Journal of Bioscience and Bioengineering

VOL. xx No. xx, 1e4, 2014

www.elsevier.com/locate/jbiosc

Synthesis of milligram quantities of proteins using a reconstituted in vitro protein

synthesis system

Yasuaki Kazuta,1 Tomoaki Matsuura,1, 2 Norikazu Ichihashi,1 and Tetsuya Yomo1, 3, 4, *

Japan Science and Technology (JST), ERATO, Dynamical Microscale Reaction Environment Project, 1-5 Yamadaoka, Suita, Osaka 565-0871, Japan,1 Graduate School of Engineering,
Osaka University, 2-1 Yamadaoka, Suita, Osaka 565-0871, Japan,2 Graduate School of Information Science and Technology, Osaka University, 1-5 Yamadaoka, Suita, Osaka 565-0871,
Japan,3 and Graduate School of Frontier Biosciences, Osaka University, 1-5 Yamadaoka, Suita, Osaka 565-0871, Japan4

Received 20 February 2014; accepted 23 April 2014

Available online xxx
In this study, the amount of protein synthesized using an in vitro protein synthesis system composed of only highly
purified components (the PURE system) was optimized. By varying the concentrations of each system component, we
determined the component concentrations that result in the synthesis of 0.38 mg/mL green fluorescent protein (GFP) in
batch mode and 3.8 mg/mL GFP in dialysis mode. In dialysis mode, protein concentrations of 4.3 and 4.4 mg/mL were
synthesized for dihydrofolate reductase and b-galactosidase, respectively. Using the optimized system, the synthesized
protein represented 30% (w/w) of the total protein, which is comparable to the level of overexpressed protein in
Escherichia coli cells. This optimized reconstituted in vitro protein synthesis system may potentially be useful for various
applications, including in vitro directed evolution of proteins, artificial cell assembly, and protein structural studies.
Ó 2014, The Society for Biotechnology, Japan. All rights reserved.

[Key words: In vitro protein synthesis; PURE system; Green fluorescent protein; Dialysis; Synthetic biology]

