Supramolecular amphiphiles based on cyclodextrin and hydrophobic drugs

Article abstract of DOI:10.1039/c6tb03272a

Herein we report a novel “supra-prodrug-type” superamphiphile design: via a redox-sensitive self-immolative linker, a hydrophobic drug molecule was labeled with an azobenzene moiety, which was designed to be capped by a hydrophilic cyclodextrin (CD) molec

Full text of DOI:10.1039/c6tb03272a

View Article Online
View Journal
Journal of
Materials Chemistry B
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: T. Sun, Q. Wang,
Y. Bi, X. Chen, L. Liu, C. Ruan, Z. Zhao and C. Jiang, J. Mater. Chem. B, 2017, DOI: 10.1039/C6TB03272A.
This is an Accepted Manuscript, which has been through the
Volume
4 Number 1 7 January 2016 Pages 1–178
Royal Society of Chemistry peer review process and has been
accepted for publication.
Journal of
Materials Chemistry B
Accepted Manuscripts are published online shortly after
Materials for biology and medicine
www.rsc.org/MaterialsB
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
author guidelines.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the ethical guidelines, outlined
in our author and reviewer resource centre, still apply. In no
event shall the Royal Society of Chemistry be held responsible
for any errors or omissions in this Accepted Manuscript or any
consequences arising from the use of any information it contains.
ISSN 2050-750X
PAPER
Guoping Chen et al.
Regulating the stemness of mesenchymal stem cells by tuning
micropattern features
rsc.li/materials-b

Page 1 of 29
Journal of Materials Chemistry B
View Article Online
DOI: 10.1039/C6TB03272A
Supramolecular amphiphiles based on cyclodextrin and
hydrophobic drugs
Tao Sun,a Qingbing Wang,b Yunke Bi,c Xinli Chen,a Lisha Liu,a Chunhui Ruan,a Zhifeng Zhao,a
Chen Jiang*a
aKey Laboratory of Smart Drug Delivery (Ministry of Education), State Key Laboratory of
Medical Neurobiology, Department of Pharmaceutics, School of Pharmacy, Fudan University, 826
Zhangheng Road, Shanghai 201203, P. R. China. Tel. & Fax: +8621ꢀ51980079. Eꢀmail:
jiangchen@shmu.edu.cn
bDepartment of Chemistry, Fudan University, 220 Handan Road, Shanghai 200433, P. R. China
cDepartment of Neurosurgery, Shanghai First People’s Hospital, School of Medicine, Shanghai
Jiao Tong University, 100 Haining Road, Shanghai 201620, P. R. China

Journal of Materials Chemistry B
Page 2 of 29
View Article Online
DOI: 10.1039/C6TB03272A
ABSTRACT
Herein we report a novel “supraꢀprodrugꢀtype” superamphiphile design: via a redoxꢀsensitive
selfꢀimmolative linker, a hydrophobic drug molecule was labeled with an azobenzene moiety,
which was designed to be capped by a hydrophilic cyclodextrin (CD) molecule. Four clinical
hydrophobic drugs 7ꢀethylꢀ10ꢀhydroxycamptothecin (SNꢀ38), doxorubicin (DOX), phenytoin
and aliskiren were investigated to directly participate in building the novel class of
superamphiphile, in which the CD moiety plays as the hydrophilic head and the drug as
hydrophobic tail. The novel type of superamphiphile can further selfꢀassemble into vesicular
or tubular structures, characterized by transmission electron microscope (TEM), scanning
electron microscope (SEM) and dynamic light scattering (DLS). The possible selfꢀassembly
mechanism was given based on multiple evidences, including nuclear magnetic resonance
(NMR), Fourier transform infrared spectroscopy (FTꢀIR), Xꢀray diffraction (XRD) and
Ultraviolet–visible spectroscopy (UVꢀvis) results. The reconversion kinetics of the prodrug as
a function of glutathione (GSH) in presence or absence of UV irradiation is presented. The
cell experiments indicate the “supraꢀprodrug” can be facially endowed with a function by a
simple substitution with another functionalized host. We hope this work can provide new
reference in the field of drug screening, formulation and delivery.
KEYWORDS
Superamphiphile, SelfꢀImmolation, Supramolecular Chemistry, Prodrug, StimuliꢀResponsive

Page 3 of 29
Journal of Materials Chemistry B
View Article Online
DOI: 10.1039/C6TB03272A
1. INTRODUCTION
We should always salute to the innovation of amphiphiles, which provides more opportunities
in material sciences,1 including intelligent machines, template synthesis, drug formulation and
targeted delivery. Conventionally, amphiphile refers to a term of chemicals with both
hydrophilic and hydrophobic segments conjugated by covalent bonds.2 During the past decade,
an emerging concept is “supramolecular amphiphile (also named “superamphiphile”)”, where
the hydrophilic and hydrophobic moieties are linked by noncovalent interactions,3 such as
hostꢀguest recognition, hydrogen bonds, metalꢀligand coordination, πꢀion interaction, etc..
The rudiment of superamphiphile can date back to the work by Chen et. al. in 2007,4 where
they reported a selfꢀassembled vesicles made from a CD/1ꢀnaphthylammonium chloride
supramolecular complex. Compared with conventional amphiphiles based on covalent bonds,
superamphiphile is usually synthesized through relatively facile and environmentꢀfriendly
supramolecular strategies. With the aim of satisfying different requirements, various
superamphiphiles in terms of different compositions and corresponding structureꢀfunction
relationship have been well studied, among which superamphiphile based on hostꢀguest
molecular recognition is particularly attractive due to its unique stimuliꢀresponsiveness.3, 5 To
be specific, the formation and disruption of the superamphiphile can be exquisitely tailored
via influencing the hostꢀguest recognition phenomenon.6 Hostꢀguest type superamphiphile
consists of a hydrophilic host and a guested hydrophobic tail. Great efforts have been devoted
to investigating the innovations of the supramolecular hosts, including cyclodextrin,7
cucurbituril,8 pillararene,9 calixarene,10 calixpyrrole,11 crown ether,12 calixresorcin arene,13
biphenarene14 and tetraphenylethylene.15 On the contrary, however, much less attention has
been paid to the elaborate design of guested hydrophobic tails.3 To date, most scientists just
employ simple alkyl chains as the hydrophobic tails with the aim to serve as the hydrophobic
volume.16 It is believed that exploring the topology of the hydrophobic guests is at least
equally important to the decoration of the hydrophilic hosts in expanding the diversity,
versatility and adaptability of the superamphiphile concept.3, 6
Smallꢀmolecular drugs are still playing the most important role in fighting against diseases.
However, about 40% marketing drugs are poorly waterꢀsoluble and the percentage is believed

