Application of ring-closing metathesis to the synthesis of rigidified amino acids and peptides

Article abstract of DOI:10.1021/ja961626l

Ruthenium complexes 1a and 1b have been applied to the ring-closing metathesis (RCM) reactions of a number of dienic substrates. The substrate scope includes rings of 6 to 20 members. In addressing macrocyclic peptides, a class of tetrapeptide disulfides

Full text of DOI:10.1021/ja961626l

9606
J. Am. Chem. Soc. 1996, 118, 9606-9614
Application of Ring-Closing Metathesis to the Synthesis of
Rigidified Amino Acids and Peptides
Scott J. Miller, Helen E. Blackwell, and Robert H. Grubbs*
Contribution from Arnold and Mabel Beckman Laboratories of Chemical Synthesis,
DiVision of Chemistry and Chemical Engineering, California Institute of Technology,
Pasadena, California 91125
ReceiVed May 15, 1996X
Abstract: Ruthenium complexes 1a and 1b have been applied to the ring-closing metathesis (RCM) reactions of a
number of dienic substrates. The substrate scope includes rings of 6 to 20 members. In addressing macrocyclic
peptides, a class of tetrapeptide disulfides inspired the synthesis of the carbon-carbon bond analogs. Replacement
of cysteine residues with allylglycines resulted in the acyclic precursors which were subjected to RCM to afford the
corresponding macrocycles. In addition, several macrocycles were prepared which were not based upon disulfide-
bridge-containing species found in nature. The method was found to function on dienic peptides which were either
dissolved in organic solvents or bound to solid supports.
Introduction
cyclic amino acids and peptides, containing multiple hetero-
atoms, acidic protons, and hydrogen bonds.5 In addition,
Ghadiri and co-workers have recently demonstrated that these
structural features can be engineered into systems to allow for
efficient cross-metathesis processes, enabling efficient dimer-
ization and template driven inter- and intramolecular olefin
metathesis.6 In this paper, we describe our findings concerning
the scope of RCM in the synthesis of cyclic amino acids and
peptides.
Given the prominent role of conformationally restricted amino
acids and peptides in the design of peptidomimetics, we sought
to apply the functional group tolerant catalysts 1 to the synthesis
of this class of molecules.7 Many approaches to the synthesis
of a vast array of these compounds have been described. Amide
linkage modification, side chain modification, and de noVo
synthesis of particular structural motifs are among the strategies
employed to date. Cyclic substructures are likewise frequently
incorporated since these motifs serve to restrict the conforma-
tional space of the molecules. Such modifications often result
in increased affinity for a given biological receptor, with
simultaneously diminished sensitivity to cellular peptidases.
Because RCM based on catalysts 1a and 1b was emerging as
a powerful method for carbocycle and heterocycle synthesis in
Previous reports from these laboratories have demonstrated
that ruthenium complexes 1a and 1b efficiently catalyze ring-
closing metathesis (RCM) reactions to form five-, six-, seven-,
and eight-membered carbocycles and heterocycles (eq 1).1-3
Metathesis-based strategies for carbocycle and heterocycle
synthesis, based on alkylidenes of ruthenium and molybdenum,
are now being applied with some regularity in natural products
synthesis.4 The extraordinary functional group tolerance of the
ruthenium-based catalysts 1 has also enabled the synthesis of
X Abstract published in AdVance ACS Abstracts, September 1, 1996.
(1) For a recent reviews on applications of olefin metathesis and related
processes in organic synthesis, see: (a) Grubbs, R. H.; Pine, S. H. In
ComprehensiVe Organic Synthesis; Trost, B. M., Ed.; Pergamon: New York,
1991; Vol. 5, Chapter 9.3. (b) Grubbs, R. H.; Miller, S. J.; Fu, G. C. Acc.
Chem. Res. 1995, 28, 446-552. (c) Schmalz, H.-G. Angew. Chem., Int.
Ed. Engl. 1995, 34, 1833-1836.
(4) For recent applications of metathesis-based methods to natural product
synthesis, see: (a) Martin, S. F.; Liao, Y.; Rein, T. Tetrahedron Lett. 1994,
35, 691-694. (b) Borer, B. C.; Deerenberg, S.; Bieraugel, H.; Pandit, U.
K. Tetrahedron Lett. 1994, 35, 3191-3194. (c) Morken, J. P.; Didiuk, M.
T.; Visser, M. S.; Hoveyda, A. H. J. Am. Chem. Soc. 1994, 116, 3123-
3124. (d) Martin, S. F.; Wagman, A. S. Tetrahedron Lett. 1995, 36, 1169-
1170. (e) Houri, A. F.; Xu, Z.; Cogan, D.; Hoveyda, A. H. J. Am. Chem.
Soc. 1995, 117, 2943-2944. (f) Huwe, C. M.; Blechert, S. Tetrahedron
Lett. 1995, 36, 1621-1624. (g) Overkleeft, H. S.; Pandit, U. K. Tetrahedron
Lett. 1996, 37, 547-550. (h) Randall, M. L.; Tallarico, J. A.; Snapper, M.
L. J. Am. Chem. Soc. 1995, 117, 9610-9611. (i) Huwe, C. M.; Kiehl, O.
C.; Blechert, S. Synlett 1996, 65-66. (j) Fu¨rstner, A.; Langemann, K. J.
Org. Chem. 1996, 61, 3942-3943.
(5) (a) Miller, S. J.; Grubbs, R. H. J. Am. Chem. Soc. 1995, 117, 5855-
5856. (b) Garro-He´llion, F.; Guibe´, F. J. Chem. Soc., Chem. Commun. 1996,
641-642.
(6) Clark, T. D.; Ghadiri, M. R. J. Am. Chem. Soc. 1995, 117, 12364-
12365.
(7) For an overview of the field, see: (a) Gante, J. Angew. Chem., Int.
Ed. Engl. 1994, 33, 1699-1720. (b) Tetrahedron Symposia-in-Print Number
50, M. Kahn, Ed.; 1993; Vol. 49, number 17. (c) Giannis, A.; Kolter, T.
Angew. Chem., Int. Ed. Engl. 1993, 32, 1244-1267. (d) Hruby, V. J.; Al-
Obeidi, F.; Kazmierski, W. Biochem. J. 1990, 268, 249-262. (e) Toniolo,
C. Int. J. Pept. Protein Res. 1990, 35, 287-300.
(2) For previous reports on RCM from this laboratory, see: (a) Fu, G.
C.; Grubbs, R. H. J. Am. Chem. Soc. 1992, 114, 5426-5427. (b) Fu, G. C.;
Grubbs, R. H. J. Am. Chem. Soc. 1992, 114, 7324-7325. (c) Fu, G. C.;
Grubbs, R. H. J. Am. Chem. Soc. 1993, 115, 3800-3801. (d) Fu, G. C.;
Nguyen, S. T.; Grubbs, R. H. J. Am. Chem. Soc. 1993, 115, 9856-9857.
(e) Fujimura, O.; Fu, G. C.; Grubbs, R. H. J. Org. Chem. 1994, 59, 4029-
4031. (f) Kim, S. H.; Bowden, N.; Grubbs, R. H. J. Am. Chem. Soc. 1994,
115, 10680-10681. (g) Miller, S. J.; Kim, S. H.; Chen, Z.-R.; Grubbs, R.
H. J. Am. Chem. Soc. 1995, 117, 2108-2109. (h) Coates, G. W.; Grubbs,
R. H. J. Am. Chem. Soc. 1996, 118, 229-230. (i) Fujimura, O.; Grubbs, R.
G. J. Am. Chem. Soc. 1996, 118, 2499-2500. (j) Kim, S. H.; Zuercher, W.
J.; Bowden, N. B.; Grubbs, R. H. J. Org. Chem. 1996, 61, 1073-1081.
(3) For the preparation and characterization of catalyst 1a, see: (a)
Nguyen, S. T.; Johnson, L. K.; Grubbs, R. H.; Ziller, J. W. J. Am. Chem.
Soc. 1992, 114, 3974-3975. (b) Nguyen, S. T.; Grubbs, R. H.; Ziller, J.
W. J. Am. Chem. Soc. 1993, 115, 9858-9859. For the preparation and
characterization of catalyst 1b, see: (c) Schwab, P.; Grubbs, R. H.; Ziller,
J. W. J. Am. Chem. Soc. 1996, 118, 100-110. (d) Schwab, P.; France, M.
B.; Ziller, J. W.; Grubbs, R. H. Angew. Chem., Int. Ed. Engl. 1995, 34,
2039-2041.
