Conformational stability studies of a stapled hexa-β3- peptide library

Article abstract of DOI:10.1039/c2ob06617c

A library of 14-helical hexa β³-peptides was synthesized in order to determine the influence of sequence variation as well as staple size and location on conformational stability. From this study we show that appropriately stapled hexa-β³-peptides can allow for a number of variations without significant perturbation of the 14-helix.

Full text of DOI:10.1039/c2ob06617c

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Organic &
Biomolecular
Chemistry
Cite this: Org. Biomol. Chem., 2012, 10, 1802
www.rsc.org/obc
PAPER
Conformational stability studies of a stapled hexa-β3-peptide library†‡
Romila D. Gopalan§,a,b Mark P. Del Borgo§,b Ylva E. Bergman,a Sharon Unabia,b Roger J. Mulder,c
Matthew C. J. Wilce,a Jacqueline A. Wilce,b Marie-Isabel Aguilar*b and Patrick Perlmutter*a
Received 22nd September 2011, Accepted 6th December 2011
DOI: 10.1039/c2ob06617c
A library of 14-helical hexa β3-peptides was synthesized in order to determine the influence of sequence
variation as well as staple size and location on conformational stability. From this study we show that
appropriately stapled hexa-β3-peptides can allow for a number of variations without significant
perturbation of the 14-helix.
staples in stabilizing the helical conformation in solution. Staples
Introduction
employed so far include disulfides,10 salt-bridges11,12 and
Acyclic β3-peptides show a remarkable propensity to form a
variety of stable conformations in solution.1 A major focus of
interest has been on one particular conformation, the 14-helix,
which has a pitch of ∼4.8 Å and almost three residues per turn.2
An important consequence of this is that side chains along each
face are aligned perfectly (see Fig. 1). This feature has been
exploited by several groups for the development of new
materials.3–8
lactams.7 Recently, we introduced the use of ring closing metath-
esis (RCM) to staple a series of hexa-β3-peptides.9
In this paper we describe studies designed to probe the
efficacy of our staple in stabilizing the 14-helix. We also show
how the alkene, which rests across the surface of the helix, can
be functionalised without disruption of the helical conformation
or decomposition of the peptide backbone. In particular, we
designed a library of hexa-β3-peptides in order to determine the
influence of various structural features on the conformational
stability of the 14-helix. These included:
Several strategies have been reported for the generation of
“stapled” residues along one face.7–10 The intention of these
prior studies was to establish the relative efficacy of the new
(a) sequence variation
(b) staple location
(c) staple ring size
(d) chemical modifications of the staple
Results and discussion
Fig. 1 A schematic representation of a 14-helical hexa-β3-peptide illus-
trating the alignment of residues at i, i + 3 providing three surfaces on
the peptide.
All peptides were synthesised on Wang resin using standard
SPPS protocols. RCM was performed on resin using Hoveyda–
Grubbs 2nd generation catalyst (35 mol%) utilising the same
conditions as described previously.9,13,14 All peptide RCMs gave
the E-alkene product with high selectivity and conversion in all
cases was >90% as evidenced by HPLC (Scheme 1).
NMR spectroscopy was used to determine the solution struc-
ture of a sample pair of N-acetyl-β3-hexapeptides. Thus NOESY
experiments of diene 3f and stapled 4f in d3-MeOH revealed all
the expected i/i + 2 and i/i + 3 connectivities for a 14-helix
(Fig. 2). These results strongly support the assignment of 14-
helical conformations for these hexa-β3-peptides.
CD analysis of each stapled and unstapled peptide pair was
employed to determine the influence of various structural
changes on conformational stability. (NOESY experiments were
also used in select cases, vide infra). Typically, for 14-helical β3-
peptides, a minimum is observed around 211–214 nm with a
maximum between 195 and 198 nm.1 The CD spectrum of each
aSchool of Chemistry, Monash University, Clayton, VIC 3800, Australia.
