A toolbox for controlling the properties and functionalisation of hydrazone-based supramolecular hydrogels

Article abstract of DOI:10.1039/c5tb01870f

In recent years, we have developed a low molecular weight hydrogelator system that is formed in situ under ambient conditions through catalysed hydrazone formation between two individually non-gelating components. In this contribution, we describe a molecular toolbox based on this system which allows us to (1) investigate the limits of gel formation and fine-tuning of their bulk properties, (2) introduce multicolour fluorescent probes in an easy fashion to enable high-resolution imaging, and (3) chemically modify the supramolecular gel fibres through click and non-covalent chemistry, to expand the functionality of the resultant materials. In this paper we show preliminary applications of this toolbox, enabling covalent and non-covalent functionalisation of the gel network with proteins and multicolour imaging of hydrogel networks with embedded mammalian cells and their substructures. Overall, the results show that the toolbox allows for on demand gel network visualisation and functionalisation, enabling a wealth of applications in the areas of chemical biology and smart materials.

Full text of DOI:10.1039/c5tb01870f

Journal of
Materials Chemistry B
View Article Online
View Journal
PAPER
A toolbox for controlling the properties and
functionalisation of hydrazone-based
Cite this:DOI: 10.1039/c5tb01870f
supramolecular hydrogels†
Jos M. Poolman,‡a Chandan Maity,‡a Job Boekhoven,ab Lars van der Mee,a
Vincent A. A. le Sage,a G. J. Mirjam Groenewold,c Sander I. van Kasteren,c
Frank Versluis,a Jan H. van Esch*a and Rienk Eelkema*a
In recent years, we have developed a low molecular weight hydrogelator system that is formed in situ
under ambient conditions through catalysed hydrazone formation between two individually non-gelating
components. In this contribution, we describe a molecular toolbox based on this system which allows
us to (1) investigate the limits of gel formation and fine-tuning of their bulk properties, (2) introduce
multicolour fluorescent probes in an easy fashion to enable high-resolution imaging, and (3) chemically
modify the supramolecular gel fibres through click and non-covalent chemistry, to expand the
functionality of the resultant materials. In this paper we show preliminary applications of this toolbox,
enabling covalent and non-covalent functionalisation of the gel network with proteins and multicolour
imaging of hydrogel networks with embedded mammalian cells and their substructures. Overall, the
results show that the toolbox allows for on demand gel network visualisation and functionalisation,
enabling a wealth of applications in the areas of chemical biology and smart materials.
Received 9th September 2015,
Accepted 21st December 2015
DOI: 10.1039/c5tb01870f
www.rsc.org/MaterialsB
non-covalent interactions.22–28 Nevertheless, it remains difficult
and laborious to modulate the properties of the obtained hydrogels,
Introduction
Supramolecular systems consist of self-assembled structures as it requires complete synthesis of a new hydrogelator.
that are held together by non-covalent interactions. By carefully Furthermore, introduction of functional groups on the fibrillar
balancing the strength of the individually weak interactions, networks typically requires multi-step syntheses of function-
relatively robust architectures can be devised, retaining the ability alised monomers,29–31 rendering gel functionalisation an inefficient
to respond to environmental changes such as temperature,1 process. To overcome these challenges, we present a modular
light2 and pH.3 In recent years, a wealth of systems capable hydrogelator system that spontaneously forms in situ by chemical
of forming supramolecular fibres and hydrogels has been bond formation between non-gelating precursors. The chemical
reported.4,5 Due to the reversible nature of the bonds that hold structure and properties of the resulting materials can be easily
the network together, supramolecular hydrogels find applica- tuned by slight modifications of the precursor molecules. These
tions as biomaterials,6 sensors7 and host–guest systems.8 The precursor molecules are synthesised in a few, straightforward
gelators are often derived from biological materials such as steps from readily available starting materials. Moreover, by
peptides,9–12 nucleobases,13–15 saccharides16–18 and hybrids mixing in small amounts of functionalised precursors with the
thereof.19–21 Alternatively, small synthetic organic compounds standard precursors, hydrogel networks are obtained that are
have been used extensively to form hydrogels based on molecularly designed to display specific functions. In all, this
strategy allows for easy tuning of hydrogel properties and
further modification of the fibre network.
a Advanced Soft Matter, Department of Chemical Engineering,
Recently, we have developed a supramolecular hydrogelator
Delft University of Technology, Julianalaan 136, 2628 BL Delft, The Netherlands.
that is formed in situ through hydrazone formation between
water-soluble hydrazide H1 and aldehyde A11 under ambient
conditions (Fig. 1).32,33 Gels will form spontaneously at room
temperature after mixing of the precursors, without the need
for a heating step to aid dissolution of the gelator. The rate
of hydrazone formation can be increased through the use of
catalysts such as aniline or acid, reducing gelation times from
E-mail: r.eelkema@tudelft.nl, j.h.vanesch@tudelft.nl
b Institute for Advanced Study and Department of Chemistry,
¨
¨
Technische Universitat Munchen, Lichtenbergstrasse 2A, 85748 Garching, Germany
c Leiden Institute of Chemistry and The Institute for Chemical Immunology,
Leiden University, Einsteinweg 55, 2333 CC Leiden, The Netherlands
† Electronic supplementary information (ESI) available: Synthesis, characterisa-
tion, and experimental procedures. See DOI: 10.1039/c5tb01870f
‡ These authors contributed equally.
