Structural development of cell-penetrating peptides containing cationic proline derivatives

Article abstract of DOI:10.1248/cpb.c18-00079

We designed and synthesized a series of cell-penetrating peptides containing cationic proline derivatives (Pro^Gu) that exhibited responsive changes in their secondary structures to the cellular environment. Effects of the peptide length and steric arrangement of the side chain in cationic proline derivatives [Pro^4SGu and Pro^4RGu] on their secondary structures and cell membrane permeability were investigated. Moreover, peptides 3 and 8 exhibited efficient intracellular delivery of plasmid DNA.

Full text of DOI:10.1248/cpb.c18-00079

Vol. 66, No. 5
Chem. Pharm. Bull. 66, 575–580 (2018)
575
Regular Article
Highlighted Paper selected by Editor-in-Chief
Structural Development of Cell-Penetrating Peptides Containing Cationic
Proline Derivatives
,a
Hiroyuki Kobayashi,a,b Takashi Misawa,* Makoto Oba,c Naoya Hirata,d Yasunari Kanda,d
,a
Masakazu Tanaka,c Kenji Matsuno,b and Yosuke Demizu*
a Division of Organic Chemistry, National Institute of Health Sciences; 3–25–26 Tonomachi, Kawasaki-ku, Kawasaki
b
210–9501, Japan: Department of Chemistry and Life Science, Kogakuin University; 2665–1 Nakano, Hachioji, Tokyo
c
192–0015, Japan: Graduate School of Biomedical Sciences, Nagasaki University; 1–14 Bunkyo-machi, Nagasaki
d
852–8521, Japan: and Division of Pharmacology, National Institute of Health Sciences; 3–25–26 Tonomachi,
Kawasaki-ku, Kawasaki 210–9501, Japan.
Received February 1, 2018; accepted February 21, 2018
We designed and synthesized a series of cell-penetrating peptides containing cationic proline derivatives
(ProGu) that exhibited responsive changes in their secondary structures to the cellular environment. Effects
of the peptide length and steric arrangement of the side chain in cationic proline derivatives [Pro4SGu and
Pro4RGu] on their secondary structures and cell membrane permeability were investigated. Moreover, pep-
tides 3 and 8 exhibited efficient intracellular delivery of plasmid DNA.
Key words cell-penetrating peptide; cationic proline derivative; cellular environment; drug delivery system
(DDS) carrier
Cell penetrating peptides (CPPs) are useful tools in deliver- and their cell-penetrating ability were evaluated. The studies
ing hydrophilic cargo molecules such as proteins and nucleic revealed that the peptides 1 and ent-1, which formed a right-
acids into cells. It is well known that human immunodefi- handed (P) and a left-handed (M) helical structure, respec-
ciency virus (HIV)-1 Tat and oligoarginine (Rn), which are tively, showed higher cell-penetrating ability than peptide 2,
representative CPPs contain cationic amino acids.1,2) Recently, which formed a random structure. These results demonstrated
it has been reported that the CPPs exert their cell membrane that formation of a helical structure is favorable to exert the
permeability based on the interaction between the side-chain cell permeability of Arg based CPPs. Furthermore, we re-
guanidino groups of arginine and acidic groups existing on ported that peptides containing cationic cyclic dAA residue
the cell surface.3,4) However, the influences of the second- ApiC2Gu [FAM-βAla-(L-Arg-L-Arg-ApiC2Gu)3-NH2] formed a
ary structure of CPPs on their cell membrane permeability stable helical structure and efficiently delivered plasmid DNA
has not been fully elucidated. To data, several methods to (pDNA) into cells.9) Moreover, we also reported that a series
control peptide secondary structures have been developed. of CPPs containing the cationic proline derivative FAM-βAla-
The introduction of unnatural amino acids, such as α,α- (L-Arg-L-Arg-Pro4SGu)3-NH2 (3, Pro4SGu: a proline derivative
disubstituted amino acids (dAAs) or cyclized β-amino acids containing a guanidine group at the 4-position) efficiently
have been shown to enhance peptide helical structures.5–7) delivered pDNA into cells, with secondary structure of the
To investigate the effects of the secondary structure of CPPs peptide changing in response to the environment.10) Although
on their cell membrane permeability, we designed and syn- peptide 3 formed a random structure in phosphate buffered
thesized novel CPP derivatives containing dAAs residues saline (PBS) buffer, it formed a helical structure in 1% sodium
such as α-amino-isobutyric acid (Aib).8) For example, the dodecyl sulfate (SDS) PBS buffer, which mimiced the envi-
nonapeptides FAM-βAla-(L-Arg-L-Arg-Aib)3-NH2 (1, FAM: ronment as near the cell membrane (Fig. 1).