E-mail: patrick.perlmutter@monash.edu; Fax: +613 9905 4597;
Tel: +613 9905 4522
bDepartment of Biochemistry & Molecular Biology, Monash University,
Clayton, VIC 3800, Australia. E-mail: mibel.aguilar@monash.edu;
Fax: +613 9902 9500; Tel: +613 9905 3723
cCSIRO, Clayton, VIC 3169, Australia. E-mail: roger.mulder@csiro.au;
Fax: +613 9545 2552Tel: +613 9545 2550
†This article is part of an Organic & Biomolecular Chemistry web
theme issue on Foldamer Chemistry.
‡Electronic supplementary information (ESI) available: Full description
of peptide synthesis and functionalization; HPLC and MS data; CDs and
thermal melts as well as a histogram of CD minima for all peptides in
different solvents; 1H NMR and NOESY spectra for 3m and 4m. See
DOI: 10.1039/c2ob06617c
§These authors contributed equally to the study
1802 | Org. Biomol. Chem., 2012, 10, 1802–1806
This journal is © The Royal Society of Chemistry 2012

View Article Online
Scheme 1 Synthetic scheme for peptides. (i) HG-II, 20–35 mol%,
TFE/CH2Cl2 (4 : 1), rt, 48 h; (ii) Pip., DMF; (iii) Ac2O, DMF; (iii) (iv)
TFA, 95%.
Fig. 3 CD spectra (upper) and thermal melts (lower) of selected pep-
tides with varying sequences in 25% phosphate buffer/AcCN (pH 7.4).
generally greater than in TFE (See Fig. S4†). Moreover, all
stapled and unstapled peptides showed more consistent elliptici-
ties in MeOH than in the other two solvents. Closer inspection
of the effect of specific sequence changes on the ellipticity
revealed some trends. For example, 3/4a–c (central residues are
Leu/Val) showed somewhat higher ellipticity in all solvents than
the closely related 3/4d–f in which the central residues contain
Ala. Peptides 3/4g–j, in which the central Val/Leu residues have
been replaced with Ile or Phe, show similar ellipticities in MeOH
and MeCN and lower values in TFE. While it is not possible to
analyse the spectra more quantitatively, these results suggest that
the degree of 14-helical structure is strongly influenced by
solvent, irrespective of whether the peptides are stapled or
unstapled. The CD signal at 215 nm was also monitored
between 10 and 80 °C and shown in Fig. 3 (See also Fig. S5†).
The data show that those peptides with significant structure at
lower temperatures underwent some thermally-induced unfolding
up to 80 °C although a substantial degree of structure still
remained (see, eg, Fig. 3).
Scheme 2 lists the peptides (5–8) synthesized with different
staple locations and the CD spectra compared to the “parent”
peptides 3a and 4a. Again, all peptides exhibited CD spectra
(Fig. 4) consistent with the adoption of a 14-helix in all three
solvents. In general, 3a and 4a (with the staple centrally located)
exhibited the least intense CD signals suggesting that the place-
ment of the staple at either the N- or C-terminus induces a
higher degree of structure. The peptide with a staple at the C-ter-
minus, 8, showed the highest ellipticity in all three solvents
Fig. 2 Summary of NOE correlations observed for the stapled and
unstapled pair 3f and 4f consistent with the formation of a 14-helix.
Grey dots or squares represent NOEs ambiguously assigned due to spec-
tral overlap.
hexa-β3-peptide was measured in 25% 5 mM phosphate buffer
(pH 7.4) in MeCN, 100% TFE and 100% MeOH at a peptide
concentration of 60 μM. Thermal stabilities of each peptide were
also determined by following the change in ellipticity at 215 nm
as the temperature was increased from 10 °C to 80 °C in 25%
5 mM phosphate buffer/MeCN.
