First interchain peptide interaction detected by ESR in fully synthetic, template-assisted, two-helix bundles

Article abstract of DOI:10.1021/ja992079h

We have designed and synthesized by solution methods two simple two-helix bundles based on a conformationally constrained cyclo-dipeptide template to the side chains of which two short, 2,2,6,6-tetramethylpiperidine-1-oxyl-4-amino-4-carboxylic acid spin-monolabeled, 3₁₀-helical peptides are covalently tethered. The preferred conformation of the appended chains has been assessed by FTIR absorption. The conclusions are corroborated by an X-ray diffraction analysis of one of the terminally blocked pentapeptide tails. For the first time, a solvent-dependent, inter-helix interaction has been monitored by conventional ESR spectroscopy on fully synthetic peptide systems. Half-field ESR measurements of these side-chain-substituted templates provided an experimental average distance between the two labels that is in good accord with that determined in a molecular modeling study.

Full text of DOI:10.1021/ja992079h

J. Am. Chem. Soc. 1999, 121, 11071-11078
11071
First Interchain Peptide Interaction Detected by ESR in Fully
Synthetic, Template-Assisted, Two-Helix Bundles
†
‡
,‡
†
Alessandra Polese, D. Joe Anderson, Glenn Millhauser,* Fernando Formaggio,
†
†
,†
Marco Crisma, Fernando Marchiori, and Claudio Toniolo*
Contribution from the Department of Chemistry and Biochemistry, UniVersity of California at Santa Cruz,
Santa Cruz, California 95064, and Biopolymer Research Center, CNR, Department of Organic Chemistry,
UniVersity of PadoVa, 35131 PadoVa, Italy
ReceiVed June 18, 1999. ReVised Manuscript ReceiVed September 17, 1999
Abstract: We have designed and synthesized by solution methods two simple two-helix bundles based on a
conformationally constrained cyclo-dipeptide template to the side chains of which two short, 2,2,6,6-
tetramethylpiperidine-1-oxyl-4-amino-4-carboxylic acid spin-monolabeled, 310-helical peptides are covalently
tethered. The preferred conformation of the appended chains has been assessed by FTIR absorption. The
conclusions are corroborated by an X-ray diffraction analysis of one of the terminally blocked pentapeptide
tails. For the first time, a solvent-dependent, inter-helix interaction has been monitored by conventional ESR
spectroscopy on fully synthetic peptide systems. Half-field ESR measurements of these side-chain-substituted
templates provided an experimental average distance between the two labels that is in good accord with that
determined in a molecular modeling study.
Introduction
Nitroxide-based, spin mono-labeled peptide and protein
variants are currently extensively investigated by electron spin
resonance (ESR) and fluorescence quenching techniques to
assess a variety of properties, including 3D structure; confor-
mational transition; self-association tendency; local and global
mobilities; conservation of subunit interface and overall topology
in multiple-helix bundles; accessibility of functional groups;
interaction with small ligands, metal ions, surfactants, and
receptors; immersion depth and orientation into biological and
Figure 1. Structures of a typical side-chain spin label Cys (A) and
TOAC (B).
stability of turns and helices and on intrahelix residue-residue
2
1
distances (ESR spectroscopic ruler).
model membranes; and gating mechanism.
Commonly, the nitroxide label is introduced into a flexible
Cys side chain through a flexible spacer arm (Figure 1A) after
total chemical synthesis or site-directed mutagenesis of the
peptide or protein target. More recently, we have shown that
significantly stronger spin-spin interactions may be provided
by polypeptide systems based on two residues of the side-chain
On the other hand, ESR studies of spin-spin interactions of
double nitroxide-labeled peptides and proteins have recently
provided relevant information on the precise conformation and
†
University of Padova.
University of California at Santa Cruz.
‡
(
1) Millhauser, G. L. Trends Biochem. Sci. 1992, 17, 448-452. (b)
R
Pertinhez, T. A.; Nakaie, C. R.; Paiva A. C. M.; Schreier, S. Biopolymers
conformationally restricted, strongly helicogenic, C -tetrasub-
stituted R-amino acid 2,2,6,6-tetramethylpiperidine-1-oxyl-4-
1
997, 42, 821-829. (c) Altenbach, C.; Greenhalgh, D. A.; Khorana, H. G.;
Hubbell, W. L. Proc. Natl. Acad. Sci. U.S.A. 1994, 91, 1667-1671. (d)
Hubbell, W. L.; Altenbach, C. Curr. Opin. Struct. Biol. 1994, 4, 566-573.
(2) (a) Miick, S. M.; Martinez, G. V.; Fiori, W. R.; Todd, A. P.;
Millhauser, G. L. Nature 1992, 359, 653-655. (b) Millhauser, G. L.
Biochemistry 1995, 34, 3873-3877. (c) Smithe, M. L.; Nakaie, C. R.;
Marshall, G. R. J. Am. Chem. Soc. 1995, 117, 10555-10562. (d) He, M.
M.; Voss, J.; Hubbell, W. L.; Kaback, H. R. Biochemistry 1997, 36, 13682-
13687. (e) Rabenstein, M. D.; Shin, Y. K. Proc. Natl. Acad. Sci. U.S.A.
1995, 92, 8239-8243. (f) Saxena, S.; Freed, J. H. Chem. Phys. Lett. 1996,
251, 102-110. (g) Sankarapandi, S.; Sukumar, M.; Balaram, P.; Manoharan,
P. T. Biochem. Biophys. Res. Commun. 1995, 213, 439-446. (h) Wertz, S.
L.; Savino, Y.; Cafiso, D. S. Biochemistry 1996, 35, 11104-11112. (i)
Monaco, V.; Formaggio, F.; Crisma, M.; Toniolo, C.; Hanson, P.;
Millhauser, G. L.; George, C.; Deschamps, J. R.; Flippen-Anderson J. L.
Bioorg. Med. Chem. 1999, 7, 119-131. (j) Hanson, P.; Martinez, G.;
Millhauser, G. L.; Formaggio, F.; Crisma, M.; Toniolo, C.; Vita, C. J. Am.
Chem. Soc. 1996, 118, 271-272. (k) Hanson, P.; Millhauser, G.; Formaggio,
F.; Crisma, M.; Toniolo, C. J. Am. Chem. Soc. 1996, 118, 7618-7625. (l)
Toniolo, C.; Valente, E.; Formaggio, F.; Crisma, M.; Pilloni, G.; Corvaja,
C.; Toffoletti, A.; Martinez, G. V.; Hanson, M. P.; Millhauser, G.; George,
C.; Flippen-Anderson, J. L. J. Pept. Sci. 1995, 1, 45-57. (m) Hanson, P.;
Anderson, D. J.; Martinez, G.; Millhauser, G. L.; Formaggio, F.; Toniolo,
C.; Vita, C. Mol. Phys. 1998, 95, 957-966.
(e) Ottemann, K. M.; Thorgeirsson, T. E.; Kolodziej, A. F.; Shin, Y.-K.;
Koshland, D. E., Jr. Biochemistry 1998, 37, 7062-7069. (f) Jackson, A.