6-fluorescein), FAM-βAla-(L-Arg-D-Arg-Aib)3-NH2 (2) and
To further investigate the effects of the secondary structure
FAM-βAla-(D-Arg-D-Arg-Aib)3-NH2 (ent-1) were synthesized, on their cell membrane permeability, we designed and synthe-
Fig. 1. Chemical Structures of Homochiral (1, ent-1 and 3), Heterochiral Peptides (2) and Pro4SGu
*To whom correspondence should be addressed. e-mail: misawa@nihs.go.jp; demizu@nihs.go.jp
© 2018 The Pharmaceutical Society of Japan

576
Chem. Pharm. Bull.
Vol. 66, No. 5 (2018)
Fig. 2. Chemical Structures of the Peptides (4–8) Synthesized in This Study
Reagents and conditions: (a) Cbz-Cl, NaHCO3 aq., THF, room temperature (r.t.), 7h, 100% (b) BnBr, TBAI, TEA, THF, r.t., 19h, 83% (c) LiBr, DMEAD, PPh3, THF, r.t.,
24h, 70%; (d) NaN3, DMF, 70°C, 5h, 83% (e) PPh3, H2O, THF, 80°C, 15h, 96% (f) N,N′-Bis(tert-butoxycarbonyl)-1H-pyrazole-1-carboxamide, DIPEA, CH2Cl2, r.t., 15h,
70% (g) 10% Pd-C, H2, MeOH, r.t., 4h, 100% (h) Fmoc-OSu, Na2CO3 aq., 1,4-dioxane, 0°C to r.t., 30min, 72%.
Chart 1
sized a set of peptide 3 derivatives (4, 5), which were com- displayed negative maxima at approximately 208 and 222nm
posed of a different ratio of L-Arg and D-Arg, and shortened in PBS buffer containing 1% SDS (Fig. 3c), whereas D-Arg
peptides (6, 7). It has also been reported that the chirality of based peptide 5 showed positive maxima at approximately
the amino acid side-chain in a peptide sequence can affect its 208 and 222nm (Fig. 3a). These results indicated that L-Arg
secondary structure.11,12) Therefore, we also synthesized pep- based peptides 3 and 7 and D-Arg based peptide 5 possessed
tide 8 containing Pro4RGu and investigated the effects of the a right-handed (P) and a left-handed (M) screw sense, respec-
chirality at the side chain of proline residues on their second- tively.11,14) In addition, these results revealed that the chirality
ary structure and cell-membrane permeability (Fig. 2). More- at the side chain of the proline residues did not affect its sec-
over, to demonstrate the utility of the synthesized peptides, ondary structure (Fig. 3e). Moreover, the shortened peptides 6
the pDNA transfection assay was performed.
and heterochiral peptide 4 formed random structures both in
PBS buffer with and without 1% SDS (Figs. 3a, c). These data
demonstrated that a peptide length longer than six amino acid
Results and Discussion
The Pro4RGu was prepared as previously reported (Chart residues and homochirality were required to change their sec-
1).13) The amino and carboxyl groups of (2S,4R)-hydroxy- ondary structures from a random to helical structure depend-
proline were protected with carboxybenzyl (Cbz) and benzyl ing on the environments.
groups, respectively. The hydroxy group of 9 was converted to
Next, we evaluated the cell membrane permeability of the
the azide group by the Mitsunobu reaction to yield 10. After synthesized peptides against adhesive HeLa cells. The cells
the reduction of azide group by Staudinger reaction, the amino were treated with 1µM synthesized peptides 3–8 at 37°C for
group was guanidinylated. Subsequently, the Cbz and benzyl 2h, and the intracellular uptake was analyzed by flow cytom-
groups were deprotected, and the resulting amino group was etry. Cellular uptakes of the tested peptides were represented
protected with fluorenylmethyloxycarbonyl (Fmoc) group to as the ratio to that of R9 [FAM-βAla-(Arg)9-NH2]. As shown
obtain the target compound Pro4RGu
.
