Expressed protein ligation for the preparation of fusion proteins with cell penetrating
peptides for endotoxin removal and intracellular delivery
Hao-Hsin Yu, Ikuhiko Nakase, Sílvia Pujals, Hisaaki Hirose, Gen Tanaka, Sayaka Katayama,
Miki Imanishi, Shiroh Futaki⁎
Institute for Chemical Research, Kyoto University, Uji, Kyoto 611-0011, Japan
a b s t r a c ta r t i c l ei n f o
Received 1 December 2009
Received in revised form 29 January 2010
Accepted 3 February 2010
Available online 17 February 2010
Expressed protein ligation (EPL)
Cell penetrating peptide
Expressed protein ligation (EPL) is a useful method for the native chemical ligation of proteins with other
proteins or peptides. This study assessed the practicability of EPL in the preparation of fusion proteins of
enhanced green fluorescent protein (EGFP) with chemically synthesized cell-penetrating peptides (CPPs) for
intracellular delivery. Using intein-mediated purification with an affinity chitin-binding tag (IMPACT)
system, the thioester of EGFP (EGFP-SR) was prepared. Optimization of the ligation of EGFP-SR with arginine
12-mer (R12) produced the fusion protein in high yield. The EPL procedure also allows the preparation of
EGFP-R12 containing a low level of endotoxin (ET), via the satisfactory ET removal of EGFP-SR prior to
ligation with the R12 peptide. Fusion proteins of EGFP with R12 and the D-isomer of R12 prepared by EPL
showed similar levels of cellular uptake compared to the fusion protein directly expressed in Escherichia coli.
© 2010 Elsevier B.V. All rights reserved.
Cell penetrating peptides (CPPs) have been employed as vectors
for the intracellular delivery of various molecules which have
difficulty in entering cells by themselves [1,2]. One of the most
desirable applications of CPPs would be the delivery of bioactive
proteins having therapeutic prospects . However, contamination of
CPP-fusion proteins by the endotoxins (ETs) can be a potential
problem when the fusion proteins are expressed in Escherichia coli
(E. coli) .
ETs (lipopolysaccharides, LPS) are a major component of the outer
response in the immune system. Disruption of bacterial membranes
upon the harvesting of recombinant proteins from E. coli leads to the
release of ETs from the membranes and the contamination of ETs in the
proteins of interest. Misleading interpretation of the results by the
contamination of ETsmay be obtained if usingimmune-associated cells
in the assay [5–7]. When ETs enter the blood circulation, they induce
various undesired physiological responses, such as the release of pro-
inflammatory materials, fever, coagulopathy, septic shock, and even
mortal outcome . Therefore, the preparation of biological products
charged, due to the phosphate groups in lipid A, that is a major
Tat peptide, comprise a major class of CPPs. When these positively
charged peptides are employed as CPPs and CPP-fusion proteins are
expressed in E. coli, this concern of contamination of the ETs would also
Expressed protein ligation (EPL) is a posttranslational splicing of
protein segments that accompanies exclusion of an internal segment
(Intein) followed by spontaneous attachment of a lateral segment .
Intein-mediated EPL has already been employed to form conjugates of
proteins or peptides with proteins, peptides, unnatural amino acids,
fluorophores and dendrimers [12–16]. This methodology is also
promising for the preparation of novel CPP-fusion proteins bearing
CPPs with various chemical structures including those composed of
for the preparation of the CPP–protein conjugates [12–16].
In this report, we assessed the practicability of EPL through the
preparation of fusion proteins of the enhanced green fluorescent
protein (EGFP) with the arginine 12-mer (R12) peptide, a represen-
tative oligoarginine CPP , as a model. Optimization of the ligation
efficiency, procedures to prepare the EGFP-R12 of low ET content, and
the internalization efficiency of the obtained EGFP-R12 were studied.
