An extended β-strand mimic for a larger artificial β-sheet

Article abstract of DOI:10.1021/ja963843s

This paper reports the development of β-strand mimic B, which duplicates the hydrogen-bonding functionality of one edge of a tetrapeptide β-strand. When attached to a tripeptide by a suitable linking group, β- strand mimic B forms a hydrogen-bonded antiparallel β-sheet structure, artificial β-sheet 2. β-Strand mimic B is based upon a 5-hydrazino-2- metboxybenzoic acid building block. The first half of the paper describes synthetic, IR and ¹H NMR spectroscopic, X-ray crystallographic, and molecular modeling studies of 5-hydrazino-2-methoxybenzoic acid derivatives and related molecules. These studies establish that hydrazide derivatives of 5-hydrazino-2-methoxybenzoic acid adopt a conformation similar to that of a peptide β-strand and are suitable for use as β-strand mimics. The second half of the paper describes synthetic and ¹H NMR spectroscopic studies of artificial β-sheet 2 and of controls 20 and 21, which resemble the peptidomimetic and peptide strands of 2. These experiments indicate that 2 adopts a conformation and hydrogen-bonding pattern similar to that of an antiparallel β-sheet and establish that β-strand mimic B can induce β- sheet formation in an attached peptide strand.

Full text of DOI:10.1021/ja963843s

J. Am. Chem. Soc. 1997, 119, 5413-5424
5413
An Extended â-Strand Mimic for a Larger Artificial â-Sheet
James S. Nowick,* Mason Pairish, In Quen Lee, Darren L. Holmes, and
Joseph W. Ziller
Contribution from the Department of Chemistry, UniVersity of California, IrVine,
IrVine, California 92697-2025
ReceiVed NoVember 5, 1996X
Abstract: This paper reports the development of â-strand mimic B, which duplicates the hydrogen-bonding
functionality of one edge of a tetrapeptide â-strand. When attached to a tripeptide by a suitable linking group,
â-strand mimic B forms a hydrogen-bonded antiparallel â-sheet structure, artificial â-sheet 2. â-Strand mimic B is
based upon a 5-hydrazino-2-methoxybenzoic acid building block. The first half of the paper describes synthetic, IR
and 1H NMR spectroscopic, X-ray crystallographic, and molecular modeling studies of 5-hydrazino-2-methoxybenzoic
acid derivatives and related molecules. These studies establish that hydrazide derivatives of 5-hydrazino-2-
methoxybenzoic acid adopt a conformation similar to that of a peptide â-strand and are suitable for use as â-strand
mimics. The second half of the paper describes synthetic and 1H NMR spectroscopic studies of artificial â-sheet 2
and of controls 20 and 21, which resemble the peptidomimetic and peptide strands of 2. These experiments indicate
that 2 adopts a conformation and hydrogen-bonding pattern similar to that of an antiparallel â-sheet and establish
that â-strand mimic B can induce â-sheet formation in an attached peptide strand.
Introduction
to peptide strands by suitable linking groups, this compound
serves as a template to induce â-sheet formation. By measuring
the relative stabilities of these â-sheets, Kemp and co-workers
were able to determine the â-sheet forming propensities of
different amino acids. Two years later, Martin and co-workers
developed 1,2,3-trisubstituted cyclopropanes as peptide isosteres
that mimic the â-strand conformation of peptides bound by
proteolytic enzymes.4 Renin and collagenase inhibitors incor-
porating these â-strand mimics were prepared and studied.
Smith, Hirschmann, and co-workers invented 3,5-linked pyr-
roline-4-ones, which duplicate the main-chain conformation and
side-chain placement of â-strands, and applied these compounds
to the creation of peptidomimetic inhibitors of HIV protease
and renin.5 At the beginning of 1996, Michne and Schroeder
reported that a 3-amino-4(1H)-quinoline template, which is
The mimicry of peptide and protein structures has emerged
as a focal point of bioorganic and medicinal chemistry. This
emergence has been marked by the appearance of at least 15
review articles since 1990 on different aspects of peptidomimetic
chemistry.1 As a result of the recent activity in this field, there
is now a growing appreciation that peptidomimetic chemistry
can serve as a basis for drug development and that peptidomi-
metic model systems can provide fundamental insights into the
folding and structure of proteins.
Recently, compounds that mimic the conformational and
hydrogen-bonding properties of peptide â-strands have begun
to attract attention. In 1988, Kemp and Bowen reported that
2,8-diaminoepindolidione duplicates the hydrogen-bonding func-
tionality of one edge of a peptide â-strand.1a,2,3 When attached
(3) (a) Kemp, D. S.; Bowen, B. R. In Protein Folding: Deciphering the
Second Half of the Genetic Code; Gierasch, L. M., King, J., Eds.; American
Association for the Advancement of Science: Washington, DC, 1990; pp
293-303. (b) Kemp, D. S.; Muendel, C. C.; Blanchard, D. E.; Bowen, B.
R. In Peptides: Chemistry, Structure and Biology: Proceedings of the
EleVenth American Peptide Symposium; Rivier, J. E., Marshall, G. R., Eds.;
ESCOM: Leiden, 1990; pp 674-676. (c) Kemp, D. S.; Blanchard, D. E.;
Muendel, C. C. In Peptides: Chemistry, Structure and Biology: Proceedings
of the Twelfth American Peptide Symposium; Smith, J. A., Rivier, J. E.,
Eds.; ESCOM: Leiden, 1992; pp 319-322.
(4) (a) Martin, S. F.; Austin, R. E.; Oalmann, C. J. Tetrahedron Lett.
1990, 31, 4731. (b) Martin, S. F.; Austin, R. E.; Oalmann, C. J.; Baker, W.
R.; Condon, S. L.; deLara, E.; Rosenberg, S. H.; Spina, K. P.; Stein, H. H.;
Cohen, J.; Kleinert, H. D. J. Med. Chem. 1992, 35, 1710. (c) Baker, W. R.;
Jae, H.-S.; Martin, S. F.; Condon, S. L.; Stein, H. H.; Cohen, J.; Kleinert,
H. D. Bioorg. Med. Chem. Lett. 1992, 2, 1405. (d) Martin, S. F.; Oalmann,
C. J.; Liras, S. Tetrahedron 1993, 49, 3521.
(5) (a) Smith, A. B., III; Keenan, T. P.; Holcomb, R. C.; Sprengeler, P.
A.; Guzman, M. C.; Wood, J. L.; Carroll, P. J.; Hirschmann, R. J. Am.
Chem. Soc. 1992, 114, 10672. (b) Smith, A. B., III; Holcomb, R. C.;
Guzman, M. C.; Keenan, T. P.; Sprengeler, P. A.; Hirschmann, R.
Tetrahedron Lett. 1993, 34, 63. (c) Smith, A. B., III; Hirschmann, R.;
Pasternak, A.; Akaishi, R.; Guzman, M. C.; Jones, D. R.; Keenan, T. P.;
Sprengeler, P. A.; Darke, P. L.; Emini, E. A.; Holloway, M. K.; Schleif,
W. A. J. Med. Chem. 1994, 37, 215. (d) Smith, A. B., III; Guzman, M. C.;
Sprengeler, P. A.; Keenan, T. P., Holcomb, R. C.; Wood, J. L.; Carroll, P.
J.; Hirschmann, R. J. Am. Chem. Soc. 1994, 116, 9947. (e) Smith, A. B.;
Akaishi, R.; Jones, D. R.; Keenan, T. P.; Guzman, M. C.; Holcomb, R. C.;
Sprengeler, P. A.; Wood, J. L.; Hirschmann, R.; Holloway, M. K.
Biopolymers 1995, 37, 29. (f) Smith, A. B.; Hirschmann, R.; Pasternak,
A.; Guzman, M. C.; Yokoyama, A.; Sprengeler, P. A.; Darke, P. L.; Emini,
E. A.; Schleif, W. A. J. Am. Chem. Soc. 1995, 117, 11113.
X Abstract published in AdVance ACS Abstracts, May 15, 1997.
(1) Kemp, D. S. Trends Biotechnol. 1990, 8, 249. (b) Ho¨lzemann, G.
Kontakte 1991, 3. (c) Ho¨lzemann, G. Kontakte 1991, 55. (d) Hirschmann,
R. Angew. Chem., Int. Ed. Engl. 1991, 30, 1278. (e) Olson, G. L.; Bolin,
D. R.; Bonner, M. P.; Bo¨s, M.; Cook, C. M.; Fry, D. C.; Graves, B. J.;
Hatada, M.; Hill, D. E.; Kahn, M.; Madison, V. S.; Rusiecki, V. K.; Sarabu,
R.; Sepinwall, J.; Vincent, G. P.; Voss, M. E. J. Med. Chem. 1993, 36,
3039. (f) Wiley, R. A.; Rich, D. H. Med. Res. ReV. 1993, 13, 327. (g)
Chorev, M.; Goodman, M. Acc. Chem. Res. 1993, 26, 266. (h) Zuckermann,
R. N. Curr. Opin. Struct. Biol. 1993, 3, 580. (i) Kahn, M. Synlett 1993,
821. (j) Giannis, A.; Kolter, T. Angew. Chem., Int. Ed. Engl. 1993, 32,
1244. (k) Gante, J. Angew. Chem., Int. Ed. Engl. 1994, 33, 1699. (l) Liskamp,
R. M. J. Recl. TraV. Chim. Pays-Bas 1994, 113, 1. (m) Adang, A. E. P.;
Hermkens, P. H. H.; Linders, J. T. M.; Ottenheijm, H. C. J.; van Staveren,
C. J. Recl. TraV. Chim. Pays-Bas 1994, 113, 63. (n) Beeley, N. Trends
Biotechnol. 1994, 12, 213. (o) Schneider, J. P.; Kelly, J. W. Chem. ReV.
1995, 95, 2169. (p) Goodman, M.; Shao, H. Pure Appl. Chem. 1996, 68,
1303. (q) Nowick, J. S.; Smith, E. M.; Pairish, M. Chem. Soc. ReV. 1996,
25, 401.
(2) (a) Kemp, D. S.; Bowen, B. R. In Tetrahedron Lett. 1988, 29, 5077.
(b) Kemp, D. S.; Bowen, B. R. Tetrahedron Lett. 1988, 29, 5081. (c) Kemp,
D. S.; Bowen, B. R.; Muendel, C. C. J. Org. Chem. 1990, 55, 4650.
S0002-7863(96)03843-7 CCC: $14.00 © 1997 American Chemical Society

