A self-assembled delivery platform with post-production tunable release rate

Article abstract of DOI:10.1021/ja3051876

Self-assembly of three molecular components results in a delivery platform, the release rate of which can be tuned after its production. A fluorophore-conjugated gelator can be hydrolyzed by an enzyme, resulting in the release of a fluorescent small molecule. To allow the release to be tunable, the enzyme is entrapped in liposomes and can be liberated by heating the system for a short period. Crucially, the heating time determines the amount of enzyme liberated; with that, the release rate can be tuned by the time of heating.

Full text of DOI:10.1021/ja3051876

Communication
pubs.acs.org/JACS
A Self-Assembled Delivery Platform with Post-production Tunable
Release Rate
Job Boekhoven,† Mathijs Koot, Tim A. Wezendonk, Rienk Eelkema, and Jan H. van Esch*
Department of Chemical Engineering, Delft University of Technology, Julianalaan 136, 2628 BL Delft, The Netherlands
S
* Supporting Information
To develop a model delivery platform with a tunable release
ABSTRACT: Self-assembly of three molecular compo-
nents results in a delivery platform, the release rate of
which can be tuned after its production. A fluorophore-
conjugated gelator can be hydrolyzed by an enzyme,
resulting in the release of a fluorescent small molecule. To
allow the release to be tunable, the enzyme is entrapped in
liposomes and can be liberated by heating the system for a
short period. Crucially, the heating time determines the
amount of enzyme liberated; with that, the release rate can
be tuned by the time of heating.
rate, we designed a system comprised of three major
components: a hydrogelator, a proteolytic enzyme, and a
phospholipid. The low-molecular-weight hydrogelator 1,11
based on a cyclohexane-tris-amide core,12 is covalently
connected to the fluorogenic molecule 6-aminoquinoline (6-
AQ) via an enzymatically hydrolyzable amide linker (Figure
1a). The excitation wavelength of 6-AQ shifts drastically when
the amide is hydrolyzed to an amine, which makes it highly
suitable for spectroscopic release studies.13 Above its critical
gelation concentration (cgc) of 1.9 mM, gelator 1 assembles
into fibrous aggregates, forming a hydrogel and thereby
immobilizing the gelator-conjugated 6-AQ in the material.
Gelator 1 can be hydrolyzed by the protease α-chymotrypsin
(α-chy), resulting in the release of the small molecule 6-AQ.14
In order to tune the release rate after production of the gel,
the α-chy should initially be isolated from 1 in a separate
compartment. Gelator 1 and liposome-forming lipids can self-
assemble orthogonally within a single system, with each
preserving its own characteristics.10a Thus, loading a gel of 1
with liposomes that contain α-chy in their inner aqueous
compartment physically isolates the enzyme and its substrate,
gelator 1 (Figure 1b). In such a system the triggered liberation
of α-chy from its compartment will initiate the hydrolysis of 1
and thus the release of the fluorescent 6-AQ. The liberation of
liposomal content can be brought about by a series of triggers,
such as heating the liposomes to their phase transition
temperature (Tm),15 changing the pH,16 using an external
magnetic field,17 adding co-surfactants such as Triton-X,18 or
stimulating with ultrasound.19 In the latter examples, controlled
release comes from the intrinsic properties of the material or
the method of preparation. As a result, the release rate cannot
be tuned after the preparation of the delivery platform.
Examples of platforms, the drug release rate of which can be
tuned post-productionally do exist, but they often require
permanent changes in the environment of the platform, such as
the concentration of plasticizer20 or the pH.21 Here we used
thermosensitive liposomes to encapsulate α-chy (Figure 1b).
We postulated that controlled heating of the samples would
liberate α-chy into the gel matrix, after which hydrolysis of 1 by
the liberated enzyme can take place, also at lower temperatures
(T < Tm). The amount of liberated enzyme, and thus the small-
molecule release rate, can be controlled by the heating time.
