Engineering the Binding Kinetics of Synthetic Polymer Nanoparticles for siRNA Delivery

Article abstract of DOI:10.1021/acs.biomac.9b00611

The affinity of a synthetic polymer nanoparticle (NP) to a target biomacromolecule is determined by the association and dissociation rate constants (k_on, k_off) of the interaction. The individual rates and their sensitivity to local environmental influences are important factors for the on-demand capture and release a target biomacromolecule. Positively charged NPs for small interfering RNA (siRNA) delivery is a case in point. The knockdown efficacy of siRNA can be strongly influenced by the binding kinetics to the NP. Here, we show that k_on and k_off of siRNA to NPs can be individually engineered by tuning the chemical structure and composition of the NP. N-Isopropylacrylamide-based NPs functionalized with hydrophobic and amine monomers were used. k_off decreased by increasing the amount of amine groups in the NP, whereas k_on did not change. Importantly, NPs showing a low k_off at pH 5.5 together with a high k_off at pH 7.4 showed high knockdown efficiency when NP/siRNA complexes were packaged in lipid nanoparticles. These results provide direct evidence for the premise that the efficacy of an siRNA delivery vector is linked with the strong affinity to the siRNA in the endosome and low affinity in the cytoplasm.

Full text of DOI:10.1021/acs.biomac.9b00611

Subscriber access provided by University of Texas Libraries
Article
Engineering the binding kinetics of synthetic
polymer nanoparticles for siRNA delivery
Hiroyuki Koide, Tatsuya Fukuta, Anna Okishima, Saki Ariizumi, Chiaki Kiyokawa,
Hiroki Tsuchida, Masahiko Nakamoto, Keiichi Yoshimatsu, Hidenori Ando,
Takehisa Dewa, Tomohiro Asai, Naoto Oku, Yu Hoshino, and Kenneth J. Shea
Biomacromolecules, Just Accepted Manuscript • DOI: 10.1021/acs.biomac.9b00611 • Publication Date (Web): 13 Sep 2019
Downloaded from pubs.acs.org on September 14, 2019
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 35
Biomacromolecules
1
2
3
4
5
6
7
8
9
Engineering the binding kinetics of synthetic
polymer nanoparticles for siRNA delivery
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Hiroyuki Koide,† Tatsuya Fukuta,† Anna Okishim,† Saki Ariizumi,† Chiaki Kiyokawa,† Hiroki
Tsuchida,† Masahiko Nakamoto,‡ Keiichi Yoshimatsu,§ Hidenori Ando,† Takehisa Dewa, ǁ
Tomohiro Asai,† Naoto Oku,† Yu Hoshino, *‡ and Kenneth J. Shea*§
†
Department of Medical Biochemistry, School of Pharmaceutical Sciences, University of
Shizuoka, 52-1 Yada, Suruga-ku, Shizuoka, Shizuoka 422-8526, Japan; ‡ Department of
Chemical Engineering, Kyushu University, 744 Motooka, Fukuoka 819-0395, Japan; §
Department of Chemistry, University of California Irvine, Irvine, CA 92697 USA; ǁ Department
of Life Science and Applied Chemistry, Graduate School of Engineering, Nagoya Institute of
Technology, Gokiso-cho, Showa-ku, Nagoya, Aichi 466-8555, Japan
KEYWORDS: polymer nanoparticle, binding kinetics, siRNA, cell penetrate peptide, lipid
nanoparticle
ACS Paragon Plus Environment
1

Biomacromolecules
Page 2 of 35
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
ABSTRACT
The affinity of a synthetic polymer nanoparticle (NP) to a target biomacromolecule is
determined by the association and dissociation rate constants (kon, koff) of the interaction. The
individual rates and their sensitivity to local environmental influences are important factors for
the on demand capture and release a target biomacromolecule. Positively charged NPs for siRNA
delivery is a case in point. The knockdown efficacy of siRNA can be strongly influenced by the
binding kinetics to the NP. Here, we show that kon and koff of siRNA to NPs can be individually
engineered by tuning the chemical structure and composition of the NP. N-isopropylacrylamide-
based NPs functionalized with hydrophobic and amine monomers were used. koff decreased by
increasing the amount of amine groups in the NP, whereas kon did not change. Importantly, NPs
showing a low koff at pH 5.5 together with a high koff at pH 7.4 showed high knockdown
efficiency when NP/siRNA complexes were packaged in lipid nanoparticles. These results
provide direct evidence for the premise that efficacy of a siRNA delivery vector is linked to
strong affinity to the siRNA in the endosome and low affinity in the cytoplasm.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment
2

Page 3 of 35
Biomacromolecules
1
2
3
4
5
6
7
8
9
INTRODUCTION
There are now a number of examples of synthetic polymer hydrogel nanoparticles (NPs), that
have been designed to capture and/or release target biomacromolecules on the basis of their
1-5
intrinsic chemical composition. The performance of abiotic NPs can be significantly influenced
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
6-9
by not only their biomacromolecule affinity and selectivity but also the binding kinetics (kon, koff
)
7, 8, 10
of the NP-biomacromolecule interaction.
For example, when NPs are injected into the
bloodstream, they immediately become associated with serum proteins. This protein corona can
6-8, 10
significantly diminish the binding affinity of the NP to its intended target. However, if the NP-
plasma protein association and dissociation rates are fast, these proteins can be exchanged,
allowing NPs to interact with target proteins. It was found that both association and dissociation
7
4
5
11
12
rates depend on the protein identity, pH, temperature, flexibility of NPs and polymer density.
The dissociation rate constants of NPs to the targets can also be tuned by optimizing the
combination and density of charged, hydrogen-bond forming, aliphatic, and aromatic functional
4, 6, 13-15
monomers.
We have been developing poly N-isopropylacrylamide (pNIPAm)-based polymer
16
6
17, 18
nanoparticles (NPs) that bind to target small molecules, peptides and proteins both in vitro
and in vivo by incorporating combinations of hydrophobic and charged monomers in the
polymer. Recently, we reported NPs that bind to the vascular endothelial growth factor
(
VEGF165), an important signaling protein, by the inclusion of multiple sulfated 3,4,6S GlcNAc
19
20
into the NPs. The NPs inhibited tumor growth by the intravenous injection in mice. It should be
possible therefore, to engineer the binding kinetics of NP-biomacromolecule interactions by
varying monomer composition.
Small interfering RNA (siRNA) disrupts the expression of specific genes with complementary
21, 22
nucleotide sequences and is being utilized in biomedical research and drug development. The
packaging and delivery of siRNA remains challenging. There have been a number of reports of
23-32
siRNA delivery vectors incorporating amine groups with significant knockdown efficacy. It is
important to form stable siRNA-vector complexes by multi-point electrostatic interactions
between ammonium cations in the vector and phosphate anions in the siRNA for siRNA
ACS Paragon Plus Environment
3

