Three-stranded mixed artificial β-sheets

Article abstract of DOI:10.1016/S0040-4020(01)01092-4

This paper describes the design, synthesis, and structural evaluation of a pair of compounds, comprising molecular templates and attached peptide strands, that mimic a small three-stranded mixed β-sheet (6a and b). Each of these artificial β-sheets is com

Full text of DOI:10.1016/S0040-4020(01)01092-4

TETRAHEDRON
Pergamon
Tetrahedron 58 .2002) 727±739
Three-stranded mixed arti®cial b-sheets
²
James S. Nowick,p Eric M. Smith, Joseph W. Ziller and A. J. Shaka
Department of Chemistry, University of California, Irvine, CA 92697-2025, USA
Received 16 July 2001; revised 13 September 2001; accepted 14 September 2001
AbstractÐThis paper describes the design, synthesis, and structural evaluation of a pair of compounds, comprising molecular templates and
attached peptide strands, that mimic a small three-stranded mixed b-sheet .6a and b). Each of these arti®cial b-sheets is composed of two
different molecular templates and two dipeptide strands. One of the templates is based on a 5-amino-2-methoxybenzoic hydrazide group that
mimics the hydrogen-bonding functionality of a peptide b-strand and serves as the top strand. This template forms a pattern of hydrogen
bonds similar to that of an antiparallel b-sheet with the middle peptide strand. The middle peptide strand forms a pattern of hydrogen bonds
similar to that of a parallel b-sheet with the bottom peptide strand. The other template holds the three strands next to each other and is based
upon a triurea. In one arti®cial b-sheet .6a), both the upper and middle urea groups and the middle and lower urea groups are linked by
dimethylene .CH2CH2) chains; in the other .6b), the upper and middle urea groups are linked by a dimethylene chain, while the middle and
lower urea groups are linked by a trimethylene .CH2CH2CH2) chain. 1H NMR chemical shift and NOE studies establish that both of these
compounds fold to adopt a hydrogen-bonded b-sheetlike structure in CDCl3 solution. Chemical shift studies establish that three-stranded
mixed arti®cial b-sheets 6 are more tightly folded than their smaller two-stranded homologues, arti®cial parallel b-sheet 1 and arti®cial
antiparallel b-sheet 3, as well as their three-stranded homologues with truncated b-strand mimics, arti®cial b-sheets 5. These studies show
that the folding of arti®cial b-sheets 6 is cooperative, with the interactions between the upper and middle strands and between the middle and
lower strands reinforcing each other. q 2002 Elsevier Science Ltd. All rights reserved.
Protein b-sheets come in two basic varietiesÐparallel and
antiparallel. Parallel b-sheets are characterized by adjacent
parallel peptide strands that form a series of twelve-
membered hydrogen-bonded rings; antiparallel b-sheets
are characterized by adjacent antiparallel peptide strands
that form an alternating series of 10- and 14-membered
hydrogen-bonded rings. Mixed b-sheets contain both
parallel and antiparallel peptide strands with both types of
hydrogen-bonding patterns. All of these structures occur
widely in proteins, although mixed b-sheets are somewhat
less common than antiparallel and parallel b-sheets.
Researchers have developed a variety of small peptidelike
molecules that fold to resemble protein b-sheets to gain
enhanced understanding of protein folding and structure,
invent new molecules and materials with proteinlike proper-
ties, and develop novel biologically active compounds.
Pioneering efforts in this important and popular area have
come from the research groups of Mutter,1 Feigel,2 Kemp,3
Kelly,4 Gellman,5,6 and Seebach,7,8 as well as from our own
research group. These studies have lead to the creation of
structures that resemble both antiparallel and parallel
b-sheets. Related efforts have focused upon the creation
of peptides that fold into antiparallel b-sheets,9 with current
studies focusing on the development of multistranded
Keywords: peptidomimetic; peptide; mixed beta sheets; molecular
templates.
p
structures.10
Author to whom correspondence regarding synthetic and structural
studies should be addressed. Fax: 11-949-824-8571;
²
Author to whom correspondence regarding experimental NMR proce-
dures should be addressed.
e-mail: jsnowick@uci.edu; ajshaka@uci.edu
While these efforts have yielded structures that resemble
both parallel and antiparallel b-sheets, none, to our
0040±4020/02/$ - see front matter q 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020.01)01092-4

728
J. S. Nowick et al. / Tetrahedron 58 )2002) 727±739
Figure 1. X-Ray crystallographic structure of 5-amino-2-methoxybenzoic hydrazide b-strand mimic 7.
knowledge, have lead to structures that resemble mixed
b-sheets. Our research group has actively been pursuing
1. Results and discussion
both parallel and antiparallel b-sheetlike structures since
the early 1990s. We have termed these compounds, in
which molecular templates induce b-sheet structure in
attached peptide strands, arti®cial b-sheets. In 1995, we
reported a small parallel arti®cial b-sheet .1) containing a
diurea turn template that allowed the attachment of two
peptide strands by their N-termini in a parallel fashion.11
Shortly thereafter, we reported a series of two-stranded anti-
parallel b-sheets .e.g. 2 and 3) containing the diurea turn
structure and b-strand mimics designed to rigidify the sheet
and block aggregation.12 Concurrently, we reported a triurea
homologue of the turn template .4) designed for the creation
of three-stranded arti®cial b-sheets.13 This work has paved
the way for creating larger and more complex arti®cial
b-sheets.14 In the current study, we set out to determine
whether we could create three-stranded mixed .parallel
and antiparallel) arti®cial b-sheets by combining our triurea
scaffold with two parallel peptide strands and a b-strand
mimic.15 This paper describes our design, synthesis, and
evaluation of these structures.
1.1. Design of three-stranded arti®cial b-sheets
We designed three-stranded mixed arti®cial b-sheets 6
based on our earlier studies of two-stranded arti®cial
b-sheets 1 and 3 and triureas 4. Arti®cial b-sheets 6 contain
the same dipeptide groups .valylalanine and phenylalanyl-
leucine) as arti®cial parallel b-sheet 1 and the same
5-amino-2-methoxybenzoic hydrazide b-strand mimic as
arti®cial antiparallel b-sheet 3. Arti®cial b-sheet 6a
contains a two-carbon linker between its lower and middle
urea groups, while arti®cial b-sheet 6b contains a three-
carbon linker between these groups. We chose to evaluate
both linkers, because X-ray crystallographic and molecular
modeling studies raised concerns whether the shorter two-
carbon linker would allow the lower peptide strand to
properly align with the middle peptide strand in the three-
stranded structures.13 Previous studies in our laboratory
had shown that while both two- and three-carbon linkers
allowed oligoureas to form hydrogen-bonded turnlike
structures, the two-carbon linked structures are more stably
folded.16,17
For comparison, we also designed arti®cial b-sheets 5.
These arti®cial b-sheets contain the truncated 5-amino-2-
methoxybenzamide b-strand mimic that is present in arti-
®cial b-sheet 2. The former b-strand mimic is long enough
to hydrogen bond to the entire valylalanine dipeptide strand;
the latter is only long enough to hydrogen bond to half of it.
Like arti®cial b-sheets 6a and 6b, arti®cial b-sheet 5a

J. S. Nowick et al. / Tetrahedron 58 )2002) 727±739
729
Figure 2. X-Ray crystallographic structure of triurea template 4a.13
contains a two-carbon linker, and arti®cial b-sheet 5b
contains a three-carbon linker.
To gain insight into the structures of the b-strand mimic and
triurea templates, we performed single-crystal X-ray
crystallographic studies on simple crystalline compounds
containing these building blocks.18 Compound 7 contains
the 5-amino-2-methoxybenzoic hydrazide b-strand mimic
bearing a single urea group.12c X-Ray crystallography illus-
trates that it adopts an extended structure, with the urea,
aromatic, and diacylhydrazine groups roughly coplanar
.Fig. 1). This structure provides a linear arrangement of
three hydrogen-bonding groups that can hydrogen bond to
a tripeptide. We have previously reported the X-ray crystal-
lographic structures of triureas 4a .R1, R2, R3ˆPh) and 4b
.R1, R2, R3ˆMe).13 .The structure of 4a is reproduced here
in Fig. 2.) These studies have shown that the three urea
groups hydrogen bond together and align the substituents
Scheme 1.
R1±R3 in roughly the right orientation to allow b-sheet
formation.
1.2. Synthesis of arti®cial b-sheets 5 and 6
Arti®cial b-sheets 5 and 6 were ef®ciently synthesized from
diamines 813 by the procedures we reported previously for
the synthesis of triureas 4.19 Arti®cial b-sheets 5a and 6a
were prepared from diamine 8a, which contains a two-
carbon linker between its amino groups; arti®cial b-sheets
5b and 6b were prepared from diamine 8b, which contains a
three-carbon linker. Scheme 1 illustrates their syntheses.
Diamines 8a and b were converted to diureas 9a and b by
treatment ®rst with 1 equiv. of valylalanine methyl ester
isocyanate and then with a second equivalent of phenyl-
alanylleucine methyl ester isocyanate.20 The reactions of
diamines 8a and b with the isocyanates are highly regio-
selective, because the amino group at the middle position is
aliphatic and is relatively reactive, while the amino group at
the bottom position is aromatic and is less reactive.13 A third
amino group, at the top position, is protected as the Boc

