Enamino ester reduction: A short enantioselective route to pyrrolizidine and indolizidine alkaloids. Synthesis of (+)-laburnine, (+)-tashiromine, and (-)-isoretronecanol

Article abstract of DOI:10.1021/jo982169p

Various chiral pyrrolidine tetrasubstituted β-enamino esters were reduced catalytically or chemically with good to moderate diastereoselectivity owing to a chiral induction originated from (S)-α- methylbenzylamine. With endocyclic double bond compounds, t

Full text of DOI:10.1021/jo982169p

3122
J . Org. Chem. 1999, 64, 3122-3131
En a m in o Ester Red u ction : A Sh or t En a n tioselective Rou te to
P yr r olizid in e a n d In d olizid in e Alk a loid s. Syn th esis of
(+)-La bu r n in e, (+)-Ta sh ir om in e, a n d (-)-Isor etr on eca n ol
Olivier David, J e´roˆme Blot, Christian Bellec, Marie-Claude Fargeau-Bellassoued,
Gjergj Haviari, J ean-Pierre Ce´le´rier, Ge´rard Lhommet,* J ean-Claude Gramain,† and
Daniel Gardette†
Laboratoire de Chimie des He´te´rocycles, UMR 7611, Universite´ P. et M. Curie (Boˆıte 43), 4 Place J ussieu,
75252 Paris Cedex 05, France
Received October 28, 1998
Various chiral pyrrolidine tetrasubstituted â-enamino esters were reduced catalytically or chemically
with good to moderate diastereoselectivity owing to a chiral induction originated from (S)-R-
methylbenzylamine. With endocyclic double bond compounds, the best result was obtained using
PtO2 as hydrogenation catalyst and led to a major syn addition product (e.d. 90%). In the case of
exocyclic double bond compounds, hydrogenation over Pd/C gave rise to the higher diastereoselec-
tivity and mainly afforded the unexpected anti addition product (e.d. 84%). The scope of these
reductions has been extended to the synthesis of three pyrrolizidine or indolizidine alkaloids: (+)-
tashiromine, (+)-laburnine, and (-)-isoretronecanol. Syntheses of these natural products, starting
from chiral â-enamino diesters, were achieved in a short and convenient manner, leading to
enantiopure compounds in good overall yields.
In tr od u ction
esters directly instead of triflates in the Eschenmoser
coupling reaction.5 The presence of a chiral center on the
â-enamino ester is required to create a facial differen-
ciation during the reduction process, to induce a diaste-
reoselectivity in this pivotal step. This chirality can be
localized either on the ester moiety or at the R-position
of the nitrogen atom. For our part, we retained this last
solution in which the initial source of chirality was
supplied by R-methylbenzylamine (ee 99%) which is
commercially available under both enantiomeric forms.
In short preceding papers, we described the synthesis
with a good diastereoisomeric excess of chiral cyclic
amino esters from tetrasubstituted â-enamino esters with
an endocyclic6 or exocyclic7 double bond. These amino
esters, with an appropriate functionalized carbon chain
(such as a second ester moiety), are potential precursors
of pyrrolizidine and indolizidine alkaloids. Thus, we
described in a preliminary paper an asymmetrical syn-
thesis of (-)-isoretronecanol 1 by diastereoselective re-
duction of enamino diester with an endocyclic double
bond.8 We wish to report in this full account a short and
convenient enantioselective synthesis of two other alka-
loids: (+)-tashiromine 2 and (+)-laburnine 3 in which the
key step is the diastereoselective reduction of a â-enam-
ino diester with an exocyclic ethylenic moiety.
Enamino esters are recognized as very useful inter-
mediates for the synthesis of alkaloids.1 The specific
interest of these compounds resides in their polyfunc-
tionality: they carry simultaneously a nitrogen atom, a
double bond (the reduction of which can create one or
two stereocenters), and an ester moiety which potentially
allows extension of the side chain or ring closure.2 The
â-enamino ester moiety is particularly suitable for the
synthesis of the nitrogen-containing fused bicyclic system
present in indolizidine3 or pyrrolizidine alkaloids, whose
chirality is generally located in the R- and â-position of
the nitrogen atom. In this context, the reduction of the
double bond which generates the chiral center(s) will be
the key step of the sequence in order to obtain the desired
stereochemistry of these alkaloids.
Several synthetic approaches to natural products have
been reported, using trisubstituted enamino esters
intermediates.1a-c,4 However, with these substrates, a
single stereocenter is generated during the reduction.
Aiming to create simultaneously two stereogenic centers
in this reduction step, we have been interested for some
time in the â-enamino ester synthesis with a tetrasub-
stituted endocyclic or exocyclic double bond; in particular,
we developed a simple route using secondary R-halogeno
Resu lts a n d Discu ssion
† Laboratoire de Chimie des Substances Naturelles, UMR 6504,
Universite´ B. Pascal, 63177 Aubie`re Cedex, France.
Syn th esis of Ch ir a l â-En a m in o Ester s. The syn-
thesis of chiral â-enamino esters 6a -e with an exocyclic
(1) (a) Gerrans, G. C.; Howard, A. S.; Orlek, B. S. Tetrahedron Lett.
1975, 4171. (b) Fujimoto, R.; Kishi, Y.; Blount, J . F. J . Am. Chem. Soc.
1980, 102, 7154. (c) Shiosaki, K.; Rapoport, H.; J . Org. Chem. 1985,
50, 1229. (d) Cook, G. R.; Beholz, L. G.; Stille, J . R. Tetrahedron Lett.
1994, 35, 1669. (e) Hernandez, A.; Marcos, M.; Rapoport, H. J . Org.
Chem. 1995, 60, 2683.
(5) Marchand, P.; Fargeau-Bellassoued, M.-C.; Bellec, C.; Lhommet,
G. Synthesis 1994, 1118.
(2) Hart, D. J .; Hong, W.-P.; Hsu, L.-Y. J . Org. Chem. 1987, 52, 4665.
(3) (a) Kiddle, J . J .; Green, D. L. C.; Thompson, C. M. Tetrahedron
1995, 51, 2851. (b) Michael, J . P.; Gravestock, D. Synlett 1996, 981.
(4) (a) Hart, D. J .; Kanai, K. J . Am. Chem. Soc. 1983, 105, 1255. (b)
Hart, D. J .; Hong, W. P. J . Org. Chem. 1985, 50, 3670. (c) Michael, J .
P.; Chang, S.-F.; Wilson, C. Tetrahedron Lett. 1993, 34, 8365. (d)
Munchhof, M. J .; Meyers, A. I. J . Am. Chem. Soc. 1995, 117, 5399.
(6) Haviari, G.; Ce´le´rier, J .-P.; Petit, H.; Lhommet, G.; Gardette,
D.; Gramain, J .-C. Tetrahedron Lett. 1992, 33, 4311.
(7) Blot, J .; Bardou, A.; Bellec, C.; Fargeau-Bellassoued, M.-C.;
Ce´le´rier, J .-P.; Lhommet, G.; Gardette, D.; Gramain, J .-C. Tetrahedron
Lett. 1997, 38, 8511.
(8) Haviari, G.; Ce´le´rier, J .-P.; Petit, H.; Lhommet, G. Tetrahedron
Lett. 1993, 34, 1599.
10.1021/jo982169p CCC: $18.00 © 1999 American Chemical Society
Published on Web 04/02/1999

Pyrrolizidine and Indolizidine Alkaloids
J . Org. Chem., Vol. 64, No. 9, 1999 3123
Sch em e 1
double bond was carried out, transposing the modified
Eschenmoser coupling reaction previously described for
N-benzyl substrates.5 Thus a mixture of triethylamine
and triphenylphosphine was slowly added to a solution
of 1-(1(S)-phenylethyl)pyrrolidine-2-thione9 4 and com-
mercially available secondary R-bromo esters 5a -e in
refluxing acetonitrile. Compounds 6a -e were obtained
in good yields as a mixture of E/ Z isomers (Scheme 1)
in which the E-isomer was the major one. E/ Z ratios
were determined by 1H NMR measurements. For 6e, only
the E isomer was detected.
Sch em e 2
Enamino ester 7, with an endocyclic double bond, was
obtained in 74% yield by condensation of (S)-R-methyl-
benzylamine with methyl 2-acetylcyclopropanecarboxy-
late in refluxing toluene, following a procedure previously
described with various achiral amines (Scheme 2).10
Red u ction of Ch ir a l â-En a m in o Ester s. Due to the
decisive nature of this step, a systematic study of endo
and exocyclic â-enamino esters reduction was carried out
catalytically or chemically in order to investigate the
influence of the reducing agent on the diastereoselectivity
of this step. Catalytic reductions were performed under
atmospheric pressure in ethanol or methanol, using PtO2,
10% Pt/C, and 10% Pd/C as catalysts, whereas chemical
reductions were realized with triacetoxyborohydride in
acetic acid.10
Sch em e 3
For the enamino ester 7 with an endocyclic double
bond, chemical reduction was previously reported by
Palmieri et al.;11 therefore only catalytic hydrogenation
was considered with this compound. With 0.04 equiv of
Pd/C, the hydrogenation proceeded very slowly and the
debenzylation rate was faster than the reduction rate,
making this catalyst inappropriate. Reduction in the
presence of Pt/C yielded mainly two diastereoisomers 8
and 9 in 5.8/1 ratio. X-ray analysis performed on the
major diastereomer permitted the assignment of 2S,3S
absolute configurations to the stereocenters created,
corresponding to a cis stereochemistry for compound 8.
A debenzylation reaction on the diastereomeric mixture
effected with 0.6 equiv of Pd/C allowed us to assign the
same cis geometry to the second major diastereomer 9
whose configurations are 2R,3R (Scheme 3). A very small
percentage of one of the trans isomers (<5%) was
detected by gas chromatography. With PtO2 as catalyst,
the percentages of 8 and 9 were respectively 89% and
8% while about 3% of the trans isomer was detected.
