A Cationic Micelle as In Vivo Catalyst for Tumor-Localized Cleavage Chemistry

Article abstract of DOI:10.1002/anie.202106526

The emerging strategies of accelerating the cleavage reaction in tumors through locally enriching the reactants is promising. Yet, the applications are limited due to the lack of the tumor-selectivity for most of the reactants. Here we explored an alternative approach to leverage the rate constant by locally inducing an in vivo catalyst. We found that the desilylation-induced cleavage chemistry could be catalyzed in vivo by cationic micelles, and accelerated over 1400-fold under physiological condition. This micelle-catalyzed controlled release platform is demonstrated by the release of a 6-hydroxyl-quinoline-2-benzothiazole derivative (HQB) in two cancer cell lines and a NIR dye in mouse tumor xenografts. Through intravenous injection of a pH-sensitive polymer micelles, we successfully applied this strategy to a prodrug activation of hydroxyl camptothecin (OH-CPT) in tumors. Its “decaging” efficiency is 42-fold to that without cationic micelles-mediated catalysis. This micelle-catalyzed desilylation strategy unveils the potential that micelle may act beyond a carrier but a catalyst for local perturbing or activation.

Full text of DOI:10.1002/anie.202106526

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Cationic Micelle as An In Vivo Catalyst for Tumor-Localized
Cleavage Chemistry
Authors: Chunhong Wang, Hanyu Hong, Mengqi Chen, Zexuan Ding,
Yuchen Rui, Jianyuan Qi, Zi-Chen Li, and Zhibo Liu
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: Angew. Chem. Int. Ed. 10.1002/anie.202106526
Link to VoR: https://doi.org/10.1002/anie.202106526

10.1002/anie.202106526
Angewandte Chemie International Edition
RESEARCH ARTICLE
Cationic Micelle as An In Vivo Catalyst for Tumor-Localized
Cleavage Chemistry
Chunhong Wang,[a] Hanyu Hong, [a] Mengqi Chen, [a] Zexuan Ding, [a] Yuchen Rui, [a] Jianyuan Qi, [a] Zi-
Chen Li, [c] Zhibo Liu*[a,b]
Abstract: The emerging strategies of accelerating the cleavage
reaction in tumor through locally enriching the reactants is promising.
Yet, the applications are limited due to the lack of the tumor-selectivity
for most of the reactants. Here we explored an alternative approach
to leverage the rate constant by locally inducing an in vivo catalyst.
We found that the desilylation-induced cleavage chemistry could be
catalyzed in vivo by cationic micelles, and accelerated over 1400-fold
under physiological condition. This micelle-catalyzed controlled
release platform is demonstrated by the release of a 6-hydroxyl-
quinoline-2-benzothiazole derivative (HQB) in two cancer cell lines
and a NIR dye in mouse tumor xenografts. Through intravenous
injection of a pH-sensitive polymer micelles, we successfully applied
this strategy to a prodrug activation of hydroxyl camptothecin (OH-
CPT) in tumors. Its “decaging” efficiency is 42-fold to that without
cationic micelles–mediated catalysis. This micelle-catalyzed
desilylation strategy unveils the potential that micelle may act beyond
a carrier but a catalyst for local perturbing or activation.
tumor-selectivity of the two reactants, that have limited their
applications. Here we explored the idea of whether the reaction
rate could be leveraged by locally manipulating the rate constant,
which would be feasible to catalyze many bioorthogonal reactions
in vivo.
Micelles or liposomes have been widely used in the deliveries
of small molecule drugs[6], nucleic acids[7], bacterial effectors[8]
and proteins[9]. These nano-carriers could be featured with tissue-
selectivity by tuning up their compositions[10] or decorating
receptor-based targets[11]. Interestingly, micelle has also been
widely used in accelerating a couple of organic reactions in
water[12], which often face the loss of reactivity when induced into
the water phase from the organic solvent. It is a combinational
effect synergizing from hydrophobic effect, enrichment effect and
solvent effect[13]. If the local rate constant of a bioorthogonal
cleavage chemistry could be increased by a tumor-enriched
micelle, it would provide an ideal means to release the “client”
molecule or rescue the function of protein in a tumor-selective
manner.
Tuning the reaction rate in a tissue-selective manner is highly
demanding for both biomedical research and clinical usage, yet is
still an unmet challenge. [1] In the lab, a chemist could adjust the
reaction rate by changing the temperature, reaction medium,
substrate concentration or adding catalyst and etc.[2]
Nevertheless, the strategies that could control the reaction rate in
vivo are limited, let alone in a tissue-selective manner. Most of the
bioorthogonal cleavage chemistries[3] involve two reaction
components, a “caged” functional molecule and a “decaging”
reagent[4], therefore their reaction rates often conform to the law
of the second-order reaction. Recently, strategies that could co-
locally enrich two reaction components in the tissue or organelle
Previous work:
This work:
X
Without micelle
Cationic micelle
of interest (e.g. tumor[5] or mitochondrion[1c]
) have been
Scheme 1. Micelle-catalyzed Phe-BF3-mediated desilylation enabled controlled
release of functional molecules and is remarkably faster than that without
micellar catalysis.
established, allowing to leverage the reaction rate in a tumor-
selective manner. However, these strategies depend on the
Recently, we established a novel bioorthogonal chemistry, in
which a cell-enterable amino acid mimetic Phe-BF3 [14] specifically
and efficiently desilylates and “cleaves” an optimized silyl ether-
containing carbamate linker, which enables tumor-selective
activation of gasdermin and then triggers potent T cell-mediated
antitumor immunity[5]. Here we are pleased to offer a more
advanced strategy that utilizing micellar catalysis to manipulate
the local rate constant to control the bioorthogonal cleavage
chemistry in the tumor. Under physiological conditions, TBDPS
(tert-Butyldiphenylsilyl) is generally the most robust silyl ether
protecting group towards fluoride-mediated desilylation.[15] The
desilylation rate constant of TBDPS is about 1/1,000,000 that of
de-protecting TMS and 1/100 that of de-protecting TBS,
respectively. Herein, we choose Phe-BF3-mediated TBDPS
deprotection as a proof of concept to study the catalytic efficacy
of micelle in test tubes, living cells and tumor-bearing mice. We
found that micelle-catalyzed desilylation is highly efficient
(Scheme 1). Without micellar catalysis, desilylation of a TBDPS-
[a]
C. H. Wang, H. Y. Hong, M. Q. Chen, Dr. Z. X. Ding, Y. C. Rui, J. Y.
Qi, Prof. Dr. Z. B. Liu
College of Chemistry and Molecular Engineering, Peking University,
Beijing, 100871 (China)
Radiation Chemistry Key Laboratory of Fundamental Science,
Beijing National Laboratory for Molecular Sciences, E-mail:
zbliu@pku.edu.cn.