Cell-free protein synthesis enables in vitro protein synthesis
without the requirement of living cells (1e3). Because the cell-free
system is independent of cell growth, the synthesis of proteins that
may affect cell growth is possible. Furthermore, rapid protein
production is possible using a cell-free system, as protein synthesis
typically requires only a few hours. Because of these favorable
properties, cell-free systems have been used for various applica-
tions and basic research.
Among cell-free systems, the PURE system is the only fully
reconstituted system in which the minimal components for protein
translation are individually purified and reconstituted in vitro (4,5).
Using this reconstituted system, components can be added or
omitted, and the concentration of each component can be altered.
This system has enabled the efficient incorporation of unnatural
amino acids into peptides (6,7). Because of the inherently low
nuclease and protease activity, this system has been used for the in
vitro evolution of proteins and enzymes (8e10). An additional
advantage of the PURE system is the lack of amino acid scrambling,
which enables nuclear magnetic resonance structural studies
(11,12). However, despite its wide application, there have been few
reports on the enhancement of the yield of proteins synthesized
using this reconstituted in vitro protein synthesis system (13,14).
In this study, we report the optimization of the PURE system,
which resulted in the synthesis of greater than 4 mg/mL protein.
This optimization was achieved by modifying the component
* Corresponding author at: Department of Bioinformatic Engineering, Graduate
School of Information Science and Technology, Osaka University, 1-5 Yamadaoka,
Suita, Osaka 565-0871, Japan. Tel.: þ81 6 6879 4171; fax: þ81 6 6879 7433.
E-mail address: yomo@ist.osaka-u.ac.jp (T. Yomo).
concentrations and utilizing the system in dialysis mode. We have
improved the protein yield of this system by including additional
components and altering the component concentrations (15,16).
Attempts to improve the yield of in vitro protein synthesis have
been described (13,14). However, the development of a recon-
stituted in vitro protein synthesis system capable of synthesizing a
protein at concentrations of greater than 1 mg/mL has not yet been
achieved. We evaluated the performance of the optimized system
using three different sets of component concentrations determined
from the optimization process and demonstrate that milligram
quantities of protein can be synthesized.
MATERIALS AND METHODS
DNA and RNA preparation The plasmid DNA used for the in vitro tran-
scription-translation system was purified using the Qiagen Plasmid Mini Kit
(Qiagen) according to the manufacturer’s instructions. Under the control of a T7
promoter, pETG5tag encoding green fluorescent protein (GFP), pET-dhfr (Gene
Frontier) encoding dihydrofolate reductase (DHFR), and pET-bgal encoding b-
galactosidase were used as expression plasmids. For in vitro transcription, pETG5tag
was amplified using the primers T7F (50 -TAATACGACTCACTATAGGG-30 ) and T7R (50 -
GCTAGTTATTGCTCAGCGG-30 ) and KOD FX DNA polymerase (Toyobo, Osaka, Japan).
The PCR product was purified using the QIAquick PCR Purification Kit (Qiagen) ac-
cording to the manufacturer’s instructions. RNA was prepared from this PCR product
using the MEGAscript T7 Kit (Life Technologies) and purified using an RNeasy kit
(Qiagen).
Component preparation Versions 1 and 2 of the system and their compo-
nents were prepared as previously described (5,15). The protein components
constituting version 7 of the system were prepared with additional treatment.
Components, with the exception of the ribosome, were stored in stock buffer
(50 mM HEPES-KOH, pH 7.6, 100 mM KCl, 10 mM MgCl2, 30% glycerol, and 7 mM
2-mercaptoethanol), and the ribosome was stored in 70S buffer (20 mM HEPES-
KOH, pH 7.6, 6 mM Mg(OAc)2, 30 mM KCl, and 7 mM 2-mercaptoethanol). All of
these components were dialyzed against buffer A (50 mM HEPES, pH 7.6, 10 mM

1389-1723/$ e see front matter Ó 2014, The Society for Biotechnology, Japan. All rights reserved.
http://dx.doi.org/10.1016/j.jbiosc.2014.04.019

Please cite this article in press as: Kazuta, Y., et al., Synthesis of milligram quantities of proteins using a reconstituted in vitro protein synthesis
system, J. Biosci. Bioeng., (2014), http://dx.doi.org/10.1016/j.jbiosc.2014.04.019
2 KAZUTA ET AL. J. BIOSCI. BIOENG.,

10 obtain a standard curve of the relationship between the volume of 35S-Met and the
signal intensity obtained in the autoradiogram. Subsequently, an appropriate vol-
ume of 35S-Met (e.g., p ml/mL) was added to the reaction mixture in which a large
excess of cold Met (0.3 mM) was present. After the synthesis reaction, a portion of
8 the sample was subjected to SDS-polyacrylamide gel electrophoresis followed by
PURE ver 1
GFP concentration ( M)

autoradiography. The volume of 35S-Met (e.g., q ml/mL) was determined from the
PURE ver 2 band intensity in the autoradiogram using the standard curve. The concentration of
incorporated Met was calculated using the equation 0.3  (q/p) (mM), and we ob-
6 PURE ver 7
Fluorescent

tained the concentration of the synthesized proteins by dividing this value by the
number of Met residues in the sequence of each synthesized protein.

