Supramolecular polymers as dynamic multicomponent cellular uptake carriers

Article abstract of DOI:10.1021/ja3029075

Supramolecular synthesis represents a flexible approach to the generation of dynamic multicomponent materials with tunable properties. Here, cellular uptake systems based on dynamic supramolecular copolymers have been developed using a combination of differently functionalized discotic molecules. Discotics featuring peripheral amine functionalities that endow the supramolecular polymer with cellular uptake capabilities were readily synthesized. This enabled the uptake of otherwise cell-impermeable discotics via cotransport as a function of supramolecular coassembly. Dynamic multicomponent and multifunctional supramolecular polymers represent a novel and unique platform for modular cellular uptake systems.

Full text of DOI:10.1021/ja3029075

Communication
pubs.acs.org/JACS
Supramolecular Polymers as Dynamic Multicomponent Cellular
Uptake Carriers
Katja Petkau-Milroy,† Michael H. Sonntag,† Arthur H. A. M. van Onzen, and Luc Brunsveld*
Laboratory of Chemical Biology, Department of Biomedical Engineering, Eindhoven University of Technology, Den Dolech 2, 5612
AZ Eindhoven, The Netherlands
S
* Supporting Information
Chart 1. (a) Concept of Supramolecular Cellular Uptake
Carriers; (b) Library of Amine- And Ligand-Functionalized
Discotics
ABSTRACT: Supramolecular synthesis represents a
flexible approach to the generation of dynamic multi-
component materials with tunable properties. Here,
cellular uptake systems based on dynamic supramolecular
copolymers have been developed using a combination of
differently functionalized discotic molecules. Discotics
featuring peripheral amine functionalities that endow the
supramolecular polymer with cellular uptake capabilities
were readily synthesized. This enabled the uptake of
otherwise cell-impermeable discotics via cotransport as a
function of supramolecular coassembly. Dynamic multi-
component and multifunctional supramolecular polymers
represent a novel and unique platform for modular cellular
uptake systems.
he highly selective permeability of the plasma membrane,
Twhile essential for cell function and survival, represents as
well a major challenge for the intracellular delivery of cargo,
such as imaging agents, therapeutics, and reporter molecules.1
For cargo delivery, two strategies have been widely used: one is
based on ligand−receptor interactions,2 and the other uses cell-
penetrating peptides (CPPs).3,4 A key feature of CPPs5,6 is their
positive charge due to their high arginine and lysine content,
which favors binding to cell-membrane-bound proteoglycans
and leads to cellular uptake via endocytosis.7,8 This
phenomenon is not restricted only to polycationic peptides;
polyamine dendrimers,9,10 foldamers,11,12 and polymers13,14 are
similarly capable of crossing the cell membrane. Synthetic
supramolecular polymers15 have yet to be explored as
intracellular carriers, although their unique self-assembly
properties combined with the modular nature of their synthesis,
as in liposome-based systems,16 makes supramolecular
polymers attractive systems for cellular uptake. Dynamic and
adaptable multifunctional and multivalent self-assembled
scaffolds can be conveniently prepared without the need for
repeated syntheses,17,18 as is the case for conventional polymer-
based systems. We envisaged that the supramolecular synthesis
of heterogeneous noncovalent supramolecular polymers
comprising cell-penetrating monomers and non-cell-permeable
monomers would enable cellular uptake of the intermixed
system (Chart 1a). Here we demonstrate the use of
supramolecular synthesis to access in a rapid manner dynamic
multicomponent polymers of diverse composition with power-
ful cell-penetrating properties.
The supramolecular polymers used herein are C3-symmetric
amphiphilic discotics that self-assemble into columnar stacks at
dilute micromolar concentrations in water.19 The monomers
consist of an aromatic core shielded by nine biocompatible and
water-soluble poly(ethylene oxide) (PEO) chains, which can be
readily functionalized at the peripheral regions to enable, for
example, the attachment of bioactive ligands.20,21 The self-
assembly of these discotic molecules into columnar assemblies
induces strong autofluorescence with a large Stokes shift22,23
that is convenient for cellular imaging. For this work, we
designed a library of structurally related discotic molecules
differing only in the number of peripheral amine groups
(1NH2-Disc, 3NH2-Disc, and 9NH2-Disc; Chart 1b) with the
purpose of studying their cellular uptake properties. Together
with intrinsically non-cell-permeable discotic monomers
featuring either a single fluorescein fluorophore, or three
biotins, or only inert glycol side chains (1Fluorescein-Disc,
Received: March 26, 2012
Published: April 27, 2012
© 2012 American Chemical Society
8086
dx.doi.org/10.1021/ja3029075 | J. Am. Chem. Soc. 2012, 134, 8086−8089

Journal of the American Chemical Society
Communication
3Biotin-Disc, or Inert-Disc, respectively), a library of six
supramolecular discotic monomers was synthesized.
