Modification of N-terminal α-amino groups of peptides and proteins using ketenes

Article abstract of DOI:10.1021/ja208009r

A method of highly selective N-terminal modification of proteins as well as peptides by an isolated ketene was developed. Modification of a library of unprotected peptides XSKFR (X varies over 20 natural amino acids) by an alkyne-functionalized ketene (1) at room temperature at pH 6.3 resulted in excellent N-terminal selectivity (modified α-amino group/modified ε-amino group = >99:1) for 13 out of the 20 peptides and moderate-to-high N-terminal selectivity (4:1 to 48:1) for 6 of the 7 remaining peptides. Using an alkyne-functionalized N-hydroxysuccinimide (NHS) ester (2) instead of 1, the modification of peptides XSKFR gave internal lysine-modified peptides for 5 out of the 20 peptides and moderate-to-low N-terminal selectivity (5:1 to 1:4) for 13 out of the 20 peptides. Proteins including insulin, lysozyme, RNaseA, and a therapeutic protein BCArg were selectively N-terminally modified at room temperature using ketene 1, in contrast to the formation of significant or major amounts of di-, tri-, or tetra-modified proteins in the modification by NHS ester 2. The 1-modified proteins were further functionalized by a dansyl azide compound through click chemistry without the need for prior treatment.

Full text of DOI:10.1021/ja208009r

Article
pubs.acs.org/JACS
Modification of N-Terminal α-Amino Groups of Peptides and Proteins
Using Ketenes
Anna On-Yee Chan,† Chi-Ming Ho,† Hiu-Chi Chong,‡ Yun-Chung Leung,‡ Jie-Sheng Huang,†
Man-Kin Wong,*,‡ and Chi-Ming Che*,†
†Department of Chemistry, State Key Laboratory of Synthetic Chemistry, and Open Laboratory of Chemical Biology of the Institute
of Molecular Technology for Drug Discovery and Synthesis, The University of Hong Kong, Pokfulam Road, Hong Kong, China
‡Department of Applied Biology and Chemical Technology, The Hong Kong Polytechnic University, Hung Hom, Kowloon, Hong
Kong, China
S
* Supporting Information
ABSTRACT: A method of highly selective N-terminal
modification of proteins as well as peptides by an isolated
ketene was developed. Modification of a library of unprotected
peptides XSKFR (X varies over 20 natural amino acids) by an
alkyne-functionalized ketene (1) at room temperature at pH 6.3
resulted in excellent N-terminal selectivity (modified α-amino
group/modified ε-amino group = >99:1) for 13 out of the 20
peptides and moderate-to-high N-terminal selectivity (4:1 to
48:1) for 6 of the 7 remaining peptides. Using an alkyne-
functionalized N-hydroxysuccinimide (NHS) ester (2) instead
of 1, the modification of peptides XSKFR gave internal lysine-modified peptides for 5 out of the 20 peptides and moderate-to-
low N-terminal selectivity (5:1 to 1:4) for 13 out of the 20 peptides. Proteins including insulin, lysozyme, RNaseA, and a
therapeutic protein BCArg were selectively N-terminally modified at room temperature using ketene 1, in contrast to the
formation of significant or major amounts of di-, tri-, or tetra-modified proteins in the modification by NHS ester 2. The 1-
modified proteins were further functionalized by a dansyl azide compound through click chemistry without the need for prior
treatment.
amino group to generate an α-keto group7 or preformation of
INTRODUCTION
Covalent incorporation of small molecule(s) of unique
■
peptide chain pair in close proximity.5 One of the challenges is
to develop a relatively simple and general method for the
function(s) into proteins is of paramount importance in
chemical biology.1,2 These sophisticated chemical reactions
incorporation of a functional molecule into a natural amino acid
residue of peptides or proteins with high site-selectivity.
for protein modification need to be conducted in aqueous
media at low concentrations (typically <100 μM of protein)
under mild conditions and with high selectivity. Well-
documented strategies include the following: (i) direct
In this endeavor, a potentially useful approach is through
acylation of the N-terminal α-amino group, considering its
considerably lower pKa value (∼8) than that of the ε-amino
group of lysine (pKa value ∼10).1 In the literature, trans-
amination methods are applicable for selectively removing the
α-amino group of any N-terminal residue to generate an α-keto
group,2k,8,9 using pyridoxal at 100 °C8 or using glyoxylate in the
presence of divalent metal ion and base at room temperature.9
The biomimetic transamination under mild conditions with
subsequent oxime formation7 can efficiently modify various N-
terminal residues except histidine, tryptophan, lysine, proline,
cysteine, serine, and glutamine. Common reagents for amine
acylation, such as N-hydroxysuccinimide (NHS) esters, usually
afford heterogeneous bioconjugates due to facile modification
of −OH group(s) or internal lysine residue(s),10,11 although at
pH 6.5 a moderate N-terminal selectivity of 3:1 (ratio of α-
amino- to ε-amino-modified peptide) has been reported for
modification of natural amino acid residue(s) such as lysine,3a−e
cysteine,3f−k,4b,c,e,f tryptophan,3l,m,4a,d and serine,3n,4a including
those at the N-terminus,4 in preintroduced sequence,3f,g,i,n or in
paired peptide chains;5 (ii) installation of unnatural functional
group(s)6,7 at the natural amino acid residue(s) including the
N-terminal residue,7 followed by reactions such as condensa-
tion,6a−c,7 Staudinger ligation,6d−g olefin metathesis,6h and 1,3-
dipolar cycloaddition.6i−n Protein modification based on these
strategies usually relies on the presence of one or few particular
types of amino acid residues. Notably, the amino acid residue to
be modified can vary over a range of amino acids and yet can
retain high site-selectivity in peptide modification at N-terminal
residue by a biomimetic transamination with subsequent oxime
formation7 and in peptide modification at the side chain of an
internal amino acid residue by a molecular recognition
combined with metal-catalyzed carbenoid transfer reaction.5
The two methods feature preremoval of the N-terminal α-
Received: August 26, 2011
Published: January 30, 2012
© 2012 American Chemical Society
2589
dx.doi.org/10.1021/ja208009r | J. Am. Chem.Soc. 2012, 134, 2589−2598

Journal of the American Chemical Society
Article
modification of peptide IQKVAGTWYSLA with NHS−LC−
biotin.11b,12 It thus would be of interest to seek an amine
acylation reagent suitable for modification of peptides and
proteins with high N-terminal selectivity.
copper-catalyzed click reaction with an organic azide21 (Scheme
1). Remarkably, in contrast to an alkyne-functionalized NHS
ester (2), ketene 1 exhibited good-to-excellent N-terminal
selectivity in modifying almost all of a library of unprotected
peptides XSKFR with X varying over 20 natural amino acids.
In our efforts to develop new bioconjugation reactions,13,14
we have reported a “[Mn(2,6-Cl2TPP)Cl]/alkyne/H2O2”
protocol14 [H2(2,6-Cl2TPP) = meso-tetrakis(2,6-
dichlorophenyl)porphyrin] for N-terminal α-amino group
ligation of peptides through oxidative amide bond forma-
tion.15,16 Using this protocol, acylation of the N-terminal amino
group of six peptides was achieved with their internal lysines
residues remaining intact;14 however, oxidation at cysteine and
methionine residues was found. Such oxidation may alter the
structure of proteins, thus hindering the application of the
protocol in protein modification. Mechanistic studies suggested
that ketenes (RR′CCO) are the key intermediates
accounting for the selective N-terminal modification of peptides
EXPERIMENTAL SECTION
■
Preparation of Ketene 1. Lithium bis(trimethylsilyl)amide (1 M
in hexane, 20 mL) was added dropwise to a solution of phenylacetic
acid methyl ester (2.73 g, 18.2 mmol) in dried tetrahydrofuran (40
mL) at −78 °C. After 1 h, the reaction mixture was warmed to 0 °C,
and 6-iodo-hex-1-yne (4.16 g, 20 mmol) in dried tetrahydrofuran (5
mL) was added dropwise to the solution. After stirring at 0 °C for 1.5
h, the reaction mixture was quenched with water, washed with a
saturated ammonium chloride solution, and extracted with ether twice.
