Development and conformational analysis of a pseudoproline-containing turn mimic

Article abstract of DOI:10.1021/jo026233l

The liquid-phase synthesis and the conformational analysis of a small library of fully protected tetramers containing L-pyroglutamic acid (L-pGlu), (4S,5R)-4-methyl-5-carboxybenzyloxazolidin-2-one (L-Oxd), or (4R,5S)-4-methyl-5-carboxybenzyloxazolidin-2-o

Full text of DOI:10.1021/jo026233l

Develop m en t a n d Con for m a tion a l An a lysis of a
P seu d op r olin e-Con ta in in g Tu r n Mim ic†
Gianluigi Luppi, Donato Lanci, Valerio Trigari, Marco Garavelli, Andrea Garelli, and
Claudia Tomasini*
Dipartimento di Chimica “G. Ciamician”, Alma Mater Studiorum, Universita` di Bologna, Via Selmi 2,
I-40126 Bologna, Italy
tomasini@ciam.unibo.it
Received J uly 24, 2002
The liquid-phase synthesis and the conformational analysis of a small library of fully protected
tetramers containing L-pyroglutamic acid (L-pGlu), (4S,5R)-4-methyl-5-carboxybenzyloxazolidin-
2-one (L-Oxd), or (4R,5S)-4-methyl-5-carboxybenzyloxazolidin-2-one (D-Oxd) as residue i + 1 are
reported to test the tendency of these oligomers to assume a â-hairpin conformation. The most
promising molecule is Boc-L-Val-D-Oxd-Gly-L-Ala-OBn, which assumes a preferential â-turn
1
conformation in CDCl3, as shown by IR and H NMR analysis. These findings have been confirmed
by DFT calculations, which provide an interpretation for the available experimental data and agree
with the reported observations.
In tr od u ction
ing D-Pro-Gly units.6 Imperiali studied sequences includ-
ing an L-Pro at the (i + 1) position and a D-amino residue
at the (i + 2) position and demonstrated that Ac-L-Val-
L-Pro-D-Ser-L-His-NH2 in particular adopts a significant
amount of reverse-turn character both in water and in
dimethyl sulfoxide.7
In this paper, we will describe a study on the synthesis
and the conformational analysis of fully protected tet-
ramers containing pseudoprolines as i + 1 residues
(Figure 1).
Reverse turns are common motifs in protein structures
and account for up to one-third of the residues in globular
proteins.1 A turn can be defined as the site where the
peptide changes its overall direction, and different types
of turns have been described depending on how many
residues are involved in the loop.2 Among them, the most
common naturally occurring type is the â-turn, which
involves at least four residues, with a hydrogen-bonded
amide proton between the carbonyl of residue i and the
NH group of residue i + 3. The γ-turns occur less often
and usually involve a hydrogen-bonded amide proton
between the carbonyl of residue i and the NH group of
residue i + 2. In the de novo design of a reverse-turn
conformation, the introduction of conformationally con-
strained analogues can be very useful.3 For instance,
Lubell pursued two strategies for generating peptide
mimics: the first employs the use of bicycles to constrain
a dipeptide unit, while the second uses the steric interac-
tions of bulky ring substituents to influence the geometry
and conformation of peptide amide bonds.4 A different
approach consists of the introduction of a D-amino acid
in the peptide chain: Gellman demonstrated that switch-
ing from D-Pro to L-Pro in the linker prevents parallel
sheet interactions between L-strand residues,5 and Bal-
aram built a four-stranded â-sheet structure by introduc-
We have recently reported the synthesis and the
conformational analysis of homooligomers of L-pyro-
(3) For some recent examples see: (a) Fisk, J . D.; Powell, D. R.;
Gellman, S. H. J . Am. Chem. Soc. 2000, 122, 5443. (b) Langenhan, J .
M.; Fisk, J . D.; Gellman, S. H. Org. Lett. 2001, 3, 2559. (c) Chung, Y.
J .; Huck, B. R.; Christianson, L. A.; Stanger, H. E.; Krautha¨user, S.;
Powell, D. R.; Gellman, S. H. J . Am. Chem. Soc. 2000, 122, 3995. (d)
Woll, M. G.; Lai, J . R.; Guzei, I. A.; Taylor, S. J . C.; Smith, M. E. B.;
Gellman, S. H. J . Am. Chem. Soc. 2001, 123, 11077. (e) Gennari, C.;
Mielgo, A.; Potenza, D.; Scolastico, C. Eur. J . Org. Chem. 1999, 379.
(f) Belvisi, L.; Gennari, C.; Mielgo, A.; Potenza, D.; Scolastico, C. Eur.
J . Org. Chem. 1999, 389. (g) Belvisi, L.; Gennari, C.; Madder, A.;
Mielgo, A.; Potenza, D.; Scolastico, C. Eur. J . Org. Chem. 2000, 695.
(h) Nowick, J . S. Acc. Chem. Res. 1999, 32, 287. (i) Nowick, J . S.; Tsai,
J . H.; Bui, Q.-C. D.; Maitra, S. Soc. J . Am. Chem. Soc. 1999, 121, 8409.
(j) Holmes, D. L.; Smith, E. M.; Nowick, J . S. J . Am. Chem. Soc. 1997,
119, 7665. (k) Nowick, J . S.; Insaf, S. J . Am. Chem. Soc. 1997, 119,
10903. (l) Bolm, C.; Mu¨ller, D.; Hackenberger, C. P. R. Org. Lett. 2002,
893. (m) Golebiowski, A.; Klopfenstein, S. R.; Shao, X.; Chen, J . J .;
Colson, A.-O.; Grieb, A. L.; Russell, A. F. Org. Lett. 2000, 2, 2615.
(4) (a) Halab, L.; Gosselin, F.; Lubell, W. D. Biopolymers (Peptide
Sci.) 2000, 55, 101. (b) Polyak, F.; Lubell, W. D. J . Org. Chem. 2001,
66, 1171. (c) Gosselin, F.; Lubell, W. D. J . Org. Chem. 2000, 65, 2163.
(d) Halab, L.; Lubell, W. D. J . Am. Chem. Soc. 2002, 124, 2474.
(5) (a) Fisk, J . D.; Gellman, S. H. J . Am. Chem. Soc. 2001, 123, 343.
(b) Fisk, J . D.; Powell, D. R., Gellman, S. H. J . Am. Chem. Soc. 2000,
122, 5443.
* To whom correspondence should be addressed. Phone: +39-051-
2099511. Fax: +39-051-2099456.
† Dedicated to Professor Gianfranco Cainelli in the occasion of his
70th birthday.
(1) (a) Lewis, P. N.; Momany, F. A.; Scheraga, H. A. Biochim.
Biophys. Acta 1973, 303, 211. (b) Chou, P. Y.; Fasman, G. D. J . Mol.
Biol. 1977, 115, 135. (c) Wilmot, C. M.; Thornton, J . M. J . Mol. Biol.
1988, 203, 211.
(2) For leading references, see: (a) Sibanda, B. L.; Thornton, L. M.
Nature 1985, 316, 170. (b) Sibanda, B. L.; Blundell, T. L.; Thornton,
L. M. J . Mol. Biol. 1989, 206, 759. (c) Sibanda, B. L.; Thornton, L. M.
J . Mol. Biol. 1993, 229, 428.
(6) Das, C.; Raghothama, S.; Balaram. P. Chem. Commun. 1999,
967.
(7) (a) Imperiali, B.; Kapoor, T. M. Tetrahedron 1993, 49, 3501. (b)
Imperiali, B.; Fisher, S. L. J . Org. Chem. 1992, 57, 757. (c) Imperiali,
B.; Fisher, S. L.; Moats, R. A.; Prins, T. J . J . Am. Chem. Soc. 1992,
114, 3182. (d) McDonnel, Kevin A.; Imperiali B. J . Am. Chem. Soc.
2002, 124, 428.
10.1021/jo026233l CCC: $25.00 © 2003 American Chemical Society
Published on Web 02/07/2003
1982
J . Org. Chem. 2003, 68, 1982-1993

