Poly-L-proline type II peptide mimics based on the 3-azabicyclo[3.1.0]hexane system

Article abstract of DOI:10.1021/jo001201o

This paper describes conformational studies of proline-templated amino acids (PTAAs) based on the 3-azabicyclo[3.1.0]hexane system as well as conformational studies on short peptides composed of these PTAAs. NOE data, coupling constants, and molecular modeling are consistent with a flattened boat conformation for monomeric and oligomeric residues based on this bicyclic system. NMR studies on dimeric and trimeric oligomers are consistent with a populated poly-L-proline type II conformation in CDCl₃ and D₂O. Solution studies and molecular modeling predicts φ ～ -70°, ψ ～ 131°, χ1 ～ -57°, and χ2 ～ -158° for oligomeric residues.

Full text of DOI:10.1021/jo001201o

J . Org. Chem. 2001, 66, 455-460
455
P oly-L-p r olin e Typ e II P ep tid e Mim ics Ba sed on th e
3-Aza bicyclo[3.1.0]h exa n e System
Ahmed Mamai, Rui Zhang, Amarnath Natarajan, and J ose S. Madalengoitia*
Department of Chemistry, University of Vermont, Burlington, Vermont 05405
jmadalen@zoo.uvm.edu
Received August 7, 2000
This paper describes conformational studies of proline-templated amino acids (PTAAs) based on
the 3-azabicyclo[3.1.0]hexane system as well as conformational studies on short peptides composed
of these PTAAs. NOE data, coupling constants, and molecular modeling are consistent with a
flattened boat conformation for monomeric and oligomeric residues based on this bicyclic system.
NMR studies on dimeric and trimeric oligomers are consistent with a populated poly-L-proline type
II conformation in CDCl3 and D2O. Solution studies and molecular modeling predicts φ ∼ -70°, ψ
∼ 131°, ø1 ∼ -57°, and ø2 ∼ -158° for oligomeric residues.
In tr od u ction
Over the past decade, poly-L-proline type II (PPII)
secondary structure has been shown to play a critical role
in mediating several cellular signaling pathways.1,2 Due
to the potential utility of agents capable of inhibiting
these signaling mechanisms, we are developing a pro-
gram focused on the design and synthesis of mimics of
rotation about the RC - âC bond, ø2 defining rotation
about the âC - γC bond, etc. Examination of a number
the PPII conformation. PPII helices are characterized by
an extended left-handed fold with φ ∼ -75°, ψ ∼ 145°,
and ω ∼ 180°. Since many PPII helices possess amino
acids other than proline, PPII mimics must not only
adopt the desired backbone conformation, but also be able
to account for the nonprolyl side chains which are often
critical for receptor recognition of the PPII helix. Our
strategy for mimicking this secondary structure involves
first the synthesis of proline-templated amino acids
(PTAAs) and then the synthesis of oligopeptides from
these PTAAs.3 OligoPTAAs are thus expected to prefer-
entially populate the PPII conformation in solution since
(1) φ ∼ -75° due to the constraints of the pyrrolidine ring,
(2) ψ ∼ 145° due to pseudo A(1,3)-strain, and (3) ω ∼ 180°
because the amide trans conformation is favored. One of
the advantages of the PTAA approach lies in that not
only the peptide backbone, but also the amino acid side
chains can be constrained. Side chain conformation is
described by the parameters ø1, ø2, ø3... with ø1 defining
of receptor-bound PPII helices reveals that residues
possessing ø1 ∼ gauche(-) and ø2 ∼ trans are common,
thus necessitating a strategy to mimic this geometry.
With this goal in mind, we envisioned that a PTAA based
on the 3-azabicyclo[3.1.0]hexane system would approxi-
mate a ø1 angle ∼ gauche(-) (compare Newman projec-
tions 3 and 4, Figure 1).4 Furthermore, due to the greater
p-character in the cyclopropane C-C bonds, we expected
that ø2 would approximate the desired trans geometry.
This paper describes the synthesis and conformational
evaluation of PPII mimics based on the 3-azabicyclo-
[3.1.0]hexane system in CHCl3 and D2O.5-7
* To whom correspondence should be addressed.
(1) ) For a PPII review, see: Siligardi, G.; Drake, A. F. Pept. Sci.
1995, 37, 281.
(2) For examples of proteins which bind PPII helices, see: (a) Raj,
P. A.; Marcus, E.; Edgerton, M. Biochemistry 1996, 35, 4314. (b) Lee,
C.-H.; Saksela, K.; Mirza, U. A.; Chait, B. T.; Kuriyan, J . Cell 1996,
85, 931. (c) Peng, S.; Kasahara, C.; Rickles, R. J .; Schreiber, S. L. Proc.
Natl. Acad. Sci. U.S.A. 1995, 92, 12408. (d) J ardetzky, T. S.; Brown,
J . H.; Gorga, J . C.; Stern, L. J .; Urban, R. G.; Strominger, J . L.; Wiley:
D. C. Proc. Natl. Acad. Sci. U.S.A. 1996, 93, 734. (e) Zeile, W. L.;
Purich, D. L.; Southwick, F. S. J . Cell Biol. 1996, 133, 49. (f) Mahoney,
N. M.; J anmey, P. A.; Almo, S. C. Nat. Struct. Biol. 1997, 4, 953. (g)
Zhu, X.; Lamango, N. S.; Lindberg, I. J . Biol. Chem. 1996, 271, 23582.
(h) Viguera AR, Arrondo J . L., Musacchio, A.; Saraste, M.; Serrano, L.
Biochemistry 1994, 33, 10925. (i) Feng, S.; Chen, J . K. Yu, H.; Simon,
J . A.; Schreiber, S. L. Science 1994, 266, 1241. (j) Yu, H.; Chen, J . K.;
Feng, S.; Dalgarno, D. C.; Brauer, A. W.; Schreiber, S. L. Cell 1994,
76, 933.