730
J. S. Nowick et al. / Tetrahedron 58 )2002) 727±739
1
1.3. H NMR spectroscopic characterization and
assignment of arti®cial b-sheets 6
Arti®cial b-sheets 6a and 6b were thoroughly characterized
by 1H NMR spectroscopy in CDCl3 solution. First, the one-
dimensional spectra of these compounds were recorded at
500 MHz. The resonances were then numbered sequen-
tially, beginning at the up®eld region of the spectra, and
tabulated. When resonances were subsequently identi®ed
to comprise a group of overlapping peaks, prime .0) and
double-prime .00) designations were added to the numbers
to re¯ect the individual components of the peaks. Two-
dimensional PFG-COSY and Tr-ROESY21 spectra were
then recorded and used to assign the resonances. Molecular
modeling was used to aid in the assignments of resonances
from diastereotopic methyl and methylene groups and guide
the analysis of the Tr-ROESY data. Molecular models of 6a
and 6b were generated using MacroModel V5.0 with the
AMBERp force ®eld.22 The starting conformations .before
minimization) of the amino acid side-chains were chosen to
re¯ect measured coupling constants and NOEs when pos-
sible, but are largely arbitrary. Fig. 3a and b show these
molecular models and the assignments of the resonances
to their respective protons.
On the basis of these models and the NOE and coupling
constant data, the diastereotopic methyl and methylene
resonances for the valine and phenylalanine side chains of
both 6a and 6b were assigned with reasonable con®dence.
The diastereotopic leucine methylene resonances of 6a and
6b were also assigned with reasonable con®dence, as were
the leucine methyl resonances of 6a. Assignments of the
diastereotopic methylene resonances of the triamine back-
bones of 6a and 6b were not possible, with the exception
of the lower dimethylene group of 6a .PhNCH2CH2N).
Analysis of the coupling constants and NOEs associated
with this group strongly suggested it to adopt a single
gauche conformation .shown in Fig. 3a).
1
1.4. H NMR NOE studies of arti®cial b-sheets 6
The Tr-ROESY studies described earlier provide strong
evidence that arti®cial b-sheets 6a and 6b fold into b-sheet-
like structures. Most compelling among the data is an
extensive network of NOEs between the b-strand mimic
.upper strand) and the valylalanine dipeptide group .middle
strand) and between the valylalanine dipeptide group and
the phenylalanyl dipeptide group .lower strand) in both
compounds.23 Both arti®cial b-sheets 6a and 6b exhibit
NOEs between the 6-position of the aromatic ring of the
b-strand mimic and the valine a-proton and alanine side
chain, the hydrazide NHi proton and the alanine methyl
group, the isobutyryl methyl groups and alanine methyl-
amide methyl group, the phenylalanine a-proton and the
valine NH and side chain, the phenylalanine side chain
and the valine side chain, the alanine a-proton and the
leucine NH and side chain, and the alanine methylamide
NH group and the leucine methylamide methyl group.
Chart 1 succinctly shows these NOEs with arrows.
Figure 3. Models of arti®cial b-sheets 6a and 6b showing the assignments
of the 500 MHz 1H NMR spectral resonances of the compounds in 25 mM
CDCl3 at 298 K. The models were generated using MacroModel V5.0 with
the AMBERp force ®eld. The starting conformations .before minimization)
of the amino acid side-chains were chosen to re¯ect measured coupling
constants and NOEs when possible, but are largely arbitrary. The spectral
resonances for each compound were numbered 1±45, from up®eld .low
ppm) to down®eld .high ppm).
derivative and is not reactive. The Boc protective groups of
9a and b were removed by treatment with tri¯uoroacetic
acid .TFA), and the resulting primary amines were
cyanoethylated with acrylonitrile to form secondary amines
10a and b. The methyl ester groups of 10a and b were
converted to methylamide groups by aminolysis with
methylamine in methanol, affording secondary diamides
11a and b. Reaction of 11a and b with isocyanate 1215a
afforded arti®cial b-sheets 5a and b, while reaction of 11a
and b with isocyanate 1315a afforded arti®cial b-sheets 6a
and b.
A number of minor differences among the interstrand NOEs
observed for the two compounds are present. Arti®cial
b-sheet 6a exhibits an NOE between the alanine and leucine

J. S. Nowick et al. / Tetrahedron 58 )2002) 727±739
731
Chart 1.
side chains and between the alanine and leucine a-protons;
arti®cial b-sheet 6b exhibits an NOE between the hydrazide
NHi proton and the alanine NH group, and between the
phenylalanine side chain and the alanine methylamide
methyl group. These differing NOEs are generally weak
and may re¯ect small differences in the conformations or
conformational populations of the two compounds. The
presence of a weak NOE between the alanine and leucine
a-protons of 6a may, for example, indicate small deviation
from ideal b-sheetlike structure. The presence of a weak to
moderate NOE between the phenylalanine side chain and
the alanine methylamide methyl group in 6b may re¯ect
conformations in which the molecule twists in the fashion
characteristic of b-sheets to bring these groups into
proximity. While these differences in observed NOEs may
re¯ect meaningful conformational differences, it is also
possible that they merely re¯ect insigni®cant differences
between the signal-to-noise levels of the Tr-ROESY
experiments on the two compounds.
Chart 2.
this NH proton and the proton at its 4-position in both 6a and
6b. The intraresidue and interresidue NOEs between the NH
and a-protons of the phenylalanylleucine peptide strand are
comparable in strength in both compounds, suggesting that
the conformation of this strand is less b-strandlike than the
valylalanine peptide strand. The 3JHNa coupling constants of
the four amino acids in both compounds are consistent with
b-strandlike conformations, ranging from 8.3±9.3 Hz.24
Collectively, these NOE and coupling constant data support
a model in which both 6a and 6b adopt structures that are
basically b-sheetlike, with minor deviation from b-sheetlike
structure occurring in the lower peptide strand.
1.5. 1H NMR chemical shift studies of arti®cial b-sheets 6
1
Comparison of the H NMR chemical shifts of the NH
Intrastrand NOEs between the NH and a-protons of the
valylalanine peptide strand of 6a and 6b suggest that this
peptide strand adopts a b-strandlike conformation in both
compounds.23 Thus, this peptide strand shows strong inter-
residue NOEs between the NH and a-protons and weaker
intraresidue NOEs between the NH and a-protons in both
compounds. In a similar fashion, the upper b-strand mimic
shows a relatively strong NOE between the NH proton Hg
and the proton at its 6-position and little or no NOE between
groups of arti®cial b-sheets 6 to appropriate non-b-sheetlike
controls provides quantitative insight into their folding. In
CDCl3 solution, non-hydrogen-bonded peptide amide
protons resonate at about 6 ppm and non-hydrogen-bonded
peptide urea protons resonate at about 4.5 ppm.11,16,17 When
hydrogen bonded, these protons resonate ca. 2±2.5 ppm
further down®eld.11,16,17 A fully hydrogen-bonded peptide
amide proton typically appears slightly down®eld of 8 ppm;
a fully hydrogen-bonded peptide urea proton typically
appears at ca. 7 ppm. Amide and urea protons that are in
rapid equilibrium between hydrogen-bonded and non-
hydrogen-bonded states or are weakly hydrogen bonded
appear at intermediate chemical shifts. Other types of NH
protons .e.g. aromatic ureas and hydrazides) appear at
different positions and also shift down®eld upon hydrogen
bonding, with their degree of down®eld shifting roughly
re¯ecting their degree of hydrogen bonding.
Table 1. 1H NMR chemical shifts of the NH protons of 1±3, 5±7, and 14±
18
Ha
Hb
Hc
Hd
He
Hf
Hg
Hh
Hi
1
2
3
4.61 7.37 6.23 6.68 6.75 6.96
5.02 8.20 5.45
4.82 8.29 5.79
10.00 8.11
9.93 10.97 10.45
5a 4.62 7.16 6.53 6.57 8.46 7.00
5b 4.67 7.03 6.27 6.27 8.41 6.84 10.05
6a 4.38 8.16 6.72 6.17 8.53 7.94 10.01 10.87 11.07
6b 4.75 8.22 5.87 6.96 8.79 7.36 10.42 11.04 11.22
9.98
8.14
8.13
For these studies, the chemical shifts of arti®cial b-sheets 6
were recorded at 1.0 mM in CDCl3 at 295 K and compared
to controls 7 and 15±18.25 Control 7 resembles the b-strand
mimic .upper strand) of 6.12c Controls 15 and 16 respec-
tively resemble the entire middle .valylalanine) peptide
strand and the right-hand portion of this peptide strand;
controls 17 and 18 respectively resemble the entire lower
.phenylalanylleucine) peptide strand and the right-hand
portion of this peptide strand.11,12a Two compounds are
7
14
6.31 10.85
6.31 7.92
8.78
15
16
4.75 6.34 6.58
6.01 6.01
17 4.43 6.37 6.71
5.82 6.00
18
Spectra were recorded at 295 K in 1.0 mM CDC13 solution.

732
J. S. Nowick et al. / Tetrahedron 58 )2002) 727±739
Comparison of the chemical shifts of the NH groups of
arti®cial b-sheets 6a and 6b to those of controls 7 and
15±18 indicates that protons Hb, Hd±Hg, and Hi are largely
or wholly hydrogen bonded while protons Ha and Hc are
largely not hydrogen bonded. Chart 3 summarizes these
chemical shift differences and also shows the chemical
shift differences between arti®cial b-sheets 1±3 and 5 and
the corresponding controls, which are discussed subse-
quently. In arti®cial b-sheets 6a and 6b, protons Hb, He,
and Hi appear 1.79±2.45 ppm down®eld of their respective
controls, indicating that they are largely or wholly hydrogen
bonded. Proton Hd appears 2.21 ppm down®eld of control
15 in arti®cial b-sheet 6b but only 1.42 ppm down®eld in
arti®cial b-sheet 6a. This unusually small down®eld shifting
may re¯ect that the NH group in this compound must hydro-
gen bond to the p-face of the carbonyl, rather than to the
lone pair of electrons, to allow the lower dimethylene group
of 6a .PhNCH2CH2N) to adopt a gauche conformation
.discussed earlier and shown in Fig. 3a).
Proton Hf appears 1.93 ppm down®eld of control 16 in
arti®cial b-sheet 6a but only 1.35 ppm down®eld in arti®cial
b-sheet 6b. The unusually small down®eld shifting of this
proton in 6b suggests that it is hydrogen bonded slightly
more than half of the time and may re¯ect some fraying
of the lower right-hand corner of the b-sheet. This deviation
in shift mirrors some of the variations observed in the intra-
and interresidue NOEs, discussed earlier.
Proton Hg appears exceptionally far .3.70±4.41 ppm) down-
®eld of control 7 in arti®cial b-sheets 6. The unusually large
down®eld shifting of this proton may result from anisotropic
effects associated with changes in the torsion angle about
the bond between the urea group and the aromatic ring of the
b-strand mimic or may re¯ect an exceptionally strong
hydrogen bond in which the carbonyl group of the middle
urea is jammed into the NHg group of the upper urea.12a
Proton Ha appears within 20.05±0.32 ppm of control 17,
indicating it to not be signi®cantly hydrogen bonded in
either compound. Proton Hc appears within 20.13 ppm of
control 18 in arti®cial b-sheet 6b, while it appears 0.72 ppm
down®eld of this control in arti®cial b-sheet 6a. This down-
®eld shifting of Hc in 6a provides further evidence for the
bottom peptide strand's partial deviation from b-sheetlike
structure in this compound. Proton Hh, which is hydrogen
bonded in control 7, is also hydrogen bonded in arti®cial
b-sheets 6 and appears at nearly the same chemical shifts in
all three compounds.
1.6. Comparison of arti®cial b-sheets 6 to other arti®cial
b-sheets
Chart 3.
1
Comparison of the H NMR chemical shifts of the NH
used as controls for each of the peptide strands, because the
longer compounds tend to adopt b-turnlike conformations
in which the methylamide NH groups are hydrogen bonded
to the urea carbonyl groups. Thus, the shorter compounds
serve as better controls for the methylamide NH groups.
Table 1 summarizes the chemical shifts of the NH reso-
nances of these compounds. Also included are shift data
for arti®cial b-sheets 1±3, and 5 and control 14, which are
discussed subsequently.11,12a,c Chart 2 shows the assign-
ments of these resonances to their respective protons,
which are designated Ha±Hi.
protons of arti®cial b-sheets 6 to those of arti®cial b-sheets
1 and 5 provides insight into the effects of the b-strand
mimic upon the structure of 6. Two-stranded arti®cial
b-sheet 1 contains the two peptide strands that are present
in 6 with a two-carbon linker .like 6a). Three-stranded arti-
®cial b-sheets 5a and 5b contain the same two peptide
strands and a truncated version of the b-strand mimic that
lacks the hydrogen-bond donor .NHi) of the hydrazide
group. Chart 3 provides a convenient comparison of the
down®eld shifting of the NH protons in these compounds