Biochimica et Biophysica Acta 1798 (2010) 2249–2257
Abbreviations: α-MEM(+), α-minimum essential medium supplemented with 10%
heat-inactivated bovine serum; CPP, cell penetrating peptide; CBD, chitin-binding
domain; CLSM, confocal laser scanning microscopy; E. coli, Escherichia coli; EGFP,
enhanced green fluorescent protein; EPL, expressed protein ligation; ET, endotoxin; EU,
endotoxin unit; FACS, fluorescence-activated cell sorting; IMPACT, Intein mediated
purification with an affinity chitin-binding tag; IPTG, isopropyl-β-D-thiogalactopyrano-
side; LAL, Limulus amebocyte lysate; MALDI-TOF MS, matrix-assisted laser desorption
ionization time of flight mass spectrometry; MESNA, sodium 2-sulfanylethanesulfonate;
PBS, phosphate buffered saline; PMX, polymyxin B
⁎ Corresponding author. Institute for Chemical Research, Kyoto University, Uji, Kyoto
611-0011, Japan. Tel.: +81 774 38 3210; fax: +81 774 32 3038.
E-mail address: email@example.com (S. Futaki).
0005-2736/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
Contents lists available at ScienceDirect
Biochimica et Biophysica Acta
journal homepage: www.elsevier.com/locate/bbamem
2. Materials and methods
the Peptide Institute (Osaka, Japan) and Novabiochem (Läufelfingen,
Switzerland). Other chemicals unless specifically noted were from Wako
Pure Chemicals (Osaka, Japan).
2.2. Synthesis of peptides
CG-R12 [NH2-Cys-Gly-(Arg)12-amide] and CG-dR12 [NH2-Cys-
Gly-D(Arg)12-amide] were synthesized by 9-fluorenylmethyloxycar-
bonyl (Fmoc)-solid-phase peptide synthesis on a Rink amide resin as
previously reported . For the preparation of FITC-CG-R12 [FITC-
GABA-Cys-Gly-(Arg)12-amide] and FITC-R12 [FITC-GABA-(Arg)12-
amide], a γ-aminobutyryl (GABA) residue was employed as a linker
connecting N-terminal FITC moiety to the arginine peptides. Depro-
tection of the peptides and cleavage from the resin were conducted by
the treatment of trifluoroacetic acid/1,2-ethanedithiol (95:5) for 3 h
at room temperature. The synthesized peptides were purified by high
performance liquid chromatography followed by lyophilization. The
fidelity of the synthesized peptides was confirmed by matrix-assisted
laser desorption ionization time of flight mass spectrometry (MALDI-
TOF MS) using a Voyager-DE STR (Applied Biosystems).
2.3. Conjugation of EGFP to synthetic peptide using EPL
2.3.1. Preparation of EGFP-thioester (EGFP-SR)
A fusion protein of EGFP and mini-intein (derived from Mycobacte-
rium Xenopi gyrase A, Mxe GyrA) combined with a chitin-binding
domain (EGFP-Intein-CBD) was prepared by the intein-mediated
purification with an affinity chitin-binding tag (IMPACT) system (New
England Biolab, Ipswich, MA) . The Intein-CBD element permits
the purification of the fusogenic EGFP using a chitin bead column
followed by the on-column cleavage to generate a thioester of EGFP as
follows: DNA encoding EGFP was amplified from pEGFP-N1 using the
forward primer 5′-GCAGTCGACGGTACC-3′ and the reverse primer 5′-
amplified cDNA was inserted into the NcoI and SpeI site on the pTXB3
vector to create an expression vector encoding the EGFP-Intein-CBD
was grown in LB medium (800 ml) and the expression of the EGFP-
Intein-CBD was induced by the treatment with 0.4 mM isopropyl-β-D-
thiogalactopyranoside(IPTG)at20 °Cfor16 h.TheE.coliwasharvested
by centrifugation and lysed in chitin binding buffer (20 mM Tris–HCl
on the chitin bead column to capture the EGFP-Intein-CBD protein. The
beads were washed with 20-bed volumes of the chitin binding buffer
and then incubated with 2.5-bed volumes of the chitin binding buffer
containing100 mMsodium2-sulfanylethanesulfonate (MESNA) at 4 or
20 °C for 24 h to allow on-column cleavage of the EGFP. The generated
EGFP-SR was eluted from the column with the ligation buffer (0.1 M
Tris–HCl containing 500 mM NaCl and 10 mM MESNA, pH 8.5), and
EGFP-SR was concentrated using a Vivaspin 4 centrifugal concentrator
PES (10,000 MWCO) (Sartorius Stedim Biotech, Goettingen, Germany)
at 4 °C. For the analysis of the MESNA induced cleavage efficiency, the
remaining resin (50 µl) after the protein elution was sampled and
subjected to SDS-PAGE using a 12.5% acrylamide running gel. The
purified EGFP-SR was quantified on the basis of the molar extinction
coefficient at 488 nm (55,000 M−1cm−1) .
