Site-specific protein backbone and side-chain NMR chemical shift and relaxation analysis of human vinexin SH3 domain using a genetically encoded 15N/19F-labeled unnatural amino acid

Article abstract of DOI:10.1016/j.bbrc.2010.10.046

SH3 is a ubiquitous domain mediating protein-protein interactions. Recent solution NMR structural studies have shown that a proline-rich peptide is capable of binding to the human vinexin SH3 domain. Here, an orthogonal amber tRNA/tRNA synthetase pair for ¹⁵N/¹⁹F-trifluoromethyl-phenylalanine (¹⁵N/¹⁹F-tfmF) has been applied to achieve site-specific labeling of SH3 at three different sites. One-dimensional solution NMR spectra of backbone amide (¹⁵N)¹H and side-chain ¹⁹F were obtained for SH3 with three different site-specific labels. Site-specific backbone amide (¹⁵N)¹H and side-chain ¹⁹F chemical shift and relaxation analysis of SH3 in the absence or presence of a peptide ligand demonstrated different internal motions upon ligand binding at the three different sites. This site-specific NMR analysis might be very useful for studying large-sized proteins or protein complexes.

Full text of DOI:10.1016/j.bbrc.2010.10.046

Biochemical and Biophysical Research Communications 402 (2010) 461–466
Contents lists available at ScienceDirect
Biochemical and Biophysical Research Communications
journal homepage: www.elsevier.com/locate/ybbrc
Site-specific protein backbone and side-chain NMR chemical shift
and relaxation analysis of human vinexin SH3 domain using a
genetically encoded 15N/19F-labeled unnatural amino acid
a,
Pan Shi a,b,1, Zhaoyong Xi c,1, Hu Wang c, Chaowei Shi a,b, Ying Xiong b, , Changlin Tian
⇑
⇑
a National Laboratory for Physical Science at Microscale, University of Science and Technology of China, Hefei, Anhui 230026, PR China
b School of Life Science, University of Science and Technology of China, Hefei, Anhui 230026, PR China
c School of Chemistry, University of Science and Technology of China, Hefei, Anhui 230026, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 7 October 2010
Available online 12 October 2010
SH3 is a ubiquitous domain mediating protein–protein interactions. Recent solution NMR structural
studies have shown that a proline-rich peptide is capable of binding to the human vinexin SH3 domain.
Here, an orthogonal amber tRNA/tRNA synthetase pair for 15N/19F-trifluoromethyl-phenylalanine
(
15N/19F-tfmF) has been applied to achieve site-specific labeling of SH3 at three different sites. One-
dimensional solution NMR spectra of backbone amide (15N)1H and side-chain 19F were obtained for
SH3 with three different site-specific labels. Site-specific backbone amide (15N)1H and side-chain 19F
chemical shift and relaxation analysis of SH3 in the absence or presence of a peptide ligand demonstrated
different internal motions upon ligand binding at the three different sites. This site-specific NMR analysis
might be very useful for studying large-sized proteins or protein complexes.
Keywords:
Unnatural amino acids
Trifluoromethyl-phenylalanine
Site-specific labeling
Solution NMR
Relaxation analysis
Ligand binding
Ó 2010 Elsevier Inc. All rights reserved.
1. Introduction
vides a good mechanism for introducing an NMR-active spin at a
designated site in a protein. Several different stable-isotope-
Proteins are intrinsically flexible and dynamic systems. These
characteristics are critical in allowing proteins to perform various
biological functions. Nuclear magnetic resonance (NMR) has been
proven to be a unique tool in allowing a detailed, site-specific
description of internal motions in proteins. A variety of NMR pulse
sequences have been developed to study the internal mobility of
the backbone and side chains of a protein [1]. Most protein back-
bone internal motion analyses have focused on longitudinal (T1)
and transverse (T2) relaxation of backbone amide 15N and dipolar
coupling of the 15N–1H bond, while protein side-chain internal mo-
tion has been analyzed through relaxation measurements of the
nuclei 13C or 2H. However, analysis of protein relaxation data for
either backbone or side-chain nuclei is greatly hampered by diffi-
culties in peak assignment and rapid signal decay due to slow rota-
tional motion of large-sized proteins.
labeled unnatural amino acids have been incorporated into
proteins for chemical shift analysis in 1D and 2D NMR spectra [5–
7]. However, to the best of our knowledge, there have not hitherto
been any reports on relaxation analysis of proteins with site-specif-
ically incorporated unnatural amino acids with active NMR spins, in
particular, with both backbone and side-chain double labeling.
