Effect of progressive benzyl substitution on the conformations of aminocaproic acid-cyclized dipeptides

Article abstract of DOI:10.1021/jo001308b

The constraint of dipeptides into a β-turn conformation can be accomplished by linking the two ends of a standard dipeptide with a linker derived from aminocaproic acid (aca). To elucidate the possibility of using substituted Aca linkers in peptidomimetic design, a series of five macrocycles composed of a monobenzylated Aca linker (containing the benzyl group on each of the five methylene groups of the parent linker) and Gly-gly were synthesized. The requisite linkers were made by regiochemically controlled ring expansion techniques (for substitution on Aca positions C-3, C-4, or C-5), an Evans alkylation route (for C-2), or by chain extension of L-phenylalanal (for C-6). The solution-phase conformations of the macrocycles were examined by Nmr and Cd techniques; in addition, crystal structures of the C-4- and C-6-benzyl-substituted linkers were obtained. Four out of the five macrocycles were found to exist with the dipeptide portion taking up either a type Ii or Ii′ β-turn conformation, but the Gly-gly unit in the compound derived from 4-benzyl-aca did not correspond to one of the standard β-turn types.

Full text of DOI:10.1021/jo001308b

2636
J . Org. Chem. 2001, 66, 2636-2642
Effect of P r ogr essive Ben zyl Su bstitu tion on th e Con for m a tion s of
Am in oca p r oic Acid -Cyclized Dip ep tid es
Mary MacDonald, David Vander Velde, and J effrey Aube´*
Department of Medicinal Chemistry, University of Kansas, Lawrence, Kansas 66045-2506
jaube@ukans.edu
Received August 30, 2000
The constraint of dipeptides into a â-turn conformation can be accomplished by linking the two
ends of a standard dipeptide with a linker derived from aminocaproic acid (Aca). To elucidate the
possibility of using substituted Aca linkers in peptidomimetic design, a series of five macrocycles
composed of a monobenzylated Aca linker (containing the benzyl group on each of the five methylene
groups of the parent linker) and Gly-Gly were synthesized. The requisite linkers were made by
regiochemically controlled ring expansion techniques (for substitution on Aca positions C-3, C-4,
or C-5), an Evans alkylation route (for C-2), or by chain extension of L-phenylalanal (for C-6). The
solution-phase conformations of the macrocycles were examined by NMR and CD techniques; in
addition, crystal structures of the C-4- and C-6-benzyl-substituted linkers were obtained. Four out
of the five macrocycles were found to exist with the dipeptide portion taking up either a type II or
II′ â-turn conformation, but the Gly-Gly unit in the compound derived from 4-benzyl-Aca did not
correspond to one of the standard â-turn types.
In tr od u ction
are attached to resins to allow for solid-phase syntheses
which permit the development of â-turn-inspired pepti-
domimetic libraries.12-16
Understanding the bound conformations of bioactive
peptides is an important goal of rational drug design.
Because peptides, per se, can be difficult to develop as
drugs, the use of peptidomimetics has greatly expanded
in recent years. Besides ameliorating the problems of
poor oral bioavailability and proteolytic degradation, such
molecules are often designed to bias the structure toward
the active conformation, thus increasing its affinity to
the target. Peptidomimetics based on â-turns1 are im-
portant because numerous peptides require such a con-
formation to produce a biological response.2-6 Many turn
replacements consist of rigid heterocyclic units that are
incorporated into a polypeptide chain in place of the
central dipeptide of the reported turn.7 An alternative
method of mimicry constrains the dipeptide in question
with a tether, often incorporating heteroatoms8,9 or
aromatic groups10,11 into the carbon chain. Some tethers
Woody, Scheraga, and co-workers first described the
use of 6-aminocaproic acid (Aca) as a linker to constrain
L-alanylglycine into a â-turn (e.g., see compound 1a for
an example).17-20 Conformational analysis of these com-
pounds were carried out using molecular mechanics,
nuclear magnetic resonance (NMR), circular dichroism
(CD), and infrared (IR) spectroscopies. In this work, they
found that the amino acid stereochemistry had a pro-
found influence on the subtype of â-turn adopted by these
macrocycles. In general, macrocycles derived from L,L-
dipeptides preferred a type I turn conformation, whereas
L,D-dipeptides took up a type II solution structure.
Earlier work in this laboratory has focused on the effect
of substitution within the Aca tether on the conforma-
(11) Feng, Y.; Wang, Z.; J in, S.; Burgess, K. J . Am. Chem. Soc. 1998,
120, 10768-10769.
* Corresponding author: Tel 1.785.864.4496. Fax 1.785.864.5326.
(1) Rose, G. D.; Gierasch, L. M.; Smith, J . A. In Advances in Protein
Chemistry; Anfinsen, C. B.; Edsall, J . T.; Richards, F. M., Eds.;
Academic: Orlando, 1985; Vol. 37, pp 1-109.