Prof. Dr. Z. B. Liu
Peking University-Tsinghua University Center for Life Sciences,
Beijing, 100871 (China)
[b]
[c]
Prof. Dr. Z. C. Li,
Beijing National Laboratory for Molecular Sciences, Key Laboratory
of Polymer Chemistry & Physics of Ministry of Education,
Department of Polymer Science & Engineering, College of
Chemistry and Molecular Engineering, Center for Soft Matter
Science and Engineering, Peking University
Beijing 100871 (China)
Supporting information for this article is given via a link at the end of
the document.
1
This article is protected by copyright. All rights reserved.

10.1002/anie.202106526
Angewandte Chemie International Edition
RESEARCH ARTICLE
protecting molecule hardly occur with 10 mM Phe-BF3 in 12 hours,
but readily finishes in 30 min when treated with cationic
micelles+10 mM Phe-BF3 (Scheme 1 and Figure S1). Besides,
in the aspect of prodrug design, the more stable the “caging”
group, the less off-target release it would cause to normal
tissues.[16] Thus, we expect that micellar catalysis–activated
prodrug, especially caged by TBDPS, would exhibit less side-
effect and more selectivity for tumor-targeting drug release.
micelle and reactant are often counter-charged: anionic surfactant
micelles often accelerate metal cation–mediated coupling
reaction[17] and cationic micelles often catalyze the fluoride-
mediated desilylation[18]
.
Though trifluoroborate (-BF3) is
negatively charged, Phe-BF3 is neutral. To find the “perfect match”
that could catalyze Phe-BF3-mediated desilylation, we screened
four micelles formed by the surfactants with different net charges,
including cationic CTAB (Cetyltrimethylammonium bromide),
anionic SDS (Sodium dodecyl sulfate), zwitterionic DPC
A
(Dodecylphosphocholine)
and
nonionic
Brij®
C10
(Polyoxyethylene (10) cetyl ether). To ensure micelle formation,
the concentration of these surfactants in phosphate buffer saline
(PBS) is 2 mM or 10 mM, which is higher than their corresponding
PBS
19]
critical micelle concentrations (CMC).[13, , The stability of
TBDPSO-coumarin was evaluated in the above micelles prior to
treated by Phe-BF3 (Figure S2, dotted curve), indicating that the
reaction was triggered by Phe-BF3 not the micelle itself. As shown
in Figure 1B, intense fluorescence was detected within 10 min
after treating TBDPS-coumarin+Phe-BF3 with cationic CTAB
micelle. Meanwhile, almost no fluorescence was observed from
other samples that were treated with anionic SDS micelle,
zwitterionic DPC micelle, and non-ionic Brij® C10 micelles or
without micelle in PBS. More quantitatively, the kinetic profiles of
micelle-catalyzed desilylation were further investigated according
to the fluorescence increase (Figure 1C). The curves fit well with
the second-order reaction and the rate constants are summarized
in Figure 1D. The second-order rate constant of Phe-BF3-
desilylation in cationic CTAB micelle is 3.13 ± 0.31 M-1·s -1, which
is over 3-magnitude larger than that without micellar catalysis (i.e.
0.0021 ± 0.0002 M-1·s -1). The rate constants in anionic SDS
micelle, non-ionic Brij® C10 micelle and zwitterionic DPC micelle
are 0.0048 ± 0.0023 M-1·s -1, 0.014 ± 0.003 M-1·s -1 and 0.027 ±
0.001 M-1·s -1, respectively. These values are 2-12 folds larger
than that of micelle free systems, yet are much smaller than that
of cationic micelle–catalyzed system. As summarized in the
bottom line of Figure 1D, if the rate constant without micellar-
catalysis is standardized as 1.00, the catalytic efficacy of cationic,
zwitterionic, non-ionic and anionic micelles would be 1490 ± 150,
12.8 ± 0.5, 6.67 ± 1.42 and 2.28 ± 1.09, respectively. This result
corroborates with our hypothesis that cationic micelle might be the
perfect match to catalyze Phe-BF3-mediated desilylation.
The stability of the "caging" group is another key issue in
developing the cleavage chemistry needed for in vivo
manipulation.[20] In general, a stable “caged” molecule should
remain >90% in 24 hours.[5] In the above study, we found that
though TBDPSO-coumarin is stable in PBS (Figure S1), its
stability in CTAB is insufficient (Figure S2). Then, we screened
various TBDPS-protected phenols to study the structure-stability
relationship. As shown in Figure 2A, the substitutions of these
compounds include strong electron-withdrawing groups
(compound 1-4), mild or weak electron-withdrawing groups
(compound 5-13), no substitution (compound 14), and electron-
donating groups (compound 15). The relationship between a
given substituent and its effect on the pKa of a phenol represents
one of the fundamental correlations in physical organic chemistry.
We wondered if the hydrolytic stability of TBDPS-protected
phenol was related to its pKa value. If so, knowledge of the pKa
of a given phenol can be used to rationally design TBDPS-
protected phenol with predictable stability.
