A facile system for genetic incorporation of two different noncanonical amino acids into one protein in Escherichia coli

Article abstract of DOI:10.1002/anie.201000465

Two's company: Using a wild-type or evolved PylRS-pylT_UUA pair to suppress ochre mutation and an evolved MjTyrRSMjtRNA_CUA ^Tyr pair to suppress amber mutation, two different noncanonical amino acids (NAAs) have been concomitantly incorporated into one protein in E. coli with high efficiency (see picture, with NAAs 1-4; GFP = green-fluorescent protein). "Chemical equation presented"

Full text of DOI:10.1002/anie.201000465

Angewandte
Chemie
DOI: 10.1002/anie.201000465
Genetic Code Expansion
A Facile System for Genetic Incorporation of Two Different
Noncanonical Amino Acids into One Protein in Escherichia coli**
Wei Wan, Ying Huang, Zhiyong Wang, William K. Russell, Pei-Jing Pai, David H. Russell, and
Wenshe R. Liu*
Since the first report in 1998,[1] genetic incorporation of
noncanonical amino acids (NAAs) into proteins at amber
UAG codons in living cells using orthogonal aminoacyl-tRNA
synthetase (aaRS)–amber suppressor tRNA (tRNACUA) pairs
We attempted to combine the PylRS–pylT pair and the
Tyr
MjTyrRS–MjtRNA
pair to incorporate two different
CUA
NAAs into a single protein because of their proven efficiency
in NAA incorporation.[2d–h,4] To use both pairs in a single
E. coli cell, we first demonstrated that they are orthogonal to
each other. When co-expressed in E. coli, either one of the
aaRSs could not efficiently charge tRNACUA from the other
pair with its cognate amino acid (Ne-Boc-l-lysine for PylRS[2d]
and l-tyrosine for MjTyrRS) to suppress an amber mutation
at position 112 of a chloramphenicol acetyltransferase gene to
give detectable chloramphenicol resistance. We then tested
whether mutated pylT can be used to suppress other blank
codons. Although naturally encoded by the UAG codon, it
has been reported that pyrrolysine is not hardwired for co-
translational insertion at UAG codon positions.[5] The crystal
structure of the PylRS–pylT complex also revealed no direct
interaction between PylRS and the anticodon region of
pylT.[6] We suspected that mutation of the anticodon of pylT
might not affect its interaction with PylRS, and the mutated
PylRS-pylT pair could be used to suppress a blank codon such
as opal UGA codon, ochre UAA codon, or even a four-base
codon UAGA. A pAcKRS–pylT–GFP1Amber plasmid from
our previous work[3,7] was employed. This plasmid contains
genes coding a mutant PylRS (AcKRS) specific for AcK,
pylT, and GFPUV. The GFPUV gene has one amber mutation at
position 149. Growing cells transformed with this plasmid in
lysogeny broth (LB) medium supplemented with 5 mm AcK
led to full-length GFPUV expression that was easily detected
by fluorescent emission of GFPUV when excited (Figure 1).
When the anticodon of pylT was mutated to the complimen-
tary one for opal, ochre, or a four-base UAGA codon and the
has flourished. Evolved Methanococcus jannaschii tyrosyl-
Tyr
CUA
tRNA synthetase (MjTyrRS)–MjtRNA
pairs together
with the naturally occurring wild-type or evolved pyrrolysyl-
tRNA synthetase (PylRS)–tRNAPyl (pylT) pairs have enabled
the incorporation of more than 30 NAAs into proteins in E.
coli.[2] Although the incorporation of these NAAs, which have
unique chemical properties and reactivities, into proteins has
dramatically increased our ability to manipulate protein
structure and function, there still exists one major limitation
for the technique. Namely, the technique in general only
allows the incorporation of a single NAA into a single protein
because the amber codon is the only one available for the
incorporation of NAAs and the nonsense suppression rate in
living cells is low. We have previously demonstrated the
incorporation of up to three Ne-acetyl-l-lysines (AcKs) at
amber mutation sites of GFPUV in E. coli under an enhanced
amber suppression condition.[3] However, to incorporate two
different NAAs into a single protein, the use of an additional
blank codon is required. Herein, we show that the PylRS–
pylT pair can be mutated to suppress the ochre UAA codon
and the mutated pair can be coupled with an evolved
Tyr
CUA
MjTyrRS–MjtRNA
pair to efficiently incorporate two
different NAAs at two defined sites of a single protein in
E. coli by both amber and ochre suppressions. This develop-
ment greatly expands the diversity of modifications we can
introduce to proteins.
[*] Dr. W. Wan,[+] Y. Huang,[+] Dr. Z. Wang, Dr. W. K. Russell, Dr. P.-J. Pai,
Prof. D. H. Russell, Prof. W. R. Liu
Department of Chemistry, Texas A&M University
College Station, TX 77843 (USA)
Fax: (+1)979-845-4719
E-mail: wliu@chem.tamu.edu
Dr. W. Wan[+]
Department of Biology, Texas A&M University
College Station, TX 77843 (USA)
[+] These authors contributed equally to this work.
[**] We thank Prof. Peter G. Schultz of The Scripps Research Institute for
providing us with the pEVOL-AzFRS plasmid and Prof. Daniel
Romo, Prof. Kevin Burgess, and Prof. Jiong Yang of Texas A&M
University for using their instruments. This work was supported by
Welch Grant A-1715 and the New Faculty Startup Fund to W.R.L.
from the Chemistry Department of Texas A&M University.
Figure 1. The suppression levels of UAG, UAA, UGA, and UAGA
mutations at position 149 of GFPUV by their corresponding mutant
pylT suppressors. Suppression levels are represented by the relative
fluorescent emission intensity of expressed full-length GFPUV proteins.
Excitation wavelength: 385 nm, observation wavelength: 505 nm.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201000465.
Angew. Chem. Int. Ed. 2010, 49, 3211 –3214
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3211

Communications
corresponding codon was introduced to position 149 of
GFPUV in the pAcKRS–pylT–GFP1Amber plasmid, cells
transformed with the plasmid and grown in LB medium
supplemented with 5 mm AcK also exhibited detectable
GFPUV expression for all three mutated pylTs (Figure 1). In
comparison to wild-type pylT, both pylTUCA that suppresses
opal codon and pylTUUA that suppresses ochre codon gave
significantly higher suppression levels. This result indicates
that both the PylRS–pylTUCA pair and the PylRS–pylTUUA
pair can be used to efficiently incorporate NAAs into proteins
at their corresponding suppressed codons. Furthermore, it
might also be feasible to couple the PylRS–pylTUCA (or
pylTUUA
MjtRNA
)
pair together with an evolved MjTyrRS–
Tyr
pair to incorporate two different NAAs into a
CUA
single protein in E. coli by both amber and opal (or ochre)
suppressions. Although the suppression of opal mutation in
GFPUV by pylTUCA led to a higher full-length GFPUV
expression level than in the case with the suppression of
ochre mutation in GFPUV by pylTUUA, we eventually chose to
use the ochre suppressor for NAA incorporation because the
anticodon of an opal suppressor may form a wobble pair with
the UGG codon in mRNA and decrease the fidelity of
tryptophan incorporation.[8]
To demonstrate the utility of the PylRS-pylTUUA pair
Figure 2. a) Structures of four NAAs. b) The expression level of full-
length GFPUV with amber mutation at position 1 and ochre mutation at
position 149 from cells transformed with pEVOL-AzFRS and pPylRS-
pylT-GFP1TAG149TAA under different conditions. All the NAAs were
supplemented into media at 1 mm. nd=not determined. c) ESI-MS
analysis of purified GFPUV(1+4), GFPUV(2+4), and GFPUV(3+4). Only
deconvoluted mass spectra are presented.
Tyr
together with an evolved MjTyrRS–MjtRNA
pair to
CUA
incorporate two different NAAs into a single protein in
E. coli by both amber and ochre suppressions, two plasmids,
pEVOL–AzFRS and pPylRS–pylT–GFP1TAG149TAA
(Supporting Information, Figure S1), were used to transform
E. coli BL21 cells. The pEVOL–AzFRS plasmid contains
Tyr
CUA
genes encoding an optimized MjtRNA
and two copies of
an evolved MjTyrRS (AzFRS) specific for p-azido-l-phenyl-
alanine (4; Figure 2a). This plasmid provides an enhanced
amber suppression in E. coli.[9] The pPylRS–pylT–
GFP1TAG149TAA plasmid contains genes encoding wild-
type M. mazei PylRS, pylTUUA, and GFPUV. The GFPUV gene
has an amber mutation at position 1, an ochre mutation at
149, an N-terminal Met-Ala leader dipeptide in front of the
amber mutation, and an opal stop codon at the C-terminal
end. Growing the transformed cells in 2YT medium supple-
mented with 1 mm Ne-Boc-l-lysine (1; Figure 2a) and 1 mm 4
afforded full-length GFPUV with a yield of 11 mgLÀ1 (Fig-
ure 2b, lane 1). No cellular toxicity owing to strong amber and
ochre suppressions was observed. Exclusion of either NAA
from the medium led to no detectable full-length GFPUV
expression (Figure 2b, lanes 4,7). The results indicate that
the suppressions of amber and opal mutations are dependent
on the presence of their corresponding NAAs. The ESI-MS of
the purified full-length GFPUV incorporated with 4 at
position 1 and 1 at position 149 (GFPUV(1+4)) confirmed
the expected incorporations (Figure 2c). The detected mass
(28085 Da) agrees within 70 parts per million with the
calculated mass (28083 Da) of full-length GFPUV(1+4) with-
out N-terminal methionine. The cleavage of N-terminal
methionine from expressed GFPUV in E. coli has been
observed in related studies.[3,10] A mass peak (28059 Da)
that is 26 Da smaller than the major peak is probably due to
the decomposition of the azide group in 4 to form the
corresponding amine during ESI-MS analysis, which has been
observed previously.[11] As wild-type PylRS also charges pylT
with Ne-propargyloxycarbonyl-l-lysine[2h] (2; Figure 2a) and
Ne-cyclopentyloxycarbonyl-l-lysine[12] (3; Figure 2a), the
incorporation of either of these two NAAs together with 4
into GFPUV was also tested in cells transformed with pEVOL–
AzFRS and pPylRS–pylT–GFP1TAG149TAA. Growing cells
in 2YT medium supplemented with 1 mm 2 (or 3) and 1 mm 4
afforded full-length GFPUV in good yields (Figure 2b). No
full-length GFPUV expression was detected when only one
NAA was present in the medium. ESI-MS analysis of the
purified proteins confirmed the expected incorporations
(GFPUV incorporated with 4 and 2 (GFPUV(2+4)): 28065 Da
(calculated), 28067 Da (detected); GFPUV incorporated with
4 and 3 (GFPUV(3+4)): 28095 Da (calculated), 28096 Da
(detected)). Similar decomposition of the azide group during
MS analysis was observed in both cases (Figure 2c).
As GFPUV(2+4) contains both an alkyne group and an
azide group, we tested the feasibility of separately labeling
this protein with different fluorescent dyes by performing
click reactions on these two functional groups.[2h,13] The
reaction of GFPUV(2+4) with 3-azido-7-hydroxycoumarin[14]
(5; Figure 3) in the presence of a CuI catalyst[13b] led to a
labeled GFPUV that emitted strong blue fluorescence under
long wavelength UV light (365 nm) after the protein was
denatured and analyzed in a SDS-PAGE gel (Figure 3,
lane 2). The same labeling reaction between wild-type
GFPUV (wtGFPUV) did not give any detectable blue fluores-
cence when excited. As both proteins were denatured prior to
3212
www.angewandte.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 3211 –3214

Angewandte
Chemie
In summary, we have developed a facile system for genetic
incorporation of two different NAAs at two defined sites of a
single protein in E. coli with moderate to high protein
production yields. This technique will greatly expand the
scope of potential applications for the genetic NAA incorpo-
ration approach, and it can be applied to install a FRET pair
to a protein for conformation and dynamics studies, synthe-
size proteins with two different post-translational modifica-
tions for functional analysis, or generate phage-displayed
peptide libraries with the expanded diversity of the displayed
peptides.[16,17]
Figure 3. Labeling wtGFPUV and GFPUV(2+4) with 5 and 6. Top panel:
proteins stained with Coomassie blue in a SDS-PAGE gel; bottom
panel: fluorescent imaging of the same gel under UV light (365 nm).
The image shows real colors captured by a regular camera. The faint
protein band (at top left) was due to the partial precipitation of
wtGFPUV during the labeling reaction.
Experimental Section
The construction of pPylRS-pylT-GFP1TAG149TAA and other
plasmids all followed standard cloning and QuikChange site-directed
mutagenesis procedures using Platinum Pfx (Invitrogen) and Pfu-
Turbo (Stratagene) DNA polymerases. Sequences of all the con-
structed plasmids were verified by DNA sequencing. For details of
plasmid construction, synthesis and characterization of NAAs and
fluorescent dyes, GFPUV expression, ESI-MS analysis and fluorescent
labeling of expressed proteins, please see the Supporting Information.
SDS-PAGE analysis, the endogenous fluorophore of GFPUV
was quenched and did not interfere with the analysis. The
similar click chemistry reaction was also carried out between
GFPUV(2+4) and a propargyl-conjugated fluorescein (6;
Figure 3). The labeled protein emitted strong green fluores-
cence when excited at 365 nm (Figure 3, lane 4). The control
reaction on wtGFPUV again yielded no detectable fluores-
cently labeled protein (Figure 3, lane 3). These experiments
demonstrate that both side-chain functional groups of 2 and 4
are active after their incorporation into proteins and can be
used separately to effect site-specific protein modifications.
As a large excess of 5 or 6 relative to the protein were used
during labeling experiments, the self-coupling reaction
between the azide and alkyne groups from two GFPUV(2+4)
molecules was prevented. No GFPUV(2+4) dimer was
observed after the reactions.
Received: January 26, 2010
Revised: February 19, 2010
Published online: March 25, 2010
Keywords: amber suppression · genetic code expansion ·
.
noncanonical amino acids · protein engineering
[1] R. Further, Protein Sci. 1998, 7, 419.
[2] a) J. Xie, P. G. Schultz, Methods 2005, 36, 227; b) L. Wang, J. Xie,
P. G. Schultz, Annu. Rev. Biophys. Biomol. Struct. 2006, 35, 225;
c) L. Wang, P. G. Schultz, Angew. Chem. 2005, 117, 34; Angew.
Chem. Int. Ed. 2005, 44, 34; d) T. Mukai, T. Kobayashi, N. Hino,
T. Yanagisawa, K. Sakamoto, S. Yokoyama, Biochem. Biophys.
Res. Commun. 2008, 371, 818; e) T. Fekner, X. Li, M. M. Lee,
M. K. Chan, Angew. Chem. 2009, 121, 1661; Angew. Chem. Int.
Ed. 2009, 48, 1633; f) H. Neumann, S. Y. Peak-Chew, J. W. Chin,
Nat. Chem. Biol. 2008, 4, 232; g) T. Yanagisawa, R. Ishii, R.
Fukunaga, T. Kobayashi, K. Sakamoto, S. Yokoyama, Chem.
Biol. 2008, 15, 1187; h) D. P. Nguyen, H. Lusic, H. Neumann,
P. B. Kapadnis, A. Deiters, J. W. Chin, J. Am. Chem. Soc. 2009,
131, 8720.
To generalize the method, we tested the feasibility of
using other evolved PylRS–pylTUUA and MjTyrRS–
Tyr
CUA
MjtRNA
pairs to genetically incorporate their cognate
NAAs into one protein in E. coli. We replaced the PylRS gene
in the pPylRS–pylT–GFP1TAG149TAA plasmid with the
AcKRS gene[2f] and then transformed E. coli BL21 cells with
the modified plasmid together with pEVOL–AzFRS. Grow-
ing the transformed cells in 2YT medium supplemented with
1 mm 4 and 5 mm AcK led to full-length GFPUV expression
with a yield of 6.4 mgLÀ1 (Supporting Information, Fig-
ure S2). Similarly, we also replaced AzFRS in the pEVOL–
AzFRS with an evolved MjTyrRS (sTyrRS)[15] that is specific
for O-sulfo-l-tyrosine (sTyr) and used the resulting plasmid
and pPylRS–pylT–GFP1TAG149TAA to transform E. coli
BL21 cells. Growing cells in 2YT medium supplemented with
1 mm sTyr and 1 mm 2 led to full-length GFPUV expression
with a yield of 0.4 mgLÀ1 (Supporting Information, Fig-
ure S2). The low GFPUV expression yield in this case is most
likely due to the low efficiency of the evolved sTyrRS.[15] In
both cases, no full-length GFPUV was expressed when only
one NAA was provided in the media. These results indicate
[3] Y. Huang, W. K. Russell, W. Wan, P. J. Pai, D. H. Russell, W. Liu,
Mol. Biosyst. 2010, 6, 683.
[4] a) J. Xie, P. G. Schultz, Nat. Rev. Mol. Cell Biol. 2006, 7, 775;
b) X. Li, T. Fekner, J. J. Ottesen, M. K. Chan, Angew. Chem.
2009, 121, 9348; Angew. Chem. Int. Ed. 2009, 48, 9184; c) D. P.
Nguyen, M. M. G. Alai, P. B. Kapadnis, H. Neumann, J. W. Chin,
J. Am. Chem. Soc. 2009, 131, 14194.
[5] A. Ambrogelly, S. Gundllapalli, S. Herring, C. Polycarpo, C.
Frauer, D. Soll, Proc. Natl. Acad. Sci. USA 2007, 104, 3141.
[6] K. Nozawa, P. OꢀDonoghue, S. Gundllapalli, Y. Araiso, R.
Ishitani, T. Umehara, D. Soll, O. Nureki, Nature 2009, 457, 1163.
[7] Y. Huang, W. Wan, W. K. Russell, P. J. Pai, Z. Wang, D. H.
Russell, W. Liu, Bioorg. Med. Chem. lett. 2010, 20, 878.
[8] a) F. H. Crick, J. Mol. Biol. 1966, 19, 548; b) G. Eggertsson, D.
Soll, Microbiol. Rev. 1988, 52, 354.
that our method can be generally applied to combine any two
Tyr
CUA
evolved PylRS–pylTUUA and MjTyrRS–MjtRNA
pairs to
genetically incorporate their cognate NAAs into a single
protein in E. coli.
[9] T. S. Young, I. Ahmad, J. A. Yin, P. G. Schultz, J. Mol. Biol. 2010,
395, 361.
Angew. Chem. Int. Ed. 2010, 49, 3211 –3214
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3213

Communications
[10] P. R. Chen, D. Groff, J. Guo, W. Ou, S. Cellitti, B. H. Geier-
[14] K. Sivakumar, F. Xie, B. M. Cash, S. Long, H. N. Barnhill, Q.
Wang, Org. Lett. 2004, 6, 4603.
[15] C. C. Liu, P. G. Schultz, Nat. Biotechnol. 2006, 24, 1436.
[16] F. Tian, M. L. Tsao, P. G. Schultz, J. Am. Chem. Soc. 2004, 126,
15962.
[17] Note added in proof: While our original submission was under
review, Chin et al. published work that utilized both amber and
quadruple codon suppressions for genetic incorporation of two
noncanonical amino acids into one protein: H. Neumann, K.
Wang, L. Davis, M. Garcia-Alai, J. W. Chin, Nature, 2010, 464,
441.
stanger, P. G. Schultz, Angew. Chem. 2009, 121, 4112; Angew.
Chem. Int. Ed. 2009, 48, 4052.
[11] a) W. Liu, A. Brock, S. Chen, P. G. Schultz, Nat. Methods 2007, 4,
239; b) K. L. Kiick, E. Saxon, D. A. Tirrell, C. R. Bertozzi, Proc.
Natl. Acad. Sci. USA 2002, 99, 19.
[12] C. R. Polycarpo, S. Herring, A. Berube, J. L. Wood, D. Soll, A.
Ambrogelly, FEBS Lett. 2006, 580, 6695.
[13] a) H. C. Kolb, M. G. Finn, K. B. Sharpless, Angew. Chem. 2001,
113, 2056; Angew. Chem. Int. Ed. 2001, 40, 2004; b) S.
Schoffelen, M. H. L. Lambermon, M. B. van Eldijk, J. C. M.
van Hest, Bioconjugate Chem. 2008, 19, 1127.
3214
www.angewandte.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 3211 –3214