Journal of Materials Chemistry B
Page 4 of 29
View Article Online
DOI: 10.1039/C6TB03272A
much higher among candidates currently under development.17 The formulation of these
drugable candidates with poor water solubility is a critical challenge in pharmaceutical
sciences. Very recently, directly employing modified drug molecules to build nanomaterials
for “selfꢀdelivery” was found to be more efficient and controllable in drug formulation and
delivery. Cui18 and Shen19 et. al. reported “drug amphiphiles” composed of camptothecin and
hydrophilic chains. In 2015, Cui concluded this class of prodrug as “oneꢀcomponent
nanomedicine” in another review, and named them as “drug amphiphiles”20. EZNꢀ2208 and
NK012 as “drug amphiphiles” are now in clinical II trial. Meanwhile, “drug amphiphiles”
also brought about several problems. Long and tedious chemical linking is necessary to
functionalize the parent drug, which is not suitable for highꢀscale drug screening. The
covalent conjugation might weaken the stimuliꢀresponsiveness.
A supramolecular
functionalization based on molecular recognition could be a promising approach to solve the
problems.
The supramolecular hostꢀguest (HꢀG) pair, βꢀCD/azobenzene is widely used as a
remoteꢀtailored lightꢀdriven switch in the field of controllable release. Wu et. al. used βꢀCD
and azobenzene to construct a supramolecular hydrogel for photoꢀcontrolled protein release in
deep tissue.21 Wu et. al. reported a βꢀCD/azobenzene and aptamerꢀmodified substrate for
lightꢀtriggered specific cancer cell release.22 Yu and Chen employed the βꢀCD/azobenzene
HꢀG couple to prepare a supramolecular platform with switchable multivalent affinity for
photoꢀreversible capture/release of bacteria.23 Inspired by the strategies mentioned above,
through a selfꢀimmolative linker, four hydrophobic parent drugs were conjugated with a
classic and releasable supramolecular guest moleculeꢀazobenzene moiety that exhibits strong
binding affinity to βꢀCD (hydrophilic host) to form a novel superamphiphile. In this design,
the waterꢀinsoluble drug were designed to directly play as the hydrophobic tail of a
superamphiphile, which can further selfꢀassemble into tubular and vesicular structures in
aqueous solution. The hostꢀguest recognition phenomenon and secondary selfꢀassembly
mechanism were investigated in detail based on multiple evidences. The dual
stimuliꢀresponsiveness to photo and reductant was studied. The reconversion kinetics of the
prodrug as a function of GSH in presence or absence of UV irradiation is presented. Not only
can the “supramolecular drug amphiphiles” strategy proposed in this work enrich the concept

Page 5 of 29
Journal of Materials Chemistry B
View Article Online
DOI: 10.1039/C6TB03272A
of superamphiphile, but also provide new reference to highꢀthroughput drug screening and
combinatorial chemistry.
Figure 1 (A) Structures of βꢀCD and azoꢀmodified guested prodrugs; (B) The selfꢀassembly
mechanism and the dualꢀresponsiveness to UV irradiation and reductant.

Journal of Materials Chemistry B
Page 6 of 29
View Article Online
DOI: 10.1039/C6TB03272A
2 EXPERIMENTAL PROCEDURE
2.1 Materials
βꢀCD was purchased from Zhiyuan Biotechnology Co. Ltd. (Binzhou, China). βꢀCD was
recrystallized twice in water, and dried under vacuum at 50 °C for 12 h before using.
pꢀHydroxyazobenzene, 1,4ꢀdibromobutane thiourea, DTT and GSH were purchased from Aladdin
(Shanghai, China). 2, 2'ꢀDithiobispyridine was from TCI Europe (Eschborn, Germany). Reagents
including DMSOꢀδ6, CD3OD and CDCl3 for nuclear magnetic resonance (NMR) spectroscopy
were from J&K Scientific Ltd. (Beijing, China). HCl, KOH and all the solvents mentioned in this
work were from Sinopharm Chemical Reagent (Shanghai, China). Preparative TLC boards and
silica gel for flash column chromatography (SilicaFlash F60, 230~400 mesh) were from Haiyang
Chemical Co., Ltd. (Qingdao, China). N2 and Ar were purchased from Chenggong Gas (Shanghai,
China). SNꢀ38 was from Knowshine Pharmachemicals Inc. (Shanghai, China), DOX, aliskiren and
phenytoin were from Meilun Biotechnology Co. Ltd. (Dalian, China). All chemicals were used as
received unless indicated otherwise.
2.2 Equipment
Protonꢀ1 NMR (1H NMR) spectra were recorded on a Varian Oxford NMR spectrometer (Palo
Alto, California, USA) operating at 400 MHz/25 °C, while carbonꢀ13 NMR (13C NMR) and NMR
ROESY spectra were recorded on a Bruker NMR Ascend 600 spectrometer (Billerica,
Massachusetts, USA) operating at 600 MHz/25 °C. FTꢀIR spectra were obtained using ATR
geometry on a Spectrum 65 infrared spectrophotometer (PerkinꢀElmer, Waltham, Massachusetts,
USA) at 25 °C. Mass spectrometry (MSꢀESI) data were obtained from Agilent LCMSD (Santa
Clara, California, USA). All TEM observations were carried on a JEMꢀ100CX electron
microscope from JEOL Ltd. (Tokyo, Japan). SEM images were obtained with a Hitachi (Tokyo,
Japan) Sꢀ4800 scanning electron microscope by coating the vesicular solution to the base plate and
then dried and sputterꢀcoated with gold. DLS and Z potential measurements were carried out with
a Wyatt QELS Technology DAWN HELEOS instrument (Santa Barbara, USA) poised at constant
room temperature (25 °C) by using a 12ꢀangle replaced detector in a scintillation vial and a 50
mW solidꢀstate laser (λ = 658.0 nm). MM2 energy minimization and molecular size calculation
were undertaken on ChemBio 3D Ultra (14.0 version, Cambridge Software, Massachusetts, USA).

Page 7 of 29
Journal of Materials Chemistry B
View Article Online
DOI: 10.1039/C6TB03272A
Drugꢀrelease was monitored by analytical HPLC (Agilent Technologies Inc., California, USA),
equipped with gradient flow control pump (1260 Quat PumpꢀG1311B), autosampler (1260
ALSꢀG1329B), diode array detector (1260 DADꢀG4212B), fluorescence detector (FLDꢀG1321A)
and column oven (1290 TCCꢀG1316C). A 250ꢀ4.6 column (RPꢀC18 (5 ꢁm), Agilent Technologies
Inc., California, USA) was employed for the HPLC analysis of SNꢀ38 and DOX. Cellular uptake
of 231 cells were visualized and photographed using a fluorescence microscope (Leica, Wetzlar,
Germany).