4 RESULTS AND DISCUSSION

We prepared three versions of the reconstituted in vitro protein

2
0 system was evaluated by synthesizing GFP using 200 nM RNA as a
0 1 2 3 4 template at 37 C (Fig. 1A). Version 7 demonstrated 49-fold higher
Time (h)
FIG. 1. GFP synthesis using various versions of the reconstituted in vitro protein syn-
thesis system. Time course of GFP synthesis in batch mode. RNA was used as a tem-
plate at a concentration of 200 nM. GFP concentrations were determined from
fluorescence intensities using the relationship between the fluorescence and concen-
tration of GFP purified from E. coli. Purified GFP was assumed to be 100% active.
Mg(OAc)2, 100 mM potassium glutamate (K-Glu), and 7 mM 2-mercaptoethanol)
using a microscale dialyzer (Nippon Genetics) with a molecular weight cut-off of
3000 Da at 4 C overnight. When the components precipitated as a result of
dialysis, 3 M K-Glu was added until the pellets were dissolved. All protein
components were concentrated to greater than 10 mg/mL using an Amicon
concentrator (Millipore). Protein concentrations were calculated from their
absorbances at 280 nm. The concentrated proteins were flash frozen in liquid
nitrogen and stored at 80 C. These protein components were subsequently used
to assemble version 7 of the system.
Protein synthesis using the reconstituted in vitro protein synthesis
system For proteins synthesized in batch mode, plasmid DNA was added to
the reaction mixture supplemented with 4 units of RNasin (Promega). For dialysis
reactions, 50 mL of the reaction mixture was placed in a microscale dialyzer (Nippon
Genetics) and dialyzed against 200 mL of buffer, which contained all system com-
ponents except for proteins and tRNAs. All reactions were performed at 37 C. Time
course measurements of GFP synthesis were performed using a temperature-
controlled fluorometer (Mx3005P; Agilent) and a 492/516-nm excitation/emission
filter set.
When necessary, in vitro protein synthesis was performed using 35S-methionine
(35S-Met), and the quantity of synthesized protein was determined from the in-
tensity of the corresponding band in autoradiograms of SDS-polyacrylamide gels.
Briefly, various volumes of radiolabeled 35S-Met were spotted onto a filter paper to
synthesis system (versions 1, 2, and 7), each of which contain
different component concentrations. The performance of each
GFP fluorescence after 3 h compared with that of version 1. A list of
component concentrations is provided in Table S1. Below, we
describe the details of each version of the system.
Version 1 was prepared according to the original study by
Shimizu et al. (4), and version 2 was prepared based on two pre-
vious studies (15,16). HrpA (ATP-dependent RNA helicase) and Tig
(trigger factor) were decided to be included in version 2 of the
system based on a comprehensive analysis using the PURE system
that indicated a beneficial role for these components in GFP syn-
thesis (16). Additionally, the concentrations of 69 system compo-
nents were systematically altered to improve the GFP yield (15),
resulting in the set of concentrations used in version 2. Beginning
with version 2, we first omitted chloride and glycerol from the
system because an absence of these components has been shown to
increase GFP yield (unpublished results). We then altered the
concentration of the ribosome by threefold and increased the
concentration of the other protein components, with the exception
of EF-Tu (34 components in total), by approximately fourfold.
Additionally, peptidyl-tRNA hydrolase (Pth) (17) was included
because it was found to improve the yield by 1.2-fold. As a result of
these modifications, an 8-fold increase in fluorescent GFP yield was
observed for version 7 compared with that of version 2 (Fig. 1). The
translation rate was determined for version 7 (Fig. 2), which syn-
thesized GFP at a rate of 4 amino acid residues/min/RNA. We
postulate that the majority of the improvement is derived from the
improvement in the elongation step. This is because the increase in
the EF-Tu concentration of the version 1 system to that of version 2,

FIG. 2. The rate of protein synthesis using the optimized system. (A) GFP synthesis was performed using different concentrations of RNA, and the initial rate was calculated from the
band intensity in the autoradiogram. All experiments were performed in duplicate. (B) The relationship between RNA concentration and the initial rate. The translation rate was
calculated from the slope of the linear regression of the data, resulting in an elongation rate of 0.96  0.06 GFP molecules/min/RNA. Because GFP is 240 residues long, the rate is
equivalent to 3.84 amino acid residues/s/RNA.