S0002-7863(96)01626-5 CCC: $12.00 © 1996 American Chemical Society

Ring-Closing Metathesis
J. Am. Chem. Soc., Vol. 118, No. 40, 1996 9607
the presence of a wide range of functional groups, we
endeavored to apply this strategy to the synthesis of peptide-
based structures.
by the ring strain inherent to the cyclic product, which requires
the reactions be run at high dilution to minimize competing
intermolecular oligomerization processes.11
A notable limitation of the RCM strategy employing 1 for
the synthesis of cyclic amino acids is illustrated by our
unsuccessful attempt to prepare the dehydroproline 9 by this
method (Scheme 2). Initially, the vinyl glycine derivative 8
was projected to undergo facile, rapid RCM with 1a to afford
the 5-membered ring in analogy to the other 5-membered rings
we had studied.12 However, upon exposure to the ruthenium-
based catalyst 1a, only acyclic, R,â-unsaturated esters were
recovered. Although no rigorous mechanistic studies were
carried out, we attribute this undesired reactivity to the labile
acidic CR proton associated with the vinylglycine structure. The
synthesis of the dehydroproline derivative was not pursued
further.
Results and Discussion
Cyclic Amino Acids. Our initial objective was the synthesis
of simple amino acid derivatives containing various ring sizes
which might subsequently be introduced into peptides. Depend-
ing on the desired ring size, our approach required the
introduction of alkene functionality at either CR or at the amide
nitrogen of the amino acid. Alkene incorporation at CR is
facilitated by the commercial availability of (+/-)-, (+)-, and
(-)-allylglycine. In addition, highly efficient methodology for
the asymmetric synthesis of this unnatural amino acid is also
available.8 Introduction of alkene functionality at the amide
nitrogens has also been addressed. In particular, Seebach has
shown that peptides can be converted to the poly(N-allylamides)
upon exposure to allyl bromide in the presence of the P4-
phosphazene base.9 Synthesis of the RCM precursors 2-4 then
proceeded in a straightforward fashion employing conventional
peptide-coupling and derivatization techniques (see Experimen-
tal Section).
Scheme 2
The results of the RCM experiments are shown in Scheme
1.10 Treatment of modified amino acid 2 with catalyst 1a under
conditions analogous to those described in our earlier reports
(5 mol % 1a, 0.20 M, CHCl3, 25 °C) furnished the dehydro-
pipicolinate 5 in good yield (91%) within 1 h. In contrast,
substrates 3 and 4 required more vigorous conditions for ring
closure, and the isolated yields for the corresponding cyclizations
were reduced. Nevertheless, subjection of acyclic dienes 3 and
4 to the above RCM conditions afforded the seven-membered
ring 6 in 52% yield and the eight-membered ring 7 in 51% yield,
respectively. The latter two transformations appear to be limited
Macrocyclic Peptides. At this stage, we turned our attention
to the synthesis of macrocyclic peptide structures. Our initial
approach to the problem was based on the incorporation of
alkene functionality at the amide nitrogens (Scheme 3). Syn-
thesis of N,N,N-(triallyl)peptide 10 was accomplished by
conventional peptide-coupling chemistry, followed by allylation
Scheme 1
Scheme 3
(8) Myers, A. G.; Gleason, J. L.; Yoon, T. Y. J. Am. Chem. Soc. 1995,
117, 8488-8489 and references therein.
(9) Pietzonka, T.; Seebach, D. Angew. Chem. Int. Ed. Engl. 1992, 31,
1481-1482.
(11) We have discussed this issue in more detail in ref 2g.
(12) Vinyl glycine derivative 8 was prepared by derivatization of N-allyl-
N-BOC-methionine methyl ester in analogy to the procedure of Rapoport,
see: (a) Afzali-Ardakani, A.; Rapoport, H. J. Org. Chem. 1980, 45, 4817-
4820. (b) Meffre, P.; Voquang, L.; Voquang, Y.; Le Goffic, F. Synth.
Commun. 1989, 19, 3457-3468.
(10) Compounds were identified on the basis of their 1H NMR, 13C NMR,
IR, and mass spectral characteristics (see Experimetal Section). The cases
involving C-allylglycine in Scheme 1 were performed on the racemic
mixtures.

9608 J. Am. Chem. Soc., Vol. 118, No. 40, 1996
Miller et al.
Figure 1. Naturally occurring disulfide-stabilized â-turns.
with the P4-phosphazene base. Our initial projection was that
the substrate, upon treatment with the metathesis catalyst, would
undergo a bicyclization reaction wherein a rapid six-membered
ring closure would be followed by a macrocyclization. How-
ever, treatment of 10 with 1a resulted in only monocyclization
and no macrocyclic compounds were obtained.13 While the
reasons for this lack of macrocyclization were not definitively
established, we attributed the low reactivity of the structure to
either the ring strain inherent to the bicyclic product or kinetic
problems associated with its formation. As a result, we sought
an alternative approach to macrocyclic peptide formation.
Balaram and co-workers have reported the disulfide-stabilized
â-turn 11 as an analog of the redox active â-turns above, and
this substrate stimulated us to attempt to prepare the analogous
tetrapeptide olefin (Figure 2).18 Significantly, Balaram has
shown that both 11 and the corresponding acyclic bis(benzyl)
thioether 12 possess the illustrated hydrogen bond in CHCl3
solvent, and we felt that this feature might help dispose the
acyclic diene analog toward a conformation suitable for ring
closure.
Dicarba Analogs of Naturally Occurring Macrocyclic
Disulfides. In order to contend with the potential thermody-
namic and kinetic problems associated with peptide macrocy-
clizations, we turned our attention to examples of macrocyclic
peptides which were found in nature. While examples of such
peptides are numerous, we were attracted to a class of disulfide-
stabilized tetrapeptides found in a number of redox active
proteins (Figure 1).14 Importantly, each member of this class
contains a disulfide bridge which locks the tetrapeptide into a
â-turn type structure.15
Our objective was to replace the disulfide bridges in these
known systems with carbon-carbon bonds. So-called “dicarba”
peptides of this nature have themselves been prepared previously
in peptidomimetic research.16 Replacement of the disulfide
bridge in biologically active compounds can provide drugs
which have increased metabolic stability, as the covalent cross-
link is inert to conditions which can reduce disulfide-based
cross-links.17 Our approach was to design a laboratory model
for the oxidation of cysteines to disulfide bridges. Replacement
of cysteines with allylglycines in acyclic peptides sets up RCM
instead of thiol oxidation. In the presence of a metathesis
catalyst, a carbon-carbon bond is then formed in place of the
disulfide bridge.
Figure 2. Balaram’s disulfide-stabilized â-turn abd tge CdC bond
analog.
Since disulfide bridges possess different dihedral angle
requirements relative to olefins,19 it was not obvious at the outset
of our study whether (S)-allylglycine would be the optimal
stereoisomer for replacement of the L-cysteine. Therefore, in
our initial study of this system, we prepared a statistical mixture
of the four stereoisomers of the acyclic dienes 13. As reported
in our earlier communication, when this mixture of the four
diastereomers was treated with catalyst 1a (20 mol %, 0.002
M, CH2Cl2, 40 °C), a single macrocycle (14 diastereomer was
obtained. The majority of the reaction mixture was composed
of unreacted dienes. When diastereomerically pure (S,S,S)-13
was subjected to the reaction conditions, (S,S,S)-14 was obtained
in 60% yield and the product was identical to that obtained from
the analogous experiment on the mixture (eq 2).
(13) A monocyclized product was identified by FABMS. No ions
corresponding to the bicyclization product could be detected. The mono-
cyclized product was formed at approximately the same rate as the six-
membered cyclic product 5, and no other product was formed after elongated
reaction time. The formation of cyclic product 5 was previously found to
be fast and effectively irreversible. Thus, we believe that treatment of
substrate 10 under the analogous reaction conditions leads to the six-
membered monocyclization product as opposed to the 11-membered
monocyclization product.
In order to explore the scope of the process, and in order to
assay the role of preorganization in facilitating ring closure, we
prepared additional tetrapeptides where we systematically
removed the conformationally constrained amino acids proline
and aminoisobutyric acid (Aib). The Pro-Aib sequence is
(14) (a) Musiol, H.-J.; Siedler, F.; Quarzago, D.; Moroder, L. Biopolymers
1994, 34, 1553-1562. (b) Siedler, F.; Quarzago, D.; Rudolph-Bohner, S.;
Moroder, L. Biopolymers 1994, 34, 1563-1572.
