Simultaneous and site-directed incorporation of an ester linkage and an azide group into a polypeptide by in vitro translation

Article abstract of DOI:10.1039/b909188b

A method is presented by which an azide-containing side chain can be introduced into any internal position of a polypeptide chain by in vitro translation. For this, 2′-deoxy-cytidylyl-(3′→5′)- adenosine was acylated on the 3′(2′)-hydroxyl group of adenosine with 6-azido-2(S)-hydroxyhexanoic acid (AHHA), an α-hydroxy- and ε-azide derivative of l-lysine. The acylated dinucleotide was enzymatically ligated with a tRNA transcript to provide chemically stable E. coli suppressor AHHA-tRNA^Cys(CUA). The esterase 2 gene from Alicyclobacillus acidocaldarius was modified by the amber stop codon (UAG) on position 118. Using AHHA-tRNA^Cys(CUA) in an E. coli in vitro translation/transcription system, the site-directed introduction of an azide group linked to a backbone ester into the esterase polypeptide was achieved. The yield of the synthesized modified protein reached 80% compared to translation of the native esterase. Subsequently, azide coupling with an alkyne-modified oligodeoxynucleotide demonstrated the feasibility of this approach for conjugation of polypeptides.

Full text of DOI:10.1039/b909188b

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Simultaneous and site-directed incorporation of an ester linkage and an azide
group into a polypeptide by in vitro translation†
Martin Humenik, Yiwei Huang, Igor Safronov and Mathias Sprinzl*
Received 11th May 2009, Accepted 17th July 2009
First published as an Advance Article on the web 12th August 2009
DOI: 10.1039/b909188b
A method is presented by which an azide-containing side chain can be introduced into any internal
position of a polypeptide chain by in vitro translation. For this, 2¢-deoxy-cytidylyl-(3¢→5¢)-adenosine
was acylated on the 3¢(2¢)-hydroxyl group of adenosine with 6-azido-2(S)-hydroxyhexanoic acid
(AHHA), an a-hydroxy- and e-azide derivative of L-lysine. The acylated dinucleotide was enzymatically
ligated with a tRNA transcript to provide chemically stable E. coli suppressor AHHA-tRNACys(CUA)
.
The esterase 2 gene from Alicyclobacillus acidocaldarius was modified by the amber stop codon (UAG)
on position 118. Using AHHA-tRNACys(CUA) in an E. coli in vitro translation/transcription system, the
site-directed introduction of an azide group linked to a backbone ester into the esterase polypeptide
was achieved. The yield of the synthesized modified protein reached 80% compared to translation of
the native esterase. Subsequently, azide coupling with an alkyne-modified oligodeoxynucleotide
demonstrated the feasibility of this approach for conjugation of polypeptides.
mutant-containing E. coli IVT mixture showed increased protein
Introduction
production due to decreased competition between RF1 and
the suppressor tRNA.11 However, yields were still less than
20 mg/mL and suppression efficiencies (24–68%) depended on
the incorporated uAA. Purified translation systems, with all
the components required for proper translation, have also been
developed.12 Although this method is especially attractive for
producing proteins with multiple uAAs, achieved yields were
lower than 80 mg/mL. The in vivo approach of engineered tRNA
and ARS pairs was also adopted for cell-free translations with
suppression efficiencies of 30–50%.13-15
Recently, we reported on esterase 2 from thermophilic bacteria
Alicyclobacillus acidocaldarius, which can be produced at up to
200 mg/mL by an IVT system derived from Escherichia coli.16
Deactivation of RF1 by specific antibodies allows translation of
the UAG amber codon by seryl-tRNASer(CUA) in high yield.17
Here, we describe the application of this system for incorpo-
ration of potentially any uAA into polypeptides. As an exam-
ple, suppression of the amber codon with orthogonal suppres-
sor tRNACys(CUA) acylated by a a-hydroxy-e-azido-derivative of
L-lysine (4, Scheme 1) provided an azide group linked to a back-
bone ester of the neoprotein. Combination of these components
leads to an in vitro system that affords an exceptional suppression
efficiency and high fidelity of the neoprotein synthesis.
The nonsense suppression of the amber codon (UAG) by a
suppressor tRNA(CUA) is used in vivo and in vitro for incorporation
of unnatural amino acids (uAAs) into proteins (neoproteins).1,2
Aminoacyl-tRNA synthetase (ARS) variants capable of aminoa-
cylating orthogonal suppressor tRNA(CUA) with uAA are applied
in vivo.3 Many incorporated uAAs carried functional groups,
which can be further modified e.g. with fluorescence labels, biotin
and polyethylene glycols.4 However, the protein concentrations
obtained in this system (5–20 mg/mL) as well as the suppression
efficiencies (25–50%, yield of neoprotein compared to the yield of
a native protein) are low.2,5,6 Several limitation complicate further
improvements of the in vivo system. First, an unnatural amino
acid has to adhere to requirements for cellular uptake (mainly
lipophilic, uncharged residues are favored)7 and cytotoxicity.8
Second, the competition of suppressor tRNA(CUA) with release
factor 1 (RF1) for the amber stop codon negatively influences
the suppression efficiency. Improvements achieved by increasing
tRNA(CUA) gene copy9 or by using an orthogonal ribosome-mRNA
pair,6 were partially successful but a profound manipulation
of involved translation factors can substantially disturb cell
physiology.
