Stereoretentive synthesis and chemoselective amide-forming ligations of C-terminal peptide α-ketoacids

Article abstract of DOI:10.1021/ja800053t

C-Terminal peptide cyanosulfur ylides are readily converted to C-terminal peptide α-ketoacids, poised for chemoselective amide-forming reactions with hydroxylamines. These easily prepared and bench stable ylides are quickly and selectively oxidized with a

Full text of DOI:10.1021/ja800053t

Published on Web 03/12/2008
Stereoretentive Synthesis and Chemoselective Amide-Forming Ligations of
C-Terminal Peptide r-Ketoacids
Lei Ju, Alexander R. Lippert, and Jeffrey W. Bode*
Roy and Diana Vagelos Laboratories, Department of Chemistry, UniVersity of PennsylVania,
Philadelphia, PennsylVania 19104
Received January 3, 2008; E-mail: bode@sas.upenn.edu
Scheme 1. Preparation of R-Ketoacid Peptides via Chemically
and Configurationally Stable C-Terminal Sulfur Ylides
Chemoselective ligation reactions make possible the coupling
of unprotected molecular fragments via selective bond forming
reactions of unique functional groups. Particularly prized are ligation
processes that lead to natural bond connections, such as the elegant
and powerful native chemical ligation of C-terminal thioesters and
N-terminal cysteines,1 which allows access to large peptides and
proteins with synthetic modifications.2 In our own work, we have
recently documented a novel amide-forming reaction3 through the
chemoselective coupling of R-ketoacids4 and hydroxylamines under
mild conditions, without reagents and without the need for side
chain protecting groups (eq 1).5 Although this reaction shows great
Scheme 2. Acyl Cyanides Are Configurationally Labile
Intermediates in the Formation of R-Ketoacids by Oxidation of
Cyanophosphorus or Cyanosulfur Ylides
potential for general chemoselective ligation, a key obstacle to its
practical utilization is the preparation of enantioenriched peptide-
derived R-ketoacids.6
In seeking to develop a strategy that would make possible the
synthesis of enantiopure, side chain deprotected peptide R-ketoacids
under mild, aqueous conditions, we turned to the use of sulfur
ylides, which we had recently found to be useful in a different
context.10,11 Preliminary studies suggested that the C-terminal pep-
tide ylides could be formed under a variety of convenient protocols
without epimerization of the R-stereocenter (Vida infra) and that
both the acylation reactions and subsequent oxidations of the cyano-
sulfur ylide occurred more rapidly and cleanly than the phosphorus
variant. The cyanosulfur ylides were not only rapidly oxidized by
ozone and DMDO12 but also could be transformed into R-ketoacids
using mild and easy to handle Oxone13 under aqueous conditions.
Although Wasserman had previously shown that Oxone could
oxidize certain highly activated phosphorus ylides, these reactions
suffered poor chemoselectivity.8f,14 Furthermore, our peptide-derived
cyanophosphorous ylides were inert to aqueous Oxone.
Sulfur ylide 1 was chosen as a model substrate for the
optimization of the Oxone oxidation conditions.15 Due to the
difficulty in isolating and analyzing the R-ketoacids, which exist
as mixtures of the keto and hydrated forms and can also undergo
decarboxylation upon concentration, we assayed the degree of
epimerization by ligating the resulting R-ketoacid to HONH-Gly-
Phe-OtBu (3) under a set of standard but unoptimized conditions
(Scheme 3). Oxidation conditions were screened with the goal of
maximizing the yield of the R-ketoacid in preference to the
formation of the corresponding carboxylic acid, which arises from
oxidative decarboxylation of the initially formed R-ketoacid,
probably via the generation of hydrogen peroxide.16
A viable method for R-ketoacid synthesis should address three
key challenges: (1) it must provide the C-terminal R-ketoacids
without epimerization; (2) it must be compatible with the prepara-
tion of side chain unprotected peptide R-ketoacids; and (3) it must
interface with established methods and reagents used for iterative
peptide synthesis. Herein, we document the first stereoretentive
synthesis of unprotected C-terminal peptide R-ketoacids through
an operationally friendly approach (Scheme 1).
