Controlled synthesis of (S,S)-2,7-diaminosuberic acid: A method for regioselective construction of dicarba analogues of multicystine-containing peptides

Article abstract of DOI:10.1021/jo0606913

(Chemical Equation Presented) A method to facilitate regioselective formation of multiple dicarba isosteres of cystine is described. A sequence of ruthenium-catalyzed cross metathesis and rhodium-catalyzed hydrogenation of nonproteinaceous allylglycine de

Full text of DOI:10.1021/jo0606913

Controlled Synthesis of (S,S)-2,7-Diaminosuberic Acid: A Method
for Regioselective Construction of Dicarba Analogues of
Multicystine-Containing Peptides
Jomana Elaridi, Jim Patel, W. Roy Jackson, and Andrea J. Robinson*
School of Chemistry, P.O. Box 23, Monash UniVersity, Victoria 3800, Australia
andrea.robinson@sci.monash.edu.au
ReceiVed March 31, 2006
A method to facilitate regioselective formation of multiple dicarba isosteres of cystine is described. A
sequence of ruthenium-catalyzed cross metathesis and rhodium-catalyzed hydrogenation of nonproteina-
ceous allylglycine derivatives has been developed to achieve high-yielding and unambiguous formation
of diaminosuberic acid derivatives. Allylglycine derivatives readily undergo ruthenium-catalyzed metathesis
and hydrogenation to yield diaminosuberic acid derivatives in near quantitative yield. Under the same
experimental conditions, prenylglycine was found to be inert to both Grubbs’ and Wilkinson’s catalyzed
metathesis and hydrogenation, respectively, but was readily activated for metathesis via cross metathesis
with Z-butene. Subsequent cross metathesis of the metathesis-formed crotylglycine derivative, followed
by hydrogenation, yielded the second diaminosuberic acid derivative in excellent yield.
Introduction
tides,9 has been reported. These dicarba analogues were shown
to be biologically active and many showed improved pharma-
cokinetic properties relative to their cognate peptides.
Cystine bridges are common structural motifs found in
peptides and proteins. In some cases, the cystine bridge
constitutes part of a peptide binding domain or active site where
reduction results in the release of metal-chelating thiol groups.1
In many other peptides, however, the cystine bridge serves only
a structural role to maintain secondary and tertiary structure.
Here, cystine may be replaced with isosteric units, such as a
nonreducible all-carbon -(CH2)4- bridge, without significantly
affecting biological activity. Replacement of cystine (1) by (S,S)-
2,7-diaminosuberic acid (2, (S,S)-2,7-diaminooctanedioic acid)2
in several naturally occurring and synthetic cyclic peptides, such
as vasopressin,3 natriuretic â-ANP,4 octreotide,5 oxytocin,6,
calcitocin,7 bradykinin antagonists,8 and hematoregulatory pep-
Until recently, synthesis and incorporation of (S,S)-2,7-
diaminosuberic acid into a peptide was challenging. Several
elegant approaches have recently been reported, which include
Scho¨llkopf bislactam ether methodology,10 Kolbe coupling of
(1) (a) Pons, M.; Albericio, F. R.; Royo, M.; Giralt, E. Synlett 2000, 2,
172-181. (b) Jones, R. M.; Bulaj, G. Curr. Opin. Drug DiscoVery DeV.
2000, 3, 141-154.
(2) (a) Walker, R.; Yamanaka, T.; Sakakibara, S. Proc. Natl. Acad. Sci.
U.S.A. 1974, 71, 1901-1905. (b) Nutt, R. F.; Verber, D. F.; Saperstein, R.
J. Am. Chem. Soc. 1980, 102, 6539-6545. (c) Collier, P. N.; Campbell, A.
D.; Patel, I.; Raynham, T. M.; Taylor, R. J. K. J. Org. Chem. 2002, 67,
1802-1815.
(3) Hase, S.; Morikawa, T.; Sakakibara, S. Experientia 1969, 25, 1239-
1240.
(4) Kambayashi, Y.; Nakajima, S.; Ueda, M.; Inouye, K. FEB 1989, 248,
28-34.
(5) Whelan, A.; Elaridi, J.; Harte, M.; Smith, S.; Jackson, W. R.;
Robinson, A. J. Tetrahedron Lett. 2004, 45, 9545-9547.
(6) (a) Jost, K.; Sorm, F. Collect. Czech. Chem. Commun. 1971, 36, 234-
245. (b) Stymiest, J. L.; Mitchell, B. F.; Wong, S.; Vederas, J. C. Org.
Lett. 2003, 5, 47-49.
(7) Cerovsky, V.; Wunsch, E.; Brass, J. Eur. J. Biochem. 1997, 247,
231-237.
(8) Lange, M.; Cuthbertson, A. S.; Towart, R.; Fischer, P. M. J. Peptide
Sci. 1998, 4, 289-293.
(9) Bhatnagar, P. K.; Agner, E. K.; Alberts, D.; Arbo, B. E.; Callahan,
J. F.; Cuthbertson, A. S.; Angelsen, S. J.; Fjerdingstad, H.; Hartmann, M.;
Heerding, D.; Hiebl, J.; Huffman, W. F.; Hysben, M.; King, A. G.;
Kremminger, P.; Kwon, C.; LoCastro, S.; Lovhaug, D.; Pelus, L. M.;
Petteway, S.; Takata, J. S. J. Med. Chem. 1996, 39, 3814-3819.
10.1021/jo0606913 CCC: $33.50 © 2006 American Chemical Society
Published on Web 09/02/2006
7538
J. Org. Chem. 2006, 71, 7538-7545

Controlled Synthesis of (S,S)-2,7-Diaminosuberic Acid
SCHEME 1
SCHEME 2
glutamic acid derivatives,11 and the use of chiral auxiliaries.12
The use of ruthenium-catalyzed metathesis has also emerged
as a powerful tool for the preparation of this cystine isostere.