E.; Harris, T. M.; Puett, D. J. Protein Chem. 1987, 6, 497-515. (g)
Brandenburg, D.; Fabry, M.; Engels, N.; Kleinjung, J.; Strack, U.; Thevis,
W. In Peptides 1996; Ramage, R., Epton, R., Eds.; Mayflower Sci.:
Kingswinford, England, 1998; pp 101-104. (h) Stopar, D.; Jansen, K. A.
J.; Pali, T.; Marsh, D.; Hemminga, M. Biochemistry 1997, 36, 8261-8268.
(i) Victor, K.; Cafiso, D. S. Biochemistry 1998, 37, 3402-3410. (j)
Johansson, J. S.; Gibney, B. R.; Rabanal, F.; Reddy, K. S.; Dutton, P. L.
Biochemistry 1998, 37, 1421-1429. (k) Koteiche, H. A.; Berengian, A.
R.; Mehaourab, H. S. Biochemistry 1998, 37, 12681-12688. (l) Klug, C.
S.; Eaton, S. S.; Eaton, G. R.; Feix, J. B. Biochemistry 1998, 37, 9016-
9
023. (m) T o¨ r o¨ k, M.; Hideg, K.; Dux, L.; Horvath, L. J. Mol. Struct. 1997,
4
08/409, 177-180. (n) Subczynski, W. K.; Lewis, R. N. A. H.; McElhaney,
R. N.; Hodges, R. S.; Hyde, J. S.; Kusumi, A. Biochemistry 1998, 37, 3156-
3172. (o) Griffith, O. H.; Waggoner, A. S. Acc. Chem. Res. 1969, 2, 17-
4. (p) Pispisa, B.; Palleschi, A.; Stella, L.; Venanzi, M.; Toniolo, C. J.
2
Phys. Chem. 1998, B102, 7890-7898. (q) Monaco, V.; Formaggio, F.;
Crisma, M.; Toniolo, C.; Hanson, P.; Millhauser, G. Biopolymers 1999,
5
0, 239-253.
1
0.1021/ja992079h CCC: $18.00 © 1999 American Chemical Society
Published on Web 11/18/1999

11072 J. Am. Chem. Soc., Vol. 121, No. 48, 1999
Polese et al.
amino-4-carboxylic acid (TOAC), which is characterized by a
chromatic reaction as appropriate. All compounds were obtained in a
chromatographically homogeneous state. Mass spectra were recorded
for all TOAC-containing peptides by means of a time-of-flight Reflex
mass spectrometer using the MALDI ionization tecnique. IR absorption
spectra were obtained in KBr pellets on a Perkin-Elmer model 580-B
spectrophotometer equipped with a Perkin-Elmer model 3600 IR data
station and a model 660 printer.
stable nitroxide moiety partially incorporated into the six-
membered ring side chain (Figure 1B).2i-m,3
In the present study, we monitored for the first time by ESR
the onset of an intramolecular inter-helix spin-spin interaction
in fully synthetic, simple models based on multiple (two)-helix
peptide bundles. The models are characterized by a semi-rigid
The free amino acid TOAC and its fluorenyl-9-methyloxycarbonyl
4
template, the 2,5-diketopiperazine (DKP) cyclo-(Glu-Glu), and
R
(
Fmoc) N -protected derivative were prepared according to published
by two identical, aligned in parallel, putative helical peptide
tails. The DKP bis(carboxylic acid) platform, because of the
cis geometry imparted by the two homo-chiral components of
the six-membered ring, provides a unique opportunity to
systematically alter the side-chain substituents while holding
the orientation between them fixed relative to acyclic analogues.
The N-terminal amino function of each tail is covalently linked
to a Glu γ-carboxylic function of the DKP scaffold via an amide
bond. In the two pentapeptide chains of 2, a single, guest TOAC
residue was incorporated in the central position of an R-ami-
noisobutyric acid (Aib) host homo-peptide chain. The presence
2l,8a
procedures.
Since the acidic and reducing conditions required to
remove the tert-butyloxycarbonyl (Boc) and benzyloxycarbonyl (Z)
groups, respectively, are not compatible with the full integrity of the
nitroxide moiety,8b the Fmoc N -protecting group was chosen for the
stepwise elongation of the TOAC-containing peptides. The Fmoc group
R
was removed by treatment with a 25% diethylamine solution in CH
Cl . After evaporation of the solvent, the N-terminal free peptide was
dissolved in CHCl and isolated by elution through a 3-cm bed of silica
gel using a CHCl -EtOH 9:1 mixture.
2
-
2
3
3
The Aib and Ala residues were incorporated using the symmetrical
anhydride approach (method I), while the TOAC residues were
introduced by either the 1-hydroxy-7-aza-benzotriazole (HOAt)-medi-
R
of five C -tetrasubstituted R-amino acids in this pentapeptide
9a
9b
ated carbodiimide method (method II) or the acyl fluoride method
(method III). Fmoc-TOAC-F was prepared from the N -protected amino
allowed us to be confident that a rather stable 310-helical
R
5
6
conformation forms under appropriate experimental conditions.
In the side-chain bis-substituted template, 3, two Aib residues
in positions 2 and 4) were replaced by the weaker helix-
acid, cyanuric fluoride, and pyridine in CH
2
Cl
2
as described in ref 5.
The substituted linear template Ac-TOAC-Aib
2
-TOAC-Aib-OtBu (1)
was synthesized from its N-terminal free analogue by treatment with
acetic anhydride in CH Cl
(
R
6
2
2
.
inducing, C -trisubstituted R-amino acid Ala. For a comparative
analysis, the linear pentapeptide model Ac-TOAC-Aib2-TOAC-
Aib-OtBu (1) (Ac, acetyl; OtBu, tert-butoxy) was also synthe-
sized. Our conformational study of peptides 1-3 was forced to
rely heavily on infrared (IR) absorption and X-ray diffraction
techniques (the latter for the linear model peptide 1). Indeed,
For the synthesis of the side-chain-substituted cyclic templates 2
R
and 3, the linear N -deprotected pentapeptides were coupled to the
4
cyclo-(Glu-Glu) template by activating the γ-carboxylic groups of the
Glu side chains of the DKP with HOAt/carbodiimide (method IV).
The physical properties of the amino acid derivatives and peptides
are listed in Table 1. The substituted linear template 1 and cyclic
templates 2 and 3 were additionally characterized by amino acid analysis
1
for TOAC-containing short peptides, neither H NMR (nuclear
magnetic resonance) nor CD (circular dichroism) analysis is
informative due to a paramagnetic (nitroxide) line broadening
(C. Erba model 3A30 amino acid analyzer, Rodano, Milan, Italy). It is
worth noting that TOAC is unstable under the acidic conditions required
for the hydrolysis of the -CONH-, -OCONH-, and -COO- bonds,
but, after hydrolysis, an area proportional to the quantity of TOAC
can still be measured [1, Aib, 2.8, Xxx (TOAC), 2.2; 2, Glu, 2.0, Aib,
7.9, Xxx (TOAC), 2.1; 3, Glu, 1.9, Ala, 3.8, Aib, 4.2, Xxx (TOAC),
1
7
effect in the H NMR spectra and a serious overlapping of
optically active electronic transitions from the nitroxide and
2l
peptide chromophores in the CD curves. The intramolecular
inter-helix spin-spin interaction and distance were determined
in peptides 2 and 3 by conventional and half-field measurements,
respectively. Finally, the experimental average distance was
compared to that extracted from a molecular modeling inves-
tigation.