Interestingly, the presence of the fibrillar network of 1 may
xternal control over catalyst activity could play a major role
Ein the development of responsive materials and controlled
release applications. Recent developments in soft matter self-
assembly show several opportunities for application in
medicine.1 For instance, self-assembling peptide amphiphiles
have been successfully applied in regenerative medicine,2 and
thermosensitive liposomes can be used for localized3 or
targeted4 drug delivery as well as other applications.5
Furthermore, nanotechnology has been exploited to tune
drug release rates from polyelectrolyte films,6 gels,7 block-
copolymer micelles,8 and vesicles.9 In the latter examples,
controlled release comes from the intrinsic properties of the
material or the method of preparation. As a result, the release
rate cannot be tuned after the preparation of the delivery
platform. It remains a challenge to develop a delivery system,
the release rate of which can be tuned after its production. Such
a material would be useful for applications in personalized
medicine and could potentially lead to the development of
customizable medications where the end user, either a health-
care professional or the patient, can tune the drug release rate
as needed. Likewise, this concept has potential for the
development of implants where the drug release rate can be
set externally, or in materials science for controlling the
reactivity of multicomponent adhesives.
In this work we show that multicomponent self-assembly10
can be exploited to develop a delivery platform where the
release rate can be tuned, after production and shortly before
use, to the user’s needs. In the platform described in this paper,
controlled heating of the platform liberates an enzyme into the
system, which hydrolyzes a gelator and thereby releases
detectable small molecules from the scaffold. The amount of
liberated enzyme determines the small-molecule release rate
and, crucially, can be controlled by the length of the heating
period.
Received: May 29, 2012
Published: July 23, 2012
© 2012 American Chemical Society
12908
dx.doi.org/10.1021/ja3051876 | J. Am. Chem. Soc. 2012, 134, 12908−12911

Journal of the American Chemical Society
Communication
Figure 2. (a) Rate of hydrolysis of 1 as a function of substrate
concentration. The rate of hydrolysis remains constant above 1.5 mM
1 at a rate of 50 μM h−1. Black line, added to guide the eye; red line:
theoretical Michaelis−Menten curve calculated from Km and Vmax. (b)
Rate of hydrolysis of 1 as a function of enzyme concentration. The rate
of hydrolysis is first-order in α-chy, with rate constant k = 1.0 h−1. (c)
Concentration of 6-AQ in gels against time for different heating times
(indicated in minutes at each data set) (black markers) and their linear
fit (red lines). (d) 6-AQ release rate, derived from (c) and its duplicate
experiment, against heating time (black markers). Red line added to
guide the eye. Error bars: standard deviation over two individual
experiments. All amide hydrolysis experiments were performed at 25
°C.
Figure 1. (a) Structure of gelator 1 bearing fluorogenic 6-AQ. α-
Chymotrypsin can selectively hydrolyze the amide bond between L-
phenylalanine and 6-AQ, resulting in release of the fluorescent
molecule. (b) Self-assembled multicomponent controlled-release
system: a solution of 1 (green) in DMSO is added to an aqueous
solution of the enzyme α-chy (blue) loaded in liposomes (red),
resulting in the formation of a gel network with embedded liposomes.
Briefly heating the gel at the Tm of the liposomes results in partial
liberation of the enzyme into the gel matrix, allowing for release of 6-
AQ through enzymatic hydrolysis of the amide bond. The heating time
dictates the amount of enzyme liberated and thus the small-molecule
release rate.
Supporting Information for details). Addition of these enzyme-
loaded liposomes to a concentrated solution of gelator 1 in
DMSO resulted in the rapid formation of turbid gels (Figure
3a), typically within seconds.
To liberate the enzyme from the liposomes, and thus initiate
the release of 6-AQ, the gels containing the enzyme-loaded
liposomes were heated for 5 min at 42 °C,23 which matches the
phase transition temperature of the liposomes (vide infra).3a
After heating, the gels were cooled to room temperature,
stabilized for 30 min, and studied by fluorescence spectroscopy.
The release rate of 6-AQ was found to be 8.6 μM h−1 (Figure
2c,d). From the previously found k value (1.0 h−1), it can be
calculated that the concentration of released α-chy is 8.6 μM.