Hexa-β3-peptides 3/4a–m exhibited spectra in all three sol-
vents at 20 °C, consistent with a 14-helix, with a minimum
between 211 and 214 nm and a maximum between 195 and
198 nm (selected examples shown in Fig. 3 (others are included
in the Supplementary Information†). Considering the ellipticities
in the CD spectra to be a measure of helicity of peptides in sol-
ution, ellipticities of all peptides 4a–m are generally greater in
25% phosphate buffer/MeCN, than MeOH which in turn were
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 1802–1806 | 1803

View Article Online
was substituted with one and two allylglycine residues to give
peptides 9a/10a and 9b/10b, which reduced the ring size by two
and four atoms to give a 19 and 17 membered ring respectively.
These substitutions were repeated in peptides 3l and 4l in which
polar residues (arginine and lysine) were introduced at positions
1 and 4 (see Scheme 2) to give rise to peptides 9c/10c and 9d/
10d.
The CD analysis showed that all the β3-hexapeptides exhibited
14-helical spectra in all three solvents (25% 5mM phosphate
buffer/MeCN, TFE and MeOH) as shown in Fig. 3. Peptide 10a,
which has a 19-membered ring, shows significantly greater struc-
ture in MeOH and TFE when compared to 4a. Conversely,
peptide 10b shows a significant loss in structure in all solvents,
(see Fig. 4). In contrast, polar peptide 4l (21-membered ring)
shows greater structure by CD than all other peptides in MeOH
and greater structure than the shorter stapled peptides in all sol-
vents tested. This may indicate that a 21-membered ring is ideal
for stabilising the 14-helical structure in β3-peptides. Interest-
ingly though, the unstapled polar peptides seem to have greater
structure than their stapled partners in TFE and phosphate buffer.
CD analysis of polar peptides with shorter staples (9d and 10d)
shows significant loss in structure. This may be due to the steric
constraints exerted by the staple on the backbone, creating a shift
or change in orientation of the atoms involved in intramolecular
H-bonding that holds the 14-helix in place. Therefore, the 21-
membered ring is optimum to avoid perturbation of 14-helical
structure within a hexa-β3-peptide. Again, the CD spectra
revealed some degree of unfolding at 80 °C.
Scheme 2 Peptides synthesized to evaluate the influence of staple ring
size and/or location on conformational stability.
The alkene staple offers obvious opportunities for functionali-
zation and a series of standard transformations was carried out. It
was of interest to establish whether or not β3-peptides could tol-
erate conditions typically employed for processes including dihy-
droxylation and dibromination. Scheme 3 summarises the
transformations successfully carried out on alkene 4a.
Fig. 4 CD spectra (upper) and thermal melts (lower) of selected pep-
tides with varying staple ring size or location in 25% phosphate buffer/
AcCN (pH 7.4).
indicating that it is more structured. These CD results indicate
that while the position of the staple is well tolerated in the
peptide structure without affecting the 14-helical conformation,
the structure can be manipulated by the location of the staple.
Moreover, the thermal unfolding spectra showed some degree of
unfolding at 80 °C (see Fig. 4).
The third series of peptides was synthesised to assess the
ability of the ring size of the staple to stabilise or hinder structure
within the hexapeptide. Peptide 3a/4a (with a 21 membered
ring) was used as the model sequence and either β3-O-allylserine
or β3- allylglycine residues were used for RCM. β3-O-allylserine
Scheme 3 Synthetic scheme for functionalization of the alkene staple.
(i) Pip., DMF; (ii) Ac2O, DMF; (iii) TFA, 95%; (iv) 10% Pd-C, H2,
MeOH, rt, 4h, 94%; (v) Acetone, H2O, t-BuOH, NMO, K2OsO4·2H2O,
14%; (vi) Br2, CH2Cl2, 98%; (vii) NaN3, DMF, 100 °C, 18%.