SH3 is a ubiquitous domain mediating protein–protein interac-
tions. The SH3 domain of human vinexin has been extensively stud-
ied recently, and it was shown to be bound by a proline-rich peptide
(GEVPPPRPPPPEE, P868) at several specific sites with high affinity
[8]. Previous data have shown that residues Phe7 and Tyr51 are lo-
cated in the SH3 binding pocket for peptide ligand P868, while res-
idue Tyr26 is located away from the binding pocket [8]. In the
work described herein, a 15N/19F-labeled unnatural amino acid, tri-
fluoromethyl-phenylalanine (15N/19F-tfmF), has been incorporated
at several aromatic residue sites, namely Phe7, Tyr26, or Tyr51.
Chemical shift and T1, T2 relaxation analyses of backbone amide
Unnatural amino acids were first incorporated into a protein
using an orthogonal amber tRNA/tRNA synthetase (tRNA/RS) pair
about a decade ago [2–4]. In this method, unnatural amino acids
can be specificallyincorporated into a proteinat the amber nonsense
codon (TAG) in the protein’s coding DNA sequence [4], which pro-
(
15N)1H or side-chain 19F spins at these three sites of SH3 have been
conducted in the presence or absence of peptide ligand P868.
2. Materials and methods
⇑
2.1. Synthesis of the 15N-labeled trifluoromethyl-phenylalanine
Corresponding authors. Address: School of Life Science, University of Science
and Technology of China, Hefei, Anhui 230026, PR China (C. Tian). Fax: +86 551
3600441.
To synthesize 15N-labeled trifluoromethyl-phenylalanine,
15N-labeled diethyl acetaminomalonate was first synthesized as
E-mail addresses: yxiong73@ustc.edu.cn (Y. Xiong), cltian@ustc.edu.cn (C. Tian).
These authors contributed equally to this work.
1
0006-291X/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved.
doi:10.1016/j.bbrc.2010.10.046

462
P. Shi et al. / Biochemical and Biophysical Research Communications 402 (2010) 461–466
an intermediate. A mixture of Na15NO2 (8.75 g), diethyl malonate
(16 mL), and water (0.75 mL) was placed in a three-necked, round-
bottomed flask, with toluene (24 mL) as solvent. The stirred mixture
was cooled in an ice bath. Glacial acetic acid (10 mL) was then slowly
added. The reaction mixture was warmed to 40–50 °C and
maintained at this temperature for 6 h. The organic layer was then
separated using a separating funnel. An yellow oil (diethyl oximino-
malonate) was obtained after removing the solvent. Reaction of
diethyl oximinomalonate with glacial acetic acid was initiated by
the addition of zinc powder (22 g) and the stirred mixture was kept
at 40–50 °C for 4 h. The 15N-labeled diethyl acetaminomalonate was
obtained as a pure white product by recrystallization from water.
The required 15N-labeled trifluoromethyl-phenylalanine was
synthesized from 15N-labeled diethyl acetaminomalonate and 4-tri-
fluoromethylbenzyl bromide. A solution of diethyl acetamidomalo-
nate (10.0 g) in anhydrous ethanol (125 mL) was added to a sodium
ethoxide solution and the mixture was heated at 50 °C for 1.5 h.
4-Trifluoromethylbenzyl bromide (11.0 g) was then added portion-
wise to the reaction mixture and the reaction was allowed to pro-
ceed for 20 h. The reaction mixture was then cooled to room
temperature and the precipitate was collected by filtration and
added to a mixture of 98% formic acid (50 mL) and concentrated
HCl (50 mL). The resulting mixture was refluxed at 100 °C for 24 h
before the final product of 15N-labeled trifluoromethyl-phenylala-
nine (tfmF) was dried. Details of verification of the identity of the
product and assessment of its purity can be found in the Supporting
Information.
20 min before packing onto a gravity-flow column. Impurities were
washed out by applying 50 mL of washing buffer [50 mM Tris,
500 mM NaCl, 20 mM imidazole, pH 8.0] to the column. The target
proteins were then eluted from the column using elution buffer
[50 mM Tris, 500 mM NaCl, 250 mM imidazole, pH 8.0].