in Fig. 4, LD- and D-based peptides 4 and 5 showed the lower
The designed peptides were synthesized using solid-phase intracellular uptake compared to L-based peptide 3, and the
methods, purified by reverse-phase HPLC, and were charac- shortened peptides 6 and 7 showed mild intracellular uptake.
terized by LC-MS. The preferred secondary structures of syn- These data indicated that the peptide helicity and length
thesized peptides 3–8 were analyzed based on their circular played an important role in permeating the cells. In addition,
dichroism (CD) spectra, shown in Fig. 3 [peptide concentra- peptides 3 and 8 exhibited the same extent of intracellular up-
tion of 100µM in PBS buffer (pH=7.4) with or without 1% take, demonstrating that the chirality of the side chains on the
SDS]. All peptides formed random structures in PBS buffer proline residues did not affect their intracellular uptake.
(Figs. 3b, d, f). However, the homochiral L-peptides 3 and 7
Next, we assessed the internalization mechanisms of pep-

Vol. 66, No. 5 (2018)
Chem. Pharm. Bull.
577
Fig. 3. Preferred Secondary Structures of the Synthesized Peptides in PBS Buffer (pH=7.4) with (a, c, e) or without 1% SDS (b, d, f) by CD Spectra
Analysis
Peptide Concentration: 100µM.
tides 8 by using HeLa cells. It has been previously reported (amiloride,16) an inhibitor of macropinocytosis; nystatin,17) an
that peptide 3 internalized into the cells via macropinocyto- inhibitor of caveolae-mediated endocytosis; and sucrose,18) an
sis.15) Therefore, in order to confirm the internalization path- inhibitor of clathrin-mediated endocytosis), and the intracel-
way of peptide 8, the cells were incubated with the tested lular uptake was measured by flow cytometry (Fig. 5a). As a
peptides in the presence of several endocytosis inhibitors result, the intracellular uptake of peptide 8 was inhibited by

578
Chem. Pharm. Bull.
Vol. 66, No. 5 (2018)
treatment with amiloride, whereas nystatin and sucrose exhib- peptides 3–8 at concentrations of 1, 4, and 8µM. None of the
ited mild effects on intracellular uptake of peptide 8. More- peptides exhibited the significant cytotoxicity.
over, the intracellular distribution of peptides 3, 8, and R9 in
Finally, we performed a pDNA intracellular delivery experi-
HeLa cells was investigated by confocal laser scanning mi- ment using the synthesized peptides 3–8. The peptide-pDNA
croscopy (CLSM). HeLa cells were treated with 10µM peptide complexes were prepared at a 4:1 charge ratio, because the
solution. After incubation for 2h at 37°C, nuclei were stained residual molar ratio of the amino and/or guanidine groups in
with Hoechst 33342. As shown in Fig. 5b, R9 was observed the peptide corresponds to the number of phosphate groups in
as small green spots, whereas peptides 3 and 8 were widely the pDNA. Huh-7 cells were incubated with the peptide-pDNA
distributed in the cytosol. We speculate that the peptide 8 complexes, encoding luciferase, complex and the delivery effi-
internalized into cells via not only macropinocytosis, but also ciency was assessed by chemiluminescence. As shown in Fig.
direct penetration of cell membrane, which is similar to that 7, peptides 3 and 8 exhibited higher transfection efficiency
observed for peptide 3.
than R9, and they exhibited the same extent of intracellular
Next, we performed the cell viability assay using a cell delivery efficiency with jetPEI, which is a commercially avail-
counting kit-8 following the manufacturer’s protocol. It has able transfection reagent. However, peptides 6 and 7 did not
been previously reported that R9 does not exhibit cytotoxic- exhibit intracellular uptake. Although heterochiral peptides
ity against several cell lines.9,10) The results of the cytotoxicity 4 and 5 exhibited a lower intracellular uptake compared to
analysis are shown in Fig. 6. The HeLa cells were treated with homochiral peptides 3 and 8, they exerted the potent intracel-
lular delivery efficiency. It has been reported that the intro-
duction of D-amino acid residues would improve their stabil-
ity against the degrading enzymes.19) Therefore, heterochiral
peptides 4 and 5 may exert continuous pDNA delivery, and
exhibit the same extent of transfection efficacy compared to
homochiral peptides 3 and 8.
Conclusion
In conclusion, we designed and synthesized a series of
homochiral and heterochiral CPP derivatives containing cat-
ionic proline derivatives Pro4SGu and Pro4RGu, and evaluated
their cell membrane permeability against HeLa cells. It was
demonstrated that peptide 8 showed responsive changes in
the secondary structure depending on its environment, with
the same extent of cell membrane permeability as peptide 3.