5414 J. Am. Chem. Soc., Vol. 119, No. 23, 1997
Nowick et al.
related to Kemp’s 2,8-diaminoepindolidione, mimics a â-strand
and induces a â-strand conformation in an adjacent peptide
strand.6 Molecules designed as peptidomimetic decoys that
duplicate a portion of the intracellular adhesion molecule 1
(ICAM-1) and contain the 3-amino-4(1H)-quinoline template
were synthesized; however, the desired biological activity was
not detected. Pallai, Rebek, and co-workers described bicyclic
vinylamides, which mimic aspects of the main-chain and side-
chain functionality of a peptide â-strand, and used these
molecules to mimic a four-residue segment of the CD4 receptor.7
Very recently, Schrader and Kirsten reported that 3-aminopy-
razole stabilizes a â-strand conformation in dipeptides by means
of intermolecular hydrogen bonding.8 Many other peptidomi-
metic compounds that duplicate different aspects of peptide
strand structure have been developed. These include azapep-
tides,9 vinylogous polypeptides,10 azatides,11 and various other
unnatural biopolymers.12
Several years ago, we became interested in combining
â-strand mimics, other peptidomimetic templates, and peptide
strands to create compounds that mimic the structure and
hydrogen-bonding patterns of â-sheets. We envision these
artificial â-sheets as model systems in which to study â-sheet
structure and stability.1q,13 From these model studies, we hope
to gain insight into protein folding, enzyme-substrate interac-
tions, and â-amyloid deposition in Alzheimer’s disease. We
are also interested in developing artificial â-sheets as building
blocks for molecular receptors and catalysts.
developed by other researchers meets these requirements.
â-Strand mimic A provides the right pattern of hydrogen-
bonding groups along one edge and is blocked along the other
edge, but it is too short. We reasoned that it might be possible
to extend the 5-amino-2-methoxybenzamide â-strand mimic by
linking two or more of these units end-to-end, by the nitrogen
atoms, to generate 5-hydrazino-2-methoxybenzoic acid â-strand
mimic B. This paper reports the development of this extended
â-strand mimic and its successful application to the creation of
artificial â-sheet 2.
Since 1992, we have reported the development of peptido-
mimetic templates, which we have termed molecular scaffolds.14
These templates are based upon a simple oligourea structure
and resemble, in some ways, â-turns. The oligourea molecular
scaffolds are designed to hold two or more peptide strands in
proximity and induce the formation of hydrogen-bonded â-sheet
structures. In 1995, we reported an artificial parallel â-sheet
containing a diurea molecular scaffold and two attached peptide
strands.15 Early in 1996, we described artificial antiparallel
â-sheet 1, in which 5-amino-2-methoxybenzamide â-strand
mimic A and a diurea molecular scaffold stabilize â-sheet
structure in an attached peptide strand.16
With the goal of preparing larger artificial â-sheets, we
required an extended â-strand mimic. We wanted the â-strand
mimic to duplicate the hydrogen-bonding functionality of one
edge of a peptide â-strand and lack hydrogen-bonding func-
tionality along the other edge. None of the â-strand mimics
Results and Discussion
The first half of this section describes synthetic and structural
studies of methoxy-substituted derivatives of hydrazinobenzoic
acid. These studies establish that hydrazide derivatives of
5-hydrazino-2-methoxybenzoic acid adopt a conformation simi-
lar to that of a peptide â-strand and are suitable for use as
â-strand mimics. The second half describes the coupling of
5-hydrazino-2-methoxybenzoic acid â-strand mimic B to a
tripeptide strand by means of an ethylenediamine diurea
molecular scaffold to form artificial â-sheet 2. Structural studies
indicate that 2 adopts a hydrogen-bonded conformation similar
to that of an antiparallel â-sheet and establish that â-strand
mimic B can induce â-sheet formation in an attached peptide
strand.
(6) Michne, W. F.; Schroeder, J. D. Int. J. Peptide Protein Res. 1996,
47, 2.
(7) Boumendjel, A.; Roberts, J. C.; Hu, E.; Pallai, P. V.; Rebek, J., Jr.
J. Org. Chem. 1996, 61, 4434.
(8) Schrader, T.; Kirsten, C. Chem. Commun. 1996, 2089.
(9) Gante, J. Synthesis 1989, 405.
(10) Hagihara, M.; Anthony, N. J.; Stout, T. J.; Clardy, J.; Schreiber, S.
L. J. Am. Chem. Soc. 1992, 114, 6568.
(11) Han, H.; Janda, K. D. J. Am. Chem. Soc. 1996, 118, 2539.
(12) (a) Liskamp, R. M. J. Angew. Chem., Int. Ed. Engl. 1994, 33, 633.
(b) Moran, E. J.; Wilson, T. E.; Cho, C. Y.; Cherry, S. R.; Schultz, P. G.
Biopolymers 1995, 37, 213.
(13) (a) Nowick, J. S.; Lee, I. Q.; Mackin, G.; Pairish, M.; Shaka, A. J.;
Smith, E. M.; Ziller, J. W. In Molecular Design and Bioorganic Catalysis;
Wilcox, C. S., Hamilton, A. D., Eds.; Kluwer Academic: Dordrecht,
Netherlands, 1996; pp 111-136. (b) Smith, E. M.; Pairish, M.; Nowick, J.
S. In Self-Assembly in Synthetic Chemistry; Wuest, J. D., Ed.; Kluwer
Academic: Dordrecht, Netherlands, in press.
(14) (a) Nowick, J. S.; Powell, N. A.; Martinez, E. J.; Smith, E. M.;
Noronha, G. J. Org. Chem. 1992, 57, 3763. (b) Nowick, J. S.; Abdi, M.;
Bellamo, K. A.; Love, J. A.; Martinez, E. J.; Noronha, G.; Smith, E. M.;
Ziller, J. W. J. Am. Chem. Soc. 1995, 117, 89. (c) Nowick, J. S.; Mahrus,
S.; Smith, E. M.; Ziller, J. W. J. Am. Chem. Soc. 1996, 118, 1066.
(15) Nowick, J. S.; Smith, E. M.; Noronha, G. J. Org. Chem. 1995, 60,
7386.
Design of the â-Strand Mimic. We envisioned m-hydrazi-
nobenzoic acid as a building block for extended â-strand mimics.
If the hydrazinobenzoic acid units are coupled by means of
hydrazide linkages, an oligomeric â-strand mimic is formed.
In this oligomer, each hydrazinobenzoic acid unit is equivalent
in length to the main-chain of a dipeptide (six atoms), and the
CdO and NH groups are in the same positions as the CdO
and NH groups of one edge of a peptide â-strand. Chart 1
shows the relationship between this oligomer and a peptide
strand; key hydrogen-bonding groups are highlighted in bold-
face.
(16) Nowick, J. S.; Holmes, D. L.; Mackin, G.; Noronha, G.; Shaka, A.
J.; Smith, E. M. J. Am. Chem. Soc. 1996, 118, 2764.

An Extended â-Strand Mimic for a Larger Artificial â-Sheet
J. Am. Chem. Soc., Vol. 119, No. 23, 1997 5415
Table 1. Spectroscopic Properties of NH Groups of Hydrazides 7
and 8 (22 °C, 10 mM in CHCl3 or CDCl3)
Chart 1
IR (cm-1
NHa
)
1H NMR (ppm)
compd
NHb
Ha
Hb
7a
7b
7c
8a
8b
8c
3436
3437
3437
3402
3402
3404
3329
3344
3338
3336
3352
3342
7.94
7.95
7.89
9.45
9.43
9.43
6.30
6.79
6.62
6.40
6.85
6.71
Chart 2
was brominated using N-bromosuccinimide (NBS).17 Halogen-
metal exchange using 2-3 equiv of n-butyllithium, followed
by reaction of the resulting organolithium compounds with di-
tert-butylazodicarboxylate, gave di-tert-butyloxycarbonyl (Boc)
protected hydrazinobenzoic acids 6a-c.18 Although the conver-
sions proceed in modest yields (typically 30-50%), the 5-hy-
drazino-2-methoxybenzoic acid derivative (6a) is readily purified
to analytical purity by precipitation and washing, making its
preparation especially convenient. 1H NMR analysis of 6a-c
is complicated by the presence of multiple conformers that
interconvert slowly on the NMR time scale. The Boc-protected
derivatives 6a-c can be converted to hydrazinobenzoic acids
3a-c (as the mono- or dihydrochloride salts) by treatment with
a mixture of concentrated aqueous HCl and isopropyl alcohol.
1
Scheme 1
The H NMR spectra of these salts show single conformers,
simplifying their spectroscopic analysis.
IR and 1H NMR Spectroscopic Studies of Hydrazinoben-
zoic Acid Derivatives. Hydrazides 7a-c and 8a-c were
synthesized and studied spectroscopically, to determine the
effects of methoxy substituents on the conformations of m-
hydrazinobenzoic acid derivatives. tert-Butyloxycarbonyl com-
pounds 6a-c were converted to hydrazides 7a-c and 8a-c by
successive methyl esterification with diazomethane, removal of
the tert-butyloxycarbonyl groups with HCl, and acylation with
either benzoyl chloride or o-anisoyl chloride.
We hypothesized that methoxy substitutuents ortho to either
the carboxyl group, the hydrazino group, or both might enforce
an extended conformation by means of intramolecular hydrogen
bonding. Thus, we considered oligomers of 5-hydrazino-2-
methoxybenzoic acid, 3-hydrazino-4-methoxybenzoic acid, and
5-hydrazino-2,4-dimethoxybenzoic acid for further study. Chart
2 illustrates these structures and the hypothesized patterns of
hydrogen bonding.
Compounds 7 and 8 exhibit significant differences in their
infrared and 1H NMR spectra in CHCl3 or CDCl3 solution (Table
1). These studies were performed at 10 mM, a concentration
at which little or no self-association occurs. In the infrared
spectrum, the amidic NHa stretch appears as a band centered at
ca. 3437 cm-1 in benzoyl derivatives 7a-c, and at ca. 3403
cm-1 in o-anisoyl derivatives 8a-c. The weakened NH stretch
in o-anisoyl derivatives 8 provides evidence that the o-methoxy
group of the anisoyl fragment is hydrogen bonded to the amide
The Cambridge Structural Database provided us with insight
into some of the features of these structures. 2-Methoxy-
benzamides adopt intramolecularly hydrogen bonded conforma-
tions. Various hydrazides in the database adopt either a linear
or a twisted geometry about the N-N bond. No data on the
effect of o-methoxy substituents upon the conformation of
aromatic hydrazides is available. To gain further insight into
the structures of methoxy-substituted m-hydrazinobenzoic acid
derivatives, we turned to synthetic and structural studies.
1
NH group. In the H NMR spectrum, the amidic NHa group
appears at 7.89-7.95 ppm in 7a-c and at 9.43-9.45 ppm in
8a-c. The substantial downfield shift of this resonance in 8a-c
provides further evidence for hydrogen bonding of the o-
methoxy group.
Synthesis of Hydrazinobenzoic Acid Derivatives. Deriva-
tives of 5-hydrazino-2-methoxybenzoic acid (3a), 3-hydrazino-
4-methoxybenzoic acid (3b), and 5-hydrazino-2,4-dimethoxy-
benzoic acid (3c) were prepared for further study. These
compounds were synthesized from commercially available
2-methoxy-, 4-methoxy-, and 2,4-dimethoxybenzoic acids 4a-c
as shown in Scheme 1. The 2- and 4-methoxybenzoic acids
were brominated using bromine; the 2,4-dimethoxy compound
The IR stretching frequencies of the NHa groups of com-
pounds 7a-c differ little, as do the 1H NMR chemical shifts of
1
Ha in these molecules. The IR and H NMR properties of Ha
(17) Auerbach, J.; Weissman, S. A.; Blacklock, T. J.; Angeles, M. R.;
Hoogsteen, K. Tetrahedron Lett. 1993, 34, 931.
(18) Demers, J. P.; Klaubert, D. H. Tetrahedron Lett. 1987, 28, 4933.