For example, the facile convergent synthesis of 9NH2-Disc
(Scheme 1) commenced with the reaction of methyl 3,4,5-
density of peripheral positive charge upon self-assembly into
columnar stacks. This explanation is supported by recent
studies of polyarginine peptides, which revealed that in the case
of linear peptides, a minimum of seven or eight arginines is
necessary for successful cellular uptake,25,26 whereas three
arginines presented on self-assembling vesicles were as effective
as the natural TAT sequence.27 Supramolecular assembly thus
allows for the generation of high-density positive charge
sufficient for effective cellular uptake using discotic monomers
featuring only a limited number of peripheral amine groups.
The supramolecular homopolymers were nontoxic toward
HeLa cells up to 10 μM (MTT assay; Figure S3).28 The cellular
uptake of the supramolecular polymers and their potential
cytotoxicity were further investigated for two additional cell
lines: HEK 293 and the difficult to transfect suspension cell line
THP-1. In both cases, similar effective uptake and low toxicity
was observed (Figures S4−S7). These results underline the
general cellular applicability of the supramolecular polymers.
Time-lapse confocal microscopy was performed to determine
the subcellular localization of the supramolecular polymers
(Figure 2a). Within 10 min of incubation with 3NH2-Disc or
a
Scheme 1. Synthesis of 9NH2-Disc
a
Reagents and conditions: (a) K2CO3, DMF, 70 °C, 10 h, 86%; (b)
KOH, 1:1 EtOH/H2O, 80 °C, 3 h, 97%; (c) 1-chloro-N,N-2-
trimethylpropenylamine, CH2Cl2, rt, 2 h, quant.; (d) 2,2′-bipyridine-
3,3′-diamine, NEt3, CH2Cl2, rt, 18 h, 61%; (e) 5, NEt3, CH2Cl2, 5 °C,
18 h, 32%; (f) TFA, CH2Cl2, rt, 2 h, 97%.
trihydroxybenzoate (1) with tosylate 2 and subsequent
saponification to yield wedge molecule 3 in an efficient 85%
yield over two steps. Acid chloride formation24 followed by
monoacylation of 4 afforded bipyridine wedge 5, whose
subsequent reaction with trimesic chloride 6 followed by
amine deprotection using trifluoroacetic acid (TFA) afforded
the desired 9NH2-Disc [for the syntheses of 1NH2-Disc and
3NH2-Disc, see the Supporting Information (SI)].
The cellular uptake of the supramolecular homopolymers by
adherent HeLa cells was evaluated using live-cell multiphoton
fluorescence microscopy. HeLa cells were incubated for 1 h
with different concentrations of Inert-Disc, 1NH2-Disc, 3NH2-
Disc, and 9NH2-Disc and imaged after washing. At high
concentrations (5 μM), no uptake could be observed for Inert-
Disc or 1NH2-Disc, which might be explained by the low
number of amines of these discotics. In contrast, cells incubated
with 3NH2-Disc or 9NH2-Disc featured cellular uptake, as
determined by the detection of discotic fluorescence, at both
high (5 μM) and low (0.5 μM) concentrations (Figure 1 and
Figure S2 in the SI). The effectiveness of the cellular uptake of
3NH2-Disc and 9NH2-Disc did not significantly differ in the
range tested (0.5−10 μM). The surprising excellent uptake
efficiency of 3NH2-Disc might be explained by the increased
Figure 2. (a) Cellular uptake of 3NH2-Disc (5 μM) by live HeLa cells
over time. (b) Confocal microscopy image of live HeLa cells incubated
with 3NH2-Disc (5 μM) overnight and co-stained with DiO, a marker
for cellular membranes that stains early endosomes as well as the cell
membrane.