(15) For discussions of â-turns in proteins, see: (a) Rizo, J.; Gierasch,
L. M. Ann. ReV. Biochem. 1992, 61, 387-418. (b) Rose, G. D.; Gierasch,
L. M.; Smith, J. A. AdV. Protein Chem. 1985, 37, 1-109. For a review of
â-turn mimetics, see: ref 7b.
(16) Nutt, R. F.; Strachan, R. G.; Veber, D. F.; Holly, F. W. J. Org.
Chem. Am. Chem. Soc. 1980, 45, 3078-3080.
(17) Nutt, R. F.; Veber, D. F.; Saperstein, R. J. Am. Chem. Soc. 1980,
102, 6539-6545.
(18) Ravi, A.; Balaram, P. Tetrahedron 1984, 40, 2577-2583.
(19) The disulfide dihedral angle in 10 has been shown to be 82° with
right-handed chirality in the solid state. Ravi. A.; Prasad, B. V.; Balaram,
P. J. Am. Chem. Soc. 1983, 105, 105-109.

Ring-Closing Metathesis
J. Am. Chem. Soc., Vol. 118, No. 40, 1996 9609
macrocyclization to afford cyclic tetrapeptide 18 (eq 4). These
known to restrict the conformational space of peptides.20
Therefore, we examined substrates where these amino acids
were replaced with less rigidifying residues.
Replacement of the Aib residue with protected tyrosine
resulted in the formation of peptide 15. The Pro-Tyr dipeptide
sequence spans the cysteine residues in the glutaredoxin active
site. Compound 15 therefore represented an attempt to syn-
thesize a carbon-carbon bond mimic of the active site of this
protein. Exposure of 15 to the ruthenium catalyst 1b under
our standard macrocyclization conditions (0.004 M, CH2Cl2, 40
°C) resulted in clean formation of the macrocycle 16 in 80%
yield (eq 3).21 This experiment demonstrates that two consecu-
results establish that there is no rigorous requirement for the
Pro-Aib sequence as the bridging residues for the synthesis of
tetrapeptide macrocycles by RCM.
Conformational Analysis of Cyclic Dicarbatetrapeptides.
In order to ascertain the extent to which the macrocyclic
products containing carbon-carbon double bonds resemble their
disulfide bridge analogs, an investigation of the solution
conformations of the cyclic tetrapeptides was undertaken. As
indicated above, Balaram’s Cys-Pro-Aib-Cys system 11 was
found by a series of IR and NMR experiments to possess an
intramolecular, transannular H-bond. The corresponding data
on the carbon-carbon bond analogs of these systems is
discussed below.
Table 1 lists the NH region of the IR spectra of the cyclic
peptides 14, 16, and 18. In each case, the characteristic bands
for both non-hydrogen-bonded NH groups (>3400 cm-1) and
intramolecular CdO‚‚‚H-N hydrogen bonds appear (<3400
cm-1).22 Although it is difficult to definitively establish from
the IR data alone which of the amide N-H groups is involved
in intramolecular hydrogen bonding, these data are consistent
with the possibility of transannular intramolecular H-bonding
in the cyclic structures.
Table 1. IR Data (cm-1) for Cyclic Tetrapeptides (Concentration
∼0.002 M, CH2Cl2)
14
16
18
3424
3324
3426
3354
3419
3339
tive conformationally constrained amino acids are not required
for the synthesis of tetrapeptide macrocycles by RCM.
Replacement of both the proline and the Aib residues with
leucines resulted in peptide 17, which is devoid of conforma-
tionally restricted amino acids in the bridging positions. Once
again, exposure to the catalyst 1b resulted in very efficient
To further understand the H-bonding network in the cyclic
tetrapeptides, we turned to H NMR techniques (Figure 3).
Compound 14 was studied to identify the presence of any
transannular hydrogen bonding. The participation of specific
NH groups in hydrogen bonds was established by examining
the temperature dependence of the NH chemical shifts (-∆δ/
1
(20) (a) Venkatachalapathi, Y. V.; Balaram, P. Biopolymers 1981, 20,
1137-1145. (b) Prasad, B. V. V.; Balaram, H.; Balaram, P. Biopolymers
1982, 21, 1261-1273.
(21) A related cyclization was performed with unprotected tyrosine which
afforded the analogous macrocycle in 70% yield. The free phenol therefore
does not interfere with catalyst activity.
(22) For interpretation of N-H stretches in CH2Cl2, see: (a) Gellman,
S. H.; Dado, G. P.; Liang, G.-B.; Adams, B. R. J. Am. Chem. Soc. 1991,
113, 1164-1173. (b) Liang, G.-B.; Desper, J. M.; Gellman, S. H. J. Am.
Chem. Soc. 1993, 115, 925-938. (c) Gardner, R. R.; Liang, G.-B.; Gellman,
S. H. J. Am. Chem. Soc. 1995, 117, 3280-3281. (d) Rao, C. P.; Nagaraj,
R.; Rao, C. N. R.; Balaram, P. Biochemistry 1980, 19, 425-431.

9610 J. Am. Chem. Soc., Vol. 118, No. 40, 1996
Miller et al.
1
Figure 3. (a) Amide NH and olefin region of the 500 MHz H NMR spectrum of 14 in 10% (CD3)2SO/CDCl3. (b) Temperature dependence of
chemical shift for amide protons in (CD3)2SO solution (-∆δ/∆T). (c) Dependence of chemical shift for amide protons in CDCl3 solution with
increasing (CD3)2SO concentration.
∆T) in (CD3)2SO and the solvent dependence of the chemical
Nevertheless, at high dilution (0.002 M) it undergoes successful
RCM to yield macrocyclic peptide 20.
shifts in CDCl3/(CD3)2SO mixtures (Figure 3).23 Proton as-
1
signments are shown on the illustrated H NMR plot showing
the olefinic and amide regions of the spectrum and were
established using conventional decoupling techniques. For
cyclic peptide 14, an intramolecular transannular H-bond was
identified. The temperature dependence (-∆δ/∆T) of the key
allylglycine NH was found to be 0.003 ppm/K. In contrast,
the (-∆δ/∆T) for all the other amide resonances are >0.007
ppm/K, reflecting full solvent exposure. Likewise, in the solvent
titration measurements, the allylglycine NH shows very minor
changes in chemical shift with increasing (CD3)2SO concentra-
tion, while the other amide resonances shift steadily downfield.
These results are very similar to the observations of Balaram
for the disulfide 11, where the illustrated Cys(CdO‚‚‚H-N)-
Cys was established by these methods.24
Additional Examples. In an effort to understand the
generality of the method for introducing carbon-carbon bond
cross-links into peptide systems, we have explored several
examples which are not based on the tetrapeptide framework.
In these cases, we have also studied examples wherein alkene
functionality is not solely based on allylglycine. For example,
the bis(O-allyl)tyrosine dimer 19 undergoes efficient RCM to
produce the 20-membered ring macrocycle 20 in 70% yield.25
It is interesting to note that acyclic peptide 19 has no apparent
structural preorganization by intramolecular hydrogen bonding.
Likewise the serine-allylglycine dipeptide derivative 21 can
be converted to macrocycle 22 by RCM in 56% yield (eq 6).
Interestingly, this system required extremely high-dilution
conditions to minimize dimer formation. When this cyclization
was performed at conventional cyclization concentration (0.002
M), dimeric products were observed. However, further dilution
to the somewhat extreme level of 0.0005 M resulted in the
exclusion of dimer.
Solid Support. Our initial efforts to study the RCM of
alkene-containing peptides wherein we incorporate >5 amino
acid residues have been thwarted by the low solubility of the
(23) See ref 19.
(24) See ref 18.
(25) The olefin geometry of 22 was not assigned.

Ring-Closing Metathesis
J. Am. Chem. Soc., Vol. 118, No. 40, 1996 9611
substrates in the organic solvents where catalysts 1 display their
highest activity. As a result, we have begun to examine the
feasibility of performing the ring-closing reactions on solid-
support-bound peptides.26 To demonstrate the compatibility of
the catalyst with solid phase techniques, we revisited one of
the tetrapeptide macrocycles that was found to undergo efficient
RCM in solution (Scheme 4). We resynthesized the solid-
Scheme 4
Figure 4. (a) HPLC trace of mixture of acyclic and cyclic glutaredoxin
analogs synthesized on solid support. (b) Low-resolution FABMS of
mixture of acyclic and cyclic glutaredoxin analogs.
substrate was cleaved from the resin, HPLC analysis revealed
the presence of two peaks in a 3:2 ratio, the minor peak
corresponding to the acyclic peptide (Figure 4a). Analysis by
FABMS confirmed the identity of the two peaks as the starting
material and the ring-closed product 24 (Figure 4b). This
experiment demonstrates that catalysts 1 react efficiently with
solid-support-bound substrates. We now intend to investigate
the scope of this method for the synthesis of larger conforma-
tionally rigidified systems.