5416 J. Am. Chem. Soc., Vol. 119, No. 23, 1997
Nowick et al.
Figure 1. X-ray crystallographic structures of hydrazides 7a,b and 9-13. Hydrazide 7b has three crystallographically-independent molecules per
asymmetric unit. The molecules differ little in geometry, and only one is shown above. Hydrazide 11 has two crystallographically-independent
molecules per asymmetric unit. The molecules differ little in geometry, and only one is shown above.
Chart 3
in compounds 8a-c are also similar to each other. These data
indicate that methoxy groups at the 4-positions (R2) of these
molecules do not hydrogen bond to the amidic hydrogens (Ha).
Collectively, these IR and 1H NMR studies indicate that
2-methoxy groups help enforce the desired extended conforma-
tion in chloroform solution, while 4-methoxy groups do not.
X-ray Crystallographic Studies of Hydrazinobenzoic Acid
Derivatives. X-ray crystallography provides insight into the
conformations of the methoxy-substituted hydrazinobenzoic acid
derivatives in the solid state. We prepared and attempted to
crystallize a variety of methoxy-substituted m-hydrazinobenzoic
acid derivatives. Of these derivatives, compounds 7a, 7b, and
9-13 afforded crystals suitable for X-ray crystallography. The
X-ray crystallograpic structures of these compounds are shown
in Figure 1.
bonded to this group (Chart 3, structure b). When a methoxy
group is at the 4-position of an aromatic ring, a hydrazide group
at the 3- or 5-position is rotated so that the amine NH group is
directed toward the methoxy group (Chart 3, structure c). These
observations are consistent with the solution-phase studies
described above and indicate that methoxy substitution at the
2-position of the hydrazinobenzoic acids stabilizes a desired
conformation, while methoxy substitution at the 4-position
stabilizes an undesired conformation.
X-ray crystallography provides insight into the hydrogen-
bonding properties of the hydrazinobenzoic acid derivatives as
well as their conformations. Thus, hydrazide derivatives 7a,
12, and 13 form hydrogen-bonded dimers in which two
molecules are antiparallel to each other, and the carbonyl oxygen
atoms and the amine NH groups participate in hydrogen-bonded
ten-membered rings (Figure 2). Because of the antiparallel
orientations of the molecules and the geometry of the hydrogen-
bonded rings, these dimers resemble antiparallel â-sheets. In
each of these dimers, the amine groups lack ortho methoxy
substitution. In contrast, hydrazides 7b, 9, 10, and 11 do not
form hydrogen-bonded dimers of this sort and have methoxy
substituents ortho to the amine NH of the hydrazide group.
These observations indicate that methoxy-substituted m-hy-
drazinobenzoic acid derivatives form â-sheetlike hydrogen-
bonded dimers if there is a methoxy substituent at the 2-position,
but not if there is a methoxy substituent at the 4-position. The
Three trends are evident from these structures. The hydrazide
groups are twisted, with C-N-N-C torsion angles ranging
from 70° to 97°. Structure a in Chart 3 summarizes this finding
pictorially. When a methoxy group is at the 2-position of an
aromatic ring, an amide group at the 1-position is hydrogen
1
X-ray studies as well as the IR and H NMR studies indicate
that the 5-hydrazino-2-methoxybenzoic acid building block may

An Extended â-Strand Mimic for a Larger Artificial â-Sheet
J. Am. Chem. Soc., Vol. 119, No. 23, 1997 5417
Figure 2. Hydrogen-bonded dimers in the X-ray crystallographic structures of hydrazides 7a, 12, and 13.
Figure 3. Model of a peptide strand docked to hydrazide 14. The model for hydrazide 14 was generated from the crystallographic coordinates of
7a with C-N-N-C torsion angles of -82° and 82°, respectively. The geometry of 14 was held fixed, while Ac-Ala-Ala-Ala-NHMe was docked
to it and energy minimized using MacroModel V5.0 with the AMBER* force field.
be suitable for use as a â-strand mimic, while the 3-hydrazino-
4-methoxybenzoic acid and 5-hydrazino-2,4-dimethoxybenzoic
acid building blocks are not.
Molecular Modeling Studies. Many other hydrazides that
were prepared did not yield crystals suitable for X-ray crystal-
lography. Of these, we were particularly interested in hydrazide
14, because the IR, 1H NMR, and X-ray crystallographic studies
Using the crystallographic coordinates of 13, we created a
model for 14 with C-N-N-C torsion angles of -82° and 82°,
respectively. In this model, the distance between the carbonyl
groups is 7 Å, which is comparable to that of a peptide in a
â-strand conformation. Docking a peptide strand (Ac-Ala-Ala-
Ala-NHMe) to this model reveals excellent complementarity
between the peptidomimetic template and the peptide strand;
after energy minimization, the peptide strand adopts a â-strand
conformation and hydrogen bonds to the template (Figure 3).
These modeling studies suggest that hydrazide derivatives of
5-hydrazino-2-methoxybenzoic acid are suitable templates for
inducing a â-strand conformation in peptides.
Synthesis of Artificial â-Sheet 2. To further evaluate the
5-hydrazino-2-methoxybenzoic acid template, we combined this
unit with a 1,2-diaminoethane diurea molecular scaffold and a
tripeptide strand to form artificial â-sheet 2. Artificial â-sheet
2 was synthesized as shown in Scheme 2. Amine 15 was
converted to urea 16 by reaction with phenylalanylisoleucylleu-
cine methyl ester isocyanate19 and subsequent aminolysis of the
peptide methyl ester group by treatment with methylamine in
methanol. The Boc protective group was removed by treatment
with HCl in methanol, and the free amine was formed upon
treatment of the resulting amine hydrochloride salt with aqueous
base. Reaction with isocyanate 17 then afforded urea 18.
Hydrogenolysis of the benzyl ester protective group, followed
by coupling of the resulting carboxylic acid with N-methyl-5-
hydrazino-2-methoxybenzamide hydrochloride (19) using 1-ethyl-
of hydrazides 7-13 suggested that it should adopt a â-strand
conformation and be complementary to a tripeptide in length
and hydrogen-bonding functionality. In the absence of a
crystallographic structure of this compound, we turned to
molecular modeling to predict its structure and examine its
complementarity to peptides.
Molecular mechanics calculations using MacroModel V5.0
and the AMBER* force field predict the hydrazide groups of
14 to have 180° C-N-N-C dihedral angles. Adequate
parameters for this group are lacking, however, and this value
is unreliable in light of the overwhelming body of crystal-
lographic evidence that a geometry of ca. 80° or 90° is preferred.
Ab initio calculations on N′-phenylformohydrazide (PhNHNH-
CHO) at the 6-31G* level using Spartan corroborate the
crystallographic studies and indicate that a 90° C-N-N-C
dihedral angle is preferred.
(19) Nowick, J. S.; Holmes, D. L.; Noronha, G.; Smith, E. M.; Nguyen,
T. M.; Huang, S.-L. J. Org. Chem. 1996, 61, 3929.

5418 J. Am. Chem. Soc., Vol. 119, No. 23, 1997
Nowick et al.
Table 2. 1H NMR Chemical Shifts of the NH Protons of 2, 20,
and 21 (30 °C, 1 mM in CDCl3)
Scheme 2
Ha
Hb
Hc
Hd
He
Hf
Hg
Hh
2
20
21
9.95
6.27
9.55
9.54
8.45
6.41
8.10
7.87
4.80
8.14
6.07
8.02
4.50
6.48
7.52
6.87
Table 3. 1H NMR Chemical Shifts of the Aromatic Protons of 2
and 20 (30 °C, 1 mM in CDCl3)
Hi
Hj
Hk
Hl
Hm
Hn
2
20
8.66
7.66
8.40
8.08
6.98
6.99
8.16
7.84
7.07
7.02
6.90
6.88
protons are fully hydrogen bonded and that 2 predominantly
adopts a hydrogen-bonded â-sheet conformation. Amide proton
Hg appears at 6.07 ppm in 2, indicating that it is not hydrogen
bonded. In control 21, however, Hg appears downfield at 7.52
ppm. We interpret the shift of Hg in 21 to mean that it forms
a hydrogen-bonded â-turn with the carbonyl of the urea group
(vide infra), a phenomenon which we have observed in other
peptide ureas.15,16 In summary, these shift data suggest that in
2 the â-strand mimic hydrogen bonds to the peptide strand,
forcing it to adopt an extended conformation and preventing it
from adopting a hydrogen-bonded â-turn conformation.
The aromatic ring protons of artificial â-sheet 2 also differ
in chemical shift from those of control 20. â-Strand mimic
proton Hi of artificial â-sheet 2 is shifted 1.00 ppm downfield
of the corresponding proton in control 20 (Table 3). The
downfield shifting of both this proton and Ha apparently results
from proximity to the carbonyl group of the urea on the bottom
half of 2; that the shifts are exceptionally large suggests that
the carbonyl group is exceptionally close to both protons.22
â-Strand mimic protons Hj and Hl are each shifted downfield
by 0.32 ppm in 2. These shifts may reflect enforced proximity
to the carbonyl groups of the upper urea and the isoleucine.
Similar downfield shifting has been observed in other peptido-
mimetic â-sheets,1a,3 and in protein â-sheets,23 and may also
reflect proximity of carbonyl groups in these compounds.
1H NMR NOE Studies of Artificial â-Sheet 2. 1H NMR
nuclear Overhauser effect (NOE) studies indicate that the
peptidomimetic and peptide strands of 2 are near each other
and provide strong evidence that 2 adopts an antiparallel â-sheet
conformation in CDCl3 solution.24 At 500 MHz and 21 °C, 2
exhibits very small NOEs, presumably as a result of its high
molecular weight (989). At 30 °C, the NOEs are larger,
however, allowing us to perform a series of 1-D NOE studies,
in which 22 of the molecule’s 39 distinct resonances were
irradiated, and 83 NOEs were observed and quantified. In these
experiments, most of the NOEs detected were between 1 and
4%, and it was possible to be confident that NOEs of 0.4% or
greater were real rather than artifacts. These studies are
discussed below and are described in detail in the Experimental
3-(3′-dimethylaminopropyl)carbodiimide hydrochloride (EDC),
generated the artificial â-sheet. Artificial â-sheet 2 was
subsequently prepared by solid-phase synthesis on Merrifield
resin using a slightly modified version of the route shown in
Scheme 2. This procedure will be described in a forthcoming
paper on the solid-phase synthesis of artificial â-sheets.20
1H NMR Chemical Shift Studies of Artificial â-Sheet 2.
1H NMR chemical shift studies indicate that 2 forms a hydrogen-
bonded antiparallel â-sheet structure in CDCl3 solution. In these
studies, the chemical shifts of the NH groups of artificial â-sheet
2 were compared to those of analogous protons in compounds
20 and 21. These two compounds serve as controls, resembling
respectively the upper (peptidomimetic) and lower (peptide)
halves of artificial â-sheet 2.
(21) Under the conditions of these studies, compounds 2, 20, and 21 do
not self-associate to any appreciable extent. Dilution titration studies indicate
that these compounds self-associate with dimerization constants of ca. 1
M-1 at ambient temperature in CDCl3 solution. In artificial â-sheet 2, protons
Hg and He exhibit the largest concentration-dependent changes in chemical
shift. This observation suggests that these protons are most available for
intermolecular hydrogen bonding and is consistent with the pattern of
intramolecular hydrogen bonding seen in this compound. In controls 20
and 21, protons Ha, Hc, Hf, and Hh exhibit the largest concentration-
dependent changes in chemical shift.
(22) Wagner, G.; Pardi, A.; Wu¨thrich, K. J. Am. Chem. Soc. 1983, 105,
5948.
(23) (a) Wishart, D. A.; Sykes, B. D.; Richards, F. M. J. Mol. Biol. 1991,
222, 311. (b) Wishart, D. A.; Sykes, B. D.; Richards, F. M. Biochemistry
1992, 31, 1647, and references contained therein.
In 1 mM CDCl3 solution, protons Ha, Hc, Hf, and Hh appear
1.15-3.68 ppm downfield of the analogous protons of controls
20 and 21 (Table 2).21 In this solvent, hydrogen-bonded NH
protons generally appear about 2-3 ppm downfield of NH
protons that are not hydrogen bonded. Thus, non-hydrogen-
bonded aliphatic amide protons generally appear at about 6 ppm,
while hydrogen-bonded aliphatic amide protons generally appear
at about 8 ppm.15,16 The chemical shifts of Hf and Hh of
artificial â-sheet 2 (8.14 and 8.02 ppm) suggest that these
(20) Holmes, D. L.; Smith, E. M.; Nowick, J. S. Submitted to J. Am.
Chem. Soc.
(24) Wu¨thrich, K. NMR of Proteins and Nucleic Acids; Wiley: New
York, 1986; pp 124-129.

