Stereochemical control of hairpin formation in β-peptides containing dinipecotic acid reverse turn segments

Article abstract of DOI:10.1021/ja993416p

We examine the relationship between covalent structure and conformational propensity among a series of β-amino acid tetramers. These experiments focus on the hairpin folding motif. Among conventional peptides, the minimum increment of β-sheet secondary structure is a 'β-hairpin,' in which two strands are connected via a short loop. The present studies are aimed at optimizing hairpin stability among β-peptides. Previous work from our laboratory has identified optimal substitution patterns for residues that form strands in an antiparallel β-peptide sheet (Krauthauser et al. J. Am. Chem. Soc. 1997, 119, 11719), and we have shown that a dinipecotic acid segment can promote sheet-type interactions between attached strand residues (Chung et al. J. Am. Chem. Soc. 1998, 120, 10555). Here we compare all four possible configurations of the dinipecotic acid segment, (R,S), (S,R), (R,R) and (S,S), for the ability to induce sheet formation with a constant set of enantiomerically pure strand residues. We show that both heterochiral dinipecotic acid segments promote hairpin formation, although one is distinctly superior. Neither of the homochiral dinipecotic acid supports hairpin folding. When the strand residues are β-alanine (achiral), the heterochiral dinipecotic acid segment is again superior to the homochiral segment, but we find a difference between hairpin conformations in solution and in the solid state.

Full text of DOI:10.1021/ja993416p

J. Am. Chem. Soc. 2000, 122, 3995-4004
3995
Stereochemical Control of Hairpin Formation in â-Peptides
Containing Dinipecotic Acid Reverse Turn Segments
Yong Jun Chung, Bayard R. Huck, Laurie A. Christianson, Heather E. Stanger,
Susanne Krautha1user, Douglas R. Powell, and Samuel H. Gellman*
Contribution from the Department of Chemistry, UniVersity of Wisconsin, Madison, Wisconsin 53706
ReceiVed September 20, 1999. ReVised Manuscript ReceiVed February 23, 2000
Abstract: We examine the relationship between covalent structure and conformational propensity among a
series of â-amino acid tetramers. These experiments focus on the hairpin folding motif. Among conventional
peptides, the minimum increment of â-sheet secondary structure is a “â-hairpin,” in which two strands are
connected via a short loop. The present studies are aimed at optimizing hairpin stability among â-peptides.
Previous work from our laboratory has identified optimal substitution patterns for residues that form strands
in an antiparallel â-peptide sheet (Krautha¨user et al. J. Am. Chem. Soc. 1997, 119, 11719), and we have
shown that a dinipecotic acid segment can promote sheet-type interactions between attached strand residues
(Chung et al. J. Am. Chem. Soc. 1998, 120, 10555). Here we compare all four possible configurations of the
dinipecotic acid segment, (R,S), (S,R), (R,R) and (S,S), for the ability to induce sheet formation with a constant
set of enantiomerically pure strand residues. We show that both heterochiral dinipecotic acid segments promote
hairpin formation, although one is distinctly superior. Neither of the homochiral dinipecotic acid supports
hairpin folding. When the strand residues are â-alanine (achiral), the heterochiral dinipecotic acid segment is
again superior to the homochiral segment, but we find a difference between hairpin conformations in solution
and in the solid state.
Introduction
represent important progress toward a long-range goal of
creating unnatural tertiary structures.
Unnatural oligomers with well-defined folding behavior
(“foldamers”) are subjects of increasing interest.1 Biological
systems rely almost exclusively on well-folded polymers
(proteins and RNA) for complex molecular functions such as
catalysis, and it might be possible to engineer comparable
functions into unnatural foldamers. Numerous recent reports
have described oligomers that adopt specific secondary
structures,2-21 especially helices; collectively these efforts
(5) Computer simulations of â-peptides: (a) Daura, X.; van Gunsteren,
W. F.; Rigo, D.; Jaun, B.; Seebach, D. Chem. Eur. J. 1997, 3, 1410. (b)
Christianson, L. A., Ph.D. Thesis, University of Wisconsin-Madison, 1997.
(c) Daura, X.; Jaun, B.; Seebach, D.; van Gunsteren, W. F.; Mark, A. J.
Mol. Biol. 1998, 280, 925. (d) Daura, X.; Gademann, K.; Jaun, B.; Seebach,
D.; van Gunsteren, W. F.; Mark, A. E. Angew. Chem., Int. Ed. 1999, 38,
236. (e) Daura, X.; van Gunsteren, W. F.; Mark, A. E. Proteins: Struct.,
Funct., Genet. 1999, 34, 269. (f) Wu, Y. D.; Wang, D. P. J. Am. Chem.
Soc. 1998, 120, 13485. (g) Mo¨hle, K.; Gu¨nther, R.; Thormann, N.; Hofmann,
H.-J. Biopolymers 1999, 50, 167.
(6) Hairpin-forming â-peptides: (a) Krautha¨user, S.; Christianson, L. A.;
Powell, D. R.; Gellman, S. H. J. Am. Chem. Soc. 1997, 119, 11719. (b)
Chung, Y. J.; Christianson, L. A.; Stanger, H. E.; Powell, D. R.; Gellman,
S. H. J. Am. Chem. Soc. 1998, 120, 10555. (c) Seebach, D.; Abele, S.;
Gademann, K.; Jaun, B. Angew. Chem., Int. Ed. 1999, 38, 1595.
(7) â-Peptide nucleic acid analogues: Diederichsen, U.; Schmitt, H. W.
Eur. J. Org. Chem. 1998, 827.
(8) Vinylogous peptide folding: Hagihara, M.; Anthony, N. J.; Stout,
T. J.; Clardy, J.; Schreiber, S. L. J. Am. Chem. Soc. 1992, 114, 6568.
(9) Nucleic acid analogue folding: Beier, M.; Reck, F.; Wagner, T.;
Krishnamurthy, R.; Eschenmoser, A. Science 1999, 283, 699 and references
therein.
(10) Aedamer folding: (a) Lokey, R. S.; Iverson, B. L. Nature 1995,
375, 303. (b) Nguyen, J. Q.; Iverson, B. L. J. Am. Chem. Soc. 1999, 121,
2639.
(11) Sulfonamide oligomer folding: Gennari, C.; Salom, B.; Potenza,
D.; Longari, C.; Fioravanzo, E.; Carugo, O.; Sardone, N. Chem. Eur. J.
1996, 2, 644 and references therein.
(12) γ-Peptide folding: (a) Hintermann, T.; Gademann, K.; Jaun, B.;
Seebach, D. HelV. Chim. Acta 1998, 81, 983. (b) Hanessian, S.; Luo, X.;
Schaum, R.; Michnick, S. J. Am. Chem. Soc. 1998, 120, 8569. (c) Hanessian,
S.; Luo, X.; Schaum, R. Tetrahedron Lett. 1999, 40, 4925.
(13) δ-Peptide folding: (a) Szabo, L.; Smith, B. L.; McReynolds, K.
D.; Parrill, A. L.; Morris, E. R.; Gervay, J. J. Org. Chem. 1998, 63, 1074.
(b) Smith, M. D.; Claridge, T. D. W.; Tranter, G. E.; Sansom, M. S. P.;
Fleet, G. W. J. J. Chem. Soc., Chem. Commun. 1998, 2041. (c) Long, D.
D.; Hungerford, N. L.; Smith, M. D.; Brittain, D. E. A.; Marquess, D. G.;
Claridge, T. D. W.; Fleet, G. W. J. Tetrahedron Lett. 1999, 40, 2195. (d)
Claridge, T. D. W.; Long, D. D.; Hungerford, N. L.; Aplin, R. T.; Smith,
M. D.; Marquess, D. G.; Fleet, G. W. J. Tetrahedron Lett. 1999, 40, 2199.
(1) Recent reviews: (a) Gellman, S. H. Acc. Chem. Res. 1998, 31, 173.
(b) Kirshenbaum, K.; Zuckermann, R. N.; Dill, K. A. Curr. Opin. Struct.
Biol. 1999, 9, 530. (c) Stigers, K. D.; Soth, M. J.; Nowick, J. S. Curr.
Opin. Chem. Biol. 1999, 3, 714. (d) Barron, A. E.; Zuckermann, R. N. Curr.
Opin. Chem. Biol. 1999, 3, 681.
(2) Helical â-peptide oligomers: (a) Seebach, D.; Overhand, M.; Ku¨hnle,
F. N. M.; Martinoni, B.; Oberer, L.; Hommel, U.; Widmer, H. HelV. Chim.
Acta 1996, 79, 913. (b) Seebach, D.; Ciceri, P.; Overhand, M.; Juan, B.;
Rigo, D.; Oberer, L.; Hommel, U.; Amstutz, R.; Widmer, H. HelV. Chim.
Acta 1996, 79, 2043. (c) Seebach, D.; Matthews, J. L. J. Chem. Soc., Chem.
