Site-specific labeling of proteins with a chemically stable, high-affinity tag for protein study

Article abstract of DOI:10.1002/chem.201202495

Site-specific labeling of proteins with paramagnetic lanthanides offers unique opportunities by virtue of NMR spectroscopy in structural biology. In particular, these paramagnetic data, generated by the anisotropic paramagnetism including pseudocontact shifts (PCS), residual dipolar couplings (RDC), and paramagnetic relaxation enhancement (PRE), are highly valuable in structure determination and mobility studies of proteins and protein-ligand complexes. Herein, we present a new way to label proteins in a site-specific manner with a high-affinity and chemically stable tag, 4-vinyl(pyridine-2,6-diyl) bismethylenenitrilo tetrakis(acetic acid) (4VPyMTA), through thiol alkylation. Its performance has been demonstrated in G47C and E64C mutants of human ubiquitin both in vitro and in a crowded environment. In comparison with the published tags, 4VPyMTA has several interesting features: 1)it has a very high binding affinity for lanthanides (higher than EDTA), 2)there is no heterogeneity in complexes with lanthanides, 3)the derivatized protein is stable and potentially applicable to the in situ analysis of proteins. NMR tag: Site-specific labeling of proteins with a high affinity, chemically stable lanthanide-binding tag for structural biology is presented (see figure). The protein-tag construct is an ideal system for the study of protein stability and self-assembly processes under in situ conditions. Copyright

Full text of DOI:10.1002/chem.201202495

FULL PAPER
DOI: 10.1002/chem.201202495
Site-Specific Labeling of Proteins with a Chemically Stable, High-Affinity
Tag for Protein Study
Yin Yang, Qing-Feng Li, Chan Cao, Feng Huang, and Xun-Cheng Su*[a]
Abstract: Site-specific labeling of pro-
teins with paramagnetic lanthanides
offers unique opportunities by virtue of
NMR spectroscopy in structural biolo-
gy. In particular, these paramagnetic
data, generated by the anisotropic pa-
ligand complexes. Herein, we present a
new way to label proteins in a site-spe-
cific manner with a high-affinity and
chemically stable tag, 4-vinyl(pyridine-
been demonstrated in G47C and E64C
mutants of human ubiquitin both in
vitro and in a crowded environment. In
comparison with the published tags,
4VPyMTA has several interesting fea-
tures: 1) it has a very high binding af-
finity for lanthanides (higher than
EDTA), 2) there is no heterogeneity in
complexes with lanthanides, 3) the de-
rivatized protein is stable and poten-
tially applicable to the in situ analysis
of proteins.
2,6-diyl)bismethylenenitrilo
tetrakis-
AHCTUNGTRENGN(UN acetic acid) (4VPyMTA), through
thiol alkylation. Its performance has
A
N
shifts (PCS), residual dipolar couplings
(RDC), and paramagnetic relaxation
enhancement (PRE), are highly valua-
ble in structure determination and mo-
bility studies of proteins and protein–
Keywords: alkylation · conjugation ·
lanthanides · NMR spectroscopy ·
protein modifications
Introduction
Because most proteins do not bind lanthanides specifical-
ly, generation of these paramagnetic data are usually ach-
ieved through site-specific labeling of proteins with lantha-
nide-binding ligands.[3] Recently, much effort has been di-
rected towards tagging proteins with lanthanides, including
attaching a chemically synthesized ligand through one or
two disulfide bonds,[4] insertion of a lanthanide-binding pep-
tide into a protein,[5] and non-covalently but specifically
binding a protein with a lanthanide complex.[6] Fusion of a
lanthanide-binding peptide into a protein avoids chemical
modification of the protein and disulfide bond formation be-
tween the protein and tag, but suitable insertion sites in a
protein are rare and limited. The specific interaction be-
tween a lanthanide complex and a protein has been shown
to be an attractive way of obtaining such paramagnetic data,
but engineering a specific binding motif capable of recogniz-
ing the lanthanide complex is needed for many proteins.[6b]
To date, disulfide-bond tethers between the protein and the
chemically synthesized ligand is still the most common way
to tag proteins due to the unique reactivity of the thiol
group and the flexibility of engineering the ligation site in a
protein. However, the instability of the disulfide-bond tether
Lanthanide ions are chemically similar but have very differ-
ent magnetic and optical properties. Paramagnetic lantha-
nides with anisotropic paramagnetism provide a wealth of
structural restraints including pseudocontact shifts (PCS),
paramagnetic relaxation enhancement (PRE), and residual
dipolar couplings (RDCs) of biomolecules.[1] These para-
magnetic effects offer great advantages in the study of struc-
tures and dynamics of proteins and protein–ligand com-
plexes.[2]
PCS are usually described by
1
PCS ¼
½Dcaxð3 cos2 q ꢀ 1Þ þ 1:5Dcrh sin2 q cos 2ꢀ
12pr3
in which r, q, and f are the polar coordinates of the nucle-
ar spin with respect to the principal axes of the Dc tensor.
