Enhancing the proline effect: Pseudo-prolines for tailoring Cis/ trans isomerization

Article abstract of DOI:10.1021/ja973966s

Cis-trans isomerization of proline-like oxazolidines and thiazolidines, denoted pseudo-prolines, is investigated spectrophotometrically with a chymotrypsin coupled assay and by ¹H NMR. A series of peptides carrying substituents of varying stere

Full text of DOI:10.1021/ja973966s

2714
J. Am. Chem. Soc. 1998, 120, 2714-2720
Enhancing the Proline Effect: Pseudo-Prolines for Tailoring Cis/
Trans Isomerization
Michael Keller,† Ce´dric Sager,† Pascal Dumy,† Mike Schutkowski,‡ Gunter S. Fischer,‡ and
Manfred Mutter*,†
Contribution from the Institute of Organic Chemistry, UniVersity of Lausanne, BCH-Dorigny,
CH-1015 Lausanne, Switzerland, and Max-Planck-Research Unit “Enzymologie der Proteinfaltung”,
Kurt-Mothes Str. 3, D-06120 Halle/Saale, Germany
ReceiVed NoVember 19, 1997
Abstract: Cis-trans isomerization of proline-like oxazolidines and thiazolidines, denoted pseudo-prolines, is
investigated spectrophotometrically with a chymotrypsin coupled assay and by 1H NMR. A series of peptides
carrying substituents of varying stereochemistry at the 2-C position of ΨPro was prepared for evaluating
kinetic and thermodynamic data pertaining to the isomerization of the imidic bond. ΨPro are shown to exhibit
an enhanced proline effect, allowing the cis content along the imide bond to be tailored between 5 and about
100%. These Pro surrogates may serve as â-turn type VI mimetics and can be used to introduce specifically
cis-imide bonds into peptides and proteins. ΨPro derived from Ser and Thr show a marked difference in the
rate of isomerization about the imide bond compared to Cys derived thiazolidines. As with Pro itself, the cis
content for the (S) epimers of oxazolidines and thiazolidines can be raised from about 40-60% using LiCl/
TFE as a solvent, allowing more accurate measurements of the isomerization kinetics. In general, peptides
containing ΨPro exhibit enhanced isomerization rate constants for cis/trans isomerization compared to their
proline analogues depending on stereochemistry and degree of substitution at the 2-C ΨPro position. For
example, the thermodynamic barrier of the isomerization process for 2-C dimethylated ΨPro is decreased by
about 2-4 kcal/mol in comparison to Pro. In summary, ΨPro represent versatile Pro surrogates enhancing
the conformational and structural effects of Pro and offer a wide range of applications in peptide design and
engineering.
Introduction
lidine ring δ-position and the preceding residue.6 This raises
the free energy difference between cis and trans conformers
relative to peptide bonds devoid of Pro. The energy difference
between cis and trans isomers of a normal amide bond is thought
to be due to steric conflicts between adjacent R-carbon substit-
uents. This provides an explanation for the common occurrence
of cis Xaa-Pro bonds among protein structures.7 During the
refolding process of denaturated proteins, slow cis/trans isomer-
ization is often the rate-limiting step due to high rotational
energy barriers of cis/trans interconversions.8 For a better
understanding of this biologically most relevant conformational
change, model peptides containing Pro or Pro-surrogates
represent a valuable tool to investigate the origin of isomeriza-
tion. In this context, we recently introduced the pseudo-proline
(ΨPro) concept as a structure disrupting protection technique
in peptide chemistry and for modulating biological and phar-
maco-kinetical properties of peptides.9,10
A better understanding of the biological activity of peptide
lead structures requires a detailed knowledge of their structure
and conformation. Peptide chains exhibit a variety of confor-
mations which are energetically competitive. Long-range
cooperative effects stabilize specific conformations required for
protein-protein or ligand-receptor interactions.1 Oligopeptides
lack these long-range effects, and covalent restrictions have been
introduced to reduce their flexibility.2 Conformationally con-
strained peptide analogues are a valuable tool in deciphering
the receptor-bound conformation of biologically active peptides.3
For example, alkylated amino acids have been proposed as a
means of restricting the conformational flexibility of peptides
targeted to a specific recognition site.4 Being the only cyclic
proteinogenic amino acid, Pro plays a particular role in
determining the structural and conformational properties of
peptides and proteins.5 In particular, increased cis content along
the imidic bond of Xaa-Pro is often observed. trans-Xaa-Pro
peptide bonds introduce steric interactions between the pyrro-
The straightforward synthetic access allows for the preparation
of a variety of Pro surrogates tailored to the structural require-
ments of the desired target molecules. Most notably, the cis
content along the imidic bond together with their differential
† University of Lausanne.
‡ Max-Planck-Research Unit “Enzymologie der Proteinfaltung”.
(1) Flippen-Anderson, J. L.; Gilardi, R.; Karle, I. L.; Frey, M. H.; Opella,
S. J.; Gierasch, L.; Goodman, M.; Madison V.; Delaney, N. G. J. Am. Chem.
Soc. 1983, 105, 6609-6615.
(6) Mikhailov, D.; Daragan, V. A.; Mayo K. H. Biophysical J. 1995,
68(4), 1540-1550.
(7) Wu¨thrich, K.; Billeter, M.; Braun, W. J. Mol. Biol. 1984, 180, 715-
(2) Toniolo, C. Int. J. Peptide Protein Res. 1990, 35, 287-300.
(3) Marshall, G. R. Tetrahedron 1993, 49, 3547-3558.
(4) (a) Balaram, P. Curr. Opin. Struct. Biol. 1992, 2, 845-851. (b)
Giannis, A.; Kolter, T. Angew. Chem. 1993, 105, 1303-1326. (c) Burgess,
K.; Ho, K.-K.; Pal, B. J. Am. Chem. Soc. 1995, 117, 3803-3819.
(5) Schmid., F. X.; Mayr, L. M.; Mu¨cke, M.; Scho¨nbrunner, E. R. AdV.
Prot. Chem. 1993, Vol.44, 25-66.
740.
(8) Nall, B. T. In Protein Refolding; Gierasch, L. M., King, J., Eds.;
AAAS Press: Washington, DC, 1990; pp 198-207.
(9) Wo¨hr, T.; Wahl, F.; Nefzi, A.; Rohwedder, B.; Sato, T.; Sun, X.;
Mutter, M. J. Am. Chem. Soc. 1996, 118, 9218-9227.