The organic layers were combined and dried over anhydrous
magnesium sulfate to give 4.17 g (76%) of intermediate A. Colorless
oil, analytical TLC (silica gel 60) (10% EtOAc in n-hexane), Rf = 0.33.
1H NMR (300 MHz, CDCl3): δ 7.34−7.25 (m, 5H), 3.65 (s, 3H),
3.54 (t, J = 7.5 Hz, 1H), 2.16 (dt, J = 7.0, 2.5 Hz, 2H), 2.10−2.07 (m,
1H), 1.92 (t, J = 2.5 Hz, 1H), 1.80−1.75 (m, 1H), 1.57−1.49 (m, 2H),
1.42−1.32 (m, 2H). 13C NMR (100.62 MHz, CDCl3): δ 174.7, 139.3,
128.9, 128.2, 127.5, 84.5, 68.6, 52.2, 51.7, 33.2, 28.4, 26.9, 18.5. IR
(KBr, neat, cm−1): 2117, 1736 cm−1. EIMS: m/z 230 (M+). HRMS
(EI) for C15H18O2: calcd 230.1307, found 230.1304. A solution of
intermediate A (4.17 g, 18.1 mmol) in methanol (100 mL) and H2O
(2 mL) was treated with potassium hydroxide pellets (1.5 g, 27 mmol)
and refluxed overnight. The reaction mixture was evaporated to
dryness using a rotary evaporator. Water was added to the reaction
mixture, and the reaction mixture was washed with diethyl ether. The
aqueous layer was collected and acidified with hydrochloric acid and
then extracted with ether twice. The organic layers were combined,
dried over anhydrous magnesium sulfate, and evaporated to give
intermediate B (3.32 g, 87%) as a colorless oil: analytical TLC (silica
by the “[Mn(2,6-Cl2TPP)Cl]/alkyne/H2O2” protocol.14
A
ketene in situ generated from the photolysis of diazo
compound PhC(O)CHN2 could selectively acylate the N-
terminal amino group of three peptides.14 These findings
prompted us to examine the possibility of selective N-terminal
amino group acylation of peptides and proteins using an
isolated ketene.
Ketenes have long been employed as versatile synthetic
intermediates in organic synthesis.17,18 However, reports on
their application in protein modification are sparse.
H2CCO, generated by thermal decomposition of acetone
vapor, was applied for extensive acetylation of amino groups of
pepsin, insulin, and ovalbumin.19 Nonspecific labeling of
proteins at lysine or histidine residue(s) using ketenes
generated in situ from photochemical decomposition of diazo
compounds was also reported.20 The nonspecific labeling might
be due to the generation of reactive carbene intermediates in
the photochemical reactions. A small amount of specific
labeling of lysozyme upon photolysis of diazo N-acetylglucos-
amine (NAG) was observed, but the precise site of
modification was not well characterized.20a To the best of our
knowledge, highly selective N-terminal modification of proteins
using ketene compounds has not been reported.
1
gel 60) (20% EtOAc in n-hexane), Rf = 0.25. H NMR (300 MHz,
CDCl3): δ 7.31−7.24 (m, 5H), 3.54 (t, J = 7.5 Hz, 1H), 2.14 (dt, J =
7.0, 2.5 Hz, 2H), 2.09−2.04 (m, 1H), 1.90 (t, J = 2.5 Hz, 1H), 1.83−
1.74 (m, 1H), 1.56−1.48 (m, 2H), 1.43−1.33 (m, 2H). 13C NMR
(75.48 MHz, CDCl3): δ 180.3, 138.3, 128.7, 128.0, 127.5, 84.2, 68.4,
51.4, 32.4, 28.1, 26.5, 18.1. IR (KBr, neat, cm−1): 2117, 1703 cm−1.
EIMS: m/z 216 (M+). HRMS (EI) for C14H16O2: calcd 216.1150,
found 216.1148. Oxalyl chloride (2 M in dichloromethane, 5 mL) was
added to a solution of intermediate B (1.08 g, 5 mmol) in dried
dichloromethane (5 mL) at rt, and the reaction mixture was stirred for
2 h. The solvent was distilled under nitrogen atmosphere to give a light
yellow oil. The light yellow oil was dissolved in dried tetrahydrofuran
(10 mL), and dried triethylamine (6 mL, 20 mmol) was added
dropwise to the solution at 0 °C. The resulting mixture was stirred at 0
°C for 2 h. The salt formed was filtered under nitrogen atmosphere,
and the filtrate was distilled at 110 °C (1 mmHg) to give 1 (0.26 g,
35%) as a bright yellow oil. 1H NMR (400 MHz, CDCl3): δ 7.33−7.25
(m 3H), 7.07−7.00 (m, 2H), 2.41 (t, J = 7.0 Hz, 2H), 2.22 (dt, J = 7.0,
2.5 Hz, 2H), 1.94 (t, J = 2.5 Hz, 1H), 1.72−1.60 (m, 4H). 13C NMR
(100.62 MHz, CDCl3): δ 204.6, 132.5, 129.0, 124.3, 124.1, 84.0, 68.6,
28.0, 27.1, 23.1, 18.1. IR (KBr, neat, cm−1): 2116, 2096 cm−1. ESI-MS
for C14H14O: calcd 198.1045, found 198.1264.
In the present paper, an alkyne-functionalized ketene (1) was
synthesized and used for selective N-terminal modification of
peptides and proteins (Scheme 1). The proteins modified
Scheme 1
Modification of Peptides XSKFR Using 1. In a 1.0 mL
eppendorf tube, XSKFR solution in H2O (1 μmol/mL, 10 μL), 1
(10 equiv, 10 μL of a 10 μmol/mL stock solution of 1 in dried THF),
and phosphate buffer (pH 6.3, 80 μL) were mixed. The reaction
mixture was kept at rt for 15 min. The conversions of the 20 XSKFR
peptides were determined from TIC (total ion count) of LC−MS/MS
(liquid chromatography−tandem mass spectrometry) analysis of the
reaction mixtures. Using EIC (extracted ion chromatogram) analysis,
we determined the N-terminal selectivity.
Scale-up Modification of Peptide YTSSSKNVVR Using 1. A
solution of peptide YTSSSKNVVR (4 mg) and 1 (8 mg, 10 equiv) in
pH 6.3 phosphate buffer (40 mL) was prepared and placed into a 50
mL centrifugal tube. The reaction mixture was kept at rt overnight.
include insulin, lysozyme, RNaseA, and a therapeutic protein,
BCArg. By utilizing the alkyne group in 1, the 1-incorporated
proteins were further functionalized in situ via a known type of
2590
dx.doi.org/10.1021/ja208009r | J. Am. Chem.Soc. 2012, 134, 2589−2598

Journal of the American Chemical Society
Article
The aqueous phase was freeze-dried, and the residue was purified by
preparative reverse-phase HPLC equipped with a C18 column using
CH3CN/H2O/TFA as the solvent system. The 1-modified
YTSSSKNVVR was isolated in 40% yield (3.2 mg). The sequence of
the 1-modified peptide was confirmed by LC−MS/MS analyses.
Modification of Insulin Using 1. In a 1.0 mL eppendorf tube,
insulin solution in H2O (1 μmol/mL, 10 μL), 1 (10 equiv, 10 μL of a
10 μmol/mL stock solution of 1 in dried THF), and phosphate buffer
(pH 6.3, 80 μL) were mixed at rt for 15 min. The conversion was
determined from TIC of LC−MS/MS analysis of the reaction mixture.
To reduce the disulfide bond between chain A and chain B,
dithiothreitol (10 equiv, 1 μL of a 10 μmol/100 μL stock solution
of dithiothreitol in H2O) was added to the 1-modified insulin solution
(100 μL). The reaction mixture was incubated at 56 °C for 3 h. The
N-terminal modification of chain B of insulin by 1 was confirmed by
LC−MS/MS analysis.
modifying the N-terminus of a library of 20 unprotected
peptides, XSKFR [Scheme 2 and Table 1, X = residues of Asp
Scheme 2
Modification of Lysozyme Using 1. In a 1.0 mL eppendorf tube,
lysozyme solution in H2O (1 μmol/mL, 10 μL) was mixed with
phosphate buffer (pH 9.2, 84 μL) at 37 °C. Ketene 1 (6 equiv, 6 μL of
a 10 μmol/mL stock solution of 1 in dried THF) was added in three
equal portions at 15 min interval. The mixture was kept at 37 °C
overnight. The conversion was determined from TIC of LC−MS/MS
analysis of the reaction mixture.