Micelle
B
D
C
1.2
1.0
0.8
0.6
0.4
0.2
0.0
CTAB
DPC
Brij® C10
SDS
1
2
3
4
5
NONE
0
10 20 30
Time / min
Rate constants of Phe-BF3-mediated desilylation
Brij®
Types
CTAB
DPC
SDS
NONE
C10
k
3.13
0.027 0.014 0.0048 0.0021
± 0.31 ± 0.001 ± 0.003 ± 0.0023 ± 0.0002
/ M-1.s-1
1490
± 150
12.8
6.67
2.28
k / k NONE
1.00
± 0.5 ± 1.42 ± 1.09
Figure 1. A cationic CTAB micelle was identified as the catalyst that can
accelerate Phe-BF3-mediated desilylation by >1000-fold. (A) Schematic
representation of micelle-catalyzed desilylation releases coumarin from
TBDPSO-Coumarin. (B) Photographs showing the fluorescence from TBDPSO-
Coumarin samples treated with 0.50 mM Phe-BF3 and cationic CTAB micelle
(1), zwitterionic DPC micelle (2), neutral Brij® C10 micelle (3), anionic SDS
micelle (4) and no surfactant (5), respectively. The concentrations of the
surfactants 1-3 and 4 are 2.0 mM and 10.0 mM, respectively. And, the reaction
time is 10 min. (C) Fluorescence-time curve of TBDPSO-Coumarin treated by
0.50 mM Phe-BF3 and the indicated micelles (2 mM or 10 mM surfactants in
PBS, pH=7.4) (D) Rate constants summary of micelle-catalyzed Phe-BF3-
mediated desilylation.
To explore this micelle-catalyzed desilylation strategy, we
firstly constructed a TBDPS-“caged” coumarin, in which the
TBDPS group is directly linked to the electron-donating phenolic
oxygen (Figure 1A). Free coumarin, as opposed to the oxygen-
blocked coumarin in the silyl-phenolic ether system, is highly
fluorescent, thus providing a simple and quantitative assay for
micelle-catalyzed desilylation. Phe-BF3-mediated desilylation
follows a mechanism that involves no free fluoride intermediate,
and the fluoride is transferred directly from the trifluoroborate to
the empty d-orbital of the silicon atom on the silyl ether. Therefore,
an environment that could stabilize the fluoride-transfer transition
state, cationic micelle might be preferred, would decrease the
activation energy and accelerate the desilylation.
As reported previously, the net charge of micelle surfactant
often greatly impacts the efficacy of micellar catalysis. The paired
2
This article is protected by copyright. All rights reserved.

10.1002/anie.202106526
Angewandte Chemie International Edition
RESEARCH ARTICLE
1.2
1.0
0.8
0.6
0.4
0.2
0.0
A
B
C
6
7
8
9
11
12
13
14
1
2
3
4
5
1
2
3
4
10
15
0
3 24
36
48
60
5
9
6
7
8
Time / h
2000
1000
5 6
13
8
11
10
9
7
10
11
12
100
3
2
1
0
3
2
4
13
14
15
1
D
6
7
8
9
10
11
pKa of various phenols
The stability of TBDPS-protected phenols and the pKa values of corresponding phenols
No. pKa
kw (h-1)
t1/2 (h)
0.23 ± 0.03
1.5 ± 0.4
1.6 ± 0.5
1.2 ± 0.2
No. pKa
kw (h-1)
t1/2 (h)
No. pKa
11 9.41
12 9.65
13 9.89
14 9.99
15 10.26
kw (h-1)
t1/2 (h)
3.1
0.45
0.42
0.58
± 0.2
6
7
8
9
8.73
8.95
9.29
9.33
(7.1 ± 0.4) ×10-4 (9.8 ± 1.4) ×102
(2.5 ± 0.4) ×10-3 (3.5 ± 1.4) ×102
(5.1 ± 0.4) ×10-4 (1.3 ± 0.5) ×103
(1.0 ± 0.1) ×10-3 (6.8 ± 2.6) ×102
(8.5 ± 1.0) ×10-4 (8.1 ± 2.9) ×102
(7.2 ± 0.5) ×10-4 (9.6 ± 1.2) ×102
1
2
3
4
5
7.15
± 0.05
± 0.05
± 0.05
7.23
n.d.
n.d.
7.4
(4.9 ± 0.4) ×10-4 (1.4 ± 0.3) ×103
7.8
n.d.
n.d.
n.d.
n.d.
± 0.8) ×10-4 (8.8 ± 3.1) ×102
8.51 (7.9
10 9.37
Figure 2. A structure-stability relationship study of TBDPS-protected phenols in CTAB. (A) Chemical structures of TBDPS-protected phenols with different
substituents. (B) Time-stability curves of compound 1-15 in CTAB solution. (C) A correlation-analysis about the half-lives of TBDPS-protected phenols in CTAB
and the pKa values of the corresponding phenols. (D) Data summary. The pKa values were referred to “CRC Handbook of Chemistry and Physics, the 91st Edition”
[21]. The general experimental condition is 0.10 mM of substrates in PBS of 2mM CTAB solution. n.d., not detected.
A systematic kinetic study was performed to investigate their
hydrolytic stabilities. The hydrolysis curve follows a pseudo first-
order reaction (Figure 2B). The half-lives and kinetic constants of
TBDPS-protected phenols and the pKa values of their
corresponding phenols are summarized in Figure 2D. If a
molecule remains more than 90% after 24 hours, it means that
the corresponding half-life should be over 170 hours. The half-
lives of compound 1-4 in CTAB are 0.23 ± 0.03 h, 1.5 ± 0.4 h, 1.6
± 0.5 h, 1.2 ± 0.2 h, respectively (Figure 2B and Figure 2D). They
cannot afford the stability required for the in vivo cleavage
chemistry. To our satisfaction, the hydrolysis rates of rest
compounds are remarkably slower (Figure 2B). The half-lives of
compound 5-15 are above 300 hours, and hydrolytic products of
12, 14, and 15 are almost negligible (Figure 2D). Interestingly,
the pKa values of corresponding phenols of compounds 1-4 are
below 8, while the rest are above 8.5 (Figure 2C). Therefore, we
conclude that a TBDPS-protected phenol should be stable
enough for micelle-catalyzed desilylation when its corresponding
pKa is >8.5.
