Reverse turn induced π-facial selectivity during polyaniline-supported cobalt(II) salen catalyzed aerobic epoxidation of N-cinnamoyl L-proline derived peptides

Article abstract of DOI:10.1021/jo020352j

A novel chemo- and diastereoselective aerobic epoxidation of the N-cinnamoyl peptides catalyzed by polyaniline-supported cobalt(II) salen (PASCOS) is described. The N-cinnamoyl proline derived peptides 1 show a high π-facial selectivity during these epoxidations. The origin of this diastereoselectivity in I has been attributed to (i) the propensity of the N-cinnamoyl proline amide to exist predominantly as trans rotamer in CDCl₃, DMSO-d₆, and CH₃CN medium and (ii) existence of these peptides as organized structures (γ- and β-turns) due to the presence of intramolecular hydrogen bonds. An extensive solution NMR and MD simulation study on 1d and 1f indicates that the origin of the high π-facial selectivity is due to the well-defined γ- and β-turns which result in the hindrance of one face of the cinnamoyl double bond in the transition state of the epoxidation reaction.

Full text of DOI:10.1021/jo020352j

Rever se Tu r n In d u ced π-F a cia l Selectivity d u r in g
P olya n ilin e-Su p p or ted Coba lt(II) Sa len Ca ta lyzed Aer obic
Ep oxid a tion of N-Cin n a m oyl L-P r olin e Der ived P ep tid es
J yoti Prokosh Nandy,† E. N. Prabhakaran,† S. Kiran Kumar,‡ A. C. Kunwar,‡ and
J aved Iqbal*,†,§
Department of Chemistry, Indian Institute of Technology, Kanpur 208 016, India, and Indian Institute of
Chemical Technology, Hyderabad 500 007, India.
javediqbaldrf@hotmail.com
Received May 29, 2002
A novel chemo- and diastereoselective aerobic epoxidation of the N-cinnamoyl peptides catalyzed
by polyaniline-supported cobalt(II) salen (PASCOS) is described. The N-cinnamoyl proline derived
peptides 1 show a high π-facial selectivity during these epoxidations. The origin of this diastereo-
selectivity in 1 has been attributed to (i) the propensity of the N-cinnamoyl proline amide to exist
predominantly as trans rotamer in CDCl3, DMSO-d6, and CH3CN medium and (ii) existence of
these peptides as organized structures (γ- and â-turns) due to the presence of intramolecular
hydrogen bonds. An extensive solution NMR and MD simulation study on 1d and 1f indicates that
the origin of the high π-facial selectivity is due to the well-defined γ- and â-turns which result in
the hindrance of one face of the cinnamoyl double bond in the transition state of the epoxidation
reaction.
In tr od u ction
an oxo-transfer agent to olefins. The polyaniline-sup-
ported cobalt acetate catalyst was recyclable, resistant
to decomposition in organic solvents as well as in aqueous
medium, and applicable for epoxidation on a wide variety
of olefins including chalcone and R,â-unsaturated amides
with good to excellent yields. Moreover, unlike most of
the epoxidation catalysts known in the literature, this
catalyst did not require any peroxo or any other hazard-
ous oxygen-rich compound as an oxo source. Reactions
were carried out in the presence of isobutyraldehyde and
simply in ambient oxygen pressure (using an oxygen
balloon), using molecular oxygen as the oxo source. Soon
after the successful introduction of polyaniline as a
support for an epoxidation catalyst, cobalt-salen was
attached to it to develop another heterogeneous epoxi-
dation catalyst, which came out to be equally as robust
as polyaniline-supported cobalt acetate and effected the
epoxidation reactions on cinnamoyl amide derivatives in
excellent yields.4 The most significant difference between
other polymer-linked catalysts and polyaniline-supported
cobalt(II) salen (PASCOS) is that in this case the polymer
itself may coordinate to the central metal ion, which will
make the catalyst very stable and less prone to leaching
of the polymer from the catalyst. A tentative structural
arrangement of cobalt(II) salen in the polyaniline matrix
is shown in Figure 1.
After extensive work on the development of homoge-
neous cobalt catalysts for biomimetic aerobic oxidation
of various functionalized/unfunctionalized hydrocarbons
and epoxidation of a diverse range of olefinic systems,1
we introduced polyaniline-supported cobalt acetate as a
simple and novel heterogeneous epoxidation catalyst in
which a low-cost, easily synthesizable polyaniline was
used for the first time as the polymeric support to
generate a heterogeneous epoxidation catalyst.2 Earlier,
polyaniline was used as a polymeric support to palladium
for generation of a heterogeneous catalyst for hydrogena-
tion of alkenes,3 but there was no report of it being used
as a heterogeneous support to a catalyst that acted as
† Indian Institute of Technology.
‡ Indian Institute of Chemical Technology.
§ Present address: Discovery Research, Dr. Reddy’s Laboratories
Ltd., Bollaram Road, Miyapur, Hyderabad 500 050, India.
(1) (a) Punniyamurthy, T.; Bhatia, B.; Iqbal, J . J . Org. Chem. 1994,
59, 850. (b) Punniyamurthy, T.; Iqbal, J . Tetrahedron Lett. 1994, 35,
4003 and 4007. (c) Reddy, M. M.; Punnyamurthy, T.; Iqbal, J .
Tetrahedron Lett. 1995, 36, 159. (d) Reddy, M. M.; Punnyamurthy, T.;
Iqbal, J . Tetrahedron Lett. 1995, 36, 159. (e) Punnyamurthy, T.; Reddy,
M. M.; Kalra, S. J . S.; Iqbal, J . J . Pure Appl. Chem. 1996, 68, 619. (f)
Punnyamurthy, T.; Kalra, S. J . S.; Iqbal, J . Tetrahedron Lett. 1995,
36, 8497. (g) Punnyamurthy, T.; Asthana, P.; Kalra, S. J . S.; Iqbal, J .
Proc. Indian Acad. Sci. 1996, 107, 335. (h) Mandal, A. K.; Khanna, V.;
Iqbal, J . Tetrahedron Lett. 1996, 37, 3769. 17.
(2) For other polymer-supported catalysts see: (a) Karjalainen, J .
K.; Hormi, O. E. O.; Sherington, D. C. Molecules 1998, 3, 51. (b)
Karjalainen, J . K.; Hormi, O. E. O.; Sherington, D. C. Tetrahedron:
Asymmetry 1998, 9, 1563. (c) Karjalainen, J . K.; Hormi, O. E. O.;
Sherington, D. C. Tetrahedron: Asymmetry 1998, 9, 2019. (d) Kar-
jalainen, J . K.; Hormi, O. E. O.; Sherington, D. C. Tetrahedron:
Asymmetry 1998, 9, 3895. (e) Zhang, W.; Loebach, J . L.; Wilson, S. R.;
J acobsen, E. N. J . Am. Chem. Soc. 1990, 112, 2801. (f) Canali, L.;
Cowan, E.; Deleuze, H.; Gibson, C. L.; Sherington, D. C. J . Chem. Soc.,
Chem. Commun 1998, 2561. (g) Yao, X.; Huilin, C.; Lu, W.; Pan, G.;
Hu, X.; Zheng, Z. Tetrahedron Lett. 2000, 41, 10267.
Recently it has been shown from our group that
N-cinnamoyl amides 1 can be converted to â-phenyliso-
serine derivatives 3 via a tandem polyaniline-supported
cobalt(II) salen (PASCOS) catalyzed aerobic epoxidation
and amination protocol (Scheme 1). The intermediate
(3) Das, B. C.; Iqbal, J . Tetrahedron Lett. 1997, 38, 1235.
(4) Sobczak, J . W.; Lesaik, B.; J ablonski, A.; Palczewska, W. Pol. J .
Chem. 1995, 69, 1732
10.1021/jo020352j CCC: $25.00 © 2003 American Chemical Society
Published on Web 02/01/2003
J . Org. Chem. 2003, 68, 1679-1692
1679

Nandy et al.
epoxide 4 that was converted to the corresponding
carboxylic acid 5 by ruthenium catalyzed oxidation of the
primary alcohol group.6 The carboxylic acid 5 was sub-
sequently transformed to the corresponding peptide 2c
by mixed anhydride coupling with the methyl ester of
L-proline-L-leucine dipeptide (Scheme 2). The specific
rotation of 2c prepared by Sharpless’s procedure (RD
)
-191°) and that by PASCOS catalyzed epoxidation of 1c
were similar in sign and magnitude. Also, the proton
NMR of epoxide 2c obtained by both routes was identical.
Thus, based on this correlation, the absolute stereochem-
istry of epoxide 2c is assigned 2R,3S. A similar correla-
tion was also carried out for 2d and 2e.
F IGURE 1.
cinnamoyl epoxide 2 undergoes a stereoselective SN2 type
opening with several aromatic amines leading to a highly
selective synthesis of the anti diastereomer of the â-phe-
nylisoserine derivatives.
It is noteworthy that the epoxidation of tripeptides
1f-h was found to be more diastereoselective compared
with that of peptides 1c-e as they underwent highly
selective aerobic epoxidation, leading to corresponding
epoxides 2f-h in high yields under the same reaction
conditions (Table 2). The ratio in favor of the major
diastereomer was found to be >85:<15 (HPLC), and they
were separated by column chromatography to afford the
epoxide whose purity was ∼90% (chiral HPLC). The
absolute stereochemistry of the major diastereomer was
assigned to be 2R,3S by converting diastereomerically
pure 2c (obtained by aerobic epoxidation) to 2f-h (fol-
lowing a sequence of alkaline hydrolysis of 2c and mixed
anhydride coupling with the corresponding amine/esters
of amino acids) and comparing with same epoxides
obtained by direct epoxidation (Sharpless) of the corre-
sponding olefins.
The moderate yields of the epoxides 2 during PASCOS
catalyzed epoxidations and the formation of other oxi-
dized products led us to modify the reaction conditions.
We reasoned that buffering the reaction medium might
help in reducing the reaction period, which may reduce
the side reactions also. Hence the epoxy peptides 2b, 2c,
2f, 2i, and 2j were obtained with the following epoxida-
tion conditions in high chemical and optical purity (Table
3). To a solution of peptide 1 in acetonitrile was added
2-methylpropanal (3 equiv) and PASCOS catalyst (10
mg), followed by sodium acetate (10 equiv), and the
reaction mixture was stirred for 12 h under aerobic
condition. Expectedly, 1a and 1c underwent complete
conversion to the corresponding epoxide 2b and 2c,
respectively, in excellent yields. On the other hand, an
additional amount of aldehyde (2 equiv) and catalyst were
required after 12 h to achieve the synthesis of epoxides
2f, 2i, and 2j in excellent conversion and yields during
20-22 h of stirring. It is noteworthy that there was no
significant change in the rotation indicating thereby that
the diastereoselectivity is not affected as a result of
buffering the reaction medium. Thus, buffering the
epoxidation reactions with sodium acetate resulted in
faster reactions leading to excellent chemical yields, with
no adverse effect on the diastereoselectivity of the cor-
responding epoxides.
Thus, epoxidation of the cinnamoyl amides, followed
by opening of the epoxides with amines in the same pot,
catalyzed by polyaniline-supported cobalt salen, led to
the formation of vicinal amino alcohols which are â-phe-
nylisoserne deivatives, a novel class of unnatural â-amino
acid residues. This is an excellent example of protecting
group functionalization where N-cinnamoylated peptides
are used as synthons to generate novel â-amino acid
residues in a very selective manner. PASCOS is a very
specific catalyst for epoxidaion of cinnamoyl double
bonds. Allyl and crotonyl amides remain unaffected under
the same reaction conditions. To achieve a stereoselective
outcome of the two-step process, both steps (that is, the
epoxidation and the opening) have to be stereoselective
in nature. It already has been established5b that the ring
opening in glycidyl peptides proceeds via an SN2 pathway,
giving only the anti diastereomers as the products under
these reaction conditions. To achieve an overall chiral
synthesis of â-phenylisoserine peptides, we attempted
stereoselective epoxidation on a number of N-cinnamoy-
lated peptides using cobalt salen polyaniline as catalyst
and also conducted a detailed study to find out the origin
of this diastereoselectivity, which is described in the
following section.
Resu lts a n d Discu ssion s
To achieve a chiral synthesis of these derivatives we
have attempted a diastereoselective aerobic epoxidation
of N-cinnamoyl-L-proline derivatives. The findings are
listed in Table 1, which indicates that the epoxidation of
methyl-N-cinnamoyl L-prolinate 1a exhibits very poor
diastereoselectivity during epoxidation; however, the
corresponding allylic amide 1b showed better selectivity
in formation of the corresponding epoxide 2b than the
former. Interestingly, the diastereoselectivity increases
for substrates having an additional C-terminal amino
acid residue in 1a . Thus, PASCOS catalyzed aerobic
epoxidation of dipeptides 1c-e afforded the correspond-
ing epoxides 2c-e, respectively, with higher diastereo-
selectivities. The ratio in favor of the major enantiomer
was found to be >70:<30.
The absolute stereochemistry for the major enantiomer
of epoxides 2a -e was assigned as 2R,3S, which was
established by following a correlation with the epoxide
obtained by Sharpless’s epoxidation procedure.5 Thus the
cinnamoyl alcohol was subjected to Sharpless’s epoxida-
tion by using (+)-DET to give enantiomerically pure
Or igin of Dia ster eoselectivity. To understand the
cause for the diastereoselectivity during these epoxida-
tions, we have focused on the population of the cis/ trans
conformation of the cinnamoyl-Proline amide bond
which may have a profound role in dictating the π-facial
selectivity of the cinnamoyl group during epoxidation. It
(5) (a) Das, B. C.; Iqbal, J . Tetrahedron Lett. 1997, 38, 2903. (b) De,
A.; Basak, P.; Iqbal, J . Tetrahedron Lett. 1997, 38, 8383.
(6) Hanson, R. M.; Sharpless, K. B. J . Org. Chem. 1986, 51, 1922.
1680 J . Org. Chem., Vol. 68, No. 5, 2003