Purified recombinant SH3 proteins were analyzed using
SDS–PAGE (12%, w/v) and concentrated using an Amicon Ultra
15 mL device (5000 MWCO, Millipore). Two rounds of concentra-
tion and dilution in NMR buffer [50 mM NaH2PO4–Na2HPO4, pH
6.5] were applied to the eluted protein sample for imidazole re-
moval and buffer exchange. D2O was added to the SH3 sample to
give a 10% (v/v) final concentration before solution NMR analysis.
2.5. Backbone amide 1H and side-chain 19F chemical shift and
relaxation data analysis
All one-dimensional 15N-filtered 1H NMR spectra were acquired
at 298 K on a Bruker Avance 500 MHz spectrometer equipped with
a triple-resonance cryo-probe. A heteronuclear single-quantum
correlation (HSQC) pulse sequence was applied to implement
15N magnetization filtering. The first free-induction decay (FID)
data, with 2048 complex points for the HSQC experiment, were col-
lected and processed with an exponential window function (line
broadening = 10 Hz) using Bruker data-processing software. Back-
bone amide 15N T1 and T2 relaxation data were acquired using a
1D mode standard HSQC-based pulse sequence on the same Bruker
500 MHz spectrometer [10]. Peak intensities of a total of seven lon-
gitudinal relaxation durations (61.33, 141.6, 242.0, 362.42, 523.0,
753.82 and 1145.2 ms) were measured and regressed against the
durations as a single-exponential function for the T1 data, while peak
intensities of six transverse relaxation durations (17.6, 35.2, 52.8,
70.4, 105.6 and 140.8 ms) were measured and regressed for the T2
data of backbone amide 1H in tfmF–SH3 using Origin software
(OriginLab, Co.).
2.2. Constructs
A DNA fragment encoding the human vinexin SH3 domain was
PCR-amplified from the previously described expression plasmid
[8]. The amplified fragment was inserted into pBAD vector (Invitro-
gen Co.) between the NcoI and XhoI sites. The protein sequence
derived from the plasmid contained a His6-tag attached to the
N terminus of the recombinant SH3 domain.
All one-dimensional 19F NMR spectra were acquired at 298 K on
a Bruker Avance 400 MHz spectrometer equipped with a broad-
band double-resonance probe and the observation channel was
tuned to 19F (376 MHz). One-dimensional 19F chemical shift data
were acquired with 16384 complex points using a simple 1D pulse,
and processed with an exponential window function (line broad-
ening = 10 Hz) using Bruker data-processing software. Side-chain
19F T1 relaxation data were collected with eight longitudinal relax-
ation durations (50, 100, 200, 500, 800, 1000, 1500 and 2000 ms)
using a standard Bruker one-dimensional inverse-recovery pulse
sequence. Side-chain 19F T2 relaxation data were collected with
eight transverse relaxation durations (100, 150, 200, 400, 600,
800, 1200 and 1600 ms) using a standard Bruker 1D Carr–Purcel–
Meiboom–Gill (CPMG) pulse sequence. The intensities of the series
of peaks were measured and regressed for T1 or T2 data for the
side-chain 19F in 15N/19F-tfmF–SH3.
2.3. Incorporation of 15N/19F-labeled trifluoromethyl-phenylalanine
into the SH3 domain of human vinexin protein
In plasmid pBAD-His6-SH3, codons corresponding to Phe7,
Tyr26, and Tyr51 of the SH3 domain were respectively, mutated
to amber stop codon TAG using quick-change site-directed muta-
genesis (Stratagene Co.). The pBAD plasmid coding the mutated
SH3 domain and plasmid pDule-tfmF (containing DNA sequences
coding tRNACUA and tfmF-specific aminoacyl-tRNA synthetase,
kindly provided by Dr. R.A. Mehl, Department of Chemistry, Frank-
lin and Marshall College, Pennsylvania, USA) were co-transformed
into Escherichia coli host cells TOP10 in the presence of 15
lg/mL
tetracycline and 100 g/mL amphicillin, in a similar manner as re-
l
ported previously [7,9]. The transformed bacteria were incubated
in LB medium overnight, then transferred to fresh 2ꢀ YT medium
containing 1 mM 15N/19F-tfmF at 37 °C. Expression of the SH3 pro-
tein was induced using 0.2% arabinose when the OD600 reached 1.0.