Analysis of a Pseudoproline-Containing Turn Mimic
To study and to compare their conformational prefer-
ences, conformational analysis and DFT calculations
have been performed on the most interesting molecules.
Resu lts a n d Discu ssion
Syn th esis. We have identified some oligomeric struc-
tures as good candidates to check whether the imide
moiety favors or disfavors the formation of a â-hairpin
conformation. First, we have prepared some oligomers
containing L-pGlu as residue i + 1, and we have varied
the residue i (L-Ala and L-Val) and the residue i + 2 (Gly
and Aib) (Aib-OH ) 2-aminoisobutanoic acid). Gly was
chosen because of its high occurrence into natural
reverse-turn conformations and Aib was chosen because
it is highly helycogenic.11
F IGURE 1. General structure of the fully protected tetramers.
F IGURE 2. Preferential conformation of the imidic bond,
which accounts for the anomalous chemical shift of the CH-R
proton chemical shift. This effect is due to the tendency of the
two carbonyls to lie apart one from the other and is enhanced
by the formation of a CdO‚‚‚H-C hydrogen bond, which
stabilizes the structure by approximately 1.4 kcal/mol.8
L-Pyroglutamic acid is often present in natural polypep-
tides, but only as an N-terminal amino acid, owing to its
low reactivity of the lactam nitrogen.12 We have recently
developed a straightforward method for the N-derivati-
zation of this compound, simply by reacting the lithium
salt of H-L-pGlu-OBn (OBn ) benzyloxy) with activated
esters of protected R-amino acids.13 The acylation reac-
tions usually occur in high yield. Following this protocol,
we have now prepared the tetramers Boc-L-Ala-L-pGlu-
Gly-L-Ala-OBn 4a , Boc-L-Val-L-pGlu-Gly-L-Ala-OBn 4b,
and Boc-L-Val-L-pGlu-Aib-L-Ala-OBn 5 (Scheme 1). After
transformation of L-pGlu-OH into L-pGlu-OBn 1, the
product was acylated with Boc-L-Ala-OPfp (OPfp )
pentafluorophenyl) or with Boc-L-Val-OPfp to obtain the
fully protected dimers 2a and 2b, which were hydrogeno-
lyzed by catalyzed reaction with H2/Pd. The couplings
with H-Gly-L-Ala-OBn and H-Aib-L-Ala-OBn afforded the
desired tetramers 4a ,b and 5 in satisfactory overall
yields.
A similar molecule was prepared from L-Oxd by acyl-
ation of H-L-Oxd-OBn 6 with Boc-L-Val-OPfp in the
presence of diisopropylethylamine (DIEA) and dimeth-
ylaminopyridine (DMAP) in DMF (Scheme 2). The yield
of this reaction is not very high, but any attempt to
ameliorate this result failed. For instance, the protocol
utilized for the acylation of H-L-pGlu-OBn 1 (LiHMDS,
OPfp ester in dry THF) afforded only a complex mixture.
The deprotection of the ester moiety, followed by coupling
with H-Gly-L-Val-OBn, afforded the fully protected tet-
ramer 9 in good yield.
glutamic acid (L-pGlu)8 and of (4S,5R)-4-methyl-5-car-
boxybenzyloxazolidin-2-one (L-Oxd),9 which have been
synthesized up to the tetramer and to the pentamer level,
respectively. We have demonstrated that those new
polyimides behave as rigid spacers, owing to the presence
of the endocyclic carbonyl, which strictly imparts a trans
conformation to the adjacent peptide bond. This effect is
due to the tendency of the two carbonyls to lie apart one
from the other and is enhanced by the formation of a
CdO‚‚‚H-C hydrogen bond, which stabilizes the struc-
ture of about 1.4 kcal/mol (Figure 2). X-ray diffraction
analysis of Boc-(L-pGlu)2-OH (Boc ) tert-butyloxycar-
bonyl, pGlu-OH ) pyroglutamic acid) and DFT calcula-
tions furnished a confirmation of this finding.8-10
Here, we extend our studies on the imidic bond
conformational behavior to some short oligopeptides
containing L-pGlu, L-Oxd, or D-Oxd (trans-(4R,5S)-4-
carboxy-5-methyloxazolidin-2-one) as residue i + 1. In
naturally occurring reverse turns, this residue is often
L-Pro, so by introducing pseudoprolines, we want to check
whether these motifs favor or disfavor the formation of
a â-hairpin secondary structure. Some L-Pro-containing
tetrapeptides have also been synthesized as reference
compounds.
A similar approach was followed for the preparation
of tetramers Boc-L-Val-D-Oxd-Gly-L-Ala-OBn 13 and Boc-
L-Val-D-Oxd-Aib-L-Ala-OBn 14 (Scheme 3).
Finally, we prepared two fully protected tetrapeptides,
containing L-Pro as residue i + 1: Boc-L-Val-L-Pro-Gly-
L-Ala-OBn 15 and Boc-L-Val-L-Pro-Aib-L-Ala-OBn 16
(Figure 3). These molecules have been synthesized by
liquid-phase synthesis, and the couplings have been
carried out in good yields following the HOBt/NMM/EDCI
protocol [HOBt ) 1-hydroxybenzotriazole; NMM ) N-
(8) Bernardi, F.; Garavelli, M.; Scatizzi, M.; Tomasini, C.; Trigari,
V.; Crisma, M.; Formaggio, F.; Peggion, C.; Toniolo, C. Chem. Eur. J .
2002, 8, 2516.
(9) (a) Lucarini, S.; TomasiniC. J . Org. Chem. 2001, 66, 727. (b)
Tomasini, C.; Trigari, V.; Lucarini, S.; Bernardi, F.; Garavelli, M.;
Peggion, C.; Formaggio, F.; Toniolo, C. Eur. J . Org. Chem. 2003,
259.
(10) Frisch, M. J .; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J . R.; Zakrzewski, V. G.; Montgomery, J . A.,
J r.; Stratmann, R. E.; Burant, J . C.; Dapprich, S.; Millam, J . M.;
Daniels, A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J .;
Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo,
C.; Clifford, S.; Ochterski, J .; Petersson, G. A.; Ayala, P. Y.; Cui, Q.;
Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;
Foresman, J . B.; Cioslowski, J .; Ortiz, J . V.; Stefanov, B. B.; Liu, G.;
Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R. L.;
Fox, D. J .; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.;
Gonzalez, C.; Challacombe, M.; Gill, P. M. W.; J ohnson, B. G.; Chen,
W.; Wong, M. W.; Andres, J . L.; Head-Gordon, M.; Replogle, E. S.;
Pople, J . A. Gaussian 98, revision A.9; Gaussian, Inc.: Pittsburgh, PA,
1998.
(11) (a) Toniolo, C.; Benedetti, E. Macromolecules 1991, 24, 4004.
(b) Toniolo, C.; Crisma, M.; Formaggio, F.; Valle, G.; Cavicchioni, G.;
Pre´cigoux, G.; Aubry, A.; Kamphuis, J . Biopolymers 1993, 33, 1061.
(12) (a) Ramasubbu, N.; Parthasarathy, R. Int. J . Paptide Res. 1989,
33, 328. (b) Paul, P. K. C.; Osguthorpe, D. J .; Campbell, M. M. J . Chem.
Soc., Perkin Trans. 1 1990, 3363. (c) Paul, P. K. C.; Burney, P. A.;
Campbell, M. M.; Osguthorpe, D. J . Bioorg. Med. Chem. Lett. 1992, 2,
141.
(13) Tomasini, C.; Villa M. Tetrahedron Lett., 2001, 42, 5211.
J . Org. Chem, Vol. 68, No. 5, 2003 1983

Luppi et al.
SCHEME 1. Syn th etic Rou tes for th e Tetr a m er s
4a ,b a n d 5a
SCHEME 3. Syn th etic Rou tes for th e Tetr a m er s
13 a n d 14a
a
Reaction conditions: (i) DIEA (4 equiv), DMAP (0.7 equiv),
Boc-L-Val-OPfp (1.5 equiv), dry DMF, rt, 2 h; (ii) Pd/C 10% (cat.),
MeOH, rt, 2 h; (iii) H-Gly-L-Ala-OBn or H-Aib-L-Ala-OBn (2 equiv),
NMM (4 equiv), HOBt (1.2 equiv), EDCl (1.2 equiv), dry DMF, rt,
16 h.
a
Reaction conditions: (i) LiHMDS (1.2 equiv), dry THF, 0 °C,
30 min; (ii) Boc-L-Ala-OPfp or Boc-L-Val-OPfp (1.5 equiv), dry THF,
0 °C, 1 h; (iii) Pd/C 10% (cat.), MeOH, rt, 2 h; (iv) H-Gly-L-Ala-
OBn (2 equiv), NMM (4 equiv), HOBt (1.2 equiv), EDCl (1.2 equiv),
dry DMF, rt, 16 h; (v) H-Aib-L-Ala-OBn (2 equiv), NMM (4 equiv),
HOBt (1.2 equiv), EDCl (1.2 equiv), dry DMF, rt, 16 h.
SCHEME 2. Syn th etic Rou tes for th e Tetr a m er 9a
F IGURE 3. Tetrapeptides 15 and 16 contain L-Pro as residue
i + 1 and have been prepared as reference compounds by
liquid-phase synthesis, following the HOBt/NMM/EDCI pro-
tocol.
OBn 13, Boc-L-Val-D-Oxd-Aib-L-Ala-OBn 14; and the
reference molecules containing L-Pro at the i + 1 position
were Boc-L-Val-L-Pro-Gly-L-Ala-OBn 15 and Boc-L-Val-
L-Pro-Aib-L-Ala-OBn 16.
a
Reaction conditions: (i) DIEA (4 equiv), DMAP (0.7 equiv),
Boc-L-Val-OPfp (1.5 equiv), dry DMF, rt, 2 h; (ii) Pd/C 10% (cat.),
MeOH, rt, 2 h; (iii) H-Gly-L-Ala-OBn (2 equiv), NMM (4 equiv),
HOBt (1.2 equiv), EDCl (1.2 equiv), dry DMF, rt, 16 h.
IR An a lysis. The IR absorption spectra were obtained
as a 3 mM solution in methylene chloride: at this con-
centration, the intramolecular aggregation is usually
unimportant. Figure 4 shows the absorption bands
between 3500 and 3200 cm-1 of all the synthesized com-
pounds and helps us to see if we have non-hydrogen-
bonded amide proton bands (above 3400 cm-1) or hydro-
gen-bonded amide proton bands (below 3400 cm-1).15 The
IR spectra of tetramers 4a ,b and 9 are very similar, with
a greater band at about 3430 cm-1 (non-hydrogen-bonded
methylmorpholine; EDCI ) 1-(3-dimethylaminopropyl)-
3-ethylcarbodiimide hydrochloride].14
Con for m a tion a l An a lysis. Once we obtained a small
library of fully protected tetrapeptides, their preferential
conformation was analyzed by 1H NMR, IR, and DFT
computational modeling. Moreover, some interesting
information was obtained by electrospray ionization mass
(ESI-MS) analysis. The molecules analyzed were Boc-
L-Ala-L-pGlu-Gly-L-Ala-OBn 4a , Boc-L-Val-L-pGlu-Gly-L-
Ala-OBn 4b, Boc-L-Val-L-pGlu-Aib-L-Ala-OBn 5, Boc-L-
Val-L-Oxd-Gly-L-Ala-OBn 9, Boc-L-Val-D-Oxd-Gly-L-Ala-
(15) For an interpretation of N-H stretches in CH2Cl2, see: (a)
Gardner, R. R.; Liang, G.-B.; Gellman, S. H. J . Am. Chem. Soc. 1995,
117, 3280. (b) Rao, C. P.; Nagaraj, R.; Rao, C. N. R.; Balaram, P.
Biochemistry 1980, 19, 425. (c) Yang, J .; Christianson, L. A.; Gellman,
S. H. Org. Lett. 1999, 1, 11.
(14) Ko¨nig, W.; Geiger, R. Chem. Ber. 1970, 103, 788.
1984 J . Org. Chem., Vol. 68, No. 5, 2003