N-Cinnamoly L-Proline Derived Peptides
SCHEME 1
TABLE 1. P olya n ilin e-Su p p or ted Coba lt(II) Sa len
(P ASCOS) Ca ta lyzed Aer obic Ep oxid a tion of
N-cin n a m oyl-L-p r olin e P ep tid es 1
TABLE 2. P olya n ilin e-Su p p or ted Coba lt(II) Sa len
(P ASCOS) Ca ta lyzed En a n tioselective Aer obic
Ep oxid a tion of N-Cin n a m oyl-L-p r olin e Con ta in in g
P ep tid esa
a
The isolated yields of the epoxides were 75-78% in all the
b
cases. The ratios of the diastereomers were determined by chiral
HPLC and the absolute configuration (2R,3S) and specific rotations
(CH2Cl2) of the major epoxide is shown in all cases.
acid at the N-terminal of Proline. It would be interesting
to see in our case as the Proline-Hδ and the R- or â-vinyl
proton would be the controlling factor in determining the
π-facial selectivity during epoxidation. To address that,
we have studied the solution NMR spectra of the peptides
having the cinnamoyl group in both cis and trans
orientations. The possible interactions in cis and trans
rotamers in the cinnamoyl Proline system are the fol-
lowing (Figure 2).
a
The isolated yields of the epoxides were 80-90% in all the
cases. b Ratio of the diastereomers determined by chiral HPLC and
the absolute configuration (2R,3S) and specific rotations (in
CH2Cl2) of the major epoxide are shown in all cases.
SCHEME 2. Assign m en t of Ep oxid e
Ster eoch em istr y in 2c
Two olefin peptides viz. allyl N-cinnamoyl-L-Prolyl-L-
Leucinate 1d and allyl N-cinnamoyl-(L)-Prolyl-(L)-Leucine
amide 1f (differing only in the linkage pattern at the
C-terminal of the leucine residue) were chosen for
detailed study of their solution NMR spectra. Distinction
between the cis and trans rotamers of Proline in both
the peptides was done by the help of NOESY experiment.
The trans isomer was characterized by the presence of
NOEs between pro δ-H and cinnamoyl R-H whereas the
cis isomer showed NOE between Proline CRH and the
cinnamoyl CRH (Figure 3). Both peptides, when they exist
with the trans-cinnamoyl-Proline linkage, have the pos-
sibility of forming stable intramolecular hydrogen bonds
between the oxygen of the cinnamoyl carbonyl and the
amide NH(s), thereby existing as highly organized struc-
tures. These kinds of intramolecular hydrogen bonds are
salient features of the proteins and dictate the formations
of reverse turns (â- and γ-turns) in the peptide chains
which are important motifs of the protein secondary
structures. A seven-membered ring in the peptide chain
formed due to intramolecular hydrogen boding between
the carbonyl of the ith residue and the NH of the (i+2)nd
residue makes a γ-turn while similar kinds of hydrogen
is well-known that Proline amides are capable of existing
as two rotamers and their population is controlled by the
polarity of the solvent. The population of the cis/trans
ratio is also guided by Proline-Hδ and acyl group interac-
tion. For example, population of the cis/trans rotamer
ratio in the peptide bond between Proline and an R-amino
acid will be governed by the noncovalent interaction
between the Proline-Hδ and the side chain of the R-amino
J . Org. Chem, Vol. 68, No. 5, 2003 1681

Nandy et al.
TABLE 3. Com p a r ison betw een Bu ffer ed a n d
Un bu ffer ed Rea ction s Con d u cted u n d er P ASCOS
Ca ta lyzed Aer obic Ep oxid a tion of N-Cin n a m oyl P r olin e
P ep tid es 1
F IGURE 3. NOE’s observed in 1d .
a
Yield of the isolated diastereomer. The ratios of diastereo-
meters were not determined. The purities of the isolated epxodies
are in the range of 90-95%. Reaction conditions: To a solution
F IGURE 4. NOE’s observed in 1f.
b
In our case, for both peptides 1d and 1f, Proline can
be considered as occupying the (i+1) position, so it is only
the trans-cinnamoyl-Proline bond that can promote the
formation of the turns in both cases. For peptide 1d , a
seven-membered ring (γ-turn) can form due to intramo-
lecular hydrogen bonding between the cinnamoyl carbo-
nyl and the NH of the Leucine residue (Figure 3) while
for peptide 1f there are possibilities for the formation of
both a 7-membered ring, i.e. a γ-turn (due to similar kind
of bonding as in 1d ), as well as a 10-membered ring (â-
turn) due to intramolecular hydrogen bonding between
the cinnamoyl carbonyl and the NH of the allyl amide
(Figure 3). The solution NMR of 1f in DMSO-d6 shows
two sets of signals in the ratio of 45:55 due to cis and
trans rotamers, respectively, about the Proline-amide
bond. Detailed analysis of NMR spectra was undertaken
for both isomers and the spectral data for 1f with trans
and cis rotamers of Proline are presented in Tables 4 and
5, respectively.
To probe the existence and extent of intramolecular
hydrogen bonding, variable-temperature NMR experi-
ments are a very useful tool. The low magnitude of the
chemical shift/temperature coefficient (∆δ/∆T) for the
amide proton is an indicator of its participation in
H-boding. The (∆δ/∆T) values for the cis isomer of 1f have
a large magnitude (ca. -6 ppb/°C) for both the NHs
(Table 4) while it is moderately small (Table 3) for the
trans isomer (∆δ/∆T ) -3.4 ppb/°C for allyl NH and -4.5
ppb/°C for Leu NH). It suggests that a sizable fraction of
the trans isomer has allyl NH participating in H-bonding.
The presence of such an H-bond coupled with the exist-
ence of NOESY peaks between Leu NH/allyl NH and
strong Proline CRH/Leucine NH suggest the preference
of a â-turn around the Proline-Leucine residue (Figure
4).
of peptide 1 in acetonitrile was added 2-methylpropanal (3 equiv)
and PASCOS catalyst, followed by sodiuim acetate (10 equiv), and
the reaction mixture was stirred for 12-20 h under oxygen balloon.
F IGURE 2.
bonding between the carbonyl of the ith residue and NH
of the (i+3)rd residue generate a 10-membered ring
known as a â-turn. These turns are stable in suitable
media and dilution and keep the peptides in highly
organized form.8 Interestingly, it is only the carbonyl of
the trans rotamer of the amide linkage between the ith
and (i+1)st residue that can participate in these turns.
In the corresponding cis rotamer, this particular carbonyl
points in the opposite direction and no intramolecular
hydrogen bonding is feasible.
(7) (a) Carlsen, P. H. J .; Katsuki, T.; Martin, V. S.; Sharpless, K. B.
J . Org. Chem. 1981, 46, 3936. (b) Thanks to Professor B. Zwanenburg,
Department of Organic Chemistry, University of Nijmegen, The
Netherlands, for providing the exact experimental details of the
synthesis of optically pure (3S,2R)-3-phenylglycidic acid using RuCl3‚
H2O catalyst.
(8) For â-turn see: (a) Ball, J . B.; Hughes, R. A.; Alewood, P. L.;
Andrews, P. R. Tetrahedron 1993, 49, 3467 and references therein.
(b) Haubner, R.; Finsinger, D.; Kessler, H. Angew. Chem., Int. Ed. Engl.
1997, 36, 1374. (c) Kim, K.; Germanas, J . P. J . Org. Chem. 1997, 62,
2853. (d) J ones, I. G.; J ones, W.; North, M. J . Org. Chem. 1998, 63,
1505. (e) Krauthauser, S.; Christianson, L. A.; Powell, D. R.; Gellman,
S. H. J . Am. Chem. Soc. 1997, 119, 11719.
1682 J . Org. Chem., Vol. 68, No. 5, 2003

N-Cinnamoly L-Proline Derived Peptides
TABLE 4. Ch em ica l Sh ifts (p p m ), Cou p lin g Con sta n ts (Hz),a a n d Tem p er a tu r e Coefficien ts (p p b/°C) (DMSO-d 6) for
Com p ou n d 1f (Tr a n s)
protons
Pro
Leu
allyl
NH
7.94 (d)
7.80 (t)
(J NH-RH ) 8.5)
(J NH-RH ) 5.6)
CRH
4.36 (dd)
(J R-â1 ) 4.0, J R-â2 ) 8.5)
2.13 (m)
4.26 (m)
3.69 (m)
(J R-â(pro S) ) 5.0, J R-â(pro R) ) 9.8)
Câ1H
Câ2H
1.5-1.6 (m)b
5.78 (m)
2.10 (m)
(J â-γ-cis )10.5, J â-γ-trans ) 17.2,
J R1-â ) 5.0, J R2-â ) 5.0)
5.13 (m)
(J R-γ ) 1.7, J γ-γ ) 1.7)
5.02 (m)
Cγ1H
1.92 (m)
1.5-1.6 (m)
Cγ2H
Cδ1H
1.83 (m)
3.84 (m)
0.87 (d)
(CH3, J γ-δ(pro R) ) 6.3)
0.82 (d)
Cδ2H
3.72 (m)
(CH3, J γ-δ(pro S) ) 5.9)
others: 7.47 (d, J ) 15.5, cinnamoyl âH), 7.02 (d, J ) 15.5, cinnamoyl RH), 7.46-7.72 (phenyl)
∆δ/∆T (ppb/°C) -4.5
-3.4
a
b
All J values are expressed in Hz. Individual J values for both CâH and Leu residue could not be traced.
TABLE 5. Ch em ica l Sh ifts (p p m ), Cou p lin g Con sta n ts (Hz),a a n d Tem p a r etu r e Coefficien ts (p p b/°C) (DMSO-d 6) for
Com p ou n d 1f (Cis)
protons
Pro
Leu
allyl
NH
8.33 (d)
8.11 (t)
(J NH-RH ) 8.5)
(J NH-RH ) 5.6)
CRH
4.71(dd)
(J R-â1)1.8, J R-â2 ) 8.6)
2.22 (m)
4.32 (m)
3.65 (m, 2H)
(J R-â(pro S) ) 5.2, J R-â(pro R) ) 10.2)
1.45(ddd)
(J â(pro R)-â(pro S) ) 13.3, J â(pro S)-γ ) 10.1)
Câ1H
5.74 (ddt)
(J â-γ-cis ) 10.5, J â-γ-trans ) 15.4,
J R-â ) 5.0)
Câ2H
Cγ1H
1.92 (m)
1.91 (m)
1.35 (m)
1.55 (m)
5.07 (dq, H-trans)
(J R1-γ-trans ) 1.7, J γ-cis,γ-trans ) 1.7,
J R2-γ-trans ) 1.7, J R2-γ-cis ) 1.7)
5.01 (dq, H-cis)
Cγ2H
Cδ1H
Cδ2H
1.81 (m)
3.55 (m)
3.47 (m)
(J â-γ-cis ) 11.9)
0.66 (d)
(CH3, J γ-δ(pro R) ) 6.6)
0.61 (m)
(CH3, J γ-δ(pro S) ) 6.3)
others: 7.44 (d, J ) 15.5, cinnamoyl âH), 6.74 (d, J ) 15.5, cinnamoyl RH), 7.46-7.72 (phenyl)
∆δ/∆T (ppb/ °C)
-6.1
-6.3
a
All J values are expressed in Hz.
We also observed that in both cis and trans isomers
the Leucine side chain is fairly rigid. For the cis isomer
For the olefin 1d , in addition to studying the solution
NMR of both rotamers in CDCl3 we studied the NMR in
DMSO-d6 solvent also where the spectrum shows the
presence of cis and trans isomers in equal proportion.
NOSEY/TOCSY experiments were used in assigning the
resonances to both isomers. The ∆δ/∆T for the Leucine
NH (in DMSO-d6) has a large magnitude, which rules
out the presence of intramolecular H-bonding in 1d .
Experiments were then performed in CDCl3 solution
where a majority of the population of the molecules in
1d (∼80%) are found to possess trans conformation. The
Leucine NH appeared at a low field with a δ value of 7.64
ppm, which indicates its possible participation in H-
bonding (Table 6).
Additional support for such an H-bond in 1d was
gained from solvent titration as the addition of up to 33%
DMSO (v/v) shifted the resonance signal by only 0.4 ppm
(Chart 1). The most likely acceptor would be the carbonyl
of the cinnamoyl group. A strong ProlineRH/Leucine NH
NOESY peak in addition to the above-mentioned H-
bonding implies the propensity of a γ-turn in the mol-
the coupling constants J R-â(pro ) 10.2 Hz, J ø-â(pro
)
R)
S)
5.2 Hz, J â(pro R)-γ ) 4.7 Hz, and J â(pro S)-γ ) 10.1 Hz and
strong NOE peaks between Leucine CRH-Leucine CδH-
(pro S) and Leucine NH-Leucine â (pro R) support such
conclusion. The values show that a very large fraction of
rotamers (∼70%) have an N-CR-Câ-Cγ angle about
-60°(g-) and about 70% of the rotamers about the Câ-
Cγ bond have an anti relationship between âH (pro S)
and γH in both isomers. However, observations suggest
very similar behavior of the Leucine side chain for both
the cis and trans isomers (Tables 3 and 4). The NMR
study was also performed by dissolving 1f in a compara-
tively less polar solvent (CDCl3), where it showed a
preponderance of the trans isomer to a large extent (of
∼90%). Intramolecular hydrogen bonds are generally
more stable in solvents with lower polarity. So, it can be
safely assumed that a similar kind of structural organi-
zation as in DMSO-d6 prevails in CDCl3 also for com-
pound 1f.
J . Org. Chem, Vol. 68, No. 5, 2003 1683

