The genetic incorporation of thirteen novel non-canonical amino acids

Article abstract of DOI:10.1039/c3cc49068h

Thirteen novel non-canonical amino acids were synthesized and tested for suppression of an amber codon using a mutant pyrrolysyl-tRNA synthetase-tRNAPylCUA pair. Suppression was observed with varied efficiencies. One non-canonical amino acid in particular contains an azide that can be applied for site-selective protein labeling. The Royal Society of Chemistry 2014.

Full text of DOI:10.1039/c3cc49068h

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
The genetic incorporation of thirteen novel
non-canonical amino acids†
Cite this: Chem. Commun., 2014,
50, 2673
Alfred Tuley, Yane-Shih Wang, Xinqiang Fang,‡ Yadagiri Kurra,
Received 27th November 2013,
Accepted 9th January 2014
Yohannes H. Rezenom and Wenshe R. Liu*
DOI: 10.1039/c3cc49068h
www.rsc.org/chemcomm
Thirteen novel non-canonical amino acids were synthesized and tested each NCAA. Herein we demonstrate that PylRS(N346A/C348A) has
for suppression of an amber codon using a mutant pyrrolysyl-tRNA an even broader substrate scope than previously reported.
Pyl
CUA
synthetase–tRNA
pair. Suppression was observed with varied effici-
Our previous studies revealed a large active site pocket in
encies. One non-canonical amino acid in particular contains an azide PylRS(N346A/C348A).30 Removal of the N346 side chain amide
that can be applied for site-selective protein labeling.
dismisses the steric clash that prevents the binding of the aromatic
side chain of phenylalanine and the loss of the C348 thiol yields a
Site-selective installation of non-canonical amino acids (NCAAs) at cavernous pocket capable of binding the para- or meta-substituted
an amber codon is an efficient approach to synthesize proteins with phenylalanine described above. Interestingly, although phenyl-
unique functionalities; applications span from basic studies such as alanine derivatives with small para-substituents have shown to be
protein cellular localization and protein–protein interaction analysis, ineffective substrates for PylRS(N346A/C348A), their isomers with
to biotechnological applications such as the synthesis of heat stable meta-substituents act as highly efficient substrates of PylRS(N346A/
enzymes and therapeutic protein manufacturing.1–5 Two aminoacyl- C348A) for their genetic incorporation at amber codons.30,31 In
tRNA synthetase–tRNACUA pairs have been well adapted for the other words, phenylalanine derivatives with para-substituents can
genetic incorporation of NCAAs at amber codons in bacteria. One only be incorporated when they possess large side chains. Upon
Tyr
is the tyrosyl-tRNA synthetase–tRNA
pair that was derived from further inspection, it appears that a majority of the vacancy in the
CUA
Methanocaldococcus jannaschii.6–8 The other is the pyrrolysyl-tRNA active site pocket of PylRS(N346A/C348A) exists near the meta
Pyl
synthetase (PylRS)–tRNA
pair that naturally occurs in some position of phenylalanine. Encouraged by our preliminary work,
CUA
methanogenic archaea.9–12 Due to its broad-spectrum orthogonality we reasoned that PylRS(N346A/C348A) could incorporate phenyl-
from bacteria to human cells and the fact that it can be easily alanine derivatives with more sterically demanding side chains.
engineered to target a large variety of NCAAs, including natural
Our investigation began with the synthesis and genetic incorpora-
tion of four different para-substituted phenylalanine derivatives (1–4
Pyl
amino acids with posttranslational modifications, the PylRS–tRNA
CUA
pair has captivated researchers for the past several years.13–28 in Fig. 1A), each with a unique functionality and steric requirement.