This journal is ©The Royal Society of Chemistry 2016
J. Mater. Chem. B

View Article Online
Paper
Journal of Materials Chemistry B
Critical gelation concentration experiments
Gels were formed by preparing stock solutions of the aldehyde
(240 mM) and hydrazide (40 mM) in 100 mM sodium phosphate
buffer (Na2HPO4/NaH2PO4, pH 5.0). Appropriate amounts of the
stock solutions were mixed and vortexed vigorously for three
seconds. The mixed samples were allowed to gel overnight and
gelation was checked using the inverted vial test. The critical gel
concentration (CGC) is expressed as the initial concentration
(in mM) of the hydrazide precursor in the mixture and defined as
the point between the lowest concentration at which gels were
formed and the highest concentration at which the solvent
cannot be supported.
Fig. 1 Schematic representation of catalysed gel formation. Soluble, non-
gelating starting components H1 and A11 react to give a self-assembling
hydrogelator molecule, which assembles into fibres and finally a network,
gelling the solvent.
Confocal laser scanning microscopy
hours to minutes. Upon reaching a critical minimum concen-
tration, the hydrazone gelator self-assembles into fibres, which in
due course form a network capable of retaining water. The use of
catalysis supplies us with a handle through which the rate of
formation and mechanical properties of the gel network can be
controlled.33 Additionally, we have shown that through the catalyst
the spatial distribution of gel formation can be controlled, using
micro-patterned catalytic surfaces34 or a light triggered catalyst.35 In
this contribution we present a molecular toolbox which is aimed at:
(1) investigating the molecular constraints of gel formation, (2)
covalently functionalising the network with a variety of fluorescent
probes and (3) the integration of reactive groups in the hydrogel
network, allowing for further modification of the gel fibres (Fig. 2).
Our approach enables easy tuning of the physical properties of the
gel network and its further functionalisation. We accomplish this
either using derivatives of the aldehyde and hydrazide precursors as
the majority components in the gelation mixture, or by mixing in
small amounts of functionalised derivatives of the benzaldehyde
precursor with the standard aldehyde precursor A11. This makes
our gel system highly versatile, enabling the incorporation of a wide
variety of benzaldehyde-derived functional compounds into the gel
network. Additionally, the synthesis of these derivatives is generally
straightforward and can be achieved in a few steps from commonly
available precursors.
Stock solution of H1, A11, functionalised aldehydes (A14–A23,
A25) and reactive compounds (azide-fluor 545, FITC-SH, Fluor
488-alkyne, Cy5-labelled azide-ovalbumin and fluorescein labelled
Concanavalin A) were prepared in phosphate buffer of pH 5.0 (see
ESI† for details). Gel samples were prepared by mixing an appro-
priate amount of stock solutions of H1, A11 and the function-
alised aldehyde. The mixture was immediately deposited into a
polydimethylsiloxane cuvette or an imaging chamber (diameter ꢀ
thickness = 20 mm ꢀ 0.6 mm), closed off with a glass cover slide
and was allowed to stand overnight. Confocal laser scanning
micrographs were obtained in the fluorescence mode on a Zeiss
LSM 700. The laser beam was focused on a 40ꢀ oil immersion
objective and the sensitivity of detectors and filters was adjusted
accordingly to obtain the maximum signal to noise ratio. CLSM
micrographs (Fig. 7 and ESI†) were recorded using the same
settings (apart from the excitation wavelength) to allow compar-
ison of the fluorescence intensity for analysis purposes.
Results and discussion
In a first study, we examined the influence of the molecular
design of the precursors on gelation behaviour. To do so, we
synthesised a small library of aldehyde and hydrazide deriva-
tives, and tested all possible combinations for their gelation
behaviour in a simple mixing + reaction experiment (Fig. 3 and
Table 1). The benzaldehyde derivatives were functionalised with
one or two ethylene glycol (EG) chains of varying length (Fig. 3),
Experimental section
Synthesis
Detailed synthetic procedures and characterisation are found
in the ESI.†
Fig. 2 Concept of functionalised gel fibre network formation via partial
replacement of the original aldehyde (A11) by a functionalised aldehyde
creating a gel with tailored functionalities.
Fig. 3 Molecular structures of the hydrazide and aldehyde derivatives.