Biomacromolecules
Page 4 of 35
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
delivery. After these complexes are taken up into cells via endocytosis, proton buffering capacity
of the vector is also important for endsomal escape of siRNA-vector complex by the proton-
33-35
sponge effect. Under ideal conditions, siRNA should detach from the vector and make a
complex with argonaute2 in the cytoplasm for the degradation of target mRNA. Therefore, the
vector requires orthogonal functions; high affinity for siRNA and fast release into the cytoplasm.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
To address this challenge, several groups have suggested that the number and pK
a
of the amine
26, 28, 36, 37
groups in the vector are critical factors for efficient gene-silencing.
However, to our
knowledge, there have been no reports of the relationship between binding kinetics of the siRNA
vector complex (kon and koff) and the knockdown efficiency of target mRNA. In this report, we
focused on fine-tuning the binding kinetics by varying the chemical structure and functional
monomer feed ratio of the NP carrier. We show direct evidence for the relationship between
binding kinetics (kon and koff) of the vector against siRNA and gene-silencing effect.
EXPERIMENTAL SECTION
Materials.
N-isopropylacrylamide
(NIPAm),
N,N,N’,N’-
tetramethylethylenediamine,
hexadecyltrimethylammonium bromide (CTAB), 2,2'-azobis(isobutyronitrile) (AIBN)
ethylenediamine (EDA), diethylenetriamine (DETA) methacryloyl chloride, di-tert-butyl
dicarbonate, ethyl trifluoroacetate, 2M HCl/Et O phenylmethylsulfonyl fluoride (PMSF), 2
2
μg/mL leupeptin, 2 μg/mL aprotinin, and 2 μg/mL pepstatin A were purchased from Sigma
Aldrich. N,N’-methylenebis(acrylamide) (BIS) and N-tert-butylacrylamide (TBAm) were from
ACROS ORGANICS. Cholesterol-conjugated siRNA was purchased from Hokkaido System
Science Co. (Hokkaido, Japan). The cholesterol was conjugated with siRNA at the 3'-end of the
sense strand. The nucleotide sequences of siRNA with a 2-nucleotide overhang (underline) wereꢀ
5
ʹ-GGCUACGUCCAGGAGCGCACC-3ʹ (sense) and 5ʹ-UGCGCUCCUGGACGUAGCCUU-3ʹ
ACS Paragon Plus Environment
4

Page 5 of 35
Biomacromolecules
1
2
3
4
5
6
7
8
9
(
antisense)
and
5ʹ-GGCUACGUCCAGGAGCGCACC-3ʹ
(sense)
and
5ʹ-
UGCGCUCCUGGACGUAGCCUU-3ʹ (antisense, against Luc2). B16F10 murine melanoma
cells were purchased from ATCC (Verginia, USA) and B16F10-Luc2 Bioware Ultra Cell Line (a
luciferase-expressing cell line stably transfected with the firefly luciferase gene (Luc2, B16F10-
Luc2)) were purchased from Caliper Life Sciences (Hopkinton, MA). n-octyl-β-D-glucoside was
purchased from Dojindo (Kumamoto, Japan).
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
8
Synthesis of methacrylate ethylenediamine
Di-tert-butyl dicarbonate (37.2 g, 618 mmol) in dichloromethane (DCM) (40 mL) was
added dropwise to DCM containing ethylenediamine (15.0 g, 68.7 mmol) at 0 °C. Twenty-four
h after stirring at r.t, DCM was evaporated and water insoluble products were removed by the
filtration after the addition of water (100 mL). After the extraction with DCM (100 mL x 3), the
organic layer was then dried with anhydrous Na
2
SO and evaporated to obtain an oily product.
4
Then, DCM (50 mL) containing methacryloyl chloride (5.38 g, 51.43 mmol) and TEA (15.6 g,
54.3 mmol) added dropwise to the oily product (8.23 g, 51.43 mmol) and stirring for 2 h at 0
C. After the extraction of the organic layer with water (50 mL x 3) and drying with anhydrous
1
°
Na
ether/hexane 3:5. In addition, recrystallization was also preformed with chloroform/diethyl ether/
hexane1:55:55. Yield: 8.0 g (47.0 mmol, 68.4 %); H1 NMR (500 MHz, CDCl ): δ=1.40 (s, 9H),
δ=1.98 (s, 3H), δ=3.40 (d, 4H), δ=5.40 (s, 1H), δ=5.75 (s, 1H), m/z calcd for C11 ON Na:
2
SO
4
, evaporation of DCM was performed and the product was recrystallized with diethyl
3
H
2
2
O
3
+
+
2
51.1372 (M Na ); found: 251.1376
DCM (30 mL)-containing product (3.6 g, 9.8 mmol) and 2 M HCl/Et O (30 ml) was
2
stirred at r.t. for 24 h. The precipitated hygroscopic salt was filtered and washed with diethyl
1
ether. Yield: 2.0 g (9.7 mmol, 99.6 %); H NMR (500 MHz, CDCl
3
): δ=2.00 (s, 3H), δ=3.13 (t,
ACS Paragon Plus Environment
5

