Oxime Ligation via in situ Oxidation of N-Phenylglycinyl Peptides

Article abstract of DOI:10.1021/acs.orglett.8b00713

Mild conditions for oxime ligations via in situ generation of α-imino amide intermediates are reported. The evaluation of a variety of N-terminal N-phenylglycine residues revealed that a metal-free, chemoselective oxidation was possible using oxygen as the only oxidant in buffer at pH 7.0. Moreover, selective unmasking of an inert residue by addition of potassium ferricyanide is demonstrated. These simple and mild conditions, which can be fine-tuned by the electronic properties of the N-phenylglycine residue, offer unique advantages over conventional approaches for oxime ligations.

Full text of DOI:10.1021/acs.orglett.8b00713

Letter
pubs.acs.org/OrgLett
Cite This: Org. Lett. XXXX, XXX, XXX−XXX
Oxime Ligation via in situ Oxidation of N‑Phenylglycinyl Peptides
Quibria A. E. Guthrie and Caroline Proulx*
Department of Chemistry, North Carolina State University, Raleigh, North Carolina 27695-8204, United States
S
* Supporting Information
ABSTRACT: Mild conditions for oxime ligations via in situ generation of α-imino amide intermediates are reported. The
evaluation of a variety of N-terminal N-phenylglycine residues revealed that a metal-free, chemoselective oxidation was possible
using oxygen as the only oxidant in buffer at pH 7.0. Moreover, selective unmasking of an inert residue by addition of potassium
ferricyanide is demonstrated. These simple and mild conditions, which can be fine-tuned by the electronic properties of the N-
phenylglycine residue, offer unique advantages over conventional approaches for oxime ligations.
cleavage from the resin and purification, such as periodate
hemoselective ligation reactions are utilized in applica-
Ctions ranging from total chemical protein synthesis to site- oxidation of serine residues, are required to access glyoxylyl-
selective bio-orthogonal protein modifications and beyond.1
For example, α-oxo aldehydes have been used in ligation
substituted peptides. These conditions have been shown to
degrade Met, Cys, Tyr, and Trp residues.9 To circumvent these
issues, we report the mild and chemoselective oxidation of an
N-terminal N-phenylglycine residue, generating the highly
reactive α-imino amide intermediates in situ (Scheme 1b).
Chemoselective oxidation of N-phenylglycine residues has
been exploited in the past for the synthesis of unnatural amino
acids and peptide derivatives.10 However, these transformations
often require redox-active catalysts and strong chemical
oxidants in stoichiometric amounts.10 Similarly, electron-rich
N-terminal N-arylglycines were shown to oxidize to the α-oxo
aldehyde in the presence of AgClO4.11 Recently, auto-oxidative
coupling reactions with N-phenylglycine esters was reported
under aerobic conditions with trace amounts of acid in a
MeCN/DCE solvent mixture.12 Herein, we report that
oxidation of N-arylglycines and coupling with aminooxy
functional groups to give oxime linkages can occur in water,
with the electronics of the N-arylglycines dictating optimal
conditions.
reactions with aminooxy and hydrazine functional groups to
give oximes and hydrazones, respectively.2 α-Effects in aminoxy
groups lead to the equilibrium favoring the oxime,3 which has
been found to be stable for up to 65 h at pH 8.4 Although the
optimal pH for the ligation was found to be near 5,5 addition of
p-methoxyaniline catalyzes this reaction at pH 7 by up to 40-
fold via formation of a protonated aniline Schiff base
intermediate (Scheme 1a).6 However, high concentrations of
aniline can be toxic to cells7 and may competitively react with
endogenous aldehydes and ketones, rendering oxime ligations
biorestricted.8 Moreover, additional synthetic steps following
Scheme 1. (a) Nucleophilic Catalysis6 and (b) in situ
Oxidation of N-Arylglycines for Oxime Ligations (This
Work)
N-(p-MeO-Ph)G-LYRAG13 was synthesized as a test peptide
sequence. The installation of the N-arylglycine was accom-
plished by bromoacetylation of the resin-bound peptide
followed by displacement with p-methoxyaniline using standard
submononer peptoid synthesis conditions.14,11 To probe the
oxidation conditions for in situ generation of the α-imino amide
intermediate and ligation, N-(p-MeO-Ph)G-LYRAG (1 mM)
and O-benzylhydroxylamine (1 mM) were combined in 100
mM ammonium acetate or phosphate buffer (Figure 1).