Analysis of a Pseudoproline-Containing Turn Mimic
F IGURE 4. N-H stretch region FT-IR data for 3 mM samples of 4a ,b, 5, 9, and 13-16 in pure CH2Cl2 at room temperature,
after subtraction of the spectrum of pure CH2Cl2.
amide proton) and another band at about 3390 cm-1
(weakly hydrogen-bonded amide proton). We gather that
in these systems we can both vary the nature of the
L-amino acid as residue i and utilize L-Oxd or L-pGlu as
residue i + 1, without a substantial variation of the IR
spectrum and, possibly, of the preferred conformation of
the molecule. The IR spectrum of the reference compound
Boc-L-Val-L-Pro-Gly-L-Ala-OBn 15 shows a strong band
at 3433 cm-1 and another longer band at 3326 cm-1
(hydrogen-bonded amide proton). The Aib moiety was
J . Org. Chem, Vol. 68, No. 5, 2003 1985

Luppi et al.
TABLE 1. Ch em ica l Sh ifts of th e CHr of Resid u e i for
Tetr a m er s 4a ,b, 5, 9, a n d 13-16
entry
product
residue i
residue i + 1
δ CHR-i (ppm)
1
2
3
4
5
6
7
8
4a
4b
5
Ala
Val
Val
Val
Val
Val
Val
Val
pGlu
5.34
5.37
5.37
5.40-5.43
5.33
5.26
4.25
PGlu
PGlu
L-Oxd
D-Oxd
D-Oxd
L-Pro
L-Pro
9
13
14
15
16
4.25
introduced as residue i + 2 in two tetramers: Boc-L-Val-
L-pGlu-Aib-L-Ala-OBn 5 and the reference compound Boc-
L-Val-L-Pro-Aib-L-Ala-OBn 16. The IR spectra show the
presence of a strong band at about 3420 cm-1 and a weak
shoulder at 3367 cm-1, thus suggesting that in both cases
an equilibrium between a â-turn conformation and an
open form is present.
All these results are quite disappointing because they
suggest that the introduction of the pseudoprolines
L-pGlu and L-Oxd in the tetramer as residue i + 1 does
not facilitate the formation of a reverse-turn conformation
more than the introduction of L-Pro.
Much better results have been obtained with the
tetramers containing D-Oxd as residue i + 1: the
spectrum of Boc-L-Val-D-Oxd-Gly-L-Ala-OBn 13 shows
three bands of nearly the same intensity at 3435 cm-1
(non-hydrogen-bonded amide proton), 3390 cm-1 (weakly
hydrogen-bonded amide proton), and 3331 cm-1 (strongly
hydrogen-bonded amide proton), while the IR spectrum
of Aib-containing 14 shows a strong band at 3419 cm-1
and a weak band at 3356 cm-1, suggesting that in this
case the Aib moiety disfavors the presence of strong
hydrogen bonds. From a general insight, the best can-
didate for a â-hairpin structure should be 13.
F IGURE 5. Significant NOE enhancements of 4b, 5, 9, and
13-16 obtained by performing the NOESY experiments on 5
mM solutions in CDCl3 and utilizing a mixing time of 900 ms.
proton, it will be expected to dramatically move its
chemical shift downfield. In the case of a â-turn confor-
mation, the amide proton (i + 3)-NH should be hydrogen
bonded, while for a γ-turn conformation it should be
(i + 2)-NH. The evaluation of inaccessible NH groups by
1H NMR was performed by adding increasing amounts
of DMSO-d6 to 1 mM tetramer solutions in CDCl3. The
results are reported in Figure 6.
The results obtained for 4a ,b and 9 are very similar
and show that only the i-NH is nearly insensitive to
DMSO, thus suggesting that this proton is hydrogen-
bonded. This is in agreement with the IR result and show
that we can both vary the nature of the L-amino acid at
the i position and utilize L-Oxd or L-pGlu, without a
substantial variation of the preferred conformation of the
molecule. The results obtained from Boc-L-Val-L-Pro-Gly-
L-Ala-OBn 15 show a similar trend. From these data we
can gather that this group of molecules do not show any
preference for a â-hairpin conformation, although they
should take a compact conformation, as shown by the
NOESY experiments.
The introduction of the Aib moiety gives better re-
sults: Boc-L-Val-L-pGlu-Aib-L-Ala-OBn 5 possesses two
hydrogen-bonded amide protons i-H and (i + 3)-NH, i.e.,
NH-Ala and NH-Val and is in agreement with a â-turn
structure. The reference molecule Boc-L-Val-L-Pro-Aib-
L-Ala-OBn 16 shows a similar trend.
The two compounds of the D series behave in a very
different way: Boc-L-Val-D-Oxd-Aib-L-Ala-OBn 14 gives
quite good results (∆δ i-NH: 0.24 ppm; ∆δ (i + 3)-NH:
0.38 ppm), while the titration of 13 affords a very unusual
1H NMR An a lysis. All the spectra of the molecules
containing L-pGlu, L-Oxd, or D-Oxd show a signal between
5 and 6 ppm, which is due to the CHR of the residue i
(Table 1). We have already observed and studied this
outcome,8,9 and it was attributed to the presence of the
heterocycle carbonyl, which strongly deshields the hy-
drogen. The reference compounds 15 and 16 (entries 7
and 8), which contain L-Pro as residue i + 1, in fact do
not produce this effect.
An evaluation of the NOESY spectra (CDCl3) of the
compounds (Figure 5) shows that in any case a significant
NOE was observed between the CHR of the residue i +
1 (L-pGlu for 4b and 5; L-Oxd for 9; D-Oxd for 13 and 14;
L-Pro for 15 and 16) and NH of the residue i + 2 (Gly or
Aib). Furthermore, D-Oxd containing compounds (13 and
14) show an NOE enhancement between NH-Gly and
NH-Ala for 13 and between NH-Aib and NH-Ala for 14
(weak signal).
Those results show that a kind of turn should be
formed; in particular, the proximity of the two amide
protons is indicative of a compact, rather than extended
structure. This outcome has been further detected by
investigation of the DMSO-d6 dependence of NH proton
chemical shift.16 This solvent has a strong hydrogen-
bonding acceptor character and, if it bound to a free NH
(16) (a) Kopple, K. D.; Ohnishi, M.; Go, A. Biochemistry 1969, 8,
4087. (b) Martin, D.; Hauthal, G. In Dimethyl Sulphoxide; Van
Nostrand-Reinhold: Wokingham, U.K., 1975.
1986 J . Org. Chem., Vol. 68, No. 5, 2003

Analysis of a Pseudoproline-Containing Turn Mimic
F IGURE 6. Variation of NH proton chemical shift (ppm) of 4a ,b, 5, 9, and 13-16 as a function of increasing percentages of
DMSO-d6 to the CDCl3 solution (v/v) (concentration: 1 mM).
plot of the three NHs (∆δ ) 0.44, 0.68, and 0.69 ppm),
thus showing that an equilibrium is going on and all the
amide protons are involved. This outcome is very disap-
pointing, as 13 shows both promising IR spectrum and
NOE enhancements and is the best candidate for a stable
â-hairpin conformation. So further analyses were needed
to explain the behavior of 13 in solution.
Figure 7 shows the variation of the NH proton chemical
shift as a function of the temperature for 13 and for its
epimer 9 (used as reference compound) in CDCl3 and in
DMSO-d6. This test points out the presence of hydrogen-
bonded amide protons: in CDCl3 a small temperature
coefficient (e3 ppb/K in absolute value) indicates the
presence of a hydrogen-bonded amide proton,17 while
J . Org. Chem, Vol. 68, No. 5, 2003 1987