Trypsin Digestion of 1-Modified Lysozyme. In a 1.0 mL
eppendorf tube, 1-modified lysozyme mixture (100 μL) was mixed
with 1:1 butanol/H2O (100 μL) at 65 °C for 15 min. The reaction
mixture was diluted 4-fold by ammonium bicarbonate solution (50
mM, 600 μL). Trypsin solution [1 mg/mL, 2.86 μL; ratio of trypsin to
lysozyme = 1:50 (w/w)] was added to the eppendorf tube at 0 °C.
The reaction mixture was incubated at 37 °C overnight, and the
trypsin digested mixture was analyzed using LC−MS/MS analysis.
Modification of RNaseA Using 1. This was performed using
RNaseA and 1 (15 equiv) by a procedure similar to that for the
modification of lysozyme (see the Supporting Information).
Trypsin Digestion of 1-Modified RNaseA. This was performed
using 1-modified RNaseA mixture and trypsin solution (1 mg/mL,
2.74 μL) by a procedure similar to that for the trypsin digestion of 1-
modified lysozyme (see the Supporting Information).
Modification of BCArg Using 1. In a 1.0 mL eppendorf tube, 1
(500 equiv, 13 μL of a 100 μmol/10 μL stock solution of 1 in dried
THF) was added to BCArg solution in H2O (0.5 μmol/mL, 500 μL).
The reaction mixture was kept at 37 °C overnight. The modified
protein was subjected to ESI−MS analysis.
Trypsin Digestion of 1-Modified BCArg. In a 1.0 mL eppendorf
tube, 1-modified protein mixture was mixed with trypsin [BCArg/
trypsin 1:20 (w/w)] and incubated at 56 °C. After incubation for 2 h,
trypsin was added to the reaction mixture for replenishment, and the
reaction mixture was incubated at 56 °C for 1 h. The insoluble protein
fraction was collected by centrifugation at 10 000 × g for 3 min. The
pellet was solubilized in 0.1% aqueous TFA with 30% CH3CN. The
reaction mixture was concentrated using C18 ZipTip and applied to
MALDI−TOF (matrix-assisted laser desorption/ionization−time-of-
flight) MS/MS analysis.
(D), Glu (E), Gly (G), Ala (A), Pro (P), Ser (S), Thr (T), Tyr
(Y), Trp (W), Met (M), Gln (Q), Cys (C), Lys (K), His (H),
Asn (N), Val (V), Phe (F), Leu (L), Ile (I), or Arg (R)].
Previous studies14 demonstrated modification of N-terminal
residues glycine, serine, threonine, tyrosine, and histidine of six
other peptides using this protocol. The modification of XSKFR
(100 μM) was conducted with the alkyne phenylacetylene (2
equiv, Table 1; 10 equiv, Table S1 in the Supporting
Information). The conversions are in the range of 8−100%
(Table 1), as revealed by LC-MS/MS analysis. Mono-modified
peptides (Scheme 2, R = Ph, R′ = H) were obtained for all the
20 peptides (for KSKFR and RSKFR, 3% and 2%, respectively,
of di-modified products depicted in Scheme 2 were also found).
Excellent N-terminal selectivity (>99:1, determined by LC−
MS/MS analysis) was achieved for 14 of the 20 peptides
XSKFR23 (Table 1, entries 1−14), and moderate-to-high N-
terminal selectivities (3:1 to 15:1) were obtained for the others
(Table 1, entries 15−20). Oxidation at residues tryptophan,
methionine, and cysteine was observed (Table 1, entries 9, 10,
12).
Synthesis of an Alkyne-Functionalized Ketene 1 and
NHS Ester 2. In view of the results described above, we
envisaged that isolated ketenes should be promising reagents
for N-terminal modification of peptides and proteins. Ketene 1
was synthesized. The synthetic route (Scheme 3) includes
alkylation of phenylacetic acid methyl ester with 6-iodo-hex-1-
yne to give alkyne-functionalized ester A followed by its
hydrolysis to afford carboxylic acid B. Reaction of B with oxalyl
chloride followed by treatment with triethylamine afforded 1,
which could be isolated in reasonably pure form in 35% yield by
Azide−Alkyne Cycloaddition of 1-Modified Proteins (In-
sulin, Lysozyme, or RNaseA). In a 1.0 mL eppendorf, the 1-
modified protein mixture (10 μL), a solution of N-(3-azidopropyl)-5-
(dimethylamino)-1-naphthalenesulfonamide (3)22 in DMSO (10
μmol/mL, 1 μL), a solution of tris(2-carboxyethyl)phosphine
hydrochloride (TCEP) in H2O (10 μmol/mL, 10 μL), a solution of
tris(benzyltriazolymethyl)amine (TBTA) in 1:4 DMSO-t-butanol
(13.5 μmol/mL, 10 μL), a CuSO4 solution in H2O (15.7 μmol/mL,
10 μL), and PBS buffer (pH 8.4, 59 μL) were mixed. The reaction
mixture was kept at rt for 2 h. The dansyl-tagged proteins were
characterized by LC−MS analysis.
RESULTS
■
Before starting to synthesize ketene 1, we examined the
applicability of the “[Mn(2,6-Cl2TPP)Cl]/alkyne/H2O2” pro-
tocol, which would generate ketene RR′CCO in situ, for
2591
dx.doi.org/10.1021/ja208009r | J. Am. Chem.Soc. 2012, 134, 2589−2598

Journal of the American Chemical Society
Article
vacuum distillation (110 °C, 1 mmHg). The isolated ketene 1
can be stored at −20 °C for 1 week without significant
decomposition. For comparing the N-terminal selectivity of 1
with that of a common type of reagent used for amino group
modification of peptides and proteins, NHS ester 2 was
synthesized in 78% yield by the coupling reaction of B with N-
hydroxysuccinimide (Scheme 3).
Table 1. Modification of Peptides XSKFR Using Ketene 1
Compared with That Using NHS Ester 2 and the “[Mn(2,6-
a
Cl2TPP)Cl]/Phenylacetylene/H2O2” Method (I)
N-terminal selectivity of
b
c
conversion (%)
mono-modified XSKFR
peptide
sequence
entry
I
1
2
I
1
2
Modification of Peptides XSKFR Using Ketene 1. The
N-terminal selectivity of 1 was examined using the foregoing
library of 20 peptides XSKFR. Modification of XSKFR (100
μM) with 1 (1 mM; 10 equiv) was conducted in 100 μL of pH
6.3 phosphate buffer/THF (9:1) for 15 min at room
temperature. The results are listed in Table 1.
As depicted in Table 1, the conversions range from 9%
(ASKFR) to 94% (QSKFR). Mono-modified peptides (Scheme
2, R = Ph, R′ = (CH2)4CCH) were obtained for all 20
peptides, of which KSKFR and HSKFR also gave 8% and 19%,
respectively, of di-modified peptides (Scheme 2).
Peptides with N-terminal Asp, Glu, Gly, Ala, Pro, Ser, Thr,
Tyr, Trp, Met, Gln, Asn, or Val gave excellent N-terminal
selectivity (>99:1; Table 1, entries 1−11, 15, 16). Moderate-to-
high N-terminal selectivities (4:1 to 48:1) were obtained for the
N-terminal Cys, Lys, Phe, Leu, Ile, and Arg peptides (entries
12, 13, 17−20). However, a low N-terminal selectivity of 1:1
was observed for HSKFR having N-terminal histidine residue
(Table 1, entry 14).
We have also studied the effect of the amount of 1 on the
modification of peptide LSKL (Table S4 in the Supporting
Information). With 10 equiv of 1, the ratio of unreacted LSKL
to modified LSKL was 48:52 with 28:1 N-terminal selectivity,
and no di-modified LSKL was observed. Upon increase of the
amount of 1 from 10 to 50 equiv, the ratio of LSKL to mono-
modified LSKL to di-modified LSKL was found to be 4:87:9.