Commercially available allylglycine residues (Hag) are used in
place of cysteine and cyclized under mild experimental condi-
tions to yield carbocyclic structures.5,6b,13 The initial product
of the metathesis is an unsaturated C4-bridge, where both E-
and Z-isomers are possible, and hydrogenation is readily
achieved to give the dicarba analogue. This provides scope for
the probing of conformational preferences around the replaced
cystine bridge and is also useful for library generation.
Furthermore, all of the synthesis, both linear peptide construc-
tion, cyclization, and reduction, can be accomplished on a solid
support.
SCHEME 3
both Grubbs’ catalyst ((PCy3)2(Cl)2RudCHPh)18 and second
generation Grubbs’ catalyst ((IMesH2)(PCy3)(Cl)2RudCHPh),19
whereas the electron-deficient R-N-acyldienamide (3) should be
considerably less reactive. Subsequent asymmetric hydrogena-
tion of the dienamide moieties would lead to reactive allyl-
glycine units and a subsequent ring closing metathesis reaction
would then produce the second carbocycle. The last step
involves hydrogenation of the unsaturated carbocycles, if
required, to afford the saturated cystine isosteres.
Many peptides, such as conotoxins14 and cyclotides,15 contain
multiple cystine bridges within a single peptide chain. Chal-
lenges associated with selective cystine formation and the facile
reduction of disulfide bonds to reactive thiol groups have led
us to investigate the use of homogeneous catalysis for the
regioselective construction of dicarba analogues of multicystine-
containing peptides.
Results and Discussion
To validate the proposed strategy, we needed to show that
(i) the dienamide would not react under conditions required for
the ring closing metathesis of allylglycine residues, (ii) asym-
metric hydrogenation of the dienamide would proceed in a
highly regioselective and stereoselective manner, (iii) ring
closing metathesis of the resulting allylglycine units would
proceed in the presence of an unsaturated carbocycle (without
resulting in mixed cross metathesis products), and (iv) the
unsaturated carbocycles could be reduced to afford saturated
dicarba bridges. We therefore conducted a series of independent
experiments that would serve as a model to the peptide system.
The allylglycine derivative 4 was prepared via Rh(I)-
DuPHOS20 catalyzed asymmetric hydrogenation of the dien-
amide 316 with excellent stereoselectivity (95% ee) and yield
(97%) and readily cross metathesized using a ruthenium
benzylidene catalyst, Grubbs’ catalyst, to produce dimer 5
(Scheme 2). Importantly, the dienamide 3 did not react under
these cross metathesis conditions. Cross metathesis of allyl-
Our initial strategy planned to capitalize on the use of R-N-
acyldienamide (3), a masked precursor to allylglycine deriva-
tives.16 We devised a strategy involving a double metathesis-
hydrogenation sequence (Scheme 1). This requires a selective
ring closing metathesis (RCM) of allylglycine units in the
presence of dienamide functionalities. Grubbs and co-workers
have previously reported that selective cross metathesis can be
accomplished with olefins of varying reactivity.17 Terminal
olefins (e.g., allylglycine) undergo rapid homodimerization with
(10) Lange, M.; Fischer, P. M. HelV. Chim. Acta 1998, 81, 2053-2061.
(11) Hiebl, J.; Blanka, M.; Guttman, A.; Hollman, H.; Leitner, K.;
Mayrhofer, G.; Rovenszky, F.; Winkler, K. Tetrahedron 1998, 54, 2059-
2074.
(12) Williams, R. M.; Yuan, C. J. Org. Chem. 1992, 57, 6519-6527.
(13) (a) Gao, Y.; Lane-Bell, P.; Vederas, J. J. Org. Chem. 1998, 63,
2133-2143. (b) Williams, R. M.; Lui, J. J. Org. Chem. 1998, 63, 2130-
2132. (c) Miller, S. J.; Blackwell, H. E.; Grubbs, R. H. J. Am. Chem. Soc.
1996, 118, 9606-9614. (d) Creighton, C. J.; Reitz, A. B. Org. Lett. 2001,
3, 893-895. (e) See also: Aguilera, B.; Wolf, L. B.; Nieczypor, P.; Rutjes,
F. P. J. T.; Overkleeft, H. S.; van Hest, J. C. M.; Schoemaker, H. E.; Wang,
B.; Mol, J. C.; Furstner, A.; Overland, M.; van der Marel, G. A.; van Boom,
J. H. J. Org. Chem. 2001, 66, 3584-3589 and references therein.
(14) Adams, D. J.; Alewood, P. F.; Craik, D. J.; Drinkwater, R. D.; Lewis,
R. J. Drug DeV. Res. 1999, 46, 219-234.
(17) (a) Chatterjee, A. K.; Choi, T.-L.; Sanders, D. P.; Grubbs, R. H. J.
Am. Chem. Soc. 2003, 125, 11360-11370. (b) Chatterjee, A. K.; Sanders,
D. P.; Grubbs, R. H. Org. Lett. 2002, 4, 1939-1942.
(18) Schwab, P.; France, M. B.; Ziller, J. W.; Grubbs, R. H. Angew.
Chem., Int. Ed. Engl. 1995, 34, 2039-2041.
(19) Louie, J.; Bielawski, C. W.; Grubbs, R. H. J. Am. Chem. Soc. 2001,
123, 11312-11313.
(20) Burk, M. J.; Feaster, J. E.; Nugent W. A.; Harlow, R. L. J. Am.
Chem. Soc. 1993, 115, 10125-10138.
(15) Craik, D. J.; Daly, N. L.; Bond, T.; Waine, C. J. Mol. Biol. 1999,
294, 1327-1336.
(16) Teoh, E.; Campi, E. M.; Jackson, W. R.; Robinson, A. J. Chem.