Commun. 1997, 2015-2022. (d) Seebach, D.; Abele, S.; Gademann, K.;
Guichard, G.; Hintermann, T.; Jaun, B.; Matthews, J. L.; Schreiber, J.;
Oberer, L.; Hommel, U.; Widmer, H. HelV. Chim. Acta 1998, 81, 932. (e)
Seebach, D.; Abele, S.; Sifferlen, T.; Ha¨nggi, M.; Gruner, S.; Seiler, P.
HelV. Chim. Acta 1998, 81, 2218.
(3) Helical â-peptide oligomers: (a) Appella, D. H.; Christianson, L.
A.; Karle, I. L.; Powell, D. R.; Gellman, S. H. J. Am. Chem. Soc. 1996,
118, 13071. (b) Appella, D. H.; Christianson, L. A.; Klein, D. A.; Powell,
D. R.; Huang, X.; Barchi, J. J.; Gellman, S. H. Nature 1997, 387, 381. (c)
Applequist, J.; Bode, K. A.; Appella, D. H.; Christianson, L. A.; Gellman,
S. H. J. Am. Chem. Soc. 1998, 120, 4891. (d) Appella, D. H.; Barchi, J. J.;
Durell, S.; Gellman, S. H. J. Am. Chem. Soc. 1999, 121, 2309. (e) Appella,
D. H.; Christianson, L. A.; Karle, I. L.; Powell, D. R.; Gellman, S. H. J.
Am. Chem. Soc. 1999, 121, 6206. (f) Appella, D. H.; Christianson, L. A.;
Klein, D. A.; Richards, M. R.; Powell, D. R.; Gellman, S. H. J. Am. Chem.
Soc. 1999, 121, 7574. (g) Wang, X.; Espinosa, J. F.; Gellman, S. H., J.
Am. Chem. Soc., in press.
(4) Gung, B. W.; Zou, D.; Stalcup, A. M.; Cottrell, C. E. J. Org. Chem.
1999, 64, 2176.
10.1021/ja993416p CCC: $19.00 © 2000 American Chemical Society
Published on Web 04/12/2000

3996 J. Am. Chem. Soc., Vol. 122, No. 17, 2000
Chung et al.
Oligomers of â-amino acids (“â-peptides”) are among the
most thoroughly studied unnatural foldamers at present.2-7 Early
work on â-amino acid homopolymers suggested that both sheet22
and helix23 secondary structures are accessible. Helix formation
by short â-peptide oligomers in organic solvents has been
demonstrated via NMR by Seebach et al.2 and by us3; we have
been able to correlate solution data with crystal structures.3
Recently, we have shown that â-peptide hexamers can adopt
stable helical conformations in aqueous solution if the residues
are properly selected.24
tions. Hairpin-forming peptides have proven to be a good source
of catalysts for stereoselective chemical transformations.28
We have identified an optimal â-amino acid substitution
pattern for the strand residues in an antiparallel sheet through
the use of hairpin-forming molecules that contain non-â-peptide
loop segments.6a R,â-Disubstituted residues, with the appropriate
configurations at the R- and â-carbons, are expected to favor
an anti torsion angle about the NCâ-CRC(dO) bond, and we
showed that such residues are predisposed to form a â-peptide
sheet.6a We have subsequently focused on replacing the non-
â-peptide loop in our initial hairpins6a with a â-peptide loop.6b
In a preliminary report, we described a loop segment containing
two nipecotic acid residues that adopts a reverse turn conforma-
tion and promotes hairpin formation.6b (This â-peptide reverse
turn is reminiscent of the common â-turn in conventional
peptides and proteins.29) The resulting tetra-â-peptide represents
the first autonomously folding â-peptide sheet. This accomplish-
ment completed the demonstration that each of the three types
of regular secondary structure observed in proteins, helix, sheet,
and reverse turn can also be formed by â-peptides.
It is more difficult to generate small increments of sheet than
small increments of helix because helices form along a co-
valently continuous segment, while sheets form from strand
segments that can be widely separated along the polymer
backbone.25 To create a unimolecular sheet, one must identify
residues that have a high propensity to form strands and residues
that have a high propensity to form a reverse turn or loop that
will induce sheet interactions between attached strand segments.
Among conventional peptides and proteins, the strand-loop-
strand motif is designated a “â-hairpin”,26 and we retain the
term “hairpin” in our discussion of â-peptides. A hairpin
represents the smallest increment of autonomously folding sheet
secondary structure. Considerable effort has been devoted to
the development of conventional peptides26 (R-amino acid
residues) and peptidomimetics27 that display hairpin conforma-
Here, we present a systematic evaluation of dinipecotic acid
turn segment configuration with regard to hairpin formation.
We have examined two types of strand residue, enantiomerically
pure R,â-disubstituted residues, and â-alanine, because prior
results indicate that each strand type favors a distinct form of
â-peptide sheet.6a We also use the current body of literature on
â-peptide hairpins,6 which includes a â-peptide hairpin very
recently reported by Seebach et al.,6c to formulate general rules
for design and analysis of autonomously folding â-peptide
sheets.
(14) Oligourea folding: Nowick, J. S.; Mahrus, S.; Smith, E. M.; Ziller,
J. W. J. Am. Chem. Soc. 1996, 118, 1066.
(15) Anthranilic acid oligomer folding: Hamuro, Y.; Geib, S. J.;
Hamilton, A. D. J. Am. Chem. Soc. 1997, 119, 10587 and references therein.
(16) Peptoid folding: (a) Armand, P.; Kirshenbaum, K.; Goldsmith, R.
A.; Farr-Jones, S.; Barron, A. E.; Truong, K. T.; Dill, K. A.; Mierke, D. F.;
Cohen, F. E.; Zuckermann, R. N.; Bradley, E. K. Proc. Natl. Acad. Sci.
U.S.A. 1998, 95, 4309. (b) Kirshenbaum, K.; Barron, A. E.; Goldsmith, R.
A.; Armand, P.; Bradley, E. K.; Truong, K. T.; Dill, K. A.; Cohen, F. E.;
Zuckermann, R. N. Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 4303.
(17) m-Phenylacetylene oligomer folding: (a) Nelson, J. C.; Saven, J.
G.; Moore, J. S.; Wolynes, P. G. Science 1997, 277, 1793. (b) Prince, R.
B.; Okada, T.; Moore, J. S. Angew. Chem., Int. Ed. 1999, 38, 233. (c) Gin,
M. S.; Yokozawa, T.; Prince, R. B.; Moore, J. S. J. Am. Chem. Soc. 1999,
121, 2643.
Results
Synthesis. The tetra-â-peptides discussed below were pre-
pared in solution via standard coupling reactions, which involved
1-[3-(dimethylamino)propyl]-3-ethylcarbodiimide hydrochloride
(EDC) and 1-hydroxybenzotriazole (HOBt) in dimethylform-
amide (DMF). The Boc group was used for amino protection.
We prepared the optically active R-methyl-â-ethyl strand residue
by following the route of Jefford and McNulty;30 straighforward
extrapolation from this route provided the R-benzyl-â-(thio-
phenyl)methyl derivative.6a (Complete experimental details for
these two R,â-disubstituted residues are provided in the Sup-
porting Information for ref 6a.) Racemic nipecotic acid is
commerically available. Resolution of enantiomers can be
achieved by cocrystallization of racemic ethyl nipecotate (Nip-
OC2H5) with enantiomerically pure tartaric acid, and repeated
recrystallization of the resulting salt.31,32 We devised an alterna-
tive resolution in which racemic nipecotic acid itself is cocrys-
(18) R-Aminooxy acid oligomer folding: Yang, D.; Qu, J.; Li, B.; Ng,
F.; Wang, X.; Cheung, K.; Wang, D.; Wu, Y. J. Am. Chem. Soc. 1999,
121, 589 and references therein.
(19) Guanidine oligomer folding: Tanatani, A.; Yamaguchi, K.; Azumaya,
I.; Fukutomi, R.; Shudo, K.; Kagechika, H. J. Am. Chem. Soc. 1998, 120,
6433 and references therein.
(20) Oligopyrrolinone conformations: Smith, A. B.; Guzman, M. C.;
Sprengler, P. A.; Keenan, T. P.; Holcomb, R. C.; Wood, J. L.; Carroll, P.
J.; Hirschmann, R. J. Am. Chem. Soc. 1994, 116, 9947.
(21) Pyridine/pyrimidine oligomer folding: (a) Bassani, D. M.; Lehn,
J.-M.; Baum, G.; Fenske, D. Angew. Chem., Int. Ed. Engl. 1997, 36, 1845.
(b) Bassani, D. M.; Lehn, J.-M. Bull. Soc. Chim. Fr. 1997, 134, 897.
(22) (a) Bestian, H. Angew. Chem., Int. Ed. Engl. 1968, 7, 278. (b)
Glickson, J. D.; Applequist, J. J. Am. Chem. Soc. 1971, 93, 3276. (c) Narita,
M.; Doi, M.; Kudo, K.; Terauchi, Y. Bull. Chem. Soc. Jpn. 1986, 59, 3553.