Heating the gels for longer times did not result in higher release
rates, but instead decreased the release rate. From this
observation we conclude that, after 5 min, all the available
enzyme has been released, and heating for longer than 10 min
most likely results in autodigestion of the enzyme.23 Crucially,
if the gels were heated for less than 5 min, the observed 6-AQ
release rates were lower (Figure 2c). For instance, heating a gel
for 1 min resulted in a release rate of 1.6 μM h−1. Moreover,
between 0 and 5 min of heating, the release rate increased
linearly with heating time (Figure 2d). These results clearly
show that it is possible to set the small-molecule release rate to
any required value between 0 and 8.6 μM h−1 by heating the
system for a predetermined amount of time. Thus, the
orthogonal self-assembly of a molecular hydrogelator with
liposomes containing enzymes has been successfully exploited
stabilize the enzyme-loaded liposomes, thereby increasing the
stability (and thus the shelf life) of the platform.
We first investigated the kinetics of the enzymatic cleavage of
1 by α-chy using fluorescence spectroscopy. In line with
previous results,14 addition of 50 μM α-chy to a gel of 1
resulted in the release of 6-AQ. This reaction follows
Michaelis−Menten kinetics as long as the substrate concen-
tration ([1]) remains below 1.5 mM (Figure 2a). Above 1.5
mM, a value that nicely agrees with the cgc of 1, the hydrolysis
remains zeroth-order in substrate 1, and the rate (Vh) remains
constant at 50 5 μM h−1 with increasing concentration of 1.
All further experiments were done in the concentration regime
[1] = 7.5 mM. We also confirmed that, at substrate
concentrations above the cgc, the rate of hydrolysis is linearly
proportional to the concentration of α-chy, with a first-order
rate constant of 1.0 h−1 (Figure 2b). This linear relationship
between the hydrolysis of 1 and the concentration of α-chy
allows us to tune the small-molecule release rate by controlling
the concentration of substrate-accessible enzyme.
Next, we synthesized the controlled-release platform as
depicted in Figure 1. Thermosensitive liposomes, designed to
have a phase transition temperature of 42 °C,3a,22 were
prepared and loaded with 1.6 mM α-chy in their interior (see
12909
dx.doi.org/10.1021/ja3051876 | J. Am. Chem. Soc. 2012, 134, 12908−12911

Journal of the American Chemical Society
Communication
Confocal laser scanning microscopy (CLSM) was used to
further investigate the structure of the self-assembled multi-
component platform and to corroborate the release mechanism
proposed above. For this study, fluorescent probe-labeled
versions of the three components (gelator, fluorescein; lipid,
nitrobenzoxadiazole (NBD); enzyme, rhodamine) were used
(see Supporting Information for details). To ensure the
visibility of the liposomes, giant multilamellar liposomes were
prepared, rather than the 100 nm liposomes used in all other
studies. Excitation of the lipids at 458 nm clearly shows the
presence of giant multilamellar liposomes (Figure 3b).
Excitation of the probe for 1 reveals the presence of fibers,
and excitation of the probe for α-chy at 558 nm shows localized
domains of higher enzyme concentration (Figure 3c,d).
Overlaying these images clearly shows the successful
encapsulation of the α-chy by the liposomes (Figure 3e).
Also, CLSM confirms that the presence of neither 1 nor the
enzyme visually interferes with the liposomes. Vice versa, the
presence of the enzyme-loaded liposomes does not prevent
gelator 1 from forming fibers, confirming that self-assembly of 1
and the liposomes takes place orthogonally. Finally, CLSM
shows that all observed liposomes are filled with α-chy, which
confirms that spontaneous leaking of liposomal content does
not take place to any significant extent.