J. S. Nowick et al. / Tetrahedron 58 )2002) 727±739
733
relative to controls. Chart 3 also provides a comparison to
two-stranded arti®cial b-sheets 2 and 3, which are smaller
homologues of 5 and 6 that contain the same b-strand
mimics and phenylalanylleucine peptide strand but lack
the middle valylalanine peptide strand.
b-strandlike conformation. Most notably, the leucine NHb
proton of 3 appears 1.92 ppm down®eld of the correspond-
ing proton in control 17, while in 1 it appears only 1.00 ppm
down®eld. Protons Hg and Hi in 3 are also more strongly
shifted down®eld of their respective controls than are the
spatially related protons Hd and Hf in 1, however these
protons differ too much in environment and type to allow
meaningful comparison.
Particularly notable is that protons Hb and Hf are shifted
much further down®eld in 6a and 6b than in 5a and 5b or
in 1. In 6a, Hb and Hf appear 1.79 and 1.93 ppm down®eld of
the corresponding protons in controls 17 and 16; in 6b Hb
and Hf appear 1.85 and 1.35 ppm down®eld. In 5a, Hb and
Hf appear only 0.79 and 0.99 ppm down®eld; in 5b, these
protons appear only 0.66 and 0.83 ppm down®eld. In 1, they
appear 1.00 and 0.95 ppm down®eld. The down®eld shifting
of these protons suggests that the peptide strands of arti®cial
b-sheets 6 are largely hydrogen bonded while those of 1
and 5 are a little less than half hydrogen bonded .assuming
a dynamic equilibrium in which 2±2.5 ppm down®eld
shifting corresponds to complete hydrogen bonding).
2. Conclusion
These synthetic and structural studies establish a paradigm
for the creation of small peptidelike molecules that fold in
chloroform solution to resemble three-stranded mixed
b-sheets. Central to this paradigm is the use of a triurea
molecular scaffold to hold the three strands next to each
other in suitable orientations. The middle and lower strands
are peptides and are attached at their N-termini, while the
upper strand is a peptidomimetic template that resembles a
peptide b-strand that is attached by its C-terminus. The
b-strand mimic may be viewed as acting in conjunction
with the triurea template to form a `corner bracket' that
stabilizes b-sheet structure in both of the attached peptide
strands, and the folding of the arti®cial b-sheets may be
viewed as being cooperative, with the b-strand mimic
helping to organize the middle peptide strand in a b-strand-
like conformation and thus augmenting its interaction with
the lower peptide strand.
Also notable is that Hi is shifted further down®eld in 6a and
6b than in 3. In 6a and 6b, Hi appears 2.29±2.44 ppm down-
®eld of the corresponding proton in control 7, while in 3, Hi
appears only 1.67 ppm down®eld. This greater degree of
down®eld shifting suggests more complete hydrogen bond-
ing between the upper and middle strands of three-stranded
arti®cial b-sheets 6 than between the strands of two-
stranded arti®cial b-sheet 3. Further support for the greater
degree of folding of the three stranded structures comes
from comparison of the down®eld shifting of the alanine
NHe proton in arti®cial b-sheets 5 and 6 to the down®eld
shifting of the leucine NHb proton in arti®cial b-sheets 2 and
3. In arti®cial b-sheets 5 and 6, He appears 2.07±2.45 ppm
down®eld of the corresponding proton in control 15, while
in 2 and 3, Hb appears only 1.83±1.92 ppm down®eld of the
corresponding proton in control 17. This comparison is
fairly reasonable, because the spatial relationship of the
protons .He and Hb) to the b-strand mimics of the different
compounds is the same, even though they belong to different
amino acid residues.26
Although substantial differences between the three-stranded
mixed arti®cial b-sheets containing two- and three-carbon
linkers between the lower and middle urea groups were
anticipated, only minor differences were found. Both the
two- and three-carbon linkers afford arti®cial b-sheets in
which the middle peptide strand is well ordered and the
lower peptide strand is slightly less ordered. While the
peptide strands in the current study are relatively short
.dipeptides), previous studies in our laboratories have
shown that it is possible to prepare two-stranded arti®cial
b-sheets containing longer peptide strands and extended
versions of the b-strand mimic or the b-strand mimic linked
to a peptide.12b,d The preparation of three-stranded mixed
arti®cial b-sheets containing longer strands such as these
should also be possible. Collectively, these studies show
that our modular approach to the creation of structures
that fold to resemble protein b-sheets is robust and pave
the way for even larger and more complex structures that
mimic the complex topologies of proteins.
Together, these data indicate that arti®cial b-sheets 6 are
more completely folded than arti®cial b-sheets 1, 3, or 5.
The lower two peptide strands of 6 are more thoroughly
paired than the two peptide strands in 1, the upper two
strands of 6 are more thoroughly paired than the two strands
in 3, and the fully templated three-stranded structure of 6 is
better folded than the partially templated three-stranded
structure of 5. That arti®cial b-sheets 6 are better folded
than arti®cial b-sheets 1, 3, or 5 indicates that the folding
of 6 is cooperative. These ®ndings of cooperativity in this
mixed b-sheet system match the ®ndings of cooperativity in
the folding of three-stranded antiparallel peptide b-sheets
3. Experimental
by the groups of Searle,10a,i Gellman,10c and Kortemme and
3.1. Data for compounds
10e
Â
Ramõrez-Alvarado.
3.1.1. Diamine 8b. A 500-mL, three-necked, round-
bottomed ¯ask, equipped with a nitrogen inlet adapter, a
glass stopper, a gas inlet adapter ®tted with a hydrogen
balloon, and a magnetic stirring bar was charged with
250 mL of methanol, potassium acetate .8.80 g, 89.7
mmol), Boc-glycinal27 .8.28 g, 28.2 mmol), N-phenyl-
propylenediamine17 .6.70 g, 44.6 mmol), and 7.61 g of
10% Pd/C. The ¯ask was evacuated and ®lled with nitrogen
The current studies do not establish whether the extra
stability of 6 arises only because of cooperativity, or
whether the relatively rigid b-strand mimic of 6 also imparts
extra stability beyond that which would be provided by a
third peptide strand. A comparison of arti®cial b-sheets 1
and 3 suggests that the b-strand mimic does impart some
special stability associated with its preorganization in a

