Influence of chirality of the preceding acyl moiety on the cis/trans ratio of the proline peptide bond

Article abstract of DOI:10.1021/jo0159439

We report that the cis/trans ratio of the proline peptide bond can be strongly influenced by the chirality of the acyl residue preceding proline. Acyl moieties derived from (2S)-2,6-dimethyl-3-oxo-3,4-dihydro-2H-1,4-benzoxazine-2-carboxylic acid (8) and (

Full text of DOI:10.1021/jo0159439

7044
J . Org. Chem. 2001, 66, 7044-7050
In flu en ce of Ch ir a lity of th e P r eced in g Acyl Moiety on th e
cis/tr a n s Ra tio of th e P r olin e P ep tid e Bon d
Matej Breznik,† Simona Golicˇ Grdadolnik,‡ Gerald Giester,§ Ivan Leban,| and Danijel Kikelj*,†
University of Ljubljana, Faculty of Pharmacy, Asˇkercˇeva 7, 1000 Ljubljana, Slovenia, National Institute
of Chemistry, Ljubljana, Slovenia, Institut fu¨r Mineralogie und Kristallographie, Universita¨t Wien,
Austria, and University of Ljubljana, Faculty of Chemistry and Chemical Technology,
Ljubljana, Slovenia
danijel.kikelj@ffa.uni-lj.si
Received J uly 18, 2001
We report that the cis/trans ratio of the proline peptide bond can be strongly influenced by the
chirality of the acyl residue preceding proline. Acyl moieties derived from (2S)-2,6-dimethyl-3-oxo-
3,4-dihydro-2H-1,4-benzoxazine-2-carboxylic acid (8) and (2R)-3-methoxy-2-methyl-2-(4-methyl-2-
nitrophenoxy)-3-oxopropanoic acid (5) in acyl-Pro molecules influence isomerization of the proline
peptide bond constraining the ω dihedral angle to the trans orientation. Structures of benzyl (2S)-
1-{[(2S)-2,6-dimethyl-3-oxo-3,4-dihydro-2H-1,4-benzoxazin-2-yl]carbonyl}-2-pyrrolidinecarboxy-
1
late (3) derived from 2D H NMR conformational analysis and crystallographic data exhibit only
the trans conformation of proline peptide bond. On the other hand the diastereomer 4, which contains
1
an (R) acyl moiety, exhibits two sets of signals in H NMR spectra. The signals were assigned to
trans (72%) and cis (28%) conformers. Crystallographic analysis of 4 showed that only the cis
1
conformation is present in the crystalline state. The H NMR chemical shift pattern of three sets
of signals observed in 2 was observed also in benzyl (2S)-1-[(2R/ S)-3-methoxy-2-methyl-2-(4-methyl-
2-nitrophenoxy)-3-oxopropanoyl]-2-pyrrolidinecarboxylate. (R)-Carboxylic acid 5, after coupling with
(S)-ProOBn, yielded benzyl (2S)-1-[(2R)-3-methoxy-2-methyl-2-(4-methyl-2-nitrophenoxy)-3-oxo-
propanoyl]-2-pyrrolidinecarboxylate (6), which in DMSO-d6 exhibited only the trans conformation
of the proline peptide bond. These results suggest that in these particular cases acyl-Pro peptide
bond isomerization is strongly influenced by the stereochemistry of the acyl residue preceding
proline. (2S)-2,6-Dimethyl-3-oxo-3,4-dihydro-2H-1,4-benzoxazine-2-carboxylic acid (8) and (2R)-3-
methoxy-2-methyl-2-(4-methyl-2-nitrophenoxy)-3-oxopropanoic acid (5) are promising chiral pep-
tidomimetic building blocks that can be used as acyl moieties to force the proline peptide bond into
the trans conformation in a variety of acyl-Pro molecules.
In tr od u ction
naturally occurring biologically active and pharmaceuti-
cally important peptides and peptide mimetics.6
Naturally occurring amino acids form predominantly
trans peptide bonds, with the exception of proline whose
ability to form cis peptide bonds1 and undergo cis/trans
isomerization is well-known.2 The isomerization barrier
of the acyl-Pro bond in proline-containing molecules is
lowered from ca. 20 to ca. 13 kcal/mol as a result of
concomitant pyrrolidine puckering changes;3 additionally,
the trans bond, involving a nitrogen atom of the pyrro-
lidine ring, is favored only slightly (ca. 2 kcal/mol) over
the cis bond.4 Because of these unique properties proline
plays an important role in biological processes5 and in
Determination of bioactive conformations of biologically
active compounds, followed by optimization of structural
and spatial arrangement of groups important for biologi-
cal activity with the aim to enhance the affinity and the
selectivity of a molecule toward the sites of action, is a
crucial step in the development of potent therapeutics.7
Conformational freedom and thus the number of confor-
mations accessible to the molecule can be reduced by
introducing constraints that may directly influence the
pharmacological effect of the molecule.8 Proline incorpo-
(4) (a) Tanaka, S.; Scheraga, H. A. Macromolecules 1974, 7, 698-
705. (b) Momany, F. A.; McGuire, R. F.; Brurgess, A. W.; Scheraga, H.
A. J . Phys. Chem. 1975, 79, 2361-2381. (c) Ne´methy, G.; Pottle, M.
S.; Scheraga, H. A. J . Phys. Chem. 1983, 87, 1883-1887.
(5) Yaron, A.; Naider, F. Crit. Rev. Biochem. Mol. Biol. 1993, 28,
31-81.
(6) (a) Moore, G. J .; Smith, J . R.; Baylis, B. W.; Matsoukas, J . M.
Adv. Pharmacol. (San Diego) 1995, 6, 91-141. (b) Kern, D.; Schut-
kowski, M.; Drakenberg, T. J . Am. Chem. Soc. 1997, 119, 8403-8408.