Biomacromolecules
Page 6 of 35
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
H), δ=3.57 (t, 2H), δ=5.48 (s, 1H), δ=5.84 (s, 1H), m/z calcd for C
6
H
12
N ONa: 151.0847
2
+
+
(
M Na ); found .151.0852
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Synthesis of methacrylate diethylenetriamine.
Diethylentriamiane (DETA, 10 g, 96.9 mmol) was protected on a primary amine by
reaction with ethyl trifluoroacetate (13.7 g, 96.9 mmol, MeOH, -78 °C, for 1h). The remaining
amino functional groups were next protected with di-tert-butyl dicarbonate (46.5 g, 213.2 mmol,
0
°C, for 2h), then, triethylamine (21.6 g, 213.2 mmol, 0 °C, for 2h) was added. The TFA
protecting group was then removed by increasing the pH to 11 with conc. aq. ammonia, stirring
25 °C, overnight). The di-boc protected DETA was purified by silica gel column
(
chromatography (DCM: MeOH: NH OH= 100: 50 to 100: 20: 2, v/v/v) for two times. Yield: 13.1
3
g (43.18 mmol, 44.6%); H1 NMR (500 MHz, CDCl3): δ=1.41 (s, 9H), δ=1.45 (s, 9H), δ=3.26 (t,
H), δ=3.40 (t, 4H), δ=3.46 (t, 2H)
2
M/z calcd for C14
H
29
N
3
O Na: 326.2056: (M+Na+); found. 326.2049
4
Methacryloyl chloride (3.3 g, 32.0 mmol) and TEA (9.1 g, 90.0 mmol) were dissolved in
chloroform (20 mL) and added dropwise to the di-boc-DETA (5 g, 16.5 mmol) over a period of 2
h at 0 °C. The organic layer was extracted with water (50 mL x 3) and dried over anhydrous
Na
2
SO . Chloroform was evaporated and the product was purified by silica gel column
4
chromatography (DCM: MeOH = 100: 5 to 10: 1, v/v). Yield: 3.6 g (9.8 mmol, 59.4 %); H1
NMR (500 MHz, CDCl3): δ=1.42 (s, 9H), δ=1.45 (s, 9H), δ=1.94 (s, 3H), δ=3.31 (t, 4H),
δ=3.47 (t, 4H), δ=5.36 (s, 1H) , δ=5.75 (s, 1H) m/z calcd for C18
H
33
N
3
O Na: 394.2318: (M+Na+);
5
found .394.2314
A solution-containing product (3.6 g, 9.8 mmol) in DCM 30 mL and 2 M HCl/Et
2
O (30
ml) was stirred at room temperature for 24 h. Then, the precipitated hygroscopic salt was filtered
1
and washed with diethyl ether. Yield: 2.0 g (9.7 mmol, 99.6 %); H NMR (500 MHz, CDCl
3
):
6
ACS Paragon Plus Environment

Page 7 of 35
Biomacromolecules
1
2
3
4
5
6
7
8
9
δ=2.02 (s, 3H), δ=3.39 (sextet, 6H), δ=3.66 (t, 2H), δ=5.50 (s, 1H) , δ=5.89 (s, 1H) m/z calcd
+
for C
8
H
18
N O: 172.1450: (M+H ); found .172.1457
3
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Preparation of NPs.
3
9
NPs were prepared by a modified precipitation polymerization of N-isopropylacrylamide
(NIPAm), N,N’-methylenebisacrylamide (Bis), N-tert-butylacrylamide (TBAm) and N-(3-
aminopropyl)methacrylamide hydrochloride, methacrylated ethylendiamine (EDA) or
methacrylated diethylentriamine (DETA). NIPAm, amine monomer (APM, EDA or DETA),
TBAm, BIS, and CTAB (10 mg) were dissolved in ultrapure water (total monomer
concentration; 65 mM). TBAm was dissolved in a small amount of ethanol before the addition.
The monomers and surfactant-containing solutions were filtered and degassed by the nitrogen
purge for 30 min. Then, azobisisobutyronitrile (AIBN, 6 mg/10 mL) was added to the solution
under a nitrogen atmosphere for the starting the polymerization (65°C for 3 h). The synthesized
NPs were purified by dialysis (changed twice a day) for 4 days.
Characterization of NPs
The hydrodynamic diameter of NP was determined in aqueous solution by dynamic light
scattering (DLS) (Zetasizer Nano ZS). The temperature of the NP samples was controlled via
Peltier device at 25 ± 0.1 °C. Yield and concentration of NPs were determined by measuring
weight of NP after lyophilization.
Titration assay
ACS Paragon Plus Environment
7

Biomacromolecules
Page 8 of 35
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
5
ml of each NP solution (1 mg/ml in water) was bubbled by nitrogen for 30 min. Then,
1
0 µl of 0.01 N HCl was added into the NP solution. pH change was monitored after each
injection.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Electrophoresis assay
siRNA and amine NP complex was incubated for 30 min at r.t. Then, free siRNA
dissociated from NPs was separated by electrophoresis in 15% acrylamide gel. The gel was
stained for 30 min in Gel Red, and siRNA was detected by a LAS-3000 mini system (Fuji Film,
Tokyo, Japan).
Cell culture
B16F10 and B16F10-Luc2 cells were cultured in DME/Ham F-12 medium (WAKO,
Osaka, Japan) containing 10% fetal bovine serum (FBS, Sigma-Aldrich), 100 units/mL penicillin
(
MP Biomedicals, Irvine, CA), and 100 μg/mL streptomycin (MP Biomedicals) in a CO
2
incubator.
siRNA uptake
4
0
siRNA uptake studies were performed as previously described. Briefly, B16F10 cells
4
(
1.0 × 10 cells/well) were incubated with naked FITC-siRNA (100 nM) or FITC-labeled siRNA-
C formulated in NPs or NP/siRNA-C-LNP-CPP for 24 h. After washing the cells with PBS, the
cells were lysed (1 w/v% n-octyl-β-D-glucoside containing 1 mM PMSF, 2 μg/mL leupeptin, 2
μg/mL aprotinin, and 2 μg/mL pepstatin A). The FITC fluorescence intensity and protein amount
were measured by a Tecan Infinite M200 microplate reader (Salzburg, Austria) (ex. 485 nm, em.
ACS Paragon Plus Environment
8

Page 9 of 35
Biomacromolecules
1
2
3
4
5
6
7
8
9
5
35 nm) and BCA Protein Assay Reagent Kit (PIERCE Bio- technology, Rockford, IL),
respectively.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Knockdown effect
4
0
Knockdown studies were performed as previously described. Briefly, B16F10-luc2 cells
3
(
5 × 10 cells/well) were incubated with sample (50 pmol of siRNA-C (siLuc2)) in the medium
without antibiotics. After washing the cells, the cells were cultured in the fresh medium (with
antibiotics) for 24 h. Then, luminescence and cell viability were measured with ONE-Glo
Reagent (Promega, Madison, WI) and Cell Titer-Fluor Cell Viability Assay kit (Promega),
respectively.
Preparation of CPP-modified NPs and siRNA complex-encapsulated lipid nanoparticles
(
NP/siRNA-C-LNPs-CPP)
,2-Dipalmitoyl-sn-glycero-3-phosphoglycerol,
1
sodium
salt
(DPPG)
dioleoylphosphatidylethanolamine (DOPE) and cholesterol (DPPG: DOPE: Cho = 1:1:1 as a
molar ratio) were dissolved in tert-butyl alcohol and freeze-dried. LNPs were prepared by the
hydration with DEPC-treated RNase-free water and sized by extrusion 10 times through a
polycarbonate membrane filter (100-nm pores, Nucleopore, Maidstone, UK). NP (75 µg) and
siRNA-C (250 pmol) were incubated at 37 °C for 30 min (total 250 µL). NPs and siRNA
complex-encapsulated lipid nanoparticles (NP/siRNA-C-LNPs) was prepared by freeze-thawing
40
method. The NP/siRNA-C complex (250 µL) was added to LNPs (1 mM, 750 µL) and freeze-
thawed for 3 times. CPP-DOPE was modified onto the NP/siRNA-C-LNP surface (6 mol%
against total lipid as molar ratio) by incubation at 37 °C for 30 min.
ACS Paragon Plus Environment
9