3. Results and discussion
3.1. Unnatural amino acid tfmF synthesis
2.4. Protein purification
The reactions shown in Scheme 1 were carried out to incorpo-
rate 15N spin into the unnatural amino acids. 15N spin was intro-
duced at the amide nitrogen of 19F-trifluoromethyl-phenylalanine
Cells were harvested by centrifugation at 6000 rpm for 8 min at
4 °C. Cell pastes were suspended in 40 mL of lysis buffer [50 mM
Tris–HCl, 500 mM NaCl, 3 mM imidazole, pH 8.0]. Cell suspensions
in lysis buffer were probe-sonicated (VC500, Sonics and Materials,
Danbury, CT) at a power level of 30%, 2.0 s pulse on and 4.0 s pulse
off, for a total of 10 min on ice. The lysate was then centrifuged at
16,000 rpm for 20 min at 4 °C. Pellets from the centrifugation were
discarded and the supernatant was mixed with 5 mL of Ni2+-NTA
resin (QIAgen, Valencia, CA). The mixture was rotated at 4 °C for
(
19F-tfmF) from 15N-labeled NaNO2, which provided the original
source of 15N spin (Scheme 1). Before synthesizing 15N/19F-tfmF,
15N-labeled diethyl acetamidomalonate (DEAM), an amino acid
intermediate, was synthesized (Scheme 1, reaction 1). This
15N-amide-labeled DEAM could be used as a common amino acid
intermediate for the further synthesis of many different natural
or unnatural amino acids. Here, 4-trifluoromethylbenzyl bromide

P. Shi et al. / Biochemical and Biophysical Research Communications 402 (2010) 461–466
463
sion yield was not as high as that of the wild-type protein under
the same expression conditions; the yield of purified SH3 protein
was about 2 mg/L of culture. SDS–PAGE analysis of the purified pro-
tein after Ni–NTA affinity chromatography indicated a high purity
for further solution NMR studies (data not shown).
3.3. Different chemical shift changes at different sites in the one-
dimensional NMR spectra
Since 15N/19F-tfmF was site-specifically introduced at three
different sites, only one peak was observed in each 15N-filtered
1H spectrum and 19F spectrum (Figs. 1 and 2). Therefore, resonance
assignment is straightforward. In the HSQC-based 15N-filtered
1H one-dimensional spectra (Fig. 1) and directly observed 19F
one-dimensional spectra (Fig. 2), the single peaks can be assigned
to the 15N/19F-tfmF substituted residue at the respective sites: the
Phe7 site in Figs. 1A, B, 2C and D; the Tyr26 site in Figs. 1C, D, 2E
and F; and the Tyr51 site in Figs. 1E, F, 2G and H. At the same time,
in both the amide 1H and 19F spectra, pronounced chemical shift
changes were observed for 15N/19F-tfmF incorporation at the
Phe7 (Figs. 1A, B, 2C and D) and Tyr51 sites (Figs. 1E, F, 2G and
H) with the absence or presence of peptide ligand P868, while no
chemical shift changes were observed for compound tfmF alone
(Fig. 2A and B) or for 15N/19F-tfmF incorporation at the Tyr26 site
(Fig. 2E and F). The observed chemical shift changes upon ligand
binding are consistent with previous reports that Phe7 and Tyr51
are located in the ligand-binding pocket of SH3, while Tyr26 is
located away from the binding pocket [8]. No 19F chemical shift
changes were observed for the 15N/19F-tfmF compound in the
absence or presence of the peptide ligand, which excluded any
non-specific interaction between the peptide and the labeled
residues in proteins.
Scheme 1. Synthesis of 15N-labeled trifluoromethyl-phenylalanine starting from
15N-labeled NaNO2. The 15N incorporation yields were 48.5% in reaction 1 and 46.9%
in reaction 2.
was added to pre-activated 15N-labeled diethyl acetamidomalo-
nate. Consequently, trifluoromethyl-phenylalanine was synthe-
sized with 15N labeling at its backbone amide nitrogen and a
trifluoromethyl group attached to the phenylalanine side chain
(Scheme 1, reaction 2). Since there is only one neutron difference
between 15N and the naturally abundant 14N, this minor difference
will not affect the size or chemical properties of the unnatural ami-
no acid tfmF. Therefore, tfmF-specific aminoacyl-tRNA synthetase
can recognize 15N-labeled tfmF and catalyze charging the unnatu-
ral amino acid to a specific tRNACUA. Any protein with site-specific
15N/19F-labeled tfmF incorporation will contain both 15N spin in its
backbone and 19F spin in a side chain at this specific site.