(4) In discussion of the PTAAs, the numbering of the bicyclo[3.1.0]-
hexane system is used.
(5) A preliminary account of this work has appeared: Zhang, R.;
Madalengoitia, J . S. J . Org. Chem. 1999, 64, 330.
(6) For cyclopropane ring constraints of amino acids, see: (a) 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. (b) Burgess, K.; Ho, K.-K.; Pettitt,
B. M. J . Am. Chem. Soc. 1994, 116, 799.
(3) (a) Zhang, R.; Brownewell, F.; Madalengoitia, J . S. J . Am. Chem.
Soc. 1998, 120, 3894. (b) Zhang, R.; Madalengoitia, J . S. Tetrahedron
Lett. 1996, 37, 6235.
10.1021/jo001201o CCC: $20.00 © 2001 American Chemical Society
Published on Web 12/29/2000

456 J . Org. Chem., Vol. 66, No. 2, 2001
Mamai et al.
F igu r e 1. Comparison of side chain orientations.
Sch em e 1
F igu r e 2. Selected NOEs for 24 and 25.
Con for m a tion of Mon om er s. Previous studies have
shown that the bicyclo[3.1.0]hexane system populates a
flattened boat conformation.10,11 Although the pucker is
slight, the degree of the pucker will influence φ, prompt-
ing first an investigation of the bicyclic ring system
conformation. NOE experiments on the model compounds
Ac-PTL-OBn, 17, and Ac-PTE(Me)-OBn, 18 (Figure 2),
were utilized to explore the ring pucker since a populated
boat or chair conformation would result in unequal
distances from the acyl methyl group to 4cH and 4tH
(c and t denote a cis and trans relationship relative to
the ester carbonyl).12 NOESY spectra of compounds 17
and 18 reveal that the acyl methyl group exhibits a
higher NOE intensity with 4tH than 4cH as expected for
a populated boat conformation. This experiment is de-
pendent on a planar amide bond geometry for which
reason, the sterically demanding trimethylacetyl substi-
tuted monomers were not utilized. To investigate if the
magnitude of the pucker is the same in both PTL 17 and
PTE 18, we determined and compared the volumes of the
4tH-CH3 and 4cH-CH3 NOESY cross-peaks. For PTL
17, the ratio of the 4tH-CH3/4cH-CH3 cross-peak vol-
umes is 1.3, while the analogous value for PTE 18 is 2.8,
consistent with a more pronounced pucker for PTE 18.
A more pronounced pucker for PTE 18 is also supported
by coupling constant data (J 4cH-5H ) 5.2 Hz for 17, J 4cH-5H
) 3.8 Hz for 18).
To gain better insights into the exact geometries of the
monomer units, we also investigated Ac-PTL-OMe and
Ac-PTE(Me)-OMe by molecular modeling. Geometry op-
timizations were performed on Ac-PTL-OMe and Ac-PTE-
(Me)-OMe using molecular mechanics (MMX force field)
and ab initio calculations (HF/6-31g* level). Both molec-
ular mechanics and ab initio methods reproduce the
slight pucker (Figure 3). Table 1 compares some torsional
angles for the PTL and PTE structures generated by both
approaches. A cursory examination of these parameters
reveals that in contrast to the NMR data, molecular
modeling predicts essentially identical geometries for the
PTE and PTL bicyclic framework. This discrepancy can
be described by the inherent differences and limitations
Resu lts a n d Discu ssion
Syn th esis. The synthesis of the oligomeric PTAAs was
achieved by a modular assembly described in Scheme 1.8,9
The synthesis of oligomeric leucine PTAAs have already
been reported.5 The dimeric and trimeric proline-tem-
plated glutamic acids (PTE) were synthesized from Boc-
PTE(Me)-OH (8). Protection of the carboxy terminus with
BnOH, diisopropylcarbodiimide (DIC), and DMAP gave
the benzyl ester 9 (82%). Boc-deprotection (TFA) and
reprotection of the resultant amine afforded the trim-
ethylacetamide 10 in 95% yield. Debenzylation of the
ester gave the carboxylic acid 11 which underwent
smooth DIC-HOBt-mediated coupling with amino ester
12 to give the dimer 13 in 77% yield. The trimer 14 was
obtained by debenzylation of 13 followed by DIC-HOBt
coupling with the amine fragment 12 in 91% yield. The
ester functionalities were hydrolyzed with LiOH to afford
the water soluble dimer 15 (93%) and trimer 16 (83%).
(10) (a) Fujimoto, Y.; Irreverre, F.; Karle, J . M.; Karle, I. L.; Witkop,
B. J . Am. Chem. Soc. 1971, 93, 3471. (b) Abraham, R. J .; Gatti, G.
Org. Magn. Reson. 1970, 2, 173.
(11) For some applications of the flattened boat conformation in the
bicyclo[3.1.0]hexane system: (a) Marquez, V. E.; Ezzitouni, A.; Russ,
P.; Siddiqui, M. A.; Ford, H., J r.; Feldman, R. J .; Mitsuya, H.; George,
C.; Barchi, J . J ., J r. J . Am. Chem. Soc. 1998, 120, 2780. (c) Marquez,
V. E.; Siddiqui, M. A.; Ezzitouni, A.; Russ, P.; Wang, J .; Wagner, R.
W.; Matteucci, M. D. J . Med. Chem. 1996, 39, 3739. (c) Rodriguez, J .
B.; Marquez, V. E.; Nicklaus, M. C.; Mitsuya, H.; Barchi, J . J ., J r. J .
Med. Chem. 1994, 37, 3389.
(7) For a review of cyclopropane amino acids see: Stammer, C. H.
Tetrahedron 1990, 46, 2231.