734
J. S. Nowick et al. / Tetrahedron 58 )2002) 727±739
twice, and then evacuated, ®lled with hydrogen, and main-
tained under a hydrogen atmosphere for 13 h. The reaction
mixture was ®ltered and concentrated by rotary evaporation.
The residue was dissolved in 150 mL of 1.0 M NaOH and
the solution was extracted with three 100-mL portions of
methylene chloride. The combined organic layers were
dried over Na2SO4 and concentrated by rotary evaporation
to 13.77 g of a yellow oil. The crude product was puri®ed by
column chromatography on silica gel .EtOAc±i-PrOH, 1:1)
to afford 3.05 g of 8b as a pale yellow oil .23%): IR .neat)
3365, 1697 cm21; 1H NMR .500 MHz, CDCl3) d 7.15 .dd,
Jˆ8.5, 7.3 Hz, 2H), 6.67 .tt, Jˆ7.3, 1.0 Hz, 1H), 6.59 .dd,
Jˆ8.6, 1.0 Hz, 2H), 5.06 .br. s, 1H), 3.20 .br. q, Jˆ5.6 Hz,
2H), 3.16 .t, Jˆ6.6 Hz, 2H), 2.71 .t, Jˆ6.7 Hz, 2H), 2.70 .t,
Jˆ5.9 Hz, 2H), 1.76 .quintet, Jˆ6.6 Hz, 2H), 1.44 .s, 9H);
13C NMR .125 MHz, CDCl3) d 156.0, 148.4, 129.0, 116.9,
112.5, 79.0, 49.1, 47.6, 42.4, 40.1, 29.3, 28.3; HRMS m/e
calcd for C16H28N3O2 .M1H)1, 294.2181 found: 294.2181;
Anal. Calcd for C16H27N3O2: C, 65.50; H, 9.28; N, 14.32.
Found: C, 65.16; H, 9.04; N, 14.08.
3.1.3. Diurea 9b. Diamine 8b was converted to diurea 9b in
84% yield, as described earlier: IR .CHCl3) 3421, 3338,
1
1741, 1672, 1645 cm21; H NMR .500 MHz, CDCl3) d
7.36 .appar t, Jˆ7.2 Hz, 2H), 7.31 .appar t, Jˆ7.3 Hz,
1H), 7.23±7.19 .m, 3H), 7.04±7.01 .m, 4H), 6.93 .d, Jˆ
6.9 Hz, 1H), 6.83 .d, Jˆ7.8 Hz, 1H), 5.93 .d, Jˆ7.9 Hz,
1H), 5.30 .appar t, Jˆ5.0 Hz, 1H), 4.60±4.52 .m, 3H),
4.49 .td, Jˆ8.2, 5.3 Hz, 1H), 4.18 .t, Jˆ7.4 Hz, 1H), 3.71
.s, 3H), 3.70 .s, 3H), 3.69±3.58 .m, 2H), 3.35±3.17 .m, 6H),
2.99 .dd ABX pattern, JABˆ14.0 Hz, JAXˆ5.6 Hz, 1H), 2.91
.dd ABX pattern, JABˆ14.0 Hz, JBXˆ6.9 Hz, 1H), 2.21±
2.15 .m, 1H), 1.78±1.65 .m, 2H), 1.62±1.45 .m, 3H), 1.41
.s, 9H), 1.37 .d, Jˆ7.2 Hz, 3H), 0.95 .appar t, Jˆ7.5 Hz,
6H), 0.89 .d, Jˆ6.2 Hz, 3H), 0.87 .d, Jˆ6.2 Hz, 3H); 13C
NMR .125 MHz, CDCl3) d 173.2, 173.0, 172.2, 171.5,
157.6, 156.3, 140.5, 136.6, 130.0, 129.1, 128.2, 128.1,
127.8, 126.4, 79.1, 59.8, 54.9, 52.0, 50.6, 47.6, 46.7, 46.3,
45.1, 40.7, 39.4, 37.6, 30.7, 28.1, 27.4, 24.5, 22.5, 21.7,
19.2, 18.2, 17.6; HRMS .FAB) m/e for C43H66N7O10
.M1H)1, calcd 840.4871 found: 840.4882; Anal. Calcd
for C43H65N7O10: C, 61.48; H, 7.80; N, 11.67. Found: C,
61.34; H, 7.90; N, 11.47.
3.1.2. Diurea 9a. A solution of diamine 8a13 .0.178 g,
0.637 mmol) and 0.224 g of valylalanine methyl ester iso-
1
cyanate .70% purity by H NMR analysis, 0.69 mmol) in
3.1.4. Secondary amine 10a. A solution of diurea 9a
.0.317 g, 0.383 mmol) in 15 mL of methylene chloride
and 1.2 mL of tri¯uoroacetic acid was stirred for 90 min
and then concentrated by rotary evaporation. The residue
was partitioned between 20 mL of methylene chloride and
20 mL of saturated aqueous sodium bicarbonate. The
aqueous layer was extracted with two 20-mL portions of
methylene chloride, and the combined organic layers were
dried over Na2SO4 and concentrated by rotary evaporation
to afford 0.278 g .100%) of PhN.CO±Phe±Leu±OMe)-
CH2CH2N.CO±Val±Ala±OMe)CH2CH2NH2 as a colorless
oil.
30 mL of methylene chloride was stirred under nitrogen for
1 h and then concentrated by rotary evaporation. The
residue was puri®ed by column chromatography on silica
gel .EtOAc±hexanes, 2:1) to afford 0.319 g .99%) of
PhNHCH2CH2N.CO±Val±Ala±OMe)CH2CH2NHBoc as a
white foam.
A
mixture of PhNHCH2CH2N.CO±Val±Ala±OMe)-
CH2CH2NHBoc .0.296 g, 0.583 mmol) and phenylalanyl-
leucine methyl ester isocyanate20 .0.225 g, 0.710 mmol) in
13 mL of 1,2-dichloroethane was heated at re¯ux under
nitrogen for 36 h and then concentrated by rotary evapo-
ration. Radial thin-layer chromatography on silica gel
.sequential elution with EtOAc±hexanes, 3:2, EtOAc±
hexanes, 2:1, and EtOAc±i-PrOH, 95:5) afforded 0.359 g
.75%) of diurea 9a as a sticky white foam: IR .CHCl3)
A solution of PhN.CO±Phe±Leu±OMe)CH2CH2N.CO±
Val±Ala±OMe)CH2CH2NH2 .0.278 g, 0.383 mmol) in
20 mL of methanol and 0.5 mL acrylonitrile .0.4 g, 8
mmol) was stirred under nitrogen for 7 h and then concen-
trated by rotary evaporation. Column chromatography on
silica gel .EtOAc±i-PrOH, 9:1) afforded 0.249 g .84%) of
secondary amine 10a as a white foam: IR .CHCl3) 3419,
3419, 3338, 1741, 1697, 1674, 1645 cm21 1H NMR
;
.500 MHz, CDCl3) d 7.38±7.35 .m, 2H), 7.34±7.30 .m,
1H), 7.25±7.23 .m, 3H), 7.10±7.08 .m, 2H), 7.01±6.99
.m, 2H), 6.96 .d, Jˆ7.0 Hz, 1H), 6.77 .br s, 1H), 6.45 .br
s, 1H), 5.23 .br t, Jˆ4.3 Hz, 1H), 4.76 .d, Jˆ7.7 Hz, 1H),
4.61 .q, Jˆ7.2 Hz, 1H), 4.57 .q, Jˆ7.3 Hz, 1H), 4.48 .td,
Jˆ8.3, 5.3 Hz, 1H), 4.15 .t, Jˆ7.4 Hz, 1H), 3.78±3.71 .m,
1H), 3.71 .s, 3H), 3.70 .s, 3H), 3.62±3.55 .m, 1H), 3.49±
3.36 .m, 3H), 3.31±3.20 .m, 2H), 3.19±3.12 .m, 1H), 3.02
.dd, ABX pattern JABˆ13.9 Hz, JAXˆ6.7 Hz, 1H), 2.96 .dd
ABX pattern, JABˆ14.0 Hz, JBXˆ7.4 Hz, 1H), 2.29 .octet,
Jˆ6.7 Hz, 1H), 1.61±1.54 .m, 1H), 1.52±1.45 .m, 2H),
1.388 .s, 9H), 1.385 .d, Jˆ6.8 Hz, 3H), 1.04 .d, Jˆ
6.8 Hz, 3H), 1.02 .d, Jˆ6.8 Hz, 3H), 0.88 .d, Jˆ6.2 Hz,
3H), 0.86 .d, Jˆ6.2 Hz, 3H); 13C NMR .125 MHz,
CDCl3) d 173.5, 172.9, 172.2, 171.2, 158.0, 156.6, 156.3,
141.1, 136.7, 130.2, 129.2, 128.4, 128.3, 128.0, 127.5,
126.7, 79.0, 60.8, 55.1, 52.0, 50.7, 48.9, 47.6, 47.3, 46.2,
40.8, 39.7, 37.8, 30.4, 28.2, 24.7, 22.6, 21.7, 19.4, 18.6,
18.1; HRMS .FAB) m/e for C42H64N7O10 .M1H)1, calcd
826.4714, found: 826.4717; Anal. Calcd for C42H63N7O10:
C, 61.07; H, 7.69; N, 11.87. Found: C, 61.19; H, 7.72; N,
11.77.
1
3325, 2251, 1741, 1672, 1643 cm21; H NMR .500 MHz,
CDCl3) d 7.36 .t, Jˆ7.4 Hz, 2H), 7.32±7.29 .m, 1H), 7.25±
7.19 .m, 3H), 7.08±7.06 .m, 2H), 7.04 .d, Jˆ7.7 Hz, 2H),
6.88 .br d, Jˆ5.8 Hz, 1H), 6.80 .d, Jˆ8.4 Hz, 1H), 6.65 .br
s, 1H), 4.75 .d, Jˆ7.4 Hz, 1H), 4.59±4.52 .m, 2H), 4.48 .td,
Jˆ8.0, 5.5 Hz, 1H), 4.09 .t, Jˆ7.5 Hz, 1H), 3.78±3.63 .m,
2H), 3.71 .s, 3H), 3.70 .s, 3H), 3.50±3.44 .m, 1H), 3.43±
3.35 .m, 2H), 3.33±3.28 .m, 1H), 3.01 .dd ABX pattern,
JABˆ13.9 Hz, JAXˆ6.4 Hz, 1H), 2.95 .dd ABX pattern,
JABˆ14.0 Hz, JBXˆ7.5 Hz, 1H), 2.93±2.87 .m, 2H), 2.80±
2.72 .m, 2H), 2.56±2.45 .m, 2H), 2.16 .sextet, Jˆ6.7 Hz,
1H), 1.68 .br s, 1H), 1.61±1.55 .m, 1H), 1.54±1.46 .m, 2H),
1.36 .d, Jˆ7.2 Hz, 3H), 0.98 .d, Jˆ 6.7 Hz, 6H), 0.89 .d,
Jˆ6.2 Hz, 3H) 0.86 .d, Jˆ6.2 Hz, 3H); 13C NMR
.125 MHz, CDCl3) d 173.4, 173.0, 172.2, 171.2, 158.6,
156.6, 141.3, 136.8, 130.2, 129.3, 128.4, 128.0, 127.7,
126.7, 118.7, 60.5, 55.2, 52.20, 52.17, 50.8, 48.8, 48.4,
47.8, 46.4, 45.2, 41.0, 37.7, 30.7, 24.7, 22.7, 21.8, 19.4,
18.5, 18.1; HRMS .FAB) m/e for C40H59N8O8 .M1H)1,
calcd 779.4455, found: 779.4456; Anal. Calcd for

