Photoregulation of Gene Expression with Amantadine-Modified Caged siRNAs through Host–Guest Interactions

Article abstract of DOI:10.1002/chem.202003084

RNA interference is an essential and powerful tool for targeting and verifying specific gene functions. Conditional control of small interfering RNA (siRNA) activity, especially using light activation, is a potential method for regulating target gene expr

Full text of DOI:10.1002/chem.202003084

Accepted Article
Accepted Article
Title: Photoregulation of Gene Expression with Amantadine-modified
Caged siRNAs Through Host/Guest Interaction
Authors: Jinhao Zhang, Nannan Jing, Xinli Fan, and XinJing Tang
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.202003084
Link to VoR: https://doi.org/10.1002/chem.202003084
01/2020

10.1002/chem.202003084
Chemistry - A European Journal
FULL PAPER
Photoregulation of Gene Expression with Amantadine-modified
Caged siRNAs Through Host/Guest Interaction
Jinhao Zhang[a], Nannan Jing[a], Xinli Fan[a] and Xinjing Tang *[a]
[a]
J. Zhang, N. Jing, X. Fan, Prof. X. Tang
State Key Laboratory of Natural and Biomimetic Drugs, School of Pharmaceutical Sciences, Peking University
No. 38 Xueyuan Rd. Beijing, China
E-mail: xinjingt@hsc.pku.edu.cn
Abstract: RNA interference (RNAi) is an essential and powerful tool
for targeting and verifying specific gene functions. Conditional control
of siRNA activity, especially using light activation, is a potential
method to regulate target gene expression and functions. Here, a
series of photolabile siRNAs with amantadine modification were
rationally designed and developed through host/guest interaction of
amantadine and β-cyclodextrin derivatives to enhance the blocking
effect of siRNA binding and/or RISC processing. These caged siRNAs
with amantadine modification at the 5’ end of antisense-strand RNA
were efficiently inactivated due to the host/guest interaction of
amantadine and β-cyclodextrin. Photomodulation of gene silencing
activity of these amantadine-modified caged siRNAs targeting both
exogenous and endogenous genes was successfully achieved, which
indicated that host/guest interaction could be a new strategy for
developing new caged siRNAs for gene photoregulation with low
leaking activity.
Light, as a noninvasively and easily operated stimulus, shows
great potential for highly efficient conditional regulation with
spatiotemporal resolution12b. With the introduction of caging
technology, caged siRNAs are expected to precisely regulate
gene functions. Many different caged siRNAs have been
developed through nucleobase caging, phosphate caging, caging
hairpins and siRNA cyclization[12]. Among them, caged siRNAs
involving the attachment of caging groups to the phosphate
backbone were the first type of photolabile siRNAs developed
through a statistical caging strategy. However, the gene silencing
effect of these caged siRNAs was not effectively abrogated due
to their leakage activity[13]. Previously, we developed caged
siRNAs with site-specific attachment of 2-(2-nitrophenyl) propyl
(NPP) to each phosphate backbone of siRNA and found that the
modification at 5' terminus of the antisense RNA strand of siRNA
duplex had a more efficient inhibitory effect on siRNA activity,
while modification at the 3' terminus of antisense RNA strand was
partially tolerated[14]. Different modifications have been further
applied to 5′ terminus of siRNA antisense strand through a
photolinker, and these modifications showed effective
photomodulation of siRNA activities and gene silencing.
Introduction
Small interfering RNAs (siRNAs), double-stranded RNA duplexes
consisting of 20-25 base pairs, play a crucial role in RNA
interference (RNAi)-induced gene silencing[1]. Considering the
high failure rates of traditional drug discovery[2], RNAi-based
nucleic acid drugs demonstrate significant potential for precise
therapeutics, especially for currently undruggable targets[3].
Although patisiran, the first siRNA drug, was approved by FDA in
2018[4], there still exist some challenges associated with siRNA
drugs[5], such as off-target effects[6], low stability[7], and difficult
cellular delivery[8]. To solve these problems, many researchers
have focused on the chemical modification of siRNAs. A variety
of chemical modifications have been applied in recent decades to
optimize the gene silencing effect of siRNA, including siRNA
backbone modification and terminal conjugation of specific
moieties[9]. These modifications achieved great enhancement of
siRNA delivery efficiency and therapeutic effect. In particular,
terminal conjugation of N-acetyl galactosamine (GalNAc) clusters
showed great improvement in targeting hepatocytes due to the
highly specific and efficient interaction of the moiety with its
According to previous reports, large blocking groups on the
5' end of antisense RNA strands could greatly enhance the
photomodulation of caged siRNA activity[15] [16]. Host/guest
interaction may be a new strategy for developing caged siRNAs
only with easy modification of a small guest molecule. -
Cyclodextrin belongs to the cyclodextrin family of polysaccharides,
containing an appropriate hydrophobic cavity that can host
various compounds with high size selectivity and binding affinity
through host/guest interaction, especially amantadine[17]. Both β-
cyclodextrin and amantadine show good biocompatibility and
have been used in clinical applications[18]. Herein, we rationally
designed and developed a new host/guest siRNA caging strategy
involving interaction of
a β-cyclodextrin moiety and an
amantadine-modified photolabile antisense RNA strand at 5
terminus. The binding of amantadine with β-cyclodextrin through
host/guest interaction greatly enhances the blocking effect of
RISC formation and/or RNAi processing until light activation
removes amantadine and the β-cyclodextrin complex.. Based on
this new strategy both exogenous gene (firefly luciferase) and
endogenous gene (Eg5) were targeted to demonstrate
photomodulation of siRNA gene silencing ability using these
caged siRNAs through host/guest interaction.
corresponding
receptor
through
clathrin-dependent
endocytosis[10]. This finding indicates that siRNAs conjugated with
receptors or targeting ligands have considerable potentials for
cellular uptake and distribution.
However, most covalent conjugation of siRNAs may lead to
a partial decrease in siRNA activity if the conjugated moieties
were not removed. To address this problem, conditional
regulation of siRNA activity has received much attention, and
Results and Discussion
many stimulus-responsive moieties have been introduced[11]
.
1
This article is protected by copyright. All rights reserved.