Journal of Materials Chemistry B
Page 8 of 29
View Article Online
DOI: 10.1039/C6TB03272A
2.3 Synthesis
Figure 2 Synthesis of the azoꢀlabeled guested prodrugs (azoꢀSNꢀ38, azoꢀDOX, azoꢀphenytoin and
azoꢀaliskiren) and DHAꢀβꢀCD.
Four prodrugs were obtained and fully characterized. Key compounds 7 and 10 were prepared
from similar procedures as reported in our previous work.24 The detailed steps and spectra are
shown in S. I. (Fig. S27ꢀ66).
2.4 Supramolecular complexation procedure

Page 9 of 29
Journal of Materials Chemistry B
View Article Online
DOI: 10.1039/C6TB03272A
The host/guest supramolecular inclusion complex was prepared following previous literatures.5
Typically, the complex were prepared by slowly adding drips of guest’s ethanol solution into the
host (herein, the CDs solution) aqueous solution under vigorous stirring over 5 min at 10 °C.
Bluish opalescence (but not precipitation) appeared immediately, indicating the formation of
assembled particles. The mixture was treated by vortex and then sonicated for 30 min, and ethanol
was removed under vacuum or dialysis at room temperature. The vesicular samples for TEM and
DLS tests should be aged at 25 °C for 12 h at dark. The dried supramolecular inclusion complex
was obtained by a 12ꢀh freezeꢀdrying of the aqueous solution.
2.5 Preparation of the βꢀCD/guest inclusion complex and the physical mixture
The sample of βꢀCD/guest inclusion complex for FTꢀIR and XRD characterizations were isolated
by freezeꢀdrying the βꢀCD/guest aqueous suspension. The physical mixture was collected by
simply mixing the guested compound and βꢀCD together with exactly the same molar ratio as the
inclusion. The mixture should be freshly prepared to avoid the possible complexion activity
between the host and guest molecules in solid state. It should be noted that for FTꢀIR spectra, the
physical mixture sample βꢀCD and guested molecule should be grinded with dry KBr separately,
then mixed together rapidly and tableted into a single cake, in order to avoid the inclusion during
the grinding procedure.7 The increased solubility of βꢀCD/azoꢀdrugs was measured employing
HPLC after a 12ꢀh extraction by Soxhlet extractor with acetone and excess DTTꢀtreatment.
2.6 Preparation of the TEM and DLS samples
All TEM samples were freshly prepared via the phosphotungstic acid staining technique: the
copper mesh was dripped with the sample solution and airꢀdried, then a drop of phosphotungstic
acid aqueous solution (0.5% wt) was dripped onto the mesh and baked under the infrared lamp for
30 min. The phase contrast was enhanced by underꢀfocus.
2.7 Preparation of samples for Job’s plots and complex constant.
The complex stoichiometry of βꢀCD with the guest compound in water was determined by Job’s
continuous variation method using a UVꢀvis spectrometer. A set of working solutions were
obtained by mixing Vg mL of the stock guest solution (10ꢀ4 mol/L) with (VtꢀVg) mL of the stock

Journal of Materials Chemistry B
Page 10 of 29
View Article Online
DOI: 10.1039/C6TB03272A
βꢀCD solution (10ꢀ4 mol/L), where Vt is a fixed total volume and Vg is a variable value (from 0 to
10 mL, 0 ≤ Vg ≤ Vt). To obtain the complex constant, the concentration of guest solution (2 × 10ꢀ5
mol/L) was kept constant with the concentration of βꢀCD ranging from 0 eq to 10 eq.
2.8 Photoꢀresponsive property investigation
UV irradiation (254 nm, 60 mW·cmꢀ2) and irradiation by visible light (λ = 434 nm, 20 mW·cmꢀ2)
were carried out with a ZFꢀ20D UV analyzer (Chengxian Instrument Co., Ltd., Shanghai, China).
It should be noted that the samples for DLS, TEM, NMR or UVꢀvis measurement should be well
sealed in order to avoid possible solvent evaporation. The treated samples should be vortexed
again for 3 min before measured.
2.9 Redoxꢀresponsive property investigation
Before DLS and TEM characterization, excess DTT or GSH solid was added to the sample and
vortexed vigorously for 15 min at 300 K. Notably, no filtrate screening procedure by 0.45 ꢁm
filter was carried out before DLS measurement unless indicated.
2.10 HPLC qualification for the drugs24
To determine SNꢀ38, a fluorescence detector with λex/λem=377/550 nm was used. An isocratic
mobile phase of 25ꢀ60% acetonitrile in 10 mM phosphate buffer (pH 3.0) in 18 min at a 1.0
mL/min flow rate was used. The retention time for SNꢀ38 is 8.3 min under these circumstances.
For the determination of DOX, a fluorescence detector with λex/λem=480/560 nm was used. A
mobile phase of KH2PO4 solution (aq. 10 mM): ACN:AcOH (70:30:0.3, v/v/v) in 10 min at a 1.0
mL/min flow rate was used. The retention time for DOX is 2.5 min under these circumstances. For
phenytoin prodrug analysis, a UV detector was used and the absorbance was recorded at 210 nm.
An isocratic mobile phase of 40% acetonitrile in 10 mM phosphate buffer (pH 3.0) in 18 min at a
0.8 mL/min flow rate was used. The retention time for phenytoin is 6.7 min. It should be noted
that the released proꢀphenytoin will go through a rapid reconversion to phenytoin with a halfꢀlife
o
of 1.6 s at 37 C in aqueous solution as shown below. As predicted, no peak for the intermediate
was found from the HPLC experiments. A mobile phase of KH2PO4 solution (aq. 10 mM + 0.1%

Page 11 of 29
Journal of Materials Chemistry B
View Article Online
DOI: 10.1039/C6TB03272A
TFA):ACN:MeOH (50:40:10, v/v/v) in 10 min at a 1.0 mL/min flow rate was used. The retention
time for DOX is 5.2 min under these circumstances.
2.11 Cell experiments
Cellular uptake was tested using MDAꢀMBꢀ231 cells, which were seeded in 24ꢀwell culture plates
(CorningꢀCoaster, Tokyo, Japan) at a density of 3 × 104 cells per well. When achieving a 70%–80%
confluence, cells were incubated with DOX, azoꢀDOX, βꢀCD/azoꢀDOX, DHAꢀβꢀCD (20% in
βꢀCD)/azoꢀDOX, DHAꢀβꢀCD/azoꢀDOX and UVꢀpretreated DHAꢀβꢀCD (20% in
βꢀCD)/azoꢀDOX (254 nm, 60 mW·cmꢀ2, 30 min), at the concentration of 10 ꢁg/mL DOX in 200
ꢁL per well of isotonic PBS 7.4. After a 30ꢀmin incubation, cells were washed with isotonic PBS
7.4 for three times and then visualized and photographed under the fluorescent microscope
(DMI400D, Leica, Germany). For the flow cytometry analysis, cells were digested, centrifuged,
and washed twice with PBS (pH 7.2). The fluorescence intensity of DOX in cells was analyzed
using flow cytometer (BD Biosciences, Bedford, MA, USA) at 488 nm excitation with 1 × 104
events collected for each analysis. Cells without any treatment were used as control.
2.12 In vitro cytotoxicity study
In vitro cytotoxicity was assessed by the classic MTT assay (n = 4). Briefly, MDAꢀMBꢀ231 cells
were seeded in a 96ꢀwell plate at a density of 5 × 103 cells/well and incubated at 37 °C for 24 h.
When 60~70% confluence was achieved, cells were washed and preꢀincubated with
Dꢀglucoseꢀfree Hank’s at 37 °C for a quarter. Then the cells were incubated with free DOX,
azoꢀDOX, βꢀCD/azoꢀDOX and DHAꢀβꢀCD/azoꢀDOX at various DOX concentrations in medium
containing 2 mM Dꢀglucose at 37 °C. After 48 h, the medium was removed and the cells were
washed for three times. Then MTT solution (0.5 mg/mL, 100 ꢁL) was added into every well and
cells were incubated at 37 °C for 4 h. Next, the medium was removed and DMSO (100 ꢁL) was
added into every well, followed by a 10ꢀmin shake. The absorbance of formazan crystals was

Journal of Materials Chemistry B
Page 12 of 29
View Article Online
DOI: 10.1039/C6TB03272A
recorded at 570 nm using microplate spectrophotometer. Cells without treatment were served as
the control.