In vitro translation (IVT) systems use an orthogonal aminoacyl-
tRNA(CUA) prepared in a combination of organic and enzymatic
synthesis.10 Since the prerequisite of cell viability is irrelevant
in this case, an IVT method offers broader control over the
translation components. Several attempts to improve translation
efficiency have been made. Mild heating of a thermolabile RF1
Results and discussion
L-Lysine
1
was modified in three steps, desamination,18
amine–azide nucleophilic substitution19 and carboxylic acid
activation20 (Scheme 1A). The cyanomethyl ester of 6-azido-2(S)-
hydroxyhexanoic acid (4) was prepared in 52% overall yield.
Dinucleotide 2¢-deoxy-cytidylyl-(3¢→5¢)-adenosine (pdCpA, 5)
was prepared using phosphoramidite chemistry, acylated with
cyanomethyl ester 420,21 (Scheme 1B), purified by HPLC and
characterized by NMR and MALDI-TOF spectroscopy (ESI†).
Laboratorium fu¨r Biochemie, Universita¨t Bayreuth, Universita¨tstrasse
30, 95440 Bayreuth, Germany. E-mail: mathias.sprinzl@uni-bayreuth.de;
Fax: +49-921-55-2066
† Electronic supplementary information (ESI) available: TOF MS ES+
1
spectrum, NMR structural assignment, H NMR spectrum, HH-COSY
NMR spectrum and 31P NMR spectrum of compound 6. See DOI:
10.1039/b909188b
4218 | Org. Biomol. Chem., 2009, 7, 4218–4224
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
Scheme 1 A) Preparation of cyanomethyl ester of a-hydroxy-e-azido-modified L-lysine 4. B) Preparation of suppressor AHHA-tRNACys(CUA) in two
steps: (a) chemical acylation of dinucleotide pdCpA 5 on the 3¢(2¢) position of adenosine using activated ester 4 and (b) enzymatic ligation of acylated
dinucleotide 6 to cytidine 74 of truncated transcript tRNACys(CUA) (C74) using T4 RNA ligase. The position of ligation (C74) as well as the changed
anticodon loop (CUA) are shown in bold letters in the secondary structure of the suppressor tRNA.
The ratio of 3¢-O and 2¢-O esters on the adenosine moiety of 6 was
2.3:1. tRNACys(CUA) charged with 6-azido-2(S)-hydroxyhexanoic
acid (AHHA, 3) was prepared in almost quantitative yield by
enzymatic ligation of the truncated transcript of E. coli suppressor
tRNACys(CUA) (C74) with pdCpA 6 (Scheme 1B, Fig. 1A). The
suppressor AHHA-tRNACys(CUA) possesses several useful attributes
for use in the in vitro translation. First, the ester bond of the
a-hydroxycarboxylic acid attached to dinucleotide 6 or to the
related tRNA is considerably more stable than the corresponding
aminoacyl derivative.22 This simplifies the synthetic protocols and
increases the yield of the acylated suppressor tRNA.20 Second,
endogenous suppressor tRNAs used in E. coli IVT systems
are more active in in vitro protein synthesis, as compared to
exogenous suppressors.23 AHHA-tRNACys(CUA) was derived from
E. coli tRNACys(GCA). Substitutions in the anticodon loop (GCA
for CUA) decrease the cysteinylation efficiency by three orders of
magnitude.24 The crystal structure of the complex between E. coli
tRNACys(GCA) and CysRS revealed that the anticodon represents the
essential recognition element.25 This makes AHHA-tRNACys(CUA)
(Scheme 1B) a suitable orthogonal candidate for an E. coli IVT
system, i.e. it is not recognized by E. coli CysRS.
The formation of an aa-tRNA·EF-Tu·GTP ternary complex is
essential in protein translation since this step controls the entry
of aminoacyl-tRNAs to the ribosomal A-site.26 The complex
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 4218–4224 | 4219
©

View Article Online
Fig. 1 A) Ligation of the truncated transcript tRNACys(CUA) (C74)
with acylated dinucleotide 6; lane 1: tRNA transcript; lane 2: ligation
of truncated tRNA and 6 by T4 RNA ligase affording quantitative
yield of suppressor AHHA-tRNACys(CUA); analyzed by urea-PAGE. B)
EF-Tu·GTP complex formation with studied tRNAs; lane 1: suppressor
AHHA-tRNACys(CUA), lanes 2, 3 and 4: AHHA-tRNACys(CUA) (3 mM) in the
presence of 1, 2, and 3 mM EF-Tu·GTP, respectively, lane 5: EF-Tu, lane
6: native Phe-tRNAPhe (3 mM) and EF-Tu·GTP (3 mM), lane 7: uncharged
tRNACys(CUA) pdCpA, lanes 8, 9 and 10: uncharged tRNACys(CUA) pdCpA
(3 mM) in the presence of 1, 2 and 3 mM EF-Tu·GTP, respectively; analyzed
by Native PAGE.