J. Mater. Chem. B
This journal is ©The Royal Society of Chemistry 2016

View Article Online
Journal of Materials Chemistry B
Paper
Table 1 Critical gel concentration (CGC) tests with hydrazide and alde-
hyde derivatives in a pH 5 sodium phosphate buffer under ambient
conditions. The tests were performed by mixing the appropriate amount
of hydrazide and aldehyde precursors in a sample vial and allowing the
samples to stand overnight. Results after inverting the vial were classified
into precipitation (P), homogeneous solution (S), opaque gel (G) and weak
gel (WG), which is a gel which is not capable of supporting its own weight
for 430 seconds. The given numbers are CGC values in mM, where CGC is
defined as the minimum concentration of the hydrazide precursor to form
a gel that withstands gravity when turned upside down
instead of gelation. In contrast, benzaldehyde derivatives bear-
ing more than four EG monomers (A12, A13) yielded too water-
soluble hydrazone materials, maintaining clear solutions after
reaction (Table 1). In between these two extremes, however,
opaque gels with a water content of up to 99.5% (w/w) were
obtained from benzaldehyde derivatives modified with 2 to 4
EG, pointing towards a balance between the crystallisation and
solubilisation processes. para-Functionalised benzaldehyde
derivatives (A2–A5) provided stable hydrogels. The optimum
gelation efficiency (i.e. lowest CGC, 3.5 mM) was obtained using
three EG units (A3), whereas benzaldehyde derivatives with
shorter (A2) and longer EG (A4, A5) needed higher gelator
concentrations (8, 11 and 13 mM respectively). In contrast,
meta-functionalised benzaldehyde derivatives (A7–A9) showed
poor solubility in water compared to their para-functionalised
counterparts. This effect is presumably due to higher aromatic
surface exposure towards the aqueous environment. Reacting
H1 with either A7 or A9 (containing 2 and 4 EG, resp.) yielded
gels at the respective concentrations of 5.5 mM and 16 mM,
whereas A8 with 3 EG provided a hydrazone material that
precipitated in the form of oil droplets at the bottom of the vial.
The meta- and para-functionalised aldehyde (A11) with four EG
provided a hydrogel with a CGC of 4.5 mM. Furthermore, a
which increases the water solubility of the compounds due to the benzaldehyde derivative with a terminal hydroxy group on the
hydrophilic nature of EG and possibly prevents precipitation of EG (A5) gelated less efficiently than the corresponding derivative
the corresponding hydrazone through favourable interactions with a methyl terminated chain (A3).
between the supramolecular fibres and the solvent.36 These
Mixing benzaldehyde derivatives with dihydrazide compounds
functions of the EG chains are assumed to be of paramount (H2–H7) generally afforded gels at higher minimum gelation
importance for the self-assembly of the hydrazone materials into concentrations (14.5–25 mM) than for H1 (3.5–16 mM). Also,
fibres and subsequent gelation. The hydrazide series contained a the range of benzaldehyde derivatives that would yield a gel was
trishydrazide (H1) and six bishydrazide (H2–H7) compounds, with narrower (Table 1). Remarkably, dihydrazides with a spacer
a varying spacer length between the two hydrazide moieties, in the composed of an even number of carbon atoms (H2, H4, and
range of 2–7 carbon atoms. The hydrazides (H1–H7) were either H6) yielded hydrogels, whereas precipitation was observed for
commercially available or easily synthesised in one step from their the dihydrazides with an odd numbered carbon spacer (H3, H5,
corresponding methyl esters in quantitative yield without any and H7). Upon reaction with meta-functionalised benzalde-
purification steps. The aldehyde derivatives (A2–A4, A7–A9, and hydes (A7–A9), dihydrazides with an even carbon atom spacer
A11–A13) were synthesised in good yield (78–99%) from the yielded crystalline precipitates whereas hydrazone materials
corresponding hydroxybenzaldehydes (A1, A6 or A10) by reaction with odd carbon atom spacer phase separated in the form of
with the relevant tosylated ethylene glycol monomethyl ether. oil droplets. Increasing the length of the carbon atom spacer
Compound A5 was synthesised in a single step by reacting A1 and thereby decreasing hydrophobicity has a favourable effect
and 2-[2-(2-chloroethoxy)ethoxy]ethanol.
on the gelation of the formed hydrazones as lower gelator
Critical gel concentration (CGC) tests were performed to concentrations are observed with increasing chain length,
investigate the influence of the molecular design of the starting eventually limited by the solubility of the hydrazide derivative.
aldehydes and hydrazides on the gelation behaviour of the
For selected samples, we determined the stiffness of the
resulting hydrazone products. In a typical experiment, the aldehyde obtained gels using oscillatory rheology (Fig. 4a). The elastic
and hydrazide derivatives were dissolved separately in a phosphate modulus G0 varied over three orders of magnitude, between
buffer of pH 5. Whereas the starting materials did not self- 45 Pa (H2 + A3) and 1.2 ꢀ 105 Pa (H1 + A11). In general, an inverse
assemble under these conditions, mixing the two solutions resulted relationship between the observed elastic modulus and the CGC
in conversion to the expected gelating hydrazone products.32 can be observed (Fig. 4b), showing that a lower CGC in general
To promote the full conversion of hydrazides into hydrazones, predicts a higher elastic modulus of the obtained material. As all
two equivalents of aldehyde were added for each hydrazide samples were measured at a fixed 20 mM concentration, this
group. Evaluation of the CGC tests was performed by the general trend partly originates from the higher fibre density for
inverted vial test after the solution had been allowed to stand lower-CGC samples, leading to higher G0. Fig. 4b also shows a
overnight at room temperature.
bifurcation in the CGC-G0 dependence, with a steeper dependence
Mixing hydrazide H1 with non-EG-modified benzaldehyde for H1 derived gels compared to those derived from H2–6. Such a
derivatives (A1, A6, A10) yielded precipitated hydrazone materials difference in behaviour may point to a different mode of gelation.