These results show that a high selectivity is observed in
catalytic reduction of 7; the major diastereomer obtained
is that corresponding to the expected syn addition on the
less overcrowded face (Re face at C2 and C3) on the
enamino ester moiety and is consistent with theoretical
calculations.6
inseparable by chromatographic methods. In the case of
Pd/C, the catalyst with which the parasite debenzylation
reaction can be envisaged, we adapted a catalyst amount
to 0.04 equiv to avoid this competitive route. However,
in these conditions, 6e was not reduced and the increase
of catalyst amount only gave rise to debenzylation.
In all cases, amino esters were obtained in good
chemical yields (80-98%) as a mixture of two, three, or
four diastereoisomers noted 10, 11, 12, and 13, according
to their elution order in gas chromatography. Diastereo-
isomer ratios of hydrogenated compounds according to
the reducing agents were determined both by CPV
analysis and by 13C NMR measurements and were
reported in Table 1.
In most cases, the main diastereoisomer formed was
12 whereas 11 was always present in a very small
proportion or not detected. However for Pt/C reduction
of 6e and for PtO2 reduction of 6a , 6c, and 6e, the major
diastereomer was 13. With the exception of the hydro-
genation with this last catalyst whose results did not
show any regularity depending on the side chain size,
the best diastereoselectivity was observed for 6a reduc-
tion independent of the nature of the reducing agent.
Besides, the diastereoselectivity decreases when ethyl
replaces the methyl group and then does not significantly
vary with more chain extension. For all compounds with
an alkyl substituent, the highest diastereoselectivity was
obtained with 10% Pd/C (e.d. 68-84%); with 10% Pt/C
as catalyst, the diastereomeric excess varies between 40
and 60% and with NaBH(OAc)3/AcOH from 22 to 50%.
It was essential to determine absolute configuration
of the two chiral centers created for hydrogenated
For compounds with an exocyclic double bond, 6a -e,
all reductions were achieved on the E/ Z isomer mixture,
(9) Nikiforov, T.; Stanchev, S.; Milenkov, B.; Dimitrov, V. Hetero-
cycles 1986, 24, 1825.
(10) Ce´le´rier, J .-P.; Haddad, M.; J acoby, D.; Lhommet, G. Tetrahe-
dron Lett. 1987, 28, 6597.
(11) Cimarelli, C.; Palmieri, G. J . Org. Chem. 1996, 61, 5557.

3124 J . Org. Chem., Vol. 64, No. 9, 1999
David et al.
Ta ble 1. Red u ction of En a m in o Ester s 6a -e to Am in o Ester s 10-13
H2 PtO2 H2 Pt/C H2 Pd/C
11 12 11 12
NaBH(OAc)3/AcOH
redn
10
13
10
11
12
13
10
13
10
11
12
13
6a , R ) Me
6b, R ) Et
6c, R) Pr
6d , R ) Bu
6e, R ) Ph
19
6
9
37
85
35
49
3
44
9
56
38
97
10
12
10
11
1
80
77
71
70
40
9
11
14
14
60
5
10
4
3
92
84
89
90
7
14
13
9
3
73
64
61
67
63
17
22
26
24
31
6
7
6
5
5
13
4
5
Sch em e 4
Sch em e 5
Sch em e 6
compounds in the diastereoisomeric mixture. The stereo-
chemical assignments for compounds 12a and 13a (R )
Me) were proven from X-ray analysis.7,12 Diffraction
measurements carried out on 12a as its picrate salt
showed the 2S,2′R configuration while 13a had a 2R,2′R
configuration. By comparing 13C NMR chemical shifts of
some significant carbon atoms of both major diastereo-
isomers 12a and 13a with those of the series b-d (R )
Et, n-Pr, n-Bu), the same absolute configurations can be
respectively assigned7 to all compounds 12b -d and
13b-d (Scheme 4). For diastereomers resulting from the
reduction of 6e the correlation was more critical due to
the perturbation induced by the phenyl ring. The at-
tribution proposed in Table 1 was confirmed by X-ray
measurements on the picrate salt of the major diastere-
omer 13e, obtained in the PtO2 reduction, whose con-
figurations were assigned to be 2R,2′R.
In such catalytic reductions, a stereoselective syn
hydrogenation process is generally expected. However,
for 6a -d , only the minor 6Z stereoisomer would lead to
12a -d which is the major reduced product in the
mixture. Hence, it is surprising that starting from an
enamino esters E/Z mixture, both present in notable
proportion, a main diastereoisomer is yet obtained up to
92%. Such a result (presence of both syn and anti
addition products) is often explained by a double bond
migration during the reducing course.13 But in our case,
the too important conformational liberty in the endo- or
exocyclic intermediate resulting from a hydrogen addi-
tion-elimination sequence cannot simply explain the
good diastereoselectivity observed. A more attractive
hypothesis has sometimes been advanced to explain such
results with diesters substrates.14 It involved the forma-
tion of an enolate intermediate by addition of a MH2
equivalent on the enamino ester double bond. The chiral
amino group would favor the addition on the front face7
(Si face at C2′) leading to an R configuration for this
center. The more favored transition state structure
represented in Scheme 5 was based on the conformation
in which 1,3-allylic strain was minimized and the C-N
bond became oriented perpendicular to the π system.
Stereoselective protonation on the rear face (Re face at
C2) would lead to the S configuration for this carbon. It
should be remarked that the hydrogen atom used for C2
protonation may arise either from the catalyst surface
or protic solvent.
Such a model could satisfactorily explain the good
diastereoselectivity observed for compounds 12a -d when
hydrogenation is carried out in the presence of palladium
or platinum. This model could also explain the production
of both syn and anti addition compounds in the reduction
of only the E stereomer of 6e. Contrary to compounds
6a -d , it should be remarked that the expected syn
addition product was the main stereomer obtained with
catalytic reductions of this compound. In the case of R )
Ph, an important steric strain between aromatic and
pyrrolidine rings disfavors the conformation shown in
Scheme 5 whereas a π-π interaction between the two
phenyl groups promotes the conformation represented in
Scheme 6, in which however remains a 1,3-allylic strain.
Favored protonation on the upper face (Si face at C2)
principally leads to the R configuration at this center and
yields the main 2R,2′R diastereoisomer 13e.
Molecu la r Mod elin g Stu d y. The catalytic hydroge-
nation of the double bond of the studied amino esters
must arise from the less crowded face of the double bond.
Therefore we undertook a molecular modeling study to
determine the most stable conformers of compounds 6a
(E) and 7 (absolute configuration S) and their geometrical
conformations in order to evalue the relative steric
(12) Bardou, A.; Ce´le´rier, J .-P.; Lhommet, G. Tetrahedron Lett. 1997,
38, 8507.
(13) (a) House, H. O. Modern SyntheticReactions; W. A. Benjamin
Inc., 1972; pp 20-22. (b) Hudlicky, M. Reductions in Organic Chem-
istry; J ohn Wiley: New York, 1984;
p 40. (c) Rylander, P. N.
Hydrogenation Methods; Academic Press: New York, 1985; pp 29-
52.
(14) Siegel, S. Comprehensive Organic Synthesis; Pergamon Press:
1991; Vol. 8, p 427.

Pyrrolizidine and Indolizidine Alkaloids
J . Org. Chem., Vol. 64, No. 9, 1999 3125
F igu r e 1. Curves E ) f(Θ)1) for compounds 6a , 6g, and 6f (E) a n d (Z).
crowding for each face of the double bond. These calcula-
tions were performed using the SYBYL 6.3 software
package15 as described in the Experimental Section. All
the generated geometries were minimized using the
Tripos force field and optimized using AM116 calculations.
The conformational spaces of 6a and 7 were explored
using the SYBYL search facility.
CH3 group. The conformational spaces of each compound
were explored using SYBYL 6.3 as described above. The
two bonds C*-N (θ1) and C*-Ph (θ2) were allowed to
rotate respectively from 0 to 360° and 0 to 180° by 15°
increments. The obtained results were analyzed by
graphical representation in three dimensions (E, θ1, θ2).
Their two-dimensional projections (E, θ1) were then
examined.
In the case of compound 7 with an endocyclic double
bond, rotation around the N-C* (θ1) and the C*-Ph (θ2)
bonds gave a three-dimensional (E, θ1, θ2) graph. Its two-
dimensional projection on the E, θ1 plane gave a curve
with a single wide minimum with a width of 120°
between -50° and -170°. The bottom of the curve was
rather flat with a minimum between -80° and -120°
(depth of 8.4 kJ mol-1) indicating the presence of only
one conformer17 in which the rear face of the unsaturated
ring is severly crowded by the phenyl ring. This explains
the very high stereospecificity of the hydrogenation
reaction (de > 90%) which proceeds by the less crowded
face of the molecule, leading to the 2R,3R stereochemistry
(see scheme in ref 6). The presence of only one conformer
is clearly due to the steric interaction between the C2-
CH3 and the C* substituents. This interaction is mini-
mized when the smaller group (C*-H) is in front of the
bulky C2-CH3 group (“gear effect”).
For compound 6f(E) the curve E ) f(θ1) has three
marked minima (Figure 1) at nearly 120° from each other
as in CRAM models. On the other hand, in the case of
compound 6f(Z) the curve E ) f(θ1) shows a single deep
minimum corresponding to a unique conformer in which
the rear face of the molecule is overcrowded by the phenyl
group. This calculation demonstrates the importance of
(15) Tripos Associates, Inc., 1699 S. Harley Rd., Suite 303, St. Louis,
MO 63144.