Despite the formation of di-modified LSKL, the N-terminal
selectivity for mono-modified LSKL increased to 68:1. In
addition, N-terminal modification of LSKL in buffers of
different pH was also conducted. The N-terminal selectivity
was 15:1 at pH 7.4 and 14:1 at pH 9.2 (Table S4 in the
Supporting Information).
Modification of Peptides XSKFR Using NHS Ester 2.
For comparison, the library of 20 peptides XSKFR were
modified using NHS ester 2 (Scheme 2 and Scheme S1 in the
Supporting Information) under the same conditions as those
for the modification with ketene 1.
Poor N-terminal selectivity was obtained for most of the
peptides when the NHS ester 2 was used (Table 1). For the N-
terminal alanine, proline, threonine, tyrosine, and valine
peptides, the product detected was the lysine-modified peptide,
instead of the corresponding N-terminally modified peptide
(Table 1, entries 4, 5, 7, 8, 16). Unexpectedly, the N-terminal
selectivity for N-terminal tryptophan and methionine peptides
is exceptionally high (>99:1 and 30:1, respectively; Table 1,
entries 9, 10).
d
1
DSKFR
ESKFR
GSKFR
ASKFR
PSKFR
SSKFR
TSKFR
YSKFR
WSKFR
MSKFR
QSKFR
CSKFR
KSKFR
HSKFR
NSKFR
VSKFR
FSKFR
LSKFR
ISKFR
43
33
62
33
9
78
93
81
84
64
81
72
73
21
58
>99:1
>99:1
>99:1
>99:1
>99:1
>99:1
>99:1
>99:1
>99:1
>99:1
>99:1
>99:1
>99:1
>99:1
4:1
>99:1
>99:1
>99:1
>99:1
>99:1
>99:1
>99:1
>99:1
>99:1
>99:1
>99:1
8:1
1:1
2:1
2
100
100
14
3
5:1
4
<1:99
<1:99
1:1
5
39
44
44
37
47
19
29
94
10
48
52
17
23
29
23
37
28
6
24
7
23
<1:99
<1:99
>99:1
30:1
4:1
8
100
e
9
17
f
10
11
12
13
14
15
16
17
18
19
20
8
13
100
25
g
74
1:1
h
i
j
34
55
6:1
1:1
k
l
35
19
19
15
25
37
79
1:1
1:4
48
72
41
71
67
>99:1
>99:1
29:1
2:1
13:1
<1:99
1:1
15:1
8:1
29:1
2:1
5:1
48:1
2:1
m
n
RSKFR
11
19
3:1
4:1
1:1.5
a
Conditions: XSKFR (0.01 μmol) and 1 or 2 (0.1 μmol) in phosphate
buffer of pH 6.3 (100 μL), 15 min, rt, or XSKFR (0.01 μmol),
phenylacetylene (0.02 μmol, 2 equiv), [Mn(2,6-Cl2TPP)Cl] (2 nmol),
H2O2 (1.18 μmol), NaHCO3 (0.1 μmol) in CH3CN−H2O (3:2 v/v,
b
100 μL), 6 h, rt. Determined by total ion count (TIC) of LC−MS
c
analysis. Ratio of N-terminal α-amino group modified peptide to
lysine ε-amino group modified peptide; this ratio was determined by
extracted ion chromatogram (EIC) of LC−MS analysis. Di-modified
DSKFR accounted for 10%. The indole group of tryptophan was
oxidized and presumably gave 1,3-dihydro-indol-2-one. Methionine
was oxidized to its corresponding sulfoxide. Cysteine was oxidized to
give disulfide-bonded peptide. Di-modified KSKFR accounted for
3%. Di-modified KSKFR accounted for 8%. Di-modified KSKFR
accounted for 4%. Di-modified HSKFR accounted for 19%. Di-
modified HSKFR accounted for 8%. Di-modified RSKFR accounted
for 2%. Di-modified RSKFR accounted for 7%.
d
e
f
g
h
i
j
k
l
m
n
Scheme 3
Modification of Cysteine-Containing Peptide
STSSSCNLSK Using Ketene 1 or NHS Ester 2. In previous
work,14 exclusive N-terminal ligation of a cysteine-containing
peptide STSSSCNLSK (N-terminal serine), without cysteine
modification, was achieved using the “[Mn(2,6-Cl2TPP)Cl]/
alkyne/H2O2” protocol. This is attributed to the oxidation of
the cysteine thiol group to disulfide under oxidizing conditions.
In the present work, the ligation of STSSSCNLSK (100 μM)
using 1 (1 mM, 10 equiv) in 100 μL of phosphate buffer/THF
(9:1) at pH 6.3 for 1 h at room temperature afforded mono-
2592
dx.doi.org/10.1021/ja208009r | J. Am. Chem.Soc. 2012, 134, 2589−2598

Journal of the American Chemical Society
Article
modified STSSSCNLSK. The ratio of unmodified
STSSSCNLSK to mono-modified STSSSCNLSK was 80:20.
The ratio of N-terminally modified STSSSCNLSK to cysteine-
modified STSSSCNLSK was about 1:1, and no lysine-modified
peptide was detected. The STSSSCNLSK with thioester linkage
(10 μM) resulting from cysteine modification can be
hydrolyzed by hydroxylamine (500 μM, 50 equiv) in 200 μL
of phosphate buffer of pH 9.2 at 37 °C overnight to afford
disulfide-bonded N-terminally modified STSSSCNLSK under
alkaline reaction conditions.
Scheme 5
In contrast, no N-terminal modification was observed in the
modification of STSSSCNLSK using NHS ester 2 under the
same conditions as those using ketene 1. The modification
using 2 gave a 54:46 ratio of unmodified STSSSCNLSK to
mono-modified STSSSCNLSK, the latter being a cysteine-
modified product.
Scale-Up Modification of Peptide YTSSSKNVVR Using
Ketene 1. Modification of peptides with 1 could be scaled up.
An 8 mg scale one-pot modification of peptide YTSSSKNVVR
with 1 gave 3.2 mg of the N-terminally modified
YTSSSKNVVR (40% isolated yield, Scheme 4) after
purification by preparative reversed-phase HPLC.
Scheme 4
temperature for 15 min. Product analysis by LC−MS (Figure
1a) revealed a peak at 6005 Da, which was assigned to mono-
modified insulin (4, Scheme 5) on the basis of the mass shift
from that (5807 Da) of unmodified insulin (unmodified
insulin/mono-modified insulin = 2:1). Only a trace amount of
di-modified insulin was detected in the reaction mixture. LC−
MS/MS analysis of the reaction mixture, after treatment with
dithiothreitol to reduce the internal disulfide bonds between
chain A and chain B of the 1-modified insulin, revealed that the
modification of insulin by 1 selectively occurred at the N-
terminal phenylalanine of chain B. For comparison, mod-
ification of insulin using NHS ester 2 was conducted in
phosphate buffer of pH 6.3. Although 2 also resulted in the
formation of mono-modified insulin as the major product, a
significant amount of di-modified and trace amount of tri-
modified insulin were found by LC−MS analysis (see Table
S13 in the Supporting Information).
For lysozyme and RNaseA, modification was performed in
100 μL of phosphate buffer of pH 9.2 at 37 °C overnight. The
amount of 1 used is 6 equiv for lysozyme (100 μM) and 15
equiv for RNaseA (100 μM); in both cases, 1 was added in
three equal portions at 15 min intervals. LC−MS analysis of the
reaction mixtures showed peaks at 14502 Da (Figure 1b) and
13879 Da (Figure 1c), which were assigned to the mono-
modified lysozyme (5, Scheme 5) and mono-modified RNaseA
(6, Scheme 5), respectively (unmodified lysozyme/mono-
modified lysozyme = 62:38; unmodified RNaseA/mono-
modified RNaseA = 77:23). A trace amount of di-modified
product was also detected. Upon trypsin digestion, the
modification by 1 was found to selectively occur at the N-
terminus, as revealed by LC−MS/MS analysis (see, for
Modification of Insulin, Lysozyme, RNaseA, and
BCArg Using Ketene 1. To demonstrate the applicability of
ketene 1 in protein modification, we examined the reactivity of
1 toward the following proteins (for their sequences, see Chart
S1 in the Supporting Information): insulin (carrying N-terminal
glycine on chain A and N-terminal phenylalanine and one
internal lysine on chain B), lysozyme (carrying N-terminal
lysine and five internal lysines), RNaseA (carrying N-terminal
lysine and nine internal lysines), and BCArg (carrying N-
terminal methionine and 14 internal lysines). Strikingly, these
proteins were modified by 1 selectively at their N-terminus
(Scheme 5).