Moreover, these results revealed that the chirality at the side
Fig. 4. Intracellular Uptake of Synthesized Peptides 3–8 Using Flow
Cytometry
Values are represented as means standard deviation (S.D.) of three independent
chain of proline residues did not affect their secondary struc-
experiments. Vertical axis represents the relative intracellular uptake to that of R9.
Fig. 5. (a) Analysis of the Internalization Mechanism of Peptides 3, 8, and R9 by Treatment with Endocytosis Inhibitors, and (b) Intracellular Distri-
bution of Peptides 3, 8, and R9 in HeLa Cells by CLSM

Vol. 66, No. 5 (2018)
Chem. Pharm. Bull.
579
peptides were synthesized using solid-phase methods on
NovaPEG Rink amide resin following the standard Fmoc
chemistry. The following describes a representative coupling
and deprotection cycle at a 25µmol scale. First, 65mg No-
vaPEG Rink amide resin (loading: 0.5mmol/g) was soaked
for 1h in CH2Cl2. After the resin had been washed with
N,N-dimethylformamide (DMF), Fmoc-amino acid (4eq),
1-[bis(dimethylamino)methylene]-1H-1,2,3-triazolo[4,5-b]-
pyridinium 3-oxide hexafluorophosphate (HATU) (4eq) and
1-hydroxy-7-azabenzotriazole (HOAt) (4eq) dissolved in 2mL
N-methyl-2-pyrrolidone (NMP) were added to the resin. Then,
N,N-diisopropylethylamine (8eq) was added for the coupling
reaction and the resulting mixture was shaken for 1h at room
temperature. Fmoc-deprotection was carried out by 2mL of
20% piperidine in NMP (2mL) for 15min at room tempera-
ture. After the peptide elongation, the resin was suspended
in cleavage cocktail [1.9mL trifluoroacetic acid (TFA), 50µL
water, 50µL triisopropylsilane; final concentration: 95% TFA,
2.5% water, 2.5% triisopropylsilane] for 3h at room tem-
perature. The TFA solution was evaporated to a small volume
under a stream of N2 and dripped into cold ether to precipi-
tate the peptides. The dried crude peptides were dissolved in
1.3mL of 50% acetonitrile in water and then purified by
reversed-phase HPLC using a Discovery® BIO Wide Pore C18
column (25cm×21.2mm). After being purified, the peptide
solutions were lyophilized. Peptide purity was assessed using
analytical HPLC and a Discovery® BIO Wide Pore C18 col-
umn (25cm×4.6mm; solvent A: 0.1% TFA/water, solvent B:
0.1% TFA/MeCN, flow rate: 1.0mL·mL−1, gradient: 10–100%
gradient of solvent B over 30min). The peptides were charac-
terized by LCMS-IT-TOF spectroscopy.
Fig. 6. Cytotoxicity of the Tested Peptides 3–8 against HeLa Cells
Values are represented as means S.D. of three independent experiments.
CD Spectrometry CD spectra were recorded with a
Jasco J-720W spectropolarimeter using a 1.0mm path length
cell. The data are expressed in terms of [θ]; i.e., total molar
ellipticity (degcm2 dmol−1). 20mM phosphate buffer (pH=7.4)
and 1% SDS in 20mM phosphate buffer (pH=7.4) were used
as solvents. Peptide concentration; 100µM.
Fig. 7. Transfection Efficiency of Peptides 3–8 and R9–pDNA Com-
plex in Huh-7 Cells
Values are represented as means S.D. of three independent experiments
(* p<0.05 versus control).
Cell Culture HeLa cells were cultured in Dulbecco’s
modified Eagle medium (DMEM) containing 10% fetal bovine
ture and cell-membrane permeability. Furthermore, peptide 8 serum (FBS). HeLa cells was grown on 10-mm dishes and in-
functioned as a carrier of pDNA. These results indicated that cubated at 37°C under 5% CO2 humidified atmosphere.
the peptides containing cationic proline derivatives could be
Flow Cytometry HeLa cells were seeded in 6-well dishes
a useful tool to deliver the hydrophilic biomolecules into the at a density of 2.0×105 cells/well and cultured in DMEM for
cells.