Differential scanning calorimetry (DSC) experiments were
carried out to study the extent of orthogonal self-assembly of
the different components. DSC on thermosensitive liposomes
without enzyme showed a sharp phase transition temperature
of 41.9 °C, in agreement with literature.3a In contrast, the
enzyme-loaded liposomes displayed a broadened phase
transition temperature with a maximum at 43.4 °C, indicating
that the enzyme somehow interacts with the liposome bilayer
(Figure 3f). Because the Tm shifted only slightly and became
broader in the presence of α-chy, heating at 42 °C, as used for
the controlled enzyme release experiments, is suitable to
partially melt the bilayer and induce leakage of the liposomes as
described above. Comparing DSC traces of enzyme-loaded
liposomes embedded in a gel network to DSC traces of the
same liposomes in solution showed no significant differences,
from which we can conclude that the self-assembly of the
enzyme-loaded liposomes and gelator 1 is fully orthogonal, in
agreement with the CLSM results. It should be noted that the
increased stability of the enzyme-loaded liposomes in a gel
network is most likely the result of mechanical interactions
between the liposomes and the fibrous network at the sub-
micrometer length scale, rather than interactions at the
molecular level.10a
Figure 3. (a) Photograph of a gel of 1 embedded with enzyme-loaded
liposomes. (b) Confocal laser scanning micrograph of α-chy entrapped
in giant multilamellar liposomes embedded in a gel matrix.
Micrographs are taken with an excitation wavelength (λexc) of (b)
458, (c) 493, and (d) 558 nm for the lipid probe, probe for 1, and
probe for α-chy, respectively. (e) An overlay of all three micrographs.
(f) DSC traces of liposomes (black), enzyme-loaded liposomes (red),
and enzyme-loaded liposomes in a gel of 1 (blue).
to control catalyst activity through addressable compartmen-
talization. Coupling the catalytic activity to the release of a
detectable small molecule shows that such a platform could be
applied to tunable drug delivery systems in the future.
In conclusion, we have shown that orthogonal multi-
component self-assembly can be used to create a functional
system in which the catalytic activity can be tuned after its
production. With the catalytic activity coupled to the release of
a small molecule, this prototype system can be envisioned as a
future delivery platform with post-production tunable release
rate. The small-molecule release rate can be tuned using a
simple trigger: the heating time at a constant temperature. This
system is created by compartmentalizing an enzyme in
liposomes which in turn are entrapped in a fibrillar network
of a self-assembled gelator. A fluorescent small molecule is
connected to the hydrogelator by an amide bond, which can be
hydrolyzed selectively by the enzyme. Controlled heating of the
sample around the Tm of the liposomes results in liberation of
the enzyme into the gel, leading to release of the fluorophore.
Crucially, the amount of liberated enzyme, and thus the
The previous results showed that no significant amount of 6-
AQ was released after preparation of the gels, which is a good
indication that the enzyme remains entrapped in the liposomes
during preparation of the gel. We observed, however, that
freshly prepared gels made using a solution of 4-day-old
liposomes showed a small but significant release of 6-AQ (see
Supporting Information). Further experiments showed that the
6-AQ release rate of the gels increased with the age of the
liposome solution, suggesting that α-chy-loaded liposomes in
solution tend to be unstable and will partially leak their
contents over time. In contrast, freshly prepared liposomes
entrapped in a gel network of 1 did not show any significant
release of 6-AQ, even after standing for 2 weeks, indicating that
the fibrillar network of 1 serves as a stabilizing matrix for the
liposomes, potentially increasing the shelf life of these and
other liposome-based materials and drug formulations.
12910
dx.doi.org/10.1021/ja3051876 | J. Am. Chem. Soc. 2012, 134, 12908−12911

Journal of the American Chemical Society
Communication
Kowlgi, K. N.; Koper, G. J. M.; Eelkema, R.; van Esch, J. H. Angew.
Chem., Int. Ed. 2010, 49, 4825. (g) Li, X.; Kuang, Y.; Shi, J.; Gao, Y.;
Lin, H. C.; Xu, B. J. Am. Chem. Soc. 2011, 133, 17513.
(12) (a) Hanabusa, K.; Kawakami, A.; Kimura, M.; Shirai, H. Chem.
Lett. 1997, 26, 191. (b) van Bommel, K. J.; van der Pol, C.; Muizebelt,
I.; Friggeri, A.; Heeres, A.; Meetsma, A.; Feringa, B. L.; van Esch, J.