This journal is ©The Royal Society of Chemistry 2016
J. Mater. Chem. B

View Article Online
Paper
Journal of Materials Chemistry B
with the aid of a fluorescent probe, illustrating the need for
flexibility in the choice of probe to simultaneously image the
hydrogel network structure. Therefore, we synthesised fluorescent
styryl, coumarin, fluorescein, rhodamine and cyanine derivatives
that contain an aldehyde functionality (Fig. 5).
Styryl labelled aldehydes (A14, A15) were synthesised from
terephthalaldehyde via a condensation reaction with the corres-
ponding pyridinium (yielding A14) and indolium (yielding A15)
sulfonate salts. Coumarin labelled aldehyde (A16) was synthe-
sised via allylic oxidation of the corresponding methyl derivative.
Fluorescein (A17), rhodamine (A18) and cyanine (A19) labelled
aldehydes were acquired via reaction of 4-hydroxybenzaldehyde
with fluorescein isothiocyanate, rhodamine B isothiocyanate and
IR-783, respectively. After obtaining compounds A14–A19, gels were
formed using hydrazide H1 and aldehyde A11, incorporating
0.1 mol% (with respect to aldehyde A11) of the fluorescently
labelled aldehyde. Subsequently, the structure of the resulting
gel networks were analysed using CLSM (Fig. 6). In all cases, a
branched network of fibre bundles is observed, showing that
the fluorescently labelled aldehydes are incorporated into the
gel network, without observably disturbing fibre formation.
Furthermore, exciting the various probes was achieved using
lasers with different wavelengths, depending on their absorp-
tion characteristics. The styryl (A14, A15) and coumarin (A16)
derivatives were excited at 405 nm, whereas the fluorescein
(A17), rhodamine (A18), and cyanine derivatives (A19) required
488, 543 and 633 nm lasers, respectively. Therefore, this toolbox
of fluorescently labelled aldehydes should allow for the intro-
duction of other species with orthogonal fluorescent labels into
our gel system (e.g. cells).
Fig. 4 Rheological data of selected hydrogels. (a) Elastic moduli (G0),
measured for 20 mM gels, 20 mM of any hydrazide derivative, 120 mM
of aldehyde derivative (for H1) or 80 mM of aldehyde derivative (for H2–6).
(b) A plot of G0 versus the inverse of the CGC. For each point, the
corresponding gel mixture is indicated. The H2 + A3 data point is indicated
in parentheses, as its CGC was determined at 425 mM (see Table 1).
To get some insight into the morphology of these gels, we
collected TEM micrographs for the H1 + A2, H1 + A7, H1 + A11
and H6 + A2 gels. All H1 derived gels gave dense fibre networks
consisting of fibres with diameters o10 nm. The H6 + A2 gel,
however, showed plate-type structures with dimensions in the
micrometre range (see the ESI†). This change in morphology
may account for the marked difference in both CGC and
the relationship between CGC and G0, when compared to H1
derived gels.
As shown above, our supramolecular gel system enables the
tuning of gelation properties such as CGC, elastic modulus or
network morphology by changing the aldehyde or hydrazide
structure. Next, we made use of this modularity to incorporate
chemical labels in the structure. For these experiments, we
used the well-characterised H1–A11 system32,33 as a common
gelator. In a first example, we replaced small fractions of the
reacting aldehyde derivatives with aldehyde-derived fluorophores
(Fig. 5). As we observed earlier that the trishydrazide (H1) showed
gelation after reaction with a wide range of aldehyde derivatives
(Table 1), we anticipated that small percentages of other aldehyde
derivatives can be incorporated into the supramolecular gel
network without hindering gel formation. Using fluorescently
labelled aldehyde derivatives allows for imaging of the structure
of the supramolecular network using confocal laser scanning
microscopy (CLSM). We used various fluorescent probes, with
different excitation wavelengths. This strategy increases our
flexibility with respect to future applications of our hydrogel
system, for instance as scaffolds for 3D cell culturing. There,
typically, imaging of cell membranes or nuclei will be performed
Encouraged by the above results, we set out to demonstrate the
use of our fluorescent gel network in combination with fluores-
cently tagged mammalian cells (mouse brain derived astrocytes).