Fig. 2 A) Substitution of the sense glutamine codon (GAG) with the
amber codon (UAG) in the position 118 of esterase 2 mRNA. B) Structure
of Pro117, Glu118 and His119 in the esterase 2 (data taken from PDB ID
1EVQ). C) Possible structure of the esterase-neoprotein with introduced
AHHA 3 on position 118. The substitution of a peptide bond with an
ester bond in the protein backbone is highlighted by a red circle. The azide
group linked to the backbone ester is depicted in blue. Other colors code for
oxygen (red), carbon (grey), nitrogen (blue) and hydrogen on heteroatoms
(yellow).
neoprotein synthesized under optimized conditions (5 mM) and the
unmodified esterase 2 translated in the presence of RF1 (6.2 mM).
The suppression efficiency reached 80%, producing 160 mg/mL of
active protein.
The ester bond and the azide group can be specifically modified
without side reactions on the rest of the esterase polypeptide.
This orthogonality was exploited for: (a) test of the stability of
the backbone ester under alkaline conditions, (b) availability of
the linked azide group for a copper catalyzed conjugation and (c)
determination of translation fidelity—the ratio between amounts
of the protein with incorporated AHHA 3 and total amount of
full length protein.
An ester bond, representing an isostere of the amide group
in the polypeptides, is prone to hydrolysis under mild basic
conditions, while peptide bonds remain stable.29 We tested this
in experiments presented in Fig. 3A. The in vitro synthesis of
native esterase and esterase-neoprotein was performed in the
presence of [14C]Leu. Native esterase 2 (Fig. 3A, lanes 1–3) and
esterase-neoprotein (lanes 4–6) were incubated under alkaline
conditions in Tris/HCl buffer, pH 9.5 or 0.1 M NaOH. Only
neoprotein with ester bond-118 was hydrolyzed, under formation
of a 192 amino acid long polypeptide with apparent mass of
20 kDa (Fig. 3A, lane 6). However, strong base (0.1 M NaOH)
was required to draw hydrolysis to completion. This unusually
high stability of the backbone ester is, however, consistent with a
previous observation, where branched and sterically demanding
residues at the position preceding the ester bond influence the rate
of ester bond cleavage.29 Pro117 of the modified esterase possesses
such sterical requirements to hinder ester group accessibility. This
enhanced stability of the incorporated backbone ester allows
convenient handling, PAGE analyses and enzymatic assaying of
the neoprotein using common buffers and pH <9.
formation of acyl-tRNAs with EF-Tu·GTP was analyzed by
polyacrylamide gel electrophoresis (PAGE), as shown in Fig. 1B.
The AHHA-tRNACys(CUA) treated with different concentrations of
EF-Tu·GTP (lanes 2–4) formed a complex comparable with that
obtained with Phe-tRNAPhe (lane 6). As a control, nonacylated
tRNACys(CUA)pdCpA, prepared by ligation of dinucleotide 5 with
truncated tRNACys(CUA) (C74), was used (lane 7). No binding of
EF-Tu·GTP to uncharged tRNACys(CUA) (lanes 8–10) was observed.
The activity of the AHHA-tRNACys(CUA) was then tested in an E.
coli in vitro transcription/translation system. For this, a plasmid
pIVEX-Est2_Amb118 with the amber stop codon UAG 118 in-
stead of GAG (Glu118) coding for esterase 2 from Alicyclobacillus
acidocaldarius was prepared (Fig. 2). The amber stop codon,
which is normally recognized by RF1, was assigned for suppressor
AHHA-tRNACys(CUA). For the termination of translation, the opal
codon (UGA), recognized by RF2, was used.
The formerly estimated optimal concentration for suppressor
Ser-tRNASer(CUA), obtained by enzymatic aminoacylation with
endogenous SerRS, was 2.5 mM in the presence of RF1. The
inactivation of RF1 by specific antibody led to full suppression
of the activity of Ser-tRNASer(CUA) and the optimal concentration
was found to be 45 nM.17 Using the same IVT with depleted
RF1 and plasmid pIVEX-Est2_Amb118, addition of 10 mM
AHHA-tRNACys(CUA) led to the highest yield of the esterase-
neoprotein. The difference in the optimal concentration of Ser-
tRNASer(CUA) and AHHA-tRNACys(CUA) can be explained by two
factors. First, orthogonal suppressor AHHA-tRNACys(CUA) cannot
be recycled as native aa-tRNAs are. Second, higher dissociation
constants are observed for interaction between EF-Tu·GTP and
a-hydroxyacyl-tRNAs as compared to a-aminoacyl-tRNAs.27 An
esterase activity test28 was used to compare yields of the esterase-
4220 | Org. Biomol. Chem., 2009, 7, 4218–4224
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
in sodium phosphate buffer (Na-Pi), pH 7.9 (Scheme 2B).32 The
formation of a product with higher apparent molecular mass can
be conveniently visualized by SDS-PAGE. The gel (Fig. 3B) was
first stained for esterase activity33 showing an active enzyme as a
dark red band on a blue background (Coomassie Blue stained
proteins of IVT mixture). Thus both, the esterase-neoprotein
(lane 1) and the esterase-ODN conjugate (lane 2) were enzymati-
cally active. The same gel was analyzed by [14C] autoradiography
(Fig. 3C) to calculate the ratio between conjugation product and
unreacted polypeptide (Fig. 3C, lane 2). The evaluation of the band
intensities revealed 80% conversion of the azide-modified esterase
into esterase-ODN conjugate. The yield of the conjugation could
be assumed to be quantitative since the same part of the neoprotein
also underwent alkaline hydrolysis (Fig. 3A, lane 6). In accord
with these results, it is plausible that the unconjugated and non-
cleavable part of the synthesized protein represents polypeptides
consisting of natural amino acids, probably resulting from frame-
shifts or misincorporation during translation. These results mean
also that the translation fidelity, with which the a-hydroxy e-azido
L-lysine 3 was introduced into the neoprotein by the presented IVT
system, reached 80%.
