Solution and solid-state models of peptide CH...O hydrogen bonds

Article abstract of DOI:10.1021/ja0257366

Fumaramide derivatives were analyzed in solution by ¹H NMR spectroscopy and in the solid state by X-ray crystallography in order to characterize the formation of CH...O interactions under each condition and to thereby serve as models for these interactions in peptide and protein structure. Solutions of fumaramides at 10 mM in CDCl₃ were titrated with DMSO-d₆, resulting in chemical shifts that moved downfield for the CH groups thought to participate in CH...O=S(CD₃)₂ hydrogen bonds concurrent with NH...O=S(CD₃)₂ hydrogen bonding. In this model, nonparticipating CH groups under the same conditions showed no significant change in chemical shifts between 0.0 and 1.0 M DMSO-d₆ and then moved upfield at higher DMSO-d₆ concentrations. At concentrations above 1.0 M DMSO-d₆, the directed CH...O=S(CD₃)₂ hydrogen bonds provide protection from random DMSO-d₆ contact and prevent the chemical shifts for participating CH groups from moving upfield beyond the original value observed in CDCl₃. X-ray crystal structures identified CH...O=C hydrogen bonds alongside intermolecular NH...O=C hydrogen bonding, a result that supports the solution ¹H NMR spectroscopy results. The solution and solid-state data therefore both provide evidence for the presence of CH...O hydrogen bonds formed concurrent with NH...O hydrogen bonding in these structures. The CH...O=C hydrogen bonds in the X-ray crystal structures are similar to those described for antiparallel β-sheet structure observed in protein X-ray crystal structures.

Full text of DOI:10.1021/ja0257366

Published on Web 08/31/2002
Solution and Solid-State Models of Peptide CH‚‚‚O Hydrogen
Bonds
Paul W. Baures,* Alicia M. Beatty,† Muthu Dhanasekaran,‡ Brian A. Helfrich,
Waleska Pe´rez-Segarra, and John Desper
Contribution from the Department of Chemistry, 111 Willard Hall, Kansas State UniVersity,
Manhattan, Kansas 66506
Received January 28, 2002
Abstract: Fumaramide derivatives were analyzed in solution by 1H NMR spectroscopy and in the solid
state by X-ray crystallography in order to characterize the formation of CH‚‚‚O interactions under each
condition and to thereby serve as models for these interactions in peptide and protein structure. Solutions
of fumaramides at 10 mM in CDCl3 were titrated with DMSO-d6, resulting in chemical shifts that moved
downfield for the CH groups thought to participate in CH‚‚‚OdS(CD3)2 hydrogen bonds concurrent with
NH‚‚‚OdS(CD3)2 hydrogen bonding. In this model, nonparticipating CH groups under the same conditions
showed no significant change in chemical shifts between 0.0 and 1.0 M DMSO-d6 and then moved upfield
at higher DMSO-d6 concentrations. At concentrations above 1.0 M DMSO-d6, the directed CH‚‚‚OdS(CD3)2
hydrogen bonds provide protection from random DMSO-d6 contact and prevent the chemical shifts for
participating CH groups from moving upfield beyond the original value observed in CDCl3. X-ray crystal
structures identified CH‚‚‚OdC hydrogen bonds alongside intermolecular NH‚‚‚OdC hydrogen bonding, a
result that supports the solution 1H NMR spectroscopy results. The solution and solid-state data therefore
both provide evidence for the presence of CH‚‚‚O hydrogen bonds formed concurrent with NH‚‚‚O hydrogen
bonding in these structures. The CH‚‚‚OdC hydrogen bonds in the X-ray crystal structures are similar to
those described for antiparallel â-sheet structure observed in protein X-ray crystal structures.
transfer, dispersion, and exchange repulsion forces.8 These
individual energies can contribute different percentages toward
Introduction
Hydrogen bonding in biological systems has long been
associated with contributing to the structure and function of
biomolecules.1 The participation of C-H bonds in the formation
of intermolecular complexes to electron donor atoms has been
long established,2-4although disagreements regarding the nature
of these interactions persist today.5-7 In general, these inter-
actions are described as CH‚‚‚O hydrogen bonds for which
Steiner8 has recently proposed the following definition:
the total energy of a hydrogen bond, especially when strong,
partially covalent hydrogen bonds (15-40 kcal/mol) are con-
sidered vs those that are moderate (4-15 kcal/mol) or weak
(<4 kcal/mol).8 Generally, distance and angle criteria are used
to classify heteroatom contacts as hydrogen bonds. For weak
hydrogen bonds such as those described by CH‚‚‚O contacts,
the distance between the carbon and oxygen atoms are generally
shorter than the sum of their van der Waals (vdW) radii [3.25
Å] and the angles at both the hydrogen and the oxygen are
greater than 110°.8-11 It has been pointed out, however, that
these geometric criteria are far too restrictive and should no
longer be applied.8 Indeed, there are examples of CH‚‚‚O
contacts that either do not fit these geometric criteria6,10,12,13 or
do not show the spectroscopic changes expected for hydrogen-
bonded atoms.14
An X-H‚‚‚A interaction is called a “hydrogen bond” if (1)
it constitutes a local bond and (2) X-H acts as a proton
donor to A.
Hydrogen-bond energies can vary from 0.2 to 40 kcal/mol
and include contributions from electrostatic, polarization, charge
* Corresponding author: e-mail baures@signaturebio.com. Present ad-
dress: Signature Bioscience, Inc., 1240 South 27th Street, Richmond, CA
94804.
In recent years there has been mounting evidence regarding
the importance of CH‚‚‚O hydrogen bonds in peptide and protein
structure.5,15-29 The presence of CH‚‚‚O hydrogen bonds in
antiparallel â-sheets (Figure 1) has been acknowledged on the
† Present address: University of Notre Dame.
‡ Present address: University of Arizona.
(1) Jeffrey, G. A.; Saenger, W. In Hydrogen Bonding in Biological Structures;
Springer-Verlag: New York, 1991.
(2) Sutor, D. J. Nature 1962, 195, 68-69.
(9) Desiraju, G. R.; Steiner, T. In The Weak Hydrogen Bond in Structural
Chemistry and Biology; Desiraju, G. R., Steiner, T., Eds.; Oxford University
Press: New York, 1999.
(10) Taylor, R.; Kennard, O. Acc. Chem. Res. 1984, 17, 320-326.
(11) Taylor, R.; Kennard, O. J. Am. Chem. Soc. 1982, 104, 5063-5070.
(12) Gavezzotti, A. Crystallogr. ReV. 1998, 7, 5-121.
(13) Steiner, T.; Saenger, W. J. Chem. Soc., Perkin Trans. 2 1998, 371-377.
(14) Chaney, J. D.; Goss, C. R.; Folting, K.; Santarsiero, B. D.; Hollingsworth,
M. D. J. Am. Chem. Soc. 1996, 118, 9432-9433.
(3) Allerhand, A.; Schleyer, P. v. R. J. Am. Chem. Soc. 1963, 85, 1715-1723.
(4) Green, R. D. In Hydrogen Bonding by C-H Groups; John Wiley & Sons:
New York, 1974.
(5) Steiner, T. J. Chem. Soc., Perkin Trans. 2 1995, 1315-1319.
(6) Steiner, T.; Desiraju, G. R. J. Chem. Soc., Chem. Commun. 1998, 891-
892.
(7) Dunitz, J. D.; Gavezzotti, A. Acc. Chem. Res. 1999, 32, 677-684.
(8) Steiner, T. Angew. Chem., Int. Ed. 2002, 41, 48-76.
9
10.1021/ja0257366 CCC: $22.00 © 2002 American Chemical Society
J. AM. CHEM. SOC. 2002, 124, 11315-11323
11315

A R T I C L E S
Baures et al.
Figure 2. Similarities of an N-acylpeptide with a fumaramide derivative
in the s-cis, s-cis conformation, including two rotatable bonds and both
NH and CH groups adjacent to a carbonyl group.
Figure 1. Antiparallel â-sheet structure illustrating concurrent NH‚‚‚O and
CH‚‚‚O hydrogen bonds.
or triethylamine is found downfield in comparison to a nonpolar
medium such as cyclohexane, a result consistent with the
formation of CH‚‚‚O hydrogen bonds to the polar solvents.
The CH‚‚‚O hydrogen bond has been used in crystal
engineering38-41 and has also been described in the context of
drug design and molecular recognition.1,9,42-45 Yet it has rarely
been used in the deliberate design of biologically active
compounds.46,47 We are unaware of any reports to date sug-
gesting the presence of CH‚‚‚O hydrogen bonds in the peptoids
[N-alkylglycine oligomers], though the similar bioactivity48-51
and secondary structures observed for some peptoids as
compared to the corresponding peptide sequences is note-
worthy.52-55 In designing these oligomers, a peptide is aligned
with the retropeptoid backbone in a manner that superimposes
the carbonyl groups and the side chains of each structure. This
alignment effectively replaces the NH groups in the peptide
backbone for backbone CH2 groups in the peptoid.