Page 13 of 29
Journal of Materials Chemistry B
View Article Online
DOI: 10.1039/C6TB03272A
3
RESULTS AND DISCUSSION
3.1 A modular design
To explore the scope, a modular design24 that enables an attachable integration was employed
to construct the superamphiphile. We started by selecting appropriate drug candidates. Four
principles were followed for the drug selection:
(i) the candidate should be among the clinical drugs;
(ii) the drug should be with low aqueous solubility and greatly rely on chemical or physical
means for formulation and delivery. This also permits the key role for them to play as the
hydrophobic ends for the superamphiphile;
(iii) the drug should contain at least one functional group for the modification (for phenytoin,
an aliphatic hydroxyl can be facially preꢀattached);
(iv) in order to adapt the strategy, drugs with high complexing constants with cyclodextrins
(CDs) were not selected. For instance, paclitaxel (PTX) can be well recognized by CDs,25
which will result the competition between CD/azo and CD/PTX. Notably, the selected drug
SNꢀ38 can also be partly included by βꢀCD, but the complex constant is much lower than
CD/azo
.
We selected four drugs with different functional groups as the models: SNꢀ38 (phenolic
hydroxylꢀcontaning), DOX (aminoꢀcontaining), phenytoin (aliphatic hydroxylꢀcontaining)
and aliskiren (aminoꢀcontaining) for azoꢀSNꢀ38, azoꢀDOX, azoꢀphenytoin and azoꢀaliskiren,
respectively. SNꢀ38 and DOX meanwhile represent a class of insoluble drugs with
πꢀconjugated systems.
CDs, as sugarꢀbased biocompatible materials,26 have been extensively explored as a
supramolecular host.27 It is known that βꢀCD/trans azobenzene has a much higher complexing
constant than βꢀCD/cis azobenzene.28 The trans
ꢀcis transformation of azobenzene can be
driven by UV or visible light,29 which enables us to regulate the formation and disintegration
of the superamphiphile in a noninvasive and reversible manner.3 Thus, the supramolecular
complex of βꢀCD/azobenzene was selected as the lightꢀtailored “OnꢀOff” switch for the
superamphiphile.
A disulfideꢀbased selfꢀimmolative linker30 was employed to conjugate the azobenzene and the

Journal of Materials Chemistry B
Page 14 of 29
View Article Online
DOI: 10.1039/C6TB03272A
parent drug together. A selfꢀimmolation procedure is normally driven by cyclisation31 or
elimination reaction.32 Compared with cyclisation, the 1,6ꢀbenzyl elimination is faster, more
independent and less reliable on the drug structure.33 Thus, 1,6ꢀbenzyl elimination24, 29 was
selected as the selfꢀimmolation process that is generally driven by the cleavage of disulfide
bond triggered by reductant including glutathione.34 It was reported that the optimal length of
linker should possess 10~12 carbon atoms,6 we selected a linker with a suggested length.
A retrosynthesis for the guested prodrugs was undertaken (Fig. S1). The key compound
containing the disulfide bond was synthesized by a classic thiol exchange reaction. Then the
drugs were easily conjugated via a carbonate or carbamate bond to obtain the key
intermediates, followed by a substitution of the 2ꢀthiopyridyl group with the thiolꢀmodified
azobenzene. All compounds were fully characterized (Fig. S24 to S60).
3.2 Hostꢀguest molecular recognition of CD to azo segment
CD was designed to play as the hydrophilic head, while the azoꢀdrug as the hydrophilic tail. Thus,
the successful hostꢀguest recognition of CD to azo segment is crucial in this strategy.5 We firstly
studied the supramolecular inclusion phenomenon of the CD to the azo segment on the prodrug.
The supramolecular inclusion was prepared followed a classic procedure. Multiple means were
employed to study the supramolecular hostꢀguest recognition phenomenon. In order to confirm the
inclusion as a new species, we also prepared a host/guest physical mixture (HGP) with exactly the
same molarꢀratio composition. The HGP’s XRD (Fig. S2) and FTꢀIR (Fig. S3) spectra are
obviously a simple overlap of CD and the guested compounds, while in the inclusion’s spectra, the
position and intensity of the key peaks changed obviously, and some peaks that can be easily
found in the HGP’s spectra even disappeared. In the inclusion complex’s FTꢀIR spectrum, the
stretching vibration of the 3350 cmꢀ1 peak suggests the formation of hydrogen bonds.35 The
comparison suggests the formation of new species and new interactions (eg. hydrogen bonds, Fig.
S3) between the components in the supramolecular inclusion.
The synthesized azoꢀdrugs have such a low aqueous solubility that it is thus impossible to be
1
monitored by H NMR in D2O no matter how we amplify the signal. However, clear peaks
1
belonging to azoꢀSNꢀ38 can be found in the βꢀCD/azoꢀSNꢀ38’s H NMR spectrum in D2O

Page 15 of 29
Journal of Materials Chemistry B
View Article Online
DOI: 10.1039/C6TB03272A
(Fig. S4), demonstrating a favorable solubilizing ability of βꢀCD to azoꢀSNꢀ38. As anticipated,
the βꢀCDꢀcapped supraꢀprodrug can greatly increase the aqueous solubility of SNꢀ38 from
7.2ꢀ36 ꢁg/mL36 to 371.2 ꢁg/mL (10.3ꢀ51.5 folds), DOX from 46 ꢁg/mL37 to 1.93 mg/mL (42
folds), phenytoin from 19.9 ꢁg/mL38 to 1.55 mg/mL (78 folds) and aliskiren from 2.1 ꢁg/mL39
to 83 ꢁg/mL (40 folds), all determined by HPLC.
The comparison of 1H NMR spectra of βꢀCD and βꢀCD/azoꢀSNꢀ38 was undertaken to provide
more detailed information in analyzing the interactions.40 It was found that protons at 3 and 6
position of CD molecule show much more significant shifts (Table S1, Fig. S5). It is known
that Hꢀ3 locates inside the CD cavity, while Hꢀ6 locates at the rim side (Fig. 4).41 The
evidence suggests an inclusion of βꢀCD to the guested molecule. Furthermore, in the NMR
ROESY spectrum42 of βꢀCD/azoꢀSNꢀ38, clear crossed correlation between the signals of the
Hꢀ3 and the phenylꢀprotons of azoꢀSNꢀ38 was clearly monitored (Fig. S6), demonstrating the
insertion of phenyl into βꢀCD’s cavity. Meanwhile, nearly no interactions were detected
between azoꢀSNꢀ38 and Hꢀ1 of βꢀCD (locating outside the cone), further evidencing the
supposed inclusion model.
1
Table S1. H NMR: comparison of the chemical shifts of single βꢀCD and βꢀCD/azoꢀSNꢀ38a in
D2O
Entry
δ(βꢀCD)
Hꢀ1
Hꢀ2
Hꢀ3
Hꢀ4
Hꢀ5
Hꢀ6
5.020
5.022
ꢀ0.002
3.599
3.600
ꢀ0.001
3.950
3.937
0.013
3.552
3.556
ꢀ0.004
3.867
3.868
ꢀ0.001
3.821
3.816
0.005
δ(βꢀCD/azoꢀSNꢀ38
)
ꢂδb
aDHO peak (δ = 4.790) is as the interior label.
bꢂδ = δ(βꢀCD) ꢀ δ(βꢀCD/azoꢀSNꢀ38).
We then excluded the possibility of the possible inclusion occurred between CD cavity and
the drug moiety. The complex stoichiometry of βꢀCD/azoꢀdrugs in water was measured by
the continuous variation method (Job’s method) using a UVꢀvis spectrophotometer.43 The
Job’s plots for the binding of βꢀCD with azoꢀdrug in water showed a maximum value at a
molar fraction of 0.5 (Fig. S7), suggesting that both supramolecular inclusion complexes all