(16) Merz, K. M.; Besler, B. H. MOPAC version 5.0 ESC, QCPE 589,
1990, with AM1 force field parameters.
For compounds with an exocyclic double bond, we
studied first the simplified (and no synthesized) models
6f(E) and 6f(Z) in order to show the influence of the C2′-

3126 J . Org. Chem., Vol. 64, No. 9, 1999
David et al.
Sch em e 7
F igu r e 2.
the “gear effect” between the C2′-CH3 and the substituents
of the C* and allows the prediction of a high diastereo-
selectivity at the C2′ for the hydrogenation of the double
bond and a R configuration of the C2′ stereogenic center
of the reaction product.
The curves corresponding to functionalized compounds
6g and 6a are rather similar to those of 6f(E) and 6f(Z)
showing also the importance of the C2-CH3 group and
allowing the prediction of a higher stereoselectivity of a
catalytic hydrogenation reaction for 6a than for 6g. In
fact catalytic hydrogenation of 6g was already reported9
and gave a mixture of two diastereoisomers at C2′ in a
87/13 ratio (de ) 74%). As expected, the stereoselectivity
of the catalytic hydrogenation of 6a is higher: the ratio
of diastereoisomers at C2′ is 98/2 and the de 84% and can
reach 92% (ratio 96/4) (Table 1) with propyl (6c) or butyl
groups (6d ) instead of methyl.
This study shows the interest in molecular modeling
as a predicting tool for the stereochemistry of the
catalytic hydrogenation reaction. A conformational analy-
sis of the enol 6e′, obtained after the first step of the
hydrogenation process, was performed using SYBYL 6.3
in order to predict the stereochemistry of the hydrogena-
tion product of 6e (R ) Ph). The optimized structure
obtained as previously described was subjected to a
restrained molecular dynamic simulation lasting 10 ps
at a temperature of 1000 K. Structures were selected
after analysis of dynamic trajectory and the lowest energy
conformers minimized in vacuo by Maximin 2 Tripos
force field. Only one conformer was obtained (Figure 2)
showing a π-π interaction between the two phenyl
groups which promotes the conformation represented in
Scheme 6 and thus explaining the stereochemistry at the
C2 stereogenic center of the reaction product.
Ap p lica tion to th e Syn th esis of (+)-Ta sh ir om in e
2 a n d (+)-La bu r n in e 3. This efficient exocyclic enamino
ester reduction was applied to alkaloid synthesis. We
exploit the main point of the hydrogenation course which
is the high level of diastereoselectivity observed despite
starting from an unfavorable E/ Z mixture of 6. The scope
of this key step can be extended to the synthesis of two
natural products, (+)-tashiromine and (+)-laburnine, as
it will lead to the correct relative stereochemistry at the
two stereogenic centers in a very simple manner. To form
the second ring present in these indolizidine and pyr-
rolizidine skeletons, we chose to start directly from
â-enamino esters with appropriate carbon chains.
Tashiromine was isolated in 1990 from the stems of a
leguminous plant Maackia tashiroi, a deciduous shrub
of subtropical Asia.18 Only three syntheses have been
reported todate, a racemic one19 and two enantioselective
approaches: a short route described by Nagao and co-
workers20 and a recent synthesis in 13 steps from
L-glutamic acid by B. P. Branchaud et al.21 Absolute
configurations of both chiral centers are unknown; only
a trans relationship between H8 and H9 was ascribed.18
Consequently tashiromine was determined to be (5S,6R)-
5-hydroxymethylindolizidine or its enantiomer. As for
laburnine, this pyrrolizidine alkaloid was first isolated22
from Cytisus laburnum in 1949 by F. Galinovsky and
then, among others, from the leaves of Planchonella
anteridifera, a tree found in the rain forest area of New
Guinea.23 Only two enantioselective syntheses are re-
ported at this time.24
The first step of our synthetic route (Scheme 7) is the
access to chiral â-enamino diesters readily obtained
through modified Eschenmoser sulfide reaction. There-
fore we have transposed to R-bromo diesters with desired
side chain size, the conditions described above for second-
ary R-bromo esters. Thus 1-(1(S)-phenylethyl)pyrrolidine-
2-thione9 4 was condensed with dimethyl bromobutane-
dioate25 14 and dimethyl bromopentanedioate26 15 to give
respectively chiral â-enamino diesters 16 and 17 in 84%
and 85% yields. Enamino diesters 16 and 17 were
(18) Ohmiya, S.; Kubo, H.; Otomasu, H.; Saito, K.; Murakoshi, I.
Heterocycles 1990, 30, 537.
(19) Paulvannan, K.; Stille, J . R. J . Org. Chem. 1994, 59, 1613.
(20) Nagao, Y.; Dai, W.-M.; Ochiai, M.; Tsukagoshi, S.; Fujita, E. J .
Org. Chem.1990, 55, 1148.
(21) Gage, J . L.; Branchaud, B. P. Tetrahedron Lett. 1997, 38, 7007.
(22) Galinovsky, F.; Goldberger, H.; Po¨hm, M. Monatsh.Chem. 1949,
80, 550.
(23) Hart, N. K.; Lamberton, J . A. Aust. J . Chem. 1966, 19, 1259.
(24) (a) Robins, D..J .; Sakdarat, S. J . Chem. Soc., Perkin Trans. 1
1981, 909. (b) Arai, Y.; Kontani, T.; Koizumi, T. J . Chem. Soc., Perkin.
Trans. 1 1994, 15.
(17) The curve shows another minimum at θ ) 140° for an energy
of nearly 28 kJ mol-1. The population of the corresponding conformer
can be neglected due to its high energy.

Pyrrolizidine and Indolizidine Alkaloids
J . Org. Chem., Vol. 64, No. 9, 1999 3127
Ta ble 2. Red u ction of En a m in o Diester 17 to Am in o
Ta ble 3. Assign m en t of m in or d ia ster eom er s 18 a n d 19
Diester s 18-21
H2 Pd/C
H2 PtO2
H2 Pt/C
NaBH(OAc)3 AcOH
18
19
20
21
2
2
96
8
2
54
36
6
1
82
11
8
3
68
21
obtained as an inseparable diastereomeric mixture of
E/ Z isomers in respective 75/25 and 65/35 ratios, deter-
1
mined by H NMR measurements.
For our studies with model enamino esters 6 showing
that catalytic hydrogenation with 10% Pd/C gave rise to
the best diastereoselectivity, we first naturally used this
catalyst in the case of enamino diesters 16 and 17.
However, that reductions performed on 17 with PtO2, Pt/
C, or NaBH(OAc)3 /AcOH were less selective than with
Pd/C was tested. For 17 every reduction produces a
diastereomeric mixture of three or four hydrogenated
compounds 18, 19, 20, and 21, the percentages of which
are reported in Table 2.
amino diesters
lactam esters
confign %
%
8
Ca
C2′
S
C2
18
S
R
SR + RS
SS + RR
98
2
20
19
90
2
S
S
R
S
S
S
21
0
S
R
R
Since Pd/C can induce an undesirable debenzylation
reaction, we modified the catalyst amounts in order to
obtain a significant reaction rate without appreciable
debenzylation. When 10% Pd/C amounts decreased from
0.2 to 0.04 equiv, no debenzylation reaction was observed
and simultaneously the fraction of the major diastereoi-
somer 20 in the mixture surprisingly increased from 58%
to 96%. For enamino ester 16, reduction and debenzyla-
tion rates are very close and it is impossible to avoid a
debenzylation reaction even with Pd/C amounts lower
than 0.04 equiv. So we used in this case 10% Pt/C as
catalyst. Unfortunately the percentage of major diaste-
reomer 24 in the mixture 22-25 was only 77% but no
debenzylation was observed. Stereomers 22, 23, and 25
were present as 8%, 2%, and 13% in the mixture,
respectively. To separate main diastereoisomers 20 and
24, each diastereomeric mixture was transformed to
picrate salts, and after several recrystallizations, dia-
stereomers 20 and 24 were respectively obtained in 98%
and 96% de.
2S,2′R configurations can be assigned to stereoisomers
20 and 24 while 21 and 25 have 2R,2′R configurations
at both created centers. Once again, these assignments
were based on comparison of their chemical shifts in 13C
NMR with those of compounds of the series 12a -d and
13a -d .7 Thus, chemical shifts of Ca, Cb, and Cc were
respectively 57.0, 46.6, and 23.2 ppm for 20, 59.2, 48.1,
and 24.1 for 21, 56.8, 46.3, and 23.1 for 24, and 58.7, 48.3,
and 23.7 for 25. X-ray analysis performed on the picrate
salt of 20 confirms the absolute configurations proposed
by NMR assignments. In the case of 24, the stereochem-
ical assignments will be further proved by the conversion
of this stereomer into laburnine whose specific rotation
is known. An attribution of absolute configurations of
both minor stereomers 18 and 19 will be proposed below
using results of the amino diester cyclization step (see
Table 3).
of 20 and 24 was achieved in MeOH under atmospheric
pressure of H2 with a higher amount of catalyst (0.6
equiv) than those used in the reduction step. Thus, chiral
lactams 26 and 27 were respectively obtained in quan-
titative yields as a mixture of two diastereoisomers in
the same ratios (99/1 and 98/2) as those of the â-amino
diesters they are produced from. The minor diastereomer
was separated by chromatography, and compounds 26
and 27 were then obtained in a pure enantiomeric form
respectively in 78% and 75% yields.
The investigation of the debenzylation step allows us
to suggest absolute stereochemistry for minor amino
diester diastereomers 18 and 19 issued from the reduc-
tion of 17, and 22 and 23 resulting from those of 16. The
stereochemistry assignments were proved by comparison
of chromatograms, for example of the amino diester
mixture 18, 19, 20, and 21 with those of debenzylated
diastereomers in the next step. Thus, debenzylation of a
mixture, 18 (8%), 19 (2%), 20 (90%), 21 (0%), yields two
diastereomeric lactam esters in a 98/2 ratio (Table 3).