Nandy et al.
TABLE 6. Ch em ica l Sh ifts (p p m ), Cou p lin g Con sta n ts (Hz),a a n d Tem p er a tu r e Coefficien ts (p p b/°C) (CDCl3) for
Com p ou n d 1d (Tr a n s)
protons
Pro
Leu
allyl
NH
CRH
7.64 (d, J NH-RH ) 7.4)
4.50 (m)
4.78 (dd)
(J R-â1 ) 1.9, J R-â2 ) 8.0)
4.63 (dq)
(J R1-â ) 3.2, J R1-R2 ) 11.2, J R1-γ ) 1.6)
3.62 (dq)
(J R2-â ) 3.2, J R2-γ ) 1.6)
5.91 (m)
(J â-γ-trans ) 17.2, J â-γ-cis ) 10.5)
Câ1H
2.49 (m)
1.6-1.7 (m)
Câ2H
Cγ1H
Cγ2H
Cδ1H
Cδ2H
1.86 (m)
2.18 (m)
2.05 (m)
3.76 (m)
3.64 (m)
1.6-1.7 (m)
1.6-1.7 (m)
5.35 (m)
5.24 (m)
0.88 (d)
0.914 (d)
(CH3, J γ-δ ) 6.1)
others: 7.75 (d, J ) 15.5, cinnamoyl CâH), 6.75 (d, J ) 15.5, cinnamoyl CRH), 7.3-7.6 (phenyl)
∆δ/∆T (ppb/°C)
-4.0
a
All J values are expressed in Hz.
CHART 1. 1H NMR Titr a tion Stu d y on 1d in
CDCl3-DMSO-d 6
F IGURE 5. (a) The hindrance of one of the π-faces in 1f due
to formation of a â-turn. (b) The crossed arrow points out the
hindered face of the olefin in 1f while the uncrossed arrow
points out the unhindred face and explains the observed
stereoselectivity.
γ-turns in amide 1f as well as ester 1d . It is evident from
MD study that for olefin 1f, there is a strong indication
for the presence of a â-turn and that this turn is more
efficient in inducing selectivity toward epoxidation with
very effective hindrance of one of the π-faces of the olefin
(Figure 5). The low-energy conformer, which explains this
selectivity was chosen from the MD trajectory files in
which the hydrogen bond diststance (between Leucine
NH and cinnamoyl carbonyl) of 2.1 Å is indicative of the
presence of a â-turn around the Proline-Leucine system
(Figure 5). Again, it is clear from the diagram that the
structural organization due to the presence of this â-turn
in the peptide renders only one face of the olefin free for
the approach of the oxygen during epoxidation to give
rise to the observed diastereoelectivity (2R,3S).
The NMR studies in CDCl3 and DMSO-d6 clearly
indicate that these peptides are existing as organized
structures in solution. It is interesting to note that even
in a highly polar solvent like DMSO and considerably
polar solvent like CDCl3 the peptides 1d and 1f exhibit
the presence of γ- and â-turns, respectively. Furthermore,
we have carried out NMR studies in CD3CN, which also
happens to be the medium for these epoxidation reac-
tions. The trans:cis ratios of the Proline rotamers in CD3-
CN for both peptides (1d and 1f) are ∼9:1 which clearly
support that the majority of the peptides exist with the
trans rotamer of Proline (the NOESY spectrum of 1f in
ecule. The NOE pattern for 1d is described in Figure 4.
The observations obtained from NMR studies are con-
sistent with Molecular Dynamics (MD) calculation stud-
ies discussed subsequently.
Molecular dynamics (MD) calculations for compounds
1d and 1f were carried out with the Cerius2 program on
a Silicon graphics Indigo2 workstation. Charges were
calculated by using the charge-equilibration method and
the CFF9 force field with default parameters were used
throughout the simulations. To understand the confor-
mational freedom, Simulated Annealing Molecular Dy-
namics calculations were carried out. The temperature
was varied between 300 and 1200 K in steps of 50 K for
100/150 cycles. The molecules were allowed to equilibrate
for 0.5 ps for every change in temperature. Minimizations
were done first with steepest descent, followed by con-
jugate gradient methods for a maximum of 1000 itera-
tions each or a root-mean-sauare deviation of 0.01 kcal/
mol, whichever was earlier. The energy-minimized
structures were then subjected to MD simulations. Vari-
ous conformers obtained in each MD run were minimized
by using the above-mentioned minimization protocol. The
majority of peptides having trans conformations at the
Proline residue in CDCl3 indicate the propensity of â- or
1684 J . Org. Chem., Vol. 68, No. 5, 2003

N-Cinnamoly L-Proline Derived Peptides
CHART 2. NOESY Sp ectr a of 1f in CD3CN
F IGURE 6. The structural organization in 1d and 1f that
explains the observed diastereoselectivity.
SCHEME 3
F IGURE 7. Hindrance of one of the π-faces of the cinnamoyl
bond in 6 and 7 due to the formation of a â-turn.
â-turn conformations (Figure 6B) and will render such
epoxidations facially selective.
That the high enantioselectivity is dictated by the
â-turn is again seen during the epoxidation of N-cin-
namoyl peptides 6 and 7, which are constrained to have
an intramolecular ten-membered hydrogen bond (Scheme
3). The constraint present due to aziridine and dehydro-
phenylalanine residues in 6 and 7, respectively, is
responsible for coaxing the carbonyl group of the cin-
namoyl and amino group of the Leucine residue to adopt
a geometry, which encourages the intramolecular hydro-
gen bond, whose presence is also established by NMR
titrations.8 This intramolecular hydrogen bonding pre-
organizes9 the cinnamoyl group to undergo facially selec-
tive aerobic epoxidation under catalysis of polyaniline-
supported cobalt salen to give predominantly diastereo-
mers 6a and 7a , respectively, with a de >90, as indicated
by HPLC. The absolute configuration for 6a and 7a is
assigned as 2R,3S by analogy with epoxides 2f-h .
The selected low-energy conformers of 6 and 7 from
the MD trajectory files are shown in Figure 7. It is
evident from these figures that both structures are
preorganized as â-turns due to the presence of intramo-
lecular hydrogen bonds.
The hydrogen bond distances in these structures are
typical (2.3-2.0 Å). The structures clearly show that one
of the π-faces of the cinnamoyl double bond in both cases
of aziridine peptide 6 and dehydrophenylalanine derived
tripeptide 7 is hindered by the ester part and the side
chain of the Leucine residue, respectively. The diaste-
reoselectivity of the epoxidations is expected to be
achieved from the less hindered side of the cinnamoyl
double bond. The NMR, MD studies clearly vindicate the
diastereoselectivity observed during epoxidation of these
tripeptides.
CD3CN is shown in Chart 2). The presence of such highly
organized structures in acetonitrile medium suggests
that the transition state geometries of these molecules
during epoxidation may be controlled by the formation
of intramolecular hydrogen bonds leading to organized
γ- or â-turns.
Hence, the 2D solution NMR of the olefins (1d and 1f)
and their MD simulation studies clearly suggest that the
epoxidations may be occurring on preorganized structures
formed due to the presence of γ- or â-turns. Thus, the
high facial selectivity in epoxidation of 1c-e can be
explained by invoking the possibility of these reactions
taking place on preorganized structures resulting from
the formation of γ- and â-turns due to intramolecular
H-bonding. The tripeptides 1c-e can adopt a seven-
membered γ-turn thereby forcing the cinnamoyl group
to exist with trans, s-cis geometry (Figure 6, A) as the
corresponding s-trans geometry will be disfavored due
to nonbonding interactions. The seven-membered γ-turn
will render only one face of the cinnamoyl double bond
exposed and thus epoxidation via such organized struc-
tures will be facially selective. A similar explanation can
be offered for the high enantioselectivity during epoxi-
dation of 1f-h , which would take place on well-organized
Interestingly, optically pure aziridine peptide 8, which
also does not have features to preorganize itself forming
(9) Prabhakaran, E. N.; Nandy, J . P.; Shukla, S.; Tewari, A.; Das,
S. K.; Iqbal, J . Tetrahedron Lett. 2002, 43, 6461.
J . Org. Chem, Vol. 68, No. 5, 2003 1685

Nandy et al.
SCHEME 4
In conclusion, we have developed a novel diastereose-
lective aerobic epoxidation of the cinnamoyl peptides,
catalyzed by polyaniline-supported cobalt salen. The
origin of this diastereoselectivity has been attributed to
the propensity of the N-cinnamoyl Proline amide to exist
predominantly as the trans rotamer in CH3CN medium.
Existence of these peptides as organized structures (γ-
and â-turns) is due to the presence of intramolecular
hydrogen bonds. By extensive solution NMR and MD
simulation studies we have concluded that the origin of
the high facial selectivity is due to the well-defined γ-
and â-turns which result in the hindrance of one face of
the cinnamoyl double bond in the transition state of the
epoxidation reaction. The proposal is finally supported
by carrying out a highly diastereoselective epoxidation
on peptides, which are constrained to form intramolecu-
lar hydrogen bonding leading to a â-turn.
SCHEME 5
Exp er im en ta l Section
Ma ter ia ls a n d Meth od s. Acetonitrile, ethyl acetate, hex-
ane, THF, and all other solvents were purified by standard
procedures. All the amino acids were purchased from Spec-
trochem, India Limited and used as such. Polyaniine supported
Co(salen) was prepared according to the procedure developed
in our laboratory. Column chromatography was performed on
ACME silica gel eluant. TLC was performed on ACME silica-G
coated glass plates that were irradiated with iodine or Kie-
selgel 60 GF254 coated glass plates that were irradiated with
a UV lamp. 1H NMR spectra were recorded using J EOL PMX-
60, Bruker WP-80, J EOL 300 FT NMR, J NM-LA 400 FTNMR
Varian Gemini 2000 NMR, or J EOL 200 FT NMR machines
in CDCl3 or in DMSO-d6. Chemical shifts are given relative
to TMS in ppm (δ). Multiplicity is indicated by the following
abbreviations: singlet (s); br s (broad singlet); doublet (d); dd
(doublet of a doublet); dt (doublet of a triplet); td (triplet of a
doublet); ddd (doublet of a doublet of a doublet); q (quartet);
and m (multiplet). The FAB mass spectra were recorded on a
J EOL SX 102/DA 6000 mass spectrometer data system using
argon (6 kV, 10 mA) as the FAB gas and on Hewlett-Packard,
5989A mass spectrometer and PE SICEX API 300 LC-MS
machines with isobutane as the gas. IR spectra were recorded
on a Perkin-Elmer 683 spectrophotometer and a Perkin-Elmer
1650 spectrophotometer using either a neat sample or a
solution in CDCl3/CH2Cl2 and solids were examined as KBr
pellets and the values are reported in νmax (cm-1). HPLC
analyses were done with a Rainin system fitted with a
Dynamax SD-200 pump and detected with a Groton PDA
solonet Diode Array Detector.
P r ep a r a tion of P olya n ilin e. Freshly distilled aniline (10
mL, 109.5 mmol) was dissolved in 125 mL of 1.5 M HCl and a
solution of ammoniumpersulfate (54.8 mmol) in 1.5 M HCl (125
mL) was added to it at 0 °C. Since aniline polymerization is
strongly exothermic, the oxidant was added slowly over a
period of 1 h. After the addition of the oxidant was over the
reaction mixture was stirred for an additional 4 h. The
polyaniline hydrochloride precipitated and was separated by
filtration and rinsed consecutively with water (3 × 30 mL),
methanol (2 × 25 mL), and diethyl ether (2 × 15 mL) to remove
the oligomers and any possible side products. The polymer was
then vacuum-dried until constant mass. Deprotonation of
polyaniline hydrochloride was achieved with aqueous ammonia
(3 wt %). Deprotonated polymer was again washed with water,
methanol, and diethyl ether and dried until constant mass (∼3
g). Polyaniline is quiet stable in air and can be stored
indefinitely in closed glass vials.
SCHEME 6
a turn, also undergoes a nondiastereoselective epoxida-
tion at its cinnamoyl end under the same reaction
conditions giving an ∼1:1 mixture of diastereomeric
epoxides 8a (Scheme 4).
It is noteworthy, though not particulary surprising,
that PASCOS catalyzed diastereoselectivity is very spe-
cific to the presence of Proline residue in the i+1 position
as any other amino acid in this place does not exercise
any control on the facial selectivity. This is amply clear
from the epoxidation of peptides 1k , 1l, and 1m (Proline
in the i+2 position), which showed no diastereoselectivity
during PASCOS catalyzed oxidation, as a diastereomeric
mixture of the corresponding epoxides 2k , 2l, and 2m ,
respectively, was obtained in good yields (Scheme 5). The
1H NMR studies on 1k , 1l, and 1m revealed the absence
of intramolecular hydrogen bonds in these peptides.
It is also interesting to note here that the chirality of
the Proline residue controls the diastereoselectivity in 1
as replacing the L-Proline in 1f with D-Proline gives rise
to 9 whose epoxidation under polyaniline catalyzed
reaction leads to the corresponding epoxide 9a . The sign
and magnitude of optical rotation ([R]D +121; CH2Cl2) of
9a indicates that the epoxide stereochemistry in it is
opposite to that present in 1f (Scheme 6). This result
further confirms the role of γ- or â-turn in structural
preorganization of the N-cinnamoyl-L-proline derived
peptides during the epoxidation reaction.
P r epar ation of P olyan ilin e-Su ppor ted Cobalt(II) Salen
(P ASCOS). Cobalt(II) salen (200 mg) and polyaniline (200 mg)
were added to a solution of acetic acid (25 mL) in acetonitrile
(25 mL) and the mixture was stirred at ambient temperature
1686 J . Org. Chem., Vol. 68, No. 5, 2003