J. S. Nowick et al. / Tetrahedron 58 )2002) 727±739
735
C40H58N8O9: C, 61.68; H, 7.51; N, 14.39. Found: C, 61.68;
H, 7.87; N, 14.02.
6.5 Hz, 1H), 3.79±3.74 .m, 1H), 3.49±3.38 .m, 3H), 3.29±
3.16 .m, 2H), 3.01 .dd ABX pattern, JABˆ13.8 Hz, JAX
ˆ
6.1 Hz, 1H), 2.94 .td, Jˆ6.6, 1.3 Hz, 2H), 2.88±2.77 .m,
3H), 2.73 .d, Jˆ4.8 Hz, 3H), 2.70 .d, Jˆ4.8 Hz, 3H), 2.60
.dt, Jˆ16.8, 6.7 Hz, 1H), 2.55 .dt, Jˆ16.8, 6.5 Hz, 1H), 2.14
.octet, Jˆ6.7 Hz, 1H), 1.86±1.65 .m, 4H), 1.59±1.45 .m,
2H), 1.37 .d, Jˆ7.1 Hz, 3H), 0.95 .appar t, Jˆ6.4 Hz, 6H),
0.90 .d, Jˆ6.6 Hz, 3H), 0.87 .d, Jˆ6.5 Hz, 3H); 13C NMR
.125 MHz, CDCl3) d 172.8, 172.6, 172.5, 171.8, 159.0,
156.7, 140.5, 136.2, 130.2, 128.9, 128.5, 128.1, 126.7,
118.7, 60.4, 55.8, 51.6, 48.8, 48.6, 48.4, 46.9, 45.3, 45.2,
40.1, 37.4, 30.7, 27.5, 26.2, 26.1, 24.7, 23.1, 21.5, 19.4,
18.3, 17.6; HRMS .FAB) m/e for C41H63N10O6 .M1H)1,
calcd 791.4932, found: 791.4927; Anal. Calcd for
C41H62N10O6: C, 62.26; H, 7.90; N, 17.71. Found: C,
62.18; H, 8.08; N, 17.43.
3.1.5. Secondary amine 10b. Diurea 9b was converted to
secondary amine 10b in 82% yield, as described earlier: IR
1
.CHCl3) 3423, 3294, 2251, 1741, 1670, 1645 cm21; H
NMR .500 MHz, CDCl3) d 7.36 .appar t, 7.1 Hz, 2H),
7.31 .appar t, Jˆ7.3 Hz, 1H), 7.23±7.19 .m, 3H), 7.05±
7.01 .m, 4H), 6.85 .d, Jˆ7.7 Hz, 1H), 6.65 .d, Jˆ 7.2 Hz,
1H), 6.64 .br s, 1H), 4.57 .td, Jˆ7.6, 6.3 Hz, 1H), 4.52±4.45
.m, 3H), 4.07 .dd, Jˆ8.5, 6.6 Hz, 1H), 3.73 .s, 3H), 3.71 .s,
3H), 3.71±3.66 .m, 1H), 3.60±3.55 .m, 1H), 3.47±3.34 .m,
2H), 3.25±3.19 .m, 2H), 3.00±2.89 .m, 4H), 2.86±2.76 .m,
2H), 2.63 .dt, Jˆ16.8, 6.9 Hz, 1H), 2.55 .dt, Jˆ16.9, 6.7 Hz,
1H), 2.08 .octet, Jˆ6.7 Hz, 1H), 1.84 .br s, 1H), 1.69
.quintet, Jˆ7.1 Hz, 2H), 1.62±1.50 .m, 3H), 1.37 .d, Jˆ
7.2 Hz, 3H), 0.95 .d, Jˆ6.8 Hz, 3H), 0.93 .d, Jˆ6.8 Hz,
3H), 0.90 .d, Jˆ6.0 Hz, 3H), 0.87 .d, Jˆ6.0 Hz, 3H); 13C
NMR .125 MHz, CDCl3) d 173.0, 172.9, 172.3, 171.6,
158.7, 156.0, 140.4, 136.6, 129.8, 129.0, 128.00, 127.97,
127.6, 126.2, 118.6, 59.4, 54.6, 51.8, 50.5, 48.3, 47.6,
46.5, 45.2, 44.9, 40.5, 37.5, 30.8, 27.1, 24.4, 22.4, 21.5,
19.0, 18.1, 18.0, 17.2; HRMS .FAB) m/e for C41H61N8O8
.M1H)1, calcd 793.4612, found: 793.4635; Anal. Calcd for
C41H60N8O8: C, 62.10; H, 7.63; N, 14.13. Found: C, 61.99;
H, 7.83; N, 13.89.
3.1.8. Arti®cial b-sheet 5a. A solution of diamide 11a
.0.104 g, 0.134 mmol) and isocyanate 1215a .0.030 g,
0.145 mmol) in 15 mL of methylene chloride was stirred
under nitrogen for 6 h and then concentrated by rotary
evaporation. Radial thin-layer chromatography on silica
gel .EtOAc±i-PrOH, 9:1) followed by preparative reverse-
phase HPLC .C18 column, 40:60 acetonitrile±water eluant)
afforded 0.118 g .90%) of arti®cial b-sheet 5a as a glassy
white solid: IR .CHCl3) 3410, 3294, 2251, 1657,
1645 cm21 1H NMR .500 MHz, CDCl3) d 9.92 .br s,
;
3.1.6. Diamide 11a. A solution of secondary amine 10a
.0.224 g, 0.288 mmol) in 20 mL of a 47% .w/w) solution
of methylamine in methanol was stirred for 45 min and then
concentrated by rotary evaporation to afford 0.227 g .101%)
of diamide 11a as a sticky white foam: IR .CHCl3) 3417,
1H), 8.422 .d, Jˆ1.5 Hz, 1H), 8.418 .br s, 1H), 8.26 .dd,
Jˆ8.8, 1.8 Hz, 1H), 8.15 .br q, Jˆ4.7 Hz, 1H), 7.37±7.29
.m, 4H), 7.20±7.17 .m, 3H), 7.05±6.99 .m, 5H), 6.89 .d,
Jˆ9.1 Hz, 1H), 6.73±6.61 .m, 2H), 4.70 .d, Jˆ6.9 Hz, 1H),
4.59 .q, Jˆ6.7 Hz, 1H), 4.44±4.34 .m, 2H), 4.11 .t, Jˆ
8.3 Hz, 1H), 3.93 .s, 3H), 3.78±3.67 .m, 2H), 3.60±3.52
.m, 2H), 3.47±3.41 .m, 1H), 3.37±3.24 .m, 5H), 3.05 .dd,
ABX pattern, JABˆ13.8 Hz, JAXˆ6.1 Hz, 1H), 3.00 .d, Jˆ
1
3338, 2251, 1659 cm21; H NMR .500 MHz, CDCl3) d
7.37±7.27 .m, 4H), 7.20±7.17 .m, 4H), 7.09 .br s, 1H),
7.03 .d, Jˆ7.0 Hz, 2H), 6.98±6.97 .m, 2H), 6.79 .d, Jˆ
6.5 Hz, 1H), 6.76 .br s, 1H), 6.69 .d, Jˆ6.0 Hz, 1H),
4.56±4.52 .m, 1H), 4.50 .quintet, Jˆ7.2 Hz, 1H), 4.40 .td,
Jˆ8.9, 4.7 Hz, 1H), 4.08 .t, Jˆ6.7 Hz, 1H), 3.83±3.79 .m,
1H), 3.70±3.65 .m, 1H), 3.59±3.53 .m, 1H), 3.36±3.27 .m,
4.7 Hz, 3H), 2.95 .dd, ABX pattern, JABˆ13.7 Hz, JBX
ˆ
7.2 Hz, 1H), 2.70 .d, Jˆ5.0 Hz, 3H), 2.69 .d, Jˆ5.0 Hz,
3H), 2.69±2.63 .m, 1H), 2.58 .dt, Jˆ16.9, 6.1 Hz, 1H),
2.16±2.08 .m, 1H), 1.66 .ddd, Jˆ13.9, 8.6, 5.4 Hz, 1H),
1.53 .ddd, Jˆ14.5, 9.7, 5.6 Hz), 1.46±1.38, .m, 1H), 1.22
.d, Jˆ7.0 Hz, 3H), 1.08 .d, Jˆ6.5 Hz, 3H), 1.04 .d,
Jˆ6.6 Hz, 3H), 0.84 .d, Jˆ6.6 Hz, 3H), 0.81 .d, Jˆ
6.5 Hz, 3H); 13C NMR .125 MHz, CDCl3) d 173.0, 172.7,
172.4, 171.5, 166.0, 159.0, 157.4, 155.4, 152.5, 141.3,
136.3, 134.4, 130.2, 129.0, 128.4, 128.1, 127.4, 126.7,
122.6, 122.1, 120.5, 118.8, 111.6, 62.1, 56.0, 55.8, 51.8,
49.7, 49.1, 48.5, 47.6, 47.3, 45.2, 39.8, 37.6, 30.0, 26.4,
26.1, 25.9, 24.6, 22.8, 21.6, 19.5, 19.3, 17.5, 16.8; HRMS
.FAB) m/e calcd for C50H71N12O9 .M1H)1, 983.5467
found: 983.5472; Anal. Calcd for C50H70N12O9: C, 61.08;
H, 7.18; N, 17.10. Found: C, 61.43; H, 7.10; N, 17.07.
2H), 3.22±3.17 .m, 1H), 3.02 .dd ABX pattern, JAB
13.7 Hz, JAXˆ6.2 Hz, 1H), 2.94 .dd ABX pattern, JAB
ˆ
ˆ
13.8 Hz, JBXˆ7.3 Hz, 1H), 2.88 .t, Jˆ6.5 Hz, 2H), 2.79±
2.69 .m, 2H), 2.71 .d, Jˆ5.1 Hz, 3H), 2.70 .d, Jˆ5.4 Hz,
3H), 2.53±2.46 .m, 2H), 2.15 .octet, Jˆ6.6 Hz, 1H), 1.72
.ddd, Jˆ13.9, 9.0, 5.0 Hz, 1H), 1.62±1.55 .m, 1H), 1.52±
1.44 .m, 1H), 1.36 .d, Jˆ7.0 Hz, 3H), 0.97 .appar t, Jˆ
5.9 Hz, 6H), 0.89 .d, Jˆ6.6 Hz, 3H), 0.86 .d, Jˆ6.5 Hz,
3H); 13C NMR .125 MHz, CDCl3) d 172.9, 172.5, 172.3,
171.7, 158.7, 156.8, 141.4, 136.3, 130.1, 129.0, 128.4,
128.0, 127.5, 126.7, 118.6, 60.8, 55.6, 51.8, 48.8, 48.6,
48.2, 48.0, 45.9, 45.1, 40.0, 37.4, 30.6, 26.1, 26.0, 24.6,
23.0, 21.6, 19.4, 18.4, 18.3, 18.1; HRMS .FAB) m/e for
C40H61N10O6 .M1H)1, calcd 777.4775, found: 777.4773;
Anal. Calcd for C40H60N10O6: C, 61.83; H, 7.78; N, 18.03.
Found: C, 62.14; H, 7.97; N, 18.14.
3.1.9. Arti®cial b-sheet 5b. Diamide 11b was converted to
arti®cial b-sheet 5b in 93% yield by treatment with iso-
cyanate 12,15a as described earlier: IR .CHCl3) 3410,
1
3294, 2251, 1659 cm21; H NMR .500 MHz, CDCl3) d
3.1.7. Diamide 11b. Secondary amine 10b was converted to
diamide 11b in 98% yield as described earlier: IR .CHCl3)
3415, 3325, 2251, 1657 cm21; 1H NMR .500 MHz, CDCl3)
d 7.39±7.33 .m, 3H), 7.22±7.15 .m, 3H), 7.07±7.04 .m,
3H), 6.94±6.92 .m, 2H), 6.91 .d, Jˆ8.7 Hz, 1H), 6.81 .br
q, Jˆ4.7 Hz, 1H), 6.68 .br q, Jˆ4.5 Hz, 1H), 6.63 .br s, 1H),
4.55 .d, Jˆ6.5 Hz, 1H), 4.50±4.38 .m, 3H), 4.03 .dd, Jˆ7.3,
10.05 .br s, 1H), 8.43 .br s, 1H), 8.42 .d, Jˆ2.4 Hz, 1H),
8.32 .dd, Jˆ8.9, 2.4 Hz, 1H), 8.14 .q, Jˆ4.8 Hz, 1H), 7.41
.appar t, Jˆ7.1 Hz, 2H), 7.37 .appar t, Jˆ7.2 Hz, 1H),
7.22±7.20 .m, 3H), 7.13 .br d, Jˆ7.8 Hz, 1H), 7.06±7.04
.m, 2H), 7.01±6.99 .m, 2H), 6.91 .br s, 1H), 6.90 .d, Jˆ
9.1 Hz, 1H), 6.39 .br s, 1H), 6.34 .br s, 1H), 4.68 .d, Jˆ
7.0 Hz, 1H), 4.63 .q, Jˆ7.0 Hz, 1H), 4.42±4.38 .m, 1H),