Ambient atmosphere was used as a mild oxidant by leaving
the reaction vessels open to air at different pH. The progress of
the reactions was monitored by LCMS at 214 nm, following
Received: March 2, 2018
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.8b00713
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Figure 1. Oxidation and coupling of N-(p-MeO-Ph)G-LYRAG (1) to
O-benzylhydroxylamine to give oxime 3 using 1 mM reactant
concentration. The extent of the reaction was monitored by LCMS
at 214 nm. Each experiment was performed in triplicate.
Figure 2. Percent oxidation of substituted N-phenylglycine-LYRAG
analogues 1a−g for pH 4.5−7 after 24 h. Each experiment was
performed in triplicate.
quenching of a small aliquot with NaOH. Although < 2%
ligation product was observed at pH 7 after 24 h, decreasing the
pH to 4.5 afforded oxime 3 in 77% conversion, with no trace of
the α-oxo aldehyde (Figure 1). The optimal pH was found to
be about one unit below the pKa of p-methoxyaniline (pKa 5.3),
suggesting that protonation of the N-arylglycine may facilitate
its oxidation to 2.
pH conditions, which is essential for biomolecule stability, and
(2) prevents acid-promoted decomposition during resin
cleavage and purification. Moreover, it greatly simplifies the
synthesis of the required precursors for oxime ligation,
precluding the incorporation and periodate oxidation of N-
terminal serine residues.
Upon closer examination, we found that oxidation of 1f was
complete after 8 h. However, a 70:30 mixture of the oxime and
α-oxo aldehyde was observed under these conditions (Table 1,
entry 1). Although the aldehyde would convert to the desired
oxime after 24 h, to circumvent the competing hydrolysis of the
α-imino amide intermediate, the ratio of O-benzylhydroxyl-
amine/N-phenylG-LYRAG peptide was increased to 5:1. We
noted inconsistencies in our results under these conditions
along with reduced rates of oxidation, which may be due to
excess O-benzylhydroxylamine sequestering oxygen and in-
hibiting N-phenylglycine oxidation. To address these issues, the
reaction was performed under an O2 atmosphere, providing
oxime 3 in 91% conversion after only 8 h and decreasing the
population of the α-oxo aldehyde to ∼8% (Table 1, entry 2).
Of note, current aniline-catalyzed oxime ligation procedures
require 24 h to go to completion at neutral pH. To more
closely mimic typical concentrations of biomolecules, the
tandem oxidation/ligation reaction sequence was attempted
under dilute conditions, namely 0.1 mM (1f) and 0.5 mM (O-
benzylhydroxylamine). While this 10-fold dilution led to an
expected increase in aldehyde byproduct (40%), the desired
oxime remained the major product (60%) under these low
concentrations (Table 1, entry 3).
Pleased to bypass NaIO4 in the synthesis of α-oxo aldehydes,
we explored oxidation of other N-arylglycines to achieve faster
kinetics at neutral pH. Seven N-aryl-G-LYRAG analogues (1a−
g) were synthesized, and their conversion to the oxime product
was quantified after 24 h at different pH (Figure 2). Inspired by
aniline catalysis in oxime ligation of α-oxo aldehydes15 and
having noticed a relationship between aniline pKa and oxidation
rate, electron-rich N-arylglycines were targeted. Moreover,
because anthranilic acid derivatives were previously shown to
be superior catalysts in ligations with α-oxo aldehydes, we
included analogue 1g, which was envisioned to protonate the
N-phenylglycine via an intramolecular proton transfer.15a−c 4-
(NMe2)aniline is known to be unstable in air-saturated
water;15a however, analogue 1f was anticipated to oxidize to
the desired α-imino amide intermediate. Most analogues were
either unreactive under the entire pH range tested (1d,e,g)16 or
exhibited similar reactivity trends as analogue 1b (1c), where a
pH of ∼4.5 was required to achieve good conversions after 24
h. In contrast, we were pleased to find that N-(p-Me2N-Ph)G-
LYRAG (1f) was quantitatively oxidized to a mixture of oxime
3 (major) and α-oxo aldehyde (minor) at pH 6.5 and 7.0. This
analogue displayed a reverse reactivity trend where lower pH
values decreased the oxidation rate, presumably due to