Please cite this article in press as: Kazuta, Y., et al., Synthesis of milligram quantities of proteins using a reconstituted in vitro protein synthesis
system, J. Biosci. Bioeng., (2014), http://dx.doi.org/10.1016/j.jbiosc.2014.04.019
VOL. xx, 2014 OPTIMIZATION OF A RECONSTITUTED CELL-FREE SYSTEM 3

and the increase in the ribosome concentration of the version 2
system to that of version 7 were one of the most effective changes.
Fig. 1 indicates that the batch reaction terminated after a few
hours. The lack of an energy source, such as ATP, hinders long-term
protein synthesis in cell-free systems (18). Dialysis mode, in which
synthesis is performed during dialysis against a buffer containing
an energy source, has been reported to be an effective strategy to
prolong synthesis and to improve the yield of the synthesized
proteins (19,20). Therefore, we utilized dialysis mode in our opti-
mized system (Fig. 3A). The batch reaction yielded 0.38 mg/mL GFP,
whereas the dialysis reaction yielded 3.8 mg/mL GFP after 46 h.
Furthermore, 89% and 69% of the GFP produced was functional
when synthesized in batch and dialysis mode, respectively. We
hypothesize that the lower percentage of functional protein
observed in dialysis mode is due to incomplete GFP folding at 46 h.
Indeed, the amount of functional GFP increases from 30 to 46 h
(Fig. 3A, closed squares). In addition, synthesized GFP was clearly
visible (Fig. 3B). We also evaluated the synthesis of DHFR (18 kDa)
and b-galactosidase (120 kDa) in dialysis mode and obtained final
protein concentrations of 4.3 mg/mL and 4.4 mg/mL, respectively
(Fig. 3C). Bands for these synthesized proteins were clearly
observed on a Coomassie-stained SDS-polyacrylamide gel (Fig. 3D,
Fig. S1). Although cell extract-based, cell-free protein synthesis
systems have been reported to synthesize milligram quantities of
proteins (1e3), this is the first report to achieve these levels using a
reconstituted (synthetic) system.
The PURE system is an Escherichia coli-based reconstituted
system in which protein components, with the exception of the
energy regeneration reaction, are derived from E. coli. Therefore, we
compared the performance of our optimized system to that of an
E. coli cell. The rate of protein synthesis in vivo is 20 amino acid
residues/s (21); in contrast, our system synthesized proteins at a
rate of 4 amino acid residues/s, which is only 1/5 of the in vivo rate.
However, the yield of the optimized in vitro system is comparable to
that of an E. coli cell if the percentage of newly synthesized proteins
is considered. The optimized system consisted of 2.17 mg/mL ri-
bosomal proteins and 7.21 mg/mL additional proteins (e.g., initia-
tion, elongation, and termination factors). This system enabled the
synthesis of at most 4.4 mg/mL protein. Thus, newly synthesized
protein represented 32% [¼4.4/(2.17 þ 7.21 þ 4.4)] of the entire
protein in the system. When a particular protein is overexpressed in
an E. coli cell, newly synthesized proteins represent at most
approximately 30% of the total protein. In this regard, the optimized
system has achieved the level of protein synthesis in an E. coli cell.
The optimized system may be useful for various applications,
including in vitro directed evolution of proteins (22,23), artificial

FIG. 3. Protein synthesis in batch and dialysis modes using DNA as a template. (A) Time course of GFP synthesis in batch and dialysis modes. Total protein concentrations were
calculated from the band intensities in the autoradiogram, and the concentrations of active GFP were determined from the GFP fluorescence, assuming that the GFP synthesized and
purified from E. coli was 100% active. (B) GFP synthesis using the optimized system. 1, no DNA; 2, synthesis in batch mode using version 7 of the system; and 3, synthesis in dialysis
mode using version 7 of the system. (C) Time course of b-galactosidase (b-gal) and DHFR synthesis in dialysis mode. Total concentrations were calculated from the band intensities
in the autoradiogram. (D) Representative data of a Coomassie-stained SDS-polyacrylamide gel. The arrows indicate the bands of the synthesized proteins. The leftmost lane
represents the molecular weight marker. The result of the repeated experiment is shown in Fig. S1. A DNA concentration of 10 ng/mL was used for all experiments.