Biomacromolecules
Page 10 of 35
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Confocal microscopy
B16F10 cells were seeded onto eight-well chamber slides (Nunc, Rochester, NY, USA)
4
at a density of 1.0ꢀ×ꢀ10 cells/well and transfected with TAMRA-labeled siRNA-C (TAMRA-
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
siRNA-C), which was complexed with NPs or NP/siRNA-C-LNP-CPP (TAMRA-siRNAC
concentration was 30 pmol/0.3ꢀml/well). Twenty-four h after transfection with each sample, the
cells were washed three times with PBS (pH 7.4) containing heparin (5 units/ml). Then, the cells
were incubated in fresh medium containing Lysotracker in accordance with the manufacturer’s
instructions and fixed with 4% paraformaldehyde. Nuclei were stained with 4ʹ,6-diamino-2-
phenylindole (DAPI; 10ꢀµg/ml; Invitrogen) in PBS containing 3% BSA and 0.1% saponin
(Sigma-Aldrich). The localization of FITC-labeled siRNA-C in B16F10 was observed by
LSM510 META confocal laser-scanning microscope (Carl Zeiss, Jena, Germany).
Quartz crystal microbalance (QCM) analysis
8
Interactions between the NPs and siRNA was quantified by an Affinix Q QCM instrument
12, 18
(Initium Co. Ltd., Tokyo, Japan). Gold surfaces in QCM cell were incubated with piranha
solution for 5 min, twice fir the surface cleaning. 3,3’-Dithiodipropionic acid (1 mM, 0.1 mL)
was added to the gold surface and incubated for overnight. Then, the gold surfaces were washed
with ultrapure water and added 1-ethyl-3-(3-dimethylaminopropyl)-carbodiimide (100 mg/ml)
and N-hydroxysuccinimide (100 mg/mL) (1:1, 0.1 mL) to form N-hydroxysuccinimidyl esters.
Amine NPs were loaded on the cells to give NPs-immobilized cells. Then, the QCM cells were
washed with PBS for 3 times. siRNA-C was added onto NPs immobilized QCM cells at the
concentration of 30, 100 or 300 nM. Interactions between NPs and siRNA-C were measured at
(37±0.1)°C in PBS (pH 7.4). The apparent dissociation constant of NPs to siRNA-C was
calculated under the assumption that all particles have the same affinity to siRNA-C.
ACS Paragon Plus Environment
1
0

Page 11 of 35
Biomacromolecules
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment
1
1

Biomacromolecules
Page 12 of 35
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
RESULTS AND DISCUSSION
Preparation of synthetic polymer NPs for siRNA affinity
To engineer the binding kinetics of NPs to siRNA, we evaluated the contribution of both
the number and spacing of amine residues on NP−siRNA binding and both dissociation rate and
dissociation equilibrium constants. It was reported that spacing of positively charged amino
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
8
groups in the side chain plays a key role for gene-silencing. Amine monomers of methacrylated
ethylenediamine (EDA) and diethylenetriamine (DETA) were used to introduce positively
charged functionality (Figure 1). Since the synthetic NP is a poly-N-acrylamide derivative,
DETA functionalized NPs have a different number of amine residues per side chain and the
length of amine side chain differs from EDA functionalized NPs. To establish the influence of
carbon number, (3-aminopropyl)methacrylamide hydrochloride (APM) was also used as an
amine source. The amine groups in the NPs contribute to the electrostatic attraction to siRNA at
a pH lower than the pK
a
of their conjugate ammonium ions. It is also known that incorporation of
6
, 7
hydrophobic groups in the NP is effective for creating high affinity for hydrophobic targets and
4
1
for tuning the pK
a
of ammonium ions in the NPs. Thus, siRNA was conjugated with cholesterol
4
2, 43
(siRNA-C) at the 3’ position of the sense strand
and 40 mol% of N-tert-butyl acrylamide
(TBAm, hydrophobic monomer) was incorporated into all NPs to create the hydrophobic
interactions. Cholesterol conjugation to siRNA at the 3’ position improves stability of siRNA
4
2
against nuclease in the bloodstream without reduction of the knockdown effect. The NPs were
synthesized by free-radical copolymerization of the functional monomers with a cross-linking
monomer (N,N’-methylene-bis(acrylamide) (BIS), 2 mol %) in aqueous solution containing
small amounts of CTAB. AIBN was used as initiator (Figure 1). Summaries of NP
compositions, particle sizes and ζ-potentials are given in Table 1. To measure protonation
capacity and apparent pK of the NPs, an acid-base titration assay was performed (Figure S1a-
a
2
6
c). Both mono-amine (APM and EDA) NPs showed a single and clear neutralization point.
Interestingly, APM and EDA NPs with 5 mol% of the amine monomers (NP1 and 5) showed
lower average pK
a
’s (~7.5) than those with more than 15 mol% (~8-10). This may be due to a
ACS Paragon Plus Environment
1
2

Page 13 of 35
Biomacromolecules
1
2
3
4
5
6
7
8
9
less polar microenvironment around isolated amine groups in NP1 and 5 than those of NPs
41, 44
containing a larger number of amines. Di-amine (DETA) NPs showed a characteristic gradual
decrease in pH upon addition of 0.01N HCl due to the two different pK’s of the DETA
monomer. Due to the electrostatic repulsion of neighboring ammonium ions, apparent pK’s of
a
a
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
DETA NPs (<7.5) are lower than that of APM and EDA NPs. Importantly, proton capacity of
NPs with the same percentage of amines was not significantly different. In addition the degree of
protonation was not significantly different for each NP, ~1 at pH5.5 and ~ 0.4-0.5 at pH7.4. It
was reported that both ionization behavior and proton buffering capacity are important for the
26, 45-47
proton sponge effect.
be similar. ”
These results suggest that the proton sponge effect of each NP should
O
O
H
N
H
N
HN
HN
O
O
N-isopropyl acrylamide N-t-butyl acrylamide N,N'-methylenebis
(NIPAm)
(TBAm)
acrylamide (Bis)
CTAB, AIBN in H O
2
7
0 °C, 3 h
O
O
O
H
N
NH2
N
H
NH2
N
H
N
H
NH2
Polyamine NPs
(APM)
(EDA)
(DETA)
Figure 1. Preparation of amine functionalized NPs. NPs were synthesized by free radical
copolymerization of functional (NIPAm, TBAm and Bis) and amine monomers in the presence
of cetyltrimethylammonium bromide (CTAB, 0.694 mM) in water. Following the addition of
2,2'-azobis(isobutyronitrile) (AIBN, 2.63 mM), the polymerization was carried out at 70 °C for 3
h. NPs were purified by dialysis using 100K MW cutoff dialysis tubing. NP numbering is
correlated with the functional amine monomer.
Table 1. Monomer compositions, sizes, ζ-potentials and electrophoretic mobilities of NPs.
ACS Paragon Plus Environment
1
3

