A genetically encoded spin label for electron paramagnetic resonance distance measurements

Article abstract of DOI:10.1021/ja411535q

We report the genetic encoding of a noncanonical, spin-labeled amino acid in Escherichia coli. This enables the intracellular biosynthesis of spin-labeled proteins and obviates the need for any chemical labeling step usually required for protein electron paramagnetic resonance (EPR) studies. The amino acid can be introduced at multiple, user-defined sites of a protein and is stable in E. coli even for prolonged expression times. It can report intramolecular distance distributions in proteins by double-electron electron resonance measurements. Moreover, the signal of spin-labeled protein can be selectively detected in cells. This provides elegant new perspectives for in-cell EPR studies of endogenous proteins.

Full text of DOI:10.1021/ja411535q

Communication
pubs.acs.org/JACS
A Genetically Encoded Spin Label for Electron Paramagnetic
Resonance Distance Measurements
Moritz J. Schmidt, Julia Borbas, Malte Drescher,* and Daniel Summerer*
Department of Chemistry, Zukunftskolleg, and Konstanz Research School Chemical Biology, University of Konstanz,
Universitatsstraße 10, 78457 Konstanz, Germany
̈
S
* Supporting Information
tRNAPyl/pyrrolysyl-tRNA-synthetase (PylRS) pairs in E.
coli.13−24 Compared to previously genetically encoded ncAA,
a particular challenge of this approach is the reported reactivity
of nitroxides in vivo, which results in lifetimes far below the
requirements of typical protein expressions in E. coli.25−28 We
chose the five-membered 2,2,5,5-tetramethyl-pyrrolin-1-oxyl
moiety as spin-label for amino acid design (1a and 2a, Figure
1A).26,27 To avoid reactions that might interfere with the design
ABSTRACT: We report the genetic encoding of a
noncanonical, spin-labeled amino acid in Escherichia coli.
This enables the intracellular biosynthesis of spin-labeled
proteins and obviates the need for any chemical labeling
step usually required for protein electron paramagnetic
resonance (EPR) studies. The amino acid can be
introduced at multiple, user-defined sites of a protein
and is stable in E. coli even for prolonged expression times.
It can report intramolecular distance distributions in
proteins by double-electron electron resonance measure-
ments. Moreover, the signal of spin-labeled protein can be
selectively detected in cells. This provides elegant new
perspectives for in-cell EPR studies of endogenous
proteins.
lectron paramagnetic resonance (EPR) spectroscopy in
Ecombination with site-directed spin labeling (SDSL) is a
powerful tool to study structure, dynamics, and interactions of
proteins.1,2 Of particular interest are SDSL-EPR distance
measurements that can be performed using two identical,
small spin labels. Such measurements rely on the dipole−dipole
coupling between the spin labels (that is inversely proportional
to the cube of their distances) and are precise over a broad
range of distances.3−5 These dipole−dipole interactions can be
separated from other contributions of the spin Hamiltonian
using pulsed methods6 such as the widely used double-electron
electron resonance [DEER, an acronym synonymously used
with PELDOR (pulsed electron double resonance)].
SDSL traditionally relies on the chemical conjugation of
unique cysteine residues in proteins with sulfhydryl-reactive
reagents such as the methanethiosulfonate spin label
(MTSSL).7 Alternatively, nitroxide amino acids can be
introduced by protein total synthesis or semisynthesis,8
nonsense codon suppression with chemically aminoacylated
tRNAs,9,10 or oxime formation with the noncanonical amino
acid (ncAA) p-acetyl-L-phenylalanine.11 However, these
approaches rely on obligate chemical labeling steps that require
multistep procedures and result in limitations such as the need
for cystein-free mutants and for surface-accessible sites of the
protein.
