A nanocarrier based on a genetically engineered protein cage to deliver doxorubicin to human hepatocellular carcinoma cells

Article abstract of DOI:10.1039/c3cc44508a

Herein, we report the preparation of genetically engineered protein cages (HspG41C-SP94), taken up selectively by human hepatocellular carcinoma (HCC) cells. An engineered protein cage-doxorubicin (DOX) conjugate was as cytotoxic as free DOX against HCC cells but much less cytotoxic against normal hepatocytes.

Full text of DOI:10.1039/c3cc44508a

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
A nanocarrier based on a genetically engineered
protein cage to deliver doxorubicin to human
hepatocellular carcinoma cells†
Cite this: Chem. Commun., 2013,
4
9, 7442
Received 16th June 2013,
Accepted 26th June 2013
ab
acd
d
ac
a
Riki Toita,z Masaharu Murata,*
Kana Abe, Sayoko Narahara, Jing Shu Piao,
b
acd
Jeong-Hun Kang and Makoto Hashizume
DOI: 10.1039/c3cc44508a
www.rsc.org/chemcomm
Herein, we report the preparation of genetically engineered protein and an inner diameter of 6.5 nm through self-organization of
7
cages (HspG41C-SP94), taken up selectively by human hepatocellular 24 subunit proteins. The cage possesses eight pores with a
7
carcinoma (HCC) cells. An engineered protein cage–doxorubicin diameter of 3 nm that link the interior and exterior environments,
(
DOX) conjugate was as cytotoxic as free DOX against HCC cells but enabling encapsulation of drugs and imaging agents in the
6,8
much less cytotoxic against normal hepatocytes.
cage’s interior. Furthermore, the presence of exposed C-terminal
regions on the outer surface of the Hsp cage makes it easy
1
Drug delivery systems using drug carriers such as liposomes,
to introduce peptide ligands through a genetic engineering
2
3
4
7
polymers, micelles, and proteins have been extensively studied approach. Despite the cage’s unique and well-defined structure,
as a means to improve therapeutic effects and reduce undesired biocompatibility, and easy fabrication by a genetic methodology,
side effects of anticancer agents, which exhibit no selectivity its lack of cell specificity makes it difficult to use as a drug
toward cancer cells. Therefore, the specific delivery of therapeutic carrier. We have recently reported the preparation of Hsp cages
9
agents to targeted cells is a fundamental issue, and modification with SP94 peptide ligands bound to their surface via poly-
of drug carriers with peptide ligands that bind to specific ethylene glycol (PEG) linkers. These Hsp cages were taken up by
5
receptors of targeted cells is one of the most attractive strategies.
human hepatocellular carcinoma (HCC) cells but not by other
1
0
However, conjugation of peptide ligands to the surface of drug types of cells such as normal hepatocytes. However, it was
carriers, especially polymeric drug carriers, through covalent difficult to control the densities of SP94 peptides and PEG on
bonds requires complicated multi-step chemical reactions. the cage surface because of the required multi-step chemical
In contrast, naturally occurring proteins can be modified with reactions.
peptide ligands much more easily using a genetic modification
Herein we describe the simple preparation of a novel HCC-
6
strategy. In addition, naturally occurring proteins can have targetable Hsp cage (HspG41C-SP94) by a genetic engineering
many advantages such as biodegradability, low toxicity, and approach involving the addition of SP94 peptides to C-terminal
4
,6
ready availability.
regions exposed on the outer surface of the cage (Scheme 1).
Here, we focused on a naturally occurring small heat shock One of the advantages of a genetic engineering approach is
protein (Hsp 16.5) produced by Methanococcus jannaschii. Hsp the much easier acquisition of completely identical structures
16.5 forms a ‘‘cage’’ structure with an outer diameter of 12 nm and chemical compositions (e.g. the number of peptide
ligands). To assess the cell specificity of the engineered Hsp
a
cages, we prepared fluorophore-labeled HspG41C-SP94 cages
and investigated uptake by five cell types, including HCC
cells and normal hepatocytes, using fluorescence microscopy.
Department of Advanced Medical Initiatives, Faculty of Medical Sciences,
Kyushu University, 3-1-1 Maidashi, Higashi-ku, Fukuoka 812-8582, Japan.