Since 20 has the 2′R,2S configuration, 21 is 2′R,2R, and
18 and 20 on the one hand and 19 and 21 on the other
hand lead after cleavage of the chiral auxiliary to two
pairs of enantiomers, 18 has a 2′S,2R configuration and
19 is 2′S,2S.
The synthesis was completed in a final step with the
reduction of both the lactam carbonyl group and ester
moieties of 26 and 27. These reductions were simulta-
neously achevied using LiAlH4 in THF at room temper-
ature as no epimerization was observed in these condi-
tions. Thus, (+)-tashiromine 2 and (+)-laburnine 3 were
1
isolated in 87% and 85% yields, respectively. H and 13C
NMR spectra were in complete agreement with those
previously described.19,20,24
The optical rotation measured for (+)-tashiromine was
+44.7 (c 1.1, EtOH) but no value for the natural product
was reported; hence no comparison with our result was
possible. It should be remarked that Y. Nagao et al.20
The stereocenters are now well established with the
desired configuration. Having performed its function of
chiral control, the R-methylbenzyl group would be re-
moved, enabling the formation of the second ring present
in the target alkaloids. For this purpose, in a third step,
the hydrogenolysis of the benzylic amine was performed
in the presence of Pd/C and the subsequent ring closure
gave rise to the bicyclic lactam esters. The debenzylation
mentioned [R]20 ) -25.9 (c 1.16, EtOH) for the (-)
D
enantiomer rotation, while B. P. Branchaud21 obtained
[R]23D ) -36.3 (c 0.88, CHCl3) for a product with 92% ee.
For (+)-laburnine the measured optical rotation [R]20
)
D
+15.4 (c 1.20, EtOH) was in total concordance with that
of the natural product: [R]20 ) +15.5 (EtOH) in Cytisus
D

3128 J . Org. Chem., Vol. 64, No. 9, 1999
David et al.
Sch em e 8
The first step of our approach (Scheme 8) is the
synthesis of dihydropyrrole 29 adapting the procedure
described previously for 7 and starting from a properly
functionalized cyclopropane 28. Thus, compound 288
reacted with (S)-R-methylbenzylamine in refluxing tolu-
ene to afford â-enamino ester 29 with 83% yield. Dihy-
dropyrrole catalytic hydrogenation over PtO2 gave the cis
diastereoisomers 30 and 31 with a diastereoisomeric
excess of 90%, measured by chromatography (87% chemi-
cal yield). The absolute configurations of the two created
stereocenters have been established in the last step by
comparison of the specific rotation of our synthetic
sample with those of the natural product. Hydrogenolysis
of the R-methylbenzyl group allowed the construction of
the second ring. The mixture of 30 and 31 hydrogenated
over 10% Pd/C afforded, after heating in toluene at 90
°C, bicyclic lactam esters 32 and 33 in 90% yield.
The lactam moiety was transformed into thiolactam
with Lawesson reagent, and optically pure compound 34
was obtained, after recrystallization, in 46% yield. Selec-
tive reduction of the thiolactam moiety was realized by
its conversion with MeI in CH2Cl2 into the corresponding
thioiminium salt which was reduced with NaBH4 in
MeOH.30 This sequence supplied the intermediate (-)-
chysine 35 in 77% yield. The optical rotation measured
for this compound [[R]20D ) -65 (c 2.16, CHCl3)] is in good
agreement with that of its enantiomer, the natural (+)-
chysine ([R]25 ) +64 (c 1.1, CHCl3).29 Finally, ester
D
functional group of 35 was reduced with LiAlH4 in THF
at -80 °C. In fact, the reduction must be carried out very
smoothly at low temperature to avoid epimerization at
C8. (-)-Isoretronecanol 1 was then isolated in an enan-
tiomerically pure form in 83% yield after distillation. The
optical rotation measured [R]20 ) -77.0 (c 2.1, EtOH)
D
is in good agreement with those of the natural product
[R]20D ) -77.5 (c 2.5, MeOH).29 This pyrrolizidine alkaloid
1 was then obtained in seven steps from activated
cyclopropane 28 in 20% overall yield.
Con clu sion
laburnum22 and +15.4 (c 1.44, EtOH) in Planchonella
anteridifera.23
In conclusion, we have shown that the use of R-meth-
ylbenzylamine as a control element provides in both cases
(enamino esters with endo- or exocyclic double bonds) a
highly diastereoselective catalytic reduction of the eth-
ylenic moiety. The stereochemistry of hydrogenated
compounds depends strongly on the position of the double
bond with regard to pyrrolidine ring. Endocyclic double
bond enamino esters undergo expected syn hydrogenation
while, surprisingly, with exocyclic double bond com-
pounds, the main diastereoisomer formed formally re-
sults from an anti-addition of hydrogen on the major E
enamino ester. We applied this stereodirecting reduction
step to the enantioselective synthesis of some alkaloids
which was performed in a very short and convenient
route. This strategy enables access to 1-hydroxymeth-
ylpyrrolizidine or indolizidine in good yields in optically
active form.
Therefore (+)-tashiromine and (+)-laburnine were
synthesized in pure enantiomeric form, in four steps from
chiral thiolactam 4 in 25% and 27% overall yields,
respectively.
Ap p lica tion to th e Syn th esis of (-)-Isor etr on -
eca n ol 1. In the same manner, the good diastereoselec-
tivity observed for the cis hydrogenation of endocyclic
enamino ester 7 has been exploited for the pyrrolizidine
alkaloid synthesis, the (-)-isoretronecanol.
Isoretronecanol 1 was first isolated from a Plan-
chonella species27 and then from Phalaenopsis equestris,
an orchidaceae from Philippines.28 Many syntheses, both
racemic and enantioselective, were reported in the lit-
erature.29 This alkaloid, with a cis relationship between
H1 and H8, is a diastereoisomer of laburnine 3 in which
these two protons have a trans relative stereochemistry.
Exp er im en ta l Section
(25) Anschu¨tz, R.; Bennert, C. Liebigs Ann. Chem. 1889, 254, 161.
(26) Hoye, T. R.; Caruso, A. J .; Magee, A. S. J . Org. Chem. 1982,
47, 4152.
(27) Hart, N. K.; J ohns, S. R.; Lamberton, J . A. Aust. J . Chem. 1968,
21, 1393.
(28) Branda¨nge, S.; Lu¨ning, B.; Moberg, C.; Sjo¨strand, E. Acta Chem.
Scand. 1972, 26, 2558.
Gen er a l Meth od s. Melting points are uncorrected. Capil-
lary gas chromatography was performed on a chromatograph
with a flame ionization detector using CP-SIL 5 (Chromapack)
columns. Column chromatography was performed on silica gel
(29) Southon, I. W.; Bukingham, J . Dictionary of Alkaloids; Chap-
man and Hall Ltd.: London, 1989, pp 543 and references therein.
(30) Raucher, S.; Klein, P. Tetrahedron Lett. 1980, 21, 4061.

Pyrrolizidine and Indolizidine Alkaloids
J . Org. Chem., Vol. 64, No. 9, 1999 3129
60 (Merck 230-400 mesh), and reactions were monitored by
TLC on Kieselgel 60 F254 (Merck); detection was effected by
examination under UV light. 1H and 13C NMR spectra were
recorded at 250 and 62.5 MHz, respectively, in CDCl3 refer-
enced to Me4Si for the proton spectra and the solvent for the
carbon spectra. The microanalyses were performed by the
Service de Microanalyse, Universite´ Pierre et Marie Curie. IR
spectra were recorded as thin films on NaCl plate. Dichlo-
romethane was distilled from P2O5. Diethyl ether was distilled
from CaH2 and redistilled from LiAlH4. Tetrahydrofuran was
freshly distilled from sodium benzophenone ketyl. Methanol
and ethanol were dried by refluxing with magnesium meth-
oxide or ethoxide and distilled. (S)-R-Methylbenzylamine was
purchased from Aldrich (99% ee).
Gen er a l P r oced u r e for Esch en m oser Cou p lin g Rea c-
tion . A mixture of 1-[1(S)-phenylethyl]pyrrolidine-2-thione (4)
(5.06 g, 24 mmol), bromo esters 5a -e, 14, or 15 (72 mmol),
and sodium iodide (1.82 g, 12 mmol) was refluxed in MeCN
(20 mL). A solution of PPh3 (12.63 g, 48 mmol) and triethy-
lamine (6.7 mL, 48 mmol) in MeCN (100 mL) was added
dropwise during 1-2 h. After about 6 to 7 h of stirring under
reflux, the reaction mixture was concentrated under reduced
pressure, and the residue was diluted with CH2Cl2 (15 mL)
and then extracted with 1 M HCl (8 × 70 mL). The aqueous
layers were combined, neutralized with Na2CO3, and then
extracted with AcOEt (8 × 70 mL). The combined organic
phases were dried (Na2SO4), and the solvent was removed. The
crude product was purified by chromatography on a silica gel
column (AcOEt/cyclohexane 1/3).