Commun. 2002, 978-979.
J. Org. Chem, Vol. 71, No. 20, 2006 7539

Elaridi et al.
FIGURE 1. 1H NMR binding studies between Grubbs’ catalyst and dienamide 3.
SCHEME 4
SCHEME 5
to dimerize allylglycine 4 in the presence of dienamide 3 were
unsuccessful. H NMR binding studies between the catalyst,
1
dienamide 3, and allylglycine 4 (ratio of 1:1:1) showed that the
dienamide rapidly and preferentially coordinated to the ruthe-
nium center forming a [Ru]-vinylalkylidene complex 10 (Figure
1, spectrum a), which slowly transformed into a more stable
ruthenium species 11 (spectrum b) with the elimination of
tricyclohexylphosphine. This transformation was complete after
2 h and the resulting complex was unreactive over an 18 h period
(spectrum c).21 Unfortunately, attempts to isolate this complex
were unsuccessful. One possible explanation for these results
is that the first formed [Ru]-vinylalkylidene slowly coordinates
to the ester carbonyl group leading to the formation of the cyclic
compound 11.
Furthermore, attempts to regenerate the dienamide (3) from
the Ru-carbonyl chelate (11), via reaction with ethyl vinyl ether
and formation of the Fischer-type carbene complex,22 failed due
to conjugate addition of liberated tricyclohexylphosphine to the
dienamide substrate (3). This highlighted the sensitivity of 3 to
N- and P-based nucleophiles and potential problems arising
glycine derivative 6 in the presence of unsaturated dimer 5
proceeded with Grubbs’ catalyst to afford dimer 7 (Scheme 3).
No mixed cross metathesis product 8 was observed. Use of
second generation Grubbs’ catalyst did, however, lead to a
mixture of cross metathesis products 7 and 8. Homogeneous
hydrogenation of dimer 5 with Wilkinson’s catalyst, Rh(I)-
(PPh3)3Cl, under mild experimental conditions, gave the satu-
rated dicarba analogue 9 in quantitative yield (Scheme 2).
These results looked very promising; however, our attempts
(21) Fu¨rstner, A.; Thiel, O. R.; Lehmann, C. W. Organometallics 2002,
21, 331-335.
(22) Louie, J.; Grubbs, R. H. Organometallics 2002, 21, 2153-2164.
7540 J. Org. Chem., Vol. 71, No. 20, 2006

Controlled Synthesis of (S,S)-2,7-Diaminosuberic Acid
SCHEME 6
SCHEME 7
during peptide synthesis, where piperidine is routinely used to
facilitate FMOC-cleavage from residues prior to coupling.
A revised strategy was then investigated centering on the use
of nonproteinaceous, terminally functionalized allylglycine units.
This alternate route involved (i) metathesis of allylglycine units
in the presence of the phenyl-substituted dienamide 12 and (ii)
subsequent hydrogenation of the dienamide 12 to yield a more
reactive olefin 13 for the second ring closing metathesis (Scheme
4). We postulated that the presence of a phenyl substituent at
the olefin terminus might impede binding of the metathesis
catalyst. Significantly, 1H NMR binding studies of 1:1 mixtures
of Ru-benzylidene catalyst and dienamide 12 showed no
alkylidene formation. Hence, the poor chelating properties of
the modified dienamide 12 to Grubbs’ catalyst now facilitated
dimerization (28%) of allylglycine 4 to dimer 5 (Scheme 4).
Rh(I)-DuPHOS catalyzed asymmetric hydrogenation of diena-
mide 12 under mild conditions (75 psi H2) gave 13 in 97% yield
and 99% ee. Disappointingly, cross metathesis of 13 with
Grubbs’ catalyst was unsuccessful. After 13 h, 1H NMR
spectroscopy showed no conversion to the desired dimer.
Conditions to facilitate CM were found by using the more
reactive second generation Grubbs’ catalyst. After treatment of
13 with a 5 mol % solution of second generation Grubbs’
catalyst over 20 h, 33% conversion to 5 was achieved. This
reaction was not optimized, however, because use of this catalyst
makes the previously formed unsaturated carbocycle vulnerable
to further CM and would result in mixed cross metathesis
products.
genation of the unsaturated dimer 5 (Scheme 4). Unfortunately,
these conditions resulted in reduction of all CdC double bonds
to give 14 (Scheme 4). This disappointing observation was
perhaps not unexpected. The rate of olefin reduction by
Wilkinson’s catalyst is profoundly influenced by steric hindrance
about the CdC double bond, but related reductions involving
styrene have previously shown that electronic effects override
these steric effects and that the aromatic substituent enhances
the rate of reduction.23
This led to a final revision of our strategy which would enable
the hydrogenation of an unsaturated carbocycle in a peptide
while maintaining a reactive olefin for metathesis (Scheme 5).
We decided to capitalize on the slow reactivity of trisubstituted
olefins to Wilkinson’s hydrogenation and their reduced reactivity
to metathesis. 1,1-Disubstituted olefins, for example, do not
undergo homodimerization and only react with more reactive
olefins.17
This differential reactivity would therefore facilitate
the CM of allylglycine units and subsequent hydrogenation
without interference from the 1,1-disubstituted olefin residues.
A simple transformation can then render the trisubstituted olefin
more reactive to metathesis and complete the second cyclization.
The prenyl olefin 15 was prepared via asymmetric hydroge-
nation of the corresponding dienamide 16 in quantitative yield
and excellent enantioselectivity (Scheme 6). This enamide 15
was subjected to the hydrogenation conditions that quantitatively
reduce the dimer 5 to 9 and, encouragingly, 94% of the starting
enamide 15 was recovered (Scheme 6). This was a very
promising result that prompted us to further investigate cross
metathesis reactions involving this substrate 15.