This paper describes the use of fumaramides as models of
the peptide bond (Figure 2) that are useful for studying CH‚‚‚O
hydrogen bonds in both solution and the solid state. The
fumaramide reverses the position of the strong hydrogen-bond
donor (NH) and the associating CH group as compared with
an N-acylpeptide bond. Though the fumaramide retains two
rotatable bonds, the s-cis and s-trans orientations are recognized
as nearly equivalent in energy and of lower energy than other
noncoplanar bond orientations.56 Thus, it was expected that the
basis of X-ray crystal structures where the distances between
heteroatoms were used in their identification.21,23 In this manner,
the CH‚‚‚O hydrogen bond has also been shown to be involved
in the capping of R-helices in several peptide sequences that
terminate in the Schellman motif28 and in a peptide containing
a γ-amino acid that forms an intramolecular CH‚‚‚O hydrogen
bond reminiscent of a hydrogen-bonded â-turn conformation.29
It is reasonable to expect that foldamers containing unnatural
amino acids will also utilize CH‚‚‚O hydrogen bonds as
stabilizing elements of their secondary structure.30-37 Ab initio
studies of formamide dimers have been used to estimate a
possible energetic contribution of CH‚‚‚O hydrogen bonds in
protein structure.19 These calculations suggest that CH‚‚‚O
hydrogen bonds could be significant contributors to protein
folding and structure due to their individual contribution to
stability as well as the large number of contacts common in a
typical protein. To date, there have been no experimental
quantifications of CH‚‚‚O hydrogen-bond energies.
Early solution studies of C-H group interactions often
employed freezing-point depression, enthalpies of mixing,
vibrational spectroscopy, and NMR spectroscopy in order to
verify and characterize the formation of complexes.4 For
example, the chloroform 1H NMR signal in acetone, ethyl ether,
(15) Senes, A.; Ubarretxena-Belandia, I.; Engelman, D. M. Proc. Natl. Acad.
Sci. U.S.A. 2001, 98, 9056-9061.
(16) Felcy Fabiola, G.; Bobde, V.; Damodharan, L.; Pattabhi, V.; Durani, S. J.
Biomol. Struct. Dyn. 2001, 18, 579-594.
(38) Anthony, A.; Desiraju, G. R.; Jetti, R. K. R.; Kuduva, S. S.; Madhavi, N.
N. L.; Nangia, A.; Thaimattam, R.; Thalladi, V. R. Cryst. Eng. 1998, 1,
1-18.
(17) Olson, C. A.; Shi, Z.; Kallenbach, N. R. J. Am. Chem. Soc. 2001, 123,
6451-6452.
(18) Thakur, A. K.; Kishore, R. Biopolymers 2000, 53, 447-454.
(19) Vargas, R.; Garza, J.; Dixon, D. A.; Hay, B. P. J. Am. Chem. Soc. 2000,
122, 4750-4755.
(39) Desiraju, G. R. Acc. Chem. Res. 1996, 29, 441-449.
(40) Desiraju, G. R. Angew. Chem., Int. Ed. Engl. 1995, 34, 2311-2327.
(41) Aakero¨y, C. B. Acta Crystallogr. 1997, B53, 569-586.
(42) Burley, S. K.; Petsko, G. A. AdV. Protein Chem. 1988, 39, 125-189.
(43) Wahl, M. C.; Sundaralingam, M. Trends Biochem. Sci. 1997, 22, 97-102.
(44) Pascard, C. Acta Crystallogr. 1995, D51, 407-417.
(45) Glusker, J. P. Acta Crystallogr. 1995, D51, 418-427.
(46) Baures, P. W.; Wiznycia, A. V.; Beatty, A. M. Bioorg. Med. Chem. 2000,
8, 1599-1605.
(20) Nagarajan, V.; Pattabhi, V.; Johnson, A.; Bobde, V.; Durani, S. Int. J. Pept.
Protein Res. 1997, 49, 74-79.
(21) Fabiola, G. F.; Krishnaswamy, S.; Nagarajan, V.; Pattabhi, V. Acta
Crystallogr. D 1997, 53, 316-320.
(22) Bella, J.; Berman, H. M. J. Mol. Biol. 1996, 264, 734-742.
(23) Derewenda, Z. S.; Lee, L.; Derewenda, U. J. Mol. Biol. 1995, 252, 248-
262.
(47) Baures, P. W. Org. Lett. 1999, 1, 249-252.
(24) Panneerselvam, K.; Chacko, K. K. Int. J. Pept. Protein Res. 1990, 35, 460-
(48) Low, C. M. R.; Black, J. W.; Broughton, H. B.; Buck, I. M.; Davies, J. M.
R.; Dunstone, D. J.; Hull, R. A. D.; Kalindjian, S, B.; McDonald, I. M.;
Pether, M. J.; Shankley, N. P.; Steel, K. I. M. J. Med. Chem. 2000, 43,
3505-3517.
464.
(25) Panneerselvam, K.; Chacko, K. K.; Veena, R. Int. J. Pept. Protein Res.
1989, 33, 191-194.
(26) Parthasarathy, R.; Fridey, S. M.; Srikrishnan, T. Int. J. Pept. Protein Res.
1989, 33, 308-312.
(49) Heizmann, G.; Hildebrand, P.; Tanner, H.; Ketterer, S.; Pansky, A.;
Froidevaux, S.; Beglinger, C.; Eberle, A. N. J. Recept. Signal Transduct.
Res. 1999, 19, 449-466.
(27) Ramachandran, G. N.; Chanrasekharan, R. Biopolymers 1968, 6, 1649-
1658.
(50) Tran, T.-A.; Mattern, R.-H.; Afargan, M.; Amitay, O.; Ziv, O.; Morgan,
B. A.; Taylor, J. E.; Hoyer, D.; Goodman, M. J. Med. Chem. 1998, 41,
2679-2685.
(28) Aravinda, S.; Shamala, N.; Pramanik, A.; Das, C.; Balaram, P. Biochem.
Biophys. Res. Commun. 2000, 273, 933-936.
(29) Cheung, E. Y.; McCabe, E. E.; Harris, K. D. M.; Johnston, R. L.; Tedesco,
E.; Raja, K. M. P.; Balaram, P. Angew. Chem., Int. Ed. 2002, 41, 494-
496.
(51) Figliozzi, G. M.; Goldsmith, R.; Ng, S. C.; Banville, S. C.; Zuckermann,
R. N. Methods Enzymol. 1996, 267, 437-447.
(52) Armand, P.; Kirshenbaum, K.; Goldsmith, R. A.; Farr-Jones, S.; Barron,
A. E.; Truong, K. T. V.; Dill, K. A.; Mierke, D. F.; Cohen, F. E.;
Zuckermann, R. N.; Bradley, E. K. Proc. Natl. Acad. Sci., U.S.A. 1998,
95, 4309-4314.
(30) Seebach, D.; Brenner, M.; Rueping, M.; Bernhard, J. Chemistry 2002, 8,
573-584.
(31) Etezady-Esfarjani, T.; Hilty, C.; Wuethrich, K.; Rueping, M.; Schrieber, J.
Seebach, D. HelV. Chim. Acta 2002, 85, 1197-1209.
(32) Hill, D. J.; Mio, M. J.; Prince, R. B.; Hughes, T. S.; Moore, J. S. Chem.
ReV. 2001, 101, 3893-4011.
(53) Kirshenbaum, K.; Barron, A. E.; Goldsmith, R. A.; Armand, P.; Bradley,
E. K.; Truong, K. T. V.; Dill, K. A.; Cohen, F. E.; Zuckermann, R. N.
Proc. Natl. Acad. Sci., U.S.A. 1998, 95, 4303-4308.
(33) Woll, M. G.; Lai, J. R.; Guzei, I. A., Taylor, S. J. C.; Smith, M. E. B.;
Gellman, S. H. J. Am. Chem. Soc. 2001, 123, 11077-11078.
(34) Cheng, R. P.; Gellman, S. H.; DeGrado, W. F. Chem. ReV. 2001, 101,
3219-3232.
(54) Armand, P.; Kirshenbaum, K.; Falicov, A.; Dunbrack, R. I., Jr.; Dill, K.
A.; Zuckermann, R. N.; Cohen, F. E. Folding Des. 1997, 2, 369-375.
(55) Sanborn, T. J.; Wu, C. W.; Zuckerman, R. N.; Barron, A. E. Biopolymers
2002, 63, 12-20.