2-[1-(1(S)-P h en yleth yl)p yr r olid in -2-ylid en e]p r op a n o-
ic Acid Eth yl Ester (6a ). Condensation of ethyl 2-bromopro-
panoate (5a ) with 4 gave 6a as a mixture of 90/10 E/ Z
diastereomers: yield 89%; oil; 1H NMR 7.35-7.15 (m, 5H), 5.50
(q, 0.1H, J ) 6.9 Hz), 5.40 (q, 0.9H, J ) 6.9 Hz), 4.10 (q, 2H,
J ) 6 Hz), 3.25-3.10 (m, 1H), 3.05-2.95 (m, 2H), 2.90-2.80
(m, 1H), 2.00 (s, 2.7H), 1.85 (s, 0.3H), 1.85-1.60 (m, 2H), 1.55
(d, 3H, J ) 7 Hz), 1.25 (t, 3H, J ) 6 Hz); 13C NMR 170.1, 167.8,
163.8, 161.5, 141.6, 141.3, 128.0, 127.8, 126.7, 126.4, 89.6, 87.0,
58.5, 55.7, 54.5, 47.8, 46.1, 35.2, 34.9, 21.4, 20.0, 16.8, 16.3,
15.6, 14.7, 14.3; IR (neat) 1675, 1575 cm-1. Anal. Calcd for
C17H23NO2: C, 74.69; H, 8.48; N, 5.12. Found: C, 74.92; H, 8.22;
N, 5.37.
141.7, 128.5, 128.2, 127.2, 127.1, 126.9, 126.6, 95.4, 93.9, 58.8,
56.1, 55.2, 48.0, 46.7, 35.8, 34.4, 33.0, 32.8, 31.1, 28.1, 22.9,
22.7, 22.3, 20.5, 17.4, 16.0, 14.6, 14.4, 13.9; IR (neat) 1670,
1570 cm-1. Anal. Calcd for C20H29NO2: C, 76.15; H, 9.27; N,
4.44. Found: C, 76.03; H, 9.35; N, 4.40.
2-P h en yl-2-[1-(1(S)-p h en yleth yl)p yr r olid in -2-ylid en e]-
eth a n oic Acid Eth yl Ester (6e). Condensation of ethyl
2-bromo-2-phenylethanoate (5e) with 4 gave 6e (E isomer):
1
yield 66%; oil; H NMR 7.2-7.0 (m, 10H), 4.3 (q, 1H, J ) 6.9
Hz), 4.0 (m, 2H), 3.2 (m, 2H), 3.1-2.7 (m, 2H), 1.7 (m, 2H),
1.1-1.0 (m, 6H); 13C NMR 169.8, 162.9, 141.0, 139.5, 132.0,
128.5, 128.3, 128.0, 127.1, 126.8, 126.1, 96.4, 59.0, 52.9, 46.9,
35.9, 21.3, 14.9, 14.6; IR (neat) 1670, 1550 cm-1. Anal. Calcd
for C22H25NO2: C, 78.77; H, 7.51; N, 4.18. Found: C, 78.90; H,
7.45; N, 4.11.
1-(1(S)-P h en ylet h yl)-2-m et h yl-4,5-d ih yd r op yr r ole-3-
ca r boxylic Acid Meth yl Ester (7). A mixture of 1-acetylcy-
clopropanecarboxylic acid methyl ester (1.42 g, 10 mmol) and
(S)-R-methylbenzylamine (1.21 g, 10 mmol) in 10 mL of toluene
was heated during 16 h at 140 °C. After complete elimination
of water in a Dean Stark apparatus, the solvent was evapo-
rated under reduced pressure and the residue was diluted with
diethyl ether. The solution was cooled at 0 °C, and the
hydrochloride of 7 was prepared by bubbling HCl during 15
min. The salt was extracted with water (20 mL), and the
aqueous layer was neutralized until pH 6.8-7.0 with
a
saturated KHCO3 solution. The organic phase was dried (Na2-
SO4), and solvent was evaporated. The crude compound was
distillated under vacuo (135 °C, 0.05 mmHg): yield 74%; oil;
1H NMR 7.4-7.15 (m, 5H), 4.85 (q, 1H, J ) 6.8 Hz), 3.15 (s,
3H), 3.4 (q, 1H, J ) 10 Hz), 3.25-3.05 (m, 1H), 2.75-2.6 (m,
2H), 2.3 (s, 3H), 1.55 (d, 3H, J ) 6.8 Hz); 13C NMR 167.8, 161.0,
141.1, 128.6, 127.3, 126.5, 95.6, 52.5, 50.0, 45.1, 26.5, 17.5, 12.1;
IR (neat) 1665, 1590 cm-1. [R] 20D ) +39.9 (c 1.96, EtOH). Anal.
Calcd for C15H19NO2 : C, 73.44; H, 7.81; N, 5.71. Found: C,
73.11; H, 7.89; N, 5.58.
(2S,3S)-[1-(1(S)-P h en yleth yl)-2-m eth ylp yr r olid in -3-yl]-
ca r boxylic Acid Meth yl Ester (8-9). To a degassed solution
of dihydropyrrole 7 (12.35 g, 50 mmol) in MeOH (50 mL) was
added PtO2 (50 mg), and the mixture was hydrogenated under
atmospheric pressure for 24 h. The solution was filtered, the
insoluble material was rinsed with MeOH, and the combinated
filtrates were evaporated to give a mixture of two cis diaster-
eomers 8 and 9 in an 11/1 ratio which was recrystallized from
2-[1-(1(S)-P h en ylet h yl)p yr r olid in -2-ylid en e]b u t yr ic
Acid Eth yl Ester (6b). Condensation of ethyl 2-bromobu-
tanoate (5b) with 4 gave 6b as a 75/25 E/ Z mixture: yield
1
n-hexane: yield 90%; mp 62 °C; H NMR 7.4-7.15 (m, 5H),
1
63%; oil; H NMR 7.3-7.2 (m, 5H), 5.4 (q, 0.3H, J ) 6.9 Hz),
3.7 (s, 3H), 3.65-3.4 (m, 2H), 3.2-3.0 (m, 1H), 2.8-2.6 (m,
1H), 2.6-2.45 (m, 1H), 2.3-2.1 (m, 1H), 2.0-1.8 (m, 1H), 1.35
(d, 3H, J ) 6.8 Hz), 0.80 (d, 3H, J ) 6.8 Hz); 13C NMR 173.7,
5.2 (q, 0.7H, J ) 6.9 Hz), 4.1 (q, 2H, J ) 7.0 Hz), 3.15-2.80
(m, 2H), 2.9 (m, 2H), 2.65-2.10 (m, 2H), 1.80-1.55 (m, 2H),
1.5 (d, 3H, J ) 6.9 Hz), 1.2 (q, 3H, J ) 7.0 Hz), 1.1 (t, 3H, J )
7.3 Hz); 13C NMR 169.8, 167.4, 162.5, 160.9, 141.5, 141.3,
128.0, 127.8, 127.6, 126.6, 126.5, 126.4, 126.1, 126.0, 96.1, 94.6,
58.1, 55.5, 54.9, 47.5, 46.2, 35.3, 33.7, 24.1, 21.5, 20.9, 20.1,
16.8, 15.3, 14.6, 14.5, 14.1; IR (neat) 1730, 1560 cm-1. Anal.
Calcd for C18H25NO2: C, 75.22; H, 8.77; N, 4.87. Found: C,
75.33; H, 8.72; N, 4.90.
145.3, 128.2, 127.4, 126.9, 60.8, 57.0, 51.5, 48.8, 47.5, 24.9, 20.9,
20
12.5; IR (CHBr3) 1725 cm-1. [R]
) -27.7 (c 2.03, EtOH).
D
Anal. Calcd for C15H21NO2: C, 72.84; H, 8.56; N, 5.66. Found:
C, 72.51; H, 8.61; N, 5.62.
Gen er a l P r oced u r e for Ca ta lytic Hyd r ogen a tion w ith
P la tin u m Oxid e. To a degassed solution of â-enamino ester
6 (0.75 mmol) in EtOH (10 mL) was added PtO2 (36 mg). The
resulting suspension was hydrogenated under atmospheric
pressure. The reaction was monitored by gas chromatography.
At the end of the reaction, after consumption of 1 mol of H2
per mole of 6, the solution was filtered. The filter cake was
rinsed with EtOH and the solvent removed. The crude mixture
of diastereomers 10-13 was directly analyzed by gas chro-
matography and 13C NMR.
Gen er a l P r oced u r e for Ca ta lytic Hyd r ogen a tion w ith
P la tin u m on 10% Ch a r coa l. To a degassed solution of
â-enamino ester 6 (or 7) (0.75 mmol) in EtOH (10 mL) was
added 37 mg of platinum on charcoal (10% w/w). The resulting
suspension was hydrogenated and worked up according to the
general procedure used for PtO2 reduction.
2-[1-(1(S)-P h en yleth yl)p yr r olid in -2-ylid en e]p en ta n o-
ic Acid Eth yl Ester (6c). Condensation of ethyl 2-bromopen-
tanoate (5c) with 4 gave 6c as a 75/25 E/ Z mixture: yield 60%;
oil; 1H NMR 7.3-7.2 (m, 5H), 5.3 (q, 0.3H, J ) 6.9 Hz), 5.1 (q,
0.7H, J ) 6.9 Hz), 4.0 (m, 2H), 3.1-2.8 (m, 2H), 2.9 (m, 2H),
2.6-2.1 (m, 2H), 1.7 (m, 2H), 1.55 (d, 3H, J ) 6.9 Hz), 1.50-
1.25 (m, 2H), 1.2 (t, 3H, J ) 7.0 Hz), 0.8 (m, 3H); 13C NMR
171.0, 168.7, 163.1, 161.7, 141.8, 128.5, 128.2, 127.2, 127.1,
126.9, 126.6, 95.4, 93.8, 58.8, 56.1, 55.3, 48.1, 46.8, 35.8, 34.4,
33.6, 30.5, 23.9, 21.9, 20.6, 17.5, 16.0, 14.6, 14.3, 14.2; IR (neat)
1670, 1560 cm-1. Anal. Calcd for C19H27NO2: C, 75.71; H, 9.03;
N, 4.65. Found: C, 75.59; H, 9.12; N, 4.56.