Luppi et al.
F IGURE 7. Variation of NH proton chemical shift (ppm) of 9 and 13 of as a function the temperature (°C) for 1 mM samples
measured both in CDCl3 and DMSO-d6 (concentration: 1 mM).
TABLE 2. Ca lcu la ted En er gies a n d Geom etr ica l
P a r a m eter s for Con for m er s IA a n d IB of
MeOOC-(L-Va l)-(L-Oxd )-Gly-(L-Ala )-OMe a n d for
Con for m er s IIA a n d IIB of
MeOOC-(L-Va l)-(D-Oxd )-Gly-(L-Ala )-OMe
F IGURE 8. Preferred conformations of 13 proposed on the
basis of the spectroscopic data.
distance
protons that participate to an equilibrium between
hydrogen-bonded and non-hydrogen-bonded state show
a larger temperature coefficient. In DMSO-d6, the tem-
perature coefficients can be bigger, so that an hydrogen-
bonded amide proton can have a value of 4 ppb/K or
smaller in absolute value.18
For 13, in the two solvents we obtain opposite results.
In CDCl3, both protons NH-Val and NH-Ala are hydrogen
bonded, thus showing that this molecule assumes a
â-turn conformation, like 13A (Figure 8) in structure-
supporting solvents such as chloroform. This outcome is
in full agreement both with the IR spectrum (in CH2Cl2)
con-
entry former
Eabs
(hartree)
Erel
da O1-H4b
âc
(kcal mol-1
)
(Å)
(Å)
(deg)
1
2
3
4
IA
IB
-1598.92130
-1598.92274
+1.46
+0.55
0.00
5.18
9.03
5.31
8.81
1.93
5.55
1.99
5.61
+8.5
+161.5
-29.6
IIA -1598.92362
IIB -1598.92227
+0.85
-176.6
a
b
â-Turns are characterized by d less than 7.00 Å. â-Turns are
characterized by a O1-H4 distance less than 2.50 Å. c The virtual
dihedral angle â indicates a tight â-turn in the range of 0 ( 30°.
and with the NOESY results (in CDCl3). On the contrary,
in a competitive solvent like DMSO-d6, only NH-Gly is a
hydrogen-bonded amide proton, thus suggesting that in
this solvent a γ-turn conformation is preferred (13B or
13C, Figure 8). Again, DMSO-d6 favors a γ-turn confor-
mation of 9, while CDCl3 does not favor any turn of 9,
because only NH-Val is hydrogen-bonded.
The titration outcome of 13 with DMSO-d6 (Table 2,
entry 6) can be easily explained: as CDCl3 favors the
formation of a â-turn conformation and DMSO-d6 favors
(17) (a) Boussard, G.; Marraud, M. J . Am. Chem. Soc. 1985, 107,
1825. (b) Stevens, E. S.; Sugawara, N.; Bonora, G. M.; Toniolo, C. J .
Am. Chem. Soc. 1980, 102, 7048. (c) Gellman, S. H.; Adams, B. R.
Tetrahedron Lett. 1989, 30, 3381. (d) Gellman, S. H.; Dado, G. P.; Liang,
G.-B.; Adams B. R. J . Am. Chem. Soc. 1991, 113, 1164. (e) Yang, J .;
Gellman, S. H. J . Am. Chem. Soc. 1998, 120, 9090-9091. (f) J ones, I.
G.; J ones, W.; North, M. J . Org. Chem. 1998, 63, 1505.
(18) Alonso, E.; Lopez-Ortiz, F.; del Pozo, C.; Peralta, E.; Mac`ıas,
A.; Gonzalez, J . J . Org. Chem. 2001, 66, 6333.
1988 J . Org. Chem., Vol. 68, No. 5, 2003

Analysis of a Pseudoproline-Containing Turn Mimic
a γ-turn conformation, in a mixture of structure-support-
ing and competitive solvents an equilibrium is going on
between open and closed conformations.
ESI-MS An a lysis. Thermospray high-performance
liquid chromatography/mass spectrometry (LC/ESI-MS)
has been successfully used to analyze nonvolatile and
thermally labile compounds. More recently, electrospray
ionization (ESI)19 linked with multiple tandem mass
spectrometry has been employed in the structural char-
acterization of a series of nonvolatile products, including
oligopeptides.20
We have undertaken a systematic analysis of our
tetramers to check their purity and to obtain some
information regarding their behavior in the gas phase.
This information could indeed be very useful for our
research because it has been recently demonstrated that
the gas-phase proteins should be viewed as tools to
examine the interactions that are present in solution.21
From the mass spectral analysis, the presence of the
M + H2O peak is outlined by a main or an abundant peak
for the molecules containing L-pGlu, L-Oxd, or D-Oxd. The
tests were repeated avoiding the contact of the samples
with water (using dry methanol as solvent), and similar
results were obtained. Thus, the M + H2O peaks should
be ascribed to the presence of a water molecule strongly
bonded to the tetrapeptide between two or more func-
tional groups. This result is not surprising; indeed, it has
been recently observed that the presence of cooperative
effects between more than one functional group is crucial
for the hydration of a polypeptide chain. Only the
molecules containing L-pGlu, L-Oxd, and D-Oxd (4a ,b, 5,
9, 13, and 14) show this peak, regardless the presence of
Gly or Aib. On the contrary, molecules containing L-Pro
never show the peak, thus suggesting that these mol-
ecules are unable to bond to the water and that the
heterocycle carbonyl is crucial for this effect.
If we try to draw a hypothetical structure to explain
this result, we can presume that the water is bonded
somehow between the i-NH and the carbonyl of the
heterocycle. This outcome could explain the DMSO-d6
titration trends (Table 2) which highlight that i-NH is
always hydrogen-bonded. Further studies are currently
being made to check the nature of the interaction
between water and oligomers containing L-pGlu, L-Oxd,
or D-Oxd and to analyze if the water can be replaced by
other molecules.
DF T Com p u ta tion a l Mod elin g. In this section, the
results of fully unconstrained DFT optimizations are
presented for tetrameric systems analogous to 9 and 13
(I, MeOCO-L-Val-L-Oxd-Gly-L-Ala-OMe; II, MeOCO-L-
Val-D-Oxd-Gly-L-Ala-OMe, respectively) (see also the
Experimental Section).
F IGURE 9. DFT-optimized structures for the tetrameric
systems I (MeOCO-L-Val-L-Oxd-Gly-L-Ala-OMe) analogous to
9 with partial atom labeling. For clarity, only the H-bonded
hydrogen atoms are shown.
in general, and weak hydrogen bonds in particular (such
as C-H‚‚‚OdC interactions), have to be described, which
can be involved in the stabilization of complex molecular
architectures. In previous works,8,9 we have shown how
weak intramolecular hydrogen bonds (i.e., C-H‚‚‚OdC)
play an important role in the stabilization of conforma-
tional minima for foldameric-type homooligomers. As
shown above, 1H NMR analysis has unveiled the presence
of these features also in the oligomers investigated here
(Table 1). Moreover, the hydrogen bonding network plays
a very important role in the design and stability of â-turn
mimics. Therefore, despite the increased computational
costs, we decided to carry out our calculations at a fully
correlated level (i.e., density functional theory) using both
a well tested functional and basis set.
At first, a reasonable set of starting structures has been
constructed to match the geometrical information re-
vealed by NOESY spectra. Preliminary AM1 optimiza-
tions have been performed on this molecular set, and the
minimum energy points located within this procedure
provided the starting structures for the DFT optimiza-
tions and refinements. Although this procedure does not
guarantee the location of the lowest energy minima,22 as
we will show below the DFT simulations are in agree-
ment with the experimental evidence, and also provide
an interpretation for the available observations. This
provides a further validation for this approach and for
the reliability of the aforementioned optimized structures.
Molecular mechanics and molecular dynamics are the
conventional tools used for modeling oligomers and for
conformational search and analysis. This is a very
convenient approach when big molecular systems are
under investigation. However, the reliability of the
results depends heavily on the quality of the force field
used. This is particularly crucial when hydrogen bonds
(19) (a) Fenn, J . B.; Mann, M.; Meng, C. K.; Wong, S. F.; Whitehouse
C. M. Mass Spectrom. Rev. 1990, 9, 37. (b) Whitehouse, C. M.; Dreyer,
R. N.; Yamashuta, M.; Fenn, J . B. Anal. Chem. 1985, 57, 675.
(20) Karas, M.; Hillenkamp, F. Anal. Chem. 1988, 60, 2299.
(21) J arrold, M. F. Acc. Chem. Res. 1999, 32, 360.
(22) Extended molecular dynamics computations on these tetrameric
oligomers are currently under investigation and will be reported soon.
J . Org. Chem, Vol. 68, No. 5, 2003 1989