(c) Perera, L.; Darden, T. A.; Pedersen, L. G. Biochemistry 1998, 37,
10920-10927.
(7) Doucet, J . P.; Weber, J . Computer-Aided Molecular Design:
Theory and Applications; Academic Press LTD: London, 1997; pp 197-
200, 363-404.
† Faculty of Pharmacy, University of Ljubljana.
‡ National Institute of Chemistry.
§ Universita¨t Wien.
| Faculty of Chemistry and Chemical Technology, University of
Ljubljana.
(1) (a) Brandts, J . F.; Halvorson, H. R.; Brennan, M. Biochemistry
1975, 14, 4953-4963. (b) Hetzel, R.; Wu¨trich, K. Biopolymers 1979,
18, 2589-2606. (c) Grathwohl, C.; Wu¨trich, K. Biopolymers 1981, 20,
2623-2633. (d) Stewart, D. E.; Sarkar, A.; Wampler, J . E. J . Mol. Biol.
1990, 214, 253-260.
(2) (a) Schmid, F. X. Annu. Rev. Biophys. Biomol. Struct. 1993, 22,
123-143. (b) Fischer, S.; Dunbrack, R. L., J r.; Karplus, M. J . Am.
Chem. Soc. 1994, 116, 11931-11937. (c) Reimer, U.; El Mokdad, N.;
Schutkowski, M.; Fischer, G. Biochemistry 1997, 36, 13802-13808.
(3) (a) Schulz, G. E.; Schirmer, R. H. Principles of Protein Structure;
Springer-Verlag: New York, 1961; p 25. (b) Tonelli, E. A. J . Am. Chem.
Soc. 1973, 95, 5946-5948.
(8) (a) Patrick, G. L. An Introduction to Medicinal Chemistry; Oxford
University Press: New York, 1995; pp 96-101. (b) Nogrady, T.
Medicinal Chemistry: A Biochemical Approach; Oxford University
Press: New York, 1988; pp 47-49, 459.
10.1021/jo0159439 CCC: $20.00 © 2001 American Chemical Society
Published on Web 09/27/2001

cis/trans Ratio of the Proline Peptide Bond
J . Org. Chem., Vol. 66, No. 21, 2001 7045
Sch em e 1
rated into a biologically active compound restricts the
local conformational freedom,9 which leads to changes in
activity due to altered molecular flexibility and the
number of possible conformers. An example demonstrat-
ing the importance of the cis/trans isomerization of the
proline peptide bond is provided by the tripeptidomimetic
thrombin inhibitors10 derived from the peptide sequence
D-Phe-Pro-Arg, in which one or more amino acids are
replaced by a mimetic or another amino acid. The
majority of potent thrombin inhibitors containing proline
possess the trans conformation of the proline amide bond
in order to form a hydrogen bond to Gly 216, which is
one of residues defining the S1 specificity pocket of the
thrombin active site.10,11
In small linear peptides the cis/trans ratio of the acyl-
Pro bond favors the trans conformation12 but depends
considerably on the amino acid adjacent to proline.12,13
The probability of the cis conformation is higher when
an aromatic residue such as phenylalanine precedes
proline, probably as a result of interactions between the
aromatic and pyrrolidine rings.1b,12a
Numerous surrogates, mimetics, and analogues of
proline have been developed and used in the synthesis
of biologically active compounds, with the aim of altering
the cis/trans ratio of acyl-Pro bonds,14 constraining the
conformation of the peptide bond10a,15 and producing
proline-like turns.16 Because of their ability to enforce
the trans conformation of the amide bond, some have
been used as peptidomimetic scaffolds and introduced
into thrombin inhibitors. (R)-Thiazolidine-4-carboxylic
acid6b,17 and (3R)-6-amino-5-oxohexahydro-5H-[1,3]thia-
zolo[3,2-a]pyridine-3-carboxylic acid11a were incorporated
in a series of thrombin inhibitors that showed high
affinity and selectivity.
Here we report the synthesis, X-ray structure analysis,
and NMR conformational study of benzyl (2S)-1-{[(2S)-
2,6-dimethyl-3-oxo-3,4-dihydro-2H-1,4-benzoxazin-2-yl]-
carbonyl}-2-pyrrolidinecarboxylate (3) and the corre-
sponding (S,R)-diastereomer 4, providing evidence that
the (2S)-2,6-dimethyl-3-oxo-3,4-dihydro-2H-1,4-benzox-
azin-2-carbonyl moiety constrains the acyl-Pro bond of 3
to trans. Additionally, we report an NMR conformational
analysis of benzyl (2S)-1-[(2R)-3-methoxy-2-methyl-2-(4-
methyl-2-nitrophenoxy)-3-oxopropanoyl]-2-pyrrolidine-
carboxylate (6), an intermediate in the stereoselective
synthesis of 3, which also possesses the trans conforma-
tion of the proline peptide bond. These results demon-
strate that the cis/ trans ratio of the proline peptide bond
in acyl-Pro molecules is strongly influenced by the
chirality of the acyl moiety derived from 3, 4, and 6 and
can be fixed to trans.
(9) (a) Schimmel, P. R.; Flory, P. J . J . Mol. Biol. 1968, 34, 105-
120. (b) Tonelli, A. E. J . Mol. Biol. 1974, 86, 627-635. (c) Tonelli, A.
E. Biopolymers 1976, 15, 1615-1622.
(10) Rewinkel, J . B. M.; Adang, A. E. P. Curr. Pharm. Des. 1999, 5,
1043-1075.
(11) (a) Wagner, J .; Kallen, J .; Ehrhardt, C.; Evenou, J . P.; Wagner,
D. J . Med. Chem. 1998, 41, 3664-3674. (b) Hilpert, K.; Akermann, J .;
Banner, D. W.; Gast, A.; Gubernator, K.; Hadva´ry, P.; Labler, L.;
Mu¨ller, K.; Schmid, G.; Tschopp, T. B.; Van der Waterbeemd, H. J .