Journal of Materials Chemistry B
Page 16 of 29
View Article Online
DOI: 10.1039/C6TB03272A
stay in a 1:1 stoichiometry mode. From the BenesiꢀHildebrand equation (1/ꢂA =
1/α+1/αKap[βꢀCD]n),44 the stoichiometry was further confirmed as 1:1 and the inclusion
constants were calculated by UV doubleꢀreciprocal method as 30253 L/mol for
βꢀCD/azoꢀSNꢀ38, 11759 L/mol for βꢀCD/azoꢀDOX, 17598 L/mol for βꢀCD/azoꢀphenytoin,
22258 L/mol for βꢀCD/azoꢀaliskiren (Fig. S8). The inclusion constant values are also close to
the ones by βꢀCD/azobenzene (trans, 1.24 ± 0.12×105 L/mol).45 The same stoichiometry and
similar inclusion constant reveal that βꢀCD shall recognize the common group of the
azoꢀdrugs in a 1:1 mode.
Admittedly, many drugs (including drugs mentioned here) can be included or partially
included by βꢀCD.46 For instance, it was reported that the inclusion constants for
hydroxypropylꢀβꢀCD and camptothecin (analogue of SNꢀ38) is 260 ± 18 L/mol (camptothecin
in lactoneꢀform) and 113 ± 7 L/mol (camptothecin in carboxylateꢀform).47 However, these
constants are much weaker than those found in the pair of βꢀCD/azoꢀSNꢀ38 (30253 L/mol).
The comparison suggests, in our case, the molecular recognition mainly occurs between
βꢀCD/azoꢀsegment but not βꢀCD/drugꢀsegment. Based on all evidences above, we proposed
an optimized inclusion model shown as Fig. 1.
3.2 Selfꢀassembly of the superamphiphiles in aqueous solution
Figure 3 TEM images of the samples’ micromorphology (scale bars = 200 nm, 10−4 mol/L) of
(A) βꢀCD/azoꢀSNꢀ38; (B) βꢀCD/azoꢀDOX; (C) βꢀCD/azoꢀphenytoin and (D)

Page 17 of 29
Journal of Materials Chemistry B
View Article Online
DOI: 10.1039/C6TB03272A
βꢀCD/azoꢀaliskiren in water at room temperature all with phosphotungstic acid as the
negative staining agent. (E) SEM images of the micromorphology (βꢀCD/azoꢀDOX, scale
bars = 200 nm, 10−4 mol/L). (F) DLS and (G) Zꢀpotential results of the above samples in
aqueous solution (10−4 mol/L) at room temperature.
With the knowledge of the supramolecular inclusion configuration, we started seeking the
advanced selfꢀaggregated microstructures from the superamphiphiles. In water, azoꢀSNꢀ38
exhibits as cloudy precipitation and βꢀCD as a true solution. Typical Tyndall phenomenon of
the βꢀCD/azoꢀSNꢀ38 suspension was observed when a laser pointer was used to light the
sample (Fig. S9), evidencing the formation of microstructures. To reveal the microstructures
of the selfꢀaggregation, transmission electron microscopy (TEM) and dynamic light scattering
(DLS) were employed to study the morphologies and sizeꢀdistribution of βꢀCD/azoꢀdrugs
samples (Fig. 3). In βꢀCD/azoꢀSNꢀ38 sample, regular singleꢀlayered tubular architecture48, 49
(5 ꢁm × 0.3 ꢁm) were observed by TEM. The πꢀπ stacking between SNꢀ38 molecules could
result a more rigid tubular structure with relatively lower axial curvity. Meanwhile, βꢀCD can
include SNꢀ38 molecule somehow (even the complex constant is much weaker than
βꢀCD/azobenzene), which reduces the uniformity. This example suggests that we should
cautiously consider the potential inclusion competition in the initial strategy design. In the
βꢀCD/azoꢀDOX, βꢀCD/azoꢀphenytoin and βꢀCD/azoꢀaliskiren samples, vesicles ranging
from 40 to 200 nm were found. The hydrodynamic diameters given by DLS results are in
accordance with the TEM sizes. Trace large vesicles (>1 ꢁm) were found in freshlyꢀprepared
samples. As aged, the large vesicles would transform into small vesicles (40~200 nm). Thus,
we deduce the ultraꢀlarge vesicles are kinetic stable, while the smaller vesicles are
thermodynamically stable. SEM was also used to further characterize the morphology.
Wellꢀdefined spherical structures with a diameter around 180 nm were found. After harsh
goldꢀsputtering treatment for the SEM samples, the vesicles can still keep rigid, suggesting a
o
fine mechanical stability. The vesicles can stay stable for 7 days at 25 C as monitored by
TEM and DLS. In PBS (10 mM), similar aggregates and sizes were found by TEM and DLS,
suggesting the selfꢀaggregation is not dependent on the salt concentration. There are nearly no
o
morphological differences between 25 and 37 C. The vesicles show a slightly positive Z

Journal of Materials Chemistry B
Page 18 of 29
View Article Online
DOI: 10.1039/C6TB03272A
potential in water, which is beneficial to maintain the stability.
3.3 Selfꢀassembly mechanism
The “knowꢀhow” knowledge of the nanoaggregates’ formation mechanism is important in
tailoring the properties. βꢀCD is macrocyclic cavitand consisting of seven glucopyranose units,
and can trap hydrophobic substrate to form a supramolecular inclusion complex in aqueous
solution.42 In our case, βꢀCD’s cavity can encapsulate the azobenzene moiety of azoꢀdrugs,
with hostꢀguest molecule recognition and hydrogen bonds as the driving force. 3, 5 βꢀCD acts
as the hydrophilic head of the superamphiphile, while the azoꢀcontaining guest bearing with
waterꢀinsoluble drug segment as the hydrophobic tail. The hydrophilic βꢀCD heads interact
with surrounding water, while the hydrophobic tails minimize their exposure to water by
aggregating together possibly via πꢀπ stacking and hydrophobic interactions, in which way the
superamphiphile can further selfꢀassemble into nanotubular or vesicular bilayered structures.
The rough molecular sizes were calculated (azoꢀSNꢀ38 3.4 nm, azoꢀDOX 3.6 nm,
azoꢀphenytoin 3.1 nm, azoꢀaliskiren 3.7 nm and βꢀCD 0.6×0.8×1.5 nm, Fig. S10) after MM2
energy minimization.35 The axial length of the disulfideꢀcontaining linker is within the
recommended length (10~12 cabon atoms) for the superamphiphile design.6 As reported by
Ma50 and Chu et. al.51, the hostꢀguest interaction between the hosted βꢀCD unit and the
hydrophobic part of a guested molecule plays a crucial role in driving the spontaneous
selfꢀassembly. The mechanism is illustrated in Fig. 1 (B).
3.4 Stimuliꢀresponsiveness and drugꢀrelease
Regulating the therapeutic reagents’ activity of a drug exclusively at the target site represents
a very difficult but appealing goal. Light as an activating trigger is more attractive than other
approaches for it is precise and spatioꢀtemporally controllable, leading fewer damages to
normal and fragile tissues and increase the drug concentration at the selected sites.52 The
photo and reductant responsiveness of the superamphiphile system was investigated. Upon
treatment by UV irradiation (10~80 min),
a continuous structural variation of
βꢀCD/azoꢀSNꢀ38 and βꢀCD/azoꢀphenytoin sample from well regular nanotubes or vesicles to