The modification of insulin (100 μM) using 1 (10 equiv) was
conducted in 100 μL of phosphate buffer of pH 6.3 at room
2593
dx.doi.org/10.1021/ja208009r | J. Am. Chem.Soc. 2012, 134, 2589−2598

Journal of the American Chemical Society
Article
found at the N-terminal methionine by MALDI−TOF MS/MS
analysis (Figure 2b).
Without further purification, the crude 1-modified proteins
(insulin, lysozyme, and RNaseA, 10 μM) containing an alkyne
moiety were further modified with a dansyl azide 322 (100 μM
in DMSO) in the presence of the triazolyl ligand TBTA (100
μM in 1:4 DMSO/t-butanol), tris-carboxyethylphosphine
(TCEP) (1 mM), and CuSO4 (1 mM) in 100 μL of PBS
buffer of pH 8.4 (Scheme 6) in a one-pot reaction (see the
Experimental Section). The resulting 3-modified proteins (7−
9, Scheme 6) were characterized by LC−MS analysis.
DISCUSSION
■
Peptide Modification by Ketene 1. Bioconjugation
reaction using ketene conducted in aqueous media has to
face competitive hydration of ketene, which affords carboxylic
acid as a side product.17 Since the hydration rate of ketene is
slower than the amination rate,17 amination of ketenes in the
presence of water has previously been reported in the literature
and used for amide bond formation in organic synthesis.17
To synthesize a relatively stable ketene with functional
diversity for modification of peptides and proteins, we designed
the alkyne-linked ketene 1. The key features of 1 include the
following: (1) a terminal aliphatic alkyne group is installed for
further modification with different biophysical probes through
bio-orthogonal click chemistry;6k−n (2) an aryl ring is installed
for enhancing the stability of ketene for vacuum distillation. To
the best of our knowledge, 1 is the first isolated alkyne-
functionalized ketene.25
The library of 20 unprotected peptides XSKFR was chosen
mainly for examining the generality of the N-terminal
modification of peptides by ketenes. The serine and lysine
residues in the sequence of XSKFR are for studying the chemo-
and site-selectivity in the presence of both −OH group (from
serine) and amino groups (from lysine and the N-terminus)
(see Scheme 2). The aromatic phenylalanine residue was
incorporated for assisting HPLC analysis, and the C-terminal
residue, arginine, was incorporated for obtaining better signal in
MS analysis.
Amination of almost all of the peptides XSKFR by 1
preferentially occurred at the N-terminal α-amino group
despite the presence of nucleophilic −OH group and more
basic lysine ε-amino groups. At pH 6.3, excellent N-terminal
selectivity (>99:1) was obtained in 13 out of the 20 XSKFR
peptides, and moderate (4:1) to high (48:1) N-terminal
selectivity was obtained in the modification of the remaining 6
XSKFR peptides (Table 1). No oxidation of amino acid
residues was observed, in contrast to the oxidation at the
cysteine, methionine, and tryptophan side chains by the
“[Mn(2,6-Cl2TPP)Cl]/alkyne/H2O2” protocol (Table 1).
Except for the oxidation problem, the “[Mn(2,6-Cl2TPP)Cl]/
alkyne/H2O2” protocol, which would generate ketenes in situ,
resulted in moderate-to-excellent N-terminal selectivity for the
20 peptides XSKFR (Table 1).
No −OH modification on serine residue was observed for
the modification of peptides XSKFR using 1 or the “[Mn(2,6-
Cl2TPP)Cl]/alkyne/H2O2” protocol (Table 1). The absence of
−OH modification might be due to the lower nucleophilicity of
the −OH group than that of the α- or ε-amino groups of the N-
terminus and lysine residue.
Figure 1. Mass reconstruction spectrum in LC−MS analysis of the
reaction mixture in the modification of (a) insulin, (b) lysozyme, and
(c) RNaseA using 1.
example, Figure 2a). At pH 9.2, modification of lysozyme or
RNaseA using 10 equiv of ketene 1 at room temperature for 15
min produced mono-modified product only (see Tables S14
and S16 in the Supporting Information), while that using NHS
ester 2 mainly gave tri- or tetra-modified product (see Tables
S15 and S17 in the Supporting Information).
BCArg, carrying 14 internal lysine residues, is a therapeutic
protein for cancer treatment.24 The coupling reaction between
BCArg (2.6 mM) and 1 (500 equiv) was performed in 100 μL
of H2O at 37 °C overnight. In MALDI−TOF MS analysis of
the reaction mixture, the peaks at 33256 and 33455 Da are
assigned to unmodified BCArg and mono-modified BCArg,
respectively. By trypsin digestion, the site of modification was
In the modification of cysteine-containing peptide
STSSSNCLSK using 1, both cysteine-modified and N-
terminally modified peptides were obtained. This is presumably
2594
dx.doi.org/10.1021/ja208009r | J. Am. Chem.Soc. 2012, 134, 2589−2598

Journal of the American Chemical Society
Article
Figure 2. The MS/MS spectrum of the N-terminal 1-modified fragment of (a) lysozyme and (b) BCArg.
due to the higher nucleophilicity of the cysteine thiol group,
which results in the formation of thioester linkage. The
thioester linkage can be reduced back to a free cysteine thiol
group by treatment with hydroxylamine under basic conditions.
Comparison of Peptide Modification by Ketenes with
Literature Methods. The high N-terminal selectivity of
ketenes in bioconjugation can be revealed by comparison with
the result of modification of peptides XSKFR using NHS ester
2. As depicted in Table 1, 2 gave moderate-to-low N-terminal
selectivity (i.e., from 5:1 to <1:99) in modifying 18 out of the
20 peptides XSKFR. The observed modification of
STSSSCNLSK by 2 at the cysteine residue but not at the N-
terminus further indicates that ketene 1 is more selective
toward N-terminal α-amino group than NHS ester 2.
Several methods have been reported for selective N-terminal
modification of peptides or proteins. Apart from the trans-
amination methods described in the Introduction section,7−9
other methods each suitable for modifying one or two types of
N-terminal residues have been developed, such as modification
of serine or threonine residues by periodate oxidation,4a,9b
modification of tryptophan residues by Pictet−Spengler
reaction,4d and modification of cysteine residues by native
2595
dx.doi.org/10.1021/ja208009r | J. Am. Chem.Soc. 2012, 134, 2589−2598

Journal of the American Chemical Society
Article
Modification of proteins RNaseA and lysozyme were
conducted in phosphate buffer at pH 9.2 and at 37 °C in
order to have higher conversions. Although most of the N-
terminal α-amino groups should have been protonated under
basic conditions, excellent N-terminal selectivity was still
observed in the modification of RNaseA and lysozyme using
ketene 1. No internal-lysine-modified protein was observed in
either case. The modification of RNaseA and lysozyme using
NHS ester 2 gave multiple site-modified proteins (up to +7
modified RNaseA was formed using NHS ester 2, as depicted in
Table S17 in the Supporting Information). These results
further reveal the high selectivity of ketenes toward the N-
terminal α-amino group.
Scheme 6
The N-terminal selectivity in the modification of insulin (N-
terminal phenylalanine), lysozyme (N-terminal lysine), and
RNaseA (N-terminal lysine) using ketene 1 is higher than that
for similar reactions with peptides carrying the same N-terminal
amino acid residues. The peptides XSKFR with N-terminal
lysine and phenylalanine residues gave N-terminal selectivity of
6:1 and 29:1, respectively (Table 1, entries 13 and 17). One
possibility is that the tertiary structure of proteins leads to
microenvironment that reduces the accessibility of ketene to
the internal lysine residues, thus resulting in an increase of N-
terminal selectivity.