Med. Chem. 1994, 37, 3889-390. (c) Maryanoff, B. E.; Qiu, X.;
Padmanaphan, K. P.; Tulinsky, A.; Almond, H. R. J r.; Andrade-Gordon,
P.; Greco, M. N.; Kauffman, J . A.; Nicolau, K. C.; Liu, A.; Brungs, P.
H.; Fusetani, N. Proc. Natl. Acad. Sci. U.S.A. 1990, 90, 8048-8052.
(12) (a) Kang, Y. K.; J hon, J . S.; Han, S. J . J . Peptide Res. 1999, 53,
30-40. (b) Reimer, U.; Scherer, G.; Drewello, M.; Kruber, S.; Schut-
kowski, M.; Fischer, G. J . Mol. Biol. 1999, 279, 449-460.
(13) (a) Zimmerman, S. S.; Scheraga, H. A. Biopolymers 1977, 16,
811-843. (b) MacArthur, M. W.; Thornton, J . M. J . Mol. Biol. 1991,
218, 397-412.
(14) (a) Delaney, N. G.; Madison, V. J . Am. Chem. Soc. 1982, 104,
6635-6641. (b) Montelione, G. T.; Hughes, P.; Clardy, J .; Scheraga,
H. A. J . Am. Chem. Soc. 1986, 108, 6765-6773. (c) Beausoleil, E.;
Lubell, W. D. J . Am. Chem. Soc. 1996, 118, 12902-12908. (d)
Beausoleil, E.; Sharma, R.; Michnick, S. W.; Lubell, W. D. J . Org.
Chem. 1998, 63, 6572-6578. (e) Keller, M.; Seger, C.; Dumy, P.;
Schutkowski, M.; Fischer, G. S.; Mutter, M. J . Am. Chem. Soc. 1998,
120, 2714-2720.
(15) (a) Ward, P.; Evan, J . B.; J ordan, C. C.; Ireland, S. J .; Hagan,
R. M.; Brown, J . R. J . Med. Chem. 1990, 33, 1848-1851. (b) Zabrocki,
J .; Dunbar, J . B., J r.; Marshall, K. W.; Toth, M. V.; Marshall, G. R. J .
Org. Chem. 1998, 63, 6572-6578.
(16) (a) Hinds, M. G.; Welsh, J . H.; Brennand, D. M.; Fisher, J .;
Glennie, M. J .; Richards, N. G. J .; Turner, D. L.; Robinson J . A. J .
Med. Chem. 1991, 34, 1777-1789. (b) Genin, M. J .; J ohnson, R. L. J .
Am. Chem. Soc. 1992, 114, 8778-8783. (c) Gramberg, D.; Weber, C.;
Beeli, R.; Inglis, J .; Bruns, C.; Robinson, J . A. Helv. Chim. Acta 1995,
78, 1588-1606. (d) No¨teberg, D.; Brånalt, J .; Kvarnstro¨m, I.; Classon,
B.; Samuelsson, B.; Nillroth, U.; Danielson, U. H.; Karle´n, A.; Hallberg,
A. Tetrahedron 1997, 53, 7975-7984.
Resu lts a n d Discu ssion
Syn th esis. The carboxylic acid 1, providing the acyl
residues of 3 and 4, was synthesized as described
previously.18 Alkylation and concomitant cyclization to
benzoxazinone proceeded in a nonstereoselective manner,
providing racemic 1 which, after coupling with S-ProOBn,
afforded 2 as a mixture of diastereomers. Diastereomers
3 and 4 of (2S)-1-[(2,6-dimethyl-3-oxo-3,4-dihydro-2H-1,4-
benzoxazin-2-yl)carbonyl]-2-pyrrolidinecarboxylate (2) were
obtained after separation by column chromatography
(Scheme 1).
The diastereomer 3 was also prepared by a stereo-
selective synthesis via PLE-catalyzed hydrolysis of di-
methyl 2-methyl-2-(4-methyl-2-nitrophenoxy)malonate,
affording (2R)-monomethyl 2-methyl-2-(4-methyl-2-nitro-
phenoxy)malonate (5) as the key intermediate19 (Scheme
2). Palladium-catalyzed reduction of 5, in the presence
of an equivalent amount of NaOH in water and following
in situ cyclization, yielded 8,19 which was easily converted
to 3 via EDC-catalyzed coupling with S-ProOBn. EDC-
catalyzed coupling of 5 with S-ProOBn yielded 6, which
because of its lability was not purified but was used
immediately to obtain 7, which was used directly without
purification to obtain 3 after esterification with benzyl
(17) (a) Shuman, R. T.; Rothenberger, R. B.; Campbell, C. S.; Smith,
G. F.; Gifford-Moore, D. S.; Paschal, J . W.; Gesellchen, P. D. J . Med.
Chem. 1995, 38, 4446-4453. (b) Dumy, P.; Keller, M.; Ryan, D. E.;
Rohwedder, B.; Wo¨hr, T.; Mutter, M. J . Am. Chem. Soc. 1997, 119,
918-925.
(18) Kikelj, D.; Suhadolc, E.; Urleb, U.; Zˇbontar U. J . Heterocycl.
Chem. 1993, 30, 597-602.
(19) Breznik, M.; Hrast, V.; Mrcina, A.; Kikelj, D. Tetrahedron:
Asymmetry 1999, 10, 153-167.

7046 J . Org. Chem., Vol. 66, No. 21, 2001
Breznik et al.
Sch em e 2
indicating the existence of two conformations of proline
peptide bond in DMSO-d6 solution at 302 K. For the more
abundant species, NOE peaks were observed between
δCH2Pro/C2-CH3 and δCH2Pro/HetC8H (marked with
arrow in right spectrum in Figure 1) characterizing the
trans conformation.