(35) Abele, S.; Vogtli, K.; Seebach, D. HelV. Chim. Acta 1999, 82, 1539-1558.
(36) Gellman, S. H. Acc. Chem. Res. 1998, 31, 173-180.
(37) Seebach, D.; Matthews, J. L. Chem. Commun. 1997, 2015-2022.
(56) Eliel, E. L.; Wilen, S. H.; Doyle, M. P. In Basic Organic Stereochemistry;
Eliel, E. L., Wilen, S. H., Doyle, M. P., Eds.; John Wiley & Sons: New
York, 2001; p 394.
9
11316 J. AM. CHEM. SOC. VOL. 124, NO. 38, 2002

Models of Peptide CH‚‚‚O Hydrogen Bonds
A R T I C L E S
Scheme 1
Chart 2
Scheme 2
Chart 1. Starting Materials Used for Library Synthesis
Chart 3
Table 1. Alkene CH Chemical Shift Comparisonsa
c
compound
observedb
calculated
1
6.96, 6.83
6.96, 6.56
6.90, 6.85
7.06, 6.62
7.66, 6.47
7.46, 6.72
6.90
7.63, 7.01
7.63, 7.01
7.61, 6.98
7.60
2
4
6
8
6.90
6.85
7.04
6.71
9
10
11
6.90d
7.60
7.60
7.60
7.29, 7.08e
7.97f
a Chemical shifts are in parts per million (ppm). b Values were determined
at a concentration of 10 mM (1-8) in CDCl3 or DMSO-d6 (values shown
in italic type). c Reference 58. d D2O, ref 59. e CDCl3, ref 60. f DMSO-d6,
ref 61.
fumaramides could form concurrent NH‚‚‚O and CH‚‚‚O
hydrogen bonds in solutions containing strong hydrogen-bond
acceptors such as DMSO-d6. The solid-state packing of fuma-
ramides was expected to result from concurrent NH‚‚‚O and
CH‚‚‚O hydrogen bonds and we report herein the solid-state
structures of a peptide, two comparative fumaramides, and two
additional model compounds.
carbodiimide. Solids formed in some of the reaction vials as
the solutions were allowed to concentrate to approximately half
of the initial volume. Crystallization of products 4-7 was
observed and these compounds were then synthesized on a larger
scale for detailed analysis (Chart 2).
Results and Discussion
1
Solution H NMR Studies. We attempted to make initial
Synthesis. The synthesis of 1 and 2 were accomplished from
monoethyl fumarate via the acid chloride (Scheme 1). Similarly,
the peptide analogue 3 was synthesized by first forming the
acid chloride of phenylacetic acid and then reacting this
intermediate with glycine ethyl ester hydrochloride and diiso-
propylethylamine (Scheme 2). As a means to identify crystalline
derivatives that could be added to this study, a library of 40
compounds was synthesized in parallel fashion by using two
monocarboxylic acids and two dicarboxylic acids and combining
these with 10 different amino acid derivatives (Chart 1).
Solutions of the starting materials in ethanol were mixed with
N-methylmorpholine and an aqueous solution of a water-soluble
alkene CH chemical shift assignments for 1, 2, and 4 on the
basis of standard chemical shift models for substituted ethyl-
enes,57,58 along with the addition of 8-11 (Chart 3) as examples
for comparison. However, the observed chemical shifts are
significantly different than the calculated chemical shifts based
on these models (Table 1). There are also differences in the
alkene chemical shift depending upon the solvent, although only
one of the two alkene CHs in each of the compounds 1, 2, and
4 shifts dramatically upfield in DMSO-d6 as compared to CDCl3.
(57) Pascual, C., Meier, J.; Simon, W. HelV. Chim. Acta 1966, 49, 164-168.
(58) Matter, U. E.; Pascual, C.; Pretsch, E.; Pross, A.; Simon, W.; Sternhell, S.
Tetrahedron 1969, 25, 691-697.
9
J. AM. CHEM. SOC. VOL. 124, NO. 38, 2002 11317

A R T I C L E S
Baures et al.
Table 2. Chemical Shift Values of 1a Observed in Varying
Proportions of CDCl3 and DMSO-d6
[DMSO-d6]
NH
anti
NH
syn
H b
H b
a
b
0.00 mM
5.00 mM
10.0 mM
20.0 mM
50.0 mM
100 mM
200 mM
500 mM
1.00 M
5.66
5.69
5.69
5.75
5.81
6.40
6.64
7.00
7.16
7.86
7.90
5.66
5.69
5.69
5.75
5.81
6.02
6.06
6.12
6.16
7.42
7.51
6.96
6.94
6.94
6.94
6.95
6.98
7.00
7.03
7.04
6.98
6.96
6.83
6.83
6.83
6.84
6.84
6.84
6.83
6.82
6.80
6.58
6.56
Figure 3. Hydrogen bonding (NH‚‚‚O) of 1 to DMSO-d6 with a concurrent
CH‚‚‚O hydrogen bond to Ha.
By comparison, the alkene CH in 6 is found at 6.90 ppm in
both CDCl3 and DMSO-d6, a value consistent with the CH
observed furthest downfield in 1 and 2. Diethyl fumarate, 8,
was also used for comparison with the alkene CH in this diester
observed at 6.85 ppm in CDCl3 (as well as in solution containing
up to 10.0 M DMSO-d6) and then found at 6.71 ppm in DMSO-
d6. These values are consistent with the upfield shift for one of
the two alkene CHs in both 1 and 2. Compound 9 in D2O has
an alkene chemical shift at 6.90 ppm,59 while 10 in CDCl3 has
two alkene CH chemical shifts observed at 7.29 and 7.08 ppm.60
It is likely that the two shifts are due to conformational s-cis
and s-trans isomers, as two sets of 13C NMR signals were also
observed for 10. The alkene chemical shift values for 1 and 2
places these compounds in a very select category since the
deviations between observed and predicted values are greater
than 0.50 ppmsa category into which only 0.3% of 4298
alkenes fell in an early compilation.58 Compound 11, though
very similar to 6 and 9 in structure, has an alkene chemical
shift significantly higher and closer to its predicted value.61 We
do not have a satisfactory explanation for the chemical shift
behavior of these alkenes at this time, although similar R,â-
unsaturated systems also have alkene chemical shift values
disparate from those calculated on the basis of the earlier
models.62-65
Confirmation of the alkene chemical shift assignments was
attempted by using NOESY NMR spectroscopy on solutions
of 1 and 2 in CDCl3 containing 100 mM and 1.0 M DMSO-d6,
respectively, and looking for the NHanti-Ha (or comparable)
nuclear Overhauser effect (NOE) cross-peak. Fumaramides 1
and 2 were expected to associate with a strong hydrogen-bond
acceptor and thereby result in an observable effect on the NH
and CH chemical shift values (Figure 3). Unfortunately, we were
unable to observe any cross-peaks from an NH to either alkene
CH, presumably due to rapid bond rotation on the NMR time
scale that occurs even in the presence of the 1.0 M DMSO-d6.
The solution NMR data are therefore only supportive of the
interactions shown in Figure 3, and evidence of the concurrent
CH‚‚‚O hydrogen bond in solution is not definitive at this time.
The extent to which these data do support the presence of this
CH‚‚‚O hydrogen bond over other possibilities, such as solvent
polarity and conformation as previously described for R,â-
unsaturated carbonyl compounds,66 is included in the following
10.0 M
14.1 M
a Compound 1 (10 mM): Et2OC-CHbdCHa-CONH2. Chemical shifts
are in parts per million. b Alkene chemical shift assignments are tentative
as discussed in the text.
descriptions of chemical shift behavior for each compound
measured as a function of increasing DMSO-d6 concentration.
The CDCl3 solutions containing DMSO-d6 were referenced
to tetramethylsilane (TMS) since the observed CHCl3 and
DMSO chemical shift values are dependent upon the concentra-
tion of one another. The chemical shift of CHCl3 was observed
at 7.26(2) ppm in the solutions containing 0.0-1.0 M DMSO-
d6, 8.19(3) ppm at 10.0 M DMSO-d6, and 8.31(2) ppm in the
14.1 M DMSO-d6 solutions. Trace CHCl3 in DMSO-d6 is
observed at 8.32 ppm.67
When a 10 mM solution of 1 in CDCl3 is titrated with DMSO-
d6, a downfield shift for the amide NH resonance is observed
due to hydrogen bonding with the DMSO-d6 (Table 2). Two
amide NH resonances are observed at and above 100 mM
DMSO-d6, while at low concentrations of DMSO-d6 (<100
mM) the NH chemical shift values are identical to one another
due to either the fast dynamics of bond rotation at room
temperature or coincidence. Increasing the concentration of a
strong hydrogen-bond acceptor such as DMSO-d6 could be
expected to slow bond rotation due to the additional energetic
penalty of breaking this hydrogen bond, though the on and off
rate of DMSO-d6 hydrogen-bond formation could still be fast
on the NMR time scale. The higher DMSO-d6 concentrations
may have an additional influence on the kinetics of bond rotation
due to the significant change in solvent polarity over the CDCl3
solutions. If the chemical shift of the carboxamide NHs are
identical due to coincidence, then increasing the concentration
of DMSO-d6 could still differentiate the NHs due to the alternate
environments formed in the hydrogen-bonded complex. The
alkene and amide hydrogen chemical shift data for 1 from the
0.0-1.0 M DMSO-d6 solutions is graphed in Figure 4. The
amide NH chemical shifts have the greatest frequency separation
(1.00 ppm) at 1.0 M DMSO-d6 in the solutions tested, but the
separations in 10.0 M DMSO-d6 (0.44 ppm) and in neat DMSO-
d6 (0.39 ppm) are significantly less.