An Extended â-Strand Mimic for a Larger Artificial â-Sheet
J. Am. Chem. Soc., Vol. 119, No. 23, 1997 5419
Figure 4. Model of artificial â-sheet 2 illustrating long-range NOEs. The model was generated using MacroModel V5.0 with the AMBER* force
field. The starting geometry (before minimization) was chosen to reflect the preferred (anti) conformation of the 1,2-diaminoethane diurea backbone.14b,c
The starting conformations of the amino acid side chains were chosen to reflect measured coupling constants and NOEs when possible but are
largely arbitrary. Adequate parameters for the C-N-N-C torsion angle of the hydrazide group were not available, and this torsion was constrained
to the crystallographically observed value of 80° during minimization. Only small differences in the hydrogen-bonded â-sheet structure occur if the
C-N-N-C torsion angle is varied; in other modeling studies the angle was constrained to -80° or unconstrained (180°) with little effect upon the
â-sheet structure. Mirror-image diastereomeric geometry of the N-C-C-N diurea backbone and different starting conformations of the amino
acid side chains were also employed, with little effect on the â-sheet structure.
Chart 4
between the Ile NH (7) and Phe R (18) protons, the Leu NH
(17) and Ile R (21) protons, and the lower methylamide NH (9)
and Leu R (20) protons. These interresidue NOEs are stronger
than the intraresidue NOEs between the NH and R protons of
these residues. Strong NOEs between the R-proton of one
residue and the NH group of the next residue are characteristic
of the â-strand conformation.24 NOEs between the NH groups
of the peptide strand (19, 7, 17, and 9) were not detected. The
absence of these NOEs provides additional support for an
extended (â-strand) conformation, in which these groups are
far apart.
Short-range NOEs within the â-strand mimic indicate that
this unit also adopts an extended conformation and further
support the model shown in Figure 4. Thus, urea proton 1
exhibits a much stronger NOE to â-strand mimic proton 3 than
to â-strand mimic proton 5; hydrazide proton 4 exhibits a
stronger NOE to â-strand mimic proton 6 than to â-strand mimic
proton 12; and hydrazide proton 2 exhibits a much stronger NOE
to â-strand mimic proton 12 than to â-strand mimic proton 6.
A number of NOEs are not wholly consistent with the model
shown in Figure 4. Urea proton 1 exhibits strong NOEs to all
four protons of the 1,2-diaminoethane backbone (NCH2CH2N,
24-26). Molecular modeling indicates that the urea proton
cannot be close to all four backbone protons at the same time.
Instead, the backbone adopts two mirror-image diastereomeric
anti conformations, each of which places two of 1,2-diamino-
ethane protons next to the urea NH group.14b,c The amino acid
side-chains are also conformationally mobile. Leucine methyl
group 36 exhibits small NOEs to both the phenylalanine phenyl
group and to the upper methylamide methyl (27), although
molecular modeling indicates that methyl 36 cannot be close
to both groups at one time. The 3JRâ coupling constants of the
Phe and Leu groups are 7-9 Hz, supporting the analysis that
these sidechains have little conformational preference. These
studies show that parts of artificial â-sheet 2 are conformation-
Section. Chart 4 will aid in this discussion, illustrating the 39
resonances that were observed (numbered in order of decreasing
chemical shift) and showing their assignments to the various
protons of 2. These assignments were determined by means
of COSY and NOE studies.
Long-range NOEs between the peptide and peptidomimetic
strands of 2 provide compelling evidence for a â-sheet structure.
Strong NOEs were observed between â-strand mimic proton 3
and the Phe R proton (18) and between â-strand mimic proton
6 and the Leu R proton (20). An NOE from the Ile NH (7) to
â-strand mimic proton 3 and an NOE from the lower methy-
lamide NH (9) to â-strand mimic proton 6 were also observed.
In addition, there were NOEs between the lower methylamide
NH (9) and the upper methylamide methyl group (27). Figure
4 provides a three-dimensional representation illustrating these
NOEs. These long-range NOE data are characteristic of an
antiparallel â-sheet structure and show that the peptide and
peptidomimetic strands of 2 are aligned.24
Short-range NOEs within the peptide strand provide further
evidence for â-sheet structure. Strong NOEs were observed

5420 J. Am. Chem. Soc., Vol. 119, No. 23, 1997
Nowick et al.
1
Table 4. Temperature Dependence of the H NMR Chemical Shifts
Table 5. 1H NMR Coupling Constants (Hz) of the NH Protons of
2, 20, and 21 (30 °C, 1 mM in CDCl3)
(ppb/K) of the NH Protons of 2, 20, and 21 (1 mM in CDCl3)a
Ha
Hb
Hc
Hd
He
Hf
Hg
-2.7b -8.5
-0.6 -0.8b -6.8
-4.2
Hh
Hb
Hc
4a
5.2
He
Hf
Hg
2
-1.5 -2.2 -3.3 -3.1 -0.5 -2.1
20 -1.6 -3.1 -0.3 -3.5
2
20
21
4.0
5.1
8.5
9.1
7.2
21
b
6.5
8.1
a Spectra recorded at 10 °C intervals from -30 °C to 30 °C.
b Temperature-dependence of these NH shift data exhibited poor linear
correlation.
a Value is approximate, because peak appears as a broad doublet.
b Value could not be determined due to overlap with phenylalanine R
proton.
coupling constants of 7-8 Hz. The coupling constants of
hydrazide protons Hb and Hc are small (4-5 Hz) in both 2 and
21. If the dependence of coupling constants upon dihedral angle
for hydrazides is similar to that of peptides, this value may
reflect that the hydrazide group is twisted, rather than linear.
This geometry would be reasonably consistent with that which
is observed crystallographically for hydrazides 7a,b and 9-13.
ally mobile and that no single structure (e.g., Figure 4) can
provide a complete picture of this molecule.
Variable-Temperature 1H NMR Studies of Artificial
â-Sheet 2. The temperature dependence of the 1H NMR
chemical shifts of amide NH protons reflects their state of
hydrogen bonding. In CDCl3, peptide amide protons that are
either not hydrogen bonded or locked in a hydrogen-bonded
conformation exhibit small temperature dependencies in chemi-
cal shifts, while protons that participate in an equilibrium
between a hydrogen-bonded state and a non-hydrogen-bonded
state exhibit large temperature dependencies in their chemical
shifts.25,26 Temperature dependencies are typically -2 or -3
ppb/K in the former case and -4 to -8 ppb/K in the latter.
In artificial â-sheet 2, all protons exhibit small temperature
dependencies of chemical shifts, with the exception of methy-
lamide proton Hh (Table 4). These observations are consistent
with a model in which artificial â-sheet 2 adopts a hydrogen-
bonded â-sheet conformation, but where the hydrogen bond to
the leucine methylamide NH proton “frays” upon warming. We
hesitate to place too much emphasis on the interpretation of
these data: much of what is known about temperature depen-
dence of NH shifts has been established for peptide amide NH
groups, but 2 contains not only peptide amide NH groups but
also urea, amine, and aromatic amide NH groups.
Conclusions and Outlook
These studies establish that â-strand mimic B duplicates the
hydrogen-bonding functionality of one edge of a peptide strand
in a â-strand conformation and can template â-sheet formation
in an attached peptide strand. 5-Hydrazino-2-methoxybenzoic
acid â-strand mimic B provides an array of four hydrogen-bond
donor and acceptor groups along its hydrogen-bonding edge. It
is twice as long as 5-amino-2-methoxybenzoic acid â-strand
mimic A, which has two hydrogen-bond donor and acceptor
groups. It is also longer than Kemp’s epindolidione â-strand
mimic, which has three hydrogen-bond donor and acceptor
groups. Because the N-to-C directionality of â-strand mimic
B is opposite to that of a peptide strand, direct comparison to
a peptide strand is difficult. Nevertheless, â-strand mimic B is
equivalent in length to the main chain of a tetrapeptide (12
atoms) and may best be described as a tetrapeptide mimic.
Although â-sheets have a reputation for self-associating
strongly and being insoluble, compounds containing â-strand
mimic B are extremely tractable. Artificial â-sheet 2 has greater
than 100 mM solubility in CDCl3 and control 20 has greater
than 25 mM solubility, and each of these compounds self-
In controls 20 and 21, NH groups Ha-Hf exhibit small
temperature dependencies. In contrast, the leucine NH (Hg)
exhibits a large temperature dependence, and the leucine
methylamide proton (Hh) exhibits a moderate temperature
dependence. These temperature dependencies suggest that Hg
and Hh of control 21 equilibrate between hydrogen-bonded and
non-hydrogen-bonded states. The large temperature dependence
of Hg provides further support that it forms a hydrogen-bonded
â-turn with the carbonyl of the urea group (vide supra).
1H NMR Coupling Constant Studies of Artificial â-Sheet
2. The 3JHNR coupling constants reflect the dihedral angle about
the CR-NH bond of the amino acids, which in turn reflects the
conformation of the amino acids in peptides and proteins.
Coupling constants greater than 7 Hz are generally considered
to be consistent with â-sheet structure, while coupling constants
of less than 6 Hz are consistent with R-helical structure.27,28
Consistent with â-sheet structure, the 3JHNR coupling constants
of the Phe, Ile, and Leu of artificial â-sheet 2 are 7-9 Hz (Table
5, He, Hf, and Hg). These values should not be regarded as
proof of â-sheet structure, however, since peptides lacking well-
associates with dimerization constants of about 1 M-1 21
In
.
contrast, Kemp’s 2,8-diaminoepindolidione, which has hydrogen-
bonding functionality along both edges, is only soluble in
concentrated sulfuric acid.2 Unlike a peptide strand, the other
edge of â-strand mimic B is blocked by methoxy groups and
offers no hydrogen-bonding functionality. The blockage of this
edge may account for the low tendency of derivatives containing
the â-strand mimic to self-associate and the high solubility of
these compounds in organic solvents.29
â-Strand mimic B is readily prepared by the coupling of a
5-amino-2-methoxybenzoic acid unit with a 5-hydrazino-2-
methoxybenzoic acid unit. â-Strand mimics of even greater
length (e.g., C and D) might be prepared by combining the
3
defined structures (e.g., control 21) typically exhibit JHNR
(25) (a) Ribeiro, A. A.; Goodman, M.; Naider, F. Int. J. Peptide Protein
Research 1979, 14, 414. (b) Stevens, E. S.; Sugawara, N.; Bonora, G. M.;
Toniolo, C. J. Am. Chem. Soc. 1980, 102, 7048. (c) Gellman, S. H.; Adams,
B. R. Tetrahedron Lett. 1989, 30, 3381.
(26) In competitive solvents, such as water and dimethyl sulfoxide, the
chemical shifts of peptide amide protons exhibit different temperature
dependencies: Solvent-exposed protons exhibit large temperature depend-
encies, while intramolecularly hydrogen-bonded (solvent-shielded) protons
exhibit small temperature dependencies.
(27) (a) Kessler, H. Angew. Chem., Int. Ed. Engl. 1982, 21, 512. (b)
Dyson, H. J.; Wright, P. E. Annu. ReV. Biophys. Biophys. Chem. 1991, 20,
519.
(28) Wu¨thrich, K. NMR of Proteins and Nucleic Acids; Wiley: New
York, 1986; pp 166-168.
5-amino-2-methoxybenzoic acid unit with two or more 5-hy-
drazino-2-methoxybenzoic acid units. In future studies, we will