736
J. S. Nowick et al. / Tetrahedron 58 )2002) 727±739
.dt, Jˆ17.0, 6.3 Hz, 1H), 2.54 .dt, Jˆ17.0, 6.1 Hz, 1H), 1.99
.octet, Jˆ6.9 Hz, 1H), 1.74 .ddd, Jˆ14.1, 8.5, 5.7 Hz, 1H),
1.65 .ddd, Jˆ14.5, 9.5, 5.7 Hz, 1H), 1.27 .d, Jˆ6.9 Hz, 3H),
1.24 .d, Jˆ6.9 Hz, 3H), 1.23±1.17 .m, 1H), 1.12 .d, Jˆ
6.8 Hz, 3H), 1.03 .d, Jˆ6.7 Hz, 3H), 0.95 .d, Jˆ6.7 Hz,
3H), 0.78 .d, Jˆ6.7 Hz, 3H), 0.72 .d, Jˆ6.6 Hz, 3H); 13C
NMR .125 MHz, CDCl3) d 173.1, 173.0, 172.2, 172.0,
159.8, 159.2, 158.1, 155.3, 152.2, 142.1, 136.7, 134.7,
130.1, 129.5, 128.2, 127.8, 127.3, 126.5, 122.8, 121.6,
118.8, 111.4, 61.0, 56.2, 55.4, 51.7, 50.4, 47.9, 47.5, 47.3,
46.9, 45.1, 38.7, 37.1, 32.8, 31.2, 25.93, 25.87, 24.4, 22.9,
21.5, 19.5, 19.3, 19.1, 18.0, 17.5; HRMS .FAB) m/e calcd
for C53H76N13O10 .M1H)1, 1054.5838 found: 1054.5835;
Anal. Calcd for C53H75N13O10: C, 60.38; H, 7.17; N, 17.27.
Found: C, 60.58; H, 7.38; N, 17.13.
Figure 4. Pulse sequences used in this work. .a) PFG COSY with P-type
selection. A ®xed delay Dˆ25 ms was used to allow the spectra to be
recorded with fewer t1 increments. The PFGs were 3 G cm21 in strength
and 1 ms in duration. Connectivities can be quickly determined but
coupling constants cannot be measured. .b) Tr-ROESY pulse sequence.
The 908.^y) purging pulses allow a phase-sensitive two-dimensional
spectrum to be obtained. Throughout the mixing period a train of phase
alternating 1808 pulses spin locks y-magnetization while inhibiting TOCSY
transfer. The RF ®eld strength was 6 kHz during tm. As in conventional
ROESY, weak spurious COSY-type cross peaks can arise from zero-
quantum coherence.
3.1.11. Arti®cial b-sheet 6b. Diamide 11b was converted
to arti®cial b-sheet 6b in 98% yield by treatment with iso-
cyanate 13,15a as described above: IR .CHCl3) 3423, 3304,
1
2249, 1655, 1630 cm21; H NMR .500 MHz, CDCl3) d
11.18 .d, Jˆ6.5 Hz, 1H), 11.02 .d, Jˆ7.8 Hz, 1H), 10.41
.s, 1H), 8.78 .d, Jˆ9.3 Hz, 1H), 8.72 .s, 1H), 8.36 .dd,
Jˆ9.0, 2.8 Hz, 1H), 8.18 .d, Jˆ8.9 Hz, 1H), 7.41 .appar t,
Jˆ7.5 Hz, 2H), 7.35 .appar t, Jˆ7.4 Hz, 1H), 7.32 .br q,
Jˆ4.4 Hz, 1H), 7.23±7.17 .m, 3H), 7.12 .appar dd, Jˆ7.4,
1.7 Hz, 2H), 7.07 .appar d, Jˆ7.5 Hz, 2H), 6.94 .d, Jˆ
9.0 Hz, 1H), 6.91 .d, Jˆ9.1 Hz, 1H), 5.94 .br s, 1H), 4.95
.td, Jˆ8.4, 6.4 Hz, 1H), 4.75 .d, Jˆ8.3 Hz, 1H), 4.67 .dq,
Jˆ9.3, 7.1 Hz, 1H), 4.49 .t, Jˆ9.0 Hz, 1H), 4.37 .td, Jˆ9.1,
5.8 Hz, 1H), 4.02 .s, 3H), 3.81 .dt, Jˆ14.0, 4.6 Hz, 1H),
3.65±3.60 .m, 2H), 3.58±3.52 .m, 3H), 3.47±3.43 .m,
1H), 3.42±3.37 .m, 2H), 3.26 .appar dt, Jˆ15.2, 7.6 Hz,
1H), 2.99 .dd ABX pattern, JABˆ13.5 Hz, JAXˆ9.0 Hz,
1H), 2.96±2.89 .m, 2H), 2.79 .d, Jˆ4.6 Hz, 3H), 2.71 .dt,
Jˆ17.0, 6.2 Hz, 1H), 2.67 .dt, Jˆ17.0, 6.2 Hz, 1H), 2.59 .d,
Jˆ4.7 Hz, 3H), 2.29±2.20 .m, 1H), 2.11±2.02 .m, 1H),
1.73±1.66 .m, 2H), 1.59 .ddd, Jˆ13.8, 9.4, 5.8 Hz, 1H),
1.54±1.44 .m, 1H), 1.26 .d, Jˆ6.8 Hz, 3H), 1.21 .d, Jˆ
6.9 Hz, 6H), 1.09 .d, Jˆ6.6 Hz, 3H), 0.97 .d, Jˆ6.7 Hz,
3H), 0.88 .d, Jˆ6.7 Hz, 3H), 0.84 .d, Jˆ6.6 Hz, 3H); 13C
NMR .125 MHz, CDCl3) d 173.2, 172.50, 172.47, 172.1,
171.3, 159.1, 158.8, 156.8, 155.5, 152.1, 141.1, 136.8,
134.8, 130.2, 129.1, 128.11, 128.05, 127.7, 126.3, 122.8,
121.7, 119.0, 118.5, 111.3, 61.3, 56.1, 55.6, 50.9, 48.5,
48.2, 47.7, 47.5, 46.5, 45.3, 39.8, 38.2, 32.4, 31.3, 28.8,
25.9, 25.8, 24.7, 22.7, 22.0, 19.5, 19.4, 19.3, 19.2, 17.9,
17.6; HRMS .FAB) m/e calcd for C54H78N13O10 .M1H)1,
1068.5994 found: 1068.5986; Anal. Calcd for
C54H77N13O10: C, 60.71; H, 7.27; N, 17.04. Found: C,
60.65; H, 6.97; N, 16.91.
4.31 .quintet, Jˆ7.1 Hz, 1H), 4.16 .t, Jˆ8.5 Hz, 1H), 3.94
.s, 3H), 3.83 .dt, Jˆ14.2, 6.2 Hz, 1H), 3.58 .t, Jˆ6.3 Hz,
2H), 3.55±3.51 .m, 2H), 3.46±3.38 .m, 3H), 3.33 .dt, Jˆ
15.4, 6.8 Hz, 1H), 3.24 .dt, Jˆ15.3, 6.7 Hz, 1H), 3.01 .d,
Jˆ4.8 Hz, 3H), 3.01±2.97 .m, 1H), 2.86 .dd ABX pattern,
JABˆ13.8 Hz, JAXˆ7.3 Hz, 1H), 2.69 .t, Jˆ6.3 Hz, 2H),
2.67 .d, Jˆ4.8 Hz, 6H), 2.22±2.14 .m, 1H), 1.87±1.78
.m, 1H), 1.73±1.65 .m, 2H), 1.55±1.46 .m, 2H), 1.22 .d,
Jˆ7.1 Hz, 3H), 1.05 .d, Jˆ6.6 Hz, 3H), 1.04 .d, Jˆ6.5 Hz,
3H), 0.99 .d, Jˆ6.4 Hz, 3H), 0.97 .d, Jˆ6.4 Hz, 3H); 13C
NMR .125 MHz, CDCl3) d 173.4, 172.8, 172.3, 171.6,
166.0, 158.8, 156.5, 155.4, 152.3, 140.2, 136.3, 134.5,
130.2, 129.0, 128.3, 128.1, 126.6, 122.4, 121.9, 120.3,
118.9, 111.6, 62.4, 56.0, 55.5, 51.4, 49.4, 48.4, 48.0, 46.5,
45.6, 45.0, 40.4, 37.9, 29.7, 28.0, 26.4, 26.0, 25.7, 24.6,
22.8, 21.6, 19.6, 19.4, 17.5, 16.6; HRMS .FAB) m/e calcd
for C51H73N12O9 .M1H)1, 997.5623 found: 997.5632;
Anal. Calcd for C51H72N12O9: C, 61.43; H, 7.28; N, 16.85.
Found: C, 61.66; H, 7.47; N, 16.72.
3.1.10. Arti®cial b-sheet 6a. Diamide 11a was converted to
arti®cial b-sheet 6a in 88% yield by treatment with iso-
cyanate 13,15a as described earlier: IR .CHCl3) 3423,
;
3304, 2249, 1655, 1630 cm21 1H NMR .500 MHz,
CDCl3) d 11.03 .br d, Jˆ5.3 Hz, 1H), 10.86 .d, Jˆ7.1 Hz,
1H), 10.01 .s, 1H), 8.58 .d, Jˆ2.6 Hz, 1H), 8.52 .d, Jˆ
9.2 Hz, 1H), 8.34 .dd, Jˆ9.0, 2.8 Hz, 1H), 8.09 .d, Jˆ
8.6 Hz, 1H), 7.85 .br q, Jˆ4.3 Hz, 1H), 7.30 .appar t, Jˆ
7.6 Hz, 2H), 7.25 .appar t, Jˆ7.4 Hz, 1H), 7.20±7.16 .m,
3H), 7.09 .d, Jˆ7.5 Hz, 2H), 7.05±7.04 .m, 2H), 6.90 .d,
Jˆ9.1 Hz, 1H), 6.72 .br q, Jˆ3.4 Hz, 1H), 6.25 .d, Jˆ
8.5 Hz, 1H), 4.78 .ddd, Jˆ8.4, 6.9, 5.5 Hz, 1H), 4.68 .dq,
Jˆ8.1, 6.9 Hz, 1H), 4.42 .d, Jˆ8.7 Hz, 1H), 4.32±4.27 .m,
2H), 4.18 .ddd, Jˆ15.5, 11.1, 4.9 Hz, 1H), 4.01 .s, 3H), 3.98
.dd, Jˆ14.3, 4.2 Hz, 1H), 3.46 .br t, Jˆ6.0 Hz, 2H), 3.40
.ddd, Jˆ14.3, 11.2, 5.3 Hz, 1H), 3.26 .br s, 2H), 3.16 .dd
ABX pattern, JABˆ13.4 Hz, JAXˆ5.1 Hz, 1H), 3.15 .br s,
2H), 3.09 .dd, Jˆ15.6, 4.5 Hz, 1H), 3.03 .dd ABX pattern,
JABˆ13.5 Hz, JBXˆ7.0 Hz, 1H), 2.88 .septet, Jˆ6.9 Hz,
1H), 2.81 .d, Jˆ4.6 Hz, 3H), 2.70 .d, Jˆ4.8 Hz, 3H), 2.63
1
3.2. Two-dimensional H NMR spectroscopic studies of
arti®cial b-sheets 6a and 6b
Studies were performed at 500 MHz using a 25 mM sample
in CDCl3 that was degassed by ®ve freeze-pump-thaw
cycles on a high-vacuum line .,0.001 mmHg) and sealed
1
under vacuum. Two-dimensional H NMR spectra were
recorded at 298 K, as were one-dimensional reference
spectra. Magnitude-mode COSY spectra were obtained
using P-type selection with bipolar pulsed ®eld gradients
.PFGs) bracketing the mixing pulse, as shown in Fig. 4a.28
A ®xed delay D in both t1 and t2 served to improve