(d) Kovacs, J.; Ballina, R.; Rodin, R. L.; Balasubramanian, D.; Applequist,
J. J. Am. Chem. Soc. 1965, 87, 119. (e) Chen, F.; Lepore, G.; Goodman,
M. Macromolecules 1974, 7, 779. (f) Yuki, H.; Okamoto, Y.; Taketani,
Y.; Tsubota, T.; Marubayashi, Y. J. Polym. Sci. Polym. Chem. Ed. 1978,
16, 2237.
(23) (a) Ferna´ndez-Santin, J. M.; Aymam´ı, J.; Rodr´ıguez-Gala´n, A.;
Mun˜oz-Guerra, S.; Subirana, J. A. Nature 1984, 311, 53. (b) Ferna´ndez-
Santin, J. M.; Mun˜oz-Guerra, S.; Rodr´ıguez-Gala´n, A.; Aymam´ı, J.;
Lloveras, J.; Subirana, J. A.; Giralt, E.; Ptak, M. Macromolecules 1987,
20, 62. (c) Bella, J.; Alema´n, C.; Ferna´ndez-Santin, J. M.; Alegre, C.;
Subirana, J. A. Macromolecules 1992, 25, 5225. (d) Lo´pez-Carrasquero,
F.; Alema´n, C.; Mun˜oz-Guerra, S. Biopolymers 1995, 36, 263. (e) Bode,
K. A.; Applequist, J. Macromolecules 1997, 30, 2144.
(27) (a) Nowick, J. S. Acc. Chem. Res. 1999, 32, 287. (b) Hanessian, S.;
McNaughton-Smith, G.; Lombart, H.-G.; Lubell, W. D. Tetrahedron 1997,
53, 12789. (c) Belvisi, L.; Gennari, C.; Mielgo, A.; Potenza, D.; Scolastico,
C. Eur. J. Org. Chem. 1999, 389. (d) Halab, L.; Lubell, W. D. J. Org.
Chem. 1999, 64, 3312.
(28) (a) Gilbertson, S. R.; Pawlick, R. V. Angew. Chem., Int. Ed. Engl.
1996, 35, 902. (b) Jarvo, E. R.; Copeland, G. T.; Papaioannou, N.;
Bonitatebus, P. J.; Miller, S. J. J. Am. Chem. Soc. 1999, 121, 11638.
(29) Rose, G. D.; Gierasch, L. M.; Smith, J. A. AdV. Protein Chem. 1985,
37, 1.
(24) High-resolution structural characterization of â-peptide folding in
water is described in refs 3d and 3 g. For other studies in aqueous solution,
see refs 2e and 4.
(25) Nesloney, C. L.; Kelly, J. W. Bioorg. Med. Chem. 1996, 4, 739.
(26) (a) Gellman, S. H. Curr. Opin. Chem. Biol. 1998, 2, 717. (b) Lacroix,
E.; Kortemme, T.; de la Paz, M. L.; Serrano, L. Curr. Opin. Struct. Biol.
1999, 9, 487.
(30) Jefford, C. W.; McNulty, J. HelV. Chim. Acta 1994, 77, 2142.
(31) Akkerman, A. M.; de Jongh, D. K.; Veldstra, H. Recl. TraV. Chim.
Pay-Bas 1951, 70, 899.
(32) Magnus, P.; Thurston, L. S. J. Org. Chem. 1991, 56, 1166.

Optimization of â-Peptide Hairpin Stability
J. Am. Chem. Soc., Vol. 122, No. 17, 2000 3997
tallized with enantiomerically pure camphorsulfonic acid.
Enantiomeric purity was determined via HPLC (Chiracel OD
column) after the nipecotic acid had been Boc-protected on the
nitrogen and converted to the 2-naphthylmethyl ester. HPLC
analysis indicated that both resolution methods provide each
enantiomer of nipecotic acid in g98% e.e.
Conformational Analysis of Tetra-â-peptides Containing
r,â-Disubstituted Strand Residues. Diastereomeric â-peptides
1-4 were examined to elucidate the effect of loop segment
configuration on hairpin conformational stability. In this series,
all four configurations of the dinipecotic acid reverse turn
segment [(R,S), (S,R), (R,R), and (S,S)] are evaluated while the
strand residues are kept constant. The comparison was carried
out in a relatively nonpolar solvent, methylene chloride. Under
these conditions, intramolecular hydrogen bonding provides a
modest but not overwhelming driving force for adoption of
compact conformations, as we have previously shown in our
study of â-hairpin formation by conventional tetrapeptides.33
Thus, comparing the folding of 1-4 in methylene chloride
allows us to determine whether the turn-forming propensity of
each dinipecotic acid segment is compatible with sheet-type
interactions between the attached R,â-disubstituted residues.
here by three other measurements: amide proton NMR chemical
shifts, N-H stretch region IR spectra, and tertiary amide rotamer
ratios detected by NMR. These three measurements provide a
basis for comparing the folding behavior of diastereomers 1-4
in solution.
Amide proton chemical shifts (δNH) are very sensitive
indicators of hydrogen bond formation in nonpolar solvents such
as methylene chloride.33,34 Amide protons engaged in N-H- -
OdC hydrogen bonds are typically shifted downfield by ∼2
ppm relative to non-hydrogen bonded amide protons. If the
NMR measurements are carried out at sufficient dilution to
preclude intermolecular interactions, then hydrogen bonding
detected in this way is strictly intramolecular. Figure 1 shows
the concentration dependence of δNH for the three amide
protons of 1 in CD2Cl2. There is little variation between 0.01
mM and 1.0 mM, which suggests that there is no intermolecular
hydrogen bonding at or below 1 mM. The slight upward
curvature observed for NH1 above 1 mM suggests that some
hydrogen bond-mediated self-association occurs in this con-
centration range. Equilibration between non-hydrogen bonded
and hydrogen bonded forms is usually rapid on the NMR time
scale for small oligoamides such as those discussed here;
therefore, observed δNH values are population-weighted aver-
ages of the contributing hydrogen bonded and non-hydrogen
bonded forms.
N-H stretch region IR data also provide insight on hydrogen
bond formation in nonpolar solvents.33,34 Non-hydrogen bonded
secondary amide N-H stretch bands typically appear between
3400 and 3500 cm-1, while formation of N-H- -OdC hydrogen
bonds leads to N-H stretch bands between 3250 and 3400 cm-1.
The time scale of the IR measurement is short enough that
hydrogen bonded and non-hydrogen bonded states give rise to
distinct signals, in contrast to NMR measurements. IR spec-
troscopy has a disadvantage relative to NMR spectroscopy,
however, in that N-H signals from different secondary amides
within a molecule tend to overlap in the IR but not the NMR
spectrum
Interconversion between tertiary amide rotamers is generally
slow enough that each rotamer will give rise to a distinct set of
signals in the NMR spectrum at room temperature.35 This
interconversion is fast enough, however, for equilibrium to be
reached rapidly at room temperature. Determination of rotamer
proportions among 1-4 provides indirect information on the
Two-dimensional NMR data for 1 demonstrate that a hairpin
conformation is highly populated in CD2Cl2 (these data are
available in graphical summary form in ref 6b, and in complete
detail in the Supporting Information for ref 6b). Numerous
NOEs are observed between the two strand residues. NOEs
between adjacent residues show that the tertiary amide group
linking the first and second residues has a Z configuration and
the tertiary amide group linking the two nipecotic acid residues
has an E configuration. The crystal structure of tetra-â-peptide
5, a synthetic precursor of 1 containing a Boc group at the
N-terminus, shows that the same hairpin conformation forms
in the solid state.6b The high-resolution structural evidence for
hairpin formation by 1 and close analogue 5 is complemented
(34) Leading references on the use of NMR and IR data to determine
internal hydrogen bonding patterns of oligoamides in nonpolar solvents:
(a) Gellman, S. H.; Dado, G. P.; Liang, G.-B.; Adams, B. R. J. Am. Chem.
Soc. 1991, 113, 1164. (b) Dado, G. P.; Gellman, S. H. J. Am. Chem. Soc.
1994, 116, 1054. (c) Tsang, K. Y.; Diaz, H.; Graciani, N.; Kelly, J. W. J.
Am. Chem. Soc. 1994, 116, 3988. (d) Nowick, J. S.; Abdi, M.; Bellamo, K.
A.; Love, J. A.; Martinez, E. J.; Noronha, G.; Smith, E. M.; Ziller, J. W. J.
Am. Chem. Soc. 1995, 117, 89. (e) Gung, B. W.; Zhu, Z. J. Org. Chem.
1996, 61, 6482. (f) Yang, J.; Christianson, L. A.; Gellman, S. H. Org. Lett.
1999, 1, 11.
(33) (a) Haque, T. S.; Little, J. C.; Gellman, S. H. J. Am. Chem. Soc.
1994, 116, 4105. (b) Haque, T. S.; Little, J. C.; Gellman, S. H. J. Am.
Chem. Soc. 1996, 118, 6975. (c) Stanger, H. L.; Gellman, S. H. J. Am.