An Extended â-Strand Mimic for a Larger Artificial â-Sheet
J. Am. Chem. Soc., Vol. 119, No. 23, 1997 5421
289.1796. Anal. Calcd for C16H23N3O2: C, 66.41; H, 8.01; N, 14.52.
Found: C, 66.43; H, 7.86; N, 14.38.
determine whether â-strand mimics such as these can be
synthesized and whether they template â-sheet formation in
peptide strands. We are particularly interested in learning
whether the unprotected aniline-type amino groups will com-
plicate coupling reactions during the preparation of higher
oligomers and whether the twist about the hydrazide N-N bonds
will interfere with the mimicry of extended â-strand structures.
We will address these issues in subsequent papers.
Urea 16. Phenylalanylisoleucylleucine methyl ester hydrochloride
(0.444 g, 1.00 mmol) was converted to phenylalanylisoleucylleucine
methyl ester isocyanate as described in ref 19. A solution of the
isocyanate and amine 15 (1.09 g, 3.76 mmol) in 2 mL of dichlo-
romethane was stirred under nitrogen for 20 h and then concentrated
by rotary evaporation. Column chromatography of the residue on silica
gel (EtOAc-hexanes, 1:1) afforded 0.543 g (75%) of PhN(CO-Phe-
Ile-Leu-OMe)CH2CH2N(Boc)CH2CH2CN as a colorless oil: IR (CHCl3)
3425, 3325, 2252, 1741, 1687, 1674 cm-1; 1H NMR (500 MHz, CDCl3)
δ 7.34 (m, 3 H), 7.21 (br s, 3 H), 6.99 (m, 4 H), 6.80 (d, J ) 8.5 Hz,
1 H), 6.43 (d, J ) 7.4 Hz, 0.5 H), 6.35 (d, J ) 7.4 Hz, 0.5 H), 4.63-
4.47 (m, 3 H), 4.26 (t, J ) 7.3 Hz, 1 H), 3.72 (s, 3 H), 3.83-3.62 (m,
2 H), 3.54-3.32 (m, 4 H), 3.04 (dd, J ) 14.1, 5.3 Hz, 1 H), 2.96-
2.88 (m, 1 H), 2.64-2.54 (m, 2 H), 1.95-1.87 (m, 1 H), 1.70-1.60
(m, 2 H), 1.60-1.54 (m, 1 H), 1.42 (br s, 5 H), 1.25 (s, 5 H), 1.12-
1.03 (m, 1 H), 0.94 (appar. t, J ) 5.3 Hz, 6 H), 0.89 (t, J ) 7.3 Hz, 3
H), 0.86 (d, J ) 6.8 Hz, 3 H); HRMS (FAB) m/e for C39H57N6O7
(M + H)+, calcd 721.4288, found 721.4292. Anal. Calcd for
C39H56N6O7: C, 64.98; H, 7.83; N, 11.66. Found: C, 64.80; H, 7.74;
N, 11.52.
Experimental Section
5-Bromo-2-methoxybenzoic Acid (5a).17 A mixture of bromine
(5.6 mL, 109 mmol), 100 mL of dichloromethane, 100 mL of water,
and o-anisic acid (15.2 g, 100 mmol) was stirred for 19 h. Excess
bromine was destroyed by addition of NaHSO3 (1.15 g, 11.0 mmol),
the layers were separated, and the aqueous layer was extracted
sequentially with a 100-mL and a 50-mL portion of methylene chloride.
The combined organic layers were dried over MgSO4, filtered, and
concentrated by rotary evaporation. The resulting solid residue was
suspended in 40 mL of dichloromethane and triturated with 500 mL
of ice-cold hexanes. The suspension was filtered, and the precipitate
was washed with ice-cold hexanes and dried to afford 22.1 g (96%) of
acid 5a as a white solid: mp 119-120 °C; IR (CHCl3) 3307, 1738
A solution of PhN(CO-Phe-Ile-Leu-OMe)CH2CH2N(Boc)CH2CH2-
CN (0.252 g, 0.350 mmol) in 10 mL of a saturated solution of
methylamine in methanol was allowed to react under nitrogen for 2 h
and then concentrated by rotary evaporation to afford 0.255 g of a white
solid. Column chromatography on silica gel (EtOAc) afforded 0.237
g (94%) of urea 16 as a foamy, white solid: IR (CH2Cl2) 3388, 3307,
1
cm-1; H NMR (300 MHz, CDCl3) δ 8.27 (d, J ) 2.6 Hz, 1 H), 7.67
(dd, J ) 8.8, 2.6 Hz, 1 H), 6.97 (d, J ) 8.9 Hz, 1 H), 4.07 (s, 3 H);
HRMS (EI) m/e for C8H7BrO3 (M+), calcd 229.9579, found 229.9576.
Anal. Calcd for C8H7BrO3: C, 41.59; H, 3.05. Found: C, 41.70; H,
2.89.
1
2252, 1664, 1659 cm-1; H NMR (500 MHz, CDCl3) δ 7.33 (br s, 4
Acid 6a. A 250-mL, three-necked, round-bottomed flask equipped
with a nitrogen inlet adapter, a glass stopper, a rubber septum, and a
magnetic stirring bar was charged with ca. 100 mL of dry tetrahydro-
furan, cooled to -78 °C, and charged with 27.5 mL of a 1.6 M solution
of n-butyllithium in hexanes (44 mmol). 5-Bromo-2-methoxybenzoic
acid (4.62 g, 20.0 mmol) was added in a single portion, and the reaction
mixture was stirred for 15 min. Di-tert-butylazodicarboxylate (5.53
g, 24.0 mmol) was added in a single portion, the reaction mixture stirred
for 5 min, and the cooling bath was removed. After 20 min, 100 mL
of water was added, the layers were separated, and the aqueous phase
was washed with 50 mL of ether. Ethyl acetate (100 mL) and 1 M
aqueous HCl (44 mL) were added to the aqueous phase, the layers
were separated, and the aqueous layer was extracted with 30 mL of
ethyl acetate. The combined ethyl acetate layers were dried over
MgSO4, filtered, and concentrated by rotary evaporation to a viscous
yellow oil. The residue was dissolved in ether and allowed to stand
overnight. The resulting suspension was filtered, and the solid residue
was washed with ether to afford 2.89 g (38%) of acid 6a as a white
H), 7.23-7.14 (m, 4 H), 7.01-7.03 (m, 1 H), 7.02-6.95 (m, 2 H),
6.95-6.88 (m, 2 H), 4.82-4.73 (m, 1 H), 4.62 (br s, 0.5 H), 4.54 (br
s, 1 H), 4.48 (br s, 0.5 H), 4.38-4.29 (m, 1 H), 3.86-3.27 (m, 6 H),
3.11 (dd, J ) 13.9, J ) 4.6 Hz, 1 H), 2.80 (dd, J ) 13.6, J ) 8.1 Hz,
1 H), 2.73-2.47 (m, 5 H), 1.98-1.87 (br s, 1 H), 1.86-1.76 (m, 1 H),
1.65-1.53 (m, 2 H), 1.53-1.45 (m, 1 H), 1.42 (s, 4.5 H), 1.24 (s, 4.5
H), 1.04-1.16 (m, 1 H), 0.90 (m, 12 H); HRMS (FAB) m/e for
C39H58N7O6 (M + H)+, calcd 720.4448, found 720.4446.
2-Methoxy-5-nitrobenzoic Acid.30 To an ice-cooled suspension of
o-anisic acid (1.52 g, 10.0 mmol) in 20 mL of concentrated sulfuric
acid was added NH4NO3 (0.890 g, 11.0 mmol) in small portions over
20 min. The ice bath was removed, and the reaction mixture was
allowed to warm to room temperature over 20 min and then poured
into 100 mL of water. The resulting suspension was filtered, and the
precipitate was washed with 300 mL of water, dissolved in methanol,
and reisolated by rotary evaporation to afford 1.49 g (75%) of
2-methoxy-5-nitrobenzoic acid as a white solid: mp 161-162 °C; IR
1
(CHCl3) 3358, 1745, 1346 cm-1; H NMR (300 MHz, CDCl3) δ 9.00
1
solid: mp 174-175 °C dec; IR (CHCl3) 3400, 3305, 1738 cm-1; H
(d, J ) 2.9 Hz, 1 H), 8.46 (dd, J ) 9.1, 2.9 Hz, 1 H), 7.19 (d, J ) 9.3
Hz, 1 H), 4.17 (s, 3 H); HRMS (CI, isobutane) m/e for C8H8NO5 (M
+ H)+, calcd 198.0402, found 198.0398. Anal. Calcd for C8H7NO5:
C, 48.74; H, 3.58; N, 7.10. Found: C, 48.64; H, 3.46; N, 7.00.
Benzyl 2-Methoxy-5-nitrobenzoate. A suspension of 2-methoxy-
5-nitrobenzoic acid (1.01 g, 5.12 mmol) in 5.0 mL of thionyl chloride
and 0.02 mL of N,N-dimethylformamide was heated at reflux for 2 h
and then concentrated by rotary evaporation to afford 2-methoxy-5-
nitrobenzoyl chloride as a yellow solid. A solution of benzyl alcohol
(0.259 mL, 2.50 mmol), triethylamine (0.348 mL, 2.50 mmol),
4-(dimethylamino)pyridine (0.031 g, 0.25 mmol), and a portion of the
crude acid chloride (0.457 g, 2.12 mmol) in 10 mL of dichloromethane
was stirred under nitrogen for 20.5 h. Dichloromethane (10 mL) and
20 mL of water was added to the reaction mixture, and the aqueous
phase was then acidified to pH 4 with 1 M aqueous HCl. The phases
were separated, the aqueous phase was extracted with 5 mL of
dichloromethane, and the combined organic layers were washed with
5 mL of saturated aqueous NaHCO3. The aqueous layer was extracted
with 5 mL of dichloromethane, and the combined organic layers were
dried over Na2SO4, decanted, and concentrated by rotary evaporation.
Column chromatography of the residue on silica gel (EtOAc-hexanes,
1:1) afforded 0.490 g (80%) of benzyl 2-methoxy-5-nitrobenzoate as a
NMR (300 MHz, CDCl3) δ 10.77 (br s, 1 H), 8.18 (br s, 1 H), 7.80 (br
s, 0.7 H), 7.63 (br s, 0.3 H), 7.01 (d, J ) 9.0 Hz, 1 H), 6.82 (br s, 0.8
H), 6.64 (br s, 0.2 H), 4.07 (s, 3 H), 1.49 (s, 18 H); HRMS (FAB) m/e
for C18H26N2O7 (M+), calcd 382.1740, found 382.1747. Anal. Calcd
for C18H26N2O7: C, 56.54; H, 6.85; N, 7.33. Found: C, 56.72; H,
6.74; N, 7.11.
Amine 15. A solution of C6H5NH(CH2)2NH(CH2)2CN14b (0.189 g,
1.00 mmol) and di-tert-butyl dicarbonate (0.218 g, 1.00 mmol) in 10
mL of dichloromethane was stirred for 2.3 h under N2 and then
concentrated by rotary evaporation. Column chromatography of the
residue on silica gel (EtOAc-hexanes, 3:5) afforded 0.217 g (75%) of
amine 15 as a colorless, viscous oil: IR (CHCl3) 3444, 3429, 3402,
2252, 1689 cm-1; 1H NMR (500 MHz, CDCl3) δ 7.18 (t, J ) 7.6 Hz,
2 H), 6.71 (br s, 1 H), 6.60 (d, J ) 7.0 Hz, 2 H), 4.16 (br s, 0.5 H),
3.91 (br s, 0.5 H), 3.53 (br s, 2 H), 3.48 (t, J ) 6.5 Hz, 2 H), 3.33
(appar. t, J ) 5.4 Hz, 2 H), 2.65 (br s, 1 H), 2.54 (br s, 1 H), 1.49 (s,
9 H); HRMS (FAB) m/e for C16H23N3O2 (M+), calcd 289.1790, found
(29) (a) Rajarathnam, K.; Sykes, B. D.; Kay, C. M.; Dewald, B.; Geiser,
T.; Baggiolini, M.; Clark-Lewis, I. Science 1994, 264, 90. (b) Ghadiri, M.
R.; Kobayashi, K.; Granja, J. R.; Chadha, R. K.; McRee, D. E. Angew.
Chem., Int. Ed. Engl. 1995, 34, 93. (c) Kobayashi, K.; Granja, J. R.; Ghadiri,
M. R. Angew. Chem., Int. Ed. Engl. 1995, 34, 95. (d) Nesloney, C. L; Kelly,
J. W. J. Am. Chem. Soc. 1996, 118, 5836.
1
pale yellow solid: mp 95-96 °C; IR (CHCl3) 1728, 1346 cm-1; H
NMR (500 MHz, CDCl3) δ 8.69 (d, J ) 2.8 Hz, 1 H), 8.33 (dd, J )
9.3, 2.9 Hz, 1 H), 7.45 (d, J ) 7.4 Hz, 2 H), 7.39 (t, J ) 7.4 Hz, 2 H),
(30) de Paulis, T.; Janowsky, A.; Kessler, R. M.; Clanton, J. A.; Smith,
H. E. J. Med. Chem. 1988, 31, 2027.