The alkyne functional group on ketene 1 allows further
modification of the 1-modified proteins under in vitro or in vivo
conditions through bio-orthogonal reactions in click chemis-
try,6k−n such as a copper-catalyzed [3 + 2] cycloaddition of
azide with alkyne.21 In this work, the copper-catalyzed [3 + 2]
cycloaddition reaction of dansyl azide 3 with 1-modified
proteins was achieved without prior purification of the 1-
modified proteins.
BCArg is a protein that displays anticancer activity. The
excellent N-terminal selectivity obtained in the modification of
BCArg by ketene 1 demonstrates the applicability of alkyne-
linked ketene in the modification of pharmaceutical proteins.28
Therapeutic proteins are of increasing global demand.
However, therapeutic proteins suffer from poor stability, low
solubility, and short circulating half-life. PEGylation is the most
widely used method to improve the pharmacological profiles of
protein drugs.29 The alkyne-incorporated BCArg would allow
further ligation with PEG moiety through click chemistry. The
high N-terminal selectivity accomplished by 1 would allow the
bioconjugation of pharmaceutical proteins to occur at a precise
location resulting in homogeneous pharmaceutical proteins
with higher consistency in clinical performance.
There is a growing need to develop methods for modifying
proteins at two distinct sites for fluorescence resonance energy
transfer (FRET) studies.30 The high selectivity of ketenes to
the N-terminus of proteins opens a new way for site-selective
incorporation of one functional group into proteins. The
bioconjugation reaction conducted using ketenes is expected to
be compatible with most of the literature-reported bioconju-
gation reactions.31 Thus, the second functional group could be
installed on proteins through modification of cysteine, tyrosine,
or tryptophan without the need for prior treatment of the
reaction mixture of the ketene-modified proteins.32
chemical ligation.4b,c,e,f Transamination with subsequent
Pictet−Spengler reaction can selectively modify N-terminal
glycine residues.26 The N-terminal modification using ketene 1
not only exhibits impressive generality but also directly
incorporates a functional molecule into the N-terminal α-
amino group through a relatively simple acylation reaction,
without preferentially modifying the side chain(s) of the N-
terminus.
Protein Modification by Ketene 1. In the modification of
peptide LSKL using 1, the conversion of the modification
reaction increased with increasing amount of 1 used (from 10
to 50 equiv). Also, increasing pH of the reaction mixture
resulted in higher reaction conversion, while the N-terminal
selectivity remained high (14:1 at pH 9.2). On the other hand,
lowering the pH value of the reaction mixture could increase
the N-terminal selectivity but could slightly decrease the
reaction conversion. Increase in reaction conversion may be
compromised by multisite modification; the latter could be
minimized by adding the ketene in portions.
Accordingly, the reaction conversion for the modification of
proteins was raised by choosing the right amount of ketene and
the right pH value of the reaction mixture while attaining
excellent N-terminal selectivity. Proteins including insulin,
lysozyme, RNaseA, and BCArg were N-terminally modified
using 1, as confirmed by trypsin digestion and LC−MS/MS
analysis.27 Selective N-terminal modification on chain B of
insulin was found in the modification of insulin using 1. In this
modification, the N-terminal glycine of chain A of insulin
remained intact. One rationale for this finding is the higher
CONCLUSION
solvent-accessible area of the N-terminal phenylalanine of chain
■
13b
B, which was calculated to be 31.19 Å2
by GETAREA 1.4.
We have observed highly selective N-terminal modification of
peptides and proteins by an isolated, alkyne-functionalized
ketene 1. This method can directly incorporate a functional
molecule into the N-terminus of peptides XSKFR under mild
The solvent-accessible area of N-terminal glycine of chain A
was found to be 18.31 Å2,13b rendering the N-terminus of chain
A less exposed for modification using 1.
2596
dx.doi.org/10.1021/ja208009r | J. Am. Chem.Soc. 2012, 134, 2589−2598

Journal of the American Chemical Society
Article
47, 10030. (h) Bernardes, G. J. L.; Castagner, B.; Seeberger, P. H. ACS
Chem. Biol. 2009, 4, 703. (i) Kurpiers, T.; Mootz, H. D. Angew. Chem.,
Int. Ed. 2009, 48, 1729. (j) Lin, Y. A.; Chalker, J. M.; Davis, B. G.
ChemBioChem 2009, 10, 959. (k) Sletten, E. M.; Bertozzi, C. R. Angew.
Chem., Int. Ed. 2009, 48, 6974. (l) Afagh, N. A.; Yudin, A. K. Angew.
conditions (pH 6.3, room temperature), with excellent terminal
selectivity for the majority of the 20 natural amino acid residues
(X) and with moderate-to-high terminal selectivity for the
remaining X’s except X = His. Studies on peptides HSKFR and
STSSSCNLSK reveal that the selectivities for N-terminal
histidine and serine are comparable to those for the coexistent
internal lysine and cysteine, respectively. Proteins including
insulin, lysozyme, RNaseA, and therapeutic protein BCArg have
been modified by 1 at their N-terminus under conditions of pH
6.3 (room temperature) or 9.2 (37 °C). Further functionaliza-
tion of the 1-incorporated proteins with a dansyl azide 3 has
been achieved through click chemistry. Ketene 1 can
circumvent the oxidation problem encountered using the
“[Mn(2,6-Cl2TPP)Cl/alkyne/H2O2” protocol14 for peptide
modification. The present work lends credence to practical
application of ketenes as a highly site-selective acylation reagent
for N-terminal modification of peptides and proteins.
Chem., Int. Ed. 2010, 49, 262. (m) Basle,
Chem. Biol. 2010, 17, 213. (n) Canalle, L. A.; Lowik, D. W. P. M.; van
Hest, J. C. M. Chem. Soc. Rev. 2010, 39, 329. (o) Kalia, J.; Raines, R. T.
́
E.; Joubert, N.; Pucheault, M.
̈
Curr. Org. Chem. 2010, 14, 138.
(3) For examples, see: (a) Reference 1a. (b) Reference 1b.
(c) Jentoft, N.; Dearborn, D. G. J. Biol. Chem. 1979, 254, 4359.
(d) Tanaka, K.; Kamatani, M.; Mori, H.; Fujii, S.; Ikeda, K.; Hisada,
M.; Itagaki, Y.; Katsumura, S. Tetrahedron 1999, 55, 1657. (e) Tanaka,
K.; Fujii, Y.; Fukase, K. ChemBioChem 2008, 9, 2392. (f) Griffin, B. A.;
Adams, S. R.; Tsien, R. Y. Science 1998, 281, 269. (g) Adams, S. R.;
Campbell, R. E.; Gross, L. A.; Martin, B. R.; Walkup, G. K.; Yao, Y.;
Llopis, J.; Tsien, R. Y. J. Am. Chem. Soc. 2002, 124, 6063.
(h) Rademann, J. Angew. Chem., Int. Ed. 2004, 43, 4554. (i) Cao,
H.; Xiong, Y.; Wang, T.; Chen, B.; Squier, T. C.; Mayer, M. U. J. Am.
Chem. Soc. 2007, 129, 8672. (j) Kim, Y.; Ho, S. O.; Gassman, N. R.;
Korlann, Y.; Landorf, E. V.; Collart, F. R.; Weiss, S. Bioconjugate Chem.
2008, 19, 786. (k) Chalker, J. M.; Bernardes, G. J. L.; Lin, Y. A.; Davis,
B. G. Chem.Asian J. 2009, 4, 630. (l) Antos, J. M.; Francis, M. B. J.
Am. Chem. Soc. 2004, 126, 10256. (m) Antos, J. M.; McFarland, J. M.;
Iavarone, A. T.; Francis, M. B. J. Am. Chem. Soc. 2009, 131, 6301.
(n) Halo, T. L.; Appelbaum, J.; Hobert, E. M.; Balkin, D. M.;
Schepartz, A. J. Am. Chem. Soc. 2009, 131, 438.