(8) For the synthesis of PTAAs see: Zhang, R.; Mamai, A.; Mad-
alengoitia, J . S. J . Org. Chem. 1999, 64, 547.
(9) For the synthesis of 3,4-methanoproline: Sagnard, I.; Sasaki,
A.; Chiaroni, A.; Riche, C.; Potier, P. Tetrahedron Lett. 1995, 36, 3149.
(12) The monomers 17 and 18 exist as 7: 3 mixture of trans/cis
amide bond rotamers, respectively.

Poly-L-proline Type II Peptide Mimics
J . Org. Chem., Vol. 66, No. 2, 2001 457
F igu r e 3. Edge-on view of Ac-PTL-OMe and Ac-PTE(Me)-
OMe showing the slight puckering of the flattened boat
conformation.
Ta ble 1. Com p a r ison of Som e Ac-P TE(Me)-OMe Wh ich
Wer e Geom etr y Op tim ized (MMX for ce field a n d by a b
in itio ca lcu la tion s)
F igu r e 4. Overlays of Leu/PTL, Glu/PTE, and Arg/PTR. Dark
bonds denote the PTAAs.
φai
φMMX
ø1ai
ø1MMX ø2ai ø2MMX ø2′ai ø2′MMX
PTE2 -71.9 -68.7 -55.4 -56.9 -158 -155
-
-
PTL2 -71.1 -67.7 -57.0 -59.9 -161 -159 -16.9 -14.9
a
Torsional angles are in deg.
of NMR and molecular modeling methods. For example,
since molecular modeling provides single point minimum
energy conformations, it is feasible that the PTL and PTE
frameworks have additional local minima which differ
in the degree of the pucker. It is thus possible that the
geometry optimizations converged on the same conforma-
tion and missed the other minima of minimum. However,
considering the inherent rigidity of the system, it seems
unlikely that there are additional minima that vary in
the degree of the pucker. We believe that a more plausible
explanation reflects the inherent property that NMR data
are time averaged. With this in mind, we are of the
opinion that the discrepancy between the NMR and
molecular modeling results reflects a different time
averaged behavior for PTL and PTE in which the C7-
methyl hinders PTL motion toward a more prominent
boat conformation. Since the distance between the PTL
C7-hydrogen (closest to nitrogen) and N3 is ∼2.5 Å it easy
to extrapolate that motion toward a more prominent
pucker is less favored for PTL than for PTE and that this
different time averaged behavior is reflected in the NMR
data. The minima, however, appear to be nearly identical.
Further examination of the dihedral angles reveals that
the 3-azabicyclo[3.1.0]hexane system serves well to mimic
the desired dihedral angles with φ ∼ -70°, ø1 ∼ -57°
(gauche(-)) and ø2 ∼ -158° (trans). It should be noted
that the greater p-character in the cyclopropane bonds
do help to approximate the “trans” geometry about the
âC-γC bond as evidenced in that a ø2 value ∼ -158°
comes reasonably close to - 180°. We also explored how
the cyclopropane influences the angle ø3 as this would
be critical in PTAAs with longer side chains (for example
PTR 7). For PTR 7 there are three minima found at ø3
values of -37°, 85°, and -152° with relative energies
0.88, 0.0, and 0.0 kcal/mol, respectively. The minima are
at shallow potential energy wells so that deviations of
ø3 ( 20° from the minima result in an increase ∼ 0.4
kcal/mol. As such, the three minima are similar to the
gauche(-), gauche(+), and trans values expected for the
natural amino acid. To evaluate how well our PTAAs
mimic the desired geometry, overlays of Leu/PTL, Glu/
PTE, and Arg/PTR were generated. The overlays (Figure
4) show that indeed the PTAAs fit remarkably well with
the natural amino acids (RMSLeu/PTL ) 0.30 Å, RMSGlu/PTE
) 0.16 Å, RMSArg/PTR ) 0.12 Å).9,10
F igu r e 5. Selected NOEs and theoretical and experimental
distances for oligomeric PTAAs.
that the minimization of pseudo A(1,3)-strain between
adjacent prolines can define ψ ∼ 145° as required in PPII
helices. Geometry optimization of dimer 19 (Piv-PTL-
PTL-OBn) affords a ψ value ) 131°, consistent with the
conformational analysis prediction. For Piv-PTL-PTL-
OBn the minimization of pseudo A(1,3)-strain should
place 2H(i) in close spatial relationship to 4cH(i + 1) and
4tH(i + 1) when the PTAA-PTAA amide bond is in the
trans conformation (Figure 5).13,14 Indeed molecular
modeling provides a 2H(i)-4cH(i + 1) distance ) 2.5 Å
and a 2H(i)-4tH(i + 1) distance ) 2.2 Å. More signifi-
cantly, for Piv-PTL-PTL-OBn the 2H(i)-4cH(i + 1)
distance and the 2H(i)-4tH(i + 1) distance calculated
from the NOESY cross-peak volumes are in excellent
agreement at 2.5 and 2.0 Å, respectively. These data
indicate that a unit as small as a dimer is conformation-
ally constrained to the φ/ψ angles of a PPII helix. It
Con for m a tion of P TL Oligom er s in CDCl3. The
conformational behavior of the PTL dimer 19 and trimer
20 was next studied in CHCl3. We previously have argued
(13) All oligomeric PTAAs (15, 16, 19, 20) exhibited only the
presence of trans amide bond rotamer.
(14) The designation (i) denotes the N-terminal residue.