5422 J. Am. Chem. Soc., Vol. 119, No. 23, 1997
Nowick et al.
7.34 (t, J ) 7.1 Hz, 1 H), 7.05 (d, J ) 9.3 Hz, 1 H), 5.36 (s, 2 H) 3.99
(s, 3 H); HRMS (EI) m/e for C15H13NO5 (M+), calcd 287.0794, found
287.0782. Anal. Calcd for C15H13NO5: C, 62.72; H, 4.56; N, 4.88.
Found: C, 62.60; H, 4.53; N, 4.92.
Hz, 1 H), 1.81-1.70 (m, 1 H), 1.66 (ddd, J ) 13.7, 8.1, 5.8 Hz, 1 H),
1.54-1.47 (m, 1 H), 1.41 (ddd, J ) 13.6, 9.4, 6.0 Hz, 1 H), 1.36-
1.30 (m, 1 H), 1.03-0.95 (m, 1 H), 0.85 (d, J ) 6.6 Hz, 3 H), 0.82 (d,
J ) 6.5 Hz, 3 H), 0.77 (d, J ) 6.8 Hz, 3 H), 0.74 (t, J ) 7.4 Hz, 3 H);
HRMS (FAB) m/e for C50H63N8O8 (M + H)+, calcd 903.4768, found
903.4768. Anal. Calcd for C50H62N8O8: C, 66.50; H, 6.92; N, 12.41.
Found: C, 66.47; H, 6.60; N, 12.39.
Benzyl 5-Amino-2-methoxybenzoate. A suspension of benzyl
2-methoxy-5-nitrobenzoate (0.179 g, 0.623 mmol) and stannous chloride
dihydrate (0.703 g, 3.11 mmol) in 6 mL of ethyl acetate was heated at
reflux for 19.5 h and then partitioned between 20 mL of ethyl acetate
and 45 mL of half-saturated aqueous sodium bicarbonate. The aqueous
layer was extracted with 20 mL of ethyl acetate, and the combined
organic layers were washed with 5 mL of saturated aqueous NaCl, dried
over MgSO4, filtered, and concentrated by rotary evaporation to give
0.151 g of a yellow oil. Column chromatography on the residue on
silica gel (EtOAc-hexanes, 1:1) afforded 0.115 g (72%) of benzyl
5-amino-2-methoxybenzoate as a pale yellow oil: IR (CHCl3) 3444,
N-Methyl-5-hydrazino-2-methoxybenzamide Hydrochloride (19).
A solution of acid 6a (0.382 g, 1.00 mmol), 1-hydroxybenzotriazole
hydrate (0.230 g, 1.50 mmol), and 1-ethyl-3-(3′-dimethylaminopropyl)-
carbodiimide hydrochloride (0.230 g, 1.20 mmol) in 10 mL of methanol
was allowed to react under nitrogen for 24 min, and 2 mL of a saturated
solution of methylamine in methanol was then added. After 2 h, the
mixture was concentrated by rotary evaporation, and the residue was
partitioned between 20 mL of water and 20 mL of dichloromethane.
The aqueous layer was extracted with two 5-mL portions of dichlo-
romethane, and the combined organic layers were dried over MgSO4,
filtered, and concentrated by rotary evaporation to give 0.460 g of an
orange solid. Column chromatography on silica gel (EtOAc-hexanes,
3:1) afforded 0.335 g (85%) of the methylamide derivative of acid 6a
as a white solid: mp 183-184 °C dec; IR (CHCl3) 3421, 1741, 1714,
1651 cm-1; 1H NMR (300 MHz, CDCl3) δ 8.23 (br s, 0.3 H), 8.19 (br
s, 0.7 H), 7.84 (br s, 1 H), 7.66 (br s, 0.7 H), 7.47 (br s, 0.3 H), 6.88-
7.01 (m, 1.7 H), 6.60 (br s, 0.3 H), 3.95(s, 3 H), 3.01(d, J ) 4.8 Hz,
3 H) and 1.48 (s, 18 H); HRMS (FAB) m/e for C19H30N3O6 (M + H)+,
calcd 396.2134, found 396.2146. Anal. Calcd for C19H29N3O6: C,
57.71; H, 7.39; N, 10.63. Found: C, 57.74; H, 7.18; N, 10.45.
A solution of the methylamide derivative of acid 6a (0.190 g, 0.480
mmol) in 5 mL of a ca. 2.5 M solution of HCl in methanol was
permitted to react for 26 h and was then concentrated by rotary
evaporation to afford 0.122 g (110%) of N-methyl-5-hydrazino-2-
methoxybenzamide hydrochloride (or dihydrochloride) as an orange
1
3369, 1718 cm-1; H NMR (300 MHz, CDCl3) δ 7.38 (m, 5 H), 7.16
(s, 1 H), 6.80 (s, 2 H), 5.32 (s, 2 H), 3.81 (s, 3 H), 3.48 (s, 2 H);
HRMS (CI) m/e for C15H16NO3 (M+), calcd 257.1052, found 257.1059.
Anal. Calcd for C15H15NO3: C, 70.02; H, 5.88; N, 5.44. Found: C,
70.30; H, 5.66; N, 5.28.
Benzyl Ester 18. A mixture of urea 16 (0.488 g, 0.678 mmol) and
7 mL of a ca. 2.5 M solution of HCl in methanol was allowed to react
under nitrogen for 19 h and then concentrated by rotary evaporation.
The residue was dissolved in 10 mL of water, and the solution was
made basic (ca. pH 9) by addition of 6 M aqueous NaOH and extracted
with three 10-mL portions of dichloromethane. The combined organic
layers were dried over MgSO4, filtered, and concentrated by rotary
evaporation to afford 0.422 g of a foamy, white solid. Column
chromatography on silica gel (MeOH-EtOAc, 1:20) afforded 0.343 g
(82%) of PhN(CO-Phe-Ile-Leu-OMe)CH2CH2NHCH2CH2CN as a
foamy, white solid: mp 81-82 °C; IR (CHCl3) 3419, 3385, 3305, 2251,
1
1657 cm-1; H NMR (500 MHz, CDCl3) δ 7.42-7.35 (m, 4 H), 7.23
solid: mp 154-156 °C dec; IR (Nujol mull) 3600-2500, 1641 cm-1
;
(t, J ) 7.3 Hz, 1 H), 7.18 (t, J ) 7.2 Hz, 2 H), 7.03 (s, 1 H), 7.02 (d,
J ) 2.4 Hz, 1 H), 6.94 (appar. q, J ) 4.3 Hz, 1 H), 6.88 (d, J ) 7.2
Hz, 2 H), 6.79 (d, J ) 6.1 Hz, 1 H), 4.67 (appar. s, 1 H), 4.50 (ddd, J
) 10.9, 8.2, 3.8 Hz, 1 H), 4.29 (dt, J ) 8.4, 4.2 Hz, 1 H), 4.23 (dd, J
) 6.6, 4.4 Hz, 1 H), 3.76 (dt, J ) 14.1, 6.2 Hz, 1 H), 3.66 (dt, J )
14.1, 6.0 Hz, 1 H), 3.13 (dd, J ) 14.0, 4.7 Hz, 1 H), 2.89-2.81 (m, 2
H), 2.79 (d, J ) 4.7 Hz, 3 H), 2.73 (dd, J ) 14.1, 9.1 Hz, 1 H), 2.69
(t, J ) 6.1 Hz, 2 H), 2.43 (t, J ) 6.4 Hz, 2 H), 2.09-2.01 (m, 1 H),
1.86 (ddd, J ) 14.0, 10.0, 4.0 Hz, 1 H), 1.70-1.58 (m, 2 H), 1.50-
1.41 (m, 1 H), 1.17-1.07 (m, 1 H), 0.94 (d, J ) 7.2 Hz, 6 H), 0.91 (d,
J ) 6.3 Hz, 3 H), 0.90 (t, J ) 7.4 Hz, 3 H); HRMS (FAB) m/e for
C34H50N7O4 (M + H)+, calcd 620.3924, found 620.3911. Anal. Calcd
for C34H49N7O4: C, 65.89; H, 7.97; N, 15.82. Found: C, 65.64; H,
7.93; N, 15.66.
[CAUTION: PHOSGENE IS VOLATILE AND HIGHLY
TOXICsUSE HOOD]. A 1.93 M solution of phosgene in toluene (0.56
mL, 1.1 mmol) was added to a rapidly stirred, ice-cooled solution of
benzyl 5-amino-2-methoxybenzoate (0.166 g, 0.645 mmol) in 5 mL of
methylene chloride and 5 mL of saturated aqueous NaHCO3 solution.
After 15 min, the phases were separated, the organic phase was extracted
with two 5-mL portions of methylene chloride, and the combined
organic phases were dried over MgSO4, filtered, and concentrated by
rotary evaporation. The residue was dissolved in 2 mL of methylene
chloride, and a solution of PhN(CO-Phe-Ile-Leu-OMe)CH2CH2NHCH2-
CH2CN (0.333 g, 0.537 mmol) in 3 mL of dichloromethane was added.
The mixture was allowed to react under nitrogen for 2 h and was then
concentrated by rotary evaporation to afford a foamy, pale yellow solid.
Column chromatography on silica gel (EtOAc-dichloromethane, 3:1)
afforded 0.434 g (89%) of benzyl ester 18 as a white powder: mp
199-200 °C; IR (CHCl3) 3412, 3307, 2251, 1711 (sh), 1660 cm-1; 1H
NMR (500 MHz, CDCl3) δ 9.07 (s, 1 H), 8.16 (d, J ) 2.6 Hz, 1 H),
7.91 (dd, J ) 9.0, 2.7 Hz, 1 H), 7.48 (d, J ) 7.4 Hz, 2 H), 7.41-7.35
(m, 5 H), 7.32 (t, J ) 7.4 Hz, 1 H), 7.22-7.18 (m, 3 H), 7.05-7.01
(m, 2 H), 6.99-6.93 (m, 4 H), 6.39 (appar. q, J ) 4.7 Hz, 1 H), 6.20
(d, J ) 8.4 Hz, 1 H), 5.40 (s, 2 H), 4.76 (d, J ) 7.0 Hz, 1 H), 4.57 (q,
J ) 7.1 Hz, 1 H), 4.39 (td, J ) 8.6, 5.9 Hz, 1 H), 4.16 (t, J ) 6.7 Hz,
1 H), 3.88 (s, 3 H), 3.81-3.68 (m, 3 H), 3.61-3.54 (m, 2 H), 3.45-
3.39 (m, 1 H), 2.97 (dd, ABX pattern, JAB ) 14.0 Hz, JAX ) 6.4 Hz,
1 H), 2.89 (dd, ABX pattern, JAB ) 13.9 Hz, JBX ) 7.8 Hz, 1 H),
2.77-2.70 (m, 1 H), 2.72 (d, J ) 4.7 Hz, 3 H), 2.62 (dt, J ) 17.1, 5.9
1H NMR (300 MHz, D2O) δ 7.46 (d, J ) 2.9 Hz, 1 H), 7.23 (dd, J )
9.0, 2.9 Hz, 1 H), 7.13 (d, J ) 9.0 Hz, 1 H), 3.90 (s, 3 H) and 2.88 (s,
3 H); HRMS (CI) m/e for C9H12N3O2 (M - HCl - H)+, calcd 194.0929,
found 194.0923.
Artificial â-Sheet 2. A two-necked, 50-mL, round-bottomed flask
equipped with a gas inlet adapter connected to a nitrogen/vacuum
manifold, a gas inlet adapter fitted with a balloon filled with hydrogen,
and a magnetic stirring bar was evacuated, filled with nitrogen, and
then charged with 0.230 g of 10% Pd/C, 7.7 mL of tetrahydrofuran,
and benzyl ester 18 (0.426 g, 0.472 mmol). The flask was evacuated,
filled with hydrogen gas, and maintained under a hydrogen atmosphere
for 14 h. The reaction mixture was then filtered through Celite, and
the filtrate was concentrated by rotary evaporation to afford 0.383 g
(100%) of the carboxylic acid derivative of 18 as a foamy, light yellow
solid. This compound was used directly in the next step without further
purification. HRMS (FAB) m/e for C43H57N8O8 (M + H)+, calcd
813.4299, found 813.4306.
1-Ethyl-3-(3′-dimethylaminopropyl)carbodiimide hydrochloride (0.106
g, 0.552 mmol) was added to a solution of the product of the preceding
reaction (0.359 g, 0.442 mmol), 19 (0.137 g, 0.590 mmol), triethylamine
(0.092 mL, 0.66 mmol), and 1-hydroxybenzotriazole hydrate (0.075 g,
0.55 mmol) in 5 mL of methanol. The reaction mixture was stirred
under nitrogen for 19 h and then concentrated by rotary evaporation.
The residue was partitioned between 25 mL of water and 50 mL of
dichloromethane, and the organic layer was extracted with 25 mL of
saturated aqueous NaHCO3, dried over MgSO4, filtered, and concen-
trated by rotary evaporation. Column chromatography of the residue
on silica gel (twice, first with i-PrOH-EtOAc-dichloromethane, 5:45:
50, then with i-PrOH-EtOAc, 5:95) afforded 0.265 g (61%) of artificial
â-sheet 2 as a pale yellow solid. Reverse-phase HPLC analysis on a
C18 column with an eluant of 70:30 CH3OH-H2O and 254 nm UV
detection indicated 2 to be 98.4% pure: mp 142-144 °C; IR (CHCl3)
3410, 3305, 2251, 1651 cm-1; 1H NMR (500 MHz, CDCl3) δ 9.94 (s,
1 H), 9.57 (d, J ) 5.9 Hz, 1 H), 8.65 (d, J ) 2.4 Hz, 1 H), 8.47 (d, J
) 5.8 Hz, 1 H), 8.39 (dd, J ) 8.9, 2.5 Hz, 1 H), 8.19-8.15 (m, 2 H),
8.12 (appar. q, J ) 4.7 Hz, 1 H), 8.08 (appar. q, J ) 4.4 Hz, 1 H),
7.38-7.29 (m, 3 H), 7.14-6.96 (m, 6 H), 6.94-6.88 (m, 3 H), 6.28
(d, J ) 7.0 Hz, 1 H), 5.00 (q, J ) 7.0 Hz, 1 H), 4.85 (d, J ) 8.5 Hz,
1 H), 4.41 (q, J ) 7.2 Hz, 1 H), 4.14 (t, J ) 9.1 Hz, 1 H), 4.11 (s, 3