J. S. Nowick et al. / Tetrahedron 58 )2002) 727±739
Table 2. H NMR spectral data for arti®cial b-sheets 6
737
1
Resonance
Arti®cial b-sheet 6a
Arti®cial b-sheet 6b
1
2
3
4
5
6
7
8
0.72 .d, Jˆ6.6 Hz, 3H)
0.84 .d, Jˆ6.6 Hz, 3H)
0.78 .d, Jˆ6.7 Hz, 3H)
0.88 .d, Jˆ6.7 Hz, 3H)
0.95 .d, Jˆ6.7 Hz, 3H)
0.97 .d, Jˆ6.7 Hz, 3H)
1.03 .d, Jˆ6.7 Hz, 3H)
1.09 .d, Jˆ6.6 Hz, 3H)
1.12 .d, Jˆ6.8 Hz, 3H)
1.21 .d, Jˆ6.9 Hz, 6H)
1.23±1.17 .m, 1H)
1.24 .d, Jˆ6.9 Hz, 3H)
1.26 .d, Jˆ6.8 Hz, 3H)
1.54±1.44 .m, 1H)
1.59 .ddd, Jˆ13.8, 9.4, 5.8 Hz, 1H)
1.27 .d, Jˆ6.9 Hz, 3H)
9
1.65 .ddd, Jˆ13.9, 9.5, 5.7 Hz, 1H)
1.74 .ddd, Jˆ13.9, 8.5, 5.7 Hz, 1H)
1.99 .octet, Jˆ6.9 Hz, 1H)
2.54 .dt, Jˆ17.0, 6.1 Hz, 1H)
2.63 .dt, Jˆ17.0, 6.3 Hz, 1H)
2.70 .d, Jˆ4.8 Hz, 3H)
1.73±1.66 .m, 2H)
2.11±2.02 .m, 1H)
2.29±2.20 .m, 1H)
10
11
12
13
14
15
16
17
18
19
190
20
21
22
23
24
25
26
260
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
2.59 .d, Jˆ4.7 Hz, 3H)
2.67 .dt, Jˆ17.0, 6.2 Hz, 1H)
2.71 .dt, Jˆ17.0, 6.2 Hz, 1H)
2.79 .d, Jˆ4.6 Hz, 3H)
2.81 .d, Jˆ4.6 Hz, 3H)
2.88 .septet, Jˆ6.9 Hz, 1H)
3.03 .dd ABX pattern, JABˆ13.5 Hz, JAXˆ7.0 Hz, 1H)
3.09 .dd, Jˆ15.6, 4.5 Hz, 1H)
3.15 .br s, 2H)
2.96±2.89 .m, 2H)
2.99 .dd ABX pattern, JABˆ13.5 Hz, JAXˆ9.0 Hz, 1H)
3.26 .appar dt, Jˆ15.2, 7.6 Hz, 1H)
3.42±3.37 .m, 2H)
3.16 .dd ABX pattern, JABˆ13.4 Hz, JBXˆ5.1 Hz, 1H)
3.26 .br s, 2H)
3.47±3.43 .m, 1H)
3.58±3.52 .m, 3H)
3.40 .ddd, Jˆ14.3, 11.2, 5.3 Hz, 1H)
3.46 .br t, Jˆ6.0 Hz, 2H)
3.98 .dd, Jˆ14.3, 4.2 Hz, 1H)
4.01 .s, 3H)
3.65±3.60 .m, 2H)
3.81 .dt, Jˆ14.0, 4.6 Hz, 1H)
4.02 .s, 3H)
4.37 .td, Jˆ9.1, 5.8 Hz, 1H)
4.49 .t, Jˆ9.0 Hz, 1H)
4.18 .ddd, Jˆ15.5, 11.1, 4.9 Hz, 1H)
4.29 .appar td, Jˆ9.0, 5.7 Hz, 1H)
4.31 .appar t, Jˆ8.1 Hz, 1H)
4.42 .d, Jˆ8.7 Hz, 1H)
4.67 .dq, Jˆ9.3, 7.l Hz, 1H)
4.75 .d, Jˆ8.3 Hz, 1H)
4.95 .td, Jˆ8.4, 6.4 Hz, 1H)
5.94 .br s, 1H)
6.91 .d, Jˆ9.1 Hz, 1H)
6.94 .d, Jˆ9.0 Hz, 1H)
7.07 .appar d, Jˆ7.5 Hz, 2H)
7.12 .appar dd, Jˆ7.4, 1.7 Hz, 2H)
7.23±7.17 .m, 3H)
7.32 .br q, Jˆ4.4 Hz, 1H)
7.35 .appar t, Jˆ7.4 Hz, 1H)
7.41 .appar t, Jˆ7.5 Hz, 2H)
8.18 .d, Jˆ8.9 Hz, 1H)
8.36 .dd, Jˆ9.0, 2.8 Hz, 1H)
8.72 .s, 1H)
4.68 .dq, Jˆ8.1, 6.9 Hz, 1H)
4.78 .ddd, Jˆ8.4, 6.9, 5.5 Hz, 1H)
6.25 .d, Jˆ8.5 Hz, 1H)
6.72 .br q, Jˆ3.4 Hz, 1H)
6.90 .d, Jˆ9.1 Hz, 1H)
7.05±7.04 .m, 2H)
7.09 .d, Jˆ7.5 Hz, 2H)
7.20±7.16 .m, 3H)
7.25 .appar t, Jˆ7.4 Hz, 1H)
7.30 .appar t, Jˆ7.6 Hz, 2H)
7.85 .br q, Jˆ4.3 Hz, 1H)
8.09 .d, Jˆ8.6 Hz, 1H)
8.34 .dd, Jˆ9.0, 2.8 Hz, 1H)
8.52 .d, Jˆ9.2 Hz, 1H)
8.58 .d, Jˆ2.6 Hz, 1H)
8.78 .d, Jˆ9.3 Hz, 1H)
10.01 .s, 1H)
10.86 .d, Jˆ7.1 Hz, 1H)
10.41 .s, 1H)
11.02 .d, Jˆ7.8 Hz, 1H)
11.18 .d, Jˆ6.5 Hz, 1H)
11.03 .br d, Jˆ5.3 Hz, 1H)
Spectra were recorded at 298 K in 25 mM CDC13 solution.
the ef®ciency of the data collection by allowing the purely
antiphase cross peaks to come partially into phase, so that
the cross peak multiplet has a non-zero integral.29 In this
way, fewer t1 increments need be recorded before the cross
peaks become discernible. The diagonal peaks gain some
antiphase character by the same token, making them some-
what less intrusive, and allowing a milder ®lter function to
be used. The bipolar gradients effectively cancel eddy-
current effects, and avoid disturbing the ®eld-frequency
lock, making them ideal for very fast repetition rates.30
zero-quantum coherence in the rotating frame. As key
contacts arise from spins with no J coupling, the small
COSY-type peaks present little practical problem. Tables
2 and 3 summarize the one-dimensional reference spectra
and Tr-ROESY NMR data from these studies.
Acknowledgements
This work was supported by the National Institutes of
Health Grant GM-49076 and National Science Foundation
Grant CHE-9813105. E. M. S. thanks the National Institute
on Aging for support in the form of a training grant
.National Research Service Award AG00096). J. S. N.
thanks the following agencies for support in the form of
awards: the Camille and Henry Dreyfus Foundation
.Teacher±Scholar Award), the American Chemical Society
The Tr-ROESY sequence .Fig. 4b) was used to obtain
information on spatial proximity as measured by cross
relaxation.21 This sequence gives phase-sensitive two-
dimensional spectra by the usual method of States et al.31
A mixing time tmˆ300 ms gave adequate cross peak
intensity and only small COSY-type peaks arising from

738
J. S. Nowick et al. / Tetrahedron 58 )2002) 727±739
Table 3. 1H NMR Tr-ROESY crosspeaks for arti®cial b-sheets 6
Resonance
Arti®cial b-sheet 6a
6 .s), 9 .m), 10 .s), 26 .s)
6 .s), 9 .s), 10 .s), 26 .m), 36 .w), 37 .m)
Arti®cial b-sheet 6b
7 .s), 8 .s), 9 .s), 25 .s), 33 .w)
7 .s), 8 .s), 9 .s), 25 .m), 33 .w)
11 .s), 26 .s), 32 .w), 34 .m), 35 .m), 42 .m)
11 .s), 26 .s), 32 .s), 34 .m), 35 .m)
15 .m), 16 .s), 27 .s), 36 .w), 41 .m), 42 .s), 45 .w)
15 .m), 16 .s)
1
2
3
4
5
6
7
11 .s), 260 .s), 29 .w), 30 .m), 41 .w)
11 .s), 260 .s), 29 .m), 30 .s), 41 .w)
28 .s), 34 .w), 37 .w), 41 .m), 42 .m)
26 .w), 27 .w), 39 .w)
15 .m), 16 .s)
1 .s), 2 .s), 25 .m), 39 .m)
8
9
90
15 .m), 16 .s)
1 .m), 2 .s), 5 .w), 6 .w), 10 .m), 26 .m), 28 .m), 39 .m)
N/a
1 .s), 2 .s), 5 .w), 9 .m), 26 .m), 28 .m), 39 .w)
3 .s), 4 .s), 260 .s), 29 .w), 30 .s)
13 .s), 22 .s)
12 .s), 22 .s)
31.s), 38 .w)
7 .m), 8 .m), 38 .s)
7 .s), 8 .s), 45 .s)
N/a
190 .s), 27 .m), 29 .s), 33 .s)
23 .m), 25 .s)
43 .m)
17 .s), 27 .m), 29 .s), 33 .s)
43 .m)
18 .m), 23 .s), 34 .m)
N/a
N/a
12 .s), 13 .s)
N/a
18 .m), 21.s)
32 .s), 44 .m)
18 .s), 30 .s)
1 .s), 2 .m), 6 .m), 9 .m), 10 .s), 31 .m), 39 .m)
3 .s), 4 .s), 11 .s), 30 .m), 41 .s), 42 .s)
17 .m), 190 .m), 33 .m), 34 .s), 39 .m)
3 .w), 5 .s), 9 .s), 10 .m), 26 .w), 38 .s), 39 .m), 41 .m)
3 .m), 4 .m), 11 .w), 17 .s), 190 .s), 27 .s), 30 .w), 33 .m), 39 .s)
3 .m), 4 .s), 11 .m), 25 .s), 260 .m), 34 .w)
14 .s), 26 .m)
1 .s), 2 .s), 9 .s), 25 .m), 27 .m), 39 .m)
1 .s), 2 .s), 8 .s), 25 .m), 27 .m), 39 .w)
10 .s), 19 .m), 2100 .m), 23 .m), 32 .m), 33 .w)
90 .s), 18 .m), 19 .m), 23 .m), 32 .w)
3 .s), 4 .s), 26 .m), 29 .m), 32 .s), 34 .m)
25 .w), 30 .s), 34 .w), 35 .m), 36 .m)
210 .s), 22 .s)
10
11
12
13
14
15
16
160
17
18
19
190
20
21
210
2100
22
220
23
24
25
26
260
27
28
29
30
31
32
33
210 .s), 22 .s)
5 .s), 6 .m), 35 .w), 36 .s)
5 .s), 6 .s), 45 .s)
28 .s), 29 .s), 34 .s), 35 .w)
28 .s), 29 .s), 34 .s), 35 .w)
10 .m), 2100 .s), 23 .m), 32 .s)
90 .m), 10 .m), 23 .s), 33 .m)
43 .m)
43 .m)
220 .s), 43 .m)
13 .s), 14 .s)
90 .m), 18 .s), 32 .s)
13 .s), 14 .s)
21 .s), 43 .s)
90 .m), 10 .m), 18 .m), 19 .s), 33 .m)
31 .s), 44 .s)
1 .s), 2 .s), 7 .m), 8 .m), 9 .m), 30 .s), 39 .s)
3 .s), 4 .s), 11 .m), 32 .m), 41 .s), 42 .s)
N/a
5 .s), 8 .m), 9 .m), 36 .s), 39 .m), 42 .m)
160 .s), 17 .s), 29 .s), 33 .s), 34 .m), 39 .w)
11 .m), 160 .s), 17 .s), 28 .s), 32 .w), 34 .s), 39 .s)
12 .s), 25 .s), 39 .m)
12 .w), 24 .s), 40 .s)
24 .s), 40 .s)
1 .w), 3 .w), 4 .w), 17 .s), 190 .s), 27 .s), 29 .s)
3 .w), 4 .s), 90 .m), 10 .w), 11 .s), 18 .s), 2100 .s), 26 .m), 29 .w)
1 .s), 2 .m), 7 .w), 90 .m), 10 .m), 18 .w), 19 .s), 23 .m), 28 .s),
29 .m), 37 .s), 38 .s)
34
35
36
37
38
39
40
41
42
43
44
45
2 .w), 5 .w), 21 .s), 27 .s), 30 .s), 39 .m)
1 .m)
2 .w)
2 .m), 5 .w)
14 .w), 15 .s), 28 .s)
6 .w), 9 .m), 10 .w), 26 .m), 27 .m), 28 .w), 29 .m), 34 .w)
32 .s)
3 .w), 5 .m), 260 .s), 28 .m)
5 .m), 260 .s), 43 .s)
19 .m), 20 .m), 42 .s)
24 .m)
3 .s), 4 .m), 11 .m), 12 .w), 160 .s), 17 .s), 28 .m), 29 .s), 35 .s)
3 .m), 12 .m), 15 .m), 160 .m), 17 .m), 29 .m), 30 .w), 34 .s)
5 .w), 12 .m), 15 .s), 27 .s)
29 .w), 33 .s)
1 .w), 2 .w), 29 .m), 33 .s)
7 .w), 8 .m), 9 .m), 25 .m), 27 .m), 28 .m), 29 .s), 30 .w)
31.s), 43 .w)
5 .w), 26 .s), 43 .s)
3 .m), 5 .s), 26 .s), 27 .m), 45 .w)
190 .m), 20 .m), 21 .m), 220 .s), 41 .s)
5 .w), 24 .m)
5 .w), 16 .m)
5 .w), 16 .m), 42 .w)
Spectra were recorded at 298 K in 25 mM CDC13 solution. Crosspeaks in the f1 dimension are tabulated for each resonance in the f2 dimension.
2. .a) Wagner, G.; Feigel, M. Tetrahedron 1993, 49, 10831±
10842. .b) Brandmeier, V.; Sauer, W. H. B.; Feigel, M.
Helv. Chim. Acta 1994, 77, 70±85.
.Arthur C. Cope Scholar Award), and the Alfred P. Sloan
Foundation .Alfred P. Sloan Research Fellowship).
3. .a) Kemp, D. S.; Bowen, B. R. Tetrahedron Lett. 1988, 29,
5077±5080. .b) Kemp, D. S.; Bowen, B. R. Tetrahedron Lett.
1988, 29, 5081±5082. .c) Kemp, D. S.; Bowen, B. R.;
Muendel, C. C. J. Org. Chem. 1990, 55, 4650±4657.
.d) Kemp, D. S.; Li, Z. Q. Tetrahedron Lett. 1995, 36,
4175±4178.
References
1. .a) Mutter, M.; Altmann, E.; Altmann, K.-H.; Hersperger, R.;
Koziej, P.; Nebel, K.; Tuchscherer, G.; Vuilleumier, S.;
Gremlich, H.-U.; MuÈller, K. Helv. Chem. Acta 1988, 71,
835±847. .b) Mutter, M.; Vuilleumier, S. Angew. Chem.,
4. .a) Diaz, H.; Tsang, K. Y.; Choo, D.; Espina, J. R.; Kelly, J. W.
J. Am. Chem. Soc. 1993, 115, 3790±3791. .b) Tsang, K. Y.;
Diaz, H.; Graciani, N.; Kelly, J. W. J. Am. Chem. Soc.
1994, 116, 3988±4005. .c) Nesloney, C. L.; Kelly, J. W.
È
Int. Ed. Engl. 1989, 28, 535±554. .c) Tuchscherer, G.; Dorner,
B.; Sila, U.; Kabmer, B.; Mutter, M. Tetrahedron 1993, 49,
3559±3575.