Biomacromolecules
Page 14 of 35
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
ꢀ
NIPAm APM TBAm Bis
Size (nm)
PDI
ζ-potential (mV) Mobility (µmcm/Vs)
NP1
(5% APM)
NP2
10% APM)
NP3
15% APM)
NP4
20% APM)
53
48
43
38
5
40
40
40
40
2
2
2
2
56 ± 8
62 ± 8
61 ± 7
74 ± 19
0.19 ± 0.12
0.16 ± 0.10
0.22 ± 0.05
0.27 ± 0.06
47 ± 4
39 ± 4
42 ± 2
24 ± 2
3.7 ± 0.3
3.1 ± 0.3
3.3 ± 0.2
1.9 ± 0.2
(
(
(
10
15
20
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
ꢀNIPAm EDA TBAm Bis
Size (nm)
PDI
ζ-potential (mV) Mobility (µmcm/Vs
NP5
(
5% EDA)
NP6
53
48
43
38
5
40
40
40
40
2
2
2
2
58 ± 5
44 ± 6
57 ± 7
78 ± 10
0.25 ± 0.04
0.22 ± 0.03
0.24 ± 0.07
0.29 ± 0.13
36 ± 8
34 ± 5
35 ± 10
37 ± 4
2.8 ± 0.6
2.7 ± 0.4
2.7 ± 0.8
2.9 ± 0.3
(
10% EDA)
NP7
15% EDA)
NP8
10
15
20
(
(
20% EDA)
ꢀ
NIPAm DETA TBAm Bis
Size (nm)
PDI
ζ-potential (mV) Mobility (µmcm/Vs)
7
8
9
0
1
2
3
4
NP9
5% DETA)
NP10
10% DETA)
NP11
15% DETA)
NP12
(
53
48
43
38
5
40
40
40
40
2
2
2
2
53 ± 4
52 ± 6
52 ±16
53 ± 4
0.27 ± 0.04
0.20 ± 0.04
0.24 ± 0.10
0.27 ± 0.10
44 ± 3
37 ± 9
40 ± 4
36 ± 3
3.4 ± 0.2
2.9 ± 0.7
3.1 ± 0.3
2.8 ± 0.2
(
(
10
15
20
5 (20% DETA)
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment
1
4

Page 15 of 35
Biomacromolecules
1
2
3
4
5
6
7
8
9
Evaluation of NP affinity to siRNA-C
We next evaluated the siRNA interaction with NPs by electrophoretic analysis. (Figure
2a). After incubating NPs (5 µg) and siRNA-C (0.35 µg) in water (pH 5.5), the unbound siRNA-
C was separated on a 15% polyacrylamide gel. siRNA incubated with NPs which have only 5
mol% of APM, EDA, and DETA (NP1 (5% APM), NP5 (5% EDA) and NP9 (5% DETA),
respectively) showed a strong band at the same migration distance as the one without NPs (lane
C), indicating the NPs had little affinity for siRNA-C. However, NPs with more than 10 mol% of
amine monomers (NP2-4, NP6-8 and NP10-12, respectively) captured the siRNA-C almost
quantitatively, indicating the importance of the density of charged groups. It was reported that
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
5
capture of siRNA by a NP requires inclusion of a critical amount of cationic groups, indicating
charge density is important for siRNA binding. We suggest that the positively charged
monomers are uniformly dispersed in the NPs due to electrostatic repulsion. Therefore, the
charge density of 5 mol% cationic monomer containing NPs is insufficient for siRNA binding.
To confirm the contribution of the hydrophobic interaction to NP-siRNA binding, an
electrophoretic assay was performed with siRNA that was not conjugated with cholesterol or
NPs containing a low percentage of hydrophobic monomer (TBAm). The association of
cholesterol free siRNA with DETA NPs was lower than siRNA-C (Figure S2a). However, APM
and EDA NPs still had a strong interaction indicating electrostatic interactions dominate siRNA
interactions with these NPs. In the case of low TBAm containing NPs, we focused on the DETA
NPs. When the TBAm percentage decreased from 20% to 0% (Table S1), free siRNA bands
gradually appeared (Figure S2b). Since hydrophobic interactions between cholesterol and
TBAm should assist the interaction between siRNA and DETA containing NPs, the combination
of electrostatic and hydrophobic interactions are important to achieve high affinity between these
NPs and siRNA-C (Figure S2c,d,e) and highlights the difference between di-amine and mono-
amine containing NPs.
ACS Paragon Plus Environment
1
5

Biomacromolecules
Page 16 of 35
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
To account for the difference in binding affinity, we measured association and
dissociation rate constants of NPs against siRNA-C at pH 5.5 (endosomal pH) and pH 7.4
(
cytoplasm pH). siRNA-C was injected into amine NPs immobilized on quartz crystal
microbalance (QCM) cells at several concentrations (Figure 2b). Both the binding rate
relaxation time of the binding process) and the binding amount of siRNA-C depended on the
concentration of siRNA-C. The time courses of the frequency changes of the QCM in the
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(
48
siRNA-C binding events were fitted to a single exponential function (Figure S3). ΔF, ΔF ;
∞
Frequency change at time t or ∞ after injection, τ; relaxation time.
In the Langmuir adsorption model, the binding process is defined as eq. (1), and the
increasing amount of siRNA/binding-site complex per unit time is given by eq. (2). In this study,
the amount of siRNA in solution is in large excess relative to the number of binding sites on the
surface of the QCM (eq. (3)), so the number of siRNA/binding-site complexes at time t after
injection is given by eq. (4). Then, eq. (2) is rewritten as eq. (5); here, we defined τ as eq. (6).
Thus, the amount of siRNA/binding-site complex formed at time t after injection is given by eq.
(7). In the QCM measurement, the frequency change at time ∞ (∆Fmax) corresponds to eq. (8)
and the frequency change at time t after injection (∆F) corresponds to the number of
siRNA/binding-site complexes. Thus, eq. (7) is used synonymously with eq. (9). Therefore, the
relaxation time (τ) of siRNA binding can be obtained from curve fitting by a single exponential
function using eq. (9) at different siRNA concentrations (30–300 nM). Association and
dissociation rate constants (kon and koff, respectively) are given as the slope and y-intercept of the
linear correlations between siRNA concentrations and the reciprocal of the binding relaxation
time according to eq. (6). The dissociation constants (K ) are given as koff/kon.
d
Frequency changes in the siRNA binding processes were fit well by the single exponential
function (Figure S3). The reciprocals of the binding relaxation times (τ), which were obtained
from the exponential function, were proportional to the concentrations of siRNA (Figure 2).
ACS Paragon Plus Environment
1
6