(4) For examples, see: (a) Geoghegan, K. F.; Stroh, J. G. Bioconjugate
Chem. 1992, 3, 138. (b) Dawson, P. E.; Muir, T. W.; Clark-Lewis, I.;
Kent, S. B. H. Science 1994, 266, 776. (c) Dawson, P. E.; Kent, S. B. H.
Annu. Rev. Biochem. 2000, 69, 923. (d) Li, X.; Zhang, L.; Hall, S. E.;
Tam, J. P. Tetrahedron Lett. 2000, 41, 4069. (e) Yeo, D. S. Y.;
Srinivasan, R.; Chen, G. Y. J.; Yao, S. Q. Chem.Eur. J. 2004, 10,
4664. (f) Flavell, R. R.; Muir, T. W. Acc. Chem. Res. 2009, 42, 107.
(5) (a) Zaykov, A. N.; MacKenzie, K. R.; Ball, Z. T. Chem.Eur. J.
2009, 15, 8961. (b) Popp, B. V.; Ball, Z. T. J. Am. Chem. Soc. 2010,
132, 6660. (c) Popp, B. V.; Ball, Z. T. Chem. Sci. 2011, 2, 690.
(6) For examples, see: (a) Cornish, V. W.; Hahn, K. M.; Schultz, P.
G. J. Am. Chem. Soc. 1996, 118, 8150. (b) Dirksen, A.; Hackeng, T. M.;
Dawson, P. E. Angew. Chem., Int. Ed. 2006, 45, 7581. (c) Carrico, I. S.;
Carlson, B. L.; Bertozzi, C. R. Nat. Chem. Biol. 2007, 3, 321. (d) Saxon,
E.; Bertozzi, C. R. Science 2000, 287, 2007. (e) Lemieux, G. A.; de
Graffenried, C. L.; Bertozzi, C. R. J. Am. Chem. Soc. 2003, 125, 4708.
(f) Soellner, M. B.; Nilsson, B. L.; Raines, R. T. J. Am. Chem. Soc. 2006,
128, 8820. (g) Hangauer, M. J.; Bertozzi, C. R. Angew. Chem., Int. Ed.
2008, 47, 2394. (h) Lin, Y. A.; Chalker, J. M.; Davis, B. G.
ChemBioChem 2009, 10, 959. (i) Song, W.; Wang, Y.; Qu, J.; Madden,
M. M.; Lin, Q. Angew. Chem., Int. Ed. 2008, 47, 2832. (j) Song, W.;
Wang, Y.; Qu, J.; Lin, Q. J. Am. Chem. Soc. 2008, 130, 9654. (k) Agard,
N. J.; Baskin, J. M.; Prescher, J. A.; Lo, A.; Bertozzi, C. R. ACS Chem.
Biol. 2006, 1, 644. (l) Baskin, J. M.; Prescher, J. A.; Laughlin, S. T.;
Agard, N. J.; Chang, P. V.; Miller, I. A.; Lo, A.; Codelli, J. A.; Bertozzi,
C. R. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 16793. (m) Codelli, J. A.;
Baskin, J. M.; Agard, N. J.; Bertozzi, C. R. J. Am. Chem. Soc. 2008, 130,
11486. (n) Laughlin, S. T.; Baskin, J. M.; Amacher, S. L.; Bertozzi, C.
R. Science 2008, 320, 664.
ASSOCIATED CONTENT
* Supporting Information
■
S
Experimental section including general, biological mass
spectrometry, and procedures not included in the text, NMR
spectra of 1−3, results of modification of peptide LSKL using 1,
2, and the “[Mn(2,6-Cl2TPP)Cl]/alkyne/H2O2” protocol, and
characterization of modified peptides and proteins (such as
MS/MS spectra). This material is available free of charge via
the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
cmche@hku.hk; bcmkwong@inet.polyu.edu.hk
■
ACKNOWLEDGMENTS
■
We are grateful for the financial support of The University of
Hong Kong (University Development Fund), Hong Kong
Research Grants Council (PolyU 7052/07P), The Hong Kong
Polytechnic University (PolyU Departmental General Research
Funds and Competitive Research Grants for Newly Recruited
Junior Academic Staff), and the Areas of Excellence Scheme
established under the University Grants Committee of the
Hong Kong Special Administrative Region, China (AoE/P-10/
01).
REFERENCES
■
(1) (a) Aslam, M.; Dent, A. Bioconjugation: Protein Coupling
Techniques for the Biomedical Sciences; Macmillan Reference Ltd.:
London, 1998. (b) Lundblad, R. L. Chemical Reagents for Protein
Modification, 3rd ed.; CRC Press: Boca Raton, FL, 2005. (c) Walsh, C.
T. Post-translational Modification of Proteins: Expanding Nature’s
Inventory; Roberts and Company Publishers: Greenwood Village,
Colorado, 2006. (d) Francis, M. B. In Chemical Biology: From Small
Molecules to Systems Biology and Drug Design; Schreiber, S. L., Kapoor,
T. M., Wess, G., Eds.; WILEY-VCH Verlag GmbH & Co. KGaA:
Weinheim, Germany, 2007; Vol. 2, p 593. (e) Hermanson, G. T.
Bioconjugate Techniques, 2nd ed.; Academic Press: San Diego, CA,
2008.
(7) (a) Gilmore, J. M.; Scheck, R. A.; Esser-Kahn, A. P.; Joshi, N. S.;
Francis, M. B. Angew. Chem., Int. Ed. 2006, 45, 5307. (b) Scheck, R. A.;
Francis, M. B. ACS Chem. Biol. 2007, 2, 247. (c) Scheck, R. A.; Dedeo,
M. T.; Lavarone, A. T.; Francis, M. B. J. Am. Chem. Soc. 2008, 130,
11762.
(8) Cennamo, C.; Carafoli, B.; Bonetti, E. P. J. Am. Chem. Soc. 1956,
78, 3523.
(9) (a) Dixon, H. B. F. Biochem. J. 1964, 92, 661. (b) Dixon, H. B. F.
J. Protein Chem. 1984, 3, 99.
(10) Abello, N.; Kerstjens, H. A. M.; Postma, D. S.; Bischoff, R. J.
(2) For recent reviews, see: (a) Prescher, J. A.; Bertozzi, C. R. Nat.
Chem. Biol. 2005, 1, 13. (b) van Swieten, P. F.; Leeuwenburgh, M. A.;
Kessler, B. M.; Overkleeft, H. S. Org. Biomol. Chem. 2005, 3, 20.
(c) Antos, J. M.; Francis, M. B. Curr. Opin. Chem. Biol. 2006, 10, 253.
(d) van Maarseveen, J. H.; Reek, J. N. H.; Back, J. W. Angew. Chem.,
Int. Ed. 2006, 45, 1841. (e) Carrico, I. S. Chem. Soc. Rev. 2008, 37,
1423. (f) Gauthier, M. A.; Klok, H.-A. Chem. Commun. 2008, 2591.
(g) Hackenberger, C. P. R.; Schwarzer, D. Angew. Chem., Int. Ed. 2008,
Proteome Res. 2007, 6, 4770.
2597
dx.doi.org/10.1021/ja208009r | J. Am. Chem.Soc. 2012, 134, 2589−2598

Journal of the American Chemical Society
Article
(11) (a) Gaudriault, G.; Vincent, J.-P. Peptides 1992, 13, 1187.
their existence in undetectable amounts could not be excluded. In view
of the low N-terminal selectivity of 1:1 for 1-mono-modified peptide
HSKFR (which bears histidine) and the occurrence of cysteine
modification in ∼50% of the 1-modified peptide STSSSCNLSK, the
unobserved modification of histidine or cysteine in 1-labeled proteins
could be rationalized as follows: (i) The moderate-to-excellent N-
terminal selectivity for 1-modified peptides XSKFR (X ≠ His) suggests
a low reactivity of internal lysine (relative to the N-terminal X
residues) toward 1. Because the mono-modification of HSKFR with 1
gave a 1:1 mixture of N-terminal histidine and internal lysine modified
HSKFR, we believe that N-terminal histidine is relatively unreactive
toward labeling with ketene 1. On going from N-terminal histidine to
internal histidine, its reactivity toward 1 should be further reduced.