3.2. Protein expression and purification
Human vinexin SH3 domain containing site-specific
15N/19F-labeled tfmF was over-expressed from a double-plasmid
system. Site-specific 15N/19F-tfmF incorporation was accomplished
under two specificities: (1) specific recognition of tfmF by tfmF–
tRNA synthetase and charging the unnatural amino acid to the 30-tail
of tfmF–tRNA; (2) complementation between the amber stop codon
and the tfmF–tRNA anti-codon sequence. Due to the presence of the
stop codon in the SH3 sequence in pBAD plasmid, the SH3 expres-
3.4. Different changes of internal motions at different sites upon
peptide ligand binding
Fig. 3 shows backbone amide 15N and side-chain 19F longitudi-
nal T1 and transverse T2 relaxation analyses of 15N/19F-tfmF7–SH3
(
15N/19F-tfmF incorporation at the Phe7 site). Upon ligand bind-
ing, backbone amide 15N T2 relaxation times at the Phe7 site were
Fig. 1. One-dimensional 15N-filtered amide 1H spectra of 15N/19F-tfmF site-specifically labeled SH3 in the absence (A, C, E) or presence (B, D, F) of peptide ligand P868. Upon
ligand binding, different amide 1H chemical shifts were observed at the Phe7 (A, B) and Tyr51 (E, F) sites, while no chemical shift changes were observed at the Tyr26 site (C,
D).

464
P. Shi et al. / Biochemical and Biophysical Research Communications 402 (2010) 461–466
Fig. 2. One-dimensional side-chain 19F spectra of 15N/19F-tfmF site-specifically labeled SH3 in the absence (A, C, E, G) or presence (B, D, F, H) of peptide ligand P868. Upon
ligand binding, different side-chain 19F chemical shifts were observed at the Phe7 (C, D) and Tyr51 (G, H) sites, while no chemical shift changes were observed at the Tyr26
site (E, F) or for the 15N/19F-tfmF compound alone (A, B).
observed to be halved (Fig. 3B cf. D), while no pronounced differ-
ences in T1 were observed (Fig. 3A cf. C). At the same time, T2
relaxation times of side-chain 19F at the Phe7 site increased by
a factor of 10 (Fig. 3F cf. H), while the T1 relaxation time was ob-
served to increase 1.5-fold (Fig. 3E and G) upon ligand binding.
According to relaxation theory and the spectral density function
of 15N and 19F spins, the T1/T2 ratio is a function of internal
mobility [11–13]. An increased T1/T2 ratio or pronounced de-
crease in T2 indicates a restrained internal motion at a specific
site, while a decreased T1/T2 ratio or pronounced increase in T2
indicates
a liberated internal motion [14]. Therefore, the
markedly reduced backbone amide 15N T2 time and increased
side-chain 19F T2 time at the Phe7 site indicated that backbone
internal motion at the Phe7 site was restrained, while side-chain
internal motion at the Phe7 site was liberated, upon binding of
the peptide ligand P868 to SH3.
Besides relaxation analysis at the Phe7 site of SH3, site-specific
backbone and side-chain T1, T2 relaxation data at the Tyr51 and
Tyr26 sites were also collected, and are summarized in Table 1. No
pronounced backbone or side-chain T1, T2 relaxation changes were
observed at the Tyr26 site, which is consistent with previous reports
that Tyr26 is located away from the P868 binding pocket. Relaxation
analysis at the Tyr51 site showed a decreased T2 relaxation time and
an increased backbone T1/T2 ratio in both the backbone and side-
chain, indicating that internal motions in both the backbone and
side-chain regions at the Tyr51 site were restrained after peptide li-
gand binding to human vinexin SH3, which is in marked contrast to
the internal motion changes at the Phe7 site.
3.5. Site-specific isotope labeling and relaxation analysis for other
proteins
Fig. 3. Longitudinal T1 (A, C, E, G) and transverse T2 (B, D, F, H) relaxation analysis of
backbone amide 15N (A–D) and side-chain 19F (E–H) of 15N/19F-tfmF site-specifically
incorporated at the Phe7 site of SH3, in the absence (A, B, E, F) or presence (C, D, G,
H) of peptide ligand P868.