Page 17 of 35
Biomacromolecules
1
2
3
4
5
6
7
8
9
From those results, we analyzed the binding kinetics by approximating them by the Langmuir
adsorption model.
kon
(
1)
2)
siRNA + Binding site ꢀ siRNA
ꢁ
Binding site
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
k
off
(
d[siRNA
ꢁBinding site]/dt = k [siRNA][Binding site]-k [siRNAꢁBinding site]
on off
(3)
[siRNA] >> [ Binding site]0
0
(
4)
5)
[Binding site]=[Binding site] - [siRNA
ꢁ
Binding site]
0
(
d[siRNA
ꢁ
Binding site]/dt = -1/ τ ( [siRNA
ꢁ
Binding site]-k [siRNA] [Binding site] /-k [siRNA] + k
)
off
on
0
0
on
0
(6)
1/ τ =k [siRNA] + k
on 0 off
(
7)
[siRNA
ΔFmax = k [siRNA] [Binding site] / (k [siRNA] +k )
off
ꢁ
Binding site] = (1-exp(-t/ τ ))k [siRNA] [Binding site] / (k [siRNA] +k
)
on
0
0
on
0
off
(
8)
9)
on
0
0
on
0
(
ΔF = ΔFmax(1-exp(-t/ τ ))
Plotting the reciprocal of the binding relaxation time (t) against siRNA-C concentration
−1
gives a linear correlation between siRNA-C concentration and τ for each NP (Figure 2b). The
binding process can be approximated as Langmuir-type binding isotherm shown in (10).
Apparent association/dissociation rate constants (kon and koff) of NPs are given as the slope and y-
11, 12
intercept of the linear correlations (11).
siRNA + NP � siRNA・NP
(10)
(11)
-
1
τ = kon [siRNA] +koff
Apparent association and dissociation rate constants (kon and koff) and dissociation
equilibrium constants K (=koff/kon) at pH 5.5 and pH 7.4 are shown in Table 2. The kinetic analysis
revealed that kon of all NPs shows approximately 10×10 M s at pH 5.5 and 20×10 M s at pH
d
3
-1 -1
3
-1 -1
7
.4 although the amine percentage and number of amine groups in each NP were different. koff
ACS Paragon Plus Environment
1
7

Biomacromolecules
Page 18 of 35
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
could be reduced by increasing the percent of amine monomers, this results in a stronger affinity
(
smaller K value) to the siRNA. Both kon and koff of siRNA to APM and EDA NPs (NP2 (10%
d
APM), 4 (20% APM), 6 (10% EDA), 8 (20% EDA)) can be equally accelerated (2~3 times) by
increasing pH. However, koff of siRNA to DETA NPs (NP10 (10% DETA), 12 (20% DETA))
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
increased 6-8 times by the pH change despite a 1.5-fold increase in kon. Eventually, K of DETA
d
NPs was dramatically increased compared with that of APM and EDA NPs. It is possible
therefore to engineer NPs, which have the same binding affinity but a different exchange rate by
changing the pH. We find it is possible to use pH to accelerate only the dissociation rate constant
(koff) by optimizing amine structure and percentage. The result is to increase binding affinity by
about one order of magnitude.
a
C
1
2
3
4
5
6
7
8
9
10 11 12 (NP)
APM NPs
EDANPs
DETANPs
b
NP2 (10% APM)
pH 5.5 pH 7.4
NP4 (20% APM)
NP6 (10% EDA)
pH 5.5 pH 7.4
1
0
8
6
4
2
0
10
10
8
pH 5.5 pH 7.4
8
6
4
2
0
6
τ
4
2
0
0
100
200
300
0
100
200
300
0
100
200
300
siRNA (nM)
siRNA (nM)
siRNA (nM)
c
NP8 (20% EDA)
pH 5.5 pH 7.4
NP10 (10% DETA)
pH 5.5 pH 7.4
NP12 (20% DETA)
pH 5.5 pH 7.4
1
0
10
8
10
8
8
6
4
2
0
6
6
4
4
2
2
0
0
0
100
200
300
0
100
200
300
0
100
200
300
siRNA (nM)
siRNA (nM)
siRNA (nM)
ACS Paragon Plus Environment
1
8

Page 19 of 35
Biomacromolecules
1
2
3
4
5
6
7
8
9
Figure 2. Binding affinity of amine NPs to siRNA-C. (a) Electrophoretic analysis of NP-siRNA-
C interaction. siRNA-C which is not attached to 5 µg of NPs was separated by electrophoresis in
15% polyacrylamide gel. C; Naked siRNA-C. Remaining lanes siRNA-C + indicated NPs. (b,c)
Plots of binding relaxation time (τ) of NP2 (10% APM), 4 (20% APM), 6 (10% EDA) 8 (20%
EDA), 10 (10% DETA) and 12 (20 % DETA) at pH 7.4 (open circle) or 5.5 (black circle).
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment
1
9

Biomacromolecules
Page 20 of 35
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table 2. Association and dissociation rate constants and equilibrium constants of NPs.
pH 5.5
pH 7.4
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
k
on
k
off
K
d
k
on
k
off
K
d
3
-
(10 M
1
-1
-3 -1
3
-1
-1
-3 -1
� s )
(10 s )
(nM)
82
(10 M � s )
(10 s )
(nM)
114
NP2
13
1.06
23
2.63
(10% APM)
NP4
1
1
3
2
0.30
1.26
0.38
0.50
0.23
23
23
22
21
17
15
0.78
2.53
0.89
3.16
1.92
34
(
20% APM)
NP6
10% EDA)
105
35
115
42
(
NP8
11
(20% EDA)
NP10
1
1
1
1
45
186
128
(
10% DETA)
NP12
20% DETA)
21
(
ACS Paragon Plus Environment
2
0