N-Cinnamoly L-Proline Derived Peptides
for 36 h. The resultant catalyst was filtered off and washed
first with acetic acid (3 × 10 mL) and then thoroughly with
acetonitrile until the filterate was colorless. The resulting
residue was dried in an air oven at 110 °C for 2 h to afford the
black (or blakish brown) catalyst (325 mg). The presence of
cobalt(II) species in this catalyst was confirmed by UV and
EPR spectroscopy. PASCOS is stable to atmosphere and can
be stored indefinitely in closed vials.
yielded the crude product, which was further purified by
column chromatography (EtOAc: Hexane ) 1:1.5) to yield
methyl-N-cinnamoyl-(L)-proline-(L)-leucinate 1c as a solid (mp
110-11 °C) in good yield (80%); [R]25D -170.5° (c ) 0.01, CH2-
Cl2).
1H NMR (400 MHz, CDCl3): δ 7.76 (d, J ) 15.3 Hz, 1H),
7.65 (d, J ) 7.32 Hz, 1H), 7.56-7.53 (m, 2H), 7.39-7.35 (m,
3H), 6.76 (d, J ) 15.6 Hz, 1H), 4.78 (d, J ) 6.84 Hz, 1H); 4.57-
4.47 (m, 2H); 3.73 (s, 3H); 3.71-3.59 (m, 2H); 2.51-2.46 (m,
1H); 2.06-2.03 (m, 1H); 2.21-2.12 (m, 1H); 1.89-1.81 (m, 3H);
1.67-1.56 (m, 2H); 0.91 (d, J ) 5.6 Hz, 3H); 0.88 (d, J ) 5.6
Hz, 3H). FT-IR (CH2Cl2): 3278, 3059, 2956.5, 2872.7, 1744.8,
Syn th esis of Meth yl-N-cin n a m oyl (L)-P r olin a te (1a ). To
a stirring, ice-cold solution of cinnamic acid (1.15 g, 10 mmol)
and triethylamine (1.4 mL, 10 mmol) in THF (15 mL) was
added isobutyl chloroformate (1.29 mL, 10 mmol) and the
mixture was stirred vigorously for 1 min, after which a solution
of the methyl-(L)-prolinate hydrochloride (1.82 g, 11 mmol) in
DMSO (4-5 mL) was added followed by triethylamine (3.1 mL,
22 mmol) dissolved in THF (15 mL). The reaction vessel was
allowed to warm to room temperature and vigorously stirred
for an additional 3-4 h. Triethylamine hydrochloride was
filtered off on a sintered funnel under suction and washed with
THF. Removal of solvent from the filtrate in vacuo yielded a
residue, which was taken in EtOAc (20 mL) and washed with
a saturated aqueous solution of NaHCO3 (2 × 10 mL) and brine
(1 × 10 mL). The organic layer was isolated and dried over
anhydrous Na2SO4 and after evaporation of solvent in vacuo
the resulting mass was subjected to column chromatography
(EtOAc:hexane 1:2) to afford 1a as a crystalline solid (mp 58
°C) in good yields (74%).
1649.6, 1598.1, 1542.1, 1498.0, 1425.3 cm-1
.
Syn th esis of Allyl-N-cin n a m oyl-(L)-p r olin e-(L)-leu ci-
n a te (1d ). To a solution of N-cinnamoyl-(L)-proline-(L)-leucine
(1.25 g, 3.5 mmol) in acetone (20 mL) was added K2CO3 (0.351
g, 3.85 mmol) and allyl bromide (0.47 g, 3.85 mmol) and the
reaction mixture was set to reflux for 8 h, at which point
complete conversion had taken place. The inorganic salt was
filtered off and solvent was removed in vacuo. The resulting
residue was subjected to purification by column chromatog-
raphy (Silica gel-EtOAc:hexane 40:60) to afford 1d in good
yield (84%).
1H NMR data (500 MHz) for compound 1d in DMSO-d6 are
presented in Table 5 for the trans rotamer of the proline
residue.
Syn th esis of Meth yl-4-N-cin n a m oyl- (L)-p r olin e-(L)-
leu cin yl)-cr oton a te (1e). To a solution of N-cinnamoyl-(L)-
proline-(L)-leucine (0.54 g, 1.5 mmol) in acetone (7.5 mL) was
added K2CO3 (0.23 g, 1.65 mmol) and methyl-4-bromocrotonate
(0.27 g, 1.5 mmol) and the reaction mixture was set to reflux
for 8 h during which time the reaction is almost complete. The
inorganic salts were filtered off on a sintered funnel under
suction and solvent was removed in vacuo. The resulting
residue was taken up in EtOAc (20 mL) and washed with a
saturated aqueous solution of NaHCO3 (2 × 10 mL), water (2
× 10 mL), and brine (1 × 10 mL). Drying the organic layer
(anhyd. Na2SO4) and concentration in vacuo yielded a thick
residue that was further purified by column chromatography
(EtOAc in hexane, 40%) (TLC: Rf ) 0.5; hexane:ethyl acetate
1:1) to yield methyl-4-(N-cinnamoyl-(L)-proline-(L)-leucinyl)-
crotonate 1e as a white solid (mp 92 °C) in good yield (68%).
1H NMR (400 MHz, CDCl3): δ 7.79 (d, J ) 7.3 Hz, 1H); 7.75
(d, J ) 15.5 Hz, 1H); 7.56-7.53 (m, 2H); 7.41-7.39 (m, 3H);
6.94 (td, J ) 15.9, 4.64 Hz, 2H); 6.78 (d, J ) 15.6 Hz, 1H);
6.08 (td, J ) 15.9, 1.9 Hz, 1H); 4.79 (dd, J ) 4.9, 1.9 Hz, 2H);
4.64 (d, J ) 3.2 Hz, 1H); 4.51 (dd, J ) 12.9, 5.6 Hz, 1H); 3.79-
3.74 (m, 1H); 3.75 (s, 3H); 3.68-3.61 (m, 1H); 2.53-2.48 (m,
1H); 2.29-2.28 (m, 1H); 2.18-2.02 (m, 1H); 1.91-1.81 (m, 1H);
1.70-1.62 (m, 3H); 0.93 (d, J ) 5.8 Hz, 3H); 0.89 (d, J ) 5.8
Hz, 3H).
1H NMR (CDCl3, 60 MHz): δ 7.55 (d, J ) 14.6 Hz, 1H); 7.2-
7.11 (m, 5H); 6.6 (d, J ) 14.8 Hz, 1H); 4.6-4.5 (m, 1H); 3.68
(br s, 3H); 3.52-3.41 (m, 2H); 2.2-1.8 (m, 4H).
Syn th esis of Allyl-N-cin n a m oyl (L)-P r olin e Am id e (1b).
To a stirring ice cold solution of N-cinnamoyl-proline (2.45 g,
10 mmol) and triethylamine (1.4 mL, 10 mmol) in THF (15
mL) was added isobutylchloroformate (1.29 mL, 10 mmol) and
the mixture was stirred vigorously for 50-60 s, after which a
solution of allyl amide (1.2 mL, 15 mmol) in THF (10 mL) was
added followed by triethylamine (2.1 mL, 15 mmol) dissolved
in THF (10 mL). The reaction vessel was allowed to warm to
room temperature and vigorously stirred for further 3-4 h.
Triethylamine hydrochloride was filtered off on a sintered
funnel under suction and solvent was removed in vacuo. The
resulting residue was taken in EtOAc (40 mL) and washed
with a saturated aqueous solution of NaHCO3 (2 × 15 mL),
water (2 × 10 mL), and brine (1 × 10 mL). Drying the organic
layer (anhyd. Na2SO4) and concentration in vacuo yielded a
thick residue that was further purified by column chromatog-
raphy (Silicagel-EtOAc:hexane 1:1.5) to yield allyl-N-cin-
namoyl-proline amide 1b in good yield (81%).
1H NMR (CDCl3, 400 MHz): δ 7.75 (d, J ) 15.4 Hz, 1H);
7.56-7.53 (m, 2H); 7.49 (br s, 0.8H); 7.40-7.38 (m, 3H); 6.76
(d, J ) 15.4 Hz, 1H); 6.27 (m, 0.2H); 5.88-5.79 (m, 1H); 5.18
(dd, J ) 17.9, 1.2 Hz, 1H); 5.11 (d, J ) 8.56 Hz, 1H); 4.76 (d,
J ) 7 Hz, 1H); 3.93-3.84 (m, 2H); 3.81-3.74 (m, 1H); 3.66-
3.61 (m, 1H); 2.54-2.51 (m, 1H); 2.18 (dt, J ) 18.3, 9.3 Hz,
1H); 2.08-2.04 (m, 1H); 1.91-1.81 (m, 1H).
Syn th esis of Meth yl-N-cin n a m oyl-(L)-p r olin e-(L)-leu ci-
n a te (1c). A stirring solution of N-cinnamoyl-(L)-proline (2.45
g, 10 mmol) and triethylamine (1.4 mL, 10 mmol) in THF (15
mL) was cooled to -10 °C in an ice-salt bath and to it was
added isobutylchloroformate (1.29 mL, 10 mmol) with vigorous
stirring for 50-60 s. Then a solution of methyl-(L)-leucinate
hydrochloride (2.00 g, 11 mmol) in DMSO (0.5 mL) was added
followed by a solution of triethylamine (3.1 mL, 22 mmol) in
THF (15 mL). The mixture was warmed to room temperature
by removal of the ice bath and vigorously stirred for further 4
h. Triethylamine hydrochloride was filtered off on a sintered
funnel under suction and washed twice with THF. Removal
of solvent from the filtrate in vacuo yielded a residue, which
was dissolved in EtOAc (30 mL) and washed with a saturated
aqueous solution of NaHCO3 (2 × 10 mL) and brine (1 × 10
mL). Drying (Na2SO4) and evaporation of solvent in vacuo
Syn th esis of Allyl-N-cin n a m oyl-(L)-p r olin e-(L)-leu cin e
Am id e (1f). To a stirring ice-cold solution of N-cinnamoyl-
(L)-proline-(L)-leucine (1.79 g, 5 mmol) and triethylamine (0.7
mL, 5 mmol) in THF (10 mL) was added isobutyl chloroformate
(0.65 mL, 5 mmol) and the mixture was stirred vigorously for
50-60 s, after which a solution of allylamine (0.6 mL, 7.5
mmol) in THF (10 mL) was added followed by triethylamine
(1.1 mL, 8 mmol) dissolved in THF (10 mL). The reaction
mixture was allowed to come to room temperature and further
stirred for 3-4 h. Inorganic salt was filtered off followed by
removal of the solvent in vacuo. The residue then was taken
up in EtOAc (35 mL) and washed with a saturated solution of
NaHCO3 (2 × 10 mL) and brine (1 × 10 mL). The organic layer
was separated and concentrated in vacuo to a residue, which
was dried over Na2SO4 and subjected to purification by column
chromatography (Silicagel-EtOAc-hexane 2:3) to afford allyl-
N-cinnamoyl-(L)-proline-(L)-leucine amide 1f as a crystalline
solid (mp 123-24 °C) in excellent yield (88%).
1H NMR data (500Mz) for compound 1f in DMSO-d6 are
presented in Table 3 and Table 4 for trans and cis rotamers of
proline residue, respectively.
J . Org. Chem, Vol. 68, No. 5, 2003 1687