Reduction of the first-formed unsaturated carbocycle prior
to the second metathesis reaction would, however, eliminate
the chance of mixed cross metathesis (Scheme 5). We therefore
subjected the phenyl-substituted olefins 12 and 13 to the
hydrogenation conditions previously developed for the hydro-
(23) (a) Osborn, J. A.; Jardine, F. H.; Young, J. F.; Wilkinson, G. J.
Chem. Soc. A 1966, 1711-1736. (b) Jardine, J. H.; Osborn, J. A.; Wilkinson,
G. J. Chem. Soc. A 1967, 1574-1580.
J. Org. Chem, Vol. 71, No. 20, 2006 7541

Elaridi et al.
SCHEME 8
Cross metathesis of allylglycine unit 4 to dimer 5 in the
presence of the prenyl enamide 15 proceeded smoothly in
excellent yield (Scheme 6); the starting prenyl enamide 15 was
recovered unchanged. The prenyl compound 15 was subjected
to ethenolysis to convert it to the more reactive allylglycine
derivative 4 (Scheme 6). Exposure of 15 to 20 mol % of second
generation Grubbs’ catalyst under an atmosphere of ethylene
(15 psi) resulted in only poor conversions to 4 (<32%). We
postulated that this result may be due to the unstable nature of
the in situ generated ruthenium-methylidene intermediate at
elevated temperature24 or unfavorable competition between the
rising concentration of terminal olefins and 15 for binding to
the ruthenium catalyst.25 To circumvent this problem, the prenyl
enamide 15 was instead exposed to an atmosphere of cis-2-
butene (15 psi) thereby facilitating the catalysis via the more
stable [Ru]-ethylidene complex. Butenolysis of 15 in the
presence of 5 mol % second generation Grubbs’ catalyst gave
the expected crotylglycine derivative 17 in quantitative yield.
This olefin was then readily cross metathesized to the expected
dimer 5 with 5 mol % of second generation Grubbs’ catalyst
(Scheme 6).
Finally, a one-pot equimolar mixture of olefins 15 and 6 was
exposed to a tandem sequence of the five homogeneous catalytic
reactions described above (Scheme 7). Quantitative conversion
of the reactive substrate occurred in each step and ultimately
yielded 9 and 18 as the only products in 81% and 73% isolated
yields, respectively. No intermediate isolation was conducted
during this catalytic sequence and no reaction between 15 and
6 was observed.
The above-described methodology was last applied to a
simple resin-bound pentapeptide sequence 19. Standard solid-
phase peptide synthesis, using HATU-NMM activation and
Fmoc-protected amino acids, was used to construct the linear
peptide sequence 19 on Wang resin (Scheme 8). This sequence
possesses three nonproteinaceous residues, two L-allylglycine
(Hag) residues and one L-prenylglycine (Pre) residue, to facilitate
construction of two dicarba bridges. Linear peptide intermediates
were carried through without purification or characterization
up to the pentapeptide 19. A sample of the linear peptide was
obtained by cleavage from the resin and determined to be of
>95% purity by reverse-phase analytical chromatography. Mass
spectral analysis showed the required molecular ion peak for
19 at m/z 813.5 [M + H+].
The first catalytic step involved selective ring closing
metathesis of the allylglycine residues in the presence of the
less reactive prenyl side chain. RCM of the resin-tethered
pentapeptide was performed with 40 mol % second generation
Grubbs’ catalyst in dichloromethane and 10% lithium chloride
in dimethylformamide and, as expected, incorporation of
prenylglycine did not hinder cyclization (Scheme 8). Mass
spectral analysis of a cleaved aliquot of peptide confirmed
formation of the intramolecular dicarba bond and unsaturated
carbocycle 20 with the appearance of a molecular ion peak at
m/z 785.4 (M + H)+. Importantly, the prenylglycine residue
remained inert to these metathesis conditions and no mixed cross
metathesis products were observed. Attempts to decrease
reaction time and catalyst loading led to incomplete reaction;
however, decreasing peptide loading on the resin (from 0.9 to
0.3 mmol g-1) enabled complete RCM with 10 mol % of second
generation Grubbs’ catalyst. Selective hydrogenation of the
resin-bound unsaturated carbocycle 20 was performed under 80
psi of H2 with homogeneous Wilkinson’s catalyst, Rh(I)P-
(PPh3)3Cl, in a mixture of dichloromethane: methanol (9:1).
After 22 h, a small aliquot of peptide was cleaved and analyzed
by mass spectrometry. The appearance of a peak at m/z 787.3
(M + H)+ was consistent with formation of the saturated
carbocycle 21. Importantly, the prenyl group remained stable
to these reducing conditions and was consistent with the
observed reactivity of prenylglycine in the solution phase model
studies.
Activation of the prenyl group was achieved via butenolysis
of the resin-bound pentapeptide 21. The peptide was exposed
to an atmosphere of cis-2-butene (15 psi) and 40 mol % second
generation Grubbs’ catalyst in dichloromethane for 62 h. This
led to a mixture of the desired product 22 and the starting peptide
21. This activation step was unexpectedly and inexplicably slow.
The recovered resin-peptide was re-subjected to the butenolysis
conditions to achieve complete conversion to the target crotyl-
(24) (a) Schwab, P.; Grubbs, R. H.; Ziller, J. W. J. Am. Chem. Soc. 1998,
118, 100-110. (b) Adlhart, C.; Chen, P. J. Am. Chem. Soc. 2004, 126,
3496-3510. (c) Hong, S. H.; Day, M. W.; Grubbs, R. H. J. Am. Chem.
Soc. 2004, 126, 7414-7415.
(25) Burdett, K. A.; Harris, L. D.; Margl, P.; Maughon, B. R.; Mokhtar-
Zadeh, T.; Saucier, P. C.; Wasserman, E. P. Organometallics 2004, 23,
2027-2047.
