Homooligomers of substituted prolines and β-prolines: Syntheses and secondary structure investigation

Article abstract of DOI:10.1039/c3nj00127j

Homooligomers of enantiomerically pure (2S,3R)-3-methyl-proline, (3R,4R)-4-methyl-β-proline and (3R,4S)-3,4-dimethyl-β-proline were synthesized and studied using circular dichroism (CD) in water, methanol and propanol and using NMR in water. Changes in th

Full text of DOI:10.1039/c3nj00127j

New Journal of Chemistry
A journal for new directions in chemistry
www.rsc.org/njc
Volume 37 | Number 5 | May 2013 | Pages 1267–1632
ISSN 1144-0546
PAPER
Philippe Karoyan et al.
Homooligomers of substituted prolines and β-prolines: syntheses and
secondary structure investigation

NJC
View Article Online
View Journal | View Issue
PAPER
Homooligomers of substituted prolines and b-prolines:
syntheses and secondary structure investigation†
Cite this: NewJ.Chem., 2013,
37, 1312
a
a
b
a
´
Cecile Caumes, Nicolas Delsuc, Redouane Beni Azza, Isabelle Correia,
Fabrice Chemla,b Franck Ferreira,b Ludovic Carlier,a Alejandro Perez Luna,b
a
a
a
´
Roba Moumne, Olivier Lequin and Philippe Karoyan*
Homooligomers of enantiomerically pure (2S,3R)-3-methyl-proline, (3R,4R)-4-methyl-b-proline and
(3R,4S)-3,4-dimethyl-b-proline were synthesized and studied using circular dichroism (CD) in water,
methanol and propanol and using NMR in water. Changes in the far-UV CD spectrum were observed
from dimers to hexamers, but little change was observed from hexamers to octa- or nonamers, both in
water and methanol. CD and NMR data allowed us to conclude that oligomers of 3-substituted prolines
with more than six residues adopt a characteristic PPII secondary structure both in water and aliphatic
alcohols. Oligomers of (3R,4R)-4-methyl-b-proline bear the same CD signature as non-substituted
b-proline oligomers, suggesting that substitution at position 3 is not sufficient to reduce conformational
heterogeneity in b-proline oligomers. In the case of 3,4-disubstituted-b-proline oligomers, an atypical signature
with an extra negative band at around 225 nm was observed, together with a concentration dependent
CD spectrum indicating association properties. Nevertheless, NMR studies of 13C labelled oligomers of
3,4-disubstituted-b-prolines revealed a complex mixture of cis–trans conformers even for longer oligomers.
Received (in Montpellier, France)
6th November 2012,
Accepted 13th February 2013
DOI: 10.1039/c3nj00127j
www.rsc.org/njc
significant one is that like a-peptides, stabilisation of their
secondary structures almost always relies on specific backbone
Introduction
Biopolymers display a broad variety of structures that are
directly linked to their biological function. Native a peptides
made of natural a amino acids are often poor drug candidates
owing to bad bioavailability and high susceptibility to protease
degradation in biological media. In the last decade, foldamers,
i.e. short length oligomers with unnatural backbones that adopt
well-defined secondary structures, have been developed to
mimic biopolymers.1–4 They are useful tools for understanding
the folding propensities and the associated biological functions
of biopolymers. Even though foldamers could also represent
promising drug candidates since they may overcome the lack of
structural and biological stability of natural biopolymers, only a
few examples of foldamers that indeed interfere with biological
processes by mimicking natural protein have been reported
so far.5,6 This is probably due to two major drawbacks associated
with most of the foldamer backbones reported to date. The most
hydrogen-bonding patterns, which dramatically limit their
folding propensity in the aqueous biological medium. The
second issue is related to the functionalization of the foldamers
backbone with substituents that could participate in molecular
recognition, which is often rather tricky.
An interesting strategy to access synthetic backbones cap-
able of folding in water is to design monomers that have the
intrinsic propensity to fold without the contribution of hydrogen
bond stabilisation. This point has been addressed for example with
N-alkylglycine oligomers, i.e. peptoids, an important class of
foldamers that are able to adopt a wide range of secondary struc-
tures (PPI,7,8 PPII helices,9...), thanks to the control of the amide
bond cis–trans isomerism. This is for instance also done by Nature
that uses proline to build polyproline helical scaffolds. Among the
twenty natural amino acids this cyclic amino acid is unique in its
structure since the secondary amine group forms a tertiary amide
when involved in a peptidic bond thus preventing hydrogen-bond
formation. Although the pyrrolidine ring restricts the flexibility of
f and C peptide backbone dihedral angles, the formed tertiary
amide bond is more susceptible to cis–trans isomerism extending
the accessible conformational space. Despite the lack of hydrogen
bond, polyproline or proline rich peptide sequences can fold
into a helical conformation, i.e. the polyproline II (PPII) helix,
a
´
UMR7203, Laboratoire des Biomolecules, UPMC-ENS-CNRS, FR2769 Chimie
´
´
Moleculaire, Ecole Normale Superieure de Chimie, 24 rue Lhomond, 75252 Paris
Cedex 05, France. E-mail: philippe.karoyan@upmc.fr; Tel: +33 144322447
b
´
UMR 7201, Institut Parisien de Chimie Moleculaire, UPMC-CNRS, FR2769,
4 Place Jussieu, 75252 Paris Cedex 05, France
† Electronic supplementary information (ESI) available: Experimental details and
additional CD spectra. See DOI: 10.1039/c3nj00127j
c
1312 New J. Chem., 2013, 37, 1312--1319
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013

View Article Online
Paper
NJC
Fig. 1 Substituted proline and b-proline derivatives.