Dcax and Dcrh are the axial and rhombic components of the
Dc tensor, respectively. It is evident that PCS are unique
structural restraints containing both distances and angular
orientations of nuclear spins relative to the paramagnetic
center.
ꢀ
precludes the subsequent use of -S S- derivatives in the
presence of reducing agents and in situ conditions.
In designing a lanthanide-binding tag, several issues have
to be taken into account: 1) the affinity for lanthanide metal
ions, 2) rigidity, 3) chirality, 4) size, and 5) side-effects on the
protein. Lanthanide metal ions have eight to nine coordina-
tion sites and a suitable tag needs to possess multiple coor-
dinating atoms to form a stable complex with lanthanide. In
addition, generation of the enantiomer in coordination with
lanthanide should be avoided so that a single paramagnetic
species is present in the NMR spectrum. Because a low-
[a] Y. Yang,+ Q.-F. Li,+ C. Cao, F. Huang, Prof. X.-C. Su
State Key Laboratory of Elemento-organic Chemistry
Nankai University
Weijin Road 94, Tianjin 300071 (P.R. China)
Fax : (+86)22-23500623
E-mail: xunchengsu@nankai.edu.cn
[+] These authors contributed equally to this work.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201202495.
Chem. Eur. J. 2013, 19, 1097 – 1103
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1097

binding-affinity tag does not capture the lanthanide tightly,
a certain amount of free paramagnetic ion will be present,
which can result in nonspecific PRE or additional binding to
the protein. Side-effects on proteins may also be caused by
nonspecific interactions between the tag and the protein, es-
pecially for tags with high molecular weight. 4-mercapto-
obtain 4VPyMTA with a total yield of approximately 11%
(Scheme 1). In comparison with the synthesis of published
tags,[3,4] large quantities of 4VPyMTA is practically achieva-
ble. It is noteworthy to point out that only a trace amount
of 4VPyMTA was obtained from the fluorine-free cross-cou-
pling between 1 and triethoxyvinylsilane in the presence of
meth
E
PdACTHNGUTERNNUG
(OAc)2 and NaOH in aqueous solution,[10] which is proba-
acid (3MDPA),[7b] 4-mercaptodipicolinic acid (4MDPA),[7c]
nitrilotriacetic acid (NTA),[4f] and 4-vinyl dipicolinic acid
(4-vinylDPA)[8] have been shown to have minimum pertur-
bation on proteins due to their small sizes, but the insufficient
coordinating numbers of these tags require the coordination
of one acidic amino acid close to the ligation site in proteins.
To overcome the above limitations, we developed an ap-
proach through which site-specific labeling of proteins could
be achieved with a high-affinity lanthanide-binding tag in a
chemically stable manner.[8] We first synthesized an interest-
ing lanthanide-binding tag, 4-vinyl(pyridine-2,6-diyl)bis-
bly due to the inactivation of palladium in coordination with
the hydrolyzed 1 in basic aqueous solution.
Tagging reaction and NMR assignments: The reaction of
1
4VPyMTA and l-cysteine was studied by 1D H NMR spec-
troscopy at 298 K; the NMR spectral changes are shown in
Figure 2. The peak intensity of 1.0 mm 4VPyMTA in 20 mm
phosphate at pH 7.5 was gradually decreased after addition
of 3.0 mm l-cysteine, and the reaction was completed within
six hours. Notably, no line broadening effects were observed
in the NMR spectra of the mixture of 4VPyMTA and l-cys-
teine, suggesting that no radicals were produced in the reac-
tion (Figure 2). In contrast, no NMR spectral changes of
4VPyMTA were detected upon incubation of 1.0 mm
4VPyMTA with 6.0 mm l-lysine or l-methionine for 24 h
under the same conditions (Figure S1), indicating that the
amino group of lysine and the N-terminal methionine do
not react with 4VPyMTA. The reaction of 4VPyMTA and
cysteine is specific for the sulfhydryl group. In the case of
4VPyMTA and ubiquitin mutants, the reaction rate was
slower than with free cysteine. At pH 7.8, thiol alkylation
was monitored by 15N-HSQC spectroscopic analysis and the
cross peaks corresponding to the free protein completely
disappeared after 24 h, implying that the tag was quantita-
tively attached to the protein both for G47C and E64C (Fig-
ure S2 in the Supporting Information).
ACHTUNGTRENNUNGmethACHTUNGTRENNUNGylenenitrilo tetrakis(acetic acid) (4VPyMTA; Figure 1).
Figure 1. Chemically stable and high affinity lanthanide tag, 4VPyMTA.