Figure 1. Genetic encoding of spin-labeled amino acids. (A) Structures
of nitroxide and hydroxylamine amino acids. (B) Growth assay of E.
coli expressing CAT_Q98TAG and tRNAPyl/PylRS-SL1 in the
presence and absence of 2a on Cam media. (C) Expression of GFP-
Y39→2a and GFP-Y39→2b with C-terminal His6 tag under
coexpression of tRNAPyl/PylRS-SL1. Top: Cellular GFP fluorescence.
Bottom: SDS-PAGE analysis of proteins purified by Ni-NTA
chromatography. (D) SDS-PAGE analysis of TRX-R74→2a expres-
sions under coexpression of tRNAPyl/PylRS-SL1 purified as in panel C.
process, we employed the corresponding, expectedly stable
hydroxylamines 1b and 2b as surrogates for evolution of PylRS
mutants with potential cross-reactivity toward nitroxides 1a and
2a.
We subjected a library of ∼108 PylRS active site mutants to
an in vivo selection process [(see the Supporting Information
(SI)].12,22 No PylRS mutant was identified for 1b, whereas for
2b, the sampled diversity converged to the three mutants
PylRS-SL1−3 (SI Table 1). Clones coexpressing tRNAPyl
To obviate the need for chemical labeling in SDSL-EPR
studies of proteins, we aimed to directly biosynthesize spin-
labeled proteins in vivo. Our approach is based on the genetic
encoding of nitroxide amino acids in response to the amber
stop codon (TAG)12 by evolutionary design of orthogonal
Received: November 12, 2013
Published: January 15, 2014
© 2014 American Chemical Society
1238
dx.doi.org/10.1021/ja411535q | J. Am. Chem. Soc. 2014, 136, 1238−1241

Journal of the American Chemical Society
Communication
/PylRS-SL1−3 and an amber mutant of chloramphenicol
(Cam)-acetyltransferase (CAT_Q98TAG) exhibited growth on
Cam media in the presence of both 2b and 2a, but did not grow
in their absence (SI Figure 6). PylRS-SL1 exhibited the
strongest growth (Figure 1B), suggesting that it indeed exhibits
the envisaged cross-reactivity for 2a and 2b and promotes their
incorporation into proteins with high efficiency and fidelity. For
quantitative analysis, we coexpressed tRNAPyl/PylRS-SL1 with
an amber mutant of green fluorescent protein with a C-terminal
His6 tag (GFP-Y39TAG)21 in the presence or absence of 2a or
2b. Cellular fluorescence as well as SDS-PAGE analysis of the
protein (purified by Ni-NTA chromatography via the C-
terminal His6 tag) revealed the expression of full-length protein
only in the presence of 2a or 2b to similar yields (4.6 and 4.3
mg/L, Figure 1C). Moreover, electrospray ionization tandem
mass spectrometry (ESI-MS/MS) of DTT-reduced, trypsin-
digested GFP-Y39→2a revealed the expected presence of the
reduced form 2b at position 39 (SI Figure 7). In addition, an
equivalent expression/purification employing 2a and different
amber mutants of E. coli thioredoxin revealed expression of full-
length protein only in the presence of 2a (Figure 1D and SI
Figure 11). In summary, these data indicate high efficiency and
fidelity of incorporation promoted by PylRS-SL1, i.e., protein
labeling degrees of >99%.12 PylRS-SL1 contains three
mutations, Y306A, Y384F, and I413L. The Y306A mutation
has previously been shown to facilitate the acceptance of large,
Nε-carbamate-linked lysine derivatives,18,21,22 presumably due
to an opening of the binding pocket.18,29 I413 is positioned
behind Y306; hence, the mutation to leucine may further
modulate the hydrophobic surface of the rear part of the
opened pocket (SI Figure 5). The Y384F mutation has been
shown to increase aminoacylation rates for distinct ncAA,
possibly by facilitating access of tRNAPyl.18
Nitroxides have been reported to be highly unstable in all
previously studied intracellular contexts,26−28 but their
reactivity in E. coli is unknown. Though high labeling degrees
are offered by our approach, its applicability is determined by
the integrity degree (i.e., the percentage of intact, paramagnetic
2a present in a protein) as a function of time. We expressed
GFP-Y39→2a in E. coli Top10 with time-resolved monitoring
by SDS-PAGE analysis and EPR measurements (Figure 2A). In
view of potential DEER measurements that require the
incorporation of multiple ncAA, we also tested E. coli JX33
that provides increased amber suppression efficiencies.30,31
Expression levels increased up to an induction time of 6 h, with
higher levels for JX33. Likewise, the purified proteins featured a
nitroxide-characteristic EPR signal that emerged at 3 h and
reached its maximum at 6 h (Figure 2B). The spectral shape for
2a incorporated into protein revealed a relatively low mobility
of the nitroxide moiety (Figure 2C, rotational correlation time
of τ = 250 ps). In contrast, the spectrum of free 2a
corresponded to a fast-tumbling small molecule (Figure 2D, τ
= 40 ps; for full characterization of the EPR spectrum of 2a, see
SI). Quantification of the integrity degree by EPR measure-
ments revealed the presence of 49% of intact spin label, even
after 8 h induction time in the presence of 1 mM 2a (Figure
2E).
Figure 2. Direct biosynthesis of spin-labeled proteins in E. coli. (A)
Expression kinetics of GFP-Y39→2a in E. coli strains Top10 (TT) and
JX33 analyzed by SDS-PAGE. (B) EPR measurements of GFP-Y39→
2a samples from panel A (from Top10 and JX33 shown in red and
black, respectively). (C) EPR spectrum of purified TRX-D14→2a in
buffer. (D) EPR spectrum of 2a in buffer. Spectrum and spectral
simulation for C and D are shown in red and black, respectively. (E)
Quantification of the integrity degree of GFP-Y39→2a. Concentration
of GFP protein was quantified with a BCA assay (light gray bars) and
of 2a by EPR measurements (dark gray bars). Integrity degree was
obtained as the ratio of the concentration of 2a and the concentration
of GFP in percent.
in previous in-cell SDSL-EPR experiments.26 The data further
show that the stability of 2a in E. coli is generally much higher
than previously reported for nitroxides in other intracellular
environments, possibly reflecting differences in the types and
abundances of reducing agents between previously studied
organisms and E. coli.26−28 This suggests that the potential of
nitroxides for in vivo applications has previously been
underestimated and that our approach could be transferred to
other organisms.
A particular advantage of DEER over fluorescence resonance
energy transfer (FRET)-based methods for protein structural
analysis is the ability to obtain precise, absolute distance
distributions. Moreover, DEER should be more accessible in
terms of label introduction by genetic encoding: In contrast to
FRET studies that necessitate the selective incorporation of two
different, large (and potentially perturbative) chromophores
into a protein, DEER studies require only two small, identical
spin labels. Since the scope of aminoacyl-tRNA-synthetases
(aaRS) to accept large chromophores is still limited,32−35 site-
selective double labeling usually relies on the incorporation of
small, reactive ncAA in response to individual codons followed
by orthogonal conjugation in vitro.14,36 In contrast, the direct in
vivo double incorporation of 2a requires only a single tRNA/
aaRS pair and could be achieved with high efficiency under
exclusive use of the amber codon in E. coli JX33. We expressed
the doubly labeled protein mutant TRX-D14/R74→2a in JX33
(Figure 3A). The protein was purified as the single mutants and
dialyzed into aqueous buffer. Following addition of 20% (v/v)
glycerol, the protein solution was shock-frozen in order to trap
the macromolecular conformation. The distance measurements
were performed at a temperature of 50 K (for details see the
SI). TRX-D14→2a was used as a single labeled control. The
DEER data of TRX-D14→2a is in full agreement with a
When 3 mM 2a was employed, integrity degrees of 68% after
8 h and 52% after 20 h were obtained (the partial loss of
paramagnetism is likely due to a reduction of 2a to the
hydroxylamine 2b; see SI Figure 8). This shows that our
approach results in high yields of protein with a labeling degree
of >99% and integrity degrees comparable to the ones observed
1239
dx.doi.org/10.1021/ja411535q | J. Am. Chem. Soc. 2014, 136, 1238−1241

Journal of the American Chemical Society
Communication
ester, esterified to tRNAPyl, and potentially incorporated into
amber-terminated, genome-encoded host proteins could
prevent a selective detection of the protein under study.
We performed expressions of E. coli TRX-R74→2a as above,
subjected the cells to a short wash protocol and used the cell
pellets for EPR measurements. To assess the individual signal
intensities caused by the different forms of 2a, we included
controls for all three orthogonal translation components
required for genetic encoding, i.e., for 2a, the tRNAPyl/PylRS-
SL1 pair and the amber codon at position 74 (Figure 4A). No
Figure 3. EPR distance measurements between two amino acids 2a
incorporated into E. coli thioredoxin (TRX). (A) Cartoon
representation of the crystal structure of TRX (PDB entry 2TRX).38
Amino acids at both incorporation sites of 2a are shown as red sticks,
and the corresponding Cα−Cα distance is indicated as a red line. (B)
DEER raw data of doubly labeled TRX-D14/R74→2a (green) and
TRX-D14→2a as control (black). (C) DEER data of TRX-D14/
R74→2a after background correction (black) and fit based on a
model-free analysis (red). (D) Distance distribution for TRX-D14/
R74→2a corresponding to the fit shown in C (black) compared to the
theoretically predicted distance distribution between conventional
MTSSL labels (red).
Figure 4. Selective detection of endogenous, spin-labeled proteins in
E. coli. (A) EPR signal intensities for EPR measurements of washed E.
coli cells with negative controls omitting 2a, the amber codon at
position 74, or the tRNAPyl/PylRS-SL1 pair as shown in the figure.
A.U. = arbitrary units. (B) EPR spectrum of washed E. coli cells
expressing E. coli TRX-R74→2a (column 4 of panel A) at 4 °C. The
EPR signal corresponds to a spin-label concentration of 13.4 μM.
homogeneous 3D distribution of spin labels as expected for a
single labeled protein in solution (Figure 3B). For TRX-D14/
R74→2a, the background-corrected DEER data (Figure 3C)
resulted in the corresponding distance distribution by a model-
free analysis using DEERAnalysis2013 (Figure 3D; see the
SI).37
A systematic analysis of the influence of all steps of data
postprocessing was performed (see the SI). In general, the
width of the distance distribution reflects both the flexibility of
TRX and the reorientational degree of freedom of 2a, while the
errors of EPR distance measurements are significantly smaller.5
The influence of the linker flexibility of the spin labels can be
predicted via a rotamer approach. The experimental distance
distribution for TRX-D14/R74→2a is in full agreement with
the theoretically predicted distance distribution between
conventional MTSSL spin labels (Figure 3D; see the SI) and
matches distances expected from published crystal structures
(2.1 Å for the Cα−Cα distance, Figure 3A).38 Hence, we can
conclude that 2a is a useful structural probe with characteristics
similar to those of MTSSL.
The applicability of 2a for in vivo protein biosynthesis raises
the possibility, that the EPR signal of endogenous, spin-labeled
proteins could even be selectively detected in cells? Since in-cell
SDSL-EPR studies ususally rely on the microinjection of in vitro
spin-labeled molecules into Xenopus oocytes,26,39−42 this would
provide a basis for future in-cell studies with increased
biorelevance, i.e., for studying proteins directly in the natural
host environment where they are biosynthesized and processed.