24h, respectively. The cells were treated with each peptide
(peptide concentration; 1µM) and incubated for 2h. Then, the
cells were washed three times with PBS supplemented with
Experimental
General All coupling reagents were obtained from Wata- heparin (20units/mL) and detached by treatment of trypsin–
nabe Chemical Industries, Ltd. (Japan) and were used as ethylenediaminetetraacetic acid (EDTA). The collected cells
supplied without further purification. Fmoc-protected amino were pelleted by centrifugation at 3000rpm for 5min and
acids were obtained from Tokyo Chemical Industry Co., Ltd. the supernatant was removed. The cells were washed twice
(Japan) and Watanabe Chemical Industries, Ltd. Analytical with PBS buffer. Then, the collected cells were suspended in
TLC was performed on Merck Silica Gel F254. High-resolution 500µL PBS containing 3% FBS and mean fluorescence inten-
mass spectra were recorded with SHIMAZU LCMS-ion trap- sity in cells was measured by flow cytometer.
time-of-flight (LCMS-IT-TOF) spectrometer. The purified
Cytotoxicity HeLa cells were seeded onto 96-well culture
peptides were characterized using LCMS-IT-TOF spectros- plate (5000cells/well) and incubated for 24h in DMEM con-
copy (Shimadzu, Japan). Fluorescence intensity was recorded taining 10% FBS. The medium was replaced and peptide solu-
with FACSCalibur (Becton Dickinson Co., Ltd., U.S.A.) and tion in fresh DMEM was added at each concentration (1, 4,
FACSAria II cell sorter (Becton Dickinson Co., Ltd.).
8µM). After 24h, cell viability was evaluated using cell count-
Synthesis and Purification of Peptides The proline ing kit-8 (DOJINDO, Japan) following to the manufacturer’s
derivatives Fmoc-Pro4SGu-OH and Fmoc-Pro4RGu-OH were protocol. The results are presented as the mean and standard
synthesized following the reported synthetic methods.13) The deviation values obtained from 3 samples.

580
Chem. Pharm. Bull.
Vol. 66, No. 5 (2018)
Inhibition of Endocytosis The cells were seeded onto
Acknowledgments This work was supported in part by
6-well culture plates (400000cells/well) and incubated over- a JSPS KAKENHI Grant number 17k08385 (Y.D.) 15K18905
night in 2mL of DMEM containing 10% FBS. After the me- (T.M), and by a Grant from the Takeda Science Foundation
dium had been replaced with fresh medium containing 10% (Y.D.), a Grant from the Terumo Life Science Foundation
FBS in the absence or presence of amiloride (5mM), sucrose (Y.D.), a Grant from the Suzuken Memorial Foundation, and a
(0.4 M), or nystatin (25µg/mL), the cells were pre-incubated at Grant from the Naito Foundation (Y.D.).
37°C for 30min. Peptide solution was applied to each well at
a concentration of 1µM. After the cells had been incubated for
Conflict of Interest The authors declare no conflict of
1h, the medium was removed, and the cells were washed 3 interest.
times with PBS supplemented with heparin (20units/mL) and
detached by treatment of trypsin–EDTA. Then, fluorescence
Supplementary Materials The online version of this ar-
intensity in the cells was measured as above. The results are ticle contains supplementary materials.
presented as the mean and standard deviation values obtained
from 3 samples.
References
1) Brooks H., Lebleu B., Vivès E., Adv. Drug Deliv. Rev., 57, 559–577
(2005).
Fluorescence Microscope HeLa cells were seeded onto
glass bottom dish (Matsunami, Osaka, Japan) (100000cells/
well) and incubated overnight in 2mL of DMEM containing
10% FBS. The medium was then replaced with fresh medium
containing 10% FBS, and peptide solution was applied to
well at a concentration of 10µM. After the cells had been in-
cubated for 2h, the medium was removed, and the cells were
washed 3 times with ice-cold PBS supplemented with heparin
(20units/mL). The intracellular distribution of the complexes
was observed by CLSM after staining late endosomes/lyso-
somes with LysoTracer Red and nuclei with Hoechst 33342.
The CLSM observations were performed using a BZ-9000
(Keyence, Osaka, Japan) equipped with a 40× objective lens.
Direct Cell Permeability by Flow Cytometry Analysis
at 37°C The cells were seeded onto 6-well culture plates
(200000cells/well) and incubated overnight in 2mL of
DMEM containing 10% FBS. After the medium had been
replaced with fresh medium containing 10% FBS and peptide
solution was applied to each well at a concentration of 1µM.