(10) Dumy, P.; Keller, M.; Ryan, D. E.; Rohwedder, B.; Wo¨hr, T.; Mutter
M. J. Am. Chem. Soc. 1997, 119, 918-925.
S0002-7863(97)03966-8 CCC: $15.00 © 1998 American Chemical Society
Published on Web 03/17/1998

Enhancing the Proline Effect
J. Am. Chem. Soc., Vol. 120, No. 12, 1998 2715
and 16 °C. For each temperature, at least three measurements
per compound were recorded until less than 1% deviation was
reached. Measuring times varied between 400 and 600 s
depending on the completion of isomerization. The absorption
of the released p-nitroaniline (ꢀ ) 11 814 M-1 cm-1) was
monitored on a Hewlett-Packard 8452 A diode array spectro-
photometer with a thermostated cuvette holder to give a final
absorbance between 0.4 and 0.6 (390 nm). Constant temperature
was maintained within the cell (d ) 1 cm) by water circulated
from a Cryostat Haake D8 (Haake Fisons, Germany). The
Hewlett-Packard 89531 A MS-DOS-UV/vis operation software
or Sigma Plot Scientific Graphing System Vers. 5.01 (Jandel
Corp., U.S.A.) were used for data analysis.
Figure 1. ΨProlines Xaa(ΨproR1,R2) are obtained by condensation re-
action of aldehydes or ketones with Xaa ) Ser, Thr, and Cys. X ) O,
R ) CH3: Thr-derived oxazolidines; X ) O, R ) H: Ser-derived ox-
azolidines; X ) S, R ) H: Cys-derived thiazolidines, R: see Table 1.
acid lability can be modulated within a wide range. Further-
1
more, structural studies by H NMR are facilitated due to the
heteroatom at the C1-position of the ring. The characteristic
difference of the proton chemical shift of the 2-C protons of
the cis and the trans conformers allows for the measurement of
the cis content as well as the determination of the thermody-
namic parameters ∆G‡, ∆H‡, and ∆S‡ and the rate constant of
the isomerization of the imidic bond. To test the degree of
tolerance of enzymes toward ΨPro as surrogates for Pro, the
chymotrypsin coupled assay developed by Fischer et al.12 was
used to measure cis-trans isomerization of ΨPro containing
substrates. This enzyme coupled assay exploits the isomer
specificity of chymotrypsin mediated hydrolysis of an amide
bond between the P2′-P3′ position.13 Short model peptides of
type Suc-Val-Pro-Phe-pNA are hydrolyzed by chymotrypsin
between Phe and pNA only if the Val-Pro peptide bond is in
the trans conformation.14 After an initial Phe-pNA hydrolysis
termed burst phase of the present trans isomers, only the cis
isomers are left in solution. Subsequently, this temporary
disturbance of the conformational equilibrium relaxes allowing
for the kinetics and thermodynamics of the equilibrium to be
analyzed by following the UV absorption of the liberated pNA.
For our purpose, Pro was replaced by various ΨPro derivatives
in order to investigate the influence of bulkyness and stereo-
chemistry at 2-C upon the isomerization process. We focus
especially on Ser and Cys-derived ΨPro so as to avoid further
steric complications of oxazolidines derived from Thr.
NMR Spectroscopy. Spectra were recorded at 400 and 600
MHz using a Bruker ARX and AMX2 spectrometer, respec-
tively. Samples of about 2-5 mg were dissolved in 0.4-0.5
mL of DMSO-d6. Chemical shifts were calibrated with respect
to the residual DMSO signal (1H: 2.49 ppm; 13C: 39.5 ppm).
2D experiments typically were acquired using 2K × 512
matrixes over a 4000 Hz sweep width in both dimensions.
Quadrature detection in the indirect dimension was achieved
by using the TPPI procedure.17 Scalar connectivities were
recovered from 2D double quantum filtration (DQF) COSY
experiments.18 Dipolar connectivities were obtained either by
the conventional NOESY19 or ROESY20 sequence with mixing
times from 150 to 200 ms. A randomization of the mixing
length ((5%) was introduced in the NOESY experiments in
order to minimize coherence transfer. The spin lock mixing
interval of the ROESY sequence was applied by coherent CW
irradiation at γB2/2π ) 2 kHz. Experimental data processing
was performed using the Felix software package.21 The standard
sinebell squared routine was employed for apodization with a
shift range of 60-90° and zero filling in both dimensions before
2D transformations were applied to result in square matrixes
of 2K × 2K real point data. Proton resonance assignments of
the two isomeric forms and assignment of the stereochemistry
at the 2-C position were made using (DQF) COSY and ROESY
experiments according to previously published protocols.10 The
determination of kinetic parameters was performed by NOESY
spectra (600 MHz) recorded at variable temperature using values
of 5 s for the recovery delay to ensure restoration of thermal
equilibrium. This allowed for the measurement of the exchange
(Ict ) Itc) and diagonal peak (Icc and Itt) intensity. The kinetic
constants kc-t and kt-c were calculated using formula (I)
established under the assumptions of insignificant dipolar cross
relaxation with neighboring protons as well as similar external
relaxation rates between the exchanging spins.11 These as-
sumptions were satisfactorily verified within a short mixing time
τm value (60 ms) selected from a series of NOESY (30, 40, 60
80, 120 ms) experiments showing absence of NOE and good
signal-to-noise ratios for integration.
Methods
Peptide Synthesis. The ΨPro containing peptides were
synthesized according to methodologies described earlier.9
Mono- and disubstituted oxazolidine containing peptides (1 and
3) were obtained by postinsertion15 using dipeptide Fmoc-Val-
Ser-OBzl. Mono- and disubstituted thiazolidine containing
peptides (2 and 4) were synthesized by condensation of Cys
with the corresponding aldehyde or ketone followed by coupling
Fmoc-Val-F.16 Peptide 5 was obtained by condensing aqueous
formaldeyhde with Ser and subsequent trapping by Fmoc-Val-
F. Suc-Val-Pro-Phe-pNA (6) was purchased from BACHEM
(Bubendorf, Switzerland) and used without further purification.
The Chymotrypsin Coupled Assay. Typically, a stock
solution of the substrate in DMSO (10 mg/mL) was prepared,
of which 3 µL were pipetted to a solution of 50 µL of
chymotrypsin (25 mg/mL, 1 mM HCl) and 1150 µL of buffer
(HEPES 0.035 M, pH 7.8) at various temperatures between 3.6
2x
Ict
kt-c
)
c arctanh
(1)
(
)
τm
xt Icc + xc Itt
Xt, Xc are the molar fraction, Ict, Icc the intensities of the cross-
peaks in the NOESY spectrum, and kt-c the rate constant for
trans to cis isomerization.
(11) (a) Perrin, C. L.; Dwyer, T. Chem. ReV. 1990, 90, 935-967. (b)
Macura, S.; Farmer, B. T.; Brown, L. R. J. Magn. Reson. 1986, 70, 493-
499.
