Convenient access to glutamic acid side chain homologues compatible with solid phase peptide synthesis

Article abstract of DOI:10.1021/ol0520222

(Chemical Equation Presented) Preparation of several side chain length variants of glutamic acid is achieved via olefin cross metathesis of allyl glycine derivatives. The products are suitably protected for direct use in Fmoc solid-phase peptide synthesis, as demonstrated by successful synthesis of test sequences.

Full text of DOI:10.1021/ol0520222

ORGANIC
LETTERS
2005
Vol. 7, No. 21
4765-4767
Convenient Access to Glutamic Acid
Side Chain Homologues Compatible
with Solid Phase Peptide Synthesis
Shannon J. Ryan, Yongda Zhang, and Alan J. Kennan*
Department of Chemistry, Colorado State UniVersity, Fort Collins, Colorado 80523
kennan@lamar.colostate.edu
Received August 22, 2005
ABSTRACT
Preparation of several side chain length variants of glutamic acid is achieved via olefin cross metathesis of allyl glycine derivatives. The
products are suitably protected for direct use in Fmoc solid-phase peptide synthesis, as demonstrated by successful synthesis of test sequences.
The utility of olefin metathesis reactions in synthetic organic
chemistry has been validated across a broad expanse of
molecular architectures, from the most simple to the most
dauntingly complex.1 The resulting efficient routes to previ-
ously inconvenient targets have facilitated numerous inves-
tigations. In the course of peptide molecular recognition
studies, we sought to prepare glutamic acid analogues of
varying side chain length. Here we report the use of olefin
cross metathesis reactions to obtain derivatives having
between one and three extra side chain methylenes from a
common precursor, and their successful incorporation into
peptides via Fmoc solid-phase methods.
The increased flexibility in protein design conferred by
unnatural amino acid incorporation has numerous benefits,
both structural and functional.2 We recently undertook an
investigation into the impact of side chain length on complex
stability in self-assembling peptide systems, particularly with
respect to interfacial interactions of glutamic acid-lysine
pairs. Although the appropriate lysine analogues are readily
commercially available, the glutamic acid variants must be
independently synthesized. As such, we explored several
alternatives, hoping to identify a route that was sufficiently
modular to permit preparation of several different targets,
efficient enough to generate multigram quantities, and
compatible with standard side chain protecting groups for
direct use in solid-phase peptide synthesis.
The use of allyl glycine as a common synthetic intermedi-
ate addresses all of these goals. It is easily prepared on gram
scale and further elaborated via olefin cross metathesis. The
documented use of allyl glycines in diverse metathesis reac-
tions, and the prospects for facile access to suitable coupling
partners, further supported the viability of this route.3
Allyl glycine was synthesized and N-protected according
to methods developed by Myers and co-workers (Scheme
(3) (a) Nolen, E. G.; Kurish, A. J.; Potter, J. M.; Donahue, L. A.; Orlando,
M. D. Org. Lett. 2005, 7, 3383-3386. (b) Boyle, T. P.; Bremner, J. B.;
Coates, J. A.; Keller, P. A.; Pyne, S. G. Tetrahedron 2005, 61, 7271-
7276. (c) Nolen, E. G.; Kurish, A. J.; Wong, K. A.; Orlando, M. D.
Tetrahedron Lett. 2003, 44, 2449-2453. (d) Biswas, K.; Coltart, D. M.;
Danishefsky, S. J. Tetrahedron Lett. 2002, 43, 6107-6110. (e) Dondoni,
A.; Giovannini, P. P.; Marra, A. J. J. Chem. Soc., Perkin Trans. 1 2001,
2380-2388. (f) 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, 535-540. (g) Biagini, S. C. G.; Gibson, S. E.; Keen, S. P. J. Chem.
Soc., Perkin Trans. 1 1998, 2485-2499.
(1) (a) Nicolaou, K. C.; Bulger, P. G.; Sarlah, D. Angew. Chem., Int.
Ed. 2005, 44, 4490-4527. (b) Grubbs, R. H. Tetrahedron 2004, 60, 7117-
7140. (c) Schrock, R. R.; Hoveyda, A. H. Angew. Chem., Int. Ed. 2003,
42, 4592-4633.
(2) (a) Wang, L.; Schultz, P. G. Angew. Chem., Int. Ed. 2005, 44, 34-
66. (b) Hendrickson, T. L.; de Crecy-Lagard, V.; Schimmel, P. Annu. ReV.
Biochem. 2004, 73, 147-176. (c) Bilgic¸er, B.; Kumar, K. J. Chem. Educ.
2003, 80, 1275-1281.
10.1021/ol0520222 CCC: $30.25
© 2005 American Chemical Society
Published on Web 09/15/2005