Indeed, upon trypsin digestion of the 1-modified protein BCArg, the
histidine-containing fragments (such as LMHHHHHH) detected by
LC−MS/MS were not found to be labeled by 1. (ii) STSSSCNLSK
contains a reactive free thiol group in the cysteine side chain, but the
protein BCArg is devoid of cysteine, and in the structures of insulin,
lysozyme, and RNaseA, all the cysteine residues are covalently paired
through disulfide bridge. In previous work,14 we observed exclusive N-
terminal modification of STSSSCNLSK, without cysteine modifica-
tion, by in situ generated ketene using the “[Mn(2,6-Cl2TPP)Cl]/
alkyne/H2O2” protocol, a phenomenon that can be attributed to the
oxidation of the cysteine thiol group to disulfide under oxidizing
conditions, as mentioned above in the Results section.
(b) Sel
́ ́ ́
o, I.; Negroni, L.; Creminon, C.; Grassi, J.; Wal, J. M. J.
Immunol. Methods 1996, 199, 127.
(12) For peptide YGGFL, which bears no internal lysine residue,
selective modification at the N-terminal α-amino group has been
obtained.11a
(13) (a) Shiu, H.-Y.; Chan, T.-C.; Ho, C.-M.; Liu, Y.; Wong, M.-K.;
Che, C.-M. Chem.Eur. J. 2009, 15, 3839. (b) Ho, C.-M.; Zhang, J.-
L.; Zhou, C.-Y.; Chan, O.-Y.; Yan, J. J.; Zhang, F.-Y.; Huang, J.-S.; Che,
C.-M. J. Am. Chem. Soc. 2010, 132, 1886.
(14) Chan, W.-K.; Ho, C.-M.; Wong, M.-K.; Che, C.-M. J. Am. Chem.
Soc. 2006, 128, 14796.
(15) For a review on amide bond formation, see: Montalbetti, C. A.
G. N.; Falque, V. Tetrahedron 2005, 61, 10827.
(16) For examples of amide bond formation, see: (a) Gunanathan,
C.; Ben-David, Y.; Milstein, D. Science 2007, 317, 790. (b) Shen, B.;
Makley, D. M.; Johnston, J. N. Nature 2010, 465, 1027.
(17) Tidwell, T. T. Ketenes, 2nd ed.; John Wiley & Sons, Inc.:
Hoboken, NJ, 2006.
(18) For reviews, see: (a) Tidwell, T. T. Angew. Chem., Int. Ed. 2005,
44, 5778. (b) Paull, D. H.; Weatherwax, A.; Lectka, T. Tetrahedron
2009, 65, 6771.
(19) (a) Herriott, R. M.; Northrop, J. H. J. Gen. Physiol. 1934, 18, 35.
(b) Stern, K. G.; White, A. J. Biol. Chem. 1938, 122, 371. (c) Neuhaus,
O. W.; Miller, L. Arch. Biochem. Biophys. 1957, 71, 162.
(20) (a) Burkhardt, A. E.; Russo, S. O.; Rinehardt, C. G.; Loudon, G.
M. Biochemistry 1975, 14, 5465. (b) Henderson, B. S.; Larsen, B. S.;
Schwab, J. M. J. Am. Chem. Soc. 1994, 116, 5025. (c) Banzon, J. A.;
Kuo, J. M.; Miles, B. W.; Fischer, D. R.; Stang, P. J.; Raushel, F. M.
Biochemistry 1995, 34, 743. (d) Fleming, S. A. Tetrahedron 1995, 51,
12479. (e) Pedone, E.; Brocchini, S. React. Funct. Polym. 2006, 66, 167.
(f) Givens, R. S.; Yousef, A. L.; Yang, S.; Timberlake, G. T. Photochem.
Photobiol. 2008, 84, 185.
(28) The unreacted ketene is likely to be removed by ultrafiltration
using, for example, commercially available PALL Nanosep centrifugal
devices (which have different molecular cut off such as 3K according
to manufacturing instruction).
(29) Harris, J. M.; Chess, R. B. Nat. Rev. Drug Discovery 2003, 2, 214.
(30) For reviews, see: (a) Truong, K.; Ikura, M. Curr. Opin. Struct.
Biol. 2001, 11, 573. (b) Heyduk, T. Curr. Opin. Biotechnol. 2002, 13,
292. (c) Roy, R.; Hohng, S.; Ha, T. Nat. Methods 2008, 5, 507.
(31) The bioactivity of protein before and after treatment with
ketene 1 was examined using lysozyme as an example by lysozyme
assay purchased from Invitrogen. No significant influence on the
lysozyme activity by labeling with 1 was observed (see Figure S111 in
the Supporting Information). Note that THF (which may cause
protein denaturation) is not a solvent required for protein
modification with 1; it was used in this work for preparing a stock
solution of 1, and the reaction mixture contained only 2.5−15% v/v
THF with water being the predominant solvent.
(21) Rostovtsev, V. V.; Green, L. G.; Fokin, V. V.; Sharpless, K. B.
Angew. Chem., Int. Ed. 2002, 41, 2596.
(22) Compound 3 was prepared according to the literature method:
Díaz, D. D.; Rajagopal, K.; Strable, E.; Schneider, J.; Finn, M. G. J. Am.
Chem. Soc. 2006, 128, 6056.
(23) For KSKFR, the exact ligation site at the N-terminal lysine (α-
amino or ε-amino group) could not be identified using MS/MS alone.
Note that the pKa value of the N-terminal α-amino group is
considerably lower than that of the ε-amino group of lysine (see the
Introduction section).
(24) (a) Cheng, P. N. M.; Leung, Y. C.; Lo, W. H.; Tsui, S. M.; Lam,
K. C. Cancer Lett. 2005, 224, 67. (b) Cheng, P. N.-M.; Lam, T.-L.;
Lam, W.-M.; Tsui, S.-M.; Cheng, A. W.-M.; Lo, W.-H.; Leung, Y.-C.
Cancer Res. 2007, 67, 309. (c) Lam, T. L.; Wong, G. K. Y.; Chong, H.
C.; Cheng, P. N. M.; Choi, S. C.; Chow, T. L.; Kwok, S. Y.; Poon, R.
T. P.; Wheatley, D. N.; Lo, W. H.; Leung, Y. C. Cancer Lett. 2009, 277,
91. (d) Tsui, S.-M.; Lam, W.-M.; Lam, T.-L.; Chong, H.-C.; So, P.-K.;
Kwok, S.-Y.; Arnold, S.; Cheng, P. N.-M.; Wheatley, D. N.; Lo, W.-H.;
Leung, Y.-C. Cancer Cell Int. 2009, 9, 9.
(32) Lundblad, R. L. Techniques in Protein Modification; CRC Press:
Boca Raton, FL, 1995.
(25) For examples of other types of isolated ketenes, see: (a) Baigrie,
L. M.; Seiklay, H. R.; Tidwell, T. T. J. Am. Chem. Soc. 1985, 107, 5391.
(b) Hodous, B. L.; Fu, G. C. J. Am. Chem. Soc. 2002, 124, 1578.
(c) Hodous, B. L.; Fu, G. C. J. Am. Chem. Soc. 2002, 124, 10006.
(d) Wiskur, S. L.; Fu, G. C. J. Am. Chem. Soc. 2005, 127, 6176.
(26) Sasaki, T.; Kodama, K.; Suzuki, H.; Fukuzawa, S.; Tachibana, K.
Bioorg. Med. Chem. Lett. 2008, 18, 4550.
(27) The N-terminal selectivity in protein modification by 1 is
consistent with (i) the mono-modification revealed by Figure 1, (ii)
the presence of readily accessible reactive α-NH2 group at the N-
terminus of proteins (compared with internal amino acid residues),
(iii) the moderate-to-excellent N-terminal selectivity observed in
modification of the library of XSKFR peptides (X ≠ His) with 1
(Table 1), and (iv) the detection of N-terminal modified fragment
(Figure 2). We have made many efforts to look for fragments
corresponding to protein modification at internal amino acid residues,
but these fragments have not been detected, though the possibility of
2598
dx.doi.org/10.1021/ja208009r | J. Am. Chem.Soc. 2012, 134, 2589−2598