For the less abundant species, the aromatic ring and
C2-CH3 exhibited no NOEs with δ protons of proline
(Figure 1), indicating that the proline peptide bond of the
species in solution adopts the cis conformation. Inter-
conversion between these two compounds was observed
by exchange cross-peaks between RCHPro resonances of
the cis and trans isomers in the NOESY experiments
measured at 302 K (Figure 1). Chemical exchange cross-
peaks were identified on the basis of their phases, which
were in phase with the diagonal peaks but out of phase
with NOE cross-peaks.
The same phenomenon could be observed when a less
conformationally constrained acyl residue, i.e., 3-meth-
oxy-2-methyl-2-(4-methyl-2-nitrophenoxy)-3-oxopropano-
ic acid was used. The 1H NMR chemical shift pattern was
similar to that of 2. When the (2R)-isomer (acid (5)) was
introduced into the Acyl-Pro molecule, the corresponding
amide (6) (Scheme 2) possessed only the trans conforma-
tion according to the structure determined in solution by
bromide. The diastereomeric purity of 3-7 was checked
1
by H NMR spectroscopy.
NMR Sp ectr oscop y. The proton chemical shifts of 3,
4, 6, and 7 were assigned following standard procedures
using homonuclear COSY and TOCSY experiments in
combination with 1H13C-HMQC experiments. For 3, 6,
1
2D H NMR. Cyclization of 6 and concomitant cleavage
of the benzyl ester yielded 7, whose 1H NMR spectra
exhibited only one set of resonances; conformational
analysis revealed the trans conformation of the proline
peptide bond.
1
and 7 only one set of resonances was observed in H NMR
spectra, demonstrating the existence of a single confor-
mation of the proline peptide bond in DMSO-d6 solution.
NOE peaks were observed between δCH2Pro/C2-CH3 and
δCH2Pro/HetC8H (marked with arrow in left spectrum
in Figure 1) characterizing the trans conformation of the
peptide bond of the species. In 6 NOEs between δCH2-
Pro/C2-CH3 and ArC6H/C2-CH3 were used as relevant
restraints for the structure elucidation. Two distinct sets
of signals were observed in the one-dimensional 1H
spectrum of 4 with relative populations of 28% and 72%,
The proton-proton distances (Table 1) were calculated
from the NOESY spectra measured at 150 ms using the
two-spin approximation and the integrated intensity of
a geminal pair of protons assumed to have a distance of
1.8 Å. All compounds were in the positive NOE regime
when studied in DMSO at 302 K.
Molecu la r Dyn a m ics. The experimentally derived
distances were incorporated as conformational restraints
F igu r e 1. Expanded regions of NOESY spectra of 3 (left) and 4 (right). NOEs that are consistent with the trans orientation of
the proline peptide bond are marked with arrows. NOE cross-peaks have the opposite sign to the diagonal peaks and are shown
by dotted contours. The resonances of the cis and trans conformers of 4 overlap in the aromatic region.

cis/trans Ratio of the Proline Peptide Bond
J . Org. Chem., Vol. 66, No. 21, 2001 7047
F igu r e 2. Stereoviews of the solution structures of 3 (a), 4-trans (b), 4-cis (c), and 6 (d), observed in DMSO.
Ta ble 1. NOE Restr a in ts Used To Deter m in e Str u ctu r es
3, 4, 6, a n d 7 in DMSO a n d Dista n ce Viola tion s fr om
Restr a in ed Sim u la tion s [Å]
average NOE-distance violation from the restrained
simulation of each compound is less than 0.2 Å, and the
largest NOE-distance violation is less than 0.3 Å (Table
1). In addition, the dihedral angle variations of this part
of the molecules are less than 15°.
The conformation of the proline peptide bond of result-
ing structures is consistent with the observed NOEs. The
benzyl part of the molecules is flexible during simulation,
not being restrained experimentally by the aromatic part
of benzoxazinone ring and proline ring, and the dihedral
angles of the methylene group that connects the benzyl
ring to the rest of the molecule vary greatly in conse-
quence.
Cr yst a llogr a p h y. Figure 3 was produced using
ORTEPII29 and Figure 4 using PLATON.30 Molecule 3 is
shown to contain a trans peptide bond [torsion angle C2-
C10-N21-C22 ) 179.0(2)°], whereas the bond in mol-
ecule 4 is cis [torsion angle C2-C10-N21-C22 ) 0.53-
(3)°] (Figure 3). There is also a difference in the chirality
of the C2 atom, the S configuration being present in 3,
but the R configuration in compound 4. A stereoscopic
view of the molecular packing is presented in Figure 4.
The bond lengths and angles are normal and in agree-
ment with the values for the related compounds. L-
Proline pyrrolidine ring puckering exists close to Câ-exo/
Cγ-endo in both structures. The packing of the molecules
in both structures is dictated by an intermolecular, nearly
linear hydrogen bond N2-H4‚‚‚O3 [O3 (x - 1,y,z), 2.917-
(2) Å] in 3 and [O3 (1 - x, -1/2 + y, 3/2 - z), 2.788(2) Å]
in 4.
3
4-trans
4-cis
6
7
δCH2Pro/C2-CH3
RCHPro/âCH2Pro
δCH2Pro/γCH2Pro
HetC8H/δCH2Pro
âCH2Pro/γCH2Pro
RCH2Pro/δCH2Pro
ArC6H/C2-CH3
3.2
1.8
2.4
2.9
2.6
3.6
3.0
2.1
2.6
2.7
no NOE 3.4 3.0
2.1
2.1 2.2
2.3 2.4
2.7
2.5
no NOE
2.8
0.1 0.3
0.1 0.1
largest distance violation
0.3
0.3
0.1
0.1
0.1
average distance violation 0.1
in molecular dynamics simulations carried out in explicit
DMSO.28 Important energetic features of the molecule
are included by application of the force field used in the
molecular dynamics simulations. Moreover, the inclusion
of explicit solvent molecules improves the energetic
refinement given the large surface area of the molecules.