Nandy et al.
Syn th esis of Allyl-N-cin n a m oyl-(L)-p r olin e-(L)-leu cin e-
glycin a te (1g). To a solution of N-cinnamoyl-(L)-proline-(L)-
leucine-glycine (0.83 g, 2 mmol) in acetone (15 mL) was added
K2CO3 (0.455 g, 3.3 mmol) and allyl bromide (0.268 g, 2.2
mmol) and the reaction mixture was set to reflux for 8 h, at
which point complete conversion had taken place. The inor-
ganic salt was filtered off and solvent was removed in vacuo.
The resulting residue was subjected to purification by column
chromatography (Silica gel-EtOAc:hexane 45:55) to afford 1g
in good yield (78%).
1H NMR (400 MHz, CDCl3): δ 7.72 (d, J ) 15.4 Hz, 1H);
7.55-7.53 (m, 2H); 7.39-7.27 (m, 3H); 6.84 (d, J ) 8.6 Hz,
1H); 6.75 (d, J ) 15.4 Hz, 1H); 5.90 (ddd, J ) 22.9, 11.7, 5.8
Hz, 1H); 5.33 (dd, J ) 17.5, 1.4 Hz, 1H); 5.24 (dt, J ) 10.5, 1.2
Hz, 1H); 4.74 (d, J ) 6.1 Hz, 1H); 4.63 (d, J ) 5.6 Hz, 2H);
4.48-4.45 (m, 1H); 4.14 (dd, J ) 8.1,, 5.5 Hz, 1H); 3.97 (dd, J
) 18.1, 5.5 Hz, 1H); 3.81-3.79 (m, 1H); 3.70 (dd, J ) 8.8, 7
Hz, 1H); 2.37-2.24 (m, 1H); 2.14-1.99 (m, 3H); 1.83-1.81 (m,
1H); 1.59-1.52 (m, 2H); 0.89 (d, J ) 6.1 Hz, 3H); 0.86 (d, J )
6.1 Hz, 3H).
(1b) (0.426 g, 1.5 mmol) in CH3CN was added 2-methylpro-
panal (0.272 mL, 3 mmol) and PASCOS (∼0.005 g) and the
mixture was stirred under oxygen atmosphere at ambient
temperature for 12 h. After this time, a fresh sample of the
catalyst and 2-methylpropanal (0.272 mL, 3 mmol) were added
to the reaction mixture and allowed to stir, until (22 h)
complete conversion of the olefin (TLC: Rf ) 0.5; hexane:ethyl
acetate 3:2). The catalyst was filtered off on a sintered funnel
and acetonitrile was removed in vacuo. The resulting residue
was taken up in EtOAc (30 mL) and washed successively with
a saturated solution of NaHCO3 (2 × 10 mL), water (2 × 10
mL), and brine (1 × 10 mL). Separating the organic phase,
drying (Na2SO4), and concentration in vacuo yielded the
corresponding oxirane containing peptide in high purity and
yield (HPLC). This was further subjected to purification by
column chromatography (silica gel-EtOAc:hexane 1:1.5) to get
the pure epoxide 2b (40%) as a gum. [R]25D -180° (c 0.01, CH2-
Cl2).
1H NMR (CDCl3, 400 MHz): δ 7.97 (d, J ) 7.32 Hz, 0.2H);
7.34-7.36 (m, 2H); 7.32-7.29 (m, 3H); 7.14 (m, 0.8H); 5.85-
5.8 (m, 1H); 5.18 (ddd, J ) 15.4, 3.4, 1.8 Hz, 1H); 5.11 (ddd, J
) 10.2, 4.16, 1.48 Hz, 1H); 4.67-4.65 (m, 1H); 4.1 (d, J ) 1.92
Hz, 0.3H); 4.09 (d, J ) 2.04 Hz, 0.7 Hz); 3.90-3.78 (m, 2H);
3.69-3.66 (m, 1H); 3.63 (J ) 1.8 Hz, 0.8H); 3.61 (d, J ) 1.9
Hz, 0.2H); 3.54-3.50 (m, 1H); 2.47-2.4 (m, 1H), 2.22-2.12 (m,
1H); 2.04-1.97 (m, 1H); 1.95-1.84 (m, 1H). MS (m/z): 301
(M+), 285, 243, 27, 216, 199, 154.
Syn th esis of Allyl-N-cin n a m oy-(L)-p r olin e-(L)-leu cin e-
(L)-isoleu cin a te (1h ). To a solution of N-cinnamoyl-(L)-
proline-(L)-leucine-(L)-isoleucine (1.41 g, 3 mmol) in acetone
(15 mL) was added K2CO3 (0.455 g, 3.3 mmol) and allyl
bromide (0.403 g, 3.3 mmol) and the reaction mixture was set
to reflux for 8 h during which time the reaction was almost
complete. The inorganic salts were filtered off on a sintered
funnel under suction and solvent was removed in vacuo. The
resulting residue was taken up in EtOAc (20 mL) and washed
with a saturated aqueous solution of NaHCO3 (2 × 10 mL),
water (2 × 10 mL) ,and brine (1 × 10 mL). Drying the organic
layer (anhyd. Na2SO4) and concentration in vacuo yielded a
thick residue that was further purified by column chromatog-
raphy (EtOAc in hexane, 50%) (TLC: Rf ) 0.5; hexane:ethyl
acetate 1:1) to yield methyl-4-(N-cinnamoyl-(L)-proline-(L)-
leucinyl)-crotonate 1h as a white solid in good yield (68%).
Syn th esis of Meth yl-N-(3-p h en ylglycid yl)-(L)-p r olin e-
(L)-leu cin a te (2c). To a solution of the methyl-N-cinnamoyl-
(L)-proline-(L)-leucinate (1c) (0.744 g, 2 mmol) in CH3CN (10
mL) was added 2-methylpropanal (0.363 mL, 4 mmol) and
PASCOS (∼0.005 g) and the mixture was stirred under oxygen
atmosphere at room temperature for 12 h. After this time, a
fresh sample of the catalyst and 2-methylpropanal (0.363 mL,
4 mmol) were added to the reaction mixture which was allowed
to stir until complete (21 h) conversion of the olefin (TLC: Rf
) 0.5; hexane:ethyl acetate 1:1). The catalyst was filtered off
on a sintered funnel and acetonitrile was removed in vacuo.
The resulting residue was taken up in EtOAc (20 mL) and
washed successively with a saturated solution of NaHCO3 (2
× 10 mL), water (2 × 10 mL), and brine (1 × 10 mL).
Separating the organic phase, drying, and concentration in
vacuo yielded the corresponding oxirane containing peptide
in high purity and yield (HPLC). This was further subjected
to column chromatography (silica gel; EtOAc:hexane 2:3)
purification to afford the pure epoxide 2c (93%; HPLC) as a
[R]25 -171° (c 0.0075, CH2Cl2).
D
1H NMR (400 MHz, CDCl3): δ 7.72 (d, J ) 15.4 Hz, 1H);
7.56-7.49 (m, 2H); 7.41-7.35 (m, 3H); 6.84 (d, J ) 8.56 Hz,
1H); 6.74 (d, J ) 15.6 Hz, 1H); 5.96-5.85 (m, 1H); 5.33 (td, J
) 17.3, 1.5 Hz, 1H); 5.23 (td, J ) 9.3, 1.2 Hz, 1H); 4.79-4.74
(m, 1H); 4.64-4.49 (m, 3H); 4.39-4.34 (m, 1H); 3.79-3.76 (m,
1H); 368-3.63 (m, 1H); 2.46-2.41 (m, 1H); 2.17-1.74 (m, 5H);
1.64-1.42 (m, 4H); 0.97-0.80 (m, 12H).
Syn th esis of Meth yl-N-(3-p h en ylglycid yl)-(L)-p r olin a te
(2a ). To a solution of methyl-N-cinnamoyl-(L)-prolinate (1a )
(0.52 g, 2 mmol) in CH3CN (10 mL) was added 2-methylpro-
panal (0.365 mL, 4 mmol) and PASCOS (∼0.005 g) and the
mixture were stirred under oxygen atmosphere at room
temperature for 12 h. After this time, a fresh sample of the
catalyst and 2-methylpropanal (0.365 mL, 4 mmol) were added
to the reaction mixture and allowed to stir, until complete
conversion of the olefin (TLC: Rf ) 0.5; hexane:ethyl acetate
1:2). The catalyst was filtered off on a sintered funnel and
acetonitrile was removed in vacuo. The resulting residue was
taken in EtOAc (25 mL) and washed successively with a
saturated solution of NaHCO3 (2 × 10 mL), water (2 × 10 mL),
and brine (1 × 10 mL). Separating the organic phase, drying
(Na2SO4), and concentration in vacuo yielded the correspond-
ing oxirane containing peptide in high purity and yield
(HPLC). This was further subjected to purification by column
chromatography (silica gel; EtOAc:hexane 2:3) to afford the
solid (45%; mp 89-90 °C). [R]25 -183° (c 0.01, CH2Cl2).
D
1H NMR (400 MHz, CDCl3): δ 7.34-7.20 (m, 5H); 7.12 (d,
J ) 7.32 Hz, 1H); 4.61-4.57 (m, 1H); 4.44-4.38 (m, 1H); 4.12
(br s, 0.2H); 4.09 (d, J ) 1.9 Hz, 0.8H); 3.78-3.71 (m, 1H);
3.73 (s, 3H); 3.60 (d, J ) 1.9 Hz, 1H); 3.56-3.47 (m, 1H); 2.42-
2.38 (m, 1H); 2.2-2.14 (m, 1H); 2.08-1.84 (m, 5H); 0.95 (d, J
) 5.6 Hz, 3H); 0.92 (d, J ) 5.6 Hz, 3H). MS (m/z): 389 (M+),
339, 307, 281, 269, 244, 216, 209, 181, 154, 136. IR νmax: 3200
(br), 3030, 2910, 2880, 1760, 1655 cm-1
Syn th esis of Allyl-N-(3-p h en ylglycid yl)-(L)-p r olin e-(L)-
leu cin a te (2d ). To a solution containing allyl-N-cinnamoyl-
(L)-proline-(L)-leucinate (1d ) (0.796 g, 2 mmol) in CH3CN (10
mL) was added 2-methylpropanal (0.363 mL, 4 mmol) and
PASCOS (∼0.005 g) and the mixture was stirred under oxygen
atmosphere at room temperature for 12 h. After that a fresh
sample of catalyst and 2-methylpropanal (0.363 mL, 4 mmol)
were added to the reaction mixture and stirring was continued
until complete conversion of the olefine to epoxide, the catalyst
was filtered off, and solvent was removed. The residue then
was taken up in EtOAc (35 mL) and washed with a saturated
solution of NaHCO3 (2 × 10 mL) and brine (1 × 10 mL). The
organic layer was separated and concentrated in vacuo to a
residue, which was dried over Na2SO4 and subjected to
purification by column chromatography (silica gel-EtOAc-
hexane 2:3) to isolate the major diastereomer of allyl-N-(3-
pure epoxide 2a (94%; HPLC) as a gum in 40% yield. [R]25
-201° (c 0.01, CH2Cl2).
D
1H NMR (400 MHz, CDCl3): δ 7.36-7.2 (m, 5H); 4.60 (dd,
J ) 8.32, 4 Hz, 0.3H); 4.55 (dd, J ) 8.32, 4 Hz, 0.7H); 4.14
(d,J ) 1.92 Hz, 0.1 H); 4.07 (d, J ) 1.88 Hz, 0.7H); 3.98 (d, J
) 1.88 Hz, 0.2H); 3.74 (s, 3H); 3.71-3.61 (m, 2H); 3.59 (d, J )
1.9 Hz, 1H); 2.29-1.90 (m, 4H). IR (Neat) νmax 3440-3280 (br),
1790, 1725, 1660, 1520 cm-1
.
Syn t h esis of Allyl-N-(3-p h en ylglycid yl)-(L)-p r olin e
Am id e (2b). To a solution of allyl-N-cinnamoyl-proline amide
1688 J . Org. Chem., Vol. 68, No. 5, 2003