Chem. Soc. 1998, 120, 4236. (d) For related work, see: Raghothama, S.
R.; Awasthi, S. K.; Balaram, P. J. Chem. Soc., Perkin Trans. 2 1998, 137.
(35) Stewart, W. E.; Siddall, T. H. Chem. ReV. 1970, 70, 517.

3998 J. Am. Chem. Soc., Vol. 122, No. 17, 2000
Chung et al.
involves the 12-membered N-H- -OdC interaction that is
associated with reverse turn formation across the dinipecotic
acid reverse turn segment. The hydrogen bonding at NH3
(“outer”) involves the 20-membered ring N-H- -OdC interac-
tion that defines the hairpin conformation. The IR data for 1
show that most absorbance in the N-H stretch region occurs
in the hydrogen bonded region. The small band in the non-
hydrogen bonded region (3421 cm-1) must arise largely from
NH1, which does not form an intramolecular hydrogen bond
in the hairpin conformation, as established by two-dimensional
NMR and crystallographic data.6b
The data for 2, which contains the alternative heterochiral
turn segment, are similar to the data for 1; therefore, the (S,R)
dinipecotic acid unit of 2 also promotes hairpin formation. The
principal difference between 2 and 1 is the larger population of
a minor rotamer in 2, which suggests that the (S,R) turn segment
of 2 is slightly less effective than the (R,S) segment of 1 as a
hairpin inducer. The δNH pattern in Table 1 for 2 is very similar
to that for 1, suggesting that the major rotamers of both
diastereomers display similar intramolecular hydrogen bonding
patterns. This conclusion is reinforced by the N-H stretch IR
data for 2, which are quite similar to the corresponding data
for 1; the single broad hydrogen bonded band at 3354 cm-1
observed for 2 corresponds to the partially resolved bands at
3373 and 3333 cm-1 observed for 1. The small non-hydrogen
bonded N-H band at 3425 cm-1 for 2 matches a very similar
feature in the spectrum of 1.
Figure 1. Amide proton NMR chemical shifts of 1 in CD2Cl2 at room
temp, as a function of the logarithm of â-peptide concentration: (O),
NH-1; (4), NH-2; (×), NH-3.
Table 1. NMR Data for Diastereomeric Tetra-â-peptides
diastereomer
% major rotamer
δNH1a
δNH2a
δNH3a
1
2
3
4
94b
78b
82c
67c
5.80
5.81
5.90
5.73
7.71
7.32
6.14
6.46
7.49
7.41
6.15
5.88
a δNH values for major rotamer. b One minor rotamer. c Two minor
rotamers.
The data for 3 and 4 are similar to one another and quite
distinct from the data for 1 and 2. For both 3 and 4, all δNH
values of the major rotamers are relatively far upfield, which
suggests that intramolecular hydrogen bonding is limited in each
case and that there is no single preferred internal hydrogen
bonding pattern for either 3 or 4. This conclusion is supported
by the IR data for 3 and 4, which show two major bands in the
non-hydrogen bonded region, at 3455 and 3425 cm-1. The band
at 3455 cm-1 can be assigned to NH3 in the non-hydrogen
bonded state on the basis of extensive precedent. Secondary
amide NH groups bearing a methyl group, like NH3, display
non-hydrogen bonded N-H stretch bands that are typically 20
cm-1 higher in energy than those of secondary amide NH groups
with an adjacent alkyl branch point,37 as is the case for NH1
and NH2. Thus, the 3425 cm-1 bands for 3 and 4 are assigned
to non-hydrogen bonded NH1 and/or NH2. The data for 3 and
4 indicate that neither homochiral reverse turn segment is an
effective promoter of hairpin formation.
The comparison among diastereomers 1-4 shows that
â-peptide hairpin formation is very sensitive to reverse turn
configuration. These findings are consistent with earlier studies,
in which we found that the heterochiral dinipecotic acid segment
allows 12-membered ring hydrogen bond formation (i.e., this
segment can adopt a reverse turn conformation), while the
homochiral dinipecotic acid segment does not allow formation
of the 12-membered ring hydrogen bond.6b
Figure 2. N-H stretch FT-IR data for 1 mM samples in CH2Cl2 at
room temp, after subtraction of the spectrum of pure CH2Cl2. From
left to right: 1, maxima at 3421, 3373, and 3333 cm-1; 2, maxima at
3425 and 3354 cm-1; 3, maxima at 3454, 3425, 3378 (shoulder), and
3337 cm-1; 4, maxima at 3455, 3425, and 3338 cm-1
.
folding of these molecules because the hairpin conformation
observed for 1 requires the nitrogens of the first and second
nipecotic acid residues to be involved in tertiary amide bonds
with Z and E configurations, respectively (numbering from N-to-
C, as with conventional peptides).
Table 1 provides δNH values for diastereomers 1-4 and
indicates the population of the major rotamer in CD2Cl2. N-H
stretch region IR data for diastereomeric tetra-â-peptides 1-4
are shown in Figure 2.
Conformational Analysis of Tetra-â-peptides Containing
â-Alanine Strand Residues. Two types of antiparallel sheet
interactions are possible for â-peptide strand residues, depending
on whether the strand residues favor anti or gauche NCâ-CRC-
(dO) torsion angles;6a diastereomers 6 and 7 were examined
to determine whether turn segment configuration influences the
formation of a hairpin with gauche strand residues. We have
previously pointed out that anti NCâ-CRC(dO) torsion angles,
as observed for the R,â-disubstituted residues in 1-4, result in
The data for 1 are consistent with the hairpin conformation
revealed by two-dimensional NMR analysis.6b A single rota-
meric form of 1 predominates in CD2Cl2,36 and the amide proton
chemical shift pattern reflects the expected intramolecular
hydrogen bonding pattern. Thus, both δNH2 (7.71) and δNH3
(7.49) occur in a range that suggests extensive intramolecular
hydrogen bond formation, while δNH1 (5.80) indicates little
hydrogen bonding. The hydrogen bonding at NH2 (“inner”)
(36) Solvent effects provide evidence that the multiplicity of the NMR
resonances arises from amide rotamers. In CD3OH, 1 displays three sets of
NH resonances, with a ratio of 4.3:2.8:1 at 15 °C and 3.3:2.1:1 at 24 °C.
(37) Boussard, G.; Marraud, M. J. Am. Chem. Soc. 1985, 107, 1825 and
references therein.

Optimization of â-Peptide Hairpin Stability
J. Am. Chem. Soc., Vol. 122, No. 17, 2000 3999
Figure 4. N-H stretch FT-IR data for 1 mM samples in CH2Cl2 at
room temp, after subtraction of the spectrum of pure CH2Cl2. From
left to right: 6, maxima at 3453 and 3342 cm-1; 7, maxima at 3449
and 3338 cm-1
.
â-alanine strands, at least to some extent, while the homochiral
turn segment (in 7) does not promote such interactions at all.
NMR data indicate the presence of at least three amide rotamers
in slow exchange for 6 in CD2Cl2, but one of these rotamers is
predominant ( ∼70%). For the major rotamer of 6, δNH1 (7.49)
and δNH2 (7.87) both indicate amide protons that are intramo-
lecularly hydrogen bonded to a large extent, while δNH3 (5.84)
indicates an amide proton that experiences little hydrogen
bonding. These data are consistent with the interstrand hydrogen
bonding pattern expected if the â-alanine residues adopt gauche
NCâ-CRC(dO) torsion angles and engage in nonpolar sheet
formation (Figure 3); this proposed conformation is indicated
by the “solution” structure drawn for 6. The major rotamer
conformation for 6 in CD2Cl2 is not as securely established as
for 1, because resonance overlap inhibits two-dimensional NMR
analysis. Nevertheless, it is clear from the data that the major
rotamer of 6 and the major rotamer of 1 have dramatically
different internal hydrogen bonding patterns in CD2Cl2. The
δNH values of the minor rotamers of 6 fall between 6.00 and
7.00, suggesting that there is relatively little internal hydrogen
bonding in these species.
Figure 3. Two possible types of antiparallel sheet interaction between
â-peptide strands: (A) polar sheet, because all carbonyls are oriented
similarly; (B) nonpolar sheet, because carbonyl orientations alternate
along each strand.
a sheet with a strong dipole because all of the strand carbonyls
point in approximately the same direction (Figure 3).6a In
contrast, gauche NCâ-CRC(dO) torsion angles prevent a strong
dipole from developing because the carbonyls of adjacent
residues are oriented in opposite directions. â-Alanine strand
residues have been shown to adopt gauche NCâ-CRC(dO)
torsion angles and form the nonpolar type of antiparallel sheet
when attached to a non-â-peptide turn segment.6a Tetra-â-
peptides 6 and 7 allow us to determine how dinipecotic acid
linkers affect sheet interactions between â-alanine strands; since
the strand residues are achiral in this series, only two diaster-
eomeric forms are possible.