However, this is not a trivial task: microinjection allows
sensitive and background-free measurements because of high
concentrations of the labeled and purified protein. In contrast,
the encoded biosynthesis of spin-labeled proteins is limited in
intracellular concentrations and results in particular selectivity
challenges: signal of intracellular 2a as free amino acid, as AMP-
nitroxide signal was detected in the absence of 2a (column 1).
In the presence of 2a, similar intensities were detected for cells
that did not provide an amber codon in TRX (column 2 and 3).
This was independent of the presence (column 3) or absence
(column 2) of the tRNAPyl/PylRS-SL1 pair. In both cases, 2a
cannot be incorporated into TRX in response to the amber
codon, and in the latter case, 2a additionally cannot be
activated or charged to tRNAPyl. These data reveal free 2a as
the major source of EPR signal in the absence of spin-labeled
target protein. However, in its presence, i.e., in cells expressing
TRX-R74→2a as a result of coexpression of the tRNAPyl/
PylRS-SL1 pair and the presence of the amber codon and 2a,
the recorded EPR signal increased 2-fold (column 4).
This indicates that our approach indeed affords sufficient
expression levels of endogenous, spin-labeled proteins for in-
cell SDSL-EPR measurements and that free 2a can be
selectively washed out of cells under conditions that leave the
spin-labeled protein intact and localized in the cells. Moreover,
free 2a and 2a incorporated into a protein can be differentiated
by distinct spectral shapes (Figure 2C,D). Figure 4B shows the
first EPR spectrum of an endogenous, spin-labeled protein in its
natural host (sample from Figure 4A, column 4).
In conclusion, we demonstrate the intracellular, genetically
encoded biosynthesis of spin-labeled proteins. This eliminates
the need for in vitro spin-labeling and thus overcomes current
limitations of SDSL-EPR, including the restriction to cystein-
free mutants and the need for multistep labeling protocols. 2a
can be incorporated at multiple, user-defined sites within a
protein and exhibits characteristics similar to those of MTSSL,
making it a valuable probe for the measurement of intra-
molecular distance distributions by DEER. Additionally, we
1240
dx.doi.org/10.1021/ja411535q | J. Am. Chem. Soc. 2014, 136, 1238−1241

Journal of the American Chemical Society
Communication
(16) Fekner, T.; Li, X.; Lee, M. M.; Chan, M. K. Angew. Chem., Int.
Ed. 2009, 48, 1633.
(17) Chen, P. R.; Groff, D.; Guo, J. T.; Ou, W. J.; Cellitti, S.;
Geierstanger, B. H.; Schultz, P. G. Angew. Chem., Int. Ed. 2009, 48,
4052.
(18) Yanagisawa, T.; Ishii, R.; Fukunaga, R.; Kobayashi, T.;
Sakamoto, K.; Yokoyama, S. Chem. Biol. 2008, 15, 1187.
(19) Neumann, H.; Peak-Chew, S. Y.; Chin, J. W. Nat. Chem. Biol.
2008, 4, 232.
(20) Polycarpo, C. R.; Herring, S.; Berube, A.; Wood, J. L.; Soll, D.;
Ambrogelly, A. FEBS. Lett. 2006, 580, 6695.
make a first step toward future in-cell SDSL-EPR studies of
endogenous proteins. We analyze fundamental aspects of EPR
signal stability, sensitivity, and selectivity in E. coli and
demonstrate that spin-labeled proteins can be selectively
detected in living cells. Future efforts will focus on the transfer
of the approach to other organisms, on in vivo DEER
measurements, and on the development of nitroxide amino
acids with further improved stabilities and conformational
characteristics.
ASSOCIATED CONTENT
* Supporting Information
Syntheses and analytical data, construction of PylRS selection
systems, libraries and selection, expression and purification of
GFP and TRX proteins, cw EPR measurements, DEER
measurements, and rotamer simulations. This material is
available free of charge via the Internet at http://pubs.acs.org.
(21) Plass, T.; Milles, S.; Koehler, C.; Schultz, C.; Lemke, E. A.