protonation of the 4-dimethylamino group, rendering it
electron poor. This trend (1) allows reactivity at physiological
This convenient aerobic one-pot oxidation/oxime ligation
reaction sequence should find broad applicability in bio-
conjugations, in particular, where exposure to periodate
B
DOI: 10.1021/acs.orglett.8b00713
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Table 1. Oxidation Conditions for N-Phenylglycine-Terminated Peptides 1f and 1b and Conversions (%) as Determined by
LCMS
a
entry
substrate
oxidation conditions at pH 7
air
time
initial aminooxy/peptide ratio
% oxidation
aldehyde/oxime ratio
% oxime product
1
2
3
4
5
6
7
8
1f
8 h
1:1
5:1
>99
>99
30:70
8:92
70
91
b
1f
O2
8 h
c
c
c
c
1f
air
8 h
5:1
>99
40:60
60
d
e
1f
1 mM K3[Fe(CN)6]
5 min
24 h
24 h
24 h
4 h
5:1
1:1
5:1
5:1
5:1
84
<2
14:86
N/A
7:93
70
1b
1b
1b
1b
air
<2
66
1 mM K3[Fe(CN)6]
10 mM K3[Fe(CN)6]
10 mM K3[Fe(CN)6]
71
f
>99
96
6:94
77
12:87
84
a
b
O-Benzylhydroxylamine concentration was 1 or 5 mM, and N-phenylglycine-terminated peptide concentration was 1 mM. Reaction was
c
performed under O2 atmosphere (oxygen balloon). Reaction was performed with a 10-fold dilution, with reactant concentration of either 0.1 or 0.5
mM. Reaction was performed under argon. Appearance of a byproduct is observed when >1 mM K3[Fe(CN)6] is used (see the Supporting
Information). Appearance of an over-oxidation byproduct (M + 16) is observed after 24 h.
d
e
f
reagents is problematic (e.g., in the presence of Met, Ser, Trp)
and/or where competing aldehyde functional group may exist.
To demonstrate this, N-(p-Me2N-Ph)G-LYRAM, possessing a
C-terminal methionine, was subjected to our mild ligation
conditions (O2, pH 7) with O-benzylhydroxylamine (5 mM)
for 6 h, giving the oxime product in 90% conversion with no
sign of methionine oxidation over that time (see the Supporting
Information).
In the oxime ligation of two peptides, namely 1f and
aminooxyacetyl-GRGDSGG (4), >99% oxidation of 1f was
observed after only 4 h, albeit affording an oxime/α-oxo
aldehyde ratio of ∼1:1 (Scheme 2). As a point of comparison,
by several orders of magnitude (5 min vs 8 h). To the best of
our knowledge, these conditions represent one of the fastest
reported oxime ligation reactions. Furthermore, oxidation of a
methionine side chain was not observed under these conditions
(1 mM K3[Fe(CN)6], 5 min) using N-(p-Me2N-Ph)G-LYRAM
as the peptide substrate.
In addition to identifying substituted N-phenylglycines that
could undergo mild in situ oxidation at pH 7, we were also
interested in pursuing them as masked electrophiles, where
residues that are inert at pH 7 could be selectively unmasked in
the presence of chemical oxidants. In principle, formation of α-
imino amide intermediates may be confined to environments
under oxidative stress, such as cancer cells, leading to
environmentally triggered ligations. Peptide 1b, which exhibited
very low reactivity at pH 7 under aerobic conditions, was
treated with 1−10 mM potassium ferricyanide (Table 1, entries
6−8). Gratifyingly, with 1 mM K3[Fe(CN)6], oxime 3 was
obtained in 66% conversion after 24 h, with the major
byproduct being unreacted peptide 1b (Table 1, entry 6).
Increasing K3[Fe(CN)6] concentration by 10-fold led to >99%
oxidation after 24 h; however, appearance of a byproduct with
M + 16 relative to the oxime was observed over time, likely a
result of tyrosine oxidation (Table 1, entry 7). Quenching this
reaction after 4 h led to 84% oxime formation with no
detectable overoxidation byproducts (Table 1, entry 8). The
selective unmasking of N-(p-MeO-Ph)glycines with potassium
ferricyanide is a unique feature of this chemistry, rendering it
orthogonal to aldehydes at pH 7. To confirm this, glucose was
selectively coupled to O-benzylhydroxylamine in the presence
of peptide 1b and aniline catalyst (100 mM) at pH 7 to give
oxime 6 (Scheme 3a).18
Scheme 2. Ligation of 1f with Aminooxyacetyl-GRGDSGG
(4) to Give 5 under O2 Atmosphere at pH 7
oxime 5 was previously reported to proceed to 78% conversion,
requiring 22 h reaction time, 100 mM p-methoxyaniline
catalyst, and synthesis and isolation of glyoxylyl-LYRAG.