Diastereomer 7 shows evidence of at least two rotamers in
CD2Cl2, but none is predominant. All δNH values for 7 are
below 7.00, which suggests that no amide proton is internally
hydrogen bonded to a large extent in either of the rotamers.
N-H stretch region IR data for 6 and 7 (Figure 4) support the
conclusions derived from NMR data. The large band at 3342
cm-1 observed for 6 indicates substantial intramolecular N-H- -
OdC hydrogen bonding, while the presence of only a small
band at 3338 cm-1 for 7 shows that this molecule experiences
very little internal hydrogen bonding.
The crystal structure of 6 (Figure 5) reveals a hairpin
conformation that is different from the major conformation
indicated by the δNH data for 6 in methylene chloride solution.
In the solid state, the â-alanine strand residues display anti
NCâ-CRC(dO) torsion angles, and the intramolecular hydrogen
bonding pattern is analogous to that observed for 5 in the solid
state and for 1 and 2 in solution. This result is a reminder that
the conformation of a flexible molecule in the solid state may
not reflect the conformational preferences of that molecule in
solution.
Solid-State Structural Comparisons. The availability of
crystal structures for three hairpin-forming molecules, 5, 6, and
8 (ref 6a), provides an opportunity for detailed structural com-
parisons that could help identify key features of sheet secondary
structure among â-peptides. Table 2 provides backbone torsion
NMR and IR data reveal that the heterochiral dinipecotic acid
turn segment (in 6) promotes sheet interactions between

4000 J. Am. Chem. Soc., Vol. 122, No. 17, 2000
Chung et al.
Table 2. Backbone Torsion Angles of â-Amino Acid Residues in Crystalline Hairpin Moleculesa,b
residue 1
residue 2
residue 3
residue 4
â-peptide
φ
θ
ψ
φ
θ
ψ
φ
θ
ψ
φ
θ
ψ
5
94.4
176.2
-171.5
134.4
-164.9
174.7
171.4
-178.7
171.2
-122.3
-173.4
-168.3
-117.9
-142.9
115.7
98.3
99.3
-
-175.2
-179.5
-175.3
-
81.7
82.3
81.3
-
-117.4
-119.3
-124.2
-
-176.6
179.4
179.7
-
-71.2
-59.6
-59.3
-
130.1
-161.1
-176.9
110.5
173.1
-178.4
175.6
174.8
174.8
-135.6
160.5
-172.9
-132.7
-132.7
6a
6b
8a
8b
148.5
-
-
-
-
-
-
110.5
a There were two independent molecules in crystalline 6 and 8. b The backbone torsion angles are defined as:
Table 3. Intramolecular Hydrogen Bond Parameters in Crystalline
Hairpin Moleculesa
inner H-band
â-peptide d(H‚‚‚O)b (N-H‚‚‚O)c d(H‚‚‚O)b
outer H-band
(N-H‚‚‚O)c
5
2.15
170.8
162.1
161.2
149.7
2.20
2.02
2.02
2.20
172.3
147.0
150.4
144.7
6a
6b
8
2.12
2.17
2.09
a H-bond definitions:
Figure 5. Ball-and-stick representation of the solid state conformation
of 6. All hydrogen atoms, except those attached to nitrogen, have been
omitted. Hydrogen bonds are indicated with dotted lines. The nitrogen
and oxygen atoms are labeled.
b O‚‚‚H separation in Å. c N-H‚‚‚O angle in degrees.
torsion angles of the strand residues. The nipecotic acid residues
of 5 and 6 also display anti θ torsion angles, presumably
enforced by the six-membered ring. The φ torsion angles of
the nipecotic acid residues are not anti and vary significantly,
reflecting the sp2 hybridization of the nitrogen atom.
Table 3 provides geometric parameters for the intramolecular
hydrogen bonds in the crystal structures of 5, 6, and 8. The
average O- -H distance and N-H- -O angle of the “inner” and
“outer” hydrogen bonds in these structures are similar. The
N-H- -O angles of these â-peptide sheet hydrogen bonds are
comparable to the average N-H- -O angles (160°) for inter-
strand hydrogen bonds in protein antiparallel â-sheets.40 The
â-peptide hydrogen bonds are somewhat longer, and therefore
perhaps less favorable, than those in protein â-sheets (average
O- -H ) 1.96 Å).40
angles for each â-amino acid residue in each of the three struc-
tures. We follow conventional peptide nomenclature in using φ
to designate the intraresidue torsion adjacent to N (i.e., (Od)-
CN-CâCR) and ψ to designate the intraresidue torsion adjacent
to CdO (i.e., CâCR-C(dO)N). We follow Seebach et al.38 in
using θ to designate the central NCâ-CRC(dO) torsion angle
in each residue. There were two independent molecules in the
crystals of 6 and 8, and both sets of torsion angle data are
provided. All strand residues display anti θ torsion angles (near
180°). In contrast, there is significant variation among φ and ψ
Discussion
Our results show that the reverse turn conformation adopted
by a heterochiral dinipecotic acid segment promotes antiparallel
sheet interactions between attached strand residues, while a
homochiral dinipecotic acid segment is unable to induce sheet
formation. This distinction holds for both R,â-disubstituted
strand residues, which prefer anti NCâ-CRC(dO) torsion angles,
and â-alanine strand residues, which prefer gauche NCâ-CRC-
(38) Abele, S.; Seiler, P.; Seebach, D. HelV. Chim. Acta 1999, 82, 1559.
(39) Baker, E. N.; Hubbard, R. E. Prog. Biophys. Mol. Biol. 1984, 44,
(40) Wu¨thrich, K. NMR of Proteins and Nucleic Acids; Wiley: New
97.
York, NY, 1986.

Optimization of â-Peptide Hairpin Stability
J. Am. Chem. Soc., Vol. 122, No. 17, 2000 4001
Seebach reverse turn, formed across a di-â-peptide segment
composed of an R-substituted residue followed by a â-substi-
tuted residue, involves a 10-membered ring hydrogen bond
between CdO of residue i and N-H of residue i - 1.6c As
shown in Figure 6, this difference in reverse turn structure leads
to a switch in the directionality of the sheet dipole relative to
the N- and C-termini of the â-peptide.
Comparisons among the data now available for several
different â-peptide hairpins in organic solvents reveal general
trends in sheet secondary structure formed by â-amino acid
residues and suggest guidelines for the design of larger sheets
that will fold in aqueous solution. The regular secondary
structures of conventional peptides can be identified via NOEs
involving backbone atoms of residues that are not adjacent in
sequence;40 NOEs characteristic of standard antiparallel â-sheet
secondary structure include CRH- -CRH between non-hydrogen
bonded residue pairs and NH- -NH between hydrogen bonded
residue pairs. Antiparallel sheets involving â-peptide residues
also appear to display a characteristic pattern of backbone NOEs.
In an antiparallel sheet formed by â-amino acid residues with
anti NCâ-CRC(dO) torsion angles, all of the residues in one
strand are either exclusively hydrogen bond donors or exclu-
sively hydrogen bond acceptors with regard to a neighboring
strand. The three hairpins with â-peptide strands for which NOE
data are available, 1 (ref 6b), 8 (ref 6a) and the hairpin of
Seebach et al.,6c all show backbone NOEs between CâH on the
hydrogen bond acceptor strand and CRH on the hydrogen bond
donor strand (Figure 7). The different strand juxtapositions that
result from the different reverse turn segments used by us6a,b
and by Seebach et al.6c can be distinguished by which of the
interstrand CâH- -CRH NOEs is observed. Our reverse turn, with
a 12-membered ring hydrogen bond, leads to a CâH- -CRH NOE
in the N-to-C direction, while the reverse turn of Seebach et
al., with a 10-membered ring hydrogen bond, leads to a CâH- -
CRH NOE in the C-to-N direction.
Figure 6. Comparison of â-peptide hairpin from our laboratory (left,
ref 6b) and from Seebach et al. (right, ref 6c), showing how the
difference in reverse turn hydrogen bonding pattern leads to a difference
in the orientation of the sheet dipole.
(dO) torsion angles. We have previously shown that hairpin
formation among conventional peptides (R-amino acid residues)
is also strongly influenced by turn configuration: D-Pro-Xxx
segments promote hairpin formation between L-residue strands,
but L-Pro-Xxx segments discourage hairpin formation.26,33 The
origin of these effects differs between the two peptide classes.
For conventional peptides, the critical issue is matching the local
twist of the turn segment to the intrinsic twist preferred by the
strand segments.26,33 For the dinipecotic acid â-peptide reverse
turns, on the other hand, the critical issue is whether the ends
of the turn can come together to form the 12-membered ring
hydrogen bond.