Though the NH chemical shift value for 1 changes at DMSO-
d6 concentrations below 100 mM, there are no significant
changes in either alkene CH chemical shifts in these solutions.
Concurrent formation of a CH‚‚‚O hydrogen bond along with
the NHanti‚‚‚O hydrogen bond is suggested at concentrations
between 100 mM and 1.0 M DMSO-d6 as evidenced by the
downfield chemical shift of Ha at the same time that Hb is not
(59) Arora, P. S.; Van, Q. N.; Famulok, M.; Shaka, A. J.; Nowick, J. S. Bioorg.
Med. Chem. 1998, 6, 1421-1428.
(60) Bla¨ser, E.; Kolar, P.; Fenske, D.; Goesmann, H.; Waldmann, H. Eur. J.
Org. Chem. 1999, 329-333.
(61) Ko¨nig, S.; Lohberger, S.; Ugi, I. Synthesis 1993, 1233-1234.
(62) Suhara, Y.; Maruyama, H. B.; Kotoh, Y.; Miyasaka, Y.; Yokose, K.; Shirai,
H.; Takano, K.; Quitt, P.; Lanz, P. J. Antibiot. 1975, 28, 648-655.
(63) de Pooter, H.; van Rompaey, L.; van Sumere, C. F. Bull. Soc. Chim. Belg.
1976, 85, 647-656.
(64) Canoira, L.; Rodriguez, J. G. J. Heterocycl. Chem. 1985, 22, 1511-1518.
(65) Horne, S.; Taylor, N.; Collins, S.; Rodrigo, R. J. Chem. Soc., Perkin Trans.
1 1991, 3047-3051.
(66) Savin, V. I.; Temyachev, I. D.; Kitaev, Y. P. Zh. Org. Khim. 1974, 10,
1590-1593.
(67) Gottlieb, H. E.; Kotlyar, V.; Nudelman, A. J. Org. Chem. 1997, 62, 7512-
7515.
9
11318 J. AM. CHEM. SOC. VOL. 124, NO. 38, 2002

Models of Peptide CH‚‚‚O Hydrogen Bonds
A R T I C L E S
Table 4. Chemical Shift Values of 3a Observed in Varying
Proportions of CDCl3 and DMSO-d6
[DMSO-d6]
H C
2
CH
2
NH
CH
2
0.00 mM
5.00 mM
10.0 mM
20.0 mM
50.0 mM
100 mM
200 mM
500 mM
1.00 M
4.18
4.18
4.18
4.18
4.18
4.18
4.18
4.18
4.17
4.10
4.07
4.00
3.99
4.00
3.99
3.99
3.99
3.99
3.98
3.98
3.84
3.82
5.89
5.92
5.92
5.93
5.97
5.97
6.04
6.32
6.69
8.38
8.46
3.63
3.64
3.64
3.64
3.63
3.63
3.63
3.63
3.62
3.50
3.48
Figure 4. 1H NMR chemical shift data for alkene (tentative assignments)
and amide hydrogens in 1 [Et2OC-CHbdCHa-CONH2] as CDCl3 solution
containing varying concentrations of DMSO-d6.
10.0 M
14.1 M
Table 3. Chemical Shift Values of 2a Observed in Varying
Proportions of CDCl3 and DMSO-d6
a Compound 3 (10 mM): H3CH2CO2CCH2NHCOCH2Ph. Chemical
shifts are in parts per million.
[DMSO-d6]
CH
2
NH
H b
H b
a
b
Table 5. Chemical Shift Values of 4a Observed in Varying
Proportions of CDCl3 and DMSO-d6
0.00 mM
5.00 mM
10.0 mM
20.0 mM
50.0 mM
100 mM
200 mM
500 mM
1.00 M
4.55
4.55
4.55
4.55
4.55
4.54
4.53
4.52
4.50
4.41
4.39
6.02
6.17
6.19
6.23
6.34
6.60
6.85
7.61
7.98
8.95
9.02
6.90
6.92
6.92
6.92
6.93
6.95
6.98
7.04
7.07
7.07
7.06
6.85
6.85
6.85
6.85
6.85
6.85
6.85
6.84
6.83
6.64
6.62
[DMSO-d6]
CH
2
NH
CH b
CH b
a
b
0.00 mM
5.00 mM
10.0 mM
20.0 mM
50.0 mM
100 mM
200 mM
500 mM
1.00 M
4.18
4.18
4.18
4.18
4.18
4.17
4.17
4.15
4.12
3.97
3.96
6.14
6.16
6.17
6.19
6.25
6.40
6.53
6.94
7.40
8.49
8.53
6.47
6.47
6.47
6.47
6.48
6.50
6.58
6.57
6.62
6.71
6.72
7.66
7.66
7.66
7.66
7.66
7.65
7.65
7.64
7.62
7.46
7.46
10.0 M
14.1 M
a Compound 2 (10 mM): Et2OC-CHbdCHa-CONHCH2Ph. Chemical
shifts are in parts per million. b Alkene chemical shift assignments are
tentative as discussed in the text.
10.0 M
14.1 M
a Compound 4 (10 mM): EtO2CCH2NHCO-CHadCHb-Ph. Chemical
shifts are in parts per million. b Alkene chemical shift assignments from
ref 63.
significantly affected. Higher concentrations of DMSO-d6 (>1.0
M) result in Hb moving significantly upfield, a result that we
attribute to increased frequency of largely random interactions
of the Hb with the DMSO-d6. In contrast, Ha is by and large
protected from random DMSO-d6 interactions at these high
concentrations due to the specific and directional NHanti‚‚‚O
hydrogen bond and concurrent CH‚‚‚O hydrogen bond. This
orientation of the DMSO-d6 keeps the shielding cone of the
carbonyl group directed at Ha. The observed chemical shift in
pure DMSO-d6 returns to the value observed in CDCl3, likely
due to the slight increase in random orientation of the DMSO-
d6 around Ha resulting from increased frequency of DMSO-d6
exchange at higher concentrations. The magnitude of the
difference in the alkene chemical shift values for the DMSO-
d6 versus CDCl3 solutions cannot be interpreted in terms of the
strength of the CH‚‚‚O hydrogen bond67 and is simply taken as
a qualitative observation consistent with complex formation.
The chemical shift changes observed for 2 are similar to those
observed for 1 with the NH chemical shift value moving
downfield upon addition of DMSO-d6 (Table 3). The Ha
chemical shift value in 2 also moves downfield with increasing
concentrations of DMSO-d6, whereas Hb remains unchanged
at DMSO-d6 concentrations below 1.0 M and then moves upfield
at higher concentrations. The maximum observed ∆δ for Ha is
0.17 ppm in 2 as compared to a ∆δ for Ha of 0.08 ppm in 1.
The ester CH2 behaves similarly to Hb, with its observed
chemical shift moving upfield at DMSO-d6 equal to and above
1.0 M.
the CH2s in 3 behaved similarly to one another, with no
significant chemical shift changes observed below 1.0 M
DMSO-d6 and upfield shifts observed at higher concentrations
of DMSO-d6. The lack of a similar upfield chemical shift for
the benzyl CHs of 3 could result from several differences
between 3 and 1 or 2. There are two benzyl CHs in 3 and each
could interact weakly with a DMSO-d6 that hydrogen bonds to
the NH group, or alternatively, the two CHs could be dynami-
cally interchanging in such an interaction. There is also a
difference in hybridization of the carbon atom adjacent to the
amide carbonyl bond in 3 (sp3) as compared with the carbon
atom in 1 or 2 (sp2), plus there is an altered conformational
flexibility in 3. The magnetic anisotropy of the nearby aromatic
ring may also influence the chemical shift values in these solvent
mixtures.68,69 It is not possible at this time to identify which
change is the most important or even whether all of the changes
make contributions to the observed differences in solution.