(12) Fischer, G.; Bang, H.; Mech, C. Biomed. Biochim. Acta 1984, 43,
1101.
(13) Berger, A.; Schechter, I. Philos. Trans R. Soc. 1970, B257, 249.
(14) Fischer, G. S. Angew. Chem., Int. Ed. Engl. 1994, 33, 1415-1436.
(15) Wo¨hr, T.; Mutter, M. Tetrahedron Lett. 1995, 36, 3847-3848.
(16) Activation of Fmoc-Val-OH with cyanuric fluoride (10 equiv) in
the presence of pyridine (1 equiv) in DCM for 2 h under reflux. Filtration
of the precipitate and removing of solvent gave Fmoc-Val-F quantitatively.
(17) Drobny, G.; Pines, A.; Sinton, S.; Wertekamp, D.; Wemmer, D.
Faraday DiV. Chem. Soc. Symp. 1979, 13, 48-58.
(18) Piantini, V.; Sørensen, O. W.; Ernst, R. R. J. Am. Chem. Soc. 1982,
104, 6800-6801.
(19) Jeener, J.; Meier, B. H.; Bachmann, P.; Ernst, R. R. J. Chem. Phys.
1979, 71, 4546-4553.
(20) Bothner-By, A. A.; Stephen, R. R.-L.; Lee, J., Warren, C. D.;
Jeanloz, R. W. J. Am. Chem. Soc. 1984, 106, 811-813.
(21) Biosym Technologies Inc., San Diego.

2716 J. Am. Chem. Soc., Vol. 120, No. 12, 1998
Keller et al.
Cis/trans ratios were determined by integration of the
characteristic 2-C protons in the case of 1S, 1R, 2S, 2R and for
compounds 3 and 4 by integration of the NH-pNA protons.22
Results
Chymotrypsin Coupled Assay. Substrates 1-6 exhibiting
different substituents (R1, R2) and stereochemistry at the 2-C
position of the ΨPro ring were used to determine the kinetic
and thermodynamic parameters of cis/trans isomerization about
the Val-ΨPro imidic bond.
In a preliminary assay, the degree of tolerance of chymot-
rypsin for the ΨPro unit was tested. It was found that for the
investigated ΨPro containing substrates 1-5 as well as for the
Pro containing substrate 6, no affection of neither the cis content
nor the kinetic data could be observed by a 3-fold increase of
the standard enzyme concentration ([Est] ) 20 µM; substrate
standard concentration [Sst] ) 40 µM). Consequently, compa-
rable affinities of the enzyme for the artificial substrates as for
the Pro substrate can be assumed.
(2-C)-Monosubstituted ΨPro Containing Substrates 1 and
2. Interestingly, even ΨPro with aromatic substituents at 2-C
position were recognized by chymotrypsin as prolines enabling
the rate constant and associated thermodynamic data for the
cis-trans isomerization of each compound to be determined.
They were found to be in good agreement with previously
published data on similar systems.14,34 2-C (S)-configurated
ΨPro generally show considerably higher cis contents (37%)
than their 2-C (R)-configurated stereoisomers (5%). A pro-
nounced influence of the heteroatom at position 1 of the ΨPro
ring on the kinetics was observed. Oxazolidine derivative 1S
isomerizes two times faster than its thiazolidine analogue 2S
(Figure 3).
Figure 2. Isomer specific cleavage of the trans-Phe-pNA amide bond
of ΨPro containing substrates by chymotrypsin. The trans-Val-ΨPro
unit represents a nonnative cleavage site recognized by the serine
protease chymotrypsin. Legend: CT ) chymotrypsin, pNA ) p-
nitroaniline, Suc ) succinate.
Table 1. Substrates Investigated in This Study (see Figure 2)a
substrate
Xaa
X
R1
R2
1S
1R
2S
2R
3
4
5
Ser
Ser
Cys
Cys
Ser
Cys
Ser
Pro
O
O
S
S
O
S
pmp
H
pmp
H
CH3
CH3
H
H
pmp
H
pmp
CH3
CH3
H
O
CH2
6
H
H
a Legend: pmp ) p-methoxyphenyl, X: heteroatom at position 1
of ΨPro unit. R1, R2: substituents at the 2-C position of ΨPro.
A severe limitation of the chymotrypsin coupled assay became
apparent while measuring the kinetics of the 2-C (R)-epimers
(22) Integration of CR protons affords the same cis content. Due to signal
overlap in some cases, the NH-pNA was chosen for all compounds to
determine their cis/trans ratios.
(23) Kofron, J. L.; Kuzmic, P.; Kishore, V.; Colon-Bonilla, E.; Rich, D.
H. Biochemistry 1991, 30, 6127-6134.
(24) At 9 °C, the cis/trans isomerization rate for Suc-Val-Ser(Ψpmppro)-
Phe-pNA (1R) was measured to be kct ≈ 18 × 10-3 s-1, while its (S) isomer
isomerizes with a value of kct ≈ 15 × 10-3 s-1
.
(25) Stein, R. L. AdV. Protein Chem. 1993, 44, 1.
(26) Dissolvation of substrate 5 in DMSO before slowly adding NaOH
1
(1 N) until a maximum of cis in H NMR was observed.
(27) (a) Nachtergaele, W. Y. Anteunis, J. O. Bull. Soc. Chim. Belg. 1980,
89, 9, 749-758. (b) Toppet, S.; Claes, P.; Hoogmartens, J. Org. Magn.
Reson. 1974, 6, 48-52. (c) Benedetti, F.; Ferrario, F.; Sala, A.; Soresinetti,
P. R. J. Heterocycl. Chem. 1994, 31, 1343-1347.
(28) (a) Radzicka, A.; Acheson, S. A.; Wolfenden, R. Bioorg. Chem.
1992, 20, 382. (b) Eberhardt, E. S.; Loh, S. N.; Hinck, A. P.; Taines, R. T.
J. Am. Chem. Soc. 1992, 114, 5437.
Figure 3. Cis/trans isomerization of (2-C)-dimethylated ΨPro contain-
ing substrates 3 (a) and 4 (b) as determined by the chymotrypsin coupled
assay at λ ) 390 nm.
(29) (a) La Planche, L. A.; Rogers, M. T. J. Am. Chem. Soc. 1964, 86,
337. (b) Madison, V.; Kopple, K. D. J. Am. Chem. Soc. 1980, 102, 4855.
(c) Ke, H.; Mayrose, D.; Cao, W. Proc. Natl. Acad. Sci. U.S.A. 1993, 90,
3324-3328.