458 J . Org. Chem., Vol. 66, No. 2, 2001
Mamai et al.
should be noted that the unequal 2H(i)-4cH(i + 1)/2H-
(i)-4tH(i + 1) distances are due to the flattened boat
conformation and that both residues populate the flat-
-70°, ø1 ∼ -57°, and ø2 ∼ -158°. Molecular modeling
estimates that oligomeric PTAAs based on this ring
system exhibit ψ ∼ 131°. Solution studies of oligoPTLs
in CDCl3 and oligoPTEs in D2O preferentially populate
the PPII conformation. These PTAAs should prove useful
for the synthesis of conformationally constrained PPII
mimics which possess residues with ø1 ∼ gauche(-) and
a ø2 ∼ trans. Work is ongoing to synthesize and evaluate
PPII mimics for a number of pharmacologically relevant
receptors.
tened boat conformation as evidenced by J 5H(i)-4cH(i)
)
J 5H(i+1)-4cH(i+1) ) 5.2 Hz (consistent with J 4cH-5H ) 5.2 Hz
for Ac-PTL-OBn). The trimer Piv-PTL-PTL-PTL-OBn is
an important compound for our studies since three
residues define one turn of a PPII helix. A NOESY
spectrum of trimer 20 also reveals the key NOEs between
2H(i)-4H(i + 1) and 2H(i + 1)-4H(i + 2). A statement
about the distances and relative NOE intensities between
2H(i)-4cH(i + 1) and 2H(i)-4tH(i + 1) cannot be made
in this case because 4cH(i + 1) and 4tH(i + 1) are not
resolved. However, since these NOEs are the strongest
inter-PTAA NOEs it may be stated that qualitatively
trimer 20 preferentially populates a conformation in
which ψ ∼ 131°. It should also be noted that the three
residues maintain the flattened boat conformation as
evidenced by 5H-4cH coupling constants which are 5.0,
5.2, and 5.2 Hz for the i, i + 1, and i + 2 PTLs,
respectively.
Exp er im en ta l Section
Gen er a l. Unless otherwise specified, all reagents were
purchased from commercial sources and were used without
further purification. THF was distilled from sodium benzophe-
none ketyl and CH2Cl2 was distilled from CaH2. All reactions
were carried out under an N2 or Ar atmosphere. Flash
chromatography was carried out on Selecto silica gel (230-
400 mesh). 1D and 2D NMR spectra were collected in CDCl3
using standard pulse sequences provided by Bruker. NOESY
spectra were obtained using a 2 s repetition delay and a 400
ms mixing time with TD1 ) 4096 and TD2 ) 512. NOESY
cross-peak volumes were calculated by the method described
by Prestegard.15 Characterization data as well as 1H and 13C
NMR spectra of the PTL series have been included as Sup-
porting Information in ref 4. To avoid duplicity, these have
not been included here.
Molecu la r Mod elin g. Structure building was performed
with Chem 3D. The structure generated from Chem 3D was
then subjected to geometry optimization with Gaussian 98 at
the HF/6-31G* level. The Chem 3D structure was also geom-
etry optimized with PCMODEL 7.0 using the MMX force field.
The minimum energy value for ψ was obtained using the
dihedral driver function of PCMODEL. For the Leu/PTL
overlay, the following leucine values were used ø1 ) -60°, ø2
) 180°, ø2′ ) -60°. CR, Câ, Cγ, and Cδ as well as the nitrogen
and carbonyl carbon of leucine were chosen to be overlaid with
the corresponding atoms on PTL. For the Glu/PTE overlay,
the following glutamic acid values were used ø1 ) -60°, ø2 )
180°. CR, Câ, Cγ, and Cδ as well as the nitrogen and carbonyl
carbon of glutamic acid were chosen to be overlaid with the
corresponding atoms on PTE. For the Arg/PTR overlay, the
following Arg values were used ø1 ) -60°, ø2 ) 180°, ø3 )
180° while for PTR, ø3 ) -152°. CR, Câ, Cγ, and Cδ as well
as the nitrogen and carbonyl carbon of arginine were chosen
to be overlaid with the corresponding atoms on PTR.
Con for m a tion of P TE Oligom er s in D2O. With the
establishment that PTL oligomers populate the PPII
conformation in solution, we next investigated the con-
formational behavior of the PTE oligomers in D2O.
Geometry optimization of dimer 15 (Piv-PTE-PTE-OH)
also afforded a ψ value ) 131°. Furthermore, the opti-
mized structure also exhibited the 2H(i)-4cH(i + 1) and
2H(i)-4tH(i + 1) distances as 2.5 Å and ) 2.2 Å,
respectively. For Piv-PTE-PTE-OH the 2H(i)-4cH(i + 1)
and 2H(i)-4tH(i + 1) distances calculated from the
NOESY cross-peak volumes are 2.4 and 1.8 Å respec-
tively. We were somewhat surprised by the shorter 2H-
(i)-4tH(i + 1) distance. To determine whether this were
a function of the PTAA or the solvent (D2O), we used
dimer 21 in CDCl3 as a reference compound which is
similar to Piv-PTE-PTE-OH in that it does not possess
the C7 methyl as the PTL analogues do. The 2H(i)-4cH-
(i + 1) and 2H(i)-4tH(i + 1) distances for 21 calculated
from the NOESY cross-peak volumes are 2.5 and 1.9 Å,
respectively suggesting that the shorter 2H(i)-4tH(i +
1) distance is a function of the PTAA. We suspect that
again this reflects the more “pronounced averaged” boat
in the PTE framework vs the PTL framework. Coupling
constants (J 5H(i)-4cH(i) ) J 5H(i+1)-4cH(i+1) ) 3.9 Hz) indicate
the same conformational behavior for the bicyclic frame-
works as is found for the monomer Ac-PTE(Me)-OBn.