Conclusion
Ruthenium alkylidenes have been shown to demonstrate high
metathesis activity for the RCM of amino acids and peptides
containing olefins. A range of substrates was examined and
ring sizes ranging from 6 to 20 were prepared. The method
was found to be particularly well-suited for the synthesis of
tetrapeptide, carbon-carbon double bond mimics of naturally
occurring tetrapeptide disulfides. Solution conformational
analysis of one of the cyclic tetrapeptides revealed the presence
of intramolecular H-bonding which was analogous to that
reported for the corresponding disulfide-bridged system. In
addition, RCM using ruthenium-based alkylidenes was dem-
onstrated to be compatible with substrates bound to solid
supports. These experiments illustrate a very straightforward
synthetic strategy for the synthesis of conformationally rigidified
cyclic peptides which contain carbon-carbon bond cross-links.
support-bound analog 23 of tetrapeptide 15 using conventional
SPPS (solid phase peptide synthesis) techniques. To effect
RCM, the beads were swelled in CH2Cl2, the solvent of choice
for maximum activity of 1. Catalyst 1b was then introduced,
and the mixture was heated to 40 °C for 22 h, in analogy to the
conditions for the solution phase RCM reaction. When the
(26) For literature on cyclization reactions performed on solid-support-
bound substrates, see: (a) Hiroshige, M.; Hauske, J. R.; Zhou, P. J. Am.
Chem. Soc. 1995, 117, 11590-11591. (b) Andreu, D.; Albericio, F.; Sole,
N. A.; Munson, M. C.; Ferrer, M.; Barany, G. Methods Mol. Biol. (Totowa,
NJ) 1994, 35 (Peptide Synthesis Protocols), 91. (c) Bunin, B. A.; Plunkett,
M. J.; Ellman, J. A. Proc. Natl. Acad. Sci. U.S.A. 1994, 91, 4708-4712.
(d) Virgilio, A. A.; Ellman, J. A. J. Am. Chem. Soc. 1994, 116, 11580-
11581. (e) Zhao, Z.; Felix, A. M. Pept. Res. 1994, 7, 218-223. (f) Albericia,
F.; Hammer R. P.; Garciaechevaerria, C.; Mollins, M. A.; Chang, J. L.;
Munson, M. C.; Pons, M.; Giralt, E.; Barany, G. Int. J. Pept. Protein Res.
1991, 37, 402-413. (g) Apsimon, J. W.; Dixit, D. M. Can. J. Chem. 1982,
60, 368-370.
Experimental Section
General. NMR spectra were recorded on a General Electric QE-
300 or Bruker AM-500 spectrometer. Chemical shifts are reported in
parts per million (ppm) downfield form tetramethylsilane (TMS) with

9612 J. Am. Chem. Soc., Vol. 118, No. 40, 1996
Miller et al.
reference to internal solvent. Multiplicities are abbreviated as fol-
lows: singlet (s), doublet (d), triplet (t), quartet (q), and multiplet (m).
Major rotamer peaks are reported in spectra that are not fully coalesced.
The olefin configurations for 14, 16, 18, and 22 were assigned by
sequential irradiation of the Câ protons and analysis of the resulting
patterns. The notation d(m) refers to the value of the primary J value
(coupling constant) extracted from this analysis. Infrared spectra were
obtained on a Perkin-Elmer 1600 Series FT-IR. Optical rotations were
recorded on a Jasco DIP-181 digital polarimeter at 589 nm and are
reported as [R]λ (concentration in grams/100 mL of solvent). Low-
and high-resolution mass spectra were provided by either the Chemistry
and Biology Mass Spectrometry Facility (Caltech) or the Southern
California Mass Spectrometry Facility (University of California,
Riverside).
Analytical thin-layer chromatography (TLC) was performed using
silica gel 60 F254 precoated plates (0.25 mm thickness) with a
fluorescent indicator. Flash column chromatography was performed
using silica gel 60 (230-400 mesh) from EM Science.27 Catalysts
1a,b were prepared according to the published procedures.28
Peptide Synthesis. Peptides 13, 15, and 17 were synthesized by
standard solution phase peptide-coupling protocols, using N,N-dicy-
clohexylcarbodiimide (DCC)/1-hydroxybenzotriazole (HOBT) as pep-
tide-coupling agent.29
allylglycine (400 mg, 1.86 mmol), DCC (383 mg, 1.86 mmol), and
4-(dimethylamino)pyridine (DMAP) (25 mg, 0.121 mmol). The
mixture was stirred for 1 h as a white precipitate formed. The mixture
was filtered, concentrated, and chromatographed (1.5 cm × 10 cm silica
gel, 20% EtOAc/hexanes to 50% EtOAc/hexane) to afford 566 mg
(93%) of 3 as a clear oil: TLC Rf 0.40 (30% EtOAc/hexane); 1H NMR
(CDCl3, 500 MHz, not coalesced) δ 5.81-5.70 (m, 2H), 5.29-5.04
(m, 5H), 4.65 and 4.40 (q, J ) 8.2 Hz, 1H), 4.21 (d, J ) 7.2 Hz, 1H),
4.05-3.80 (m, 3H), 3.82 (d, J ) 7.2 Hz, 1H), 3.71 and 3.68 (2 × s,
3H), 2.50-2.40 (m, 1H), 2.34-2.30 (m, 1H), 1.38 and 1.37 (2 × s,
9H); 13C NMR (CDCl3, 125 MHz, not coalesced) δ 172.3, 172.1, 169.5,
169.3, 155.2, 155.0, 132.9, 132.6, 132.2, 118.5, 118.4, 118.1, 79.6,
79.5, 52.3, 52.0, 51.2, 49.9, 49.7, 49.4, 48.2, 46.9, 37.6, 37.3, 28.2,
28.1; IR (neat, cm-1) 3320, 2979, 1747, 1713, 1650, 1514, 1454; HRMS
calcd for C16H27N2O5 (MH+) 327.1920, found 327.1914.
Methyl (()-3-[(tert-Butoxycarbonyl)amino]-2,3,4,7-tetrahydro-
2-oxo-1H-azepine-1-acetate (6). To a 50 °C solution of acyclic diene
3 (160 mg, 0.491 mmol) in 60 mL of CHCl3 was added ruthenium
catalyst 1a (29 mg, 0.032 mmol, 5 mol %). The orange-brown solution
was stirred at this temperature for 4 h before it was concentrated and
applied directly to a silica gel column. Chromatography (1.5 cm ×
12 cm silica gel, 25% EtOAc/hexane) afforded 76 mg (52%) of 6 as a
1
clear oil: TLC Rf 0.15 (30% EtOAc/hexane); H NMR (CDCl3, 500
All ring-closing metathesis (RCM) reactions were carried out under
an argon atmosphere with dry, degassed solvents under anhydrous
conditions.
MHz) δ 5.76-5.68 (m, 2H), 4.91 (m, 1H), 4.52 (br d, J ) 17 Hz, 1H),
4.35 (d, J ) 17.4 Hz, 1H), 4.03 (d, J ) 17.4 Hz, 1H), 3.70 (s, 3H),
3.70 (m, 1H), 3.32 (dd, J ) 17.6, 7.2 Hz, 1H), 2.63 (dd, J ) 18.1, 4.1
Hz, 1H), 2.23 (m, 1H); 13C NMR (CDCl3, 125 MHz) δ 172.9, 169.4,
154.9, 129.9, 123.6, 79.5, 52.2, 50.1, 50.0, 47.3, 33.2, 28.3; IR (neat,
cm-1) 3272, 2973, 1759, 1715, 1650, 1538, 1487, 1454; HRMS calcd
for C14H23N2O5 (MH+) 299.1607, found 299.1603.
N-Allyl-N-(tert-butoxycarbonyl)-4,5-didehydro-DL-norvaline, Meth-
yl Ester (2). To a 0 °C solution of (+/-)-N-BOC-allylglycine methyl
ester (1.4 g, 6.1 mmol) in 30 mL of DMF was added allyl bromide
(581 µL, 6.7 mmol) follwed by sodium hydride (160 mg, 6.7 mmol).