The parallel syntheses were designed to identify additional
model compounds for combined solution and solid-state analy-
sis. trans-Cinnamic acid, monoethylfumarate, and fumaric acid
were chosen as model carboxylic acids that retain an alkene
CH adjacent to the amide carbonyl, whereas succinic acid was
included as a model for peptides containing sp3-hybridized
carbon atoms adjacent to the amide carbonyl bond.
Independent titrations of 4 (Table 5) and 6 (Table 6) with
DMSO-d6 resulted in chemical shift changes that parallel those
observed for 1 and 2. The NH chemical shift values for each
Peptide 3 was synthesized in order to determine whether
chemical shift changes similar to those observed for 1 and 2
would occur with the addition of DMSO-d6. The NH chemical
shift for 3 moves downfield with DMSO-d6 addition, in likewise
fashion to the observed shift for 1 and 2 (Table 4). All three of
(68) Wiley, G. R.; Miller, S. I. J. Am. Chem. Soc. 1972, 94, 3287-9293.
(69) Johnston, M. D., Jr.; Gasparro, F. P.; Kuntz, I. D., Jr. J. Am. Chem. Soc.
1969, 91, 5715-5724.
9
J. AM. CHEM. SOC. VOL. 124, NO. 38, 2002 11319

A R T I C L E S
Baures et al.
Table 6. Chemical Shift Values of 6a Observed in Varying
Proportions of CDCl3 and DMSO-d6
bonds, changes in hybridization, or altered conformational
flexibility between the compounds.
[DMSO-d6]
R-CH
NH
H
a
E−H
X-ray Crystallography. The single-crystal X-ray structures
of the model compounds were sought in order to determine
whether CH‚‚‚O hydrogen bonds formed concurrently with
NH‚‚‚O hydrogen bonds in the solid state. The presence of
CH‚‚‚O hydrogen bonds would support the solution NMR results
since crystallization occurs by molecular recognition forces
analogous to those governing solution interactions. The hydrogen-
bond acceptor during crystallization is an amide carbonyl rather
than a sulfoxide, but both have similar hydrogen-bond-accepting
abilities.70 Crystals suitable for single-crystal X-ray diffraction
were obtained for 1-3, 5, and 6 and the crystal data for each
structure are given in Table 8. Each compound forms concurrent
intermolecular NH‚‚‚O and CH‚‚‚O hydrogen bonds. These
intermolecular interactions are consistent with the solution NMR
results and are illustrated in Figure 5. The corresponding
distances and angles for these interactions are given in Table
9. Though crystalline solids formed for both 4 and 7, the crystals
were not suitable for the determination of their crystal structures.
There are two strong intermolecular interactions in 1: the
NHanti‚‚‚Oamide hydrogen bond, listed in Table 9 and illustrated
in Figure 5a, as well as a second hydrogen bond between the
carboxamide NH and ester carbonyl oxygen that is not shown
in this illustration [NHsyn‚‚‚Oester, N‚‚‚O separation ) 3.00(1)
Å, NH‚‚‚O ) 168(1)°]. Formation of these two NH‚‚‚O
hydrogen bonds results in different planes being occupied by
nearby molecules in the crystal lattice. A CH‚‚‚O hydrogen bond
is formed concurrent with the NHanti‚‚‚Oamide hydrogen bond.
The distances between heteroatoms [C‚‚‚O, 3.34(1) Å] and for
the hydrogen to acceptor oxygen [H‚‚‚O, 2.57(1) Å] are less
than the vdW distances often used to distinguish these
interactions.8-11
0.00 mM
5.00 mM
10.0 mM
20.0 mM
50.0 mM
100 mM
200 mM
500 mM
1.00 M
4.73
4.73
4.73
4.73
4.73
4.72
4.70
4.67
4.64
4.40
4.37
6.67
6.69
6.69
6.71
6.76
6.83
7.02
7.45
7.83
8.75
8.81
6.90
6.90
6.90
6.91
6.92
6.94
6.96
7.00
7.03
6.92
6.90
3.76
3.76
3.75
3.76
3.75
3.75
3.74
3.73
3.72
3.65
3.64
10.0 M
14.1 M
a Compound 6 (10 mM): E-Leu-NHCOCHadCHa′-CONH-Leu-E.
Chemical shifts are in parts per million.
Table 7. Chemical Shift Values of 7a Observed in Varying
Proportions of CDCl3 and DMSO-d6
[DMSO-d6]
R-CH
NH
CH
2
EH
0.00 mM
100 mM
500 mM
1.00 M
4.58
4.57
4.55
4.54
4.54
6.53
6.67
6.95
7.21
7.21
2.55
2.56
2.56
2.56
2.56
3.73
3.73
3.72
3.71
3.71
14.1 M
a Compound 7 (10 mM): E-Leu-NHCO-CH2-CH2-CONH-Leu-E.
Chemical shifts are in parts per million.
compound moved downfield with increasing DMSO-d6 con-
centration. From 100 mM to 1.0 M DMSO-d6, the chemical
shift for alkene Ha in 4 moves downfield. The downfield
chemical shift changes continued for 4 at higher DMSO-d6
concentrations, while in 6 the chemical shift values return to
that observed in CDCl3. This may be due to the need to only
interact with one DMSO-d6 molecule to create the complex and
full change in 4, while two DMSO-d6 molecules must bind to
6 (a less favorable ternary complex) for the full effect to be
observed. Other CHs in 4 and 6 did not have their chemical
shift values change significantly below 1.0 M and then shifted
upfield at increasing DMSO-d6 concentrations.
Compound 5 was not studied in solution due to the low
solubility of this compound in CDCl3. Only a few solutions
containing DMSO-d6 were prepared and analyzed for 7 as no
chemical shift differences were observed for any CHs at either
1.0 M or neat DMSO-d6 as compared to the values in neat
CDCl3 (Table 7). The difference between 6 and 7 could again
be due to dynamic exchange between possible CH‚‚‚O hydrogen
In contrast, the concurrent intermolecular NH‚‚‚O and CH‚‚‚O
hydrogen bonds found in the crystal structure of 2 occur between
molecules translated within the same plane. The NH‚‚‚O
hydrogen bond is longer in 2 [3.07(1) Å] as compared to 1 [2.90-
(1) Å], but the CH‚‚‚O hydrogen bond is shorter in 2 [3.25(1)
Å] as compared to 1 [3.34(1) Å]. We attribute these differences
to a compromise between the strength of the two interactions
in the context of one another and the remaining crystal packing
effects. Importantly, the presence of CH‚‚‚O hydrogen bonds
in the solid states of both 1 and 2, despite differences in the
crystal packing of each, supports the hypothesis of their
Table 8. Selected Crystal Data Collection and Refinement Data for 1, 2, 3, 5, and 6
crystal data
formula
weight (g mol-1
1
2
3
5
6
C6H9NO3
C13H15NO3
233.26
C12H15NO3
221.25
C14H18N2O2
246.30
C18H30N2O6
370.44
)
143.14
crystal size (mm)
crystal system
space group
a (Å)
b (Å)
c (Å)
R (deg)
â (deg)
γ (deg)
0.50 × 0.40 × 0.20
orthorhombic
Pna21
20.880(5)
3.928(1)
8.610(2)
90.0
0.30 × 0.20 × 0.10
orthorhombic
Pna21
17.632(1)
14.097(1)
4.920(1)
90.0
0.50 × 0.20 × 0.10
orthorhombic
Pna21
9.975(2)
12.130(2)
9.553(2)
90.0
0.30 × 0.15 × 0.10
monoclinic
C2
13.514(6)
4.797(2)
21.238(10)
90.0
0.30 × 0.20 × 0.15
orthorhombic
P212121
14.440(5)
4.852(2)
27.377(11)
90.0
90.0
90.0
90.0
90.0
90.0
90.0
97.14(1)
90.0
90.0
90.0
Z
4
4
4
4
4
temp (K)
R/Rw2 (obs data)
S
203(2)
0.0257/0.0625
1.104
293(2)
0.0436/0.0729
1.105
203(2)
0.0333/0.0688
1.079
293(2)
0.0400/0.0805
1.128
293(2)
0.0687/0.1676
0.977
9
11320 J. AM. CHEM. SOC. VOL. 124, NO. 38, 2002

Models of Peptide CH‚‚‚O Hydrogen Bonds
A R T I C L E S
Figure 5. Diagrams illustrating the intermolecular packing interactions involving hydrogen bonding as found in the X-ray crystal structures of (a) 1, (b) 2,
(c) 3, (d) 5, and (e) 6.
distance often accepted for these interactions.8-11 It may be
possible that this distance alone prevents a chemical shift change
Table 9. Hydrogen-Bonding Distances and Angles in 1-3, 5, and
6
X−H
X‚‚‚O (Å)
H‚‚‚O (Å)
XH‚‚‚O (deg)
and thus explains our solution results. The CH‚‚‚O hydrogen
bond in 3 may also have less significance for the molecular
recognition events during crystal formation than the analogous
interactions in 1 and 2, or it may be an artifact of stronger forces
that take over during crystal packing. The data in this study,
though consistent and supportive of a weaker CH‚‚‚O hydrogen
bond in 3, are not sufficient to differentiate these possibilities.