which is often involved in the recognition interface of protein–
protein interactions.10–12 Functionalized PPII helices could be
designed by using prolinoamino acids, i.e. proline chimeras,13
combining the pyrrolidine conformational restriction together
with amino acid side chain functionalities.
We previously reported an efficient synthetic route to access
to b-substituted proline analogues14–17 (Fig. 1, 1) by means of
the intramolecular carbozincation of zinc enolates derived
from N-homoallyl a-amino esters. A significant feature of this
approach is that it enables the introduction of various func-
tional groups at position 3. In particular, the synthesis of
proline chimeras with proteinogenic amino acid side chains
has been demonstrated.13,14,18–22
Scheme 1 Synthesis of b-ProMe2 monomer 3.
amino-enoate 5, readily obtained from N-allyl-N-((S)-1-phenyl-
ethyl)-amine and methyl 2-bromomethyl-acrylate following our
previously reported procedure.26 Carbocyclization of b-aminoester
6a by treatment with LDA followed by transmetalation with ZnBr2
yielded pyrrolidine 7a diastereo- and enantiomerically pure in 88%
yield at a gram-scale. Access to enantiopure (3R,4R)-3,4-dimethyl-
b-proline 3 from 7a was achieved efficiently by a three-step
sequence involving saponification of the methyl ester and
reductive cleavage of the 1-phenylethyl chiral auxiliary by
catalytic hydrogenation over Pd/C in methanol.
By a similar strategy, access to (3R,4R)-4-methyl-b-proline 2,
i.e. b-prolinovaline, devoid of the a methyl substituent proved
troublesome as reproducibility problems were met at the hydro-
genolysis stage. Thus, we developed an alternate route involving
benzyl ester derivatives (Scheme 2). Cyclization precursor 6b
was obtained by reaction of amine 4 with benzyl acrylate in the
presence of DBU. Cyclization of 6b provided methylpyrrolidine
7b in 54% yield, diastereo- and enantiomerically pure. Depro-
tection leading to cyclic amino acid 2 was then readily per-
formed in a single step by catalytic hydrogenation over Pd/C.
Finally, we also synthesized a 13C-labelled analogue of com-
pound 3, i.e. 3* that is 13C-labelled on a specific methyl group
(Scheme 3). This compound was prepared in order to help the
These monomers allow combining the conformational con-
straints and lack of hydrogen-bond donor of the proline residue
in a peptide while keeping the opportunity to decorate them
with proteinogenic side chains. We have shown that these
residues can be used to build functionalized b-turns that are
stable in water,23 or to introduce a specific functional group in
PPII folded peptides that inhibit peptide–protein24 or protein–
protein interaction.25 Recently the carbocyclization synthetic
strategy has been applied to the preparation of b-proline
analogues substituted at position 4 (Fig. 1, 2) and di-substituted
at positions 3 and 4 (Fig. 1, 3).17,26 We report here a structural
study using CD and NMR of the folding abilities of our
substituted proline and b-proline derivative oligomers, i.e.
prolinovalines (1 RQMe, Fig. 1) and b-prolinovalines (2 and 3,
RQMe, Fig. 1) in order to evaluate the influence of the
substitution of the pyrrolidine ring on the folding of the
polymers. This work represents the first step toward the rational
design of water-stable functionalized polyproline-like foldamers.
The foremost goal of this work is to use these new foldamers
to mimic proteins that interact with their partner through
an extended domain in order to target interesting biological
interactions.
Results and discussion
Synthesis
Amino acid 1, i.e. prolinovaline, was synthesized according to
the procedure previously described.14 By this method the cis-
(2S,3R)-prolinovaline 1 was obtained in enantiomerically pure
form and at a gram-scale.