Gas evolution was observed, and the reaction mixture assumed a pale
yellow color. The solution was stirred for 1.5 h at 0 °C and 30 min at
25 °C before the reaction was quenched by addition of 20 mL of dilute
aqueous NH4Cl. The product was extracted with three 20 mL portions
of Et2O, dried over MgSO4, and purified by flash chromatography (1.5
cm × 12 cm of silica gel, 20% EtOAc/hexanes) to afford 900 mg (55%)
N-Allyl-N-[N-allyl-N-(tert-butoxycarbonyl)-L-alanyl]glycine, Meth-
yl Ester (4). To a -78 °C solution of N-BOC-Ala-Gly methyl ester
(180 mg, 0.66 mmol) in 5 mL of THF was added allyl bromide (125
mL, 1.45 mmol) followed by P4-phosphazene base (968 µL, 1.38
mmol). The mixture was stirred at -78 °C for 1 h and then warmed
to 25 °C for 30 min. The reaction mixture was concentrated and
purified by chromatography (1.5 cm × 12 cm silica gel, 25% EtOAc/
Hexane) to afford 44 mg (20%) of 4 as a colorless oil: TLC Rf 0.55
(50% EtOAc/hexane); 1H NMR (toluene-d8, 300 MHz, 80 °C, not fully
coalesced) δ 5.8 (br m, 1H), 5.65 (br m, 1H), 5.30-4.80 (m, 5H), 4.20-
1
of 2 as a colorless oil: TLC Rf 0.50 (30% EtOAc/hexane); H NMR
(CDCl3, 500 MHz, not coalesced) δ 5.77-5.60 (br m, 2H), 5.10-4.95
(br m, 4H), 4.5-3.6 (br m, 3H), 3.65 (s, 3H), 2.80-2.65 (br m, 1H),
2.60-2.45 (br m, 1H); 13C NMR (CDCl3, 125 MHz, not coalesced) δ
171.7, 155.4, 154.6, 135.0, 134.5, 134.4, 134.2, 120.4, 117.6, 117.3,
116.1, 80.5, 80.4, 59.0, 58.2, 51.9, 50.5, 49.0, 34.7, 33,8, 28.2; IR (neat,
cm-1) 3079, 2978, 1745, 1697, 1643, 1453; HRMS calcd for C14H24N1O4
(MH+) 270.1705, found 270.1698.
3.70 (m, 6H), 3.40 (s, 3H), 1.40 (s, 9H), 1.28 (d, J ) 6.8 Hz, 3H); 13
C
NMR (toluene-d8, 125 MHz, 80 °C, not fully coalesced) δ 172.0, 169.7,
155.8, 136.9, 134.1, 117.3, 115.7, 80.1, 51.4, 47.6, 46.5, 28.6, 16.4;
IR (neat, cm-1) 2980, 1754, 1660, 1450; HRMS calcd for C17H29N2O5
(MH+) 341.2076, found 341.2064.
1-tert-Butyl 2-Methyl (()-3,6-Dihydro-1,2(2H)-pyridinedicar-
boxylate (5). To a 25 °C solution of acyclic diene 2 (170 mg, 0.636
mmol) in 7 mL of C6H6 was added ruthenium catalyst 1a (29 mg, 0.032
mmol, 5 mol %). The orange-brown solution was stirred at this
temperature for 2 h before it was concentrated and applied directly to
a silica gel column. Chromatography (1.5 cm × 12 cm silica gel,
solvent gradient from 5% EtOAc/hexane to 20% EtOAc/hexane)
afforded 139 mg (91%) of 5 as a clear oil: TLC Rf 0.40 (30% EtOAc/
hexane); 1H NMR (CDCl3, 300 MHz, not coalesced) δ 5.77-5.65 (br
m, 1H), 5.01-4.80 (br m, 1H), 4.20-3.70 (br m, 2H), 3.71, 3.70 (2 ×
s, 3H), 2.65-2.50 (br m, 2H), 1.49, 1.46 (2 × s, 9H); 13C NMR (CDCl3,
75 MHz, not coalesced) δ 172.3, 155.9, 124.5, 124.2, 122.4, 122.0,
Methyl (3S,6Z)-4-(tert-Butoxycarbonyl)-3,4,5,8-tetrahydro-3-
methyl-2-oxo-1,4-diazocine-1(2H)-acetate (7). C6H6 was added ru-
thenium catalyst 1a (7 mg, 0.008 mmol, 16 mol %) as a solution in 5
mL of C6H6. The orange-brown solution was stirred at this temperature
for 24 h before it was concentrated and applied directly to a silica gel
column. Chromatography (1.5 cm × 12 cm silica gel, 25% EtOAc/
Hexane) afforded 8 mg (51%) of 7 as a clear oil: TLC Rf 0.30 (50%
EtOAc/hexane); 1H NMR (toluene-d8, 300 MHz, 80 °C, not fully
coalesced) δ 5.60 (m, 1H), 5.37 (m, 1H), 5.00 (br m, 1H), 4.40-3.00
(br m, 6H), 3.35 (s, 3H), 1.40 (br d, J ) 7 Hz), 1.38 (s, 9H); 13C NMR
(C6D6, 125 MHz, 70 °C, not fully coalesced) δ 171.2, 169.3, 154.3,
134.4 (br), 125.2 (br), 79.9, 51.2, 49.9, 45.4, 43.2 (br), 30.0, 28.4, 17.1;
IR (neat, cm-1) 2926, 1755, 1693, 1659, 1436, 1393; HRMS calcd for
C15H24O5N2 (M+) 312.1685, found 312.1678.
80.4, 52.4, 52.3, 51.0, 42.3, 41.6, 28.4, 26.7, 26.6; IR (neat, cm-1
)
2976, 1746, 1694, 1454, 1403; HRMS calcd for C12H18N1O4 (M - H)
240.1236, found 240.1236.
N-Allyl-N-[N-(tert-butoxycarbonyl)-4,5-didehydro-DL-norval-
yl]glycine, Methyl Ester (3). To a solution of N-allyl-N-BOC-Gly
methyl ester (1.17 g, 5.10 mmol) in 20 mL of CH2Cl2 was added 10
mL of trifluoroacetic acid (TFA). The mixture was stirred for 1 h before
it was concentrated and redissolved in 25 mL of CH2Cl2. The solution
was washed with 100 mL of saturated NaHCO3 solution, dried over
Na2SO4, filtered, and concentrated. The unpurified amine was then
dissolved in 15 mL of CH2Cl2 and treated with (+/-)-N-BOC-
N-Benzyl-N2-[N-[1-[N-(tert-butoxycarbonyl)-4,5-didehydro-L-nor-
valyl]-L-prolyl]-2-methylalanyl]-4,5-didehydro-L-norvalinamide (13).
Tetrapeptide 13 was prepared according to the standard solution protocol
described in the general experimental above. For 13: TLC Rf 0.45
1
(100% EtOAc); [R]D -53.0 (c 0.7, CH2Cl2); H NMR (CD3CN, 500
MHz) δ 7.74 (br t, 1H), 7.29 (m, 5H), 7.22 (m, 1H), 7.0 (br s, 1H),
5.83-5.78 (m, 2H), 5.55 (br d, 1H), 5.18-5.05 (m, 4H), 4.42-4.21
(m, 5H), 3.99 (br t, 1H), 3.76 (m, 1H), 3.60 (m, 1H), 2.75 (m, 1H),
2.5-1.1.7 (m, 8H), 1.40 (s, 12H), 1.35 (s, 3H); 13C NMR (CD3CN,
125 MHz) δ 175.5, 174.0, 173.5, 172.6, 157.0, 140.8, 136.6, 135.0,
129.7, 128.7, 128.1, 119.3, 118.1, 80.5, 63.6, 58.3, 54.8, 53.8, 49.0,
43.9, 36.9, 36.6, 30.0, 29.0, 27.6, 26.3, 24.8; IR (CH2Cl2, cm-1) 3425,
(27) Still, W. C.; Kahn, M.; Mitra, A. J. Org. Chem. 1978, 43, 2923-
2925.
(28) See ref 3.
(29) Bodansky, M. Peptide Chemistry; Springer-Verlag: New York,
1988, pp 55-146 and references therein.

Ring-Closing Metathesis
J. Am. Chem. Soc., Vol. 118, No. 40, 1996 9613
3339, 2982, 2934, 1668, 1498, 1439; HRMS calcd for C31H46N5O6
(MH+) 584.3448, found 584.3471.