An Extended â-Strand Mimic for a Larger Artificial â-Sheet
J. Am. Chem. Soc., Vol. 119, No. 23, 1997 5423
H), 3.91 (s, 3 H), 3.86-3.78 (m, 1 H), 3.68-3.55 (m, 2 H), 3.54-
3.40 (m, 3 H), 3.01 (d, J ) 4.7 Hz, 3 H), 2.88 (dd, ABX pattern, JAB
) 14.1 Hz, JAX ) 6.6 Hz, 1 H), 2.81 (d, J ) 4.4 Hz, 3 H), 2.78 (dd,
ABX pattern, JAB ) 14.0 Hz, JBX ) 9.3 Hz, 1 H), 2.62-2.59 (m, 2 H),
1.67-1.53 (m, 3 H), 1.45 (ddd, J ) 12.6, 8.0, 5.9 Hz, 1 H), 1.42-
1.33 (m, 1 H), 1.06-0.96 (m, 1 H), 0.93 (d, J ) 6.2 Hz, 3 H), 0.81 (d,
J ) 6.5 Hz, 3 H), 0.79 (d, J ) 6.9 Hz, 3 H), 0.58 (t, J ) 7.3 Hz, 3 H);
13C NMR (125 MHz, CDCl3) δ 172.9, 171.8, 170.8, 166.1, 165.1, 157.2,
155.5, 152.5, 151.6, 143.2, 140.9, 136.4, 134.7, 130.2, 129.0, 128.2,
128.0, 127.3, 126.5, 123.0, 122.3, 121.5, 119.3, 118.8, 118.5, 116.2,
112.1, 111.6, 57.6, 56.3, 55.6, 52.2, 50.5, 48.1, 45.1, 41.1, 38.1, 36.5,
26.4, 26.0, 24.63, 24.55, 22.49, 22.2, 17.4, 15.3, 10.6; HRMS (FAB)
m/e for C52H68N11O9 (M + H)+, calcd 990.5201, found 990.5191.
Control 20. [CAUTION: PHOSGENE IS VOLATILE AND
HIGHLY TOXICsUSE HOOD]. A 1.93 M solution of phosgene in
toluene (0.46 mL, 0.89 mmol) was added to a rapidly stirred ice-cooled
solution of benzyl 5-amino-2-methoxybenzoate (0.115 g, 0.447 mmol)
in 5 mL of methylene chloride and 5 mL of saturated aqueous NaHCO3
solution. After 15 min, the phases were separated, and the aqueous
phase was extracted with two 5-mL portions of methylene chloride.
The combined organic phases were dried over MgSO4, filtered, and
concentrated by rotary evaporation. The residue was dissolved in 5
mL of methylene chloride, and diethylamine (0.092 mL, 0.89 mmol)
was added. The mixture was allowed to react for 0.75 h and was then
concentrated by rotary evaporation to give 0.160 g of a colorless oil.
Column chromatography on silica gel (EtOAc-hexanes, 1:1) afforded
0.135 g (84%) of 5Et2NCONH-2MeO-C6H3CO2Bn as a colorless
oil: IR (CHCl3) 3462, 1720, 1657 cm-1; 1H NMR (500 MHz, CDCl3)
δ 7.77 (dd, J ) 9.0, 2.8 Hz, 1 H), 7.59 (d, J ) 2.8 Hz, 1 H), 7.45 (d,
J ) 7.3 Hz, 2 H), 7.38 (t, J ) 7.4 Hz, 2 H), 7.33 (t, J ) 7.2 Hz, 1 H),
6.93 (d, J ) 9.0 Hz, 1 H), 6.21 (s, 1 H), 5.34 (s, 2 H), 3.88 (s, 3 H),
3.35 (q, J ) 7.1 Hz, 4 H), 1.21 (t, J ) 7.1 Hz, 6 H); HRMS (CI) m/e
for C20H25N2O4 (M+), calcd 356.1736, found 356.1752. Anal. Calcd
for C20H24N2O4: C, 67.40; H, 6.79; N, 7.86. Found: C, 67.09; H,
6.80; N, 7.80.
164-166 °C; IR (CHCl3) 3460, 3410, 1657, 1651 cm-1; 1H NMR (300
MHz, CDCl3) δ 9.55 (d, J ) 4.3 Hz, 1 H), 8.02 (dd, J ) 9.0, 2.6 Hz,
1 H), 7.90 (appar. q, J ) 4.6 Hz, 1 H), 7.82 (d, J ) 2.7 Hz, 1 H), 7.70
(d, J ) 2.6 Hz, 1 H), 7.00 (dd, J ) 8.9, 2.8 Hz, 1 H), 6.96 (d, J ) 9.1
Hz, 1 H), 6.85 (d, J ) 8.8 Hz, 1 H), 6.55 (s, 2 H), 4.00 (s, 3 H), 3.88
(s, 3 H), 3.35 (q, J ) 7.1 Hz, 4 H), 2.97 (d, J ) 4.8 Hz, 3 H), 1.19 (t,
J ) 7.1 Hz, 6 H); 13C NMR (125 MHz, CDCl3) δ 165.8, 165.5, 154.8,
153.2, 152.3, 142.5, 133.5, 125.7, 123.0, 122.1, 119.5, 118.1, 117.7,
112.5, 112.2, 56.5, 56.3, 41.5, 26.5, 13.9; HRMS (CI) m/e for
C22H30N5O5 (M + H)+, calcd 444.2247, found 444.2238.
Control 21. A solution of PhN(Et)CO-Phe-Ile-Leu-OMe19 (0.102
g, 0.184 mmol) in 15 mL of a saturated solution of methylamine in
methanol was allowed to react under nitrogen for 1 h and then
concentrated by rotary evaporation to afford 0.101 g of urea 21 as a
tan film. Reverse-phase HPLC analysis on a C18 column with an eluant
of 80:20 CH3OH-H2O and 254 nm UV detection indicated 21 to be
99.3% pure: mp 225-226 °C; IR (CHCl3) 3421, 3388, 3300, 1669
1
(sh), 1655 cm-1; H NMR (500 MHz, CDCl3) δ 7.58 (d, J ) 8.1 Hz,
1 H), 7.42-7.35 (m, 3 H), 7.29-7.21 (m, 1 H), 7.18 (t, J ) 7.5 Hz, 2
H), 6.91-6.88 (m, 3 H), 6.84 (d, J ) 7.1 Hz, 2 H), 6.48 (d, J ) 6.4
Hz, 1 H), 4.53-4.48 (m, 2 H), 4.22 (dd, J ) 6.6 Hz, 3.5 Hz, 1 H),
4.16 (ddd, J ) 9.2, 4.2, 2.3 Hz, 1 H), 3.71-3.65 (m, 1 H), 3.59-3.54
(m, 1 H), 3.16 (dd, J ) 14.1, 4.4 Hz, 1 H), 2.82 (d, J ) 4.6 Hz, 3 H),
2.64 (dd, J ) 14.1, 9.7 Hz, 1 H), 2.17-2.10 (m, 1 H), 1.90 (ddd, J )
13.9, 10.6, 3.6 Hz, 1 H), 1.73 (ddd, J ) 13.9, 11.9, 4.0 Hz, 1 H),
1.69-1.53 (m, 2 H), 1.52-1.44 (m, 1 H), 1.15-1.07 (m, 1 H), 1.03 (t,
J ) 7.1 Hz, 3 H), 0.97-0.95 (m, 6 H), 0.92 (d, J ) 7.9 Hz, 3 H), 0.90
(d, J ) 7.2 Hz, 3 H); 13C NMR (125 MHz, CDCl3) δ 173.3, 173.0,
170.6, 157.1, 139.7, 135.5, 130.5, 129.1, 128.5, 128.4, 128.0, 127.3,
59.1, 57.5, 52.0, 44.0, 39.6, 37.0, 35.5, 26.3, 25.0, 24.6, 23.4, 20.5,
16.2, 13.5, 11.7; HRMS (FAB) m/e calcd for C31H46N5O4 (M + H)+,
calcd 552.3550, found: 552.3554. Anal. Calcd for C31H45N5O4: C,
67.49; H, 8.22; N, 12.69. Found: C, 67.09; H, 8.29; N, 12.62.
1H NMR NOE Studies of Artificial â-Sheet 2. A 25-mM solution
of artificial â-sheet 2 in dry CDCl3 was flame-sealed in an NMR tube
after five freeze-pump-thaw cycles on a high-vacuum line. The 500
MHz 1H NMR spectra were recorded at 30 °C and 39 resonances were
identified as follows: 1, δ 9.93 (s, 1 H); 2, δ 9.55 (s, 1 H); 3, δ 8.65
(d, J ) 2.8 Hz, 1 H); 4, δ 8.44 (br s, 1 H); 5, δ 8.39 (dd, J ) 9.0, 2.8
Hz, 1 H); 6, δ 8.15 (d, J ) 3.0 Hz, 1 H); 7, δ 8.14 (d, J ) 9.2 Hz, 1
H); 8, δ 8.10 (q, J ) 4.7 Hz, 1 H); 9, δ 8.01 (q, J ) 4.6 Hz, 1 H); 10,
δ 7.37-7.30 (m, 3 H); 11, δ 7.10 (tt, J ) 7.1, 1.7 Hz, 1 H); 12, δ 7.08
(dd, J ) 8.8, 3.0 Hz, 1 H); 13, δ 7.05-6.99 (m, 4 H); 14, δ 6.98 (d,
J ) 9.1 Hz, 1 H); 15, δ 6.91 (dd, J ) 8.0, 1.2 Hz, 2 H); 16, δ 6.90 (d,
J ) 8.7 Hz, 1 H); 17, δ 6.17 (d, J ) 7.2 Hz, 1 H); 18, δ 5.00 (td, J )
8.8, 6.8 Hz, 1 H); 19, δ 4.83 (d, J ) 8.5 Hz, 1 H); 20, δ 4.44 (q, J )
7.3 Hz, 1 H); 21, δ 4.12 (t, J ) 9.0 Hz, 1 H); 22, δ 4.04 (s, 3 H); 23,
δ 3.91 (s, 3 H); 24, δ 3.85-3.80 (m, 1 H); 25, δ 3.68-3.56 (m, 2 H);
26, δ 3.54-3.41 (m, 3 H); 27, δ 3.00 (d, J ) 4.8 Hz, 3 H); 28, δ 2.87
(dd, ABX pattern, JAB ) 14.1 Hz, JAX ) 6.6 Hz, 1 H); 29, δ 2.81 (d,
A two-necked, 50-mL, round-bottomed flask was equipped with a
gas inlet adapter connected to a nitrogen/vacuum manifold, a gas inlet
adapter was fitted with a balloon filled with hydrogen, and a magnetic
stirring bar was evacuated, filled with nitrogen, and then charged with
0.070 g of 10% Pd/C, 4 mL of tetrahydrofuran, and 5Et2NCONH-
2MeO-C6H3CO2Bn (0.137 g, 0.386 mmol). The flask was evacuated,
filled with hydrogen gas, and maintained under a hydrogen atmosphere
for 21 h. The reaction mixture was then filtered through Celite, and
the filtrate was concentrated by rotary evaporation. The residue was
dissolved in a mixture of 10 mL of water, 3 mL of saturated aqueous
NaHCO3, 1 mL of 1 M aqueous NaOH, and 20 mL of dichloromethane.
The layers were separated, and the aqueous layer was acidified with 5
mL of 1 M aqueous HCl and extracted with two 8-mL portions of
dichloromethane. The combined organic layers were dried over Na2-
SO4 and concentrated by rotary evaporation to afford 0.094 g (92%)
of 5Et2NCONH-2MeO-C6H3CO2H as a foamy, white solid: mp 154-
J ) 4.7 Hz, 3 H); 30, δ 2.78 (dd, ABX pattern, JAB ) 14.1 Hz, JBX
)
1
156 °C, IR (CHCl3) 3462, 3379, 3290, 1732, 1657 cm-1; H NMR
9.3 Hz, 1 H); 31, δ 2.72-2.59 (m 2 H); 32, δ 1.67-1.54 (m, 3 H); 33,
δ 1.45 (ddd, J ) 12.6, 8.0, 5.9 Hz, 1 H); 34, δ 1.42-1.34 (m, 1 H);
35, δ 1.06-0.95 (m, 1 H); 36, δ 0.93 (d, J ) 6.4 Hz, 3 H); 37, δ 0.81
(d, J ) 6.3 Hz, 3 H); 38, δ 0.79 (d, J ) 6.7 Hz, 3 H); 39, δ 0.59 (t,
J ) 7.4 Hz, 3 H).
(300 MHz, CDCl3) δ 8.21 (d, J ) 7.6 Hz, 1 H), 7.93 (s, 1 H), 7.00 (m,
2 H), 4.05 (s, 3 H), 3.42 (q, J ) 7.1 Hz, 4 H), 1.23 (t, J ) 7.1 Hz, 6
H); HRMS (EI) m/e for C13H18N2O4 (M)+, calcd 266.1266, found
266.1270. Anal. Calcd for C13H18N2O4: C, 58.64; H, 6.81; N, 10.52.
Found: C, 58.37; H, 6.77; N, 10.22.
Twenty-two one-dimensional NOE experiments were performed with
3 s irradiation time and a collection of 512 transients. Percentage NOEs
were calculated by comparing the area of the enhanced peak and are
normalized to reflect the irradiation of multiple hydrogens but are not
normalized to reflect enhancements of more than one proton. Enhanced
peaks involving more than one proton are marked with an asterisk.
Enhancements of 0.4% or greater are reported as follows: Irradiation
of 1 enhanced 3 (4.4), 5 (0.7), 24 (1.5), 25 (4.1*), 26 (1.9*). Irradiation
of 2 enhanced 6 (0.7), 12 (3.5), 22 (1.1*). Irradiation of 3 enhanced
1 (1.6), 18 (1.8). Irradiation of 4 enhanced 6 (2.1), 12 (0.9). Irradiation
of 6 enhanced 18 (0.9, NOE attributed to nonselective irradiation in
which 7 was also irradiated) 20 (1.1). Irradiation of 7 enhanced 3 (0.6),
18 (3.9), 21 (0.4), 32 (1.1), 34 (0.4). Irradiation of 8 enhanced 23
(0.8*), 27 (6.6*). Irradiation of 9 enhanced 6 (0.8), 20 (3.4), 27 (0.7*),
29 (6.9*), 32 (0.6). Irradiation of 17 enhanced 20 (1.1), 21 (3.3), 32
(2.4*), 33 (1.4). Irradiation of 18 enhanced 3 (2.2), 7 (3.4), 13 (5.4*),
1-Ethyl-3-(3′-dimethylaminopropyl)carbodiimide hydrochloride (0.092
g, 0.480 mmol) was added to a solution of 5Et2NCONH-2MeO-C6H3-
CO2H (0.082 g, 0.309 mmol), 19 (0.111 g, 0.480 mmol), triethylamine
(0.074 mL, 0.53 mmol), and 1-hydroxybenzotriazole hydrate (0.065 g,
0.48 mmol) in 5 mL of methanol. The mixture was allowed to react
under nitrogen for 4 h and was then concentrated by rotary evaporation.
The residue was partitioned between 20 mL of water and 20 mL of
dichloromethane. The aqueous phase was then acidified to pH 1 with
1 M aqueous HCl, and the layers were separated. The organic layer
was extracted with 20 mL of saturated aqueous NaHCO3, dried over
MgSO4, filtered, and concentrated by rotary evaporation. Column
chromatography of the residue on silica gel (i-PrOH-CHCl3, 1:9)
afforded 0.112 g (82%) of urea 20 as an orange solid: Reverse-phase
HPLC analysis on a C18 column with an eluant of 50:50 CH3OH-
H2O and 254 nm UV detection indicated 21 to be 98.5% pure: mp