The required cyclic b-amino acids 2 and 3 were prepared in
diastereo- and enantiomerically pure forms by intramolecular
carbocyclization of zinc enolates derived from N-allyl b-amino
esters (Scheme 1).26 For gram-scale synthesis, precursor 6a
was best prepared by reduction with magnesium in MeOH of Scheme 2 Synthesis of b-prolinovaline monomer 2.
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013
New J. Chem., 2013, 37, 1312--1319 1313

View Article Online
NJC
Paper
Scheme 3 Synthesis of 13C-labelled b-ProMe2 monomer 3*.
oligomer NMR signal interpretation. Indeed, NMR study of
proline derived oligomers can be complicated due to cis–trans
interconversion of tertiary amide bonds giving rise to
numerous sets of NMR signals, hampering spectra interpreta-
tion. Thus, the preparation of 13C-labelled amino acid 3* was
also achieved via the carbocyclization methodology (Scheme 3).
The 13C-labelled methyl group was introduced with labelled
iodomethane by methylation of the lithium enolate derived
from 6b. Pyrrolidine 7c was then obtained by carbocyclization
of enoate 6c and the amino acid 3* was obtained after catalytic
hydrogenation over Pd/C.
A final Fmoc protection step of amino groups yielded
monomers 8, 10a, 10a* and 10b suitable for the solid phase
synthesis of the oligomers (Scheme 4). For monomers of
N-Fmoc-(2S,3R)-prolinovaline 8 and N-Fmoc-(3R,4R)-3,4-dimethyl-
b-proline 10a and 10a*, coupling reactions were achieved after
conversion to the corresponding acyl chloride. Indeed, the
steric hindrance of these compounds hampered getting satis-
factory yields using standard peptide coupling reagents (HBTU,
HATU or PyBroP). In the case of 3,4-dimethyl-b-proline 10a and
10a*, the activation of the carboxylic function through acyl
chloride preparation was required only for the coupling of the
first amino acid to the resin, HATU giving satisfactory yields in
the next steps (Scheme 4). Since charges at both extremities of
Fig. 2 Oligomers prepared for CD studies. (a) Pentamer 14d was prepared with
the enantiomer of monomer 3 (Fig. 1) and is thus the enantiomer of the
compound drawn.
polyproline oligomers can dramatically influence the PPI–PPII
conformational equilibrium,27 the oligomers were acetylated at
the N-terminus and produced as carboxamides. After cleavage
from the resins and purification by RP-HPLC, three series of
oligomers (i.e. 12, 13, 14, Fig. 2) were obtained, respectively, for
prolinovaline derivatives (from dimer to nonamer, 12a to 12h),
for b-prolinovaline derivatives (from dimer to heptamer, 13a to
13f), and for 3,4-dimethyl-b-proline derivatives (from dimer to
octamer, 14a to 14g).
Three additional compounds in series 14 have been syn-
thesized for NMR studies, namely 14h, 14i and 14j (Fig. 2). They
are homooligomers of increasing chain length (respectively
trimer, heptamer and nonamer), incorporating one 13C-labelled
monomer into the central position. The 13C will be used as a
reporter of the cis–trans conformational heterogeneity.
Circular dichroism and NMR conformational studies
Circular Dichroism (CD) is a commonly used technique for
rapid identification of secondary structures in proteins. It also
provides insight into the folding propensities of non-natural
oligomers or polyproline oligomers when other techniques fail.
For example, PPII conformations are difficult to establish using
NMR since the extended conformation prevents the observa-
tion of through space correlation (NOEs) between non-adjacent
residues. The existence of cis–trans interconversion equilibria
also often hampers the interpretation of NMR spectra.
Oligomers built from proline adopt extended conformations
in solution, i.e. PPII conformation in water and PPI conforma-
tion in aliphatic alcohols such as MeOH or PrOH. In the case of
polyproline conformations, characteristic CD signatures are
observed: a weak maximum at 226 nm and a strong minimum
at 206 nm in water for PPII conformation, and weak minima at
200 and 232 nm, together with a strong maximum at 215 nm in
aliphatic alcohols for PPI conformation.10 PPII–PPI helix inter-
conversion has been studied using non-natural proline surro-
gates substituted at various positions.28–32
Scheme 4 Suitable N- and C-derivatisation of monomers for oligomers solid
phase syntheses.
c
1314 New J. Chem., 2013, 37, 1312--1319
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013

View Article Online
Paper
NJC
To perform a comparative conformational study in water and is nevertheless consistent with the establishment of a PPII
and aliphatic alcohols, we ensured that the oligomers were well conformation.33 We used NMR spectroscopy to confirm the
soluble. Interestingly, all oligomers showed a solubility 4mM presence of major conformers in water. The 1H NMR spectra of
in all solvent systems. The water solubility of these oligomers is oligomers 12a–h cannot be completely assigned owing to the
promising for construction of foldamers having stable confor- poor chemical dispersion observed for such homopolymers.
mation in biological media.