(30) The cis/trans ratios of all compounds were determined by the
integration of the 2-C protons which for each of the cis and the trans form
give separate singlet-signals at around 6.1 and 6.2 ppm, respectively.
1
(31) The H NMR spectrum in DMSO shows one predominant confor-
mation. For certain protons such as the NH-pNA proton, traces of another
conformation can be detected. However, due to a relative intensity of less
than 3% with respect to the major conformation, an unambiguous assignment
is impossible. It was yet detected by the CT coupled assay that very few
percent of the trans form exist depending on temperature and solvent.
(32) Beausoleil, E.; Lubell, D. W. J. Am. Chem. Soc. 1996, 118, 12902-
12908.
Figure 4. Cis-trans isomerization of monosubstituted ΨPro containing
substrates 1S (a) and 2S (b) as determined by the chymotrypsin coupled
assay at λ ) 390 nm.
1R and 2R as well as the 2-C dihydrooxazolidine derivative 5.
Containing a vast excess of the trans form, the initial burst phase
is too dominant to accurately measure the following isomer-
ization kinetics. Since the content of cis isomers remains very
low after the initial hydrolysis of pNA, a detection of the
(33) Magaard, V. W.; Sanchez R. M.; Bean, J. W.; Moore M. L.
Tetrahedron Lett. 1993, Vol. 34, No. 3, 381-384.
(34) Kern, D.; Schutkowski, M.; Drakenberg, T. J. Am. Chem. Soc. 1997,
119, 8403-8408.
(35) Nefzi, A.; Schenk, K.; Mutter, M. Protein Pept. Lett. 1994, 1, 66-
69.

Enhancing the Proline Effect
J. Am. Chem. Soc., Vol. 120, No. 12, 1998 2717
Table 2. First-Order Rate Constants kct and Thermodynamic Values for the Cis to Trans Isomerization of Compounds 1-6 in 0.035 M
HEPES (pH 7.8) as Determined by the Chymotrypsin Coupled Assay12
a
substrate
cis [%]
kct × 10-3 s-1 ∆G° b [kcal/mol] ∆Hq c [kcal/mol/K] ∆Sq c [cal/mol/K] ∆Gqct[kcal/mol] ∆Gq d [kcal/mol]
ct
ct
tc
1S
2S
3
4
6
37 ( 1
37 ( 1
85 ( 1
99 ( 1
5 ( 1e
17.3
8.5
12.3
37
-0.32
-0.32
1
2.7
-1.76
21.4 ( 0.9
21.8 ( 1
9.3 ( 3.3
18.6
19
19
9.3 ( 3.7
7.9 ( 1.4
5.4 ( 3
19.4
17.9
15.5
21.3
20.2 ( 0.3
19.9 ( 0.9
21.5 ( 0.8
18.9
18.3
19.5
4
6.8 ( 2.6
a Measured at 282.5 K at substrate concentrations of 40 µM. b From ∆G°ct ) -lnKRT [300 K]. c Determined from the Eyring equation ln(k/T)
) [-∆Hq/R][1/T]+ ∆Sq/R + ln(kB/h). d Calculated from the Gibbs equation ∆Gq ) ∆Hq - T∆Sq where R is the universal gas constant, kB is the
Boltzmann constant, and h is Planck’s constant. e Cis content in HEPES (35 mM, pH 7.8) starting from the equilibrium in DMSO. The rate constant
for this substrate was determined using LiCl/TFE as solvent.
Table 3. Effect of LiCl (0.5 M in TFE) on the Cis Content of
ΨPro Containing Substrates (see Table 1) as Determined by the
Chymotrypsin Coupled Assay23
In contrast to the findings for the monosubstituted ΨPro
substrates 1S and 2S, substrate 4 has an isomerization rate three
times higher than its oxazolidine analogue 3 (Table 2). The
transition state energies ∆G‡ct for 1-5 are typically around 1-2
kcal/mol lower than that of the Pro substrate 6.
cis [%]
substrate
DMSO
TFE (0.5 M LiCl)
1H NMR. The present kinetic and thermodynamic data
obtained by the biological assay are in full agreement with the
1H NMR data. However, due to the low solubility (e1 mg/
mL), the 1H NMR measurements were limited to organic
solvents which are known to accelerate cis/trans isomeriza-
tion.25
1S
1R
2S
2R
37
5
37
5
60
5
60
5
subsequent isomerization becomes inaccurate due to low cis/
trans ratios. Attempts to increase the cis content failed, thus
restricting the investigation on the 2-C (R)-isomers 1R and 2R
(2-C)-Unsubstituted ΨPro Containing Substrate 5. 1H
NMR spectrum of compound 5 in DMSO shows broad signals
for the trans form preventing adequate integration measure-
ments. By applying lower temperatures (282 K) and a mixture
of water/NaOH/DMSO (pH 8),26 the cis content could be
increased up to 14% while simultaneously reducing the rate of
isomerization. The rate constant of 2.15 s-1 (Table 4) indicates
a decrease of the free energy of activation ∆∆G‡ct (∼2.8 kcal/
mol) compared to Pro containing peptide 6 (Table 2). As the
steric interactions between the valine residue and the unsub-
stitued oxazolidine derivative 5 are similar to those found in
the Pro containing peptide 6, it is reasonable to assume that the
cis ground states of 5 and 6 are comparable in energy.
Consequently, the decrease of the free energy barriers ∆∆G‡
can be attributed to a stabilization of the transition state.
(2-C)-Monosubstituted ΨPro Containing Substrates 1 and
1
to dynamic H NMR studies. For the compounds 1S and 2S,
rate constants of 17.3 × 10-3 and 8.5 × 10-3 s-1 were found,
respectively. Compared to the Pro containing substrate 6 (kct
) 4 × 10-3 s-1), all ΨPro containing substrates show
remarkably increased isomerization rates.
