Nitrone protecting groups for enantiopure N-hydroxyamino acids: Synthesis of N-terminal peptide hydroxylamines for chemoselective ligations

Article abstract of DOI:10.1039/c004490c

The synthesis of enantiopure N-benzylidene nitrones of N-hydroxy-α- amino acids and their incorporation using standard Fmoc-based peptide chemistry into solid-supported peptide chains is described. Deprotection and resin cleavage affords N-terminal peptid

Full text of DOI:10.1039/c004490c

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Nitrone protecting groups for enantiopure N-hydroxyamino acids: synthesis of
N-terminal peptide hydroxylamines for chemoselective ligations†
S. Irene Medina,a Jian Wua and Jeffrey W. Bode*b
Received 19th March 2010, Accepted 13th May 2010
First published as an Advance Article on the web 22nd June 2010
DOI: 10.1039/c004490c
The synthesis of enantiopure N-benzylidene nitrones of N-hydroxy-a-amino acids and their
incorporation using standard Fmoc-based peptide chemistry into solid-supported peptide chains is
described. Deprotection and resin cleavage affords N-terminal peptide hydroxylamines, which are the
key substrates for chemoselective ligations with C-terminal peptide a-ketoacids. This general route is
applicable to a variety of different N-terminal residues and provides a general approach to the solid
phase synthesis of peptide hydroxylamines.
The greatest obstacle to the widespread application of this
reaction to the synthesis of larger peptides has been the lack of syn-
We have recently reported a novel chemoselective amide-forming
thetic approaches for the preparation of the requisite C-terminal
Introduction
ligation of a-ketoacids and N-alkyl hydroxylamines (Scheme 1).1
a-ketoacids and N-terminal hydroxylamines. In each case, general
This reaction does not require any reagents, proceeds under mild
methods that not only preserve adjacent stereochemistry and
conditions, tolerates unprotected functional groups, and produces
tolerate unprotected functional groups but also interface with
only carbon dioxide and water as byproducts. In preliminary
studies, we have demonstrated that this amide-formation has the
potential to serve as a general peptide ligation for the joining of
standard, established technologies for peptide synthesis must be
identified. For the synthesis of the C-terminal a-ketoacids, we have
reported a stereoretentive approach based on cyanosulfur ylides
unprotected a-peptide fragments.2 Our efforts, and those of other
groups,3 suggest that this process will be useful for the preparation
of medium to large peptides by ligation of shorter fragments at
numerous different ligation sites. This uniquely selective reaction
therefore has the potential to compliment the powerful native
chemical ligation4,5 by providing a general amide-forming ligation
reaction that is not limited to peptides or substrates possessing an
N-terminal cysteine residue or sulfur-containing derivatives.6
that is compatible with Fmoc-based solid phase synthesis.7 We
have also recently disclosed a novel reagent for the preparation of
solid-supported N-terminal hydroxylamines.8 This is a convenient
method for the rapid preparation of peptide hydroxylamines, but
suffers from modest yields and is best suited for exploratory, rather
than preparative, applications. The preparation of N-terminal
glycine hydroxylamines is possible via nucleophilic displacement
on N-terminal bromoacetates, a protocol we have recently em-
ployed in our synthesis of the human GLP-1 (7-36) peptide via
the ketoacid–hydroxylamine ligation of two fully unprotected 15-
mer peptide fragments.9 In this case, the choice of an N-terminal
glycine as the ligation site was governed by the synthetic methods
available for the preparation of the hydroxylamine rather than any
particular advantage of the sterics of the glycine hydroxylamine.
On the contrary, our experience to date suggests that N-terminal
aRoy and Diana Vagelos Laboratories, Department of Chemistry, University
of Pennsylvania, Philadelphia, PA, USA 19104
bLaboratorium fu¨r Organische Chemie, ETH–Zu¨rich, HCl F-315 Wolf-
gang Pauli Strasse 10, 8093, Zu¨rich, Switzerland. E-mail: bode@org.
chem.ethz.ch; Fax: +41 446331235; Tel: +41 446332103
† Electronic Supplementary Information (ESI) available: Experimental
procedures, spectra and characterization data. See DOI: 10.1039/c004490c
Scheme 1 Ketoacid–hydroxylamine amide ligations of unprotected peptide fragments.
The Royal Society of Chemistry 2010 Org. Biomol. Chem., 2010, 8, 3405–3417 | 3405
This journal is
©

View Article Online
Scheme 2 Synthesis of N-terminal peptide hydroxylamines by monomer coupling.
Scheme 3 Confirmation of stereochemical integrity of chiral N-hydroxylamines during ligations with a-ketoacids.
glycine hydroxylamines are more prone to decomposition than
substituted residues.
ligations can occur at N-terminal residues containing unprotected
side chains.
The most straightforward approach to introducing enantiopure
N-terminal peptide hydroxyamino acid residues into a synthetic
peptide would be the coupling, under standard conditions, of a
suitably protected N-hydroxyamino acid to the peptide N-terminal
amine. The successful implementation of this strategy requires four
discrete, but interconnected successes: 1) the preparation of enan-
tiomerically pure peptide hydroxylamines; 2) the identification of
a chemically and configurationally stable protecting group for the
peptide hydroxylamines; 3) the coupling of the N-hydroxyamino
acid monomers without epimerization; and 4) deprotection of
the hydroxylamine and isolation of the side-chain deprotected
N-terminal peptide hydroxylamines. A further complication is
that N-terminal peptide hydroxylamines are somewhat unstable
towards oxidation or elimination, constraining the methodologies
that can be employed for their synthesis, manipulation, and
purification. Once prepared and isolated in pure form, they can
be stabilized as their TFA or oxalate salts, but the relative lability
of intermediates necessary for the synthesis of the appropriate
amino acid monomer as well as the final peptides demands careful
consideration and innovative solutions.
Results and discussion
Stereoretention of peptide N-hydroxylamines during ligations with
a-ketoacids
In our preliminary studies on the a-ketoacid–hydroxylamine lig-
ation, we had occasionally detected peptide epimers at the former
hydroxylamine stereocenter, but we were unsure if epimerization
had occurred during the preparation of the peptide hydroxylamine
or during the ligation itself. In order to determine at which
step epimerization occurred, enantiomerically enriched (>99%
ee as determined by chiral HPLC) dipeptide 2 was carefully
prepared and subjected to ligation with a-ketoacid dipeptide 1 to
afford tetrapeptide 3 (Scheme 3). Comparison of the a-ketoacid–
hydroxylamine ligation product with diastereomeric standard 4 by
SFC chiral column revealed preservation of stereochemistry. This
result, and related observations in our laboratory, suggested that
epimerization occurred during the preparation of the peptide hy-
droxylamine. Indeed, epimerization at some stage of the synthesis
We now report a general, comprehensive approach to the
synthesis of N-terminal peptide hydroxylamines by the coupling of
enantiomerically pure, nitrone-protected N-hydroxyamino acids
to solid-supported peptides (Scheme 2). This chemistry is com-
patible with common reagents, resins, and protection strategies
used in Fmoc-based solid phase peptide synthesis. We have
rigorously established the preservation of stereochemistry in the
synthesis of the amino acid monomers as well as during their
coupling using typical protocols. Several approaches to the unique
challenges of deprotecting, isolating, and purifying the N-terminal
peptide hydroxylamines are described. In preliminary work, these
studies have allowed us to demonstrate that chemoselective peptide
of certain hydroxylamines was confirmed by their reduction with
R
R ꢀ
Zn or RANEYꢀ Ni and analysis of the resulting amines.10
Synthesis of enantiopure N-hydroxyamino acids
Our first task therefore became the identification of a re-
liable, general method for the preparation of enantiopure
N-hydroxyamino acids. Our previous forays into the syn-
thesis of enantiopure N-terminal peptide hydroxylamines re-
lied on the late-stage oxidation of the N-terminal amine of
fully assembled peptides using a 3-step protocol reported
by Fukuyama.11 For example, monoalkylation of N-terminal
3406 | Org. Biomol. Chem., 2010, 8, 3405–3417
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Scheme 4 Synthesis of an N-terminal peptide hydroxylamine by late-stage application of the Fukuyama protocol.
amine 5 with bromoacetonitrile followed by regioselective
oxidation of the resulting secondary amine with mCPBA afforded
“cyano-nitrone” 7 that was hydrolyzed to the corresponding hy-
droxylamine with excess hydroxylamine hydrochloride in MeOH
(Scheme 4). The N-hydroxyamino peptide 8 was isolated by
precipitation as its oxalate salt. This process, which is best executed
in solution phase, works remarkably well for the preparation of
short peptide hydroxylamines but suffers from the inconvenience
of extensive manipulations of often sparingly soluble side-chain
protected peptides and attendant poor yields.