However the methyl resonance is a convenient probe to assess
cis–trans isomerism. Indeed, for each oligomer, major methyl
resonances are observed (Fig. 4). Their number corresponds to
the number of prolinovaline units, as can be seen in dimer 12a,
trimer 12b and tetramer 12c (Fig. 4). This indicates that a major
conformer is populated in solution. The analysis of 2D ROESY
spectra (Fig. 5) shows strong Ha–Hd correlations and weak
Ha–Ha correlations, demonstrating that the trans conforma-
tion of peptide bonds predominates.
Upon chain elongation, variation of the normalized intensity
of the negative band on CD spectra is not linear and reaches a
plateau at six residues (Fig. 3a, inset). This suggests that from
six residues the CD spectra do not depict anymore to the sum of
the chiral contribution of each monomer but a defined chiral
secondary structure in water. For the octamer, upon increasing
the temperature from 20 to 90 1C, the shape of the CD signal
remains unchanged, only a slight decrease in intensity is
observed (see ESI†). This indicates that the secondary structure
adopted by the prolinovaline octamer is stable in this tempera-
ture range. In MeOH, the negative band also shifts from 200 to
210 nm upon chain elongation (Fig. 3b). As compared to water,
an additional positive band is observed at 235 nm for the longer
oligomers in MeOH and PrOH. The spectra of heptamer 12f
(Fig. 3c) are superimposable in alcohols and very close to the
spectrum obtained in water. As we already reported, the substitu-
tion at position 3 of the pyrrolidine ring can be accommodated in
PPII conformations,24,25 thus these data all together suggest that
as in water, the PPII conformation is also reached in alcohols.
Prolinovaline series 12a to 12h
Since prolinovaline and proline have closely related structures
(only an additional methyl at position 3) similar behaviour for
the oligomers in solution was expected. Whatever the solvent
used, the dimer’s CD spectra show a negative band at 200 nm
and a positive band at 218 nm. CD spectra of longer oligomers
recorded in water (Fig. 3a) show only one strong negative band
that shifts from 200 to 210 nm. The positive band at 215 nm
disappeared and more surprisingly, no band is observed at
around 225 nm where polyprolines usually show a small
positive band. This phenomenon has already been reported
b-Prolinovaline series 13a to 13f
The folding properties of oligomers built from non-substituted
b-proline monomers have already been investigated in solution.
Although a first study by Yuki et al. suggested that poly-
b-prolines exist in random coiled conformation,34 Kim et al.35
described an ordered conformation for the b-proline decamer
in 10 mM phosphate buffer (pH 7.0) on the basis of its CD
spectrum exhibiting a large positive band at 215 nm and a
negative band at 198 nm. Moreover, Gellman and coworkers36
have shown that the mean residue ellipticity of b-proline
oligomers is independent of the chain length from four resi-
dues suggesting that a stable secondary structure is established
in MeOH. This behaviour is similar to that of a-proline oligo-
mers. Nevertheless, further investigation by NMR on these
oligomers showed conformational heterogeneity in solution
since a mixture of cis–trans amide bond rotamers is observed.37
So far, conformational stabilisation in b-proline oligomers has
only been achieved by restriction of the conformational flexi-
bility of the monomers, using di-substituted derivatives at
position 5 of the pyrrolidine ring,38 or using a,d ethano-bridge
b-proline derivatives, both favouring cis rotamers,39 or a,g methano-
bridge b-proline selecting trans rotamers.40
Fig. 3 CD spectra of prolinovaline oligomers 12a–h recorded at 20 1C using a
cell length of 0.1 cm. Concentrations are 400 mM for 12a–c, 300 mM for 12d,
200 mM for 12e–g and 100 mM for 12h (a) 12a–h in H2O, inset: [y] at 211 nm as a
function of chain length, (b) 12a–h in MeOH, inset: [y] at 211 nm as a function of
chain length. (c) Superimposition of the heptamer (12f) CD spectra recorded in
water, MeOH and PrOH.
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013
New J. Chem., 2013, 37, 1312--1319 1315

View Article Online
NJC
Paper
Fig. 6 CD spectra of b-Proval series, i.e. oligomers 13a–f recorded at 20 1C using
a 0.1 cm cell length. Concentrations were set at 500 mM for 13a–b, 200 mM for
13c–d and 100 mM for 13e–f. (a) In H2O, inset: [y] at 215 nm as a function of
chain length. (b) Superimposition of the 13f heptamer CD spectra recorded in
water, MeOH and PrOH.