Fig. 3 SDS-PAGE of in vitro translation products. A) [14C] Autoradio-
graphy of the products after alkaline cleavage; lanes 1–3: native esterase
2; lanes 4–6: esterase-neoprotein; lanes 1 and 4: samples analyzed directly
after in vitro synthesis, lanes 2 and 5: samples analyzed after hydrolysis in
Tris/HCl buffer pH 9.5 at 37 ◦C, for 30 min, lanes 3 and 6: samples analyzed
after hydrolysis in 0.1 N NaOH at 37 ◦C, for 30 min. B) Activity staining
of esterase-neoprotein (lane 1) and esterase-neoprotein conjugated with
5-alkyne-ODN (lane 2); esterase activity staining results in dark red bands
of active enzyme. In addition the gels were stained by Coomassie Blue to
visualize other proteins in the IVT mixture. C) [14C] Autoradiography of
the gel shown in B.
Conclusion
Cycloaddition reaction between alkynes and azides in the
presence of CuI ions do not affect other reactive groups
such as amines, carboxyls and sulfhydryls, commonly present
in biological systems.30,31 Thus, the presence of an azide
group in position 118 of the esterase was used for con-
jugation with an alkyne-oligodeoxynucleotide. The 5¢-alkyne-
oligodeoxynucleotide (ODN) was prepared from a 10 nucleotides
long 5¢-amino-ODN 7 and 4-pentynoic acid succinimidyl ester (8)
(Scheme 2A) as described previously.32 The 5¢-alkyne-ODN 9 was
conjugated with azide modified esterase in the presence of whole
IVT system, CuSO4 and tris(2-carboxyethyl)phosphine (TCEP)
We established a new orthogonal AHHA-tRNACys(CUA) for site-
specific protein modifications in an E. coli in vitro translation
system. Under conditions of RF1 depletion a high yield of in
vitro synthesis of an azide-modified neoprotein was achieved. High
suppression efficiency, together with the general applicability of
the azide product for a click reaction makes the described approach
superior to previously known methods for posttranslational
modification of polypeptides. The presented in vitro synthesis of
neoproteins necessitates the incorporation of an ester bond into
the polypeptide chain, which is, however, sufficiently stable and
does not affect the activity of the neoprotein.
Scheme 2 A) Preparation of 5¢-alkyne modified ODN. B) Conjugation of the azide modified esterase and 5¢-alkyne-ODN using CuI catalyzed
cycloaddition.
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 4218–4224 | 4221
©

View Article Online
=
2120 (N3), 1728 (C O), 1458, 1350, 1262 (C(=O)-OH), 1116
Experimental
(>C-OH), 897, 735, 665.
All reagents were of analytical grade and were purchased from
Sigma-Aldrich (Taufkirchen, Germany). DNase I was from
Peqlab Biotechnologie GmbH (Erlangen, Germany), T4 RNA
ligase, T4 DNA ligase and restriction enzymes were from New
England Biolabs (Frankfurt, Germany). 5¢-Amine modified ODN
was purchased from Biomers (Ulm, Germany). The transcrip-
tion/translation kits were a gift from RiNA (Berlin, Germany).
1H and 31P NMR spectra were recorded on Bruker Avance-360
NMR (Bruker GmbH, Rheinstetten, Germany) in CDCl3 using
tetramethylsilane as internal standard. The autoradiography of
dry gels was performed on the PhosphorImager SI (Molecular
Dynamics, Sunnyvale, USA). The MALDI-TOF measurements
were carried out with a Bruker Reflex III mass spectrometer
(Bremen, Germany).