The crystal structure of 5 provides additional support for the
importance of concurrent CH‚‚‚O hydrogen bonds in peptide
and protein structure. Three strong hydrogen-bond donors (NHs)
and two hydrogen-bond acceptors (amide oxygens) are present
in 5 and, as expected on the basis of hydrogen-bonding rules
for small molecules,71,72 each NH is hydrogen-bonded to an
amide carbonyl oxygen. Two of the NH‚‚‚O hydrogen bonds
are between molecules related by translation, where the amide
carbonyl oxygens also make concurrent CH‚‚‚O hydrogen bonds
to the alkene CH (Ha) and CRH (Hc). The distances for the
NH‚‚‚O and CH‚‚‚O hydrogen bonds in 5 are similar to the
values observed for 1 and 2. The NH‚‚‚O hydrogen bond not
shown in Figure 5d occurs between the carboxamide and the
secondary amide carbonyl oxygen [NHsyn‚‚‚O2°amide, N‚‚‚O
separation ) 2.99(1) Å, NH‚‚‚O ) 163(1)°]. The ψ torsion
angle within the valinamide [134°] is consistent with the angle
expected for antiparallel â-sheet structure [135°].73
1, Et2OC-CHbdCHa-CONH2
N-Hanti
C-Ha
2.90(1)
3.34(1)
1.95(1)
2.57(1)
170(1)
139(1)
2, Et2OC-CHbdCHa-CONHCH2Ph
N-H
C-Ha
3.07(1)
3.25(1)
2.11(1)
2.42(1)
159(1)
149(1)
3, H3CH2CO2CCH2NHCOCH2Ph
N-H
C-H2
2.79(1)
3.45(1)
1.90(1)
2.73(1)
156(1)
131(1)
5, Ph-CHbdCHa-CONH-CHc[CH(CH3)2]-CONH2
N-H
2.97(1)
3.20(1)
2.88(1)
3.31(1)
2.02(1)
2.56(1)
1.91(1)
2.41(1)
152(1)
135(1)
157(1)
141(1)
C-Ha
N-Hanti
C-Hc
(6) E-Leu-NHCO-CHadCHa′-CONH-Leu-E
N-H
C-Ha
N-H
C-Ha′
2.96(1)
3.35(1)
2.95(1)
3.32(1)
2.04(1)
2.56(1)
2.04(1)
2.54(1)
156(1)
142(1)
162(1)
138(1)
(secondary) involvement in the intermolecular recognition
leading to crystallization.
The NMR study of 3 did not provide evidence of CH‚‚‚O
hydrogen bonding in solution. In the crystal structure of 3, a
short NH‚‚‚O hydrogen bond [2.79(1) Å] is observed along with
a concurrent CH‚‚‚O hydrogen bond [3.45(1) Å] to one of the
two benzylic CHs having a C‚‚‚O separation near the maximum
The symmetric fumaric acid derivative 6 forms concurrent
NH‚‚‚O and CH‚‚‚O hydrogen bonds in the solid state with
distances similar to those observed in 1 and 2. The melting point
(70) Arnett, E. M.; Mitchell, E. J.; Murty, T. S. S. R. J. Am. Chem. Soc. 1974,
96, 3875-3891.
(71) Etter, M. C. J. Phys. Chem. 1991, 95, 4601-4610.
(72) Etter, M. C. Acc. Chem. Res. 1990, 23, 120-126.
(73) Creighton, T. E. Proteins, Structure and Molecular Properties, 2nd ed.;
W. H. Freeman: New York, 1984; p 183.
9
J. AM. CHEM. SOC. VOL. 124, NO. 38, 2002 11321

A R T I C L E S
Baures et al.
in DMSO-d6, with DMSO as the internal reference for 1H (δ 2.50) and
DMSO-d6 as the internal reference for 13C (δ 39.50). For 5, a small
volume of MeOH-d4 was added to the CDCl3 in order to aid solubility.
4-Amino-4-oxo-(E)-2-butenoic Acid Ethyl Ester (1). Monoethyl
fumarate (0.200 g, 1.39 mmol) was added to 10 mL of benzene, thionyl
chloride (0.10 mL, 1.37 mmol)) was added dropwise with stirring, and
the resulting solution was refluxed for 12 h. The solvents were removed
under vacuum, and residual SOCl2 was removed by azeotropic
formation with added CH2Cl2 (3×) that was removed under vacuum.
The resulting oil was dissolved in THF, and 0.40 mL of ammonia in
THF (∼30% v/v) was added dropwise at 0 °C. The reaction was stirred
for 5 min at room temperature before removal of THF under vacuum.
The resulting material was partitioned between EtOAc and a 10% citric
acid solution. The organic fraction was washed consecutively with a 1
M NaHCO3 solution, water, and brine. The solution was dried (MgSO4)
and filtered, and the solvent was removed under vacuum to give 0.195
g of 1 (98%). An analytical sample was obtained by crystallization
from ethyl acetate/hexanes: mp 91-94 °C [lit. mp 94 °C (ref 81), 93-
of 6 (202-204 °C) is significantly higher in comparison to 7
(106-108 °C), though it is possible that 7 adopts a folded solid-
state conformation, on the basis of the X-ray structures of ring-
constrained analogues.74
The CH‚‚‚O hydrogen bonds in these crystal structures can
be defined as either attractive and stabilizing, repulsive and
stabilizing, or repulsive and destabilizing.8,75 It is reasonable to
expected that the interactions are stabilizing, but it is not possible
to know whether they are attractive or repulsive in the solid
state. Other intermolecular forces such as NH‚‚‚O hydrogen
bonds or aromatic ring interactions with themselves or other
functional groups may influence the CH‚‚‚O hydrogen bond and
manifest these influences on the distances and angles associated
with the CH‚‚‚O contact. Indeed, significant energy is associated
with the solid-state packing of aromatic rings75 and the
interaction of aromatic rings with π-CHs has also been shown
to influence conformation in small molecules.76,77 The evidence
for participation by the various C-H groups in the DMSO-d6
titrations is consistent with the observation of similar solid-
state CH‚‚‚O hydrogen bonds, although definitive assignments
of alkene chemical shifts in 1 and 2 were not made in these
experiments.
1
95 °C (ref 82)]; H NMR (10 mM in CDCl3) δ 6.96 (d, J ) 15.6 Hz,
1 H), 6.83 (d, J ) 15.6 Hz, 1 H), 5.66 (br s, 2 H), 4.26 (q, J ) 7.2 Hz,
2 H), 1.32 (t, J ) 7.2 Hz, 3 H); 1H NMR (10 mM in DMSO-d6) δ 7.90
(br s, 1 H), 7.51 (br s, 1 H), 6.96 (d, J ) 15.6 Hz, 1 H), 6.56 (d, J )
15.6 Hz, 1 H), 4.18 (q, J ) 7.2 Hz, 2 H), 1.24 (t, J ) 7.2 Hz, 3 H); 13
C
NMR (CDCl3) δ 166.2, 165.5, 135.7, 131.3, 61.3, 14.0; FAB MS m/z
144 [M + H]+. Anal. Calcd for C6H9NO3: C, 50.34; H, 6.34; N, 9.79.
Found: C, 50.09; H, 6.32; N, 9.57.
Conclusion
The formation of a CH‚‚‚O hydrogen bond concurrent with
NH‚‚‚O hydrogen bonding is a motif previously reported to
occur in the solid-state cocrystal between barbital and acet-
amide.78,79 Replacement of a NH‚‚‚O hydrogen bond with a
CH‚‚‚O hydrogen bond within the context of an intermolecular
dimer has reported80 that could thereby serve as an isofunctional
replacement for some NH‚‚‚O hydrogen bonds. The participation
of C-H groups in hydrogen bonding warrants further empirical
studies aimed at determining the strength of these interactions,
thereby better understanding their contributions to molecular
recognition and protein structure. In this way, the fumaramides
and related compounds represent models of the N-acylpeptide
bond that are useful for both solution and solid-state studies of
CH‚‚‚O hydrogen bonds.