The resulting conformations of 3, 4-trans, 4-cis, and 6
are represented in Figure 2.
The part of these molecules that consists of acyl and
proline residues adopts one conformation in DMSO. The
(20) (a) Aue, W. P.; Bartholdi, E.; Ernst, R. R. J . Chem. Phys. 1975,
64, 2229-2246. (b) Hurd, R. E. J . Magn. Reson. 1990, 87, 422-428.
(21) Braunschweiler, L.; Ernst, R. R. J . Magn. Reson. 1983, 53, 521-
528.
(22) (a) Mu¨ller, L. J . Am. Chem. Soc. 1979, 101, 4481-4484. (b)
Hurd, R. E.; J ohn, B. K. J . Magn. Reson. 1991, 91, 648-653.
(23) J eener, J .; Meier, B. H.; Bachmann, P.; Ernst, R. R. J . Chem.
Phys. 1979, 71, 4546-4553.
(24) COLLECT, data collection software, KappaCCD Reference
Manual; Nonius, B. V.: Delft, The Netherlands, 1999.
(25) Otwinowski, Z.; Minor, W. In Methods in Enzymology; Carter,
C. W., J r., Sweet, R. M., Eds.; Academic Press: New York, 1996; pp
276, 307.
(26) Sheldrick, G. M. Acta Crystallogr., Sect. A 1990, 46, 467-473.
(27) Sheldrick, G. M. SHELXL-97, Program for crystal structure
refinement; University of Go¨ttingen: Germany, 1997; pp 2-5, 2-6.
(28) Mierke, D. F.; Kessler, H. J . Am. Chem. Soc. 1991, 113, 9466-
9470.
Only trans conformers of the peptide bond were found
in compounds 3, 6, and 7, showing that in these cases
(29) (a) J ohnson, C. K. ORTEP II, ORNL Report-3794, revised; Oak
Ridge National Laboratory: Tennessee, 1971. (b) Farrugia, L. J .
ORTEP-3, Implementation for Windows. Version 1.05; University of
Glasgow: Scotland, 1999.
(30) Spek, A. L. PLATON; University of Utrecht: The Netherlands,
1998.

7048 J . Org. Chem., Vol. 66, No. 21, 2001
Breznik et al.
F igu r e 3. Structures of 3 (left) and 4 (right), showing the atomic numbering scheme. Ellipsoids are drawn at 30% probability
level.
F igu r e 4. Stereoviews of the molecular packing for 3 (left) and 4 (right). Hydrogens are omitted for clarity.
the S conformation of the acyl residue in 3 and 7 and
the R conformation of the acyl residue in 6 constrain the
proline amide bond angle and consequently the confor-
mation of the proline peptide bond to trans. This might
be due to the steric effect of the methyl group on the
stereogenic center. We anticipate that both acids 5 and
8 can be used as peptidomimetic building blocks that lock
the proline peptide bond to trans in different types of acyl-
Pro molecules. Conformational analysis of benzyl (2S)-
1-{[(2R)-2,6-dimethyl-3-oxo-3,4-dihydro-2H-1,4-benzoxazin-
2-yl]carbonyl}-2-pyrrolidinecarboxylate (4) (Figures 1, 2b,
2c) revealed 78% trans and 28% cis conformer of proline
peptide bond, showing that the cis/ trans ratio is only
slightly affected in this case.
is due to the stereoisomerism and not to hydrophobic
collapse31 of the molecule.
Exp er im en ta l Section
Melting points were taken on a Reichert hot stage micro-
scope and are uncorrected. IR spectra were recorded on a
Perkin-Elmer 1600 series FTIR spectrometer as KBr disks.
Optical rotations were measured on a Perkin-Elmer 1241 MC
polarimeter. The reported specific rotations are average values
of 10 successive measurements using an integration time of 5
s. Elemental analyses were performed at the Faculty of
Chemistry and Chemical Technology, University of Ljubljana
on a Perkin-Elmer C,H,N Analyzer 240 C. Mass spectra were
obtained on a Autospec Q, VG-analytical mass spectrometer
using ionizing voltage of 70 eV by electron impact.
Crystallographic data of 4 showed only the cis confor-
mation of the proline peptide bond, indicating that, in
the solid state, the cis conformation is energetically
favored over trans. The pyrrolidine ring puckering of all
of the structures of diastereomeres obtained in solution
and solid state was Câ-exo/Cγ-endo, indicating that this
most favorable conformation is not affected by any
structural changes. The results of NMR conformational
analysis suggest that the benzyl part of the molecule is
loose, allowing the aromatic ring to move around freely.
No NOE peaks were found between the protons of the
aromatic ring and the benzyl group in 3, 4, and 6,
confirming that enhancement of the trans conformation
P r ep a r a tion of Ben zyl (2S)-1-[(2R/S)-(2,6-Dim eth yl-3-
oxo-3,4-d ih yd r o-2H-1,4-ben zoxa zin -2-yl)ca r bon yl]-2-p yr -
r olidin ecar boxylate (2). To a stirred solution of 2,6-dimethyl-
3-oxo-3,4-dihydro-2H-1,4-benzoxazine-2-carboxylic acid (1) (663
mg, 3.0 mmol), synthesized as described previously,18 and
benzyl (2S)-2-pyrrolidinecarboxylate hydrochloride (725 mg,
3.0 mmol) in dry DMF (12 mL) at 0-2 °C were added DPPA
(0.81 mL, 3.78 mmol) and triethylamine (1.05 mL, 7.56 mmol).