Luppi et al.
These computational results provide a rationalization
for the observed preference of 13 to give a â-hairpin
structure, while 9 has been shown to prefer a less
structured conformation. Furthermore, since both â-hair-
pin and open-chain conformations have similar energies
(Table 2), tuning between the two structures can be easily
induced, in principle, by changing external conditions
(i.e., the environment). This can be accomplished, for
example, by changing the solvents: structure-supporting
solvents (such as chloroform) can increase the stability
of more compact conformations such as â-turns, while
highly coordinating competitive solvents (like DMSO) can
invert this stability pattern, as we observed for 13.
Con clu sion s
In this paper, we have shown the liquid-phase synthe-
sis and the conformational analysis of a small library of
fully protected tetramers containing L-pGlu, L-Oxd, D-
Oxd, or L-Pro as residue i + 1 and Gly or Aib as residue
i + 2. These compounds have all been analyzed by IR,
1H NMR, and ESI-MS. The molecules containing D-Oxd
showed a good propensity to the formation of a â-hairpin
conformation. Among them, Boc-L-Val-D-Oxd-Gly-L-Ala-
OBn assumed a preferential â-turn conformation in
chloroform and a preferential γ-turn conformation in
DMSO, while its epimer Boc-L-Val-L-Oxd-Gly-L-Ala-OBn
showed a minor propensity to assume ordered forms.
These findings have been confirmed by DFT calculations,
which provide an interpretation for the available experi-
mental data, and agree with the reported observations.
On the other hand, the introduction of an Aib moiety as
residue i + 2 enhanced the propensity for the â-turn
conformation in the L-series while restraining it in the
D-series.
Furthermore, by LC/ESI-MS analysis, we have high-
lighted that only the tetramers containing L-pGlu, L-Oxd,
or D-Oxd show a strong tendency to chelate a molecule
of water in the gas phase. This outcome suggests that
this behavior is maintained in the liquid phase and that
oligopeptides containing these heterocycles show a re-
markable tendency to chelate water molecules. On the
contrary tetramers containing L-Pro do not follow this
behavior.
F IGURE 10. DFT-optimized structures for the tetrameric
systems II (MeOCO-L-Val-D-Oxd-Gly-L-Ala-OMe) analogous to
13 with partial atom labeling. For clarity, only the H-bonded
hydrogen atoms are shown.
Table 2 reports the results for the DFT-optimized
lowest energy minima of these conformers: two different
minima exist for both molecular systems I (Figure 9) and
II (Figure 10) and are very close in energy. Compounds
IA and IIA represent stable and compact â-hairpin
conformations (Figures 9a and 10a), while IB and IIB
are γ-turn like open structures (Figures 9b and 10b). The
topography of â-turns and their mimic can be described
in terms of a single virtual tortion angle â, defined by
CR1, CR2, CR3, and N4 of the tetrapeptide model and
the interatomic distance d between CR1 and CR4. The
DFT geometry-optimized minima IA, IB, IIA, and IIB
were evaluated according to the rules fixed by Ball:23 (1)
â-turns are characterized by d less than 7.00 Å; (2) the
distance between O1 and H4 should be less than 2.50 Å;
(3) the virtual torsion angle â (defined by the atoms C1-
C2-C3-N4) should be in the range of 0 ( 30°. All these
criteria are fulfilled by conformations IA and IIA, so they
are both tight â-turn conformations. Very remarkably,
the energies comparison shows that the â-turn structure
IIA is more stable than the open form IIB; on the
contrary, for isomer I we obtain the opposite result: the
open form IB is slightly preferred to IA.
Exp er im en ta l Section
Gen er a l Meth od s. Routine NMR spectra were recorded
with spectrometers at 300 or 200 MHz (1H NMR) and at 75 or
50 MHz (13C NMR). Chemical shifts are reported in δ values
relative to the solvent peak of CHCl3, set at 7.27 ppm. Infrared
spectra were recorded with an FT-IR spectrometer. Melting
points were determined in open capillaries and are uncor-
rected. Dry CH2Cl2 and dry DMF were purchased from a
commercial supplier. Peptide bond formation was accom-
plished via standard solution-phase procedures with 1-hy-
droxybenzotriazole, N-methylmorpholine, and 1-(3-dimethyl-
aminopropyl)-3-ethylcarbodiimide hydrochloride.10 The amino
groups and the carboxy groups were orthogonally protected
with tert-butyloxycarbonyl and benzyl groups. H-Gly-L-Ala-OH
was purchased from a commercial supplier.
IR Stu d ies. High-quality infrared spectra (64 scans) were
obtained at 2 cm-1 resolution using a 1 mm NaCl solution cell
and a Nicolet 210 FT-infrared spectrometer. All spectra were
obtained in 3 mM solutions in dry CH2Cl2 at 297 K. All
compounds were dried in vacuo, and all the sample prepara-
tions were performed in a nitrogen atmosphere.
(23) Ball, J . B.; Hughes, R. A.; Alewood, P. F.; Andrews, P. R.
Tetrahedron 1993, 49, 3467.
1990 J . Org. Chem., Vol. 68, No. 5, 2003