The â-peptide hairpin results presented here and in previous
papers from our laboratory6a,b are complemented by a â-peptide
hairpin study very recently reported by Seebach et al.6c The
design strategy employed by Seebach and co-workers6c closely
follows our earlier efforts6a,b in two respects. First, Seebach et
al. have used the hairpin folding motif to generate soluble sheets;
these workers report that individual â-peptide strand segments
are insoluble,6c which mirrors the well-known behavior of
conventional hydrophobic peptides that are prone to adopt
extended â-strand conformations. Second, Seebach et al. have
used strand residues with the R,â-disubstitution pattern that we
previously showed to be optimal for sheet secondary structure.6a
The Seebach hairpin differs from our hairpins, however, in the
reverse turn segment (Figure 6). Our dinipecotic acid reverse
turn involves a 12-membered ring hydrogen bond between Cd
O of residue i and N-H of residue i + 3.6b In contrast, the
Side chain-side chain NOEs are often observed in â-sheets
formed by conventional peptides; these NOEs help one refine
the structure of the â-sheet, and they frequently indicate cross-
strand interactions that stabilize the sheet interaction. For sheets
formed by R,â-disubstituted â-amino acid residues, there are
two different sets of side chains that could display NOEs for
each interstrand residue pairing, the â-side chain on the
hydrogen donor residue and the R-side chain on the hydrogen
bond acceptor residue, or the R-side chain on the donor residue
and the â-side chain on the acceptor residue. Comparison of
NOE data for hairpins 1 (ref 6b) and 8 (ref 6a) indicates that
only one or the other of these pairings leads to side chain-side
Figure 7. Summary of backbone-backbone and side chain-side chain NOEs observed for 8 (A, ref 6a), 1 (B, ref 6b) and the hairpin of Seebach
et al. (C, ref 6c). In each case, there is an interstrand CâH- -CRH NOE, but the directionality varies CâH on the N-terminal side/CRH on the
C-terminal side (A, B) vs CâH on the C-terminal side/CRH on the N-terminal side (C). In each case, NOEs are observed between one pair of side
chains (designated RR or Râ), but in A this pair is attached to the carbons bearing the hydrogens that show the backbone-backbone NOE, while
in B and C this pair of side chains is attached to the “other” carbons.

4002 J. Am. Chem. Soc., Vol. 122, No. 17, 2000
Chung et al.
Nicolet 740 FT-infrared spectrometer. High-resolution electron impact
ionization mass spectroscopy was performed using a Kratos MS-25
spectrometer.
IR Studies. High quality infrared spectra (128 scans) were obtained
at 2 cm-1 resolution using a 1 mm CaF2 solution cell and a Nicolet
740 FT-infrared spectrometer. All spectra were obtained in 1 mM
solutions in anhydrous CH2Cl2 at 297 K. All compounds were dried in
vacuo at elevated temperatures in the presence of P2O5. CH2Cl2 was
distilled from CaH2 and stored over activated 4 Å molecular sieves.
All sample preparations were performed in a nitrogen atmosphere.
Resolution of Racemic HN-Nip-OC2H5 with Enantiomerically
Pure Tartaric Acid. This resolution was carried out according to the
published procedure.31,32 Racemic HN-Nip-OC2H5 (71 mL) and
D-(-)-tartaric acid (67 g) were dissolved in hot EtOH (750 mL). The
solution as cooled to room temp and stored at 4 °C overnight. The
resulting crystals were isolated by filtration and washed with cold EtOH.
Redissolution of these crystals in hot EtOH (650 mL) followed by
cooling as before gave a second batch of crystals (mp 110-130 °C).
Redissolution in hot EtOH (600 mL) and cooling as before gave a third
batch of crystals (mp 145-147 °C; [R]D ) -45.7 (c 2, 0.2% aqueous
ammonium molybdate). Redissolution in hot EtOH (600 mL) and
cooling as before gave a fourth batch of crystals (mp 153-154 °C;
[R]D ) -51.0 (c 2, 0.2% aqueous ammonium molybdate). Redissolution
in hot EtOH (650 mL) and cooling as before gave a fifth batch of
crystals (mp 155-156 °C; [R]D ) -51.0 (c 2, 0.2% aqueous ammonium
molybdate) (lit.32,42 mp 155-156 °C, [R]D ) -51.0).
Figure 8. Nonbonded carbon-carbon distances observed in the crystal
structures of 8 (left, ref 6a) and 5 (right, ref 6b). These distances are
consistent with the pattern of side chain-side chain NOEs observed
for 8 and 1 in solution (Figure 7A and B). Thus, the smaller separation
for 8 involves the CR substituent of the N-terminal strand and the Câ
substituent of the C-terminal strand (there are two distances because
of disorder in this part of the structure), corresponding to the observed
side chain-side chain NOE. The smaller separation for 5 involves the
Câ substituent of the N-terminal strand and the CR substituent of the
C-terminal strand, and there is a corresponding side chain-side chain
NOE for 1.
chain NOEs (Figure 7). An NOE is observed between the
methylene of the â-side chain of residue 1 and the methylene
of the R-side chain of residue 4 in 8 (ref 6a) while NOEs are
observed between the methylene of the R-side chain of residue
1 and the methylene of the â-side chain of residue 4 of 1 (ref
6b). The crystal structures of 5 (a close analogue of 1) and 8
reveal a variation in side chain separations that mirrors the side
chain-side chain NOE differences observed between 1 and 8
in solution (Figure 8). Further inspection of the crystal structures
indicates that the variation in interstrand juxtaposition between
the two molecules arises at least in part from variations in the
φ torsion angles in each strand residue (Table 2).
Resolution of Racemic Nipecotic Acid with Enantiomerically
Pure Camphorsulfonic Acid. (1R)-(+)-10-Camphorsulfonic acid
(11.62 g, 0.05 mol) was added to a stirred solution of racemic nipecotic
acid (6.46 g, 0.05 mol) in acetone (100 mL). The solution was heated
to reflux, and H2O (15 mL) was added until all solids dissolved. The
solution was cooled to room temperature and allowed to stir overnight.
The precipitate was isolated by filtration and recrystallized three times
from 6:1 acetone:H2O to afford 1.72 g (10% yield) of the desired
product as a white solid: mp 221-223 °C; [R]D -25.9 (c 1.0, MeOH).
Subsequent results indicated this salt to contain (R)-nipecotic acid.
Boc-(R)-Nip-OH. To a 0 °C solution of (R)-nipecotic acid (1R)-
(+)-10-camphorsulfonic acid salt (3.62 g, 10 mmol) in MeOH (40 mL)
were added Et3N (7 mL, 50 mmol) and (Boc)2O (2.62 g, 12 mmol).
After stirring 4 h at 50 °C, the solution was concentrated under reduced
pressure. The residue was dissolved in water (20 mL) and acidified to
pH 2-3 by adding saturated KHSO4 solution. The solution was
extracted with EtOAc, and the organic layer was washed with 1 N
HCl and brine. The organic extract was dried over MgSO4, concentrated,
and dried under vacuum to afford Boc-(R)-Nip-OH (2.28 g, 99%) as a
The side chain-side chain interaction trends noted here have
important ramifications for design of â-peptide hairpins that fold
in aqueous solution. Hairpin formation in water is likely to
require favorable interactions between large hydrophobic side
chains on neighboring strands, as is the case for conventional
peptides that adopt â-hairpin conformations in aqueous solu-
tion.41 None of the small â-peptide hairpins reported so far
provides the opportunity for favorable interstrand pairing of
hydrophobic side chains: the side chain pairs in these examples
all have a methyl or an ethyl group on one side,6 and these
moieties are too small to provide the necessary hydrophobic
driving force. We are currently using this and other design
principles in an effort to generate â-peptides that display stable
hairpin conformations in water.
1
white solid: mp 166-167 °C (lit.42 mp 165.8-166.6 °C); H NMR
(CDCl3) δ 11.11(br s), 4.13 (br), 3.88 (dt, J ) 13.2, 7.6 Hz), 3.01 (br),
2.87 (m), 2.48 (m), 2.06 (m), 1.69 (m), 1.46 (s); [R]D ) -52.17 (c
1.15, CH3OH) (lit.42 [R]D ) -41.6 (c 5, CH3OH)).
HPLC Analysis of Enantiomeric Purity. Boc-Nip-OH, generated
from either of the two resolution methods described, was converted to
the corresponding 2-naphthylmethyl ester as follows. Boc-Nip-OH (0.25
g, 1.1 mmol) was dissolved in N,N-dimethylformamide (DMF) (5 mL),
Cs2CO3 (0.36 g, 1.1 mmol) and 2-(bromomethyl)naphthalene (0.28 g,
1.2 mmol) were added, and the solution was stirred at room temp for
24 h. The solution was then concentrated, and the residue was dissolved
in H2O. The aqueous solution was extracted with CH2Cl2. The organic
layer was dried over MgSO4 and concentrated to give an oil. The crude
product was purified by silica chromatography eluting with 1:2 ethyl
acetate:hexanes to afford 0.34 g (84% yield) of the desired product as
a white solid: mp 70-71 °C; 1H NMR (CDCl3, 300 MHz) δ 7.9-7.8
(m), 7.55-7.45 (m), 5.3 (s), 4.40-4.05 (br s), 4.00-3.85 (br d), 3.10-
2.90 (br s), 2.85-2.75 (dt), 2.60-2.45 (m), 2.15-2.05 (m), 1.75-
1.55 (m), 1.55-1.40 (s); FAB-MS m/z (M + Na+) calcd for
C22H27NO4Na 392.0, obsd 392.6.