Fig. 4 1D 1H NMR spectra of oligomers 12a–c and 12g recorded in D2O at 25 1C
showing the methyl resonances. Methyl signals appear as doublets owing to 3J
coupling with the Hb proton.
band at 197 nm and one positive band at 215 nm (Fig. 6). This
shape is identical to that observed for unsubstituted b-proline
oligomers.36 The normalised intensity of both bands increases
with chain elongation in a non-linear manner, reaching a
plateau from 5 residues. CD spectra of the heptamer were
recorded at different temperatures and upon heating to 90 1C
no conformational transition was observed, evidenced by only
slight changes in the spectra (see ESI†). These data would
possibly indicate the establishment of a stable secondary
structure. However, the same behaviour was observed for
non-substituted b-proline oligomers suggesting that these CD
spectra likely reflect equilibrium between several secondary
structures rather than the presence of a unique secondary
structure in solution. The same trend is observed in MeOH
and PrOH. The superimposition of the CD spectra recorded in
water, MeOH and PrOH suggests that this conformational
equilibrium is not influenced by solvation effects (Fig. 6).
Me2-b-Pro series 14a to 14g
Before investigating the conformations adopted by all oligomers
of each series described herein (12, 13 and 14), the CD spectra of
the tetramers (i.e. 12c, 13c and 14c) were recorded at different
concentrations ranging from 25 mM to 1 mM. For compounds 12c
and 13c, the tetramer signals were superimposable meaning that
Fig. 5 2D 1H–1H ROESY spectrum of oligomer 12h recorded in D2O at 25 1C
(300 ms mixing time) showing the strong Ha–Hd correlations characteristic of
trans peptide bond conformation.
In an attempt to investigate the influence of the substitution no association occurred for these oligomers whatever the con-
at position 4 of the pyrrolidine ring of b-proline on the secondary centrations were (see ESI†). A different behaviour was evidenced
structure, CD experiments were performed at 20 1C in water for tetramer 14c, in both water and MeOH (Fig. 7). Indeed, in this
(Fig. 6) and MeOH (see ESI†). b-Prolinovaline oligomers have a case, a concentration dependent signal was observed. At 25 mM,
CD signal bearing two bands of equal intensity: one negative the CD spectrum exhibited two bands: a minimum at 195 nm
c
1316 New J. Chem., 2013, 37, 1312--1319
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013

View Article Online
Paper
NJC
Fig. 8 1D 13C NMR spectra of oligomers 14h–j in D2O at 25 1C showing the
methyl resonance.
The conformational properties of this series were further
analyzed using NMR spectroscopy. The NMR spectra exhibit low
chemical shift dispersion and high complexity due to chemical
shift heterogeneity. The NMR spectra of dimer 14a could be
unambiguously assigned. Four forms with roughly identical
populations are observed, corresponding to cis–trans rotamers
of both amide bonds in slow exchange on the NMR time scale
(ESI†). In order to facilitate the conformational analysis of larger
oligomers, we incorporated a 13C-labelled methyl group into a
single unit occupying the central position in oligomers. 1D 13C
NMR spectra were recorded to probe the cis–trans interconver-
sions of amide bonds around the central unit. The 1D 13C NMR
spectrum of trimer 14h (Fig. 8) shows eight signals of compar-
able intensities that correspond to cis–trans isomerism of the
three amide bonds. The similar populations prove that there is
no conformational preference around the amide bonds in the
trimer. An even larger number of peaks is observed on the
13C NMR spectra of heptamer 14i and nonamer 14j (Fig. 8).
Thus the methyl 13C signal appears to be very sensitive to cis–
trans isomerism of several amide groups, and not only those
directly connecting the labelled unit. The high complexity of
13C NMR spectra clearly demonstrates that there is no stabilisa-
tion of an amide bond conformer, even in longer oligomers.
Fig. 7 CD spectra of Me2-b-Pro series, i.e. oligomers 14a–g recorded at 20 1C
using a 0.1 cm cell length. (a) CD spectra of 14c at different concentrations
in H2O. (b) CD spectra of 14c at different concentrations in MeOH. (c) CD spectra
of 14a–g in H2O. Concentrations were set at 400 mM for 14a–c and 200 mM
for 14d–g.
and a maximum at 210 nm, which was similar to that of
b-prolinovaline oligomers. From 25 to 100 mM, the intensities
of both bands decreased and a new band at 225 nm appeared in
H2O. Above 100 mM, no change was observed anymore. These
results suggest that the tetramer self-assembles above 75 mM in
H2O. In MeOH, the same trend was observed and self-assembling
seems to be completed at around 500 mM.
Since self-assembling is often associated with the establish-
ment of a more stable secondary structure, we decided to
perform the experiments at 100 mM, the lower concentration
at which no more variation is observed on the CD spectrum.
Upon elongating the chain, the signals intensity at 210 and
225 nm increases whereas it decreases at 195 nm (Fig. 7c).