Analysis of a Pseudoproline-Containing Turn Mimic
1H NMR Stu d ies. High-quality 1H NMR spectra were
recorded with a Varian Mercury 400. Measurements were
carried out in CDCl3 and in DMSO-d6 using tetramethylsilane
as internal standard. Samples for conformational analysis
were prepared on the benchtop and were typically between 5
and 10 mM. Proton signals were assigned by COSY spectra.
Data for conformational analysis came from NOESY spectra
with typical mixing times of 900 ms.
LC-MS An a lysis. Acetonitrile and methanol for HPLC
were purchased from a commercial supplier. All the samples
were prepared by diluting 1 mg in 5 mL of a 1:1 mixture of
H2O and acetonitrile in pure acetonitrile or in pure methanol.
The samples were analyzed with a liquid chromatography
Agilent Technologies HP1100 equipped with a Zorbax Eclipse
XDB-C8 Agilent and Technologies column (flow rate 0.5 mL/
min) and equipped with a diode-array UV detector (220 and
254 nm). The MSD1100 mass detector was utilized under the
following conditions: mass range 100-2500 uma, positive
scanning, energy of fragmentor 50 V, drying gas flow (nitrogen)
10.0 mL/min, nebulizer pressure 45 psig, drying gas temper-
ature 350 °C, capillary voltage 4500 V.
for C13H20N2O6: C, 51.99; H, 6.71; N, 9.33. Found: C, 51.90;
H, 6.67; N, 9.25.
Boc-L-Va l-L-p Glu -OH (3b). For the synthetic procedure
from 2b, see the preparation of Boc-L-Ala-L-pGlu-OH (3a )
above: yield 97%; mp ) 52-57 °C; [R]D ) -16.1 (c ) 1.0, CH2-
Cl2); IR (CH2Cl2) ν 3430, 1751, 1722, 1698 cm-1 1H NMR
;
(CDCl3) δ 0.82 (d, 3 H, J ) 6.9 Hz), 1.08 (d, 3 H, J ) 6.6 Hz),
1.46 (s, 9 H), 2.11-2.25 (m, 2 H), 2.40-2.55 (m, 1 H), 2.55-
2.80 (m, 2 H), 4.85 (dd, 1 H, J ) 3.6, 9.0 Hz), 5.22 (d, 1 H,
J ) 9.9 Hz), 5.42 (dd, 1 H, J ) 3.3, 9.9 Hz), 9.21 (bs, 1 H); 13
C
NMR (CDCl3) δ 15.9, 20.1, 21.7, 28.5, 30.3, 32.2, 58.0, 58.2,
80.2, 156.4, 174.0, 174.3, 175.1. Anal. Calcd for C15H24N2O6:
C, 54.87; H, 7.37; N, 8.53. Found: C, 54.92; H, 7.42; N, 8.46.
Boc-L-Ala -L-p Glu -Gly-L-Ala -OBn (4a ). H-Gly-L-Ala-OBn
(0.26 mmol, 61 mg), NMM (0.52 mmol, 57 µL), HOBt (0.16
mmol, 22 mg), and EDCI (0.16 mmol, 31 mg) were added to a
stirred solution of Boc-L-Ala-L-pGlu-OH (3a ) (0.13 mmol, 39
mg) in dry DMF (2 mL) under nitrogen atmosphere at room
temperature. The mixture was stirred at room temperature
for 24 h, ethyl acetate (10 mL) was added, and the mixture
was washed with 1 M aqueous HCl (2 × 7 mL) and with an
aqueous saturated solution of NaHCO3 (1 × 10 mL). The
combined organic layers were dried over sodium sulfate, and
the solvent was removed under reduced pressure. The residue
was purified by silica gel chromatography (cyclohexane/ethyl
acetate 2:8 as eluant), and product 4a was obtained pure as a
white solid in 45% yield: mp ) 120-124 °C; [R]D ) -16.8
Boc-L-Ala -L-p Glu -OBn (2a ). LiHMDS (2.2 mmol, 1M sol.
in THF, 2.2 mL) was added to a stirred solution of benzyl (S)-
pyroglutamate 124 (2.0 mmol, 438 mg) in dry THF (7 mL) under
nitrogen atmosphere at 0 °C. The mixture was stirred for 20
min at 0 °C and 40 min at room temperature. A solution of
Boc-L-Ala-L-pGlu-OPfp (2.4 mmol) in dry THF (4 mL) was
added dropwise at 0 °C. The mixture was stirred 20 min at 0
°C and 3 h at room temperature. Water (10 mL) was added,
and the mixture was concentrated (rotary evaporator) to
remove THF, the aqueous layer was extracted with ethyl
acetate (3 × 10 mL), the combined organic layers were dried
over sodium sulfate, and the solvent was removed under
reduced pressure. The residue was purified by silica gel
chromatography (cyclohexane/ethyl acetate 8:2 as eluant) and
obtained as a liquid in 80% yield: [R]D -54.2 (c ) 1.1 CH2-
(c ) 0.2, CH2Cl2); IR (CH2Cl2) ν 3431, 3390, 1746, 1680 cm-1
;
1H NMR (CDCl3) δ 1.37 (d, 3 H, J ) 7.3 Hz), 1.44 (s, 9 H),
1.45 (d, 3 H, J ) 7.2 Hz), 2.08-2.38 (m, 2 H), 2.54-2.66 (m, 1
H), 2.75-2.90 (m, 1 H), 3.81 (dd, 1 H, J ) 4.8, 16.8 Hz), 4.18
(dd, 1 H, J ) 6.6, 16.8 Hz), 4.62 (dq, J ) 6.9 Hz), 4.70 (dd, 1
H, J ) 3.0, 8.4 Hz), 5.19 (AB, J ) 12.3 Hz), 5.20 (bs, 1 H),
5.34 (dq, J ) 7.2 Hz), 7.01 (d, J ) 7.2 Hz, 1 H), 7.25 (bs, 1 H),
7.21-7.43 (m, 5 H); 13C NMR (CDCl3) δ 18.0, 21.9, 28.6, 32.3,
43.2, 48.6, 50.2, 59.5, 67.5, 80.1, 128.3, 128.7, 128.9, 135.5,
155.8, 168.7, 171.4, 173.1, 174.8, 175.9. Anal. Calcd for
1
Cl2); IR (Nujol) ν 3397, 1739, 1700 cm-1; H NMR (CDCl3) δ
1.30 (d, 3H, J ) 6.9 Hz), 1.41 (s, 9H), 2.01-2.18 (m, 1H), 2.30-
2.45 (m, 1H), 2.52-2.75 (m, 2H), 4.85 (dd, 1H, J ) 3.3, 9.6
Hz), 5.07 (d, 1H, J ) 6.9 Hz), 5.16 (AB, 2H, J ) 12.3 Hz), 5.35
(dq, 1H, J ) 6.9 Hz), 7.32-7.40 (m, 5H); 13C NMR (CDCl3) δ
17.6, 21.3, 28.2, 31.7, 49.7, 57.7, 67.4, 79.8, 128.2, 128.5, 134.8,
155.1, 170.4, 173.7, 174.5. Anal. Calcd for C20H26N2O6: C,
61.53; H, 6.71; N, 7.18. Found: C, 61.43; H, 6.79; N, 7.22.
Boc-L-Va l-L-p Glu -OBn (2b). For the synthetic procedure
from 1, see the preparation of Boc-L-Ala-L-pGlu-OBn (2a )
above: liquid, 83% yield; [R]D ) -18.5 (c ) 1.0, CH2Cl2); IR
C
25H34N4O8: C, 57.90; H, 6.61; N, 10.80. Found: C, 57.99; H,
6.57; N, 10.72.
Boc-L-Va l-L-p Glu -Gly-L-Ala -OBn (4b). For the synthetic
procedure from 3b, see the preparation of Boc-L-Ala-L-pGlu-
Gly-L-Ala-OBn (4a ) above: yield 50%; mp ) 53-58 °C; [R]D
)
+9.6 (c ) 0.5, CH2Cl2); IR (CH2Cl2) ν 3431, 3383, 1748, 1681
cm-1; H NMR (CDCl3) δ 0.77 (d, 3 H, J ) 6.8 Hz), 1.05 (d, 3
1
H, J ) 6.8 Hz), 1.43 (s, 9 H), 1.44 (d, 3 H, J ) 7.6 Hz), 2.10-
2.36 (m, 3 H), 2.56-2.66 (m, 1 H), 2.74-2.86 (m, 1 H), 3.77
(dd, 1 H, J ) 4.8, 17.0 Hz), 4.17 (dd, 1 H, J ) 6.8, 17.0 Hz),
4.56-4.66 (m, 2 H), 5.03 (d, 1 H, J ) 9.6 Hz), 5.18 (AB, 2 H,
J ) 12.4 Hz), 5.37 (dd, 1 H, J ) 3.2, 9.6 Hz), 6.64 (bs, 1 H),
6.77 (d, 1 H, J ) 6.4 Hz), 7.20-7.44 (m, 5 H); 13C NMR (CDCl3)
δ 16.0, 17.8, 19.8, 21.6, 28.3, 30.3, 32.2, 43.0, 48.3, 58.1, 59.4,
67.2, 79.8, 128.0, 128.6, 135.3, 159.1, 168.2, 170.8, 172.5, 174.5.
Anal. Calcd for C27H38N4O8: C, 59.33; H, 7.01; N, 10.25.
Found: C, 59.40; H, 7.07; N, 10.18.
1
(CH2Cl2) ν 3456, 3377, 1746, 1686 cm-1; H NMR (CDCl3) δ
0.76 (d, 3 H, J ) 6.9 Hz), 1.05 (d, 3 H, J ) 6.9 Hz), 1.43 (s, 9
H), 1.98-2.15 (m, 1 H), 2.15-2.30 (m, 1 H), 2.30-2.48 (m, 1
H), 2.51-2.75 (m, 2 H), 4.85 (dd, 1 H, J ) 3.6, 9.6 Hz), 5.09
(bs, 1 H), 5.18 (s, 2 H), 5.41 (dd, 1 H, J ) 3.3, 9.6 Hz), 7.30-
7.41 (m, 5 H); 13C NMR (CDCl3) δ 15.9, 20.1, 21.7, 28.6, 30.3,
32.2, 58.1, 58.5, 67.7, 79.9, 128.5, 128.8, 128.9, 155.8, 171.0,
171.7, 174.1. Anal. Calcd for C22H30N2O6: C, 63.14; H, 7.23;
N, 6.69. Found: C, 63.23; H, 7.27; N, 6.73.
Boc-L-Va l-L-p Glu -Aib-L-Ala -OBn (5). H-Aib-L-Ala-OBn
(0.26 mmol, 69 mg), NMM (0.52 mmol, 57 µL), HOBt (0.16
mmol, 22 mg), and EDCI (0.16 mmol, 31 mg) were added to a
stirred solution of Boc-L-Val-L-pGlu-OH (3b) (0.13 mmol, 43
mg) in dry DMF (2 mL) under nitrogen atmosphere at room
temperature. The mixture was stirred at room temperature
for 24 h, ethyl acetate (10 mL) was added, and the mixture
was washed with 1 M aqueous HCl (2 × 7 mL) and with an
aqueous saturated solution of NaHCO3 (1 × 10 mL). The
combined organic layers were dried over sodium sulfate, and
the solvent was removed under reduced pressure. The residue
was purified by silica gel chromatography (cyclohexane/ethyl
acetate 2:8 as eluant) and product 4a was obtained pure as a
white solid in 58% yield: mp ) 176-178 °C; [R]D ) -11.3
(c ) 0.9, CH2Cl2); IR (CH2Cl2) ν 3417, 3370, 1746, 1713, 1693
Boc-L-Ala -L-p Glu -OH (3a ). Pd/C (10%) on charcoal (50 mg)
was added to a solution of 2a (0.51 g, 1.3 mmol) in methanol
(10 mL), and the mixture was stirred in a Parr apparatus with
hydrogen (3 atm) for 3 h. Then, the catalyst was filtered on a
Celite pad, and the mixture was concentrated. The acid (3a )
was obtained as a waxy solid in 98% yield (0.43 g) without
any further purification: [R]D ) -49.5 (c ) 1.0, CH2Cl2); IR
1
(CH2Cl2) ν 3435, 1745, 1720, 1698 cm-1; H NMR (CDCl3) δ
1.32 (d, 3 H, J ) 6.6 Hz), 1.46 (s, 9 H), 2.05-2.20 (m, 2 H),
2.41-2.55 (m, 1 H), 2.60-2.82 (m, 2 H), 4.85 (dd, 1 H, J )
3.9, 9.0 Hz), 5.20 (d, 1 H, J ) 9.9 Hz), 5.36 (dd, 1 H, J ) 3.3,
9.9 Hz), 9.52 (bs, 1 H); 13C NMR (CDCl3) δ 16.9, 21.7, 28.7,
31.3, 48.3, 58.0, 80.5, 155.4, 174.3, 174.7, 175.1. Anal. Calcd
cm-1; H NMR (CDCl3) δ 0.75 (d, 3 H, J ) 6.8 Hz), 1.04 (d, 3
H, J ) 7.2 Hz), 1.40 (d, 3 H, J ) 6.8 Hz), 1.43 (s, 9 H), 1.52 (s,
3 H), 1.56 (s, 3 H), 1.95-2.31 (m, 3 H), 2.56 (ddd, 1H, J ) 5.2,
1
(24) Ryan, A. A.; Khan, J . A.; Moody, C. M.; Young, D. W. J . Chem.
Soc., Perkin Trans. 1 1996, 507.
J . Org. Chem, Vol. 68, No. 5, 2003 1991