In shifting our goal to developing a suitable protection and
coupling strategy for enantiopure N-hydroxyamino acids, we re-
evaluated the available methods for the preparation and coupling
of the requisite monomers. Polon´ski, Miller and others have
reported the synthesis of non-racemic N-benzylidene nitrones
derived from a-amino acids including alanine, phenylalanine
and leucine by oxidations of the corresponding benzylidene
imines.12 This method offers the most direct route to our targeted
monomers, but its execution is often capricious and not readily
extended to amino acids containing diverse, protected side-
chain functionalities. Other methods for the synthesis of non-
racemic peptide hydroxylamines include SN2 displacements of
activated, chiral a-hydroxyacids, which require the preparation
of corresponding enantiopure starting materials13 and the direct
oxidation of amino esters with dimethyldioxirane at cryogenic
temperatures.14 We therefore elected to focus our attention on
refinements of the Fukuyama procedure to provide a reliable, gen-
eral method for the preparation of enantiopure N-hydroxyamino
acids. This approach would also allow us to readily test a variety
of protecting group strategies for the peptide hydroxylamines.
The Fukuyama protocol has proven to be a remarkably robust
approach to the synthesis of primary hydroxylamines. It is readily
executed on a variety of reaction scales and substrates. In applying
this approach to the synthesis of peptide hydroxylamines, we
did occasionally detect epimerized products, the origin of which
we eventually traced to the “cyano-nitrone” intermediate. If
not immediately hydrolyzed to the hydroxylamine, these “cyano-
nitrones” undergo epimerization, presumably due to the highly
electron-deficient nature of this moiety. Fortunately, as long as
this intermediate is used promptly the stereochemical integrity is
maintained. In general, the alkylation and oxidation steps occur
without difficulty. In order to avoid epimerization at the cyano-
nitrone stage, we usually conducted the oxidation, hydrolysis and
isolation of the hydroxylamine oxalates without purification of
intermediates.15 In some batches, we have occasionally found the
precipitation of the salts to be difficult or low yielding. For this
reason, we recommend purification of the cyanomethylated amine
prior to the oxidation step, despite the fact that this reaction
usually proceeds cleanly and in excellent yield. The resulting
hydroxylamine salts were proven to be enantiomerically pure by
conversion to the N-benzylidene nitrone derivatives followed by
HPLC-analysis on chiral columns (vide infra).
With this information, we developed a general route to the
synthesis of nitrone protected hydroxylamines (Scheme 5 and
Table 1). For the preparation of N-hydroxyamino acids lacking
side-chain functionality, we began with O-tert-butyl protected
a-amino acids. For amino acids with side-chain functionality,
such as L-tyrosine and L-lysine, we employed O-allyl protected
amino acids. In either case, N-alkylation with bromoacetonitrile
proceeded smoothly and the resulting secondary amine (e.g. 10)
was isolated by column chromatography. Oxidation with mCPBA
provided the stereochemically labile cyano-nitrone (e.g. 11), which
was immediately hydrolyzedtothe free hydroxylamine andisolated
as its mono oxalate salt. The oxalate salts are chemically and
Scheme 5 Preparation of enantiopure L-alanine-derived a-N-hydroxyamino acid.
The Royal Society of Chemistry 2010 Org. Biomol. Chem., 2010, 8, 3405–3417 | 3407
This journal is
©

View Article Online
Table 1 Enantiopure hydroxylamines prepared by the Fukuyama protocola
b
cyanomethylation
oxidation and hydrolysis
99
97
97
80
60
cyanomethylation
oxidation and hydrolysis
90
56
86
98
86
90
a See Scheme 5 reaction conditions. b This hydroxylamine amine was prepared from tert-butylbromoacetate and O-(2-methoxyisopropyl)hydroxylamine,
followed by treatment of this product with oxalic acid (See Supporting Information for details†).
configurationally stable and can be stored without difficulty for
several months or longer.
peptide coupling. Ominously, Bentley and Brooks indicated that
they experienced difficulty in the peptide coupling when using
carbodiimide reagents. The advantage of this approach, however,
was the simplicity of the protection step and the ability to modulate
the properties of the protected hydroxylamines by variation of the
aldehyde used for nitrone formation.
Protecting groups for a-N-hydroxyamino acids
In selecting a protecting group for the N-hydroxyamino acid
monomers, we wished to identify an approach that would be chem-
ically orthogonal to side-chain protection strategies employed
in Fmoc-based solid phase peptide synthesis.16 We were further
constrained by potential reactions of the hydroxylamines, which
are prone to N–O bond cleavage under reductive conditions or by
elimination reactions, particularly when dealing with a-peptide-
derived hydroxylamines.17 Although we briefly explored either
N-monoprotection or N,O-bis-protection with carbamate pro-
tecting groups,18 we found some of these derivatives to be
unstable towards deprotection and migration of the carbamates
was occasionally observed.19 In certain cases, O-protection with
silyl groups,20 trityl derivatives21 or as acetals22 was viable, but in
our hands we could not develop this into a general strategy for the
protection and coupling of N-hydroxyamino acid monomers.
During the course of these studies, our attention turned to a
single report of the protection and coupling of N-hydroxyamino
acid monomers to amines by Bentley and Brooks.23 This survey of
a number of protection and coupling strategies revealed that the
N-hydroxyamino acids were best protected as their corresponding
nitrones. These amino acid monomers could be coupled to a
primary amine by the formation of a mixed anhydride with
isobutyl chloroformate and deprotected using aqueous acid.
Lacking in this study was the use of enantiomerically pure
N-hydroxyamino acids. We were initially hesitant to adopt the
nitrone protecting group due to fears that the use would lead
to epimerization, particularly as we had already established that
the cyano-nitrone intermediates employed in the Fukuyama hy-
droxylamine synthesis were configurationally labile. Furthermore,
even if the nitrone-protected N-hydroxyamino acid monomers
could be prepared in enantiomerically pure form, we were
concerned that they would undergo epimerization during the
Protection of the hydroxylamines as nitrones was achieved
simply by mixing the oxalate salts with benzaldehyde and NEt3
in DMF or CH2Cl2 (Scheme 6). The stereochemical integrity
of the N-benzylidene nitrones was confirmed by HPLC-analysis
on chiral columns by comparison to a racemic standard. We
were pleased to find that the O-tert-butyl, N-benzylidene ni-
trones were chemically and configurationally stable and, like
the hydroxylamine salts, could be stored without difficulty or
epimerization. Importantly, deprotection of the tert-butyl esters
with TFA occurred smoothly without affecting the nitrone, which
is labile towards aqueous acid. The resulting N-benzylidene
nitrone a-amino acid (e.g. 20) was also stable and amenable to
storage. The same procedures were applied to the synthesis of N-
benzylidene protected, N-hydroxyamino acids from glycine (22),
L-leucine (23), L-phenylalanine (24), L-tyrosine (25), and L-lysine
(26) (Table 2).
Peptide coupling of nitrone-protected N-hydroxyamino acids
The goal of this research project was the identification of a method
for the introduction of a suitably protected, enantiomerically pure
N-hydroxyamino acid to the N-terminus of a solid-supported syn-
thetic peptide. The work of Bentley and Brooks had demonstrated
that amide-formation with the N-benzylidene nitrone protected
hydroxylamines was feasible,23 but did not address the potential for
epimerization during this reaction. Even under the best conditions,
epimerization of protected amino acids during peptide coupling is
an issue; the use of the electron-deficient nitrones and the potential
for the Lewis basic oxygen atom to react with the intermediate
activated esters amplified this concern.
3408 | Org. Biomol. Chem., 2010, 8, 3405–3417
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Scheme 6 Protection of L-alanine-derived hydroxylamine and confirmation of stereochemical integrity.
Table 2 Protection a-N-hydroxyamino acid monomers as benzylidene nitronesa
Nitrone
formation
deprotection
99
93
86
84
96
99
75
80
59b
50b
a See Scheme 6 for reagents and conditions. b Deprotection conditions: PdCl2(PPh3) [10 mol %], PPh3 [30 mol %], morpholine (excess).