Variations of the normalised intensity of the bands at 195, 210
and 225 nm show different behaviours: a saturation of the
Experimental
Peptide synthesis
signal seems to be reached at 195 and 225 nm whereas a linear Oligomers were synthesized manually on a solid support using an
trend is observed at 210 nm (see ESI†). These findings pre- Fmoc-rink amide resin (substitution level = 0.43–0.57 mmol gÀ1
vented us to conclude regarding the establishment of a unique resin, 100–200 mesh). The resin was swollen in DCM for 15 min
stable conformation in water.
prior to use. Fmoc removal was performed by using 20% (v:v)
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013
New J. Chem., 2013, 37, 1312--1319 1317

View Article Online
NJC
Paper
piperidine in 1-methyl-2-pyrrolidone (NMP) once for 1 min and solutions were evaporated. The peptide was triturated with cold
then once for 15 min. The resin was washed five times Et2O, collected by centrifugation, and the pellet was washed
with NMP.
three times with cold Et2O. The pellet was then dried for
Fmoc-amino acid 8 couplings. Acyl chloride 9 was formed 3 hours under vacuum, redissolved in deionized water and freeze
using 1-chloro-N,N-2-trimethyl-1-propenylamine (1.1 eq.) in dried. The peptide dissolved in deionized water containing 0.1%
anhydrous DCM (0.2 M) under an Ar atmosphere. The reaction of TFA (v:v) was purified by reversed phase HPLC using a C18
mixture was stirred at room temperature for 1 h 30 min and was column with an acetonitrile–water gradient both solvents con-
directly used for amino acid coupling on the resin: the resin taining 0.1% TFA.
was suspended in anhydrous DCM, N,N-diisopropylethylamine
(4 eq.) then the solution of acyl chloride (2 equivalents) was
Sample preparation of circular dichroism spectroscopy
added. After 1 hour the solution was removed by filtration and Stock solutions of each peptide were prepared in deionised H2O
the resin was washed with DCM five times then rinsed with and were in the millimolar range. The peptide concentration
NMP 3 times. Reaction completion was checked by the Kaiser was determined by NMR using sodium 4,4-dimethyl-4-silapentane-
or chloranil test for primary or secondary amines, respectively. 1-sulfonate (DSS) as an internal reference compound (see ESI†
In the case of a positive test, the same amino acid was condensed for more details).
again until getting a negative test. Cycles of deprotection-washing–
coupling-washing were repeated until the desired sequence was prepared by dilution of stock solutions. To prepare MeOH
achieved.
solutions, the required amount of stock solution was freeze-
CD solutions of desired concentration in H2O were further
Fmoc-amino acid 10a (and 10a*) couplings. Except for the dried and MeOH was added. MeOH samples were heated at
first residue, all couplings were performed using 3 equivalents 50 1C for 1 hour and then allowed to stand at room temperature
of Fmoc-amino acid 10a (or 10a*), 3 equivalents of HATU and for 24 hours prior to recording CD spectra.
HOAt and 6 equivalents of DIEA dissolved in NMP (0.15 M).
Circular dichroism spectroscopy
Each coupling reaction was allowed to proceed for 0.5 h at room
temperature. The solution was then removed by filtration and CD measurements were performed using a Jasco J 815 spectro-
the resin was washed 5 times with NMP. Reaction completion polarimeter equipped with a Peltier to control the temperature,
was checked by performing the chloranil test. In the case of a using quartz cells with a path length of 0.1 cm, a scan speed of
positive test, the coupling was repeated. Cycles of deprotection- 50 nm minÀ1 and a resolution of 0.2 nm. Spectra were recorded
washing–coupling-washing were performed until the desired as an average of four scans. For each spectrum, the background
sequence was achieved. The first coupling required prior for- was subtracted prior to smoothing and processing. All the
mation of the acyl chloride 11. 11 was obtained using SOCl2 measured ellipticities are presented normalised as a function
(47 eq.) in toluene. The reaction was allowed to proceed at 90 1C of the concentration of solutions and the number of chromo-
for 0.5 hour and the solvents were removed. After Fmoc phores (amide bonds).
removal, the resin was rinsed three times with dried DCM
and kept under an Ar atmosphere. Anhydrous DCM and dried
NMR spectroscopy
DIEA were added followed by the solution of the acyl chloride in NMR experiments were performed on a Bruker Avance III
dried DCM. The resin was shaken at room temperature for 500 MHz spectrometer equipped with a TCI cryoprobe. NMR
0.5 hour, rinsed five times with DCM. And coupling completion spectra were recorded in D2O at temperatures between 25 1C
was monitored by the Kaiser test.
and 35 1C. 1H and 13C chemical shifts were referenced to
Fmoc-amino acid 10b couplings. Stepwise couplings were sodium 4,4-dimethyl-4-silapentane-1-sulfonate (DSS). NMR
accomplished with 3 equivalents of Fmoc-amino acid 10b, studies of small oligomers were based on the analysis of 2D
3 equivalents of HBTU and HOBt and 6 equivalents of DIEA in homonuclear COSY, TOCSY and NOESY spectra, and 2D 1H–13C
HSQC, HMBC, and HSQC-TOCSY spectra. 2D NOESY experi-
NMP (0.15 M). Each coupling reaction was allowed to proceed for
0.5 h at room temperature. The solution was then removed by ments were recorded with a mixing time of 1 s.
filtration and the resin was washed with NMP five times.