In a preliminary set of experiments, we first fluorescently
Fig. 6 Confocal laser scanning fluorescence micrographs (CLSM) of
gel networks made using various aldehyde-derived fluorescent probes
(scale bar = 20 mm): (a) styryl-pyridinium derivative (A14), (b) styryl-indolium
derivative (A15), (c) coumarin derivative (A16), (d) fluorescein derivative (A17),
(e) rhodamine derivative (A18), and (f) cyanine derivative (A19). General
conditions: [H1] : [A11] = 5 : 30 mM, a 1 : 2 ratio of hydrazide to aldehyde
functional groups at pH 5, fluorophore labelled aldehyde (30 mM, 0.1 mol%
with respect to A11).
Fig. 5 Molecular structures of fluorophore labelled aldehydes.
J. Mater. Chem. B
This journal is ©The Royal Society of Chemistry 2016

View Article Online
Journal of Materials Chemistry B
Paper
Fig. 7 Merged confocal laser scanning micrographs (CLSM) of astrocytes
embedded in a gel network made of H1, A11 and A18 (scale bar = 20 mm).
(a) The green channel shows calcein-AM-stained cells (lexc. = 488 nm) and
the red channel gel fibres (lexc. = 543 nm), (b) nuclei are shown in blue
(lexc. = 404 nm), the green channel are actin filaments (lexc. = 488 nm) and
the red channel represents gel fibres (lexc. = 543 nm). Concentrations:
[H1] : [A11] = 10 : 60 mM, [A18] = 30 mM.
labelled these cells with calcein-AM, a green marker for viable
cells. These cells were combined with our gel network that was
functionalized with the red fluorescent A18 dye. Using fluores-
cence microscopy, we imaged the cells in the green channel and
the fibres in the red channel and found the cells embedded in the
gel network (Fig. 7a). Secondly, we used the same astrocytes, but
Fig. 8 Synthesis of aldehyde derivatives with functional groups for fibre-
modification. Reagents and conditions: (i) Br–CH2CRCH, NaH, THF, rt;
(ii) ClCOCHQCH2, Et3N, CH2Cl2, 0 1C - rt; (iii) Ts-Cl, Et3N, CH2Cl2, rt;
(iv) NaN3, DMF, D; (v) 2,3,4,6-tetraacetylpropargyl-a-mannoside, CuI, Et3N,
CH2Cl2, rt; (vi) NaOMe, MeOH, rt.
in this case the cells were fixed and permeabilised. Subsequently, attempted by forming gels with H1 and A11, mixed with A20 or
their actin filaments were labelled with a fluorescein-conjugated A23 (0.3 mol%, with respect to A11). Styryl labelled aldehyde
phalloidin (green) and their nuclei with DAPI (blue). Finally, the A15 (0.05 mol%) was used for imaging the gel network (Fig. 9).
gel network with the red fluorescent A18 dye was added to During gel formation, azide functionalised rhodamine (azide-
demonstrate the orthogonality of the fluorescent dyes. Using fluor 545) was added to the gelation mixture containing acetylene
confocal microscopy, we imaged the system and found the derivatised aldehyde A20, whereas acetylene functionalised rhod-
nucleus, actin filaments and gel network individually marked amine (fluor 488-alkyne) was added to the gelation mixture
blue, green and red, respectively (Fig. 7b).
containing azido-aldehyde A23. CuI was added to catalyse the
We further expanded the toolbox of functional precursors by click reaction. After gelation, confocal imaging showed colocalisa-
synthesising aldehyde derivatives with reactive groups, allow- tion of the rhodamine fluorescence with the styryl fluorescence of
ing for modification of the fibre network, either through the the gel network (Fig. 9). These findings clearly demonstrate suc-
formation of permanent covalent bonds via click chemistry,37,38 cessful functionalisation of styryl stained fibres with a rhodamine
or by non-covalent interactions with biomolecules. Three alde-
hyde derivatives were synthesised containing functionalities for
either Michael addition or alkyne–azide click chemistry, and an
a-mannose derived aldehyde was synthesised to decorate the
fibres with sugar groups, and to show non-covalent binding
of the lectin Concanavalin A (ConA) to the gel fibres.39 The
aldehydes with reactive groups were synthesised from 4-(2-(2-(2-
hydroxyethoxy)ethoxy)ethoxy)benzaldehyde (A5) (Fig. 8). Alkyne
(A20) and enone (A21) labelled aldehyde derivatives were synthe-
sised via nucleophilic substitution of A5 to propagyl bromide and
acryloyl chloride respectively. Alternatively, the hydroxy group of
A5 was converted to a tosylate group (A22) via the treatment with
tosyl chloride. An azide derivative (A23) was obtained from A22
via nucleophilic substitution with the azide ion. A Cu-catalysed
cycloaddition of A22 with the alkyne-functionalised 2,3,4,6-
tetraacetylpropargyl-a-mannoside provided the acetate-protected
mannose-benzaldehyde conjugate A24, which afforded the mannose
alised gel networks (scale bar = 20 mm) in the presence of A20 and
Fig. 9 Confocal laser scanning fluorescence micrographs of function-
derivative A25 via removal of the acetate groups.