Cyanomethyl ester of 6-azido-2(S)-hydroxyhexanoic acid (4)
Compound 3 (0.43 g, 2.5 mmol) was coevaporated with dry
benzene and dry toluene and redissolved in dry DMF (3 ml)
under N2. The solution was placed on ice and treated with
chloroacetonitrile (1 mL, 1.2 g, 17.6 mmol), triethylamine (0.4 mL,
0.3 g, 3 mmol). The solution was stirred at room temperature for
3 days, diluted with diethyl ether (20 ml), washed with an aqueous
solution of 0.1 M HCl (60 ml) and a saturated aqueous solution
of NaCl. The aqueous layer was extracted with two portion of
diethyl ether (2 ¥ 20 ml). The combined organic solutions were
washed with saturated aqueous solutions of NaHCO3 (30 ml) and
NaCl. The organic layer was dried over MgSO4, concentrated
under reduced pressure and the residue was purified on a silica
gel column (11 ¥ 3 cm, 80 ml) equilibrated with CH2Cl2. Elution
with diethyl ether (10% in CH2Cl2) was applied to the column
and fractions containing product 4 were combined, concentrated
under reduced pressure and dried over P2O5. Compound 4 (0.45
Preparation of 6-amino-2(S)-hydroxyhexanoic acid (2)
L-Lysine·H2O 1 (3.28 g, 20 mmol) was dissolved in 50 mL of
10% H2SO4, heated to 45 ◦C and an aqueous solution of NaNO2
(5.18 g, 72 mmol, 20 ml) was added dropwise during 2 h and
stirred additionally at 45 ◦C for 7 h. The reaction was stopped
by addition of an aqueous solution of urea (7.0 g, 0.12 mol,
20 ml) The reaction mixture was stirred further for 30 min, diluted
with 140 mL H2O and applied onto an ion exchange column of
1
g) was obtained in 84% yield. H-NMR (270 MHz, CDCl3) dH
1.40–1.90 (m, 6H, -CH2- CH2- CH2-), 2.64 (d, 1H, J = 5.8 Hz,
-CH(OH)), 3.29 (t, 2H, J = 6.5 Hz, -CH2-N3), 4.25–4.35 (m, 1H,-
CH(OH)), 4.03 (s, 2H, -CH2-CN); IR (n (cm-1)): 3450 (O-H), 2912
≡
=
and 2870 (C-H), 2120 (N3 and C N), 1800 (C O), 1460, 1430,
1363, 1257 and 1160 (C(=O)-OR), 1120 (>C-OH), 1048, 1008,
897, 735.
+
Dowex 50W ¥ 8 (Serva, Heidelberg, Germany) in NH4 -form,
(22 ¥ 2 cm, 70 ml). The column was washed with 50 mM NH4HCO3
(100 ml) and product was eluted stepwise with NH4HCO3 (75 mM,
150 mL; 100 mM, 250 mL; 150 mM, 250 ml). Fractions containing
compound 2 were combined, concentrated under reduced pressure
and co-evaporated 3 times with water to remove NH4HCO3.
Product 2 (2.36 g) was obtained in 80% yield. 1H-NMR (270 MHz,
CDCl3) dH 1.30–1.75 (m, 6H, -CH2- CH2- CH2-), 2.95–3.00 (m, 2H,
-CH2-NH2), 3.95–4.00 (m, 1H,-CH(OH)).
Tetrabutylammonium salt of pdCpA dinucleotide 5
Dinucleotide 5 (3090 OD260, 0.134 mmol) in 100 mL of 10 mM
acetic acid was treated with 1.5 mM tetrabutylammonium hydrox-
ide. The solution was lyophilized, the residue was redissolved in
15 mL of water and the pH of the resulting solution was adjusted to
5.5 with 0.1 M tetrabutylammonium hydroxide. The lyophilization
was repeated and the residue was co-evaporated with absolute
ethanol under reduced pressure. The tetrabutylammonium salt of
dinucleotide 5 was dried over P2O5 for 16 hours.
Preparation of 6-azido-2(S)-hydroxyhexanoic acid (3)
A suspension of NaN3 (6.8 g, 95.4 mmol) in 17 mL H2O and 9 mL
of CH2Cl2 was cooled on ice and triflic (Tf) anhydride (7.42 g,
5.6 mL, 26.3 mmol) was added dropwise to a vigorously stirred
suspension. The reaction mixture was stirred at room temperature
for 2 h. The organic layer was separated and the aqueous layer
was extracted with 15 mL of CH2Cl2. The organic solutions
were combined and washed with saturated aqueous Na2CO3. The
resulting solution of TfN3 in CH2Cl2 was added to the mixture of
compound 2 (1.93 g, 13.1 mmol), K2CO3 (2.17 g, 15.7 mmol) and
CuSO4·5H2O (26 mg, 1.1 mmol) in H2O/methanol (100 mL, 2:3).