(2) Farmer, P. S. In Drug Design; Arie¨ns, E. J ., Ed.; Academic: New
York, 1980; Vol. 10, pp 119-143.
(3) Giannis, A.; Kolter, T. Angew. Chem., Int. Ed. Engl. 1993, 32,
1244-1267.
(4) Olson, G. L.; Bolin, D. R.; Bonner, M. P.; Bo¨s, M.; Cook, C. M.;
Fry, D. C.; Graves, B. J .; Hatada, M.; Hill, D. E.; Kahn, M.; Madison,
V. S.; Rusiecki, V. K.; Sarabu, R.; Sepinwall, J .; Vincent, G. P.; Voss,
M. E. J . Med. Chem. 1993, 36, 3039-3049.
(5) Gante, J . Angew. Chem., Int. Ed. Engl. 1994, 33, 1699-1720.
(6) Aube´, J . In Advances in Amino Acid Mimetics and Peptidomi-
metics; Abell, A., Ed.; J AI Press: Greenwich, 1997; Vol. 1, pp 193-
232.
(12) Feng, Y.; Pattarawarapan, M.; Wang, Z.; Burgess, K. Org. Lett.
1999, 1, 121-124.
(13) Virgilio, A. A.; Ellman, J . A. J . Am. Chem. Soc. 1994, 116,
11580-11581.
(14) Virgilio, A. A.; Schu¨rer, S. C.; Ellman, J . A. Tetrahedron Lett.
1996, 37, 6961-6964.
(15) Virgilio, A. A.; Bray, A. A.; Zhang, W.; Snyder, M.; Morrissey,
M. M.; Ellman, J . A. Tetrahedron 1997, 53, 6635-6644.
(16) Eguchi, M.; Lee, M. S.; Nakanishi, H.; Stasiak, M.; Lovell, S.;
Kahn, M. J . Am. Chem. Soc. 1999, 121, 12204-12205.
(17) Bandekar, J .; Evans, D. J .; Krimm, S.; Leach, S. J .; Lee, S.;
McQuie, J . R.; Minasian, E.; Ne´methy, G.; Pottle, M. S.; Scheraga, H.
A.; Stimson, E. R.; Woody, R. W. Int. J . Pept. Protein Res. 1982, 19,
187-205.
(18) Deslauriers, R.; Evans, D. J .; Leach, S. J .; Meinwald, Y. C.;
Minasian, E.; Ne´methy, G.; Rae, I. D.; Scheraga, H. A.; Somorjai, R.
L.; Stimson, E. R.; Van Nispen, J . W.; Woody, R. W. Macromolecules
1981, 14, 985-996.
(19) Maxfield, F. R.; Bandekar, J .; Krimm, S.; Evans, D. J .; Leach,
S. J .; Ne´methy, G.; Scheraga, H. A. Macromolecules 1981, 14, 997-
1003.
(20) Ne´methy, G.; McQuie, J . R.; Pottle, M. S.; Scheraga, H. A.
Macromolecules 1981, 14, 975-985.
(21) Kitagawa, O.; Vander Velde, D.; Dutta, D.; Morton, M.; Takusa-
gawa, F.; Aube´, J . J . Am. Chem. Soc. 1995, 117, 5169-5178.
(7) Hanessian, S.; McNaughton-Smith, G.; Lombart, H.-G.; Lubell,
W. D. Tetrahedron 1997, 53, 12789-12854.
(8) Shao, H.; Wang, S. H. H.; Lee, C.-W.; O¨ sapay, G.; Goodman, M.
J . Org. Chem. 1995, 60, 2956-2957.
(9) Shao, H.; Lee, C.-W.; Zhu, Q.; Gantzel, P.; Goodman, M. Angew.
Chem., Int. Ed. Engl. 1996, 35, 90-92.
(10) Bach, A. C., II.; Espina, J . R.; J ackson, S. A.; Stouten, P. F. W.;
Duke, J . L.; Mousa, S. A.; DeGrado, W. F. J . Am. Chem. Soc. 1996,
118, 293-294.
10.1021/jo001308b CCC: $20.00 © 2001 American Chemical Society
Published on Web 03/29/2001

Aminocaproic Acid-Cyclized Dipeptides
J . Org. Chem., Vol. 66, No. 8, 2001 2637
Sch em e 1
Sch em e 2
tional preference of similar macrocycles.21 Compounds
consisting of L-alanylglycine cyclized with various dia-
stereomers of 3,5-dimethyl-Aca were synthesized. In
addition to exhibiting the general dependence of confor-
mation on dipeptide stereochemistry expected from the
Woody/Scheraga results, the methyl group configurations
also influenced the type of â-turn adopted. While cyclo-
(L-Ala-Gly-Aca) had been previously shown to prefer the
type II orientation,18,19 some stereoisomers containing a
disubstituted linker existed largely in the type I confor-
mation. Interestingly, cyclo(Gly-Gly-(3R)-methyl-Aca) (1b),
which contains only a single stereocenter, had a CD
spectrum that bore a great resemblance to more highly
substituted compounds. Thus, it appeared that the pres-
ence of a single methyl group on the Aca linker could
persuade the dipeptide to reside predominantly in one
orientation.