azide-fluor 545 (a, f and k), A21 and FITC-SH (b, g and l), A23 and fluor
488-alkyne (c, h and m), A20 and Cy5-azide-Ova (d, i and n), and A25 with
FITC-ConA (e, j and o): (a–d) lexc. = 405 nm (exciting the A15 styryl dye),
(e and g) lexc. = 488 nm (exciting FITC), (f) lexc. = 543 nm (exciting fluor-
545), (h) lexc. = 488 nm (exciting fluor-488), (i) lexc. = 633 nm (exciting
Cy5), (j) lexc. = 543 nm (exciting A18 rhodamine), (k) lexc. = 405 & 543 nm,
(l) lexc. = 405 & 488 nm, (m) lexc. = 405 & 488 nm, (n) lexc. = 405 & 633 nm
and (o) lexc. = 488 & 543 nm. For general conditions, see the ESI.†
Incorporation of alkyne bearing aldehyde A20 into the net-
work allows for covalent addition of azide compounds, whereas
the azide containing compound A23 enables the Cu mediated
azide alkyne Huisgen cycloaddition with any alkyne derivative.
Incorporation of alkene A21 allows for Michael addition of
thiols. Gel network functionalisation with A20 and A23 was
This journal is ©The Royal Society of Chemistry 2016
J. Mater. Chem. B

View Article Online
Paper
Journal of Materials Chemistry B
dye using click chemistry. Similar to the Cu-catalysed click reaction were observed (see the ESI†). These experiments show that it is
to functionalise azide or alkyne derived gelators, a thia-Michael possible to functionalise the fibres with aldehyde-derived sugar
click reaction was performed during gelation to generate function- groups, and that such functional groups can subsequently be
alised gel fibres. During gelation, 0.5 mol% enone-derived A21 used for non-covalent binding to other (protein) biomolecules.
(with respect to A11) and fluorescein labelled thiol (FITC-SH) Overall, these examples clearly illustrate that modification of the
were added to the mixture. The gel network was visualised gel network can be achieved using these functional aldehydes and
using incorporated A15 and again, the fluorescence colocalisa- shows the utility of the various fluorescently labelled aldehydes
tion of fluorescein and A15 indicated functionalisation of the that we developed.
fibres using the click reaction (Fig. 9).
To determine the scope of the approach, we also applied this
methodology to functionalise the gel network with various Conclusions
biomolecules. We first attempted the direct attachment of a
In this paper, we present a molecular toolbox for controlling the
protein to the gel by performing a click-reaction of a model
gelation properties and functionalisation of hydrazone-derived
antigenic protein (ovalbumin,40 Ova), with the gel matrix. Ova is
supramolecular gels. These gels are made in the gelation medium
the most commonly used model antigen in immune assays. In
under ambient conditions by hydrazone formation between simple,
future, ovalbumin-conjugation to the gel matrix would allow
non-gelating building blocks. First we show that by varying the
the study of any immunogenicity of these gel matrices using
chemical structure and hydrophilicity of the individually non-
standard T-cell assays.41 The protein was functionalised with
gelating starting components the gelation behaviour of the result-
azides by expression in a methionine auxotrophic strain of
ing hydrazone compounds can be controlled. Next, we show how
E. coli to replace all methionines with azidohomoalanines42–44
replacing a small part of the regular aldehyde by an aldehyde
without affecting the structure and function of the protein.45
carrying a functional tag can be used to introduce functional
The protein was further functionalised with Cy5 dyes through
groups in the supramolecular gels. Introduction of various fluor-
lysine modification, to allow facile visualisation using confocal
microscopy. By adding the azide-Ova to a gelation mixture
containing Cu(I) and the acetylene-aldehyde A20, as well as the
styryl-aldehyde probe A15 for visualisation of the fibre network, an
Ova-functionalised gel was obtained. CLSM showed extensive
colocalisation of the styryl and Cy5 dyes indicating binding of
Ova to the fibres (Fig. 9d, i and n). Controls with wild-type Ova
escent probes in the gel network was performed using fluorescently
labelled aldehydes. Confocal imaging showed that these fluores-
cent probes can be used for imaging of the gel network at a wide
range of excitation wavelengths. This enables multicolour imaging
which allows for the visualisation of hydrogel networks in conjunc-
tion with cells and cellular substructures. Next, aldehyde derivatives
with various functional groups were synthesised, allowing for facile
(without azide tag) or by leaving out the acetylene-aldehyde from
covalent and non-covalent modification of the supramolecular
the gelation mixture showed a severely weakened intensity of the
fibres. In particular, various labels for click-chemistry were intro-
fluorescence intensity of the Cy5 probe, as well as a diminished
duced, as well as a sugar-derived functional group capable of non-
colocalisation with the styryl probe, indicating a lack of binding to
the fibres (see Fig. S7, ESI†). As a second class of biomolecules, we
functionalised the fibres with sugar groups. To demonstrate that
aldehyde A25 allows for the functionalisation of the gel network
with mannose moieties, gels were formed using trishydrazide H1
and aldehyde A11, mixed with 1 mol% (with respect to aldehyde
A11) mannose derivative A25. Furthermore, rhodamine labelled
aldehyde A18 was incorporated (0.02 mol%) to enable imaging of
covalent binding to a lectin protein. The click groups were used for
in situ connection to fluorescent labels or protein tags. The results
show that the gel system can be modified on demand without
hindering gel formation, enabling applications in the area of smart
materials and chemical biology. Moreover, the modular nature of
the gel formation and the facile synthesis of the various building
blocks make this a straightforward method to construct complex
functional materials from simple building blocks, eliminating the
the gel network using CLSM (Fig. 9). During gel formation, a
fluorescein derivative of the lectin Concanavalin A (ConA) was
need for long and difficult synthetic procedures.