Since this assay is severely limited in the case of substrates
exhibiting low contents of cis conformation, considerable efforts
have been devoted to increase the cis population of Xaa-Pro
bonds. As reported by Rich et al.,23 the Xaa-Pro amide bond
in a solution of LiCl/TFE (or THF) exists predominantly in the
cis conformation. When a LiCl/TFE solution of substrate is
added to a biological buffer, the Xaa-Pro cis/trans conforma-
tional equilibrium is restored, and the enzymatic catalysis of
this process can be monitored. The same holds for the 2-C
(S)-stereoisomers for both thiazolidine and oxazolidine sub-
strates. Solvation in 0.5 M LiCl/TFE increases the cis content
to approximately 60% (Table 3). Surprisingly, no effect on the
2-C (R)-isomers is observed. Although no precise measure-
ments on the 2-C (R)-isomers could be carried out over a broad
range of temperature, preliminary assays indicate faster isomer-
ization rates compared to their 2-C (S)-counterparts, as revealed
1
2. The H NMR signals for the peptides 1S and 2S can be
attributed to the two conformations cis and trans in slow
exchange. The stereochemistry at position 2 for the two isomers
3
was determined by using JHR-Hâ as previously described.10
The substituent of the (R)-isomers syn to the ΨPro-CO
induces mainly trans, whereas the (S)-epimer (anti to ΨPro-
CO) adopts the cis form up to 50%. The assignment of the
two epimers allowed for the determination of both rate constants
and thermodynamic values in DMSO and CDCl3 (Table 4). As
reported earlier, hydrophobic solvents accelerate cis/trans
isomerization due to the hydrophobic nature of the transition
state.28 Similar to Pro containing peptide 5, the rate constants
for the ΨPro containing compounds 1 and 2 were considerably
faster in CDCl3 and DMSO compared to water. As expected,
by H NMR measurements.24
1
(2-C)-Dimethylated ΨPro Containing Substrates 3 and 4.
As depicted in Table 2, the cis content of the 2-C dimethylated
thiazolidine ΨPro containing compound 4 is higher by about
15% compared to the corresponding oxazolidine derivative 3.
Surprisingly, the content of cis of compound 4 approaches up
to >97%.
Table 4. First-Order Rate Constants and Thermodynamic Values for the Cis to Trans Isomerization of Compounds 1 and 2 at 300 K in
1
DMSO and CDCl3 as Determined by Dynamic H NMR Measurements
substrate
kcis-transa [s-1
]
∆Hq b [kcal/mol]
∆Sq b [cal/mol/K]
∆Gq c [kcal/mol]
ct
ct
ct
1R
1S
2R
2S
5
6.2 (8.2)
0.4 (6)
2.2 (6)
0.7 (4.5)
2.15
18 ( 1 (18 ( 1)
20 ( 1 (20 ( 1)
20 ( 1 (22 ( 1)
16 ( 1 (17 ( 1)
4 ( 4 (7 ( 3)
6 ( 2 (10 ( 3)
10 ( 3 (19 ( 3)
-5 ( 3 (1 ( 3)
16.5 ( 0.1 (16.3)
18.3 ( 0.1 (16.6)
17.2 ( 0.1 (16.6)
18 ( 0.1 (16.7)
16.7 ( 0.1d
a-c See Table 2. Values measured in CDCl3 in brackets. Standard errors according to Sandstro¨m et al.: 38 ∆∆Gq/∆Gq ) {[∆T/T(ln kBT/h/k +
1)]2 + (∆k/k)2}1/2/(ln kBT/h/k); assumed errors ∆T ) 0.5 K, k ( 20%. d Based on one single measurement at 282 K.

2718 J. Am. Chem. Soc., Vol. 120, No. 12, 1998
Keller et al.
Table 5. Influence of the Polarity of Solvents on the Cis Content
of Xaa-ΨPro
butylproline systems and those of Magaard for δ-dimethylated
Pro.33 A quantitative assessment of the effect of the heteroatom
of the ΨPro ring is difficult. Previous studies using 2-C-
unsubstituted oxazolidines revealed considerably increased
isomerization rates and a pronounced preference for trans
preventing the application of the chymotrypsin coupled assay.34
Decreased transition state energies and accelerated isomerization
rates compared to Pro might be due to different ring puckerings
induced by the heteroatom.10,27 However, the effects of the ring
conformation are difficult to evaluate, and with the present data
an unambiguous attribution is impossible. Being 37% cis, the
2-C (S)-isomers prove to be ideal candidates for the chymo-
trypsin coupled assay for obtaining accurate kinetic and
thermodynamic data. However, an unambiguous interpretation
of the kinetic data cannot be given at this point. A number of
parameters such as ring puckering and stereochemistry of the
substituents appear to be important. As exemplified by
compounds 1S and 2S showing transition state energies ∆G‡ of
18.6 kcal/mol (1S) and 19 kcal/mol (2S), the energy barriers
for isomerization of ΨPro are lowered by about 1-2 kcal/mol
compared to the natural Pro derivative 5. The relatively large
entropic contribution ∆S‡ of about 9 cal/mol/K suggests that
the rate determining step is physical rather than chemical, i.e.,
the implicated imide bond in the transition state gains additional
degrees of freedom compensating for the loss of amide
resonance. Despite experimental difficulties due to the domina-
tion of the trans form, 2-C (R)-configurated ΨPro were found
to exhibit increased isomerization rates compared to their 2-C
(S)-counterparts. For 2-C dimethylated peptides 3 and 4, a
dramatic influence of the heteroatom on the isomerization
kinetics is found. While for the oxygen containing oxazolidine
a rate constant in the order of the monosubstituted substrates is
observed, the sulfur containing thiazolidine derivative 4 shows
a three times higher isomerization rate compared to compound
3. The increased distance between 2-C and the sulfur atom
compared to oxygen results in a shortened 2-C-N bond.35 As
a result, the 2-C dimethyl groups come closer to the isomerizing
bond, which, as a consequence, may twist the amide bond and
stabilize the transition state by the presence of the hydrophobic
methyl groups.36 The extraordinarily high cis content of 2-C
dimethylated ΨPro 3 and 4 can be attributed to steric conflicts
induced by the methyl groups in the trans form provoking a
turn at this position of the peptide. Due to its nearly quantitative
cis content, such a turn has striking similarities with â-turn type
VI. Dimethylated ΨPro might thus find application as mimetics
of â-turns type VI. The clearly lower entropic contribution to
the free energy ∆G‡ compared to the monosubstituted substrates
indicates a lower flexibility around the isomerizable amide bond
imposing higher conformational constraints in the transition
state. Theoretical studies by Karplus showed that autocatalysis
might be an important factor in the catalytic pathway of the
Pro dipeptide.37 Twisting the ψ angle from 0° (characteristic
for a type VIa â turn) to 38° leads to a decrease of the effective
activation barrier by about 3.6 kcal/mol. Twisting the ψ angle
out of plane has been shown to be an important factor in
destabilizing the ground state. While water is thought to
stabilize the imide bond against twisting, hydrophobic com-
pounds such as organic solvents have been shown to destabilize
and thus accelerate cis/trans isomerization.25
cis [%]
b
substrate
X
H2Oa
DMSOb
CDCl3
1S
2S
1R
2R
3
4
5
O
S
O
S
O
S
O
37
37
5
5
85
100
2-3
41
34
24
13
95
100
5
55
42
11
6
90
100
5
a Measured by the chymotrypsin coupled assay in HEPES (0.035
M, pH 7.8) starting from the equilibrium in DMSO. b Determined by
1H NMR by integration of the cis and trans signals of the 2-C proton.