Please cite this article in press as: Kazuta, Y., et al., Synthesis of milligram quantities of proteins using a reconstituted in vitro protein synthesis
system, J. Biosci. Bioeng., (2014), http://dx.doi.org/10.1016/j.jbiosc.2014.04.019
4 KAZUTA ET AL.
cell assembly (24,25), and protein structural studies (26). Addi-
tional studies, both experimentally and theoretically using math-
ematical models (27,28), on the mechanism for the enhancement of
the rate of synthesis and termination of the reaction may further
increase protein yield, resulting in a system that produces protein
yields that exceed those of E. coli cells.
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.jbiosc.2014.04.019.
ACKNOWLEDGMENTS
We thank Atsuko Uyeda, Hitomi Komai, Tomomi Sakamoto, and
Ryoko Otsuki for their technical assistance. YK and TM performed
the experiments, YK, NI, and TM analyzed the data, and YK, NI, TM
and TY discussed the results. TM organized the project and wrote
the paper.
References
1. Endo, Y. and Sawasaki, T.: Cell-free expression systems for eukaryotic protein
production, Curr. Opin. Biotechnol., 17, 373e380 (2006).
2. Carlson, E. D., Gan, R., Hodgman, C. E., and Jewett, M. C.: Cell-free protein
synthesis: applications come of age, Biotechnol. Adv., 30, 1185e1194 (2012).
3. Schwarz, D., Dotsch, V., and Bernhard, F.: Production of membrane proteins
using cell-free expression systems, Proteomics, 8, 3933e3946 (2008).
4. Shimizu, Y., Inoue, A., Tomari, Y., Suzuki, T., Yokogawa, T., Nishikawa, K.,
and Ueda, T.: Cell-free translation reconstituted with purified components,
Nat. Biotechnol., 19, 751e755 (2001).
5. Shimizu, Y., Kuruma, Y., Ying, B. W., Umekage, S., and Ueda, T.: Cell-free
translation systems for protein engineering, FEBS J., 273, 4133e4140 (2006).
6. Kawakami, T., Murakami, H., and Suga, H.: Ribosomal synthesis of poly-
peptoids and peptoid-peptide hybrids, J. Am. Chem. Soc., 130, 16861e16863
(2008).
7. Seebeck, F. P. and Szostak, J. W.: Ribosomal synthesis of dehydroalanine-
containing peptides, J. Am. Chem. Soc., 128, 7150e7151 (2006).
8. Hipolito, C. J. and Suga, H.: Ribosomal production and in vitro selection of
natural product-like peptidomimetics: the FIT and RaPID systems, Curr. Opin.
Chem. Biol., 16, 196e203 (2012).
9. Fujii, S., Matsuura, T., Sunami, T., Kazuta, Y., and Yomo, T.: In vitro evolution
of alpha-hemolysin using a liposome display, Proc. Natl. Acad. Sci. USA, 110,
16796e16801 (2013).
10. Schlippe, Y. V., Hartman, M. C., Josephson, K., and Szostak, J. W.: In vitro
selection of highly modified cyclic peptides that act as tight binding inhibitors,
J. Am. Chem. Soc., 134, 10469e10477 (2012).
Please cite this article in press as: Kazuta, Y., et al., Synthesis of milligram quantities of proteins using a reconstituted in vitro protein synthesis
system, J. Biosci. Bioeng., (2014), http://dx.doi.org/10.1016/j.jbiosc.2014.04.019
J. BIOSCI. BIOENG.,
11. Sobhanifar, S., Reckel, S., Junge, F., Schwarz, D., Kai, L., Karbyshev, M.,