Page 21 of 35
Biomacromolecules
1
2
3
4
5
6
7
8
9
Preparation of NPs and siRNA-C complex-encapsulated lipid nanoparticles
To evaluate the importance of the binding kinetics of NPs against siRNA for efficient
knockdown, it is important to design experimental conditions in which the amount of siRNA-C
uptake into the cells is equal in each NP/siRNA-C complex. To equalize the siRNA uptake,
NP/siRNA-C complexes were encapsulated in lipid nanoparticles (LNPs, 1,2-Dipalmitoyl-sn-
glycero-3-phosphorylglycerol sodium salt (DPPG) : Cholesterol : 1,2-Dioleoyl-sn-glycero-3-
phosphoethanolamine (DOPE)=1:1:1) by freeze-thawing NP/siRNA-C and LNP complexes three
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
0, 49, 50
times (NP/siRNA-C-LNP, Figure 3a).
Incorporation of DOPE, fusogenic lipids, into the LNP
has been shown to promote endosomal release by increasing the interaction between the LNP
5
1-54
and endosomal membranes. To deliver the NP/siRNA-C-LNP into the cell, the NP/siRNA-C-
LNP was modified with DOPE-conjugated cell penetrate peptide (CPP, DOPE-conjugated
RRRRRRGGRRRRG, NP/siRNA-C-LNP-CPP). It was reported that the peptide delivers LNPs
5
5
to the cell via heparan sulfate-mediated endocytosis and macropinocytosis. Initially NPs were
approximately 80 nm but after freeze-thawing the NP/siRNA-C complex, they were no longer
visible (Table 3, Figure S4). The results indicate that most of the NPs were encapsulated into
the LNPs.
ACS Paragon Plus Environment
2
1

Biomacromolecules
Page 22 of 35
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table 3. Size, PDI and ζ-potential and electrophoretic mobilities of NP/siRNA-C-LNP-CPP.
ꢀ
Size
d.nm)
PDI
ζ-potential Mobility
(
(mV)
(µmcm/Vs)
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
NP2 (10% APM)/siRNA-C-LNPs-CPP 342 ± 25 0.21 ± 0.05 33 ± 6
NP4 (20% APM)/siRNA-C-LNPs-CPP 335 ± 23 0.20 ± 0.02 34 ± 7
2.6 ± 0.5
2.7 ± 0.5
2.3 ± 0.2
2.8 ± 0.2
2.6 ± 0.4
3.0 ± 0.1
NP6 (10% EDA)/siRNA-C-LNPs-CPP
NP8 (20% EDA)/siRNA-C-LNPs-CPP
364 ± 14 0.21 ± 0.07 30 ± 2
305 ± 21 0.21 ± 0.04 36 ± 3
NP10 (10% DETA)/siRNA-C-LNPs-CPP 366 ± 15 0.22 ± 0.03 33 ± 5
NP12 (20% DETA)/siRNA-C-LNPs-CPP 357 ± 19 0.31 ± 0.03 38 ± 1
ACS Paragon Plus Environment
2
2

Page 23 of 35
Biomacromolecules
1
2
3
4
5
6
7
8
9
Importance of binding kinetics for siRNA release into cytoplasm and significant knockdown
To demonstrate cellular uptake of siRNA-C wrapped by LNP-CPP, FITC-conjugated siRNA-C
was used. FITC-siRNA-C mixed with only DOPE-conjugated CPP was used as a control.
Twenty-four hours after the addition of NP/siRNA-C-LNP-CPP onto B16F10 cells (siRNA
concentration: 50 nM), fluoresce intensity of siRNA was measured. siRNA uptake into the cells
was not significantly different in each NP (Figure S5). Importantly, the NP/siRNA and
NP/siRNA-C-LNP-CPP complexes have little toxicity at 10 µg/ml NP concentration. (Figure
S6a,b). We next demonstrated the knockdown effect of NP/siRNA-C-LNP-CPP. siRNA-C
against luciferase (Luc2, siLuc2-C) and luciferase overexpressing B16F10 (B16F10-Luc2) were
used for the assay. NP/siLuc2-C-LNP-CPP was incubated with B16F10-Luc2 cells for 24 h.
Surprisingly, NP12 (20% DETA)/siLuc2-C-LNP-CPP showed higher knockdown effect (~60 %
knockdown) than other NP/siLuc2-C-LNP-CPP (~20% knockdown, Figure 3b). Although
siRNA-C mixed with only CPP was also taken up into the cells (Figure S5), there was little
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
knockdown effect. Since surface functionality (CPP modification) and particle size were not
significantly different in each of the NP/siLuc2-C-LNP-CPP complexes, we assumed the siRNA
uptake ratio should have been the same. However, the siRNA delivery ratio of NP12 (20%
DETA)/siLuc2-C-LNP-CPP was lower than that of other NP/siLuc2-C-LNP-CPP complexes.
We can only speculate at this point why this might be so. Perhaps some siRNA became detached
from the NP12 (20% DETA) in the process of encapsulation of the NP12 (20% DETA) /siLuc2-
C complex or some siRNA did not bind to NP12 (20% DETA). Furthermore, we confirmed that
“
naked” NP/siRNA-C complexes not encapsulated by LNPs were ineffective at delivering
siRNA into cells (Figure S7a,b). As a result, the NP/siRNA complex (without LNP and CPP)
showed no knockdown effect (Figure S8). These results establish that differences in the
knockdown effect for each NP/siRNA-LNP-CPP complex arises from the differences in binding
kinetics of siRNA to the NP and proton capacity of the NP. Although NP4 (20% APM) and NP8
(20% EDA) showed similar proton capacity to NP12 (20% DETA), siRNA showed relatively
ACS Paragon Plus Environment
2
3