N-Cinnamoly L-Proline Derived Peptides
phenylglycilyl)-(L)-proline-(L)-leucinate 2d (40%). [R]25D -161°
(c 0.01, CH2Cl2).
for 3 h until the disappearance of starting material. The
solvent was removed in vacuo and dichloromethane (10 mL)
was added to it followed by a small amount of water (until
the phase separation occurred). The pH of the aqueous layer
was adjusted to 5 and the aqueous layer was extracted with
dichloromethane. Acidification and extraction were continued
until the pH remained constant. The combined extract was
dried (Na2SO4) and taken up into a clean dry flask. Triethyl-
amine (0.2 mL, 1.5 mmol) was added to the reaction vessel
and the reaction vessel was cooled to 0 °C. Isobutylchlorofor-
mate (0.14 mL, 1 mmol) was added and stirring was continued
for 0.5 min. Allylamine (75 µL, 1 mmol) was added to the
solution followed by addition of triethylamine (70 µL, 0.5
mmol) and the mixture was stirred vigorously for 3-4 h.
Removal of solvent under vacuum yielded a residue, which was
taken up in ethyl acetate (20 mL) and washed with a saturated
aqueous solution of NaHCO3 (1 × 10 mL), water (1 × 10 mL),
and brine (1 × 10 mL) sequentially. The resulting organic layer
was dried and concentrated to give a residue, which was
subjected to column chromatography to give the pure com-
pound that is chemically and optically identical to 2f synthe-
sized by aerobic epoxidation of 1f.
Syn th esis of Allyl-N-(3-p h en ylglycid yl)-(L)-p r olin e-(L)-
leu cin e-glycin a te (2g). To a solution containing allyl-N-
cinnamoyl-(L)-proline-(L)-leucine-glycinate (1g) (0.546 g, 1.2
mmol) in CH3CN (10 mL) was added isobutyraldehyde (0.219
mL, 2.4 mmol) and PASCOS (∼0.005 g) and mixture was
stirred under oxygen atmosphere at room temperature for 12
h. After that a fresh sample of catalyst and isobyraldehyde
(0.219 mL, 2.4 mmol) was added to the reaction mixture and
stirring was continued until complete conversion of the olefin
to epoxide, the catalyst was filtered off, and solvent was
removed. The residue then was taken up in EtOAc (35 mL)
and washed with a saturated solution of NaHCO3 (2 × 10 mL)
and brine (1 × 10 mL). The organic layer was separated and
concentrated in vacuo to a residue, which was dried over Na2-
SO4 and subjected to purification by column chromatography
(silica gel-EtOAc-hexane 2:3) to isolate allyl-N-(3-phenylgly-
cilyl)-(L)-proline-(L)-leucine-glycinate 2g in good yield (61%).
[R]D -130° (c 0.01, CH2Cl2).
1H NMR (80 MHz, CDCl3): δ 7.45-7,20 (m, 6H); 5.85-5.80
(m, 1H); 5.30 (d, J ) 16 Hz, 1H); 5.10 (d, J ) 10 Hz, 1H); 4.9-
4.8 (m, 1H); 4.65-4.55 (m, 3H); 4.00 (d, J ) 2 Hz, 0.8H); 3.98
(s, 0.2H); 3.75-3.5 (m, 2H); 3.55 (s, 1H); 2.2-1.8 (m, 7H); 0.95
(d, J ) 6 Hz, 6H). MS (m/z): 415 (M+), 357, 326, 295, 267,
209, 181, 149.
Syn th esis of Meth yl-4-N-(3-ph en ylglycidyl)-(L)-pr olin e-
(L)-leu cin yl-cr oton a te (2e). To a solution of methyl-4-N-
cinnamoyl-(L)-leucinyl-crotonate (1e) (0.538 g, 1.5 mmol) in
CH3CN (5 mL/mmol) was added 2-methylpropanal (0.272 mL,
3 mmol) and PASCOS (∼0.005 g) and the mixture was stirred
under oxygen atmosphere at room temperature for 12 h. After
this time, a fresh sample of the catalyst and 2-methylpropanal
(0.272 mL, 3 mmol) were added to the reaction mixture whcih
was allowed to stir until complete conversion of the olefin to
the epoxide (TLC: Rf ) 0.45; hexane:ethyl acetate 1:2.5). The
catalyst was filtered off on a sintered funnel and acetonitrile
was removed in vacuo. The resulting residue was taken up in
EtOAc (20 mL) and washed successively with a standard
solution of NaHCO3 (2 × 10 mL), water (2 × 10 mL), and brine
(1 × 10 mL). Separating the organic phase, drying, and
concentration in vacuo yielded the corresponding oxirane
containing peptide as a crude residue. This crude residue was
further subjected to column chromatography (silica gel; EtOAc:
hexane 1:2) for purification to get the pure epoxide 2e (91%;
HPLC) in good yield (50%) as a gum. [R]D -108° (c 0.01, CH2-
Cl2).
1H NMR (400 MHz, CDCl3): δ 7.39 (m, 0.5H); 7.32-7.29
(m, 2H); 7.25-7.20 (m, 3H); 6.86 (td, J ) 15.6, 4.6 Hz, 1H);
6.5 (d, J ) 8.56 Hz, 0.5H); 5.98 (dd, J ) 15.9, 2 Hz, 1H); 4.72
(dd, J ) 4.6, 1.96 Hz, 2H); 4.61 (d, J ) 8 Hz, 1H); 4.48-4.44
(m, 1H); 3.99 (d, J ) 1.7 Hz, 0.8H); 3.96 (s, 0.2H); 3.69 (s, 3H);
3.67-3.59 (m, 2H); 3.54 (d, J ) 1.7 Hz, 1H); 2.35-2.29 (m,
1H); 2.17-2.07 (m, 1H); 1.96-1.95 (m, 1H); 1.89-1.84 (m, 1H);
1.60-1.54 (m, 3H); 0.91 (d, J ) 4.6 Hz, 3H); 0.87 (d, J ) 4.6
Hz, 3H). MS (m/z): 473 (M+), 391, 353, 325, 307, 289, 244,
216, 200, 181, 154.
Syn th esis of Allyl-N-(3-p h en ylglycid yl)-(L)-p r olin e-(L)-
leu cin e Am id e (2f). To a solution containing allyl-N-cin-
namoyl-(L)-proline-(L)-leucine amide (1f) (0.684 g, 1.5 mmol)
in CH3CN (8 mL) was added 2-methylpropanal (0.272 mL, 3
mmol) and PASCOS (∼0.005 g) and the content was stirred
under oxygen atmosphere at room temperature for 12 h. After
that a fresh sample of catalyst and 2-methylpropanal (0.272
mL, 3 mmol) were added to the reaction mixture and stirring
was continued until (36 h) complete conversion of the olefin
to epoxide, the catalyst was filtered off, and solvent was
removed. The residue then was taken up in EtOAc (35 mL)
and washed with a saturated solution of NaHCO3 (2 × 10 mL)
and brine (1 × 10 mL). The organic layer was separated and
concentrated in vacuo to a residue, which was dried over Na2-
SO4 and subjected to purification by column chromatography
(silica gel-EtOAc-hexane 2:3) to isolate allyl-N-(3-phenylgly-
cilyl)-(L)-proline-(L)-leucine amide 2f in good yields (60%). [R]D
)-124.5° (c 0.01, CH2Cl2).
1H NMR (400 MHz, CDCl3): δ 7.36-7.34 (m, 3H); 7.30-
7.28 (m, 2H); 6.96 (d, J ) 7.8 Hz, 0.5H); 6.45-6.35 (m, 0.5H);
5.85-5.76 (m, 1H); 5.16 (dd, J ) 17.8, 1.44 Hz, 1H); 5.1 (d, J
) 10.3 Hz, 1H); 4.59 (dd, J ) 8.3, 3.3 Hz, 1H); 4.38-4.32 (m,
1H); 4.10 (s, 0.1H); 4.08 (d, J ) 1.76 Hz, 0.8H); 4.06 (d, J )
1.6, 0.1 Hz); 3.86-3.76 (m, 3H); 3.72-3.65 (m, 1H); 3.59 (d, J
) 1.74 Hz, 0.9H); 3.44 (d, J ) 1.6 Hz, 0.1H); 2.22-2.14 (m,
1H); 2.08-2.02 (m, 1H); 2.00-1.94 (m, 2H); 1.70-1.65 (m, 1H);
1.59-1.48 (m, 2H); 0.87 (dd, J ) 6.1, 3.7 Hz 3H); 0.83 (d, J )
5.84 Hz, 1H).
1H NMR (400 MHz, CDCl3): δ 7.65 (d, J ) 8.8 Hz, 1H);
7.66-7.23 (m, 5H); 6.9 (br s, 1H); 5.92-5.84 (m, 1H); 5.29 (dd,
J ) 17.5, 1.5 Hz, 1H); 5.22 (dd, J ) 9.3, 1.5 Hz, 1H); 4.62-
4.65 (m, 4H); 4.15-3.96 (m, 2H); 4.08 (s, 1H); 3.78-3.69 (m,
22H); 3.60 (d, J ) 1.92 Hz, 0.8H); 3.58 (s, 0.2H); 2.28-1.74
(m, 4H); 1.66-1.51 (m, 3H); 0.95-0.85 (m, 6H).
Syn th esis of Allyl-N-(3-p h en ylglycid yl)-(L)-p r olin e-(L)-
leu cin e-glycin a te (2g) fr om 2c. To a stirring solution of
LiOH‚H2O (0.046 g, 1.1 mmol) in methanol-water (4:1 ratio,
20 mL) was added 2c and the mixture was stirred at room
temperature for 3 h until the disappearance of starting
material. The solvent was removed in vacuo and dichlo-
romethane (10 mL) was added to it followed by a small amount
of water (until the phase separation occurred). The pH of the
aqueous layer was adjusted to 5 and the aqueous layer was
extracted with dichloromethane. The combined extract was
dried (Na2SO4) and taken up into a clean dry flask. Triethyl-
amine (0.2 mL, 1.5 mmol) was added to the reaction vessel
and the reaction vessel was cooled to 0 °C. Isobutyl chlorofor-
mate (0.14 mL, 1 mmol) was added and stirring was continued
for 0.5 min. A solution of allyl-glycinate hydrochloride (0.15
g, 1 mmol) in DMSO (0.5 mL) was added to the solution
followed by addition of triethylamine (140 µL, 1 mmol) and
the mixture was stirred vigorously for 3-4 h. Removal of
solvent under vacuum yielded a residue, which was taken up
in ethyl acetate (20 mL) and washed with a saturated aqueous
solution of NaHCO3 (1 × 10 mL), water (1 × 10 mL), and brine
(1 × 10 mL) sequentially. The resulting organic layer was dried
and concentrated to give a residue, which was subjected to
column chromatography to give the pure compound that is
chemically and optically identical to 2g synthesized by aerobic
epoxidation of 1g.
Syn th esis of Allyl-N-(3-p h en ylglycid yl)-(L)-p r olin e-(L)-
leu cin e Am id e (2f) fr om 2c. To a stirring solution of LiOH‚
H2O (0.046 g, 1.1 mmol) in methanol-water (4:1 ratio, 20 mL)
was added 2c and the mixute was stirred at room temperature
J . Org. Chem, Vol. 68, No. 5, 2003 1689