Lohr, F., Bernhard, F., and Dotsch, V.: Cell-free expression and stable isotope
labelling strategies for membrane proteins, J. Biomol. NMR, 46, 33e43 (2010).
12. Torizawa, T., Shimizu, M., Taoka, M., Miyano, H., and Kainosho, M.: Efficient
production of isotopically labeled proteins by cell-free synthesis: a practical
protocol, J. Biomol. NMR, 30, 311e325 (2004).
13. Hillebrecht, J. R. and Chong, S.: A comparative study of protein synthesis in in
vitro systems: from the prokaryotic reconstituted to the eukaryotic extract-
based, BMC Biotechnol., 8, 58 (2008).
14. Itoh, H., Kawazoe, Y., and Shiba, T.: Enhancement of protein synthesis by an
inorganic polyphosphate in an E. coli cell-free system, J. Microbiol. Methods, 64,
241e249 (2006).
15. Matsuura, T., Kazuta, Y., Aita, T., Adachi, J., and Yomo, T.: Quantifying
epistatic interactions among the components constituting the protein trans-
lation system, Mol. Syst. Biol., 5, 297 (2009).
16. Kazuta, Y., Adachi, J., Matsuura, T., Ono, N., Mori, H., and Yomo, T.:
Comprehensive analysis of the effects of Escherichia coli ORFs on protein
translation reaction, Mol. Cell. Proteomics, 7, 1530e1540 (2008).
17. Das, G. and Varshney, U.: Peptidyl-tRNA hydrolase and its critical role in
protein biosynthesis, Microbiology, 152, 2191e2195 (2006).
18. Kim, H. C. and Kim, D. M.: Methods for energizing cell-free protein synthesis,
J. Biosci. Bioeng., 108, 1e4 (2009).
19. Spirin, A. S., Baranov, V. I., Ryabova, L. A., Ovodov, S. Y., and Alakhov, Y. B.:
A continuous cell-free translation system capable of producing polypeptides in
high yield, Science, 242, 1162e1164 (1988).
20. Aoki, M., Matsuda, T., Tomo, Y., Miyata, Y., Inoue, M., Kigawa, T., and
Yokoyama, S.: Automated system for high-throughput protein production
using the dialysis cell-free method, Protein Expr. Purif., 68, 128e136 (2009).
21. Neidhardt, F. C. and Curtiss, R.: Escherichia coli and Salmonella: cellular and
molecular biology, 2nd ed. ASM Press, Washington, D.C. (1996).
22. Matsuura, T. and Yomo, T.: In vitro evolution of proteins, J. Biosci. Bioeng., 101,
449e456 (2006).
23. Pluckthun, A., Schaffitzel, C., Hanes, J., and Jermutus, L.: In vitro selection
and evolution of proteins, Adv. Protein Chem., 55, 367e403 (2000).
24. Szostak, J. W., Bartel, D. P., and Luisi, P. L.: Synthesizing life, Nature, 409,
387e390 (2001).
25. Ichihashi, N., Usui, K., Kazuta, Y., Sunami, T., Matsuura, T., and Yomo, T.:
Darwinian evolution in a translation-coupled RNA replication system within a
cell-like compartment, Nat. Commun., 4, 2494 (2013).
26. Yokoyama, S.: Protein expression systems for structural genomics and prote-
omics, Curr. Opin. Chem. Biol., 7, 39e43 (2003).
27. Stogbauer, T., Windhager, L., Zimmer, R., and Radler, J. O.: Experiment and
mathematical modeling of gene expression dynamics in a cell-free system,
Integr. Biol. (Camb), 4, 494e501 (2012).
28. Karzbrun, E., Shin, J., Bar-Ziv, R. H., and Noireaux, V.: Coarse-grained dy-
namics of protein synthesis in a cell-free system, Phys. Rev. Lett., 106, 048104
(2011).