2.3.2. Ligation of EGFP-SR with CG-R12
To obtain the optimized ligation conditions, EGFP-SR of different
concentrations (10, 25, and 50 µM) and CG-R12 (10 equiv. to EGFP-
SR) was incubated in ligation buffer (0.1 M Tris–HCl containing
500 mM NaCl and 10 mM MESNA, pH 8.5) at various temperatures (4,
20, and 37 °C) for 36 h, and then the ligation efficiency was analyzed
from the band intensity of the SDS-PAGE using NIH Image J. The
obtained EGFP-CG-R12 was purified on a DEAE-Sepharose Fast Flow
column (0.5 ml) using 20 mM Tris–HCl containing 50 mM NaCl (pH
8.0) as the eluent. Dithiothreitol (DTT) was then added to the fraction
of EGFP-CG-R12 to yield a final DTT concentration of 10 mM and the
mixturewas incubated at room temperature for 30 min. EGFP-CG-R12
was then concentrated on a Vivaspin 4 column (10,000 MWCO) and
the buffer was exchanged by the repetitive addition of 3× phosphate
buffered saline (PBS) prior to centrifugation to prevent aggregation of
2.4. Expression and purification of 6-His-tagged EGFPs (H6-EGFPs)
H6-EGFP, H6-EGFP-R12, and H6-EGFP-CG-R12 were prepared
using pET bearing respective H6-EGFP cDNAs. The E. coli BL21 (DE3)
transformed with these plasmids was grown in LB medium (800 ml).
IPTG(0.1 mM) wasemployed for the protein overexpression.The H6-
EGFP proteins were purified by a Ni-NTA resin column (Qiagen,
Hilden, Germany) (1 ml) according to the protocol by the manufac-
turer. Briefly, E. coli cells were lysed in lysis buffer [50 mM NaH2PO4
containing 300 mM NaCl, 10 mM imidazole, and 10 mM 2-mercap-
toethanol (pH 8.0)]. The lysate was mixed with Ni-NTA resin (1 ml
for 800 ml cell culture) at 4 °C for 1 h. The resin slurry was then
transferred to a column and washed with washing buffer
[50 mM NaH2PO4containing 300 mM NaCl, 20 mM imidazole, and
10 mM 2-mercaptoethanol (pH 8.0)]. Finally, bound proteins were
eluted from the column with elution buffer [50 mM NaH2PO4
containing 300 mM NaCl, 250 mM imidazole, and 10 mM 2-mercap-
toethanol (pH 8.0)]. The fraction containing the desired protein was
repetitively concentrated using a Vivaspin 4 column and diluted with
3× PBS to replace the buffer.
2.5. Endotoxin removal from proteins
2.5.1. Ion exchange columns
Resource S (methyl sulfonate) and HiTrap Q HP (quaternary
ammonium) pre-packed columns (GE Healthcare, Amersham, UK)
(column volume, 1 ml each) operated on the ÄKTA Explorer system
(GE Healthcare) were employed for the removal of ET. H6-EGFP-CG-
R12 was diluted in buffer A (20 mM MES containing 100 mM NaCl, pH
6.0) prior to loading ontoResource S. The column was first eluted with
buffer A (10 ml) and then with a gradient of 0–100% buffer B (20 mM
MES containing 2 M NaCl, pH 6.0) over 20 min (flow rate, 1 ml/min).