2
.2].
Typical coupling procedures used were the following:
Method I. Fmoc-Aib -TOAC-Aib -OtBu. To a stirred solution of
H-Aib-TOAC-Aib -OtBu (0.51 g, 1.12 mmol) in 5 mL of anhydrous
CH Cl was added the symmetrical anhydride of Fmoc-Aib-OH (0.98
g, 1.55 mmol), followed, after 30 min, by 0.085 mL (0.78 mmol) of
-methylmorpholine (NMM). After the reaction mixture was stirred
2
2
2
11
2
2
Materials and Methods
4
for 3 days, the solvent was removed under reduced pressure, the residue
was dissolved in EtOAc, and the organic layer was washed with 10%
KHSO , water, 5% NaHCO , and water, dried over Na SO , filtered,
4 3 2 4
and concentrated under reduced pressure. The peptide was purified by
flash chromatography on a silica gel column and eluted with a 96:4
Synthesis and Characterization of Peptides. Melting points were
determined using a Leitz (Wetzlar, Germany) model Laborlux 12
apparatus and are not corrected. Optical rotations were measured using
a Perkin-Elmer (Norwalk, CT) model 241 polarimeter equipped with
a Haake (Karlsruhe, Germany) model D thermostat. Thin-layer chro-
matography was performed on Merck (Darmstadt, Germany) Kieselgel
3
CHCl /EtOH mixture. Crystallization from EtOAc/PE afforded the
product in a 61% yield.
6
0F254 precoated plates using the following solvent systems: 1, CHCl -
3
n
n
EtOH, 9:1 (EtOH ) ethanol); 2, Bu OH-AcOH-H
)
2
O, 3:1:1 (Bu OH
(7) (a) Bolin, K. A.; Hanson, P.; Wright, S. J.; Millhauser, G. L. J. Magn.
n-butanol, AcOH ) acetic acid); 3, toluene-EtOH, 7:1; 4, EtOAc-
Reson. 1998, 131, 248-253. (b) Yu, L.; Meadows, R. P.; Wagner, R.; Fesik,
S. W. J. Magn. Reson. 1994, B 104, 77-80. (c) Kopple, K. D.; Zhu, P. P.
J. Am. Chem. Soc. 1983, 105, 7742-7746. (d) Lord, S. T.; Breslow, E.
Biochemistry 1980, 19, 5593-5602.
PE, 1:3 (EtOAc ) ethyl acetate, PE ) petroleum ether). The
chromatograms were examined by using ultraviolet (UV) fluorescence
or developed by chlorine-starch-potassium iodide or ninhydrin
(8) (a) Marchetto, R.; Schreier, S.; Nakaie, C. R. J. Am. Chem. Soc. 1993,
(
3) Toniolo, C.; Crisma, M.; Formaggio, F. Biopolymers (Pept. Sci.) 1998,
7, 153-158.
4) (a) Bonomo, R. P.; Conte, E.; Impellizzeri, G.; Pappalardo, G.;
Purrello, R.; Rizzarelli, E. J. Chem. Soc., Dalton Trans. 1996, 3093-3099.
b) Impellizzeri, G.; Pappalardo, G.; Rizzarelli, E.; Tringali, C. J. Chem.
115, 11042-11043. (b) Rozantsev, E. G. In Free Nitroxyl Radicals; Ulrich
4
H., Ed.; Plenum Press: New York, 1970; pp 93-115.
(
(9) (a) Carpino, L. A. J. Am. Chem. Soc. 1993, 115, 4397-4398. (b)
Carpino, L. A.; Sadat Aalaee, D.; Chao, H. G.; DeSelms, R. H. J. Am.
Chem. Soc. 1990, 112, 9651-9652.
(10) (a) Jones, D. S.; Kenner, G. W.; Preston, J.; Sheppard, R. C. J.
Chem. Soc. 1967, 6227-6239. (b) Crisma, M.; Valle, G.; Bianco, A.;
Formaggio, F.; Toniolo, C. Z. Kristallogr. 1996, 211, 561-562.
(11) (a) Meienhofer, J.; Waki, M.; Heimer, E. P.; Lambros, T. J.;
Makofske, R.; Chang, C. Int. J. Pept. Protein Res. 1979, 13, 35-42. (b)
Valle, G.; Bonora, G. M.; Toniolo, C. Can. J. Chem. 1984, 62, 2661-
2666.
(
Soc., Perkin Trans. 2 1996, 1435-1440. (c) Bergeron, J. R.; Phanstiel, O.,
IV; Yao, G. W.; Milstein, S.; Weimar, W. R. J. Am. Chem. Soc. 1994, 116,
8
479-8484. (d) Fusaoka, Y.; Ozeki, E.; Kimura, S.; Imanishi, Y. Int. J.
Pept. Protein Res. 1989, 34, 104-110.
(
(
5) Toniolo, C.; Benedetti, E. Trends Biochem. Sci. 1991, 16, 350-353.
6) (a) Karle, I. L.; Balaram, P. Biochemistry 1990, 29, 6747-6756. (b)
Toniolo, C.; Benedetti, E. Macromolecules 1991, 24, 4004-4009.

ESR Detection of Interchain Peptide Interaction
J. Am. Chem. Soc., Vol. 121, No. 48, 1999 11073
Table 1. Physical Properties of the Amino Acid Derivatives and Peptides
TLC
mass
melting
point (°C)
recryst.
solvent
[R]20
D
c
b
F4 (M + H)+ (M + Na)+
IRc
compound
Fmoc-TOAC-F
Fmoc-Aib)
dO
Z-Aib-OtBu
Fmoc-TOAC-Aib-OtBu
(deg)
R
F1
R
F2
R
F3
R
oil
0.25
3320, 1838, 1718, 1607, 1522
3395, 1810, 1714, 1513
3372, 1711, 1584, 1563
3398, 3313, 1718, 1657, 1521
3336, 1693, 1530
(
2
150-151 EtOAc/PE
63-64 Et O/PE
170-171 EtOAc/Et
136-137 EtOAc/PE
99-100 Et O/PE
0.95
0.95 0.95
0.95 0.95 0.50
0.95 0.95
0.65
2
2
O
602
d
2
Z-Aib -OtBu
e
Z-Ala-Aib-OtBu
2
-29.8 0.75 0.95 0.40
3308, 1736, 1713, 1689, 1676,
1
550, 1535
Fmoc-Aib-TOAC-Aib-OtBu 80-82
EtOAc/PE
0.85 0.95 0.45
0.75 0.95 0.35
666
612
688
650
3357, 1732, 1702, 1552
3442, 3372, 3333, 1731, 1707,
Fmoc-TOAC-Aib -OtBu 164-165 EtOAc/PE
2
1
689, 1521
Fmoc-TOAC-Ala-Aib-OtBu 98-100 EtOAc/PE
Fmoc-Aib -TOAC-Aib-OtBu 188-190 EtOAc
Fmoc-Aib-TOAC-Aib -OtBu 206-208 EtOAc/PE
-4.9 0.70 0.95 0.35
0.70 0.95 0.35
653
749
695
675
772
717
3338, 1728, 1709, 1526
3359, 1732, 1697, 1671, 1525
3356, 3330, 1734, 1704, 1683,
2
2
0.75 0.95 0.35
1
661, 1526
Fmoc-Ala-TOAC-Ala-Aib-
OtBu
Fmoc-TOAC-Aib -TOAC-
2
113-115 EtOAc/PE
214-215 EtOAc/
-27.9 0.60 0.95 0.35
0.75 0.95 0.30
723
949
746
971
3342, 1720, 1672, 1526
3342, 1729, 1698, 1662, 1525
3422, 3335, 1684, 1670, 1527
3339, 1731, 1667, 1526
3436, 3317, 1730, 1662
3324, 1716, 1664, 1528
3323, 1729, 1657, 1537
Aib-OtBu
2 2
CH Cl /PE
2
Fmoc-Aib -TOAC-Aib
2
-
226-228 EtOAc/PE
0.60 0.95 0.30
776
OtBu
Fmoc-Aib-Ala-TOAC-Ala-
128-129 EtOAc/PE
-2.8 0.60 0.95 0.30
0.30 0.85 0.05
809
830
790
Aib-OtBu
Ac-TOAC-Aib
OtBu (1)
2
-TOAC-Aib- 273-275 CH
2
Cl
2
/PE
768
cyclo-[Glu(Aib
2
-TOAC-
(2)
176-178 EtOAc/PE
+14.5 0.10 0.80 0.05
+29.5f 0.25 0.80 0.00
1448
1393
1470
1415
Aib -OtBu)]
2
2
2
cyclo-[Glu(Aib-Ala-TOAC- 159-161 MeOH/H O
Ala-Aib-OtBu)] (3)
2
a
O ) diethyl ether, MeOH ) methanol. c ) 0.5, methanol. c The solid-state IR absorption
b
EtOAc ) ethyl acetate, PE ) petroleum ether, Et
2
spectra were obtained in KBr pellets (only bands in the 3450-3300 and 1850-1500 cm regions are reported). d Reference 10a. e Reference 10b.