CDCl3 being a more hydrophobic solvent further enhanced the
rate constants compared to DMSO.
In particular, this method proved to be successful for
measuring the rate constant values for 2-C (R)-derivatives which
were not accurately measurable by the enzymatic assay. A clear
decrease of the activation barrier for all compounds investigated
outlines the influence of the solvent. DMSO as a more polar
solvent is thought to stabilize the ground state, while CDCl3
stabilizes the hydrophobic transition state. Such solvent assisted
mechanisms for cis/trans isomerization has already been
described for Pro and secondary amides.29 While in the organic
solvents CDCl3 and DMSO, the cis content of all oxazolidine
derivatives are generally higher compared to their thiazolidine
analogues; no such differences are observed in DMSO/water
systems (Table 5).30
(2-C)-Dimethylated ΨPro Containing Substrates 3 and 4.
As previously demonstrated, 2-C dialkylated ΨPro induce
predominantly the cis form of the Val-ΨPro imide bond (Table
2).10 For the oxazolidine derivative 3, a set of signals is
observed pointing to the presence of about 10% trans as detected
too by the enzymatic assay. Due to small chemical shift
differences and overlaps of R protons, neither the assignment
of that minor form nor the calculation of the kinetic constants
could be performed in this case. Surprisingly, no clear evidence
for the trans form of the thiazolidine derivative 4 was found
by NMR. Even if small amounts of the trans isomer exist (1-
2%) according to the enzymatic measurements, its small
population disabled a detailed attribution; we can therefore not
exclude that it might as well be traces of another conformation
not due to cis-trans isomerism.31
The increased rate constant values as measured by the
chymotrypsin coupled assay indicate that the equilibrium of the
cis/trans isomerization is fast resulting in an averaging to one
single set of signals. Temperature dependent measurements of
the isomerization rates provided the corresponding values of
∆H‡ and ∆S‡ after fitting the Van’t Hoff plot. The ∆H‡ values
demonstrate that the rotational barrier is essentially enthalpic
similar to Pro. The entropy contributions are estimated to be
small; however, the limited temperature range of the measure-
ments does not allow for quantitative evaluations of ∆S‡.
Discussion
Enhanced cis/trans rate constants and decreased transition
state energies compared to Pro are characteristic for the Xaa-
ΨPro moiety. The degree of tolerance of chymotrypsin toward
the ΨPro unit suggests no direct interference of the sterically
demanding substitutents at 2-C with the substrate binding site
of the enzyme. The results show a clear correlation with those
obtained in pure water by Beausoleil and co-workers32 on 5-tert-
(36) Albers, M. W.; Walsh, C. T.; Schreiber, S. L. J. Org. Chem. 1990,
55, No. 17, 4984-4986.
(37) Fischer, S.; Dunbrack, R. L. Jr.; Karplus, M. J. Am. Chem. Soc.
1994, 116, 11931-11937.
(38) Sandstro¨m, J. Dynamic NMR Spectroscopy; Academic Press: 1982;
pp 108-109.

Enhancing the Proline Effect
J. Am. Chem. Soc., Vol. 120, No. 12, 1998 2719
pNA), 8.46 (d, 1H, J ) 8.44 Hz, NH-Phe), 8.21 (d, 2H, J ) 9.08 Hz,
o-pNA), 8.19 (d, 1H, J ) 9.1 Hz, NH-Val), 7.8 (d, 2H, J ) 9.16 Hz,
m-pNA), 7.19-7.26 (m, 5H, aromatics-Phe), 5.29 (d, 1H, J ) 3.36
Hz, 2-C ΨPro), 4.95 (d, 1H, J ) 3.04 Hz, 2-C ΨPro), 4.61 (d × d,
1H, 6.76 Hz, R-Phe), 4.46 (t, 1H, J ) 5.92 Hz, R-Ser), 4.17 (t, 1H, J
) 7.36 Hz, â-Ser), 3.95 (t, 1H, J ) 7.88 Hz, R-Val), 3.79 (d × d, 1H,
5.6 Hz, 5.12 Hz, â-Ser), 3.31 (s, H20), 3.07 (d × d, 1H, J ) 6.32 Hz,
5.76 Hz, â-Phe), 2.96 (d × d, 1H, J ) 8.6 Hz, 7 Hz, â-Phe), 2.49 (s,
DMSO) 2.3-2.49 (m, 4H, suc), 1.89 (m, 1H, â-Val), 0.86 (d, 3H, 3.16
Hz, γ-Val), 0.85 (d, 3H, 3.24 Hz, γ-Val). ESI-MS 584.2 [M]+. HPLC
single peak tR ) 9.51 min (60-100% B, 40 min, C18).
Conclusion
Isomerization of the imidic bond of peptides containing ΨPro
was studied applying a biological assay and dynamic 1H NMR.
It is shown that ΨPro are accepted by chymotrypsin as Pro
and are isomer specifically hydrolyzed similar to their Pro
containing analogues. The significant differences in transition
state energies and kinetics of cis-trans isomerizations between
Xaa-ΨPro and the Xaa-Pro systems are indicative for the
conformational effects of the substituted five-membered rings.
Moreover, ΨPro prove to be valuable tools for delineating the
impact of hydrophobic substituents near the isomerizing imide
bond on the isomerization barrier. The additional stereocenter
at the 2-C ΨPro position allows for the investigation of both
the (R)- and (S)-isomers which show pronounced differences
in isomerization rates and cis contents. For the first time, it is
shown that LiCl/TFE increases stereoselectively the cis contents
of oxazolidine and thiazolidine imide bonds up to 60% cis.
The preferable adoption of the cis conformation of 2-C
dimethylated ΨPro allows for their use as selective cis-Xaa-
ΨPro bond inducer to chemically introduce constraint into
peptides and proteins and to test the cis-imide bond as a
structural requirement for the bioactive conformation.
Suc-Val-Cys(ΨMe,Mepro)-Phe-pNA (4). To Cys(ΨMe,Mepro)-OH‚HCl
(0.2 g, 1 mmol) in THF (10 mL) was added DIEA (210 mg, 2 mmol)
and Fmoc-Val-F16 (281 mg, 0.83 mmol) to stir for 12 h at room
temperature. After removal of the solvent, the white solid was dissolved
in methanol and passed through a silica column (CHCl3/MeOH/HOAc
) 100/10/1) to give 160 mg of pure Fmoc-Val-Cys(ΨMe,Mepro)-OH,
ESI-MS 483.5 [M + H]+. Activation of the dipeptide at -10 °C with
isobutylchloroformate (54 mg, 1.2 equiv) in DCM (10 mL) using NMM
(43 µL, 1.2 equiv) as a base followed by the addition of Phe-pNA (93
mg, 1.05 equiv) gave Fmoc-Val-Cys(ΨMe,Mepro)-Phe-pNA, which was
deprotected without further workup using DBU (5% in DCM, 10 mL,
10 min). The yellow reaction mixture was evaporated to dryness and
taken up in acetonitrile/water (5 mL, 1:1/v/v) before purifying on a
HPLC preparative (C18) using a gradient of 5-100% within 30 min.