3.97 (m, 1H), 3.85 (m, 2H), 3.68 (m, 1H), 3.64 (s, 3H), 3.36 (m, 2 H),
2.80 (m, 1H), 2.46 (m, 2 H), 2.11 (m, 2H), 1.84 (m, 1H), 1.71 (m,
2H), 1.37 (s, 9H), 1.23 (m, 1H); 13C NMR ((CD3)2SO, 75 MHz) δ
171.8, 170.8, 170.2, 170.0, 168.9, 155.0, 152.7, 149.1, 136.5, 134.0,
132.7, 130.7, 130.6, 130.3, 130.1, 128.0, 127.4, 122.9, 120.5, 78.2,
69.2, 60.8, 53.1, 51.7, 51.2, 50.7, 46.5, 35.3, 34.8, 28.5, 28.0, 24.4; IR
(CH2Cl2, cm-1) 3426, 3354, 2954, 2923, 2851, 1759, 1687, 1621, 1508,
1446, 1369, 1164; HRMS calcd for C38H47N5O11Br (MH+) 828.2443,
found 828.2455.
N-[N-[N-[N-(tert-Butoxycarbonyl)-4,5-didehydro-L-norvalyl]-L-
leucyl]-L-leucyl]-4,5-didehydro-L-norvaline, Benzyl Ester (17). Tet-
rapeptide 17 was prepared according to the standard solution protocol
described in the general experimental above. For 17: TLC Rf 0.41
(80% CH2Cl2/EtOAc); [R]D -29.5 (c 1.2, CH2Cl2); 1H NMR (CDCl3,
500 MHz, 25 °C, not fully coalesced) δ 7.33 (m, 5H), 6.91 (apparent
s, 1H), 6.73 (apparent s, 1H), 6.64 (d, J ) 7 Hz, 1H), 5.74-5.63 (m,
2H), 5.19-5.02 (m, 6H), 4.66 (q, J ) 7 Hz, 1H), 4.49 (q, J ) 9 Hz,
1H), 4.39 (q, J ) 8 Hz, 1H), 4.11 (m, 1H), 2.61-2.38 (m, 4H), 1.73-
1.46 (m, 6H), 1.43 (s, 9H), 0.96-0.86 (m, 12H); 13C NMR (CDCl3,
125 MHz, 25 °C, not fully coalesced) δ 171.9, 171.8, 171.4, 135.7,
133.1, 132.6, 128.8, 128.6, 128.5, 80.7, 67.2, 52.3, 51.9, 41.3, 41.1,
36.8, 36.5, 34.2, 29.9, 28.5, 25.9, 25.1, 25.0, 24.9, 23.2, 23.0, 22.2,
22.1; IR (CH2Cl2, cm-1) 3414, 3339, 2961, 2929, 1740, 1694, 1505,
1456, 1365, 1169; HRMS calcd for C34H53N4O7 (MH+) 629.3914, found
629.3914.
tert-Butyl (6S,8E,11S,16aS)-6-(Benzylcarbamoyl)-1,2,3,4,5,6,7,10,-
11,12,14,15,16,16a-tetradecahydro-3,3-dimethyl-1,4,12-trioxopyr-
rolo[1,2-a][1,4,7]-triazacyclotetradecine-11-carbamate (14). To a
solution of acyclic diene 13 (210 mg, 0.360 mmol) in 80 mL of CH2-
Cl2 was added via cannula a solution of ruthenium catalyst 1a (67 mg,
0.072 mmol) predissolved in 20 mL of CH2Cl2. The orange-brown
solution was heated to 40 °C and stirred at this temperature for 20 h.
The solution was then concentrated under reduced pressure to afford
an oily brown mixture. Purification by chromatography (1.5 cm × 15
cm silica gel, solvent gradient from 50% EtOAc/hexane to 100%
EtOAc) afforded 120 mg (60%) of 14 as an off-white powder.
Macrocycle 14 can be recrystallized by dissolving the powder in CH2-
Cl2 and layering the solution with hexanes. White needles and prisms
result. However, upon removal of solvent the amorphous white powder
is reobtained. For 14: TLC Rf 0.20 (100% EtOAc); [R]D +63.8 (c
1
0.73, CH2Cl2); H NMR (CD2Cl2, 500 MHz) δ 7.33-7.21 (m, 5H),
7.01 (br d, J ) 7.2 Hz, 1H), 6.91 (br t, 1H), 6.53 (br s, 1H), 5.58 (d,
J ) 7.8Hz, 1H), 5.51-5.42 (d(m), J ) 15 Hz, 1H), 5.39-5.30 (d(m),
J ) 15 Hz, 1H), 4.65-4.59 (m, 2H), 4.50 (q, J ) 6 Hz, 1H), 4.29 (dd,
J ) 15, 5.0 Hz, 1H), 4.20 (t, J ) 7.2 Hz, 1H), 3.69 (m, 1H), 3.58 (m,
1H), 2.6-1.88 (m, 8H), 1.50 (s, 3H), 1.41 (s, 9H), 1.35 (s, 3H); 13C
NMR (CD2Cl2, 125 MHz) δ 175.4, 172.5, 172.1, 172.0, 155.0, 139.9,
131.0, 129.6, 128.9, 128.5, 128.3, 80.5, 62.2, 61.9, 58.4, 53.4, 48.7,
44.2, 34.9, 34.4, 29.3, 29.2, 26.8, 23.9; IR (CH2Cl2, cm-1) 3424, 3324,
3054, 2985, 1691, 1632, 1498, 1444; HRMS calcd for C29H42N5O6
(MH+) 556.3135, found 556.3145.
tert-Butyl (3S,6S,9S,11E,14S)-14-[(Benzyloxy)carbonyl]-3,6-di-
isobutyl-2,5,8-trioxo-1,4,7-triazacyclotetradec-11-ene-9-carbam-
ate (18). To a solution of acyclic diene 17 (285 mg, 0.453 mmol) in
100 mL of CH2Cl2 was added via syringe a solution of ruthenium
catalyst 1b (112 mg, 0.136 mmol) predissolved in 22 mL of CH2Cl2.
The purple solution was heated to 45 °C and turned orange-brown in
color over 20 min. The solution was stirred at 45 °C for 21 h. The
solution was then concentrated under reduced pressure to afford an
oily brown mixture. Purification by chromatography (3 cm × 15 cm
silica gel, 80% CH2Cl2/EtOAc) afforded 163 mg (60%) of 18 as an
off-white powder: TLC Rf 0.24 (80% CH2Cl2/EtOAc); [R]D -114.0
N-[N-[[O-[(o-Bromobenzyl)oxy]carbonyl]-N-[1-[N-(tert-butoxy-
carbonyl)-4,5-didehydro-L-norvalyl]-L-prolyl]-L-tyrosyl]-4,5-dide-
hydro-L-norvalyl]glycine, Methyl Ester (15). Pentapeptide 15 was
prepared according to the standard solution protocol described in the
general experimental above.30 For 15: TLC Rf 0.27 (83% EtOAc/
hexane); [R]D -38.0 (c 1.1, CH2Cl2); 1H NMR ((CD3)2SO, 500 MHz)
δ 8.34 (br t, J ) 5 Hz, 1H), 7.94 (d, J ) 8 Hz, 1H), 7.89 (d, J ) 8 Hz,
1H), 7.71 (d, J ) 8 Hz, 1H), 7.57 (d, J ) 8 Hz, 1H), 7.46 (t, J ) 8 Hz,
1H), 7.36 (t, J ) 8 Hz, 1H), 7.29 (amide NH obscured, apparent d, J
) 8 Hz, 3H), 7.13 (d, J ) 7 Hz, 2H), 6.91 (d, J ) 8 Hz, 1H), 5.83-
5.70 (m, 2H), 5.32 (s, 2H), 5.12- 5.01 (m, 4H), 4.48 (br q, J ) 9 Hz,
1H), 4.40-4.30 (br m, 2H), 4.21 (br q, J ) 7 Hz, 1H), 3.91-3.80 (m,
2H), 3.63 (s, 3H), 3.60-3.51 (m, 2H), 3.06 (dd, J ) 14, 4 Hz, 1H),
2.87 (dd, J ) 14, 9 Hz, 1H), 2.40 (m, 1H), 2.33 (m, 2H), 2.22 (m,
1H), 1.95 (m, 1H), 1.82-1.71 (br m, 3H), 1.36 (s, 9H); 13C NMR
(CDCl3, 125 MHz, 25 °C, not fully coalesced) δ 172.1, 172.0, 171.4,
170.9, 170.3, 155.6, 153.3, 150.2, 134.8, 134.4, 133.3, 127.7, 123.5,
120.0, 118.8, 118.5, 79.8, 69.6, 63.0, 60.7, 55.9, 52.9, 52.2, 47.6, 41.3,
36.9, 36.5, 28.7, 28.4, 25.2; IR (CH2Cl2, cm-1) 3415, 3333, 2923, 2851,
1759, 1672, 1605, 1503, 1441, 1364; HRMS calcd for C40H51N5O11Br
(MH+) 856.2768, found 856.2794.