The reaction mixture was stirred for 18 h and organic solvents were
evaporated under reduced pressure. Residual aqueous solution
was acidified with concentrated HCl to pH 1.7 and water was
removed under reduced pressure. The crude product was purified
on a silica gel column (15 ¥ 5 cm, 300 ml) equilibrated with 30%
ethyl acetate in n-hexane. Stepwise elution with 30% (300 ml), 50%
(400 ml) and 60% (600 ml) ethyl acetate in n-hexane was performed
and fractions containing product 3 were combined, concentrated
under reduced pressure and dried over P2O5, affording 1.59 g
Preparation of acylated dinucleotide pdCpA 6
The tetrabutylammonium salt of dinucleotide 5 (0.022 mmol,
ESI†), dissolved in 0.52 mL of 40 mM solution of tetrabutylam-
monium hydroxide in dry dimethylformamide, was concentrated
under reduced pressure and co-evaporated two times with dry
dimethylformamide. The residue was treated with a solution of
cyanomethyl ester 4 (0.3 mL, 0.24 mmol) in dry dimethylfor-
mamide. The reaction was stirred under a protective atmosphere
of N2 for 5 h. The course of the reaction was followed by
HPLC. The reaction product was isolated by preparative HPLC
affording acylated dinucleotide 6 in 10% overall yield (52 OD260nm
,
2.25 mmol). The product was concentrated under reduced pressure
and the resulting ammonium salt of 6 was lyophilized three times
from 25 mL of 10 mM acetic acid to obtain the H+-form of
compound 6. For mixture of both 3¢- and 2¢-acyl isomers in
ratio 2.3 : 1; calculated: C25H35N11O15P2, MW = 791.56; found:
1
1
(78%) of 3. H-NMR: (270 MHz, CDCl3) dH 1.30–1.95 (m, 6H,
TOF MS ES+, [M+H]+ 792.23; H-NMR (360 MHz, D2O) dH
-CH2- CH2- CH2-), 3.25–3.30 (m, 2H, -CH2-N3), 4.25–4.30 (m,
1H,-CH(OH)); IR (n (cm-1)): 3350 (O-H), 2912 and 2870 (C-H),
(tetrabutylammonium and H2O signals are excluded) 1.20–1.40
(m, 3H), 1.60–1.95 (m, 4H), 2.30–2.40 (m, 1H), 3.13 (t, 0.6H,
4222 | Org. Biomol. Chem., 2009, 7, 4218–4224
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
J = 6,5 Hz), 3.23 (t, 1H, J = 6,5 Hz), 3.88 (s, 2H), 3.95–4.05 (m,
2H), 4.18 (s, 1H), 4.25–4.35 (m, 0.6H), 4.35–4.45 (m, 1.3H), 4.98 (t,
0.7H, J = 6.5 Hz), 5.41–5.44 (m, 0.7H), 5.59 (t, 0.3H, J = 4.5 Hz),
5.95–6.05 (m, 2.7H), 6.18 (d, 0.3H, J = 4.5 Hz), 7.83 (d, 1H, J =
7,5 Hz), 8.17 (s, 1H), 8.38 (s, 0.3H), 8.45 (s, 0.7H); 31P-NMR
(145 MHz, D2O) dp 0.25 (s, 3¢-isomer), 0.94 (s, 2¢-isomer).1H,
13C, HH-COSY, 31P NMR spectra, structural assignment and
MALDI-TOF spectra of compound 6 are presented in the ESI†
(Figs. 1S–5S).
after electrophoresis at 40 mA for 12–15 hours. Transcripts
were eluted from gel slices using 0.3 M NaOAc (pH 5.2).
Concentrations of transcripts were calculated from absorbance
measurements at 260 nm, assuming that 1 absorbance unit (1 cm
pathway) corresponds to 40 mg/mL. Purified truncated transcript
tRNACys(CUA) (C74) (30 mM) was ligated with acylated dinucleotide
6 (300 mM) using 1 U of T4 RNA ligase in 50 mM Tris/HCl,
10 mM MgCl2, 1 mM ATP, 10 mM dithiothreitol, 25 mg/mL
BSA, pH 7.5 at 4 ◦C for 5 h. The aliquots of suppressor AHHA-
tRNACys(CUA) were used directly in in vitro nonsense suppression
experiments.
Analytical HPLC
Column SP 50/10 Nucleosil 100-5 C18 (Macherey-Nagel, Du¨ren,
Germany), buffer A: 50 mM ammonium acetate, pH 4.5, buffer B:
mixture of 50 mM ammonium acetate pH 4.5 and CH3CN in the
ratio of 1:9, flow 1 ml/min, detection at 260 and 280 nm, linear
gradient of buffer B from 0–100% over 30 min. Retention times
of eluted samples: tR = 8.5 min (dinucleotide 5), tR = 11.5 min
(acylated product 6).
Construction of plasmid pIVEX-Est2_Amb118
The esterase gene in pIVEX_Est2 (pEst2) plasmid was prepared
with a UGA stop codon at the end of the esterase gene,
by the overlap extension method. Two separate PCRs were
carried out using T7 promoter primer (5¢-TAATACGACTCAC-
TTATAGGG-3¢) and EstE118x_rev (5¢-CAGGGAACTT-
GTGCTACGGCGCCAGGCGGTAGTC-3¢); EstE118x_for (5¢-
CCGCCTGGCGCCGTAGCACAAGTTCCCTGCCGC-3¢) and
T7 terminator primer (5¢-CTAGTTATTGCTCAGCGGTG-3¢).
The mutated codons are italicized. Position 118 (Glu) of the
esterase amino acid sequence was replaced by RF1 stop codon
(TAG). The PCR fragments were fused in a second PCR using
T7 promoter and T7 terminator primers. The fused PCR product
was digested with NcoI/SacI and ligated into NcoI/SacI digested
pIVEX vector. The ligation mixture was transformed into
Escherichia coli strain XL-1 Blue. The plasmid DNA was isolated
from clones and sequenced. The resulting plasmid pEst2_amb118
was used for in vitro translation.
Preparative HPLC
Column SP 250/10 Nucleosil 100-5 C18 (Macherey-Nagel), buffer
A: 50 mM ammonium acetate, pH 4.5, buffer B: mixture of 50 mM
ammonium acetate pH 4.5 and CH3CN in the ratio of 3:7, flow
1 ml/min, detection at 260 and 280 nm, isocratic elution with
buffer A over 25 min and then linear gradient of buffer B from
0–100% over 80 min. Retention times of eluted samples: tR
59 min (dinucleotide 5), tR = 72 min (acylated product 6).