At the outset of our studies there was, to the best of our
knowledge, no prior report of the synthesis of enantiopure oli-
gopeptide R-ketoacids. In contrast, considerable attention had been
placed on the synthesis of peptide R-ketoesters and R-ketoamides.7
We initially hoped to directly translate a leading method for
R-ketoester formation, Wasserman’s use of cyanophosphorus ylides,
to the preparation of unprotected peptide-derived R-ketoacids.8 The
development of a solid-supported variant rendered this strategy
additionally attractive.9 Although we successfully employed phos-
phorus ylides for the synthesis of protected R-ketoacid monomers,
attempts to extend this chemistry to the synthesis of longer and
unprotected peptide-derived R-ketoacids were met with frustration.
Of particular concern were the conditions required for oxidative
conversion of the cyanophosphorus ylides into the R-ketoacids;
typical protocols required exhaustive ozonolysis at low temperatures
or highly reactive oxidants that were not compatible with unpro-
tected functionality.8b,c We also observed epimerization under these
conditions, which we attributed to slow turnover of acyl cyanide
intermediate II (Scheme 2).
9
10.1021/ja800053t CCC: $40.75 © 2008 American Chemical Society
J. AM. CHEM. SOC. 2008, 130, 4253-4255
4253

C O M M U N I C A T I O N S
Scheme 3. Bench Stable R-Peptide Sulfur Ylides Are Prepared
via Acylations with Bromide Salt 5 with Either Boc or Fmoc
Protection
Table 2. Side Chain Compatibility of Sulfur Ylide Oxidation and
Ligation
Table 1. Optimization of Oxone Oxidation of Sulfur Ylide 1 to
R-Ketoacid 2
b
a Oxidations were conducted on a 0.05 mmol scale at 0.05 M. KA )
ketoacid, SY ) sulfur ylide, CA ) carboxylic acid (from oxidative
decarboxylation), ratios determined by SFC analysis of the unpurified
mixture at 300 nm. cDetermined by SFC analysis of the unpurified ligation
product. Ligations were performed using 1.2 equiv of hydroxylamine 3.
e
d1.1 equiv of Oxone was used in the oxidation. Not determined.
Good yields of the R-ketoacid could be obtained in mixtures of
water with a variety of organic cosolvents (Table 1, entries 1-4).
We observed that the diastereomeric ratio of the ligation product
was affected by the reaction time. When the oxidation was allowed to
proceed for a longer time (entry 5), the yields of the carboxylic acid
side product increased and the diastereomeric ratios of the ligation
product decreased. This implies that epimerization originates in the
oxidation process. Variation of the solvent ratios (entries 6-8)
altered both the oxidation yield and the diastereomeric ratio of the
coupling product as a combined result of changes in the solubility
of the reactants and the nature of the solvent. Notably, 2:1 THF/
H2O (entry 6) resulted in a 50:1 ratio of diastereomers while still
retaining high yields of the R-ketoacid. In general, lower levels of
epimerization were observed at low conversion (entries 7-8).
The optimal pH range appears to be at pH 3-4 (entries 10-
11), which gives good yields and high diastereomeric ratios. When
the reaction was performed in a highly acidic environment (entry
9) or at a pH above 5 (entry 12), the R-ketoacid was rapidly
overoxidized into the carboxylic acid. Performing the reaction at a
a HPLC yields at 300 nm: R-ketoacid (KA):sulfur ylide (SY):carboxylic
acid (CA). bHPLC yields at 300 nm: ligation product (LP):sulfur ylide (SY):
c
carboxylic acid (CA) from oxidative decarboxylation. Combined yield of
sulfur ylide and carboxylic acid. dLigation performed with HONH-Ala-
OtBu. eLigation performed at 60 °C. fIsolated yield over two steps. gLigation
h
performed with HONH-Gly-Phe-OtBu. At 50 °C.
lower temperature or adding the Oxone in portions simply slowed
down the rate of the oxidation (not shown) but could not further
suppress epimerization. In choosing our standard conditions we
elected to balance the highest yield of the R-ketoacid and the lowest
9
4254 J. AM. CHEM. SOC. VOL. 130, NO. 13, 2008

C O M M U N I C A T I O N S
levels of epimerization and selected 2:1 THF/H2O (entry 4) and
1:1 DMF/H2O (entry 6), which could accommodate the solubility
of most of our substrates and give excellent conversion and high
diastereomeric ratios. Importantly, a 0.1 M solution of ketoacid 2
prepared in this manner was chemically and configurationally stable
upon standing overnight at ambient temperature.
ization. Our studies also provide further testament to the impressive
chemoselectivity of the R-ketoacid-hydroxylamine amide-forming
ligation reaction. Current efforts are aimed at translating this work
to solid-phase methods, which will allow the preparation of longer
R-peptide derived R-ketoacids suitable for use in decarboxylative
peptide ligation reactions.