After the cells had been incubated for 1h at 37°C, the medium
2) Futaki S., Suzuki T., Ohashi W., Yagami T., Tanaka S., Ueda K.,
Sugiura Y., J. Biol. Chem., 276, 5836–5840 (2001).
3) Suzuki T., Futaki S., Niwa M., Tanaka S., Ueda K., Sugiura Y., J.
Biol. Chem., 277, 2437–2443 (2002).
4) Rothbard J. B., Kreider E., VanDeusen C. L., Wright L., Wylie B.
L., Wender P. A., J. Med. Chem., 45, 3612–3618 (2002).
5) Crisma M., Toniolo C., Biopolymers, 104, 46–64 (2015).
6) Gellman S. H., Acc. Chem. Res., 31, 173–180 (1998).
7) Demizu Y., Doi M., Kurihara M., Maruyama T., Suemune H.,
Tanaka M., Chem. Eur. J., 18, 2430–2439 (2012).
8) Yamashita H., Demizu Y., Shoda T., Sato Y., Oba M., Tanaka M.,
Kurihara M., Bioorg. Med. Chem., 22, 2403–2408 (2014).
9) Yamashita H., Oba M., Misawa T., Tanaka M., Hattori T., Naito M.,
Kurihara M., Demizu Y., ChemBioChem, 17, 137–140 (2016).
10) Yamashita H., Kato T., Oba M., Misawa T., Hattori T., Ohoka N.,
Tanaka M., Naito M., Kurihara M., Demizu Y., Sci. Rep., 6, 33003
(2016).
11) Demizu Y., Tanaka M., Nagano M., Kurihara M., Doi M.,
Maruyama T., Suemune H., Chem. Pharm. Bull., 55, 840–842
(2007).
was removed, and the cells were washed 3 times with PBS 12) Tanaka M., Demizu Y., Doi M., Kurihara M., Suemune H., Angew.
Chem. Int. Ed., 43, 5360–5363 (2004).
13) Tamaki M., Han G., Hruby V. J., J. Org. Chem., 66, 1038–1042
supplemented with heparin (20units/mL) and detached by
treatment of trypsin–EDTA. Then, fluorescence intensity in
the cells was measured as above.
Intracellular Delivery of plasmid DNA Huh-7 cells were
separately seeded onto 24-well culture plates (10000cells/well)
and incubated overnight in 400µL of DMEM containing 10%
FBS. The medium was exchanged, and the peptide/pDNA
(2001).
14) Toniolo C., Polese A., Formaggio F., Crisma M., Kampphius J., J.
Am. Chem. Soc., 118, 2744–2745 (1996).
15) Nakase I., Niwa M., Takeuchi T., Sonomura K., Kawabata N., Koike
Y., Takehashi M., Tanaka S., Ueda K., Simpson J. C., Jones A. T.,
Sugiura Y., Futaki S., Mol. Ther., 10, 1011–1022 (2004).
complex solutions (33.3µgpDNA/mL) prepared at charge ratio
4:1, and JetPEI/pDNA (at charge ratio 5:1) were applied to
each well. The amount of pDNA was adjusted to 1µg per
well. After 24-h incubation, the medium was replaced with
400µL of fresh medium, followed by incubation. Luciferase
gene expression was then evaluated based on photolumines-
cence intensity using the Luciferase assay kit and a Lumi-
nometer (Gene Light GL-210A, Microtec. Co., Ltd., Chiba,
Japan). The amount of protein in each well was concomitantly
determined using a Micro BCA protein assay kit. The results
are presented as the mean and standard deviation obtained
from 4 samples.
16) Koivusalo M., Welch C., Hayashi H., Scott C. C., Kim M., Alex-
ander T., Touret N., Hahn K. M., Grinstein S., J. Cell Biol., 188,
547–563 (2010).
17) Rothberg K. G., Heuser J. E., Donzell W. C., Ying Y. S., Glenney J.
R., Anderson R. G., Cell, 68, 673–682 (1992).
18) Guo S., Zhang X., Zheng M., Zhang X., Min C., Wang Z., Cheon S.
H., Oak M. H., Nah S. Y., Kim K. M., Biochim. Biophys. Acta, 1848
(10 Pt A), 2101–2110 (2015).
19) Hong S. Y., Oh J. E., Lee K. H., Biochem. Pharmacol., 58, 1775–
1780 (1999).