=
Preparation of suppressor AHHA-tRNACys(CUA)
The plasmid ptCys_amber for in vitro transcription of tRNACys(CUA)
(C74) was constructed by overlap extension PCR. The forward
primers (1) 5¢-GAATTCTAATACGACTCACTATAG-3¢, (2) 5¢-
TAATACGACTCACTATAGGCGCGTTAACAAAGC-3¢, (3)
5¢-GGTTATGTAGCGGATTCTAAATCCGTCTAGTCC-3¢, (4)
5¢ - GGTTCGACTCCGGAACGCGCCTCCAGGTGATCC - 3¢
and the reverse primers (5) 5¢-ATCCGCTACATAACCGCTTTG-
TTAACGCGCCTA-3¢, (6) 5¢-TTCCGGAGTCGAACCGGA-
CTAGACGGATTTAGA-3¢, (7) 5¢-AAGCTTGGATGGAT-
CACCTGGAGGCGCG-3¢ were used for the PCR. The PCR
product was cloned into pGEM-T vector. The resulting plasmid
was sequenced. The correct gene of tRNACys(CUA) (C74) was
then cloned into pUC19 vector via the cleavage sites EcoRI
and HindIII, italicized in the primer sequences. The resulting
plasmid was used for in vitro transcription of tRNACys(CUA)
(C74). The DNA template of in vitro transcription was obtained
by PCR with the forward primer (1) and reverse primer (8)
5¢-GAGGCGCGTTCCGGAGTC-3¢. Run off transcripts of
tRNACys(CUA) (C74) were obtained by incubation of 400 mL
mixtures containing 40 mM Tris/HCl, pH 8.1, 22 mM MgCl2,
1 mM spermidine, 5 mM DTE, 0.01% Triton X-100, 4 mM of ATP,
CTP, GTP and UTP, 15 ng/mL T7 RNA polymerase and DNA
template at 37 ◦C for 3 hours. After transcription, 4 mL DNase I
(1 U/mL) as added to the reaction mixtures which were incubated
1 hour at 37 ◦C to remove DNA template. Reaction mixtures
were purified by preparative 20% Urea-PAGE (40 ¥ 20 ¥ 0.1 cm).
The transcription mixture (400 mL) was loaded onto the gel in a
12 cm wide pocket. Resolution of one nucleotide was obtained
Preparation of azide modified esterase-neoprotein by in vitro
translation
The translation mixture was incubated with 0.5 mg/mL purified
rabbit polyclonal antibody against RF1, obtained as described,34
at 37 ◦C for 5 minutes to deactivate endogenous RF1. The reaction
was started by adding 5 nM of template pIVEX-Est2_Amb118
and 10 mM of AHHA-tRNACys(CUA) and incubated at 37 ◦C for
2 h. For [14C]-labeling, 0.5 mM [14C]leucine (17.3 mCi/mmol) was
added. The reaction mixture was analyzed by 12.5% SDS-PAGE.35
Polyacrylamide gels were first stained for esterase activity33 and
then proteins were fixed with 15% formaldehyde in 60% methanol
and stained with Coomassie Brilliant Blue G-250. Radioactive
probes were identified by PhosphorImager SI.
Alkaline treatment of in vitro synthesized native esterase and
esterase-neoprotein
Tris-HCl buffer (1M, pH 9.5) or NaOH aqueous solution (1M)
were added into IVT mixture (10 mL) of [14C] labeled native esterase
and esterase-neoprotein, respectively, to a final concentration of
100 mM. The mixture was incubated at 37 ◦C for 30 minutes.
The samples were loaded on SDS-PAGE. The gel was stained
with Coomassie Blue and the dried gel was visualized by [14C]
autoradiography.
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 4218–4224 | 4223
©

View Article Online
Conjugation of 5¢-alkyne-ODN with the azide group of
esterase-neoprotein
11 G. F. Short, S. Y. Golovine and S. M. Hecht, Biochemistry, 1999, 38,
8808–8819.
12 Y. Shimizu, T. Kanamori and T. Ueda, Methods, 2005, 36, 299–304.
13 D. Kiga, K. Sakamoto, K. Kodama, T. Kigawa, T. Matsuda, T. Yabuki,
M. Shirouzu, Y. Harada, H. Nakayama, K. Takio, Y. Hasegawa, Y.
Endo, I. Hirao and S. Yokoyama, Proc. Natl. Acad. Sci. U. S. A., 2002,
99, 9715–9720.
14 S. Ohno, M. Matsui, T. Yokogawa, M. Nakamura, T. Hosoya, T.
Hiramatsu, M. Suzuki, N. Hayashi and K. Nishikawa, J. Biochem.,
2007, 141, 335–343.
15 J. Fukunaga, T. Yokogawa, S. Ohno and K. Nishikawa, J. Biochem.,
2006, 139, 689–696.
16 D. E. Agafonov, K. S. Rabe, M. Grote, Y. W. Huang and M. Sprinzl,
FEBS Lett., 2005, 579, 2082–2086.
17 D. E. Agafonov, Y. W. Huang, M. Grote and M. Sprinzl, FEBS Lett.,
2005, 579, 2156–2160.