4VPyMTA contains one vinyl group that reacts with SH of
cysteine and a common metal-chelator, pyridine-2, 6-diyl-
bismethylenenitrilo tetrakis (acetic acid), which has been
widely used in luminescence and MRI studies[9] and also has
very high affinity for lanthanide metal ions (Ka >1018 mꢀ1).[9c]
Two G47C and E64C mutants of human ubiquitin were suc-
cessfully
4VPyMTA through thiol al
ation reactions. Both con-
labeled
with
AHCTUNGTRENNUNG
G
structs are stable in the pres-
ence of reducing reagents and
present high-quality NMR
spectra. Sizable PCS were pro-
duced in paramagnetic sam-
ples. In addition, no heteroge-
neity was observed in the para-
magnetic coordinated protein
samples. It has been shown
that the protein–4VPyMTA
construct is an ideal system for
the study of protein stability
and self-assembly processes
under in situ conditions.
Results
Synthesis of 4VPyMTA: Start-
ing from chelidamic acid,
seven steps were required to
Scheme 1. Synthesis of 4VPyMTA. Reagents and conditions: 1) PBr5, CH3OH; 2) NaBH4, EtOH; 3) PBr3,
CHCl3; 4) K2CO3, CH3CN; 5) TBAF, PdACHTNUTRGNE(UNG OAc)2, PPh3, DMF; 6) NaOH, H2O, EtOH; 7) Dowex W50H.
1098
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 1097 – 1103

Site-Specific Labeling of Proteins
FULL PAPER
Figure 2. Reaction of 4VPyMTA and l-cysteine in 20 mm phosphate
buffer at pH 7.5, which was monitored by 1D 1H NMR spectroscopic
analysis. A) 1.0 mm 4VPyMTA; B) 1.0 mm 4VPyMTA mixed with 3.0 mm
l-cysteine; C) incubation of the mixture B at room temperature for two
hours; D) as in C for 3.5 h; E) as in C for 5h. The NMR spectra were re-
corded at 298 K with a proton frequency of 600 MHz; the arrows depict
the intensity changes of vinyl protons in 4VPyMTA during the reaction.
Figure 3. Chemical shift differences between ubiquitin and ubiquitin–
4VPyMTA derivatives; the chemical shifts of protein backbone amides
were calculated as Dd=((DdH)2 +(DdN/10)2)1/2
. A) G47C and G47C-
4VPyMTA; B) E64C and E64C-4VPyMTA.
The protein backbone amides were assigned on the basis
of 3D 15N-NOESY-HSQC and 15N-HSQC spectra. Chemical
shift changes in 4VPyMTA-derivatized G47C and E64C
ubiquitins were limited to the amides in the vicinity of
G47C and E64C, respectively. The sizes of chemical shift
changes were small and most changes were within 0.05 ppm,
as shown in Figure 3, indicating that the conjugation of
4VPyMTA to protein has negligible perturbation on protein
structure.
Table 1. The Dc-tensor parameters of ubiquitin-4VPyMTA complexed
with Tb3+, Dy3+, Tm3+, and Yb3+, respectively.[a]
Tb3+
Dy3+
Tm3+
Yb3+
G47C-4VPyMTA
E64C-4VPyMTA
Dcax
Dcrh
Dcax
Dcrh
3.4
0.9
3.8
1.2
4.3
2.3
4.4
2.1
1.7
1.1
0.9
0.4
ꢀ2.3
ꢀ1.2
The 15N-HSQC cross-peaks of the paramagnetic samples
were assigned on the basis of assignments of the diamagnet-
ic spectrum, following the chemical shift changes of an
amide group. With respect to the paramagnetic center, the
1H and 15N spins of one amide group are close, and they
have similar chemical shift changes. As a consequence, a
number of well-resolved paramagnetic peaks could readily
be assigned. For additional assignments, the 3D structure of
ubiquitin was used in comparison with the back-calculated
PCS in the paramagnetic 15N-HSQC spectrum.
ꢀ0.5
ꢀ0.2
[a] The tensor parameters are in units of 10ꢀ32 m3.
PCS were measured in the samples containing 1:1 mix-
tures of diamagnetic (Y3+) and paramagnetic (Tb3+, Dy3+
Tm3+, or Yb3+) lanthanides. The PCS measured for the
ubiquitin–4VPyMTA complexes with Tb3+, Dy3+, Tm3+
,
,
and Yb3+ were used to determine the Dc-tensors of the re-
spective lanthanides. Four sets of PCS were applied in simul-
taneous fit of the tensors with the crystal structure of ubiq-
uitin. The calculated Dc-tensor parameters are listed in
Table 1. Similar magnitudes of Dc-tensors for a paramagnet-
ic ion were observed in 4VPyMTA-tagged G47C and E64C
proteins.
PCS measurements and Dc-tensors: Significant PCS were
observed in the complexes of 4VPyMTA-derivatized G47C
and E64C ubiquitin with paramagnetic lanthanides Tb3+
,
Dy3+, Tm3+, and Yb3+ (Table 1). All the paramagnetic
metal ions produced high quality NMR spectra (Figure 4,
Figure S3, and Figure S4). For each backbone amide, a
single cross-peak was observed in the presence of diamag-
netic Y3+ or any other paramagnetic ion. No heterogeneity
of protein conformation exchange in the lanthanide loaded
protein samples was detected.