7542 J. Org. Chem., Vol. 71, No. 20, 2006

Controlled Synthesis of (S,S)-2,7-Diaminosuberic Acid
30QC5/BPX5, 150 °C for 1 min, 10 °C min-1 to 280 °C for 6
min). νmax (neat): 3258 m, 3009 w, 2956 w, 1729 s, 1663 s, 1610
m, 1560 w, 1522 s, 1440 m, 1374 m, 1338 m, 1286 s, 1255 s,
1208 s, 1156 m, 1123 s, 1041 w, 1016 m, 687 m, 896 w, 868 m,
768 s, 716 m, 658 w, 603 m, 581 w, 603 m, 561 w cm-1. 1H NMR
(400 MHz, CDCl3): δ 1.89 (s, 6H), 2.13 (s, 3H), 3.77 (s, 3H),
5.95 (d, J ) 11.8 Hz), 6.97 (br s, 1H), 7.34 (br d, J ) 11.8 Hz).
13C NMR (100 MHz, CDCl3): δ 19.3, 23.6, 27.1, 52.4, 120.8,
121.1, 130.5, 166.0, 168.0. HRMS (ESI+, MeOH): m/z calcd for
C10H15NO3Na [(M + Na)+] 220.0950, found 220.0947.
General Asymmetric Hydrogenation Procedure. In a drybox,
a Fischer-Porter tube was charged with catalyst (1-3 mg),
deoxygenated solvent (∼5 mL), and substrate (28-108 mg). Three
vacuum/argon cycles to purge the gas line of any oxygen followed
by three vacuum/argon cycles of the vessel were carried out before
the tube was pressurized with hydrogen to the specified pressure
(psi). The reaction was then stirred at room temperature for the
specified period of time. The pressure in the vessel was then
released, and the contents were evaporated under reduced pressure.
The crude product was passed through a short plug of silica (eluent
) ethyl acetate) prior to spectroscopic and chromatographic
analysis. Hydrogenation experiments are described by using the
following format: substrate, solvent, catalyst, hydrogen pressure,
reaction time, isolated yield, enantiomeric excess (assigned con-
figuration), retention time (GC conditions).
glycine-containing peptide 22. Mass spectral analysis of a
cleaved aliquot displayed the product molecular ion peak at m/z
773.2 (M + H)+. A final cross metathesis reaction between the
activated-resin bound peptide 22 and crotylglycine derivative
1726 was then performed. Microwave irradiation of a mixture
of resin-tethered peptide 22 with 40 mol % second generation
Grubbs’ catalyst, excess crotylglycine derivative 17 (∼50 equiv)
in dichloromethane, and 10% lithium chloride in dimethylfor-
mamide resulted in formation of the second dicarba bond, an
intermolecular linkage, and unsaturated peptide 23 accompanied
by dimer 5. Mass spectrometry confirmed product formation
of 23 with the appearance of a molecular ion peak at m/z 902.3
(M + H)+. Wilkinson’s hydrogenation of the unsaturated
intermolecular bridge was achieved under conditions previously
established (80 psi of H2, dichloromethane:methanol (9:1), room
temperature, 18 h) to give the target peptide 24 containing two
selectively constructed dicarba bridges.
Conclusions
In conclusion, these model studies demonstrate that through
the combination of homogeneous catalysis and judicious selec-
tion of nonproteinaceous allylglyine residues of varying reactiv-
ity, a highly efficient, unambiguous, and regioselective synthesis
of dicarba analogues of multicystine-containing peptides is
achievable. The high selectivity, efficiency, and generic nature
of this catalytic sequence, coupled with the accessibility of both
allyl-27and prenylglycine,28 is likely to ensure the widespread
applicability of this approach to peptide synthesis. Furthermore,
the disparate reactivity of the allyl and prenyl groups toward
Ru- and Rh-catalyzed metathesis and hydrogenation could be
further exploited in syntheses involving nonpeptidic substrates.
We are currently applying this methodology to the synthesis of
natural products and dicarba mimics of cystine-containing
peptides and exploring tandem catalytic sequences for achieving
the same end.
(2S)-Methyl 2N-acetylaminopent-4-enoate (4):30 (2Z)-Methyl
2N-acetylaminopenta-2,4-dienoate (3) (108.0 mg, 0.64 mmol),
benzene (7 mL), [(COD)Rh(S,S)-Et-DuPHOS]OTf (3 mg), 30 psi
H2, 3 h, 97% yield, 97% ee (2S-4), tR ) 16.5 min (S) (GC chiral
column Model C-024, 100 °C for 1 min, 5 °C min-1 to 280 °C for
9 min). [R]22D +45.0° (c 0.76, CHCl3). Spectral data were consistent
with literature data.30
General Metathesis Procedure. A Schlenk flask was charged
with catalyst (5-20 mol %), deoxygenated solvent (∼5 mL), and
substrate (10-60 mg). The reaction mixture was left to stir at 50
°C for the specified period of time. Metathesis reactions were
terminated upon exposure to oxygen and volatile species removed
under reduced pressure. The crude product was purified by flash
chromatography. Metathesis experiments are described by using
the following format: substrate, solvent, catalyst, reaction time,
reaction temperature, percent conversion. Chromatographic puri-
fication conditions (isolated yield).
Experimental Section
Preparation of Dienamides. Dienamides 3 and 12 were prepared
as described by Teoh et al.16 and Burk et al.,29 respectively.
(2Z)-Methyl 2N-acetylamino-5-methylhexa-2,4-dienoate (16):
Dienamide 16 was prepared according to modified literature
procedures.16,29 Tetramethylguanidine (0.78 mL, 6.22 mmol) was
added to a solution of methyl 2N-acetylamino-2-(dimethoxyphos-
phinyl)acetate (1.11 g, 4.64 mmol) in distilled tetrahydrofuran (50
mL) at -78 °C. After 15 min, 3-methyl-2-butenal (0.54 mL, 5.60
mmol) was added and the mixture was stirred for 2 h at -78 °C.