1804 | Org. Biomol. Chem., 2012, 10, 1802–1806
This journal is © The Royal Society of Chemistry 2012

View Article Online
All transformations were carried out on N-acetyl capped hexa-
β3-peptides. Alkene hydrogenation15 and bromination were best
carried out off resin. Thus hydrogenation of 4a was carried out
using 10% palladium on carbon under a hydrogen atmosphere
resulting in almost quantitative conversion. Bromination with a
solution of molecular bromine in dichloromethane also gave
excellent conversion, this time to the vicinal dibromide. Dihy-
droxylation was achieved on resin and without any attack on the
peptide backbone. Cleavage from the resin then gave diol 13.
Similarly for the preparation of the diazide it was found that the
double displacement of 14 to give 16 proceeded best on resin.
Cleavage then gave diazide 17. At this stage it has not proven
possible to determine if there has been any diastereoselection in
any of these transformations.
The CD analysis shows that all functionalised peptides (12,
13, 15 and 17) adopted 14-helical conformation in both 25%
phosphate buffer/MeCN and MeOH but much less structure in
TFE (see Fig. 5 and the Supplementary Information†). Two of
the functionalized peptides also exhibited higher ellipticity than
the ‘parent’ peptide 4a in MeOH, indicating an increase in struc-
ture with hydrogenation and hydroxylation, which also unfolded
to some extent at 80 °C (see Fig. 5). The two exceptions are 15
and 17, which have two bromine atoms and azide groups
attached respectively. Overall, the CD spectra indicate that
manipulation of the staple is well accommodated by the peptides
as they all adopt a 14-helical conformation.
Conclusion
In this study, we have shown through a number of hexa-β3-pep-
tides that, in most cases, stapled peptides have somewhat greater
CD minima than unstapled peptides in all solvents, suggesting a
greater propensity for the stapled peptide to form 14-helical
structure in a range of solvents. It is also obvious that an 8-atom
bridge seems to be the optimum size for a structurally stable
peptide. We were also able to successfully introduce functional-
ity onto the peptide staple for the first time, with a range of func-
tional groups, none of which affected the structure of the
peptide. Therefore, stapled β3-hexapeptides give rise to a well
defined, structurally stable template that can accommodate the
introduction of functionality. These compounds provide a tem-
plate for further assembly of larger structures.
Experimental
For a description of general experimental procedures, see the
Supplementary Information.†
Functionalization of hexa-β3-peptides
Alkene reduction. To a solution of peptide 4a (30.0 mg,
0.04 mmol) in dry methanol (10 ml) Pd/C (10% carbon, 10 mg)
was added and the resulting black solution was stirred at room
temperature for 4 h with a hydrogen balloon attached. The cata-
lyst was filtered and the solution was concentrated in vacuo to
yield peptide 12 (28.3 mg, 94.1%) as pale brown solid. HRMS
(ESI) m/z: calculated mass for C40H72N6O10: 797.0339, Found:
797.5392. RP-HPLC analysis: single peak at 19.15 min
Dihydroxylation. To a suspension of resin-bound peptide 11
(50 mg, 21.4 mmol) in acetone/water/t-BuOH (17 : 1.5 : 1 v/v),
N-methylmorpholine N-oxide (10 mg, 1.2 mmol) and
K2OsO4·2H2O (3 mg, 1.1 mmol) was added. The reaction was
stirred overnight at room temperature under Ar. The resin was
washed with DMF (3 times), DCM (2 times), DMF (3 times).
The peptide was cleaved using 95% TFA/2.5% H2O/2.5% TIPS
and purified to yield peptide 13 (2.4 mg, 13.5%) as a white
solid. HRMS (ESI) m/z: calculated mass for C40H72N6O12:
829.0327, Found: 829.5281. RP-HPLC analysis: single peak at
28.07 min
Bromination. To
a solution of peptide 4a (60.0 mg,
0.08 mmol) in dichloromethane (5 ml) a solution of Br2 was
added dropwise until a pale orange solution was obtained. The
solution was concentrated in vacuo. CH2Cl2 (2 ml) was added
twice to the resulting pale yellow residue and the solvent
removed to yield product 15 (69.3 mg, 98.0%) as a white solid.
HRMS (ESI) m/z: calculated mass for C40H70N6O10Br2:
954.8260, Found: 955.3565. RP-HPLC analysis: single peak at
40.87 min.
Azide substitution. To a suspension of peptide 14 (4.30 mg,
4.50 μmol) in dry DMF (4 ml) sodium azide (0.88 mg,
1.41 μmol) was added and the resulting solution was stirred and
refluxed overnight at 100 °C. The resin was washed with DMF
(3×) and DCM (2×). The peptide was cleaved using 95% TFA/
2.5% H2O/2.5% TIPS and purified to yield peptide 17 (2.4 mg,
13.5%) as a white solid. HRMS (ESI) m/z: calculated mass for
Fig. 5 CD spectra (upper) and thermal melts (lower) of selected pep-
tides with functionalized staples in 25% phosphate buffer/AcCN
(pH 7.4).
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 1802–1806 | 1805

View Article Online
6 M. Rueping, Y. R. Mahajan, B. Jaun and D. Seebach, Chem.–Eur. J.,
2004, 10, 1607.
7 E. Vaz and L. Brunsveld, Org. Lett., 2006, 8, 4199.
8 E. Vaz, W. C. Pomerantz, M. Geyer, S. H. Gellman and L. Brunsveld,
ChemBioChem, 2008, 9, 2254.
C40H72N12O10: 879.0752, Found: 440.5210 RP-HPLC analysis:
single peak at 14.67 min
9 Y. E. Bergman, M. P. Del Borgo, R. D. Gopalan, S. Jalal, S. E. Unabia,
M. Ciampini, D. J. Clayton, J. M. Fletcher, R. J. Mulder, J. A. Wilce,
M. I. Aguilar and P. Perlmutter, Org. Lett., 2009, 11, 4438.
10 M. Rueping, B. Jaun and D. Seebach, Chem. Commun., 2000, 2267.
11 D. Seebach, P. I. Arvidsson and M. Rueping, Chem. Commun., 2001,
649.
12 W. F. DeGrado and R. P. Cheng, J. Am. Chem. Soc., 2001, 123, 5162.
13 A. D. Bautista, J. S. Appelbaum, C. J. Craig, J. Michel and A. Schepartz,
J. Am. Chem. Soc., 2010, 132, 2904.
Notes and references
1 (a) R. P. Cheng, S. H. Gellman and W. F. DeGrado, Chem. Rev., 2001,
101, 3219; (b) D. Steer, R. Lew, P. Perlmutter, A. I. Smith and M.
I. Aguilar, Curr. Med. Chem., 2002, 811–822.
2 P. I. Arvidsson, J. Frackenpohl, N. S. Ryder, B. Liechty, F. Petersen,
H. Zimmermann, G. P. Camenisch, R. Woessner and D. Seebach, Chem-
BioChem, 2001, 2, 771.
3 R. P. Cheng and W. F. DeGrado, J. Am. Chem. Soc., 2001, 123, 5162.
4 D. A. Guarracino, H. R. Chiang, T. N. Banks, J. D. Lear, M. E. Hodsdon
and A. Schepartz, Org. Lett., 2006, 8, 807.
14 M.-O. Ebert, J. Gardiner, S. Ballet, A. D. Abell and D. Seebach,
Helv. Chim. Acta, 2009, 92, 2643.
5 S. A. Hart, A. B. Bahadoor, E. E. Matthews, X. J. Qiu and A. Schepartz,
J. Am. Chem. Soc., 2003, 125, 4022.
15 C. E. Schafmeister, J. Po and G. L. Verdine, J. Am. Chem. Soc., 2000,
122, 5891.
1806 | Org. Biomol. Chem., 2012, 10, 1802–1806
This journal is © The Royal Society of Chemistry 2012