To verify the simulation of Dc-tensors, Figure 5 depicts
the plot of experimental and calculated PCS. Good correla-
tions between the experimental PCS and back-calculated
data are observed, which are in good agreement for the
G47C and E64C constructs, respectively.
Chem. Eur. J. 2013, 19, 1097 – 1103
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1099

X.-C. Su et al.
Figure 4. A) Superposition of 15N-HSQC spectra of 0.10 mm uniformly
15N-labeled G47C-4VPyMTA in the absence (grey) and presence of a 1:1
mixture of Y3+ and Dy3+ (black). B) Superposition of 15N-HSQC spectra
of 0.10 mm uniformly 15N-labeled E64C-4VPyMTA in the absence (grey)
and presence of a 1:1 mixture of Y3+ and Dy3+ (black). The ratio of lan-
thanides to protein was about 1:1.
Figure 5. Correlations of the back-calculated PCS (Calc PCS) plotted
against the experimentally measured PCS (Exp PCS). A) Ubiquitin
G47C-4VPyMTA; B) ubiquitin E64C-4VPyMTA.
Stability of 4VPyMTA with lanthanides: To assess the ther-
mostability of 4VPyMTA-derivatized proteins with lantha-
nide complexes, we titrated ethylenediaminetetraacetic acid
(EDTA), which is a widely used high-affinity chelating
Discussion
ligand for lanthanides, into
a solution of ubiquitin–
4VPyMTA and lanthanide complex. At [EDTA]/[protein-
4VPyMTA-Dy3+] molar ratios up to 2:1, no significant
changes were observed in the 15N-HSQC spectrum of ubiq-
uitin–4VPyMTA and lanthanide complex (Figure S5), sug-
gesting that the protein–4VPyMTA has a higher binding af-
finity than EDTA.
We demonstrated a new approach to site-specific labeling of
proteins with a lanthanide-binding tag that is not only chem-
ically stable but also has high affinity for lanthanides. The
tagged protein and lanthanide complex generates a single
species so that the protein sample produces a clean and
high-quality NMR spectrum. Thiol alkylation, especially the
thiol-ene reaction, has been well-studied in material sci-
ence,[11] but few examples have been applied in protein
chemistry,[9,12] of which maleimide derivatives were mostly
used in the site-specific labeling of proteins with radicals for
EPR and fluorescence studies.[12,13] However, the reaction of
maleimide with cysteine produces a new chiral center, re-
sulting in two cross-peaks in the NMR spectrum for the pa-
Analysis of ubiquitin–4VPyMTA in crowded conditions:
The protein stability of 0.1 mm 15N-G47C-PyMTA was inves-
tigated by using 2.0 mm hen white lysozyme (HEWL) and
2.0 mm bovine serum albumin (BSA) as the crowder, respec-
tively. Paramagnetic metal ions were titrated into the above
protein solution and 15N-HSQC spectra were recorded
(Figure 6). In the presence of HEWL and BSA, the protein
construct was stable. Addition of paramagnetic lanthanide
produced similar PCS to those shown in Figure 4 and Fig-
ure S3.
ACTHUNGTRENNUG ACHUTNGRENNUG
magnetic lanthanide-bound protein.[8] Compared with
AHCTUNGTREGaNNUN cetamide tags, which are also widely used in protein
conjugations,[14] the reaction rate of 4VPyMTA with cysteine
is slower but more selective in many cases. Holoacetamide
1100
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 1097 – 1103

Site-Specific Labeling of Proteins
FULL PAPER
between protein and tag is more stable than disulfide bonds
under harsh conditions, and is also resistant to the reducing
reagents; second, 4VPyMTA has seven coordination num-
bers and forms highly stable lanthanide complexes[9c] and
also prevents the generation of enantiomers in complex with
lanthanide; third, 4VPyMTA only reacts with the solvent-
exposed cysteine and not with buried cysteine residues
within the proteins. It is noteworthy to point out that the
site-specific labeling of proteins through thiol alkylation re-
action or disulfide bond formation is not practical for pro-
teins that have multiple solvent-exposed cysteines or one
solvent-exposed cysteine that is crucial to protein function.
In these cases, insertion of a lanthanide-binding peptide into
the protein sequence[5] or the design of binding sites for pa-
ACHUTNGRENrNUG aACHTUNGTRENNUNG
cellular processes through its ability to bind a large number
of proteins specifically. Structural analysis of ubiquitin pro-
tein complexes has shown that G47 and E64 are located in a
flexible region covered by the active sites of protein when
ubiquitin binds to its partners. The difficulty of locating par-
amagnetic metal ions by fitting of Dc-tensors lies in both the
size and sign of PCS. The uniform sign of PCS generated by
each paramagnetic ion means it is hard to assign a metal po-
sition precisely even when multiple sets of PCS are used.[7b]
In Figure 5, Dy3+, Tb3+, Tm3+, and Yb3+ all generate both
positive and negative PCS in ubiquitin. Indeed, the position
of the paramagnetic ion has been precisely determined by
simultaneous fit of four sets of PCS to the protein structure.
