Kinetic assay of the michael addition-like thiol-ene reaction and insight into protein bioconjugation

Article abstract of DOI:10.1002/asia.201402095

The chemical modification of proteins is a valuable technique in understanding the functions, interactions, and dynamics of proteins. Reactivity and selectivity are key issues in current chemical modification of proteins. The Michael addition-like thiol-ene reaction is a useful tool that can be used to tag proteins with high selectivity for the solvent-exposed thiol groups of proteins. To obtain insight into the bioconjugation of proteins with this method, a kinetic analysis was performed. New vinyl-substituted pyridine derivatives were designed and synthesized. The reactivity of these vinyl tags with L-cysteine was evaluated by UV absorption and high-resolution NMR spectroscopy. The results show that protonation of pyridine plays a key role in the overall reaction rates. The kinetic parameters were assessed in protein modification. The different reactivities of these vinyl tags with solvent-exposed cysteine is valuable information in the selective labeling of proteins with multiple functional groups. Tag! You're it! The Michael addition-like thiol-ene reaction between a pyridine-substituted vinyl group and the thiol of L-cysteine is evaluated by kinetic assay. Diverse reaction rates between different vinyl tags and free thiols are assessed, and the determined reactivity offers valuable information on protein bioconjugation.

Full text of DOI:10.1002/asia.201402095

FULL PAPER
DOI: 10.1002/asia.201402095
Kinetic Assay of the Michael Addition-Like Thiol–Ene Reaction and Insight
into Protein Bioconjugation
Fei-He Ma, Jia-Liang Chen, Qing-Feng Li, Hui-Hui Zuo, Feng Huang, and
[
a]
Xun-Cheng Su*
Abstract: The chemical modification of
proteins is a valuable technique in un-
derstanding the functions, interactions,
and dynamics of proteins. Reactivity
and selectivity are key issues in current
chemical modification of proteins. The
Michael addition-like thiol–ene reac-
tion is a useful tool that can be used to
tag proteins with high selectivity for
the solvent-exposed thiol groups of
proteins. To obtain insight into the bio-
conjugation of proteins with this
method, a kinetic analysis was per-
formed. New vinyl-substituted pyridine
derivatives were designed and synthe-
sized. The reactivity of these vinyl tags
with l-cysteine was evaluated by UV
absorption and high-resolution NMR
spectroscopy. The results show that
protonation of pyridine plays a key
role in the overall reaction rates. The
kinetic parameters were assessed in
protein modification. The different re-
activities of these vinyl tags with sol-
vent-exposed cysteine is valuable infor-
mation in the selective labeling of pro-
teins with multiple functional groups.
Keywords: conjugation
spectroscopy proteins
chemistry · thiol–ene reaction
·
NMR
·
· thiol
Introduction
side of disulfide bond formation that is widely used in pro-
tein bioconjugations.
The bioconjugation of proteins with small organic probes is
a powerful technique that is used in the elucidation of the
structure, interaction, dynamics, and function of proteins.
Small functional probes can be site-specifically attached to
a protein through a chemical or biochemical method, and
the optical or magnetic properties of these small molecules
provide valuable information of the attached target pro-
To the best of our knowledge, the Michael addition-like
thiol–ene reaction was first reported in the detection of the
free thiol group of proteins by use of acrylonitrile in the
[7]
early 1960s. Reported examples of the formation of
a hetero-Michael thioether bond as applied in protein bio-
conjugation can be distinguished as either radical initiat-
[3,4]
[5–8]
ed
or a general nucleophilic addition reaction.
The
[1,2]
teins.
In recent years, the Michael addition-like thiol–ene
radical-initiated thiol–ene reaction plays a key role in pro-
ducing functional materials and polymers in materials sci-
reaction has been used in the site-specific labeling of pro-
[1c,3–6]
[9]
teins with small molecules (Scheme 1).
The hetero-Mi-
ence. Most recently, a radical-triggered thiol-ene reaction
[3,4]
chael thioether bond formed between the protein and the
tag is resistant to reducing reagents, which avoids the down-
in the modification of proteins was also reported.
Given the stability and activity of proteins, the thiol-ene
reaction without radicals is an ideal choice for the modifica-
tion of proteins. Maleimide derivatives have thus been
widely used in the bioconjugation of proteins, as they readily
[8]
react with the free thiol groups of proteins. However, the
reaction of maleimide with the thiol group generates a new
chiral center, and the formed diastereomers compromise the
fluorescence, electron paramagnetic resonance (EPR), and
Scheme 1. Michael addition-like thiol–ene reaction between the vinyl tag
and l-cysteine.
[6a]
NMR spectroscopy experimental data. The selectivity and
reactivity of the protein bioconjugation process is an impor-
tant issue for some stringent assays, and mild reaction condi-
tions are also required to maintain the structure and func-
tion of proteins. For example, high-resolution NMR spec-
troscopy analysis of a protein–tag conjugate is sensitive to
the nonspecifically labeled protein product and the stability
of the target protein, and any side tagging reaction or partial
hydrolysis would result in a clear chemical shift perturbation
in the high-resolution NMR spectra.
+
+
+
[
a] F.-H. Ma, J.-L. Chen, Q.-F. Li, H.-H. Zuo, F. Huang, Prof. X.-C. Su
State Key Laboratory of Elemento-organic Chemistry
Collaborative Innovation Center of Chemical Science and Engineer-
ing (Tianjin)
Nankai University
Weijin Road 94, Tianjin 300071 (China)
Fax : (+86)22-23500623
E-mail: xunchengsu@nankai.edu.cn
+
[
] These authors contributed equally to this work.
We have site-specifically labeled a protein in a chiral-free
manner through a Michael addition-like thiol–ene reaction
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/asia.201402095.
Chem. Asian J. 2014, 9, 1808 – 1816
1808
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemasianj.org
Xun-Cheng Su et al.
[6]
without the assistance of radicals in aqueous solution, and
the highly stable thioether between the tag and the protein
offers the possibility to measure the protein complex by
[6b]
spectroscopic methods both in vitro and in vivo.