Experimental Section
General. All melting points are uncorrected. Reagents employed
were either commercially available or prepared according to a known
procedure. Anhydrous CH2Cl2 was obtained by distillation from CaH2.
All other solvents used were reagent grade except for hexane, which
was purified by distillation. Anhydrous reaction conditions were
maintained under a slightly positive nitrogen atmosphere in oven-dried
glassware. Silica gel chromatography was performed using 230-400
mesh silica purchased from EM Science. Routine 1H NMR spectra were
obtained on a Bruker AM-300 spectrometer. NMR spectra were
referenced to TMS (0 ppm). Infrared spectra were obtained using a
Chiral HPLC was performed on the 2-naphthylmethyl ester of Boc-
Nip using a 250 × 4.6 mm Chiracel OD column (Chiral Technologies
Inc., Exton, PA) eluting with 9:1 hexanes:2-propanol at a flow rate of
1.0 mL/min. The retention times for the (R)- and (S)-isomers were 8.8
(41) (a) de Alba, E.; Rico, M.; Jime´nez, M. A. Protein Sci. 1997, 6,
2548. (b) Maynard, A. J.; Sharman, G. J.; Searle, M. S. J. Am. Chem. Soc.
1998, 120, 1996.
(42) Zheng, X.; Day, C.; Gollamudi, R. Chirality 1995, 7, 90.

Optimization of â-Peptide Hairpin Stability
J. Am. Chem. Soc., Vol. 122, No. 17, 2000 4003
and 9.4 min. After each type of resolution, optical purity of each
enantiomer g99% (i.e., g98% e.e.).
mL) and cooled to 0 °C. To this solution were added Et3N (0.11 mL,
0.80 mmol) and acetic anhydride (47 mL, 0.46 mmol). After stirring
for 15 h at room temperature, the solution was concentrated under
reduced pressure. The residue was purified by silica chromatography
(EtOAc:MeOH ) 2:1), yielding 6 (0.12 g, 68%) as a white solid: mp
Boc-(R)-Nip-(S)-Nip-OCH3. (S)-Nip-OCH3 was prepared from
S-nipecotic acid (1R)-(+)-10-camphorsulfonic acid salt with SOCl2 in
MeOH. Boc-(R)-Nip-OH (1.72 g, 7.5 mmol), (S)-Nip-OCH3 (7.5 mmol),
and HOBt (1.22 g, 9 mmol) were dissolved in DMF (100 mL). The
solution was cooled to -15 °C, and Et3N (3.13 mL, 22.5 mmol) and
EDC (1.73 g, 9 mmol) were added. After stirring for 40 h at room
temp, the solution was concentrated under reduced pressure. The residue
was dissolved in EtOAc and washed with 1 N HCl, saturated NaHCO3,
and brine. The organic extract was dried over MgSO4 and concentrated.
The crude product was purified by silica chromatography (hexanes:
EtOAc ) 1:1), yielding Boc-(R)-Nip-(S)-Nip-OCH3 (2.58 g, 97%) as
a white solid: mp (recrystallized from 1:5 EtOAc:hexane) 86.5-88
1
(recrystallized from 10:1:70 EtOAc:MeOH:hexane) 155-157 °C; H
NMR (CD2Cl2, 500 MHz) δ 7.87 (br s), 7.49 (br s), 5.84 (br s), 4.64
(d, J ) 13.0 Hz), 4.54 (d, J ) 11.5 Hz), 4.14 (d, J ) 13.0 Hz), 3.83
(d, J ) 16.0 Hz), 3.61-3.38 (m), 3.29 (m), 3.04 (td, d ) 12.1, 6.0
Hz), 2.97 (t, J ) 12.7 Hz), 2.74 (d, J ) 5.0 Hz), 2.64-2.57 (m), 2.55
(m), 2.49 (m), 2.45 (m), 2.37 (m), 1.95-1.72 (m), 1.54 (s), 1.48-1.26
(m); EI-MS m/z (M+) calcd for C21H35O5N5 437.2638, obsd 437.2644;
IR (CH2Cl2) 3453, 3342, 1662, 1633, 1530 cm-1. Crystals of 7 suitable
for X-ray analysis were grown from a solution of CH2Cl2-MeOH-
hexane.
1
°C. H NMR (CD2Cl2) δ 4.50 (br d), 4.02 (br d), 3.78 (br t), 3.66 (s),
3.42 (m), 3.09-2.94 (m), 2.81 (m), 2.66 (m), 2.56-2.52 (m), 2.05 (m),
1.81 (m), 1.65 (m), 1.44 (s); [R]D ) +23.0 (c 1.20, CHCl3); EI-MS
m/z (M+) calcd for C18H30O5N2 354.2155, obsd 354.2151.
Ac-â-Ala-(R)-Nip-(R)-Nip-ââ-Ala-NHCH3 (7). The compound was
synthesized by a route similar to that used for 6. 1H NMR (CD2Cl2) δ
6.58 (br d), 6.42 (br d), 6.18 (br d), 4.43 (br t), 4.09 (br d), 3.67 (m),
3.46 (m), 3.21 (m), 3.04 (m), 2.85 (m), 2.74 (d, J ) 4.2 Hz), 2.58 (m),
2.48 (m), 2.35 (m), 1.89 (s), 1.86-1.44 (m); EI-MS m/z (M+) calcd
for C21H35O5N5 437.2638, obsd 437.2627; IR (CH2Cl2) 3338, 1668,
Boc-(R)-Nip-(S)-Nip-OH. To a 0 °C solution of Boc-(R)-Nip-(S)-
Nip-OCH3 (1.55 g, 4.37 mmol) in THF (20 mL) was added dropwise
a solution of LiOH‚H2O (0.19 g, 4.6 mmol) in H2O (10 mL). After 4
h at room temp, the solution was concentrated, diluted with H2O, and
extracted with Et2O. The aqueous layer was acidified to pH 2-3 with
1 N HCl. The solution was extracted with EtOAc, and the organic layer
was dried over MgSO4, concentrated, and dried under vacuum to afford
Boc-(R)-Nip-(S)-Nip-OH (1.45 g, 97%) as a white solid: mp (recrystal-
1634, 1515 cm-1
.
Tetra-â-peptides 1-4 were prepared by routes similar to that
described above for 6 (â-amino acids were coupled by using EDC and
HOBt in DMF). Synthesis of the terminal â-amino acid residues
common to 1-4 has been previously described in detail (ref 30 and
Supporting Information for ref 6a).
1
lized from 1:10 EtOAc:hexane) 166-167 °C; H NMR (CD2Cl2) δ
9.55 (br), 4.44 (br d), 4.05 (br d), 3.83 (m), 3.40 (m), 3.14 (br t), 2.93
(m), 2.84-2.56 (m), 2.45 (m), 2.10 (m), 1.83 (m), 1.66 (m), 1.43 (s);
[R]D ) +17.4 (c 1.08, CHCl3); EI-MS m/z (M+) calcd For C17H28O5N2
340.1998, obsd 340.1998.
Tetra-â-peptide 1: mp (recrystallized from 1:7 CHCl3:heptane)
1
214.5-217 °C; H NMR (CD2Cl2, 500 MHz) δ 7.70 (d, J ) 9.5 Hz),
7.49(d, J ) 4.8 Hz), 7.38-7.15 (m), 5.80 (d, J ) 10.5 Hz), 4.86 (m),
4.64 (d, J ) 13.0 Hz), 4.57 (d, J ) 12.5 Hz), 4.29 (d, J ) 12.5 Hz),
4.05 (m), 3.52 (d, J ) 11.5 Hz), 3.29 (dd, J ) 17.5, 8.0 Hz), 3.17 (d,
J ) 14.0 Hz), 3.09 (t, J ) 17.5, 12.5 Hz), 2.96 (dd, J ) 19.5, 7.8 Hz),
2.90 (d, J ) 8.0 Hz), 2.74 (d, J ) 7.8 Hz), 2.72 (m), 2.54 (m), 2.40
(td, d ) 13.0, 4.3 Hz), 2.28 (t, J ) 12.0 Hz), 2.19 (t, J ) 11.5 Hz),
1.99 (m), 1.92 (s), 1.89 m), 1.76 (d, J ) 13.5 Hz), 1.70-1.60 (m),
1.53-1.26 (m), 1.15 (m), 1.14 (d, J ) 6.1 Hz), 0.93 (t, J ) 7.3 Hz),
-0.52 (m); FAB-MS m/z (M+ + Na) calcd for C38H53N5O5SNa 714.9,
obsd 714.3; IR (CH2Cl2) 3421, 3373, 3333, 1668, 1635, 1615, 1538
Boc-(R)-Nip-(S)-Nip-â-Ala-NHCH3. Boc-(R)-Nip-(S)-Nip-OH (0.50
g, 1.46 mmol), â-alanine N-methylamide hydrochloride (0.27 g, 2.0
mmol) and HOBt (0.27 g, 2.05 mmol) were dissolved in DMF (25
mL). The solution was cooled to -15 °C, and Et3N (0.6 mL, 3.60 mmol)
and EDC (0.39 g, 2.05 mmol) were added. After stirring for 36 h at
room temperature, the solution was concentrated under reduced pressure.