2-[1-(1(S)-P h en ylet h yl)p yr r olid in -2-ylid en e]h exa n o-
ic Acid Eth yl Ester (6d ). Condensation of ethyl 2-bromo-
hexanoate (5d ) with 4 gave 6d as a 75/25 E/ Z mixture: yield
61%; oil; 1H NMR 7.3-7.2 (m, 5H), 5.35 (q, 0.3H, J ) 6.9 Hz),
5.15 (q, 0.7H, J ) 6.9 Hz), 4.05 (m, 2H), 3.10-2.75 (m, 2H),
2.95 (m, 2H), 2.60-2.15 (m, 2H), 1.7 (m, 2H), 1.5 (d, 3H, J )
6.9 Hz), 1.45-1.25 (m, 4H), 1.2 (t, 3H, J ) 7.0 Hz), 0.85 (t,
3H, J ) 7.3 Hz);13C NMR 170.9, 168.5, 163.1, 161.6, 142.0,
Gen er a l P r oced u r e for Ca ta lytic Hyd r ogen a tion w ith
P a lla d iu m on 10% Ch a r coa l. To a degassed solution of
â-enamino ester 6 (or 7) (0.75 mmol) in EtOH (10 mL) was
added 35 mg of palladium on charcoal (10% w/w). The resulting
suspension was hydrogenated, and the same workup as above
gave a mixture of 10-13 which was directly analyzed.

3130 J . Org. Chem., Vol. 64, No. 9, 1999
David et al.
Gen er a l P r oced u r e for Ch em ica l Red u ction . A solution
of NaBH(AcO)3 was prepared with 140 mg of NaBH4 (3.75
mmol) and 2.1 mL of acetic acid (37.5 mmol) cooled at 0 °C.
After 30 min of stirring at room temperature, a solution of
â-enamino ester 6 (0.75 mmol) in MeCN (5 mL) was added
and the mixture stirred for 48 h. The solution was neutralized
with a satured Na2CO3 solution. The aqueous layer was
extracted with CH2Cl2 and washed with water; organic phase
was then dried (Na2SO4) and evaporated.
2-[1-(1(S)-P h en yleth yl)p yr r olid in -2-yl]p r op a n oic Acid
Eth yl Ester (10a -13a ). Reduction of 6a . Yields: 95% (PtO2),
96% (Pt/C), 90% (Pd/C), 85% (NaBH(AcO)3); oil; 1H NMR 7.30-
7.05 (m, 5H), 4.2-3.8 (m, 3H), 3.10-2.95 (m, 1H), 2.60-2.35
(m, 3H), 1.90-1.65 (m, 2H), 1.60-1.45 (m, 2H), 1.25 (d, 3H, J
) 7 Hz), 1.15 (t, 3H, J ) 7 Hz), 1.10 (d, 3H, J ) 7 Hz); 13C
NMR 12a 175.9, 144.8, 127.8, 127.9, 127.6, 126.4, 63.4, 60.0,
57.2, 46.7, 43.2, 28.1, 23.6, 14.4, 14.3, 12.2; 13a 175.8, 145.2,
128.0, 127.5, 126.5, 62.5, 60.0, 58.6, 48.0, 42.9, 27.2, 24.3, 14.4,
14.3, 10.9; IR (neat) 1730 cm-1. Anal. Calcd for C17H23NO2: C,
74.14; H, 9.15; N, 5.09. Found: C, 73.88; H, 9.02; N, 5.19.
2-[1-(1(S)-P h en ylet h yl)p yr r olid in -2-yl]b u t a n oic Acid
Eth yl Ester (10b-13b). Reduction of 6b. Yields: 95% (PtO2),
83% (Pt/C), 92% (Pd/C), 95% (NaBH(AcO)3); oil; 1H NMR 7.3-
7.0 (m, 5H), 4.2-3.8 (m, 3H), 3.25-1.8 (m, 4H), 1.75-1.30 (m,
6H), 1.25 (m, 3H), 1.15 (t, 3H, J ) 7.0 Hz), 0.75 (t, 3H, J ) 7.3
Hz); 13C NMR 12b 174.9, 144.9, 128.3-126.2, 62.4, 59.6, 56.6,
50.5, 46.8, 27.1, 23.2, 23.1, 14.0, 12.2; 13b 175.0, 144.5, 128.3-
126.2, 62.8, 59.6, 58.7, 52.0, 47.5, 27.8, 24.1, 19.7, 14.0; IR
(neat) 1730 cm-1. Anal. Calcd for C18H27NO2: C, 74.70; H, 9.40;
N, 4.84. Found: C, 74.73; H, 9.52; N, 4.79.
2-[1-(1(S)-P h en yleth yl)pyr r olidin -2-yliden e]pen tan edio-
ic Acid Dim eth yl Ester (17). Condensation of methyl 2-bro-
mopentandioate (15) with 4 gave 17 as a 65/35 E/ Z mixture:
yield 85%; oil; 1H NMR 7.3-7.2 (m, 5H), 5.35 (q, 0.4H, J )
6.9 Hz), 5.15 (q, 0.6H, J ) 6.9 Hz), 3.6-3.5 (2s, 6H), 3.35-
2.80 (m, 4H), 2.65-2.30 (m, 4H), 1.9-1.7 (m, 2H), 1.6-1.5 (m,
3H); 13C NMR 174.1, 173.2, 170.5, 168.0, 164.5, 163.3, 141.7,
141.2, 128.6, 128.2, 127.4, 127.0, 126.9, 126.6, 92.2, 90.4, 56.5,
55.6, 51.2, 50.4, 48.2, 47.1, 35.9, 34.7, 35.0, 34.8, 27.0, 24.1,
21.6, 20.3, 17.4, 16.1; IR (neat) 1730, 1670, 1550 cm-1. Anal.
Calcd for C19H25NO4: C, 68.86; H, 7.60; N, 4.23. Found: C,
68.85; H, 7.53; N, 4.21.
2-[1-(1(S)-P h en ylet h yl)p yr r olid in -2-yl]p en t a n ed ioic
Acid Dim et h yl E st er (20). To a degassed solution of â-
enamino diester 17 (3.04 g) in MeOH (120 mL) was added 10%
Pd/C w/w (121 mg, 0.04 equiv). The resulting suspension was
hydrogenated under atmospheric pressure and the reaction
monitored by GC. At the end of the reduction, after consump-
tion of 1 mol of H2 per mole of enamino diester, the solution
was filtered. The filter cake was rinsed with MeOH and solvent
removed. The crude mixture of amino diester diastereoisomers
was directly analyzed by GC and 13C NMR and then chro-
matographed on a silica gel column (AcOEt/cyclohexane 1/3).
Reduction yield was 95%. The mixture of 18, 19, 20 was
transformed into the picrate salt which was recrystallized
three times in n-butanol. The amino diester 20, liberated with
K2CO3 and extracted from the cake with CH2Cl2, was finally
obtained with a diastereomeric purity over 99% and 70%
yield: oil; 1H NMR 7.45-7.20 (m, 5H), 4.1 (q, 1H, J ) 6.9 Hz),
3.7 (s, 6H), 3.1 (m, 1H), 2.65-2.40 et 2.35-2.15 (2 m, 3H), 2.0-
1.4 (m, 8H), 1.34 (d, 3H, J ) 6.9 Hz); 13C NMR 174.8, 144.5,
2-[1-(1(S)-P h en yleth yl)p yr r olid in -2-yl]p en ta n oic Acid
Eth yl Ester (10c-13c). Reduction of 6c. Yields: 97% (PtO2),
95% (Pt/C), 90% (Pd/C), 98% (NaBH(AcO)3); oil; 1H NMR 7.4-
7.1 (m, 5H), 4.15-3.80 (m, 3H), 3.1-2.9 (m, 1H), 2.7-1.9 (m,
3H), 1.8-1.3 (m, 6H), 1.35-1.25 (m, 3H), 1.20-1.15 (m, 5H),
0.8 (t, 3H, J ) 7.3 Hz); 13C NMR 12c 175.0, 144.8, 128.3-
126.2, 62.7, 59.6, 56.8, 48.4, 46.6, 32.3, 26.9, 23.2, 21.0, 14.1,
13.9, 12.1; 13c 175.2, 144.5, 128.3-126.2, 62.9, 59.6, 58.6, 50.0,
128.5-126.5, 62.6, 57.0, 51.5, 51.2, 48.0, 46.6, 32.1, 27.1, 23.2,
20
21.4, 12.3; IR (neat) 1730 cm-1. [R]
) +26 (c 1.15, EtOH).
D
Anal. Calcd for C19H27NO4: C, 68.44; H, 8.16; N, 4.20. Found:
C, 68.48; H, 8.09; N, 4.10. 13C NMR spectra of minor diaste-
reomer 21 observed with PtO2, Pt/C, and chemical reduction:
174.9, 142.9, 128.5-126.5, 62.9, 59.2, 51.4, 51.2, 48.7, 48.1,
32.2, 27.7, 24.5, 24.1, 14.9.
47.5, 28.9, 27.8, 24.1, 21.0, 14.1,13.9, 12.1; IR (neat) 1720 cm-1
.
2-[1-(1(S)-P h en yleth yl)pyr r olidin -2-yl]bu tan edioic Acid
Dim eth yl Ester (24). With the same workup as used for 17,
the â-enamino diester 16 (4.08 g) was reduced using 10% Pt/C
w/w (204 mg, 0.05 equiv) as catalyst to obtain the crude
mixture of diastereomers 22-25 with 95% yield. Treatment
with picric acid and recrystallizations of picrate salt in
n-butanol finally gave amino diester 24 with a diastereomeric
Anal. Calcd for C19H29NO2: C, 75.20; H, 9.63; N, 4.62. Found:
C, 75.36; H, 9.52; N, 4.80.