The PTE trimer also exhibits the 2H(i)-4cH(i + 1)/2H-
(i)-4tH(i + 1) and the 2H(i + 1)-4cH(i + 2)/2H(i + 1)-
4tH(i + 2) NOEs with the 2H(i)-4tH(i + 1) and 2H(i +
1)-4tH(i + 2) NOEs of greater magnitude. Although
overlap of the 4cH-4tH NOE cross-peaks does not allow
a quantitative analysis of the proton distances, it is
possible again to make a qualitative statement that the
PTE trimer preferentially populates a conformation in
which ψ ∼ 131°. Furthermore, the bicyclic ring system
in trimer 16 also populates a flattened boat conformation
as evidenced by J 5H(i)-4cH(i) ) 3.8 Hz J 5H(i+1)-4cH(i+1) ) 3.9
Hz, J 5H(i+1)-4cH(i+1) ) 3.8 Hz. The data provided from these
studies provides compelling evidence that even in aque-
ous medium, dimeric and trimeric water soluble PPII
mimics are constrained to the φ/ψ angles of a PPII helix.
Ben zyl 1R,2S,5S,6R-3-Aza -3-ter t-bu tyl(oxyca r bon yl)-6-
(m eth oxycar bon yl)bicyclo[3.1.0]h exan e-2-car boxylate (9).
A solution of diisopropylcarbodiimide (0.56 g, 4.4 mmol), acid
8 (0.84 g, 3.0 mmol), and DMAP (0.36 g, 3.0 mmol) in CH2Cl2
(6 mL) was maintained at 0 °C for 30 min. Benzyl alcohol (0.48
g, 4.4 mmol) was added, and the cooling bath was removed.
After 12 h, the solvent was removed under reduced pressure,
and the residue was purified by flash chromatography (9:1
hexanes-EtOAc) affording the ester 9 as a colorless oil (0.91
1
g, 82%): [R]25 ) -54.3° (c 0.12, CHCl3); H NMR (500 MHz,
D
CDCl3) δ 7.23-7.35 (m, 5H), 5.19 (d, J ) 12.5 Hz, 0.6H), 5.15
(d, J ) 12.5 Hz, 0.6H), 5.10 (d, J ) 12.2 Hz, 0.4H), 5.07 (d, J
) 12.2 Hz, 0.4H), 4.49 (s, 0.4H), 4.35 (s, 0.6H), 3.67-3.69 (m,
0.4H), 3.62-3.64 (m, 0.6H), 3.60 (s, 3H), 3.59-3.57 (m, 0.6H),
3.54-3.56 (m, 0.4H), 2.15-2.18 (m, 0.6H), 2.12-2.14 (m, 0.4H),
2.05-2.07 (m, 0.6H), 2.02-2.04 (m, 0.4H), 1.58 (dd, app t, J )
3.1 Hz, 0.6H), 1.55 (dd, app t, J ) 3.2 Hz, 0.4H), 1.39 (s, 3.6H),
1.27 (s, 5.4H); 13C NMR (125 MHz, CDCl3) 171.9, 170.6, 170.5,
154.3, 153.8, 135.3, 128.5, 128.4, 128.2, 128.0, 80.4, 66.9, 61.1,
60.7, 51.8, 48.0, 47.9, 29.5, 29.1, 28.3, 28.2, 28.0, 25.6, 24.9,
23.8, 23.6 ppm; IR (film) 1744, 1730, 1728, 1708, 1665, 1655
cm-1; MS (CI) m/z 376 (MH). Anal. Calcd for C20H25O6N: C,
63.98; H, 6.71; N, 3.73. Found: C, 63.76; H, 6.72; N, 3.64.
Con clu sion s
Our studies show that PTAAs based on the 3-azabicyclo-
[3.1.0]hexane system possess torsional parameters φ ∼
(15) Holak, T. A.; Scarsdale, J . N.; Prestegard, J . H. J . Magn. Reson.
1987, 74, 546.

Poly-L-proline Type II Peptide Mimics
J . Org. Chem., Vol. 66, No. 2, 2001 459
Ben zyl 1R,2S,5S,6R-3-Aza -3-(2,2-d im eth ylp r op a n oyl)-
6-(m e t h oxyca r b on yl)b icyclo[3.1.0]h e xa n e -2-ca r b oxy-
la te (10). The ester 9 (0.50 g, 1.3 mmol) was dissolved in a
1:1 mixture of TFA/CH2Cl2 (2 mL). After 30 min at room
temperature the volatiles were removed under reduced pres-
sure. The resulting oil was dissolved in CH2Cl2 (20 mL) and
washed with saturated aqueous NaHCO3 (10 mL). The aque-
ous layer was extracted with CH2Cl2 (3 × 15 mL), and the
combined organic layers were dried over Na2SO4 and concen-
trated to afford the free amine (0.37 g) as a yellow oil which
was used without further purification: 1H NMR (500 MHz,
CDCl3) δ 7.33-7.26 (m, 5H), 5.14 (d, J ) 12.3 Hz, 1H), 5.09
(d, J ) 12.3 Hz, 1H), 3.81 (s, 1H), 3.60 (s, 3H), 3.19 (dd, J )
3.5, 10.1 Hz, 1H), 3.02 (d, J ) 10.1 Hz, 1H), 2.23 (dd, J ) 2.9,
6.7 Hz, 1H), 2.02-1.98 (m, 2H), 1.67 (dd, app t, J ) 3.0 Hz,
1H); 13C NMR (125 MHz, CDCl3) 173.1, 172.8, 135.5, 128.5,
128.3, 128.1, 66.7, 61.2, 51.6, 47.4, 29.2, 26.4, 21.9 ppm; IR
(film) 3405, 1738, 1730 cm-1. The secondary amine from above
was dissolved in dry CH2Cl2 (4 mL), the solution was cooled
to 0 °C under argon, and Et3N (0.41 g, 4.1 mmol) and (CH3)3-
CCOCl (0.25 g, 2.1 mmol) were successively added. After 1 h,
the mixture was washed with saturated aqueous NaHCO3 (5
mL). The aqueous layer was extracted with CH2Cl2 (2 × 10
mL), and the combined organic layers were dried over Na2-
SO4 and concentrated. Flash chromatography with 1.5-8.5
(dd, app t, J ) 3.1 Hz, 1H), 1.46 (dd, app t, J ) 3.1 Hz, 1H),
1.17 (s, 9H); 13C NMR (125 MHz, CDCl3) 177.5, 171.7, 171.5,
170.3, 169.6, 135.0, 128.5, 128.3, 128.1, 67.3, 61.6, 60.5, 51.9,
51.8, 49.7, 48.2, 38.7, 27.4, 27.0, 25.8, 25.5, 23.5, 23.4 ppm; IR
(KBr) 1371, 1730, 1673, 1666 cm-1; MS (CI) m/z 527 (MH).