The mixture was stirred for 1 h at 0-2 °C and then for 60 h
at room temperature. Ethyl acetate (60 mL) was added, and
the mixture was extracted successively with 10% (w/w) citric
(31) Rich, D. H. Perspectives in Medicinal Chemistry; VCH: Basel,
1993; pp 15-25.

cis/trans Ratio of the Proline Peptide Bond
J . Org. Chem., Vol. 66, No. 21, 2001 7049
acid (3 × 5 mL), H2O (3 × 5 mL), saturated NaHCO3 solution
(3 × 5 mL), H2O (3 × 5 mL), and saturated NaCl solution (3
× 5 mL). The combined organic extracts were dried over
MgSO4 and filtered, and the solvent was evaporated to give a
crude product 2 (992 mg, 81%) as a white solid foam. The
diastereomeres were separated by column chromatography on
silica gel using diethyl ether as eluant to give compounds 3
and 4.
2-(4-methyl-2-nitrophenoxy)-3-oxopropanoyl]-2-pyrrolidine-
carboxylate (6) (470 mg, 1 mmol) (because of rapid decarboxy-
lation 6 was not purified but used immediately to obtain 7)
was dissolved in MeOH and hydrogenated over 10% Pd-C (47
mg) under normal pressure for 24 h at room temperature using
99.9% hydrogen [Messer hydrogen 3.0]. The catalyst was
filtered off, and the filtrate was evaporated in vacuo yielding
crude 7 (292 mg, 92%) as brownish crystals. Crude product 7
was purified by preparative thin-layer chromatography on 2
mm Merck TLC plates coated with silica gel 60 F254 using
Ben zyl (2S)-1-{[(2S)-2,6-d im et h yl-3-oxo-3,4-d ih yd r o-
2H-1,4-ben zoxa zin -2-yl]ca r bon yl}-2-p yr r olid in eca r boxy-
la te (3) (282 mg, 28%), colorless crystals: mp 144-148 °C;
BuOH/AcOH/H2O 50/10/40 as eluant: [R]22 ) +19.6 (c 0.22,
D
[R]22 ) +14.5 (c 0.29, MeOH); IR (KBr) 3252, 2983, 1737,
MeOH); IR (KBr) 3452, 2976, 1695, 1613, 1518, 1450, 1382,
D
1
1704, 1643, 1519, 1449, 1100, 995, 824, 736, 584 cm-1; 1H NMR
[ppm] δ 1.57 (s, 3H, C2-CH3), 1.71-2.00 (m, 3H, âCHAPro,
γCH2Pro), 2.04-2.13 (m, 1H, âCHBPro), 2.22 (s, 3H, Het-CH3),
3.69-3.81 (m, 2H, δCH2Pro), 4.27 (dd, 1H, J ) 8.7 Hz, J )
4.9 Hz, RCHPro), 5.08{5.14} (ABsystem, 2H, J ) 12.8 Hz, OCH2-
ArBn), 6.70 (d, 1H, J ) 1.5 Hz, HetC5H), 6.75 (dd, 1H, J ) 8.3
1233, 1133, 813, 532 cm-1; H NMR [ppm] δ 1.59 (s, 3H, C2-
CH3), 1.66-1.73 (m, 3H, âCHPro, γCH2Pro), 1.83-1.87 (m, 1H,
âCHPro), 2.21 (s, 3H, Het-CH3), 3.62-3.70 (m, 2H, δCH2Pro),
3.98 (dd, 1H, J ) 7.9 Hz, J ) 3.9 Hz, RCHPro), 6.64 (s, 1H,
HetC5H), 6.68-6.74 (m, 1H, HetC7H), 6.94 (d, 1H, J ) 8.1 Hz,
HetC8H), 10.57 (s, 1H, HetCONH); MS [EI] m/z 318 (M+, 6.5%),
177 (100%); MS [FAB] m/z 319 (MH+, 30%), 341 [MNa+]
(100%); HRMS calcd for C16H18N2O5 318.122800, found
318.121572.
Hz, J ) 1.5 Hz, HetC7H), 6.95 (d, 1H, J ) 8.3 Hz, HetC8H),
2-6
7.30-7.42 (m, 5H, Ar
C
H), 10.81 (s, 1H, HetCONH); MS
Bn
[FAB] m/z 408 (MH+, 16%), 176 (100%). Anal. Calcd for
23H24N2O5: C, 67.63; H, 5.92; N, 6.86. Found: C, 67.38; H,
C
P r ep a r a tion of Ben zyl (2S)-1-{[(2S)-2,6-Dim eth yl-3-
oxo-3,4-d ih yd r o-2H-1,4-ben zoxa zin -2-yl]ca r bon yl}-2-p yr -
r olid in eca r boxyla te (3) via 7. (2S)-1-{[(2S)-2,6-Dimethyl-
3-oxo-3,4-dihydro-2H-1,4-benzoxazin-2-yl]carbonyl}-2-pyr-
rolidinecarboxylic acid (7) (318 mg, 1 mmol) (due to rapid
decarboxylation, 7 was used immediately for the synthesis of
3) and KF (145 mg, 2.5 mmol) were stirred and heated to 60
°C, at which point (bromomethyl)benzene (188 mg, 1.1 mmol)
was added. The mixture stirred for 6 h at 60 °C, poured on
crushed ice, and extracted with diethyl ether (3 × 30 mL). The
combined organic extracts were extracted successively with
10% (w/w) citric acid (3 × 5 mL), H2O (3 × 5 mL), saturated
NaHCO3 solution (3 × 5 mL), H2O (3 × 5 mL), and saturated
NaCl solution (3 × 5 mL). The organic layer was dried over
MgSO4, filtered, and evaporated in vacuo. The crude product
was crystallized from MeOH to give 3 (237 mg, 57%) as
colorless crystals. The product was in all respects identical to
3 obtained by separation of diastereomers of 2.