Chiral-
free modification of proteins by using the Michael addition-
like thiol–ene reaction is a simple yet promising way to tag
proteins with a fluorophore or metal-ion chelator. Given
[6]
that only a few examples have been reported in this field,
the ligation chemistry, including the reactivity and kinetics
of the reaction, needs to be elucidated in detail. To obtain
insight into the bioconjugation of proteins through the for-
mation of a thiol ether between a vinyl-substituted tag and
the solvent-exposed cysteine of a protein, we designed and
synthesized a number of vinyl-substituted pyridine-derived
polycarboxylic acids that bind lanthanide ions strongly. The
kinetic properties of these tags and l-cysteine were evaluat-
ed by means of UV absorption and high-resolution NMR
spectroscopy. Protein modifications with these tags were
also assessed. These kinetic pa-
Figure 1. Structure of the vinyl tags used in protein bioconjugation.
rameters are valuable guide-
lines in the bioconjugation of
proteins by using the Michael
addition-like thiol–ene reaction.
Results and Discussion
Design and syntheses of the
vinyl tags
Scheme 2. Synthesis of L3. Reagents and conditions: a) Pd
. HCl/H O. TBAF=tetrabutylammonium fluoride.
2 3
ACHTUNGTNERNUNG( OAc) /TBAF/PPh /DMF; b) 1. NaOH/acetone,
2
2
The L1 (4-vinyl-DPA; DPA=
dipicolinic acid) and L2 (4-
vinyl-(pyridine-2,6-diyl)bisme-
thylenenitrilo tetrakis(acetica-
cid), 4VPyMTA) tags have
been shown to be site-specifi-
cally attached to a protein with-
[6]
out radical initiation. To sys-
tematically evaluate the kinetic
properties of the Michael addi-
tion-like thiol–ene reaction, L3
and L4 (Figure 1) were de-
signed. All the tags are vinyl-
substituted pyridine derivatives
that contain a terminal vinyl
group and a metal-ion chelating
group. The synthesis of L3 and
L4 is depicted in Schemes 2 and
3, respectively. 2,2’:6’,2’’-Terpyr-
idine-6,6’’-dicarboxylic
acid
(
TDA) is an excellent lantha-
[10]
nide chelator,
and its homo-
logue L3 (4-vinyl-TDA) was de-
signed on the basis of its high
rigidity and feasibility to label
proteins in
a
site-specific
Scheme 3. Synthesis of L4. Reagents and conditions: a) AcOH, H
d) (CF CO) O; e) SOCl , f) NH CH CH NH ; g) BrCH COOC , CH
PPh /DMF, i) 1. NaOH/H O, 2. HCl/H O. DIPEA=N,N-diisopropylethylamine.
2
O
2
; b) H
2
SO
4
, HNO
3
; c) CH
3
COBr;
manner with paramagnetic lan-
3
2
2
2
2
2
2
2
2
H
5
3
CN, DIPEA; h) Pd
A
H
U
G
R
N
U
G
2
[10b]
thanide ions.
3
2
2
Chem. Asian J. 2014, 9, 1808 – 1816
1809
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemasianj.org
Xun-Cheng Su et al.
Compound 1 was synthesized in four steps by using DPA
[11]
as the starting material.
Hiyama coupling reaction be-
tween 1 and triethoxyvinylsilane resulted in the formation
of 2 in approximately 80% yield. L3 was obtained by hy-
drolysis of 2 with sodium hydroxide. Starting from 2,6-dime-
thylpyridine (3), L4 was synthesized in nine steps in a total
yield of approximately 2.0%. L4 contains a vinyl group at
the 4-position in pyridine and has six atoms that can coordi-
nate metal ions.
Kinetic assay: Chemical reaction rates and mechanism of
the reaction of L1 and L2 with l-cysteine
To assess the reactivity of a vinyl-substituted pyridine (elec-
tron-withdrawing group) derivative towards a free thiol
group in aqueous solution, we first determined the kinetic
data for the reaction of L1 with l-cysteine (Scheme 3) at
pH 7.0 in 20 mm phosphate buffer at 258C.
The chemical reaction was monitored by observing the
UV absorption change at 300 nm of L1 with an increase in
the concentration of l-cysteine (Figure 2a). In the presence
of a large excess amount of l-cysteine, the second-order re-
action between L1 and l-cysteine can be written as Equa-
tions (1) and (2):
obs
v ¼ k1 ½L1ꢀ
ð1Þ
ð2Þ
obs
obs
k1 ¼ k2 ½l-Cysꢀ
obs
in which v is the reaction rate, k
is the second-order reac-
2
obs
tion rate constant, and k1 is the pseudo-first-order reaction
rate constant.
obs
The value of k1 , relative to a certain concentration of l-
Figure 2. Kinetic evaluation of the Michael addition-like thiol–ene reac-
tion for a mixture of L1 and l-cysteine. a) The UV absorbance changes
of L1 in incubation with l-cysteine. 0.3 mmL1 was incubated with 6 mm
l-cysteine in 20 mm phosphate buffer at pH 7.0 and at 258C. b) Plot of
the UV changes of L1, ln (C /C ) at 300 nm versus the incubation time
cysteine, was determined as the slope (Figure 2b) by linear
fitting the UV absorption changes to the incubation time. As
obs
show in Figure 2c, the value of k1 increases linearly with
the concentration of l-cysteine. The second-order rate con-
0
t
obs
stant, k2 =0.03, was determined as the slope of the linear
0 t
for Figure 2a, in which C and C are the concentrations of L1 with re-
obs
spect to the incubation time. The slope of the simulated curve is the
correlation between k1 and the concentration of l-cys-
obs
1
obs
pseudo-first-order rate constant, k
. c) Correlation of k
1
versus the
teine.
concentration of l-cysteine, for which the slope gives the estimated
To evaluate the protonation effects of pyridine on the
overall reaction rates of L1 and l-cysteine, a number of ex-
periments were performed at different pH values in the
range from 6.0 to 9.0, which would affect the protonation of
pyridine. The kinetic parameters were determined. As ex-
obs
second-order rate constant, k
2
.