Nandy et al.
Syn th esis of Allyl-N-(3-p h en ylglycid yl)-(L)-p r olin e-(L)-
leu cin e-(L)-isoleu cin a te (2h ). To a solution of allyl-N-
cinnamoyl-(L)-proline-(L)-leucine-(L)-isoleucinate (1h ) (0.760 g,
1.5 mmol) in CH3CN (7.5 mL) was added 2-methylpropanal
(0.272 mL, 3 mmol) and PASCOS (∼0.005 g) and the mixture
was stirred under oxygen atmosphere at room temperature
for 12 h. After this time, a fresh sample of the catalyst and
2-methylpropanal (0.272 mL, 3 mmol) were added to the
reaction mixture that was allowed to stir until complete
conversion of the olefine (TLC: Rf ) 0.5; hexane:ethyl acetate
3:2). The catalyst was filtered off on a sintered funnel and
acetonitrile was removed in vacuo. The resulting residue was
taken up in EtOAc (30 mL) and washed successively with a
saturated solution of NaHCO3 (2 × 10 mL), water (2 × 10 mL),
and brine (1 × 10 mL). Separating the organic phase, drying
(Na2SO4), and concentration in vacuo yielded the correspond-
ing oxirane containing peptide in high purity and yields
(HPLC). This was further subjected to column chromatography
(silica gel; EtOAc:hexane -1:1) for purification to get the pure
epoxide 2h in good yield (59%) as a gum. [R]D -86.3° (c 0.01,
CH2Cl2).
purification by flash column chromatography (silica gel-
EtOAc-hexane 40:60) to afford 2i in good yield (78%) as a
gum. [R]D -106° (c 0.002, CH2Cl2).
1H NMR (400 MHz, CDCl3): δ 7.52 (d, J ) 8.3 Hz, 1H),
7.35-7.32 (m, 5H), 4.85 (dd, J ) 8, 4.1 Hz, 1H), 4.62 (t, J )
3.6 Hz, 1H), 4.14 (d, J ) 5.7 Hz, 0.5H), 4.08 (d, J ) 5.7 Hz,
0.5H), 3.82 (dt, J ) 9.8, 4.2 Hz, 1H), 3.74 (s, 3H), 3.69 (s, 3H),
3.65-3.61 (m, 1H), 3.58 (d, J ) 5.7 Hz, 1H), 2.92 (dd, J ) 20,
15.1 Hz, 2H), 2.36-2.33 (m, 1H), 2.27-2.19 (m, 1H), 2.15-
2.10 (m, 1H), 1.90 (br s, 1H). IR νmax: 3030 (br), 2789, 1762,
1610, 1587 cm-1
Syn th esis of Allyl-N-(3-p h en ylglycid yl)-(L)-p r olin e-(L)-
p h en yla la n in e Am id e 2j. To a solution containing allyl-N-
cinnamoyl-(L)-proline-(L)-phenylalanine amide (648 mg, 1.5
mmol) and X (0.20 g, 0.38 mmol) in CH3CN (7.5 mL) was added
2-methylpropanal (216 mg, 3 mmol) and PASCOS (∼0.005 g)
followed by addition of solid anhydrous sodium acetate (0.311
g, 10 mmol) and the mixture was stirred under oxygen
atmosphere at room temperature for 17 h. After that a fresh
lot of catalyst and 2-methylpropanal (86 µL, 0.95 mmol) and
anhydrous sodium acetate (0.311 g, 10 mmol) was added to
the reaction mixture and stirring was continued until complete
conversion of the olefin to epoxide (TLC: Rf ) 0.45; EtOAc:
hexane 3:2). The catalyst and the inorganic salt were filtered
off and solvent was removed. The residue then was taken up
in EtOAc (25 mL) and washed with a saturated solution of
NaHCO3 (2 × 5 mL) and brine (1 × 10 mL). The organic layer
was separated and concentrated in vacuo to a residue, which
was dried over Na2SO4 and subjected to purification by flash
column chromatography (silica gel-EtOAc-hexane 43:57) to
afford 2j in good yield (83%) as a solid (mp 96-97 °C). [R]D
-96° (c 0.001, CH2Cl2).
1H NMR (400 MHz, CDCl3): δ 7.66 (d, J ) 7.8 Hz, 0.5H);
7.60 (d, J ) 7.56 Hz, 1H); 7.36-7.27 (m, 5H); 7.12 (d, J ) 8.56
Hz, 0.5H); 5.92-5.84 (m, 1H); 5.32 (d, J ) 17.3 Hz, 1H); 5.22
(d, J ) 10.2 Hz, 1H); 44.63-4.54 (m, 5H); 4.09 (s, 0.4H); 4.07
(s, 0.6H); 3.84-3.80 (m, 1H),;3.69-3.67 (m, 1H); 3.65-3.59 (m,
1H); 2.20-1.8 (m, 3H); 1.75-1.55 (m, 2H); 1.5-1.35 (m, 1H);
1.29-1.15 (m, 1H); 0.95-0.84 (m, 12H).
Syn th esis of Allyl-N-(3-p h en ylglycid yl)-(L)-p r olin e-(L)-
leu cin e-(L)-isoleu cin a te (2h ) fr om 2c. To a stirring solution
of LiOH‚H2O (0.046 g, 1.1 mmol) in methanol-water (4:1 ratio,
20 mL) was added 2c and the mixture was stirred at room
temperature for 3 h until the disappearance of starting
material. The solvent was removed in vacuo and dichlo-
romethane (10 mL) was added to it followed by a small amount
of water (until the phase separation occurred). The pH of the
aqueous layer was adjusted to 5 and the aqueous layer was
extracted with dichloromethane. Acidification and extraction
were continued until the pH remained constant. The combined
extract was dried (Na2SO4) and taken up into a clean dry flask.
Triethylamine (0.2 mL, 1.5 mmol) was added to the reaction
vessel and the reaction vessel was cooled to 0 °C. Isobutyl
chloroformate (0.14 mL, 1 mmol) was added and stirring was
continued for 0.5 min. A solution of allyl-leucinate hydrochlo-
ride (0.21 g, 1 mmol) in DMSO (0.5 mL) was added to it
followed by addition of triethylamine (140 µL, 1 mmol) and
the mixture was stirred vigorously for 3-4 h. Removal of
solvent under vacuum yielded a residue, which was taken up
in ethyl acetate (20 mL) and washed with a saturated aqueous
solution of NaHCO3 (1 × 10 mL), water (1 × 10 mL), and brine
(1 × 10 mL) sequentially. The resulting organic layer was dried
and concentrated to give a residue, which was subjected to
column chromatography to give the pure compound that is
chemically and optically identical with 2h synthesized by
aerobic epoxidation of 1h .
Syn th esis of Meth yl-N-(3-p h en ylglycid yl)-(L)-p r olin e-
(L)-a sp a r ta te (2i). To a solution containing methyl-N-cin-
namoyl-(L)-proline-(L)-aspartate (78 mg, 0.2 mmol) in CH3CN
(2 mL) was added 2-methylpropanal (30 mg, 0.4 mmol) and
PASCOS (∼0.005 g) followed by addition of solid anhydrous
sodium acetate (131 mg, 1.6 mmol) and the mixture was stirred
under oxygen atmosphere at room temperature for 14 h. After
that a fresh sample of catalyst and 2-methylpropanal (86 µL,
0.95 mmol) and anhydrous sodium acetate (0.311 g, 10 mmol)
were added to the reaction mixture and stirring was continued
until complete conversion of the olefin to epoxide (TLC: Rf )
0.41; EtOAc:hexane 3:1.5). The catalyst and the inorganic salt
were filtered off and solvent was removed. The residue then
was taken up in EtOAc (10 mL) and washed with a saturated
solution of NaHCO3 (2 × 5 mL) and brine (1 × 5 mL). The
organic layer was separated and concentrated in vacuo to a
residue, which was dried over Na2SO4 and subjected to
Author: Please identify the missing component (X) in the
above.
1H NMR (400 MHz, CDCl3): δ 7.37-7.35 (m, 3H); 7.30-
7.25 (m, 3H); 7.22-7.18 (m, 4H); 7.09 (d, J ) 8.2 Hz, 1H); 6.91
(t, J ) 5.5 Hz, 1H); 5.73 (ddd, J ) 22.7, 10.3, 54 Hz, 1H); 5.06
(td, J ) 26.2, 1.5 Hz, 1H); 5.05 (dd, J ) 2.6, 1.4 Hz, 1H); 4.52
(dd, J ) 7.8, 3.2 Hz, 1H); 4.00 (d, J ) 2 Hz, 1H); 3.79 (dd, J )
7.6, 1.7 Hz, 2H); 3.65-3.50 (m, 2H); 3.55 (d, J ) 2 Hz, 1H);
3.23 (dd, J ) 13.9, 6.1 Hz, 1H); 3.01 (dd, J ) 13.9, 6.1 Hz,
1H); 2.05-2.02 (m, 1H); 1.97-1.88 (m, 2H); 1.84-1.79 (m, 1H);
MS (m/z) 448 (M+), 141, 328, 302, 244, 216, 200, 154, 136; IR
ν
max: 3300-3030 (br), 2880, 1760, 1600, 1575 cm-1
Syn th esis of 3-P h en yl-(2S,3S)-oxy-p r op a n ol (4). A mix-
.
ture of activated molecular sieves (0.9 g) and 50 mL of DCM
was cooled to -10 °C. L-(+)-Diethyl tertarate (0.5 g, 2.4 mmol),
t
Ti(OiPr)4 (0.45 g, 1.6 mmol), and BuOOH (7.8 mL, 48 mmol
6.2 M in CH2Cl2) were added sequentially. After 10 min, the
mixture was cooled to -20 °C and cinnamyl alcohol (4.35 g,
32.5 mmol in 10 mL of CH2Cl2) was added with stirring over
a time period of 15 min. Stirring was continued for 45 min at
-20 °C at which point the reaction mixture was allowed to
come to 0 °C and quenched with water (10 mL) and was
allowed to come to room temperature. To this was added 2.5
mL of a 30% aqueous solution of sodium hydroxide saturated
with sodium chloride and the mixture was were stirred for 10
min, which led to clear separation of the aqueous and organic
layers. The organic phase was removed and combined with
two extracts of the aqueous phase (CH2Cl2, 2 × 10 mL). The
combined organic phases were dried over Na2SO4 and filtered
through Celite to give a clear colorless solution. Concentration
followed by chromatography gave pure (94%; HPLC) 3-phenyl-
(2S,3S)-oxy-propanol 4 (85%).
1H NMR (400 MHz, CDCl3): δ 7.37-7.20 (m, 5H); 4.01 (dd,
J ) 12.7, 2.9 Hz, 1H); 3.90 (d, J ) 2.2 Hz, 1H); 3.74 (dd, J )
12, 6.8 Hz, 1H); 3.21 (td, J ) 4.1, 2.2 Hz, 1H); 2.71 (br s, 1H).
[R]25 -49° (c 0.01, CH2Cl2).
D
Syn th esis of 3-P h en yl-(2R,3S)-glycid ic Acid (5). RuIII
-
Cl3‚H2O (0.0075 g, 33 µM) was added to a stirring biphasic
mixture of epoxy alcohol 4 (0.15 g, 1 mmol), sodium periodate
(0.642 g, 3 mmol), and sodium bicarbonate (0.42 g, 5 mmol) in
1690 J . Org. Chem., Vol. 68, No. 5, 2003