Here, we have achieved site-specific labeling and relaxation
analysis of several sites in human vinexin SH3, in the absence or
presence of a peptide ligand. The chemical shift data and backbone

P. Shi et al. / Biochemical and Biophysical Research Communications 402 (2010) 461–466
465
Table 1
Backbone and side-chain chemical shifts, and T1 and T2 relaxation times of SH3 with 15N/19F-tfmF at different sites (Phe7, Tyr51 and Tyr26) (n.d.: not detected).
Amide 1H chemical shift (ppm)
w/o P868
19F chemical shift (ppm)
w/o P868
With P868
With P868
tfmF
n.d.
n.d.
ꢁ62.21
ꢁ61.97
ꢁ62.22
ꢁ61.94
n.d.
Phe7
Tyr51
Tyr26
6.93
7.72
8.42
7.03
7.68
8.42
ꢁ62.16
ꢁ61.74
ꢁ61.96
Backbone 15N T1 (ms)
w/o P868
n.d.
Backbone 15N T2 (ms)
w/o P868
With P868
n.d.
With P868
n.d.
tfmF
n.d.
Phe7
Tyr51
Tyr26
384.3 46.0
329.59 9.89
426.50 102.15
414.9 70.5
411.56 7.77
461.10 36.77
159.29 33.25
115.55 42.24
95.06 21.36
83.48 11.29
69.94 15.09
93.25 11.19
Side-chain 19F T1 (ms)
w/o P868
1470.6 62.1
1095.7 68.7
1039.5 30.5
869.98 20.46
Side-chain 19F T2 (ms)
w/o P868
969.5 199.4
89.24 5.84
105.24 12.7
78.11 3.75
With P868
n.d.
1496.0 286.7
958.69 105.15
910.40 37.44
With P868
n.d.
1088.04 217.47
8.22 2.29
tfmF
Phe7
Tyr51
Tyr26
85.47 4.29
and side-chain relaxation times in Table 1 proved to be consistent
with the previously reported ligand binding pocket of SH3. Differ-
ent internal motions were also indicated in the backbone and side
chain at the different sites upon ligand binding. These results
strongly indicate that the method of site-specific isotope spin
incorporation is very useful and reliable for analyzing conforma-
tional changes and internal motions at biologically interesting sites
of complicated proteins.
This method provides a site-by-site mechanism to analyze detailed
conformational changes or internal mobility in the backbone or
side chain of a complicated protein system in different functional
states.
Acknowledgments
The authors are grateful for the kind courtesy of provision of plas-
mid pDule-tfmF by Dr. R.A. Mehl, Department of Chemistry, Franklin
and Marshall College, Lancaster, PA, USA. This work was supported
by the Chinese Key Research Plan (2006CB910204) and a Chinese
National High-Tech Research Grant (2006AA02A321) to C.L. Tian.
Biologically interesting sites of a protein can include active sites
of enzymes, ligand binding sites of receptors, the pore-forming re-
gion of a channel, or the interface region in protein–protein
complexes. To reveal the molecular mechanisms of a complicated
protein, analogues of the protein with a series of site mutations
(even mutation scanning) or with unnatural amino acid incorpora-
tions can be prepared for site-specific dynamics analysis. Since
Appendix A. Supplementary data
15N/19F-tfmF is
a derivative of phenylalanine, only aromatic
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bbrc.2010.10.046.
residue sites are considered safe for 15N/19F-tfmF incorporation.
This does not exclude 15N/19F-tfmF incorporation at other hydro-
phobic residue sites (e.g., Leu, Ile, Val), provided that residue
replacement does not perturb the structure or function of the
target protein. To date, there have been reports of other 13C- or
15N-labeled unnatural amino acid incorporations for protein NMR
analysis [5,15], but to the best of our knowledge only 15N/19F-tfmF
contains two different spins in both the backbone and side chain
for solution NMR chemical shift and relaxation analysis.
References
[1] J.M. Lipchock, J.P. Loria, Monitoring molecular interactions by NMR, Methods
Mol. Biol. 490 (2009) 115–134.
[2] L. Wang, J. Xie, P.G. Schultz, Expanding the genetic code, Annu. Rev. Biophys.
Biomol. Struct. 35 (2006) 225–249.
[3] J. Xie, P.G. Schultz, An expanding genetic code, Methods 36 (2005) 227–238.
[4] L. Wang, P.G. Schultz, Expanding the genetic code, Chem. Commun. (Camb.)