probe to monitor the progress of the desilylation.[22] Meanwhile, to
expand the strategy to living systems, we should solve the
problem of the bioorthogonality and the biocompatible catalytic
efficiency of the micellar catalytic reaction. We firstly assayed the
bioorthogonality of micelle-catalyzed desilylation by treating
TBDPSO-HQB (2 µM) with various biologically relevant molecules
at 37 °C in 2 mM CTAB PBS solution. No false-positive
fluorophore response was observed at 2 hours in the absence of
Phe-BF3 and these will not affect the decaging reaction induced
by Phe-BF3 (Figure 3B). The impacts on catalytic efficacy from
counter anions of micelles are presented as in Figure 3C, and no
significant difference has been observed. Interestingly, as shown
in Figure3D, we find that the chain length of surfactant has great
impact on catalytic efficiency. The desilylation rate constant in
2mM dodecyl trimethylammonium bromide (DTAB, C12) dropped
to 0.06 ± 0.012 M-1·s -1, which is only half of that in 2 mM tetradecyl
trimethyl ammonium bromide (TTAB, C14). Almost no fluorophore
has been detected in octyl trimethylammonium bromide (OTAB,
C8) or butyl trimethylammonium bromide (BTAB, C4). It is known
that the CMC values are often chain length–dependent. The CMC
values of DTAB, OTAB and BTAB in PBS are 20 mM, 600 mM
Hydroxyquinoline is an important pharmaceutical synthon
and has been the core structure for a couple of FDA-approved
anti-cancer drugs, including Topotecan, Oxyquinoline, Chloroxine, and above 1.0 M, respectively, which are 1-3 magnitudes over
Tilbroquinol, etc. In addition, the pKa of 6-hydroxyquinoline is
about 8.8. To further expand the substrate scope of this micelle-
catalyzed desilylation, we replaced the model compound with the
TBDPS-“caged” hydroxyquinoline derivatives. As presented in
Figure 3A, TBDPSO-HQB was synthesized as a fluorogenic
that of CTAB (Figure S5). The above data suggests that freely
dissolved cationic unimers would not catalyze this desilylation-
mediated approach. In addition, we have also tried to use
polyethyleneimine (PEI) to catalyze the decaging reaction.
Though it carries more positive charges, no released
3
This article is protected by copyright. All rights reserved.

10.1002/anie.202106526
Angewandte Chemie International Edition
RESEARCH ARTICLE
fluorescence was observed (Figure S6). In conclusion, we would
suggest that disassembled cationic unimers can not catalyze
Phe-BF3-mediated desilylation, and their abilities to form the intact
micelles are critical for catalytic efficiency.
(grey bar, Figure 3F). Therefore, though working well in vitro,
CTAC micelle could not catalyze desilylation in vivo. To discover
a cationic surfactant with low cytotoxicity and low CMC, we have
evaluated
two
alternative
candidates:
dihexadecyl
We then wondered whether micelle-catalyzed desilylation
could be developed into a useful controlled release system,
dimethylammonium chloride (DHAC) and 1,2-dioleoyl-3-
trimethylammonium-propane chloride (DOTAP), by the same set
of assays. As shown in Figure 3E and 3F, DOTAP micelle shows
better biocompatibility and higher catalytic efficiency to Phe-BF3-
mediated desilylation than both DHAC and CTAC. In addition, it
has been widely used for gene/nucleic acid delivery in preclinical
studies and clinics[7], so DOTAP micelle was chosen as the
making micelle as
a nano-enzyme for in vivo catalysis.
Unfortunately, the progress of pursuing this idea has been
postponed by the unacceptable cytotoxicity of CTAC. As shown
in Figure 3E and Figure S7, the cell viability of 12 µM CTAC is
below 20 % for both BGC823 and U87. But its catalytic efficiency
dropped sharply when its concentration is below 100 μM (grey bar, catalyst to accelerate the Phe-BF3-mediated desilylation in living
Figure 3F). No reaction occurred in 4 hours when the Phe-BF3-
cells and tumor-bearing mice.
mediated desilylation is incubated with 12 μM or 25 µM CTAC
A
Cationic micelle
TBDPSO-HQB
HQB (neutral form)
HQB (anion form)
B
C
D
N.S.
**
0.25
0.20
0.15
0.10
0.05
0.00
0.25
0.20
0.15
0.10
0.05
0.00
2000
1500
1000
500
0
N.R.
BGC823
***
**
E
F
****
***
****
***
DOTAP
DOTAP
1.2
1.0
0.8
0.6
0.4
0.2
0.0
1.0
**
DHAC
CTAC
DHAC
CTAC
***
****
***
0.8
0.6
0.4
****
****
N.R.
N.R.
12 μM
25 μM
100 μM
100 μM
25 μM
12 μM
Figure 3. Optimization of the cationic micelles for better biocompatibility and higher catalytic efficiency to Phe-BF3-mediated desilylation. (A) Schematic
representation of Phe-BF3-induced desilylation that can release the fluorescence-“caged” 6-hydroxyl-quinoline-2-benzothiazole (HQB). (B) Fluorescence-emission
assay of possible desilylation of TBDPSO-HQB (2 μM) by various biologically relevant molecules (2 mM for Phe-BF3, Glucose, GSH, Cys; 0.5 mg/ml for HA,
hyaluronic acid, 4k; 1 ng/L for plasmid pUC19; and ‘All’ means the combination of above reagents). The HQB fluorescence intensity was determined at 37°C in 2
mM CTAB phosphate buffered solution with λex = 360 nm and λem = 535 nm (n = 3). (C) Rate constant analysis of micelle-catalyzed desilylation treated by 2 mM
cyltrimethylammonium with bromide (CTAB), sulfonate (CTAS) and chloride (CTAC) as the counter anions (n = 3). (D) Rate constant analysis of micelle-catalyzed
desilylation treated by 2 mM alkyl trimethylammonium bromide with different side chain length (n = 4). (E) Cytotoxicity of 12 μM, 25 μM and 100 μM of DOTAP,
DHAC and CTAC against BGC823 cells, respectively (n = 3). (F) The rate constant ratios of Phe-BF3-mediated desilylation catalyzed by 100 μM, 25 μM and 12 μM
of indicating surfactants to 2mM of the micelles, respectively. N.S., no significant difference, N.R. no reaction. Data are shown as mean ± s.d.; two-tailed unpaired
Student’s t-test was performed (**P < 0.01, ***P < 0.001, ****P < 0.0001). The above assays were performed at 37 ℃ without further notification.
To test whether micellar catalysis works in living cells, we
pre-treated BGC823 and U87 cancer cells with 20 μM TBDPSO-
HQB or HQB. The cells were rinsed with PBS after incubation for
2 hours, followed by treatment with 40 µM DOTAP and 200 µM
Phe-BF3. The cell viability of two cell lines with 40 µM DOTAP,
was evaluated prior to the kinetics study, and no obvious
cytotoxicity has been observed (Figure S7). In addition,
TBDPSO-HQB and Phe-BF3 have also been tested, and show no
toxic to the two cells (Figure S8). As illustrated in Figure 4A, the
combinational treatment of the TBDPSO-HQB+Phe-BF3+DOTAP
successfully release the “caged” fluorophore and highlight the
cells, while almost no fluorescence was captured in the groups
treated with TBDPSO-HQB+DOTAP, TBDPSO-HQB+Phe-BF3
and TBDPSO-HQB only. After subtracting the cell auto-
fluorescence background, the fluorescence intensity of the
TBDPS-HQB+Phe-BF3+DOTAP group is a 5.8-fold increase in
BGC823 cells (Figure 4B) and a 4.4-fold increase in U87 cells
(Figure 4C) compared to the control groups, respectively. This
result indicates that DOTAP micelle could catalyze Phe-BF3-
mediated desilylation in living cells.
We then examined whether micellar catalysis could trigger
desilylation of TBDPS-caged functional molecule and liberate the
4
This article is protected by copyright. All rights reserved.

10.1002/anie.202106526
Angewandte Chemie International Edition
RESEARCH ARTICLE
respectively. Twenty fields of view were randomly chosen for each experiment
(two-tailed unpaired Student’s t-test, ***P < 0.001).
client fluorophore in tumor-bearing mice. We prepared a TBDPS-
“caged” near-infrared dye (Figure 5A), denoted as TBDPSO-NIR,
to sense the desilylation-to-release with optical imaging in tumor-
bearing mice.[23] TBDPSO-NIR is highly stable and showed no
spontaneous desilylation even after 12-h incubation in PBS
(Figure S9). We performed a head-to-head comparison on the
dual-tumor-bearing mice. For mouse #1 in Figure 5B, as the
negative-control group, its left tumor and right tumor were
intratumorally injected with TBDPSO-NIR and TBDPSO-
NIR+DOTAP, respectively. No fluorescence release was
observed in both tumors. For mouse #2, its left tumor and right
tumor were intratumorally injected with TBDPSO-NIR and
TBDPSO-NIR+DOTAP, respectively, followed tail-vein injection
of Phe-BF3. As expected, intense fluorescence was observed on
the right tumor at 5 hours after the injection of Phe-BF3, while
there is no fluorescence on the left tumor. For mouse #3, as the
positive-control group, its left tumor and right tumor were
intratumorally injected with TBDPSO-NIR and OH-NIR+DOTAP,
respectively. No fluorescence is on the left tumor and the
fluorescence on the right tumor is relatively as strong as the right
tumor of mouse #2. The fluorescence intensities were analyzed
and summarized as in Figure 5C. After subtracting the
autofluorescence background, the fluorescence intensity of the
tumor sequentially treated with TBDPSO-NIR+DOTAP and
followed by Phe-BF3 is nearly as strong as the positive control
tumor, and is significantly higher than the tumors treated without
Phe-BF3 or micelle. This result corroborates with the optical
imaging studies, and suggests that TBDPS-“caged” molecules
are stable in vivo and DOTAP micelle could catalyze Phe-BF3-
mediated desilylation in tumor-bearing mice.
Encouraged by the success of the micelle-catalyzed
desilylation that efficiently releases the fluorophores in vitro and
in vivo, we were confident that a broader substrate scope will work
as well with this micellar catalysis–induced release strategy. Here,
we used 10-hydroxycamptothecin (OH-CPT) as
a model
anticancer molecule to invent prodrug by taking the
a
aforementioned catalysis-to-release strategy. The prodrug was
denoted as TBDPSO-CPT, and its release process is shown in
Figure 5D. The cytotoxicity of TBDPSO-CPT, OH-CPT and
catalysis-activated HO-CPT was assayed in U87 (Figure S10A),
BGC823 (Figure S10B), MC38 (Figure S10C) and CHO (Figure
S10D) cell lines, respectively. In most cases, the cell viabilities of
TBDPSO-CPT are much better than HO-CPT to the four cell lines.
The cell viability curves of cells treated by TBDPSO-CPT+DOTAP
or TBDPSO-CPT+Phe-BF3 are not remarkably different from the
curves of TBDPSO-CPT (Figure S10). The cell viability curves of
catalysis-activated HO-CPT almost coincide with the curve of OH-
CPT, indicating the effectiveness of micelle-catalyzed release.
The IC50 of OH-CPT against BGC823 cells is 1.27 ± 0.17 μM and
the IC50 of TBDPSO-CPT against BGC823 cells is 22.08 ± 6.12
μM, which is 17-fold over its parent drug. The IC50 of activated
OH-CPT against BGC823 cells is 1.66 ± 0.16 μM, which is about
the same as the IC50 of OH-CPT. The IC50s of TBDPSO-CPT,
OH-CPT and catalysis-activated HO-CPT against other cell lines
are listed in Figure S11, which are corroborative with the cell
viability curves.
To accomplish micelle-catalyzed desilylation triggered drug
release in the tumor, a sustainable tumor accumulation of micelle
is essential. The retention and local concentration of DOTAP in
the tumor were investigated by positron emission tomography
(PET)[24]. As shown in Figure S12A most of the intratumorally
injected 89Zr-labelled DOTAP-micelle (200 µM, 20 µL PBS) stayed
in the tumor within 24 hours, while almost no uptake was
observed in other tissues. According to the 3-dimensional
quantitative analysis of the radioactive signals in the region-of-
interest (ROI), over 90% of [89Zr] DOTAP-micelle remained in the
tumor at 24 hours post-injection (Figure S12B), corroborating
with the dominated tumor uptake in PET imaging. As the tumor
volumes are 30-40 mm3 at the moment, the local concentration of
DOTAP is up to 100 µM, which should be high enough to form
stable micelles and catalyze the desilylation (Figure S5B).
Besides, the long tumor retention of DOTAP also indicates that
micelles have been formed in tumor, otherwise DOTAP would
clear from tumor much faster as a small molecule.[25]
A
25 μm
HQB
TBDPSO-HQB TBDPSO-HQB
+Phe-BF3
+DOTAP
+Phe-BF3
B
C
BGC823
U87
***
***
400
300
200
100
0
400
300
200
The treatment efficacy of this prodrug activation system was
evaluated on BGC823 tumor-bearing mice (Figure 5E). DOTAP
(200 µM, 20 µL PBS) was intratumorally injected prior to the
combinational tail-vein injection of TBDPSO-CPT and Phe-BF3.
For the group of mice treated by PBS, the volume of BGC823
tumors increased by > 30-fold in three weeks (Figure 5F). For the
group treated by TBDPSO-CPT only, TBDPSO-CPT+Phe-BF3
and TBDPSO-CPT+DOTAP, the tumor volume increased by
14.9-fold, 19.4-fold and 16.2-fold, respectively (Figure 5F),
showing partial yet insufficient treatment efficacy. Upon the
treatment with TBDPSO-CPT+Phe-BF3+DOTAP, remarkable
suppressions on tumor growth have been observed (Figure 5F,
red curve), indicating that the micellar catalysis–triggered prodrug
activation is effective in the tumor. The tumors were dissected and
then weighed, which further confirmed the antitumor effect of
100
0
HQB a
b
c
d
HQB a b c d
Figure 4. Micelle-catalyzed desilylation can release the “caged”
fluorescent HQB from the robust TBDPSO-linker in living cells. (A)
Representative confocal fluorescence images (40x fold) of BGC823 and U87
cells. The cells were pretreated with HQB or TBDPS-HQB (20 μM in RPMI-
1640) for 2 hours and washed twice with PBS, followed by incubating with
DOTAP (40 μM), Phe-BF3 (200 μM) and DOTAP (40 μM)+Phe-BF3 (200 μM),
respectively. (B, C) Fluorescence intensity analysis in BGC823 (B) or U87 (C)
cells according to the examined field of view by confocal microscopy. The cells
were treated with HQB, TBDPSO-HQB+Phe-BF3+DOTAP (a), TBDPSO-
HQB+DOTAP (b), TBDPSO-HQB+Phe-BF3 (c) and TBDPSO-HQB only (d),
5
This article is protected by copyright. All rights reserved.

10.1002/anie.202106526
Angewandte Chemie International Edition
RESEARCH ARTICLE
micellar catalysis strategy (Figure 5G and Figure S12).
Importantly, mice after the treatment do not show significant
changes in mice body weight (Figure S14). And almost no tissue
damage was observed in major organs according to histology
analysis (Figure S15). However, we found that the uncaged drug
OH-CPT shows severe systemic toxicity (Figure S16), thus the
related tumor growth curve is not included. Collectively, these
data suggested the promising potential of the in vivo micellar
catalysis–induced desilylation with good biological safety.
Negative
Control
Positive
Control
Phe-BF3
A
B
C
High
***
***
***
***
1.0
0.8
0.6
0.4
0.2
0.0
t
t
t
t
t
t
TBDPSO-NIR
With micelle
N.S.
TBDPSO-NIR
TBDPSO-NIR
& DOTAP
NC Phe-BF3 PC
Without micelle (left tumor)
With micelle (right tumor)
Phe-BF3
Mouse 1
Mouse 2
Mouse 3
Low
OH-NIR (neutral form)
OH-NIR (anion form)
D
E
F
1200
1000
800
600
400
200
0
G
PBS
a
a+Phe-BF3
a+DOTAP
a+DOTAP
+Phe-BF3
PBS
Days
-7
0
3
6
a
a+Phe-BF3
Xenograft
Implantation
a+DOTAP
****
DOTAP (i.t.)
TBDPSO-CPT+Phe-BF3 (i.v.)
a+DOTAP
+Phe-BF3
TBDPSO-CPT, a
OH-CPT, b
0
3
6
9
12 15 18
Time (days)
Figure 5. Micelle-catalyzed desilylation can release the “caged” near-infrared fluorescent OH-NIR and hydroxyl camptothecin (OH-CPT) that suppressed
tumor growth in mice. (A) Schematic representation of micelle-catalyzed desilylation induced fluorescence release in the right tumors in mice. (B) Representative
optical imaging of the tumor-bearing mice. For the left mice, its left tumor and right tumor was intratumorally injected with TBDPSO-NIR and TBDPSO-NIR+DOTAP,
respectively. For the middle mice, its left tumor and right tumor was intratumorally injected with TBDPSO-NIR and TBDPSO-NIR+DOTAP, respectively, followed
tail-vein injection of Phe-BF3. For the right mice, its left tumor and right tumor was intratumorally injected with TBDPSO-NIR and OH-NIR+DOTAP, respectively. (C)
Comparison of normalized maximum fluorescence intensity of tumors (n = 3, two-tailed unpaired Student’s t-test, ***P < 0.001). Optical imaging was taken five
hours after treatment. (N.S. no significant difference, NC: negative control, PC: positive control) (D) Scheme of OH-CPT release induced by Phe-BF3 triggered
desilylation. (E) Flow chart of treatment. (F, G) The tumor-bearing mice were intratumorally injected with DOTAP or intravenously injected (i.v.) with PBS, TBDPSO-
CPT or Phe-BF3 alone or in combination. (F) Average tumor volumes of each group of mice. (G) Photograph of the BGC823 tumors on day 18 after treatment.
Since traditional cationic micelles normally clear overly fast
3
A
D
B
C
from blood circulation, they have difficulties reaching tumors
through intravenous injection.[26] To overcome this challenge, we
constructed a pH-sensitive PEO45-b-PC7A5-b-PBA25 (PC7BA,
Figure 6A) micelles as the nano-carriers to further investigate the
in vivo application of cationic micelles–catalyzed desilylation. As
shown in Figure 6B and 6C, PC7BA micelles are neutral to mildly
negative-charged at pH = 7.4, and can readily shift to positively
charged at pH = 6.5 (acidic tumor microenvironment[27]). The CMC
values of PC7BA in PBS buffer are 6.6 μg/mL (pH = 7.4) and 10
μg/mL (pH = 6.5), respectively (Figure S17). The hydrodynamic
diameters are 15 nm and 45 nm at pH = 6.5 and at pH = 7.4
(Figure S18), respectively, which corroborated the observations
under the transmission electron microscope (TEM, Figure S19).
These results suggest that PC7BA micelles can be readily
switched into positively charged while keeping intact, which are
essential for in vivo catalysis.
pH 6.5
pH 7.4
2
pH sensitive
1
0
Acidic
pH = 6.5
Neutral
pH = 7.4
E
-1
15
[
89Zr] PC7BA
tumor
blood
10
5
High
t
l
0
0
10 20
30 40 50
Time / h
60 70
80
***
F
G
100
100
80
Control
PC7BA
60
Control
PC7BA
10
40
20
0
1
Low
We have studied their pharmacokinetics by positron emission
tomography (Figure 6D) and time-activity curve (Figure 6E) of
89Zr-labelled PC7BA micelles. As expected, they gave good blood
circulation and notable tumor uptake through intravenous
injection. Besides, a comparison experiment shows that PC7BA
micelles can effectively catalyze the Phe-BF3-mediated
desilylation at pH = 6.5 (Figure 6F). Its “decaging” efficiency is
42-fold to that of PLGA-PEG micelles in tumors in mice (Figure
6G). The above results suggest that cationic micelles play an
important role in accelerating desilylation both in vitro and in vivo.
0
2
4
6
8
Time / h
Figure 6. A pH-sensitive polymeric micelle used as a catalyst for Phe-BF3-
mediated desilylation. (A) Chemical structure of PEO45-b-PC7A5-b-PBA25
(PC7BA) polymers. (B) Schematic representation of pH-responsive working
principle of PC7BA micelles. (C) Zeta potential of PC7BA micelles in a buffer of
pH = 7.4 (blue bar) and 6.5 (red bar). (D) PET imaging of [89Zr]PC7BA micelles
in BGC823-tumor bearing mouse at 36-hour time point after tail-vain injection.
t, tumor; l, liver. (E) Time-activity curve of [89Zr]PC7BA micelle in blood and
tumor in BGC823 tumor–bearing mice. (F) Time-“decaging” yield curve of
TBDPSO-CPT+Phe-BF3 reaction treated by 1 mg/mL PC7BA micelles or 1
mg/mL PLGA5k-PEG2k (as negative control) micelles at pH
= 6.5. (G)
6
This article is protected by copyright. All rights reserved.

10.1002/anie.202106526
Angewandte Chemie International Edition
RESEARCH ARTICLE
[14] a) Z. Li, Z. Kong, J. Chen, J. Li, N. Li, Z. Yang, Y. Wang, Z. Liu, Eur. J.
Nucl. Med. Mol. Imaging 2021; b) Z. Liu, H. Chen, K. Chen, Y. Shao, D.
O. Kiesewetter, G. Niu, X. Chen, Sci. Adv. 2015, 1, e1500694.
[15] T. W. W. Greene, P. G. M. , John Wiley & Sons: New York 1991.
[16] J. Rautio, N. A. Meanwell, L. Di, M. J. Hageman, Nat. Rev. Drug
Discovery 2018, 17, 559-587.
[17] D. Motoda, H. Kinoshita, H. Shinokubo, K. Oshima, Angew. Chem., Int.
Ed. Engl. 2004, 43, 1860-1862.
[18] R. Hu, J. Feng, D. Hu, S. Wang, S. Li, Y. Li, G. Yang, Angew. Chem.,
Int. Ed. Engl. 2010, 49, 4915-4918.
[19] a) I. M. Umlong, K. Ismail, Colloids Surf., A 2007, 299, 8-14; b) N.
Scholz, T. Behnke, U. Resch-Genger, J. Fluoresc. 2018, 28, 465-476;
c) S. K. Hait, S. P. Moulik, J. Surfactants Deterg. 2001, 4, 303-309.
[20] X. Ji, C. Zhou, K. Ji, R. E. Aghoghovbia, Z. Pan, V. Chittavong, B. Ke,
B. Wang, Angew. Chem., Int. Ed. Engl. 2016, 55, 15846-15851.
[21] W. M. Haynes, D. R. Lide, CRC handbook of chemistry and physics : a
ready-reference book of chemical and physical data, 91st ed. / ed.,
Boca Raton (Fla.) : CRC Press, 2010.
[22] W. Hu, L. Zeng, Y. Wang, Z. Liu, X. Ye, C. Li, Analyst 2016, 141, 5450-
5455.
[23] Y. Qi, Y. Huang, B. Li, F. Zeng, S. Wu, Anal. Chem. 2018, 90, 1014-
1020.
Quantitative analysis of OH-CPT released in tumors in mice treated by PC7BA
micelles–delivered TBDPSO-CPT+Phe-BF3 and PLGA5k-PEG2k micelles–
delivered TBDPSO-CPT+Phe-BF3, respectively.
In this work, we take DOTAP and PC7BA micelle as the
examples to show that the cationic micelles may function as the
catalyst to remarkably accelerate the desilylation reaction both in
vitro and in vivo. According to the mechanistic study, this catalytic
nature should be applicable for all cationic micelles as well as the
pH-sensitive micelles. Therefore, micelle-catalyzed desilylation
offers an exciting opportunity to tissue-selectively activate
prodrugs without intrinsic tissue-targeting. In addition, inspired by
the fact that many reactions are remarkably faster in organic
solvent or micelles than in water, we expect that micelles would
be versatile to catalyze a broad scope of reactions in vivo.
[24] D. Duan, H. Liu, Y. Xu, Y. Han, M. Xu, Z. Zhang, Z. Liu, ACS Appl.
Mater. Interfaces 2018, 10, 5278-5286.
[25] M. Yu, J. Zheng, ACS Nano 2015, 9, 6655-6674.
Acknowledgment
[26] M. J. Mitchell, M. M. Billingsley, R. M. Haley, M. E. Wechsler, N. A.
Peppas, R. Langer, Nat. Rev. Drug Discov. 2021, 20, 101-124.
[27] a) T. Volk, E. Jahde, H. P. Fortmeyer, K. H. Glusenkamp, M. F.
Rajewsky, Br. J. Cancer 1993, 68, 492-500; b) K. Zhou, Y. Wang, X.
Huang, K. Luby-Phelps, B. D. Sumer, J. Gao, Angew. Chem., Int. Ed.
Engl. 2011, 50, 6109-6114; c) J. Z. Du, T. M. Sun, W. J. Song, J. Wu, J.
Wang, Angew. Chem., Int. Ed. Engl. 2010, 49, 3621-3626; d) S. S. Han,
Z. Y. Li, J. Y. Zhu, K. Han, Z. Y. Zeng, W. Hong, W. X. Li, H. Z. Jia, Y.
Liu, R. X. Zhuo, X. Z. Zhang, Small 2015, 11, 2543-2554.
This work was financially supported by National Natural Science
Foundation of China (no. NSFC U1867209 and no. NSFC
21778003) and the Ministry of Science and Technology of the
Peoples Republic of China (2017YFA0506300).
Keywords: Micellar catalysis, desilylation, controlled release, in
vivo chemistry
[1]
a) J. A. Prescher, C. R. Bertozzi, Nat. Chem. Biol. 2005, 1, 13-21; b) V.
Kenward, T. Katsunori, in Handbook of In Vivo Chemistry in Mice, 2020,
pp. 309-353; c) Y. Zheng, X. Ji, B. Yu, K. Ji, D. Gallo, E. Csizmadia, M.
Zhu, M. R. Choudhury, L. K. C. De La Cruz, V. Chittavong, Z. Pan, Z.
Yuan, L. E. Otterbein, B. Wang, Nat. Chem. 2018, 10, 787-794; d) R.
Rossin, R. M. Versteegen, J. Wu, A. Khasanov, H. J. Wessels, E. J.
Steenbergen, W. Ten Hoeve, H. M. Janssen, A. van Onzen, P. J.
Hudson, M. S. Robillard, Nat. Commun. 2018, 9, 1484.
[2]
a) L. P. Hammett, Physical organic chemistry; reaction rates, equilibria,
and mechanisms, 1st ed. ed., McGraw-Hill Book Company, Inc., New
York and London, 1940; b) J. Wang, X. Wang, X. Fan, P. R. Chen, ACS
Cent. Sci. 2021.
[3]
[4]
J. Li, P. R. Chen, Nat. Chem. Biol. 2016, 12, 129-137.
a) J. Tu, M. Xu, S. Parvez, R. T. Peterson, R. M. Franzini, J. Am.
Chem. Soc. 2018, 140, 8410-8414; b) J. Kim, C. R. Bertozzi, Angew.
Chem., Int. Ed. Engl. 2015, 54, 15777-15781; c) R. M. Versteegen, R.
Rossin, W. ten Hoeve, H. M. Janssen, M. S. Robillard, Angew. Chem.,
Int. Ed. Engl. 2013, 52, 14112-14116; d) Q. Fu, H. Li, D. Duan, C.
Wang, S. Shen, H. Ma, Z. Liu, Angew. Chem., Int. Ed. Engl. 2020, 59,
21546-21552; e) M. Jiang, H. Li, H. Yang, H. Fu, Angew. Chem., Int.
Ed. Engl. 2017, 56, 874-879; f) Y. Li, Z. Lou, H. Li, H. Yang, Y. Zhao, H.
Fu, Angew. Chem., Int. Ed. Engl. 2020, 59, 3671-3677.
[5]
[6]
Q. Wang, Y. Wang, J. Ding, C. Wang, X. Zhou, W. Gao, H. Huang, F.
Shao, Z. Liu, Nature 2020, 579, 421-426.
L. Su, R. Li, S. Khan, R. Clanton, F. Zhang, Y. N. Lin, Y. Song, H.
Wang, J. Fan, S. Hernandez, A. S. Butters, G. Akabani, R.
MacLoughlin, J. Smolen, K. L. Wooley, J. Am. Chem. Soc. 2018, 140,
1438-1446.
[7]
[8]
M. A. Mintzer, E. E. Simanek, Chem. Rev. 2009, 109, 259-302.
S. Yang, Q. Tang, L. Chen, J. Chang, T. Jiang, J. Zhao, M. Wang, P. R.
Chen, Angew. Chem., Int. Ed. Engl. 2020, 59, 18087-18094.
a) S. Du, S. S. Liew, L. Li, S. Q. Yao, J. Am. Chem. Soc. 2018, 140,
15986-15996; b) M. Zhang, C. C. Song, F. S. Du, Z. C. Li, ACS Appl.
Mater. Interfaces 2017, 9, 25905-25914; c) S. Su, Y. Y. Wang, F. S. Du,
H. Lu, Z. C. Li, Adv Funct Mater 2018, 28, 1805287.
[9]
[10] Q. Cheng, T. Wei, L. Farbiak, L. T. Johnson, S. A. Dilliard, D. J.
Siegwart, Nat. Nanotechnol. 2020, 15, 313-320.
[11] a) H. Cabral, K. Miyata, K. Osada, K. Kataoka, Chem. Rev. 2018, 118,
6844-6892; b) K. Ulbrich, K. Hola, V. Subr, A. Bakandritsos, J. Tucek,
R. Zboril, Chem. Rev. 2016, 116, 5338-5431.
[12] a) B. H. Lipshutz, S. Ghorai, A. R. Abela, R. Moser, T. Nishikata, C.
Duplais, A. Krasovskiy, R. D. Gaston, R. C. Gadwood, J. Org. Chem.
2011, 76, 4379-4391; b) S. Klingelhofer, W. Heitz, A. Greiner, S.
Oestreich, S. Forster, M. Antonietti, J. Am. Chem. Soc. 1997, 119,
10116-10120; c) S. Wallace, E. P. Balskus, Angew. Chem., Int. Ed.
Engl. 2016, 55, 6023-6027.
[13] T. Dwars, E. Paetzold, G. Oehme, Angew. Chem., Int. Ed. Engl. 2005,
44, 7174-7199.
7
This article is protected by copyright. All rights reserved.

10.1002/anie.202106526
Angewandte Chemie International Edition
RESEARCH ARTICLE
Entry for the Table of Contents
Cationic micelle
–accelerated
desilylation
TBDPS caged Coumarin,
HQB, OH-NIR, OH-CPT etc.
Cationic micelles accelerate desilylation-induced cleavage chemistry over 1400-fold under physiological conditions. This micelle-
catalyzed desilylation strategy unveils the potential that micelle may act beyond a carrier but an in vivo catalyst for local perturbing or
activation.
8
This article is protected by copyright. All rights reserved.