c ) 0.25, methanol.
-1
f
Method II. Fmoc-TOAC-Aib
of Fmoc-TOAC-OH
2
-TOAC-Aib-OtBu. To a suspension
(0.18 g, 0.41 mmol) and HOAt (0.058 g, 0.42
mmol) in 3 mL of anhydrous CH Cl at 0 °C was added 1-(3-
dimethylamino)propyl)-3-ethylcarbodiimide hydrochloride (EDC‚HCl)
0.1 g, 0.52 mmol). When the suspension became clear, H-Aib -TOAC-
3
CHCl /EtOH mixture. Crystallization from EtOAc/PE afforded the
product in a 61% yield.
2
l,8a
2
2
FTIR Absorption. FTIR absorption spectra were recorded with a
Perkin-Elmer model 1720 X spectrophotometer, nitrogen flushed and
(
(
-
1
2
equipped with a sample-shuttle device, at 2 cm nominal resolution,
averaging 100 scans. Solvent (baseline) spectra were obtained under
the same conditions. Cells with path lengths of 0.1, 1.0, and 10 mm
Aib-OtBu (0.18 g, 0.35 mmol), dissolved in 3 mL of anhydrous CH
Cl , was added, followed, after 30 min, by 0.046 mL (0.41 mmol) of
2
2
-
NMM. The reaction mixture was allowed to stir for 3 days at room
temperature. The solvent was removed under reduced pressure, the
residue was dissolved in EtOAc, and the organic layer was washed
2
(with CaF windows) were used. Spectrograde deuteriochloroform
(99.8% D) was purchased from Fluka.
X-ray Diffraction. Pale yellow crystals of the linear pentapeptide
with 10% KHSO
4
, water, 5% NaHCO
3
, and water, dried over Na
2 4
SO ,
1 were grown from a MeOH/Et O solution by slow evaporation. A
2
filtered, and concentrated under reduced pressure. The peptide was
purified by flash chromatography on a silica gel column and eluted
single crystal of approximate dimensions 0.4 × 0.4 × 0.1 mm was
mounted on the tip of a glass capillary. Cell parameters were determined
from 48 well-centered reflections in the 11-22° θ range. Data collection
was performed by using a Philips PW1100 four-circle diffractometer
and graphite-monochromated Cu KR radiation (λ ) 1.541 84 Å), θ-2θ
scan mode up to 2θ ) 100°, h from -10 to 10, k from -18 to 20, l
from 0 to 24. The crystal did not significantly diffract above 2θ )
100° (1.0 Å resolution).
with a 95:5 CHCl
Cl /PE afforded the product in a 49% yield.
Method III. Fmoc-TOAC-Aib-OtBu. To a stirred solution of Fmoc-
TOAC-F (2.42 g, 5.53 mmol) in 10 mL of anhydrous CH Cl at 0 °C
3 2
/EtOH mixture. Crystallization from EtOAc/CH -
2
2
2
was added H-Aib-OtBu (1.46 g, 9.2 mmol) [obtained from Pd-catalyzed
hydrogenation in methanol (MeOH) of the corresponding Z-protected
10a
The structure was solved by direct methods (SHELXS 86 program).12a
The asymmetric unit turned out to be made of two independent peptide
molecules (A and B) and four cocrystallized water molecules. Refine-
amino acid ester (2.70 g, 9.2 mmol)], dissolved in 3 mL of anhydrous
CH Cl , and the solution was allowed to stir overnight at room
2
2
temperature. The solvent was removed under reduced pressure, the
residue was dissolved in EtOAc, and the organic layer was washed
2
ment was carried out on F , using the full data set and the SHELXL
12b
with 10% KHSO
4
, water, 5% NaHCO
3
, and water, dried over Na
2 4
SO ,
97 program. In both independent molecules, the C-terminal -OtBu
group is disordered, and it was refined on two sets of positions, with
population parameters of 0.60 and 0.40 for the major and minor
conformers, respectively, in molecule A, and 0.66 and 0.34, respectively,
in molecule B. For molecule B only, the disorder could be traced to
include also the CdO group of Aib(15). The refinement was carried
out by full-matrix block least-squares, with all non-hydrogen atoms
anisotropic, and allowing the positional parameters and the anisotropic
filtered, and concentrated under reduced pressure. Crystallization from
EtOAc/PE afforded the product in a 65% yield.
Method IV. Cyclo-[Glu(Aib
tion of cyclo-(Glu-Glu) (0.021 g, 0.08 mmol) and HOAt (0.028 g,
2
-TOAC-Aib
2 2
-OtBu)] (2). To a solu-
4
0
0
.21 mmol) in 2 mL of DMF at 0 °C was added EDC‚HCl (0.044 g,
.23 mmol). Then, H-Aib -TOAC-Aib -OtBu (0.11 g, 0.19 mmol),
2
2
dissolved in 2 mL of DMF, was added, followed, after 30 min, by
.018 mL (0.16 mmol) of NMM. After the solution was stirred at room
0
(
12) (a) Sheldrick, G. M. SHELXLS 86. Program for the Solution of
Crystal Structures; University of G o¨ ttingen: G o¨ ttingen, Germany, 1986.
b) Sheldrick, G. M. SHELXL 97. Program for the Refinement of Crystal
temperature overnight, the solvent was removed in vacuo, the residue
was dissolved in EtOAc, and the organic layer was washed with 10%
KHSO , water, 5% NaHCO , and water, dried over Na SO , filtered,
4 3 2 4
and concentrated under reduced pressure. The peptide was purified by
flash chromatography on a silica gel column and eluted with a 95:5
(
Structures; University of G o¨ ttingen: G o¨ ttingen, Germany, 1997. (c) Nardelli,
M. J. Appl. Crystallogr. 1995, 28, 659. (d) Spek, A. L. Acta Crystallogr.
1990, A46, C-34, MS-02.01.05.