Figure 7 depicts the calculated metal position with four sets
of paramagnetic ions; the distances of metal ion to the
amide protons of G47 and E64 are 7.0 and 4.5 ꢁ, respective-
ly.
Figure 6. A) Superposition of 15N-HSQC spectra of 0.10 mm uniformly
15N-labeled G47C-4VPyMTA and 2.0 mm HEWL in the absence (grey)
and presence of a 0.05 mm Tm3+ (black). B) Superposition of 15N-HSQC
spectra of 0.10 mm uniformly 15N-labeled G47C-4VPyMTA and 2.0 mm
BSA in the absence (grey) and presence of 0.05 mm Dy3+ (black).
Sizable PCS and sufficient numbers of PCS are equally
important in structure determination and mobility studies of
proteins. A rigid tag gives large PCS but can lead to strong
tends to react with the amino groups of proteins at a slower
rate than with the free thiol group.
Only one paramagnetic species was observed in the
sample of 4VPyMTA-derivatized protein. Below pH 8.2,
there were no detectable side-products generated by
4VPyMTA and the amino group of the N-terminal methio-
nine or the side chain of lysine in the protein. In addition,
generation of radicals due to the irradiation applied in poly-
mer science for thiol-ene addition,[11] which is harmful to
protein and DNA, was also avoided in the present study.
The reaction of 4VPyMTA and cysteine shows no radicals
produced during the reaction because the presence of radi-
cals would generate strong PRE effects on the NMR signals.
In addition, the reactivity of 4VPyMTA depends on the po-
sition of cysteine in proteins, and 4VPyMTA only reacts
with the solvent-exposed cysteine residues. No thiol alkyla-
tion was observed for Sortase A (SrtA) from Staphylococcus
aureus, a cysteine protein enzyme that has an active cysteine
that is not solvent exposed (data not shown). In comparison
with the previously reported DPA[7,8] and NTA[4f] tags,
which need to be immobilized with coordination of protein,
4VPyMTA has several advantages: first, the thioether tether
Figure 7. Ribbon drawing of ubiquitin showing the metal ion position de-
termined by fitting the PCSs of the ubiquitin-4VPyMTA and lanthanide
complexes to the ubiquitin structure (PDB code: 1ubi), and the respec-
tive distances to the amide proton of G47 and E64.
Chem. Eur. J. 2013, 19, 1097 – 1103
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1101

X.-C. Su et al.
4-Vinyl(pyridin-2,6-diyl)bismethylenenitrilo tetrakis(acetic acid)
broadening effects on protein residues. Bulky tags usually
generate significant PCS,[15] but the dynamic exchange be-
tween the tag and target protein tends to result in strong
PRE effects, especially in ms–ms motion. Because G47 and
E64 are located in a flexible part of ubiquitin, the mobility
of the loop averages the size of PCS, resulting in smaller
magnitude of Dc-tensors compared with DPA[7,8] and
NTA[4f] tags, as well as DOTA-derivatives,[4e,15a–c] peptide
tags,[4d,g,15d] and TAHA-like tags.[16] Nevertheless, a large
number of residues still experience profound PCS. Remark-
ably, only a few residues are broadened even for strong an-
(4VPyMTA): Compound 2 (0.9 g ) was first mixed with ethanol (5 mL)
and H2O (5 mL), and then 2.0m NaOH (5 mL) was added into the above
mixture. The resulting solution was stirred at room temperature over-
night. Dowex H+-ion-exchange resin (10.0 g) was added and the solution
was filtered when the pH of the suspension was decreased to 3. The solu-
tion was evaporated under reduced pressure and the solid was suspended
in acetone (10 mL) and filtered to give the product (0.62 g, 84%) as a
white solid. 1H NMR (400 MHz, 298 K, D2O): d=7.40 (s, 2H), 6.66 (dd,
J=18.8, 11.3 Hz, 1H), 6.00 (d, J=18.8 Hz, 1H), 5.43 (d, J=11.3 Hz, 1H),
3.72 (s, 4H), 3.07 ppm (s, 8H). ESI-MS: m/z calcd for C17H21N3O8: 394.13
[MꢀH]ꢀ; found: 394.20.
Construct design and expression of ubiquitin G47C and E64C: The
single-point mutants G47C and E64C of human ubiquitin were prepared
in a pET3a vector by PCR-mediated site-directed mutagenesis using
ACHTUNGTRENNUNGisoACHTUNGTRENNUNG
tropic paramagnetic ions such as Tb3+ and Dy3+
(Figure 4). The lower magnitude of Dc-tensors is beneficial
to the more observable residues in the paramagnetic bound
protein samples.
the
primers
5’-AGGCTGATCTTTGCTTGTAAACAGCTGGAA-
and 5’-TCTTCCAGCTGTTTACAAGCAAAGAT-
GATGGGC-3’
CAGCCTCTGCTGG-3’ for G47C (GGA>TGT), and primers 5’-CTA-
CAACATCCAGAAATGCTCCACCCTGCACCTGGTAC-3’ and 5’-
GTGCAGGGTGGAGCATTTCTGGATGTTGTAGTCAGAC-3’
E64C (GAG>TGC), respectively.