We were therefore pleasantly surprised to find that
N-benzylidene protected, N-hydroxyamino acids can be easily and
cleanly coupled to an N-terminal peptide amine in excellent yield
and without epimerization (Scheme 7). A number of standard
coupling reagents including HBTU, TBTU, PyBop, and DIC
may be used, with similar results. In most cases, no special
precautions or unusual conditions were needed to effect the
coupling. This success allowed us to extend these methods
to the incorporation of a variety of N-benzylidene protected
N-hydroxyamino acids onto solid-supported peptide chains. Our
further studies, as reported below, confirm that the nitrone
protecting groups are compatible with standard methods, resins,
Scheme 7 Peptide coupling of nitrone protected L-alanine-derived hydroxylamine and confirmation of stereochemical integrity.
This journal is The Royal Society of Chemistry 2010 Org. Biomol. Chem., 2010, 8, 3405–3417 | 3409
©

View Article Online
Scheme 8 On-resin nitrone deprotection and a-ketoacid–hydroxylamine peptide ligation.
reagents and protecting groups employed in Fmoc-based peptide
synthesis.
5–10 min before being repeatedly passed through a short column
of C18 silica gel and a resin-bond hydroxylamine (5 times). A final
elution with CH3CN afforded the crude peptide hydroxylamines,
which were purified by preparative HPLC (Scheme 9).
Coupling and deprotection of nitrone-protected N-terminal peptide
hydroxylamines onto solid-supported peptides
Resin cleavage/side-chain deprotection of N-benzylidene ni-
trone peptides were conducted without the use of silane scavengers,
when such scavengers were used during resin cleavage we observed
the N-benzyl hydroxylamine peptide. Unexpectedly, nitrone hy-
drolysis under these conditions is easier for the larger amino
acid residues, such as leucine, phenylalanine and tyrosine. Glycine
and alanine-derived nitrones, in particular, are considerably more
resistant to hydrolysis. Although the peptide nitrones appear
to be indefinitely stable, the free hydroxylamines are prone to
oxidation to the corresponding oximes during reaction workup
and subsequent purification. Indeed, the major side product from
the isolated peptides of the ligation reactions in general is the
corresponding peptide oximes (e.g. 38).
The remaining task in this research plan was the identification
of reaction conditions for the late stage deprotection of the hy-
droxylamine. Strategically, this could be effected either in tandem
with resin cleavage and side-chain deprotection to give the free
N-terminal hydroxylamine that, following suitable purification,
could be used in the ligation reaction. Alternatively, we envisioned
strategies whereby the nitrone protection would be preserved
while the side chain functionalities were deprotected. This would
be an important step in the eventual development of a stable,
orthogonally reactive hydroxylamine protecting group that can be
used for segment condensation approaches to the preparation of
longer peptides. Finally, there are cases in which it may be desirable
to liberate the hydroxylamine while the peptide is still bound to the
resin, a method that we have demonstrated in our prior work on
hydroxylamine synthesis.24 Although not the focus of this study, we
were pleased to find that many resin-bound peptide nitrones could
be deprotected using these conditions to afford the resin-bound
hydroxylamines suitable for on-bead peptide ligations (Scheme 8).
We have placed greater emphasis on methods that deliver the
fully unprotected peptide hydroxylamines by nitrone hydrolysis
under acidic conditions. A variety of acidic conditions, such as
aqueous HCl,25 promote nitrone hydrolysis, but are complicated
by the extremely fast recombination of the hydroxylamine and
the aldehyde upon workup.26 Thus, deprotection reactions that
appear to be successful when monitored during the reaction
course often deliver significant amounts of the starting nitrone
upon isolation! This finding, along with our observations of the
properties of the nitrone-protected peptides, led us to adopt a
somewhat unorthodox approach to the liberation of the peptide
hydroxylamines: following resin cleavage, the peptide nitrones were
precipitated from Et2O and treated with 10% aqueous TFA for
This oxidative side reaction can be mitigated by taking care
to shield the peptides from oxygen, but some amount of oxime
is usually detected. It can be removed during purification of the
final, fully deprotected peptide hydroxylamines by reverse phase
preparative HPLC but detracts from the overall yield of the desired
product.
3410 | Org. Biomol. Chem., 2010, 8, 3405–3417
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Scheme 9 Synthesis of N-terminal peptide hydroxylamines by amide coupling and deprotection.
Scheme 10 Solid phase peptide synthesis of an 11-mer peptide hydroxylamine.
There appears to be no difficulty in applying these methods
to the synthesis of longer peptide hydroxylamines. In ongoing
work, we have routinely applied these methods to the synthesis of
unprotected peptide hydroxylamines in the range of 10–20 amino
acids. For example, the 11-mer peptide hydroxylamine 40 was
prepared by Fmoc-based solid phase peptide synthesis without
complications (Scheme 10).
and chemical reactivity of a given peptide fragment are highly
dependent on its sequence and solubility. Our chemistry is no
exception and the choice of nitrone and its ease of hydrolysis
are often sequence and application specific and we do not
currently have a predictive tool as to whether or not the N-
benzylidene nitrones will be easy or difficult to hydrolyze. Our only
generalization to date is that the nitrones of glycine and alanine
are usually more resistant.
In a preliminary attempt to identify other nitrone protecting
groups that have either greater or diminished stability, depending
on the application, we have synthesized and screened a number
of different peptide nitrones and explored their behavior under
typical conditions for side-chain deprotection and purification
(Scheme 11). Our observations from this study are summarized in
Table 3. In general, more electron-deficient nitrones demonstrated
greater stability, while the electron-rich benzylidenes were more
prone to nitrone cleavage upon resin release or HPLC purification.
We attribute these counterintuitive findings to the very high affinity
Modulation of the nitrone protecting group: preliminary results
When N-benzylidene nitrones are employed, the stability of the
nitrone and its ease of deprotection appears to be correlated to the
steric properties of the amino acid residue. Thus, smaller amino
acids, such as glycine and alanine, form stable nitrones that can be
difficult to deprotect. Depending on the particular application
of the a-ketoacid–hydroxylamine ligation, peptide nitrones of
either increased or decreased hydrolytic stability are sometimes
desired. As is always the case in peptide chemistry, the properties
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 3405–3417 | 3411
©

View Article Online
Scheme 11 Studies on the stability of various nitrones as hydroxylamine protecting groups.
Table 3 Preliminary studies of nitrone stability to resin cleavage and purification.a
entry
R =
50% TFA in CH2Cl2
95% TFA in H2O
Prep-HPLC
1
2
3
4
5
6
p-CN-C6H4
trans-cinnamyl
2,6-di-Cl-C6H3
2,4,6-tri-Me-C6H2
2,5-di-(MeO)-C6H3
2,6-di-(MeO)-C6H3
stable
stable
stable
stable
stable
unstable
stable
stable
stable
unstable
unstable
unstable
stable
stable
stable
unstable
unstable
unstable
a See Scheme 11 for peptide structures. These studies are specific to this peptide sequence and may vary depending on the N-terminal hydroxylamine and
the peptide sequence.
of the electron-deficient aldehydes to the hydroxylamines. In
contrast, the electron-rich arylaldehydes are slower to recombine
should cleavage occur.
All N-terminal hydroxylamines tested to date give at least some
of the desired ligation product. The major differences appear to
be in the stability of the hydroxylamines towards oxidation. For
example, the N-terminal lysine hydroxylamine employed for the
synthesis of 48 was particularly prone to oxime formation, thereby
diminishing the yield of the ligated product. It must be noted
that these ligations were unoptimized and performed a single time
under a single set of conditions simply to test the viability of
each of the N-terminal hydroxylamine ligation sites. In our work
on the application of the a-ketoacid–hydroxylamine ligation to
longer peptides, we have found that optimization of the reaction
conditions as a function of solubility of the peptide fragments
and products is usually required. We therefore leave such specific
optimizations to synthetic applications of the ligation reaction.