Conveniently, and importantly, the peptide sulfur ylides could
be easily prepared simply by direct coupling of an N-protected
amino acid with sulfonium salt 5 under standard peptide coupling
conditions (Scheme 3). Similar protocols with the corresponding
phosphonium salts are possible, but we found these reactions to be
capricious and highly sensitive to water and other factors.9a In
contrast, couplings with the sulfonium salt require no special
handling or precautions, save that the starting sulfonium bromide
is hydroscopic and must be stored accordingly. The resulting R-keto-
cyanosulfur ylides are configurationally and chemically stable and
may be handled in air without concern. Standard deprotection
reactions, such as acidic removal of Boc groups, does not protonate
the highly stabilized ylide. Elongation of the peptide chain is readily
achieved using standard peptide deprotection and coupling condi-
tions (EDCI or HBTU).17
With standard protocols for both the synthesis of the C-terminal
peptide cyanosulfur ylides and their conversion to the R-ketoacids
established, we investigated the compatibility of these conditions
with various amino acid side chains (Table 2). We prepared a series
of dipeptides containing an unprotected side chain adjacent to a
C-terminal phenylalanine. When placed away from the ligation site,
unprotected tryptophan, tyrosine, lysine, arginine, histidine, and
glutamic acid (entries 4-9) were all found to be compatible with
sulfur ylide oxidations with aqueous Oxone (Table 2). Protected
side chains, in most cases, were also tolerated. tert-butyl carbamates
(Boc) and esters (tBu) were not affected, but we observed partial
deprotection of trityl and Pbf protecting groups, probably due to
the acidic nature of unbuffered Oxone. When placed at the ligation
site, protected tryptophan and tyrosine residues underwent smooth
oxidations (entries 1, 2). Unprotected C-terminal tryptophan or
tyrosine sulfur ylides, however, were resistant to oxidation. It should
be noted that the ligation step was unoptimized; higher yields can
be obtained under other conditions or with excess hydroxylamine.
Examination of Table 2 reveals that in all cases the R-ketoacid
is the major product. Together, the R-ketoacid product and the
carboxylic acid byproduct resulting from the oxidative decarboxy-
lation account for >85% of the material, indicating that the side
chains do not form significant amounts of oxidized byproducts.
Basic residues (entries 4, 6-8) appear to slow the rate of oxidation,
resulting in lower conversion when using the reaction times and
conditions optimized for Fmoc-Ala-Phe-SY (2). This is consistent
with our findings that the pH of buffered solutions alters the rate
of the oxidations.
Acknowledgment. This work was supported by the NIH
NIGMS [R01-GM076320]. Unrestricted support from Amgen,
Astrazeneca, Bristol-Myers Squibb, and Eli Lilly is greatly ap-
preciated. J.W.B. is a fellow of the Beckman Foundation, the
Packard Foundation, the Sloan Foundation, and a Research
Corporation Cottrell Scholar. We are grateful to Jian Wu for helpful
discussions and the preparation of one of the sulfur ylides used in
this study.
Supporting Information Available: Experimental procedures and
characterization data for all compounds. This material is available free
of charge via the Internet at http://pubs.acs.org.
References
(1) Dawson, P. E.; Muir, T. W.; Clark-Lewis, I.; Kent, S. B. H. Science 1994,
266, 776-779.
(2) For reviews on native chemical ligation, see: (a) Dawson, P. E.; Kent, S.
B. H. Annu. ReV. Biochem. 2000, 69, 923-960. (b) Macmillan, D. Angew.
Chem., Int. Ed. 2006, 45, 7668-7672. (c) Rademann, J. Angew. Chem.,
Int. Ed. 2004, 43, 4354-4556.
(3) (a) Bode, J. W.; Fox, R. M.; Baucom, K. D. Angew. Chem., Int. Ed. 2006,
45, 1248-1252. (b) Carrillo, N.; Davalos, E. A.; Russak, J. A.; Bode, J.
W. J. Am. Chem. Soc. 2006, 128, 1452-1453.
(4) Cooper, A. J. L.; Ginos, J. Z.; Meister, A. Chem. ReV. 1983, 83, 321-
358.