2-[1-(1(S)-P h en yleth yl)p yr r olid in -2-yl]h exa n oic Acid
Eth yl Ester (10d -13d ). Reduction of 6d . Yields: 95% (PtO2),
80% (Pt/C), 86% (Pd/C), 98% (NaBH(AcO)3); oil; 1H NMR 7.4-
7.1 (m, 5H), 4.1-3.8 (m, 3H), 3.1-2.9 (m, 1H), 2.8-1.8 (m, 3H),
1.75-1.30 (m, 6H), 1.30-0.95 (m, 10H), 0.85-0.75 (m, 3H);
13C NMR 12d 175.0, 145.0, 128.5-126.4, 62.8, 59.7, 57.1, 48.6,
46.8, 29.8, 27.0, 22.7, 14.1, 13.9, 12.3; 13d 175.2, 144.8, 128.5-
126.4, 63.0, 59.7, 58.7, 50.1, 47.6, 26.4, 27.9, 24.1, 14.0, 13.8,
12.3; IR (neat) 1725 cm-1. Anal. Calcd for C20H31NO2: C, 75.67;
H, 9.84; N, 4.41. Found: C, 75.81; H, 9.69; N, 4.54.
2-P h en yl-2-[1-(1(S)-p h en yleth yl)p yr r olid in -2-yl]eth a -
n oic Acid Eth yl Ester (10e-13e). Reduction of 6e. Yields:
95% (PtO2), 80% (Pt/C), 98% (NaBH(AcO)3); oil; 1H NMR 7.3-
7.0 (m, 10H), 4.2 (q, 1H, J ) 6.9 Hz), 4.0 (q, 2H, J ) 7.0 Hz),
3.65 (m, 1H), 3.4 (d, 1H, J ) 9.3 Hz), 2.5 (m, 2H), 1.5 (m, 2H),
1.25 (d, 3H, J ) 6.9 Hz), 1.0 (t, 3H, J ) 7.0 Hz); 13C NMR 12e
173.8, 144.8, 137.6-126.3, 64.7, 60.5, 58.6, 58.5, 46.5, 29.4,
24.3, 14.1, 12.5; 13e 173.1, 144.5, 137.6-126.3, 64.7, 60.3, 58.4,
57.1, 47.1, 29.8, 24.0, 14.1, 12.2; IR (neat) 1720 cm-1. Anal.
Calcd for C22H27NO2: C, 78.30; H, 8.07; N, 4.15. Found: C,
78.23; H, 8.12; N, 4.06.
2-[1-(1(S)-P h en ylet h yl)p yr r olid in -2-ylid en e]b u t a n e-
d ioic Acid Dim eth yl Ester (16). Condensation of methyl
2-bromobutanedioate (14) with 4 gave 16 as a mixture of 75/
25 E/ Z isomers: yield 84%; oil; 1H NMR 7.6-7.4 (m, 5H), 5.48
(q, 0.3H, J ) 6.9 Hz), 5.11 (q, 0.7H, J ) 6.9 Hz), 3.94-3.89
(2s, 6H), 3.63 (s, 1.5H), 3.57 (s, 0.5H), 3.57-3.47 (m, 2H), 3.36
(t, 0.5H, J ) 7.7 Hz), 3.32-3.16 (m, 1.5H), 2.85 (t, 0.5H), 2.09-
1.98 (m, 2H), 1.89 (d, 0.8H, J ) 6.8 Hz), 1.84 (d, 2.2H, J ) 6.8
Hz); 13C NMR 173.5, 173.3, 169.6, 167.0, 165.2, 164.2, 141.7,
128.9, 128.2, 127.4, 127.0, 126.9, 87.1, 85.0, 56.3, 55.4, 51.2,
50.1, 47.9, 46.8, 36.1, 35.4, 34.8, 33.9, 21.2, 19.8, 17.7, 15.3;
IR (neat) 1730, 1670, 1550 cm-1. Anal. Calcd for C18H23NO4:
C, 68.12; H, 7.31; N, 4.41. Found: C, 68.15; H, 7.39; N, 4.42.
1
purity over 98% and 72% yield: oil; H NMR 7.38-7.26 (m,
5H), 4.03 (q, 1H, J ) 6.8 Hz), 3.68-3.66 (2s, 6H), 3.17 (q, 1H,
J ) 5.8 Hz), 3.04-2.96 (m, 1H), 2.82-2.71 (m, 1H), 2.60-2.51
(m, 3H), 1.84-1.76 (m, 2H), 1.66-1.58 (m, 2H), 1.31 (d, 3H, J
) 6.8 Hz); 13C NMR 174.0, 172.3, 144.0, 127.7-126.2, 61.5,
56.6, 51.3, 51.2, 46.2, 44.4, 33.3, 27.8, 23.0, 11.9; IR (neat) 1730
20
cm-1. [R]
) +37 (c 1.20, EtOH). Anal. Calcd for C18H25-
D
NO4: C, 67.69; H, 7.89; N, 4.39. Found: C, 67.77; H, 7.91; N,
4.35. 13C NMR spectra of minor diastereomer 25: 173.8, 173.0,
142.1, 127.7-126.2, 60.9, 58.6, 51.4, 51.3, 48.2, 44.0 29.6, 26.7,
23.6, 14.9.
5-Oxoocta h yd r oin dolizin -8-ca r boxylic Acid Meth yl Es-
ter (26). To a degassed solution of amino diester 20 (720 mg)
in MeOH (30 mL) was added 10% Pd/C w/w (435 mg, 0.6
equiv). The resulting suspension was stirred under an atmo-
spheric pressure of hydrogen for 24 h. At the end of the
reduction, controlled by GC, the solution was filtered, the
insoluble material was washed with MeOH, and the solvent
was removed. The crude mixture of diastereomers (99/1) of the
lactam ester was chromatographed on a silica gel column
(AcOEt/cyclohexane 1/3) to give optically pure 26: reduction
yield 88%; white solid, mp 74 °C; 1H NMR 3.66 (s, 3H), 3.53-
3.49 (m, 1H), 3.45-3.38 (m, 2H), 3.34-2.28 (m, 2H), 2.14 (m,
1H), 2.09 (m, 1H), 1.95-1.60 (m, 4H), 1.42 (m, 1H); 13C NMR
173.0, 167.9, 60.0, 52.0, 45.8, 44.9, 32.4, 30.4, 25.1, 21.9; IR
(CHBr3) 1740, 1625 cm-1. [R] 20 ) +68 (c 1.19, EtOH). Anal.
D
Calcd for C10H15NO3: C, 60.89; H, 7.67; N, 7.10. Found: C,
60.86; H, 7.72; N, 7.12.
3-Oxoh exa h yd r op yr r olizin -1-ca r boxylic Acid Meth yl
Ester (27). With the same workup as above, used for 20,

Pyrrolizidine and Indolizidine Alkaloids
J . Org. Chem., Vol. 64, No. 9, 1999 3131
amino diester 24 (2.2 g) and 10% Pd/C (1.3 g, 0.6 equiv) gave
a mixture of diastereomers of lactam esters (98/2) which was
chromatographed to give optically pure compound 27: reduc-
tion yield 85%; white solid, mp 43 °C; 1H NMR 4.02-3.93 (m,
1H), 3.68 (s, 3H), 3.60-3.39 (m, 1H), 3.06-2.85 (m, 3H), 2.66-
2.58 (m, 1H), 2.16-1.94 (m, 3H), 1.47-1.31 (m, 1H); 13C NMR
172.4, 171.9, 63.7, 52.2, 45.7, 41.1, 38.3, 31.5, 26.6; IR (CHBr3)
1740, 1685 cm-1. [R] 20D ) +67 (c 1.25, EtOH). Anal. Calcd for
C9H13NO3: C, 59.00; H, 7.15; N, 7.65. Found: C, 59.04; H, 7.20;
N, 7.80.
(m, 2H), 2.7-2.45 (m, 1H), 2.4-2.0 (m, 4H), 1.75-1.5 (m, 1H);
13C NMR 175.4, 172.9, 63.1, 51.8, 45.2, 41.0, 33.6, 30.2, 22.3;
IR (neat) 1725, 1680 cm-1. [R] 20D ) -16.6 (c 2.08, EtOH). Anal.
Calcd for C9H13NO3: C, 59.00; H, 7.15; N, 7.65. Found: C,
58.37; H, 7.07; N, 7.67.
(1S,8S)-5-Th ioxoh exah ydr opyr r olizin -1-car boxylic Acid
Meth yl Ester (34). A mixture of bicyclic lactam 32 (4.9 g, 26.8
mmol) and Lawesson reagent (5.7 g, 14.1 mmol) was refluxed
in dry toluene (50 mL) during 1 h. The solvent was then
evaporated, and the solid obtained was recrystallized in
1
(+)-Tash ir om in e (8-Meth an ol octah ydr oin dolizin e) (2).
A mixture of lactam ester 26 (440 mg, 2.23 mmol) and LiAlH4
(340 mg, 9 mmol) was stirred in freshly distilled THF (40 mL)
for 24 h at room temperature. The residue was quenched with
440 µL of water, 440 µl of an aqueous NaOH solution (15%),
and then 1 mL of water. After filtration and evaporation of
the solvent, the crude product was distillated on a Kugelrohr
apparatus (125 °C, 0.3 mmHg): yield 87%; white solid, mp 35
°C; 1H NMR 3.88-3.81 and 3.71-3.64 (2m, 2H), 3.31-3.21 (m,
2H), 2.31-2.20 (q, 1H, J ) 17.8 Hz), 2.18-2.0 (m, 3H), 1.99-
1.78 (m, 2H), 1.76-1.63 (m, 3H), 1.32-1.15 (m, 1H); 13C NMR
Et2O: yield 56%; mp 81 °C; H NMR 4.5-4.3 (m, 1H), 4.05-
3.7 (m, 1H), 3.65 (s, 3H), 3.4-3.2 (m, 1H), 3.2-2.9 (m, 3H),
2.55-2.1 (m, 3H), 1.8-1.5 (m, 1H); 13C NMR 196.0, 172.1, 70.2,
51.9, 48.3, 44.5, 44.1, 30.6, 25.2; IR (neat) 1725 cm-1. [R] 20
)
D
-118.2 (c 0.9, EtOH). Anal. Calcd for C9H13NO3S: C, 54.27;
H, 6.53; N, 7.03. Found: C, 53.91; H, 6.47; N, 7.11.