E-mail: m-murata@dem.med.kyushu-u.ac.jp; Fax: +81-92-642-6252;
Tel: +81-92-642-6251
b
c
Department of Biomedical Engineering, National Cerebral and Cardiovascular
Center Research Institutes, 5-7-1 Fujishirodai, Suita, Osaka 565-8565, Japan
Innovation Center for Medical Redox Navigation, Kyushu University,
3
-1-1 Maidashi, Higashi-ku, Fukuoka 812-8582, Japan
d
Center for Advanced Medical Innovation, Kyushu University, 3-1-1 Maidashi,
Higashi-ku, Fukuoka 812-8582, Japan
†
Electronic supplementary information (ESI) available: Experimental details; SEC,
SDS-PAGE, cytotoxicity, and cellular uptake of Hsp cages; synthesis of DOX-EMCH;
and subcellular localization of Hsp-DOX. See DOI: 10.1039/c3cc44508a
Scheme 1 Schematic illustration of HCC-targeted DOX delivery using geneti-
cally engineered Hsp cages. Additional peptide sequences are genetically intro-
duced at the surface of cages. SP94 (HCC-binding peptide): SFSIIHTPILPL; linker
peptide: GSPSG or TSGGSGGSPSG.
‡
Present address: Department of Biomaterials, Faculty of Dental Science,
Kyushu University, 3-1-1 Maidashi, Higashi-ku, Fukuoka 812-8582, Japan.
7
442 Chem. Commun., 2013, 49, 7442--7444
This journal is c The Royal Society of Chemistry 2013

View Article Online
Communication
ChemComm
1
1
Furthermore, doxorubicin (DOX), a well-known anticancer
agent, was conjugated to the interior of HspG41C-SP94 cages
(HspG41C-SP94–DOX). Cytotoxic effects of HspG41C-SP94–DOX,
non-targeted HspG41C–DOX, and free DOX were compared
using HCC cells and normal hepatocytes. The results suggest
that this new HspG41C-SP94 cage will be a useful nanocarrier for
targeted drug delivery to HCC cells and for dramatic reduction of
undesired side effects toward normal hepatocytes.
Three types of engineered Hsp cages, including two HCC-
targetable Hsp cages modified with SP94 peptides via linkers of
different lengths (i.e. HspG41C-SP94(L5) and HspG41C-SP94(L11))
and a control Hsp cage (HspG41C), were prepared using an
E. coli protein expression system (Scheme 1; see ESI† for amino
acid sequences). To enable linkage of fluorophores and drugs,
Gly41, located at the interior of native Hsp cages, was substi-
tuted with Cys through genetic modification. Sodium dodecyl
sulfate polyacrylamide gel electrophoresis (SDS-PAGE) analysis
(
(
Fig. 1(A)) and size-exclusion chromatography (SEC) analysis
Fig. S1, ESI†) showed that all cages were successfully obtained
with sufficient purity. Also, all Hsp cages had a mean diameter
7
of 10–20 nm, which was consistent with previous results,
showing that all genetically engineered Hsp formed cage struc-
tures (Fig. 1(B)).
For cellular uptake experiments, Hsp cages labeled with a Fig. 2 (A) Fluorescence images of cellular uptake of Hsp cages. Green and blue
fluorophore (Alexa488) were prepared by the Michael addition represent Hsp cages and nucleus, respectively. (B and C) Uptake inhibition assay
of fluorophore-labeled HspG41C-SP94(L11) cages with excess unlabeled HspG41C-
reaction of Alexa Fluor 488 C5 maleimide (Invitrogen) and
Cys41 (interior of Hsp cages). Successful fluorophore modifica-
SP94(L11). Data are means ꢁ SEM of three independent experiments. ***, p o 0.001.
tions were confirmed by SDS-PAGE analysis (Fig. S2, ESI†). To
determine whether HspG41C-SP94 cages were selectively taken subsequent experiments. Cellular uptake of HspG41C-SP94(L11)
up by HCC cells, the three types of fluorophore-modified Hsp cages decreased 2- to 3-fold when HCC cells were pretreated
cages were transfected to human HCC cells (Huh-7, HepG2, and with a 100-fold excess of SP94 peptides (Fig. 2(B) and (C)),
Hep3B), normal rat hepatocytes (RLN-8), and human cervical showing that HspG41C-SP94(L11) cages were taken up by HCC
carcinoma (HeLa) cells, and cellular uptake was observed 24 h cells expressing specific receptors for SP94. However, these
later using fluorescence microscopy (Fig. 2(A) and Fig. S3, ESI†). receptors have not yet been identified. Also, both HspG41C-
Non-targeted HspG41C cages were taken up by all cell lines. On SP94(L11) and HspG41C cages taken up were localized in acidic
the other hand, both types of HspG41C-SP94 cages were taken organelles even at 48 h post-transfection (Fig. S4, ESI†).
up by HCC cells but not by RLN-8 cells or HeLa cells. The two Moreover, HspG41C-SP94(L11) had no cytotoxic effect towards
types of HspG41C-SP94 cages showed comparable levels of RLN-8 or Huh-7 cells (Fig. S5, ESI†). These characteristics (HCC
uptake by Huh-7 cells and HepG2 cells. However, in the case targetability and biocompatibility) are important for the appli-
of Hep3B cells, uptake of HspG41C-SP94(L11) cages was greater cation of HspG41C-SP94 cages as nanocarriers for targeted drug
than uptake of HspG41C-SP94(L5) cages (Fig. S3, ESI†). There- delivery to HCC cells.