11074 J. Am. Chem. Soc., Vol. 121, No. 48, 1999
Polese et al.
Table 2. Crystal Data for Pentapeptide 1 Dihydrate
empirical formula
mol wt
temp (K)
crystal system
space group
a (Å)
C
802.0
293(2)
triclinic
38
H
67
N
7
O
9
2
‚2H O
P1h
10.435(2)
20.878(3)
24.207(3)
68.4(1)
89.1(1)
82.8(1)
4862.1(13)
4
b (Å)
c (Å)
R (deg)
â (deg)
γ (deg)
3
V (Å )
Z
molecules/asymmetric unit
2
3
density (g/cm ) (calcd)
1.096
0.660
-
1
abs coeff (mm
)
F(000)
no. of collected reflcns
1744
9347
Figure 2. Normalized FTIR absorption spectra (N-H stretching
no. of independent reflcns
no. of reflcns with I g 2σ(I)
9311
3950
0.0834
0.2472
0.0264
region) of Ac-TOAC-Aib
Aib -TOAC-Aib -OtBu)]
solution, respectively.
2
-TOAC-Aib-OtBu (1) (I) and cyclo-[Glu-
2 (II) in 10 (A), 1 (B), 0.1 mM (C) CDCl
3
(
2
2
2
R
1
[on F g 4σ(F)]
2
2
wR [on F , all data]
R
int
no. of data/restraints/parameters
goodness of fit
9311/268/1070
0.896
0.712
were prepared in a glassing solvent mixture consisting of 9:1 (v/v)
MeOH/EtOH. Samples for room-temperature measurements (40 µL)
were sealed into glass capillaries (100 µL). Samples for half-field
measurements were prepared using 300 µL of this mixed solvent
solution placed in 4-mm-i.d. quartz sample tubes and quickly frozen
in liquid nitrogen to form a glass. The samples were directly and
immediately placed in the cooler spectrometer cavity for spectral
acquisition. Spectrograde MeOH and EtOH were obtained from Fischer,
2,2,2-trifluoroethanol (TFE) from Aldrich, and 1,1,1,3,3,3-hexafluoro-
2-propanol (HFIP) from Sigma.
Molecular Modeling. Computer models were built using MOL-
MOL14 and a library modified to incorporate the TOAC and Aib
residues. Calculations required to appropriately transform the coordi-
nates for inclusion and inspection of the models were aided by use of
MOLMOL on a Silicon Graphics workstation.
max shifts/esd’s
max/min ∆F (e/Å )
3
+0.424/-0.307
displacement parameters of the non-hydrogen atoms to refine at alternate
cycles. Restraints were applied to the 1-2 and 1-3 interatomic
distances involving atoms of the disordered groups, as well as to their
anisotropic displacement parameters. Hydrogen atoms of the peptide
molecules were calculated at idealized positions, and during the
refinement they were allowed to ride on their parent atom, with Uiso
set equal to 1.2 (or 1.5 for methyl groups) times the Ueq of their carrying
atom. The positions of the hydrogen atoms bound to the water molecules
were recovered from a ∆F map, and they were not refined. Other
relevant crystallographic data are listed in Table 2. Geometrical
1
2c
12d
calculations were performed with the PARST and PLATON
programs.
Results and Discussion
Electron Spin Resonance. Continuous wave spectra of the peptides
were recorded using a Bruker ESR-380 X-band spectrometer equipped
with a TE102 cavity and a ER4111 variable-temperature unit. The
temperatures at which this unit regulates were checked against a Cernox
ceramic thermocouple purchased from and calibrated by Lakeshore
Cryonics. Allowed transitions at 300 and at 118 K were recorded with
a center field of 3357 G, typically with a sweep width of 150 G, a
modulation frequency of 100 kHz, a modulation amplitude of 1 G, a
time constant of 82 ms, a conversion time of 164 ms, and a receiver
FTIR Absorption. The conformational preferences of the
substituted linear and cyclic templates 1-3 were investigated
in CDCl3 using FTIR absorption. The most significant results
are illustrated in Figure 2 and listed in Table 3.
The three spectra are characterized by weak bands in the
-
1
3457-3430 cm region (free N-H groups) and more intense
-1
bands (shoulders) at 3370-3324 cm
(H-bonded N-H
15a,b
groups).
In particular, the more complex absorption of 2
4
gain of 1.00 × 10 , at 2.7 mW, with four scans acquired for each
and 3 can be rationalized as arising from an overlapping of the
spectrum. Half-field spectra were recorded at 118 K at a center field
of 1677 G, typically with a sweep width of 200 G, a modulation
frequency of 100 kHz, a modulation amplitude of 5 G, a time constant
-1
amide N-H stretching modes of cyclo-(Glu-Glu) at 3380 cm
and the pentapeptides anchored to the cyclic skeleton at 3350
-
1
cm (data not shown). At any event, in all of the compounds,
of 328 ms, a conversion time of 328 ms, and a receiver gain of 2.00 ×
4
even at the highest dilution examined, the intensity of the 3347-
1
0 , at 84 mW, with up to 12 scans acquired for each spectrum.
-
1
Integrated intensities for both spectral regions were determined via
double integration of the recorded derivative-mode spectra. For the half-
field signals, a third-order polynomial baseline correction was deter-
mined using 35 G of baseline at both edges of the absorption mode
spectrum and subtracting this polynomial from the entire spectrum prior
to the second integration. Integrated intensities of the half-field
transitions were normalized by dividing by the number of scans
acquired, the solution concentration, and the square-root of the
3324 cm band is remarkable, thereby suggesting the occur-
rence of a large population of highly intramolecularly H-bonded
1
5a
species. This finding indicates that pentapeptide 1 and the
peptide segments covalently bound to the DKP rings are
arranged in a well-ordered conformation. Furthemore, for all
peptides the analysis of the CdO stretching region reveals a
-
1
strong band at 1664 ( 2 cm , typical of H-bonded amide
carbonyls in a 310-helica1 conformation, in addition to weaker
spectrometer power. Relative intensity values for use in calculations
-1
according to the procedure reported by Eaton and co-workers13 were
bands (shoulders) at 1729-1728 and 1679 ( 3 cm , assigned
obtained from allowed transitions recorded at 118 K and from half-
field transitions at the same temperature. The ratio of the doubly
integrated intensities was taken after normalizing for receiver gain,
number of scans, and square root of the microwave power. Samples
(
14) Koradi, R.; Billeter, M.; W u¨ trich, K. J. Mol. Graphics 1996, 14,
1-55.
(15) (a) Mizushima, S.; Shimanouchi, T.; Tsuboi, M.; Souda, R. J. Am.
5
Chem. Soc. 1952, 74, 270-271. (b) Palumbo, M.; Da Rin, S.; Bonora, G.
M.; Toniolo, C. Makromol. Chem. 1976, 177, 1477-1492. (c) Kennedy,
D. F.; Crisma, M.; Toniolo, C.; Chapman, D. Biochemistry 1991, 30, 6541-
6548.