Forty milligrams (9.2%) of NH2-Val-Cys(ΨMe,Mepro)-Phe-pNA was
isolated after lyophilization, ESI-MS 528.5 [M + H]+. Succinic an-
hydride (15 mg, 2 equiv) and NMM (17 µL, 2 equiv) in THF (5 mL)
were added to obtain 24 mg (4.6%) of pure Suc-Val-Cys(ΨMe,Mepro)-
Phe-pNA after purification on HPLC (50%-70%, 30 min, C18). HPLC
analytic: tR ) 11.29 min (60-100% B, 40 min, C18), ESI-MS 628.5
Experimental Section
Materials. All protected amino acides were purchased from Cal-
biochem-Novabiochem AG (La¨ufelfingen, Switzerland). Phe-pNA was
purchased at BACHEM (Bubendorf, Switzerland); reagents and solvents
were purchased from Fluka (Buchs, Switzerland) and used without
further purification. HPLC was performed on Waters equipment using
columns packed with Vydac Nucleosil 300 Å 5 µm C18 particles unless
otherwise stated. Analytical columns (250 × 4.6 mm) were operated
at 1 mL/min and preparative columns (250 × 21 mm) at 18 mL/min,
with UV monitoring at 214 nm. Solvent A is water (purified on a
Milli Q Ion exchange cartridge) containing 0.09% TFA, and solvent B
is acetonitrile HPLC-R (containing 10% water and 0.09% TFA),
purchased from Biosolve, Valkenswaard, Netherlands. Mass spectra
were obtained by electron spray ionization (ESI-MS) on a Finnigan
LC 710. 1H NMR spectra were obtained on a Bruker DPX-400 with
trimethylsilane internal standard for intermediate compounds. Ab-
breviations used were as follows: NMM ) N-methylmorpholine, THF
) tetrahydrofurane, DCM ) dichloromethane, DMSO-d6 ) dimethyl
sulfoxide deuterated, DMF ) dimethylformamide, DBU ) 1,8-
diazabicyclo[5.4.0]undec-7-ene(1,5-5), TFA ) trifluoroacetic acid. For
the syntheses of 1R, 1S, and 3, see ref 10.
Suc-Val-Ser(ΨH,Hpro)-Phe-pNA (5). Fmoc-Val-Ser(ΨH,Hpro)-OH
(0.35 g, 0.8 mmol) was dissolved in DCM (10 mL) followed by the
addition of isobutyl chloroformate (120 µL, 1.05 equiv) and NMM
(78 µL, 1.05 equiv) to give a white suspension which was stirred for
30 min under nitrogen at 10 °C. Phe-pNA (239 mg, 1.05 equiv)
dissolved in DCM (2 mL) was added to give a clear solution which
was stirred for 4 h to give Fmoc-Val-Ser(ΨH,Hpro)-Phe-pNA in 60%
(HPLC). All DCM was evaporated, and the remaining yellow solid
taken up in DMF/morpholine (13 mL of a 5% solution). The solution
was turning to yellow after stirring under nitrogen for 4 h. The DMF
was removed and replaced by with ethyl acetate to precipitate. Cooling
to 4 °C for 12 h completed the recrystallization. Two hundred ten
milligrams (54%) of a white substance was separated by filtration, ESI-
MS 484.5 [M + H]+. Test of ninhydrin proved positive.
1
[M + H]+, H NMR (400 MHz, DMSO-d6, 14 mg/mL, 300 K) 10.68
(s, 1H, NH-pNA), 8.29 (d, 1H, d, NH, J ) 6.64 Hz, Phe), 8.2 (d, 2H,
J ) 8.84 Hz, o-pNA), 7.87 (d, 1H, J ) 6.72 Hz, NH-Val), 7.75 (d,
2H, J ) 8.88 Hz, m-pNA), 7.27-7.16 (m, 5h, aromatics-Phe), 4.98 (d,
1H, J ) 6.2 Hz, R-Cys), 4.68 (d × d, 1H, J ) 4.2 Hz, R-Phe), 4.31 (t,
1H, J ) 6.56 Hz, 6.72 Hz, R-Val), 3.35 (d, 1H, J ) 10 Hz, â-Cys), 3.2
(d, 1H, J ) 12.21 Hz, â-Cys), 3.12 (t, 2H, J ) 6.43 Hz, 5.59 Hz,
â-Phe), 2.5-2.2 (m, 4H, suc), 1.91 (d × d, 1H, J ) 6.6 Hz, â-Val),
1.73 (s, 3H, γ-Val), 1.62 (s, 3H, γ-Val), 1.19 (d, 3H, J ) 6.84 Hz,
2-C-CH3), 0.9 (d, 3H, J ) 6.52 Hz, 2-C-CH3).