In the majority of cases, the use of the N-benzylidene nitrones
(i.e. those derived from benzaldehyde) are ideal, but in more so-
phisticated applications of the a-ketoacid-hydroxylamine ligation
we have occasionally needed to modulate the properties of the
nitrone to either enhance or suppress its hydrolysis. The ability to
change the aldehyde component offers a convenient and powerful
means of achieving this. As a caveat, it should be noted that
regardless of the nitrone protecting group employed, they are not
stable towards the ligation conditions in the presence of another
hydroxylamine. In all case studies to date, nitrone exchange occurs,
even on the most hindered of substrates. Nitrone protection is
therefore not a suitable solution to the challenge of iterative frag-
ment condensations using the a-ketoacid–hydroxylamine ligation,
thus other protecting group strategies for the hydroxylamine will
be needed for this application.27
Conclusions
Enantiomerically pure N-hydroxyamino acids can be reliably
prepared by Fukuyama’s method and protected as their N-
benzylidene nitrones. Unlike other approaches, this method is
both general and delivers stable, protected N-hydroxyamino acids
suitable for introduction into peptides with standard coupling
reagents and without epimerization. The solid-supported nitrone-
protected hydroxylamines can be cleaved from either side-chain
protected or side-chain unprotected peptides without interfer-
ence from common side-chain functional groups. Importantly,
these findings greatly expand the range of N-terminal peptide
hydroxylamines that can be prepared and isolated, an important
and necessary step for developing the a-ketoacid–hydroxylamine
amide-ligation as a general method for peptide synthesis.
Peptide-ligations at various N-terminal hydroxylamine residues
The potential impact of the a-ketoacid–hydroxylamine amide-
formation on the chemical synthesis of peptides and proteins lies
in its ability to effect peptide ligations at a wide variety of ligation
sites. In our studies on the synthesis of peptide a-ketoacids, we
have noted that certain unprotected a-ketoacids are not good
ligation partners due to interactions of the side chains with the
a-ketoacid functionality. These studies, and ongoing work in
our laboratory, have established that C-terminal glycine, alanine,
leucine, phenylalanine, proline, isoleucine, valine and glutamic
acid are all viable. Due to the prior difficulties of their synthesis,
we have performed fewer studies on the residue-specificity of
unprotected N-terminal hydroxylamines in the ligation of a-
peptide fragments. Using the materials prepared during the
synthesis of various nitrone protected peptide hydroxylamines,
we briefly explored their ligations with C-terminal peptide a-
ketoacids (Scheme 12).
Experimental
General experimental
All reactions utilizing air- or moisture-sensitive reagents were
performed in dried glassware under an atmosphere of nitrogen.
3412 | Org. Biomol. Chem., 2010, 8, 3405–3417
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Scheme 12 Ligation of unprotected N-terminal hydroxylamine peptides.
CH2Cl2 was distilled over CaH2. THF was distilled from
Na/benzophenone. CH3OH and DMF were dried by pas-
sage over molecular sieves under Ar atmosphere. N,N-
Diisopropylethylamine (DIPEA) was distilled from CaH2. Other
20 mm for preparative), or Zorbax Eclipse XDB-C8 4.6 ¥ 150 mm.
All separations utilized a gradient of iPrOH or CH3CN and
millipore H2O, each containing 0.1% TFA. Peptides were prepared
by standard Fmoc manual solid-phase synthesis protocols using
Rink amide MBHA resin with a loading of 0.70 mmol/g.
R
reagents were used without further purification. Oxoneꢀ was
purchased from Alfa Aesar. Thin layer chromatography (TLC) was
performed on Merck precoated plates (silica gel 60 F254, Art 5715,
0.25 mm) and was visualized by fluorescence quenching under UV
light or by staining with potassium permanganate or ninhydrin.
Preparative thin-layer chromatography (PTLC) was performed
using plates prepared from silica gel EMD 60 PF254 (Art 7749).
Column chromatography was performed on E. Merck Silica Gel
General procedure for the preparation of N-hydroxyamino acid
ester oxalates
(S)-tert-Butyl 2-(hydroxyamino)propanoate oxalate (12). The
following procedure is representative. A solution (0.1 M) of
(S)-1-(tert-butoxy)-1-oxopropan-2-aminium chloride (0.516 g,
2.85 mmol, 1.00 equiv) in CH3CN (27.0 mL) was treated with
BrCH2CN (0.330 mL, 3.13 mmol, 1.10 equiv) and DIPEA
(1.00 mL 5.74 mmol, 2.00 equiv), and the reaction was stirred at
rt overnight. The mixture was transferred to a separatory funnel
and saturated aq NaHCO3 (100 mL) was added. The solution
was extracted with CH2Cl2 (2 ¥ 100 mL), the organic phases
were combined, washed with brine, dried over Na2SO4, filtered,
and concentrated in vacuo. The product was purified by flash
chromatography on silica (20% EtOAc in hexanes) to provide
compound 10 as a colorless oil (0.51 g, 97% yield). (S)-tert-
butyl 2-(cyanomethylamino) propanoate (10) (1.09 g, 5.92 mmol,
1.00 equiv) was dissolved in CH2Cl2 (20 mL), stirred, and cooled to
0 ◦C. To the cooled reaction mixture, mCPBA (2.55 g, 14.8 mmol,
2.00 equiv) was added in several in portions over 30 min. The
1
60 (230–400 Mesh) using a forced flow of 0.5–1.0 bar. H NMR
(500 MHz) and 13C NMR (125 MHz) were measured on a Bruker
Avance II 500 spectrometer. Chemical shifts are expressed in parts
per million (PPM) downfield from residual solvent peaks and
coupling constants are reported as hertz (Hz). Splitting patterns
are indicated as follows: br, broad; s, singlet; d, doublet; dd, doublet
of doublets; t, triplet; q, quartet; m, multiplet. Infrared (IR)
spectra were recorded on a JASCO FT/IR-430 spectrophotometer
T
and are reported as wavenumber (cm-1). Optical rotations [a]D
were measured at temperature T on a Jasco P–2000 polarimeter
operating at the sodium D line with a 100 mm path length cell,
and are reported as follows: (concentration (g/ml ¥ 100), solvent).
Reverse phase HPLC was performed using the following columns:
YMC R-ODS-10A C-18 (250 ¥ 4.6 mm for analytical and 250 ¥
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 3405–3417 | 3413
©

View Article Online
solution was allowed to warm to rt and stirred for 45 min.
The reaction was quenched by addition of saturated aq Na2S2O3
(15 mL) and saturated aq NaHCO3 (25 mL) and the resulting
biphasic mixture stirred for an additional 30 min. The mixture
was then transferred to a separatory funnel and the organic phase
removed. The aqueous phase was extracted with CH2Cl2 (2 ¥
30 mL), the organic phases were combined, dried over Na2SO4,
filtered, and concentrated in vacuo to provide the cyano-nitrone as
a yellow oil. The yellow oil was immediately dissolved in MeOH
(57.7 mL), hydroxylamine hydrochloride (2.05 g, 29.6 mmol,
5.00 equiv) was added to the solution, and stirred overnight at
60 ◦C. The reaction was allowed to cool to rt and CH2Cl2 (20 mL)
was added to provide a white precipitate which was removed by
vacuum filtration. Concentration of the filtrate in vacuo afforded
a yellow oil. The oil was dissolved in CH2Cl2 (20 mL), transferred
to a separatory funnel and extracted with saturated aq NaHCO3
(30 mL). The aqueous phase was washed with CH2Cl2 (2 ¥ 20 mL),
theorganicphases were combined, dried over Na2SO4, filtered, and
concentrated in vacuo to a reduced volume. Oxalic acid (1.45 g,
1.18 mmol, 2.00 equiv) in MeOH (2 mL) was added to the solution,
followed by cold Et2O (50 mL), producing a white precipitate
which was collected by vacuum filtration and rinsed several times
with cold Et2O to afford compound 12 as a white solid (0.768 g,
according to the general procedure and isolated as a white solid
(1.17 g, 98% yield). mp 105.0–106.5 ◦C; [a]D -3.2 (c 0.19,
20
1
MeOH); H NMR (CD3OD) d 7.13 (d, 2H, J = 8.0 Hz, Ar),
6.93 (d, 2H J = 8.5 Hz, Ar), 5.86–5.78 (m, 1H, OCH2CHCH2),
5.27–5.17 (m, 2H, CHCH2), 4.59–4.58 (m, 2H, OCH2), 4.04
(t, 1H, J = 7.0 Hz, NHCH), 3.05 (dd, 1H, J = 14.0, 6.5 Hz,
ArCH2), 3.00–2.94 (m, 1H, ArCH2), 1.32 (s, 9H, OCMe3); 13C
NMR (CD3OD) d 169.1, 163.1, 154.7, 131.4, 129.8, 124.2, 118.1,
115.3, 78.4, 66.1, 65.6, 33.0, 27.9; IR (thin film) n 3424.9, 3258.6,
2977.0, 1739.9, 1507.1, 1365.8, 1236.6, 1162.8 cm-1; HRMS (ESI)
calcd for C16H23NO4 [M+H]+ m/z: 294.1705, found 294.1707.