It was initially assumed that the reaction between L1 and
l-cysteine first started with nucleophilic attack of deproton-
ated sulfhydryl, thiolate, to the terminal vinyl group of L1,
obs
pected, k1 was dependent on the pH value. Notably, the
obs
observed second-order rate constant, k2 [see Eq. (2)], in-
of which the pyridine was protonated. DPA has two pK
a
[13]
creased from pH 6.0 and reached a plateau at approximately
pH 7.0; it then decreased from pH 7.5 to 9.0, as shown in
Figure 3 (see also Figure S1, Supporting Information). The
values, 2.2 and 6.9, and the pK value of the thiol group
SH
14]
[
of l-cysteine is approximately 8.4. Given that L1 is similar
to DPA, the pK of pyridine in L1 was used as 6.9 in the
a
obs
[12]
profile of k2 with respect to the pH value is similar to the
data simulation. Similar to the previous analysis, the over-
previously reported reaction of 2-vinylpyridine with reduced
all rates for the reaction of L1 with l-cysteine could there-
fore be written as [Eq. (3)]:
[12]
obs
glutathione,
for which the maximum value of k2 was
measured at approximately pH 7.0. The striking deviations
in k2 at different pH values strongly suggest that the pro-
tonation of pyridine plays a key role in the overall reaction
rate.
obs
obs
k
k
Þ
obs
k2
2
¼
¼ h
i
L1
ð3Þ
½Cysꢀ
ð
pHꢁpKa
½1 þ 10ðpKSHꢁpHÞ
ꢀ
1
þ 10
Chem. Asian J. 2014, 9, 1808 – 1816
1810
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemasianj.org
Xun-Cheng Su et al.
Table 1. The simulated k
L1 with l-cysteine by using Equation (4).
1
, k
2
, k
3
, and k
4
rate constants for the reaction of
Data
Reaction rate
k
k
k
k
1
2
3
4
1.678
0.011
0.439
0.010
rate constants k , k , k , and k are listed in Table 1. As thio-
1
2
3
4
late is a stronger nucleophile than thiol, the contributions of
thiolate in the thiol-ene reaction are expected to increase
with pH. Moreover, the population of protonated pyridine
decreases as the pH is increased. The two controversial ten-
dencies thus shift the overall reaction rates of the thiol-ene
reaction in an opposite direction. In general, the protonated
pyridine in L1 dominates the overall kinetic reaction rates
in this thiol-ene addition. This is further confirmed by the
estimated values of k , k , k , and k in Table 1, of which k
1
2
3
4
1
is higher than the others.
To obtain more information of the thiol–ene addition re-
action for different vinyl tags, we performed similar kinetic
measurements on L2 and l-cysteine. L2 was successfully
Figure 3. Plot of the experimentally determined second-order rate con-
[6b]
used to study protein behavior in crowded media. In addi-
obs
stant, k
2
, with respect to the corresponding pH value for the reaction of
tion, it forms very highly stable complexes with lanthanide
L1 with l-cysteine. The solid curve was obtained by simulation of the pH
[15]
obs
value to its determined k
2
by using a) Equation (3) and b) Equa-
metal ions. Relative to those for L1, the determined reac-
tion rates for the mixture of L2 and l-cysteine were dramat-
ically slower. Below pH 7.0, there was no clear reaction be-
tween L2 and l-cysteine by incubating the mixture for 5 h at
room temperature. The rates for the reaction of L2 with l-
cysteine increased with increasing pH. The profile of the re-
action rate versus pH suggests that the content of the cys-
teine thiolate dominates the overall reaction rate (Fig-
ure S2), which is in great contrast to the situation with L1.
The low reaction rate also indicates that the pyridine in L2
is not protonated at neutral pH, which is consistent with the
tion (4). It is evident that Equation (4) gives excellent correlation be-
obs
tween pH and k
2
.
in which k is the theoretical second-order reaction rate and
2
L1
pKa and pKSH are the pK values of the pyridine in L1 and
a
the sulfhydryl group of l-cysteine, respectively. However,
obs
the nonlinear fit of the experimental k2 value to the pH
values by using Equation (3) did not converge (Figure 3a),
and this suggests that species in addition to those in Equa-
tion (3) also contribute to the overall reaction rates.
[16]
determined protonation constant.
Hence, to better de-
Taking into account these effects, we then refit the experi-
mental data to pH values by using Equation (4), in which
the chemical reaction rates of other species are included:
scribe the reaction mechanism, Equation (4) was rewritten
as Equation (5):
obs
k1
k2
"
ꢀ
#
obs
k2
k1
¼
¼
þ
ð5Þ
obs
1 þ 10ðpKSHꢁpHÞ
1 þ 10ðpHꢁpKSHÞ
k1
k1
k2
½Cysꢀ
obs
k2
¼
¼
þ
L1
pK aL 1ꢁpH
½
Cysꢀ
ð
pHꢁpKa
Þ
ð
Þ
1
þ 10
k3
1 þ 10
k4
ð4Þ
ꢁ
In Equation (5), the term for the protonated pyridine is
þ
omitted, which results in a constant concentration for L2 in
the reaction mixture. As a consequence, only the thiol and
thiolate of cysteine contribute to the overall reaction rate.
Figure 4 presents excellent agreement between the experi-
mental data and the fitted curve by using Equation (5). The
estimated values of k and k were 0.31ꢂ0.03 and ꢁ0.06ꢂ
þ 10ðpKSHꢁpHÞ
1 þ 10ðpHꢁpKSHÞ
1
L1
in which pKa and pKSH are the pK values defined as in
a
Equation (3), and k , k , k , and k are the reaction rates for
1
2
3
4
each reactant species in solution.
1
2
L1
With the known values of pK and pKSH, the experimen-
0.03, respectively, which are almost identical to the values of
a
obs
tally measured k2 data were fit to the pH values by using
k3 and k4 obtained from Equation (4) (see Table 1). The
Equation (4). Figure 3b shows excellent correlations be-
tween the experimental data and the simulated curve, and
this clearly indicates the significant contributions of other
species in addition to the protonated pyridine and thiolate
moieties in the thiol–ene reaction. The simulated reaction
negative value of k is small and negligible, which probably
2
stems from a nonlinear fitting error. The nice-fitting curve
in Figure 4 indicates that the thiolate in cysteine is the only
major species in the reaction. Figures 3 and 4 both show
obs
that at a higher pH of 9.0, the determined value of k2 in
Chem. Asian J. 2014, 9, 1808 – 1816
1811
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemasianj.org
Xun-Cheng Su et al.