The ubiquitin–4VPyMTA construct is stable against the
commonly used crowders, HEWL and BSA, in mimicking
the cellular environment.[17] In particular, the construct is
stable in the presence of 2.0 mm BSA, which contains 35 cys-
teines and also binds lanthanide strongly.[18] The 4VPyMTA
conjugated protein constructs thus offer great opportunities
to investigate protein stability and self-assembly under in
situ conditions by paramagnetic NMR spectroscopic analy-
sis.
for
The coding region in all plasmids was confirmed by DNA sequencing.
The correct plasmid was transformed into the E. coli strain Rosetta
(DE3) (Novagen).
Uniformly 15N-labeled protein was expressed by growing the cells with
high-density methods.[21] The target protein was purified from the soluble
fraction of cell lysate by ammonium sulfate precipitation, followed by
chromatography on DEAE columns (GE Healthcare Biosciences) and
G50 (GE Healthcare Biosciences) gel filtration. 30 mg of protein was
usually obtained from 250 mL media.
Tagging reactions: A ten-fold excess of 4VPyMTA in a 100 mm stock in
water was added to a solution of 0.30 mm 15N-labeled protein in 2.0 mL,
20 mm Tris and 0.30 mm TCEP at pH 7.8. The mixture was adjusted to
pH 7.8 with 1.0m NaOH and the protein solution was incubated at room
temperature for about 24 h. Excess tag was removed with a PD10
column and the sample was concentrated with a Millipore ultrafilter to a
final protein concentration of about 1.0 mm. The overall yield of purified
ligation product was usually above 80%.
Conclusion
We have demonstrated an interesting lanthanide-binding
tag, 4VPyMTA, that can be simply incorporated into pro-
teins through thiolalkylation reactions. Because of its high
binding affinity for lanthanides and chemical stability, it is
anticipated that 4VPyMTA will become a very useful tag in
site-specific labeling of proteins with lanthanides for struc-
tural biology studies.
NMR spectra: All 2D and 3D NMR experiments were performed at
298 K in 20 mm MES buffer (pH 6.5) with a 1H NMR frequency of
600 MHz with a Bruker AV600 NMR spectrometer equipped with a QXI
probe. 3D NOESY-15N-HSQC spectra (100 ms mixing time, total record-
ing time 36 h) were recorded with a 0.80 mm solution of diamagnetic 15N-
ubiquitin G47C and E64C in 90% H2O/10% D2O, respectively. All 15N-
HSQC spectra in the presence of diamagnetic and paramagnetic lantha-
nide ions (molar ratio 1:1) were recorded in 0.10 mm protein solution.
PCS were measured from 15N-HSQC spectra as differences in 1H chemi-
cal shifts between samples with paramagnetic lanthanide and diamagnetic
Y3+. In situ NMR spectroscopic analysis was performed with 0.10 mm
15N-ubiquitin-4VPyMTA in the presence of 2.0 mm hen egg white lyso-
zyme (HEWL) and 2.0 mm bovine serum albumin (BSA) as the crowder,
respectively, and 20 mm MES at pH 6.5.
Experimental Section
Synthesis of 4VPyMTA: The synthesis of 4VPyMTA is shown in
Scheme 1.
4-Bromo-2,6-bisACHTUNGTRENNUNG[N,N’-bis(ethyoxycaronylmethyl)aminomethyl]pyridine
(1): Starting from chelidamic acid, 1 was synthesized by following a simi-
lar procedure to that reported previously.[19]
Fitting of the Dc-tensors: The Dc-tensor parameters were determined by
using the Numbat program.[22] Only PCS data from residues located in re-
gions of well-defined secondary structure were included in the fits to the
crystal structure of ubiquitin (pdb code: 1ubi).[23] In particular, no data
from the flexible loops and the flexible C-terminal residues were used.