The residue was dissolved in CH2Cl2 and washed with 1 N HCl,
saturated NaHCO3 and brine. The organic extract was dried over Na2-
SO4 and concentrated. The crude product was purified by silica
chromatography (EtOAc:MeOH ) 30:4), yielding Boc-(R)-Nip-(S)-
Nip-â-Ala-NHCH3 (0.57 g, 94%) as a white solid: mp (recrystallized
cm-1
.
Tetra-â-peptide 2: mp 136-140.5 °C; 1H NMR (CD2Cl2, 500
MHz) δ 7.41 (d, J ) 8.5 Hz, with minor rotamer at 7.21), 7.36-7.10
(m), 6.80 (d, J ) 9.8 Hz, with minor rotamer at 7.32), 5.81 (d, J )
10.0 Hz, with minor rotamer at 6.20), 4.85 (m), 4.63 (d, J ) 13.0 Hz),
4.55 (d, J ) 13.5 Hz), 4.12 (m), 3.68 (d, J ) 10.5 Hz), 3.41 (m), 3.25
(m), 3.04 (m), 2.97-2.87 (m), 2.76 (d), 2.73 (m), 2.48-2.27 (m), 2.12
(t, J ) 12.3 Hz), 1.94 (m), 1.85-1.62 (m), 1.55 (s), 1.40 (m), 1.26
(m), 1.05 (d), 0.95 (t, J ) 8.3 Hz), 0.87 (m); FAB-MS m/z (M+ + Na)
calcd for C38H53N5O5SNa 714.9, obsd 714.0; IR (CH2Cl2) 3425, 3357,
1
from 1:1 EtOAc:hexane) 203-205 °C; H NMR (CDCl3) δ 7.43 (br,
with minor rotamer at 6.67), 6.39 (br, with minor rotamer at 6.09),
4.56 (m), 4.22-3.99 (m), 3.69 (m), 3.66-3.43 (m), 3.30 (m), 3.08 (m),
2.79 (t), 2.68 (m), 2.47 (m), 2.35 (t), 2.28 (m), 1.95-1.68 (m), 1.47
(s); FAB-MS m/z (M+ + Na) calcd for C21H36N4O5Na 447.5, obsd
447.7.
Boc-â-Ala-(R)-Nip-(S)-Nip-â-Ala-NHCH3. Boc-(R)-Nip-(S)-Nip-
â-Ala-NHCH3 (0.28 g, 0.66 mmol) was dissolved in 4 N HCl in dioxane
(2.5 mL). After 2 h, the solvent was removed by a stream of N2, and
the residue was dried under vacuum. The resulting solid (HN-(R)-
Nip-(S)-Nip-â-Ala-NHCH3‚HCl), Boc-â-alanine (0.19 g, 0.98 mmol)
and HOBt (0.12 g, 0.92 mmol) were dissolved in DMF (10 mL). The
solution was cooled to -15 °C, and NMM (0.18 mL, 1.58 mmol) and
EDC (0.17 g, 0.92 mmol) were added. After stirring for 36 h at room
temperature, the solution was concentrated under reduced pressure. The
residue was dissolved in CH2Cl2 and washed with 1 N HCl, saturated
NaHCO3 and brine. The organic layer was dried over Na2SO4 and
concentrated. The crude product was purified by silica chromatography
(EtOAc:MeOH ) 6:1), yielding Boc-â-Ala-(R)-Nip-(S)-Nip-â-Ala-
NHCH3 (0.31 g, 96%) as an oil. 1H NMR (CDCl3) δ 7.78 (br s), 6.15
(br s), 6.09 (br s), 4.62 (m), 4.11 (m), 3.8 (m), 3.69-3.49 (m), 3.38
(m), 3.03 (m), 2.77 (d, J ) 4.5 Hz), 2.61-2.43 (m), 1.89-1.72 (m),
1.42 (s); EI-MS m/z (M+) calcd for C24H41O6N5 495.3057, obsd
495.3049.
3355, 1668, 1635, 1615, 1533 cm-1
.
Tetra-â-peptide 3: mp 132-136 °C; 1H NMR (CD2Cl2, 500 MHz)
δ 7.34-7.12 (m), 5.90 (d, J ) 9.0 Hz, with minor rotamers at 6.86
and 5.70), 6.14 (br, with minor rotamers at 6.49 and 6.43), 6.15 (br,
with minor rotamers at 6.42 and 5.84), 4.56 (m), 4.27 (m), 4.21 (d, J
) 12.0 Hz), 3.95 (m), 3.85 (m), 3.47-3.34 (m), 3.29-3.14 (m), 2.96-
2.73 (m), 2.67 (s), 2.61 (br d), 2.37-2.14 (m), 2.12-2.06 (m), 2.04
(m), 1.86 (s), 1.81-1.58 (m), 1.53 (s), 1.50 (m), 1.40 (m), 1.29 (m),
1.27-1.09 (m), 1.07 (d, J ) 13.5 Hz), 0.95-0.85 (m); FAB-MS m/z
(M+ + Na) calcd for C38H53N5O5SNa 714.9, obsd 714.0; IR (CH2Cl2)
3454, 3425, 3378, 3337, 1672, 1629, 1510 cm-1
.
Tetra-â-peptide 4: mp 138-142 °C; 1H NMR (CD2Cl2, 500 MHz)
δ 7.31-7.10 (m), 6.45 (br t, with minor rotamers at 6.22 and 6.12),
5.73 (br, with minor rotamers at 6.08 and 5.69), 5.88 (br, with minor
rotamer at 6.42), 4.52 (d, J ) 13.0 Hz),), 4.31 (m), 3.95-3.84 (m),
3.59-3.43 (m), 3.37-3.01 (m), 2.98-2.87 (m), 2.78 (m), 2.74 (d, J )
7.2 Hz), 2.4 (m), 2.32 (m), 2.19 (m), 1.98 (m), 1.86 (d, J ) 6.5 Hz),
1.71 (m), 1.58-1.43 (m), 1.32 (m), 1.10 (d, J ) 7.0 Hz), 0.96 (m),
0.89 (m); FAB-MS m/z (M+ + Na) calcd for C38H53N5O5SNa 714.9,
obsd 714.3; IR (CH2Cl2) 3455, 3425, 3338, 1676, 1629, 1549, 1510
Ac-â-Ala-(R)-Nip-(S)-Nip-â-Ala-NHCH3 (6). Boc-â-Ala-(R)-Nip-
(S)-Nip-â-Ala-NHCH3 (0.20 g, 0.40 mmol) was dissolved in 4 N HCl
in dioxane (2 mL). After 2 h, the solvent was removed under a stream
of N2, and the residue was dried under vacuum. This residue (H2N-â-
Ala-(R)-Nip-(S)-Nip-â-Ala-NHCH3‚HCl) was dissolved in CH2Cl2 (5
cm-1
.

4004 J. Am. Chem. Soc., Vol. 122, No. 17, 2000
Chung et al.
Crystallography of 6: C21H35N5O5, fw ) 437.54, colorless prism,
0.24 × 0.10 × 0.06 mm, crystal system: triclinic, space group: P1, a
) 9.1329(9) Å, b ) 10.0437(10) Å, c ) 13.6403(13) Å, R ) 75.350-
(2)°, â ) 71.602(2)°, γ ) 71.618(2)°, V ) 1110.20(19) Å3, Z ) 2, T
) 133(2) K, R ) 0.0558, wR(all data) ) 0.1184, GOF ) 0.944, Bruker
SMART/P4 diffractometer, λ ) 0.71073 Å. The data were corrected
for absorption (Tmin ) 0.798, Tmax ) 0.976). The structure was solved
by direct methods and refined by full-matrix least squares on F2 (ref
43). The non-hydrogen atoms were refined with anisotropic displace-
ment parameters. Hydrogen atoms were originally positioned by
geometry and were refined isotropically using a riding model.
Acknowledgment. This research was supported by the
National Institutes of Health (GM56415). Y.J.C. was supported
in part by a fellowship from the Korea Science and Engineering
Foundation, H.E.S. by a National Research Service Award (T32
GM08923), and S.K. by a fellowship from the Fonds der
Chemischen Industrie. NMR equipment was purchased in part
with funds from the NIH (S1O RRO 4981), and crystallographic
equipment in part with funds from the NSF (CHE-9310428).
Supporting Information Available: Crystallographic data
for 6 (PDF). This material is available free of charge via the
Internet at http://pubs.acs.org.
(43) Sheldrick, G. M. SHELXTL, version 5, Reference Manual; Siemens
Analytical X-ray Instruments: 6300 Enterprise Dr., Madison, WI, 1994;
pp 53719-1173.
JA993416P