Luppi et al.
9.6, 15.2 Hz), 2.74-2.85 (m, 1 H), 4.50-4.60 (m, 2 H), 5.03 (d,
1 H, J ) 9.6 Hz), 5.16 (AB, 2 H, J ) 12.4 Hz), 5.37 (dd, 1 H,
J ) 3.2, 9.6 Hz), 6.71 (s, 1 H), 6.83 (d, 1 H, J ) 7.2 Hz), 7.30-
7.39 (m, 5 H); 13C NMR (CDCl3) δ 15.9, 17.7, 19.7, 21.3, 24.3,
25.7, 28.2, 30.2, 32.1, 48.4, 57.2, 57.9, 59.4, 67.0, 79.6, 128.0,
128.3, 128.5, 135.3, 155.8, 167.0, 172.7, 173.4, 174.3. Anal.
Calcd for C29H42N4O8: C, 60.61; H, 7.37; N, 9.75. Found: C,
60.50; H, 7.45; N, 9.81.
(s, 9 H), 1.50 (d, 3 H, J ) 6.3 Hz), 2.22-2.44 (m, 1 H), 4.48 (d,
1 H, J ) 3.6 Hz), 4.72 (dq, 1 H, J ) 3.6, 6.3 Hz), 5.38-5.48
(m, 1 H), 9.60-10.02 (bs, 1 H); 13C NMR (CD3OD) δ 14.3, 18.8,
20.2, 27.6, 31.1, 62.0, 65.8, 74.7, 79.5, 152.2, 154.2, 171.2, 172.9.
Anal. Calcd for C15H24N2O7: C, 52.32; H, 7.02; N, 8.13.
Found: C, 52.34; H, 6.99; N, 8.07.
Boc-L-Va l-D-Oxd -Gly-L-Ala -OBn (13). For the synthetic
procedure from 12, see the preparation of Boc-L-Ala-L-pGlu-
Boc-L-Va l-L-Oxd -OBn (7). Boc-L-Val-OPfp (2 mmol, 0.76
g) was added to a stirred solution of H-L-Oxd-OBn (6) (1.2
Gly-L-Ala-OBn (4a ) above: yield 62%; mp ) 67-71 °C; [R]D )
9
+2.3 (c ) 0.8, CH2Cl2); IR (CH2Cl2) ν 3435, 3399, 3332, 1786,
1
mmol, 0.29 g), DIEA (5 mmol, 0.87 mL), and DMAP (0.81
mmol, 99 mg) in dry DMF (2 mL) under nitrogen atmosphere
at 0 °C. The mixture was stirred at room temperature for 2 h,
ethyl acetate (10 mL) was added, and the mixture was washed
with 1 M aqueous HCl (2 × 7 mL) and with an aqueous
saturated solution of NaHCO3 (1 × 10 mL). The combined
organic layers were dried over sodium sulfate, and the solvent
was removed under reduced pressure. The residue was purified
by silica gel chromatography (cyclohexane/ethyl acetate 7:3 as
eluant) and obtained in 45% yield as a liquid: [R]D ) -21.2
1746, 1706, 1686 cm-1; H NMR (CDCl3) δ 1.04 (d, 3 H, J )
6.6 Hz), 1.07 (d, 3 H, J ) 6.8 Hz), 1.41 (s, 9 H), 1.45 (d, 3 H,
J ) 7.6 Hz), 1.55 (d, 3 H, J ) 6.2 Hz), 1.98-2.18 (m, 1 H),
3.83 (dd, 1 H, J ) 5.8, 16.8 Hz), 4.03 (dd, 1 H, J ) 6.2, 16.8
Hz), 4.50 (d, 1 H, J ) 4.4 Hz), 4.59 (dq, 1 H, J ) 7.4 Hz), 4.82
(dq, 1 H, J ) 4.4, 6.2 Hz), 5.12 (d, 1 H, J ) 5.6 Hz), 5.20 (s, 2
H), 5.33 (dd, 1 H, J ) 5.6 Hz), 6.79 (d, 1 H, J ) 7.4 Hz), 7.20-
7.40 (m, 5 H), 7.80 (bs, 1 H); 13C NMR (CDCl3) δ 17.7, 18.1,
19.2, 21.2, 28.3, 30.2, 43.4, 48.2, 57.4, 62.9, 67.0, 74.7, 80.8,
127.9, 128.3, 128.5, 135.3, 151.5, 167.7, 168.0, 172.1, 174.2.
Anal. Calcd for C27H38N4O9: C, 57.64; H, 6.81; N, 9.96.
Found: C, 57.59; H, 6.84; N, 10.01.
(c ) 1.3, CH2Cl2); IR (CH2Cl2) ν 3393, 1791, 1752, 1699 cm-1
;
1H NMR (CDCl3) δ 0.79 (d, 3 H, J ) 7.0 Hz), 1.06 (d, 3 H, J )
6.6 Hz), 1.48 (s, 9 H), 1.55 (d, 3 H, J ) 6.2 Hz), 2.05-2.20 (m,
1 H), 4.50-4.62 (m, 2 H), 5.07 (d, 1 H, J ) 9.2 Hz), 5.21 (s, 2
H), 5.42 (dd, 1 H, J ) 3.6, 9.2 Hz), 7.25-7.42 (m, 5 H); 13C
NMR (CDCl3) δ 15.7, 19.7, 21.1, 28.3, 30.5, 57.1, 61.7, 68.1,
73.5, 80.0, 128.4, 128.7, 128.8, 134.4, 151.2, 155.8, 167.5, 172.9.
Anal. Calcd for C22H30N2O7: C, 60.82; H, 6.96; N, 6.45.
Found: C, 60.71; H, 7.06; N, 6.52.
Boc-L-Va l-D-Oxd -Aib-L-Ala -OBn (14). For the synthetic
procedure from 8, see the preparation of Boc-L-Val-L-pGlu-Aib-
L-Ala-OBn (5) above: yield 53%; mp ) 64-69 °C; [R]D ) +7.4
(c ) 0.5, CH2Cl2); IR (CH2Cl2) ν 3419, 3352, 1788, 1734, 1688
1
cm-1; H NMR (CDCl3) δ 0.95 (d, 3 H, J ) 7.2 Hz), 1.06 (d, 3
H, J ) 6.4 Hz), 1.41 (d, 3 H, J ) 6.8 Hz), 1.42 (s, 9 H), 1.51 (d,
3 H, J ) 6.2 Hz), 1.52 (s, 3 H), 1.55 (s, 3 H), 2.04-2.17 (m, 1
H), 4.35 (d, 1 H, J ) 3.6 Hz), 4.59 (dq, 1 H, J ) 6.8 Hz), 4.81
(dq, 1 H, J ) 3.6, 6.2 Hz), 5.06 (d, 1 H, J ) 5.6 Hz), 5.15 (AB,
2 H, J ) 12.8 Hz), 5.26 (bs, 1 H), 6.61 (d, 1 H, J ) 6.8 Hz),
7.00 (s, 1 H), 7.22-7.37 (m, 5 H); 13C NMR (CDCl3) δ 17.4,
18.3, 19.8, 21.4, 25.2, 28.6, 30.3, 31.2, 48.6, 52.6, 57.8, 63.6,
65.7, 67.3, 74.9, 80.6, 91.2, 128.4, 128.6, 128.8, 135.8, 152.0,
156.2, 167.2, 173.0, 173.6, 174.0. Anal. Calcd for C29H42N4O9:
C, 58.97; H, 7.17; N, 9.49. Found: C, 58.89; H, 7.11; N, 9.54.
Boc-L-Va l-L-Oxd -OH (8). For the synthetic procedure from
7, see the preparation of Boc-L-Ala-L-pGlu-OH (3a ) above: yield
98%; mp ) 118-122 °C; [R]D ) -7.4 (c ) 1.3, CH2Cl2); IR (CH2-
Cl2) ν 3350, 1786, 1706 cm-1; 1H NMR (CDCl3) δ 0.85 (d, 3 H,
J ) 6.9 Hz), 1.09 (d, 3 H, J ) 6.6 Hz), 1.46 (s, 9 H), 1.62 (d, 3
H, J ) 6.3 Hz), 2.18-2.42 (m, 1 H), 4.56 (d, 1 H, J ) 5.4 Hz),
4.65-4.85 (m, 1 H), 5.27 (d, 1 H, J ) 9.3 Hz), 5.32-5.48 (m, 1
H), 9.60-10.02 (bs, 1 H); 13C NMR (CDCl3) δ 16.1, 19.9, 21.4,
28.5, 30.8, 57.6, 61.9, 74.1, 80.7, 151.8, 156.5, 170.5, 171.1.
Anal. Calcd for C15H24N2O7: C, 52.32; H, 7.02; N, 8.13.
Found: C, 52.40; H, 7.06; N, 8.18.
Boc-L-Va l-L-P r o-Gly-L-Ala -OBn (15). This compound was
obtained following the NMM, HOBt, and EDCI protocol from
commercially available amino acids: mp ) 69-71 °C; [R]D
)
Boc-L-Va l-L-Oxd -Gly-L-Ala -OBn (9). For the synthetic
procedure from 8, see the preparation of Boc-L-Ala-L-pGlu-Gly-
-23.6 (c ) 0.3, CH2Cl2); IR (CH2Cl2) ν 3423, 3328, 1750, 1716,
1690, 1624 cm-1; 1H NMR (CDCl3) δ 0.88 (d, 3 H, J ) 6.8 Hz),
0.95 (d, 3 H, J ) 6.8 Hz), 1.42 (d, 3 H, J ) 8.0 Hz), 1.43 (s, 9
H), 1.90-2.22 (m, 5 H), 3.58-3.64 (m, 1 H), 3.66 (dd, 1 H, J )
4.4, 17.2 Hz), 3.76-3.88 (m, 1 H), 4.22-4.32 (m, 1 H), 4.62
(dq, 1 H, J ) 6.8 Hz), 5.09 (d, 1 H, J ) 9.6 Hz), 5.17 (AB, 2 H,
J ) 12.0 Hz), 6.64 (bs, 1 H), 7.20-7.44 (m, 6 H); 13C NMR
(CDCl3) δ 17.6, 19.3, 25.3, 28.4, 31.3, 43.1, 47.8, 48.2, 57.1,
60.8, 67.0, 79.1, 128.1, 128.3, 128.6, 135.5, 155.8, 168.8, 171.9,
172.2, 172.8. Anal. Calcd for C27H40N4O7: C, 60.88; H, 7.57;
N, 10.52. Found: C, 60.97; H, 7.62; N, 10.47.
L-Ala-OBn (4a ) above: yield 50%; mp ) 59-63 °C; [R]D
)
+18.7 (c ) 0.8, CH2Cl2); IR (CH2Cl2) ν 3423, 3390, 1792, 1680
1
cm-1; H NMR (CDCl3) δ 0.82 (d, 3 H, J ) 7.2 Hz), 1.07 (d, 3
H, J ) 6.9 Hz), 1.43 (s, 9 H), 1.44 (d, 3 H, J ) 6.6 Hz), 1.53 (d,
3 H, J ) 6.6 Hz), 2.14-2.22 (m, 1 H), 3.83 (dd, 1 H, J ) 4.8,
16.8 Hz), 4.15 (dd, 1 H, J ) 6.9, 16.8 Hz), 4.45 (d, 1 H, J ) 6.0
Hz), 4.61 (dq, 1 H, J ) 7.5 Hz), 4.69 (dq, 1 H, J ) 6.0 Hz), 5.10
(d, 1 H, J ) 7.0 Hz), 5.19 (AB, 2 H, J ) 12.0 Hz), 5.40-5.43
(m, 1 H), 7.04 (d, 1 H, J ) 6.9 Hz), 7.30-7.42 (m, 5 H), 7.55
(bs, 1 H); 13C NMR (CDCl3) δ 16.4, 17.9, 19.9, 20.4, 28.6, 30.8,
43.3, 48.6, 57.7, 63.8, 67.5, 74.4, 80.1, 128.3, 128.8, 128.9, 135.5,
151.9, 168.1, 168.4, 173.0, 174.4. Anal. Calcd for C27H38N4O9:
C, 57.64; H, 6.81; N, 9.96. Found: C, 57.71; H, 6.88; N, 9.89.
Boc-L-Va l-L-P r o-Aib-L-Ala -OBn (16). This compound was
obtained following the NMM, HOBt, and EDCI protocol from
commercially available amino acids: mp ) 152-155 °C;
[R]D ) -50.4 (c ) 1.0, CH2Cl2); IR (CH2Cl2) ν 3425, 3364, 1739,
1
Boc-L-Va l-D-Oxd -OBn (11). For the synthetic procedure
from 10,9 see the preparation of Boc-L-Val-L-Oxd-OBn (7)
above: liquid; 56% yield; [R]D ) +3.7 (c ) 1.8, CH2Cl2); IR
1699 cm-1; H NMR (CDCl3) δ 0.89 (d, 3 H, J ) 6.4 Hz), 0.96
(d, 3 H, J ) 7.2 Hz), 1.38 (d, 3 H, J ) 7.2 Hz), 1.39 (s, 3 H),
1.43 (s, 9 H), 1.44 (s, 3 H), 1.90-2.28 (m, 5 H), 3.56-3.64 (m,
1 H), 3.74-3.82 (m, 1 H), 4.25 (dd, 1 H, J ) 6.8, 9.2 Hz), 4.35
(dd, 1 H, J ) 4.8, 7.2 Hz), 4.54 (dq, 1 H, J ) 6.8 Hz), 5.09 (d,
1 H, J ) 9.6 Hz), 5.13 (AB, 2 H, J ) 12.0 Hz), 6.78 (s, 1 H),
7.20-7.40 (m, 6 H); 13C NMR (CDCl3) δ 18.0, 18.2, 19.7, 24.8,
25.6, 26.9, 28.1, 28.7, 31.7, 48.1, 48.7, 61.2, 67.2, 80.0, 128.3,
128.4, 128.7, 154.6, 170.9, 172.5, 172.9. Anal. Calcd for
1
(CH2Cl2) ν 3383, 1799, 1760, 1706 cm-1; H NMR (CDCl3) δ
0.89 (d, 3 H, J ) 6.6 Hz), 1.12 (d, 3 H, J ) 6.9 Hz), 1.48 (s, 9
H), 1.54 (d, 3 H, J ) 6.3 Hz), 2.26-2.44 (m, 1 H), 4.48 (d, 1 H,
J ) 3.9 Hz), 4.61 (dq, 1 H, J ) 3.9 Hz), 5.10 (d, 1 H, J ) 9.1
Hz), 5.26 (s, 2 H), 5.46 (dd, 1 H, J ) 3.9, 9.1 Hz), 7.29-7.45
(m, 5 H); 13C NMR (CDCl3) δ 16.5, 19.5, 21.2, 28.2, 31.1, 58.7,
62.0, 68.0, 73.6, 80.5, 128.3, 128.7, 134.6, 155.5, 167.5, 168.6,
172.9. Anal. Calcd for C22H30N2O7: C, 60.82; H, 6.96; N, 6.45.
Found: C, 60.88; H, 7.03; N, 6.55.
C
29H44N4O7: C, 62.12; H, 7.91; N, 9.99. Found: C, 62.18; H,
7.83; N, 10.05.
Com p u ta tion a l Meth od s. The starting guess structures
used for the DFT optimization of the analogous of 9 and 13
(i.e., MeOCO-L-Val-L-Oxd-Gly-L-Ala-OMe I and MeOCO-L-Val-
D-Oxd-Gly-L-Ala-OMe II, respectively) have been preliminary
located via AM1 optimizations. A set of reasonable structures
has been initially constructed so as to match the geometrical
Boc-L-Va l-D-Oxd -OH (12). For the synthetic procedure
from 11, see the preparation of Boc-L-Ala-L-pGlu-OH (3a )
above: yield 97%; mp ) 139-143 °C; [R]D ) +23.8 (c ) 0.8,
1
MeOH); IR (CH2Cl2) ν 3350, 1786, 1706 cm-1; H NMR (CD3-
OD) δ 0.83 (d, 3 H, J ) 6.9 Hz), 1.00 (d, 3 H, J ) 6.9 Hz), 1.44
1992 J . Org. Chem., Vol. 68, No. 5, 2003