One of our major contributions to the NCAA research field has Synthesis of these derivatives followed the same strategy presented in
been the development of PylRS mutants capable of incorporating one of our previous reports of the N346A/C348A mutant,30 with the
a number of phenylalanine derivatives, which are substantially exception of NCAA 1, which was synthesized using a different
different from the structure of pyrrolysine, the native substrate of approach (see the ESI†). These four NCAAs were then tested for their
PylRS.29–31 More specifically, we have recently shown that a rationally tolerability by PylRS(N346A/C348A). An E. coli BL21(DE3) cell that
designed, N346A/C348A mutant of PylRS (PylRS(N346A/C348A)) is harbours two plasmids, pEVOL-pylT-PylRSN346A/C348A and pET-
capable of incorporating seven para- and twelve meta-substituted pylT-sfGFP2TAG, was employed for the investigation. pEVOL-pylT-
phenylalanine derivatives at amber codons in coordination with PylRSN346A/C348A contains genes coding PylRS(N346A/C348A) and
Pyl 30,31
Pyl
tRNA
.
This broad substrate scope obviates the need to tRNAPCyUlA; pET-pylT-sfGFP2TAG carries a tRNA
coding gene and a
CUA
CUA
undergo the arduous task of discovering a new mutant for non-sequence-optimized superfolder green fluorescent protein (sfGFP)
gene with an amber mutation in position S2 (sfGFP2TAG). The same
Department of Chemistry, Texas A&M University, College Station, TX 77843, USA.
cells were used in the initial test of the recognition of para-substituted
phenylalanine derivatives by PylRS(N346A/C348A).30 Growth in mini-
E-mail: wliu@chem.tamu.edu
† Electronic supplementary information (ESI) available: Synthesis, protein expres-
mal media supplemented with 1 mM IPTG and 0.2% arabinose
sion, protein labeling, and mass spectrometry analysis. See DOI: 10.1039/c3cc49068h
without NCAA afforded a minimal expression level of full-length
‡ Current address: Fujian Institute of Research on the Structure of Matter,
Chinese Academy of Sciences, Fuzhou, Fujian 350002, China.
sfGFP (o0.3 mg Lꢀ1). Addition of any of 1–4 at 2 mM to the medium
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 2673--2675 | 2673

View Article Online
Communication
ChemComm
Pyl
CUA
PylRS(N346A/C48A)–tRNA
pair. The E. coli cells used for these
compounds harboured two plasmids, pEVOL-pylT-PylRSN346A/
C348A and pET-pylT-sfGFPS2TAG0. pET-pylT-sfGFPS2TAG0 con-
tains a sequence-optimized sfGFP with an amber mutation at
its S2 position (sfGFPS2TAG0). In comparison to the sfGFP2TAG
gene in pET-pylT-sfGFP2TAG, sfGFPS2TAG0 has one more
alanine residue in front of the amber mutation. Growing this cell
in minimal media without NCAA yielded a minimal expression level
of full-length sfGFP. However, all ether NCAAs 6–10 (2 mM) in the
medium promoted the synthesis of sfGFP with a designated NCAA
incorporated (Fig. 2A). In comparison to phenylalanine derivatives
with small meta-substituents such as 5, 6–10 apparently have low
incorporation levels. Molecular weights of purified sfGFP variants
determined by ESI-MS agreed well with the theoretical values
corresponding to a designated NCAA at the S2 position and the first
methionine hydrolysed (Fig. 2B). The removal of the first methionine
is due to the insertion of alanine after it. A number of smaller signals
can be observed, but they largely correspond to common metal
adducts; the expected masses were always the major signal. Com-
pounds 8, 9, and 10 have low solubility; when added to the medium
Fig. 1 (A) Structures of 1–4 and their site-specific incorporation into
sfGFP at its S2 position. (B) Deconvoluted ESI-MS spectra of sfGFP variants
incorporated with 1–4. Their theoretical values are 27 832 Da for 1, 27873 Da
for 2, 27 886 Da for 3, and 27856 Da for 4. Satellite signals are largely due to
metal ion adducts (i.e. Li, Na, K).