5.78; N, 6.89.
Ben zyl (2S)-1-{[(2R)-2,6-d im et h yl-3-oxo-3,4-d ih yd r o-
2H-1,4-ben zoxa zin -2-yl]ca r bon yl}-2-p yr r olid in eca r boxy-
la te (4) (135 mg, 14%), colorless crystals: mp 166-169 °C;
[R]22 ) -65.5 (c 0.33, MeOH); IR (KBr) 3190, 2978, 1755,
D
1690, 1643, 1519, 1455, 1108, 985, 885, 732, 591 cm-1; 1H NMR
[ppm] δ 1.41*/1.61 (s, 3H, C2-CH3), 1.72-2.14 (m, 4H,
âCHAPro, âCHBPro, γCH2Pro), 2.16/2.22* (s, 3H, Het-CH3),
3.38-3.51* (m, δCH2Pro)/3.64-3.88 (m, 2H, δCHAPro,
δCHBPro), 4.27 (dd, J ) 8.7 Hz, J ) 5.2 Hz)/4.48-4.52* (m)
(1H, RCHPro), 4.72{4.92} (ABsystem, J ) 12.8 Hz)/4.80*{5.01*}
(ABsystem, J ) 12.4 Hz) (2H, OCH2-ArBn), 6.65-6.97 (m, 3H,
HetC5,7-8H), 7.04-7.35(m, 5H, ArBnC2-6H), 10.73(10.92)* (s,
1H, HetCONH)*; MS [FAB] m/z 408 (MH+, 11%), 176 (100%).
Anal. Calcd for C23H24N2O5: C, 67.63; H, 5.92; N, 6.86.
Found: C, 67.24; H, 5.83; N, 6.78. *Signals of protons of the
cis isomer.
Gen er a l P r oced u r e for EDC-Ca ta lyzed Cou p lin g Used
To P r ep a r e Am id es 6 a n d 3. To a stirred solution of the
acid 5 or 8 (1 mmol), benzyl (2S)-2-pyrrolidinecarboxylate
hydrochloride (242 mg, 1.0 mmol), N-methylmorpholine (0.36
mL, 3.3 mmol), and HOBT × H2O (176 mg, 1.3 mmol) in dry
DMF (10 mL) was added EDC hydrochloride (249 mg, 1.3
mmol) at -10 °C. The mixture was stirred for 1 h at -10 °C
and then for 24 h at room temperature. Ethyl acetate (70 mL)
was added, and the mixture was extracted successively with
10% (w/w) citric acid (3 × 5 mL), H2O (3 × 5 mL), saturated
NaHCO3 solution (3 × 5 mL), H2O (3 × 5 mL), and saturated
NaCl solution (3 × 5 mL). The combined organic extracts were
dried over MgSO4 and filtered, and the solvent was evaporated
in vacuo.
P r ep a r a tion of Ben zyl (2S)-1-{[(2S)-2,6-Dim eth yl-3-
oxo-3,4-d ih yd r o-2H-1,4-ben zoxa zin -2-yl]ca r bon yl}-2-p yr -
r olid in eca r boxyla te (3) via 8. (2S)-2,6-Dimethyl-3-oxo-3,4-
dihydro-2H-1,4-benzoxazine-2-carboxylic acid (8) (221 mg, 1
mmol), synthesized as previously described,19 yielded 3 (277
mg, 68%) as colorless crystals after EDC-catalyzed coupling
with (2S)-2-pyrrolidinecarboxylate hydrochloride (242 mg, 1
mmol) and crystallization from MeOH. The product was in all
respects identical to 3 obtained by separation of diastereomers
of 2.
NMR Sp ectr oscop y. 1D and 2D 1H NMR spectra were
obtained on a Bruker Avance DPX 300 instrument operating
at 300.15 MHz. All spectra were recorded in DMSO at 302 K,
and chemical shifts were calibrated using tetramethylsilane
as internal standard. Sample concentrations varied among
different experiments but were generally between 10 and 25
mM. Typically, COSY,20 TOCSY,21 and HMQC22 experiments
were acquired with 8-16 transients of 1024-2048 complex
points per transient for each of 512 d1 increments. The data
matrix of 512 d1 points was zero-filled to 1024 points.
NOESY23 spectra of 3,4, 6 and 7 were recorded using TPPI
procedure using mixing time of 150 ms. In d1 dimension 512
real data points were collected using 8 acquisitions per FID
for 3, 4, and 7, and 16 acquisitions per FID for 6, with 2.5 s
relaxation delay; 2048 real points were obtained in d2 dimen-
sion. All spectra were processed with XWINNMR on a Silicon
Graphics Indy workstation and WINNMR on a personal
computer. Distances were determined from the integral in-
tensities of NOE cross-peaks, which were measured by the
integration routine within the WINNMR software. The pseudo-
atom corrections were added for the methylene and the methyl
groups, and the ( 10% were applied to the distances, to
produce the upper and lower limit constraints.
