amino esters. Tetrahedron 66:4994–4999
Cheng RP, Gellman SH, DeGrado WF (2001) b-Peptides: from
structure to function. Chem Rev 101:3219–3232
Christoph C, Wuthrich K (1981) NMR studies of the rates of proline
cis-trans isomerization in oligopeptides. Biopolymers 20:2623–
2633
Compain P (2007) Olefin metathesis of amine-containing systems:
beyond the current consensus. Adv Synth Catal 349:1829–1846
Deber CM, Bovey FA, Carver JP, Blout ER (1970) Nuclear magnetic
resonance evidence for cis-peptide bonds in proline oligomers.
J Am Chem Soc 92:6191–6198
123

586
A. Tolomelli et al.
`
Declerck V, Ribiere P, Martinez J, Lamaty F (2004) Sequential aza-
baylis-Hillman/ring closing metathesis/aromatization as a novel
route for the synthesis of substituted pyrroles. J Org Chem
69:8372–8381
McDonald DQ, Still WC (1996) Molecular mechanics parameters and
conformational free energies of proline-containing peptides.
J Org Chem 61:1385–1391. doi:10.1021/jo951788k
Mu¨ller G (2003) Medicinal chemistry of target family-directed master
keys. Drug Disc Today 8:681–691
Pfaltz A, Lautens M (1999) Allylic substitution reactions. In:
Jacobsen et al (eds) Comprehensive asymmetric catalysis.
Springer, Berlin, pp 833–884
Roggen M, Carreira EM (2010) Stereospecific substitution of allylic
alcohols to give optically active primary allylic amines: unique
reactivity of a (P, alkene)Ir complex modulated by iodide. J Am
Chem Soc 132:11917–11919
Declerck V, Allouchi H, Martinez J, Lamaty F (2007) 2-Trimethylsi-
lylethanesulfonyl (SES) versus Tosyl (Ts) protecting group in
the preparation of nitrogen-containing five-membered rings. A
novel route for the synthesis of substituted pyrrolines and
pyrrolidines. J Org Chem 72:1518–1521
Edwige C, Lattes A, Laval JP, Mutin R, Basset JM, Nouguier R
(1980) Influence of basicity and steric effects on the metathesis
of olefinic amines. J Mol Catal 8:297–311
Evans PA, Nelson JD (1998) Conservation of absolute configuration
in the acyclic alkylation reaction: evidence foe an enyl(r?p)or-
ganorhodium intermediate. J Am Chem Soc 120:5581–5582
Fu GC, Grubbs RH (1992) Synthesis of nitrogen heterocycles via
catalytic ring-closing metathesis of dienes. J Am Chem Soc
114:7324–7325
Fu GC, Nguyen ST, Grubbs RH (1993) Catalytic ring-closing
metathesis of functionalized dienes by a ruthenium carbene
complex. J Am Chem Soc 115:9856–9857
Royer J (2009) Asymmetric synthesis of three and four membered
ring heterocycles. Wiley VCH, Weinheim
Sandvoss LM, Carlson HA (2003) Conformational behavior of b-
proline oligomers. J Am Chem Soc 125:15855–15862
Schuffenhauer A, Ertl P, Roggo S, Wetzel S, Koch MA, Waldmann H
(2007) The Scaffold tree—visualization of the scaffold universe
by hierarchical scaffold classification. J Chem Inf Model
47:47–58. doi:10.1021/ci600338x
Sugarawa M, Tonan K, Ikawa S (2001) Effect of the solvent on the
cis–trans conformational equilibrium of a proline imide bond of
short model peptides in solution. Spectr Acta Part A 57:1305–
1316
Takeuchi R, Kashio M (1997) Highly selective allylic alkylation with
a carbon nucleophile at the more substituted allylic terminus
catalyzed by an iridium complex: an efficient method for
constructing quaternary carbon centers. Angew Chem Int Ed
Engl 36:263–265. doi:10.1002/anie.199702631
¨
Gnamm C, Franck G, Miller N, Stork T, Brodner K, Helmchen G
(2008) Enantioselective iridium-catalyzed allylic aminations of
allylic carbonates with functionalized side chains. Asymmetric
total synthesis of (S)-vigabatrin. Synthesis 20:3331–3350
Grubbs RH (2004) Olefin metathesis. Tetrahedron 60:7117–7140
Kang YK, Park HS (2009) Conformational preferences and Cis–Trans
isomerization of L-3,4-dehydroproline residue. Pept Sci
92:387–398
Kirkland TA, Grubbs RH (1997) Effect of olefin substitution on the
ring-closing metathesis of dienes. J Org Chem 62:7310–7318
Leitner A, Shu C, Hartwig JF (2004) Editing the stereochemical
elements in an iridium catalyst for enantioselelctive allylic
amination. Proc Nat Acad Sci 101:5830–5833
Takeuchi R, Naoki U, Tanabe K, Yamashita K, Shiga N (2001)
Iridium complex-catalyzed allylic amination of allylic ester.
J Am Chem Soc 123:9525–9534
Trost BM, Hachiya I (1998) Asymmetric molybdenum-catalyzed
alkylation. J Am Chem Soc 120:1104–1105
Levisalles J, Rudler H, Cuzin D, Rull T (1984) A new catalytic
system for the metathesis of functionalized olefins. J Mol Catal
26:231–238
Wright DL, Schulte JP II, Page MA (2000) An imine addition/ring-
closing metathesis approach to the spirocyclic core of halichl-
orine and pinnaic acid. Org Lett 2:1847–1850
Malkov AV, Baxendale IR, Dvorak D, Mansfield DJ, Kocovsky P
(1999) Molybdenum(II)- and tungsten(II)-catalyzed allylic sub-
stitution. J Org Chem 64:2737–2750
123