1
(c 1.0, CH2Cl2); H NMR ((CD3)2SO, 500 MHz) δ 7.90 (d, J ) 9,
1H), 7.80 (d, J ) 8, 1H), 7.37 (m, 5H), 7.24 (d, 8 Hz, 1H), 5.46 (d(m),
J ) 15 Hz, 1H), 5.16 (d(m), J ) 15 Hz, 1H), 5.13 (s, 2H), 4.48 (m,
1H), 4.21 (m, 2H), 4.03 (m, 1H), 2.45-2.12 (br m, 4H), 1.66-1.45
(m, 6H), 1.38 (s, 9H), 0.87 (apparent d, J ) 6 Hz, 6H), 0.81 (apparent
t, J ) 6 Hz, 6H); 13C NMR ((CD3)2CO, 75 MHz) δ 171.9, 170.8, 170.6,
130.3, 128.1, 127.7, 127.6, 127.5, 78.7, 66.0, 53.4, 52.8, 50.6, 50.0,
40.2, 39.7, 34.6, 33.7, 24.4, 24.2, 24.1, 22.5, 22.2, 20.4, 20.3; IR (CH2-
Cl2, cm-1) 3419, 3339, 2958, 2929, 1740, 1671, 1602, 1515, 1365,
1158; HRMS calcd for C32H49N4O7 (MH+) 601.3601, found 601.3610.
O-Allyl-N-[O-allyl-N-(tert-butoxycarbonyl)-L-tyrosyl]-L-tyro-
sine, Methyl Ester (19). To a solution of N-BOC-dityrosine methyl
ester (3.26 g, 7.10 mmol) in 30 mL of acetone were added allyl bromide
(1.71 mL, 19.8 mmol) and finely powdered K2CO3 (2.94 g, 21.3 mmol).
The reaction mixture was stirred for 48 h at 25 °C before being filtered
through a Celite pad. Purification of the residue by chromatography
(3 cm × 12 cm silica gel, solvent gradient from 20% EtOAc/hexane
to 50% EtOAc/hexane) afforded 19 as a white solid: TLC Rf 0.50 (50%
tert-Butyl (3S,6S,8E,11S,16aS)-3-[p-[[[(o-Bromobenzyl)oxy]car-
bonyl]oxy]benzyl]-1,2,3,4,5,6,7,10,11,12,14,15,16,16a-tetradecahydro-
6-[[(methoxycarbonyl)methyl]carbamoyl]-1,4,12-trioxypyrrolo[1,2-
a][1,4,7]-triazacyclotetradecine-11-carbamate (16). To a solution of
acyclic diene 15 (200 mg, 0.234 mmol) in 53 mL of CH2Cl2 was added
via syringe a solution of ruthenium catalyst 1b (58 mg, 0.072 mmol)
predissolved in 10 mL of CH2Cl2. The purple solution was heated to
45 °C and turned orange-brown in color over a 30 min period. The
solution was stirred at 45 °C for 23 h. The solution was then
concentrated under reduced pressure to afford an oily brown mixture.
Purification by chromatography (3 cm × 15 cm silica gel, solvent
gradient from 80% EtOAc/hexane to 100% EtOAc) afforded 155 mg
(80%) of 16 as an off-white powder: TLC Rf 0.21 (83% EtOac/hexane);
1
EtOAc/hexane); [R]D +20.7 (c 2.0, CH2Cl2); H NMR (CDCl3, 500
MHz) δ 7.06 (d, J ) 8.4 Hz, 2H), 6.87 (d, J ) 8.4 Hz, 2H), 6.79 (d,
J ) 8.4 Hz, 2H), 6.75 (d, J ) 8.4 Hz, 2H), 6.38 (br d, J ) 7.5 Hz,
1H), 6.00 (m, 2H), 5.35 (dd, J ) 17.1, 0.6 Hz, 2H), 5.23 (br d, J ) 9.3
Hz, 2H), 4.92 (br s, 1H), 4.71 (br d, J ) 6.2 Hz, 1H), 4.46 (br t, J )
4.1 Hz, 4H), 4.29 (br s, 1H), 3.63 (s, 3H), 2.95 (m, 4H), 1.38 (s, 9H);
13C NMR (CDCl3, 125 MHz) δ 171.3, 170.7, 157.6 (br d), 155.1, 133.2,
130.2, 130.1, 128.6, 127.8, 117.3, 117.2, 114.8, 114.7, 79.9, 68.7, 68.6,
55.8, 53.3, 51.9, 37.3, 37.0, 28.1; IR (CH2Cl2, cm-1) 3420, 2981, 2932,
1742, 1713, 1681, 1610, 1510, 1361; HRMS calcd for C30H39N2O7
(MH+) 539.2757, found 539.2766.
1
[R]D -29.0 (c 1.1, CH2Cl2); H NMR ((CD3)2SO, 500 MHz) δ 8.22
(br t, J ) 6 Hz, 1H), 7.88 (br d, J ) 9 Hz, 1H), 7.71 (d, J ) 8 Hz,
1H), 7.57 (d, J ) 7 Hz, 1H), 7.46 (t, J ) 7 Hz, 1H), 7.36 (t, J ) 7 Hz,
2H), 7.22 (d, J ) 8 Hz, 2H), 7.13 (amide NH obscured, apparent d, J
) 8 Hz, 3H), 6.69 (d, J ) 9 Hz, 1H), 5.46 (d(m), J ) 15 Hz, 1H),
5.36 (d(m), J ) 15 Hz, 1H), 5.31 (s, 2H), 4.58 (m, 2H), 4.47 (m, 1H),
tert-Butyl (13S,16S)-13-(Methoxycarbonyl)-15-oxo-2,7-dioxa-14-
azatricyclo[16.2.2.28,11]tetracosa-4,8,10,18,20,21,23-heptaene-16-car-
bamate (20). To a 50 °C solution of acyclic diene 19 (130 mg, 0.241
mmol) in 100 mL of CH2Cl2 was added ruthenium catalyst 1b (29 mg,
0.072 mmol). Within 5 min, the purple solution became orange-brown,
and the solution was stirred for an additional 2.5 h, when TLC analysis
showed full disappearance of starting material. Triethylamine (1 mL)
(30) A minor impurity remains in compound 15 even after repeated
column chromatography and recrystallization. Peptide 15 is greater than
95% pure by 1H NMR. This impurity is not present in the cyclized product
16.

9614 J. Am. Chem. Soc., Vol. 118, No. 40, 1996
Miller et al.
was added to the solution to deactivate any remaining active catalyst.
The solution was then concentrated to afford an oily brown mixture.
Purification by chromatography (3 cm × 12 cm silica gel, 50% EtOAc/
hexane) afforded 83 mg (68%) of 21 as a white powder: olefin
configuration not assigned; TLC Rf 0.45 (50% EtOAc/hexane); [R]D
+46.4 (c 1.0, CH2Cl2); 1H NMR (CDCl3, 500 MHz) δ 7.04 (br s, 2H),
6.77 (br s, 2H), 6.68 (t, J ) 8.5 Hz, 4H), 5.90 (br m, 1H), 5.83 (s, 2H),
5.25 (br s, 1H), 4.85 (br s, 1H), 4.66 (m, 4H), 4.39 (br t, 1H), 3.70 (s,
3H, minor rotomer at 3.77), 3.34 (br d, J ) 12.6 H, 1H), 2.98 (br s,
2H), 2.63 (dd, J ) 14.0, 8.9 Hz, 1H), 1.49 (s, 9H); 13C NMR (CDCl3,
125 MHz, 25 °C, not fully coalesced) δ 171.2, 171.1, 156.8, 156.6,
155.5, 130.5, 130.2, 130.0, 129.7, 129.5, 129.2, 127.4, 115.6, 115.4,
115.3, 80.9, 67.0, 66.7, 64.7, 64.1, 55.6, 52.7, 52.1, 37.4, 36.2, 28.3;
IR (CH2Cl2, cm-1) 3683, 3411, 2933, 1744, 1713, 1676, 1611, 1511,
1484; HRMS calcd for C28H35N2O7 (MH+) 511.2444, found 511.2437.