(1S,8S)-Hexa h yd r op yr r olizin -1-ca r boxylic Acid Meth -
yl Ester (35). A mixture of bicyclic thiolactam 34 (2.2 g, 11
mmol) and methyl iodide (5.46 g, 38.5 mmol) in CH2Cl2 (25
mL) was stirred at room temperature for 24 h. The solvent
was removed, and the residue was dissolved in MeOH (30 mL).
NaBH4 (0.84 g, 22 mmol) was then added. After 12 h of stirring
at room temperature, the solution was neutralized with of 1
M HCl (10 mL). After 2 h, K2CO3 was added. The gel formed
was extracted with CHCl3 (3 × 60 mL). Solvent was evapo-
rated, and the crude product was distillated on a Kugelrohr
apparatus (ot 105 °C, 0.05 mmHg): yield 77%; 1H NMR 3.80-
3.6 (m, 1H), 3.65 (s, 3H), 3.85-2.7 (m, 1H), 3.2-2.85 (m, 3H),
2.65-2.45 (m, 1H), 2.2-1.95 (m, 1H), 1.95-1.55 (m, 4H), 1.4-
1.35 (m, 1H); 13C NMR 174.1, 65.9, 55.7, 53.7, 51.6, 47.4, 28.6,
26.8, 26.4; IR (neat) 1725 cm-1. [R]20D ) -65.1 (c 2.16, CHCl3).
Anal. Calcd for C9H15NO2: C, 63.88; H, 8.94; N, 8.28. Found:
C, 63.37; H, 8.73; N, 8.19.
66.5, 65.5, 54.2, 52.7, 44.6, 29.1, 27.7, 25.2, 20.8; IR (CHBr3)
20
3600-3100 cm-1. [R]
) +44.7 (c 1.1, EtOH). Anal. Calcd
D
for C9H17NO: C, 69.63; H, 11.04; N, 9.02. Found: C, 69.72; H,
11.01; N, 9.11.
(+)-La bu r n in e (1-Meth a n ol h exa h yd r op yr r olizin e) (3).
A mixture of bicyclic lactam 27 (350 mg, 1.91 mmol) and
LiAlH4 (365 mg, 9.56 mmol) in 40 mL of dried THF worked
up according to the procedure used for 26 gave (+)-laburnine
3: yield 85%; oil; 1H NMR 3.63 (d, 2H, J ) 6.2 Hz), 3.22-3.09
(m, 2H), 2.99-2.90 (m, 1H), 2.64-2.48 (m, 2H), 2.11 (m, 1H),
2.07-1.48 (m, 7H); 13C NMR 67.5, 64.9, 54.7, 52.7, 48.5, 32.0,
20
30.1, 25.7; IR (neat) 3600-3100 cm-1. [R]
) +15.4 (c 1.2,
D
EtOH). Anal. Calcd for C8H15NO: C, 68.04; H, 10.71; N, 9.92.
Found: C, 68.01; H, 10.62; N, 9.97.
(-)-Isor etr on eca n ol (1-Meth a n ol h exa h yd r op yr r oliz-
in e) (1). To 10 mL of a 1 M solution of LiAlH4 in THF cooled
at -80 °C was added dropwise a solution of 35 (1.1 g, 6 mmol)
in 14 mL of freshly distilled THF. The mixture was stirred 24
h at -80 °C and then quenched with 1 mL of water in 10 mL
of THF. After filtration and evaporation of the solvent, the
crude product was distillated on a Kugelrohr apparatus (ot
180 °C, 0.05 mmHg): yield 87%; 1H NMR 4.33 (br s, 1H), 3.85-
3.27 (d, 3H), 3.6 (s, 3H), 3.27-2.77 (m, 2H), 2.77-1.03 (m, 2H);
13C NMR 66.32, 62.97, 55.65, 54.05, 44.33, 27.20, 26.48, 25.95;
IR (neat) 3600-2500 cm-1. [R] 20D ) -76.8 (c 2.1, EtOH). Anal.
Calcd for C8H15NO: C, 68.04; H, 10.71; N, 9.91. Found: C,
67.88; H, 10.62; N, 9.75.
3-[1-(1(S)-P h en ylet h yl)-3-m et h oxyca r b on yl-4,5-d ih y-
d r op yr r ol-2-yl]p r op a n oic Acid Meth yl Ester (29). With
the same procedure used for 7, reaction between 4-oxo-4-(1-
methyloxycarbonylcyclopropyl)butanoic acid methyl ester (28)
(14.98 g, 70 mmol) and (S)-R-methylbenzylamine (8.47 g, 70
1
mmol) gave compound 29: yield 83%; oil; H NMR 7.4-7.15
(m, 5H), 4.9 (q, 1H, J ) 6.8 Hz), 3.65 (s, 3H), 3.6 (s, 3H), 3.55
(m, 1H), 3.3 (m, 1H), 3.3-2.9 (m, 3H), 2.75-2.5 (m, 4H), 1.55
(d, 3H, J ) 6.8 Hz); 13C NMR 172.9, 167.0, 162.9, 141.0, 128.6,
127.3, 126.5, 95.7, 52.4, 51.7, 50.1, 45.2, 32.4, 26.3, 20.9, 17.7;
IR (neat) 1730, 1660, 1570 cm-1. [R] 20D ) +29.6 (c 1.98, EtOH).
Anal. Calcd for C18H23NO4: C, 68.12; H, 7.31; N, 4.41. Found:
C, 67.97; H, 7.41; N, 4.51.
Molecu la r Mod elin g Stu d y. A molecular modeling study
of described molecules was performed using the SYBYL 6.3
package on a Silicon Graphics R 8000 workstation. Structures
were built within SYBYL and minimized by the Tripos force
field Maximin 2 in vacuo conditions to provide reasonable
standard geometries. The conjugate gradient method was used
for minimization, and molecules were considered to be mini-
mized when a minimum energy charge of less than 0.021 kJ
mol-1 for one iteration was reached. Semiempirical AM1
calculations were performed using the MOPAC package
implemented in SYBYL and involved the singlet state. The
conformational spaces of compounds 6a and 7 were explored
using the SYBYL search facility. Torsion angles were defined,
and a qrid search was performed allowing chosen bonds to
rotate within a 360° or 180° revolution by 15° increments. The
lowest energy conformers thus obtained were submitted to
AM1 calculations (MOPAC version 5.0) to optimize their
geometry.
(2S,3S)-3-[1-(1(S)-P h en yleth yl)-3-m eth oxyca r bon yl-4,5-
p yr r olid in -2-yl]p r op a n oic Acid Meth yl Ester (30). A
mixture of dihydropyrrole 29 (16.1 g, 50.8 mmol) and PtO2 (125
mg) in MeOH (50 mL) worked up according to the procedure
used for the reduction of 7 gave reduced compound 30: yield
87%; 1H NMR 7.4-7.1 (m, 5H), 3.8-3.7 (m, 1H), 3.65 (s, 3H),
3.60 (s, 3H), 3.25-3.15 (m, 1H), 3.05-2.95 (m, 1H), 2.95-2.85
(m, 1H), 2.75-2.65 (m, 1H), 2.35-2.25 (m, 2H), 2.2-2.1 (m,
1H), 1.95-1.85 (m, 1H), 1.7-1.6 (m, 1H), 1.45-1.35 (m, 1H),
1.35 (d, 3H, J ) 6.8 Hz); 13C NMR 173.9, 173.5, 144.5, 128.2,
127.5, 126.8, 61.9, 60.4, 51.4, 51.3, 47.2, 46.4, 30.8, 26.2, 25.7,
18.1; IR (neat) 1730 cm-1. [R] 20 ) +7.5 (c 2.08, EtOH). Anal.
D
Calcd for C18H25NO4: C, 67.69; H, 7.89; N, 4.39. Found: C,
67.61; H, 8.01; N, 4.42.
(1S,8S)-5-Oxoh exa h yd r op yr r olizin -1-ca r boxylic Acid
Meth yl Ester (32). To a degassed solution of N-substituted
pyrrole 30 (10.3 g, 32 mmol) in MeOH (30 mL) was added 5%
Pd/C w/w (1.5 g). The resulting suspension was stirred for 48
h under an atmospheric pressure of hydrogen. The solution
was filtered, the filter cake was washed with MeOH, and the
combinated filtrates were evaporated. The residue was dis-
solved in toluene (50 mL), and the solution was heated at 90
°C for 5 h. The crude product was chromatographed on a silica
The optimized structure of 6e was subjected to a restrained
molecular dynamic simulation lasting 10 ps with a retention
time step of 1 fs. The temperature was raised to 1000 K, and
molecules were captured every 5 fs. Structures were selected
after analysis of dynamic trajectory, and the lowest energy
conformers were minimized by Maximin 2.
1
gel column (THF saturated with NH3): yield 90%; H NMR
4.08 (q, 1H, J ) 10 Hz), 3.8-3.5 (m, 1H), 3.6 (s, 3H), 3.1-2.85
J O982169P