9NH2-Disc, binding to the outer leaflet of the cell membrane
was observed, most probably resulting from electrostatic
interactions with cell-surface proteoglycans.8 Additionally,
cellular internalization, as indicated by increased fluorescence
within the cell, was observed after 30 min and increased over
time, leading to more punctuated fluorescence in the cytoplasm
that localized around the perinuclear region after 90 min. Co-
staining with the membrane and endosomal marker 3,3′-
dioctadecyloxacarbocyanine perchlorate (DiO) and the lysoso-
mal marker LysoTracker Red strongly indicated that the uptake
of the positively charged supramolecular polymers occurred via
endocytosis (Figure 2b and Figure S8). In addition to the
efficient uptake of amine-functionalized discotics, the intro-
duction of guanidinium groups might also be expected to
enhance the uptake and modulate localization, as has been seen
for CPPs.29,30
A potentially significant advantage of supramolecular
polymers over conventional polymers and small molecules is
their dynamic, multicomponent nature. Coassembly of different
monomers or the dynamic intermixing of supramolecular
Figure 1. Confocal microscopy images of the cellular uptake of Inert-
Disc, 1NH2-Disc, 3NH2-Disc, and 9NH2-Disc (all 5 μM) by HeLa
cells.
8087
dx.doi.org/10.1021/ja3029075 | J. Am. Chem. Soc. 2012, 134, 8086−8089

Journal of the American Chemical Society
Communication
polymers leads to dynamic supramolecular copolymers.31,32 To
investigate the potential of this concept, supramolecular
copolymers were prepared via mixing of a cell-penetrating
discotic monomer (3NH2-Disc or 9NH2-Disc) with a non-cell-
permeable monomer. If effective, the amine-functionalized
discotics might act as carrier molecules, facilitating the cellular
uptake of different non-cell-permeable discotics by virtue of
their combined presence in the supramolecular copolymer.
1Fluorescein-Disc alone was not taken up by HeLa cells, nor
was there any unspecific binding to the outer leaflet of the cell
membrane (Figure S9). HeLa cells were therefore incubated
with different copolymers of 9NH2-Disc and non-cell-
permeable 1Fluorescein-Disc.33 In contrast to 1Fluorescein-
Disc alone, cellular imaging after 30 min coincubation showed
a clear fluorescence signal along the cell membrane for both the
discotic fluorophore and the fluorescein (Figure S10). The cells
were subsequently washed and imaged 24 h after incubation, at
which point the copolymers had been internalized and
colocalization of the discotic and fluorescein signals was
observed inside the cell (Figure 3a). Similar results were
Finally, the versatility of the approach was highlighted by
evaluating a supramolecular copolymer consisting of three
different discotics: 9NH2-Disc, 1Fluorescein-Disc, and 3Bio-
tin-Disc (80:10:10). The copolymers were added to the cells
and incubated for 1 h, after which the cells were washed and
then fixed and immunostained after 24 h. Confocal microscopy
revealed the presence and colocalization of all three fluorescent
signals corresponding to the discotic scaffold, fluorescein, and
biotin inside the cell (Figure 4). These results highlight the
Figure 4. Confocal images of HeLa cells incubated for 1 h with a 5 μM
mixture of 9NH2-Disc, 1Fluorescein-Disc, and 3Biotin-Disc
(80:10:10). After 24 h, the cells were fixed and stained with a Cy5-
labeled antibiotin antibody.
potential of the supramolecular synthesis approach as a highly
flexible and effective means to internalize differently function-
alized discotic monomers and to express their functionality in
the combined system of a copolymer.
In conclusion, a library of cell-permeable and non-cell-
permeable ligand-functionalized discotic molecules was synthe-
sized to enable the supramolecular synthesis of a variety of
dynamic multicomponent copolymers with tunable properties.