(S)-Allyl 6-(tert-butoxycarbonylamino)-2-(hydroxyamino) hex-
anoate oxalate (18). (S)-Allyl 6-(tert-butoxycarbonylamino)-2-
(cyanomethylamino)hexanoate (3.22 g, 9.90 mmol, 1.00 equiv)
was converted to the corresponding N-hydroxylamine oxalate
according to the general procedure, and isolated as a white solid
(3.49 g, in 90% yield). mp 105–106.5 ◦C; [a]D -1.1 (c 0.54,
20
MeOH); 1H NMR (CD3OD) d 6.01–5.93 (m, 1H, OCH2CHCH2),
5.37 (dd, 1H, J = 15.5, 1.5 Hz, CHCH2), 5.26 (d, 1H, J = 10.5 Hz,
CHCH2), 4.70 (d, 2H, J = 5.5 Hz, OCH2), 3.84 (br s, 1H, NHCH),
3.03 (t, 2H, J = 6.7 Hz, NHCH2), 1.76 (br d, 2H, CH2), 1.43–1.36
(m, 13H, 2 ¥ CH2 and OCMe3); 13C NMR (CD3OD) d 171.0, 164.4,
157.3, 131.8, 118.0, 78.6, 65.8, 64.5, 39.6, 29.4, 27.8, 27.5, 22.6; IR
(thin film) n 3390.2, 2936.5, 2975.6, 1746.2, 1692.7, 1522.0, 1263.2,
1175.4 cm-1; HRMS (ESI) calcd for C14H26N2O5Na [M+Na]+ m/z:
325.1739, found 325.1734.
53% yield). mp 138–140 ◦C; [a]D -19.3 (c 0.15, MeOH); 1H NMR
20
(DMSO-d6) d 9.70 (br s, 3H, OH), 3.71–3.68 (m, 1H, NHCH),
1.42 (s, 9H, OCMe3), 1.18 (d, 3H, J = 6.5 Hz, Me); 13C NMR
(DMSO-d6) d 170.8, 163.2, 81.4, 59.8, 27.7, 13.5; IR (KBr) n
3424.4, 2296.3, 2937.0, 1747.1, 1736.1, 1631.9, 1217,8, 1157.5; ESI
calcd for C7H16NO3 [M+H]+ m/z: 162.1, found 162.5.
General procedure for the preparation of N-benzylidene nitrones
(S)-N-Benzylidene-1-carboxyethanamine oxide (20). The fol-
lowing procedure is representative. (S)-tert-butyl 2-(hydro-
xyamino)propanoate oxalate (12) (0.25 g, 0.99 mmol, 1.0 equiv)
was dissolved in CH2Cl2, (9 mL). To this solution, benzaldehyde
(0.15 mL, 1.5 mmol, 1.5 equiv) and triethylamine (0.41 mL,
3.0 mmol, 3.0 equiv) were added and the mixture stirred overnight.
The reaction was transferred to a separatory funnel, and H2O
(40 mL) was added. The organic phase was removed; the aqueous
solution was extracted with CH2Cl2 (2 ¥ 40 mL), and the organic
phases were combined and washed with brine, dried over Na2SO4,
filtered and concentrated in vacuo. The product was purified by
flash column chromatography on silica (30% EtOAc in hexanes),
to afford (S)-N-benzylidene-1-tert-butoxy-1-oxopropan-2-amine
oxide (19) as white foam (0.20 g, 81% yield). (S)-N-Benzylidene-1-
tert-butoxy-1-oxopropan-2-amine oxide (19) (0.491 g, 1.98 mmol,
1.00 equiv) was dissolved in a 1 : 1 solution of CH2Cl2/TFA
(20 mL) and stirred for 2 h at rt. The solvent was reduced
to minimum volume, and cold Et2O (15 mL) was added to
the solution to provide a white precipitate. The precipitate was
collected by vacuum filtration, and washed several times with Et2O
(S)-tert-Butyl 2-(hydroxyamino)-4-methylpentanoate oxalate
(14). (S)-tert-Butyl 2-(cyanomethylamino)-4-methylpentanoate
(0.550 g, 2.28 mmol, 1.00 equiv) was converted to the correspond-
ing N-hydroxylamine oxalate according to the general procedure
and isolated as a white solid (0.650 g, 97% yield). mp 97–100 ◦C;
[a]D +1.7 (c 0.34, MeOH); H NMR (D3COD) d 3.94 (t, 1H,
J = 7.0 Hz, NHCH), 1.80–1.77 (br m, 1H, CHMe2), 1.70–1.68 (br
m, 2H, iPrCH2), 1.53 (s, 9H, OCMe3), 0.99 (t, 6H, J = 5.5 Hz,
2 ¥ Me); 13C NMR (D3COD) d 169.2, 165.0, 85.0, 64.3, 37.5, 28.2,
26.1, 23.2, 22.0; IR (KBr) n 3426.4, 2963.5, 1743.8, 1596.2, 1253.9,
1158.0 cm-1; ESI calcd for C10H22NO3 [M+H]+ m/z: 204.2, found
204.5.
20
1
(S)-tert-Butyl 2-(hydroxyamino)-3-phenylpropanoate oxalate
(15). (S)-tert-Butyl 2-(cyanomethylamino)-3-phenylpropanoate
(2.72 g, 10.5 mmol, 1.0 equiv) was converted to the corresponding
N-hydroxylamine oxalate according to the general procedure and
◦
20
isolated as a white solid (3.21 g, 93% yield). mp 89–92 C; [a]D
1
+16.7 (c 0.12, MeOH); H NMR (D3COD) d 7.34–7.26 (m, 5H,
Ph), 4.20–4.17 (br m, 1H, NHCH), 3.31–3.26 (m, 1H, PhCH2),
to afford 20 (0.30 g, 78% yield). mp 141–142 ◦C; [a]D -23.39 (c
3.03 (dd, 1H, J = 13.5, 9.5 Hz, PhCH2), 1.33 (s, 9H, OCMe3); 13
C
20
1.03, MeOH); 1H NMR (CD3OD) d 8.28 (dd, 2H, J = 7.5, 1.5 Hz,
Ph), 7.96 (s, 1H, PhCHNO), 7.50–7.46 (m, 3H, Ph), 5.03 (q, 1H,
J = 7.0 Hz, CH), 1.71 (d, 3H, J = 7.0 Hz, Me); 13C NMR (CD3OD)
d 171.0, 140.0, 132.6, 131.4, 130.7, 129.6, 73.5, 15.6; IR (KBr) n
3436.0, 29.0, 1752.0, 1633.4, 1148.4, 1082.8; HRMS (ESI) calcd
for C10H12NO3 [M+H]+ m/z: 194.0817, found 194.0819.
NMR (D3COD) d 168.6, 164.4, 135.8, 130.6, 129.7, 128.5, 84.8,
66.7, 34.7, 28.0; IR (thin film) n 2980.4, 2561.0, 2492.5, 1740.9,
1642.5, 1253.9, 1155.1, 1253.9, 1155.1 cm-1; HRMS (ESI) calcd
for C13H19NO3Na [M+Na]+ m/z 260.1263 found 260.1241.
(S)-Allyl 3-(4-tert-butoxyphenyl)-2-(hydroxyamino) propano-
ate
oxalate
(17). (S)-Allyl
3-(4-tert-butoxyphenyl)-2-
(cyanomethylamino)propanoate (1.02 g, 3.10 mmol, 1.00 equiv)
(S)-N-Benzylidene-1-carboxy-3-methylbutan-1-amine
oxide
was converted to the corresponding N-hydroxylamine oxalate
(23). (S)-tert-Butyl
2-(hydroxyamino)-4-methylpentanoate
3414 | Org. Biomol. Chem., 2010, 8, 3405–3417
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
oxalate (14) (0.96 g, 3.30 mmol, 1.00 equiv) was converted to
the corresponding N-benzylidene nitrone ester according to the
general procedure, and purified by flash column chromatography
on silica (15% EtOAc in hexanes) to provide (S)-N-benzylidene-1-
tert-butoxy-4-methyl-1-oxopentan-2-amine oxide as a white solid
(0.83 g, 86% yield). (S)-N-benzylidene-1-tert-butoxy-4-methyl-1-
oxopentan-2-amine oxide (0.340 g, 1.17 mmol, 1.00 equiv) was
dissolved in 1 : 1 CH2Cl2/TFA (12 mL) and stirred for 2 h at rt.