fore, HspG41C-SP94(L11) cages were investigated further in
We then investigated the use of HspG41C-SP94(L11) cages
for DOX delivery to HCC cells. To this end, DOX-EMCH, a DOX
analogue possessing a maleimide group and an acid-cleavable
12
hydrazone bond, was synthesized according to a previous report
detailed in ESI†). Two types of Hsp cage–DOX conjugates,
(
HspG41C-SP94–DOX and HspG41C–DOX (negative control),
were prepared by simple Michael addition reactions between
the maleimide group of DOX-EMCH and Cys41 of the Hsp cage
(
(
detailed in ESI†). Based on the absorbance of DOX-EMCH
ꢀ1
ꢀ1
8030 cm
M ), HspG41C-SP94–DOX and HspG41C–DOX
possessed comparable numbers of DOX-EMCH molecules
18 and 20 molecules of DOX-EMCH per cage, respectively).
(
After the conjugation reactions, the SEC profiles (Fig. S6A, ESI†)
and size distributions (Fig. S6B, ESI†) of the Hsp cages barely
changed, showing that both types of Hsp cages maintained the cage
structure even after conjugation reactions. Moreover, the release of
Fig. 1 (A) SDS-PAGE analysis of the Hsp cages. Lanes 1, 2, and 3 show HspG41C,
HspG41C-SP94(L5), and HspG41C-SP94(L11), respectively. Protein was stained
with Coomassie brilliant blue. (B) Size distributions of the Hsp cages. The inset
shows a logarithmic scale for the x-axis.
This journal is c The Royal Society of Chemistry 2013
Chem. Commun., 2013, 49, 7442--7444 7443

View Article Online
ChemComm
Communication
Table 1 IC50 values of DOX-conjugated Hsp cages and free DOX towards Huh-7 fluorophore-labeled HspG41C-SP94 cages by HCC cells was
cells and RLN-8 cells
inhibited by excess SP94 peptide, indicating that HspG41C-
SP94 cages were taken up through specific receptors. Further-
IC50/mM
more, selective capacity of a DOX-conjugated HspG41C-SP94
Cells
Time/h HspG41C-SP94–DOX Free DOX HspG41C–DOX
cage (HspG41C-SP94–DOX) to HCC cells dramatically reduced
cytotoxicity towards RLN-8 normal hepatocyte cell lines, but
maintained its cytotoxic effects against Huh-7 HCC cells. The
HspG41C-SP94 cage will be a useful nanocarrier that can
selectively deliver not only anticancer drugs for HCC treatment
but also imaging agents for magnetic resonance imaging,
Huh-7 24
>10
8.7
9.0
2.0
>10
5.0
0.63
0.24
>10
10
3.5
0.49
48
RLN-8 24
48
DOX from HspG41C-SP94–DOX was negligible (0.4%) after 24 h radionuclide imaging, and fluorescence imaging for diagnosis
of incubation at pH 7.2, but increased up to 15% at pH 5.0, of HCC.
indicating that DOX can be released at acidic pH.
This work was supported by a Health Labour Sciences
After transfection of HspG41C-SP94–DOX and HspG41C– Research Grant (Research on Publicly Essential Drugs and Medical
DOX to RLN-8 cells, DOX fluorescence was observed in cells Devices) from the Ministry of Health, Labour and Welfare of
incubated with HspG41C–DOX but not in cells incubated with Japan, a Grant-in-Aid for Scientific Research (No. 21300190 and
HspG41C-SP94–DOX (Fig. S7A, ESI†). In contrast, when the 23650288) from the Ministry of Education, Culture, Sports,
cage–DOX conjugates were added to Huh-7 cells, DOX fluores- Science and Technology in Japan (MEXT) (M. M.), and a
cence was observed at both cytosolic and nuclear regions Grant-in-Aid for Research Activity Start-up (No. 23800045) from
(Fig. S7B and C, ESI†). These results showed that HspG41C- the MEXT (R. T.).
SP94–DOX cages were selectively taken up by HCC cells, which
was consistent with the cellular uptake experiments (Fig. 2(A)). Notes and references
To assess the cytotoxicity of the two types of cage–DOX conju-
1
B. Cˇ eh, M. Wintherhalter, P. M. Frederik, J. J. Vallner and D. D. Lasic,
gates, the conjugates and free DOX were added to Huh-7 cells
and RLN-8 cells, and cell viability was measured 24 and 48 h
later using a CellTiter-Glo Luminescent Cell Viability Assay
Adv. Drug Delivery Rev., 1997, 24, 165.