all promoted full-length sfGFP expression (Fig. 1A). The expression at 2 mM, compound 10 was observed to precipitate after 12 h of
levels for 1–3 are comparable to that for para-propargyloxy- expression. The low sfGFP expression levels for 8, 9 and 10 may be
phenylalanine (7.8 mg Lꢀ1),30 and the electrospray ionization mass partially due to the toxicity of the compounds; indeed, smaller pellet
spectrometry analysis of four purified sfGFP variants displayed mole- sizes are observed for 8 and 9. Although the sfGFP expression level
cular weights that agreed well with the theoretical values corre- for 9 was very low, the purified sfGFP displayed an ESI-MS molecular
sponding to full-length proteins with the first methionine (Fig. 1B).
weight that still matched the theoretical value of sfGFP with 9
The results obtained for 1–4 demonstrate that PylRS(N346A/ incorporated at S2, indicating that a low concentration of 9 was still
C348A) tolerates phenylalanine derivative substrates with rigid and sufficient to observe incorporation of 9 at the amber mutation site.
bulky substituents at the para position. However, ESI-MS data for
Overall, addition of ketone derivatives 12–15 at 2 mM to the
compound 3 also show a small side peak corresponding to the medium promoted high sfGFP expression yields, and longer alkyl
incorporation of phenylalanine, a result we have observed pre- lengths had less of an impact on protein yields in comparison to
viously. The remaining satellite peaks for these compounds corre- the ether series 6–10, though the sfGFP expression levels for 12–15
spond to common metal adducts in ESI-MS. Additionally, results are lower than that for 11 (Fig. 3A). This series of NCAAs are also
obtained for 1 demonstrated that both meta and para positions can readily soluble, with no precipitation observed in the medium after
be occupied without detriment to expression levels. These results, overnight incubation. ESI-MS analysis of the purified sfGFP variants
coupled with our previous endeavours, led us to wonder if phenyl- confirmed high incorporation fidelities of 12–15 at the S2 site.
alanine derivatives with long-chain meta-substituents could serve as
substrates of PylRS(N346A/C348A) for genetic incorporation as well.
To investigate this hypothesis, a series of meta-alkoxy and meta-acyl
phenylalanines with substituent chain lengths of up to six carbons
were synthesized. We chose these specific derivatives because
the parent NCAAs meta-methoxy-phenylalanine and meta-acetyl-
phenylalanine act as efficient substrates for PylRS(N346A/C348A).
The synthesis of meta-alkoxy-phenylalanines was straightforward,
starting with a published route to obtain protected meta-tyrosine, at
which point the intermediate was subjected to various alkyl halides
to afford different derivatives. Acidic deprotection then yielded free
amino acids as racemic chloride salts. The synthesis of meta-acyl-
phenylalanines was more divergent. Alkyl Grignards were added to
a solution of meta-tolunitrile, which afforded acylbenzenes upon
acidic workup. Radical bromination and then displacement with
diethylacetamidomalonate afforded protected meta-acyl-phenyl-
alanines that were deprotected in 6 M HCl to obtain free amino
acids. More detailed synthetic routes can be found in the ESI.†
Fig. 2 (A) Structures of 5–10 and their site-specific incorporation into sfGFP
at its S2 position. (B) Deconvoluted ESI-MS spectra of sfGFP variants incorpo-
With the desired NCAAs in hand, we thenceforth tested
their incorporation efficacies at amber codons using the 27 786 Da for 7, 27800 Da for 8, 27814 Da for 9, and 27828 Da for 10.
rated with 5–10. Their theoretical values are 27 758 Da for 5, 27 772 Da for 6,
2674 | Chem. Commun., 2014, 50, 2673--2675
This journal is ©The Royal Society of Chemistry 2014

View Article Online
ChemComm
Communication
pool can also be applied to generate phage and E. coli displayed
peptide libraries with expanded chemical moieties for drug discovery,
a direction we are actively pursuing at the current stage.
This work was supported in part by the Welch Foundation (grant
A-1715), the National Science Foundation (grant CHEM-1148684),
and the National Institute of Health (grant 1R01CA161158).
Notes and references
1 W. R. Liu, Y. S. Wang and W. Wan, Mol. Biosyst., 2011, 7, 38–47.
2 A. Gautier, A. Deiters and J. W. Chin, J. Am. Chem. Soc., 2011, 133,
2124–2127.