In this paper, we report the synthesis and conforma-
tional analysis of a series of Gly-Gly-derived macrocycles
containing an Aca tether that possesses a single benzyl
substituent at each position of the tether. One purpose
of this study was to examine how a substituent at each
position of the Aca linker might influence the overall
macrocyclic conformation. Besides a role in influencing
conformation, such linkers may (1) serve as a surrogate
for a third amino acid (e.g., as needed for the mimicry of
the RGD sequence, which is a common target of â-turn
peptidomimetics22-24), (2) provide a site for attachment
to a solid support, or (3) play a role in solubilizing the
molecule. The broad application of these concepts in
medicinal chemistry will require straightforward routes
to an array of substituted linkers that can be plugged
into standard peptide synthesis protocols. Accordingly,
enantioselective syntheses of linkers containing a benzyl
group in positions 2-6 were carried out to help establish
the general synthetic routes required for future work. In
addition, preliminary conformational analyses of the
derived macrocycles using NMR, CD, and X-ray tech-
niques are described herein.25
Resu lts a n d Discu ssion
The general retrosynthetic analysis of the macrocycles,
shown in Scheme 1, followed precedent set by previous
groups working in this field.17-20 This modular approach
allows for the combination of a standard-issue dipeptide
with an appropriately substituted linker capped by one
of the commonly used amino acid protecting groups. The
linker syntheses will be described in the following section,
which will also include a brief discussion of cyclization
conditions. The paper will be concluded by presentation
and discussion of the conformational analysis of the
macrocycles.
Lin k er Syn th esis. The routes devised to make each
linker in enantiomerically enriched form varied according
to the position of the substitution (Scheme 2). Aca
derivatives bearing C-3, C-4, and C-5 substitution (14a ,
10, and 14b, respectively) were synthesized from the
corresponding lactam, which was prepared using a ste-
reospecific ring expansion sequence.26 The linker contain-
ing the benzyl group in the C-6 position was derived from
L-phenylalanal,27 whereas linker bearing the C-2 benzyl
group utilized an Evans chiral alkylation28 to establish
the stereocenter on Aca.
Enantiomerically pure 10, containing a 4-benzyl sub-
stituent, was synthesized according to the oxaziridine
ring-expansion protocol reported earlier by this group
(22) Ojima, I.; Chakravarty, S.; Dong, Q. Bioorg. Med. Chem. 1995,
3, 337-360.
(26) Lattes, A.; Oliveros, E.; Rivie`re, M.; Belzecki, C.; Mostowicz,
D.; Abramskj, W.; Piccinni-Leopardi, C.; Germain, G.; Van Meerssche,
M. J . Am. Chem. Soc. 1982, 104, 3929-3934.
(27) Fehrentz, J .-A.; Castro, B. Synthesis 1983, 676-678.
(28) Evans, D. A. In Asymmetric Synthesis; Morrison, J . D., Ed.;
Academic: Orlando, 1984; Vol. 3, pp 2-110.
(23) Samanen, J . Annu. Rep. Med. Chem. 1996, 31, 91-99.
(24) Engleman, V. W.; Kellogg, M. S.; Rogers, T. E. Annu. Rep. Med.
Chem. 1996, 31, 191-200.
(25) A portion of this work was previously communicated: Mac-
Donald, M.; Vander Velde, D.; Aube´, J . Org. Lett. 2000, 2, 1653-1655.

2638 J . Org. Chem., Vol. 66, No. 8, 2001
MacDonald et al.
Sch em e 3
(Scheme 3).29 Thus, 4-benzylcyclohexanone30 was con-
densed with (R)-R-methylbenzylamine ((R)-R-MBA), form-
ing an imine that was immediately oxidized with
m-CPBA at -78 °C to give a mixture of oxaziridines. The
stereochemistry of the major diastereomeric oxaziridine
7 was assigned as previously described.31 The oxaziridine
mixture was photolyzed (benzene or cyclohexane, 254 nm,
room temperature) to produce a mixture of lactams that
were separated by column chromatography to produce
the major isomer 8 in good overall yield. Reductive
removal of the phenethyl group upon treatment of 8 with
Na/NH3 provided 9. The enantiomeric purity of 9 was
examined by chiral HPLC (Chirobiotic T column, 35%
EtOH/hexane) and determined to be 98% ee. Subsequent
acid hydrolysis and esterification furnished the HCl salt
of 4-benzyl-Aca methyl ester (10).