Page 19 of 29
Journal of Materials Chemistry B
View Article Online
DOI: 10.1039/C6TB03272A
irregular particles was recorded (Fig. S11 and S12) by TEM at different time intervals. This
could be explained as: UV triggers a trans to cis transformation of the azobenzene,
consequently cisꢀazoꢀdrug is liberated from the βꢀCD cavity.5 The “building blocks” thus
decompose and escape from the bilayered architecture. The released azoꢀdrug rapidly
precipitates down because of its too low solubility. Both the UV absorbance and fluorescence
intensity of βꢀCD/azoꢀSNꢀ38 sample in water drop sharply upon the UV irradiation (Fig. S13
1
and S14), proving the exclusion of cisꢀazoꢀSNꢀ38 from the suspension. New H NMR peaks
can be found in the UVꢀirradiated sample azoꢀdrug (30 min), which could be ascribed to the
UVꢀtriggered cis form of azoꢀdrug (Fig. S15). It takes ca. 50 min for the transꢀcis
1
transformation (but not a complete one from H NMR) to reach the balance (Fig. S16). Fine
1
reversibility from cis back to trans was found by H NMR after a 2ꢀh treatment by visible
light. Favorable repeatability was also noticed by UVꢀvis spectrometer (Fig. S17).
The disulfide bond of the prodrugs is designed to be cleaved upon reductant.24 We employed
excess DTT or GSH to treat βꢀCD/azoꢀSNꢀ38 and βꢀCD/azoꢀDOX samples, and aggregates
with irregular shape were recorded by TEM (Fig. S18). Meanwhile, in the DTTꢀtreated
βꢀCD/azoꢀSNꢀ38 sample, much weaker fluorescence intensity and clear precipitation were
found (Fig. S19). The appeared solid was collected by filtration, reꢀdissolved in MeOH,
analyzed by HPLC, and confirmed to be the parent drug SNꢀ38 (Rt = 8.32 min).
GSH (0 mM)
GSH (10 mM)
GSH (10 mM with UV@40 min)
GSH (10 mM with UV@0 min)
100
80
UV irradiation@40 min
60
40
20
0
UV irradiation@0 min
0
50
100
150
200
250
Time/min
o
Figure 4 In vitro drug release profiles of βꢀCD/azoꢀSNꢀ38 samples at 37 C. Black: GSH (0
mM); blue: GSH (10 mM) with UV irradiated; red: GSH (10 mM). Values are represented as

Journal of Materials Chemistry B
Page 20 of 29
View Article Online
DOI: 10.1039/C6TB03272A
means ± SD (n = 3).
As designed, azoꢀdrugs tend to precipitate from the solution upon exclusion from βꢀCD’s
cavity. The supramolecular approach thus allows an “OnꢀOff” switch for the drug
reconversion by isolating the guest from the solution containing triggers. In our case, the drug
release kinetics can be obviously slowed by a preꢀUVꢀirradiation treatment. The drug release
kinetics was evaluated by HPLC in GSH45 aqueous solution (10 mM) and PBS buffer. Only
little drug leak was found when no GSH was present (11.5%, 4 h), proving a favorable
stability of the drugꢀcarrying system. Meanwhile, a rapid drug release curve (84.4%, 4 h) was
noticed in a GSH aqueous solution (10 mM). We believed the drug release is due to the
GSHꢀtriggered selfꢀimmolation process. As expected, upon the UV irradiation at 40 min, an
obviously slower drug release was monitored (49.8%/40 min to 67.6%/4 h), compared with
no UV treatment group (45.7%/40 min to 84.4%/4 h) (Fig. 4). If we treated the sample with
UV at the very beginning of the drugꢀrelease test, only 20.7% drug was released after 4 h in
GSH buffer. It is understandable that the exclusion of azoꢀSNꢀ38 from the solution would
greatly slow down its interaction with GSH. Since a fast reconversion from prodrugs will
easily lead to drug precipitation and slow reconversion will result in the easy clearance by the
metabolic organs,54 it should be of advantage for our strategy to logically and precisely
regulate the drug reconversion employing noninvasive photo (to slow down) and endogenous
GSH level (to accelerate) as the coꢀtriggers. However, considering the poor penetrating ability
of UV and large sizes, we should admit that the photoꢀregulating strategy currently is limited
among the local drug administration to surficial tumors, such as melanoma and mammary
cancer. To overcome the problem, we are now investigating the possibility of employing
upconverting nanoparticles to attract NIR light as the trigger for the system.55 One possible
approach could be loading the upconverting nanoparticles into the hollow vesicles’ core to
absorb NIR light and transform them into UV light. Another approach is to modify the
azobenzene to decrease the needed energy for the cisꢀtrans transformation.21
3.5 Versatility

Page 21 of 29
Journal of Materials Chemistry B
View Article Online
DOI: 10.1039/C6TB03272A
Compared with covalently functionalized prodrugs, the advantage for our supramolecular
design is the detachable components. The hosted and guested scaffolds can be conveniently
replaced and combined. The versatility of the strategy was sought in our work.
As to the guest, we directly linked the azobenzene to SNꢀ38 via a ꢀ(CH2)10ꢀ alkyl chain to
obtain azoꢀC10ꢀSNꢀ38 (13) with a similar size as azoꢀSNꢀ38 (Fig. S10). βꢀCD/13 complex can
also form supramolecular inclusion and selfꢀassemble into nanotubes. However, neither
GSHꢀinitiated morphology change nor drug release was found in this group monitored by
DLS/HPLC, which is due to the replacement of GSHꢀsensitive linker by the inert alkyl linker.
Figure 5 Cellular uptake and DOX release of free DOX·HCl (A), azoꢀDOX (B),
βꢀCD/azoꢀDOX (C), DHAꢀβꢀCD (20% in βꢀCD)/azoꢀDOX (D), DHAꢀβꢀCD/azoꢀDOX (E) and
UVꢀpretreated DHAꢀβꢀCD (20% in βꢀCD)/azoꢀDOX (254 nm, 60 mW·cmꢀ2, 30 min) (F), at the
concentration of 10 ꢁg/mL DOX in 200 ꢁL PBS 7.4 per well. Fluorescence of DOX (λex = 488
nm, λem = 590 nm) was observed under the fluorescence inversion microscope.
We also labeled a DHA (tumorꢀtargeting56) moiety to βꢀCD to obtain DHAꢀβꢀCD as the
functionalized host. DHAꢀβꢀCD/azoꢀDOX can form similar stimuliꢀresponsive aggregates as
βꢀCD/azoꢀDOX (Fig. S24). MDAꢀMBꢀ231 cells were used as the model for breast cancer.
DHAꢀβꢀCD (20%)/azoꢀDOX sample shows more cellular uptake and DOX release compared
with the βꢀCD/azoꢀDOX group, indicating the endowed tumorꢀtargeting ability from