3 M. Zhang, S. Lin, X. Song, J. Liu, Y. Fu, X. Ge, X. Fu, Z. Chang and
P. R. Chen, Nat. Chem. Biol., 2011, 7, 671–677.
4 Y. Tang, G. Ghirlanda, N. Vaidehi, J. Kua, D. T. Mainz, I. W. Goddard,
W. F. DeGrado and D. A. Tirrell, Biochemistry, 2001, 40, 2790–2796.
5 C. C. Liu and P. G. Schultz, Annu. Rev. Biochem., 2010, 79, 413–444.
6 L. Wang, A. Brock, B. Herberich and P. G. Schultz, Science, 2001, 292,
498–500.
7 J. Xie and P. G. Schultz, Nat. Rev. Mol. Cell Biol., 2006, 7, 775–782.
8 J. W. Chin, S. W. Santoro, A. B. Martin, D. S. King, L. Wang and
P. G. Schultz, J. Am. Chem. Soc., 2002, 124, 9026–9027.
9 G. Srinivasan, C. M. James and J. A. Krzycki, Science, 2002, 296,
1459–1462.
Fig. 3 (A) Structures of 11–15 and their site-specific incorporation into
sfGFP at its S2 position. (B) Deconvoluted ESI-MS spectra of sfGFP variants
incorporated with 11–15. Their theoretical values are 27 770 Da for 11,
27 784 Da for 12, 27 798 Da for 13, 27 812 Da for 14, and 27 826 Da for 15.
Compounds 13 and 14 show small signals corresponding to an N-terminal
methionine on sfGFP. Compound 15 has several small signals attributed to
sodium and potassium adducts.
10 S. K. Blight, R. C. Larue, A. Mahapatra, D. G. Longstaff, E. Chang,
G. Zhao, P. T. Kang, K. B. Green-Church, M. K. Chan and J. A. Krzycki,
Nature, 2004, 431, 333–335.
Among all of the novel NCAAs that can be taken by PylRS(N346A/
C348A), 2 has an active azide functionality for a click reaction with an
alkyne32 and 12–15 contain a ketone group that potentially reacts with
a hydroxylamine. Both functionalities can be applied for site-selective
labeling of proteins incorporated with 2 and 12–15. Since labeling of
sfGFP incorporated with 11 with a hydroxylamine dye was demon-
strated previously,31 we chose to demonstrate the selective labeling of
2 using a diarylcyclooctyne dye D1 in this study (Fig. 4). D1 contains a
strained alkyne that undergoes a spontaneous reaction with an
azide.33 Incubating sfGFP incorporated with 2 with D1 overnight led
to an intensely fluorescently labeled protein; however, the same
reaction with sfGFP incorporated with 3 did not yield any fluorescently
labeled final product. This result indicates that genetically incorpo-
rated 2 can be applied to site-specifically introduce biophysical and
biochemical probes to proteins for a large variety of studies.
11 H. Neumann, S. Y. Peak-Chew and J. W. Chin, Nat. Chem. Biol., 2008,
4, 232–234.
12 W. Wan, Y. Huang, Z. Wang, W. K. Russell, P. J. Pai, D. H. Russell
and W. R. Liu, Angew. Chem., Int. Ed., 2010, 49, 3211–3214.
13 S. Greiss and J. W. Chin, J. Am. Chem. Soc., 2011, 133, 14196–14199.
14 S. M. Hancock, R. Uprety, A. Deiters and J. W. Chin, J. Am. Chem.
Soc., 2010, 132, 14819–14824.
15 T. Mukai, T. Kobayashi, N. Hino, T. Yanagisawa, K. Sakamoto and
S. Yokoyama, Biochem. Biophys. Res. Commun., 2008, 371, 818–822.
16 T. Yanagisawa, R. Ishii, R. Fukunaga, T. Kobayashi, K. Sakamoto
and S. Yokoyama, Chem. Biol., 2008, 15, 1187–1197.
17 A. R. Parrish, X. She, Z. Xiang, I. Coin, Z. Shen, S. P. Briggs, A. Dillin
and L. Wang, ACS Chem. Biol., 2012, 7, 1292–1302.