Figure 4. Kinetic assay of L2 with l-cysteine: Plot of the reaction rate
obs
constant, k
2
, with respect to pH value. The solid line was produced by
obs
nonlinear curve fitting of the pH to the experimentally determined k
2
obs
value by using Equation (5). The value of k
mixture of 0.5 mm L2 and 10 mm l-cysteine in 20 mm phosphate buffer at
58C.
2
was determined for the
2
both the L1 and L2 systems are very similar, which clearly
suggests that only the reaction between the thiolate and the
terminal vinyl group is present (Figure S3). The small reac-
tion rate constants for the thiolate in cysteine and the de-
protonated pyridine in both L1 and L2 suggest that an in-
crease in the electron deficiency in the pyridine ring dramat-
ically accelerates the overall reaction rates.
Figure 5. Kinetic assay of L3 and L4 with l-cysteine. The intensity ratio
of the vinyl protons was plotted against the incubation time with l-cys-
teine. I and I represent the signal intensity of the vinyl protons in the
0 t
absence of l-cysteine and incubated with 6.0 mm l-cysteine at time point,
respectively.
Kinetic assay of L3 and L4 with l-cysteine determined by
NMR spectroscopy
Comparison of the reaction rates of the different tags with
free thiols
In the tag of L3, the terminal vinyl-connected pyridine is
substituted by two pyridine groups at the 2,6-positions. In
tag L4, the 6-position is substituted by one methyl group
and the 2-position is similar to the left of L2. The pKa
values of the pyridine groups in the 2,2’:6’,2’’-terpyridine ho-
The four vinyl tags L1–L4 share one common structural ana-
logue, and that is that the pyridine is substituted by a termi-
nal vinyl group at the 4-position. However, the terminal
vinyl group in these tags shows strikingly different reactivi-
ties with l-cysteine. Protonation of the pyridine group that
is directly connected to the vinyl group plays a key role in
the overall reaction rates of the Michael addition like thiol–
ene reaction. L4 has lower reactivity than L1 but higher re-
activity than L3 and L2. This is probably due to the partial
protonation of the pyridine ring in L4 at neutral pH. L2 has
the lowest reactivity of the four tags, and this is because at
the measured pH value the pyridine group in L2 is depro-
tonated and the reaction rate only depends on the popula-
tion of the deprotonated thiolate. Notably, these vinyl tags
only selectively react with free thiol groups; they do not
react with any of the amine groups in any of the natural
[17]
mologue were determined to be 3.55 and 4.65. At neutral
pH, it can therefore be assumed that the pyridine moieties
in L3 are mostly deprotonated. The reaction rate constants
of L3 and L4 with l-cysteine could not be measured by UV
absorption, as no significant UV absorption changes were
observed by incubation of the tag with l-cysteine. Alterna-
tively, the chemical shift of the vinyl protons for tags L3 and
L4 are well resolved in the NMR spectra, and this can be
used to record the reaction changes with l-cysteine. High-
resolution NMR spectroscopy was then applied to monitor
the signal intensities of the vinyl protons of L3 at pH 7.0
and those of L4 at pH 6.5 in 20 mm phosphate buffer and l-
1
cysteine, respectively. The 1D H NMR spectra showed that
[6b]
the signals for the vinyl protons gradually decreased as the
incubation time with l-cysteine was increased (Figures S4
and S5). The pseudo-first-order reaction rate was deter-
mined by fitting the signal intensities of the vinyl protons to
the incubation time, as shown in Figure 5. Different from
L2, the reactions of L3 and L4 with l-cysteine proceeded
quickly. For the mixture of L4 and cysteine, the reaction was
almost complete within in 1 h.
amino acids. Taken together, the reactivity order towards
free thiol can be assigned as L2<L3<L4<L1, as listed in
Table 2.
Protein bioconjugation and NMR spectroscopy analysis
As the determined reactivity order of these vinyl tags to-
wards free thiol groups can be used as a guideline for pro-
tein chemistry, we applied this concept to protein bioconju-
gation. L1 was successfully attached to the human ubiquitin
Chem. Asian J. 2014, 9, 1808 – 1816
1812
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemasianj.org
Xun-Cheng Su et al.
obs
[a]
Table 2. The second-order rate constant, k
2
, determined for the vinyl tags.
The ligation yield is in good
agreement with the reactivity
assay of these vinyl tags. Taken
together, we have shown that
the reactivity of the vinyl tags
in tagging proteins follows the
order L1, L4, L3, and L2. Nota-
bly, the cysteine in different sec-
ondary structures of a protein
displays varied reactivity towards
the vinyl tags, as shown in L2.
obs
2
ꢁ1
ꢁ1
Tag
k
[h mM
]
pH 6.0
pH 6.5
pH 7.0
1.314
pH 7.5
pH 8.0
pH 8.5
pH 9.0
L1
L2
L3
0.960
1.228
1.213
0.005
–
1.205
0.029
–
0.904
0.142
–
0.495
0.259
–
[
b]
[b]
[b]
ND
ND
ND
–
–
–
0.066
[c]
L
4
0.372
ND
–
–
–
–
[
a] The rate constants for the reactions of L1 and L2 with l-cysteine were determined by UV absorption, and
those for the reactions of L3 and L4 with l-cysteine were determined by NMR spectroscopy. [b] Not detecta-
ble owing to a reaction rate that was too slow. [c] Not detectable owing to a reaction rate that was too fast.
T22C mutant by incubation of the protein with L1 for ap-
We then further explored the reactivity of cysteine in dif-
ferent secondary structures of a protein. Two single-point
[6a]
proximately 16–48 h at pH 7.6.
The reactivities of these
15
vinyl tags with the free thiol group in the protein were as-
sessed, and the performance for the vinyl tags in the reac-
ubiquitin mutants A28C and E64C were made, and the N-
labeled protein samples were expressed and purified. In the
structure of ubiquitin, T22 resides in the beginning of the
first a helix, A28 resides in the first a helix, and E64 resides
in the long loop region after the fifth b sheet. Under the
same conditions as those used in Figure 6a, the A28C and
E64C mutants showed greatly differing reactivities in the re-
action with L3. The E64C mutant was fully converted into
E64C–L3, whereas no clear reaction was observed for
A28C, as shown in Figure 6d,e. In summary, it is plausible
that a residue in a well-structured region shows lower activi-
ty than that in a flexible loop or unstructured segment.