Ben zyl (2S)-1-[(2R)-3-Meth oxy-2-m eth yl-2-(4-m eth yl-2-
n it r op h en oxy)-3-oxop r op a n oyl]-2-p yr r olid in eca r b oxy-
late (6). (2R)-3-Methoxy-2-methyl-2-(4-methyl-2-nitrophenoxy)-
3-oxopropanoic acid (5) (283 mg, 1 mmol), synthesized as
previously described,19 yielded 6 (409 mg, 87%) as a colorless
oil after EDC-catalyzed coupling with S-proline benzyl ester
hydrochloride. Product was purified by column chromatogra-
phy on silica gel using diethyl ether/MeOH/CHCl3 200/10/1 as
eluant: [R]22 ) -18.9 (c 0.27, MeOH); IR (NaCl) 1738, 1649,
D
1530, 1411, 1257, 1120, 801 cm-1; H NMR [ppm] δ 1.60 (s,
1
3H, C2-CH3), 1.79-1.87 (m, 3H, âCHAPro, γCH2Pro), 2.25-
2.31 (m, 1H, âCHBPro), 2.28 (s, 3H, Ar-CH3), 3.53-3.67 (m,
2H, δCH2Pro), 3.76 (s, 3H, COOCH3) 4.47 (dd, 1H, J ) 8.6
Hz, J ) 5.5 Hz, RCHPro), 5.17 (s, 2H, OCH2-ArBn), 7.12 (d,
1H, J ) 7.1 Hz, ArC6H), 7.22 (dd, 1H, J ) 7.1 Hz, J ) 1.8 Hz,
ArC5H), 7.33-7.39 (m, 5H, ArBnC2-6H), 7.68 (d, 1H, J ) 1.8
Hz, ArC3H); MS [EI] m/z 470 (M+, 2.5%), 335 (100%); MS
[FAB] m/z 471 (MH+, 30%), 91 (100%); HRMS calcd for
C
24H26N2O8 470.169900, found 470.168916.
P r ep a r a tion of (2S)-1-{[(2S)-2,6-Dim eth yl-3-oxo-3,4-d i-
h yd r o-2H -1,4-b en zoxa zin -2-yl]ca r b on yl}-2-p yr r olid in e-
ca r boxylic a cid (7). Benzyl (2S)-1-[(2R)-3-methoxy-2-methyl-
Molecu la r Mod elin g. Molecu la r Dyn a m ics. MD simula-
tions were performed with DISCOVER (consistent valence
force field) and INSIGHT II from Biosym Technologies. The

7050 J . Org. Chem., Vol. 66, No. 21, 2001
Breznik et al.
molecule with all atoms treated explicitly was centered in a
box (x ) y ) z ) 40 Å) using three-dimensional periodic
boundary conditions. The dielectric constant was set to 1.0.
Neighbor lists for calculation of nonbonded interactions were
updated every 10 fs within a radius of 14 Å. The actual
calculation of nonbonded interactions was carried out up to a
radius of 12 Å without use of switching functions. A time step
of 1 fs was employed for the MD simulation. The simulation
protocol consisted of two minimization cycles (steepest descent
and conjugate gradients), first with the solute fixed and then
with all of the atoms allowed to move freely. The convergence
criterion was 1 kcal Å-1. The initial MD phase of the calcula-
tion involved a gradual heating, starting from 100 K and then
increasing to 150, 200, 250, and finally to 300 K in steps of
0.5, 0.5, 5, 1, and 5 ps, respectively, each by direct scaling of
the velocities. The NMR-derived distance restraints with a
force constant of 10 kcal mol-1 Å-1 were applied during the
complete simulation. Configurations were saved every 1 ps for
another 200 ps of dynamics. The last 100 ps of the trajectory
were used for analysis of NOE distance violations and dihedral
angles.
X-r a y Cr ysta llogr a p h y. Compounds 3 and 4 crystallize
from methanol. Colorless prismatic crystals were cut, fixed on
a glass fiber with silicon grease, and mounted in a nitrogen
stream (Oxford Cryosystem cooler 700) at 200 K on a Nonius
Kappa CCD area-detector diffractometer. Data collection was
performed using Mo KR monochromated radiation (λ ) 0.71073
Å, 55 kV, 32 mA) with the CCD detector at 30 mm from the
sample. A selection of æ and ω scans of 1° at different ψ and
κ settings using the program COLLECT24 was used in several
sets. The raw data were processed to produce conventional
data using the program DENZO-SMN.25 Both structures were
solved by standard automatic direct methods using SHELXS-
9726 and were refined by full-matrix least squares refinement
(on F2) using SHELXL-97.27 Non-hydrogens were refined with
anisotropic displacement parameters, hydrogens were refined
with isotropic displacement parameters. In the absence of
suitable anomalous scatterers for MoKR radiation, determi-
nation of the absolute configuration was not possible. It was
assigned, therefore to agree with the known chirality of the
L-proline moiety (C22 or CR in S configuration) and the Friedel
diffraction data were merged accordingly.
Con clu sion
For the reasons stated in the Introduction, it is
desirable to have the proline peptide bond constrained
in the trans conformation. It has been demonstrated that
both acids 5 and 8 can be used as peptidomimetic
building blocks that achieve this aim in different types
of acyl-Pro molecules. Both compounds are shown to be
easily accessible and can be coupled with proline in good
yield.
The solution structures of compounds 3, 4, 6, and 7
and crystal structures of 3 and 4 clearly show that the
absolute configuration of the chiral carbon at position 2
affects the cis/ trans ratio of acyl-Pro bond. Conforma-
tional analysis of compounds 3, 6, and 7, together with
crystallographic analysis of 3, proves that these com-
pounds exhibit only trans conformation of the proline
peptide bond. On the other hand the solution structure
of diastereomer 4 revealed a mixture of 72% trans and
28% cis conformer. In these cases, it is evident that the
conformation of the proline peptide bond strongly de-
pends on the stereochemistry of the asymmetric center
in the acyl residue and that the cis/ trans ratio of the
peptide bond in these types of molecules can be tailored
without modification of proline.
Ack n ow led gm en t. The financial support of the
Ministry of Science and Technology, Republic of Slov-
enia, through grants PS-511-103 and PS-787-502 is
gratefully acknowledged. The authors would like to
thank Prof. Roger Pain for critically reading the manu-
script.
Su p p or tin g In for m a tion Ava ila ble: X-ray crystallo-
graphic data for compounds 3 and 4. This material is available
free of charge via the Internet at http://pubs.acs.org.
J O0159439