(13) Eaton, S. S.; More, K. M.; Sawant B. M.; Eaton, G. R. J. Am. Chem.
Soc. 1983, 105, 6560-6567.

ESR Detection of Interchain Peptide Interaction
J. Am. Chem. Soc., Vol. 121, No. 48, 1999 11075
Table 3. Infrared Absorption Data for Peptides 1-3 in CDCl a
3
compound
NsH stretching regionb
CdO stretching regionb
1729, 1682, 1662
Ac-TOAC-Aib
cyclo-[Glu(Aib
cyclo-[Glu(Aib-Ala-TOAC-Ala-Aib-OtBu)]
2
-TOAC-Aib-OtBu (1)
3457, 3430, 3347
c
c
2
-TOAC-Aib -OtBu)]
2
2
(2)
3435, 3370, 3329
1728, 1676, 1666
c
c
2
(3)
3434, 3364, 3324
1728, 1677, 1662
a
Concentration, 1.0 mM. b Values shown in italic type represent weak bands; values shown in boldface type represent strong bands. c Shoulder.
Table 4. Relevant Backbone Torsion Angles (deg) for Molecules
A and B of Pentapeptide 1
angle
molecule A
177.3(7)
55.7(9)
29.1(8)
178.9(6)
55.4(8)
moleculeB
-175.3(7)
ω
1
0
φ
-54.2(9)
-31.9(8)
-177.6(6)
-57.3(8)
-23.1(8)
-179.9(5)
-51.9(8)
-28.8(8)
-178.6(6)
-53.4(8)
-38.7(9)
179.0(7)
ψ
ω
1
1
φ
2
ψ
ω
2
22.5(8)
2
-177.4(5)
φ
3
47.8(7)
ψ
ω
3
32.5(8)
176.2(6)
57.1(8)
35.8(8)
3
φ
4
ψ
ω
4
4
-177.8(6)
-58.7(10) a
133.0(19), 146.0(13)
a
b
b
φ
5
47.2(13), 74.0(14)
-145.0(15), -121(2)
a b
b
b
a
ψ
ω
5
a
5
177.0(12), 175.8(19)
-171.1(14), 168(2)
a
Major conformer. b Minor conformer.
Bond lengths and bond angles (deposited) are in general
17
agreement with those reported for the peptide unit, the OtBu
1
8
19
2i,l,20
ester, and the Aib and TOAC
residues.
Molecule A and molecule B are left-handed and right-handed
3
10-helices, respectively, both stabilized by three consecutive
2
1
and strong intramolecular H-bonds between the CdO group
of residue i and the N-H group of residue i + 3, typical of
â-turns. For both molecules, the first intramolecular H-bond,
starting from the N-terminus, involves the carbonyl oxygen atom
of the acetyl moiety and the peptide nitrogen atom of Aib(3)
2
2
(
O0A‚‚‚N3 for molecule A, O0B‚‚N13 for molecule B). The
Figure 3. X-ray diffraction structure of the two independent molecules
absolute values of the φ,æ backbone torsion angles, as averaged
for the four helical residues of both molecules A and B, are
5
2
(A and B) in the asymmetric unit of Ac-TOAC-Aib -TOAC-Aib-OtBu
(1) with atom numbering. The intramolecular H-bonds are indicated
4.1°,30.3°, close to those typical for a peptide 310-helix (57°,
by dashed lines.
5
3
0°). In both molecules A and B, the C-terminal -OtBu group
is disordered over two conformations. In molecule B, the
disorder also includes the CdO group of the C-terminal Aib
residue. In all of the four rotamers, the C-terminal Aib residue
adopts a semi-extended conformation (regions F/F* of the
conformational space), with the sign of the φ torsion angle
opposite to that of the preceding residues. In the two molecules,
to the tert-butyl ester carbonyl and free amide carbonyls,
respectively.15c
In summary, the present FTIR absorption investigation has
provided convincing evidence that the linear pentapeptide 1,
and the peptide segments anchored to the side chains of the
cyclo-(Glu-Glu) template adopt a helical conformation in a
solvent of low polarity.
X-ray Diffraction. The molecular and crystal structures of
the terminally blocked pentapeptide Ac-TOAC-Aib2-TOAC-
Aib-OtBu (1) were determined in the crystal state by X-ray
diffraction. A view perpendicular to the helix axis of the two
independent molecules (A and B) in the asymmmetric unit is
illustrated in Figure 3. Amino acid residues are numbered from
2
3
(
17) (a) Benedetti, E. In Chemistry and Biochemistry of Amino Acids,
Peptides and Proteins; Weinstein, B., Ed.; Dekker: New York, 1982; pp
05-184. (b) Ashida, T.; Tsunogae, Y.; Tanaka, I.; Yamane, T. Acta
Crystallogr. 1987, B43, 212-218.
18) Schweizer, W. B.; Dunitz, J. D. HelV. Chim. Acta 1982, 65, 1547-
1554.
1
(
(19) (a) Paterson, Y.; Rumsey, S. M.; Benedetti, E.; N e´ methy, G.;
Scheraga, H. A. J. Am. Chem. Soc. 1981, 103, 2947-2955. (b) Valle, G.;
Crisma, M.; Formaggio, F.; Toniolo, C.; Jung, G. Liebigs Ann. Chem. 1987,
1055-1060.
1
to 5 in molecule A and from 11 to 15 in molecule B. For
(20) Flippen-Anderson, J. L.; George, C.; Valle, G., Valente, E.; Bianco,
clarity, for each molecule, only the major conformer of the
A.; Formaggio, F.; Crisma, M.; Toniolo, C. J. Pept. Protein Res. 1996, 47,
disordered C-terminal -OtBu group is shown. Relevant back-
2
31-238.
16
bone torsion angles are given in Table 4. In Table 5, the intra-
and intermolecular H-bond parameters are listed. Owing to the
space group symmetry, in the crystal each molecule has a
centrosymmetric counterpart of opposite handedness.
(21) (a) Ramakrishnan, C.; Prasad, N. Int. J. Protein Res. 1971, 3, 209-
231. (b) Taylor, R.; Kennard, O.; Versichel, W. Acta Crystallogr. 1984,
B40, 280-288. (c) G o¨ rbitz, C. H. Acta Crystallogr. 1989, B45, 390-395.
(22) (a) Venkatachalam, C. M. Biopolymers 1968, 6, 1425-1436. (b)
Toniolo, C. CRC Crit. ReV. Biochem. 1980, 9, 1-44. (c) Rose, G. D.;
Gierasch, L.; Smith, J. P. AdV. Protein Chem. 1985, 37, 1-109.
(23) Zimmerman, S. S.; Pottle, M. S.; N e´ methy, G.; Scheraga, H. A.
Macromolecules 1977, 10, 1-9.
(
16) IUPAC-IUB Commission on Biochemical Nomenclature. Bio-
chemistry 1970, 9, 3471-3479.

11076 J. Am. Chem. Soc., Vol. 121, No. 48, 1999
Polese et al.