4-Vinyl-2,6-bisACHTUNGTRENNUNG[N,N’-bis(ethyoxycaronylmethyl)aminomethyl]pyridine (2):
Similar to the previous report,[20] 1 (2.0 g, 3.5 mmol) was dissolved in
DMF (60 mL), and then triethoxyvinylsilane (1.5 mL, 7 mmol) and
TBAF (tetrabutylammonium fluoride; 13.5 mL, 13.5 mmol) in THF
stock, PdACHTUNGTRENNUNG(OAc)2 (40 mg, 2.8 mmol), and PPh3 (140 mg, 5 mmol) were
added stepwise under argon protection. The resulting solution was stirred
at 708C for 6h and then cooled to room temperature. The solution was
diluted with water (300 mL), the mixture was extracted with ethyl acetate
(2ꢂ60 mL), and the combined organic phases were washed with brine,
dried with anhydrous sodium sulfate, filtered, and evaporated under re-
duced pressure. The resulting yellowish oil was purified by chromatogra-
phy on silica (petroleum ether (b.p. 60–908C)/ethyl acetate 1:1.5) to give
the product as a yellowish oil (0.95 g, 52%). 1H NMR (400 MHz, CDCl3,
298 K): d=7.48 (s, 2H), 6.69 (dd, J=17.7, 11.0 Hz, 1H), 6.03 (d, J=
17.7 Hz, 1H), 5.45 (d, J=11.0 Hz, 1H), 3.74 (s, 4H), 3.53 (q, J=7.08 Hz,
8H), 3.10 (s, 8H), 1.06 ppm (t, J=7.24 Hz, 12H).
Acknowledgements
Financial support by the National Science Foundation of China (grant
agreement number 21073101, 21273121, and 21121002) is gratefully ac-
knowledged.
1102
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 1097 – 1103

Site-Specific Labeling of Proteins
FULL PAPER
C. Pichon, F. Suzenet, S. Petoud, E. Tꢄth, Chem. Eur. J. 2012, 18,
1419–1431.
[10] S. Shi, Y. Zhang, J. Org. Chem. 2007, 72, 5927–5930.
[11] C. E. Hoyle, A. B. Lowe, C. N. Bowman, Chem. Soc. Rev. 2010, 39,
1355–1387.
[12] a) Z. Miao, M. R. McCoy, D. D. Singh, B. Barrios, O. L. Hsu, S. M.
Cheal, C. F. Meares, Bioconjugate Chem. 2008, 19, 15–19; b) P.
Jonkheijm, D. Weinrich, M. Kçhn, H. Engelkamp, P. C. Christianen,
J. Kuhlmann, J. C. Maan, D. Nꢆsse, H. Schroeder, R. Wacker, R.
Breinbauer, C. M. Niemeyer, H. Waldmann, Angew. Chem. 2008,
120, 4493–4496; Angew. Chem. Int. Ed. 2008, 47, 4421–4424;
c) E. M. Sletten, C. R. Bertozzi, Angew. Chem. 2009, 121, 7108–
7133; Angew. Chem. Int. Ed. 2009, 48, 6974–6998; d) J. R. Morphy,
D. Parker, R. Kataky, M. A. W. Eaton, A. T. Millican, R. Alexander,
A. Harrison, C. Walker, J. Chem. Soc. Perkin Trans. 1 1990, 2, 573–
585.
[13] P. A. Kosen, Methods Enzymol. 1989, 177, 86–120.
[14] G. T. Hermanson, Bioconjugate Techniques, 2nd ed., Academic
Press, Amsterdam, 2008.
[15] a) D. Haussinger, J. R. Huang, S. Grzesiek, J. Am. Chem. Soc. 2009,
131, 14761–14767; b) B. Graham, C. T. Loh, J. D. Swarbrick, P. Ung,
J. Shin, H. Yagi, X. Jia, S. Chhabra, N. Barlow, G. Pintacuda, T.
Huber, G. Otting, Bioconjugate Chem. 2011, 22, 2118–2125; c) P. H.
Keizers, A. Saragliadis, Y. Hiruma, M. Overhand, M. Ubbink, J.
Am. Chem. Soc. 2008, 130, 14802–14812; d) X. C. Su, K. McAn-
drew, T. Huber, G. Otting, J. Am. Chem. Soc. 2008, 130, 1681–1687.
[1] I. Bertini, C. Luchinat, G. Parigi, Prog. Nucl. Magn. Reson. Spec-
trosc. 2002, 40, 249–273.
[2] G. Otting, J. Biomol. NMR 2008, 42, 1–9.
[3] a) F. Rodriguez-CastaÇeda, P. Haberz, A. Leonov, C. Griesinger,
Magn. Reson. Chem. 2006, 44, S10–16; b) X. C. Su, G. Otting, J.
Biomol. NMR 2010, 46, 101–112; c) P. H. Keizers, M. Ubbink, Prog.
Nucl. Magn. Reson. Spectrosc. 2011, 58, 88–96; d) J. Koehler, J.
Meiler, Prog. Nucl. Magn. Reson. Spectrosc. 2011, 59, 360–389.
[4] a) V. Gaponenko, A. S. Altieri, J. Li, R. A. Byrd, J. Biomol. NMR
2002, 24, 143–148; b) A. Dvoretsky, V. Gaponenko, P. R. Rosevear,
FEBS Lett. 2002, 528, 189–192; c) T. Ikegami, L. Verdier, P. Sakhaii,
S. Grimme, B. Pescatore, K. Saxena, K. M. Fiebig, C. Griesinger, J.
Biomol. NMR 2004, 29, 339–349; d) X. C. Su, T. Huber, N. E.