Supramolecular homopolymers of amine-functionalized dis-
cotics demonstrated efficient cellular uptake that was detected
by means of their autofluorescence using live-cell multiphoton
fluorescence microscopy. Translocation through the cell
membrane by otherwise non-cell-permeable ligand-function-
alized discotics was induced by mixing them with amine-
functionalized discotics, leading to the in situ supramolecular
synthesis of cell-permeable supramolecular copolymers. The
rapid generation of copolymers with different compositions by
simple intermixing of a variety of readily synthesized discotic
monomers led to multifunctional supramolecular polymers
consisting of up to three different discotic monomers. After
efficient cellular uptake, each of the components could be
individually visualized, highlighting the potential of dynamic
multicomponent supramolecular polymers. In this example,
these polymers represent a unique and flexible platform for
modular cellular uptake systems, but other applications can be
similarly envisioned.
Figure 3. Discotic copolymers as cellular uptake carriers. (a) Confocal
images of living HeLa cells incubated for 1 h with a 5 μM mixture of
9NH2-Disc and 1Fluorescein-Disc (80:20) and imaged after 24 h. (b)
Confocal images of HeLa cells incubated for 1 h with a 5 μM mixture
of 9NH2-Disc and 3Biotin-Disc (80:20) and then fixed and stained
with a Cy5-labeled antibiotin antibody after 24 h.
observed for HEK 293 cells (Figure S11). These results
demonstrate that it is necessary for the 1Fluorescein-Disc to
assemble with 9NH2-Disc as a supramolecular copolymer in
order for cellular uptake to occur.34
ASSOCIATED CONTENT
* Supporting Information
■
The 3Biotin-Disc was non-cell-permeable when alone
(Figure S13) but successfully taken up when present as a
supramolecular copolymer with 9NH2-Disc. To visualize the
cellular uptake in this case, the cells were fixed and stained with
a Cy5-labeled antibiotin antibody (Figure 3b). Several
copolymers were prepared using different ratios of 9NH2-
Disc in combination with either 1Fluorescein-Disc or 3Biotin-
Disc (50:50, 80:20, 95:5). For each of these copolymers,
internalization of the normally cell-impermeable discotic was
observed. The 80:20 mixtures gave the best results in terms of
uptake and visualization. Similar results were obtained for
3NH2-Disc (Figure S14).
S
Materials, methods, detailed experimental procedures (syn-
thesis and cellular assays), compound characterization, and
supporting figures. This material is available free of charge via
the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
l.brunsveld@tue.nl
■
Author Contributions
†K.P.-M. and M.H.S. contributed equally.
8088
dx.doi.org/10.1021/ja3029075 | J. Am. Chem. Soc. 2012, 134, 8086−8089

Journal of the American Chemical Society
Communication
(31) Besenius, P.; Goedegebure, Y.; Driesse, M.; Koay, M.; Bomans,
P. H. H.; Palmans, A. R. A.; Dankers, P. Y. W.; Meijer, E. W. Soft
Matter 2011, 7, 7980.
Notes
The authors declare no competing financial interest.
(32) Gasiorowski, J. Z.; Collier, J. H. Biomacromolecules 2011, 12,
3549.
(33) To ensure intermixing of the discotics, the disc mixtures were
ACKNOWLEDGMENTS
■
This work was funded by ERC Grant 204554 − SupraChemBio
and STW Grant 11022 − NanoActuate. We thank Ralf Bovee
and Dr. Xianwen Lou for MALDI−TOF MS measurements
and Dr. Lech-Gustav Milroy for help with proofreading.
pre-equilibrated overnight at room temperature.
(34) As evidence that cellular uptake was caused by the supra-
molecular copolymerization of the discotic monomers and not, for
example, by permeabilization of the membrane through the addition of
9NH2-Disc, HeLa cells were incubated with a mixture of 9NH2-Disc
and the carboxyfluorescein fluorophore alone. When the same
incubation and imaging procedure as before was used, no fluorescein
signal was detected inside the cell (Figure S12).
REFERENCES
■
(1) Torchilin, V. P. Annu. Rev. Biomed. Eng. 2006, 8, 343.
̌
(2) Blanka, R. Adv. Drug Delivery Rev. 1998, 29, 273.
(3) Fonseca, S. B.; Pereira, M. P.; Kelley, S. O. Adv. Drug Delivery Rev.
2009, 61, 953.
(4) Stewart, K. M.; Horton, K. L.; Kelley, S. O. Org. Biomol. Chem.
2008, 6, 2242.
(5) Green, M.; Loewenstein, P. M. Cell 1988, 55, 1179.
(6) Frankel, A. D.; Pabo, C. O. Cell 1988, 55, 1189.
(7) Fischer, P. M. Med. Res. Rev. 2007, 27, 755.
(8) Nakase, I.; Tadokoro, A.; Kawabata, N.; Takeuchi, T.; Katoh, H.;
Hiramoto, K.; Negishi, M.; Nomizu, M.; Sugiura, Y.; Futaki, S.
Biochemistry 2007, 46, 492.
(9) Amir, R. J.; Albertazzi, L.; Willis, J.; Khan, A.; Kang, T.; Hawker,
C. J. Angew. Chem., Int. Ed. 2011, 50, 3425.
(10) Albertazzi, L.; Serresi, M.; Albanese, A.; Beltram, F. Mol.
Pharmaceutics 2010, 7, 680 and references therein.
(11) Gillies, E. R.; Deiss, F.; Staedel, C.; Schmitter, J.-M.; Huc, I.
Angew. Chem. 2007, 119, 4159.
(12) Iriondo-Alberdi, J.; Laxmi-Reddy, K.; Bouguerne, B.; Staedel, C.;
Huc, I. Chem.Eur. J. 2010, 11, 1679.
(13) Moon, J. H.; McDaniel, W.; MacLean, P.; Hancock, L. F. Angew.
Chem., Int. Ed. 2007, 46, 8223.
(14) Kolonko, E. M.; Kiessling, L. L. J. Am. Chem. Soc. 2008, 130,
5626.
(15) Brunsveld, L.; Folmer, B. J. B.; Meijer, E. W.; Sijbesma, R. P.
Chem. Rev. 2001, 101, 4071.
(16) Mulder, W. J. M.; Strijkers, G. J.; van Tilborg, G. A. F.;
Cormode, D. P.; Fayad, Z. A.; Nicolay, K. Acc. Chem. Res. 2009, 42,
904.
(17) Silva, G. A.; Czeisler, C.; Niece, K. L.; Beniash, E.; Harrington,
D. A.; Kessler, J. A.; Stupp, S. I. Science 2004, 303, 1352.
(18) Cui, H.; Webber, M. J.; Stupp, S. I. Biopolymers 2010, 94, 1.
(19) Brunsveld, L.; Lohmeijer, B. G. G.; Vekemans, J. A. J. M.;
Meijer, E. W. Chem. Commun. 2000, 2305.
(20) Muller, M. K.; Petkau, K.; Brunsveld, L. Chem. Commun. 2011,
̈
47, 310.
(21) Muller, M. K.; Brunsveld, L. Angew. Chem., Int. Ed. 2009, 48,
̈
2921.
(22) Brunsveld, L.; Zhang, H.; Glasbeek, M.; Vekemans, J. A. J. M.;
Meijer, E. W. J. Am. Chem. Soc. 2000, 122, 6175.
(23) Toele, P.; van Gorp, J. J.; Glasbeek, M. J. Phys. Chem. A 2005,
109, 10479.
(24) Ghosez, L.; Haveaux, B.; Viehe, H. G. Angew. Chem., Int. Ed.
Engl. 1969, 8, 454.
(25) Kosuge, M.; Takeuchi, T.; Nakase, I.; Jones, A. T.; Futaki, S.
Bioconjugate Chem. 2008, 19, 656.
(26) Futaki, S.; Suzuki, T.; Ohashi, W.; Yagami, T.; Tanaka, S.; Ueda,
K.; Sugiura, Y. J. Biol. Chem. 2001, 276, 5836.
(27) Yoon, Y.-R.; Lim, Y.; Lee, E.; Lee, M. Chem. Commun. 2008,
1892.
(28) The discotic scaffold itself (Inert-Disc) showed no toxicity,
probably because of the shielding of the aromatic core by the nontoxic,
biocompatible PEOs.
(29) Thoren, P. Biochem. Biophys. Res. Commun. 2003, 307, 100.
(30) Mitchell, D. J.; Steinman, L.; Kim, D. T.; Fathman, C. G.;
Rothbard, J. B. J. Pept. Res. 2000, 56, 318.
8089
dx.doi.org/10.1021/ja3029075 | J. Am. Chem. Soc. 2012, 134, 8086−8089