2 (a) S. Binauld and M. H. Stenzel, Chem. Commun., 2013, 49, 2082;
b) R. Toita, J. H. Kang, T. Tomiyama, C. W. Kim, S. Shiosaki,
(
T. Niidome, T. Mori and Y. Katayama, J. Am. Chem. Soc., 2012,
134, 15410.
(Promega). Also, IC50 values (DOX concentrations that killed
half the cells) were calculated and are summarized in Table 1.
The IC50 value of free DOX after 48 h of treatment was 40 times
lower for RLN-8 cells than for Huh-7 cells, showing that RLN-8
cells were much more sensitive to DOX than to Huh-7 cells.
On the other hand, the IC50 value of HspG41C-SP94–DOX
toward Huh-7 cells was comparable to those of free DOX and
HspG41C–DOX. Generally, DOX–polymer and –protein conju-
gates reported in several previous studies exhibit a tendency to
be much less cytotoxic (5 to 50 times) to target cells than to free
3 Y. Bae and K. Kataoka, Adv. Drug Delivery Rev., 2009, 61, 768.
4
(a) B. Elsadek and F. Kratz, J. Controlled Release, 2011, 157, 4;
b) Y. Ma, R. J. M. Nolte and J. J. Cornelissen, Adv. Drug Delivery
Rev., 2012, 64, 811.
(
5 J. D. Byrne, T. Betancourt and L. Brannon-Peppas, Adv. Drug Delivery
Rev., 2008, 60, 1615.
6
(a) M. Murata, S. Narahara, K. Umezaki, R. Toita, S. Tabata, J. S. Pia,
K. Abe, J. H. Kang, K. Ohuchida, L. Cui and M. Hashizume, Int. J.
Nanomed., 2012, 7, 4353; (b) M. Uchida, H. Kosuge, M. Terashima,
D. A. Willits, L. O. Liepold, M. J. Young, M. V. McConnell and
T. Douglas, ACS Nano, 2011, 5, 2493; (c) K. Sao, M. Murata,
Y. Fujisaki, K. Umezaki, T. Mori, T. Niidome, Y. Katayama and
M. Hashizume, Biochem. Biophys. Res. Commun., 2009, 383, 293;
1
2
DOX. Thus, our data suggest that the HspG41C-SP94–DOX
can selectively recognize target cells, without decreasing its
cytotoxic capacity toward target cells.
(
d) K. Sao, M. Murata, K. Umezaki, Y. Fujisaki, T. Mori, T. Niidome,
Y. Katayama and M. Hashizume, Bioorg. Med. Chem., 2009, 17, 85.
K. K. Kim, R. Kim and S. H. Kim, Nature, 1998, 394, 595.
7
Furthermore, HspG41C-SP94–DOX had a higher IC50 value
than free DOX or HspG41C–DOX after 24 or 48 h of treatment.
These results indicate that HspG41C-SP94–DOX is much less
cytotoxic to RLN-8 cells, primarily because of lower uptake of
HspG41C-SP94–DOX by these cells.
In summary, we have developed an HCC-targetable protein
nanocarrier, HspG41C-SP94, by expressing the HCC-binding
peptide SP94 on the surface of a naturally occurring Hsp cage.
HspG41C-SP94 cages were selectively taken up by various
types of HCC cells but not by normal hepatocytes. Uptake of
8 (a) Z. Varpness, P. A. Suci, D. Ensign, M. J. Young and T. Douglas,
Chem. Commun., 2009, 3726; (b) M. L. Flenniken, L. O. Liepold,
B. E. Crowley, D. A. Willits, M. J. Young and T. Douglas, Chem.
Commun., 2005, 447.
9 A. Lo, C. T. Lin and H. C. Wu, Mol. Cancer. Ther., 2008, 7, 579.
0 R. Toita, M. Murata, S. Tabata, K. Abe, S. Narahara, J. S. Pia,
J. H. Kang and M. Hashizume, Bioconjugate Chem., 2012, 23, 1494.
1 R. Kizek, V. Adam, J. Hrabeta, T. Eckschlager, S. Smutny, J. V. Burda,
E. Frei and M. Stiborova, Pharmacol. Ther., 2012, 133, 26.
2 (a) L. Zhou, R. Cheng, H. Tao, S. Ma, W. Guo, F. Meng, H. Liu, Z. Liu
and Z. Zhong, Biomacromolecules, 2012, 12, 1460; (b) Y. Bae,
N. Nishiyama, S. Fukushima, H. Koyama, Y. Matsumura and
K. Kataoka, Bioconjugate Chem., 2006, 16, 122.
1
1
1
7
444 Chem. Commun., 2013, 49, 7442--7444
This journal is c The Royal Society of Chemistry 2013