(5) For other emerging methods in peptide ligation, see: (a) Nilsson, B. L.;
Kiessling, L. L.; Raines, R. T. Org. Lett. 2001, 3, 9-12. (b) Liu, L.; Hong,
Z.-Y.; Wong, C.-H. ChemBioChem 2006, 7, 429-432. (c) Tam, A.;
Soellner, M. B.; Raines, R. T. J. Am. Chem. Soc. 2007, 129, 11421-
11430. (d) Brik, A.; Ficht, S.; Yang, Y.-Y.; Bennett, C. S.; Wong, C.-H.
J. Am. Chem. Soc. 2006, 128, 15026-15033. (e) Crich, D.; Sana, K.;
Guo, S. Org. Lett. 2007, 9, 4423-4426.
(6) We are aware of only one example of an amino acid derived R-ketoacid.
See entry 3 in Table 1 of ref 8a.
(7) For a recent innovation in synthesizing R-ketoamides, see: Grassot, J.-
M.; Masson, G.; Zhu, J. Angew. Chem., Int. Ed. 2008, 47, 947-950.
(8) (a) Wasserman, H. H.; Ho, W.-B. J. Org. Chem. 1994, 59, 4364-4366.
(b) Wasserman, H. H.; Baldino, C. M.; Coats, S. J. J. Org. Chem. 1995,
60, 8231-8235. (c) Wong, M.-K.; Yu, C.-W.; Yuen, W.-H.; Yang, D. J.
Org. Chem. 2001, 66, 3606-3609. (d) Wasserman, H. H.; Parr, J. Acc.
Chem. Res. 2004, 37, 687-701. (e) Wasserman, H. H.; Ennis, D. S.;
Power, P. L.; Ross, M. J. J. Org. Chem. 1993, 58, 4785-4787. (f)
Wasserman, H. H.; Vu, C. B. Tetrahedron Lett. 1990, 31, 5205-5208.
(9) (a) Weik, S.; Rademann, J. Angew. Chem., Int. Ed. 2003, 42, 2491-2494.
(b) El-Dahshan, A.; Weik, S.; Rademann, J. Org. Lett. 2007, 9, 949-
952.
(10) Lippert, A. R.; Kaeobamrung, J.; Bode, J. W. J. Am. Chem. Soc. 2006,
128, 14738-14739.
(11) For a report of a peptide derived sulfur ylide, see: Rich, D. H.;
Bernatowicz, M. S. J. Org. Chem. 1983, 48, 1999-2001.
(12) Adam, W.; Curci, R.; Edwards, J. O. Acc. Chem. Res. 1989, 22, 205-
211.
(13) Kennedy, R. J.; Stock, A. M. J. Org. Chem. 1960, 25, 1901-1906.
(14) Wasserman, H. H.; Baldino, C. M.; Coats, S. J. J. Org. Chem. 1995, 60,
8231-8235.
(15) We found the reproducibility of the oxidation was dependent on the source
of the Oxone. The Oxone used in these studies was obtained from Alfa
Aesar and generally resulted in more efficient oxidations than Oxone
obtained from other sources.
(16) (a) Jefford, C. W.; Boschung, A. F.; Bolsman, T. A. B. M.; Moriarty, R.
M.; Melnick, B. J. Am. Chem. Soc. 1976, 98, 1017-1018. (b) Isbell, H.
S.; Frush, H. L.; Orhanovic, Z. Carbohydr. Res. 1974, 36, 283-291.
(17) Peptide couplings immediately adjacent to the cyanosulfur ylide peptide
monomer (i.e., from 6 in Scheme 3) are somewhat sluggish. In solution
phase, this is best overcome by using a small excess of the deprotected
amino sulfur ylide during this coupling step. Subsequent peptide couplings
proceed without unusual difficulties, save that their low solubility can
lead to low isolated yields. This is not an issue with solid phase methods,
and we have used this chemistry extensively for the synthesis of longer
peptide R-ketoacids.
In preliminary experiments with a methionine containing sulfur
ylide, we isolated the sulfone-ligated product and HPLC and ESI-
MS data of the crude oxidations evidenced a mixture of the sulfur
ylide and R-ketoacid with the methionine sulfur at the sulfoxide
and sulfone oxidation states. Addition of DMSO, tetrahy-
drothiophene, or thioanisole as oxidation competitors as well as
varying Oxone equivalents, solvent, and temperature resulted in
conditions providing a 29% yield of the sulfoxide R-ketoacid (see
Supporting Information).
In summary, we have developed a robust, chemoselective method
to produce C-terminal peptide R-ketoacids with minimal epimer-
JA800053T
9
J. AM. CHEM. SOC. VOL. 130, NO. 13, 2008 4255