Sch em e 4
Compounds 14a and 14b were simultaneously pre-
pared from (()-3-benzylcyclohexanone, also using ox-
aziridine chemistry (Scheme 4). In this case, synthesis
of the oxaziridines yielded two major diastereomers (11a
and 11b), with some minor diastereomeric impurities
1
evident in the H NMR. The combination of a racemic,
unsymmetrical ketone with enantiomerically pure R-MBA
permits the conversion of each enantiomer of ketone to
its own particular oxaziridine, each resulting from equa-
torial addition of the oxidant and formation of the unlike
relationship between the stereogenic nitrogen and ben-
zylic stereocenters.31,32 Each of these major oxaziridines
has been established to lead to different regioisomeric
lactams, thus allowing for the simultaneous synthesis of
enantiomerically pure lactams. In practice, the oxaziri-
(29) Aube´, J .; Wang, Y.; Hammond, M.; Tanol, M.; Takusagawa, F.;
Vander Velde, D. J . Am. Chem. Soc. 1990, 112, 4879-4891.
(30) Rosowsky, A.; Papoulis, A. T.; Forsch, R. A.; Queener, S. F. J .
Med. Chem. 1999, 42, 1007-1017.
dines were subjected to photolysis as a mixture, following
which the two predominant lactams (12a and 12b) were
isolated by column chromatography and purified further
by crystallization from Et2O/hexane. X-ray analysis of
12a and 12b verified the stereochemical assignments
(Supporting Information), and removal of the phenethyl
groups from each yielded lactams 13a and 13b, respec-
tively. Acid hydrolysis and esterification of each precursor
yielded the HCl salts of the substituted linkers 14a and
14b.
Synthesis of the linker with the benzyl group adjacent
to the amine (C-6) used a chiral pool approach (Scheme
5). Wittig reaction of Boc-L-phenylalanal27 with [3-(ethoxy-
carbonyl)propyl]triphenylphosphonium bromide33 pro-
vided the olefin 15. The stereochemistry of the double
bond was not relevant but was assigned the cis config-
(31) Aube´, J .; Hammond, M.; Gherardini, E.; Takusagawa, F. J . Org.
Chem. 1991, 56, 499-508. Correction: Idem., Ibid. J . Org. Chem. 1991,
56, 4086.
(32) Aube´, J .; Hammond, M. Tetrahedron Lett. 1990, 31, 2963-2966.
(33) Bestman, H. J .; Koschatzky, K. H.; Scha¨tzke, W.; Su¨ss, J .;
Vostrowsky, O. Liebigs. Ann. Chem. 1981, 1705-1720.
(34) Evans, D. A.; Britton, T. C.; Ellman, J . A. Tetrahedron Lett.
1987, 28, 6141-6144.
(35) Sheehan, J . C.; Preston, J .; Cruickshank, P. A. J . Am. Chem.
Soc. 1965, 87, 2492-2493.
(36) Hayashi, K.; Hamada, Y.; Shioiri, T. Tetrahedron Lett. 1992,
33, 5075-5076.
(37) Freifelder, M. Practical Catalytic Hydrogenation; Wiley-Inter-
science: New York, 1971.
(38) Diago-Meseguer, J .; Palomo-Coll, A. L. Synthesis 1980, 547-
551.
(39) Qian, L.; Sun, Z.; Deffo, T.; Mertes, K. B. Tetrahedron Lett.
1990, 31, 6469-6472.

Aminocaproic Acid-Cyclized Dipeptides
J . Org. Chem., Vol. 66, No. 8, 2001 2639
Sch em e 5
The synthesis of macrocycles 3-5 is summarized in
Scheme 8. Boc-protected glycylglycine was coupled to 3-,
4-, and 5-benzyl-Aca methyl ester (14a , 10, and 14b,
respectively). Hydrolysis of the ester, Boc deprotection,
and cyclization yielded the final compounds. The syn-
thesis of 6 proceeded via a different order of attachment
of glycine residues in an effort to increase cyclization
yields (Scheme 9). Linker 15 was first deprotected and
coupled to Boc-glycine. Ester hydrolysis and the attach-
ment of the second glycine residue to the C-terminus
furnished the protected, olefinic tripeptide 27. Hydroge-
nation of the olefin produced the tripeptide 28. After
hydrolysis of the ethyl ester, 29 was treated with TFA
for Boc removal and then cyclized using the conditions
mentioned above to afford compound 6. Hydrogenation
of 25 formed 30, which was examined by chiral HPLC
(Chiralcel OD-H, 4% EtOH/hexane). The compound was
obtained in ca. 93% ee, indicating that some epimeriza-
tion probably occurred during the basic olefination step
resulting in 15.
Sch em e 6
Due to the difficulties encountered during the ring
closure step, various cyclization conditions to form 6 were
surveyed (Table 1). First, after little success in closing
the macrocycle by forming the Gly-Aca bond as was done
for compounds 3-5, we decided to investigate cyclization
forming the less sterically encumbered Gly-Gly linkage.