Angew. Chem., Int. Ed. 2004, 43, 1663.
fluorophore release rate, can be tuned by the heating time of
the platform. In future work, other enzymes, catalysts, and
associated addressable chemical bonds might be applied to
enhance biocompatibility and selectivity. In combination with
drug-carrying gelators,7,24 such a concept could find use in
future applications in personalized medicine or medical
implants, where the drug release rate can be tuned before
intake of the medicine or by locally induced hyperthermia.
(13) Brynes, P. J.; Bevilacqua, P.; Green, A. Anal. Biochem. 1981, 116,
408.
(14) van Bommel, K. J.; Stuart, M. C.; Feringa, B. L.; van Esch, J. Org.
Biomol. Chem. 2005, 3, 2917.
(15) Torchilin, V. P.; Weissig, V. Liposomes, a practical approach, 2nd
ed.; Oxford University Press: Oxford, 2003.
(16) Drummond, D. C.; Zignani, M.; Leroux, J. Prog. Lipid Res. 2000,
39, 409.
ASSOCIATED CONTENT
* Supporting Information
Experimental details and fluorescence spectroscopy data. This
material is available free of charge via the Internet at http://
pubs.acs.org.
■
S
(17) (a) Chen, H.; Langer, R. Pharm. Res. 1997, 14, 537.
(b) Babincova,
́ ́
M.; Cicmanec, P.; Altanerova, V.; Altaner, C.;
AUTHOR INFORMATION
Babinec, P. Bioelectrochemistry 2002, 55, 17. (c) Pradhan, P.; Giri, J.;
■
Rieken, F.; Koch, C.; Mykhaylyk, O.; Doblinger, M.; Banerjee, R.;
Bahadur, D.; Plank, C. J. Controlled Release 2010, 142, 108. (d) Amstad,
̈
Corresponding Author
j.h.vanesch@tudelft.nl
E.; Kohlbrecher, J.; Muller, E.; Schweizer, T.; Textor, M.; Reimhult, E.
̈
Present Address
Nano Lett. 2011, 11, 1664.
†Institute for BioNanotechnology in Medicine, Northwestern
University, Chicago, Illinois
(18) Ruiz, J.; Goni, F. M.; Alonso, A. Biochim. Biophys. Acta 1988,
̃
937, 127.
(19) (a) Huang, S. L.; MacDonald, R. C. Biochim. Biophys. Acta 2004,
Notes
1665, 134. (b) Bohmer, M. R.; Klibanov, A. L.; Tiemann, K.; Hall, C.
̈
The authors declare no competing financial interest.
S.; Gruell, H.; Steinbach, O. C. Eur. J. Radiol. 2009, 70, 242.
(20) Choucair, A.; Soo, P. L.; Eisenberg, A. Langmuir 2005, 21, 9308.
(21) Ponta, A.; Bae, Y. Pharm. Res. 2010, 27, 2330.
ACKNOWLEDGMENTS
■
The authors thank The Netherlands Organization for Scientific
Research (NWO) for financial support through a VICI grant,
and Dr. Kristina Djanashvili, Florian Mayer, and Dr. Nicholas
Stephanopoulos for valuable scientific discussions.
(22) The temperature to induce release from thermosensitive
liposomes can be set over a broad range by the bilayer gel−fluid
phase transition temperature Tm (see also ref 15). Here, a Tm of 42 °C
was chosen deliberately to ensure that the liposomes do not leak their
contents at body temperature; leakage of content can be induced by
controlled heating of the sample.
(23) At 42 °C, no denaturation of the enzyme is expected; however,
at this temperature the enzyme is more prone to auto-digestion:
Kawamura, Y.; Nakanishi, K.; Matsuno, R.; Kamikubo, T. Biotechnol.
Bioeng. 1981, 23, 1219.
(24) (a) Xing, B.; Yu, C. W.; Chow, K. H.; Ho, P. L.; Fu, D.; Xu, B. J.
Am. Chem. Soc. 2002, 124, 14846. (b) Li, X.; Li, J.; Gao, Y.; Kuang, Y.;
Shi, J.; Xu, B. J. Am. Chem. Soc. 2010, 132, 17707. (c) Casuso, P.;
Carrasco, P.; Loinaz, I.; Grande, H. J.; Odriozola, I. Org. Biomol. Chem.
2010, 8, 5455. (d) Mao, L.; Wang, H.; Tan, M.; Ou, L.; Kong, D.;
Yang, Z. Chem. Commun. 2011, 48, 395.
REFERENCES
■
(1) Riehemann, K.; Schneider, S. W.; Luger, T. A.; Godin, B.; Ferrari,
M.; Fuchs, H. Angew. Chem., Int. Ed. 2009, 48, 872.
(2) (a) Shah, R. N.; Shah, N. A.; Del Rosario Lim, M. M.; Hsieh, C.;
Nuber, G.; Stupp, S. I. Proc. Natl. Acad. Sci. U.S.A. 2010, 107, 3293.
(b) Cui, H.; Webber, M. J.; Stupp, S. I. Biopolymers 2010, 94, 1.
(3) (a) Djanashvili, K.; ten Hagen, T. L. M.; Blange,
Peters, J. A.; Koning, G. A. Bioorg. Med. Chem. 2011, 19, 1123. (b) de
Smet, M.; Langereis, S.; van den Bosch, S.; Grull, H. J. Controlled
Release 2010, 143, 120.
(4) Kelly, C.; Jefferies, C.; Cryan, S. A. J. Drug Deliv. 2011, 2011,
́
R.; Schipper, D.;
̈
No. 727241.
(5) Liu, L.; Xu, K.; Wang, H.; Tan, P. K.; Fan, W.; Venkatraman, S.
S.; Li, L.; Yang, Y. Y. Nat. Nanotechnol. 2009, 4, 457.
(6) Wood, K. C.; Boedicker, J. Q.; Lynn, D. M.; Hammond, P. T.
Langmuir 2005, 21, 1603.
(7) Zhao, F.; Ma, M. L.; Xu, B. Chem. Soc. Rev. 2009, 38, 883.
(8) Ponta, A.; Bae, Y. Pharm. Res. 2010, 27, 2330.
(9) Choucair, A.; Soo, P. L.; Eisenberg, A. Langmuir 2005, 21, 9308.
(10) (a) Brizard, A.; Stuart, M.; van Bommel, K.; Friggeri, A.; de
Jong, M.; van Esch, J. Angew. Chem., Int. Ed. 2008, 47, 2063.
(b) Moffat, J. R.; Smith, D. K. Chem. Commun. 2009, 316.
(c) Boekhoven, J.; van Rijn, P.; Brizard, A. M.; Stuart, M. C. A.; van
Esch, J. H. Chem. Commun. 2010, 3490. (d) Pal, A.; Besenius, P.;
Sijbesma, R. P. J. Am. Chem. Soc. 2011, 133, 12987. (e) Boekhoven, J.;
Brizard, A. M.; van Rijn, P.; Stuart, M. C. A.; Eelkema, R.; van Esch, J.
H. Angew. Chem., Int. Ed. 2011, 50, 12285.
(11) (a) Aggeli, A.; Nyrkova, I. A.; Bell, M.; Harding, R.; Carrick, L.;
McLeish, T. C.; Semenov, A. N.; Boden, N. Proc. Natl. Acad. Sci. U.S.A.
2001, 98, 11857. (b) George, M.; Weiss, R. G. J. Am. Chem. Soc. 2001,
123, 10393. (c) Del Guerzo, A.; Olive, A. G.; Reichwagen, J.; Hopf, H.;
Desvergne, J. P. J. Am. Chem. Soc. 2005, 127, 17984. (d) Toledano, S.;
Williams, R. J.; Jayawarna, V.; Ulijn, R. V. J. Am. Chem. Soc. 2006, 128,
1070. (e) Rodríguez-Llansola, F.; Escuder, B.; Miravet, J. F. J. Am.
Chem. Soc. 2009, 131, 11 478. (f) Boekhoven, J.; Brizard, A. M.;
12911
dx.doi.org/10.1021/ja3051876 | J. Am. Chem. Soc. 2012, 134, 12908−12911