Table 5. Intra- and Intermolecular H-Bond Parameters for Pentapeptide 1 (Molecules A and B)
distance (Å)
D‚‚‚A
symmetry
operations of A
angle (deg)
DsH‚‚‚A
donor D-H
acceptor A
H‚‚‚A
Intramolecular
molecule A
molecule B
N3-H
N4-H
N5-H
N13-H
N14-H
N15-H
O0A
O1
O2
O0B
O11
O12
x, y, z
x, y, z
x, y, z
x, y, z
x, y, z
x, y, z
2.972(7)
3.022(7)
2.922(8)
2.959(8)
3.004(7)
2.961(8)
2.13
2.17
2.18
2.13
2.15
2.21
164.4
171.2
143.7
161.7
169.2
146.2
Intermolecular
x, y, z
peptide-water
water-peptide
N1-H
O1W
O2W
O3W
O4W
O13
O14D
O14
O1D
O3
O4D
O4
2.981(7)
2.988(8)
2.938(7)
2.979(8)
2.782(7)
2.922(7)
2.800(8)
2.831(8)
2.777(7)
2.880(7)
2.806(8)
2.856(9)
2.14
2.16
2.09
2.15
1.96
2.29
2.00
2.03
2.06
2.07
1.98
2.36
164.6
162.6
168.0
161.4
173.3
133.6
159.0
163.2
144.2
162.9
167.8
119.1
N2-H
x, y, z
x, y, z
x, y, z
x, y, z - 1
x + 1, y, z - 1
x, y, z - 1
x -1, y, z
x, y, z
x -1, y, z
x, y, z
N11-H
N12-H
O1W-H1A
O1W-H1B
O2W-H2A
O2W-H2B
O3W-H3A
O3W-H3B
O4W-H4A
O4W-H4B
O11D
x + 1, y, z
a significant deviation of the amide, peptide, and ester ω torsion
angles (|∆ω| > 5°) from the ideal trans planar conformation
TOAC(11) while intermediate between the B2,5 boat and the
T2 twist-boat dispositions for TOAC(14). It should be noted
6
(
180°) is observed only for the ester ω5 torsion angles of both
that, in this classification, we considered the ring atom sequence
C -C -C -N -C -C , where C refers to the pro-S Câ
R
â1
γ1
δ
γ2
â2
â1
conformers of molecule B. For three out of the four C-terminal
conformers, the tert-butyl ester disposition with respect to the
atom. The ring-puckering parameters are the following: Q )
T
R
preceding C -N bond is intermediate between the synperiplanar
0.648(7) Å, φ ) 261.9(5)°, θ ) 89.4(6)° for TOAC(1) and
2
2
and synclinal conformations, while it is synperiplanar for the
minor conformer of molecule B.24 For each independent
molecule A and B, the main difference between the major and
the minor conformers is found in the disposition of the
C-terminal tert-butyl group with respect to the preceding C′-O
bond. This disposition is fully staggered in the major conformer,
while in the minor conformer one of the C-CH3 bonds is
eclipsed (synperiplanar).
Q ) 0.619(6) Å, φ ) 287.6(5)°, θ ) 97.6(5)° for TOAC(4)
T
2
2
of molecule A; QT ) 0.622(7) Å, φ2 ) 102.3(6)°, θ2 ) 88.2-
6)° for TOAC(11) and QT ) 0.635(6) Å, φ2 ) 74.6(6)°, θ2 )
(
25
9
6.5(6)° for TOAC(14) of molecule B.
The endocyclic torsion angles about the C -C bonds exhibit
values in the range (40-(57°, close to those of cyclohexane
â
γ
2
6a
26b
(
(54°-(55°) and piperidine ((48°-(52°). All of the
other endocyclic torsion angles are found in the range (1°-
46° (in each ring, the lowest value is observed for one of the
In each molecule, the piperidinyl rings of the two TOAC
residues, separated by one complete turn of the 310-helix, are
oriented roughly perpendicular to the helix axis and nearly
parallel to each other, the angle between normals to the average
ring plane being 12.8(2)° and 13.8(2)° in molecules A and B,
(
γ
δ
torsion angles about the C -N bonds). Such an observation is
common to other six-membered nitroxide derivatives2
i,l,20
and
δ
indicates that the piperidine rings are most flattened in the N
part.
•
respectively. The angle between the two N-O groups is
The angles between the normal to the average ring plane and
3
9.3(4)° in molecule A and 43.5(5)° in molecule B. The distance
R
R
•
the C -N and C -C′ bonds are in the range 30°-43°,
indicating that both the R-amino and the R-carbonyl ring
between midpoints of the two N-O bonds is 7.376(7) and
7.388(7) Å in molecules A and B, respectively. The latter values
27
substituents occupy a bisectional position, as usually observed
are about 0.06 Å larger than the corresponding distance found
in the X-ray diffraction structure of the 310-helical peptide
for TOAC residues in linear derivatives and peptides.2
i,l,20
pBrBz-TOAC-(L-Ala)2-TOAC-L-Ala-NHtBu (pBrBz ) p-bro-
In the packing mode of the Ac-TOAC-Aib2-TOAC-Aib-OtBu
(1) molecules, there are no direct peptide-peptide intermolecular
H-bonds. Rather, a complex network of H-bonds is observed
mobenzoyl; NHtBu ) tert-butylamino).2
l,20
In each peptide
•
molecule of the present structure, the two N-O bonds diverge
more than in the structure mentioned above,2 owing to their
involvement in H-bonds with the cocrystallized water molecules.
It is reasonable to assume that such an effect induced on the
l,20
28
involving the four cocrystallized water molecules. In addition,
in both peptide molecules A and B, the NH groups of TOAC-
(1)/TOAC(11) and Aib(2)/Aib(12), the carbonyl groups of Aib-
(3)/Aib(13) and TOAC(4)/TOAC(14), and the nitroxide oxygen
atoms of TOAC(1)/TOAC(11) and TOAC(4)/TOAC(14) act as
intermolecular H-bonding donors or acceptors.
•
N-O groups by the water molecules may be relaxed in solution.
Interestingly, in the left-handed helical molecule A, the
piperidinyl ring of TOAC(1) is found in a conformation close
2
to the T6 twist-boat disposition, while the ring conformation
2
of TOAC(4) is intermediate between the B3,6 boat and the T6
(25) Cremer, D.; Pople, J. A. J. Am. Chem. Soc. 1975, 97, 1354-1358.
(
26) (a) Bixon, M.; Lifson, S. Tetrahedron 1967, 23, 769-784. (b)
twist-boat dispositions. Conversely, the ring conformations
Cygler, M.; Markovicz, T.; Skolimowski, J.; Skowronski, R. J. Mol. Struct.
observed in the right-handed helical molecule B are intermediate
1
980, 68, 161-171.
between the 3 B boat and the T2 twist-boat dispositions for
,6
6
(27) Luger, P.; B u¨ low, R. J. Appl. Crystallogr. 1983, 16, 431-432.
(28) (a) Yang, C. Y.; Brown, J. N.; Kopple, K. D. Int. J. Pept. Protein
(
24) Dunitz, J. H.; Strickler, P. In Structural Chemistry and Molecular
Biology; Rich, A., Davidson, N., Eds.; Freeman: San Francisco, 1968; pp
95-602.
Res. 1979, 14, 12-20. (b) Jeffrey, G. A.; Maluszynska, H. Acta Crystallogr.
1990, B46, 546-549. (c) G o¨ rbitz, C. H.; Etter, M. Int. J. Pept. Protein
Res. 1992, 39, 93-110.