The mixture was warmed to 25 °C with a warm water bath and
stirred at this temperature for an additional 18 h. The reaction
mixture was then diluted with dichloromethane (100 mL) and
washed with dilute hydrochloric acid solution (1 M, 2 × 75 mL),
copper sulfate solution (1 M, 2 × 75 mL), saturated sodium
bicarbonate solution (2 × 75 mL), and saturated sodium chloride
solution (1 × 75 mL). The organic layer was dried (MgSO4) and
evaporated under reduced pressure to give an oil (0.79 g).
Purification by flash chromatography (SiO2, dichloromethane:ethyl
acetate:light petroleum, 2:1:1) gave dienoate 16 (0.61 g, 67%) as
a pale brown solid, mp 115-116 °C. tR ) 6.3 min (GC column
(2S,7S)-Dimethyl 2,7-N,N′-diacetylamino-oct-4-enedioate
(5):32 (2S)-Methyl 2N-acetylaminopent-4-enoate (4) (14.0 mg, 0.08
mmol), dichloromethane (4 mL), Grubbs’ catalyst (13.5 mg, 0.02
mmol, 20 mol %), 24 h, 50 °C, 100% conversion. Purification by
flash chromatography (SiO2, dichloromethane:light petroleum:ethyl
acetate, 1:1:1) furnished pure dimer 5 (11.4 mg, 44%) as a brown
oil, tR(E/Z) ) 12.2, 12.7 min (GC column 30QC5/BPX5, 150 °C
for 1 min, 10 °C min-1 to 280 °C for 6 min). [R]22D +92.0 (c 0.004,
CHCl3). νmax (neat): 3286 br m, 2956 m, 2931 m, 2856 w, 1742 s,
1659 s, 1542 m, 1438 m, 1375 m, 1267 m, 1220 m, 1138 w, 1017
w cm-1. 1H NMR (300 MHz, CDCl3): δ 2.04 (s, 6H), 2.40-2.50
(m, 4H), 3.74 (s, 6H), 4.64-4.70 (m, 2H), 5.36-5.40 (m, 2H),
6.34 (br d, J ) 7.2 Hz). 13C NMR (100 MHz, CDCl3): δ 23.1,
35.1, 51.7, 52.6, 128.8, 170.3, 172.6. HRMS (ESI+, MeOH): m/z
calcd for C14H22N2O6Na [(M + Na)+] 337.1376, found 337.1375.
Spectral data were consistent with literature data.32
Competitive Cross Metathesis Studies: Cross Metathesis of
(2S)-Methyl 2N-Benzoylaminopent-4-enoate (6) in the Presence
(30) Cox, R. J.; Sherwin, W. A.; Lam, L. K. P.; Vederas, J. C. J. Am.
Chem. Soc. 1996, 118, 7449-7460.
(26) Robinson, A. J.; Elaridi, J.; Patel, J.; Jackson, W. R. Chem. Commun.
2005, 5544-5545.
(27) Guillena, G.; Najera, C. J. Org. Chem. 2000, 65, 7310-7322.
(28) Smith, A. B.; Benowitz, A. B.; Favor, D. A.; Sprengeler, P. A.;
Hirschmann, R. Tetrahedron Lett. 1997, 38, 3809-3812.
(29) Burk, M. J.; Allen, J. G.; Kiesman, W. F. J. Am. Chem. Soc. 1998,
120, 657-663.
(31) Tanaka, K.; Ahu, M.; Watanabe, Y.; Fuji, K. Tetrahedron: Asym-
metry 1996, 7, 1771-1782.
(32) Bremner, J. B.; Keller, P. A.; Pyne, S. G.; Robertson, A. D.; Skelton,
B. W.; White, A. H.; Witchard, H. M. Aust. J. Chem. 2000, 53, 2027-
2047.
(33) Pattabiraman, T. N.; Lawson, W. B. Biochem. J. 1972, 126,
645-657.
J. Org. Chem, Vol. 71, No. 20, 2006 7543

Elaridi et al.
of (2S,7S)-Dimethyl-2,7-N,N′-diacetylamino-oct-4-enedioate (5).
(2S)-Methyl 2N-benzoylaminopent-4-enoate(6) (37.0 mg, 0.16
mmol), (2S,7S)-dimethyl 2,7-N,N′-diacetylamino-oct-4-enedioate (5)
(30.0 mg, 0.09 mmol), dichloromethane (4 mL), Grubbs’ catalyst
(26.1 mg, 0.03 mmol, 20 mol %), 18 h, 50 °C, 100% conversion
of 6 into dimer 7. Purification by flash chromatography (SiO2,
dichloromethane:ethyl acetate:light petroleum, 2:2:1) gave pure
dimers 5 (30.0 mg, 100% recovery) and 7 (43.8 mg, 63% yield) as
brown oils. Spectral data were in agreement with those previously
obtained.