To shorten reaction times, we evaluated mild oxidants such
as potassium ferricyanide, which was previously shown to be
compatible with biomolecules possessing free sulfhydryl groups
and glycoproteins in other oxidative coupling reactions.17
Substrate 1f (1 mM) was treated with K3[Fe(CN)6] (1 mM) in
the presence of O-benzylhydroxylamine (5 mM), and the
reaction was monitored by LCMS following quenching of a
small aliquot with tris(2-carboxyethyl)phosphine (TCEP).
Remarkably, after 5 min, peptide 1f showed 70% conversion
to oxime 3 (Table 1, entry 4). The reaction proved sensitive to
oxidant concentration, with appearance of byproducts when
treated with >1 mM potassium ferricyanide; however, careful
control over the reaction conditions reduced the reaction time
Conversely, peptide 1b was treated with 1 mM K3[Fe(CN)6]
in the presence of O-benzylhydroxylamine (1 mM) and glucose
(1 mM) with no aniline catalyst (Scheme 3b). Oxime 3 was
obtained as the only detectable product under these conditions;
however, reactivity with glucose is observed to give 6 as a minor
product when excess O-benzylhydroxylamine and glucose are
used under prolonged reaction times, likely facilitated by the
release of 1 equiv of p-anisidine upon formation of 3.
In summary, mild aqueous conditions (O2, pH 7) were
shown to provide oxime ligation products in >90% yield after 8
h using N-arylglycine-terminated-peptides in a one-pot
oxidation/ligation sequence. Under these mild conditions, a
peptide containing a C-terminal methionine residue gave the
desired oxime ligation product in 90% conversion with no sign
of side-chain oxidation to the sulfoxide or sulfone after 8 h. The
C
DOI: 10.1021/acs.orglett.8b00713
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
(4) (a) Rose, K.; Zeng, W. G.; Regamey, P. O.; Chernushevich, I. V.;
Standing, K. G.; Gaertner, H. F. Bioconjugate Chem. 1996, 7, 552−556.
(b) Kalia, J.; Raines, R. T. Angew. Chem., Int. Ed. 2008, 47, 7523−7526.
(5) Shao, J.; Tam, J. P. J. Am. Chem. Soc. 1995, 117, 3893−3899.
(6) Dirksen, A.; Hackeng, T. M.; Dawson, P. E. Angew. Chem., Int. Ed.
2006, 45, 7581−7584.
Scheme 3. Selective Ligation at pH 7 of (a) Glucose in the
Presence of 1b under Aniline Catalysis and (b) 1b in the
Presence of Glucose and Potassium Ferricyanide Oxidant
(7) Tuley, A.; Lee, Y. J.; Wu, B.; Wang, Z. U.; Liu, W. R. Chem.
Commun. 2014, 50, 7424−7426.
(8) Prescher, J. A.; Bertozzi, C. R. Nat. Chem. Biol. 2005, 1, 13−21.
(9) (a) Geoghegan, K. F.; Stroh, J. G. Bioconjugate Chem. 1992, 3,
138−146. (b) Clamp, J. R.; Hough, L. Biochem. J. 1965, 94, 17−24.
(c) Haney, C. M.; Horne, W. S. Chem. - Eur. J. 2013, 19, 11342−
11351.
(10) For select examples, see: (a) Zhao, L.; Li, C. Angew. Chem., Int.
Ed. 2008, 47, 7075−7078. (b) Zhao, L.; Basle, O.; Li, C. J. Proc. Natl.
Acad. Sci. U. S. A. 2009, 106, 4106−4111. (c) Zhang, G.; Zhang, Y.;
Wang, R. Angew. Chem., Int. Ed. 2011, 50, 10429−10432. (d) Zhu, S.
Q.; Rueping, M. Chem. Commun. 2012, 48, 11960−11962. (e) Huo, C.
D.; Wang, C.; Wu, M. X.; Jia, X. D.; Xie, H. S.; Yuan, Y. Adv. Synth.
Catal. 2014, 356, 411−415. (f) San Segundo, M.; Guerrero, I.; Correa,
A. Org. Lett. 2017, 19, 5288−5291.
reaction time can be reduced to 5 min if treated with potassium
ferricyanide, which constitutes one of the fastest reaction
conditions to yield oximes. We further demonstrated that inert
residues can be unmasked by K3[Fe(CN)6] at pH 7. This
allows for orthogonal functionalization in the presence of other
aldehydes, which is a unique feature of this chemistry. The
chemoselective oxidation of N-phenylglycines to generate
protonated aniline Schiff base intermediates in situ under
mild conditions should minimize problems associated with (1)
aniline catalyst toxicity in vivo, (2) competition with
endogeneous aldehyde and ketone groups in live cells, and
(3) over-oxidations of proteins/peptides upon prolonged
exposure to periodate reagents previously required to
synthesize α-oxo aldehydes. Further mechanistic studies and
applications of this concept are underway and will be reported
in due course.