When HiTrap Q HP was used for the purification of H6-EGFP-CG-R12,
the protein was diluted in the buffer (20 mM Tris–HCl containing
50 mM NaCl, pH 8.0) and then loaded onto the column. The column
was eluted with the same buffer (flow rate, 0.5 ml/min) and the
eluate that contained the desired protein was collected. For the
purification of the EGFP-SR, the protein was diluted in buffer A
(20 mM Tris–HCl, pH 8.0), loaded on a column of HiTrap Q HP, and
eluted with a gradient of buffer B (20 mM Tris–HCl containing 1 M
NaCl, pH 8.0) over 20 min at the flow rate of 0.5 ml/min.
2.5.2. Polymyxin B (PMX) column
Removal of the ET was conducted using a Detoxi-Gel Endotoxin
removal column (1 ml) (Pierce, Rockford, IL) following the manufac-
turer's protocol. The solution of proteins (H6-EGFP and H6-EGFP-CG-
R12 in 50 mM Tris–HCl containing 500 mM NaCl, pH 8.5; EGFP-SR in
50 mM Tris–HCl containing 500 mM NaCl and 10 mM MESNA, pH 8.5)
was applied to the column equilibrated with the same buffer and left
for 1 h at 4 °C. The protein was then eluted from the column by the
same buffer. Fractions containing the desired proteins were concen-
trated using a Vivaspin 4 column prior to the replacement of the
H.-H. Yu et al. / Biochimica et Biophysica Acta 1798 (2010) 2249–2257
assay: application for endotoxin removal from cationic proteins, Anal. Biochem.
259 (1998) 42–47.
 K. Sarter, S. Andre, H. Kaltner, M. Lensch, C. Schulze, V. Urbonaviciute, G. Schett, M.
Herrmann, H.J. Gabius, Detection and chromatographic removal of lipopolysaccha-
ride in preparations of multifunctional galectins, Biochem. Biophys. Res. Commun.
379 (2009) 155–159.
 S. Liu, R. Tobias, S. McClure, G. Styba, Q. Shi, G. Jackowski, Removal of endotoxin
from recombinant protein preparations, Clin. Biochem. 30 (1997) 455–463.
 S. Aubry, F. Burlina, E. Dupont, D. Delaroche, A. Joliot, S. Lavielle, G. Chassaing, S.
Sagan, Cell-surface thiols affect cell entry of disulfide-conjugated peptides, FASEB
J. 23 (2009) 2956–2967.
 I. Nakase, M. Niwa, T. Takeuchi, K. Sonomura, N. Kawabata, Y. Koike, M. Takehashi,
S. Tanaka, K. Ueda, J.C. Simpson, A.T. Jones, Y. Sugiura, S. Futaki, Cellular uptake of
arginine-rich peptides: roles for macropinocytosis and actin rearrangement, Mol.
Ther. 10 (2004) 1011–1022.
 J.S. Wadia, R.V. Stan, S.F. Dowdy, Transducible TAT-HA fusogenic peptide enhances
escape of TAT-fusion proteins after lipid raft macropinocytosis, Nat. Med. 10 (2004)
 I. Nakase, A. Tadokoro, N. Kawabata, T. Takeuchi, H. Katoh, K. Hiramoto, M.
Negishi, M. Nomizu, Y. Sugiura, S. Futaki, Interaction of arginine-rich peptides
with membrane-associated proteoglycans is crucial for induction of actin
organization and macropinocytosis, Biochemistry 46 (2007) 492–501.
H.-H. Yu et al. / Biochimica et Biophysica Acta 1798 (2010) 2249–2257