as a brown oil. tR(E/Z) ) 4.2 min, 4.4 min (GC column 30QC5/
BPX5, 150 °C for 1 min, 10 °C min-1 to 280 °C for 6 min). νmax
(neat): 3284 s, 2966 w, 2954 m, 2856 w, 1747 s, 1658 s, 1547 s,
1437 s, 1375 s, 1217 m, 1142 m, 1072 w, 1016 w, 968 m, 848 m
1
cm-1. H NMR (300 MHz, CDCl3): δ 1.60 (dd, J ) 6.3, 1.2 Hz,
3H), 1.95 (s, 3H), 2.36-2.44 (m, 2H), 3.67 (s, 3H), 4.55 (dt, J )
7.8 Hz, 5.9 Hz, 1H), 5.16-5.31 (m, 1H), 5.40-5.57 (m, 1H), 6.17
(br d, J ) 6.4 Hz, 1H). 13C NMR (100 MHz, CDCl3): δ 18.1,
23.3, 35.4, 52.1, 52.4, 124.6, 130.2, 169.7, 172.6. Mass spectrum
(ESI+, MeOH): m/z 208.1 C9H15NO3Na [(M + Na)+]. Spectro-
scopic data were consistent with literature data.16,29
NMR Studies of Cross Metathesis Reactions. (a) NMR Study
of Grubbs’ Catalyst with Dienamide 3. In a drybox, a Teflon-
sealed NMR tube was charged with (2S)-methyl 2N-acetylamino-
penta-2,4-dienoate (3) (21.0 mg, 0.12 mmol), Grubbs’ catalyst (102
mg, 0.12 mmol), and degassed deuterated dichloromethane (0.8 mL)
at room temperature. The NMR tube was shaken gently and reaction
progress was monitored by 1H and 31P NMR spectroscopy.
Compounds were identified by the following diagnostic resonances.
1H NMR (300 MHz, CD2Cl2): After 15 min: Grubbs’ catalyst, δ
8.61 (d, J ) 7.6 Hz, 2H), 20.05 (s, 1H); ruthenium-dienamide
complex 10, δ 7.96 (d, J ) 11.0 Hz, 1H), 20.11 (d, J ) 11.0 Hz,
1H); ruthenium-dienamide chelate 11 (trace amount), δ 15.20 (d,
J ) 4.2 Hz, 1H). Ratio of ruthenium complexes [Ru]dCHPh:10:
11 ) 1:1:0.08. After 60 min: Grubbs’ catalyst, δ 8.45 (d, J ) 7.6
Hz, 2H), 20.04 (s, 1H); ruthenium-dienamide complex 10, δ 7.96
(d, J ) 11.0 Hz, 1H), 20.10 (d, J ) 11.0 Hz, 1H); ruthenium-
dienamide chelate 11, δ 6.73 (d, J ) 3.0 Hz, 1H) (peak obscured
by liberated styrene), 15.19 (d, J ) 4.2 Hz, 1H). Ratio of ruthenium
complexes [Ru]dCHPh:10:11 ) 3:1:1. After 120 min: ruthenium-
dienamide chelate 11: δ 6.71 (d, J ) 3.0 Hz, 1H) (peak obscured
by liberated styrene), 15.19 (d, J ) 4.0 Hz, 1H). 31P NMR (300
MHz, CDCl3): δ 35.0, 11; 37.0, Grubbs’ catalyst; 38.8, 10.
(b) NMR Study of Grubbs’ Catalyst with Dienamide 12. In
a drybox, a Teflon-sealed NMR tube was charged with (2S)-methyl
2N-acetylamino-5-phenylpenta-2,4-dienoate (12) (10.0 mg, 0.04
mmol), Grubbs’ catalyst (33.6 mg, 0.04 mmol), and degassed
deuterated dichloromethane (0.8 mL) at room temperature. The
NMR tube was shaken gently and reaction progress was monitored
by 1H NMR spectroscopy. After 4 h, no alkylidene formation was
evident and only peaks corresponding to Grubbs’ catalyst and the
starting dienamide 12 were present.
Pentapeptide Transformations. (a) Cyclic Pentapeptide:
Fmoc-c[Hag-Pro-Pre-Arg-Hag]-OH (20). The resin-bound peptide
19 was subjected to the conventional RCM procedure under the
following conditions: Resin-peptide 19 (70.0 mg, 0.06 mmol),
dichloromethane (DCM) (5 mL), LiCl/DMF (0.4 M, 0.5 mL),
second generation Grubbs’ catalyst (21.6 mg, 0.03 mmol, 40 mol
%), 50 °C, 42 h, 100% conversion to 20. At the end of the reaction
period, a small aliquot of peptidyl-resin was subjected to the
standard TFA-mediated cleavage procedure. Mass spectral analysis
of the isolated residue confirmed formation of the cyclic peptide
20. Mass spectrum (ESI+, MeCN/H2O): m/z 785.4 (M + H)+,
C41H53N8O8 requires 785.4; m/z 803.3 (M + H2O + H)+,
C41H55N8O9 requires 803.4; m/z 899.4 (M + TFA + H)+,
C43H54F3N8O10 requires 899.4.
(b) Reduced Cyclic Pentapeptide: Fmoc-rc[Hag-Pro-Pre-Arg-
Hag]-OH (21). The resin-bound peptide 20 was subjected to the
general Wilkinson’s hydrogenation procedure under the following
conditions: Resin-peptide 20 (350 mg, 0.32 mmol), DCM:MeOH
(9:1, 8 mL), Wilkinson’s catalyst, 80 psi of H2, 22 °C, 22 h, 100%
conversion to 21. At the end of the reaction period, a small aliquot
of peptidyl-resin was subjected to the standard TFA-mediated
cleavage procedure. Mass spectral analysis of the isolated residue
confirmed formation of the reduced cyclic pentapeptide 21. Mass
spectrum (ESI+, MeCN/H2O): m/z 787.2 (M + H)+, C41H55N8O8
requires 787.4; m/z 805.2 (M + H2O + H)+, C41H57N8O9 requires
803.4; m/z 901.3 (M + TFA + H)+, C43H56F3N8O10 requires
901.4.