(11) Proulx, C.; Yoo, S.; Connolly, M. D.; Zuckermann, R. N. J. Org.
Chem. 2015, 80, 10490−10497.
(12) Huo, C.; Yuan, Y.; Wu, M.; Jia, X.; Wang, X.; Chen, F.; Tang, J.
Angew. Chem., Int. Ed. 2014, 53, 13544−13547.
(13) The LYRAG peptide sequence was chosen for comparison to
aniline-catalyzed oxime ligations reported in ref 6.
(14) Zuckermann, R. N.; Kerr, J. M.; Kent, S. B. H.; Moos, W. H. J.
Am. Chem. Soc. 1992, 114, 10646−10647.
(15) (a) Crisalli, P.; Kool, E. T. J. Org. Chem. 2013, 78, 1184−1189.
(b) Larsen, D.; Pittelkow, M.; Karmakar, S.; Kool, E. T. Org. Lett.
2015, 17, 274−277. (c) Crisalli, P.; Kool, E. T. Org. Lett. 2013, 15,
1646−1649. (d) Rashidian, M.; Mahmoodi, M. M.; Shah, R.; Dozier, J.
K.; Wagner, C. R.; Distefano, M. D. Bioconjugate Chem. 2013, 24,
333−342. (e) Wendeler, M.; Grinberg, L.; Wang, X. Y.; Dawson, P. E.;
Baca, M. Bioconjugate Chem. 2014, 25, 93−101. (f) Yuen, L. H.;
Saxena, N. S.; Park, H. S.; Weinberg, K.; Kool, E. T. ACS Chem. Biol.
2016, 11, 2312−2319.
ASSOCIATED CONTENT
■
S
* Supporting Information
(16) In aniline-catalyzed hydrazone ligations, a >50-fold decrease in
reactivity at pH 7 was observed when 3,5-dimethoxyaniline was
compared with 2,4-dimethoxyaniline (see ref 15f). Given the pKa of
3,5-dimethoxyaniline (3.8), it is possible that 1d and other unreactive
analogues would undergo the tandem oxidation/oxime ligation
reaction at lower pH values (< 4.5). The pKa of 3,5-dimethoxyaniline
can be found in: Bryson, A. J. Am. Chem. Soc. 1960, 82, 4858−4862.
(17) (a) Obermeyer, A. C.; Jarman, J. B.; Netirojjanakul, C.; El
Muslemany, K.; Francis, M. B. Angew. Chem., Int. Ed. 2014, 53, 1057−
1061. (b) Yamagishi, Y.; Ashigai, H.; Goto, Y.; Murakami, H.; Suga, H.
ChemBioChem 2009, 10, 1469−1472.
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.8b00713.
Experimental procedures; LC−MS analytical data for
1a−g before and after oxime ligations (PDF)
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: cproulx@ncsu.edu.
ORCID
(18) (a) Thygesen, M. B.; Munch, H.; Sauer, J.; Clo, E.; Jorgensen,
M. R.; Hindsgaul, O.; Jensen, K. J. J. Org. Chem. 2010, 75, 1752−1755.
(b) Carrasco, M. R.; Brown, R. T. J. Org. Chem. 2003, 68, 8853−8858.
Caroline Proulx: 0000-0003-3851-793X
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
C.P. gratefully acknowledges North Carolina State University
for startup support.
■
REFERENCES
■
(1) (a) Stephanopoulos, N.; Francis, M. B. Nat. Chem. Biol. 2011, 7,
876−884. (b) Kent, S. B. H. Chem. Soc. Rev. 2009, 38, 338−351.
(2) For reviews, see: (a) El-Mahdi, O.; Melnyk, O. Bioconjugate
Chem. 2013, 24, 735−765. (b) Kolmel, D. K.; Kool, E. T. Chem. Rev.
2017, 117, 10358−10376. (c) Ulrich, S.; Boturyn, D.; Marra, A.;
Renaudet, O.; Dumy, P. Chem. - Eur. J. 2014, 20, 34−41.
(3) Sander, E. G.; Jencks, W. P. J. Am. Chem. Soc. 1968, 90, 6154−
6162.
D
DOI: 10.1021/acs.orglett.8b00713
Org. Lett. XXXX, XXX, XXX−XXX