appeared to go smoothly, producing a single dominant peak
in the crude HPLC, analysis of the isolated major product
revealed that its molecular weight was 180 mass units too
high (MWobs ) 1495.9, MWcalc ) 1315.6).5 This is consistent
with bis-alkylation of HGlu-derived acylium ions by the
anisole scavenger used during cleavage with anhydrous HF
(Figure 2), a result similar to known side reactions of
glutamic acid residues.6
Scheme 1. Synthesis of Allyl Glycine Derivatives
1).4 Synthesis of pseudoephedrine glycinamide 1 from
pseudoephedrine and glycine methyl ester is straightforward.
Alkylation, hydrolytic removal of the auxiliary, and protec-
tion affords either Boc- or Fmoc-allyl glycine. Benzylation
under standard conditions produces either the Boc (3) or
Fmoc (4) substrate for cross metathesis.
Our initial target was homoglutamic acid (HGlu), prefer-
ably a derivative compatible with Boc peptide synthesis
methods (Scheme 2). Reaction of commercially available
Figure 2. Possible explanation for unsuccessful Boc synthesis.
Acylium ion formation and capture with anisole produces a peptide
whose molecular weight is 90 mass units higher per alkylation.
It is not clear why conditions that routinely produce only
minor if any glutamic acid alkylation should favor near
complete modification of HGlu. The cleavage vessel had
warmed to room temperature during HF removal, and a
repeat experiment using an ice bath throughout did improve
matters, affording the desired product as the major peak.
However, there remained significant contamination from two
other peaks with identical masses consistent with mono-
alkylation (MWobs ) 1405.73).5 As a result, the Boc synthesis
strategy was jettisoned in favor of an Fmoc one.
Scheme 2. Synthesis of BocHGlu
The preferred acid side chain protection for Fmoc synthesis
is the tert-butyl rather than cyclohexyl ester. Thus HGlu
preparation requires commercially available tert-butyl acry-
late as a metathesis coupling partner. To synthesize higher
homologues having two (HHGlu) or three (HHHGlu) me-
thylenes more than glutamic acid, we needed to prepare the
longer chain alkenes. Fortunately, both vinyl and allyl acetic
acid (12, 13) can be converted to the corresponding esters
(14, 15) by reaction with isobutylene and catalytic sulfuric
acid in methylene chloride (Scheme 3).7 Although the yields
cyclohexyl acrylate with 3 in the presence of Grubbs’ second
generation metathesis catalyst 5, followed by hydrogenation,
produced 6 in good yield (81% over two steps) and high
optical purity (methyl ester formed with diazomethane is
>99% ee by chiral HPLC).
Having successfully prepared Boc-HGlu with the side
chain protected as its cyclohexyl ester (a standard protecting
group for Boc peptide synthesis), we undertook its incor-
poration into test peptide 7 (Figure 1). Although the synthesis
Scheme 3. Synthesis of Metathesis Substrates
are only moderate, the low-cost starting materials make this
a feasible route to multigram quantities of the desired esters.
Figure 1. Test peptide sequences, with position of glutamic acid
analogues indicated (E* ) HGlu, E** ) HHGlu, E*** )
HHHGlu). All peptides are C-terminally amidated; 7-10 contain
an N-terminal acetamidobenzoic acid (Aba) group as a spectroscopic
label, while 11 has the Aba group attached to the side chain of the
underlined lysine and is N-terminally acetylated.
(4) (a) Myers, A. G.; Schnider, P.; Kwon, S.; Kung, D. W. J. Org. Chem.
1999, 64, 3322-3327. (b) Myers, A. G.; Gleason, J. L. Org. Synth. 1999,
76, 57-76. (c) Myers, A. G.; Gleason, J. L.; Yoon, T.; Kung, D. W. J. Am.
Chem. Soc. 1997, 119, 656-673.
(5) See Supporting Information for details.
4766
Org. Lett., Vol. 7, No. 21, 2005

With the alkene substrates available, preparation of all
three glutamic acid variants by cross metathesis was inves-
tigated (Scheme 4). In each case, a reasonable yield of the
the context of a more realistically useful sequence, in
particular one that incorporated multiple copies of the new
amino acid. This sequence (11, Figure 1) is derived from
GCN4, a natural coiled-coil protein, and contains no fewer
than 8 copies of HGlu among its 30 residues. Thus it serves
to highlight the feasibility of using these unnatural residues
in longer peptides, and to emphasize the need for routes to
multigram monomer quantities for some applications.
Scheme 4. Synthesis of Fmoc Derivatives
chain extended Fmoc acid was obtained, with the side chain
acid protected as the tert-butyl ester. Each product was
optically pure, as judged by chiral HPLC evaluation of the
corresponding methyl ester (>99% ee).5
To determine the feasibility of peptide synthesis using the
new glutamic acid homologues, a second test sequence was
used (8-10, Figure 1). The shorter sequence was designed
with only one modified residue, to aid in deciphering possible
side products. Fortunately, each synthesis gave the desired
peptide as the dominant product in the crude HPLC (Figure
3).5
Figure 4. A more challenging application. Crude HPLC trace of
the 30 residue peptide 11, containing 8 copies of the unnatural HGlu
residue. The desired peptide is the dominant peak.
As the crude HPLC trace makes clear (Figure 4), even
this relatively stringent test of the new methodology was
easily passed. The observed dominant major product is the
desired peptide.5
In summary, we have developed an efficient and modular
route to chain-extended glutamic acid analogues which are
suitably protected for direct use in Fmoc solid phase peptide
synthesis. The derivatives can be prepared on multigram
scale, allowing incorporation of even multiple copies into
relatively long model peptides. These results pave the way
for facile investigation of side chain length effects in designed
peptide systems.
Figure 3. HPLC analysis of crude test peptides 8-10 (Figure 1).
The major peak in each case is the desired peptide.5
Acknowledgment. This work was supported by an NSF
CAREER award to A.J.K. (CHE-0239275).
Although success with short test peptides was encouraging,
we next sought to validate the synthetic HGlu monomer in
Supporting Information Available: Detailed experi-
mental procedures and peptide characterization. This material
is available free of charge via the Internet at http://pubs.acs.org.
(6) Miranda, L. P.; Jones, A.; Meutermans, W. D. F.; Alewood, P. F. J.
Am. Chem. Soc. 1998, 120, 1410-1420.
(7) Anderson, G. W.; Callahan, F. M. J. Am. Chem. Soc. 1960, 82, 3359.
OL0520222
Org. Lett., Vol. 7, No. 21, 2005
4767