1
5
1
tion with the protein was evaluated by 2D N– H heteronu-
1
5
clear single quantum coherence ( N HSQC) spectroscopy.
To better compare the reactivity and specificity of the pro-
1
5
tein modifications with these vinyl tags, the N-labeled
human ubiquitin T22C mutant was designed, overexpressed,
and purified. In the structure of ubiquitin, residue T22 re-
sides in the beginning of the a helix and its side chain is ex-
posed to the solvent. There is only one cysteine residue in
the T22C mutant, and the reaction can be monitored by the
chemical shift changes close to the ligation site, as the chem-
ical shift is sensitive to changes in the local chemical envi-
ronment. At pH 7.6 in 20 mm tris(hydroxymethyl)aminome-
1
5
thane (Tris) buffer, 0.1 mm N-ubiquitin T22C was separate-
ly mixed with L1, L2, L3, and L4 (10 equiv.). After incuba-
tion, the excess amount of the vinyl tag was removed
through an ion-exchange column with fast protein liquid
Conclusions
The Michael addition-like thiol–ene reaction is a promising
way to label proteins in a site-specific way with fluorophore
or metal-chelating tags in a chiral-free and highly stable
manner. The rate of the reaction between the vinyl tags and
the thiol group of the cysteine residue is an important issue
that was addressed in the present study, as it determines the
ligation yield of the proteins. Without radical assistance, the
reaction between the vinyl-substituted pyridine derivatives
and cysteine proceeded in aqueous solution at different re-
action rates. Protonation of the vinyl-substituted pyridine
enhanced the thiol–ene reaction greatly at neutral pH. For
the vinyl tags substituted with unprotonated pyridine,
a higher pH was required to fulfill the reaction. Conceivably,
the reactivity assay of these vinyl tags with l-cysteine was
applied in protein bioconjugation. In addition, we showed
that the cysteine residue in different secondary structures of
a protein shows strikingly different reactivity towards the
vinyl tags. Together, we showed that these novel vinyl tags
allow differentiation of the reactivity of different cysteine
residues in a protein, which makes it possible to label pro-
teins with multiple tags at different sites. The kinetic param-
eters determined in this study will serve as a valuable guide-
line in vinyl-substituted pyridine derivatives for protein
modification.
1
5
chromatography (FPLC), and the N HSQC spectrum was
recorded for the mixture of the protein and protein–tag
1
5
complex. Figure 6 presents the superimposed N HSQC
spectra of T22C before and after incubation with vinyl tags
1
5
with different times. It is evident that in the N HSQC spec-
trum of L1 with T22C, the cross-signals of the free protein,
especially those close to the T22C residue, disappeared com-
pletely and a new set of signals appeared. This suggests that
T22C was fully converted into T22C–L1. As for the mixture
1
5
of T22C and L2, the two N HSQC spectra were almost
identical, which indicates that no reaction took place be-
tween the T22C mutant and L2. The notably different re-
sults between the ligation of L2 with T22C and the previous-
[6b]
ly reported G47C
mutants may suggest that the micro-
structural environment of cysteine in the protein affects the
reactivity of cysteine.
In Figure 6c,d, a new set of cross-signals was generated
for separate mixtures of T22C with L3 and L4, which sug-
gests that the thioether bond was formed between the pro-
tein and the vinyl tag. Moreover, the cross-signals of the
free T22C mutant decreased in intensity. The content of the
protein–tag complex was quantitatively measured by com-
parison of the cross-signal volumes for the free and tag-
modified proteins. After incubation with the vinyl tags
(
10 equiv.) for 24 h, approximately 45% of the T22C protein
was labeled with L3 and nearly 85% was labeled with L4.
Chem. Asian J. 2014, 9, 1808 – 1816
1813
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemasianj.org
Xun-Cheng Su et al.
1
5
15
Figure 6. Superimposition of the N HSQC spectra of 0.1 mm uniformly N-labeled ubiquitin mutant incubated in the absence (red) and presence of
1
.0 mm vinyl tag (black), and then the excess amount of the tags was removed by FPLC (fractions of the protein–tag complex and the free protein were
combined for NMR spectroscopy measurements) and the pH was adjusted to 6.4. a) T22C-L1 incubated at pH 7.6 in 20 mm Tris at 258C for 24 h;
b) T22C-L2 incubated at pH 8.0 in 20 mm Tris at 308C for 2.5 days; c) T22C-L3 under the same conditions as those given in a); d) T22C-L incubated at
; e) A28C-L3 under the same conditions as those given in a);
4
pH 7.6 in 20 mm Tris at 258C for 1.5 days, and the protein was fully converted into T22C-L
4
f) E64C- L3 under the same conditions as those given in a).
mide (TBAB; 0.5 mL, 0.50 mmol) in THF stock were added stepwise to
the above mixture under an atmosphere of argon. The resulting solution
was stirred at 858C for 12 h and then cooled down to room temperature.
Water (5 mL) was added to the solution, and the mixture was extracted
with ethyl acetate (2ꢁ10 mL). The combined organic phase was washed
with brine, dried with anhydrous sodium sulfate, filtered, and then con-
centrated under reduced pressure. The resulting yellowish oil was puri-
Experimental Section
Syntheses
[
11]
1
2
: Synthesized from dipicolinic acid, as previously reported.
: Compound 1 (0.114 g, 0.25 mmol), Pd(OAc) (5.61 mg, 0.025 mmol),
(13.10 mg, 0.05 mmol) were dissolved in DMF (5 mL), and then
A
H
U
G
R
N
U
G
2
and PPh
3
triethoxyvinylsilane (0.10 mL, 0.50 mmol) and tetrabutylammonium bro-
Chem. Asian J. 2014, 9, 1808 – 1816
1814
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemasianj.org
Xun-Cheng Su et al.
fied by chromatography [silica gel; petroleum ether (b.p. 60–908C)/ethyl
was removed, and the final product was purified by chromatography
1
acetate=3:1]. A white solid (0.086 g, ꢃ80%) was obtained. H NMR
[silica gel; petroleum ether (b.p. 60–908C)/ethyl acetate=4:1] to give
1
(
400 MHz, CDCl
3
): d=8.88–8.76 (m, 2H), 8.70–8.62 (m, 2H), 8.19 (dd,
a yellowish oil (0.41 g). H NMR (400 MHz, CDCl
3
): d=7.40 (s, 1H),
J=10.9, 4.2 Hz, 2H), 8.04 (dd, J=7.8, 4.4 Hz, 2H), 6.96 (dd, J=17.6,
0.9 Hz, 1H), 6.29 (d, J=17.6 Hz, 1H), 5.64 (d, J=10.9 Hz, 1H), 4.55 (q,
7.02 (s, 1H), 6.65 (dd, J=10.96, 17.76 Hz, 1H), 5.99 (d, J=17.76, 1H),
5.46 (d, J=10.96 Hz, 1H), 4.14 (m, 6H), 3.94 (s, 2H), 3.58 (s, 4H), 3.47
(s, 2H), 2.89 (m, 4H), 2.53 ppm (m, 9H).