J. S. Nowick et al. / Tetrahedron 58 )2002) 727±739
739
J. Am. Chem. Soc. 1996, 118, 5836±5845. .d) Bekele, H.;
Fendler, J. H.; Kelly, J. W. J. Am. Chem. Soc. 1999, 121,
7266±7267. .e) Lashuel, H. A.; LaBrenz, S. R.; Woo, L.;
Serpell, L. C.; Kelly, J. W. J. Am. Chem. Soc. 2000, 122,
5262±5277.
15. For preliminary accounts of these studies, see: .a) Holmes,
D. L.; Smith, E. M.; Nowick, J. S. J. Am. Chem. Soc. 1997,
119, 7665±7669. .b) Nowick, J. S. Acc. Chem. Res. 1999, 32,
287±296.
16. Nowick, J. S.; Powell, N. A.; Martinez, E. J.; Smith, E. M.;
Noronha, G. J. Org. Chem. 1992, 57, 3763±3765.
17. 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±99.
5. For a review, see: Gellman, S. H. Acc. Chem. Res. 1998, 31,
173±180.
È
6. .a) Krauthauser, S.; Christianson, L. A.; Powell, D. R.;
Gellman, S. H. J. Am. Chem. Soc. 1997, 119, 11719±11720.
.b) Chung, Y. J.; Christianson, L. A.; Stanger, H. E.; Powell,
D. R.; Gellman, S. H. J. Am. Chem. Soc. 1998, 120, 10555±
10556. .c) Chung, Y. J.; Huck, B. R.; Christianson, L. A.;
18. We have, unfortunately, never been able to crystallize any of
our arti®cial b-sheets.
19. We have also synthesized arti®cial b-sheet 6a using solid-
phase methods. For a description, see Ref. 15a.
È
Stanger, H. E.; Krauthauser, S.; Powell, D. R.; Gellman,
20. .a) Nowick, J. S.; Holmes, D. L.; Noronha, G.; Smith, E. M.;
Nguyen, T. M.; Huang, S.-L. J. Org. Chem. 1996, 61, 3929±
3934. .b) Nowick, J. S.; Holmes, D. L.; Noronha, G.; Smith,
E. M.; Nguyen, T. M.; Huang, S.-L.; Wang, E. H. J. Org.
Chem. 1998, 63, 9144.
S. H. J. Am. Chem. Soc. 2000, 122, 3995±4004. .d) Fisk,
J. D.; Powell, D. R.; Gellman, S. H. J. Am. Chem. Soc.
2000, 122, 5443±5447. .e) Fisk, J. D.; Gellman, S. H. J. Am.
Chem. Soc. 2001, 123, 343±344.
7. For a review, see: Seebach, D.; Matthews, J. L. Chem.
Commun. 1997, 2015±2022.
21. .a) Hwang, T. L.; Shaka, A. J. J. Am. Chem. Soc. 1992, 114,
3157±3159. .b) Hwang, T. L.; Shaka, A. J. J. Magn. Reson.,
Ser. B 1993, 102, 155±165.
8. .a) Seebach, D.; Abele, S.; Gademann, K.; Jaun, B. Angew.
Chem., Int. Ed. Engl. 1999, 38, 1595±1597. .b) Daura, X.;
È
Gademann, K.; Schafer, H.; Jaun, B.; Seebach, D.; van
22. The older version of the AMBERp force ®eld associated with
this version of MacroModel appears to better reproduce the
preference of the diacylhydrazine groups of 6a and 6b to adopt
a linear conformation, which is evidenced by the large 3JHNNH
Gunsteren, W. F. J. Am. Chem. Soc. 2001, 123, 2393±2404.
9. For reviews, see: .a) Smith, C. K.; Regan, L. Acc. Chem. Res.
1997, 30, 153±161. .b) Gellman, S. H. Curr. Opin. Chem.
Biol. 1998, 2, 717±725. .c) Lacroix, E.; Kortemme, T.; de la
Paz, M. L.; Serrano, L. Curr. Opin. Struct. Biol. 1999, 9, 487±
493.
1
coupling constants in their H NMR spectra.
23. WuÈthrich, K. NMR of Proteins andNucleic Acids ; Wiley: New
York, 1986 pp 125±129.
24. WuÈthrich, K. NMR of Proteins andNucleic Acids ; Wiley: New
York, 1986 pp 166±168.
10. .a) Sharman, G. J.; Searle, M. S. Chem. Commun. 1997, 1955±
1956. .b) Doig, A. J. Chem. Commun. 1997, 2153±2154.
.c) Schenck, H. L.; Gellman, S. H. J. Am. Chem. Soc. 1998,
120, 4869±4870. .d) Das, C.; Raghothama, S.; Balaram, P.
J. Am. Chem. Soc. 1998, 120, 5812±5813. .e) Kortemme, T.;
25. This concentration .1.0 mM) provides adequate signal-to-
noise ratio without signi®cant self-association. Studies of the
concentration dependence of the chemical shifts of the NH
groups of 6a from 0.5±200 mM and 6b from 0.3±89 mM in
CDCl3 reveal negligible self-association .Kassoc#1 M21) at the
concentrations used for both the chemical shift and the NOE
studies in this paper.
Â
Ramõrez-Alvarado, M.; Serrano, L. Science 1998, 281, 253±
256. .f) Das, C.; Raghothama, S.; Balaram, P. Chem.
Commun. 1999, 967±968. .g) de Alba, E.; Santoro, J.; Rico,
M.; Jimenez, M. A. Protein Sci. 1999, 8, 854±865. .h) Das, C.;
Nayak, V.; Raghothama, S.; Balaram, P. J. Peptide Res. 2000,
56, 307±317. .i) Grif®ths-Jones, S. R.; Searle, M. S. J. Am.
Chem. Soc. 2000, 122, 8350±8356.
26. This comparison should be interpreted with caution, because
the shift differences are small .several tenths of a ppm) and are
only slightly greater than the few tenths ppm down®eld
shifting that occurs to an amide or urea NH group when its
carbonyl group accepts a hydrogen bond .e.g. Hh or Ha in
Chart 3).
11. Nowick, J. S.; Smith, E. M.; Noronha, G. J. Org. Chem. 1995,
60, 7386±7387.
12. .a) Nowick, J. S.; Holmes, D. L.; Mackin, G.; Noronha, G.;
Shaka, A. J.; Smith, E. M. J. Am. Chem. Soc. 1996, 118, 2764±
2765. .b) Nowick, J. S.; Pairish, M.; Lee, I. Q.; Holmes, D. L.;
Ziller, J. W. J. Am. Chem. Soc. 1997, 119, 5413±5424.
.c) Smith, E. M.; Holmes, D. L.; Shaka, A. J.; Nowick, J. S.
J. Org. Chem. 1997, 62, 7906±7907. .d) Tsai, J. H.; Waldman,
A. S.; Nowick, J. S. Bioorg. Med. Chem. 1999, 7, 29±38.
13. Nowick, J. S.; Mahrus, S.; Smith, E. M.; Ziller, J. W. J. Am.
Chem. Soc. 1996, 118, 1066±1072.
27. Dueholm, K. L.; Egholm, M.; Buchardt, O. Org. Prep. Proc.
Int. 1993, 25, 457±461.
28. Hurd, R. E. J. Magn. Reson. 1990, 87, 422±428.
29. Bax, A.; Freeman, R. J. Magn. Reson. 1981, 44, 542±561.
30. Wider, G.; WuÈthrich, K. J. Magn. Reson., Ser. A 1994, 108,
255±258.
31. States, D. J.; Haberkorn, R. A.; Ruben, D. J. J. Magn. Reson.
1982, 48, 286±292.
14. Nowick, J. S.; Cary, J. M.; Tsai, J. H. J. Am. Chem. Soc. 2001,
123, 5176±5180.