After several low-yielding attempts using DECP/NaHCO3
in various solvents (entries 1-4), additional coupling
agents such as bis(oxazolidinyl)phosphinyl chloride (BOP-
Cl),38 diphenylphosphoryl azide (DPPA),39 and 1-ethyl-
3-(3-dimethylaminopropyl)carbodiimide hydrochloride salt
(EDCI‚HCl)35 were tried (entries 5-7). While BOP-Cl and
DPPA did not produce favorable results, EDCI‚HCl gave
promising but variable yields. Et3N was added as base
(entry 8), and eventually toluene was added to the solvent
system in the hope that greater solvent hydrophobicity
would aid cyclization (entries 9 and 10). To date, the
combination of DECP (EDCI‚HCl is not soluble in
toluene) and Et3N in the mixture of DMF/toluene (1:1)
has consistently provided the final macrocycles in the
best yield. It is worth noting that none of the cyclizations
carried out with the monosubstituted linkers discussed
herein have proved as efficient as cyclizations with either
dimethyl-substituted Aca linkers or more highly substi-
tuted dipeptide units.
uration. Although the double bond of 15 could be reduced
or alternatively modified at this stage, this material was
directly used in the ensuing macrocycle synthesis to
optimize synthetic efficiency (see Scheme 9 below).
The key step in the synthesis of the 2-benzylated linker
was an asymmetric alkylation reaction (Scheme 6). Thus,
6-bromohexanoyl chloride was coupled to (S)-(-)-4-ben-
zyl-2-oxazolidinone to form 16. After displacement of the
bromide with sodium azide, 17 was alkylated with benzyl
bromide to form 18. The azido acid 19 was isolated after
removal of the auxiliary using standard conditions and
brought to the next phase of the synthesis.34
P ep tid e Cou p lin g a n d Cycliza tion . The Aca linkers
were attached to the N- or C-terminally protected gly-
cylglycine peptide units under standard conditions.35
Cyclizations were carried out using diethyl cyanophos-
phonate (DECP)36 and triethylamine (Et3N) in a DMF/
toluene (1:1) mixture at a peptide concentration of
approximately 0.005 M. Scheme 7 illustrates how the
azido acid 19 was coupled to glycylglycine ethyl ester,
affording the tripeptide 20. Subsequent reduction of the
azide to the Boc-protected amine37 afforded 21. The short-
lived Boc-protection aided in the handling of the inter-
mediates leading to the final target. Compound 2 was
isolated after hydrolysis of the ester, Boc deprotection,
and peptide cyclization.
Con for m a tion a l An a lysis. Conformational analysis
of macrocycles 2-6 in solution were performed using
NMR and CD spectroscopy. As previously shown pru-
dent,21 NMR COSY and ROESY spectra were obtained
using samples dissolved in DMSO-d6 at approximately
1 mg/mL to minimize aggregation. In general, the NMR
Sch em e 7

2640 J . Org. Chem., Vol. 66, No. 8, 2001
MacDonald et al.
Sch em e 8
Sch em e 9
Ta ble 1. Con d ition s for Cycliza tion Rea ction s
seemed unlikely due to a lack of NOE interactions
between Gly1NH and Gly2NH, which are indicative of
such conformations. On the basis of these results, each
of the macrocycles was believed to largely reside in some
form of a type II â-turn (although other conformations
are presumably kinetically accessible by these flexible
compounds). However, because the Gly-Gly dipeptide
does not contain a stereogenic center, these data do not
indicate which enantiomorphic backbone is most preva-
lent (âII or âII′).
entry
conditions
yield (%)
1
2
3
4
5
DECP, NaHCO3, DMF
-
-
-
-
DECP, NaHCO3, DMF/CH2Cl2
DECP, NaHCO3, CH3CN
DECP, NaHCO3, CH2Cl2
EDCI‚HCl, NaHCO3, DMF
DPPA, NaHCO3, DMF
There was some hope that the conformation of the Aca
tether could be determined from NMR studies to see if a
correlation between tether and turn existed. In the
research preceding this project, it was noted that one of
the C4 protons of Aca had an interaction with the AcaNH
in each of the macrocycles incorporating the disubstituted
constraint.21 The central methylene protons were well
5-30
15
6
7
8
9
10
BOP-Cl, NaHCO3, DMF/CH2Cl2
EDCI‚HCl, Et3N, DMF
EDCI‚HCl, Et3N, DMF/toluene
DECP, Et3N, DMF/toluene
-
13
-
23-26
data were consistent with either a type II or II′ â-turn
conformation. This conclusion was primarily based on the
presence of ROE cross-peaks between Gly2NH and one
of the prochiral Gly1HR protons (specific for type II
â-turns) and between the Gly2NH and AcaNH hydrogen
atoms (observed in both type I and II â-turns). A
significant concentration of the type I or type I′ turns
1
separated in the H NMR spectra with ranges of ∆ν )
0.17-0.81 ppm. In each of the examples mentioned here,
the C-4 protons were not well-resolved, the greatest
difference being ∆ν ) 0.14 ppm. There were no obvious
ROEs observed between either of the C4 protons and
AcaNH. In general, there was significant averaging of

Aminocaproic Acid-Cyclized Dipeptides
J . Org. Chem., Vol. 66, No. 8, 2001 2641
F igu r e 1. CD spectra of macrocycles (a) 2 (solid line) and 3 (dotted line), (b) 5 (solid line) and 6 (dotted line), and (c) 4.