18 P. R. Chen, D. Groff, J. Guo, W. Ou, S. Cellitti, B. H. Geierstanger and
P. G. Schultz, Angew. Chem., Int. Ed., 2009, 48, 4052–4055.
19 C. J. Chou, R. Uprety, L. Davis, J. W. Chin and A. Deiters, Chem. Sci.,
2011, 2, 480–483.
20 Y. S. Wang, B. Wu, Z. Wang, Y. Huang, W. Wan, W. K. Russell, P. J. Pai,
Y. N. Moe, D. H. Russell and W. R. Liu, Mol. Biosyst., 2010, 6, 1557–1560.
21 Y. J. Lee, B. Wu, J. E. Raymond, Y. Zeng, X. Fang, K. L. Wooley and
W. R. Liu, ACS Chem. Biol., 2013, 8, 1664–1670.
In summary, we have shown that thirteen novel NCAAs were
genetically incorporated into protein at the amber codon in E. coli
22 T. Fekner, X. Li, M. M. Lee and M. K. Chan, Angew. Chem., Int. Ed.,
2009, 48, 1633–1635.
Pyl
CUA
using the PylRS(N346A/C348A)–tRNA
pair. This result, coupled
23 X. Li, T. Fekner, J. J. Ottesen and M. K. Chan, Angew. Chem., Int. Ed.,
2009, 48, 9184–9187.
24 T. Umehara, J. Kim, S. Lee, L. T. Guo, D. Soll and H. S. Park, FEBS
Lett., 2012, 586, 729–733.
25 C. R. Polycarpo, S. Herring, A. Berube, J. L. Wood, D. Soll and
A. Ambrogelly, FEBS Lett., 2006, 580, 6695–6700.
26 T. Plass, S. Milles, C. Koehler, C. Schultz and E. A. Lemke, Angew.
Chem., Int. Ed., 2011, 50, 3878–3881.
with our previous findings, shows a surprisingly broad substrate
scope for PylRS(N346A/C348A). Investigations are underway to deter-
mine aspects of the active site pocket of PylRS(N346A/C348A) that
lead to this broad substrate spectrum. The current study has great
implications in understanding amino acid structure tolerance of the
protein translation system. The expanded genetically encoded NCAA
27 T. Plass, S. Milles, C. Koehler, J. Szymanski, R. Mueller, M. Wiessler,
C. Schultz and E. A. Lemke, Angew. Chem., Int. Ed., 2012, 51, 4166–4170.
28 D. P. Nguyen, H. Lusic, H. Neumann, P. B. Kapadnis, A. Deiters and
J. W. Chin, J. Am. Chem. Soc., 2009, 131, 8720–8721.
29 Y. S. Wang, W. K. Russell, Z. Wang, W. Wan, L. E. Dodd, P. J. Pai,
D. H. Russell and W. R. Liu, Mol. Biosyst., 2011, 7, 714–717.
30 Y. S. Wang, X. Fang, A. L. Wallace, B. Wu and W. R. Liu, J. Am.
Chem. Soc., 2012, 134, 2950–2953.
31 Y. S. Wang, X. Fang, H. Y. Chen, B. Wu, Z. U. Wang, C. Hilty and
W. R. Liu, ACS Chem. Biol., 2013, 8, 405–415.
32 H. C. Kolb, M. G. Finn and K. B. Sharpless, Angew. Chem., Int. Ed.,
2001, 40, 2004–2021.
33 J. C. Jewett, E. M. Sletten and C. R. Bertozzi, J. Am. Chem. Soc., 2010,
132, 3688–3690.
Fig. 4 Labeling of sfGFP incorporated with 2 (sfGFP-2) and sfGFP incorporated
with 3 (sfGFP-3) with dye D1. The top panel shows the Coomassie blue stained
SDS-PAGE gel and the bottom panel shows the fluorescent image of the same
gel under UV irradiation before the gel was stained with Coomassie blue.
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 2673--2675 | 2675