Angew. Chem., Int. Ed. 2011, 50, 3878.
(22) Schmidt, M. J.; Summerer, D. Angew. Chem., Int. Ed. 2013, 52,
■
S
4690.
(23) Yu, Z.; Ohulchanskyy, T. Y.; An, P.; Prasad, P. N.; Lin, Q. J. Am.
Chem. Soc. 2013, 135, 16766.
(24) Li, F.; Zhang, H.; Sun, Y.; Pan, Y.; Zhou, J.; Wang, J. Angew.
Chem., Int. Ed. 2013, 52, 9700.
(25) Azarkh, M.; Okle, O.; Dietrich, D.; Drescher, M. J. Magn. Reson.
2011, 212, 450.
(26) Krstic, I.; Hansel, R.; Romainczyk, O.; Engels, J. W.; Dotsch, V.;
Prisner, T. F. Angew. Chem., Int. Ed. 2011, 50, 5070.
(27) Couet, W. R.; Brasch, R. C.; Sosnovsky, G.; Tozer, T. N. Magn.
Reson. Imag. 1985, 3, 83.
AUTHOR INFORMATION
Corresponding Authors
malte.drescher@uni-konstanz.de
daniel.summerer@uni-konstanz.de
Notes
■
(28) Belkin, S.; Mehlhorn, R. J.; Hideg, K.; Hankovsky, O.; Packer, L.
The authors declare no competing financial interest.
Arch. Biochem. Biophys. 1987, 256, 232.
(29) Schneider, S.; Gattner, M. J.; Vrabel, M.; Flugel, V.; Lopez-
Carrillo, V.; Prill, S.; Carell, T. ChemBioChem 2013, 14, 2114.
(30) Johnson, D. B. F.; Wang, C.; Xu, J. F.; Schultz, M. D.; Schmitz,
R. J.; Ecker, J. R.; Wang, L. ACS Chem. Biol. 2012, 7, 1337.
(31) Johnson, D. B.; Xu, J.; Shen, Z.; Takimoto, J. K.; Schultz, M. D.;
Schmitz, R. J.; Xiang, Z.; Ecker, J. R.; Briggs, S. P.; Wang, L. Nat. Chem.
Biol. 2011, 7, 779.
(32) Summerer, D.; Chen, S.; Wu, N.; Deiters, A.; Chin, J. W.;
Schultz, P. G. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 9785.
(33) Wang, J.; Xie, J.; Schultz, P. G. J. Am. Chem. Soc. 2006, 128,
8738.
ACKNOWLEDGMENTS
■
This work was supported by DFG grants (SU 726/1-1, SU
726/2-1 within SPP1623 and DR743/4-1 within SPP1601).
M.J.S. acknowledges a Hoechst Fellowship of the Aventis
Foundation. We thank Peter G. Schultz for plasmids
pLWJ17B3 and pBK-Naphtyl, Lei Wang for E. coli strain
JX33, Edward Lemke for plasmid pBAD_GFP39TAG, Artem
Fedoseev and Christian Hintze for experimental contributions,
and Martin Spitzbarth for data analysis.
(34) Lee, H. S.; Guo, J.; Lemke, E. A.; Dimla, R. D.; Schultz, P. G. J.
Am. Chem. Soc. 2009, 131, 12921.
REFERENCES
■
(35) Speight, L. C.; Muthusamy, A. K.; Goldberg, J. M.; Warner, J. B.;
Wissner, R. F.; Willi, T. S.; Woodman, B. F.; Mehl, R. A.; Petersson, E.
J. J. Am. Chem. Soc. 2013, 135, 18806.
(36) Brustad, E. M.; Lemke, E. A.; Schultz, P. G.; Deniz, A. A. J. Am.
Chem. Soc. 2008, 130, 17664.