4-Oxo-4-[(phenylmethyl)amino]-(E)-2-butenoic Acid Ethyl Ester
(2). Monoethyl fumarate (2.000 g, 13.9 mmol) was added to 100 mL
of benzene, thionyl chloride (3.00 mL, 41.1 mmol) was added dropwise
with stirringand the resulting solution was refluxed for 17 h. The
solvents were removed under vacuum, and residual SOCl2 was removed
by azeotropic formation with added CH2Cl2 (3×) that was removed
under vacuum. The resulting oil was dissolved in toluene and cooled
to -78 °C under Ar. A solution of benzylamine (3.04 mL, 27.8 mmol)
in toluene was added dropwise and the solution was stirred for 2 h at
-78 °C. The toluene was removed under vacuum, and the residue was
partitioned between EtOAc and a 10% citric acid solution. The organic
fraction was washed consecutively with a 1 M NaHCO3 solution, water,
and brine. The solution was dried (MgSO4) and filtered, and the solvent
was removed under vacuum to give 1.777 g of 2 (55%). An analytical
sample was obtained by crystallization from ethyl acetate/hexanes: mp
109-110 °C; 1H NMR (10 mM in CDCl3) δ 7.24-7.37 (m, 5 H), 6.90
(d, J ) 15.6 Hz, 1 H), 6.85 (d, J ) 15.6 Hz, 1 H), 6.02 (br s, 1 H),
4.55 (d, J ) 5.2 Hz), 4.24 (q, J ) 7.2 Hz, 2 H), 1.31 (t, J ) 7.2 Hz,
Experimental Section
General Methods. All apparatus were oven-dried and cooled in a
desiccator. Reagent-grade THF and CH2Cl2 were distilled from sodium
benzophenone ketyl and CaH2, respectively, before use. An ammonia
in THF solution was prepared by condensing gaseous ammonia to yield
an approximate 30% (v/v) solution in THF. Diethyl fumarate and all
other reagents were purchased from commercial suppliers and used
without purification. Thin-layer chromatography was performed on
Analtech 250 µm silica gel HLF Uniplates and were visualized with
UV, I2, and ninhydrin spray for amines. Melting points were obtained
on a Fisher-Johns melting point apparatus and are uncorrected.
Elemental analyses were performed by Desert Analytics in Phoenix,
1
3 H); H NMR (10 mM in DMSO-d6) δ 9.02 (s, 1 H), 7.24-7.35 (m,
5 H), 7.06 (d, J ) 15.6 Hz, 1 H), 6.62 (d, J ) 15.6 Hz, 1 H), 4.39 (d,
J ) 6.0 Hz), 4.19 (q, J ) 7.2 Hz, 2 H), 1.24 (t, J ) 7.2 Hz, 3 H); 13
C
NMR (CDCl3) δ 165.8, 163.6, 137.4, 136.4, 130.3, 128.7, 127.9, 127.6,
61.1, 43.9, 14.0; FAB MS m/z 234 [M + H]+. Anal. Calcd for C13H15-
NO3: C, 66.93; H, 6.48; N, 6.01. Found: C, 66.73; H, 6.48; N, 6.01.
N-(Phenylacetyl)glycine Ethyl Ester (3). Phenylacetic acid (2.000
g, 14.7 mmol) was added to 100 mL of benzene, thionyl chloride (2.14
mL, 29.4 mmol) was added dropwise with stirring, and the resulting
solution was refluxed for 22 h. The solvents were removed under
vacuum, and residual SOCl2 was removed by azeotropic formation with
added CH2Cl2 (3×) that was removed under vacuum. The resulting oil
was dissolved in THF and added dropwise to a solution containing
glycine ethyl ester hydrochloride (4.10 g, 29.4 mmol) and diisopro-
pylethylamine (5.12 mL, 29.4 mmol) in THF. After 20 min at room
temperature, the THF was removed under vacuum and the residue was
partitioned between EtOAc and a 10% citric acid solution. The organic
fraction was washed consecutively with a 1 M NaHCO3 solution, water,
1
AZ. H and 13C NMR spectra were measured at 400 and 50.3 MHz,
respectively, either in CDCl3, with CHCl3 as the internal reference for
1H (δ 7.26) and CDCl3 as the internal reference for 13C (δ 77.06), or
(74) Ranganathan, D.; Haridas, V.; Kurur, S.; Thomas, A.; Madhusudanan, K.
P.; Nagaraj, R.; Kunwar, A. C.; Sarma, A. V. S.; Karle, I. L. J. Am. Chem.
Soc. 1998, 120, 8448-8460.
(75) Dunitz, J. D.; Gavezzotti, A. Acc. Chem. Res. 1999, 32, 677-684.
(76) Jennings, W. B.; Farrell, B. M.; Malone, J. F. Acc. Chem. Res. 2001, 34,
885-894.
(77) To´th, G.; Murphy, R. F.; Lovas, S. Protein Eng. 2001, 14, 543-547.
(78) Berkovitch-Yellin, Z.; Leiserowitz, L. J. Am. Chem. Soc. 1980, 102, 7677-
7690.
(81) Loubinoux, B.; Ge´rardin, P.; Kunz, W.; Herzog, J. Pestic. Sci. 1991, 33,
(79) Berkovitch-Yellin, Z.; Leiserowitz, L. Acta Crystallogr., Sect. B. 1984, 40,
159-165.
(80) Schuck, C.; Lex, J. Eur. J. Org. Chem. 2001, 1519-1523.
263-269.
(82) Quiro´s, M.; Astorga, C.; Rebolledo, F.; Gotor, V. Tetrahedron 1995, 51,
7715-7720.
9
11322 J. AM. CHEM. SOC. VOL. 124, NO. 38, 2002

Models of Peptide CH‚‚‚O Hydrogen Bonds
A R T I C L E S
and brine. The solution was dried (MgSO4) and filtered, and the solvent
was removed under vacuum to give 0.342 g of 3 (11%). An analytical
sample was obtained by dissolving the product in EtOAc and allowing
the solution to diffuse against hexanes: mp 79-82 °C [lit. mp 82 °C
(ref 83)]; 1H NMR (10 mM in CDCl3) δ 7.24-7.39 (m, 5 H), 5.89 (br
s, 1 H), 4.18 (q, J ) 7.2 Hz, 2 H), 4.00 (d, J ) 4.8 Hz, 2 H), 3.63 (s,
2 H), 1.26 (t, J ) 7.2 Hz, 3 H); 1H NMR (10 mM in DMSO-d6) δ 8.46
(s, 1 H), 7.21-7.29 (m, 5 H), 4.07 (q, J ) 7.2 Hz, 2 H), 3.82 (d, J )
5.6 Hz, 2 H), 3.40-3.51 (m, 2 H), 1.17 (t, J ) 7.2 Hz, 3 H); 13C NMR
(CDCl3) δ 171.1, 169.7, 134.4, 129.4, 129.0, 127.4, 61.5, 43.4, 41.4,
14.0; FAB MS m/z 222 [M + H]+; Anal. Calcd for C12H15NO3: C,
65.14; H, 6.83; N, 6.33. Found: C, 64.99; H, 6.85; N, 6.21.
H), 4.34-4.40 (m, 2 H), 3.64 (s, 6 H), 1.50-1.65 (m, 6 H), 0.86-
0.92 (m, 12 H); 13C NMR (CDCl3) δ 173.7, 164.3, 133.1, 52.3, 50.9,
40.9, 24.8, 22.8, 21.6; FAB MS m/z 371 [M + H]+. Anal. Calcd for
C18H30N2O6: C, 58.36; H, 8.16; N, 7.56. Found: C, 58.63; H, 7.99; N,
7.47.
1,4-Dioxobutane-1,4-diylbis[N-(S)-leucine Methyl Ester] (7). The
reaction was set up and worked up as described for 6, starting with
0.096 g (0.83 mmol) of succinic acid. Crystallization from CH3OH
provided 157 mg of 7 (51%): mp 106-108 °C; [R]D -49.3 (c 0.75,
1
MeOH); H NMR (10 mM in CDCl3) δ 6.53 (d, J ) 7.6 Hz, 2 H),
4.55-4.61 (m, 2 H), 3.73 (s, 6 H), 2.49-2.62 (m, 4 H), 1.50-1.72
1
(m, 6 H), 0.92-0.94 (m, 12 H); H NMR (10 mM in DMSO-d6) δ
N-[1-Oxo-3-phenyl-2-(E)-propenyl]glycine Ethyl Ester (4). trans-
Cinnamic acid (1.06 g, 7.16 mmol) was added to 20 mL of an CH2Cl2
and glycine ethyl ester hydrochloride (1.000 g, 7.16 mmol), and then
diisopropylethylamine (1.25 mL, 7.16 mmol), and 1-[3-(dimethylami-
no)propyl]-3-ethylcarbodiimide hydrochloride (1.44 g, 7.52 mmol) were
added consecutively. After 24 h, another 20 mL of CH2Cl2 was added
and the solution was extracted with a 10% citric acid solution. The
organic fraction was washed consecutively with a 1 M NaHCO3
solution, water, and brine. The solution was dried (MgSO4) and filtered,
and the solvent was removed under vacuum to a pale yellow solid.