Ta ble 2. Selected Dih ed r a l An gles for 6 a n d 4 (d eg)
â-turn
type Ia
type II
type II′
φi+1
ψi+1
φi+2
ψi+2
-60
-60
60
71
-54
104
-30
120
-120
-115
53
-90
80
-80
-67
-80
68
0
0
0
6
-26
94
38
4 (conformer 1)
4 (conformer 2)
-77
a
Idealized values (ref 1).
II′ â-turn in agreement with the CD and NMR data. The
observed dihedral angles obtained from the X-ray data
are all within 26° of the idealized type II′ values shown
in Table 2.1 The X-ray structure also revealed the
presence of a hydrogen bond between the Aca(CdO) and
Gly1NH (O-H distance 1.892 Å). The position of the
benzyl substituent on Aca appears to occupy an equato-
rial-like position of the macrocycle.
F igu r e 2. Ball-and-stick representations of the X-ray crystal-
lographic structures of compounds (a) 6 and (b) 4.
the chemical shifts for the protons on the tether, which
indicated flexibility within that region of the macrocycle.
Circular dichroism was used to obtain additional
solution-phase conformational information and, in par-
ticular, to differentiate between type II and II′ turns for
each compound. All measurements were performed in
methanol (c ) 1 mg/mL, 0.02 cm path length). The CD
spectra for compounds 5 and 6 (Figure 1b) resemble the
standard curve a type II′ â-turn.40,41 Each curve has a
minimum ellipticity between 195 and 200 nm and a
maximum in the region of 220-230 nm. Conversely, the
CD spectra for compounds 2 and 3 (Figure 1a) indicate
a preferred conformation of a type II â-turn in that they
both contain maxima near 200 nm and minima at 220
nm. Compound 4 was the outlier in this series (Figure
1c). The maximum at 200 nm is much more intense than
any of its isomers, and a modest minimum is noted near
230 nm. This CD data is somewhat reminiscent of that
expected for a type II turn; indeed, the NMR data would
seem to suggest that at least some of this material exists
as a type II turn in solution. However, the amplitude of
this curve, when combined with the lack of a well-defined
minimum at 220 nm, suggested that the âII turn makes
up only a small component of the solution structure.
In addition, X-ray structures were obtained for com-
pounds 4 and 6 following crystallization from methanol/
CHCl3 and methanol/CH2Cl2, respectively (Figure 2;
stereoviews of these structures can be found in the
Supporting Information). Interestingly, the racemate of
6 crystallized from the 93% ee mixture obtained as
described above, as evidenced by the space group P21/a.
The crystal structure of 6 clearly demonstrates a type
Crystallographic analysis of 4 showed the presence of
two distinct conformers, neither of which resemble a
â-turn conformation. For example, the Aca(CdO)/AcaNH
distance in this compound is 5.03-5.55 Å (with the
carbonyl groups and the N-H bond pointing in opposite
directions) compared to 1.89 Å for 6. In both conforma-
tions of 4, all of the amide bonds approximately describe
a ring, or crown, structure about the plane of the
macrocycle, with each of the carbonyl oxygens pointing
in the same direction. In addition, the Aca linker has
been significantly skewed from conformations seen in
other examples. In this case, the 4-benzyl group appears
to be placed into a pseudoaxial position, with the adjacent
methine proton pointing into the macrocycle cavity. It is
likely that at least some of these observations result from
crystal-packing considerations. Overall, considering the
proton NMR data (which indicate some âII or -II′
character), the CD (which shows significant deviations
from a “pure” type II â-turn spectrum), and these X-ray
results, it would appear that this single substitution type
among all of those examined herein is least likely to adopt
one of the classically recognized â-turn subtypes.
Su m m a r y
Cyclization of a dipeptide with Aca constrains the
dipeptide into a â-turn with the amino acid residues
occupying the i + 1 and i + 2 positions. The use of a
monosubstituted Aca tether to constrain glycylglycine has
been shown to vary the â-turn conformation from type
II to type II′. However, placement of the substituent in
the central C-4 position of the tether does not lead to a
macrocycle corresponding to a known turn type. We are
(40) Woody, R. W. In The Peptides; Hruby, V. J ., Ed.; Academic: New
York, 1985; Vol. 7, pp 15-114.