(2002) 1–11.
Normally, to conduct chemical shift and relaxation analyses of a
large-sized protein or protein complex, it is prerequisite to have
resonance assignments of the protein at hand. Here, single-peak
spectra have been ensured by using site-specific labeling, thereby
providing a straightforward peak assignment, which is very helpful
in that it avoids complicated, time-consuming resonance assign-
ment of a large protein system. At the same time, if we know that
there is only one peak, there will be no need to perform too many
data accumulations to acquire an NMR spectrum, simply because
we can identify a single peak even at a relatively low signal/noise
ratio. This is also helpful for speeding up site-specific chemical
shift and relaxation analyses of a protein using solution NMR.
In summary, site-specific 15N/19F-labeling has been achieved
through genetically incorporating the unnatural amino acid
15N/19F-tfmF in the human vinexin SH3 domain. Site-specific back-
bone and side-chain chemical shift and relaxation analyses of SH3
have been conducted at several different sites. The observed NMR
results are consistent with the previously reported ligand-binding
pocket of SH3. Site-specific relaxation analysis data of SH3 indicate
different internal motions at different sites upon ligand binding.
[5] S.E. Cellitti, D.H. Jones, L. Lagpacan, X. Hao, Q. Zhang, H. Hu, S.M. Brittain, A.
Brinker, J. Caldwell, B. Bursulaya, G. Spraggon, A. Brock, Y. Ryu, T. Uno, P.G.
Schultz, B.H. Geierstanger, In vivo incorporation of unnatural amino acids to
probe structure, dynamics, and ligand binding in a large protein by nuclear
magnetic resonance spectroscopy, J. Am. Chem. Soc. 130 (2008) 9268–9281.
[6] A. Deiters, B.H. Geierstanger, P.G. Schultz, Site-specific in vivo labeling of
proteins for NMR studies, Chembiochem 6 (2005) 55–58.
[7] J.C. Jackson, J.T. Hammill, R.A. Mehl, Site-specific incorporation of a 19F-amino
acid into proteins as an NMR probe for characterizing protein structure and
reactivity, J. Am. Chem. Soc. 129 (2007) 1160–1166.
[8] J. Zhang, X. Li, B. Yao, W. Shen, H. Sun, C. Xu, J. Wu, Y. Shi, Solution structure of
the first SH3 domain of human vinexin and its interaction with vinculin
peptides, Biochem. Biophys. Res. Commun. 357 (2007) 931–937.
[9] J.T. Hammill, S. Miyake-Stoner, J.L. Hazen, J.C. Jackson, R.A. Mehl, Preparation of
site-specifically labeled fluorinated proteins for 19F-NMR structural
characterization, Nat. Protoc. 2 (2007) 2601–2607.
[10] N.A. Farrow, O. Zhang, J.D. Forman-Kay, L.E. Kay, Characterization of the
backbone dynamics of folded and denatured states of an SH3 domain,
Biochemistry 36 (1997) 2390–2402.
[11] G. Lipari, A. Szabo, Model-free approach to the interpretation of nuclear
magnetic resonance relaxation in macromolecules, 2. Analysis of experimental
results, J. Am. Chem. Soc. 104 (1982) 4559–4570.
[12] G. Lipari, A. Szabo, Model-free approach to the interpretation of nuclear
magnetic resonance relaxation in macromolecules, 1. Theory and range of
validity, J. Am. Chem. Soc. 104 (1982) 4546–4559.

466
P. Shi et al. / Biochemical and Biophysical Research Communications 402 (2010) 461–466
[13] S.L. Grage, U.H. Durr, S. Afonin, P.K. Mikhailiuk, I.V. Komarov, A.S. Ulrich, Solid-
state 19F NMR parameters of fluorine-labeled amino acids. Part II: Aliphatic
substituents, J. Magn. Reson. 191 (2008) 16–23.
[15] D.H. Jones, S.E. Cellitti, X. Hao, Q. Zhang, M. Jahnz, D. Summerer, P.G. Schultz, T.
Uno, B.H. Geierstanger, Site-specific labeling of proteins with NMR-active
unnatural amino acids, J. Biomol. NMR (2009) 89–100.
[14] A.G. Palmer 3rd, Dynamic properties of proteins from NMR spectroscopy, Curr.
Opin. Biotechnol. 4 (1993) 385–391.