Crystallization from CH3OH provided 365 mg of 4 (22%): mp 107-
108 °C [lit. mp 106-107 °C (ref 63)]; 1H NMR (10 mM in CDCl3) δ
7.66 (d, J ) 15.6 Hz, 1 H), 7.51-7.53 (m, 2 H), 7.37-7.41 (m, 3 H),
6.47 (d, J ) 15.6 Hz, 1 H), 6.13 (br s, 1 H), 4.26 (q, J ) 7.2 Hz, 2 H),
4.18 (d, J ) 5.2 Hz, 2 H), 1.31 (t, J ) 7.2 Hz, 3 H); 1H NMR (10 mM
in DMSO-d6) δ 8.53 (t, J ) 6.0 Hz, 1 H), 7.58-7.60 (m, 2 H), 7.46
(d, J ) 16.0 Hz, 1 H), 7.38-7.43 (m, 3 H), 6.72 (d, J ) 16.0 Hz, 1
H), 4.12 (q, J ) 7.2 Hz, 2 H), 3.96 (d, J ) 6.0 Hz, 2 H), 1.21 (t, J )
7.2 Hz, 3 H); 13C NMR (CDCl3) δ 170.1, 166.1, 141.6, 134.5, 129.7,
128.7, 127.8, 119.8, 61.5, 41.5, 14.0; FAB MS m/z 234 [M + H]+.
Anal. Calcd for C13H15NO3: C, 66.93; H, 6.48; N, 6.01. Found: C,
67.03; H, 6.45; N, 5.99.
N-[1-Oxo-3-phenyl-2-(E)-propenyl)-(S)-valinamide (5). trans-Cin-
namic acid (0.225 g, 1.52 mmol) was added to 5 mL of an ethanol/
water (1:1) mix, and then (S)-valinamide hydrobromide (0.300 g, 1.52
mmol), N-methylmorpholine (0.167 mL, 1.52 mmol), and 1-[3-
(dimethylamino)propyl]-3-ethylcarbodiimide hydrochloride (0.291 g,
1.52 mmol) were added consecutively. After 24 h, CH2Cl2 (20 mL)
was added and the solution was extracted with a 10% citric acid
solution. The organic fraction was washed consecutively with a 1 M
NaHCO3 solution, water, and brine. The solution was dried (MgSO4)
and filtered, and the solvent was removed under vacuum to yield a
solid. Crystallization from CH3OH provided 141 mg of 5 (38%): mp
246-248 °C; [R]D +32.9 (c 0.70, MeOH); 1H NMR (CDCl3/CD3OD)
δ 7.51 (d, J ) 15.6 Hz, 1 H), 7.42-7.45 (m, 2 H), 7.27-7.31 (m, 3
H), 6.49 (d, J ) 15.6 Hz, 1 H), 1.96-2.05 (m, 1 H), 0.88-0.92 (m, 6
H); 13C NMR (DMSO-d6) δ 173.0, 164.9, 138.8, 135.1, 129.4, 129.0,
127.5, 122.4, 57.5, 30.6, 19.4, 18.0; FAB MS m/z 247 [M + H]+. Anal.
Calcd for C14H18N2O2: C, 68.27; H, 7.37; N, 11.38. Found: C, 68.26;
H, 7.36; N, 11.20.
7.21 (d, J ) 8.0 Hz, 2 H), 4.51-4.57 (m, 2 H), 3.71 (s, 6 H), 2.52-
2.60 (m, 4 H), 1.55-1.69 (m, 6 H), 0.91-0.94 (m, 12 H); 13C NMR
(CDCl3) δ 173.8, 172.2, 52.2, 50.9, 40.9, 31.7, 24.7, 22.8, 21.6; FAB
MS m/z 373 [M + H]+; Anal. Calcd for C18H32N2O6: C, 58.04; H,
8.66; N, 7.52. Found: C, 58.28; H, 8.60; N, 7.31.
Screening by Parallel Synthesis. Stock solutions were prepared in
EtOH. The monocarboxylic acid and amino acid salts were prepared
at 0.250 M, whereas N-methylmorpholine in EtOH was prepared at
0.500 M and the dicarboxylic acids were prepared at 0.125 M. Some
samples required heating to fully dissolve. A 0.250 M solution of 1-[3-
(dimethylamino)propyl]-3-ethylcarbodiimide hydrochloride was freshly
prepared in water. The different combinations were prepared by
combining 100 µL each of the amino acid salt solution and acid
solutions, along with 200 µL of the carbodiimide solution and 75 µL
of the N-methylmorpholine solution (except for histidine where 150
µL of base was used). The solutions were allowed to react and slowly
evaporate at room temperature; they were observed periodically for
the formation of crystalline solids. From these experiments, 4-7 were
identified as crystalline and worth further analysis.
Solution NMR Analysis. Stock solutions at 100 mM were prepared
of the compound under analysis in CDCl3 and DMSO-d6. In addition,
a diluted stock (10 mM) in DMSO-d6 was also prepared. Both NMR
solvents contained 0.3% TMS (v/v) as a standard. The stock solutions
were used to create the NMR samples for analysis, containing 5, 10,
20, 50, 100, 200, and 500 mM and 1.0 and 10.0 M DMSO-d6 in CDCl3.
Solutions at 10 mM in CHCl3 as well as DMSO-d6 at 14.1 M that
contained trace CHCl3were also prepared. The CDCl3/DMSO-d6
solutions are uncorrected for any volume differences due to mixing.
X-ray Single-Crystal Diffraction. The crystal data for 1-3, 5, and
6 were collected on a Siemens P4 four-circle diffractometer equipped
with a Bruker SMART 1000 CCD and graphite-monochromated Mo-
KR (λ ) 0.710 73 Å) radiation at either 203 or 293 K. The data were
integrated with SAINT. Crystal stabilities were monitored by measuring
3 standard reflections after every 97 reflections with no significant decay
in observed intensities. A θ - 2θ scanning technique was used for
peak collection with Lorenz and polarization corrections applied.
Hydrogen atom positions were located from difference Fourier maps,
and a riding model with fixed thermal parameters [uij ) 1.2Uij(eq) for
the atom to which they are bonded] was used for subsequent
2
refinements. The weighting function applied was w-1 ) [σ2(Fo ) +
(g1P)2 + (g2P)], where P ) [Fo + 2Fc2]/3. In all structures the
2
1,4-Dioxo-2-(E)-butene-1,4-diylbis[N-(S)-leucine Methyl Ester]
(6). Fumaric acid (0.096 g, 0.83 mmol) was added to 5 mL of an
ethanol/water (1:1) mix, and (S)-leucine methyl ester hydrochloride
(0.300 g, 1.65 mmol), N-methylmorpholine (0.182 mL, 1.65 mmol),
and 1-[3-(dimethylamino)propyl]-3-ethylcarbodiimide hydrochloride
(0.317 g, 1.65 mmol) were added consecutively. Workup was done as
described for 5. Crystallization from CH3OH provided 150 mg of 6
SHELXTL PC and SHELXL-93 packages84 were used for data
reduction, structure solution, and refinement.
Acknowledgment. This work was supported in part by
NIH-COBRE Award 1-P20-RR15563 with matching support
from the state of Kansas and NSF-REU support (0097411)
for W.P.-S.
Supporting Information Available: Figures representing the
solid-state structures of 1, 2, 3, 5, and 6 showing the thermal
ellipsoids and atom labeling (PDF) and X-ray crystallographic
files for 1, 2, 3, 5, and 6 (CIF). This material is available free
of charge via the Internet at http://pubs.acs.org.
1
(49%): mp 202-204 °C; [R]D -79.6 (c 0.65, MeOH); H NMR (10
mM in CDCl3) δ 6.90 (s, 2 H), 6.67 (d, J ) 8.0 Hz, 2 H), 4.71-4.76
(m, 2 H), 3.76 (s, 6 H), 1.56-1.72 (m, 6 H), 0.93-0.95 (m, 12 H); 1H
NMR (10 mM in DMSO-d6) δ 8.81 (d, J ) 7.2 Hz, 2 H), 6.90 (s, 2
(83) Saito, T.; Nishihata, K.; Fukatsu, S. J. Chem. Soc., Perkin Trans. 1 1981,
4, 1058-1063.
(84) Sheldrick, G. M. SHELXL-93, University of Go¨ttingen, Germany.
JA0257366
9
J. AM. CHEM. SOC. VOL. 124, NO. 38, 2002 11323