added, which was expected to bind specifically to the mannose
present in the gel fibres.30 Confocal imaging showed that the
ConA fluorescence (Fig. 9) displayed extensive colocalisation with
Acknowledgements
the rhodamine fluorescence of the gel network, indicating func-
We thank the Netherlands Organization for Scientific Research
tionalisation of the fibre network with mannose and selective
(NWO VENI and VIDI grants to R. E., ECHO grant to J. P., R. E.,
binding to the lectin. Control experiments, in which the gel
J. H. v. E.) and the European Research Council (ERC Starting
sample was prepared without addition of the functional alde-
Grant #639005 to S. I. v. K.) for funding.
hydes, did not show fluorescence colocalization by CLSM from
two different fluorophores used in the sample (see Fig. S6, ESI†),
thus ruling out the possibility of non-specific binding of the
added functional groups to the gel fibres. Using TEM, we inves-
References
tigated the impact of the incorporation of a fluorophore (A18) or
the A25–ConA pair on the morphology of H1 + A11 fibres. In both
cases, no significant changes in morphology or fibre diameter
1 P. Cordier, F. Tournilhac, C. Soulie-Ziakovic and L. Leibler,
Nature, 2008, 451, 977–980.
2 Y. Zhao, Macromolecules, 2012, 45, 3647–3657.
J. Mater. Chem. B
This journal is ©The Royal Society of Chemistry 2016

View Article Online
Journal of Materials Chemistry B
Paper
3 H. Frisch and P. Besenius, Macromol. Rapid Commun., 2015, 25 Y. Kuang, J. Shi, J. Li, D. Yuan, K. A. Alberti, Q. Xu and B. Xu,
36, 346–363. Angew. Chem., Int. Ed., 2014, 53, 8104–8107.
4 L. E. Buerkle and S. J. Rowan, Chem. Soc. Rev., 2012, 41, 26 L. Albertazzi, D. van der Zwaag, C. M. A. Leenders,
6089–6102.
R. Fitzner, R. W. van der Hofstad and E. W. Meijer, Science,
2014, 344, 491–495.
27 K. J. C. van Bommel, C. van der Pol, I. Muizebelt, A. Friggeri,
A. Heeres, A. Meetsma, B. L. Feringa and J. van Esch, Angew.
Chem., Int. Ed., 2004, 43, 1663–1667.
28 M. Guo, L. M. Pitet, H. M. Wyss, M. Vos, P. Y. W. Dankers
and E. W. Meijer, J. Am. Chem. Soc., 2014, 136, 6969–6977.
29 J. D. Hartgerink, E. Beniash and S. I. Stupp, Proc. Natl. Acad.
Sci. U. S. A., 2002, 99, 5133–5138.
30 M. K. Mueller and L. Brunsveld, Angew. Chem., Int. Ed., 2009,
48, 2921–2924.
5 X. Du, J. Zhou and B. Xu, Chem. – Asian J., 2014, 9, 1446–1472.
6 S. I. Stupp, Nano Lett., 2010, 10, 4783–4786.
7 D. Bardelang, M. B. Zaman, I. L. Moudrakovski, S. Pawsey,
J. C. Margeson, D. Wang, X. Wu, J. A. Ripmeester, C. I. Ratcliffe
and K. Yu, Adv. Mater., 2008, 20, 4517–4520.
8 S. Dong, B. Zheng, F. Wang and F. Huang, Acc. Chem. Res.,
2014, 47, 1982–1994.
9 P. A. Korevaar, C. J. Newcomb, E. W. Meijer and S. I. Stupp,
J. Am. Chem. Soc., 2014, 136, 8540–8543.
10 J. Li, Y. Gao, Y. Kuang, J. Shi, X. Du, J. Zhou, H. Wang,
Z. Yang and B. Xu, J. Am. Chem. Soc., 2013, 135, 9907–9914. 31 B. S. Kim, D. J. Hong, J. Bae and M. Lee, J. Am. Chem. Soc.,
11 J. Li, Y. Kuang, Y. Gao, X. Du, J. Shi and B. Xu, J. Am. Chem.
Soc., 2013, 135, 542–545.