Reaction completion was checked by the Kaiser or chloranil test
for primary or secondary amines, respectively. In the case of a
Conclusions
positive test, the same amino acid was condensed again until Homooligomers of enantiomerically pure (2S,3R)-3-methyl-proline,
getting a negative test. Cycles of deprotection-washing–coupling- (3R,4R)-4-methyl-b-proline and (3R,4S)-3,4-dimethyl-b-proline
washing were repeated until the desired sequence was achieved. were synthesized and studied using circular dichroism (CD)
Acetylation, cleavage and purification. Final acetylation of in water and alcohols. In each series, changes in the far-UV CD
all oligomers was performed using 32 equivalents of acetic spectrum were observed from dimers to hexamers, but little
anhydride in DCM (1.4 M) for 1 hour at room temperature. change was observed from hexamers to octa- or nonamers, in
Cleavage of peptides was accomplished by treatment of the water, methanol and propanol. We conclude from CD and NMR
resin with 3 mL of TFA/triisopropylsilane/H2O (95/2.5/2.5 v:v:v) studies that oligomers of 3-substituted prolines with more than
for 3 h at room temperature. The TFA solution was collected six residues adopt a characteristic PPII secondary structure.
and the beads rinsed 3 times with neat TFA. The combined TFA CD data on b-prolinovaline oligomers show close similarities
c
1318 New J. Chem., 2013, 37, 1312--1319
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013

View Article Online
Paper
NJC
17 A. Perez-Luna, C. Botuha, F. Ferreira and F. Chemla, New J.
Chem., 2008, 32, 594–606.
18 C. Mothes, S. Lavielle and P. Karoyan, J. Org. Chem., 2008,
73, 6706–6710.
19 J. Quancard, H. Magellan, S. Lavielle, G. Chassaing and
P. Karoyan, Tetrahedron Lett., 2004, 45, 2185–2187.
20 J. Quancard, A. Labonne, Y. Jacquot, G. Chassaing,
S. Lavielle and P. Karoyan, J. Org. Chem., 2004, 69,
7940–7948.
21 P. Karoyan, J. Quancard, J. Vaissermann and G. Chassaing,
J. Org. Chem., 2003, 68, 2256–2265.
22 P. Karoyan and G. Chassaing, Tetrahedron: Asymmetry, 1997,
8, 2025–2032.
with unsubstituted b-prolines, indicative of cis–trans conforma-
tional heterogeneity in solution, with no influence of the
solvent. In the case of 3,4-disubstituted-b-prolines, the concen-
tration dependent CD spectra reveal association properties.
An atypical CD signature was also observed with an extra negative
band at around 225 nm. However NMR studies clearly show the
absence of favoured amide rotamers even in the longest oligo-
mers. Therefore the conformational restrictions brought by the
two methyl substituents are not sufficient to stabilize unique
amide bond rotamers of b-prolines. From the structural studies
performed here with these proline analogues, we can conclude
that only 3-substituted prolines appear to be valuable tools for
building short, stable and functionalized PPII helix mimetics.
23 C. Mothes, M. Larregola, J. Quancard, N. Goasdoue,
S. Lavielle, G. Chassaing, O. Lequin and P. Karoyan,
Chembiochem, 2010, 11, 55–58.
Acknowledgements
This work was supported by the Agence National de la Recherche
(ANR-10-BLAN-707). We gratefully thank Christophe Desmarets
for valuable help with the circular dichroism spectrometer.
24 S. Sagan, J. Quancard, O. Lequin, P. Karoyan, G. Chassaing
and S. Lavielle, Chem. Biol., 2005, 12, 555–565.
25 Y. Jacquot, I. Broutin, E. Miclet, M. Nicaise, O. Lequin,
´
N. Goasdoue, C. Joss, P. Karoyan, M. Desmadril, A. Ducruix
and S. Lavielle, Bioorg. Med. Chem., 2007, 15, 1439–1447.
26 F. Denes, A. Perez-Luna and F. Chemla, J. Org. Chem., 2007,
72, 398–406.