Suc-Val-Cys(Ψpmppro)-Phe-pNA (2R and 2S). L-Cys hydrochlo-
ride monohydrate (3 g, 17.1 mmol) and potassium acetate (2.5 g, 18.4
mmol) were dissolved in water (26 mL) before the addition of
p-methoxybenzaldehyde (2.86 g, 21 mmol). The reaction mixture was
allowed to stand at room temperature for 3 h. A heavy white precipitate
developed, which was filtrated and washed several times with cold
ethanol before drying in vacuo to give dry (R)- and (S)-Cys(Ψpmppro)-
OH as a diastereomeric mixture. The above synthesized thiazolidine
derivative (0.5 g, 2.09 mmol), NMM (212 mg, 2 equiv) in THF (30
mL), and Fmoc-Val-F (710 mg, 2.1 mmol) were stirred at room
temperature to give Fmoc-Val-Cys(Ψpmppro)-OH with 90% yield
(HPLC). After passing the dipeptide through a silica column, 0.55 g
(0.8 mmol, 47%) of the pure compound was obtained. Fmoc-Val-
Cys(Ψpmppro)-OH (137 mg, 0.24 mmol) was dissolved in DCM (10
mL) at -10 °C before adding NMM (49 mg, 0.48 mmol) and
isobutylchloroformate (35 mg, 0.26 mmol), and stirring continued for
5 min. Phe-pNA (72 mg, 0.25 mmol, 1.05 equiv) in DCM (1 mL)
was added, and the reaction mixture was allowed to slowly warm to
roomtemperature. Without further workup, Fmoc-deprotection was
carried out by the addition of DBU (105 µL, 2.75 equiv, 15 min). All
liquid was evaporated, and the remaining yellow oil diluted in few
acetonitrile/water (2 mL, 1:1 v/v). The desired compound NH2-Val-
Cys(Ψpmppro)-Phe-pNA obtained as an epimeric mixture after purifica-
tion on a HPLC preparative (C18), using a gradient of 5-100% B within
30 min. Treating the epimers with succinic anhydride (40 mg, 2 equiv)
in the presence of NMM (40 µL) in DCM (5 mL) gave Suc-Val-Cys-
(Ψpmppro)-Phe-pNA 2S and 2R. Separation of the stereoisomers was
carried out by means of HPLC preparative using a gradient 55-65%
B in 30 min (C18). (S)-epimer (10%): Yield 24 mg (14%). HPLC
analytic: tR ) 11.33 min (60-100% B, 40 min). HPLC analytic: (R)-
One hundrend milligrams of the above-described substance was
dissolved with DMF (3 mL) before adding succinic anhydride (2 equiv,
42 mg) and NMM (2 equiv, 42 mg) and stirring for 12 h. The desired
product, tR ) 9.59 min (60-100% B, 40 min, C18, 68% purity) was
separated from the impurities and secondary products by means of a
C18 Sep-Pak column (isocratic, 30% A, 70% B) to give Suc-Val-
Ser(ΨH,Hpro)-Phe-pNA 5 with 88% purity (HPLC analytic). Purifica-
tion to a single peak on HPLC analytic was done on a preparative HPLC
(C18, isocratic, 35% A, 65% B). Overall yield 60 mg (0.1 mmol, 13%).
1H NMR (400 MHz, 303 K, DMSO-d6, 10 mg/mL) 10.6 (s, 1H, NH-

2720 J. Am. Chem. Soc., Vol. 120, No. 12, 1998
Keller et al.
epimer (90%): Yield 8 mg (4.7%). HPLC analytic tR ) 12.03 min
(60-100% B, 40 min). ESI-MS 706.3 [M + H]+. 1H NMR (400
MHz, 10 mg/mL, DMSO-d6, 300 K).
5.2 (d, 1H, J ) 6.51 Hz, R-ΨPro cis), 5.0 (d, 1H, J ) 8.35 Hz, R-ΨPro
trans), 4.75 (d × d, 1H, J ) 7.4 Hz, 14.6 Hz, R-Phe cis), (d × d, 1H
J ) 7.8 Hz, J ) 14.0 Hz, R-Phe trans), 4.47 (d × d, 1H, J ) 4.6 Hz,
J ) 8.8 Hz, R-Val cis), 4.47 (d × d, 1H, J ) 6.2 Hz, 8.9 Hz, R-Val
trans), 3.72 (s, 3H, OMe trans), 3.71 (s, 3H, OMe cis), 3.64 (d × d,
2H, J ) 6.5 Hz, 12.3 Hz, â-ΨPro cis), 3.15 (d × d, 2H, J ) 6.5 Hz,
12.3 Hz, â-ΨPro trans), 3.12 (m, 2H, â-Phe cis), 2.97 (m, 2H, â-Phe
trans), 2.49 (m, 4H, suc trans), 2.44 (m, 4H, suc cis), 2.02 (m, 1H,
â-Val cis), 1.99 (m, 1H, â-Val trans), 0.81 (m, 6H, γ-Val cis and trans),
0.73 (m, 3H, γ-Val trans), 0.62 (d, 3H, J ) 6.74 Hz, γ-Val cis).
(R)-Stereoisomer 2R (all trans): 10.5 (s, 1H), 8.66 (d, 1H, J ) 7
Hz, NH-Phe), 8.24 (d, 1H, J ) 8.7 Hz, NH-Val), 8.2 (2H, d, J ) 9.2
Hz, m-pNA), 7.82 (d, 2H, J ) 9.3 Hz, o-pNA), 7.68 (d, 2H, J ) 8.8
Hz, o-pmp), 7.25-7.35 (m, 4H, o, m-Phe), 7.2 (t, 1H, p-Phe), 6.86 (d,
2H, 8.8 J ) Hz, m-pmp), 6.71 (s, 1H, 2-C ΨPro), 4.67 (m, 1H,
R-ΨPro), 4.65 (m, 1H, R-Phe), 3.97 (t, 1H, 9 Hz, R-Val), 3.71 (s,
3H, OMe), 3.46 (m, 1H, â-ΨPro), 3.28 (m, 1H, â-ΨPro), 3.0 (m,
2H, â-Phe), 2.43-2.34 (m, 4H, Suc), 1.77 (m, 1H, â-Val), 0.58 (d,
3H, J ) 6.6 Hz, γ-Val), 0.33 (d, 3H, J ) 6.7 Hz, γ-Val).
(S)-Stereoisomer 2S (cis 41% and trans 59%): 10.65 (s, 1H, NH-
pNA cis), 10.58 (s, 1H, NH-pNA trans), 8.43 (d, 1H, J ) 7.3 Hz,
NH-Phe cis), 8.27 (d, 1H, J ) 7.7 Hz, NH-Phe trans), 8.2 (m, 4H,
m-pNA cis and trans), 7.81 (d, 2H, 9.3 Hz, o-pNA trans), 7.76 (d, 2H,
J ) 9.3 Hz, o-pNA cis), 7.65 (d, 1H, J ) 8.6 Hz, NH-Val cis), 7.54
(d, 1H, 8.9 Hz, NH-Val trans), 7.3-7.25 (m, 10H, aromatics-Phe, cis
and trans), 7.08 (d, 2H, J ) 8.9 Hz, o-pmp cis), (d, 2H, J ) 8.8 Hz,
o-pmp trans), 6.82 (d, 2H, J ) 9.3 Hz, m-pmp trans), 6.79 (d, 2H, J
) 9.3 Hz, m-pmp cis), 6.37 (s, 1H, 2-CH trans), 6.12 (s, 1H, 2-CH
cis), 5.73 (s, 2H, DCM impurity), 5.37 (s, 2H, COOH cis and trans)
Acknowledgment. The present work was supported by the
Swiss National Science Foundation.
Supporting Information Available: Tables comparing the
presented data with those obtained by other groups and 400
MHz NMR spectrum of the 2-C region of compound 2S in
correlation with the 400 MHz ROESY spectrum (3 pages, print/
PDF). See any current masthead page for ordering and Web
access instructions.
JA973966S