N-Cinnamoly L-Proline Derived Peptides
CCl4 (2 mL), acetonitrile (2 mL), and water (3 mL). After 42 h
of stirring, additional amounts of RuCl3 (0.0076 g, 34 µM) and
sodium periodate (0.157 g, 0.72 mmol) were added and the
stirring was continued for 1 h to complete the reaction. Then
dichloromethane (8 mL) was added followed by a small amount
of water (until phase separation occurred). The pH of the water
layer was adjusted to 4 and the aqueous layer was extracted
with dichloromethane. Acidification and extraction were re-
peated until the pH remained constant. The combined layer
was dried over Na2SO4 and after evaporation of solvent
compound 5 was characterized for the presence of the car-
boxylic group by IR (3457 cm-1) and used immediately for
further coupling without complete characterization due to its
poor stability.
Syn th esis of Meth yl-N-cin n a m oyl-(L)-leu cin e-(L)-p r o-
lin a te (1m ). A stirring solution of N-cinnamoyl-(L)-leucine
(1.30 g, 5 mmol) and triethylamine (0.7 mL, 5 mmol) in THF
(10 mL) was cooled to 0 °C and to it was added isobutylchlo-
roformate (0.65 mL, 5 mmol) with vigorous stirring for 50 s.
After that, a solution of methyl-(L)-prolinate hydrochloride
(0.916 g, 5.5 mmol) in DMSO (2 mL) was added followed by a
solution of triethylamine (1.6 mL, 11 mmol) in THF (10 mL).
The reaction mixture was allowed to come to room temperature
and further stirred for 3-4 h. Inorganic salt was filtered off
followed by removal of the solvent in vacuo. The residue then
was taken up in EtOAc (35 mL) and washed with a saturated
solution of NaHCO3 (2 × 15 mL) and brine (1 × 10 mL). The
organic layer was separated and concentrated in vacuo to a
residue that was dried over Na2SO4 and subjected to purifica-
tion by column chromatography (silica gel-EtOAc-hexane
1:1.5) to isolate methyl-N-cinnamoyl-(L)-leucine-(L)-prolinate
1m as a gum in modest yield (61%).
was added to the reaction mixture, the solution was allowed
to come to room temperature, and stirring was continued at
ambient condition for 16 h. The solvent was evaporated and
the resulting mass was directly subjected to purification by
column chromatography (silica gel-ethyl acetate:hexane 38:
62) to obtain methyl-N-(N-cinnamolyl-(L)-prolyl)-3-phenylaziri-
dine-2-(L)-leucinate 6 in good yield (68%).
1H NMR (CDCl3, 300 MHz): δ 7.76 (d, J ) 15.4 Hz, 1H);
7.65 (d, J ) 7.32 Hz, 1H); 7.55-7.53 (m, 4H); 7.39-7.36 (m,
6H); 6.77 (d, J ) 15.4 Hz, 1H); 4.78 (d, J ) 6.6 Hz, 1H); 4.49
(d, J ) 7.8 Hz, 1H); 3.80-3.76 (m, 1H); 3.73 (s, 3H); 3.71 (s,
0.4 H); 3.67 (br s, 0.6H); 3.65 (m, 0.5H); 3.63 (m, 0.5H); 3.61
(br s, 0.1H); 3.55 (br s, 0.7H); 3.39 (br s, 0.2H); 2.25-2.46 (m,
1H); 2.31-2.14 (m, 2H); 2.06-1.98 (m, 1H); 1.91-1.82 (m, 1H);
1.71-1.56 (m, 2H); 1.25 (br s, 1H); 0.90 (d, J ) 6.08 Hz, 3H);
0.88 (d, J ) 6.08 Hz, 3H).
Syn th esis of Meth yl-N-(N-3′-p h en ylglycid yl-(L)-p r olyl)-
3-p h en yla zir id in e-2-(L)-leu cin a te (6a ). To a solution con-
taining methyl-N-(N-cinnamolyl-(L)-prolyl)-3-phenylaziridine-
2-(L)-leucinate (7) (0.20 g, 0.38 mmol) in CH3CN (8 mL) was
added 2-methylpropanal (86 µL, 0.95 mmol) and PASCOS
(∼0.004 g) followed by addition of solid anhydrous sodium
acetate (0.311 g, 10 mmol) and the mixture was stirred under
oxygen atmosphere at room temperature for 12 h. After that
a fresh sample of catalyst and 2-methylpropanal (86 µL, 0.95
mmol) and anhydrous sodium acetate (0.311 g, 10 mmol) was
added to the reaction mixture, stirring was continued until
complete conversion of the olefine to epoxide, the catalyst and
the inorganic salt were filtered off, and solvent was removed.
The residue then was taken up in EtOAc (25 mL) and washed
with a saturated solution of NaHCO3 (2 × 5 mL) and brine (1
× 10 mL). The organic layer was separated and concentrated
in vacuo to a residue, which was dried over Na2SO4 and
subjected to purification by column chromatography (silica gel-
EtOAc-hexane 43:57) to isolate methyl-N-(N-glycidyl-(L)-
prolyl)-3-phenylaziridine-2-(L)-leucinate 6a in good yield (91%).
Spectral data for the major diastereomer are given below.
1H NMR (CDCl3, 60 MHz): δ 7.72 (d, J ) 16 Hz, 1H); 7.52
(d, J ) 8.9 Hz, 1H); 7.50-7.16 (m, 5H); 6.47 (d, J ) 16 Hz.
1H); 4.8-4.7 (m, 1H); 4.58-4.50 (m, 1H); 3.75 (s, 3H); 3.74-
3.59 (m, 2H); 2.20-2.10 (m, 1H); 2.10-1.99 (m, 3H); 1.76-
1.49 (m, 3H); 0.96 (d, J ) 5.6 Hz, 6H).
1H NMR (CDCl3, 400 MHz): δ 7.29-7.12 (m, 10H); 4.60 (d,
J ) 7.56 Hz, 1H); 4.43-4.41 (m, 1H); 4.10 (br s, 0.1H); 4.02
(br s, 0.9H); 3.73-3.69 (m, 3H); 3.66 (s, 3H); 3.54 (br s, 0.9H);
3.49 (br s, 0.8H); 3.44 (s, 0.2H); 3.40 (s, 0.1H); 2.34-2.32 (m,
1H); 2.09-2.07 (m, 1H); 1.94-1.57 (m, 4H); 1.18 (br s, 2H);
0.89-0.86 (m, 6H). MS (m/z): 519 (M + 1)+.
Syn t h esis of Met h yl-N-cin n a m oyl-(L)-p r olin e-d eh y-
d r op h en yla la n in e-(L)-leu cin a te (7). To an ice-cooled stirred
solution of N-cinnamoyl-(L)-proline-dehydrophenylalanine (0.16
g, 0.41 mmol) in CH2Cl2 (5 mL) was added HOBt hydrate
(0.055 g, 0.41 mmol), (L)-leucine methyl ester hydrochloride
(0.074 g, 0,41 mmol), and triethylamine (0.12 mL, 0.82 mmol),
and stirring was continued for 10 min at 0 °C. After that, a
solution of DCC (0.084 g, 0.41 mmol) in CH2Cl2 (2.5 mL) was
added to the solution and the reaction mixture was allowed
to come to room temperature. Stirring was continued at
ambient condition for 16 h. The solvent was evaporated and
the resulting mass was directly subjected to purification by
column chromatography (silica gel-ethyl acetate:hexane 38:
62) to obtain methyl-N-cinnamoyl-(L)-proline-dehydrophen-
ylalanine-(L)-leucinate 7 in pure (92%; HPLC) form with good
yield (78%).
1H NMR (CDCl3, 400 MHz): δ 8.11 (s, 0.75H); 7.67 (d, J )
15.4 Hz, 1H); 7.54-7.26 (m, 11H); 6.73 (d, J ) 15.4 Hz, 1H);
4.76-4.69 (m, 1H); 4.58 (dd, J ) 7.32, 4.60 Hz, 1H); 3.86-
3.80 (m, 1H); 3.75-3.59 (m, 4H); 2.30-1.99 (m, 4H); 1.78-
1.60 (m, 3H); 1.26-1.22 (br s, 8H); 0.89 (d, J ) 6.08 Hz, 3H);
0.85 (d, J ) 6.08 Hz, 3H). MS (m/z): 519 (M + 1)+, 502, 374,
348, 260, 228, 200, 172, 131.
Syn th esis of Meth yl-N-(3-p h en ylglyid yl)-(L)-p r olin e-
d eh d r op h en ya la n in e-(L)-leu cin a te (7a ). To a solution con-
taining methyl-N-cinnamoyl-(L)-proline-dehydrophenylalanine-
(L)-leucinate (Y) (0.20 g, 0.38 mmol) in CH3CN (8 mL) was
added 2-methylpropanal (86 µL, 0.95 mmol) and PASCOS
(∼0.004 g) and the mixture was stirred under oxygen atmos-
Syn th esis of Meth yl-N-(3-p h en ylglycid yl)-(L)-leu cin e-
(L)-p r olin a te (2m ). To a solution of methyl-N-cinnamoyl-(L)-
leucine-(L)-prolinate (1m ) (0.558 g, 1.5 mmol) in CH3CN (8 mL)
was added 2-methylpropanal (0.272 mL, 3 mmol) and PASCOS
(∼0.005 g) and the mixture was stirred under oxygen atmos-
phere at room temperature for 12 h. After this time, a fresh
sample of the catalyst and 2-methylpropanal (0.272 mL, 3
mmol) were added to the reaction mixture and allowed to stir
until complete conversion of the olefine (TLC: Rf ) 0.5; hexane:
ethyl acetate 3:2). The catalyst was filtered off on a sintered
funnel and acetonitrile was removed in vacuo. The resulting
residue was taken up in EtOAc (20 mL) and washed succes-
sively with a saturated solution of NaHCO3 (2 × 10 mL), water
(2 × 10 mL), and brine (1 × 10 mL). Separating the organic
phase, drying (Na2SO4), and concentration in vacuo yielded
the corresponding oxirane containing an equal diastereomeric
mixture of peptide 2m in high chemical yields (72%) as a gum.
1H NMR (CDCl3, 300 MHz) δ: 7.36-7.32 (m, 3H); 7.29-
7.22 (m, 2H); 4.54-4.52 (m, 1H); 4.2-4.12 (m, 1H); 3.95 (d, J
) 1.92 Hz, 0.4H); 3.92 (s, 0.2H); 3.74 (s, 1.8H); 3.72 (s, 1.2 H);
3. 67 (d, J ) 1.84 Hz, 0.4H); 3.69-3.55 (m, 2H); 3.52 (d, J )
1.96 Hz, 0.4H); 3.47 (d, J ) 1.84 Hz, 0.4H); 3.40 (d, J ) 1.96
Hz, 0.2H); 2.21-2.09 (m, 1H); 2.07-1.96 (m, 2H); 1.82-1.62
(m, 2H); 1.58-1.52 (m, 2H); 0.99 (d, J ) 6 Hz, 3H); 0.95 (dd,
J ) 9, 3 Hz, 3H).
Syn th esis of Meth yl-N-(cin n a m olyl-(L)-p r olyl)-3-p h e-
n yla zir id in e-2-(L)-leu cin a te (6). To an ice-cooled stirred
solution of methyl-N-(N-cinnamolyl-(L)-prolyl)-3-phenylaziri-
dine-2-carboxylic acid (0.16 g, 0.41 mmol) in CH2Cl2 (5 mL)
was added HOBt hydrate (0.055 g, 0.41 mmol) and triethyl-
amine (0.06 mL, 0.41 mmol) followed by addition of a solution
of (L)-leucine methyl ester hydrochloride (0.074 g, 0,41 mmol)
and triethylamine (0.06 mL, 0.41 mmol) in CH2Cl2 (2.5 mL).
Stirring was continued for 10 min at 0 °C. After that, another
solution of DCM (2.5 mL) containing DCC (0.084 g, 0.41 mmol)
J . Org. Chem, Vol. 68, No. 5, 2003 1691

Nandy et al.
phere at room temperature for 12 h. After that a fresh sample
of catalyst and 2-methylpropanal (86 µL, 0.95 mmol) was
added to the reaction mixture and stirring was continued until
complete conversion of the olefine to epoxide, the catalyst was
filtered off, and solvent was removed. The residue then was
taken up in EtOAc (25 mL) and washed with a saturated
solution of NaHCO3 (2 × 5 mL) and brine (1 × 10 mL). The
organic layer was separated and concentrated in vacuo to a
residue, which was dried over Na2SO4 and subjected to
purification by column chromatography (silica gel-EtOAc-
hexane 2:3) to isolate methyl-N-(3-phenylglyidyl)-(L)-prolyl-
dehdrohenyalanine-(L)-leucinate 7a in good yield (89%). Ex-
perimental data for the major diastereomer are given below.
1H NMR (CDCl3, 400 MHz): δ 8.00 (s, 0.6H); 7.92 (s, 0.4H);
7.50-7.20 (m, 11H); 4.74-4.70 (m, 1H); 4.55-4.51 (m, 1H);
4.08 (d, J ) 1.92 Hz, 0.4H); 4.05 (d, J ) 1.6 Hz, 0.6H); 3.74 (s,
1.2H); 3.71(s, 1.8H); 3.60 (d, J ) 1.98 Hz, 0.4H), 0.4H); 3.59
(d, J ) 1.68 Hz, 0.6H) 2.30-1.99 (m, 4H); 1.78-1.60 (m, 3H);
1.26 (br s, 1H); 0.89 (d, J ) 6.08 Hz, 3H); 0.85 (d, J ) 6.08 Hz,
3H).
Syn th esis of Eth yl-N-(N-3′-p h en ylglycid yl-(L)-p r olyl)-
3-p h en yla zir id in e-2-ca r boxyla te (8a ). To a solution con-
taining ethyl-N-(N-3′-phenylglycidyl-(L)-prolyl)-3-phenylazir-
idine-2-carboxylate (9) (0.164 g, 0.39 mmol) in CH3CN (10 mL)
was added 2-methylpropanal (86 µL, 0.98 mmol) followed by
the addition of solid anhydrous sodium acetate (0.311 g, 10
mmol) and PASCOS (∼0.004 g) and the mixture was stirred
under oxygen atmosphere at room temperature for 12 h. After
that a fresh sample of catalyst and 2-methylpropanal (88 µL,
0.98 mmol) and solid anhydrous sodium acetate (0.311 g, 10
mmol) were added to the reaction mixture, stirring was
continued until complete conversion of the olefine to epoxide,
the catalyst and the inorganic salt were filtered off, and solvent
was removed. The residue then was taken up in EtOAc (30
mL) and washed with a saturated solution of NaHCO3 (2 × 5
mL) and brine (1 × 10 mL). The organic layer was separated
and concentrated in vacuo to a residue, which was dried over
Na2SO4 and subjected to purification by column chromatog-
raphy (silica gel-EtOAc-hexane 2:3) to isolate the diastereo-
mers of ethyl-N-(N-3′-phenylglycidyl-(L)-prolyl)-3-phenylazir-
idine-2-carboxylate 8a in good yield (92%). Experimental data
for one of the pure diastereomers of 9a are given below.
1H NMR (CDCl3, 400 MHz): δ 7.41-7.20 (m, 10H); 6.82 (d,
J ) 6.32 Hz, 1H); 4.77 (dd, J ) 7.8, 3.4 Hz, 1H, 0.5H); 4.61
(dd, J ) 3.64, 8.52 Hz, 0.5H); 4.30-4.12 (m, 2H); 4.06 (d, J )
1.68 Hz, 0.3H); 4.00 (d, J ) 1.96 Hz, 0.7H); 3.95 (d, J ) 1.72
Hz, 0.6 H); 3.73-3.61 (m, 1H); 3.67 (d, J ) 1.96 Hz, 0.3H);
3.52-3.42 (m, 1H); 3.50 (d, J ) 1.96 Hz, 0.4H); 3.15 (d, J )
1.92 Hz, 0.6H); 3.12 (d, J ) 2.2 Hz, 0.7H); 2.87 (d, J ) 1.96
Hz, 0.3H); 2.40-1.87 (m, 4H); 1.30 (dt, J ) 7.08, 3.68 Hz, 3H);
MS (m/z): 435 (M+), 419, 280, 269, 241, 228, 216, 200.
Syn th esis of Allyl-N-(3-p h en ylglycid yl)-(D)-p r olin e-(L)-
leu cin e Am id e (9a ). This epoxide was prepared according to
the procedure followed for 2f in 74% yield and 91% (HPLC)
purity.
1H NMR (400 MHz, CDCl3): δ 7.41-7.36 (m, 3H); 7.35-
7.34 (m, 2H); 6.37 (d, J ) 7.75 Hz, 1H); 5.92-5.86 (m, 1H);
5.21 (dd, J ) 17.5, 1.5 Hz, 1H); 5.18 (d, J ) 10.5 Hz, 1H); 4.62
(dd, J ) 8.5, 3.5 Hz, 1H); 4.40-4.33 (m, 1H); 4.16 (s, 0.2H);
4.12 (d, J ) 1.76 Hz, 0.6H); 4.16 (d, J ) 1.5 Hz, 0.2 H); 3.82-
3.765 (m, 3H); 3.63-3.56 (m, 1H); 3.65 (d, J ) 1.7 Hz, 0.8H);
3.50 (d, J ) 1.6 Hz, 0.2H); 2.19-2.11 (m, 1H); 2.13-2.08 (m,
1H); 2.05-1.21 (m, 2H); 1.68-1.59 (m, 1H); 1.51-1.49 (m, 2H);
0.85 (dd, J ) 6.1, 3.7 Hz 3H); 0.81 (d, J ) 5.84 Hz, 1H).
Ack n ow led gm en t. We thank DST, New Delhi for
the financial support to this work. We are grateful to
Dr. Ram Thaimattam for helping us on the molecular
dynamics simulation studies.
Su p p or tin g In for m a tion Ava ila ble: Spectra data 1d , 1f,
2c, 2f, 2l, 6, 6a , 7, and 7a . This material is available free of
charge via the Internet at http://pubs.acs.org.
J O020352J
1692 J . Org. Chem., Vol. 68, No. 5, 2003