5424 J. Am. Chem. Soc., Vol. 119, No. 23, 1997
Nowick et al.
28 (1.4), 30 (1.4). Irradiation of 19 enhanced 10 (0.4*), 13 (1.8*), 15
(2.5*). Irradiation of 20 enhanced 6 (2.3), 9 (3.1), 17 (0.5), 32 (1.8*),
33 (1.1), 36 (1.3*), 37 (1.8*). Irradiation of 21 enhanced 7 (0.9), 17
(1.8), 32 (1.5*), 34 (0.6), 35 (0.8), 38 (1.8*), 39 (0.9*). Irradiation of
22 enhanced 2 (0.6), 14 (6.4). Irradiation of 23 enhanced 8 (1.0), 16
(7.5). Irradiation of 27 enhanced 8 (5.6), 9 (0.4). Irradiation of 28
enhanced 13 (4.4*), 18 (3.1). Irradiation of 29 enhanced 9 (2.7).
Irradiation of 36 enhanced 9 (0.6), 11 (0.5), 13 (1.3*), 20 (2.0), 27
(0.8*), 32 (6.0), 33 (1.9). Irradiation of 37 enhanced 20 (2.0), 32 (11.5),
33 (1.2). Irradiation of 38 enhanced 7 (0.7), 17 (0.4), 20 (0.4), 21
(2.6), 32 (4.1*), 34 (0.5), 35 (0.5), 39 (2.6*). Irradiation of 39 enhanced
21 (1.2), 32 (2.0), 34 (4.6), 35 (3.0).
AG00096-12). J.S.N. thanks the following agencies for support
in the form of awards: the Camille and Henry Dreyfus
Foundation (New Faculty Award, Teacher-Scholar Award), the
National Science Foundation (Young Investigator Award,
Presidential Faculty Fellowship), and the Arnold and Mabel
Beckman Foundation (Young Investigator Award).
Supporting Information Available: 1H NMR spectra and
HPLC traces of compounds 2, 20, and 21. Thermal ellipsoid
plots, experimental details of the X-ray crystallographic structure
determination, and tables of distances, angles, fractional coor-
dinates, and thermal parameters for compounds 7a, 7b, and
9-13 (87 pages). See any current masthead page for ordering
and Internet access instructions.
Acknowledgment. This work was supported by the National
Institutes of Health Grant GM-49076, Zeneca Pharmaceuticals
Group, the Upjohn Company, and Hoffman-La Roche Inc.
D.L.H. thanks the National Institute on Aging for support in
the form of a training grant (National Research Service Award
JA963843S