(c) Reduced Cyclic PentapeptidesOlefin Activation: Fmoc-
rc[Hag-Pro-Crt-Arg-Hag]-OH (22). The resin-bound peptide 21
was subjected to the general conditions for CM with cis-but-2-ene
under the following conditions: Resin-peptide 21 (212 mg, 0.19
mmol), DCM (8 mL), second generation Grubbs’ catalyst (82 mg,
9.7 µmol, 50 mol %), cis-but-2-ene (15 psi), 50 °C, 42 h. At the
end of the reaction period, a small aliquot of peptidyl-resin was
subjected to the cleavage procedure. Mass spectral analysis of the
isolated residue indicated the presence of the starting peptide and
the desired butenolysis product. The recovered resin-peptide was
subjected to the same butenolysis conditions in order to drive the
reaction to completion. After 42 h, a small aliquot of peptidyl-
resin was subjected to the cleavage. Mass spectral analysis of the
isolated residue confirmed quantitative conversion to the activated
peptide 22. Mass spectrum (ESI+, MeCN/H2O): m/z 773.2 (M +
H)+, C40H53N8O8 requires 773.4.
(d) Reduced Cyclic PentapeptidesCM of Activated Olefin:
Fmoc-rc[Hag-Pro-(N-Ac-Dehydro-Sub-OMe)-Arg-Hag]-OH (23).
The resin-bound peptide 22 was subjected to the general microwave-
accelerated CM under the following conditions: Resin-peptide 22
(20.0 mg, 18 µmol), DCM (4 mL), LiCl/DMF (0.4 M, 0.4 mL),
second generation Grubbs’ catalyst (6.2 mg, 7.3 µmol, 40 mol %),
(2S)-methyl 2-N-acetylaminohex-4-enoate (17) (70.0 mg, 0.38
mmol), 100 °C, 2 h. At the end of the reaction period, a small
aliquot of peptidyl-resin was subjected to the TFA-mediated
cleavage procedure. Mass spectral analysis of the isolated residue
confirmed formation of the CM product, peptide 23. Mass spectrum
(ESI+, MeCN/H2O): m/z 902.4 (M + H)+, C45H60N9O11 requires
902.4.
Homogeneous Hydrogenation Reactions. (a) Wilkinson’s
Hydrogenation of Dimer 5: (2S,7S)-Dimethyl 2,7-N,N′-Diacetyl-
aminooctanedioate (9). (2S,7S)-Dimethyl-2,7-N,N′-diacetylami-
nooct-4-enedioate (5) (25.0 mg, 0.08 mmol), benzene (5 mL),
Wilkinson’s catalyst (2 mg), 50 psi H2, 4 h, 99% yield of 9 as a
brown oil. tR ) 14.4 min (GC column 30QC5/BPX5, 150 °C for 1
min, 10 °C min-1 to 280 °C for 6 min). νmax (neat): 3426 br m,
3055 w, 2932 m, 2857 w, 2360 w, 1741 s, 1666 s, 1543 w, 1438
m, 1375 w, 1266 s, 1177 w, 1120 w, 896 w, 738 w, 702 w cm-1
.
1H NMR. (300 MHz, CDCl3): δ 1.30-1.40 (m, 4H), 1.82-1.90
(m, 4H), 2.02 (s, 6H), 3.74 (s, 3H), 4.56-4.63 (m, 2H), 6.16 (bd,
J ) 7.5 Hz, 2H). 13C NMR (100 MHz, CDCl3): δ 23.3, 24.7, 32.3,
52.0, 52.5, 170.0, 173.1. HRMS (ESI+, MeOH): m/z calcd for
C14H24N2O6Na [(M + Na)+] 339.1532, found 339.1531.
(b) Ethenolysis of (2S)-Methyl 2N-Acetylamino-5-methylhex-
4-enoate (15). (2S)-Methyl 2N-acetylamino-5-methylhex-4-enoate
(15) (24.3 mg, 0.12 mmol), dichloromethane (5 mL), second
generation Grubbs’ catalyst (31.1 mg, 0.04 mmol, 30 mol %), 60
psi ethylene, 38 h, 50 °C, 32% conversion into 4.
(c) Butenolysis of (2S)-Methyl 2N-Acetylamino-5-methylhex-
4-enoate (15). (2S)-Methyl 2N-acetylamino-5-methylhex-4-enoate
(15) (16.2 mg, 0.08 mmol), dichloromethane (5 mL), second
generation Grubbs’ catalyst (3.5 mg, 0.004 mmol, 5 mol %), 15
psi cis-2-butene, 15 h, 100% conversion into (2S)-methyl 2N-
acetylaminohex-4-enoate (17). Purification by flash chromatography
(SiO2, dichloromethane:ethyl acetate:light petroleum:methanol, 1:2:
1:0.2) gave the crotylglycine derivative 1716,26 (20.3 mg, 84% yield)
(e) Reduced Cyclic PentapeptidesWilkinson’s Hydrogena-
tion: Fmoc-rc[Hag-Pro-(N-Ac-Sub-OMe)-Arg-Hag]-OH (24).
7544 J. Org. Chem., Vol. 71, No. 20, 2006

Controlled Synthesis of (S,S)-2,7-Diaminosuberic Acid
The resin-bound peptide 23 was subjected to the general Wilkin-
son’s hydrogenation procedure under the following conditions:
Resin-peptide 23 (15.0 mg, 14 µmol), DCM:MeOH (9:1, 5 mL),
Wilkinson’s catalyst, 80 psi of H2, 22 °C, 22 h, 100% conversion
to 24. At the end of the reaction period, a small aliquot of
peptidyl-resin was subjected to the TFA-mediated cleavage.
Mass spectral analysis of the isolated residue confirmed for-
mation of the reduced cyclic pentapeptide 24. Mass spectrum
(ESI+, MeCN/H2O): m/z 904.4 (M + H)+, C45H62N9O11 requires
904.5.
Acknowledgment. The corresponding author expresses
gratitude to the talented and dedicated co-workers whose names
appear on this paper. Thanks are also due to the Australian
Research Council for financial support of this research.
Supporting Information Available: Experimental procedures
and spectral data for compounds 3-7, 9, and 12-19. This material
is available free of charge via the Internet at http://pubs.acs.org.
JO0606913
J. Org. Chem, Vol. 71, No. 20, 2006 7545