Journal of Materials Chemistry B
Page 22 of 29
View Article Online
DOI: 10.1039/C6TB03272A
DHAꢀβꢀCD. The results were further confirmed by the flow cytometry quantitative analysis
(Fig. S25). As expected, if the sample was irradiated by UV for 30 min before the cellular
uptake experiment, only particles with strong fluorescence were found extracellularly. Nearly
no cellular uptake was monitored, suggesting we can apply UV as a nonꢀinvasive means to
control the drug distribution. In vitro cytotoxicity of the above DOX formulation against
MDAꢀMBꢀ231 cells was estimated by MTT assay to determine IC50 values (Fig. S26). IC50
values of free DOX, azoꢀDOX, βꢀCD/azoꢀDOX and DHAꢀβꢀCD/azoꢀDOX are 1.72, 1.906,
1.321 and 0.936 ꢁM. Apparently, the endowed tumorꢀtargeting function from DHAꢀβꢀCD
greatly enhanced the antiꢀtumor efficiency. The cell experiments indicate the superamphiphile
can be facially endowed with a function by a simple substitution with another functionalized
host.

Page 23 of 29
Journal of Materials Chemistry B
View Article Online
DOI: 10.1039/C6TB03272A
4
CONCLUSION
The dynamic immobilization and nonꢀcovalent encapsulation of parent drugs into the micro
architecture enables pharmacists to formulate drugs in a convenient and controllable way. On
the other hand, to explore more topology structures of the superamphiphile could contribute a
lot to the development of supramolecular materials. In this study, we proposed a new idea in
functionalizing superamphiphile from the sophisticated fabrication of the guest molecule. The
proofꢀofꢀconcept of the strategy was accomplished by selecting four waterꢀinsoluble drugs,
which were designed to directly act as the hydrophobic tails of the superamphiphiles.
Azobenzene was linked to the parent drug through a selfꢀimmolative linker, then recognized
by βꢀCD to form a superamphiphile. The hostꢀguest recognition phenomenon was
demonstrated, the advanced aggregates including tubularꢀlike and vesicularꢀlike
microstructures in aqueous solution were found, and the corresponding mechanism was given
based on multiple evidences. The reconversion of the prodrugs can be coꢀregulated by photo
and reductant. The supramolecular aggregates can be conveniently functionalized by a simple
substitution by another host. We hope this study will cast a new light in the design of
superamphiphile from a pharmaceutical aspect and establish a proofꢀofꢀprinciple for the
“supraꢀprodrug” strategy.

Journal of Materials Chemistry B
Page 24 of 29
View Article Online
DOI: 10.1039/C6TB03272A
CONFLICT OF INTEREST
The authors declare no conflict of interest.
ACKNOWLEDGEMENTS
This work was supported by the National Natural Science Funds of China (21602030,
81172993), National Basic Research Program of China (973 Program, 2013CB932500),
Shanghai Sailing Program (16YF1400900), Scientific Research Foundation of Fudan
University for Talent Introduction (JJF301103), National Science Fund for Distinguished
Young Scholars (81425023) and the Important Innovative Program of Shanghai Municipal
Commission of Health and Family Planning (201410017). Supporting information is available
online or from the authors.
Author contributions: TS conceived the project, carried on the research and composed the
paper. QW and YB assisted in the characterization for the compounds. XC and LL carried on
the synthesis of the compounds. CR and ZZ carried out the TEM studies. CJ supervised the
research work.

Page 25 of 29
Journal of Materials Chemistry B
View Article Online
DOI: 10.1039/C6TB03272A
REFERENCES
1. Y. Liu, P. G. Jessop, M. Cunningham, C. A. Eckert and C. L. Liotta, Science, 2006, 313,
958–960.
2. A. Sorrenti, O. Illa and R. M. Ortuño, Chem. Soc. Rev., 2013, 42, 8200–8219.
3. X. Zhang, and C. Wang, Chem. Soc. Rev., 2011, 40, 94–101.
4. B. Jing, X. Chen, X. Wang, C. Yang, Y. Xie, and H. Qiu, Chem. Eur. J., 2007, 13,
9137−9142.
5. G. Yu, K. Jie and F. Huang, Chem. Rev., 2015, 115, 7240–7303.
6. C. Wang, Z. Wang and X. Zhang, Acc. Chem. Res., 2012, 45, 608–618.
7. Y. Liu, C. Y. Yu, H. B. Jin, B. B. Jiang, X. Y. Zhu, Y. F. Zhou, Z. Y. Lu and D. Y. Yan, J.
Am. Chem. Soc., 2013, 135, 4765–4770.
8. Y. J. Jeon, P. K. Bharadwaj, S. Choi, J. W. Lee and K. Kim, Angew. Chem. Int. Ed. Engl.,
2002, 41, 4474–4776.
9. X. Chi, G. Yu, L. Shao, J. Chen and F. Huang, J. Am. Chem. Soc., 2016, 138, 3168–3174.
10. Y.ꢀX. Wang, Y.ꢀM. Zhang and Y. Liu, J. Am. Chem. Soc., 2015, 137, 4543–4549.
11. X. Chi, H. Zhang, G. I. VargasꢀZúñiga, G. M. Peters and J. L. Sessler, J. Am. Chem. Soc.,
2016, 138, 5829–5832.
12. X. Ji, S. Dong, P. Wei, D. Xia and F. Huang, Adv. Mater., 2013, 25, 5725–5729.
13. Ju. E. Morozova, V. V. Syakaev, E. Kh. Kazakova, Ya. V. Shalaeva, I. R. Nizameev, M.
K. Kadirov, A. D. Voloshina, V. V. Zobov and A. I. Konovalov, Soft Matter, 2016, 12,
5590–5599.
14. J. Zhou, G. Yu, L. Shao, B. Hua and F. Huang, Chem. Commun., 2015, 51, 4188–4191.
15. M. Zhang, X.ꢀP. Yin, T. Tian, Y. Liang, W.ꢀN. Li, Y. Lan, J. Li, M.ꢀM. Ju, Y. Zhou and
G.ꢀT. Li, Chem. Commun., 2015, 51, 10210–10213.
16. Y. Tang, L. Zhou, J. Li, Q. Luo, X. Huang, P. Wu, Y. Wang, J. Xu, J. Shen and J. Liu,
Angew. Chem., Int. Ed., 2010, 49, 3920–3924.
17. H. D. Williams, N. L. Trevaskis, S. A. Charman, R. M. Shanker, W. N. Charman, C. W.
Pouton and C. J. Porter, Pharmacol. Rev., 2013, 65, 315–499.
18. A. G. Cheetham, P. Zhang, Y. Lin, Lye L. Lock, and H. Cui, J. Am. Chem. Soc., 2013,
135, 2907–2910.