The solution was concentrated in vacuo and compound 23 was
isolated without further purification as a colorless oil (0.230 g,
84% yield). [a]D -86.9 (c 0.16, MeOH); H NMR (CD3OD) d
8.29 (d, 2H, J = 6.5 Hz, Ph), 8.00 (s, 1H, PhCHNO), 7.51–7.47
(m, 3H, Ph), 4.96 (dd, 1H, J = 10.5, 4.5 Hz, CH), 2.32–2.26
(m, 1H, iPrCH2), 1.86–1.80 (m, 1H, iPrCH2), 1.62–1.57 (m, 1H,
CHMe2), 1.02–1.00 (m, 6H, Me ¥ 2); 13C NMR (CD3OD) d 169.8,
139.5, 131.3, 130.1, 129.4, 128.4, 75.4, 37.4, 24.7, 22.2, 20.5; IR
(thin film) n 2959.2, 1672.4, 1458.4, 1199.9; HRMS (ESI) calcd
for C13H16NO3 [M-H]- m/z: 234.1, found 234.6.
and extracted with CH2Cl2 (3 ¥ 70 mL). The organic phases were
combined, dried over Na2SO4, filtered, and concentrated in vacuo.
The product was recrystallized from Et2O to provide compound
◦
20
26 as a white solid in (0.23 g, 50% yield). mp 138–140 C; [a]D
1
-22.8 (c 0.14, MeOH); H NMR (CDCl3) d 8.27 (d, 2H, J =
7.5 Hz, Ph), 7.58–7.48 (m, 4H, Ph and PhCHNO), 4.52–4.49 (m,
1H, CH), 3.10 (br s, 2H, NHCH2), 2.35–2.31 (m, 1H, CHCH2),
2.13 (br d, 1H, CHCH2), 1.56–1.53 (m, 2H, CH2), 1.48–1.40
(m, 11H, CH2 and OCMe3); 13C NMR (CDCl3) d 169.4, 156.1,
139.7, 132.5, 130.4, 128.9, 128.7, 79.2, 75.2, 40.1, 31.3, 29.4, 28.4,
23.1; IR (thin film) n 3340.5, 2932.2, 1687.4, 1520.1, 1165.7 cm-1;
HRMS (ESI) calcd for C18H26N2O5Na [M+Na]+ m/z: 373.1739,
found 373.1731.
20
1
(S)-N -Benzylidene-2-(4-tert-butoxyphenyl)-1-carboxyethan-
amine oxide (25). (S)-Allyl 3-(4-tert-butoxyphenyl)-2-(hydroxy-
amino)propanoate oxalate (17) (2.0 g, 5.2 mmol, 1.0 equiv) was
dissolved in DMF (13 mL) and converted to the correspond-
ing nitrone according to the general procedure. The product
was purified by flash column chromatography on silica (15%
EtOAc in hexanes) to afford (S)-1-(allyloxy)-N-benzylidene-3-
(4-tert-butoxyphenyl)-1-oxopropan-2-amine oxide as a colorless
oil (1.51 g, 75% yield). (S)-1-(Allyloxy)-N-benzylidene-3-(4-tert-
butoxy phenyl)-1-oxopropan-2-amine oxide (1.34 g, 3.51 mmol,
1.00 equiv) was converted to the carboxylic acid according to the
above procedure to afford compound 25 as a white solid (0.710 g,
(S)-N-Benzylidene-1-carboxy-2-phenylethanamine oxide (24).
(S)-tert-Butyl 2-(hydroxyamino)-3-phenylpropanoate oxalate (15)
(0.16 g, 0.48 mmol, 1.0 equiv) was converted to the correspond-
ing N-benzylidene nitrone according to the general procedure,
and was purified by flash column chromatography on silica
(15% EtOAc in hexanes) to provide (S)-N-benzylidene-1-tert-
butoxy-1-oxo-3-phenylpropan-2-amine oxide as a white solid
(0.150 g, 96% yield). (S)-N-Benzylidene-1-tert-butoxy-1-oxo-3-
phenylpropan-2-amine oxide (2.57 g, 7.90 mmol, 1.00 equiv) was
dissolved in 1 : 1 CH2Cl2/TFA (40 mL) and stirred for 2 h at rt.
The solution was concentrated in vacuo and compound 24 was
isolated as a white solid without further purification (2.11 g, 99%
59% yield). mp 130–132 ◦C; [a]D -211.2 (c 0.16, MeOH); H
NMR (CDCl3) d 8.01 (d, 2H, J = 7.5 Hz, Ph), 7.50–7.38 (m,
3H, Ph), 7.08 (d, 2H, J = 8.0 Hz, Ar), 6.83 (d, 2H, J = 8.5 Hz,
Ar), 6.77 (s, 1H, PhCHNO), 4.46 (dd, 1H, J = 11.5, 3.5 Hz,
CH), 3.49–3.44 (m, 1H, ArCH2), 3.36 (dd, 1H, J = 10.5, 3.5 Hz,
ArCH2), 1.22 (s, 9H, OCMe3); 13C NMR (CDCl3) d 168.9, 155.1,
140.5, 132.8, 130.5, 130.2, 129.6, 128.9, 128.0, 124.8, 78.8, 76.3,
37.8, 28.8; IR (thin film) n 2976.1, 1734.6, 1506.6, 1450.6, 1237.1,
1160.4; HRMS (ESI) calcd for C20H24NO4 [M+H]+ m/z: 342.1705,
found 342.1715.
20
1
yield). mp 155–157 ◦C; [a]D -84.4 (c 0.32, MeOH); H NMR
(CD3OD) d 8.10 (d, 2H, J = 7.5 Hz, Ph), 7.45–7.37 (m, 4H, Ph
and PhCHNO), 7.26–7.16 (m, 5H, Ph), 4.88 (dd, 1H, J = 10.5,
4.0 Hz, CH), 3.54 (m, 1H, PhCH2), 3.37 (dd, 1H, J = 14.3, 4.7 Hz,
PhCH2); 13C NMR (CD3OD) d 169.5, 140.2, 137.4, 132.1, 130.4,
129.7, 129.3, 129.2, 127.8, 79.3, 35.9; IR (KBr) n 3438.9, 3029.1,
2863.2, 1725.9, 1456.4, 1229.8, 1130.5; HRMS (ESI) calcd for
C16H16NO3 [M+H]+ m/z: 270.1130, found 270.1117.
20
1
General procedure for the synthesis of N-terminal peptide
hydroxylamines
(S)-N -Benzylidene-5-(tert-butoxycarbonylamino)-1-carboxy-
pentan-1-amine oxide (26). (S)-Allyl 6-(tert-butoxycarbonyl-
amino)-2-(hydroxyamino)hexanoate oxalate (18) (2.06 g,
5.25 mmol, 1.00 equiv) was dissolved in DMF (26 mL) and
converted to the corresponding N-benzylidene nitrone according
to the general procedure, and purified by flash chromatography
on silica (45% EtOAc in hexanes) to afford (S)-1-(allyloxy)-N-
benzylidene-6-(tert-butoxycarbonyl amino)-1-oxohexan-2-amine
oxide as white foam (1.64 g, 80% yield). To a round bottom flask
charged with a magnetic stir bar, Cl2Pd(PPh3)2 (0.53 g, 0.13 mmol,
10 mol%) PPh3 (0.11 g, 0.41 mmol, 30 mol%), and THF (7 mL)
were stirred under nitrogen for 40 min. (S)-1-(Allyloxy)-N-
benzylidene-6-(tert-butoxycarbonylamino)-1-oxohexan-2-amine
oxide was placed under an inert atmosphere of nitrogen and
dissolved in THF (6 mL). The solution was add to the palladium
slurry with a syringe and morpholine (1.2 mL) was added
dropwise to the reaction mixture. After stirring 2 h at rt, the
solution was concentrated in vacuo and the resultant yellow solid
dissolved in H2O (50 mL) and washed with CH2Cl2 (3 ¥ 50 mL).