18 D. S. Karanewsky, M. C. Badia, D. W. Cushman, J. M. Deforrest, T.
Dejneka, M. J. Loots, M. G. Perri, E. W. Petrillo and J. R. Powell,
J. Med. Chem., 1988, 31, 204–212.
An IVT mixture containing modified esterase (5 mM) was pre-
treated with 20 mM EDTA, subjected to buffer exchange on
Microcon filter device YM10 (Millipore, Schwalbach, Germany)
against 0.1 M sodium phosphate buffer, pH 7.9 and eluted with the
same buffer. The IVT mixture (20 mL, 5 mM of neoprotein) was
added to dry 5¢-alkyne ODN (2.0 nmol). The resulting mixture
was treated with 0.5 mL of CuSO4 (final concentration 2 mM) and
0.5 mL of TCEP (final concentration 3 mM). The r◦eaction mixture
was incubated in a sealed Eppendorf tube at 37 C for 6 h. The
prepared neoprotein-ODN conjugate was characterized by 12.5%
SDS-PAGE (Fig. 3B, C).
19 A. J. Link, M. K. S. Vink and D. A. Tirrell, J. Am. Chem. Soc., 2004,
126, 10598–10602.
Acknowledgements
20 P. M. England, H. A. Lester and D. A. Dougherty, Tetrahedron Lett.,
1999, 40, 6189–6192.
21 S. A. Robertson, C. J. Noren, S. J. Anthonycahill, M. C. Griffith and
P. G. Schultz, Nucleic Acids Res., 1989, 17, 9649–9660.
22 D. Hentzen, P. Mandel and J. P. Garel, Biochim. Biophys. Acta, 1972,
281, 228–232.
This work was supported by Deutsche Forschungsgemeinschaft,
DFG Sp 243/12-2. We thank RiNA GmbH for supplying us with
the in vitro transcription/translation kits.
23 S. T. Cload, D. R. Liu, W. A. Froland and P. G. Schultz, Chem. Biol.,
1996, 3, 1033–1038.
24 G. A. Komatsoulis and J. Abelson, Biochemistry, 1993, 32, 7435–7444.
25 S. Hauenstein, C. M. Zhang, Y. M. Hou and J. J. Perona, Nat. Struct.
Mol. Biol., 2004, 11, 1134–1141.
26 M. Sprinzl, Trends Biochem. Sci., 1994, 19, 245–250.
27 K. H. Derwenskus and M. Sprinzl, FEBS Lett., 1983, 151, 143–147.
28 G. Manco, E. Adinolfi, F. M. Pisani, G. Ottolina, G. Carrea and M.
Rossi, Biochem. J., 1998, 332(Pt 1), 203–212.
29 P. M. England, H. A. Lester and D. A. Dougherty, Biochemistry, 1999,
38, 14409–14415.
30 Q. Wang, T. R. Chan, R. Hilgraf, V. V. Fokin, K. B. Sharpless and
M. G. Finn, J. Am. Chem. Soc., 2003, 125, 3192–3193.
31 M. Meldal and C. W. Tornoe, Chem. Rev., 2008, 108, 2952–3015.
32 M. Humenik, Y. Huang, Y. Wang and M. Sprinzl, ChemBioChem, 2007,
8, 1103–1106.
References
1 T. L. Hendrickson, V. de Crecy-Lagard and P. Schimmel, Annu. Rev.
Biochem., 2004, 73, 147–176.
2 A. J. de Graaf, M. Kooijman, W. E. Hennink and E. Mastrobattista,
Bioconjug. Chem., 2009.
3 D. R. Liu, T. J. Magliery, M. Pasternak and P. G. Schultz, Proc. Natl.
Acad. Sci. U. S. A., 1997, 94, 10092–10097.
4 L. Wang and P. G. Schultz, Angew. Chem., Int. Ed., 2005, 44, 34–66.
5 J. M. Xie and P. G. Schultz, Methods, 2005, 36, 227–238.
6 K. H. Wang, H. Neumann, S. Y. Peak-Chew and J. W. Chin, Nat.
Biotechnol., 2007, 25, 770–777.
7 D. A. Dougherty, Curr. Opin. Chem. Biol., 2000, 4, 645–652.
8 B. Wiltschi and N. Budisa, Appl. Microbiol. Biotechnol., 2007, 74, 739–
753.
9 K. Sakamoto, A. Hayashi, A. Sakamoto, D. Kiga, H. Nakayama, A.
Soma, T. Kobayashi, M. Kitabatake, K. Takio, K. Saito, M. Shirouzu,
I. Hirao and S. Yokoyama, Nucleic Acids Res., 2002, 30, 4692–4699.
10 D. Mendel, V. W. Cornish and P. G. Schultz, Annu. Rev. Biophys. Biomol.
Struct., 1995, 24, 435–462.
33 T. B. Higerd and J. Spizizen, J. Bacteriol., 1973, 114, 1184–1192.
34 D. E. Agafonov, K. S. Rabe, M. Grote, C. S. Voertler and M. Sprinzl,
ChemBioChem, 2006, 7, 330–336.
35 U. K. Laemmli, Nature, 1970, 227, 680–685.
4224 | Org. Biomol. Chem., 2009, 7, 4218–4224
This journal is
The Royal Society of Chemistry 2009
©