5

ESR Detection of Interchain Peptide Interaction
J. Am. Chem. Soc., Vol. 121, No. 48, 1999 11077
2
Figure 4. ESR spectra (150 G scan width) of Ac-TOAC-Aib -TOAC-
Aib-OtBu (1), cyclo-[Glu(Aib
2
-TOAC-Aib
2 2
-OtBu)] (2), and cyclo-
[
Glu(Aib-Ala-TOAC-Ala-Aib-OtBu)]
2
(3) in four alcohols at 300 K.
Electron Spin Resonance. The solvent-dependent unfolding
of 1-3 was investigated at room temperature using ESR
spectroscopy. Spectra taken in each of four solvents are shown
from left to right in Figure 4 in order of each solvent’s
demonstrated ability to support helical structures in short
2k
peptides (MeOH > EtOH > TFE > HFIP). Specifically,
alcohols with greater polarity and H-bonding capability as
donors (e.g., HFIP) are known to destabilize helical structures
in these compounds. The doubly labeled pentapeptide, 1, shows
the largest solvent-dependent spectroscopic change, exhibiting
a strong, five-line J-coupling pattern in MeOH which reveals
the close inter-nitroxide spatial proximity expected for a highly
helical conformation. This coupling gradually decreases through
the solvent series until, in HFIP, no evidence of the five-line
pattern can be seen. Similar J-coupling, although weaker, is
observed for 2 in MeOH that is quickly lost in the other solvents,
while 3 shows little evidence of such coupling. More tellingly,
though, is a comparison among the three compounds in the most
destructuring solvent (HFIP). In this solvent, the ESR spectra
lines for 2 and 3 are both broader than those of 1. This uniform
broadening of all three nitroxide hyperfine lines indicates that
the total contribution of inter-nitroxide coupling (either dipolar
or exchange coupling) is stronger for 2 and 3 than for 1 (we
note that longer correlation times, as might be expected for 2
and 3 if more motionally restricted, would preferentially broaden
only the high-field hyperfine line). This increased spectral
broadening is consistent with a shorter inter-nitroxide distance
for both 2 and 3 than for 1.
Figure 5. Plots of the interspin distance vs peptide concentration for
2 2 2
cyclo-[Glu(Aib -TOAC-Aib -OtBu)] (2) (A) and cyclo-[Glu(Aib-Ala-
TOAC-Ala-Aib-OtBu)]
2
(3) (B), determined by half-field measure-
ments. In both parts A and B, 2 and (---) correspond to the interspin
distance (relative intensity), and b and (s) to the interspin distance
(normalized intensity). The lines through the data points correspond to
linear regression (least-squares) fits.
Considering a limiting case constrained only by covalent
connectivity, a completely extended backbone conformation for
1
would put the inter-nitroxide spacing r at approximately 11-
1
2 Å. This represents an extreme upper bound for the average
inter-nitroxide distance in 2 and 3 in HFIP. Local steric
interactions restrict how closely 1 can approach such an extended
conformation, so the distance is probably significantly less than
this upper bound in all three compounds.
A more quantitative measure of the average inter-nitroxide
distance results from measurements of the half-field signal
1
3,29
intensity.
The integrated intensity of this signal is propor-
-
6
tional to r , and it can arise from intramolecular as well as
intermolecular dipolar coupling. Extrapolation of signal intensi-
ties to infinite dilution reveals only the intramolecular coupling,
Figure 6. Molecular model of cyclo-[Glu(Aib
2 2 2
-TOAC-Aib -OtBu)]
(2), with Aib methyl groups omitted for clarity.
(
29) Anderson, D. J.; Hanson, P.; McNulty, J.; Millhauser, G.; Monaco,
V.; Formaggio, F.; Crisma, M.; Toniolo, C. J. Am. Chem. Soc. 1999, 121,
919-6927.
with the intermolecular contribution manifest as a negative slope
for the extrapolation. Figure 5 shows this extrapolation for 2
6

11078 J. Am. Chem. Soc., Vol. 121, No. 48, 1999
Polese et al.
and 3 in a helix-supporting 9:1 MeOH/EtOH solvent glass at
restricted template is the cyclo-homopeptide cyclo-(Glu-Glu)
containing two side-chain carboxyl functionalities, to which two
copies of the same 3 -helical pentapeptide tails are tethered
1
20 K, and similar results for a series of 310-helical peptides
29
analogous to 1 have recently been published. Two independent
calibrations for the intensity-to-distance conversion have been
employed: an instrument-independent calibration using a
1
0
through amide covalent bonds. Each pentapeptide is character-
ized by one stable free-radical-containing TOAC residue. By
exploiting these two molecular bundles, for the first time we
detected by ESR in a fully synthetic construct an inter-helix
interaction between spins located far away in the amino acid
sequence. This phenomenon is reduced by an increase of the
solvent polarity that is responsible for a destabilization of the
peptide helical structures. The experimentally determined in-
terspin distance matches well that obtained from a molecular
modeling investigation.
13
formula developed by Eaton and co-workers, and another using
measurements performed on doubly labeled TOAC-containing
peptides of known structure studied in our laboratory. The lack
of significant negative slope in any of the extrapolated lines in
Figure 5 suggests little intermolecular contribution to the half-
field signal for these samples over the concentration range
investigated. Distances determined from both calibrations agree
within experimental error. The average for 2 is 8.1 Å, while
for 3 it is slightly longer (8.7 Å). The half-field results, taken
together with evidence of both significant J-coupling in 2 and
persistent biradical broadening for both 2 and 3, indicate that
these chains do approach one another.
The results presented in this study strongly support our
contention that the semi-rigid, helicogenic TOAC residue is a
valuable tool for the detection of interactions between an
appropriately tailored, TOAC-based, synthetic peptide inhibitor
or hormone analogue and its side-chain spin-labeled, site-
directed Cys mutant enzyme or receptor counterpart.
Molecular Modeling. Figure 6 illustrates how this approach
might be arranged. The model was constructed with each helix
5
held in 310-helical conformation (φ ) -57°, ψ ) -30°) and
with the cyclo-(Glu-Glu) side-chain torsion angles (ø1 ) 30°,
ø2 ) 180°, ø3 ) -150°) arranged such that the inter-nitroxide
distance (average of the N‚‚‚N and O‚‚‚O distances) is com-
parable to that indicated by half-field measurements (8.4 Å).
No significant steric helix-helix contacts exist in this model,
allowing for the possibility that molecules with helix-helix
contact and shorter inter-nitroxide distances represent a signifi-
cant population contributing to the ensemble average.
Acknowledgment. D.J.A. and G.L.M. were supported by a
grant from the National Institutes of Health (GM 46870).
Supporting Information Available: Tables of crystal data
and structure refinement, atomic coordinates, isotropic and
anisotropic displacement parameters, bond lengths and angles,
hydrogen coordinates, torsion angles, relevant backbone and
side-chain torsion angles, and intra- and intermolecular para-
menters for 1 (PDF). X-ray crystallographic data, in CIF format,
are also available. This material is available free of charge via
the Internet at http://pubs.acs.org.
Conclusions
In this paper, we reported the design, chemical synthesis, and
conformational analysis of two simple two-helix model systems
fully based on peptide building blocks. The conformationally
JA992079H