Anal. Calcd for C28H34N2O8: C, 63.87; H, 6.51; N, 5.32.
Found: C, 63.65; H, 6.49; N, 5.18.
Syn th esis of Tr im er 14. Pd-C (10%) (15 mg) was added
to a solution of the dimeric ester 13 (134 mg, 0.25 mmol) in
EtOAc (3 mL). This mixture was stirred overnight under a
balloon of hydrogen. The catalyst was removed by filtration
over a Celite pad. The filtrate was concentrated, and the
resulting pale solid was crystallized from hexanes-EtOAc (4:
1) to give the acid (85 mg, 78%) as a colorless solid: mp 127-
129 °C; [R]25 ) -86.4 (c 0.11, CHCl3); 1H NMR (500 MHz,
D
CDCl3) δ 8.15-7.96 (br s, 1H), 4.78 (s, 1H), 4.71 (s, 1H), 4.13
(d, J ) 10.0 Hz, 1H), 4.02-3.96 (m, 2H), 3.88 (dd, J ) 4.1,
10.9 Hz, 1H), 3.65 (s, 6H), 2.35 (dd, J ) 3.0, 7.1 Hz, 1H), 2.30-
2.27 (m, 1H), 2.24-2.21 (m, 1H), 2.11 (dd, J ) 2.9, 7.3 Hz,
1H), 1.72 (dd, app t, J ) 3.0 Hz, 1H), 1.48 (dd, app t, J ) 3.0
Hz, 1H),1.18 (s, 9H); 13 C NMR (125 MHz, CDCl3) 177.8, 171.9,
171.6, 170.9, 61.7, 60.7, 51.9, 49.8, 48.3, 47.9, 38.8, 27.3, 27.2,
27.0, 25.8, 25.6, 23.5, 23.4 ppm; IR (KBr) 3174, 1737, 1730,
1673, 1666 cm-1; HRMS (CI) m/z 437.1923 (437.1927 calcd for
C
21H29N2O8, MH). The ester 9 (45 mg, 0.12 mmol) was
hexanes-EtOAc afforded the pivaloyl ester 17 (0.46 g, 95%
dissolved in a 1:1 mixture TFA/CH2Cl2 (0.5 mL). After 30 min,
the volatiles were removed under reduced pressure, and the
resulting oil was dissolved in CH2Cl2 (5 mL) and then washed
with saturated aqueous NaHCO3 (5 mL). The aqueous layer
was extracted with CH2Cl2 (3 × 5 mL), and the combined
organic layers were dried over Na2SO4 and concentrated to
afford the free amine (33 mg) as a yellow oil which was used
without further purification. This amine was dissolved in dry
CH2Cl2 (0.5 mL) and then added to a solution of DCC (25 mg,
0.12 mmol) and the acid obtained above (52 mg, 0.12 mmol)
in dry CH2Cl2 (1 mL). The mixture was maintained overnight
under argon then washed with HCl (1 N). The aqueous layer
was extracted with CH2Cl2 (3 × 5 mL), and the combined
organic layers were dried over Na2SO4 and then concentrated.
Flash chromatography of the residue on silica gel eluting with
1:1 EtOAc-hexanes afforded the trimeric ester 14 (75 mg,
91%) as a white solid: mp 114-116 °C; [R]25D ) -75.8 (c 0.58,
CHCl3); 1H NMR (500 MHz, CDCl3) δ 7.35-7.29 (m, 5H), 5.23
(d, J ) 12.2 Hz, 1H), 5.08 (d, J ) 12.2 Hz, 1H), 4.80 (s, 1H),
4.74 (s, 1H), 4.66 (s, 1H), 4.06 (d, J ) 9.9 Hz, 1H), 4.01 (d, J
) 10.0 Hz, 1H), 3.95-3.92 (m, 3H), 3.85 (dd, J ) 3.9, 9.9 Hz,
1H), 3.64 (s, 3H), 3.63 (s, 3H), 3.62 (s, 3H), 2.26-2.23 (m, 1H),
2.21 (dd, J ) 3.0, 7.1 Hz, 1H), 2.21-2.16 (m, 2H), 2.14-2.11
(m, 2H), 1.69-1.67 (m, 2H), 1.42 (dd, app t, J ) 3.1 Hz, 1H),
1.69 (s, 9H); 13C NMR (125 MHz, CDCl3) 177.4, 171.6, 171.4,
170.1, 169.5, 169.2, 134.9, 128.9, 128.4, 128.2, 67.4, 61.8, 60.7,
60.1, 51.9, 51.8, 49.7, 48.7, 48.1, 38.7, 29.6, 27.3, 27.1, 26.7,
26.1, 25.7, 25.5, 23.7, 23.6, 23.5 ppm; IR (KBr) 1741, 1737,
1732, 1728, 1672, 1655, 1651 cm-1; MS (CI) m/z 694 (MH).