1
1
3
J=7.13 Hz, 4H), 1.52 ppm (t, J=7.32 Hz, 6H). C NMR (101 MHz,
CDCl ): d=165.35, 156.35, 154.97, 147.85, 147.14, 137.80, 135.01, 125.08,
24.37, 119.40, 119.20, 61.96, 14.36 ppm.
3
L4: NaOH (0.23 g, 6 mmol) in water (2 mL) was added to a solution of
1
11 (0.6 g, 1 mmol) in a mixture of ethanol (5 mL) and H O (5 mL). The
2
+
L3. Compound 2 (0.46 g, 1.14 mmol) was dissolved in acetone (20 mL)
and 2m NaOH (10 mL, 22 mmol) in water solution was added. The mix-
ture was then stirred for approximately 2 h. 6m HCl was added to neu-
tralize the above mixture until no more white precipitate was formed.
resulting solution was stirred at room temperature for 6 h. Dowex H
ion-exchange resin (10.0 g) was added to the above mixture, and the solu-
tion was filtered after the pH of the suspension decreased to 3. The sol-
vent was removed under reduced pressure, and the powder was suspend-
The precipitate was filtered and dried. The yield was approximately
ed in acetone (10 mL) and filtered to yield a white solid (0.36 g, 75%).
1
1
7
3%. H NMR (600 MHz, D
2
O): d=5.53 (d, J=11.3 Hz, 1H), 6.23 (d,
H NMR (400 MHz, D
1
1
2
O): d=7.20 (s, 1H), 7.11 (s, 1H), 6.61 (dd, J=
9.04, 11.64 Hz, 1H), 5.96 (d, J=19.04 Hz, 1H), 5.42 (d, J=11.64 Hz,
H), 3.56 (s, 2H), 3.02 (s, 2H), 2.81(s, 4H), 2.47(m, 4H), 2.32 ppm (s,
J=16.5 Hz, 1H), 6.82 (dd, J=16.5, 11.3 Hz, 2H), 7.85 (d, J=7.85 Hz,
A
H
U
G
R
N
U
G
ACHTUNGTRENNUNG
2
4
6
H), 7.95 (t, J=7.85 Hz, 2H), 8.27 (m, 2H), 8.32 ppm (s, 2H).
[
18]
3H).
and 5: Synthesized as previously reported.
[
19]
: Similar to a previous report, acetyl bromide (35 mL, 470 mmol) was
Kinetic measurements of L1 and L2 with l-cysteine
added dropwise to a solution of 5 (2.5 g, 148 mmol) in glacial AcOH
35 mL) at 558C. The resulting mixture was stirred for approximately 5 h.
After cooling to room temperature, water (200 mL) was added to the so-
lution, and the pH was adjusted to pH 10 with K CO powder. The above
mixture was extracted with ethyl acetate (2ꢁ). The organic phase was
dried with Na SO , and the solvent was removed under reduced pressure,
A Shimadzu UV-2550 spectrometer was used to measure the reaction
rate of the vinyl tags with l-cysteine. The reactivity of L1 was probed by
recording the decrease in the UV absorption of L1 at 300 nm by incubat-
ing 0.30 mm L1 with different concentrations of l-cysteine in 20 mm phos-
(
2
3
phate-buffered saline (PBS) at different pH values at 258C. The pseudo-
2
4
obs
first-order reaction rate, k
1
, was obtained by linear fit of the UV ab-
1
3
which resulted in a yellow powder (75%). H NMR (400 MHz, CDCl ):
d=8.04 (s, 2H), 2.59 ppm (s, 6H).
sorption changes to the incubation time. In the case of L2, the UV ab-
sorption change was monitored at 290 nm.
7
: Trifluoroacetic anhydride (8 mL, 53 mmol) was added dropwise to a so-
lution of 6 (2.0 g, 9 mmol) in dry CH Cl (25 mL) at 08C. The resulting
2
2
Protein expression and purification
solution was heated at reflux for approximately 15 h with stirring. After
cooling to room temperature, the solvent was removed under reduced
pressure. The resulting powder was mixed with a saturated aqueous solu-
15
The N-labeled ubiquitin T22C mutant was designed, expressed, and pu-
[6a]
rified according to a previously published protocol.
tion of K
mixture was extracted with ethyl acetate, and the organic phase was
dried with Na SO . The solvent was removed under reduced pressure,
which resulted in a yellow solid (1.30 g, 65%). H NMR (400 MHz,
CDCl ): d=7.32 (s, 1H), 7.30 (s, 1H), 4.74 (s, 2H), 2.58 ppm (s, 3H).
: SOCl (0.6 mL, 8 mmol) was added dropwise to a solution of 7 (1.0 g,
mmol) in dry CH Cl (10 mL) at 08C. The mixture was heated at reflux
2
CO
3
(10 mL), and the mixture was stirred for 2 h. The above
Protein modification with vinyl tags
A tenfold excess amount of the vinyl tag in a 50 mm aqueous solution
2
4
1
5
1
was added to the 0.30 mm N-labeled protein in 2.0 mL 20 mm Tris and
.30 mm tris(2-carboxyethyl)phosphine (TCEP) at pH 7.6. The pH of the
0
3
protein solution was then adjusted to 7.6. The mixture was incubated at
room temperature for 24 h. The excess amount of the free vinyl tag was
removed by FPLC with an anion-exchange column, and the buffer was
exchanged to 20 mm2-(N-morpholino)ethanesulfonic acid (MES) at
pH 6.4 for the NMR spectroscopy measurements.