Analysis of a Pseudoproline-Containing Turn Mimic
information revealed by NOESY spectra. AM1 optimizations
have then been performed on this molecular set, and the
minimum energy points located within this procedure provided
the starting structures used for the DFT optimizations and
refinements. All calculations were carried out using the tools
available in the Gaussian 9810 package on a SGI Origin 3800
multiprocessor system, using the DFT/B3LYP functional (i.e.,
Becke’s three parameter hybrid functional with the Lee-
Yang-Parr correlation functional).25 This functional has been
shown to properly describe both standard hydrogen-bonds,26
as well as nonclassical weakly bound hydrogen bonds (such
as C-H‚‚‚OdC interactions).27 In particular, according to the
literature,28
a mixed basis set was used for the correct
description of the hydrogen bonding network: a 6-31+G(d)
basis set for atoms i-NH, i-CHR, i-CO, CO-endocyclic, (i + 1)-
CO, (i + 2)-NH, (i + 3)-NH, (i + 3)-CO, and a 6-31G(d) basis
set for all other atoms. The use of diffuse functions is highly
recommended to properly simulate hydrogen bond interactions
and the related energetics.28
Ack n ow led gm en t . We want to thank MURST
(Cofin 2002, Cofin 2001, and 60%), the Alma Mater
Studiorum - University of Bologna (Funds for Selected
Topics), and CINECA (Grant No. K1IBOZZ1) for finan-
cial support. We also thank Claudio Toniolo for helpful
discussions.
(25) Becke, A. D. J . Chem. Phys. 1993, 98, 5648.
(26) (a) Gonzalez, L.; Mo, O.; Yanez M. J . Comput. Chem. 1997, 18,
1124. (b) Klein, R. A. J . Comput. Chem. 2002, 23, 585. (c) Novoa, J . J .;
Sosa C. J . Phys. Chem. 1995, 99, 15837.
(27) Brunel, L.; Carre´, F.; Dutremez, S. G.; Guerin, C.; Dahan, F.;
Eisenstein, O.; Sini G. Organometallics 2001, 20, 47.
J O026233L
(28) Novoa, J . J .; Tarron, B.; Whangbo, M.-H.; Williams, J . M. J .
Chem. Phys. 1991, 95, 5179.
J . Org. Chem, Vol. 68, No. 5, 2003 1993