1
overall) as a colorless oil: [R]25 ) -47.5° (c 0.12, CHCl3); H
D
NMR (500 MHz, CDCl3) δ 7.28-7.23 (m, 5H), 5.20 (d, J ) 12.2
Hz, 1H), 5.90 (d, J ) 12.2 Hz, 1H), 4.78 (s, 1H), 4.00 (d, J )
10.3 Hz, 1H), 3.82-3.78 (m, 1H), 3.63 (s, 3H), 2.14-2.12 (m,
2H), 1.49-1.46 (m, 1H), 1.17 (s, 9H); 13C NMR (125 MHz,
CDCl3) 177.6, 171.1, 170.4, 135.3, 128.5, 128.3, 128.1, 67.0,
62.3, 51.9, 49.1, 38.8, 27.1, 26.3, 23.2 ppm; IR (film) 1737, 1730,
1637 cm-1; MS (CI) m/z 360 (MH). Anal. Calcd for C20H25NO5:
C, 66.83; H, 7.01; N, 3.89. Found: C, 66.65; H, 7.03; N,3.79.
1R ,2S ,5S ,6R -3-Az a -3-(2,2-d i m e t h y lp r o p a n o y l)-6-
(m eth oxycar bon yl)bicyclo[3.1.0]h exan e-2-car boxylic Acid
(11). Pd-C (10%) (45 mg) was added to a solution of the amide
10 (0.45 g, 1.3 mmol) in EtOAc (6 mL). This mixture was
stirred overnight under a balloon of hydrogen. The catalyst
was removed by filtration through a Celite pad. The filtrate
was concentrated, and the resulting pale solid was recrystal-
lized from hexanes-EtOAc (4:1) to give the acid 11 (0.31 g,
92%) as a colorless solid: mp 165-167 °C; [R]25 ) -38.9° (c
D
1
0.35, CHCl3); H NMR (500 MHz, CDCl3) δ 9.45-9.55 (br s,
1H), 4.26 (s, 1H), 3.95 (d, J ) 10.6 Hz, 1H), 3.81-3.87 (m,
1H), 3.61 (s, 3H), 2.31 (dd, J ) 2.7, 7.1 Hz, 1H), 2.24-2.19 (m,
1H), 1.48 (dd, app t, J ) 2.7 Hz, 1H), 1.19 (s, 9H); 13C NMR
(125 MHz, CDCl3) 178.8, 173.7, 171.3, 62.4, 51.9, 49.2, 38.9,
27.0, 26.2, 25.8, 23.1 ppm; IR (KBr) 3203, 1737, 1730, 1651
cm-1; MS (CI) m/z 270 (MH). Anal. Calcd for C13H19NO5: C,
57.98; H, 7.11; N, 5.20. Found: C, 57.74; H, 6.99; N, 5.14.
Syn th esis of Dim er 13. The ester 9 (217 mg, 0.58 mmol)
was dissolved in a 1:1 mixture TFA/CH2Cl2 (1 mL). After 30
min at room temperature, the volatiles were removed under
reduced pressure. The resulting oil was dissolved in CH2Cl2
(10 mL) and washed with saturated aqueous NaHCO3 (5 mL).
The aqueous layer was extracted with CH2Cl2 (3 × 8 mL), and
the combined organic layers were dried over Na2SO4 and
concentrated to afford the free amine (160 mg) as a yellow oil
which was used without further purification. This amine was
dissolved in dry CH2Cl2 (1 mL) and then added to a solution
of DCC (120 mg, 0.58 mmol) and the acid 11 (157 mg, 0.58
mmol) in dry CH2Cl2 (2 mL). The mixture was stirred for 4 h
under argon then washed with 1 N HCl (5 mL), and the
aqueous layer was extracted with CH2Cl2 (3 × 5 mL). The
combined organic layers were dried over Na2SO4 and then
concentrated. Flash chromatography of the residue on silica
gel eluting with 1:3 EtOAc-hexanes afforded the dimeric ester
Anal. Calcd for C36H43N3O11
: C, 62.33; H, 6.23; N, 6.06.
Found: C, 62.47; H, 6.35; N, 5.93.
Syn th esis of Dim er 15. The dimeric ester 13 (80 mg, 0.15
mmol) was dissolved in THF (1 mL) and then added to a 1 M
solution of lithium hydroxide (2 mL). After 4 h at room
temperature, the solution was concentrated to a 1 mL volume
under reduced pressure and passed over a DOWEX resin
eluting with water to give a colorless solid. This crude
compound was crystallized from 1:4 THF-hexanes to give the
dimer 15 (58 mg, 93%) as a colorless powder: mp 135-137
1
°C; [R]25 ) -47.5 (c 0.2, CHCl3); H NMR (500 MHz, D2O) δ
D
4.73 (s, 1H), 4.61 (s, 1H), 4.10 (d, J ) 10.8 Hz, 1H), 4.04 (d, J
) 10.8 Hz, 1H), 3.92 (dd, J ) 3.8, 10.8 Hz, 1H), 3.88 (dd, J )
4.1, 10.8 Hz, 1H), 2.35 (dd, J ) 3.0, 7.1 Hz, 1H), 2.28-2.33
(m, 2H), 2.11 (dd, J ) 2.9, 7.3 Hz, 1H), 1.60 (dd, app t, J ) 3.0
Hz, 1H), 1.56 (dd, app t, J ) 3.0 Hz, 1H), 1.10 (s, 9H); 13C
NMR (125 MHz, CD3OD) 179.6, 174.8, 172.6, 172.3, 63.8, 63.7,
62.1, 51.3, 39.9, 32.7, 28.6, 27.5, 26.7, 24.9, 24.7, 23.6, 23.5
ppm; FTIR (KBr) 3449, 3072, 1735, 1729, 1710, 1670, 1658,
1443, 1437 cm-1; HRMS (CI) m/z 409.1614 (409.1611 calcd for
13 (234 mg, 77%) as a colorless solid: mp 50-52 °C; [R]25
)
D
-64.1° (c 0.46, CHCl3); 1H NMR (500 MHz, CDCl3) δ 7.33-
7.23 (m, 5H), 5.22 (d, J ) 12.2 Hz, 1H), 5.08 (d, J ) 12.2 Hz,
1H), 4.82 (s, 1H), 4.69 (s, 1H), 4.11 (d, J ) 9.9 Hz, 1H), 3.99-
3.96 (m, 2H), 3.87 (dd, J ) 3.7, 9.9 Hz, 1H), 3.64 (s, 3H), 3.63
(s, 3H), 2.24-2.15 (m, 3H), 2.09 (dd, J ) 3.0, 7.4 Hz, 1H), 1.72
C
19H25N2O8, MH).