(41) Venkatachalam, C. M.; Ramachandran, G. M. Annu. Rev.
Biochem. 1969, 38, 45-82.

2642 J . Org. Chem., Vol. 66, No. 8, 2001
MacDonald et al.
2.13 (m, 1H), 2.22 (m, 1H), 2.37 (dd, J ) 8.0, 13.6 Hz, 1H),
2.63 (dd, J ) 5.4, 13.7 Hz, 1H), 2.80 (m, 1H), 3.16 (m, 1H),
3.42 (m, 2H), 3.72 (dd, J ) 6.9, 16.3 Hz, 1H), 3.86 (dd, J )
6.0, 13.3 Hz, 1H), 7.04 (dd, J ) 3.4, 7.6 Hz, 1H), 7.14 (m, 1H),
7.21 (m, 4H), 8.69 (t, J ) 6.0 Hz, 1H), 8.91 (t, J ) 5.9 Hz, 1H);
13C NMR (100.6 MHz, DMSO-d6) δ 28.5, 32.1, 33.0, 33.4, 35.8,
41.5, 43.9, 44.7, 126.5, 128.8, 130.2, 140.8, 169.2, 170.7, 176.6;
MS (CI) m/e 318 (M+ + 1), 91; HRMS calcd for C17H24N3O3
(M+ + 1): 318.1817; found 318.1814.
currently applying this methodology toward the synthesis
of biologically relevant â-turn mimics.
Exp er im en ta l Section
Gen er a l m eth od s have been previously published.21
Gen er a l P r oced u r e for Ma cr ocycliza tion . The Boc-
protected tripeptide was dissolved in a mixture of TFA/CH2-
Cl2 (1:2) and stirred at room temperature for 1 h. The solvent
was removed in vacuo, and the residue was redissolved in a
1:1 mixture of DMF/toluene (0.005 M). DECP (5 equiv) and
enough Et3N to make the solution slightly basic were added
to the solution, which was stirred at room temperature for 19
h. After removal of the solvent in vacuo, the residue was
redissolved in CH2Cl2 and washed with 10% citric acid and
brine. The organic phase was dried over Na2SO4 and concen-
trated. The crude product was purified by column chromatog-
raphy (MeOH/CH2Cl2 ) 1:50, 1:25) to obtain the cyclized
peptide.
(12R)-12-Ben zyl-1,4,7-t r ia za cyclot r id eca n e-2,5,8-t r i-
on e (5). White solid; mp 189-190 °C; [R]D +26 (c 0.05, MeOH);
1
IR (film) 3325, 3068, 1647 cm-1; H NMR (400 MHz, DMSO-
d6) δ 1.05 (m, 2H), 1.49 (m, 1H), 1.65 (m, 1H), 1.81 (m, 1H),
2.10 (m, 2H), 2.41 (d, J ) 7.3 Hz, 1H), 2.89 (m, 1H), 3.15 (m,
1H), 3.52 (dd, J ) 5.6, 13.4 Hz, 1H), 3.58 (d, J ) 6.1 Hz, 2H),
3.70 (dd, J ) 6.3, 13.4 Hz, 1H), 7.01 (t, J ) 5.9 Hz, 1H), 7.19
(m, 3H), 7.28 (m, 2H), 8.63 (t, J ) 6.1, 1H), 8.94 (t, J ) 5.9
Hz, 1H); 13C NMR (100.6 MHz, MeOH-d4) δ 18.7, 24.9, 33.0,
36.8, 38.1, 39.3, 41.9, 43.0, 124.5, 126.8, 127.3, 139.1, 169.0,
169.9, 176.5; MS (CI) m/e 318 (M+ + 1), 204, 128, 91, 60, 44;
(9R)-9-Ben zyl-1,4,7-tr iazacyclotr idecan e-2,5,8-tr ion e (2).
White solid; mp 249 °C dec; [R]D -45 (c 0.06, MeOH); IR (film)
HRMS calcd for
318.1793.
C
17H24N3O3 (M+ + 1): 318.1817; found
3262, 1647 cm-1 1H NMR (400 MHz, DMSO-d6) δ 1.03 (m,
;
1H), 1.16 (m, 2H), 1.32 (m, 1H), 1.50 (m, 2H), 2.55 (m, 2H),
2.76 (m, 1H), 2.88 (m, 1H), 3.19 (dd, J ) 4.9, 13.3 Hz, 1H),
3.29 (obscured by solvent, 1H), 3.45 (dd, J ) 5.8, 15.7 Hz, 1H),
3.65 (dd, J ) 6.5, 15.7 Hz, 1H), 3.91 (dd, J ) 7.2, 13.3 Hz,
1H), 6.94 (t, J ) 5.8 Hz, 1H), 7.15 (m, 3H), 7.25 (m, 2H), 8.57
(t, J ) 5.9 Hz, 1H), 8.73 (t, J ) 5.9 Hz, 1H); 13C NMR (125.6
MHz, DMSO-d6) δ 20.9, 26.7, 30.0, 36.0, 38.8, 43.9, 47.5, 126.4,
128.6, 129.1, 140.1, 168.9, 170.1, 177.2; MS (CI) m/e 318 (M+
+ 1), 317 (M+), 147, 74; HRMS calcd for C17H24N3O3 (M+ + 1):
318.1818; found 318.1823.
(13R)-13-Ben zyl-1,4,7-t r ia za cyclot r id eca n e-2,5,8-t r i-
on e (6). White solid; mp 287 °C dec; [R]D +9.3 (c 0.11, MeOH);
IR (film) 3281, 3087, 2951, 1658, 1635 cm-1 1H NMR (400
;
MHz, DMSO-d6) δ 1.11 (m, 2H), 1.37 (m, 3H), 1.65 (m, 1H),
2.13 (m, 2H), 2.54 (dd, J ) 7.7, 13.3 Hz, 1H), 2.80 (dd, J )
6.0, 13.3 Hz, 1H), 3.54 (m, 1H), 3.59 (m, 1H), 3.64 (m, 3H),
7.03 (d, J ) 8.0 Hz, 1H), 7.18 (m, 3H), 7.28 (m, 2H), 8.24 (t, J
) 5.9 Hz, 1H), 8.55 (t, J ) 6.0 Hz, 1H); 13C NMR (100.6 MHz,
DMSO-d6) δ 23.1, 25.0, 32.2, 34.8, 42.2, 44.4, 44.7, 51.3, 126.9,
129.0, 129.9, 139.7, 169.7, 170.4, 174.4; MS (CI) m/e 318 (M+
+ 1), 226, 128, 91; HRMS calcd for C17H24N3O3 (M+ + 1):
318.1817; found 318.1815.
(10S)-10-Ben zyl-1,4,7-t r ia za cyclot r id eca n e-2,5,8-t r i-
on e (3). White solid; mp 238 °C dec; [R]D +26 (c 0.05, MeOH);
1
IR (film) 3325, 3068, 1647 cm-1; H NMR (400 MHz, DMSO-
d6) δ 1.07 (m, 1H), 1.21 (m, 2H), 1.43 (m, 1H), 1.92 (m, 1H),
2.06 (dd, J ) 3.7, 13.2 Hz, 1H), 2.19 (m, 1H), 2.39 (dd, J )
8.6, 13.2 Hz, 1H), 2.69 (dd, J ) 5.5, 13.2 Hz, 1H), 2.85 (m,
1H), 3.31 (partially obscured by solvent, 2H), 3.47 (dd, J )
5.9, 16.1 Hz, 1H), 3.65 (dd, J ) 6.6, 13.5 Hz, 1H), 3.85 (dd, J
) 6.6, 13.5 Hz, 1H), 7.04 (m, 1H), 7.2 (m, 3H), 7.29 (m, 2H),
8.69 (t, J ) 6.1 Hz, 1H), 8.94 (t, J ) 6.0 Hz, 1H); 13C NMR
(125.6 MHz, DMSO-d6) δ 25.1, 27.8, 36.6, 36.7, 41.3, 44.3, 44.7,
126.8, 129.1, 130.0, 141.3, 169.4, 170.9, 175.9, (one carbon
signal obscured by solvent); MS (CI) m/e 318 (M+ + 1), 204,
128, 91, 60; HRMS calcd for C17H24N3O3 (M+ + 1): 318.1817;
found 318.1793.
Ack n ow led gm en t. This work was supported by the
American Heart AssociationsHeartland and the Na-
tional Institutes of Health (GM-49093). M.M. would like
to acknowledge receipt of an NIH Predoctoral Training
Grant (GM-07775). We thank Martha Morton and
Lawrence Seib for expert assistance with NMR and
X-ray work, respectively.
Su p p or tin g In for m a tion Ava ila ble: Experimental pro-
cedures for compounds 7-29, spectral data for compounds
2-29, X-ray crystal reports for compounds 4, 6, 12a , and 12b,
and pertinent ROE data. This material is available free of
charge via the Internet at http://pubs.acs.org.
(11R)-11-Ben zyl-1,4,7-t r ia za cyclot r id eca n e-2,5,8-t r i-
on e (4). White solid; mp 193-194 °C; [R]D +98 (c 0.07, MeOH);
1
IR (film) 3291, 1654, 1635 cm-1; H NMR (400 MHz, DMSO-
d6) δ 1.08 (m, 1H), 1.27 (m, 2H), 1.49 (m, 1H), 1.78 (m, 1H),
J O001308B