(37) Jeschke, G.; Chechik, V.; Ionita, P.; Godt, A.; Zimmermann, H.;
Banham, J.; Timmel, C. R.; Hilger, D.; Jung, H. Appl. Magn. Reson.
2006, 30, 473.
(38) Katti, S. K.; LeMaster, D. M.; Eklund, H. J. Mol. Biol. 1990, 212,
167.
(39) Igarashi, R.; Sakai, T.; Hara, H.; Tenno, T.; Tanaka, T.; Tochio,
H.; Shirakawa, M. J. Am. Chem. Soc. 2010, 132, 8228.
(40) Azarkh, M.; Okle, O.; Singh, V.; Seemann, I. T.; Hartig, J. S.;
Dietrich, D. R.; Drescher, M. Chembiochem 2011, 12, 1992.
(41) Azarkh, M.; Singh, V.; Okle, O.; Dietrich, D. R.; Hartig, J. S.;
Drescher, M. ChemPhysChem 2012, 13, 1444.
(1) Jeschke, G. Annu. Rev. Phys. Chem. 2012, 63, 419.
(2) Hubbell, W. L.; Cafiso, D. S.; Altenbach, C. Nat. Struct. Biol.
2000, 7, 735.
(3) Jeschke, G.; Bender, A.; Paulsen, H.; Zimmermann, H.; Godt, A.
J. Magn. Reson. 2004, 169, 1.
(4) Jeschke, G.; Abbott, R. J. M.; Lea, S. M.; Timmel, C. R.; Banham,
J. E. Angew. Chem., Int. Ed. 2006, 45, 1058.
(5) Jeschke, G. Chemphyschem 2002, 3, 927.
(6) Schweiger, A.; Jeschke, G. Principles of pulse electron paramagnetic
resonance; Oxford University Press: Oxford, 2005.
(7) Altenbach, C.; Marti, T.; Khorana, H. G.; Hubbell, W. L. Science
1990, 248, 1088.
(8) Klare, J. P.; Steinhoff, H. J. Photosynth. Res. 2009, 102, 377.
(9) Cornish, V. W.; Benson, D. R.; Altenbach, C. A.; Hideg, K.;
Hubbell, W. L.; Schultz, P. G. Proc. Natl. Acad. Sci. U.S.A. 1994, 91,
2910.
(42) Azarkh, M.; Singh, V.; Okle, O.; Seemann, I. T.; Dietrich, D. R.;
Hartig, J. S.; Drescher, M. Nat. Protoc. 2013, 8, 131.
(10) Shafer, A. M.; Kalai, T.; Liu, S. Q. B.; Hideg, K.; Voss, J. C.
Biochemistry 2004, 43, 8470.
(11) Fleissner, M. R.; Brustad, E. M.; Kalai, T.; Altenbach, C.; Cascio,
D.; Peters, F. B.; Hideg, K.; Peuker, S.; Schultz, P. G.; Hubbell, W. L.
Proc. Natl. Acad. Sci. U.S.A. 2009, 106, 21637.
(12) Liu, C. C.; Schultz, P. G. Annu. Rev. Biochem. 2010, 79, 413.
(13) Blight, S. K.; Larue, R. C.; Mahapatra, A.; Longstaff, D. G.;
Chang, E.; Zhao, G.; Kang, P. T.; Church-Church, K. B.; Chan, M. K.;
Krzycki, J. A. Nature 2004, 431, 333.
(14) Wan, W.; Huang, Y.; Wang, Z. Y.; Russell, W. K.; Pai, P. J.;
Russell, D. H.; Liu, W. R. Angew. Chem., Int. Ed. 2010, 49, 3211.
(15) Gautier, A.; Nguyen, D. P.; Lusic, H.; An, W. A.; Deiters, A.;
Chin, J. W. J. Am. Chem. Soc. 2010, 132, 4086.
1241
dx.doi.org/10.1021/ja411535q | J. Am. Chem. Soc. 2014, 136, 1238−1241