Dixon, G. Otting, ChemBioChem 2006, 7, 1599; e) P. H. Keizers, J. F.
Desreux, M. Overhand, M. Ubbink, J. Am. Chem. Soc. 2007, 129,
9292–9293; f) J. D. Swarbrick, P. Ung, X. C. Su, A. Maleckis, S.
Chhabra, T. Huber, G. Otting, B. Graham, Chem. Commun. 2011,
47, 7368–7370; g) T. Saito, K. Ogura, M. Yokochi, Y. Kobashigawa,
F. Inagaki, J. Biomol. NMR 2009, 44, 157–166.
[5] a) C. Ma, S. J. Opella, J. Magn. Reson. 2000, 146, 381–384; b) J.
Wçhnert, K. J. Franz, M. Nitz, B. Imperiali, H. Schwalbe, J. Am.
Chem. Soc. 2003, 125, 13338–13339; c) T. Zhuang, H. S. Lee, B. Im-
periali, J. H. Prestegard, Protein Sci. 2008, 17, 1220–1231; d) K. Bar-
thelmes, A. M. Reynolds, E. Peisach, H. R. A. Jonker, N. J. DeNun-
zio, K. N. Allen, B. Imperiali, H. Schwalbe, J. Am. Chem. Soc. 2011,
133, 808–819.
ˇ ˇ
[16] F. Peters, M. Maestre-Martinez, A. Leonov, L. Kovacic, S. Becker,
[6] a) X. C. Su, H. Liang, K. V. Loscha, G. Otting, J. Am. Chem. Soc.
2009, 131, 10352–10353; b) X. Jia, H. Yagi, X. C. Su, M. Stanton-
Cook, T. Huber, G. Otting, J. Biomol. NMR 2011, 50, 411–420.
[7] a) X. C. Su, B. Man, S. Beeren, H. Liang, S. Simonsen, C. Schmitz,
T. Huber, B. A. Messerle, G. Otting, J. Am. Chem. Soc. 2008, 130,
10486–10487; b) B. Man, X. C. Su, H. Liang, S. Simonsen, T. Huber,
B. A. Messerle, G. Otting, Chem. Eur. J. 2010, 16, 3827–3832; c) X.
Jia, A. Maleckis, T. Huber, G. Otting, Chem. Eur. J. 2011, 17, 6830–
6836.
R. Boelens, C. Griesinger, J. Biomol. NMR 2011, 51, 329–333.
[17] A. C. Miklos, M. Sarkar, Y. Wang, G. J. Pielak, J. Am. Chem. Soc.
2011, 133, 7116–7120.
[18] a) K. A. Majorek, P. J. Porebski, A. Dayal, M. D. Zimmerman, K. Ja-
blonska, A. J. Stewart, M. Chruszcz, W. Minor, Mol. Immunol. 2012,
52, 174–182; b) C. Li, C. Wang, J. Li, Z. Wang, J. Therm. Anal. Cal-
orim. 2007, 89, 899–905.
[19] a) H. Takalo, J. Kankare, Acta Chem. Scand., Ser. B 1987, 41, 219–
221; b) H. Takalo, P. Pasanen, J. Kankare, Acta Chem. Scand., Ser. B
1988, 42, 373–377.
[20] M. E. Mowery, P. DeShong, Org. Lett. 1999, 1, 2137–2140.
[21] J. Marley, M. Lu, C. Bracken, J. Biomol. NMR 2001, 20, 71–75.
[22] C. Schmitz, M. J. Stanton-Cook, X. C. Su, G. Otting, T. Huber, J.
Biomol. NMR 2008, 41, 179–189.
[8] Q. F. Li, Y. Yang, A. Maleckis, G. Otting, X. C. Su, Chem. Commun.
2012, 48, 2704–2706.
[9] a) C. Gunanathan, A. Pais, E. Furman-Haran, D. Seger, E. Eyal, S.
Mukhopadhyay, Y. Ben-David, G. Leitus, H. Cohen, A. Vilan, H.
Degani, D. Milstein, Bioconjugate Chem. 2007, 18, 1361–1365; b) H.
Takalo, V. M. Mukkala, H. Mikola, P. Liitti, I. Hemmilꢃ, Bioconju-
gate Chem. 1994, 5, 278–282; c) L. Pellegatti, J. Zhang, B. Drahos, S.
Villette, F. Suzenet, G. Guillaumet, S. Petoud, E. Tꢄth, Chem.
Commun. 2008, 6591–6593; d) C. S. Bonnet, F. Buron, F. Caillꢅ,
[23] R. Ramage, J. Green, T. W. Muir, O. M. Ogunjobi, S. Love, K. Shaw,
Biochem. J. 1994, 299, 151–158.
Received: July 13, 2012
Revised: September 20, 2012
Published online: November 14, 2012
ˇ
C. M. Shade, B. DrahoS, L. Pellegatti, J. Zhang, S. Villette, L. Helm,
Chem. Eur. J. 2013, 19, 1097 – 1103
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1103