N-[N-(tert-Butoxycarbonyl)-4,5-didehydro-L-norvalyl]-O-4-pen-
tenoyl-L-serine, Methyl Ester (21). N-BOC-serine methyl ester (1.20
g, 5.47 mmol) was dissolved in 75 mL of CH2Cl2 and treated with
4-pentenoic acid (559 µL, 5.47 mmol), DCC (1.13 g, 5.47 mmol), and
DMAP (100 mg, 0.82 mmol). A white precipitate formed immediatley,
and the solution was stirred for 12 h. The mixture was filtered and
washed with 50 mL of a 10% citric acid solution, followed by 50 mL
of saturated NaHCO3 solution. The solution was dried over MgSO4
and concentrated to afford 1.65 g of the crude esterified product as a
pale yellow oil with some crystalline domains. To the crude product
(1.65 g, theoretical 5.47 mmol) dissolved in 15 mL of CH2Cl2 was
added an excess of TFA (7.67 mL, 99.6 mmol). The solution was
allowed to stir at room temperature for 2.5 h after which the solution
was concentrated to an orange oil and dried under high vacuum. The
oil was then taken up in 100 mL of CH2Cl2 and treated with
triethylamine (915 µL, 6.56 mmol) at room temperature. After 15 min
of stirring, N-BOC-allylglycine (1.18 g, 5.47 mmol), HOBT (1.11 g,
8.21 mmol), and DCC (1.13 g, 5.47 mmol) were added to the solution.
A white precipitate formed immediately, and the solution was allowed
to stir at room temperature for 9 h. The mixture was then filtered,
washed with 75 mL of a 10% citric acid solution, and subsequently
washed with 75 mL of a saturated NaHCO3 solution. The product was
dried over MgSO4 and concentrated to yield a yellow oil. Purification
by column chromatography (4 cm × 15 cm silica gel, solvent gradient
from 25% EtOAc/hexane to 50% EtOAc/hexane) afforded 1.19 g (55%)
of 21 as a clear oil: TLC Rf 0.21 (75% hexane/EtOAc); [R]D +13.1 (c
(m, 1H), 4.52 (apparent t, J ) 11 Hz, 1H), 4.16 (dd, J ) 11, 4 Hz,
1H), 4.07 (m, 1H), 3.64 (s, 3H), 2.44-2.18 (m, 6H), 1.41, 1.40 (2 ×
s, 9H); 13C NMR ((CD3)2CO, 75 MHz, 25 °C, not fully coalesced) δ
172.2, 170.5, 169.5, 155.0, 131.7, 126.0, 78.5, 61.0, 55.0, 52.3, 51.9,
50.1, 35.0, 34.0, 28.1; IR (CH2Cl2, cm-1) 3691, 3422, 2929, 1736, 1720,
1683, 1517, 1485; HRMS calcd for C17H27N4O7 (MH+) 371.1818, found
371.1812.
Solid Phase Synthesis of 23. Peptide 23 was prepared by manual
solid phase peptide synthesis.31 Fmoc-Pal-PEG-PS resin (substitution
0.20 mmol/g) was used to afford C-terminal primary amides. Com-
merically available reagents and starting materials were purchased from
Sigma Chemical Co., Applied Biosystems, Peptides International, and
PerSeptive Biosystems. NR-(Fluorenylmethoxy)carbonyl (Fmoc) pro-
tection was employed for all amino acids in the solid phase synthesis,
with the tyrosine phenol protected as a tert-butyl ester. Each amino
acid was coupled sequentially to the peptide chain grown from the
C-terminal amino acid using N,N-diisopropylcarbodiimide/1-hydroxy-
benzotriazole. A complete coupling in each step was monitored by a
quantitative ninhydrin test.32 Unreacted N-termini were acetylated using
an acetic anyhdride/HOBT/diisoprolpylethylamine capping protocol.
Fmoc groups were cleaved with 20% piperidine in dimethylformamide
(DMF). The peptide was deprotected and cleaved form the resin by
treatment with with a solution of TFA/anisole/thioanisole (90:5:5) for
2 h.
Heterogeneous RCM Protocol. To a suspension of solid-support-
bound peptide 23 (300 mg of resin, 0.06 mmol (theoretical) of bound
peptide) in 22 mL of CH2Cl2 was added via syringe a solution of
ruthenium catalyst 1b (25 mg, 0.030 mmol) predissolved in 5 mL of
CH2Cl2. The solution turned from pink to orange-brown over 3 h. The
suspension was heated to 40 °C and gently stirred for 22 h. The beads
were then filtered, rinsed with CH2Cl2, DMF, and MeOH, respectively,
and dried under high vacuum. To 270 mg of the dried resin was added
3 mL of a solution of TFA/anisole/thioanisole (90:5:5). The suspension
was shaken gently at room temperature for 2 h, after which the beads
were filtered and rinsed with a minimal amount of TFA. The filtrate
was reduced in volume to ∼0.5 mL to yield a brown oil. Trituration
with 2:1 ether/hexane afforded the crude peptide mixture as an off-
white solid. The solid was dissolved in deionized H2O and freeze dried
to afford a cream powder which was a mixture of 60% 24 and 40%
23.
1
Acknowledgment. This research was generously supported
by a grant from the NIH. S.J.M. is grateful to the NSF for a
postdoctoral fellowship. H.E.B. is grateful to the NSF for a
predoctoral fellowship. In addition, Professor Andrew G. Myers
and Dr. Jim Gleason are gratefully acknowledged for providing
samples of optically pure (L)- and (R)-allylglycine, as well as
for helpful discussions. Professor Barbara Imperiali, Dr. Mary
Struthers, Dr. Ranabir Sinha Roy, and Eldon E. Baird are
gratefully acknowledged for assistance with solid phase peptide
synthesis and many helpful discussions.
1.0, CH2Cl2); H NMR (CDCl3, 500 MHz) δ 6.92 (br d, J ) 7 Hz,
1H), 5.81-5.68 (m, 2H), 5.15-4.97 (m, 5H), 4.80-4.78 (m, 1H), 4.42
(dd, J ) 11, 4 Hz, 1H), 4.36 (dd, J ) 11, 3 Hz, 1H), 4.17 (m, 1H),
3.73 (s, 3H), 2.54-2.43 (m, 2H), 2.38 (m, 2H), 2.32 (m, 2H), 1.41 (s,
9H); 13C NMR (CDCl3, 125 MHz) δ 172.6, 171.6, 169.7, 155.6, 136.6,
133.1, 119.2, 115.8, 80.4, 63.8, 52.9, 52.0, 36.8, 34.1, 33.7, 28.8, 28.4;
IR (CH2Cl2, cm-1) 3680, 3427, 2981, 1746, 1716, 1685, 1494, 1438;
HRMS calcd for C19H31N2O7 (MH+) 399.2131, found 399.2140.
tert-Butyl (3S,6S,8E)-3-(Methoxycarbonyl)-5,12-dioxo-1-oxa-4-
azacyclododec-8-ene-6-carbamate (22). To a solution of acyclic diene
21 (50 mg, 0.125 mmol) in 230 mL of CH2Cl2 was added via syringe
a solution of ruthenium catalyst 1b (30 mg, 0.036 mmol) predissolved
in 20 mL of CH2Cl2. The purple solution was heated to 45 °C and
turned orange-brown in color over 20 min. The solution was stirred
at 45 °C for 20 h. The solution was then concentrated under reduced
pressure to afford an oily brown mixture. Purification by chromatog-
raphy (3 cm × 10 cm silica gel, (50% EtOAc/hexane) afforded 28 mg
(56%) of 22 as an off-white powder: TLC Rf 0.30 (50% EtOAc/
hexane); [R]D +41.1 (c 1.0, CH2Cl2); 1H NMR ((CD3)2SO, 500 MHz,
60 °C, not fully coalesced) δ 7.47 (br d, J ) 8 Hz, 1H), 6.03 (apparent
s, 1H), 5.48 (d(m), J ) 15 Hz, 1H), 5.32 (d(m), J ) 15 Hz, 1H), 4.72
Supporting Information Available: 1H and 13C NMR
1
spectra of 2 and 5, H NMR and GC-MS spectra of 4 and 7,
1
and H NMR spectra of 3, 4, 6, and 13-22 (20 pages). See
any current masthead for ordering and Internet access information.
JA961626L
(31) Merrified, R. B. J. Am. Chem. Soc. 1963, 85, 1476-1482.
(32) Sarin, V. K.; Kent, S. B. H.; Tam, J. P.; Merrifield, R. B. Anal.
Biochem. 1981, 117, 147-157.