460 J . Org. Chem., Vol. 66, No. 2, 2001
Mamai et al.
Syn th esis of Tr im er 16. The ester 14 (61 mg, 0.09 mmol)
was dissolved in THF (1 mL) and then added to a 1 M solution
of LiOH (1.5 mL). After 4 h at room temperature, the solution
was concentrated to a 1 mL volume under reduced pressure
and passed over a DOWEX resin eluting with water to give a
colorless solid. This crude compound was crystallized from 1:4
over Na2SO4 and then concentrated. Flash chromatography
of the residue on silica gel eluting with 7:3 EtOAc-hexanes
afforded the amide 18 (45 mg, 98%) as a colorless oil: [R]25
)
D
-54.1 (c 0.22, CHCl3); 1H NMR (500 MHz, CDCl3) δ 7.39-
4.33 (m, 5H), 5.22-5.20 (m, 1.3H), 5.16 (d, J ) 12.3 Hz, 0.7H),
4.77 (s, 0.7H), 4.46 (s, 0.3H), 3.99 (d, J ) 12.0 Hz, 0.3H), 3.86
(dd, J ) 3.9, 10.1 Hz, 0.7H), 3.69-3.66 (m, 3.7H), 3.56 (dd, J
) 4.1, 12.0 Hz, 0.3H), 2.35 (dd, J ) 3.0, 7.1 Hz, 0.3H), 2.24
(dd, J ) 3.0, 7.1 Hz, 0.7H), 2.21-2.17 (m, 0.7H), 2.16-2.12
(m, 0.3H), 2.04 (s, 2.1H), 1.91 (s, 0.9H), 1.58 (dd, app t, J )
3.1 Hz, 0.7H), 1.54 (dd, app t, J ) 3.1 Hz, 0.3H); 13C NMR
(125 MHz, CDCl3) 171.6, 170.1, 170.0, 169.8, 169.6, 135.2,
134.8, 128.7, 128.5, 128.3, 128.2, 128.0, 67.6, 67.2, 62.0, 60.3,
51.9, 49.2, 47.5, 29.6, 29.2, 27.9, 25.6, 24.2, 23.8, 23.6, 22.2
ppm; IR (film) 1744, 1737, 1666, 1659 cm-1; MS (CI) m/z 318
(MH). Anal. Calcd for C17H19NO5: C, 64.96; H, 6.03; N, 4.41.
Found: C, 64.77; H, 6.08; N, 4.25.
THF-hexanes to give the trimer 16 (51 mg, 83%) as a colorless
1
powder: mp 165-167 °C; [R]25 ) -52.8 (c 0.26, CHCl3); H
D
NMR (500 MHz, D2O) δ 4.78 (s, 1H), 4.71 (s, 1H), 4.62 (s, 1H),
4.08 (d, J ) 10.1 Hz, 1H), 4.05 (d, J ) 11.0 Hz, 1H), 4.01 (d,
J ) 11.0 Hz, 1H), 3.91 (dd, J ) 3.9, 11.0 Hz, 2H), 3.87 (dd, J
) 4.1, 10.9 Hz, 1H), 2.35-2.32 (m, 3H), 2.30 (dd, J ) 3.6, 7.7
Hz, 1H), 2.23 (dd, J ) 3.0, 7.3 Hz, 1H), 2.05 (dd, J ) 2.9, 7.3
Hz, 1H), 1.65 (dd, app t, J ) 3.0 Hz, 1H), 1.59 (dd, app t, J )
3.0 Hz, 1H), 1.53 (dd, app t, J ) 3.0 Hz, 1H), 1.04 (s, 9H). 13C
NMR (125 MHz, CD3OD) 179.8, 174.8, 174.7, 172.9, 172.2,
171.2, 63.8, 62.4, 62.1, 62.0, 51.3, 50.2, 49.8, 39.9, 30.7, 28.5,
28.4, 28.0, 27.5, 26.7, 25.2, 25.0, 24.7 ppm; IR (KBr) 3416, 3085,
2858, 1735, 1729, 1722, 1703, 1670, 1665, 1657, 1463, 1430
cm-1; HRMS (CI) m/z 562.2036 (562.2037 calcd for C26H32N3O11,
MH).
Ack n ow led gm en t. Support for this work has been
provided by NSF and NIH.
Ben zyl 1R,2S,5S,6R-3-Aza -3-(et h a n oyl)-6-(m et h oxy-
ca r bon yl)bicyclo[3.1.0]h exa n e-2-ca r boxyla te (18). The
amine 12 (40 mg, 0.14 mmol) was dissolved in dry CH2Cl2 (1
mL) and then cooled to 0 °C. Et3N (44 mg, 0.44 mmol) and
AcCl (23 mg, 0.29 mmol) were added to the mixture. After 30
min at 0 °C, the solution was diluted with CH2Cl2 (10 mL)
and washed with 1 N HCl (3 mL). The organic layer was dried
Su p p or t in g In for m a t ion Ava ila b le: 1H and 13C NMR
spectra of 9, 10, 11, 12, 13, 14, 15, 16, and 18, and NOESY
spectra of 13, 15, 16, and 18. This material is available free
of charge via the Internet at http://pubs.acs.org.
J O001201O