10.1002/chem.202003084
Chemistry - A European Journal
FULL PAPER
To synthesize amantadine modified caged siRNA, we first
vectors (pGL-3 and pRL-TK) together with the native or caged
siRNA duplex were cotransfected into HEK293T cells for
photoregulation of target gene silencing.
coupled
a photolinker phosphoramidite (Scheme S1) and
aliphatic N-hydroxysuccinimide (NHS) esters of alkyl linkers
(Scheme S2 and Scheme S3) to 5’ terminus of the antisense or
sense RNA strands of anti-firefly luciferase siRNA (APAL and
APSL) during RNA solid phase synthesis, as shown in Scheme 1.
Further amantadine was conjugated to RNA on solid phase
through amide bond formation. Two amantadine-modified caged
sense or antisense RNAs with different lengths of aliphatic linker
were then synthesized. After deprotection and HPLC purification,
all obtained RNAs were characterized by ESI-MS analysis (see
Supporting Information).
Figure 1. Photomodulation of firefly luciferase expression in AD-modified siRNA
duplexes with natural sense strands and modified antisense strands
(APAL1/SL) with different concentrations of β-cyclodextrin (β-CD). The
concentration of the siRNA duplex was 5 nM, and the concentrations of β-
cyclodextrin were 0 nM, 25 nM and 250 nM (β-CD:AD ratios are 0:1,5:1 and
50:1, respectively). NC, negative control; PC, positive control.
To demonstrate the effect of the host/guest interaction of
amantadine/β-cyclodextrin on siRNA gene silencing, the
complexes of amantadine-modified caged siRNA duplexes
(APAL1/SL, 5 nM) with β-cyclodextrin at different concentrations
(25 nM and 250 nM, ratios of 1:5 and 1:50) were first applied. The
HEK293T cells transfected with the above control or caged
siRNAs were placed in the dark for 4 h and were then divided into
two groups. One group was kept in the dark, while the other was
irradiated for 3 min. The luciferase activities were measured 44 h
after co-transfection. As shown in Fig.1, cells from both negative
and positive control groups showed almost no difference in firefly
luciferase gene expression with or without light irradiation, which
indicated that 3 min of light irradiation did not have an obvious
toxic effect on cells. As expected, the positive control groups with
native siRNA targeting firefly luciferase exhibited approximately
only 10% gene expression with or without UV irradiation. Only 5′-
end amantadine-modified caged siRNA without β-cyclodextrin
incubation resulted in nearly 80% firefly luciferase activity before
UV irradiation, which suggested that this modification at 5′
phosphate moiety partially disturbed RISC formation or
processing similar to the previous observation with the caging
group[12g]. The siRNA activity was recovered after the removal of
the photolabile group with 3 min light irradiation. For cells treated
with the complexes of amantadine-modified caged siRNAs and 5-
or 50-fold concentration of β-cyclodextrin, approximately 96% and
90% luciferase expression remained before light activation,
respectively, which exhibited greater enhanced inhibition than
amantadine-modified siRNA alone. This result revealed that the
amantadine-modified caged siRNAs were almost fully inactivated
and that target firefly luciferase gene expression was scarcely
silenced in the presence of β-cyclodextrin. The recovery of firefly
luciferase gene silencing further confirmed the light-activated
gene silencing ability of amantadine-caged siRNAs. Compared
with the β-cyclodextrin-free group, the higher inhibition for gene
silencing activity of caged siRNAs modified with amantadine was
Scheme 1. Solid-phase synthesis of amantadine-modified caged
oligonucleotides
To validate the photocleavage of amantadine-modified caged
siRNA, different combination of amantadine-modified single-
stranded sense and antisense RNAs (APAL1/SL, AL/APSL1,
APAL1/APSL1, APAL2/SL, AL/APSL2, and APAL2/APSL2) were
irradiated and analysed with 20% PAGE gels (Fig.S1 and Fig.S2).
For double-stranded duplexes of caged oligonucleotides
(APSL1/SL in lane 2-5, AL/APSL1 in lane 6-9 and ALAP1/APSL1
in lane 10-13), amantadine modification led to reduced mobility in
comparison to the native RNA duplex (AL/SL in lane 1). Upon light
activation (1-5 min irradiation), a new band of uncaged siRNA was
observed (lanes 3~5, 7~9, 11~13 in Fig S1 and Fig S2), which
showed similar mobility to the native control RNA duplex. This gel
shift assay indicated that the cleavage of the photolinker and
amantadine moiety led to the recovery of native RNA. Further
ESI-MS data analysis of these uncaged RNAs confirmed the
formation of native RNA with a terminal phosphate group. In
addition, light irradiation for 3 min under our current conditions
was sufficient for the full recovery of native siRNA, which was
applied to the subsequent experiments.
To evaluate the photomodulation of target gene silencing
using amantadine-modified caged siRNA, amantadine-modified
caged antisense-strand RNA was subsequently hybridized with
its complementary sense-strand RNA to generate siRNA
duplexes. The β-cyclodextrin derivative was then incubated with
amantadine-modified siRNAs for host/guest interaction to enlarge
the blocking group of terminally modified siRNA. The reporter
2
This article is protected by copyright. All rights reserved.

10.1002/chem.202003084
Chemistry - A European Journal
FULL PAPER
probably due to higher binding affinity of amantadine and β-
cyclodextrin before illumination, which indicated that the
host/guest interaction of amantadine/β-cyclodextrin might be
useful for improving the photomodulation efficiency of caged
siRNA.
uptake efficiency of amantadine-modified caged siRNA,
suggesting that 4 μM methyl-β-cyclodextrin could actually bind the
amantadine-modified siRNA and block lipid-raft-mediated
endocytosis of amantadine-modified siRNA.
In order to study the length effect of linkers on gene silencing
of amantadine-modified siRNAs, two amantadine modified caged
antisense strand RNAs with different length alkyl linkage were
subsequently hybridized with their complementary sense strand
RNA for siRNA duplexes. Similar to above study, the reporter
vectors (pGL-3 and pRL-TK) together with native or caged siRNA
duplexes with different linkers were cotransfected into HEK293T
cells for gene silencing evaluation. As expected, the positive
control groups with native siRNA targeting firefly luciferase
exhibited a 10% level of gene expression (Fig.S3). Two
experimental groups with 5′-terminal caged modification of siRNA
(APAL1/SL and APAL2/SL) resulted in little effect on firefly
luciferase activity before UV irradiation, which suggested that the
modification at this position efficiently blocked the 5′ phosphate
moiety and then disturbed the formation or processing of RISC.
The siRNA activity was reactivated after removing the photolabile
group with light irradiation for 3 min. And 4.4- and 4.7-fold
enhancement of photomodulating firefly luciferase gene silencing
activity were achieved for APAL1/SL and APAL2/SL, respectively
(Fig S3). Thus, further cell experiments were mainly carried out
using caged siRNA with the shorter linker 2.
Figure 2. Photomodulation of firefly luciferase expression in siRNA duplexes
mixed with nine different ratios （0:1 to 10:1）of methyl-β-cyclodextrin and
Figure 3. (a) Photomodulation of firefly luciferase activity (with Renilla luciferase
as the internal control) using amantadine-modified caged siRNA (APAL2/SL)
and HACD-17 at different concentrations (CD: AD from 0:1 to 5:1); AL/SL was
used as a positive control (PC); the concentrations of all siRNAs were 5 nM. (b)
Photomodulation of firefly luciferase activity (with Renilla luciferase as the
internal control) using amantadine-modified caged siRNA (APAL2/SL) and
HACD-17. The concentrations of the siRNAs were 2.5 nM, 5 nM, 10 nM, and 15
nM, and the CD to AD ratio was 2:1.
amantadine-modified siRNA (5 nM) after 10
h of incubation before
cotransfection
As we all know, methyl-β-cyclodextrin, a derivative of β-
cyclodextrin, was previously confirmed to be an endocytosis
inhibitor[19].We then investigated the real effect of β-cyclodextrin
derivative on the entrance of amantadine modified caged siRNA.
Methyl-β-cyclodextrin and other cellular entry inhibitors
(chlorpromazine, genistein, and amiloride) were chosen for
investing the uptake pathway of caged siRNA and gene silencing
effect, respectively under inhibitory working concentration, as
shown in Fig.S4 (Supporting Information). These inhibitors
(chlorpromazine, genistein, or amiloride) had little effect on the
cellular entry of amantadine-modified siRNA, whereas the
internalization of the amantadine modified caged siRNA was
effectively decreased by 18% in the presence of 4 μM methyl-β-
cyclodextrin which was previous reported to specifically disrupt
lipid rafts[20]. Methyl-β-cyclodextrin obviously decreased the
To confirm our hypothesis that a suitable ratio of methyl-β-
cyclodextrin and amantadine-modified caged siRNA is essentially
important for the enhancement of their cellular uptake and
subsequent gene silencing, we further investigated the detailed
dose effect of methyl-β-cyclodextrin on photomodulation of the
target gene silencing at relative low concentrations in comparison
to that used in its inhibition assay. The same dual-luciferase
evaluation system with siQuant vectors was applied with firefly
luciferase (pGL-3) as the target gene and renilla luciferase (pRL-
TK) as the internal control. Nine different ratios of methyl-β-
cyclodextrin and amantadine-modified caged siRNA (from 0:1 up
3
This article is protected by copyright. All rights reserved.

10.1002/chem.202003084
Chemistry - A European Journal
FULL PAPER
interactions with HACD, which could reduce the leakage activity
efficiently.
to 10:1) were selected with the amantadine-modified caged
siRNA APAL2/SL at a fixed concentration (5 nM). After methyl-β-
cyclodextrin and the amantadine-modified siRNA were first pre-
incubated for full non-covalent interaction to form the host/guest
complex, the complex was then transfected into cells for the
evaluation of gene silencing of firefly luciferase. As shown in Fig.2,
when the concentration of methyl-β-cyclodextrin increased from 0
to 50 nM, the light activation of gene silencing efficiency induced
an obvious enhancement from 33.5% (0:1) to 14.2% (5:1) of firefly
luciferase activity with the increased ratios of Me-β-CD/AD, then
photomodulation efficiency slowly decreased with higher ratios of
Me-β-CD/AD, although little effect of Me-β-CD/AD modified caged
siRNA on gene silencing was observed before light irradiation.
These results demonstrated that β-cyclodextrin moiety could
assist the entrance of amantadine-modified siRNA at a low
concentration and large excess of β-cyclodextrin moiety actually
saturated the binding site of cell surface which reversed to the
For further verification of the photomodulation of endogenous
gene silencing, amantadine-modified caged siRNAs targeting the
endogenous Eg5 gene with or without the addition of HACD were
also evaluated. As shown in Fig.4a, cells transfected with AE/SE
(PC) exhibited obvious irregular phenotypes of cell nuclei due to
siRNA-induced Eg5 silencing and cell mitosis suspension as well
as arrested cells with two or multiple nuclei. As expected, the
normal shape of cell nuclei was observed in cells transfected with
caged siRNAs (both APAE2/SE and APAE2/SE/HACD) without
light activation and was similar to that of the negative control,
which indicated that amantadine modification of caged siRNA
could efficiently block siRNA activity. However, upon light
exposure for 3 min, irregular cellular phenotypes (bi- or
multinuclear cells) were clearly observed, which were similar to
the phenotypes of the positive control group (AE/SE) (Fig.4a).
This indicated that the amantadine-caged APAE2/SE and
APAE2/SE/HACD were optochemically activated to knockdown
Eg5 gene expression. In addition to nuclear phenotypes, cell
phase arrest was also analysed using flow cytometry analysis
(Fig.S5) and high-content analysis (Fig.6b). For cells transfected
with APAE2/SE, light activation led to over 1.8-fold increase in the
percentage of abnormal cell nuclei compared to cells kept in the
dark, which was similar to the enhancement of the positive control
group (PC, AE/SE).
inhibition of their cellular uptake.
Recently, a series of cyclodextrin-conjugated hyaluronic
acids, namely, HACD-5.6~17 (which contains 5.6~17 β-
cyclodextrin per chain), have been reported to exhibit excellent
siRNA delivery activity and weak side effects[21]. Here, we chose
HACD-17 for amantadine binding and delivery of amantadine-
modified caged siRNA and further investigated siRNA gene
silencing based on the host/guest interaction. HEK293T cells
were co-transfected with two plasmids (pGL-3 and pRL-TK), as
well as siRNA (APAL2/SL and AL/SL) and HACD-17. As shown
in Fig.3a, the expression level of firefly luciferase for all the
APAL2/SL groups without UV irradiation was similar to that of the
negative control group, displaying almost no siRNA gene
silencing activity. However, upon light activation of amantadine-
modified caged APAL2/SL, firefly luciferase was efficiently down-
regulated in cells treated with APAL2/SL. As the ratio of CD (in
HACD) to AD increased from 0 to 5, the relative expression level
(30.4%, 25.4%, 19.2%, 20.6%, 19.3%, and 18.9%) of firefly
luciferase decreased (at ratios of 0:1, 1:1 and 2:1) and reached a
minimum when the ratio of CD to AD was over 2. This observation
demonstrated that the photomodulation of amantadine-modified
siRNA was enhanced with HACD addition, and a CD to AD ratio
of 2 was sufficient to achieve excellent photomodulation, which
was consistent with the previous results shown in Fig.1.
According to the literature, the inhibitory effect of caged
siRNAs was reduced when a high concentration of caged siRNAs
was applied due to unsuccessful blocking of RISC formation
and/or processing, which led to the “leakage” activity before light
activation of caged siRNA. [13] To investigate the leakage activity
of caged siRNA based on the new strategy of host/guest
interaction of amantadine and β-cyclodextrin, we evaluated the
dose dependency of amantadine-modified caged siRNA
incubated with HACD-17 under the optimized CD to AD ratio (2 to
1). HEK293T cells were cotransfected with transfection vectors,
HACD-17 and six gradient concentrations of amantadine modified
caged siRNAs (2.5 nM, 5.0 nM, 10 nM, 15 nM, 20 nM, and 30 nM).
As shown in Fig.3b, the control siRNA effectively down-regulated
the firefly luciferase activity from 46.4% (2.5 nM) to 14.6% (15 nM)
as expected. The cells treated with caged siRNA duplexes without
irradiation maintained approximately 90% firefly luciferase activity
compared to the negative control group with siRNA
concentrations from 2.5 to 30 nM. The inhibition efficiency showed
no obvious decrease as the concentrations of APAL2/SL siRNA
increased, which demonstrated that the leakage activity of
amantadine-modified siRNAs could be inhibited through the
host/guest interactions between amantadine and HACD. In
addition, the knockdown of firefly luciferase was also enhanced
upon light activation with photomodulation ratios from 1.9-fold at
2.5 nM to 7.0-fold at 15 nM. The observed effective
photomodulation of RNAi efficiency using amantadine-caged
modified siRNA was consistent with the specific host/guest
Figure 4. (a) Light activation of the Eg5 gene using NC, PC, and amantadine-
modified caged siRNA (APAE2/SE, APAE2/SE+HACD) in HeLa cells. The cells
were fixed and stained with Hoechst 33342 (blue). (b) DNA content of Hela cells
at G2/M phase in the cell cycle after the cells were treated with AS/SE,
APAE2/SE or APAE2/SE +HACD before or after light activation.
4
This article is protected by copyright. All rights reserved.

10.1002/chem.202003084
Chemistry - A European Journal
FULL PAPER
Eg5 mRNA levels under different treatments were also
analysed to evaluate the photomodulation of amantadine-
modified caged siRNA. The dose dependence of Eg5 siRNA
activity before or after light activation was investigated. As shown
in Fig.5, with the increase of siRNA concentration, caged siRNA
(APAE2/SE) without HACD addition showed a decrease in Eg5
mRNA level down to 70% (at 50 nM) in comparison to the
negative control even without light activation, which demonstrated
the high leakage of siRNA activity with only the amantadine
modification of siRNA. However, only a slight reduction in Eg5
mRNA levels was observed with the addition of HACD (the ratio
of CD to AD was 2). Upon light irradiation, both Eg5 mRNA levels
showed a similar concentration dependence with or without the
addition of HACD, which indicated that the same active form of
siRNA was recovered and showed very similar levels of gene
silencing activity, from 40% at 5 nM caged APAE2/SE to 20% at
50 nM caged APAE2/SE, and photomodulation ratios was
enhanced from 2.4 fold to 5.2 fold at 50 nM with the addition of
HACD. These results confirmed that the host/guest interaction of
CD/AD greatly decreased the leakage of target gene silencing
activity and enhanced the photomodulation of target gene
knockdown using amantadine-modified caged siRNA with the
addition of HACD.
efficiency was not always positively correlated with the ratios of
CD to AD. The transfection complex with a 2:1 ratio of CD (in
HACD) to AD (in AD-modified caged siRNA) showed the highest
efficiency of siRNA cellular uptake among the groups (1:1 and 5:1,
CD to AD), which is consistent with the above studies. Thus, the
correct ratio of CD/AD was very important for promoting the
transfection efficiency of amantadine-modified siRNAs with the
assistance of β-cyclodextrin derivatives. The fluorescence of NC
(AE/Cy3-SE) and amantadine-modified siRNA (APAE2/Cy3-SE)
with HACD-17 incubation (CD to AD 2:1) were analysed using
confocal laser scanning microscopy. After 6 h of incubation, the
cells were fixed and stained with Hoechst 33342 to label the cell
nuclei and LysoTracker Red as a late endosome and lysosome
marker. The Pearson's correlation coefficient between Cy3
fluorescence and LysoTracker Deep Red of negative control
siRNA was 0.347, while the colocalization ability decreased with
amantadine modified siRNA with the low Pearson's correlation
coefficients (0.176 for CD to AD 2:1), which indicated the possible
dissociation of amantadine and cyclodextrin after the complex
was uptakenby cells.
Figure 5. Relative amounts of Eg5 mRNA in HeLa cells treated with 5, 25, or
50 nM APAE2/SE with or without HACD incubation before or after light
activation. GAPDH was used as an internal control.
To
understand
the
HACD-17
enhancement
of
photomodulation of RNAi silencing of endogenous genes, we
quantified the uptake of amantadine-modified caged siRNA
hybridized with Cy3-labelled sense strand RNA. After siRNA
transfection, the uptake efficiency of siRNA was quantified
through analysing Cy3 fluorescence using confocal laser
scanning microscopy and a cytofluorometer. The experimental
groups with CD/AD ratios of 1, 2 and 5 (CD from HACD-17 and
AD from APAE2/Cy3-SE) were investigated. As shown in Fig.6a,
in comparison to the control group without amantadine
modification, up to 50% improvement of cellular uptake was
observed with CD:AD=2:1 for HACD and AD-modified cage Eg5
siRNA. These results indicated that the cellular uptake efficiency
of amantadine-modified caged siRNA was enhanced based on
the non-covalent host/guest interactions between amantadine
and HACD, which suggested the great potential of host/guest-
based delivery systems for siRNA delivery. And the delivery
Figure 6. (a) The relative fluorescence of Cy3 labelled control siRNA (AE/Cy3-
SE) and caged siRNA (APAE2/Cy3-SE) with different ratios of amantadine (AD)
and HACD (CD: AD=1:1, 2:1, 5:1). (b) The fluorescence colocalization images
of Cy3 labelled caged siRNA (APAE2/Cy3-SE) or control siRNA (AE/Cy3-SE)
with HACD-17 (CD: AD=2:1).
Conclusion
In summary, we rationally designed and developed
amantadine-modified caged siRNAs with single amantadine
modification through a photolinker at the 5′ end of antisense RNA
strand. The specific host/guest interaction of amantadine and β-
cyclodextrin derivatives greatly enhanced the blocking effect on
siRNA binding and/or processing. However, more excess of
cyclodextrin (CD) actually exhibited a negative correlation with
5
This article is protected by copyright. All rights reserved.

10.1002/chem.202003084
Chemistry - A European Journal
FULL PAPER
siRNA cellular uptake and target gene silencing due to the
inhibitory effect of excess amount of CD through lipid-raft-
dependent endocytosis. Through optimization, the ratios of AD to
CD moieties from 1:2 to 1:5 were favourable, and effective
photomodulation of target gene silencing were successfully
achieved under the optimal condition. With the assistance of
HACD (CD conjugation on hyaluronic acid), the enhancement of
cellular delivery of AD-modified caged siRNA was observed. In
addition, the host/guest interaction of AD-modified siRNA with CD
on hyaluronic acid further decreased the leakage activity of caged
siRNA to target gene silencing before light activation. Upon light
RNA oligonucleotides were then purified using a Waters
HPLC system (Alliance e2695, with a C18 reversed-phase HPLC
column) with a flow rate of 1.0 mL/min. buffer A: 0.05 M
triethylammonium bicarbonate buffer (TEAB), and buffer B:
acetonitrile. Gradient B follows the procedure: 0–30% in 20 min,
30–100% in 5 min, 100% in 5 min, 100–0% in 5 min, and 0% in 5
min. Purified RNA oligonucleotides were then characterized using
ESI mass analysis in negative mode (Fig S6).
PAGE gel shift assay
irradiation, siRNA activity was restored,
and high
photomodulation efficiency of both the exogenous firefly
luciferase gene and endogenous Eg5 gene was achieved even at
relatively high concentrations of AD-modified caged siRNA in the
presence of HACD. Amantadine is not just a guest molecule but
also an anti-viral and anti-Parkinson’s agent[22], and the design
also provided the possibility of a therapeutic combination of siRNA
and small-molecule drugs. This new caging strategy for
host/guest interaction provided another novel method for
photoregulation of RNAi-induced gene silencing with the
reduction of off-target effects due to the leakage of caged-siRNA
activity.)
The mobility of amantadine-modified caged siRNAs was
studied using 20% native PAGE gels to evaluate the
photocleavage ability of caged siRNAs. Antisense RNA
oligonucleotides (AL, APAL1) and 1.0 equivalent of the
corresponding complementary sense strand (SL, APSL1) were
mixed in PBS to form four siRNA duplexes (AL/SL, APAL1/SL,
AL/APSL1, APAL1/APSL1). The mixtures were heated at 90 °C
for 5 min and then cooled to room temperature for siRNA
hybridization. Four microliters of 1 μM modified siRNA was loaded
to 20% native PAGE gels to analyse the photorelease of caged
siRNAs after light irradiation (365 nm, 1-5 min, 7 mW/cm2). Native
siRNA and caged siRNA with no light irradiation were chosen as
control. All PAGE gels were run at 100 V for 2 h and were then
subjected to ChemiDoc XRS imaging after staining with SYBR
Gold nucleic acid gel stain (Invitrogen) for 5 min. Gel shift assay
of APAL2 and APSL2 and their duplex followed the same
procedure.
Experimental Section
Organic Synthesis and Characterization of photolabile
linker and alkyl linkers
The photolabile phosphoramidite linker (PL) was synthesized
under inert N2 conditions with dry reagents and solvents
according to our previous report[18b]. Solvents were distilled over
CaH2 before use. Silica gel column chromatography was
performed on Merck silica gel. NMR spectra were recorded at 400
MHz (1H) with a Bruker Avance III 400 MHz spectrometer. 31P and
1H NMR spectra were referenced using the external standard
85% H3PO4 and internal standard (CH3)4Si, respectively.
Cell culture
HEK293T and HeLa cells were grown at 37 °C with 5% CO2
in Dulbecco’s modified Eagle’s medium (DMEM) supplemented
with 10% foetal bovine serum (Pan), 2 mM L-glutamine, 100 U/mL
penicillin, and 100 mg/mL streptomycin (Life Technologies,
Gibco). The cells were seeded into 24-well plates at a density of
1 × 105 cells/well and incubated for 24 h before transfection.
Synthesis of amantadine-modified photolabile RNA and
native unmodified RNA was performed using an ABI 394
DNA/RNA synthesizer with 1 mol scale based on standard
phosphoramidite chemistry. PL and alkyl linker phosphoramidites
were successively coupled at 5′ terminus of RNA strands. The
terminal NHS carboxylic ester of RNA strands on the solid phase
was further conjugated with 100 mol amantadine and 50 L of
DIPEA in 1 mL DMF for 24 h at room temperature. The RNA CPG
was then washed with DMF and ethanol three times to remove
the residual amantadine. The modified RNAs on the solid CPG
were then cleaved from CPG and deprotected in 400 µL of an
AMA mixture solution containing 33% ammonium hydroxide
solution and 40% methylamine ethanol solution (v/v=1) for 12 h at
room temperature. After collection and concentration of RNA
oligonucleotides, the obtained oligonucleotide residues were
dissolved in 100 μL of anhydrous DMSO and 100 μL of
triethylamine trihydrofluoride for 6 h at room temperature to
remove t-butyldimethylsilyl group on 2′-hydroxyl group of each
nucleotide. Then, 20 L of 3 M sodium acetate and 300 L of
ethanol were added to the above solutions for RNA
oligonucleotide precipitation. After cooling in a -80 °C freezer for
2 h, the precipitates were collected through centrifugation at
13000 rpm for 10 min at 4 °C.
Dual-luciferase reporter assay for caged siRNA
A dual-luciferase reporter assay was chosen for evaluation of
caged siRNA with firefly luciferase as the target gene and Renilla
luciferase as the internal control. HEK293T cells were incubated
in 24-well plates at a density of 5×104 cells/well and grown to
approximately 80% confluence in 24 h. Complementary strands
were paired to form a 10 μM corresponding siRNA duplex solution.
The prepared siRNA solution was stored at -20 oC for next use.
DMEM in the cell culture plate was replaced with 400 μL of
OptiMEM was added per well. The transfection complex was
added to the cell culture plate, and the final volume in each well
was 500 μL. The concentration of the siRNA duplex (APAL1/SL)
was 5 nM, and the concentration of β-cyclodextrin was 0 nM, 25
nM or 250 nM. Each set of samples was prepared in 6 multiple
wells for a total of 24 wells and was then separated to 2 groups.
The samples were placed in cell incubator, and after 4 hours, one
group of samples were illuminated from the bottom of the plate
with an LED ultraviolet lamp, while the other group of cells were
6
This article is protected by copyright. All rights reserved.

10.1002/chem.202003084
Chemistry - A European Journal
FULL PAPER
not treated. The OptiMEM medium was then replaced with DMEM
containing FBS.
Assay System (Promega) according to the standard protocol.
Firefly luciferase activity was normalized to Renilla luciferase
activity for each condition. All experiments were performed in at
least triplicate, and the error bars represent the standard
deviations in three independent experiments.
HACD-17 was dissolved in RNase-free water to form a 1 μM
HACD-17 solution, meanwhile the concentrations of CD residues
(in HACD) were 17 μM. HACD-17 was then added to siRNAs
(APAL2/SL, AL/SL) solutions to form transfection complexes, the
final concentration of the siRNA duplexes were 5 nM, and the
concentrations of CD residues (in HACD) were 0, 5, 10, 15, 20, or
25 nM, meanwhile the ratios of CD to AD were from 0:1 to 5:1. In
the dose effect of HACD experiment, The concentration of siRNAs
was 2.5 nM, 5 nM, 10 nM, 15 nM, 20 nM or 30 nM, and the CD to
AD ratio was fixed to 2:1 during the dose-effect experiment. The
ratio of Firefly/Renilla luciferase data with or without UV irradiation
were obtained using similar procedures described previously.
After being cultured in the cell incubator for another 20 hours,
the cells were washed with 1×PBS once. The expression levels
of firefly luciferase and renilla luciferase were examined by a
Dual-Luciferase Reporter Assay System using Centro XS3 LB
960 fluorophotometer. The ratio of Firefly/Renilla luciferase was
calculated, and the data were obtained. All experiments were
performed in at least triplicate, and the error bars represent the
standard deviations in three independent experiments.
The AD-modified siRNAs with different linkers was examined
using the same above procedures. The concentration of the
siRNA duplex (APAL1/SL and APAL2/SL) was 5 nM, the cells
transfected with caged siRNAs were placed in the dark for 4 h,
then UV illumination was introduced to irradiate cells for 3 min,
the ratio of Firefly/Renilla luciferase data were examined using
above method after another 20 h incubation.
Uptake inhibition analysis
Phenotype of Eg5 gene silencing
The inhibitors used in these assays were genistein (37 μM,
cavelin-mediated endocytosis), chlorpromazine (42 μM, clathrin-
dependent endocytosis), amiloride (13 μM, mocropinocytosis)
and methyl-beta-cyclodextrin (4 μM, lipid-raft-dependent
endocytosis). The working dose of each inhibitor was chosen
based on previous studies[18b]. Prior to incubation with the
amantadine-modified caged siRNA APAL2/Cy3-SL, cells were
pre-incubated with culture medium containing inhibitors to block
various endocytic pathways. After 1 h of treatment with these
inhibitors, the culture medium was removed and replaced with
solutions containing transfection complex performed as described
previously and fresh inhibitors at the same concentration for
further 5-6 h incubation under a 5% CO2 atmosphere at 37 °C.
Then Cy3 fluorescence intensity was analysed with flow
cytometry (BD FACSAria II).
HeLa cells were incubated in 35-mm laser confocal plates at
a density of 2× 105/well and grown to approximately 50%
confluence in 24 h. Then, cells in each plate were transfected with
natural siRNA or amantadine-modified caged siRNA with the
assistance of Lipofectamine 2000. After 6 h of incubation at 37 °C,
culture plates were divided into two groups. One group was kept
in the dark, and the other was irradiated (3 min, 365 nm). After
replacement of the medium, the cells were incubated for an
additional 42 h at 37 °C. All the cells were fixed with 3.8%
formaldehyde and stained with Hoechst 33342 (Sigma) for 30 min.
Then, the medium was replaced with 1× PBS. Laser scanning
confocal microscope (Nikon, A1R) was used to image cell nuclei
at an excitation wavelength of 405 nm.
Real-time quantitative polymerase chain reaction (RT-qPCR)
of Eg5
The ratio effect on photoregulation of caged siRNA for
firefly luciferase gene silencing with methyl-β-cyclodextrin
derivatives
HeLa cells were incubated in 12-well plates at a density of
1×105 cells/well and grown to approximately 50% confluence
within 24 h. After the similar experimental processing for cells
treated with control siRNA and caged siRNA before and after light
irradiation , all cells were then collected and their total RNAs were
extracted using Trizol reagent (Invitrogen). Eg5 cDNAs were then
reverse-transcribed with HiScript II Q RT SuperMix for PCR
(Vazyme Biotech). GoTaq qPCR Master Mix (Promega) was used
for real-time fluorescence quantitative PCR. The threshold cycles
of each sample were normalized to the GAPDH housekeeping
gene. The following primer sequences were used: Eg5 forward
primer 5′-CAGCTGAAAAGGAAACAGCC, Eg5 reverse primer 5′-
ATGAACAATCCACACCAGCA, GAPDH forward primer 5′-
TGCACCACCAACTGCTTAGC, and GAPDH reverse primer 5′-
GGCATGGACTGTGGTCATGAG.
Methyl-β-cyclodextrin together with amantadine-modified
cage siRNA (5 nM) with different ratios of CD to AD moieties
(CD:AD=0:1, 0.5:1, 1:1, 2:1, 4:1, 5:1, 6:1, 8:1, 10:1) or control
siRNA were first incubated for full complex formation before
transfection. The firefly (200 ng, pGL-3) and Renilla (50 ng, pRL-
TK) reporter plasmids together with the corresponding control
siRNA or caged siRNAs were co-transfected to cells using
Lipofectamine 2000 in OptiMEM medium, and the final
concentrations of both the control siRNA and amantadine-
modified caged siRNA were fixed at 5 nM. After 4 h of incubation,
the wells in 24-well plates with transfected HEK293T cells were
separated to two groups. One group was kept in the dark, while
the other group was irradiated with UV light (365 nm, 7 mW/cm2)
for 3 min to remove the photolabile linkers for the recovery of gene
silencing activity of caged siRNAs. The cells were washed twice
with 1× PBS, and fresh DMEM was then added for cell culture.
After 20 h of incubation, firefly luciferase activity was evaluated
using a high-sensitivity microplate luminometer (Centro XS3 LB
960, Berthold Technologies) and the Dual-Luciferase Reporter
Cell culture
HeLa cells were treated according to the same procedure as
that used for phenotype experiment. After a total 48-h incubation
at 37 °C, all cells were digested with trypsin-EDTA and collected
7
This article is protected by copyright. All rights reserved.

10.1002/chem.202003084
Chemistry - A European Journal
FULL PAPER
Sweetser, P. P. Garg, A. K. Vaishnaw, J. A. Gollob, O. B. Suhr, N. Engl.
J. Med. 2018, 379, 11-21.
through centrifugation (2,000×g, 5 min). Then, the cells were
washed once with 1× PBS and fixed in cold 70% ethanol at 4 °C
for 24 h. The fixed cells were collected by centrifugation (2,000×g,
5 min), followed by resuspension and washing with cold 1× PBS
buffer. Finally, the cells were incubated with propidium iodide and
RNase A solution at 37 °C for 30 min in the dark and were
analysed with flow cytometry (BD FACSAria II). The percentage
of cells in the G2/M phase of the cell cycle was calculated using
ModFitLT 5.0 software.
[5]
[6]
Y. Zhou, C. Zhang, W. Liang, J. Control. Release 2014, 193, 270-281.
a)A. L. Jackson, P. S. Linsley, Nat. Rev. Drug Discov. 2010, 9, 57-67;
b)A. L. Jackson, S. R. Bartz, J. Schelter, S. V. Kobayashi, J. Burchard,
M. Mao, B. Li, G. Cavet, P. S. Linsley, Nat. Biotechnol. 2003, 21, 635-
637.
[7]
a)D. Bumcrot, M. Manoharan, V. Koteliansky, D. W. Y. Sah, Nat. Chem.
Biol. 2006, 2, 711-719; b)J. Haupenthal, C. Baehr, S. Kiermayer, S.
Zeuzem, A. Piiper, Biochem. Pharmacol. 2006, 71, 702-710.
K. A. Whitehead, R. Langer, D. G. Anderson, Nat. Rev. Drug Discov.
2010, 8, 129-138.
[8]
[9]
a)S. T. Crooke, J. L. Witztum, C. F. Bennett, B. F. Baker, Cell Metab.
2018, 27, 714-739; b)C. F. Bennett, E. E. Swayze, Annu. Rev. Pharmacol.
Toxicol. 2010, 50, 259-293.
Subcellular localization analysis by confocal microscopy
A Zeiss LSM880 laser scanning confocal microscope was
used to examine the intracellular distribution of amantadine-
modified caged siRNA in living cells. All images were obtained
using a 100× oil lens and a charge-coupled device (CCD) camera.
Hoechst 33342, Cy3 and LysoTracker Deep Red were excited at
405 nm, 561 nm, and 633 nm, respectively. All images were
analysed using ZEISS ZEN Microscope Software.
Native siRNA (AE/Cy3-SE) and amantadine-modified siRNA
(APAE2/Cy3-SE) with Cy3-labelled sense-strand RNA were used
for the localization study. These duplexes were transfected into
the cells using Lipofectamine 2000 or HACD under the same
conditions. After 4 h incubation, the medium was replaced with
fresh DMEM. After an additional 2h of incubation, the transfected
cells were stained with LysoTracker Deep Red and hoechst
33342 for further imaged. Colocalization of AE/Cy3-SE or APAE2/
Cy3-SE with LysoTracker Deep Red was investigated via
fluorescence confocal microscopy.
[10] A. A. D’Souza, P. V. Devarajan, J. Control. Release 2015, 203, 126-139.
[11] a)Y. Li, J. Yang, B. Xu, F. Gao, W. Wang, W. Liu, ACS Appl. Mater.
Interfaces 2015, 7, 8114-8124; b)M. Gujrati, A. Vaidya, Z.-R. Lu,
Bioconjug Chem. 2016, 27, 19-35; c)A. Kala, S. H. Friedman, Pharm.
Res. 2011, 28, 3050-3057; d)W. Zhou, W. Brown, A. Bardhan, M.
Delaney, A. S. Ilk, R. R. Rauen, S. I. Kahn, M. Tsang, A. Deiters, Angew
Chem Int Ed Engl 2020, 59; e)E. V. Moroz-Omori, D. Satyapertiwi, M.-C.
Ramel, H. Høgset, I. K. Sunyovszki, Z. Liu, J. P. Wojciechowski, Y.
Zhang, C. L. Grigsby, L. Brito, L. Bugeon, M. J. Dallman, M. M. Stevens,
ACS Cent. Sci. 2020.
[12] a)M. Vera, H. Alexander, RNA 2007, 13, 2341-2347; b)X. Ouyang, I. A.
Shestopalov, S. Sinha, G. Zheng, C. L. W. Pitt, W.-H. Li, A. J. Olson, J.
K. Chen, J. Am. Chem. Soc. 2009, 131, 13255-13269; c)L. Zhang, D.
Liang, Y. Wang, D. Li, J. Zhang, L. Wu, M. Feng, F. Yi, L. Xu, L. Lei, Q.
Du, X. Tang, Chem. Sci. 2018, 9, 44-51; d)Q. Liu, A. Deiters, Acc. Chem.
Res. 2014, 47, 45-55; e)J. Hemphill, Q. Liu, R. Uprety, S. Samanta, M.
Tsang, R. L. Juliano, A. Deiters, J. Am. Chem. Soc. 2015, 137, 3656-
3662; f)J. Hemphill, J. Govan, R. Uprety, M. Tsang, A. Deiters, J. Am.
Chem. Soc. 2014, 136, 7152-7158; g)L. Wu, F. Pei, J. Zhang, J. Wu, M.
Feng, Y. Wang, H. Jin, L. Zhang, X. Tang, Chemistry 2014, 20, 12114-
12122.
[13] Y. Matsushita-Ishiodori, T. Ohtsuki, Acc. Chem. Res. 2012, 45, 1039-
1047.
Acknowledgements
[14] Y.-L. Chiu, T. M. Rana, Mol. Cell 2002, 10, 549-561.
[15] P. K. Jain, S. Shah, S. H. Friedman, J. Am. Chem. Soc. 2011, 133, 440-
446.
We thank Prof. Yu Liu and Prof. Yong Chen from Nankai
University for their HACD. This work was financially supported by
the National Natural Science Foundation of China (Grants No.
81821004, 21877001, 21672015), National Major Scientific and
Technological Special Project for “Significant New Drugs
Development” (Grant No. 2017ZX09303013)
[16] a)Q. N. Nguyen, R. V. Chavli, J. T. Marques, C. P. Jr, D. Wang, W. He,
B. E. Belisle, A. Zhang, L. M. Pastor, F. R. Witney, BBA - Biomembranes
2006, 1758, 394-403; b)Y. Ji, J. Yang, L. Wu, L. Yu, X. Tang, Angew.
Chem. Int. Ed. Engl. 2016, 55, 2152-2156; c)J. Yang, C. Chen, X. Tang,
Bioconjugate Chem. 2018, 29, 1010-1015; d)L. Yu, D. Liang, C. Chen, X.
Tang, Biomacromolecules 2018, 19, 2526-2534; e)L. Yu, N. Jing, Z.
Yang, L. Zhang, X. Tang, Org. Biomol. Chem. 2018, 16, 7029-7035; f)L.
Zhang, C. Chen, X. Fan, X. Tang, Chembiochem 2018, 19, 1259-1263.
[17] G. Wenz, B.-H. Han, A. Müller, Chem. Rev. 2006, 106, 782-817.
[18] a)A. Harada, Y. Takashima, H. Yamaguchi, Chem. Soc. Rev. 2009, 38,
875-882; b)T. Yamamoto, T. L. Yaksh, Pain 1992, 49, 121-128; c)W. R.
Epa, U. Greferath, A. Shafton, P. Rong, L. M. Delbridge, A. Bennie, G. L.
Barrett, Antisense Nucleic Acid Drug Dev. 2000, 10, 469-478.
[19] M. L. Torgersen, G. Skretting, D. B. van, K. Sandvig, J. Cell Sci. 2001,
114, 3737-3747.
Keywords: caged oligonucleotide• photoactivation • gene
regulation •host/guest interaction • siRNA leaking activity
[1]
[2]
[3]
[4]
a)Y. Tomari, Genes Dev. 2005, 19, 517-529; b)D. E. Golden, V. R.
Gerbasi, E. J. Sontheimer, Mol. Cell 2008, 31, 309-312; c)C. L. C. Noland,
E. E. Ma, J. A. J. Doudna, Mol. Cell 2011, 43, 110-121.
a)J. Woodcock, R. Woosley, Annu. Rev. Med. 2008, 59, 1-12; b)C. V.
Dang, E. P. Reddy, K. M. Shokat, L. Soucek, Nat. Rev. Cancer 2017, 17,
502-508.
[20] I. A. Khalil, K. Kogure, H. Akita, H. Harashima, Pharmacol. Rev. 2006,
58, 32-45.
[21] Y. Yang, Y.-M. Zhang, Y. Chen, J.-T. Chen, Y. Liu, J. Med. Chem. 2013,
56, 9725-9736.
a)M. Manoharan, American Chemical Society, 2010, pp. CARB-31; b)G.
Ozcan, B. Ozpolat, R. L. Coleman, A. K. Sood, G. Lopez-Berestein, Adv.
Drug Deliv. Rev. 2015, 87, 108-119.
[22] J. Stanicova, P. Kovalcik, L. Chinsky, P. Miskovsky, Vib. Spectrosc. 2001,
25, 41-51.
a)C. Heras-Palou, Nature 2019, 574, S7; b)S. M. Hoy, Drugs 2018, 78,
1625-1631; c)D. Adams, A. Gonzalez-Duarte, W. D. O'Riordan, C. C.
Yang, M. Ueda, A. V. Kristen, I. Tournev, H. H. Schmidt, T. Coelho, J. L.
Berk, K. P. Lin, G. Vita, S. Attarian, V. Plante-Bordeneuve, M. M. Mezei,
J. M. Campistol, J. Buades, T. H. Brannagan, III, B. J. Kim, J. Oh, Y.
Parman, Y. Sekijima, P. N. Hawkins, S. D. Solomon, M. Polydefkis, P. J.
Dyck, P. J. Gandhi, S. Goyal, J. Chen, A. L. Strahs, S. V. Nochur, M. T.
8
This article is protected by copyright. All rights reserved.

10.1002/chem.202003084
Chemistry - A European Journal
FULL PAPER
Entry for the Table of Contents
A new type of caging strategy with host/guest interaction was developed using amantadine modified
caged siRNAs. The host/guest interaction of amantadine and β-cyclodextrin derivatives greatly enhanced
the blocking effect of RISC formation and/or processing. Photoregulation of both exogenous and
endogenous gene silencing was successfully achieved with low leaking activity
9
This article is protected by copyright. All rights reserved.