8
5
2
2
2
for 3 h. After cooling to room temperature, the solvent was removed,
1
which resulted in a yellow solid (1.10 g, 87%). H NMR (400 MHz,
CDCl
3
): d=7.36 (s, 1H), 7.18 (s, 1H), 3.81 (s, 2H), 2.53 ppm (s, 3H).
NMR spectroscopy
9
: Compound 8 (4.0 g, 15 mmol) in dry CH
3
CN (30 mL) was added drop-
wise to ethanediamine (14 mL, 210 mmol) with stirring at room tempera-
ture, and the resulting mixture was stirred for 20 h. The produced solid
was filtered off. The filtrate was removed under reduced pressure, and
the remaining solid was dissolved in ethyl acetate and washed with aque-
All NMR spectra were recorded with a Bruker 600 MHz spectrometer
equipped with a QCI-cryoprobe at 298 K. The kinetic studies of tags L3
and L4 with l-cysteine were evaluated by 1D NMR spectroscopy. The
1
1
D H NMR spectra were first recorded for the 1 mm vinyl tags. 100 mm
ous K
2 3 2 4
CO . The organic phase was dried with Na SO and then concen-
l-cysteine (0.03 mL, equaled 6 mm in NMR tube) was then added to a so-
lution of 1 mm vinyl tag in 20 mm phosphate (0.55 mL) at pH 7.0 (for L3)
trated under reduced pressure, which resulted in a yellowish oil (2.9 g,
1
7
(
5%). H NMR (400 MHz, CDCl
t, 2H), 2.73 (t, 2H), 2.55 ppm (s, 3H).
0: Ethyl bromoacetate (5.5 mL, 50 mmol) in dry CH
added dropwise to a solution of 9 (2.5 g, 10 mmol) in dry CH
40 mL) and DIPEA (9 mL, 50 mmol). The resulting mixture was heated
3
): d=7.56 (brs, 2H), 3.91 (s, 2H), 2.85
1
or 6.5 (for L4). The 1D H NMR spectra were then recorded with the in-
cubation time at 298 K. The reaction rates of L4 and l-cysteine were not
determined at pH 7.0, because the reaction rates were too fast to record
the NMR spectrum. The signal intensity changes of the vinyl protons in
the vinyl tags were fitted to the incubation time with l-cysteine.
1
3
CN (20 mL) was
CN
3
(
1
5
to 608C for 10 h with stirring. After cooling to room temperature, the re-
action mixture was filtered. The solvent was removed under reduced
pressure, and the final product was purified by chromatography [silica
All 2D N HSQC spectra were performed at 298 K in 20 mm MES
buffer (pH 6.4) for 0.1 mm protein solution. The NMR spectra were pro-
[
20]
cessed with Topspin 2.1 and analyzed with Sparky.
gel; petroleum ether (b.p. 60–908C)/ethyl acetate=3:1] to give a yellow
1
oil (2.7 g, 52%). H NMR (400 MHz, CDCl
3
): d=7.56 (s, 1H), 7.21 (s,
1
2
H), 4.19 (m, 6H), 3.95 (s, 2H), 3.57 (s, 4H), 3.53 (s, 2H), 2.89 (m, 4H),
.50 (s, 3H), 1.27 ppm (m, 9H).
Acknowledgements
1
1: Compound 10 (0.8 g, 1 mmol), dry DMF (25 mL), triethoxyvinylsilane
Financial support through the 973 Program (2013CB910200) and from
the National Science Foundation of China (21073101, 21273121, and
(
0.75 mL, 1.5 mmol), and 1m TBAF (6.5 mL, 2 mmol) in THF stock were
mixed. Under an atmosphere of argon, Pd(OAc) (22 mg, 0.1 mmol) and
PPh (85 mg, 0.3 mmol) were added. The resulting mixture was heated to
08C with stirring for 3 h. The reactant mixture was poured into water
120 mL) and then extracted with ethyl acetate. The organic phase was
A
H
U
G
R
N
N
2
2
1121002) is greatly acknowledged.
3
9
(
2 4
washed with saturated NaCl and dried with Na SO . The organic solvent
Chem. Asian J. 2014, 9, 1808 – 1816
1815
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemasianj.org
Xun-Cheng Su et al.
[
1] a) N. Stephanopoulos, M. B. Francis, Nat. Chem. Biol. 2011, 7, 876–
[6] a) Q. F. Li, Y. Yang, A. Maleckis, G. Otting, X. C. Su, Chem.
Commun. 2012, 48, 2704–2706; b) Y. Yang, Q. F. Li, C. Cao, F.
Huang, X. C. Su, Chem. Eur. J. 2013, 19, 1097–1103.
[7] a) L. Weil, T. S. Seibles, Arch. Biochem. Biophys. 1961, 95, 470–473;
b) M. Friedman, J. G. Cavins, J. S. Wall, J. Am. Chem. Soc. 1965, 87,
3672–3682.
8
7
84; b) E. M. Sletten, C. R. Bertozzi, Angew. Chem. 2009, 121,
108–7133; Angew. Chem. Int. Ed. 2009, 48, 6974–6998; c) A. Don-
doni, A. Marra, Chem. Soc. Rev. 2012, 41, 573–586; d) J. M. Chalk-
er, G. J. Bernardes, B. G. Davis, Acc. Chem. Res. 2011, 44, 730–741;
e) E. Weerapana, G. M. Simon, B. F. Cravatt, Nat. Chem. Biol. 2008,
4
, 405–407; f) Y. X. Chen, G. Triola, H. Waldmann, Acc. Chem. Res.
[8] G. T. Hermanson, Bioconjugate Techniques, 2nd ed., Academic
Press, Amsterdam, 2008.
2
011, 44, 762–773; g) Y. Takaoka, A. Ojida, I. Hamachi, Angew.
Chem. Int. Ed. 2013, 52, 4088–4106; Angew. Chem. 2013, 125, 4182–
[9] a) C. E. Hoyle, A. B. Lowe, C. N. Bowman, Chem. Soc. Rev. 2010,
39, 1355–1387; b) A. Dondoni, Angew. Chem. Int. Ed. 2008, 47,
8995–8997; Angew. Chem. 2008, 120, 9133–9135; c) M. van Dijk,
D. T. Rijkers, R. M. Liskamp, C. F. van Nostrum, W. E. Hennink,
Bioconjugate Chem. 2009, 20, 2001–2016; d) A. Massi, D. Nanni,
Org. Biomol. Chem. 2012, 10, 3791–3807.