The aqueous solution was then acidified with 1 N HCl to a pH ~ 3
a-N-Hydroxy-Tyr-Ala-Lys-Pro-Ala-Leu-NH2 (34). The pep-
tide sequence Try-Ala-Lys-Pro-Ala-Leu attached to Rink amide
resin (0.110 g) was swelled in CH2Cl2 (2.0 mL) for 5 min. (S)-N-
Benzylidene-2-(4-tert-butoxyphenyl)-1-carboxyethanamine oxide
(25) (0.110 g, 0.310 mmol, 4.00 equiv) and HBTU (0.120 g,
0.310 mmol, 3.90 equiv) were dissolved in DMF (1.00 mL), fol-
lowed by addition of DIPEA (0.110 mL, 0.630 mmol, 8.00 equiv).
After agitation for 5 min, the resin was drained and the solution
containing compound 25 was added to the resin; the solution
was agitated until reaction completion was confirmed by Kaiser
test. The resin was consecutively rinsed several times with DMF,
CH2Cl2, DMF, and MeOH. The resin was dried under vacuum,
placed in a glass vial and treated with a solution of 99% TFA
in CH2Cl2. After agitation for 30 min, the resin was filtered and
washed with TFA (1.0 mL). The filtrate was placed under a stream
of N2 and reduced to a volume of 1.0 mL. Cold Et2O was added
to the solution, inducing a white precipitate that was collected
by vacuum filtration and rinsed several times with cold Et2O
to provide the N-benzylidene nitrone peptide as a white solid
(0.050 g). The peptide (0.025 g) was dissolved in a solution of
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 3405–3417 | 3415
©

View Article Online
5% TFA in H2O. The solution was passed through a column
of C-18 silica layered with hydroxylamine Wang resin (0.134 g,
2.00 mmol/g, 10.0 equiv) several times. The resulting solution
was then lyophilized and the crude hydroxylamine was purified by
preparative HPLC using a gradient of 5–50% iPrOH in H2O over
30 min with a flow rate of 10.0 mL/min and monitored at 220 nm.
Compound 34 was isolated as a white solid (0.015 g, 66% yield).
1H NMR (D3COD) d 7.10–7.02 (m, 4H), 6.73–6.69 (m, 4H), 4.59
(br m, 1H), 4.47 (br m, 1H), 4.38–4.21 (m, 4H), 4.13 (q, 1H, J =
7.0 Hz), 3.88–3.74 (m, 2H), 3.57–3.51 (m, 2H), 3.17–3.00 (m, 2H),
2.94–2.80 (m, 6H), 2.15 (br m, 1H), 2.00–1.96 (m, 3H), 1.71–1.56
(m, 8H), 1.40–1.29 (m, 7H), 1.18–1.16 (m, 2H), 0.97–0.91 (m,
6H); HPLC retention time: 13.3 min. at 220 nm, column: YMC
R-ODS-10A 250 ¥ 20 mm, flow rate: 1 mL/min, gradient: 5–50%
iPrOH in H2O over 30 min. HRMS (ESI) calcd for C41H62N9O10
[M+H]+ 840.4620, found 840.4611.
Arora, N. Kaur and O. IV. Phanstiel, “Chemoselective N-Acylation
via Condensations of N-(Benzoyloxy)amines and a-Ketophosphonic
Acids under Aqueous Conditions”, J. Org. Chem., 2008, 73, 6182–
6186.
4 (a) M. Schno¨lzer and S. B. H. Kent, “Construction Proteins by
Dovetailing Unprotected Synthetic Peptides: Backbone-Engineered
HIV Protease”, Science, 1992, 256, 221–225; (b) P. E. Dawson, T. W.
Muir, I. Clark-Lewis and S. B. H. Kent, “Synthesis of Proteins by
Native Chemical Ligation”, Science, 1994, 266, 776–779.
5 Reviews on native chemical ligation: (a) S. B. H. Kent, “Total Chemical
Synthesis of Proteins”, Chem. Soc. Rev., 2009, 38, 338–351; (b) C.
Haase and O. Seitz, “Extending the Scope of Native Chemical Peptide
Coupling”, Angew. Chem., Int. Ed., 2008, 47, 1553–1556; (c) D.
Macmillan, “Evolving Strategies for Protein Synthesis Converge on
Native Chemical Ligation”, Angew. Chem., Int. Ed., 2006, 45, 7668–
7672; (d) J. Rademann, “Organic Protein Chemistry: Drug Discovery
through the Chemical Modification of Proteins”, Angew. Chem., Int.
Ed., 2004, 43, 4554–4556; (e) P. E. Dawson and S. B. H. Kent, “Synthesis
of Native Proteins by Chemical Ligation”, Annu. Rev. Biochem., 2000,
69, 923–960.
6 For other recent work towards general, chemoselective peptide ligation
reactions, see: (a) D. Crich and I. Sharma, “Triblock Peptide and Pep-
tide Thioester Synthesis With Reactivity-Differentiated Sulfonamides
and Peptidyl Thioacids”, Angew. Chem., Int. Ed., 2009, 48, 7591–
7594; (b) R. Merkx, A. J. Brouwer, D. T. S. Rijkers and R. M. J.
Liskamp, “Highly Efficient Coupling of b-Substituted Aminoethane
Sulfonyl Azides with Thio Acids, toward a New Chemical Ligation
Reaction”, Org. Lett., 2005, 7, 1125–1128; (c) B. L. Nilsson, L. L.
Kiesling and R. T. Raines, “High-Yielding Staudinger Ligation of a
Phosphinothioester and Azide To Form a Peptide”, Org. Lett., 2001, 3,
9–12; (d) B. L. Nilsson, L. L. Kiessling and R. T. Raines, “Staudinger
Ligation: A Peptide from a Thioester and Azide”, Org. Lett., 2000,
2, 1939–1941; (e) E. Saxon, J. I. Armstrong and C. R. Bertozzi, ”A
“Traceless” Staudinger Ligation for the Chemoselective Synthesis of
Amide Bonds”, Org. Lett., 2000, 2, 2141–2143; (f) Z. Miao and J. P.
Tam, “Bidirectional Tandem Pseudoproline Ligations of Proline-Rich
Helical Peptides”, J. Am. Chem. Soc., 2000, 122, 4253–4260.
7 (a) L. Ju and J. W. Bode, “A General Strategy for the Preparation of C-
terminal Peptide a-Ketoacids by Solid Phase Peptide Synthesis”, Org.
Biomol. Chem., 2009, 7, 2259–2264; (b) L. Ju, A. R. Lippert and J. W.
Bode, “Stereoretentive Synthesis and Chemoselective Amide-Forming
Ligations of C-Terminal Peptide a-Ketoacids”, J. Am. Chem. Soc.,
2008, 130, 4253–4255.
8 T. Fukuzumi and J. W. Bode, “A Reagent for the Convenient,
Solid-Phase Synthesis of N-Terminal Peptide Hydroxylamines for
Chemoselective Ligations”, J. Am. Chem. Soc., 2009, 131, 3864–3865.
9 J. Wu, J. M. Comstock, J. Z. Dong, J. W. Bode “Synthesis of Human
GLP-1 (7–36) by Chemoselective a-Ketoacid–Hydroxylamine Peptide
Ligation of Unprotected Fragments” submitted for publication.
10 Direct assay of the enantiopurity of hydroxylamine is difficult. This
approach was established by Fukuyama et al.11 for the synthesis of
chiral, enantiopure hydroxylamines.
General procedure for the chemoselective ligation of peptide
fragments
Fmoc-Ala-Phe-Tyr-Ala-Lys-Tyr-Pro-Ala-Leu-NH2 (45)
Hydroxylamine 34 (11.5 mg, 0.013 mmol, 1.0 equiv), and a-
ketoacid 5 (0.010 g, 0.020 mmol, 1.5 equiv) were dissolved
in 0.40 mL of 90% DMF in H2O, and stirred overnight at
40 ◦C. The reaction was allowed to cool to room temperature
and concentrated in vacuo. The ligated product was purified by
preparative HPLC with a gradient of 40–50% iPrOH in H2O over
30 min with a flow of 10 mL/min, and monitoring at 254 nm.
Compound 45 was isolated as a white solid (0.007 g, 42% yield).
1H NMR (D3COD) d 8.18–8.04 (m, 2H), 7.81–7.77 (m, 2H), 7.68–
7.64 (m, 2H), 7.40–7.00 (m, 12H), 6.71–6.68 (m, 4H), 4.50–4.18
(m, 9H), 3.99–3.98 (m, 1H), 3.61 (br m, 1H), 3.49–3.45 (m, 1H),
3.05–2.85 (m, 7H), 19.4 (br m, 2H), 1.84 (br m, 1H), 1.68–1.57 (m,
7H), 1.40–1.15 (m, 12H), 0.95–0.91 (m, 6H); HPLC retention time:
12.4 min at 254 nm, column: YMC R-ODS-10A 250 ¥ 20 mm, flow
rate: 1 mL/min, gradient: 40–50% iPrOH in H2O over 32 min;
HRMS (ESI) calcd for C68H86N11O13 [M+H]+ m/z: 1264.6407,
found 1264.6427.
Acknowledgements
11 (a) H. Tokuyama, T. Kuboyama, A. Amano, T. Yamashita and
T. Fukuyama, “A Novel Transformation of Primary Amines to
N-Monoalkylhydroxylamines”, Synthesis, 2000, 1299–1304; (b) H.
Tokuyama, T. Kuboyama and T. Fukuyama, “Transformation of
Primary Amines to N-Monoalkylhydroxylamines: N-Hydroxy-(S)-1-
Phenylethylamine Oxalate”, Org. Synth., 2003, 80, 207–212.
12 (a) T. Polon´ski and A. Chimiak, “Oxidation of Amino Acids Esters
into N-Hydroxyamino Acid Derivatives”, Tetrahedron Lett., 1974, 15,
2453–2456; (b) T. Polon´ski and A. Chimiak, “Nitrones as Intermediates
in the Synthesis of N-Hydroxyamino Acid Esters”, J. Org. Chem., 1976,
41(12), 2092–2095; (c) G. Grundke, W. Keese and M. Rimpler, “Opti-
cally Active N-Hydroxy-a-L-Amino Acid Methyl Esters: an Improved
and Simplified Synthesis”, Synthesis, 1987, (12), 1115–1116; (d) Y.-M.
Lin and M. J. Miller, “Practical Synthesis of Hydroxamate-Derived
Siderophore Components by an Indirect Oxidation Method and
Syntheses of a DIG–Siderophore Conjugate and a Biotin Siderophore
Conjugate”, J. Org. Chem., 1999, 64, 7451–7458.
This work was supported by the National Institutes of Health,
National Institute of General Medical Science (NIGMS) [NIH
R01-GM76320] and the David and Lucille Packard Foundation
(Fellowship to J.W.B). Generous support from Bristol Myers
Squibb, Roche, Amgen, and Eli Lilly is gratefully acknowledged.
Notes and references
1 (a) J. W. Bode, R. M. Fox and K. D. Baucom, “Chemoselec-
tive Amide Ligations by Decarboxylative Condensations of N-
Alkylhydroxylamines and a-Ketoacids”, Angew. Chem., Int. Ed., 2006,
45, 1248–1252; (b) N. Carrillo, E. A. Davalos, J. A. Russak and
J. W. Bode, “Iterative, Aqueous Synthesis of b3-Oligopeptides without
Coupling Reagents”, J. Am. Chem. Soc., 2006, 128, 1452–1453.
2 C. P. R. Hackenberger and D. Schwarzer, “Chemoselective Ligation
and Modification Strategies for Peptides and Proteins”, Angew. Chem.,
Int. Ed., 2008, 47, 10030–10074.
13 (a) Y. Ye, M. Liu, J. L.-F. Kao and G. R. Marshall, “Novel
Trihydroxamate-Containing Peptides: Design, Synthesis, and Metal
Coordination”, Biopolymers, 2006, 84, 472–489; (b) Y. Ye, M. Liu,
J. L.-F. Kao and G. R. Marshall, “Peptide-Bond Modification for
Metal Coordination: Peptides Containing Two Hydroxamate Groups”,
Biopolymers, 2003, 71, 489–515.
3 (a) A. K. Sanki, R. S. Talan and S. J. Sucheck, “Synthesis of Small
Glycopeptides by Decarboxylative Condensation and Insight into the
Reaction Mechanism”, J. Org. Chem., 2009, 74, 1886–1896; (b) J. S.
3416 | Org. Biomol. Chem., 2010, 8, 3405–3417
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
14 M. D. Wittman, R. L. Halcomb and S. J. Danishefsky, “On the
Conversion of Biologically Interesting Amines to Hydroxylamines”,
J. Org. Chem., 1990, 55, 1981–1983.
22 K. Mori and K. Koseki, “Synthesis of Trichostatin A, a Potent
Differentiation Inducer of Friend Leukemic Cells, and its Antipode”,
Tetrahedron, 1988, 44, 6013–6020.
23 P. H. Bentley and G. Brooks, “Semi-Synthetic Penicillins and
Cephalosporins Incorporating a Hydroxyamino Group”, Tetrahedron
Lett., 1976, 17, 3735–3738.
24 J. J. D´ıaz-Mocho´n, L. Bialy and M. Bradley, “Full Orthogonality
between Dde and Fmoc: the Direct Synthesis of PNA–Peptide Conju-
gates”, Org. Lett., 2004, 6, 1127–1129.
25 (a) E. Buehler and G. B. Brown, “A General Synthesis of
N-Hydroxyamino Acids”, J. Org. Chem., 1967, 32, 265–267;
(b) W. Oppolzer and O. Tamura, “Asymmetric Synthesis of a-
\ Amino Acids and a-N-Hydroxyamino Acids via Electrophilic
Amination of Bornanesultam-Derived Enolates with 1-Chloro-1-
Nitrosocyclohexane”, Tetrahedron Lett., 1990, 31, 991–994.
26 S. M. Turega, C. Lorenz, J. W. Sadownik and D. Philip, “Target-Driven
Selection in a Dynamic Nitrone Library”, Chem. Commun., 2008, (34),
4076–4078.
27 (a) D. Bang and S. B. H. Kent, “A One-Pot Total Synthesis of
Crambin”., Angew. Chem., Int. Ed., 2004, 43, 2534–2538; (b) V. Y.
Torbeev and S. B. H. Kent, “Convergent Chemical Synthesis and
Crystal Structure of a 203 Amino Acid “Covalent Dimer” HIV-1
Protease Enzyme Molecule”, Angew. Chem., Int. Ed., 2007, 46, 1667–
1670; (c) T. Durek, V. Y. Torbeev and S. B. H. Kent, “Convergent
Chemical synthesis and High-resolution X-ray Structure of Human
Lysozyme”, Proc. Natl. Acad. Sci. U. S. A., 2007, 104, 4846–4851.
15 Fukuyama et al.11 report this procedure without isolation of the
intermediates.
16 H. C. J. Ottenheijm and J. D. M. Herscheid, “N-Hydroxy-a-amino
Acids in Organic Chemistry”, Chem. Rev., 1986, 86, 697–707.
17 D. A. Knowles, C. J. Mathews and N. C. O. Tomkinson, “Oxidation of
Primary Amines to Ketones”, Synlett, 2008, 18, 2769–2772.
18 D. W. Knight and M. P. Leese, “A Survey of Suitable Protecting
Groups for the Synthesis of Hydroxylamines by Mitsunobu Reactions”,
Tetrahedron Lett., 2001, 42, 2593–2595.
19 R. Braslau, J. R. Axon and B. Lee, “Synthesis of N-Hydroxy Peptides:
Chemical Ligation of O-Acyl Hydroxamic Acids”, Org. Lett., 2000, 2,
1399–1401.
20 M. L. Di Gioia, A. Leggio, A. Le Pera, A. Liguori and C. Siciliano, “Op-
tically Pure N-Hydroxy-O-triisopropylsilyl-a-L-amino Acid Methyl
Esters from AlCl3-Assisted Ring Opening of Chiral Oxaziridines by
Nitrogen Containing Nucleophiles”, J. Org. Chem., 2005, 70, 10494–
10501.
21 (a) W. B. Lutz, “O-Triphenylmethylhydroxylamine (Trityloxyamine),
a useful O-Protected form of Hydroxylamine”, J. Org. Chem., 1971,
36, 3835–3837; (b) S.-M. Yang, B. Lagu and L. J. Wilson, “Mild and
Efficient Lewis Acid-Promoted Detritylation in the Synthesis of N-
Hydroxy Amides: A Concise Synthesis of (-)-Cobactin T”, J. Org.
Chem., 2007, 72, 8123–8126.
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 3405–3417 | 3417
©