Notes and references
1 S. H. Gellman, Acc. Chem. Res., 1998, 31, 173–180.
2 D. J. Hill, M. J. Mio, R. B. Prince, T. S. Hughes and 27 M. Kuemin, S. Schweizer, C. Ochsenfeld and H. Wennemers,
J. S. Moore, Chem. Rev., 2001, 101, 3893–4012. J. Am. Chem. Soc., 2009, 131, 15474–15482.
3 G. Guichard and I. Huc, Chem. Commun., 2011, 47, 5933–5941. 28 M. V. Sonar and K. N. Ganesh, Org. Lett., 2010, 12,
4 S. Hecht and Y. Huc, Foldamers: structure, properties and
applications, Wiley-VCH, Weinheim, Germany, 2007.
5 T. Edwards and A. Wilson, Amino Acids, 2011, 41, 743–754.
5390–5393.
29 M. Ku¨min, L.-S. Sonntag and H. Wennemers, J. Am. Chem.
Soc., 2006, 129, 466–467.
6 C. M. Goodman, S. Choi, S. Shandler and W. F. DeGrado, 30 R. Zhang, F. Brownewell and J. S. Madalengoitia, J. Am.
Nat. Chem. Biol., 2007, 3, 252–262. Chem. Soc., 1998, 120, 3894–3902.
7 K. Kirshenbaum, A. E. Barron, R. A. Goldsmith, P. Armand, 31 M. Kuemin, Y. A. Nagel, S. Schweizer, F. W. Monnard,
E. K. Bradley, K. T. V. Truong, K. A. Dill, F. E. Cohen and
R. N. Zuckermann, Proc. Natl. Acad. Sci. U. S. A, 1998, 95,
4303–4308.
8 J. R. Stringer, J. A. Crapster, I. A. Guzei and H. E. Blackwell,
J. Am. Chem. Soc., 2011, 133, 15559–15567.
C. Ochsenfeld and H. Wennemers, Angew. Chem., Int. Ed.,
2010, 49, 6324–6327.
32 D. G. McCafferty, D. A. Friesen, E. Danielson, C. G. Wall,
M. J. Saderholm, B. W. Erickson and T. J. Meyer, Proc. Natl.
Acad. Sci. U. S. A., 1996, 93, 8200–8204.
9 N. H. Shah, G. L. Butterfoss, K. Nguyen, B. Yoo, R. Bonneau, 33 Y.-C. Chiang, Y.-J. Lin and J.-C. Horng, Protein Sci., 2009, 18,
D. L. Rabenstein and K. Kirshenbaum, J. Am. Chem. Soc.,
2008, 130, 16622–16632.
10 F. Rabanal, M. D. Ludevid, M. Pons and E. Giralt, Biopolymers,
1993, 33, 1019–1028.
11 B. K. Kay, M. P. Williamson and M. Sudol, FASEB J., 2000,
14, 231–241.
12 M. P. Williamson, Biochem. J., 1994, 297, 249–260.
1967–1977.
34 H. Yuki, Y. Okamoto and Y. Kobayashi, J. Polym. Sci., Part A:
Polym. Chem., 1979, 17, 3867–3878.
35 Y. J. Kim, D. A. Kaiser, T. D. Pollard and Y. Ichikawa, Bioorg.
Med. Chem. Lett., 2000, 10, 2417–2419.
36 B. R. Huck, J. M. Langenhan and S. H. Gellman, Org. Lett.,
1999, 1, 1717–1720.
13 P. Karoyan, S. Sagan, O. Lequin, J. Quancard, S. Lavielle and 37 B. R. Huck, J. D. Fisk, I. A. Guzei, H. A. Carlson and
G. Chassaing, in Targets in Heterocyclic Systems-Chemistry S. H. Gellman, J. Am. Chem. Soc., 2003, 125, 9035–9037.
and Properties, ed. O. A. Attanasi and D. Spinelli, Royal 38 B. R. Huck and S. H. Gellman, J. Org. Chem., 2005, 70,
Society of Chemistry, Cambridge, 2005, vol. 8, pp. 216–273. 3353–3362.
14 P. Karoyan and G. Chassaing, Tetrahedron Lett., 1997, 38, 39 M. Hosoya, Y. Otani, M. Kawahata, K. Yamaguchi and
85–88. T. Ohwada, J. Am. Chem. Soc., 2010, 132, 14780–14789.
15 E. Lorthiois, I. Marek and J.-F. Normant, Tetrahedron Lett., 40 G. R. Krow, N. Liu, M. Sender, G. Lin, R. Centafont,
1997, 38, 89–92.
16 E. Lorthiois, I. Marek and J. F. Normant, J. Org. Chem., 1998,
63, 2442–2450.
P. E. Sonnet, C. DeBrosse, C. W. Ross, P. J. Carroll,
M. D. Shoulders and R. T. Raines, Org. Lett., 2010, 12,
5438–5441.
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013
New J. Chem., 2013, 37, 1312--1319 1319