12 P. W. J. M. Frederix, G. G. Scott, Y. M. Abul-Haija,
D. Kalafatovic, C. G. Pappas, N. Javid, N. T. Hunt, R. V. Ulijn
and T. Tuttle, Nat. Chem., 2015, 7, 30–37.
13 J. Arigon, C. A. H. Prata, M. W. Grinstaff and P. Barthelemy,
Bioconjugate Chem., 2005, 16, 864–872.
14 N. Sreenivasachary and J. M. Lehn, Proc. Natl. Acad. Sci.
U. S. A., 2005, 102, 5938–5943.
2005, 127, 16333–16337.
32 J. Boekhoven, J. M. Poolman, C. Maity, F. Li, L. van der Mee,
C. B. Minkenberg, E. Mendes, J. H. van Esch and
R. Eelkema, Nat. Chem., 2013, 5, 433–437.
33 (a) J. M. Poolman, J. Boekhoven, A. Besselink, A. G. L. Olive,
J. H. van Esch and R. Eelkema, Nat. Protoc., 2014, 9,
977–988; (b) R. Eelkema and J. H. van Esch, Org. Biomol.
Chem., 2014, 12, 6292–6296.
34 A. G. L. Olive, N. H. Abdullah, I. Ziemecka, E. Mendes,
R. Eelkema and J. H. van Esch, Angew. Chem., Int. Ed., 2014,
53, 4132–4136.
15 S. M. Park and B. H. Kim, Soft Matter, 2008, 4, 1995–1997.
16 S. Matsumoto, S. Yamaguchi, S. Ueno, H. Komatsu,
M. Ikeda, K. Ishizuka, Y. Iko, K. V. Tabata, H. Aoki, S. Ito, 35 C. Maity, W. E. Hendriksen, J. H. van Esch and R. Eelkema,
H. Noji and I. Hamachi, Chem. – Eur. J., 2008, 14,
Angew. Chem., Int. Ed., 2015, 54, 998–1001.
3977–3986.
36 J. H. van Esch, Langmuir, 2009, 25, 8392–8394.
17 P. K. Vemula, J. Li and G. John, J. Am. Chem. Soc., 2006, 128, 37 W. Xi, T. F. Scott, C. J. Kloxin and C. N. Bowman, Adv. Funct.
8932–8938. Mater., 2014, 24, 2572–2590.
18 G. Wang, S. Cheuk, H. Yang, N. Goyal, P. V. N. Reddy and 38 W. Tang and M. L. Becker, Chem. Soc. Rev., 2014, 43,
B. Hopkinson, Langmuir, 2009, 25, 8696–8705. 7013–7039.
19 X. Du, J. Li, Y. Gao, Y. Kuang and B. Xu, Chem. Commun., 39 J. Voskuhl, M. C. A. Stuart and B. J. Ravoo, Chem. – Eur. J.,
2012, 48, 2098–2100. 2010, 16, 2790–2796.
20 X. Li, Y. Kuang, H.-C. Lin, Y. Gao, J. Shi and B. Xu, Angew. 40 R. Shimonkevitz, S. Colon, J. W. Kappler, P. Marrack and
Chem., Int. Ed., 2011, 50, 9365–9369. H. M. Grey, J. Immunol., 1984, 133, 2067–2074.
21 Z. Yang, Y. Kuang, X. Li, N. Zhou, Y. Zhang and B. Xu, Chem. 41 J. Karttunen and N. Shastri, Proc. Natl. Acad. Sci. U. S. A.,
Commun., 2012, 48, 9257–9259. 1991, 88, 3972–3976.
22 J. Zhou, X. Du, Y. Gao, J. Shi and B. Xu, J. Am. Chem. Soc., 42 K. L. Kiick, E. Saxon, D. A. Tirrell and C. R. Bertozzi, Proc.
2014, 136, 2970–2973. Natl. Acad. Sci. U. S. A., 2002, 99, 19–24.
23 M. M. C. Bastings, S. Koudstaal, R. E. Kieltyka, Y. Nakano, 43 K. L. Kiick and D. A. Tirrell, Tetrahedron, 2000, 56,
A. C. H. Pape, D. A. M. Feyen, F. J. van Slochteren, P. A. 9487–9493.
Doevendans, J. P. G. Sluijter, E. W. Meijer, S. A. J. Chamuleau 44 S. I. van Kasteren, H. B. Kramer, D. P. Gamblin and
and P. Y. W. Dankers, Adv. Healthcare Mater., 2014, 3, 70–78. B. G. Davis, Nat. Protoc., 2007, 2, 3185–3194.
24 J. Leckie, A. Hope, M. Hughes, S. Debnath, S. Fleming, 45 S. I. van Kasteren, H. B. Kramer, H. H. Jensen, S. J. Campbell,
A. W. Wark, R. V. Ulijn and M. D. Haw, ACS Nano, 2014, 8,
J. Kirkpatrick, N. J. Oldham, D. C. Anthony and B. G. Davis,
9580–9589.
Nature, 2007, 446, 1105–1109.
This journal is ©The Royal Society of Chemistry 2016
J. Mater. Chem. B