Biomacromolecules
Page 24 of 35
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
low affinity for NP12 (20% DETA) at pH 7.4 (cytoplasm pH, K =128 nM) compared with NP4
d
(20% APM) (K
d
=34 nM) and NP8 (20% EDA) (K =42 nM). Although 10% amine NPs
d
(NP2,6,10) showed relatively low affinity at pH 7.4 (NP2 (10% APM): 114 nM, NP6 (10%
EDA): 115 nM, NP10 (10% DETA): 186 nM) as well as NP12 (20% DETA), these NPs had low
proton capacity compared with 20% amine NPs (NP4 (20% APM), 8 (20% EDA), 12 (20%
DETA), Figure S1). Since NP12 (20% DETA) satisfied both factors, NP12 (20%
DETA)/siLuc2-C-LNPCPP shows a higher knockdown effect than other NP/siLuc2-C-LNP-CPP
complexes. Therefore, fast siRNA release kinetics from the NP at pH 7.4 and overall proton
capacity of the NP are the important factors for efficient knockdown.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
To demonstrate the effects of binding kinetics of the NPs to siRNA on intracellular distribution
of siRNA, localization of siRNA and NPs in the cells was monitored by a confocal laser
scanning microscope at 24 h after the addition (Figure 3c). TAMRA-conjugated siRNA-C was
used and FITC labeled NPs were synthesized by the inclusion of FITC monomer into NPs (1
mol% in total monomer). Although most of the siRNA and NPs were co-localized in the cell by
adding NP4 (20% APM) or 8 (20% EDA)/siLuc2-C-LNP-CPP (slow siRNA release), many
siRNA and NPs were not co-localized by using NP12 (20% DETA)/siLuc2-C-LNP-CPP (fast
siRNA release). These results indicate that NP12 (20% DETA) effectively and rapidly released
siRNA into the cytoplasm compared with NP4 and NP8 (20% APM and EDA) resulting in
significant knockdown. Since it was suggested that spacing and the number of amine residues is
38
critical for NP−siRNA binding and induction of gene silencing, we prepared EDA, APM and
DETA monomers to evaluate these variables. Both EDA and APM containing NPs showed
similar siRNA affinity and knockdown effect. However, DETA containing NPs behaved quite
differently. These results indicate that spacing of amine residue in the mono-amine structure
(APM and EDA) is not a critical factor for siRNA affinity and knockdown effect but the
proximity of amines is critical. Increase of amine from one (APM and EDA) to two (DETA)
significantly changes the siRNA affinity (kon and koff) and knockdown effect.
ACS Paragon Plus Environment
2
4

Page 25 of 35
Biomacromolecules
1
2
3
4
5
6
7
8
9
Since mono-amine functionalized NPs showed similar K between pH 5.5 (endosomal pH) and
d
7
.4 (cytoplasm pH), siRNA cannot effectively detach from NPs in the cytoplasm. However,
siRNA was effectively released from di-amine functionalized NPs into the cytoplasm, a result of
the larger dissociation equilibrium constants at pH 7.4 compared to pH 5.5. This indicates that
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
engineering kon, koff and K
d
can be important for efficient knockdown in siRNA delivery by
polycationic delivery systems. Until now, it was well known that the important factor of siRNA
delivery vectors was the pK of the amine in the vector. We clearly showed that the importance is
a
based on the change of dissociation equilibrium constants between pH 5.5 and 7.4.
ACS Paragon Plus Environment
2
5

Biomacromolecules
Page 26 of 35
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Cholesterol
a
siRNA-C
LNP
CPP
Freeze-thaw
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
NP
NP/siRNA-C
NP/siRNA-C-LNP
NP/siRNA-C-LNP-CPP
b
100
8
6
0
0
***
40
2
0
0
DAPI (Nclei)
FITC (NP)
TAMRA(siRNA)
Merged
c
NP4 (20% APM)
/siRNA-C-LNPs-CPP
NP8 (20% EDA)
siRNA-C-LNPs-CPP
/
NP12 (20% DETA)
siRNA-C-LNPs-CPP
/
ACS Paragon Plus Environment
2
6

Page 27 of 35
Biomacromolecules
1
2
3
4
5
6
7
8
9
Figure 3. Preparation of NP/siRNA-C-LNP-CPP and its knockdown effect. (a) Preparation of
NP/siRNA-C-LNP-CPP. (b) Gene-silencing effect of NP/siRNA-C-LNP-CPP on B16F10-Luc2
cells. B16F10-Luc2 cells were incubated with NP/siRNA-C-LNP-CPP at a final siRNA
concentration of 50 nM for 24 h. After the complexes were removed, the cells were cultured for
an additional 24 h. Then, luciferase activities were measured and normalized by the total protein
content. (***P<0.001, vs. NP2 (10% APM), 4 (20% APM), 6 (10% EDA), 8 (20% EDA), 10
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(10 % DETA)/siRNA-LNP-CPP). (c) Confocal imaging of FITC-siRNA-C. Twelve hours after
transfection of FITC-siRNA-C and NP4 (20% APM), NP8 (20% EDA) or NP12 (20%
DETA)/siRNA-C-LNPs-CPP, the nuclei were stained with DAPI. The fluorescence intensity of
FITC-siRNA-C taken up into the cells was observed. Colors are indicated as green for siRNA-C
and blue for nuclei. Scale barsꢀ=ꢀ100ꢀµm.
ACS Paragon Plus Environment
2
7

Biomacromolecules
Page 28 of 35
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
CONCLUSIONS
We conclude that siRNA association and dissociation rate constants (kon and koff) can be
engineered by tuning the chemical composition and the structure of the synthetic polymer
nanoparticle. An increase of the amine monomer percentage in the NP does not change the kon but
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
reduces the koff of NPs. Dissociation equilibrium constants (K
NPs did not show a significant difference between pH 5.5 and 7.4. However, di-amine
functionalized NPs showed larger K at pH 7.4 compared with pH 5.5. The relatively large
d
) of mono-amine functionalized
d
dissociation equilibrium constants at pH 7.4 and small dissociation equilibrium constants at pH
.5 results in a high knockdown effect. These results establish the importance of tuning the
5
binding kinetics of NPs for target biomacromolecules for optimum efficacy. This is a first report
to show direct evidence of the correlation between binding kinetics (kon and koff) of the siRNA
delivery vector and knockdown effect.
ACS Paragon Plus Environment
2
8

Page 29 of 35
Biomacromolecules
1
2
3
4
5
6
7
8
9
ASSOCIATED CONTENT
Supporting Information.
The following files are available free of charge via the Internet at http://pubs.acs.org.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Evaluation of NP titration, interaction between cholesterol free siRNA and amine NPs,
interaction between siRNA-C and several percentage of TBAm NPs, time courses of frequency
changes of NP-immobilized QCM after addition of siRNA-C, size distributions of NPs and
NP/siRNA-C-LNP-CPP, siRNA uptake using NP/siRNA-C-LNP-CPP and cytotoxicity of NPs
or NP/siRNA-C-LNP-CPP.
Corresponding Author
Kenneth J. Shea, Department of Chemistry, University of California Irvine, Irvine, CA 92697
USA, E-mail: kjshea@uci.edu
Yu Hoshino, Department of Chemical Engineering, Kyushu University, 744 Motooka, Nishi-ku,
Fukuoka, Fukuoka 819-0395 Japan, E-mail: yhoshino@chem-eng.kyushu-u.ac.jp
Author Contributions
The manuscript was written through contributions of all authors. All authors have given approval
to the final version of the manuscript.
Funding Sources
This research was supported by JSPS and MEXT (20106003, 23111716 and 25107726) and in
part by the National Science Foundation (DMR-1308363).
ACKNOWLEDGMENT
ACS Paragon Plus Environment
2
9