[10] a) E. S. Andreiadis, R. Demadrille, D. Imbert, J. Pecaut, M. Mazzan-
ti, Chem. Eur. J. 2009, 15, 9458–9476; b) F. Huang, Y. Y. Pei, H. H.
Zuo, J. L. Chen, Y. Yang, X. C. Su, Chem. Eur. J. 2013, 19, 17141–
17149.
4
2
200; h) C. P. Hackenberger, D. Schwarzer, Angew. Chem. Int. Ed.
008, 47, 10030–10074; Angew. Chem. 2008, 120, 10182–10228.
[
[
2] a) C. H. Kim, J. Y. Axup, P. G. Schultz, Curr. Opin. Chem. Biol.
2013, 17, 412–429; b) X. C. Su, G. Otting, J. Biomol. NMR 2010, 46,
101–112; c) K. N. Allen, B. Imperiali, Curr. Opin. Chem. Biol. 2010,
14, 247–254.
3] a) D. Weinrich, P. C. Lin, P. Jonkheijm, U. T. T. Nguyen, H. Schrod-
er, C. M. Niemeyer, K. Alexandrov, R. Goody, H. Waldmann,
Angew. Chem. Int. Ed. 2010, 49, 1252–1257; Angew. Chem. 2010,
1
22, 1274–1279; b) P. Jonkheijm, D. Weinrich, M. Koehn, H. Engel-
[11] P. Kadjane, C. Plates-Iglesias, R. Ziessel, L. J. Charbonniere, J.
Chem. Soc. Dalton Trans. 2009, 5688–5700.
[12] K. Lindorff-Larsen, J. R. Winther, Anal. Biochem. 2000, 286, 308–
310.
kamp, P. C. M. Christianen, J. Kuhlmann, J. C. Maan, D. Nuesse, H.
Schroeder, R. Wacker, R. Breinbauer, C. M. Niemeyer, H. Wald-
mann, Angew. Chem. Int. Ed. 2008, 47, 4421–4424; Angew. Chem.
2
008, 120, 4493–4496; c) S. Wittrock, T. Becker, H. Kunz, Angew.
Chem. Int. Ed. 2007, 46, 5226–5230; Angew. Chem. 2007, 119, 5319–
323.
4] a) A. Dondoni, A. Massi, P. Nanni, A. Roda, Chem. Eur. J. 2009, 15,
1444–11449; b) M. Lo Conte, S. Staderini, A. Marra, M. Sanchez-
Navarro, B. G. Davis, A. Dondoni, Chem. Commun. 2011, 47,
1086–11088; c) Y. Li, M. Yang, Y. Huang, X. Song, L. Liu, P. R.
[13] J. G. Speight, Langeꢀs Handbook of Chemistry, 16th ed., McGraw-
Hill, New York, 2005.
5
[14] R. M. C. Dawson, D. C. Elliott, W. H. Elliott, K. M. Jones, Data for
Biochemical Research, 3rd ed., Clarendon Press, Oxford, 1986.
[15] L. Pellegatti, J. Zhang, B. Drahos, S. Villette, F. Suzenet, G. Guillau-
met, S. Petoud, E. Tꢂth, Chem. Commun. 2008, 6591–6593.
[16] C. S. Bonnet, F. Buron, F. Caillꢃ, C. M. Shade, B. Draho sˇ , L. Pelle-
gatti, J. Zhang, S. Villette, L. Helm, C. Pichon, F. Suzenet, S.
Petoud, E. Tꢂth, Chem. Eur. J. 2012, 18, 1419–1431.
[17] M. K. Nazeeruddin, S. M. Zakeeruddin, R. Humphry-Baker, T. A.
Kaden, M. Grꢄtzel, Inorg. Chem. 2000, 39, 4542–4547.
[18] A. M. Achmatowicz, L. S. Hegedus, S. David, J. Org. Chem. 2003,
68, 7661–7666.
[
1
1
Chen, Chem. Sci. 2012, 3, 2766–2770; d) Y. A. Lin, J. M. Chalker, N.
Floyd, G. J. Bernardes, B. G. Davis, J. Am. Chem. Soc. 2008, 130,
9
Chem. Commun. 2009, 3714–3716; f) N. Floyd, B. Vijayakrishnan,
J. R. Koeppe, B. G. Davis, Angew. Chem. Int. Ed. 2009, 48, 7798–
7
642–9645; e) J. M. Chalker, Y. A. Lin, O. Boutureira, B. G. Davis,
802; Angew. Chem. 2009, 121, 7938–7942; g) J. M. Chalker, L.
Lercher, N. R. Rose, C. J. Schofield, B. G. Davis, Angew. Chem. Int.
Ed. 2012, 51, 1835–1839; Angew. Chem. 2012, 124, 1871–1875; h) F.
Li, A. Allahverdi, R. Yang, G. B. Lua, X. Zhang, Y. Cao, N. Korolev,
L. Nordenskiold, C. F. Liu, Angew. Chem. Int. Ed. 2011, 50, 9611–
[19] R. C. Jones, A. J. Canty, J. A. Deverell, M. G. Gardiner, R. M. Guijt,
R. Rodemann, J. A. Smith, V. A. Tolhurst, Tetrahedron 2009, 65,
7474–7481.
[20] T. D. Goddard, D. G. Kneller, Sparky 3, University of California,
San Francisco, 2002.
9
614; Angew. Chem. 2011, 123, 9785–9788; i) L. Markey, S. Giorda-
ni, E. M. Scanlan, J. Org. Chem. 2013, 78, 4270–4277.
Received: February 21, 2014
[
5] a) J. E. T. Corrie, J. Chem. Soc. Perkin Trans. 1 1994, 2975–2982;
b) V. Hong, A. A. Kislukhin, M. G. Finn, J. Am. Chem. Soc. 2009,
Published online: May 21, 2014
1
31, 9986–9994.
Chem. Asian J. 2014, 9, 1808 – 1816
1816
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim