Large structural modification with conserved conformation: Analysis of Δ3-fused aryl prolines in model β-turns

Article abstract of DOI:10.1021/ja0471222

For the first time, the influence of a fused Δ³-arylproline on peptide conformation has been studied by the synthesis and comparison of the conformations of peptides containing proline and pyrrolo-proline, 3 (PyPro). Pyrrolo-proline was demonst

Full text of DOI:10.1021/ja0471222

Published on Web 10/19/2004
Large Structural Modification with Conserved Conformation: Analysis of
∆3-Fused Aryl Prolines in Model â-Turns
Guillaume Jeannotte and William D. Lubell*
De´partement de chimie, UniVersite´ de Montre´al, C.P. 6128, Succursale CentreVille, Montre´al,
Que´bec, Canada H3C 3J7
Received May 17, 2004; E-mail: lubell@chimie.umontreal.ca
Proline serves as a turn inducer in natural peptides and proteins
because of the pyrrolidine ring, which restricts motion about the φ
dihedral angle and reduces the energy difference between the prolyl
amide cis- and trans-isomers.1a Steric and stereoelectronic effects
of alkyl- and heteroatom-substituted prolines have thus been used
Figure 1. Representative unsaturated, aryl, and heteroaryl prolines.
Table 1. cis-Isomer N-terminal to Pro and PyPro (concn = 10 mM)
to induce specific conformations for studying factors that effect
prolyl peptide activity and biology.1 In contrast, ∆3-dehydroproline
1 acts as a conservative proline replacement in peptides and proteins.
For example, little conformational change was observed after the
replacement of Pro by 1 in a 14-residue cyclic analogue of
gramicidin S that exhibited a similarly stable â-sheet structure with
two type II′ â-turns, as shown by spectroscopic and computational
analysis.2 Moreover, replacement of Pro by 1 has provided peptide
and protein analogues with similar and improved biological
activities.3
% cis (±1%)
peptide
2:1 CD2Cl2/DMSO
DMSO
4a: Ac-D-Val-Pro-Gly-D-Leu-NMe2
4b: Ac-D-Val-PyPro-Gly-D-Leu-NMe2
5a: Ac-Leu-D-Phe-Pro-Val-NMe2
5b: Ac-Leu-D-Phe-PyPro-Val-NMe2
12
15
26
44
19
18
43
47
∆3-Fused arylprolines, such as 2, have yet to be examined in
peptides primarily because of the difficulties in their synthesis.4
Like 1, ∆3-fused arylprolines may serve as conservative proline
surrogates; moreover, structural modifications may be added without
influencing conformation. Recently, we introduced an effective
methodology for synthesizing enantiopure ∆3-fused pyrrol-prolines
(PyPro) of general structure 3 (Figure 1: R1, R2 ) H, alkyl; P )
protection).4 To examine the influence of the aryl moiety and
flattened pyrrolidine ring on conformation, Fmoc-PyPro 3 (R1 )
CH3, R2 ) H, P ) Fmoc) has now been introduced into peptides
4b and 5b for comparison with Pro in model â-turns, 4a and 5a,
respectively (Table 1). Peptides 4a and 4b were selected to examine
if PyPro would accommodate itself at the i + 1 position of a
â-hairpin because studies of the enantiomeric sequence of 4a have
shown that the D-Pro-Gly residue adopted the central position of a
â-hairpin.5 Peptides 5a and 5b, analogues of the â-turn portion of
gramicidin S, were synthesized to examine PyPro at the i + 2
position because spectroscopic analysis revealed a significant â-turn
population in a peptide related to 5a.6 Sequences 4 and 5, possessing
Pro and PyPro, were synthesized in solution and examined by NMR
spectroscopy to assess the influence of the pyrrole moiety on the
hydrogen-bonding network, the prolyl amide equilibrium, and the
turn conformation.
Peptides possessing Pro (4a and 5a) and PyPro (4b and 5b) were
synthesized in solution using NR-Boc and NR-Fmoc strategies,
respectively, as described in the Supporting Information (SI). The
NMR spectra of peptides 4 and 5 were measured in CD2Cl2 at
concentrations e10 mM because aggregation was observed at
higher concentrations.7a Intramolecular hydrogen bonding was
evaluated by plotting the change in the chemical shift of the amide
and the pyrrole protons as a function of DMSO-d6 added to the
peptide in CD2Cl2 (Figure 2) because exchangeable protons engaged
in intramolecular hydrogen bonds are typically not influenced by
strong hydrogen-bonding solvents, such as DMSO-d6 relative to
exposed protons.7b,8 The prolyl amide isomer equilibrium was
measured in CD2Cl2, 2:1 CD2Cl2/DMSO, and DMSO by integration
of the signals for the isomeric amide and R-protons for 4a and 5a
and the pyrrole protons for 4b and 5b. Finally, sequential and long
distance NOEs were measured in the NOESY spectrum of each
peptide to assess the presence of a turn conformation.
The influence of DMSO on the amide proton chemical shifts
was similar in peptides 4 and 5 (Figure 2). In 4, the amide protons
for Val and Leu were downfield from that for Gly in CD2Cl2 and
were unaffected by the addition of DMSO relative to that for Gly,
which was downfield shifted (∆δ ) 1.29 ppm), indicative of its
exposure to solvent. In peptide 5, the Leu amide proton was
unaffected by the solvent change, indicative of a solvent-shielded
NH. The Phe and Val amide protons were downfield shifted with
the addition of DMSO, indicative of their exposure to the effects
of the solvent. In peptides 4b and 5b, the pyrrole-NH was
influenced most by the solvent change (∆δ ) 1.95 and 2.1 ppm,
respectively). The most significant difference between the Pro and
PyPro peptides was the magnitude of the chemical shift variation
of the Phe amide proton, which was larger for 5a (∆δ ) 1.1 ppm)
than for 5b (∆δ ) 0.6 ppm), potentially due to a different
orientation of the phenyl ring over the peptide backbone.
The population of the amide cis-isomer, N-terminal to Pro and
PyPro (Table 1), was very similar in peptides 4 and 5 and
undetectable in CD2Cl2. In comparison with 20% cis-isomer in
DMSO-d6 observed in a related â-hairpin octapeptide containing a
central D-Pro-Gly residue,8 peptides 4a and 4b exhibited similar
amounts of cis-isomer (19 and 18%, respectively). Alternatively,
relative to 39% cis-isomer observed in Ac-D-Phe-Pro-NH(Me) in
DMSO,1b peptides 5a and 5b exhibited slightly higher cis-isomer
populations (43 and 47%, respectively). When favored â-turn
conformations are assumed, the lower amount of cis-isomer in 4 is
consistent with the stabilization of the trans-isomer by an intramo-
lecular hydrogen bond.9 The significantly higher cis-isomer in 5b
may be due to a staggered π-π interaction between the phenyl
and pyrrole ring in this conformer.1a,10 The greatest difference in
9
14334
J. AM. CHEM. SOC. 2004, 126, 14334-14335
10.1021/ja0471222 CCC: $27.50 © 2004 American Chemical Society

C O M M U N I C A T I O N S
Figure 2. Amide proton NMR chemical shift (δ) as a function of the percentage of DMSO in CD2Cl2 for peptides 4 and 5 (concn = 10 mM).
importance of turns in recognition events,11 future efforts may
employ PyPro analogues to create interactions for enhancing
biological activity without perturbing conformation.
Acknowledgment. This research was supported, in part, by
Merck Frosst Canada, NSERC Canada, and FQRNT Que´bec. We
thank Ms. Sylvie Bilodeau of the Regional High-Field NMR
Laboratory for her assistance, and Mr. Dalbir Sekhon for performing
LC-MS experiments.
Supporting Information Available: Details on the syntheses,
1
concentrations, and conformational analyses, H and 13C NMR, IR,
DEPT, COSY, and NOESY spectra for 4a,b and 5a,b. This material is
available free of charge via the Internet at http://pubs.acs.org.
Figure 3. Selected 1H-1H NOESY cross-peaks in CD2Cl2 as well as
potential hydrogen bonds from DMSO titration curves for peptides 4b
and 5b.
References
(1) Halab, L.; Lubell, W. D. J. Am. Chem. Soc. 2002, 124, 2474-2484 and
references therein. (b) Halab, L.; Lubell, W. D. J. Pept. Sci. 2001, 7,
92-104. (c) Beausoleil, E.; Sharma, R.; Michnick, S. W.; Lubell, W. D.
J. Org. Chem. 1998, 63, 6572-6578 and references therein. (d) Cavelier,
F.; Vivet, B.; Martinez, J.; Aubry, A.; Didierjean, C.; Vicherat, A.;
Marraud, M. J. Am. Chem. Soc. 2002, 124, 2917-2923. (e) DeRider, M.
L.; Wilkens, S. J.; Wadell, M. J.; Bretscher, L. E.; Weinhold, F.; Raines,
R. T.; Markley, J. L. J. Am. Chem. Soc. 2002, 124, 2497-2505. (f)
Montelione, G. T.; Hughes, P.; Clardy, J.; Scheraga, H. A. J. Am. Chem.
Soc. 1986, 108, 6765-6773.
the cis-isomer population of Pro relative to PyPro was detected in
1:2 DMSO/CD2Cl2, in which 5a exhibited 26% and 5b 44% cis-
isomer, which may be due to a combination of aromatic stacking,
as described above, and enhanced steric effects from DMSO coor-
dination to the pyrrole nitrogen which disfavored the trans-isomer.
In 4 and 5 the NOESY spectra for the Pro and PyPro peptides
exhibited similar correlations for â-turns in CD2Cl2 (Figure 3). The
amide trans-isomer, N-terminal to the Pro and PyPro residues, was
identified by NOEs between their δ-protons and the R-proton of
the N-terminal residue. A long-range NOE between the Gly and
Leu amide protons was indicative of a turn conformer in 4.
Similarly, long-range NOEs between the δ-proton of Pro/PyPro and
the Val amide hydrogen were indicative of a turn conformer in 5.
The presence of a turn geometry in 4 and 5 was supported by long-
range NOEs between the acetyl and dimethyl amide singlets.
Finally, sequential NOEs were observed between the neighboring
Leu-CRH and Phe-NH as well as the Pro/PyPro-CRH and Val-NH
in 4 and between the Pro/PyPro-CRH and Gly-NH in 5, indicative
of their proximity in the major conformer.
(2) Gibbs, A. C. G.; Bjorndahl, T. C.; Hodges, R. S.; Wishart, D. S. J. Am.
Chem. Soc. 2002, 124, 1203-1213.
(3) Felix, A. M.; Wang, C.-T.; Liebman, A. A.; Delaney, C. M.; Mowles, T.;
Burghardt, B. A.; Charnecki, A. M.; Meienhofer, J. Int. J. Pept. Protein
Res. 1977, 10, 299-310 and references therein. (b) Zhao, Z.; Liu, X.;
Shi, Z.; Danley, L.; Buang, B.; Jiang, R.-T.; Tsai, M.-D. J. Am. Chem.
Soc. 1996, 118, 3535-3536.
(4) Jeannotte, G.; Lubell, W. D. J. Org. Chem. 2004, 69, 4656-4662 and
references cited therein. See, also, refs 8, 11-13, and references cited
therein.
(5) Haque, T. S.; Little, J. C.; Gellman, S. H. J. Am. Chem. Soc. 1996, 118,
6975-6985.
(6) Sato, K.; Higashijima, T.; Sugawara, R.; Nagai, U. Bull. Chem. Soc. Jpn.
1983, 56, 3699-3702.
(7) (a) See details in the Supporting Information. (b) Hydrogen bonding was
also examined in the NH stretch region of the IR spectra of 4 and 5; see
SI.
(8) Awasthi, S. K.; Raghothama, S.; Balaram, P. Biochem. Biophys. Res.
Commun. 1995, 216, 275-381.
For the first time, the influence of a fused ∆3-arylproline on
peptide conformation has been studied by the synthesis of model
peptides, 4b and 5b, containing PyPro 3. The DMSO titration
curves, the cis/trans-isomer ratios, and the NOESY spectra all
demonstrate an overall conservation of â-turn conformation on
replacement of Pro by PyPro in peptides 4 and 5. In light of the
(9) Ac-D-Ala-Pro-NH(Me) exhibits 36% cis-isomer in DMSO; see ref 1b.
(10) Yao, J.; Bruschweiler, R.; Dyson, H. J.; Wright, P. E. J. Am. Chem. Soc.
1994, 116, 12051-12052. (b) Wu, W.-J.; Raleigh, D. P. Biopolymers 1998,
45, 381-394.
(11) Rose, G. D.; Gierash, L. M.; Smith, J. A. AdV. Prot. Chem. 1985, 27,
1-109.
JA0471222
9
J. AM. CHEM. SOC. VOL. 126, NO. 44, 2004 14335