Journal of Materials Chemistry B
Page 26 of 29
View Article Online
DOI: 10.1039/C6TB03272A
19. S. Hu, E. Lee, C. Wang, J. Wang, Z. Zhou, Y. Li, X. Li, J. Tang, D. H. Lee, X. Liu and Y.
Shen, J. Control. Release, 2015, 220, 175–179.
20. H. Su, J. M. Koo and H. Cui, J. Control. Release, 2015, 219, 383–395.
21. D. Wang, Manfred Wagner, H.ꢀJ. Butt and S. Wu, Soft Matter, 2015, 11, 7656–7662.
22. Q. Bian, W. Wang, S. Wang, and G. Wang, ACS Appl. Mater. Interfaces, 2016, 8, 27360–
27367.
23. W. Zhan, T. Wei, L. Cao, C. Hu, Y. Qu, Q. Yu and H. Chen, ACS Appl. Mater. Interfaces,
2017, 9, 3505–3513.
24. T. Sun, A. Morger, B. Castagner and J.ꢀC. Leroux, Chem. Commun., 2015, 51, 5721–
5724.
25. X. Song, Y. Wen, J.ꢀL. Zhu, F. Zhao, Z.ꢀX. Zhang and J. Li, Biomacromolecules, 2016,
17, 3957–3963.
26. G. Crini, Chem. Rev., 2014, 114, 10940–10975.
27. S. M. N. Simões, A. ReyꢀRico, A. Concheiro and C. AlvarezꢀLorenzo, Chem. Commun.,
2015, 51, 6275–6289.
28. L. He, X. Yang, K. Wang, Q. Wang, F. Zhao, J. Huang and J. Liu, Chem. Commun., 2014,
50, 7803–7805.
29. N. Kameta, H. Akiyama, M. Masuda and T. Shimizu, Chem. Eur. J., 2016, 22, 7198–
7205.
30. K. Haba, M. Popkov, M. Shamis, R. A. Lerner, C. F. Barbas III and D. Shabat, Angew.
Chem. Int. Ed., 2005, 44, 716–720.
31. S. Bhuniya, S. Maiti, E.ꢀJ. Kim, H. Lee, J. L. Sessler, K. S. Hong and J. S. Kim, Angew.
Chem., Int. Ed., 2014, 53, 4469–4474.
32. S. H. Lee, E. Moroz, B. Castagner and J.ꢀC. Leroux, J. Am. Chem. Soc., 2014, 136,
12868–12871.
33. R. Labruère, A. Alouane, T. L. Saux, I. Aujard, P. Pelupessy, A. Gautier, S. Dubruille, F.
Schmidt and L. Jullien, Angew. Chem. Int. Ed., 2012, 51, 9344–9347.
34. K. Miller, R. Erez, E. Segal, D. Shabat and R. SatchiꢀFainaro, Angew. Chem. Int. Ed.,
2009, 48, 2949–2954.
35. J. H. Jung, G. John, M. Masuda, K. Yoshida, S. Shinkai and T. Shimizu, Langmuir, 2001,

Page 27 of 29
Journal of Materials Chemistry B
View Article Online
DOI: 10.1039/C6TB03272A
17, 7229–7232.
36. J. A. Zhang, T. Xuan, M. Parmar, L. Ma, S. Ugwu, S. Ali and I. Ahmad, Int. J. Pharm.,
2004, 270, 93–107.
37. G.ꢀD. Kang, S.ꢀH. Cheon and S.ꢀC. Song, Int. J. Pharm., 2006, 319, 29–36.
38. S. A. Varia, S. Schuller, K. B. Sloan and V. J. Stella, J. Pharm. Sci., 1984, 73, 1068–
1073.
39. Drug Bank Aliskiren, http://www.drugbank.ca/drugs/DB09026 (accessed 5 October
2016).
40. H. J. Schneider, F. Hacket, V. Rüdiger and H. Ikeda, Chem. Rev., 1998, 98, 1755–1786.
41. T. Sun, Q. Guo, C. Zhang, J. Hao, P. Xing, J. Su, S. Li, A. Hao and G. Liu, Langmuir,
2012, 28, 8625–8636.
42. J. Lu, N. Brummelhuis and M. Weck, Chem. Commun., 2014, 50, 6225–6227.
43. D. Bardelang, A. Rockenbauer, H. Karoui, J. P. Finet, and P. Tordo, J. Phys. Chem. B,
2005, 109, 10521ꢀ10530.
44. G. M. Escandar, Analyst, 1999, 124, 587–591.
45. B. L. May, J. Gerber, P. Clements, M. A. Buntine, D. R. B. Brittain, S. F. Lincoln, C. J.
Easton, Org. Biomol. Chem., 2005, 3, 1481–1488.
46. J. Cheng, K. T. Khin and M. E. Davis, Mol. Pharm., 2004, 1, 183–193.
47. A. M. Sætern, N. B. Nguyen, A. BauerꢀBrandl and M. Brandl, Int. J. of Pharm., 2004,
284, 61–68.
48. X. Fan, L. Wang, Q. Luo, L. Zhao, J. Xu, J. Liu and Q. Zheng, Chem. Commun., 2015, 51,
6512–6514.
49. H. Unsal, J. Schmidt, Y. Talmon, L. T. Yildirim and N. Aydogan, Langmuir, 2016, 32,
5324–5332.
50. J. Zhang, K. Ellsworth and P. X. Ma, J. Control. Release, 2010, 145, 116–123.
51. Q.ꢀD. Hu, G.ꢀP. Tang, and P. K. Chu, Acc. Chem. Res., 2014, 15, 2017–2025.
52. D. Dolmans, D. Fukumura and R. K. Jain, Nat. Rev. Cancer, 2003, 3, 380–387.
53. A. Alsbaiee, B. J. Smith, L. Xiao, Y. Ling, D. E. Helbling and W. R. Dichte, Nature,
2016, 529, 190–194.
54. S. Mura, J. Nicolas and P. Couvreur, Nat. Mater., 2013, 12, 991–1003.

Journal of Materials Chemistry B
Page 28 of 29
View Article Online
DOI: 10.1039/C6TB03272A
55. J. Liu, W. Bu, L. Pan and J. Shi, Angew. Chem. Int. Ed., 2013, 52, 4375–4379.
56. J. Yun, E. Mullarky, C. Lu, K. N. Bosch, A. Kavalier, K. Rivera, J. Roper, I. I. Chio, E. G.
Giannopoulou, C. Rago, A. Muley, J. M. Asara, J. Paik, O. Elemento, Z. Chen, D. J.
Pappin, L. E. Dow, N. Papadopoulos, S. S. Gross and L. C. Cantley, Science, 2015, 350,
1391–1396.

Page 29 of 29
Journal of Materials Chemistry B
View Article Online
DOI: 10.1039/C6TB03272A
Supramolecular amphiphiles based on cyclodextrin and
hydrophobic drugs
Tao Sun,a Qingbing Wang,b Yunke Bi,c Xinli Chen,a Lisha Liu,a Chunhui Ruan,a Zhifeng Zhao,a
Chen Jiang*a
Table of Contents
1, Cover Letter;
2, Main Text;
3, Supporting Information;
4, List-by-List Response to the Reviewers